chromium geochemistry of serpentinized ultramafic
Post on 15-Mar-2022
13 Views
Preview:
TRANSCRIPT
CHROMIUM GEOCHEMISTRY
OF
SERPENTINITES AND SERPENTINE SOILS
A DISSERTATION
SUBMITTED TO THE DEPARTMENT OF
GEOLOGICAL amp ENVIRONMENTAL SCIENCES
AND THE COMMITTEE ON GRADUATE STUDIES
OF STANFORD UNIVERSITY
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
Christopher John-Paul Oze
August 2003
copy Copyright by Christopher John-Paul Oze 2003
All Rights Reserved
ii
I certify that I have read this dissertation and that in my opinion it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy
____________________________ Dennis K Bird Principal Adviser
I certify that I have read this dissertation and that in my opinion it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy
____________________________ Scott Fendorf Principal Adviser
I certify that I have read this dissertation and that in my opinion it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy
____________________________ Robert Coleman
Approved for the University Committee on Graduate Studies
iii
ABSTRACT
Chromium persists in the environment as Cr(III) a non-hazardous species and a
micronutrient in human nutrition or as Cr(VI) a strong oxidizing agent a toxin to living
cells and a Class A human carcinogen by inhalation Serpentinites or metamorphosed
ultramafic rocks are distributed worldwide and contain Cr concentrations typically
greater than 200 mg kg-1 Additionally serpentine soils a generic term used to describe
any soil derived from serpentinite regardless of its physical or chemical properties
contain elevated Cr concentrations often exceeding their corresponding protolith
Chromium in serpentinites and serpentine soils is dominantly in the form of Cr(III)
however geochemical interactions during serpentinite weathering and soil formation
provide oxidative pathways allowing the formation of Cr(VI) The relationship between
rock and soil and the processes governing the Cr chemistry are difficult to assess due to
the inherent complexity of evaluating and correlating the mineralogy and geochemistry of
both systems Ultimately a multidisciplinary approach combining field analytical and
laboratory studies is necessary to evaluate the Cr geochemistry related to these rocks and
soils and to determine whether or not serpentinites and serpentine soils are sources of
non-anthropogenic Cr contamination
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of
California produces serpentine soils containing high concentrations of Cr as well as other
potentially toxic elements including Ni Co and Mn Chromium concentrations in
serpentine soils from Jasper Ridge Biological Preserve in the Central Coast Range are as
high as 4760 mg kg-1 nearly three times greater than the serpentinite protolith (~1800
mg kg-1) Chromium-containing minerals within the bedrock include chlorite (~03 Cr
Wt ) enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt )
magnetite (82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate
phase containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
iv
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
Despite the resistance of chromite to weathering Cr release and oxidation from
chromite is a potential environmental hazard in sediments and soils and a pathway for
soil development (weathering of primary minerals) related to ultramafic rocks and their
metamorphic derivatives (serpentinites) Birnessite is a common pedologic mineral in
these sediments and soils capable of oxidizing aqueous Cr(III) The interaction between
chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4) and birnessite was investigated with
an interest in the potential generation of Cr(VI) The rate of Cr(VI) in solution increases
with increasing chromite suspension densities and decreasing pH but is independent of
birnessite suspension densities at values greater than ~20 m2 L-1 at 25degC The overall rate
expression of Cr released and oxidized in solution from chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression--rates ranging from 9times10-4 to 44times10-3 microM Cr(VI) h-1 These
experiments therefore demonstrate that serpentine soils are a potential source of non-
anthropogenic Cr(VI)
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
v
chemistry of Cr in serpentine soils including its protolith is reviewed with a focus on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils have widely varying pH contain a variety Fe(III) oxides (magnetite and
hematite) phyllosilicates (serpentine and chlorite) and clays (smectites and
vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1) and
Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Additionally Cr(III) is the only valence state
observed in the serpentine soil solids however Cr(VI) has been identified in New
Caledonia and California serpentine soil solutions at concentrations below 30 microM The
enrichment and range of Cr concentrations in serpentine soils are directly related to the
presence of Cr-spinels specifically chromite and Cr-magnetite These spinels are
resistant to weathering and are preserved in the soil environment however oxidation of
Cr(III) from Cr-spinels by high valent Mn oxides or other strong oxidants is a potential
source of Cr(VI) identified in serpentine soil solutions Due to the chemically and
physically resistant nature of the Cr-spinels Cr-bearing silicates including clay minerals
Cr-chlorite Cr-garnet Cr-mica and Cr-epidote are more viable sources of Cr identified
in vegetation soil extractions soil solutions and related waters
vi
PREFACE
The chemistry and mineralogy of Cr related to serpentinites and serpentine soils
are assessed in this dissertation using a multidisciplinary field analytical experimental
and theoretical approach The four chapters that constitute this dissertation describe my
investigations related to these rocks and soils Additionally the common theme shared
by these chapters is that they all examine the Cr mineralogy in serpentinites and
serpentine soils and the potential of Cr to be mobilized and oxidized in near-surface
environments Chromite a Cr-bearing spinel is the dominant source of Cr in
serpentinite and serpentine soils Although chromite is resistant to weathering the
presence of birnessite a high valent Mn oxide will increase the release of Cr and the
production of Cr(VI) from chromite over a range of pH Chromium-clinochlore or
kaumlmmererite contains Cr concentrations less than 3 Wt however the rate at which Cr-
clinochlore weathers compared to chromite makes this mineral a more probable source of
Cr released into soil solutions These types of concerns are addressed in this thesis Brief
descriptions of the four individual dissertation chapters including their publication status
and a list of the co-authors are presented in the following paragraphs
Chapter 1 presents the main objectives of this dissertation The chemistry
regulations and toxicity of Cr are discussed as well as the general characteristics of
serpentinites and serpentine soils Additionally a brief summary of each chapter is
provided
Chapter 2 focuses on serpentinites and serpentine soils at Jasper Ridge Biological
Preserve Stanford CA The results have been submitted for publication in an article
entitled ldquoChromium geochemistry of serpentinized ultramafic rocks and serpentine soils
in the Franciscan Complex of Californiardquo to the American Journal of Science (March
2003) Scott Fendorf Dennis Bird and Robert Coleman are the second third and fourth
co-authors respectively
Chapter 3 is an experimental study evaluating the release of Cr and the
production of Cr(VI) from chromite in the presence of birnessite over a range of pH The
results for this study will be submitted for publication in an article entitled ldquoOxidative
promoted dissolution of chromite by manganese dioxide and concurrent production of
vii
chromaterdquo to Environmental Science and Technology The co-authors for this paper are
Scott Fendorf and Dennis Bird
Chapter 4 is a review paper that critically evaluates the chemistry of serpentine
soils and serpentinites primarily focusing on Cr geochemistry Additionally analytical
studies of serpentine soils collected from California Oregon and New Caledonia are
presented This study will be presented at the Robert Coleman Symposium and has been
submitted as an article entitled ldquoChromium geochemistry of serpentine soilsrdquo for an
International Geology Review Special Publication Scott Fendorf Dennis Bird and
Robert Coleman are the second third and fourth co-authors respectively
Appendix A is a preliminary investigation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite) a mineral common in serpentinites and present in serpentine
soils This appendix is entitled ldquoEstimation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite)rdquo The results of this chapter will be included in a study that
estimates the thermodynamic properties of Cr-bearing silicates including fuchsite (Cr-
muscovite) tawmawite (Cr-epidote) and uvarovite (Cr-garnet)
viii
ACKNOWLEDGEMENTS
Serpentinites serpentine soils and chromium are the Geologic Trinity of this
dissertationhellipeach topic is intimately related to one another The scope and breadth of
this research required a significant amount of time and funding None of this research
would have been possible without the generous support of several foundations
organizations and Stanford University I am grateful for the financial aid I have
received from the Stanford Fellowship the Stanford University McGee and Shell funds
the National Science Foundation the Geologic Society of America Graduate Research
Fund the Stanford Office of Teaching and Learning grant and the Jasper Ridge
Biological Preserve Melon Grant
I would not be where am I today if it were not for my parents Robert and Sue
Oze and my brother Rob Oze Thank you for the sandbox and all the hose-driven
volcanoeshellipthose Wookies never stood a chance While some families vacation in theme
parks my family frequented gem and mineral conventions collected rocks and shells
from mist-shrouded shores and sought lucky stones including an infamous quartz orb
Sometimes I think I was destined to be a geologist I would also like to thank everyone
in my family and all our family friends who have provided that extra little push and
support along the way These people include Tudor Bennatts Louis Oze Margaret and
John Chambers John Bennatts Nancy Skiles the Izumis the memory of Witchie Poo
hijacking Harleys from Bremerton bikers and the Man in the Sound waiting for that one
last catch Close friends I would like to acknowledge who are not related to the Stanford
crowd include Dave and Leigh Anderson and Rebecca Parker These satellite friends
helped me keep in touch with the world outside of academia and proved to me that I
wasnrsquot meant to have a real job A special thank you goes to Katherine Akako
Izumihellipyou have made this whole trip worthwhile
My PhD advisors Scott Fendorf and Dennis Bird provided an invaluable amount
of help patience and energy over the past five years Scott always made the time to
discuss and refine many of my projects making them relevant to many current
geochemical issues Additionally he helped me hone my soil science skills in the field
and in the lab Out in the field Scott can dig a pretty mean hole Dennis devoted a
significant amount of time and sweat in order to make this dissertation happen His
ix
insightful knowledge of geochemistry and of the geology of the Sierra Nevadas provided
the cornerstone for this dissertation
Robert Coleman played a key role in the production of this dissertation He
always knew what to say at the right time and helped me understand the complexity of
ophiolites and serpentinites through experienced eyes Additionally Keith Loague and
Gary Ernst deserve recognition for their outstanding support as committee members for
this dissertation over the last four years I would also like to thank Jonathan Stebbins and
Gail Mahood for providing me the opportunity to attend Stanford
Bob and Vicki Jones always offered an open ear whether it was science or solving
the worldrsquos problems Rick Hazlett (Pomona College) and John Winter (Whitman
College) both had a significant impact on shaping my future science goals and objectives
during my undergrad years and both encouraged me to pursue a career in geology They
were the catalyst for this whole experiment
I owe a lot to my research group members (both the Bird Cage and Soil
Chemistry Group) and those people who didnrsquot mind me tagging along Before I
introduce these people one person has been my friend and co-conspirator for many years
at Stanford Tim Copehellipyou are a pain in my ass Besides never picking up any of your
crap and breaking all my stuff this whole grad school experience would not have been
the same without you Other friends and acquaintances that will not be forgotten in
chronological order of me walking into my office for the first time on September 11
1998 include Anne Dyreborg Esra Inan Tom Bawden Patrick Redmond Phil
Neuhoff Thrainn Fredrickson Lara Heister Kaye Savage Andrea Fildani Mike Schultz
Trayle Kushan Kevin Theissen Angela Hessler Chris Van de Ven Kono Lemke Colin
Doyle Colleen Hansel Ben Bostick Matthew LaForce Nancy Grumet Scott Kroeker
Anders Meibom Brian Ebel Matt Ginder-Vogel Matt Pollizitto Spud Amy Weislogel
Travis Horton and Jake Waldbauer I would like to end this Acknowledgement section
and begin my dissertation with a haikuhellip
Better than outcrops
Are the people Irsquove met here
Thanks for all the beer
x
TABLE OF CONTENTS
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iv Prefacehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip vii Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip ix Table of Contentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xi List of Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xiv List of Illustrationshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xvi Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 1
Toxicity and Regulations of Chromiumhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 2 Mineralogy of Chromium in Serpentiniteshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 3 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5 Chapter Summarieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5
Chapter 2 Chromium Geochemistry of Serpentinized Ultramafic Rocks and Serpentine Soils in the Franciscan Complex of Californiahelliphelliphelliphelliphelliphellip 8
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 8 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 9 Backgroundhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 10
Geologic Background of the Serpentinite at Jasper Ridgehelliphelliphelliphellip 10 Serpentine Soils Characteristics and Reviewhelliphelliphelliphelliphelliphelliphelliphellip 11 Sources of Chromium ProtolithBedrockhelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 12 Chromium-Bearing Phases in Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 13
Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Location and Sample Collectionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Mineral Identification in Rocks and Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Soil Chemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 15 X-ray Diffraction and Clay-Size Mineral Identificationhelliphelliphelliphelliphellip 16 X-ray Absorption Spectroscopyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17 Sequential Extraction Procedurehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17
Resultshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Serpentinite Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Distribution of Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip21 Serpentine Soil Chemistry and Mineralogy at Site JR3helliphelliphelliphelliphellip 22 Sequential Extractionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 26
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 27 Concluding Remarkshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 33 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 44 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 47
xi
Chapter 3 Oxidative Promoted Dissolution of Chromite by Manganese Dioxide and Concurrent Production of Chromatehelliphelliphelliphelliphelliphellip 63
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 63 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 64 Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 65
Solid Phase Characterization and Synthesishelliphelliphelliphelliphelliphelliphelliphelliphellip 65 Extraction Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 66 Batch Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Solution Analysishelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69
Results helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite Extraction Experimentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite and Birnessite Batch Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 70 Secondary and Solid Phase Examination of Chromite and Birnessitehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 72 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 72
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 73 Chromium(VI) Pathways from Chromite and Birnessite into Solutionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 76 Serpentine Soils and Cr(VI)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 77
Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 79 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 80 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 84 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 87
Chapter 4 Chromium Geochemistry of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip100
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 100 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 101 Reviewhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102
Chromium Geochemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102 Mineralogy of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 103 Sources of Chromium for Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphellip 105 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 106 Soils Solutions and Related Waterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 108 Serpentine Soil Vegetationhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 110 Soil Organic Matterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 111
California Oregon and New Caledonia Serpentine Soilshelliphelliphelliphelliphelliphelliphellip 112 Sample Locations and General Soil Characteristicshelliphelliphelliphelliphelliphellip 112 BedrockProtolith Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 113 Soil Chemistry and Physical Propertieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 114 Soil Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116 Chromium and Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 117
xii
Chromium Spinel in Soils and Rockshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 119 Concluding Statementhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 120 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 121 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 122 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 129 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 131
Appendiceshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 141 Appendix Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 142 Appendix Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 143 Appendix A Estimation of the Thermodynamic Properties of Kaumlmmererite (Cr-Clinochlore)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip145
Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 145 Nomenclature and Crystal Chemistry of Cr-chloritehelliphelliphelliphelliphelliphelliphelliphelliphellip 146 Estimation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 147
Estimation of theStandard Molar Volume Using Unit Cell Measurementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of the Standard Molar Volume Heat Capacity and Entropyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of Standard Molar Enthalpy and Gibbs Free Energyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 150 Critique and Discussion of the Estimation Algorithms and SUPCRT92helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 151
Evaluation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 152 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 155 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 161 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 162
List of References
References to Chapter 1helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 166 References to Chapter 2helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 172 References to Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 179 References to Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 183 References to Appendix Ahelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 195
xiii
LIST OF TABLES Chapter 2 Table 21 Whole rock elemental concentrations of serpentinites chert and sandstone Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in
serpentinite rock at Jasper Ridge Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in
serpentine soil at Jasper Ridge Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25
and 26 Table 24 Bulk soil elemental concentrations for soils and bedrock collected along
the transect in Figure 21b Table 25 Chemical and physical properties of the soil at Site JR3 Table 26 Selected trace metal concentrations for each size fraction versus depth at
Site JR3 Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3 Table 28 X-ray diffraction d-spacings for common silicates Table 29 Results of the Diablo clay serpentine soil and chromite sequential
extraction experiments Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1
kg of sample Chapter 3 Table 31 Chromite and birnessite masses and concentrations used for each batch
experiment Table 32 Serpentine soil and birnessite experimental parameters Table 33 Summary of Cr(VI) rates obtained from batch experiments Table 34 Comparative analyses of batch experiment solutions and chemical
extractions of chromite and birnessite solids Chapter 4 Table 41 Measured Cr concentrations in serpentine soils worldwide Table 42 Summary of Cr Fe and Ni extractions from serpentine soils Table 43 Serpentine soil plant Cr concentrations Table 44 Chemical and textural properties of collected soils Table 45 Total major and trace element concentrations of the collected soils Table 46 Compositions for soil Cr-containing minerals identified in Figure 49 Table 47 Chromium-bearing mineral masses and volumes required to influence Cr
concentrations in soils Appendix A Table A1 Cr and Al end-members in the Cr-clinochlore solid solution
xiv
Table A2 Standard state thermodynamic properties of Cr and Al-bearing minerals Table A3 Unit cell measurements and molar volumes of clinochlore and Cr-
clinochlores Table A4 Entropy volume and heat capacity for kaumlmmererite using the Ransom and
Helgeson Method Table A5 Fictive polyhedral considered in the estimations Table A6 Estimated thermodynamic properties for kaumlmmererite
xv
LIST OF FIGURES Chapter 2 Figure 21a Soils map of Jasper Ridge Biological Preserve CA Figure 21b Serpentinite and serpentine soil sampling sites at Jasper Ridge Biological
Preserve CA Figure 22 Chromium concentrations versus depth for four serpentine soils in
Maryland California and New Foundland Figure 23 Petrographic images of serpentinite at Jasper Ridge Figure 24 Elemental concentrations relative to Cr for chlorite pyroxene magnetite
chromite and a chromite-silicate mixture in the serpentinite at Jasper Ridge
Figure 25 Backscattering electron images of minerals in the serpentinite at Jasper Ridge
Figure 26 Backscattered electron images of Cr-chlorite and Cr-magnetite assemblage in the serpentinite bedrock
Figure 27 Highest and lowest Cr concentrations at each site in the serpentine soil transect
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected for total gt2 mm sand silt and clay size fractions
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soils at Jasper Ridge
Figure 210 X-ray diffraction clay analyses of sample JR3515 Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at specific depths Figure 212 X-ray microprobe elemental maps of Cr and Fe for the serpentine soil and
serpentinite at site JR3 Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments Figure 214 Scanning electron microprobe images of the chromite standard after each
sequential extraction experiment Figure 215 Tie-lines connecting serpentinite chromite and end-member chromite with
soil chromite values Chapter 3 Figure 31 Scanning electron microprobe images of chromite and birnessite standards Figure 32 Chromite extraction experiments evaluating Cr Fe and Al Figure 33 pH and Eh versus time (hours) for each chromite-birnessite chromite and
birnessite batch experiments Figure 34 pH and Eh with Cr stabilities Figure 35 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant chromite and variable birnessite at pH 5 Figure 36 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant birnessite and variable chromite at pH 5
xvi
Figure 37 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for variable pH and constant chromite and birnessite Figure 38 Rate of Cr(VI) production (d[Cr(VI)]dt) as a function of the surface area
concentration of chromite birnessite and pH Figure 39 SEM examination of solids following the termination of the batch
experiments Figure 310 Serpentine soil and birnessite experiments Figure 311 Iron analyses and the solubility of amorphous Fe(OH)3 and goethite Figure 312 Congruent dissolution of chromite based on Fe and Cr Figure 313 Three pathways of Cr release and oxidation Chapter 4 Figure 41 Serpentine soil localities worldwide Figure 42 Field site pictures of serpentine soils Figure 43 Stability field diagram of Cr over a range of Eh and pH conditions Figure 44 Chromium and Al Wt plot and octahedral Cr and octahedral Al plots
for chromite and Cr-silicates including Cr mica (fuchsitemariposite) Cr-garnet (uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite)
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (m) for selected serpentine soils
Figure 46 pH of serpentine soils and related groundwater worldwide Figure 47 Soil pictures with Munsel color values and names from collected soils Figure 48 Cr XANES spectra for collected serpentine soils Figure 49 Backscattering images of mineral phases within each serpentine soil Figure 410 Chromite magnetite and hematite compositions from analyzed soils as
well as analyses for chromite reported from around the world Appendix A Figure A1 Compositional variation of Cr2O3 and Al2O3 Wt in Cr-chlorite Figure A2 Compositional variation of Cr and VIAl in Cr-chlorite based on 14 oxygens Figure A3 Molar volume estimation of kaumlmmererite based on XRD unit-cell
parameters Figure A4 Natural and calculated compositions of co-existing Cr-clinochlore and Cr-
spinel solid solution in the system Mg0-FeO-Cr2O3-Al2O3-SiO2-H2O at 2 kbar pressure
xvii
CHAPTER 1
INTRODUCTION
Serpentinites or metamorphosed ultramafic rocks are distributed worldwide and
are commonly associated with ophiolite complexes (Coleman 1977 Coleman and Jove
1992) These rocks weather to form serpentine soils a generic term used to describe any
soil derived from serpentinite regardless of its physical or chemical properties
Serpentinites and serpentine soils contain Cr-bearing minerals the most prevalent being
chromite and Cr released from these minerals into surface and subsurface environments
is potentially toxic and carcinogenic The main objective of this dissertation is to
evaluate the mineralogy and chemistry of Cr related to serpentinites and serpentine soils
over geologic time using an integrated analytical experimental and theoretical approach
Additionally the potential of serpentinites and serpentine soils as sources of non-
anthropogenic Cr contamination is investigated A discussion of Cr and its significance
to serpentinites and serpentine soils begins with its discovery by a Frenchman
investigating an ore named ldquored leadrdquo from the Urals
Nicolas-Louis Vauquelin discovered Cr in 1797 in the mineral crocoite PbCrO4
(Burns and Burns 1975) The metalelement was named chromium after the Greek word
chroma meaning color due to its strong influence on mineral pigmentation (Burns and
Burns 1975) For example rubies and emeralds derive their color from trace amounts of
Cr in their structures More importantly JT Lowitz in 1798 discovered Cr in chromite
or Cr-spinel the primary source of Cr for industrial purposes (Burns and Burns 1975)
Chromium is a group VIB transition metal with two common oxidation states (Cr(III) and
Cr(VI)) an average crustal abundance of 100 mg kg-1 and an abundance of 513x105 in
the solar system (relative to [H]=1x1012) (Mason and Moore 1982)
The majority of Cr on the earth exists as Cr(III) where it is an immobile cation
that complexes strongly with organic matter and adsorbs on oxides and silicates even at
low pHs (pHgt4) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) Additionally
the chemical mobility of Cr(III) is limited due to the formation of highly stable
(oxy)hydroxides (Rai and Zachara 1988 Rai et al 1989 Ball and Nordstrom 1998)
1
Chromium(III) has the highest octrahedral site preference energy (OSPE) of any trace
element and it isomorphically substitutes into octahedral sites in oxides
(oxy)hydroxides sulfides and silicates Although Cr(III) is known to substitute for both
octahedral Al(III) and Fe(III) the strongest correlation for substitution appears to be for
Al(III) in oxides and silicates as demonstrated in Chapters 2 and 4 and Appendix A
Chromium(VI) is a strongly oxidizing and weakly adsorbing oxyanion in the form of
chromate (HxCrO42-x) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995 Shiraki
1997 Ball and Nordstrom 1998) Consequently Cr(VI) is chemically mobile and
bioavailable over a wide range of pH (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Chromium(VI) is tetrahedrally coordinated and behaves similarly to
sulfate in minerals such as jarosite (Baron and Palmer 1996)
The pathways and chemistry of Cr oxidation and reduction in soils and related
waters are covered in depth by Fendorf (1995) and Ball and Nordstrom (1998) The
reduction of Cr(VI) to Cr(III) may occur by reactions with Fe(II)aq Fe(II)-containing
minerals such as magnetite organic matter H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996) Oxidation of Cr(III)
to Cr(VI) related to serpentinite-related water and serpentine soil solutions is still a matter
of conjecture Chromium(III) oxidation in serpentine soils has been hypothesized to be
the result of Mn oxides such as birnessite as suggested by both Gough et al (1989) and
Becquer et al (2003) This idea is explored in depth in Chapter 3 where birnessite is
shown to influence the release and oxidation of Cr from chromite over a range of pH (3
to 8) at 25oC
Toxicity and Regulations of Chromium
The examination of Cr based on its oxidation state is necessary due to the
juxtaposing behavior and toxicity of Cr(III) and Cr(VI) Chromium(III) is an essential
micronutrient in human nutrition used by the body to process sugar protein and fat
(Vincent 2000) Inhalation of Cr(VI) is known to cause irritation in the nose including
symptoms such as nosebleeds (epistaxis) and ulcers andor perforations in the nasal
septum (Katz and Salem 1993 Kuo et al 1997) Ingesting considerable quantities of
Cr(VI) can result in stomach ulcers convulsions kidney and liver damage and death
2
(Katz and Salem 1993) Skin contact with certain Cr(VI) compounds causes allergic
contact dermatitis (ulcers) and irritant dermatitis (Katz and Salem 1993) Chromate
induced ulcers or chromeholes are typically red swollen crusted painless lesions
(Katz and Salem 1993) The nosology of Cr is beyond the scope of this dissertation
however Lees (1991) is an excellent source for further information Overall the
Environmental Protection Agency (EPA) the Occupational Safety and Health
Administration (OSHA) the World Health Organization (WHO) and the Department of
Health and Human Services (DHHS) have determined that that Cr(VI) compounds can
increase the risk of lung cancer primarily by inhalation as well as other forms of cancer
caused by ingestion
Federal and state regulation standards appropriate for both Cr(III) and Cr(VI) are
summarized and discussed by Proctor et al (1997) Human health action levels for Cr
vary for each state and the ranges for total Cr in soils include the following 200-66000
Cr(III) mg kg-1 (residential) 02-2000 Cr(VI) mg kg-1 (residential) 2500-1000000
Cr(III) mg kg-1 (non-residential) and 64-71000 Cr(VI) mg kg-1 (non-residential)
(Proctor et al 1997) Incidentally Cr concentrations in serpentine soils and serpentinites
(gt200 Cr mg kg-1) exceed many state action levels and would be considered toxic The
EPA has set a limit of Cr(III)+Cr(VI) at 100 microg L-1 in drinking water The OSHA has
set limits of 500 microg of water soluble Cr(III) compounds per cubic meter of workplace air
or 500 microg m-3 1000 microg m-3 for metallic Cr and insoluble Cr compounds and 52 microg m-3
for Cr(VI) compounds for 8-hour work shifts and 40-hour work weeks The limits
proposed above are constantly changing and should not be used for reference without
checking current regulation standards
Mineralogy of Chromium in Serpentinites
Chromium occurs in a variety of minerals either as a major or minor constituent
(Burns and Burns 1975) Chromium-bearing minerals in serpentinites are primarily
oxides and silicates in which the Cr is present as Cr(III) Chromite is a Cr-bearing spinel
abundant in serpentinites and is resistant to low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) The chemistry and
significance of chromite is addressed throughout this dissertation Pyroxenes such as
3
augite and enstatite are primary minerals identified in serpentinites and contain lt8 Cr Wt
(Grapes 1981 Von Knorring et al 1986 Treloar 1987a Challis et al 1995
Sanchez-Vizcaino et al 1995 Deer et al 1996) Chromate minerals such as crocoite
(PbCrO4) are rare in serpentinites and the occurrence of these minerals are limited to Pb
deposits in Australia Tasmania and the Urals (Williams 1974 Crane et al 2001)
Chromium(III) released from the primary Cr-containing phases may
isomorphically substitute into oxides and silicates for Al(III) and Fe(III) due to size
(octahedral radii Cr3+ = 0615 Aring Al3+ = 061 Aring and Fe3+=0643 Aring) and charge
similarities Magnetite formed during serpentinization incorporates Cr into its structure
and serpentine minerals (antigorite lizardite and chrysotile) contain trace concentrations
of Cr (Spangenberg 1943 Page and Coleman 1967 Burkhard 1993) Eskolaite (Cr2O3)
has been identified in Finland (Von Knorring et al 1986) Metasomatic alteration of
ultramafic rocks commonly results in the formation of Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite) uvarovite (Cr-garnet) tawmawite (Cr-epidote) and
kaumlmmererite (Cr-chlorite) (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Neiva 1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981
Max et al 1983 Nutman et al 1983 Raase et al 1983 Chao et al 1986 Von
Knorring et al 1986 Kerrich et al 1987 Treloar 1987b Treloar 1987a Morand 1990
Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993 Christofides et al 1994
Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Proenza et al
1999) Additionally kyanite and andalusite in the proximity of ophiolite complexes
incorporate up 5 Cr Wt (Cooper 1980 Raase et al 1983 Kerrich et al 1987)
Chapter 4 provides an in depth discussion of these Cr-bearing silicates as well as
demonstrating mineral compositions reported in the literature based on octahedral Cr and
Al Appendix 1 is a preliminary investigation devoted to estimating the thermodynamic
properties of kaumlmmererite These properties can then be incorporated into geochemical
modeling programs to begin accessing the stability of this common serpentinite mineral
over a range of temperatures and pressures
4
Chromium in Serpentine Soils
Evaluating Cr in serpentine soils requires an examination of both the mineralogy
and chemistry of the serpentinite and soil Measured Cr concentrations in serpentinites
and serpentine soils range from ~200 mg kg-1 to 6 Wt all of which is Cr(III) (Soane
and Saunder 1959 Proctor and Woodell 1971 Shewry and Peterson 1976 Jaffre 1980
Proctor et al 1980 Rabenhorst et al 1982 Schwertmann and Latham 1986 Brooks
1987 Gough et al 1989 Gambi 1992 Lee 1992 Morrey et al 1992) In Chapters 2
and 4 chromite and Cr-magnetite are identified as the most prevalent sources of Cr
responsible for Cr-enrichment in these soils Chromite is resistant to chemical
weathering and it has been observed to undergo incongruent dissolution conserving Cr in
serpentine soils Chromium is tightly bound in the matrix of the soil less than 1 of the
total Cr can is removed using standard soil extraction methods (10 hydrogen peroxide
(H2O2) solution 01 M potassium permanganate (KMnO4) solution10 M hydrofluoric
(HF) acid 1 M hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and 1
mM barium chloride (BaCl2) solution) due to the stability of chromite Although silicates
contain minor concentrations of Cr compared to chromite they are perhaps the main
contributors of Cr identified in plants (lt200 Cr mg kg-1) and soil solutions (lt50 microg L-1)
During soil-forming processes Cr released to soil solutions by the dissolution of Cr-
bearing minerals are sequestered into a host of structural and surface sites available in
authigenic Fe(III) oxides Fe(III) hydroxides and clay minerals (Schwertmann and
Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990 Gerth 1990 Gasser and
Dahlgren 1994 Ilton and Veblen 1994 Brigatti et al 2000 Ilton et al 2000) The
clay-sized fractions in soils at Jasper Ridge Biological Preserve CA contain Cr
concentrations as high as 1000 Cr mg kg-1
Chapter Summaries
A detailed field and analytical examination of Cr in serpentinites and serpentine
soils at Jasper Ridge is presented in Chapter 2 A majority of the Cr enrichment and
variability with depth in these soils is the result of small grains of chemically resistant
chromite from the serpentinite protolith Chromite identified in the soil undergoes
incongruent dissolution conserving Cr relative to chromite compositions identified in the
5
serpentinite bedrock Sequential extraction experiments using barium chloride (BaCl2)
hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and hydrofluoric (HF)
acid were performed on a chromite standard confirming that Cr-bearing spinels are
resistant to chemical weathering X-ray diffraction analyses of the clay-size fraction in
the serpentine soils identified Cr-clinochlore smectite vermiculite and serpentine with
the smectite being the most abundant mineral in the clay-size fraction Chromium(III)
was the only identified oxidation state of Cr in the serpentine soil solids and in
serpentinites at Jasper Ridge in the solid phase This chapter represents a comprehensive
site assessment identifying the nature and behavior of Cr in serpentine soils with respect
to the bedrock mineralogy and depth
Chromite an abundant Cr-enriched spinel is responsible for a majority of the Cr
enrichment identified in ultramafic rocks serpentinites and serpentine soils The release
and oxidation of Cr from chromite is a pathway to consider when evaluating the potential
of ultramafic rocks and their related sediments and soils as sources of Cr(VI) In
Chapter 3 experiments aimed at quantifying the release and oxidation of Cr from
chromite in the presence of birnessite are described Chromite is extremely resistant to
dissolution (lt8 Cr mg kg-1 is released) when immersed in solutions including 1 mM
barium chloride solution 10 M hydrofluoric acid 10 hydrogen peroxide and 1 M
hydroxylamine hydrochloride with 25 acetic acid Highly oxidizing solutions of 01 M
potassium permanganate increased the rate and release of Cr relative to the other
solutions listed above thus birnessite a high valence Mn oxide present in serpentine
soils was chosen to determine if Mn could accelerate Cr release and oxidation from
chromite The results presented in Chapter 3 demonstrate that birnessite increases the
rate of Cr(VI) observed in solution from chromite This production of Cr(VI) is
dependent on the chromite suspension density where birnessite surface are concentrations
are greater than 20 m2 L-1 between pHs 3 and 8 at ~25degC
A comprehensive review of the chemical characteristics of serpentine soils
including topics such as the mineralogy of serpentine soils soil solutions organic matter
and vegetation is provided within Chapter 4 Additionally chemical and spectroscopic
analyses were performed on serpentine soils from a) New Caledonia b) Eight Dollar
Mountain Selma OR c) Nickel Mountain Riddle OR d) Harvard Mine Jamestown
6
CA e) Jasper Ridge Biological Preserve Stanford CA and f) Pillikin Mine El Dorado
CA Serpentinites and serpentine soils associated with ophiolite complexes cover ~1 of
the earthrsquos exposed surface and they are correlated to modern and ancient convergent
boundaries The physical and chemical characteristics of serpentine soils are directly
related to the alteration and tectonic history of the serpentinite as well as the soil-forming
factors including the parent material climate biota topography and time Serpentine
soils have Cr concentrations exceeding 200 mg kg-1 however the concentration and the
availability of Cr in these soils are not correlated with the five soil-forming factors
(climate organic material parent material topography and time) or the soil chemistry
Although only Cr(III) was identified in the soil solids Cr(VI) has been identified in soil
solutions (lt30 microg L-1) from serpentine soils in New Caledonia and California Chromite
and Cr-magnetite grains are identified in the soils listed above and these minerals are
responsible for a majority of the Cr enrichment in serpentine soils
Kaumlmmererite is a Cr-chlorite associated with serpentinized ultramafic rocks such
as those collected at Jasper Ridge Harvard Mine and Pillikin Mine Although Cr
concentrations are typically 01-2 Wt in kaumlmmererite it is modally abundant in
serpentinites (up to 5 of the volume of the rock) and a potential source of Cr to be
released into soils and groundwater In Appendix A the standard molar thermodynamic
properties of the Cr-clinochlore end-member kaumlmmererite (Mg5Cr(AlSi3)O10(OH)8) is
estimated using established estimation algorithms This initial study provides the
foundation for submitting a future paper presenting the thermodynamic properties for
major Cr-bearing silicates including fuchsite (Cr-muscovite) uvarovite (Cr-garnet) and
tawmawite (Cr-epidote) and their metamorphic significance related to serpentinized
ultramafic rocks
7
CHAPTER 2
CHROMIUM GEOCHEMISTRY IN SERPENTINIZED ULTRAMAFIC ROCKS
AND SERPENTINE SOILS FROM THE FRANCISCAN COMPLEX OF
CALIFORNIA
ABSTRACT
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of California
produces serpentine soils containing high concentrations of Cr as well as other potentially
toxic elements including Ni Co and Mn Chromium concentrations in serpentine soils
from Jasper Ridge Biological Preserve in the Central Coast Range are as high as 4760
mg kg-1 nearly three times greater than the serpentinite protolith (~1800 mg kg-1)
Chromium-containing minerals within the bedrock include chlorite (~03 Cr Wt )
enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt ) magnetite
(82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate phase
containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
8
INTRODUCTION
Serpentine soils are formed from the weathering of ultramafic rocks and their
metamorphic equivalents (serpentinites Brooks 1987) Measured Cr concentrations in
these soils range as high as 80600 mg kg-1 and are enriched compared to non-serpentine
soils (7 to 221 mg kg-1 Proctor 1992 McBride 1994) The distribution and
concentration of Cr in serpentine soils worldwide is highly variable due to the chemical
composition and mineralogy of the parent material and soil forming factors including
climate biota topography precipitation and time (Hotz 1964 Rabenhorst et al 1982
Schwertmann and Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al
1989 Gough et al 1989 Kaupenjohann and Wilcke 1995) Potential Cr-bearing phases
in serpentine soils have been assessed using extraction experiments and these phases
include organic material silicates secondary oxides and spinels (Schwertmann and
Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994 Kaupenjohann and
Wilcke 1995 Becquer et al 2003) The weathering of these Cr-bearing phases over
geologic time and how these sources govern the distribution of Cr in serpentine soils have
not been evaluated in great detail
Chromium is a group VIB transition metal present in the environment as Cr(III) a
non-hazardous species and micronutrient andor as Cr(VI) a toxin to living cells and a
Class A human carcinogen by inhalation (Daugherty 1992 Cohen et al 1993 James et
al 1997) Chromium(III) is chemically immobile due to forming low solubility
(oxy)hydroxides and stable surface complexes whereas Cr(VI) is a weakly adsorbing
(oxy)anion and thus is mobile in most aquatic environments (McBride 1994 Fendorf
1995 Ball and Nordstrom 1998) Chromite and Cr-magnetite have been cited as
primary sources of Cr in serpentine soils (Rabenhorst et al 1982 Schwertmann and
Latham 1986 Alexander et al 1989 Gough et al 1989) Although Cr is primarily
found in the form of Cr(III) in both ultramafic rocks and serpentinites Cr(VI) has been
measured in serpentine soils in California and New Caledonia (Gough et al 1989
Becquer et al 2003) Due to the abundance of Cr and the potential for Cr(VI) formation
serpentine soils and serpentinites are potential sources of non-anthropogenic Cr
contamination
9
The objectives of this study are to determine the distribution and chemistry of Cr
in naturally Cr-enriched soils using an integrated field analytical experimental and
theoretical approach A wide array of techniques and methods including X-ray
absorption spectroscopy and chemical extractions were used to identify the nature of Cr
and its availability in serpentine soils Serpentinites and serpentine soils at Jasper Ridge
Biological Preserve located near Stanford CA were used to evaluate the geochemical
behavior of Cr These soils cover approximately 50000 m2 and contain high
concentrations of the trace elements Cr Ni and Mn The soils are dominantly Mollisols
with grassland vegetation and they are pedalogically similar to those associated with the
Central Coast Range of the Franciscan Complex in the western United States The main
serpentinite body at Jasper Ridge was exposed approximately 400000 years ago allowing
pedogenesis and Cr cycling to occur over a geologically relevant time period (Burgmann
et al 1994) These soils can then be used as a proxy for evaluating the long-term
behavior of Cr
BACKGROUND
Geologic Background of the Serpentinite at Jasper Ridge
Jasper Ridge Biological Preserve is located in the foothills of the Santa Cruz
Mountains in California and is part of the Central Coast Range of the Franciscan
Complex an Early Cretaceous accretionary meacutelange containing blocks of greywacke
greenstone blueschist and serpentinized ophiolite (Figure 21) Formation
metamorphism and emplacement of serpentinite at Jasper Ridge began approximately
130-150 million years ago during Jurassic subduction along the California continental
margin (Page and Tabor 1967) Thin wedges of ultramafic rock (peridotite) were
detached from the down-going slab of oceanic crust and incorporated into the subduction
meacutelange Subsequent serpentinization of the ultramafic rock in the down-going slab
resulted in the formation of serpentinites now exposed at Jasper Ridge
Rare blueschist blocks (1 to 5 meters in width) present in the serpentinite at Jasper
Ridge attest to the degree of metamorphism and the depths from which portions of the
tectonic meacutelange has been elevated Transformation of greenstone in the meacutelange into
blueschist required high pressures in excess of ~5-7 kbars (15-20 kilometers depth) and
10
temperatures of 200-300 degC (Ernst 1971) The low density of the serpentine (245-250 g
cm-3) allowed upward diapiric migration of the serpentinite body through the denser
enclosing tectonic meacutelange (265 g cm-3) Uplift of the present day Coast Range and
activity associated with the San Andreas fault system have shaped the current topography
and exposure of serpentinite at Jasper Ridge Serpentinite is currently present as a narrow
discontinuous band of pervasively sheared rock located along the ridge top (Figure 21)
The serpentinite is in fault contact with the Whiskey Hill Formation (Cretaceous turbidic
sandstone and mudstone) and tectonically interleaved with the Franciscan Complex
Serpentine Soils Characteristics and Review
Relative to other soil types serpentine soils are characterized by higher
concentrations of Cr Ni Co and Fe lower concentrations of plant nutrients such as Ca
K N and P lower CaMg ratios and characteristic flora and physical properties (Brooks
1987) The pH of serpentine soils ranges from ~4 up to 9 (Brooks 1987 Cole 1992
Proctor 1992) Serpentine soil solutions obtained using tension-free lysimetry have Cr
concentrations between 01 and 32 micromol L-1 in which a majority of the Cr is present as
colloidal material (Gasser and Dahlgren 1994) The inhabitability of these soils for most
plants has been attributed to the imbalance of Ca to Mg (Walker 1954) the deficiency of
plant nutrients (Turitzin 1991) and the elevated concentration of heavy metals
(Robinson et al 1935) however linking the lack of plant productivity of these soils to
toxic levels of Cr has not been fully resolved (Brooks 1987 Kruckeberg 1992) The
unique flora and fauna associated with these soils have been well-documented and Cr
concentrations in the plants range as high as 600 mg kg-1 (Brooks 1987 Gough et al
1989)
Chromium concentrations as a function of depth for serpentine soils in Maryland
(Rabenhorst et al 1982) Tehama County CA (Gough et al 1989) the Klamath
Mountains CA (Hotz 1964) and New Foundland Canada (Roberts 1992) are
illustrated in Figure 22 Chromium concentrations with respect to depth at each location
are highly variable demonstrating both maximums and minimums The concentration of
Cr varies significantly at each location due to the chemical composition of the parent
11
material and other soil-forming factors nevertheless Cr concentrations typically decrease
with depth towards the bedrock
A variety of chemical extractions on serpentine soils suggest a broad host of Cr-
containing soil phases including organic material silicates secondary oxides and spinels
(Schwertmann and Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994
Kaupenjohann and Wilcke 1995) The maximum concentration of Cr extracted from a
serpentine soil reported in the literature is 226 Wt Cr from the New Caledonia laterite
using a 03 M dithionite-citrate bicarbonate (DCB) solution (Schwertmann and Latham
1986) Li and Fendorf (2000) performed extractions on the serpentine soils at Jasper
Ridge using organic acids (citric and oxalic acids) in which a 100 microM citric acid solution
was only able to remove 23 Cr microgg-1 (08 Wt Cr) from a soil containing ~3000 Cr mg
kg-1 This suggests a majority of Cr is bound in phases that are not reactive with these
solvents
Sources of Chromium ProtolithBedrock
Chromium primarily occurs within oxide and silicate phases in ultramafic rocks
and serpentinites Chromite is a common primary (igneous) Cr-bearing spinel in mafic
and ultramafic rocks and often survives low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) Isomorphic substitution of
Al3+ Fe3+ and Ti4+ into the octahedral site and Mg2+ Ni2+ Zn2+ and Mn2+ into the
tetrahedral site is common (Sack and Ghiorso 1991) however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987 Pan and Fleet 1991 Burkhard 1993 Christofides et al 1994 Challis et
al 1995 Sanchez-Vizcaino et al 1995 Deer et al 1996 Mathiesen 1999) Pyroxenes
such as augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) are primary minerals in
ultramafic rocks that incorporate varying concentrations of Cr into their structures (Deer
et al 1996)
Serpentinization is associated with the partial oxidation of Fe(II) present in
olivine and pyroxene forming magnetite (Brooks 1987) Chromium(III) derived from
primary chromite and pyroxene may isomorphically substitute into the Fe(III) site of
magnetite due to size (octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge
12
similarities Hydrothermal and CO2 metasomatism of ultramafic rocks may produce Cr-
silicate minerals including fuchsitemariposite (Cr-mica lt17 Cr Wt ) uvarovite (Cr-
garnet lt20 Cr Wt ) tawmawite (Cr-epidote lt10 Cr Wt ) and kaumlmmererite (Cr-
chlorite lt6 Cr Wt Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Phillips et al 1980 Schreyer et al 1981 Max et al 1983 Kerrich et al 1987 Treloar
1987 Pan and Fleet 1991 Christofides et al 1994 Sanchez-Vizcaino et al 1995)
These minerals common in altered ultramafic rocks have varying concentrations of Cr
that isomorphically substitute for octahedral Al in the silicate structure due to similarities
in charge octahedral preference and octahedral radii (Al3+ = 061 Aring) Any of these
silicates and magnetite are potential sources for Cr in soils produced by the weathering of
serpentinite and ultramafic rocks
Chromium-Bearing Phases in Soils
Soil minerals specifically Fe(III) (oxy)hydroxides and clays provide a host of
structural and surface sites that may sequester Cr(III) released to soil solution by
weathering of Cr-bearing phases in serpentinite and ultramafic rocks Isomorphic
substitution of Cr(III) for Fe(III) in oxides (oxy)hydroxides and amorphous compounds
is expected based on crystallographic considerations synthesis experiments and analyses
of selective dissolution of Fe(III) phases (Schwertmann et al 1989 Gerth 1990)
Additionally Cr(III) readily sorbs on surfaces of Fe(III)-containing oxides such as
magnetite and hematite via inner-sphere complexation and forms low-solubility
precipitates at pH values greater than 4 (Charlet and Manceau 1992 Peterson 1996)
Octahedral substitution of Cr(III) for Al(III) in clays (smectites and vermiculites) is
possible based on size and charge similarities Sorption of Cr(III) on edge and interlayer
sites of clay minerals has also been observed by Ilton and Veblen (1994) Soil organic
matter (SOM) is another possible host for soil Cr in which Cr can be chelated into stable
organomineral and organometallic complexes (Kaupenjohann and Wilcke 1995 Brady
and Weil 1999)
13
MATERIALS AND METHODS
Location and Sample Collection
Jasper Ridge Biological Preserve is a microcosm of accreted terranes located in
the foothills of the Santa Cruz Mountains CA Figure 21a is a soils map reported by
Kashiwagi (1985) depicting the alluvial Tertiary sandstone and Franciscan Complex
soils present at Jasper Ridge The Franciscan Complex soils are divided into two
subgroups those derived from greenstone-graywacke and those from serpentine The
serpentine soils dominantly Mollisols and serpentinite bedrock are present as a narrow
discontinuous band located along the ridge top they do not receive sediment from the
adjacent non-serpentine soils Soils and rocks were collected at Jasper Ridge from 1999
to 2002 and sample locations are shown in Figure 21b A 500 meter transect consisting
of 10 sites extends across the length of the serpentine soil body At each site along the
transect soils and bedrock were collected with respect to depth Site 1 (Diablo clay) is
not derived from serpentinite but rather from chert and greenstone Additionally Site 1
is located down slope of the main serpentinite body and has likely received sediment
from the adjacent serpentine soil Rocks were also collected at several exposed outcrops
located in the serpentinite body
Site JR3 (Figure 21b) was chosen for a detailed geochemical study to identify
factors affecting the geochemistry of Cr in serpentine soils Soil at JR3 is a loamy
magnesic thermic Lithic Haploxeroll The thickness of the soil is approximately 50 cm
deep Grass dominantly Lolium multiflorum comprises 95 of the vegetation at this site
and trees and shrubs are not present
Mineral Identification in Rocks and Soils
Mineral identification was accomplished by electron probe microanalysis on an
automated JEOL 733A electron microprobe operated at 15 kV accelerating potential and
15 nA beam current Calibration was conducted using natural geologic standards Beam
width for analysis was 10 microm Raw counts were collected over 20 s and were constant
with time indicating that elemental drift was negligible Detection limits for probe
analyses are ~002 oxide Wt Additionally backscattering electron (BSE) images of
the rocks and soils were obtained using the microprobe
14
Carbon-coated petrographic thin sections of rock samples were used for
microprobe analyses Soil samples were dried in an oven at 60degC for 24 hours mounted
onto a glass slide using epoxy and heated to 100degC until the epoxy cured An initial cut
of each section using a petrographic saw was made followed by a second infusion of
epoxy to ensure total impregnation Finally each section was cut polished and carbon-
coated for electron microprobe measurements
Soil Chemistry
Soils were analyzed for pH electrical conductivity (EC (deciSiemensm))
particle size ammonium acetate extractable cations organic carbon total nitrogen and
elemental composition The pH was determined using a 11 mixture of soil to deionized
water Electrical conductivity (EC) was determined on 2 mL of a soil solution extract
Particle size of the soil was measured using the hydrometer method as described by Gee
and Bauder (1986) These analyses were performed at the Utah State University
Analytical Laboratory
Ammonium acetate was used to measure the concentration of extractable K Ca
Mg and Na within the soil The procedure involved weighing 2 g of soil and adding 25
mL of ammonium acetate (20 M NH4OAc) followed by mixing on an orbital shaker for
30 minutes The solution was filtered using a Whatman 42 filter paper (pore size 25
microm) Potassium Ca Mg and Na concentrations were analyzed using atomic absorption
spectrophotometry (AAS)
Organic carbon was measured using the Walkley-Black titration method (Mebius
1960) Approximately 05 g of sediment was weighed and reacted with 5 mL of 20 M
potassium chromate (K2Cr2O7) followed by the rapid addition of 10 mL concentrated
sulfuric (H2SO4) acid The bottle was cooled for 30 minutes at room temperature and
diluted with 100 mL of deionized water Next 030 mL of 0025 M ortho-
phenanthroline-ferrous solution was added and the solution titrated to the endpoint using
10M ferrous sulfate (FeSO4)
Total nitrogen of soil samples was obtained using an automated combustion
method as described by Sweeny (1989) Approximately 150 mg of air-dried and sieved
15
(40 mesh) soil was placed in a tin foil encapsulating cup A Leco total nitrogen analyzer
with resistance furnace was then used to determine total nitrogen
Major and minor element concentrations for soil and rock samples were obtained
by completely dissolving the samples using a mixture of hot concentrated nitric
perchloric and hydrofluoric acids Elemental concentrations were measured in the
supernatant using inductively coupled atomic emission spectroscopy (ICP-AES)
Standards were plusmn5 mg kg-1 of their known values These analyses were completed at
Chemex Laboratory (Vancouver Canada) Soil samples were dried and sieved (lt2mm)
before analysis Additionally total element concentrations for gt2mm sand silt and clay
size fractions were obtained for each sample collected at Site JR3
X-ray Diffraction and Clay-Size Mineral Identification
X-ray diffraction (XRD) analyses were conducted on the clay-size fraction
(lt2microm) from Site JR3 using a Siemens Diffractometer D5000 at the University of Idaho
The clay-size fraction was isolated using a suspension method described by Vaniman
(2001) Organic matter and calcium carbonate were removed from the clays using a
concentrated hypochlorite solution (Kunze and Dixon 1986) Crystalline and non-
crystalline iron oxides were removed using a citrate-sodium bicarbonate-sodium
dithionite (CBD) solution (Kunze and Dixon 1986) The remaining phyllosilicate
minerals were analyzed using a series of treatments including 1) magnesium-saturation
(2 M MgCl and 2 M MgAOC) and air drying 2) magnesium-saturation and glycolation
(1 glycerol-ethanol solution) 3) potassium-saturation (1 M KCl) and air drying and 4)
potassium (1 M KCl) saturation followed by heating to 500degC (Whittig and Allardice
1986) X-ray diffraction patterns were collected between 2θ values of 20-350deg and 20-
150deg with a generator potential of 30 kV a generator current of 22 mA (using CuKα
radiation) a Ni filter and a scan speed of 1degmin The software package JADE was used
for XRD data analysis and mineral identification
X-ray Absorption Spectroscopy
In-situ fluorescence yield X-ray absorption near-edge structure (XANES)
spectroscopy at the Cr K-edge (edge inflection point is 5989 eV for Cr(0)) was conducted
16
at the Stanford Synchrotron Radiation Laboratory (SSRL) Stanford CA using
beamlines 4-1 and 4-3 A Si(220) monochromator and a Ge-detector were used and
three scans were collected for each XANES analysis Soil and rock samples from Site
JR3 were prepared by removing the gt2mm fraction and crushed to a fine powder
Samples were loaded into a plastic sample holder and covered with Mylar tape
Elemental mapping of three samples were completed at SSRL on Beamline 11-2
using X-ray absorption spectroscopy (XAS) A Si(220) monochromator with a 6 microm
focusing capillary was used and fluorescence intensity was monitored with a Ge detector
Samples were prepared similar to those used for mineral identification using the electron
microprobe however the thickness of the sample was decreased to lt20 microm
Sequential Extraction Procedure
Sequential extraction experiments were completed on 2 g lt2 mm sieved soils
(Diablo clay from Site 1 at 10 cm depth and serpentine soil from Site 2 at 4 cm depth) in
three steps The exchangeable fraction was obtained using a barium chloride (1mM
BaCl2) solution for 1 h at 25 degC (Kunze and Dixon 1986) The Fe-crystalline oxide
fraction was obtained by reacting the solids with a hydroxylamine-hydrochlorideacetic
acid (10 M NH2OHsdotHCl in 25(vv) HOAc) solution for 6 h at 60 degC (Chester and
Hughes 1967) The silicate fraction was obtained by reacting the solids with
concentrated hydrofluoric acid (10 M HF) for 16 h at 25 degC Trace element
concentrations for each extraction were measured on the supernatant using inductively
coupled plasma optical emission spectrophotometry (ICP-OES) All analyses were
completed in triplicate Additionally sequential extraction experiments were completed
on 2g 106-250 mm sieved ultrasonically cleaned commercial grade chromite (MSDS
No 302-1 WHMIS Class D2A) using the same procedures listed above The
stoichiometric composition of chromite obtained using the electron microprobe is
(Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 Chromite was imaged after each sequential
extraction using a JEOL 5600LV environmental SEM equipped with an EDAX thin-
window energy dispersive X-ray detector
17
RESULTS
Serpentinite Mineralogy
The average serpentinite at Jasper Ridge is composed of serpentine (mainly
lizardite and antigorite 70) chlorite (15) talc (10) magnetite (4) and chromite
(1) based on petrographic and microprobe observations of five samples (JR3E JR6E 5
10 and JR3) however brucite has been reported as a major constituent in related
Franciscan Complex serpentinites (Hostetle et al 1966) Other minerals identified
include olivine augite and enstatite Although the serpentinite mineralogy is relatively
simple the modes and textures of the minerals present in the rock vary significantly from
site to site In Table 21 whole rock serpentinite compositions from Sites JR1 JR3E
JR6E 2 5 10 and JR3 are listed The average Cr concentration of the serpentinites is
~1700 mg kg-1 (tab 1) Iron (~7) Mg (~11) Ni (~3000 mg kg-1) and Co (~100 mg
kg-1) concentrations are also elevated in the serpentinite relative to the chert and
sandstone bedrock adjacent the main serpentinite body (Figure 21b and Table 21)
A range of rock and mineral textures are petrographically evident in the
serpentinite at Jasper Ridge (Figure 23) No foliations or lineations are observed in the
serpentinite (Figure 23a) obtained from the bottom of the soil pit excavated at Site JR3
(Figure 21b) Subhedral chromite and magnetite grains are randomly interspersed
amongst the serpentine matrix (Figure 23a) In serpentinite obtained in outcrop at Site
JR5E (Fig 21b) a lsquocrocodile skinrsquo texture is produced by thin irregular veins of fine-
grained subhedral magnetite separating serpentine grains (Figure 23b) In Figure 23c
numerous crosscutting veins of chlorite occur in orthopyroxene and define the foliation of
the rock in outcrop at Site JR3E (Figure 21b) The numerous crosscutting microscopic
veins random foliation patterns (or the lack thereof) and subhedral oxide grains suggest
the Jasper Ridge serpentinite has undergone extensive physical and chemical alteration
typical of most serpentinites
Microprobe analyses of minerals in the serpentinite demonstrate that the main Cr-
bearing phases are chlorite enstatite augite Mg-Al chromite Cr-magnetite and a
chromite-silicate mixture (defined in the following paragraph) Selected compositions
for these minerals are listed in order of their Cr concentrations in Table 22a Chlorite
contains the least Cr whereas the chromite-silicate mixture is the most enriched with
18
respect to Cr Figure 24 shows correlations of Cr content relative to Al Fe Mg and Si in
order to provide insight into the serpentinite mineralogy and to aid in evaluating trends of
Cr substitution
A fine-grained mixture containing spinel (chromite andor Cr-magnetite) and an
unidentifiable Cr-silicate contains the highest Cr concentration in the serpentinite at
Jasper Ridge (Table 22a shown in Figure 25a and 25b) This fine-grained mixture will
be referred to here as the chromite-silicate mixture (CSM) with chemical properties
similar to those in Table 22a and Figure 24 CSM is common around the rims of
protolith chromite grains and a number of the larger (~05 mm) Cr-magnetite grains in
the serpentinite The microcrystalline phase responsible for contributing Si to these
analyses is possibly chlorite due to similarities of false colors generated by backscattering
electron (BSE) images and chlorite commonly rimming CSM Microprobe analyses of
CSM listed in Tables 22 and 23 are considered as mixtures of several mineral phases
due to the fine-grained texture (lt5 microm) Chromium concentrations range from 12 to 16
Cr Wt in this phase CSM is chemically distinct from the chromite and magnetite as
shown in Figure 24
The chemical compositions of protolith chromite exhibit little to no variation in
any of the serpentinites collected at Sites 2 JR3E JR4E JR5E JR6E JR7E and JR3
(Table 22a and Figure 24) Chromite contains low Fe (~9 Wt ) high Mg (~9 Wt
) and high Al (~26 Wt ) compared to Cr-magnetite The average Cr concentration
for chromite is ~12 Cr Wt Additionally chromite grains are irregular and subhedral
suggesting that they have undergone significant alteration during metamorphism
Chromium concentrations in the magnetite vary depending on the proximity and
degree of alteration of the chromite (Table 22a and Figure 24) Magnetite grains with
less than 2 Cr Wt are common in veinlets in serpentinite not formed adjacent
chromite whereas magnetite with 8 to 12 Cr Wt were formed proximal or adjacent
chromite Chromium-enriched magnetite is also referred to as ferrite-chromite in the
literature (Spangenberg 1943 Burkhard 1993)
Chlorite and pyroxenes contain minor concentrations of Cr (lt1 Wt Table
22a) Aluminum Fe and Mg concentrations varied in the pyroxenes however the
majority of chlorites are relatively similar in composition (Figure 24) Chlorites are
19
more abundant in the serpentinite compared to the pyroxenes and both minerals were
subhedral and randomly distributed over the entire sampling area (Figure 21b)
Backscattering electron (BSE) images of what appeared to be a petrographically
opaque single chromite grain in Figure 23a demonstrates chromite has been
metasomatically altered resulting in the formation of Cr-magnetite Cr-chlorite and CSM
(Figure 25a) Represented compositions for these minerals (indicated by letters in Figure
25a) are listed in Table 23 Magnifying the chromite and CSM (Figure 25a)
demonstrates the fine-grained texture of CSM and an alteration rim (transition zone) is
also seen along the chromite edge (Fig 25b) This alteration rim is depleted in Cr with
respect to the chromite as shown in Figure 25b
In Figures 23a and 25c Ni was identified in three phases in the bedrock
including olivine (F 03 Ni Wt ) serpentine (G and H 03 and 0 Ni Wt ) and Ni-
Fe (I sim55 Ni Wt ) alloy (Figure 25c and 25d) their compositions are indicated by
letters and listed in Table 23 Alteration of the olivine resulted in the formation of lt10
microm Ni-Fe alloys shown in Figure 25c These Ni-Fe alloys form a corona surrounding
the remaining olivine and the illustration in Figure 25d highlights the minerals and
chemical changes
Chromite and Cr-magnetite alteration leading to the paragenesis of CSM and Cr-
chlorite is common in the serpentinites at Jasper Ridge The petrographic image in
Figure 26a and its corresponding BSE image in Figure 26b illustrate Cr-magnetite
(originally protolith chromite) partially altering to CSM and Cr-chlorite The chemical
compositions of each mineral (indicated by letters in Figure 26b) are listed in Table 23
CSM appears to marginally and metasomatically replace Cr-magnetite and is enriched by
~2 Cr Wt Chromium-chlorite is present in microcracks in Cr-magnetite and
encompasses the entire Cr-magnetite and CSM grain Serpentine minerals adjacent the
Cr-chlorite contains no measurable Cr
Distribution of Chromium in Serpentine Soils
Serpentine soils and underlying bedrock were collected at ten sites along a 500
meter transect extending the length of the main serpentine body as shown in Figure 21b
20
The range of Cr concentrations at each site (center) and Cr concentrations with respect to
depth (periphery) are shown in Figure 27 and soil analyses are listed in Table 24
Bedrock Cr concentrations for Sites 2 5 and 10 are also shown in these plots (Fig 27)
Chromium enrichment does not appear to be consistently found at a particular depth or
soil horizon for any of the sites Sites 3 9 and 10 demonstrate a Cr maximum at
approximately 006 m however Sites 1 2 5 and 6 exhibit a Cr minimum at roughly the
same depth The highest soil Cr concentration in the soil along the transect is 4760 mg
kg-1 at Site 3 at 006 meters depth The lowest soil Cr concentration is 1725 mg kg-1 at
Site 10 at 0085 meters depth The average bedrock Cr concentration is ~2000 mg kg-1
(Figure 27 Table 24) It is important to note Site 1 consists of Diablo clay not
serpentine soil however this site is located adjacent and down-slope of the main
serpentinite body From these observations it is apparent that there are no systematic
trends in soil Cr concentrations with respect to the position of the site and depth and soil
samples often contain higher concentrations of Cr with respect to their corresponding
bedrock and to the average Cr concentration of ultramafic rocks (1600 mg kg-1 Green
1972)
Trace metals including Ni Mn and Co were also evaluated in the serpentine soil
for Sites 1 through 10 and the results are listed in Table 24 Nickel concentrations range
from 1590 to 3960 mg kg-1 and do not directly correspond to the Cr fluxes with respect
to depth (Table 24) Nickel concentrations are less than Cr from Sites 1 to 4 From Sites
5 to 10 Ni concentrations exceed those of Cr averaging ~3600 mg kg-1 Unlike Cr Ni
concentrations are equivalent or greater in the bedrock (~3300 mg kg-1) as compared to
the soil (~3200 mg kg-1) Soil Mn concentrations at each site are relatively uniform at
~1800 mg kg-1 with little to no variation with depth (Table 24) Cobalt concentrations
do not exceed 238 mg kg-1 (Table 24) and are relatively uniform (~200 mg kg-1) with
respect to location along the transverse and depth at each site
Major elements including Al Fe Ca and Mg are listed for the serpentine soils
along the transect with respect to depth in Table 24 Aluminum Fe Ca and Mg
concentrations in the soils at each site are similar with exception of Site 1 (Diablo clay)
Aluminum concentrations in the Diablo clay are ~46 whereas the serpentine soils
contain less than 2 Al (Table 24) Iron and Mg concentrations are also low lt65 and
21
lt411 respectively in the Diablo clay compared to Fe (gt8) and Mg (gt6)
concentrations in the serpentine soil (Table 24) Calcium concentrations are low (lt1)
for all the serpentine soils and even the Diablo clay (Table 24) With a broad
understanding of the chemical characteristics of the serpentine soil at Jasper Ridge Site
JR3 (Figure 21b) was chosen for a detailed chemical study to examine all the factors
governing the chemistry of Cr in the soil
Serpentine Soil Chemistry and Mineralogy at Site JR3
Soil chemistry
Chemical properties (pH electrical conductivity organic matter nitrogen and
ammonium acetate extractable cations) and textures for four soil samples from Site JR3
are listed in Table 25 The pH of the soil is near neutral (pH from 671 to 698) and
increases slightly with respect to depth Electrical conductivities (EC) in the soil ranges
from 03 to 09 dSM a range indicative of relatively few dissolved salts andor major
dissolved inorganic solutes The organic carbon content of the soil is 32 at the surface
and decreases with depth to 10 near the bedrock Total nitrogen for the samples ranged
from 003 to 064 Ammonium acetate extractable Ca K Mg and Na are listed in
Table 25 The extractable concentration of both Ca and K decrease with depth whereas
there is no apparent trend associated with the concentrations of Mg and Na The
serpentine soil textures are dominantly clay loams however the texture is a sandy clay
loam near the bedrock
Whole soil samples and size fractions including gt2 mm sand (2-002 mm) silt
(002-0002 mm) and clay (lt2 microm) size fractions with respect to depth were chemically
digested and analyzed for total trace metal concentrations shown in Figure 28 and listed
in Table 26 to determine if the abundance of Cr was linked to a particular size fraction
(the percent of sand silt and clay size fractions are reported in Table 25) Additionally
Ni Mn and Co concentrations with respect to depth and size fraction are also presented in
Figure 28 and Table 26 Total and gt2mm size fraction Cr concentrations are similar at
the top and bottom of the soil profile ranging from 3297 to 5976 Cr mg kg-1 Chromium
concentrations exceed detections limits (gt10000 mg kg-1) for the sand-size fraction It is
important to note that the sand-size fraction corresponds to the size of most chromite
22
grains observed in the serpentinite Chromium in the clay fractions are low (~1000 mg
kg-1) compared to the other size fractions however these concentrations are still enriched
compared to non-serpentine soils
In Figure 28 and Table 26 Ni concentrations are enriched in the clay-size
fraction with values ranging from 2840 to 2920 mg kg-1 The silt-size fractions have the
lowest Ni concentrations (~1800 mg kg-1) Manganese concentrations are slightly
elevated relative to total Mn in the sand-size fraction with values ranging from 1728 to
2701 mg kg-1 The clay-size fraction is the least Mn enriched size-fraction with
concentrations at ~1200 mg kg-1 Cobalt concentrations are elevated in the sand-size
fraction with values ranging from 159 to 241 mg kg-1 and concentrations in the gt2mm
silt and clay-size fractions are similar to the total (~150 mg kg-1)
Soil Cr-bearing minerals
Chromium-bearing minerals in the soil identified using the electron microprobe
include chromite Cr-magnetite and CSM Chromium concentrations in soil chromite
Cr-magnetite and CSM are graphically shown in Figure 29 as a function of Al Fe Mg
Si and Al+Fe in Wt Representative analyses of the Cr-bearing soil minerals are listed
in Table 22b Compositions of soil chromite in Figure 29 are chemically distinct from
those identified in the protolith in Figure 24 Chromium concentrations in soil chromites
range from 22 to 37 Cr Wt containing minor concentrations of Al and Mg Values for
chromite in the serpentinite bedrock (Table 22a and Figure 24) and the FeCr2O4 end-
member are also shown in Figure 29 The dashed lines in Figure 29 connect Jasper
Ridge bedrock chromite (Figure 24) and stoichiometric FeCr2O4 where it can be seen
that compositions of soil chromite lie close to these tie lines Overall soil chromite
demonstrates a decrease in Al and Mg and an increase in Fe relative to Cr compared with
protolith chromite Soil Cr-magnetites contain between 7 and 20 Cr Wt and are
enriched in Cr compared to Cr-magnetites identified in the bedrock (Figure 24) Cr-
magnetite however reveals a decrease in Fe relative to Cr Minor concentrations of Si
(lt2 Wt ) in the soil chromite and Cr-magnetite may represent slightly altered rims or
contamination from sample preparation Chromium concentrations in CSM maintain a
consistent Cr concentration of ~11 Cr Wt similar to those found in the bedrock (~14
23
Cr Wt ) Soil CSMs with compositions between 8 and 12 Mg Wt are more than
likely associated with analyses containing more microcrystalline chlorite compared to
soil CSMs with less than 2 Mg Wt Soil CSM appears be less enriched with Si (lt14
Wt ) relative to those identified in the serpentinite (15 to 27 Wt )
The size of the Cr-bearing minerals varied significantly for each sample ranging
from 10 microm up to 2 mm The majority of chromite grains in the soil correspond to the
sand-size fraction and also vary from euhedral to subhedral with no discernable pattern
The sum of cations based on four oxygens for the chromite and Cr-magnetite analyses
range between 297 and 301 indicating these minerals are spinels not (oxy)hydroxides
Compositions obtained for chromite and Cr-magnetite may represent alteration rims
where weathering of the mineral in the subsurface environment has been most active
BSE images of chromite and Cr-magnetite were not able to definitively identify whether
these analyses in fact represent alteration rims
Mineralogy of the Clay-Size Fraction
The clay-size fraction in the serpentine soil at Site JR3 was evaluated with respect
to depth using XRD and specific chemical and heat treatments The d-spacings (Aring) and
their relative intensities for each soil and treatment with respect to depth are listed in
Table 27 Figure 210 shows the XRD analyses for clay sample JR3515 (depth 5-15
cm) and these analyses are almost identical to the other clays examined at Site JR3
Additionally Table 28 from Whittig and Allardice (1986) is provided to interpret the X-
ray diffraction analyses in Table 27 For example a d-spacing of ~14Aring is present in all
the treatments indicating the presence of chlorite in the clay-sized fraction Using an
XRD analysis program the clay-size fraction was determined to be composed of
smectite vermiculite serpentine minerals (lizardite and antigorite) and chlorite
(clinochlore and Cr-clinochlore) Smectite is the most abundant mineral in the clay-size
fraction whereas Cr-chlorite is the least abundant Overall the d-spacings and relative
intensities are similar despite the treatments and changes in depth (Table 27) This
indicates that the mineralogy of the clay-size fraction does not vary significantly in the
soil column
24
Chromium concentrations in the clay-size fractions are ~1010 mg kg-1 (Table
26) Smectite and vermiculite are possible Cr-bearing clay minerals in which isomorphic
substitution of Cr(III) could occur in the dioctahedral sheet Serpentine minerals
(lizardite and antigorite) are not likely hosts of Cr based on crystallographic
considerations and the absence of Cr in analyzed serpentine minerals in the bedrock
(Table 23) Clinochlore may also be a host for Cr in the inner layer octahedral sheet
Chromium-clinochlore is the only Cr-containing mineral identified in the clay-size
fraction using XRD and Cr-clinochlore (Cr-chlorite) is also an abundant mineral
identified in serpentinite protolith (Tables 22 and 23 Figure 26)
XAS of chromium in relation to depth
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in serpentine soils and bedrock at Site JR3 The XANES (X-ray absorption near-edge
structure) spectra demonstrate the absence of Cr(VI) in the soil column at Site JR3
(Figure 211) based on the lack of a pre-edge feature The detection limit for Cr(VI)
using this method is approximately ~20 mg kg-1 Ultimately XAS analyses demonstrate
Cr(VI) is not detected in any of the soil solids
X-ray microprobe analyses of soil and rock samples (0-5cm depth 5-10cm depth
and serpentinite) at Site JR3 were processed to map the relative concentrations of Fe and
Cr in which areas of elemental enrichment are identified by a violet color areas of
depletion by a blue color (Figure 212) Areas of Cr and Fe enrichment in Figure 212 are
localized with well-defined boundaries corresponding to grain edges and these elements
are not correlated in a well-defined trend Minerals assigned to these areas of Cr and Fe
enrichment were made based on petrographic and microprobe observations on the same
thin section For example high Fe and no Cr is magnetite whereas high Fe and high Cr
is either chromite or Cr-magnetite Chromite magnetite and Cr-magnetite appear to be
present in the serpentine soil at a depth between 0 to 5 cm (Figure 212A) In Figure
212b CSM is a possible mineral detected in the soil (depth 5-10 cm) and the presence
of magnetite is clearly seen due to the localized enrichment of Fe and no Cr Iron and
Cr spectra for the serpentinite were much more heterogeneous as compared to the soil
samples and Cr concentrations appear be more scattered (Figure 212c) These X-ray
25
microprobe analyses provide a complimentary method for examining Cr geochemistry
(including Cr(III) and Cr(VI)) in naturally occurring soil and rock solids however
Cr(III) is the only detected valence state in serpentine soils and rocks at Jasper Ridge
Sequential Extractions
Sequential extractions were completed on 1) Diablo clay (Inceptisol) from Site 1
(Figure 21b) at a depth of 10 cm 2) serpentine soil (Mollisol) from Site 2 (Figure 21b) at
a depth of 4 cm and 3) commercial grade chromite (106-250 microm) and these results are
shown in Figure 213 and listed in Table 29 Chromium extracted from the
exchangeable fraction is minimal (lt01 mg kg-1) for all the samples (number 1 in Table
29 Fig 213) Minor Cr concentrations are associated with the Fe-crystalline oxide
fraction in the Diablo clay (19 Cr mg kg-1) and serpentine soil (78 Cr mg kg-1) whereas
chromite only released 05 Cr mg kg-1 (number 2 in Table 29 Figure 213) Increased
concentrations of Ni and Fe are present in the Fe-oxide fraction of the Diablo clay (161
Ni mg kg-1 and 4474 Fe mg kg-1) and serpentine soil (5138 Ni mg kg-1 and 5138 Fe mg
kg-1 Table 29) Chromium extracted from the Diablo clay serpentine soil and chromite
in the silicate fraction is 14 31 and 14 Cr mg kg-1 respectively (number 3 in Table 29
Figure 213) The sum of extracted Cr Fe Ni and Mn concentrations (number 4 in
Table 29) are low with respect to total metals (number 5 in Table 29) In fact only 07
12 and 00 Cr was removed from the Diablo clay serpentine soil and chromite
respectively (number 6 in Table 29)
Chromite was extremely resistant to all the extractions and less than 001 of the
total Cr was removed (Table 29) Scanning electron microprobe (SEM) images of the
chromite after each extraction are shown in Figure 214 Chromium concentrations
released from the chromite (Table 29 and Figure 214a) increased with each sequential
extraction experiment starting with the exchangeable fraction (Figure 214b) progressing
into the Fe-crystalline oxide fraction (Figure 214c) and ending with the silicate fraction
(Figure 214d) The surface of the chromite became smoother after each extraction and
crystal imperfections in defect sights located along the edges and corners progressively
decreased (Figure 214) Chromite does not appear to be an easily accessible source of
Cr based on these extraction experiments
26
DISCUSSION
Serpentinite at Jasper Ridge has undergone significant physical and chemical
alteration since its formation and emplacement off the California continental margin
Despite being subjected to high pressures (~5-7 kbars) and elevated temperatures (200-
300degC) trace element concentrations within the serpentinite remain elevated Jasper
Ridge serpentinites are enriched with Cr Co Fe Mg Mn and Ni compared to chert and
sandstone located adjacent the main serpentinite body (Table 21 and Figure 21b) Over
the past 400000 years Cr-bearing minerals in the serpentinite have been weathering and
accumulating in the soil During this time Cr concentrations in the soil have
progressively increased relative to Cr-concentrations in the serpentinite protolith This Cr
enrichment is directly linked to specific weathering-resistant minerals identified in the
soil
Chromium in Serpentinites
Three minerals in the bedrock at Jasper Ridge contain appreciable concentrations
of Cr Cr-magnetite (82-103 Cr Wt ) chromite (~108 Cr Wt ) and CSM (~133
Cr Wt ) Cr-magnetite and CSM were derived from the alteration of primary igneous
chromite CSM or Cr-spinels with high concentrations of Si have been noted by
(Burkhard 1993) as a Cr-spinel containing a possible microscopic silicate phase Our
analyses support this result in which the microcrystalline silicate is probably Cr-chlorite
due to its common occurrence with CSM (Figure 26) Definitive identification of this
microcrystalline mixture was not possible due to size of the crystals relative to the spot
size of the electron microprobe Chlorite and pyroxene contain minor concentrations of
Cr (~1 Wt ) and these were the only identified Cr-containing silicates in the
serpentinite Minor concentrations of Cr observed in some chlorites are directly related
to the metasomatic alteration of chromite and Cr-magnetite (Figure 26) Other common
metasomatically-derived Cr-bearing silicates such as fuchsitemariposite and uvarovite
are not present possibly due to the lack of aqueous K+ and Ca+2 in the alteration fluids
Overall the chemical compositions of these protolith minerals (Figure 24) demonstrate
only minor chemical variations allowing comparative analysis with Cr-bearing soil
minerals
27
Chromium in Serpentine Soils
Cr-bearing soil minerals identified using the electron microprobe include
chromite Cr-magnetite and CSM and these phases are chemically distinct from those in
the bedrock (Figure 24 and 29) The highest concentration of Cr (41 Wt ) in the soil
is in chromite with a composition approaching the FeCr2O4 end-member This represents
a ~80 increase in the Cr concentration relative to bedrock chromite (see Figure 215)
Cr-magnetite also demonstrates Cr enrichment (up to 20 Cr Wt ) with respect to
magnetite identified in the bedrock (~10 Cr Wt ) (Figures 24 and 29) The textures of
the soil Cr-bearing spinels exhibit sharp irregular edges possibly due to extensive
weathering CSM in the soil is similar to those found in the bedrock with Cr
concentrations ranging from 11 to 14 Cr Wt however Si concentrations in soil CSMs
(8 to 14 Wt ) are noticeably less than those measured in the serpentinite (15 to 20 Wt
) The weathering of CSM might preferentially release Si relative to other elements in
its structure Cr-silicates (chlorite and pyroxenes) were not identified in the soil possibly
due to complete weathering of these phases or they were too fine-grained for microprobe
analyses however Cr-clinochlore was identified in the clay-size fraction using XRD
Elemental maps shown in Figure 212 corroborate electron microprobe observations by
revealing well-defined zones of Cr and Fe enrichment in both the soils and serpentinite
These areas of Cr and Fe enrichment are either chromite Cr-magnetite CSM or
magnetite Overall these maps reveal Cr is not homogeneously distributed in the soil
confirming Cr is in specific mineral phases Additionally the observed abundance of
chromite Cr-magnetite and CSM in the soils indicates that these minerals are probable
sources for a majority of the Cr enrichment identified in the soils
Soil Spinels
Soil chromite and Cr-magnetite appear to have undergone chemical alteration
enriching the minerals with respect to Cr In Figure 215 compositions of bedrock
chromite and end-member FeCr2O4 are shown together with tie-lines like those in figure
8 connecting both minerals Reaction progress in Figure 215 is based on the assumption
that the serpentinite chromite (0 reaction progress) is proceeding toward FeCr2O4
(100 reaction progress) in the serpentine soil Soil chromites from Jasper Ridge are also
28
plotted in Figure 215 where Cr concentrations are assumed to vary linearly between the
bedrock chromite and chromite end-member which defines the reaction progress of the
mineral These tie-lines correlate extremely well to the chemical composition of
chromites identified in the soil For example chromite containing ~30 Cr Wt will
have ~15 Fe Wt ~13 Al Wt and ~5 Mg Wt similar to soil chromite in Figure 29
The retention of Cr in chromite shown in Figure 215 has significant implications towards
the availability and cycling of Cr in serpentine soils
One possible mechanism for the chromite composition trends illustrated in
Figures 29 and 215 is incongruent dissolution where Cr is retained relative to other
elements Sequential extraction experiments (Figure 213 and Table 29) support this
idea in which Fe is more easily released than Cr (1096 Fe mg kg-1 to 19 Cr mg kg-1)
from chromite Unfortunately Mg and Al were not monitored during the experiments
From these sequential extraction experiments and chromite analyses (Figure 215) Al
and Mg have been preferentially released from the chromite structure compared to Fe and
Cr Additionally the alteration of chromite in the serpentinite and the soils are analogous
to one another Chromite in the serpentinite retains a majority of its Cr compared to Mg
and Al when being altered to CSM by fluids at elevated pressures and temperatures
(Figure 26 and Table 22) Under ambient soil conditions chromite is once again
retaining Cr relative to Mg and Al causing what appears to be an increase in Cr
concentration Chromium is not easily released from chromite whether under high
pressures and temperatures or in the soil
Crystal field stabilization energies (CFSE) and octahedral site preference energy
parameters (OSPE) provide a means to explain the incongruent dissolution or Cr
preservation in the Cr-spinels CFSE is a measure of the net energy of stabilization
gained by a metal ions nonbonding d electrons as a result of complex formation This
measure also indirectly describes the energy required to remove an ion from a crystal
structure The octahedral CFSE for Cr(III) is ndash2247 kJmole and the OSPE (the
difference between the tetrahedral and octahedral CFSE) is ndash1578 kJmole (Dunitz and
Orgel 1957 McClure 1957) These values are the most negative for any transition
element and indicating the presence of Cr(III) in an oxide increases the crystalrsquos stability
compared to an analogous crystal lacking Cr(III) Chromium(III) would be among the
29
last ions to removed from the spinel structure compared to other transition elements such
as Fe2+and Mn2+
Soil Chemistry related to Chromium Availability
In Table 25 the chemical properties of the soil at Site JR3 do not appear to be a
factor when evaluating Cr in the soil The pH of the soil did not vary considerably with
depth Additionally the near neutral pH would not have a significant impact accelerating
Cr-bearing mineral dissolution Organic matter within the soil was elevated despite the
low plant productivity of the serpentine soil The role of organic matter could be a
significant factor such as the production of organic acids promoting the dissolution of the
Cr-bearing minerals however Cr concentrations near the surface (where organic matter
was the greatest) are not significantly greater or less than those deeper in the soil profile
The relative concentrations of ammonium acetate extractable Ca K Mg and Na are
characteristic of serpentine soils and did not provide any new insight Finally Cr(III)
was the only identifiable valence state of Cr based on X-ray absorption spectroscopy
(Figure 211) which is what should be expected based on the Cr-bearing minerals and the
pH of the soil
Extraction experiments summarized in Table 29 and Figure 213 demonstrate
~99 of the Cr in a serpentine soil is retained in highly resistant phases at Jasper Ridge
The chromite standard used in these experiments contributed negligible concentrations of
Cr (lt2 mg kg-1 total) for all of the sequential extraction experiments supporting earlier
evidence that Cr-spinels are retaining Cr and are potentially responsible for a majority of
the total Cr enrichment in the serpentine soil At Jasper Ridge Cr extracted from the
serpentine soil (31 mg kg-1) is likely the result of the Cr-silicates compared to the Cr-
spinels Low Cr-containing minerals (possibly chlorite enstatite augite and clay
minerals) appear to be the most probably source of Cr to soil solutions and vegetation of
the serpentine soil Approximately 30 of the serpentine soil is composed of clay-sized
minerals and this size-fraction has ~1000 Cr mg kg-1 The mineral responsible for Cr
enrichment in the clay-size fraction is more than likely Cr-chlorite identified using XRD
Chromium-chlorite in the soils suggests that this mineral is potentially the largest silicate
contributor of Cr in the serpentine soil
30
Interpreting the enrichment and behavior of Cr in serpentine soils such as in
Figure 27 can be accomplished based on the abundance Cr-spinels observed in the soil
Small grains of chromite Cr-magnetite and even CSM in the soil have the potential to
appreciably impact Cr concentrations Table 210 demonstrates the chromite masses
required to change the Cr concentration in a 1 kg sample of soil The highest Cr
concentration at Jasper Ridge was identified at Site 3 with a measured value of 4760 mg
kg-1 which means only ~1 of the soil sample (Table 210) needs to contain chromite
Additionally a minimal amount (lt011 of the sample as shown in Table 210) of
chromite is necessary to change the concentration by 500 mg kg-1 which could easily
cause the sharp Cr increases and decreases noted in Figure 27 This agrees with our
observations of chromite abundance in the serpentine soils Additionally a majority of
the chromite grains identified in the bedrock are between 2 and 002 mm which
correspond directly to the elevated Cr concentrations (gt10000 mg kg-1) in the sand-size
fraction of the serpentine soil (Table 26) Even the ~15000 Cr mg kg-1 values in soils
measured by Hotz (1964) in Figure 22 only require ~2 of the sample of the sample to
be chromite A few grains of chromite possess the ability to significantly impact
measured Cr concentrations in soil samples
Chromium and Nickel in Serpentine Soils
Chromium and Ni have often been correlated to one another due to their common
enrichment in serpentine soils (Hotz 1964 Rabenhorst et al 1982 Schwertmann and
Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al 1989 Kaupenjohann
and Wilcke 1995) A majority of the Ni in the Jasper Ridge serpentine soil originated
from olivine (Figure 25c and 25d Table 23) Chemical alteration of olivine during
serpentinization resulted in the mineral serpentine (Coleman and Jove 1992) in which
minor concentrations of Ni (lt037 NiO Wt ) substituted into its structure Continued
alteration under highly reducing conditions and the incompatibility of Ni in the serpentine
structure resulted in the formation of Ni-Fe alloy (Figure 25c) This type of alteration is
observed in many of the rocks at Jasper Ridge where the Ni-Fe alloys are ~10 microm in size
Olivine in the rocks and soil appear to be less resistant to weathering compared to
chromite therefore Ni is possibly more available to bacteria and vegetation in serpentine
31
soils Sequential extractions (Figure 213 Table 29) support these observations in which
a greater percentage of total Ni was extractable compared to the Cr in the soils Although
Cr and Ni are enriched in serpentine soils the availability and behavior of these two
elements are significantly different
CONCLUDING REMARKS
Chromium is predominantly present in primary chromite Cr-magnetite and a
chromite-silicate mixture (CSM) in both the bedrock and soil despite weathering over
geologic time Sharp variations in soil Cr concentrations with respect to depth are caused
by the presence or absence of chromite Cr-magnetite andor CSM Chromium-spinels in
the soil have undergone incongruent dissolution possibly conserving Cr(III) with respect
to Al3+ and Mg2+ Chromite is highly resistant to weathering and it is more than likely
not contributing bioavailable Cr Although Cr-spinels are a major source of Cr Cr-
silicates in the protolith (chlorite enstatite and augite) and soil (chlorite
montmorillonite and vermiculite) afford the most accessible Cr source identified Based
on our observations and analyses chlorite is the most likely Cr source of these silicate
minerals Finally Cr(VI) was not detected in any of the soil and rock samples
ACKNOWLEDGEMENTS
We thank Jasper Ridge Biological Preserve for allowing us the opportunity to
examine the serpentine soils Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund NSF Grant EAR-9902859 (DKB) GSA Graduate
Student Research Grants Office of Teaching and Learning (SF) at Stanford University is
gratefully acknowledged We appreciate the technical support from the staff of the
Stanford Synchrotron Radiation Laboratory from Bob Jones for assistance in microprobe
analyses of the serpentinites and serpentine soils and from Ben Bostick Travis Horton
and Theresa Barber for field and lab assistance
32
Table 21 Whole rock elemental concentrations of serpentinites (Figure 21b) chert and sandstone
-----------------------Location of Serpentinitesdagger----------------------- Elemental Concentrations JR1 JR3E JR6E Site 2 Site 5 Site 10 JR3
ChertDagger SandstoneDagger
Al 02 08 03 027 22 107 024 079 027 Ca 026 007 005 003 046 013 003 001 lt001 Co mg kg-1 101 99 100 99 155 217 97 1 lt1 Cr mg kg-1 580 2280 1510 1840 1945 2090 1465 10 lt1 Fe 303 71 609 581 76 1106 512 041 024 K 001 031 099 lt001 lt001 001 lt001 029 008
Mg 122 134 154 1336 524 804 112 014 034 Mn mg kg-1 495 800 155 730 1375 1860 460 110 85 Na 181 006 093 lt001 lt001 lt001 lt001 001 001 Ni mg kg-1 2037 1545 3321 3510 3020 3960 2740 20 9
Total digestion ICP-AES analyses daggerSerpentinites are dominantly composed of the minerals serpentine chlorite talc and spinels DaggerSee Figure 21b for sample locations
33
Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in serpentinite rock at Jasper Ridge
Chlorite Enstatite Augite Chromite Cr-Magnetite ChromiteSilicatedagger SiO2 3738 5463 5222 000 024 1053 Al2O3 552 477 516 5059 182 395 NiO 001 01 00 03 00 052 FeODagger 762 614 239 1446 7919 5042 MnO 009 015 009 042 140 182 Cr2O3 024 060 099 1584 1605 1939 MgO 3337 3305 1560 1869 024 605 CaO 023 050 2348 000 000 019 Na2O 000 001 050 000 001 008 K2O 000 000 001 000 000 011 Total 8446 9994 10048 10027 9894 9307
Formula Unit Compositions
Si 370 094 189 000 001 --- IVAl 030 006 011 000 000 --- VIAl 035 004 011 157 008 --- Ni 000 000 000 001 000 ---
Fe+2 063 010 007 030 232 --- Mn 001 000 000 001 004 --- Cr 002 001 003 033 048 --- Mg 493 085 084 073 001 --- Ca 002 001 091 000 000 --- Na 000 000 004 000 000 --- K 000 000 000 000 000 --- O 1400 300 600 400 400 ---
Electron microprobe analyses daggerChromitesilicate is also referred to as CSM DaggerFe+3 is present however all conversions are based on Fe+2
34
Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in serpentine soil at Jasper Ridge
CSM Chromite Cr-Magnetite SiO2 1940 057 267 Al2O3 559 1059 381 NiO 014 001 004 FeOdagger 2629 2259 5244 MnO 155 154 419 Cr2O3 1878 5577 3237 MgO 1668 916 288 CaO 015 001 005 Na2O 006 001 001 K2O 0 008 009 Total 8864 10033 9855
Si --- 002 010 IVAl --- 042 017 VIAl --- 000 000 Ni --- 000 000
Fe+2 --- 063 148 Mn --- 004 013 Cr --- 147 096 Mg --- 046 016 Ca --- 000 000 Na --- 000 000 K --- 000 000 O --- 400 400
Electron microprobe analyses dagger Fe+3 is present however all conversions are based on Fe+2
35
Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25 and 26 A B C D E F G H I J K L Chromite
Chromite CSM Magnetite Chlorite Olivine Serpentine Serpentine Ni-Fe-Metal Cr-Magnetite CSM Chlorite
SiO2 0 0 1053 0 3407 4046 3688 4074 591dagger 021 1143 2855Al2O3
5059 5162 395 034 1114 000 397 085 054 165 980 2293NiO 028 031 052 0 004 037 037 0 5302dagger 0 0 0
FeODagger 1446 1452 5042 9992
1220 901 262 453 1753dagger 8169 3498 806MnO 042 044 182 0 008 010 0 0 0 107 150 00Cr2O3 1584 1576 1939 125 071 001 0 0 0 1369 1905 039MgO 1869 1869 605 003 292 4895 3848 3755 064 1217 2828CaO 0 0 019 0 0 0 0 0 0 0 0Na2O 0 0 0 0 0 0 0 0 0 0 0K2O 0 0 0 0 0 0 0 0 0 0 0H2O --- --- --- --- 1200 --- 1600 1600 --- --- 1200Total 10027 10136 9307
10164 9944 9891 9834 9972
9897 8897 10028
Formula Unit CompositionSi
000 000 --- 000 328 100 184 200 001 --- 269IVAl 000 000 --- 000 072 000 000 000 000 --- 131VIAl 157 159 --- 002 058 000 023 005 007 --- 123Ni 001 001 --- 000 000 001 001 000 000 --- 000
Fe+2 030 029 --- 294 084 019 011 019 239 --- 059Mn 001 001 --- 000 000 000 000 000 003 --- 000Cr 033 032 --- 004 006 000 000 000 041 --- 003Mg 073 073 --- 000 419 180 285 274 004 --- 397Ca 000 000 --- 000 000 000 000 000 000 --- 000Na 000 000 --- 000 000 000 000 000 000 --- 000K 000 000 --- 000 000 000 000 000 000 --- 000O 400 000 --- 000 1400 400 700 700 400 --- 1400
36
Electron microprobe analyses dagger elemental metal Dagger Fe+3 is present however all conversions are based o
36
2321000---
10025
591000054
53021753000000
2321000000000000
n Fe+2
Table 24 Bulk soil elemental concentrations for soils and bedrock collected along the transect
(Fig 21b)
Sites Samples Depth
(meters) Cr
(mg kg-1) Ni
(mg kg-1) Mn
(mg kg-1) Co
(mg kg-1) Al
() Fe
() Ca ()
Mg ()
COI03 0015 2420 1590 1325 118 464 565 065 352 COI36 0045 2370 1630 1600 138 467 55 063 356 COI69 0075 2160 1755 1670 144 471 6 064 365 COI912 0105 2420 1785 1740 151 496 615 067 384
1
COI1215 0135 2570 1985 2340 195 506 649 069 411 COII02 001 3710 2400 1665 169 294 702 100 768 COII25 0035 3600 2290 1625 162 275 673 093 743 COII58 0065 2910 2310 1550 159 25 672 087 771
COII811 0095 3520 2350 1595 158 276 693 093 766 2
COIIBedrock 012 1840 3510 730 99 027 581 003 1336 COIII14 0025 3750 2930 1655 190 178 897 062 762 COIII47 0055 4760 3050 1850 178 178 94 056 805
COIII710 0085 4380 3410 1930 216 216 1044 07 797 3 COIII1013 0115 4320 3380 1955 201 213 1061 063 626 COIV01 0005 3710 2970 1915 214 209 909 053 781 4 COIV15 003 3540 2710 1725 195 181 801 043 792 COV14 0025 2620 2350 1330 150 246 651 063 526 COV47 0055 1775 2710 1520 165 252 766 059 517
COV710 0085 2000 3090 1830 195 225 737 057 583 5 COVBedrock 0115 1945 3020 1375 155 22 76 046 524
COVI14 0025 3040 3720 1855 224 084 1152 015 736 COVI47 0055 2400 3610 1785 218 085 1138 014 68 COVI710 0085 2960 3800 1850 232 093 1232 015 714 6
COVI1013 0115 2800 3860 1800 224 085 1164 014 739 COVII14 0025 2620 3550 1925 229 083 1128 015 718 COVII47 0055 1975 3590 1805 216 084 1085 014 655 7 COVII712 0095 2080 3600 1840 220 07 1105 011 741 COVIII14 0025 2130 3830 1990 238 085 115 017 705 COVIII47 0055 2550 3720 1865 228 085 1099 017 714 8
COVIII712 0095 2240 3910 1860 229 076 1112 014 828 COIX14 0025 2470 3480 1890 227 104 1201 023 579 COIX47 0055 3000 3830 1595 205 084 108 013 842 9 COIX713 01 2720 3710 1845 221 111 1208 017 551 COX14 0025 2810 3740 1760 218 129 114 017 675 COX47 0055 2190 3910 1835 224 131 1206 017 702
COX710 0085 1725 3370 1595 192 11 1061 014 635 10 COXBedrock 0115 2090 3960 1860 217 107 1106 013 804
Total digestion ICP-AES analyses
37
Table 25 Chemical and physical properties of the soil at Site JR3
NH4OAc-extractable ---Hydrometer---Ca K Mg Na
Sand Silt ClaySample Depth (cm) pH EC
dSm
Organic Matter ------
LECO Nitrogen ------ ---------mg kg-1---------
Texture
JR305 25 671 04 32 012 505 2366 3291 355 36 31 33 Clay LoamJR3515
10 683 09 19 064 4205 8176 3343 2392 36 30 34 Clay LoamJR31530 225 688 03 12 003 3725 4903 3750 2305 36 33 31 Clay LoamJR33045 375 698 07 10 003 2873 3665 3388 2549 46 24 30 Sandy Clay Loam
See Methods Section for more details
38
38
Table 26 Selected trace metal concentrations for each size fraction versus depth at Site JR3
Depth (cm) Total gt2mm Fraction Sand Silt Clay Sample
------------------------Chromium (mg kg-1)------------------------ JR305 25 4906 5086 gt10000 4565 988 JR3515 10 5976 4678 8284 4000 1115 JR31530 225 5222 4206 gt10000 4133 980 JR33045 375 3477 3297 gt10000 4571 956
------------------------Nickel (mg kg-1)------------------------------ JR305 25 1935 2050 1660 1305 2920 JR3515 10 2180 2050 1980 1325 2920 JR31530 225 2190 2120 1835 1370 2880 JR33045 375 2090 2140 1740 1285 2840
------------------------Manganese (mg kg-1)-------------------------- JR305 25 1319 1571 1728 1301 1210 JR3515 10 1664 1601 2701 1471 1254 JR31530 225 1639 1626 2217 1599 1231 JR33045 375 1222 1226 1781 1405 1170
-----------------------Cobalt (mg kg-1)-------------------------------- JR305 25 128 146 159 134 131 JR3515 10 155 145 241 147 132 JR31530 225 152 147 194 162 128 JR33045 375 125 124 159 141 122
Total digestion ICP-AES analyses
39
Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3
JR305CLAY Depth 0-5cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1439 100 1833 100 1274 100 1391 100 731 179 1432 431 732 645 1047 105 71 216 941 163 71 41 838 838 474 101 847 141 475 257 726 726 356 125 73 271 365 264 313 83 712 248 355 299
473 143 313 257 364 131 355 193
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1845 1000 1268 1000 1388 1000 852 68 1447 387 724 987 1010 859 733 264 944 143 703 679 836 281 712 247 903 139 471 282 723 576 475 109 850 83 362 244 365 73 733 426 352 321 356 117 712 291 313 154 314 63 475 122
364 170 355 196
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1724 1000 1297 1000 1396 1000 842 45 1397 447 741 957 1175 708 730 318 923 176 715 686 1037 944 710 255 840 118 477 271 837 139 475 97 717 465 367 400 726 375 357 133 468 182 357 386 313 91 362 194 315 286
352 229 JR33045CLAY Depth 30-45cm
Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity
1447 1000 1869 1000 1268 1000 1393 627 842 44 1476 287 737 843 1049 1000 730 207 954 107 715 539 837 190 710 182 915 107 479 157 723 111 475 69 860 61 366 303 364 69 740 332 356 292 356 74 717 256 315 315 313 57 478 111
367 141 357 115
See Methods Section for details
40
Table 28 X-ray diffraction d-spacings for common silicates from Whittig and Allardice (1986)
Diffraction Spacing (Aring) Mineral(s) Indicated Mg-saturated air-dried
14-15 Smectite vermiculite chlorite 99-101 Mica (illite) 71-73 Serpentine
715 Kaolinite chlorite Mg-saturated glycerol-solvated
177-180 Smectite 14-15 Vermiculite chlorite
99-101 Mica (illite) 71-73 Serpentinte
715 Kaolinite chlorite K-saturated air-dried
14-15 Chlorite vermiculite (with hydroxy interlayer) 124-128 Smectite 99-101 Mica vermiculite (contracted) 71-73 Serpentine
715 Kaolinite chlorite K-saturated heated (500ordmC)
14 Chlorite 99-101 Mica vermiculite (contracted) smectite (contracted) 71-73 Serpentine
715 Chlorite
41
Table 29 Results of the Diablo clay serpentine soil and chromite sequential extraction experiments
1 Exchangeable Fraction (1mM BaCl2) 4 Sum of 1 2 amp 3 (mg kg-1) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 0 0 003 Cr 159 388 19 Fe 014 011 001 Fe 24812 25532 1096 Ni 104 103 008 Ni 1854 1868 25 Mn 0 0 0 Mn 778 1172 04
2 Fe-Oxide Fraction (1M NH2OH HCl in 25 (vv) HOAc) 5 Total Cr Fe Ni and Mn (Total Digestion) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 19 78 05 Cr 2290 3310 1767864 Fe 4474 5138 38 Fe 60000 68000 1044325 Ni 1610 1506 01 Ni 1673 2345 00 Mn 231 430 00 Mn 1498 1608 3424547
3 Silicate Fraction (10M HF) 6 Percent Extractable ((45)100) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ------------------------------------- Cr 140 310 14 Cr 07 12 00 Fe 20337 20393 1058 Fe 41 38 01 Ni 234 352 22 Ni 111 80 00 Mn 547 742 04 Mn 52 73 00
ICP-AES analyses
42
Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1 kg of sample
Chromite
FeCr2O4 Cr Concentration
(mg kg-1) Grams Sample 500 11 011 1000 22 022 2000 43 043 5000 108 108
43
Figure Captions
Figure 21a Soils map of Jasper Ridge Biological Preserve (Stanford CA) with
complete descriptions of the soils shown in the legend 21b An enlarged image of the
study area shown in 22a indicates sampling sites of the soils and rocks collected A
dashed line marks the 500 m transect
Figure 22 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above
Figure 23 Petrographic images illustrating the varying serpentinite textures at Sites (A)
JR3 (B) JR5E and (C) JR3E For reference to site locations see Figure 21b
Figure 24 Correlation diagrams of Cr-bearing minerals in the serpentinite at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Si in Wt
Chromite (circle) Cr-magnetite (square) chromitesilicate or CSM (down triangle)
chlorite (hexagon) and pyroxenes (up triangle) analyses are shown on each correlation
diagram
Figure 25 Images of the serpentinite bedrock at Site JR3 demonstrating the mineralogy
and Cr-bearing phases Microprobe analyses are indicated by circled letters and these
values are in Table 23 25a Backscattering electron image demonstrating several Cr-
bearing phases present including chromite Cr-magnetite CSM and Cr-chlorite from
Figure 23a 25b Magnified image shown in Figure 25a demonstrating the fine-grained
texture of CSM Additionally Cr2O3 Wt values are shown for chromite the depletion
zone and CSM 25c Image of Ni-bearing phases present in the bedrock including
olivine serpentine and Ni-Fe alloy from Figure 23a 25d Cartoon illustrating the
mineralogical and chemical changes present in 25c
Figure 26 Images of a Cr-chlorite and Cr-magnetite assemblage in the serpentinite
bedrock 26a Petrographic image of the bedrock at Site 4 (Figure 21b) 26b
44
Backscattered electron image demonstrating the common occurrence of Cr-chlorite
encompassing chromite grains Microprobe analyses are indicated by circled letters and
are located in Table 23
Figure 27 Highest and lowest Cr concentrations (mg kg-1) at each site in the serpentine
soil 500 m transect shown in Figure 21b Samples included soils at various depths (m)
and bedrock from each site The average Cr concentration for the Jasper Ridge
serpentinite is marked on the central figure Chromium concentrations versus depth (m)
at each site (noted in the upper left corner) outline the main diagram
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected
for total gt2 mm sand silt and clay size fractions listed in Table 26
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soil at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Al+Fe in Wt for
chromite (circles) Cr-magnetite (hexagons) and CSM (diamonds) Chromite from the
bedrock is denoted by a larger circle and FeCr2O4 with a square
Figure 210 XRD clay analyses of sample JR3515 (Site JR3 Figure 21b) demonstrate
the influence of each treament Mg-saturated Mg+glycerol K-saturated and K-saturated
+ heat (500oC) The d-spacings and intensities are also listed in Table 27
Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at depths (a) 0-5cm (b) 5-15cm (c) 15-30cm (d) 30-45cm and (e) bedrock The pre-edge
area indicative of Cr(VI) peaks is shown on the right Note that no peaks are present
Figure 212 X-ray microprobe elemental maps of Cr and Fe for (212a) serpentine soil
at a depth of 0-5cm (212b) serpentine soil at a depth of 5-10cm and (212c) serpentinite
bedrock at Jasper Ridge site JR3 (Figure 21b)
45
Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments for the Diablo clay (Site 1 Figure 21b) serpentine soil (Site 2
Figure 21b) and chromite listed in Table 29 The sum refers to the summation of the
three extraction experiments
Figure 214 SEM images of the chromite standard (A) after each sequential extraction
treatment including the (B) exchangeable fraction (C) Fe-crystalline oxide fraction and
(D) silicate fraction The concentration of Cr removed by each treatment is shown
beneath each set of images
Figure 215 Tie-lines from Figure 29 connecting serpentinite chromite and end-
member FeCr2O4 are shown on this single plot Soil analyses from Figure 29 are also
plotted on this diagram For a sum of 100 Wt oxygen must be added to the sum of
Cr Al Fe and Mg Wt
46
Figure 21a
47
Figure 21b
48
Chromium Concentration (mg kg-1)1000 10000
Dep
th (m
eter
s) 01
10
Maryland (Rabenhorst et al 1982)California (Gough et al 1989)California (Hotz 1964)New Foundland (Roberts amp Proctor 1992)
Surface
Figure 22
49
Figure 23
50
Cr (Wt )0 5 10 15 20
Al (
Wt
)
0
5
10
15
20
25
30
Cr (Wt )0 5 10 15 20
Fe (W
t
)
0
20
40
60
80
100
Cr (Wt )0 5 10 15 20
Mg
(Wt
)
0
5
10
15
20
25
Cr (Wt )0 5 10 15 20
Si (W
t
)
0
5
10
15
20
25
30
Cr0 10 20 30 40 50 60 70 80 90 100
Al
0
10
20
30
40
50
60
70
80
90
100
Fe
0
10
20
30
40
50
60
70
80
90
100
Chromite
Chlorite
Magnetite
Chromite-Silicate
Pyroxenes
Chromite
Chlorite
Pyroxenes
Magnetite
Chromite-Silicate
Figure 24
51
Figure 25
52
Figure 26
53
Sites1 2 3 4 5 6 7 8 9 10
Con
cent
ratio
n (m
g kg
-1)
1500
2000
2500
3000
3500
4000
4500
5000Soil Chromium
1
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000002004006008010012014016
2
Cr Concentration (mg kg-1)
1500 2000 2500 3000 3500 4000
Dep
th (m
eter
s)
000002004006008010012014
3
Cr Concentration (mg kg-1)3500 4000 4500 5000
Dep
th (m
eter
s)
000
002
004
006
008
010
012
5
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000002004006008010012
6
Cr Concentration (mg kg-1)2200 2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000002004006008010012
7
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000
002
004
006
008
010
9
Cr Concentration (mg kg-1)
2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000
002
004
006
008
010
012
10
Cr Concentration (mg kg-1)1600 2000 2400 2800 3200
Dep
th (m
eter
s)
000002004006008010012
4
Cr Concentration (mg kg-1)3500 3600 3700 3800
Dep
th (m
eter
s)
000
001
002
003
004
8
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000
002
004
006
008
010
BEDROCK
BEDROCK
BEDROCK
Range of Cr Concentrations in Serpentinite Bedrock
Figure 27
54
Soil Chromium
Depth (cm)0 10 20 30 40 50
Cr m
g kg
-1
02000400060008000
100001200014000
Totalgt2mmSandSiltClay
Soil Nickel
Depth (cm)0 10 20 30 40 50
Ni m
g kg
-1
0
1000
2000
3000
4000
5000
Soil Manganese
Depth (cm)0 10 20 30 40 50
Mn
mg
kg-1
0
1000
2000
3000Totalgt2mmSandSiltClay
Soil Cobalt
Depth (cm)0 10 20 30 40 50
Co
mg
kg-1
0
50
100
150
200
250
300
Totalgt2mmSandSiltClay
Totalgt2mmSandSiltClay
Exceeds detection limits of 10000 mg kg-1 Cr
Figure 28
55
Cr (Wt )0 10 20 30 40 50
Al (W
t
)
0
5
10
15
20
25
30
Cr (Wt )0 10 20 30 40 50
Mg
(Wt
)
0
2
4
6
8
10
12
Cr (Wt )0 10 20 30 40 50
Si (W
t
)
0
2
4
6
8
10
12
14Cr (Wt )
0 10 20 30 40 50
Fe (W
t
)
0
10
20
30
40
50
60
70
Cr (Wt )0 10 20 30 40 50
Al +
Fe
(Wt
)
10
20
30
40
50
60
70
JRBPChromite
FeCr2O4
Chromite
Magnetite
Chromite-Silicate
Figure 29
56
Clay XRD of JR3515 Depth 5-15cm
0
200
400
600
800
1000
1200
0 5 10 15 20 25 30 35 40 45
d-spacing (Aring)
Rel
ativ
e In
tens
ity
Mg-Saturated
Mg+Glycerol
K-saturated
K-saturated+heat (500oC)
Figure 210
57
Energy (kev)
597 598 599 600 601 602 603
Nor
mal
ized
Abs
orba
nce
-4
-3
-2
-1
0
1
2
(a)
(b)
(c)
(d)
(e)
Cr(
VI)
Figure 211
58
Figure 212
59
Diablo Clay
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Serpentine Soil
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Chromite
0
20
40
60
80
100
120
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide
SilicateSum
Figure 213
60
Figure 214
61
Reaction Progress ()0 20 40 60 80 100
Elem
enta
l Wt
0
10
20
30
40
50
ChromiumAluminumIronMagnesium
FeCr2O4Jasper Ridge
Bedrock Chromite
SoilChromite
Figure 215
62
CHAPTER 3
OXIDATIVE PROMOTED DISSOLUTION OF CHROMITE BY MANGANESE
DIOXIDE AND CONCURRENT PRODUCTION OF CHROMATE
ABSTRACT
Chromium release and oxidation from chromite is a potential environmental hazard in
sediments and soils and a pathway for soil development (weathering of primary minerals)
related to ultramafic rocks and their metamorphic derivatives (serpentinites) Birnessite is
a common pedologic mineral in these sediments and soils capable of oxidizing aqueous
Cr(III) In this study the interaction between chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4)
and birnessite is investigated with an interest in the potential generation of Cr(VI)
Specifically the effects of chromite suspension density birnessite suspension density
and pH at a temperature of ~25degC in relation to the reaction kinetics were examined The
rate of Cr(VI) in solution increases with increasing chromite suspension densities and
decreasing pH but is independent of birnessite suspension densities at values greater than
~20 m2 L-1 The overall rate expression of Cr released and oxidized in solution from
chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression with Cr(VI) rates ranging from 9times10-4 to 44times10-3 microM h-1
Additionally these experiments demonstrate that serpentine soils are a source of non-
anthropogenic Cr(VI)
63
INTRODUCTION
Chromite a Cr-enriched spinel is responsible for a majority of the Cr enrichment
identified in ultramafic rocks serpentinites and serpentine soils (Rabenhorst et al 1982
Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989 Gasser and
Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003 Oze et al
2003a Oze et al 2003b) The chemical weathering of chromite is therefore an important
process when evaluating the potential of ultramafic rocks and their related sediments and
soils as sources of Cr Despite the propensity of chromite to resist chemical weathering
the dissolution of chromite and release of Cr will occur over geologic time (Bialowolska
and Salacinski 1984 Oze et al 2003b) Chromium in chromite is in the form of Cr(III)
the non-toxic and chemically immobile valence state however oxidizing conditions in
near surface environments is a potential reaction leading to the enhanced dissolution of
chromite and the formation of Cr(VI) the toxic and chemically mobile valence state
which typically exists as the oxyanion chromate (HxCrO42-x) (Daugherty 1992 Cohen et
al 1993 Fendorf 1995 Deer et al 1996) Consequently chromite may be a non-
anthropogenic source of Cr(VI) in the environment
Naturally occurring high valence Mn oxides present in soils and sediments
derived from ultramafic rocks are strong oxidizers capable of affecting the chemical
mobility of metals through redox reactions (Fendorf and Zasoski 1991 Johnson and
Xyla 1991) These insoluble mixed valent oxides (Mn(IIIIV)) are observed coating
mineral surfaces in these ultramafic sediments and soils where total Mn concentrations
are often gt100 mg kg-1 (Brooks 1987 Schreier et al 1987 Oze et al 2003a Oze et al
2003b Oze et al) Birnessite is a naturally occurring mixed valence oxide known to
oxidize aqueous Cr(III) to the more chemically mobile Cr(VI) (Eary and Rai 1987
Fendorf and Zasoski 1991) Gough et al (1989) and Becquer et al (2003) have
suggested that Mn oxides are responsible for Cr(VI) identified in California and New
Caledonia serpentine soil solutions however the dissolution and oxidation of Cr from a
Cr-bearing mineral due to the presence of birnessite has not been quantitatively defined
In this study we examine the kinetics of Cr release and oxidation when chromite and
birnessite are present in solution
64
The objective of this study is to determine a rate expression describing the
release and oxidation of Cr from chromite in the presence of birnessite Using well-
controlled batch experiments the solution chemistry and geochemistry of Cr related to
chromite and birnessite interaction was evaluated temporally over a range of pH
Conditions were chosen to simulate the slightly acidic conditions characteristic of most
serpentine soils (Rabenhorst et al 1982 Gough et al 1989 Graham et al 1990 Cole
1992 Jeffrey 1992 Lee 1992 Verger 1992 Becquer et al 2003) Ultimately these
experiments allow the examination of one potential pathway leading to Cr(VI) identified
in solutions related to ultramafic rocks and serpentine soils
MATERIALS AND METHODS
Solid Phase Characterization and Synthesis
Chromite used in the experiments was commercial-grade chromite (MSDS No
302-1 WHMIS CLASS D2A) obtained from Barnes Environmental Inc Ontario
Chromite was prepared by cleaning the sand-size material in multiple ultrasonic
isopropanol baths and rinses The chromite was then sieved using a Retsch magnetic
sieve shaker for 30 minutes in order to obtain the 160-250 microm size fraction Impurities in
the 160-250 microm chromite were removed using a slope Franz magnetic separator These
impurities included the minerals olivine augite plagioclase feldspar and quartz Visual
and scanning electron microprobe (SEM) examinations of the remaining 160-250 microm
fraction confirmed its purity Chromite was crushed to a fine powder using an agate
mortar and pestle and then rinsed multiple times in deionized destilled (DDI) water and
001 M HCl solution to remove the ultra-fine material and eliminate highly reactive sites
caused during grinding The chromite powder was inspected before (Figure 31) and
after the experiments using a scanning electron microscope (JEOL 5600LV) the surface
area was 025 m2 g-1 as determined using N2(g) and BET isotherm
An automated JEOL 733A electron microprobe operated at 15 kV accelerating
potential and 15 nA beam current was used to investigate possible zoning and to quantify
the chemical composition of the chromite grains prior to powdering No zoning was
observed and the chemical composition of the chromite was based on the average of 25
electron microprobe analyses The stoichiometric composition of chromite based on 4
65
oxygens is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 X-ray diffraction was also used to
deduce the purity of the chromite powder X-ray diffraction spectra obtained for the
chromite powder matched JCPDS card 03-0873 (unit cell a=8300 b=8300 c=8300
α=90 β=90 and γ=90) and no contaminants were evident
Birnessite was synthesized using the procedure described by Buser et al (1954)
and implemented by Fendorf and Zasoski (1992) A solution of 63 g of potassium
permanganate (KMnO4) in 1 L of DDI water was heated to 90 degC Approximately 100
mL of the heated slurry was poured into a 4 L beaker and reheated to 90degC while being
stirred vigorously The remaining 900 mL of slurry and 66 mL of concentrated
hydrochloric acid were simultaneously and proportionally dispensed into the beaker over
a period of thirty seconds The temperature of the resulting solution was maintained at
90deg C for an additional 10 minutes and then allowed to cool to room temperature The
product was washed with alternate rinses of DDI water and a 001 M HCl solution The
birnessite surface area was 125 m2 g-1 based on BET analyses Solids were inspected
before (Figure 31) and after the experiments using the SEM
Extraction Experiment Procedures
Several chemical extraction experiments were performed on 106-250 microm
chromite grains prior to the chromite-birnessite batch experiments in order to determine
the potential release of Cr Fe and Mn One gram of chromite and 25 mL of solution
were placed into a 25 mL centrifuge tube and the solids and solution were gently agitated
The solutions and parameters used for these extraction experiments include the
following 1) 10 hydrogen peroxide (H2O2) solution at 25degC and pH of ~5 for 6 h 2)
01 M potassium permanganate (KMnO4) solution at 60degC and pH of ~3 for 6 h 3) 10 M
hydrofluoric (HF) acid at 60degC for 6 h 4) 1 M hydroxylamine (NH2OHmiddotHCl) with 25
(vv) acetic acid (HOAc) at 60degC and pH of ~4 for 6 h and 5) 1 mM barium chloride
(BaCl2) solution at 60degC and pH of ~6 for 6 h Solutions were extracted with a syringe
and passed through a 02 microm filter Solids were inspected after the experiments using the
SEM
66
Batch Experiment Procedures
Batch experiments were performed in triplicate using 1 L polypropylene bottles
and chromite and birnessite ratios listed in Table 31 at ~25degC After loading chromite
and birnessite into each bottle 1 L of a pH 3 solution a 10 mM acetate solution (pH 5) a
10 mM MES solution (pH 67) or a pH 8 solution was added to the appropriate
experiment (Table 31) Chromite and birnessite were also run individually in the acetate
buffered solution (Table 31) Aliquots were extracted from the polypropylene bottles
using a syringe and then passed through a 2 microm filter before analyses Less than 40 mL
was extracted over the course of each experiment resulting in a total volume change of
lt4 After the experiments were terminated the remaining solids were resuspended in
20 mL of 1 mM phosphate (5050 KH2PO4K2HPO4 pH 7) 20 mL of 10 mM CaCl2
solution at pH 7 and a 05 M HCl solution These suspensions were agitated for 3 h
followed by centrifugation and filtration (02 microm) before analyses The purpose of the
phosphate and CaCl2 solutions was to desorb reaction produtcts (eg Cr(VI) Cr(III) Fe
and Mn) from the solid phases The HCl treatment provides a means to evaluate
secondary phases such (oxy)hydroxides formed during the experiments Although the
HCl solution may dissolve chromite and birnessite in addition to the (oxy)hydroxides the
short time frame of the extraction (3 h) should limit the dissolution of chromite
Additionally the polypropylene bottles following the termination of five batch
experiments were rinsed with DDI water and 20 mL of 05 M HCl was added to each
bottle The HCl solution was allowed to remain in the bottles for 15 h in order to
determine the concentration of Cr Fe and Mn sorbed onto the interior lining
Serpentine Soil and Birnessite Experiments
Serpentine soils collected from Jasper Ridge Biological Preserve (Stanford CA)
and New Caledonia were used to determine the influence of birnessite in soils known to
contain chromite (Chapters 2 and 4) The Cr geochemistry of these soils as well as the
detrital chromite have been evaluated in Chapters 2 and 4 and by Becquer et al (2003)
Jasper Ridge serpentine soils contain Cr-bearing minerals including chromite Cr-
magnetite and Cr-chlorite however a majority of the Cr enrichment (Table 32) is
directly related to the abundance of chromite A majority of the Cr related to New
67
Caledonia serpentine soils is attributed to Cr (oxy)hydroxides (Chapter 4) however
chromite is present in these soils as well (Becquer et al 2003) Although birnessite has
been noted in these soils the quantity of birnessite could not be determined In order to
evaluate the impact of Mn oxides on Cr within such soils 01 g of birnessite was added to
each soil
The parameters for the experiments are listed in Table 32 Ten grams of the
lt250 microm size-fraction of each soil (JR1 JR2 and NC) and 100 mL of 1 mM acetate
buffered solution (pH 5) were loaded into individual 150 mL Nalgene bottles Aqueous
solutions were evaluated as a function of time (h) Additionally 10 g of each soil 01 g
of birnessite and 100 mL of 1 mM acetate buffered solution (pH 5) were loaded into 150
mL Nalgene bottles and were allowed to react as a function of time (h) These
experiments were rotated at a rate of 2 revolutions min-1 Solutions were removed with a
syringe and passed through a 2 microm filter before analysis Approximately 8 mL of
solution was required for the analyses over the course of each experiment resulting in a
total volume change of 8 Total Cr concentrations (mg kg-1) of the soils listed in Table
32 were obtained by completely dissolving the samples using a mixture of hot (gt200degC)
concentrated nitric perchloric and hydrofluoric acids Elemental compositions were
measured in the supernatant using inductively coupled plasma optical emission
spectroscopy (ICP-OES) The mass of chromite in each soil was calculated using total Cr
concentration for each soil and the calculations provided in Chapters 2 (Table 210) and
Chapter 4 (Table 47) The chromite surface area of each soil was approximated by
comparing the surface area measured for the chromite standard and its observed grain
size from SEM images (Figure 31) to chromites observed in the soils using backscattered
electron images (Chapters 2 and 4) The surface area of the Jasper Ridge chromite is 066
of the chromite standard due to a slightly larger grain size The surface area of the
New Caledonia chromite is approximately 025 of the chromite standard Chromite
suspension densities were calculated by multiplying the chromite mass by the estimated
surface area and then dividing by the volume of the solution
68
Solution Analysis
The pH and Eh of the supernatant was measured with an Orion flat surface
combination pH electrode inserted directly into supernatant Production of soluble
Cr(VI) was measured spectrophotometrically at 540 nm using the s-diphenyl carbazide
method modified from Bartlett and James (1979) Soluble Fe(II) was monitored using
the ferrozine method adopted from Stookey (1970) Total dissolved Cr Fe Al Mn and
Mg were ascertained using ICP-OES Concentrated trace metal grade nitric acid used to
acidify the samples prior to analyses Aluminum and Mg was determined to be present in
the nitric acid at concentrations lt1 microM Magnesium and Al are not plotted in any of the
figures Chromium(III) and Fe(III) was determined by subtracting Cr(VI) and Fe(II)
obtained from colorimetric methods from total dissolved Cr and Fe analyses using ICP-
AES Results are presented as an average of the three runs for each batch experiment
with the standard deviation noted by error bars
RESULTS
Chromite Extraction Experiments
The release of Cr Fe and Mn from chromite using 10 H2O2 01 M KMnO4 10
M HF acid 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 1 mM BaCl2 solutions was
carried out to determine the relative propensity of chromite to resist chemical dissolution
and these results are summarized in Figure 32 The KMnO4 solution with aqueous
Mn(VII) was the most effective in releasing Cr from chromite yielding a concentration
of 62 Cr mg L-1 Iron released using the KMnO4 solution was 03 mg L-1 Although
H2O2 is a direct chemical oxidant the release of Cr from chromite was less as compared
to the other solutions (only ~03 Cr mg L-1 was measured) Concentrated HF was only
able to release ~13 Cr mg L-1 while releasing 106 Fe mg L-1 NH2OHmiddotHCl with 25
(vv) HOAc and CaCl2 solutions only removed lt1 Cr mg L-1 however the NH2OHmiddotHCl
removed 36 Fe mg L-1 which is comparatively higher than the H2O2 KMnO4 and BaCl2
solutions
69
Chromite and Birnessite Batch Experiments
The interaction of chromite and birnessite leading to the release and oxidation of
Cr(III) is a relatively slow process requiring study on the order of hours and days The
pH was maintained constant throughout the duration of the experiments while Eh values
ranged from 320 to 450 mV with no discernable temporal pattern (Figure 33) For pH
values between 3 and 67 Cr3+ CrOH2+ and Cr(OH)2+ are the dominant aqueous Cr(III)
species however at pH 8 conditions are on the stability line between Cr(III) and Cr(VI)
specifically Cr(OH)3 and CrO42-
Higher chromite suspension densities produced higher rates of Cr release to the
aqueous phase and all the Cr in solution was in the form of Cr(VI) (Figure 35) The
rates of oxidation based on Cr(VI) production range from 00005 to 00041 microM h-1
(Figure 35 Table 33) Comparatively Cr released from the chromite control experiment
is maintained at a relatively constant value of ~005 microM and no Cr(VI) is measured in
solution Iron concentrations initially increase to ~ 05 microM and then decrease rapidly to a
relatively constant concentration at 02 microM (Figure 35) Iron concentrations for the
chromite control experiment began at 15 microM between 0 to 50 h and decreased to ~02
microM Iron(II) and Mn concentrations are below detectable limits for all experiments
(Figure 35) A zeroth-order dependence is assumed with respect to birnessite owing to
its large excess surface area relative to chromite
Variable birnessite suspension densities and constant chromite suspension
densities were evaluated at pH 5 (Figure 36) Although suspension densities of
birnessite are 538 and 1075 m2 L-1 (Table 31) the rate of Cr and Cr(VI) measured in
solution for both experiments is 00041 microM h-1 (Figure 36 and Table 33) Iron
concentrations are below 06 microM and maintain concentrations near 02 microM with no
indication of consistently increasing or decreasing with respect to time Iron(II) and Mn
concentrations in solution are below detectable limits for both experiments
In Figure 37 the influence of pH on chromite oxidation by birnessite was
evaluated by conducting experiments at pH values of 3 5 67 and 8 chromite and
birnessite suspension densities are constant at 0075 m2 L-1 and 1075 m2 L-1 respectively
(Table 31) Chromium and Cr(VI) rates increase with decreasing pH ranging from 0
microM h-1 at pH 8 to 00062 microM h-1 at pH 3 (Table 33) Iron is present in solution at
70
concentrations less than 04 microM for all the experiments Generally Fe concentrations are
greater near the start of the experiment and decrease to values lt 05 microM with increasing
reaction time Manganese concentrations are below detectable limits
In order to determine the rate dependence on each prominent reactant (chromite
birnessite and pH) reaction rates were regressed against reactant concentrations (Figure
38) A pseudo first-order rate dependence is noted based on chromite suspension
density the equation regressed for this relationship is listed in Table 33 As the chromite
suspension density increases the rate of Cr(VI) in solution increases In contrast a
zeroth-order rate dependence is evident for birnessite suspension density (dashed line in
Figure 38b) The rate of Cr(VI) generation is expected to decrease with respect to
birnessite suspension density at very low concentrations (dotted line in Figure 38b)
The influence of pH on the rate of Cr(VI) production is well-described by a polynomial
function (Figure 38c Table 33)
Developing an overall rate expression for Cr in solution requires integrating the
factors assessed in Figure 38 and listed in Table 33 All the Cr measured in solution is
in form of Cr(VI) therefore the rate law for Cr or Cr(VI) in solution may be written as
d[Cr(VI))]dt = k sdotChromitea sdot Birnessiteb sdot (aH+)c (31)
and
log d[Cr(VI))]dt = log k + asdotlogChromite+ bsdot log Birnessite+ csdotlog (aH+) (32)
where k is the overall rate constant for Cr release and oxidation Chromite is the
chromite suspension density (Table 31) Birnessite is the birnessite suspension density
(Table 31) and aH+ is the activity of H+ Constants a b and c are the order of
dependency for each reactant The overall rate equation (developed in the Appendix) can
be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069 (33)
71
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 the expression is valid for birnessite suspension densities
greater than 20 m2 L-1 and pH values between 3 to 8
Secondary and Solid Phase Examination of Chromite and Birnessite
The solid products were evaluated to determine the abundance and type of
secondary phases formed during the course of the experiments Solids from the batch
experiment were treated with three solutions 1 mM 5050 KH2PO4K2HPO4 1 mM
CaCl2 and 05 M HCl (Table 34) In the chromite and birnessite reactions Cr was
primarily in solution (gt90 ) The HCl extract was able to remove lt5 additional Cr
while the KH2PO4K2HPO4 and CaCl2 extracts released lt3 additional Cr Aqueous Fe
concentrations were below detectable limits for the final solution extraction Treatment
with the HCl resulted in up to 0127 micromoles of Fe however 0107 micromoles of Fe were
released upon treatment of unreacted chromite with HCl Although not shown in Table
34 Mn values were greater than gt4 micromoles when treated with HCl the
KH2PO4K2HPO4 and CaCl2 removed no detectable Mn
After the termination of each experiment the solids were examined using SEM in
order to identify changes in the surface morphology and secondary phases A large
chromite grain and conglomerate of birnessite adjacent to the chromite are depicted in
Figure 38 The face and edge of the chromite grain is not noticeably altered compared to
grains examined prior the experiments Additionally the particle morphology of
birnessite has not visibly changed Even at higher magnification angular and
conchoidally fractured fine-grained edges are noted (Figure 38bc) Only a slight
rounding of the smaller features of the main chromite grain provides any evidence of
possible chromite dissolution (Figure 38c) Birnessite appears to have limited interaction
with chromite Although chromite and birnessite were allowed to interact over 500 h in
an aqueous solution no secondary phases are observed by SEM
Serpentine Soil and Birnessite Experiments
Two serpentine soils from Jasper Ridge Biological Preserve CA and one
serpentine soil from New Caledonia were used to determine whether or not the overall
72
rate expression (Equation 33) for chromite reaction with birnessite is consistent with
Cr(VI) concentrations produced by serpentine soils containing chromite Three
serpentine soils without birnessite and three soils with the addition of birnessite buffered
at pH 5 were measured for Cr(VI) generation as a function time (Figure 310) The rates
of Cr(VI) measured in solution for both Jasper Ridge soils (JR1 and JR2) are identical at
00013 microM h-1 With the addition of birnessite the rates of Cr(VI) production in solution
both increased to 00044 microM h-1 (JR1 w Bt and JR2 wBt in Figure 310) Based on the
chromite suspension density (Table 32) the rates predicted by Equation 33 range from
00035 to 00041 microM h-1 which are 10 to 20 less than those observed in the
experiments (Figure 310) The rates of Cr(VI) in solution from the New Caledonia
serpentine soil are 00009 microM h-1 and with the addition of birnessite increase up to 00013
microM h-1 (Figure 310) A rate of 00016 microM h-1 was predicted by Equation 33 for the
New Caledonia soil
DISCUSSION
The reaction of birnessite with chromite enhances the rate of Cr release into
solution The rate of Cr(VI) measured in solution (Equation 33) is directly related to the
chromite suspension density and is independent of the birnessite at suspension densities
greater than ~20 m2 L-1 Although the release and oxidation of Cr in solution are
measured the mechanisms and pathways driving this reaction need to be evaluated The
propensity of chromite to resist dissolution the effects of pH the formation of secondary
phases and ultimately the pathways of Cr release and oxidation are addressed in the
following discussion
Extraction experiments performed on chromite (Figure 32) demonstrate it is
resistant to acidic dissolution or competitive ion displacement but strong oxidants such
as high valent Mn may oxidatively dissolve chromite It is evident that the proportions of
Cr Fe and Mn vary for each of the extractions The solutions utilized in the extraction
experiments have varying chemical properties capable of preferentially releasing a
particular species andor element These experiments were also performed over a short
period of time (lt6 h) which may have only released elements from highly reactive sites
73
Finally secondary phases associated with these experiments were possible and would
distort the elemental ratios observed in solution
A majority of the Eh and pH analyses plot in the Cr(III) stability region (Figure
33) however Cr in solution is primarily in the form of Cr(VI) (Figures 35 36 and
37) Eh measurements presented in Figures 32 and 33 do not account for the
microenvironments associated with the surfaces of birnessite Additionally the kinetics
of Cr(VI) reduction are limited in reducing conditions due to electron symmetry
constraints and the absence of a suitable redox couple (Fendorf 1995) Finally highly
oxidizing birnessite in the solution is kinetically and thermodynamically controlling the
oxidation state of Cr relative to the stabilities predicted in Figure 33 (Amacher and
Baker 1982 Fendorf and Zasoski 1992)
A fractional-order dependence on the chromite suspension density (Figure 38)
can be explained by diffusional influences competing oxidation reactions and secondary
reactions involving Cr(III) and Mn(IV) Secondary phases on the surface of chromite
may inhibit the diffusion of Cr from chromite however the clean and relatively unaltered
chromite surfaces do not support this mechanism as a major cause (Figure 38)
Chromium and Fe released from chromite may be incorporated into secondary phases
related to birnessite or chromite Approximately 10 of Cr is not in the aqueous phase
(Table 34) supporting the premise that secondary phases are contributing to the
fractional-order dependence related to the chromite suspension density Also
competition from Fe(II) oxidation on birnessite deposition of Fe(II) on birnessite (or
chromite) and diffusion onwithin birnessite may further impact the rate dependence
The zeroth-order dependence based on the birnessite suspension density (Figure
38) results from its high surface area At very low birnessite suspension densities the
rate of Cr(VI) production should decrease however due to the high surface area of
birnessite it was experimentally prohibitive to determine reaction rates at masses less
than 4 mg Therefore a hypothetical trend-line depicts the expected dependence on
birnessite concentration at very low suspension densities at higher values representative
of most field settings the reaction is independent of the birnessite surface area (ie zero-
order dependence)
74
In Figure 38c the polynomial relationship defines the Cr(VI) rate dependence on
pH As the pH decreases the rate of Cr(VI) increases more rapidly The relationship is
potentially the result of lower pHs progressively increasing the rate of chromite
dissolution However the rate of Cr(VI) in solution might be influenced by higher pHs
increasing the formation of secondary Fe(III) and Cr(III) (oxy)hydroxides on the
chromite andor birnessite surfaces thereby decreasing elemental diffusion from
chromite or decreasing available birnessite sites for Cr(III) oxidation Differentiating
between these two mechanisms is not possible based on the rate equation or extraction
analyses however the polynomial relationship evident in Figure 38c is folded into the
rate constant in order to geochemically account for the influence of pH on Cr(VI)
generation
No conclusive evidence is available to determine the fate of Mn within the
reacting suspensions Aqueous Mn concentrations are consistently maintained below
detectable limits If Mn is released from birnessite then it must be strongly and quickly
incorporated into a nonaqueous phase following its release Due to Cr(III) being oxidized
by Mn(III) andor Mn(IV) Mn(II) must be produced Extractions performed on the
chromite and birnessite solids demonstrate that Mn could only be removed with HCl
however a majority of this Mn is potentially the result of HCl dissolving birnessite On
the basis of electron transfer stoichiometry at least 15 mols of Mn(II) must be produced
for each mol of Cr(VI) generated This amount can be accounted for in the HCl
extraction where gt4 micromoles of Mn were measured Additionally Fendorf et al (1993)
determined that Mn(II) should not significantly influence the extent of Cr(III) oxidation
Iron in solution does not progressively increase in solution over time (Figures 35
36 and 37) The high initial rate of Fe release compared to Cr may be the result highly
reactive sites preferentially releasing Fe Although no Fe is detected in solution from
the extractions performed on the chromite and birnessite solids 0005 to 0035 moles of
Fe were retrieved from the lining of the bottle and 0053 to 0126 moles were released
from the solids using the HCl solution (Table 34) Iron(II) could potentially be oxidized
to Fe(II) in the presence of birnessite and form stable Fe(III)-(oxy)hydroxides (Postma
1985 Myers and Nealson 1988 Villinski et al 2001) A ratio of two Fe for every one
Mn is necessary for the oxidation of Fe(II) and this amount is recoverable based on HCl
75
extractions with gt4 micromoles of Mn measured from the HCl extractions Additionally
dissolved O2 in the batch experiments will rapidly oxidize Fe(II) for all the pHs examined
in these experiments (Stumm and Morgan 1981 Langmuir 1997) The oxidative
pathway for Fe(II) provides an explanation for the lack of Fe(II) in solution for the batch
experiments (Figure 35 36 and 37)
With a detection limit of ~02 microM and equilibrium with respect to Fe(OH)3(am) and
goethite Fe concentrations should either be at or below detection limits at pHs gt 5
Assuming aqueous Fe(III) is present dominantly as the second hydrolysis product at pH
5 the dissolved concentrations approach the solubility of ferrihydrite and are much
greater than expected for goethite (Figure 311) Thus it is probably that the former solid
is generated which is further supported by the chemical extraction experiments (Table
34)
Iron and Cr solution concentrations measured experimentally are not consistent
with congruent dissolution (Figure 312a) Assuming Fe was forming secondary phases
at a constant rate (~00001 Fe h-1) multiplying this rate by the time analysis and then
adding it to the measured aqueous Fe a corrected Fe (Fe) value can be estimated (Figure
312) Additionally the 10 of Cr not in solution can be also be used to calculate a
corrected Cr value (Cr) (Figure 312) When secondary phases are taken into account a
majority of the analyses plot near the congruent dissolution of chromite This provides
evidence that chromite is undergoing congruent dissolution within the batch experiments
but that secondary Fe(III) phases disrupt the solution phase elemental ratios
Chromium(VI) Pathways from Chromite and Birnessite into Solution
Three pathways are evaluated for Cr(VI) measured in solution a) chromite
dissolution b) birnessite dissolution and c) solid-solution reaction (Figure 313) In order
to describe these pathways end-member chromite (FeCr2O4) and birnessite (δ-MnO2) are
used to simplify the analysis It should be recalled however that the composition of the
chromite used in the experiments is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4
Chromite dissolution presents a likely explanation for Cr(VI) observed in solution
based on the following discussion In chromite dissolution model Cr(III) released from
chromite adsorbs onto the surface of birnessite and oxidizes to Cr(VI) After the electron
76
transfer is complete Cr(VI) desorbs from birnessite and remains in solution As this
process continues Cr(VI) increases at a constant rate The rate limiting step with regards
to this specific pathway can be grouped into two steps a) dissolution of chromite (Step
1a) or b) Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) (Figure 313)
Although a dissolution rate for chromite at ambient temperatures and pressures is not
known the chromite extraction experiments (Figure 32) demonstrates that chromite
dissolution is slow Manganese oxides such as birnessite oxidize Cr(III) in solution
rapidly with a measured rate constant of 6300 microM-1 s-1 in fresh water (100 mg L-1 MnO2
100 g L-1 Cr(III) pH~6) (Van der Weijden and Reith 1982) This means that the rate
for Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) is on the order of
seconds therefore chromite dissolution is likely the rate limiting step controlling the
concentration of Cr(VI) in solution for this pathway
Although a zeroth-order dependence related to birnessite concentrations (Figure
37a) is observed the birnessite dissolution pathway in which aqueous Mn4+ released
from birnessite adsorbs onto the surface of chromite oxidizing Cr(III) cannot be entirely
dismissed Manganese dissolution and adsorption could occur quickly not allowing the
Mn to be measured in solution Examination of the solids after the termination of each
experiment demonstrated extremely minimal chromite and birnessite interaction
therefore the solid solution reaction is not a likely pathway leading to the production of
Cr(VI) in solution and is furthermore inconsistent with the noted rate dependencies
Despite the three pathways (Figure 313) the generation of Cr(VI) observed in these
experiments and described by the rate equation (Equation 33) is independent of these
potential pathways
Serpentine Soils and Cr(VI)
The serpentine soil experiments (Figure 310) provide direct evidence that Cr(VI)
is forming within these soils By adding birnessite to the soil the rate of Cr(VI) in
solution increased for all experiments (00013 to 00044 microM Cr(VI) h-1 for Jasper Ridge
and 00009 to 00013 microM Cr(VI) h-1 for New Caledonia) The Cr(VI) rate increase
related to the addition of birnessite is likely the result of low birnessite suspension
densities (lt20 m2 L-1) in the soil Rates predicted for Jasper Ridge and New Caledonia
77
soils using Equation 33 and values in Table 32 are similar to those measured in the
experiments with values of 00041 and 00016 microM Cr(VI) h-1 respectively Overall these
soil experiments demonstrate that a rate equation developed from laboratory batch
experiments predicts the rate of Cr(VI) in serpentine soils reasonably well
According to limits set by the Environmental Protection Agency the California
standard for total Cr in drinking water is 50 microg L-1 (~1 microM) while the federal standard is
100 microg L-1 (~2 microM) The serpentine soil and birnessite experiments produced Cr(VI)
concentrations in excess of 50 microg L-1 after only 140 h or roughly 6 days Serpentine soils
at Jasper Ridge typically remain water saturated for prolonged periods of time (gt1 week)
during the winter season increasing the time Cr(VI) can be produced The degree and
duration of water saturation is dependent on the topography of the serpentinite body
level areas will remain water saturated while sloping areas will allow the water to flow
down gradient Although Cr(VI) was generated in the soil experiments the production of
Cr(VI) in serpentine soils may be more limited The reduction of Cr(VI) to Cr(III) is
favorable by reactions with Fe(II)aq Fe(II)-containing minerals such as magnetite H2S
and organic matter including microbial processes (Ilton and Veblen 1994 Fendorf 1995
Peterson 1996) Both organic matter and magnetite are abundant in Jasper Ridge
serpentine soils (Chapters 2 and 4) however these factors did not deter Cr(VI) from
forming during the serpentine soil experiments Additionally the soils in these
experiments were constantly rotated allowing the mineral surfaces to be continuously
exposed to solution Chromite within an undisturbed soil will not always be in contact
with the solution or proximal to birnessite
Chromium(VI) produced by the interaction of chromite and birnessite is not
limited to serpentine soils Serpentinites the bedrock for serpentine soils are typically
highly sheared due to the diapiric migration of serpentinite masses along zones of
weakness in fault zones Numerous cracks and lineations which are typical of these rocks
provide a means for Cr(VI) to transported away from the serpentine soil potentially into
local groundwater In fact preferential flow will occur along the major shear plains of
the serpentinite Additionally these shear plains are often coated with high valent Mn
oxides providing a reaction pathway for Cr(VI) formation Serpentinite springs such as
those evaluated by Barnes and ONeil (1969) and Barnes et al (1972) at the Cedars
78
ultramafic body in Cazadero CA have Cr concentrations measured at 20 microg L-1 in which
all the Cr is in the form of Cr(VI) (Chapter 4) In conclusion chromite and birnessite in
serpentine soils could contribute a source of Cr(VI) in surface waters and groundwaters
Chromium release and oxidation from chromite is increased by the presence of
birnessite and the interaction between chromite and birnessite is a pathway for Cr(VI)
generation in ultramafic sediments and soils The potential of serpentine soils as a source
of non-anthropogenic Cr contamination can now be evaluated quantitatively based on the
abundance of chromite The overall rate law for Cr(VI) in solution (Equation 39)
provides a means to estimate the Cr(VI) concentration attainable for serpentine soils and
serpentine sediments when the chromite and birnessite suspension densities and the pH of
the soil are known
ACKNOWLEDGEMENTS
We gratefully acknowledge comments and assistance from Robert Jones and Dr
Guangchao Li Funding for this research was provided by the Office of Teaching and
Learning at Stanford University
79
Table 31 Chromite and birnessite batch experiment parameters with chromite and birnessite suspension densities
Experiment Chromitea (grams)
Birnessiteb (grams) pHc
Chromite Surface Aread
(m2 g-1)
Birnessite Surface Aread
(m2 g-1)
Chromite Suspension
Density (m2 L-1)
Birnessite Suspension
Density (m2 L-1)
A 08 043 5 025 125 02 54 B
08 086 5 025 125 02 108C 03 086 5 025 125 008 108D 008 086 5 025 125 002 108E 004 086 5 025 125 001 108F 03 086 3 025 125 008 108G 03 086 67 025 125 008 108H 03 086 8 025 125 008 108
Chromite 08 ---- 5 025 125 02 ----Birnessite ---- 043 5 025 125 ---- 5480
a Chromite chemical composition is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 b Synthetic birnessite (δ-MnO2) c10 mM acetate buffered solution pH = 5 non-buffered pH=3 5 mM MES buffer pH 67 non-buffered pH 8 dMeasured BET analyses
81
Table 32 Serpentine soil and birnessite experiments with soil and birnessite masses used and the estimated chromite suspension density
Samples Soil Mass (g)
Birnessite Mass (g)
Volume (L) of Solutiona
Total Crsoil
b (mg kg-1)
Chromite Mass (g) in
Soil Samplec Estimated Surface
Areac (m2 g-1)
Chromite Suspension Density
(m2 L-1) in Soild Jasper Ridge
(JR1) 10 ---- 01 5976 013 015 02 Jasper Ridge (JR1 w Bt) 10 01 01 5976 013 015 02 Jasper Ridge
(JR2) 10 ---- 01 4978 011 015 016 Jasper Ridge (JR2 w Bt) 10 01 01 4978 011 015 016
New Caledonia (NC) 10 ---- 01 827 0015 006 001
New Caledonia (NC w Bt) 10 01 01 827 0015 006 001
a1mM Acetate Buffer (pH=5) 81 bTotal Cr concentrations were obtained through total digestion and ICP-AES analyses
cEstimated chromite mass and surface area (discussed in the Methods section) was made using chromite observations and calculations in Chapters 2 and 4
dCalculated from the chromite mass in the soil sample estimated surface area of the chromite and the volume of solution
82
Table 33 Summary of Cr(VI) rates obtained from batch experiments (Figure 35 36 and 37) and used in Figure 38
Listed in Table 31
Suspension Densities (m2 L-1)
Chromite (Ch)
Birnessite (Bt)
pH Rate (microM h-1) (d[Cr(VI)]dt)
log Rate (log microM h-1) (log d[Cr(VI)]dt) Expression
Figure 35 02 108 5 00041 -239
008 108 5 00021 -268 002 108 5 00009 -305 001 108 5 00005 -330
1
Figure 36 02 54 5 00041 -239 02 108 5 00041 -239 2
Figure 37 008 108 3 00062 -221 008 108 5 00021 -268 008 108 67 00004 -340 008 108 8 0 ----
3
Expression 1 log d[Cr(VI)]dt = asdot log Chromite +log krsquorsquo log d[Cr(VI)]dt = 06908 log Chromite ndash 1899 Expression 2 log d[Cr(VI)]dt = bsdot log Birnessite + log krsquorsquorsquo
log d[Cr(VI)]dt = 0sdot log Birnessite ndash 23872 log d[Cr(VI)]dt = -23872
Expression 3 d[Cr(VI)]dt = 264E-4sdotpH2 ndash 414E-3sdotpH + 162E-2
82
Table 34 Extractions comparing total solution and extractable Cr and Fe from the solids
------------------Cr----------------- ------------------Fe---------------- microMolesdagger Total
mole Extraction
mole microMoles Total mole
Extraction mole
Chromite Ch 02 (Table 31) Terminated at 30 days Solutiona 0033 554 ---- 0 0 ---- Phosphateb 0000 05 1 0000 03 03 CaCl2
c 0013 388 87 0000 00 00 HCld 0002 53 12 0103 997 997 Bottle HCle 0000 00 0 0000 00 00
A Ch 02 Bt 54 (Table 31) Terminated at 30 days Solution 2977 910 ---- 0 0 ---- Phosphate 0048 16 18 0004 21 21 CaCl2 0055 18 20 0002 09 09 HCl 0161 54 60 0126 758 758 Bottle HCl 0004 01 2 0035 212 212
B Ch 02 Bt 108 (Table 31) Terminated at 30 days Solution 2837 914 ---- 0 0 ---- Phosphate 0039 14 16 0008 49 49 CaCl2 0060 21 25 0000 00 00 HCl 0142 50 58 0127 803 803 Bottle HCl 0004 01 2 0024 148 148
C Ch 08 Bt 108 (Table 31) Terminated at 17 days Solution 0877 894 ---- 0 0 ---- Phosphate 0020 23 22 0002 37 37 CaCl2 0030 34 32 0000 00 00 HCl 0040 46 43 0053 881 881 Bottle HCl 0003 03 3 0005 82 82
daggerMolarity values are normalized to the final solution analysis of each experiment based on volume of solution and mass of the sample in each experiment aSolution refers to the final solution analysis before the experiment was terminated bmicroMoles extracted from the batch experiment solids using a 1 mM phosphate buffer pH 7 3 h cmicroMoles extracted from the batch experiment solids using a 10 mM CaCl2 solution pH 7 3 h dmicroMoles extracted from the batch experiment solids using a 05 N HCl solution 3 h emicroMoles extracted from the batch experiment bottles using a 05 N HCl solution 15 h
83
Figure Captions
Figure 31 Scanning electron microscopic (SEM) images of chromite and birnessite at
varying magnifications prior to the batch experiments
Figure 32 A summary of Cr Fe and Mn released from chromite during extraction
experiments Solutions used for these experiments include 1) 10 H2O2 solution 2) 01
M KMnO4 solution 3) 10 M HF 4) 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 5) 1mM
BaCl2 solution () Fe concentration for the HF extraction is 106 mg L-1
Figure 33 pH and Eh measurements as a function of time (h) for the batch
experiments Chromite (Ch) and birnessite (Bt) suspension densities used for the batch
experiments (Table 31) are listed adjacent their corresponding symbol in the key
Figure 34 Eh and pH batch experiment analyses (Figure 33) plotted on a stability field
diagram for Cr (log a Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom
(1998)
Figure 35 Batch experiments evaluating variable chromite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr and Cr(VI) measured in solution are noted and illustrated with a dotted line and
included in Table 33
Figure 36 Batch experiments evaluating variable birnessite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr(VI) measured in solution are denoted and included in Table 33
Figure 37 Influence of pH on Cr(VI) generation by reaction of constant chromite (Ch
08 m2 L-1) and birnessite (Bt 108 m2 L-1) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key Rates of Cr(VI) measured in solution are included in Table
33
84
Figure 38 Rates of Cr(VI) in solution (Table 33) as a function of a) chromite
suspension density (Figure 35) b) birnessite suspension density (Figure 36) and c) pH
(Figure 37) The line in each diagram represents the best fit for these data points and the
equations for these lines are listed in Table 33 The dotted line deviating from the
dashed line in b) indicates the probable path at low birnessite concentrations
Figure 39 SEM images of the solid phases after the completion of the batch
experiments A (Ch 02 Bt 54) and B (Ch 02 Bt 108) are shown Chromite is the larger
grain with relatively flat surfaces whereas birnessite is the fine-grained rosettes located
adjacent the larger chromite grains
Figure 310 Chromium and Cr(VI) concentrations (microM) in solution as a function of
time (h) from serpentine soils obtained at Jasper Ridge (JR1 and JR2) and New Caledonia
(NC) Soils with the addition of 01 g birnessite are noted in the figure Soil and
birnessite masses used for each experiment are listed in Table 32
Figure 311 Iron analyses as a function of time (Figure 35 and 36) with the solubility
of amorphous Fe(OH)3 and goethite and the dominant Fe3+-OH complex at pH 5 Fe
solubilities are from Langmuir (1997) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key
Figure 312 a) Experimental values of Fe and Cr (Figures 35 and 36) are plotted A
dotted line denotes the ratio of Fe and Cr necessary for chromite congruent dissolution b)
Corrected FeCr ratios as a function of time (h) for data from Figures 35 and 36 taking
into account the concentration of Fe and Cr in secondary phases calculated from Table
34 c) Corrected Fe and Cr as compared to values predicted with congruent dissolution
of chromite
85
Figure 313 Three potential pathways of Cr release and oxidation for chromite-
birnessite interactions are listed
86
Figure 31
87
0
2
4
6
8
10
12
Hydrogen
Peroxid
e
K-Perman
ganate
Hydroflu
oric Acid
Hydroxy
lamine
Barium Chlorid
e
Con
cent
ratio
ns (m
g kg
-1)
Cr
Fe
Mn
Figure 32
88
Time (h)0 100 200 300 400 500
pH
2
3
4
5
6
7
8
9
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
Time (h)0 100 200 300 400 500
Eh (m
V)
100
200
300
400
500
600
Ch Bt ------(m2 L-1)------
Figure 33
89
pH2 3 4 5 6 7 8 9
Eh (m
V)
-05
00
05
10
15
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
HCrO4-
CrO42-
Cr3+
CrOH2+
Cr(OH)3Cr(
OH
) 2+
Upper limit of water stability
Lower limit of water stability
Ch Bt ------(m2 L-1)-----
Figure 34
90
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 108 008 108 002 108 001 108 02 --- --- 54
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
0
1
2
3
4
pH 5 pH 5
pH 5 pH 5 pH 5
Ch Bt ---(m2 L-1)---
Figure 35
91
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 54 02 108
00041 microM h
-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
00
02
04
06
08
10
00041 microM h
-1
pH 5
pH 5 pH 5 pH 5
pH 5 Ch Bt----(m2 L-1)----
Figure 36
92
Time (h)
0 50 100 150 200 250
Cr c
once
ntra
tion
( microM
)
00
02
04
06
08
10
pH 3pH 5pH 67pH 8
Time (h)0 50 100 150 200 250
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250M
n co
ncen
trat
ion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
00004 microM h-1
00021 microM h-1
0006
2 microM h
-1
00021 microM h-100
062 micro
M h-1
00004 microM h-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1
Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Figure 37
93
log Birnessite m2 L-1
05 10 15 20 25 30
log
d[C
r(VI
)]dt
microM
h-1
-40
-35
-30
-25
-20
pH 5
log Chromite m2 L-1-22 -20 -18 -16 -14 -12 -10 -08 -06
log
d[C
r(VI
)]dt
microM
h-1
-34
-32
-30
-28
-26
-24
-22
pH 5
pH1 2 3 4 5 6 7 8 9
d[C
r(VI
)]dt
microM
h-1
0000
0002
0004
0006
0008
0010
a)
c)
b)
Ch 008 m2 L-1 Bt 108 m2 L-1
Figure 38
94
Figure 39
95
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR1JR1 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR2JR2 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
01
02
03
04
05
NCNC w Bt
00044 microM h-1
00013 microM h-1
00044 microM h-1
00013 microM h-1
00013 microM h-1
00009microM h-1
pH 5
pH 5
pH 5
Figure 310
96
Time (h)0 100 200 300 400 500
log ΣF
e (aq)
microM
-10
-8
-6
-4
-2
0
2
4
02 54 02 108 008 108 002 108 001 108
Fe(OH)3(am)
Goethite
pH 5Ch Bt----m2 L-1----
Fe(OH)2(aq)+
Figure 311
97
Time (h)
0 100 200 300 400
FeC
r Rat
io
0
1
2
3
4
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
02 54 02 108 008 108 002 108 001 108
Chrom
ite C
ongr
uent
Dissolu
tion
Chrom
ite C
ongr
uent
Dissolu
tion
a)
b)
c)
Ch Bt----(m2 L-1)----
ChromiteCongruent Dissolution
Figure 312
98
Pathway 1 Chromite Dissolution
1a) Cr dissolution from chromite FeCr2O3(s) + 4H+ rarr 2Cr(OH)2+
aq + Fe2+(Oxyhydroxide) + H2O
1b) Adsorption of CrOH2+ on the surface of birnessite Cr(OH)2+
aq + MnO2(s) rarr Cr(III)-MnO2(s) + OH- 1c) Oxidation of CrOH2+ on the surface of birnessite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
1d) Desorption of Cr(VI) from birnessite Cr(VI)-Mn(II) + 4H2O rarr Mn2+
(aq) + HCrO4-(aq) + 7H+
Pathway 2 Birnessite Dissolution 2a) Mn4+ dissolution from Birnessite MnO2(s) + 4H+ rarr Mn4+
(aq) + 2H2O 2b) Mn4+ adsorption on the surface of chromite Mn4+
(aq) + FeCr2O4(s) rarr Mn(IV)-FeCr2O4 2c) Oxidation of Cr(III) on the surface of chromite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
2d) Desorption of CrO24- from chromite
Cr(VI)-Mn(II) + 4H2O rarr Mn2+(aq) + HCrO4
-(aq) + 7H+
Pathway 3 Solid Solution Reaction 3) Solid solution interaction between chromite and birnessite
Figure 313
99
CHAPTER 4
CHROMIUM GEOCHEMISTRY OF SERPENTINE SOILS
ABSTRACT
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
chemistry of Cr in serpentine soils including its protolith is reviewed focusing on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils are slightly acidic (average pH of ~6) contain a variety Fe(III) oxides
(magnetite and hematite) phyllosilicates (serpentine and chlorite) and clays (smectites
and vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1)
and Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Chromium(III) is the only valence state observed
in the serpentine soil solids however Cr(VI) has been identified in New Caledonia and
California serpentine soil solutions at concentrations below 30 microM The enrichment and
range of Cr concentrations in serpentine soils are directly related to the presence of Cr-
spinels specifically chromite and Cr-magnetite These spinels are resistant to weathering
and are preserved in the soil environment however oxidation of Cr(III) from Cr-spinels
by high valent Mn oxides or other strong oxidants is a potential source of Cr(VI)
identified in serpentine soil solutions Due to the chemically and physically resistant
nature of the Cr-spinels Cr-bearing silicates including clay minerals Cr-chlorite Cr-
garnet Cr-mica and Cr-epidote are more viable sources of Cr identified in vegetation
soil extractions soil solutions and related waters
100
INTRODUCTION
Serpentine soils a generic term used to describe any soil derived from an
ultramafic rock (peridotites and pyroxenites) or serpentinite (metamorphosed ultramafic
rock) are often characterized by 1) low concentrations of plant nutrients such as N P
and K 2) high concentrations of biologically toxic elements including Ni Co and Cd 3)
low ammonium acetate extractable CaMg ratios and 4) unique flora and physical
properties (Brooks 1987) Measured Cr concentrations in serpentine soils range from
~200 mg kg-1 to 6 Wt values enriched compared to non-serpentine soils lt200 mg kg-
1 (Kabata-Pendias and Pendias 1984 Schwertmann and Latham 1986 Brooks 1987)
Chromite and Cr-magnetite are common primary sources of Cr in these soils (Rabenhorst
et al 1982 Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989
Gasser and Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003)
however the plethora of factors relating the chemistry of Cr in serpentine soils and the
weathering of chromite has not been fully addressed using an integrative approach
Overall serpentinites and serpentine soils cover ~1 of the earthrsquos exposed
surface and these rocks and soils are commonly associated with ophiolite complexes
(Coleman 1977) Serpentinites and serpentine soils related to ophiolite complexes are
the primary interest of this paper A map illustrating the worldwide distribution of major
ophiolitic serpentinite bodies and their related soils is shown in Figure 41 A majority of
the sites are near convergent boundaries (Figure 41) Although these rocks and soils
cover a minimal area of the earthrsquos surface they are present in populated areas within the
Circum-Pacific Margin and the Mediterranean Due to the wide-distribution and
occurrence of serpentinites and serpentine soils weathering of these rocks and soils will
differ from location to location due to varying climatic conditions as well as the nature of
the parent material and other factors including topography biota and time (Proctor and
Nagy 1992) As a result the chemistry of these soils as well as the nature of Cr in the
serpentinites and serpentine soils will vary from site to site potentially allowing Cr to be a
source of non-anthropogenic contamination
The objective of this paper is to review the chemical characteristics of serpentine
soils globally evaluate mineralogical and compositional variations between several soils
derived from ultramafic sources and assess the geochemistry of Cr in diverse soil-
101
forming environments Additionally chemical and spectroscopic analyses were
performed on soils derived from ultramafic rocks collected from 1) New Caledonia
South Pacific 2) Eight Dollar Mountain Selma OR 3) Nickel Mountain Riddle OR 4)
Harvard Mine Jamestown CA 5) Jasper Ridge Biological Preserve Stanford CA and 6)
Pillikin Mine El Dorado CA shown geographically in Figure 41 A basalt-derived soil
from Hawaii is included in this study in order to compare a soil formed from a mafic
protolith to those from ultramafic protoliths The various field sites are shown in Figure
42 These soils allow a comparative examination of how soil-forming factors such as
climate topography biota parent material and time affect the chemistry of Cr
REVIEW
Chromium Geochemistry
Chromium is a group VIB transition metal with two common oxidation states
Cr(III) and Cr(VI) differentiated by noticeably different chemical characteristics and
toxicities It should be noted that Cr(II) has been observed in lunar basalt olivines and in
high temperature terrestrial olivines (Sutton et al 1993 Li et al 1995) however this
oxidation state is not observed in soil environments A majority of Cr commonly persists
in ultramafic rocks serpentine soils and related waters as Cr(III) a non-hazardous
species and a micronutrient in human nutrition (Daugherty 1992 Cohen et al 1993
Fendorf 1995 James et al 1997) Chromium(III) has the highest crystal field
stabilization energy for all transition metals strongly favors octahedral coordination and
has 3 d-electrons with a high spin state Chromium(III) is strongly hydrolyzed in
aqueous solutions and the predominant species based on pH redox potential and O2(aq)
are shown in Figure 43 Chromium(III) species are more prevalent under reducing
conditions and its mobility is increased only under strongly acidic conditions At pH
values greater than 9 the hydrolyzed Cr(III) species is negatively charged demonstrating
the amphoteric properties of Cr(III) (Figure 43) Due to the factors described above
Cr(III) forms low solubility compounds and adsorbs strongly on mineral surfaces and
organic matter at pHgt4 (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) For
example the adsorption of Cr(III) on HFO (hydrous ferric oxide) as a function of pH is
shown in Figure 43 illustrating an adsorption edge at a pH of ~4 (Leckie et al 1984
102
Stumm 1992) Chromium(III) oxidation to Cr(VI) has been observed in reactions
involving hydrogen peroxide (H2O2) and high valent Mn oxides (Eary and Rai 1987
Johnson and Xyla 1991 Fendorf et al 1992 Fendorf et al 1993 Rock et al 2001)
Chromium(VI) is a strong oxidizing agent a highly corrosive substance a toxin to
living cells and a Class A human carcinogen by inhalation (Daugherty 1992 Cohen et
al 1993 Fendorf 1995) The hydrolyzed Cr(VI) species are also shown in Figure 43
with respect to pH redox potential and O2(aq) Chromium(VI) is more favorable under
oxidizing conditions and its mobility is enhanced under basic conditions In solution
Cr(VI) is a weakly adsorbing (oxy)anion (HxCrO42-x) on common weathering products
meaning that it forms highly labile complexes that can be readily displaced by ions
including phosphate and sulfate (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Additionally the adsorption of CrO42- on HFO shown in Figure 43
demonstrates the anionic character in which the mol bound of CrO42- increases with
decreasing pH and an adsorption edge is at pH ~72 (Leckie et al 1980 Stumm 1992)
The reduction of Cr(VI) to Cr(III) is known to occur by reactions with Fe(II)aq Fe(II)-
containing minerals such as magnetite H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996)
Mineralogy of Serpentine Soils
Serpentine soils have been identified on every continent excluding Antarctica
(Figure 41) The occurrence and distribution of ultramafic rocks specifically related to
ophiolites are closely associated with convergent plate boundaries (Figure 41 Coleman
1977 Coleman and Jove 1992) Tectonic blocks of ultramafic rock are hypothesized to
be detached from the down-going slab of oceanic crust and incorporated into the
subduction meacutelange Subsequent serpentinization of peridotite forms low density
serpentine group minerals (245-250 g cm-3) resulting in the upward diapiric migration of
the serpentinite masses along zones of structural weakness (fault zones) and through the
denser enclosing subduction meacutelange (~265 g cm-3) For example North America has
numerous exposed ophiolites containing serpentinite located in western California and
Oregon where subduction zones were once active in the Early Cretaceous (Harden
1998)
103
Serpentinization of ultramafic rocks is a significant factor controlling mineral
assemblages and textures in serpentinite Hydration and oxidation of olivine and
pyroxene during serpentinization forms serpentine (lizardite chrysotile and antigorite)
talc chlorite tremolite brucite and magnetite (Coleman 1977 Brooks 1987 Coleman
and Jove 1992 Malpas 1992) Additionally magnetite formation creates favorable
conditions for native metal compounds (such as Fe Co and Ni alloys) to form (Coleman
and Jove 1992) As a serpentinite mass is emplaced the extensive physical and chemical
alteration of the serpentinite results in numerous crosscutting alteration veins and highly
sheared textures Ultimately the mineral assemblages and textures in serpentinites are
complex prior to pedogenesis due to the multiple phases of metamorphosis tectonic
deformation and weathering of the rock
Serpentine soils consist of a mixture of residual minerals from the soil protolith
eolian and illuvial minerals and minerals formed within the soil Additionally highly
weathered serpentine soils enriched in Fe-oxides are also termed laterites soils common
in tropical and subtropical regions Soil protolith minerals may include primary minerals
of ultramafic rocks including olivines pyroxenes and oxides (chromite and magnetite)
as well as a large variety of metamorphic minerals the most common being serpentine
(lizardite chrysotile and antigorite) chlorite talc and actinolite Silica phases occur as
eolian grains concretions andor net type structures or silica webs (Hotz 1964
Coleman 1977 Schwertmann and Latham 1986) Eolian and illuvial aluminosilicate
minerals are important sources of Al in serpentine soils an element of low abundance in
ultramafic rocks
Layer silicates observed in serpentine soils are varied in type and abundance
They include smectites (often as montmorillonite) vermiculite chlorite brucite
serpentine and a variety of intrastratified clays (Veniale and Van der Marel 1963 Hotz
1964 Schellmann 1964 Wildman et al 1968 Rabenhorst et al 1982 Schreier et al
1987 Graham et al 1990 Bulmer et al 1992 Gasser and Dahlgren 1994 Lee et al
2003) Compositional variations crystal structures and parageneses of layer silicates
have been characterized for a number of serpentine soils (Wildman et al 1968
Rabenhorst et al 1982 Schreier et al 1987 Graham et al 1990 Bulmer et al 1992
Lee et al 2003) In addition to the layer silicates goethite hematite maghemite and
104
amorphous Fe (oxy)hydroxides are soil-forming phases in serpentine soils (Coleman and
Jove 1992) Gibbsite is not a common phase in most serpentine soils however limited
isomorphic substitution of Al(III) in goethite and hematite was noted in New Caledonia
(Schwertmann and Latham 1986) Manganese oxides such as birnessite (δ-MnO2) are
also present in serpentine soils and often coat mineral surfaces
Sources of Chromium for Serpentine Soils
Chromium primarily occurs in oxide and silicate phases in ultramafic rocks and
serpentinites Chromium-spinel often referred to as chromite is a common primary
(igneous) spinel in mafic and ultramafic rocks and often resists low-grade metamorphic
processes related to serpentinization (Hoffman and Walker 1978 Malpas 1992)
Isomorphic substitution of Al3+ Fe3+ and Ti4+ into the octahedral sites and Mg2+ Ni2+
Zn2+ and Mn2+ into the tetrahedral sites is common however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987b Pan and Fleet 1991 Sack and Ghiorso 1991 Burkhard 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996 Mathiesen 1999) Serpentinite chromite compositions reported in the literature are
plotted in Figure 44 to demonstrate the wide-range of Cr concentrations possible for Cr-
spinels Isomorphic substitution of octahedral Cr and octahedral Al for these chromite
compositions are also shown in Figure 44 Serpentinite pyroxenes containing Cr include
augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) (Grapes 1981 Von Knorring et al
1986 Treloar 1987a Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996)
Magnetite and serpentine minerals may incorporate varying amounts of Cr into
their structures during serpentinization Chromium(III) from primary Cr-containing
phases may isomorphically substitute into the Fe(III) site of magnetite due to size
(octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge similarities Chromium
concentrations in magnetite have been observed as high as 13 Wt (Oze et al 2003)
Chromate minerals such as crocoite (PbCrO4) are extremely rare in serpentinites and the
occurrence of these minerals are mainly limited to specific localities such as in Australia
Tasmania and the Urals (Williams 1974 Crane et al 2001) Serpentine minerals have
105
been reported to contain as much as 1500 mg kg-1 Cr based on spectrographic analyses
by Page and Coleman (1967) however serpentine minerals typically contain only trace
amounts of Cr
Hydrothermal and CO2 metasomatism of ultramafic rocks during the evolution of
convergent plate margins commonly produce Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite lt19 Cr Wt ) uvarovite (Cr-garnet lt20 Cr Wt )
tawmawite (Cr-epidote lt13 Cr Wt ) and kaumlmmererite (Cr-chlorite lt10 Cr Wt ) as
shown in Figure 44 (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974 Neiva
1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981 Max et al
1983 Nutman et al 1983 Raase et al 1983 Von Knorring et al 1986 Kerrich et al
1987 Treloar 1987b Treloar 1987a Morand 1990 Pan and Fleet 1991 Schandl and
Wicks 1993 Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995
Mathiesen 1999 Proenza et al 1999) Figure 44 illustrates the relationship of Cr and
Al substitution for Cr-silicate compositions based on Cr and Al Wt Additionally Cr-
silicate compositions are plotted based on octahedral Cr and octahedral Al in Figure 44
Compositions for Cr-muscovite Cr-epidote and Cr-garnet plot near the tie lines
connecting Cr and Al end-member minerals Chromium isomorphically substitutes for
octahedral Al in these silicates due to similarities in charge octahedral preference and
octahedral radii (Al3+ = 061 Aring) The numerous isomorphic substitutions possible for Cr
and Al in the chlorite structure results in analyses that do not plot near the tie line
connecting clinochlore and kaumlmmererite Overall any of these minerals are potential
sources for Cr in soils produced by the weathering of serpentinites and their hydrothermal
and CO2 metasomatic derivatives
Chromium in Serpentine Soils
Chromium concentrations vary widely in serpentine soils worldwide (Table 41)
New Caledonia soils appear to be the most Cr-enriched serpentine soil with a Cr
concentration of ~60000 mg kg-1 (Schwertmann and Latham 1986) however Jaffre
(1980) report values ranging from 6300 to 80000 mg kg-1 for New Caledonia soils
Some serpentine soils have measured Cr concentrations as low as 29 and 66 mg kg-1
atypical of most values reported in Table 41 (Proctor and Woodell 1971 Brooks et al
106
1992 Jeffrey 1992) Chromium concentrations as a function of depth for serpentine soil
profiles in Maryland (Rabenhorst et al 1982) Tehama County CA (Gough et al 1989)
the Klamath Mountains CA (Hotz 1964) Newfoundland Canada (Roberts 1992)
Harvard Mine CA (this study) and Jasper Ridge CA (Oze et al 2003) are reported in
Figure 45 The pedogenic process(es) responsible for the varying Cr concentrations as a
function of depth for each soil have not been characterized however Oze et al (2003)
determined that a majority of Cr in Jasper Ridge serpentine soils is the result of
weathering-resistant residual Cr-bearing minerals from the protolith mainly chromite
and Cr-magnetite Additionally Oze et al (2003) demonstrated that soil samples from
Jasper Ridge contain higher concentrations of Cr with respect to their corresponding
bedrock due to the Cr-spinels accumulating over time The observation that Cr
concentrations decrease with depth towards the protolith is characteristic for most of the
profiles in Figure 45
In addition to examining total Cr concentrations chemical extractions performed
on serpentine soils provide additional information with regards to Cr availability and the
minerals and complexes in which Cr may reside in the soils Table 42 summarizes a
number of extraction experiments performed on serpentine soils For a majority of the
soils listed in Table 42 total Cr concentrations for the serpentine soils were not
measured however these extractions do provide a quantitative assessment of where Cr
may be bound in soil (such as silicates organic matter and secondary Fe-bearing phases)
Extractions performed by Schwertmann and Latham (1986) were able to remove ~20000
Cr mg kg-1 from the serpentine soil using a dithionite-citrate bicarbonate (DCB) solution
which preferentially dissolves crystalline Fe-oxides Solutions such as concentrated
nitric and hydrochloric acid were only able to remove a maximum of ~1400 Cr mg kg-1
from Unst serpentine soils (Shewry and Peterson 1976) which preferentially dissolve
Cr-bearing silicates and to a lesser extent the Fe-oxides Where bulk soil Cr values were
measured extraction experiments are only able to remove a small portion (lt1) of the
total Cr present in the serpentine soils (Table 42) (Gasser and Dahlgren 1994 Oze et al
2003) Based on the extractions presented in Table 42 it appears that Cr(III) resides in
minerals and phases resistant to common extraction techniques
107
In serpentine soils Fe(III) oxides and Fe(III) hydroxides provide a host of
structural and surface sites available to sequester Cr(III) released to soil solutions by
dissolution of Cr-bearing minerals during soil-forming processes Substitution of Cr(III)
for Fe(III) in Fe-oxides Fe-(oxy)hydroxide minerals and amorphous compounds is
expected based on crystallographic considerations synthesis experiments and analysis of
extracts from selective dissolution of Fe(III) compounds in serpentine soils
(Schwertmann and Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990
Gerth 1990 Gasser and Dahlgren 1994) Chromium(III) also readily sorbs on surfaces
of Fe(III) oxides and (oxy)hydroxides via inner-sphere surface complexes (Figure 43)
and forms low-solubility precipitates (Charlet and Manceau 1992) The low
concentration of Cr(III) in soil solutions has been attributed to the low solubility of
Fe(III)-Cr(III) amorphous hydroxides (Sass and Rai 1987)
Layer silicates provide structural and surface sites capable of uptaking Cr(III)
Octahedral Cr(III) in clays such as smectites and vermiculites is expected and Cr
concentrations in the clay-sized fraction can achieve concentrations of ~1000 mg kg-1 in
serpentine soils (Oze et al 2003) Sorption of Cr(III) and Cr(VI) on edge and interlayer
sites of phyllosilicates have been evaluated by Ilton and Veblen (1994) Ilton et al (2000)
and Brigatti et al (2000) indicating that layer silicates (montmorillonite chlorite biotite
phlogopite and muscovite) provide sites for both Cr(III) sorption and Cr(VI) reduction
Kaupenjohann and Wilcke (1995) concluded that Cr in serpentine soils from Bavaria is
residually bound mainly in silicate structures including clay minerals based on extraction
experiments
Soil Solutions and Related Water
The addition of water gradually changes the spatial and temporal characterization
of pore fluids in soils including the ionic strength pH and redox potential Leaching of
a soil should progressively enrich the soil with less chemically mobile elements such as
Cr and Fe Elements identified in soil solutions include Mg Ca Fe Ni and Cr and the
ionic strengths of serpentine soil solutions are typically low (lt001) (Anderson et al
1973 Johnston and Proctor 1981 Gasser and Dahlgren 1994) Soil solutions obtained
from serpentine soils using tension-free lysimetry demonstrate Cr concentrations ranging
108
from 01 to 32 micromol L-1 (Gasser and Dahlgren 1994) This study also revealed a
majority of Cr Fe Mg and Ni in extracted soil waters were in the form of colloids
suggesting that nanoparticles are an important factor of Cr translocation in soils
The pH of a soil is a dominant factor controlling the nature of many chemical and
microbial reactions including the geochemistry of Cr (Brady and Weil 1999) Ranges in
pH for serpentine soils worldwide are plotted in Figure 46 The highest soil pH was
measured in a South Africa serpentine soil with a value of 9 (Cole 1992b) whereas the
lowest pH was reported in Ireland with a value of 38 (Jeffrey 1992) This wide range of
pH is a function of the adsorbed H+ and Al3+ the base-forming cations the organic
matter and the mineralogy present within each soil The pH range of serpentine soils
shown in Figure 46 is plotted on the stability diagram in Figure 43 demonstrating that
both Cr(III) and Cr(VI) aqueous species are favorable The Eh range for the serpentine
soils in Figure 3 is based on the soils being in contact with the atmosphere as well as
being a transitional environment Although the activity of H+ varies by four orders of
magnitude the pHs of most serpentine soils are slightly acidic (pH ~6) which is near that
of average rainwater (pH~57 Figures 43 and 46) Additionally serpentine soil pH
values have been observed to increase with depth towards the protolith for a number of
soils (Rabenhorst et al 1982 Gough et al 1989 Gasser and Dahlgren 1994) The
increase in pH with respect to depth may be the result of rainwater being progressively
influenced by the mineral chemistry and hydrolysis reactions as it descends through the
soil towards the protolith
Groundwater from serpentinized ultramafic rocks have pH values ranging from 8
to 12 (Barnes et al 1972) and these waters have been divided into two categories as
shown in Figure 46 Mg bicarbonate (Mg2+-HCO3-) type and Ca hydroxide (Ca2+-OH-)
type (Barnes and ONeil 1969 Barnes et al 1972) The Mg bicarbonate waters (pH~8)
are characteristic of most surface waters and shallow groundwaters from serpentinites
and ultramafic rocks The Ca hydroxide waters (pHasymp10-11) are associated with
incompletely serpentinized peridotites and these fluids are comparatively low in Mg2+
Fe2+ Mn2+ Crtotal and SiO2total The Ca2+ in solution originates from the serpentinization
of Ca-rich orthopyroxene (Barnes et al 1972) The pH of serpentinite groundwaters are
109
plotted on the Cr stability in Figure 43 demonstrating that both Cr(III) and Cr(VI) are
potential valence states in these waters
Chromium predominates as Cr(III) in ultramafic rocks and serpentine soils
however Cr(VI) has been identified in serpentine soil solutions and serpentinite
groundwater in California and New Caledonia at concentrations of lt20 microg L-1 (Gough et
al 1989 Becquer et al 2003) Calcium hydroxide waters (pH~11) collected from the
same spring evaluated by Barnes et al (1972) in the Cedars ultramafic body (Cazadero
CA) contains 12-22 Cr(VI) microg L-1 and no Cr(III) (this study) Additionally water from
chromite-containing ultramafic rocks in the Siniktanneyak Mountain ultramafic complex
of Alaska with pH values ranging from 67 to 93 were measured with low total Cr
concentrations lt2 Cr microg L-1 (Taylor et al 1998) A mechanism for Cr(III) oxidation in
serpentine soils and related water suggested by (Gough et al 1989 Becquer et al 2003)
may be the result of high valence Mn oxides common in serpentine soils High valence
Mn-oxides such as birnessite (δ-MnO2) are naturally occurring oxidants capable of
oxidizing Cr(III) to Cr(VI) at pHlt9 (Eary and Rai 1987 Fendorf et al 1992 Fendorf et
al 1993) It is also important to note that Cr(VI) reduction is hypothesized to occur in
soils due to electron transfer involving organic matter and Fe(II)-bearing minerals such as
magnetite (Peterson 1996 Fendorf et al 2000) Whether toxic concentrations of Cr(VI)
can be present in serpentine soils and related water over geologic time is important to
consider especially when high valence Mn oxides Fe(II)-bearing minerals and organic
matter are capable of significantly influencing the oxidation state of Cr
Serpentine Soil Vegetation
Vegetation associated with serpentine soils has received considerable attention
due to the apparent infertility of these soils The inhabitability of serpentine soils for
most plants has been attributed to the imbalance of available Mg relative to Ca or low
CaMg ratios (Walker 1954) the deficiency of plant nutrients such as K N and P
(Turitzin 1991) elevated heavy metal concentrations such as Ni and Cr (Robinson et al
1935) and poor drainage (Brooks 1987) Linking plant productivity to toxic levels of
Cr in the soil has not been fully resolved (Brooks 1987 Kruckeberg 1992) however a
majority of Cr in the soils is not available for plant uptake due to being in high stable and
110
weathering resistant phases discussed in the previous sections (Table 42) In general Cr
concentrations in plants typically do not exceed 100 mg kg-1 (Brooks 1987)
Table 43 provides a global list of serpentine soil plant species and their measured
Cr concentrations with an emphasis on Cr hyper-accumulators ie plants able
incorporate higher than average concentrations of Cr A majority of these measurements
are based on the dry weight of Cr Aerobryopsis longissima from New Caledonia has the
highest reported Cr concentration with values ranging from 200 to 700 mg kg-1 (Lee et
al 1977) Other noteworthy plants include the Lophostachys villosa from Brazil (Brooks
et al 1992) Arenaria humifusa from Newfoundland (Roberts 1992) the Calluna
vulgaris from Unst (Shewry and Peterson 1976) with Cr concentrations of 203 530 and
417 mg kgndash1 respectively Even mosses such as the Funaria flavicans (Maryland)
incorporate Cr concentrations as high as 90 mg kg-1 (Shaw 1991) By examining
specific parts of the plant Gough et al (1989) determined that Cr concentrations are
highest in the basal burl (600 mg kg-1) of the manzanita bush compared to the roots (30
mg kg-1) stems (20 mg kg-1) and leaves (20 mg kg-1) The study by Gough et al (1989)
demonstrates that Cr is not homogeneously distributed throughout the plant and that
evaluating a leaf versus a stem could result in significantly different Cr concentrations
Soil Organic Matter
Soil organic matter (SOM) including the biomass (living organisms such as
bacteria) detritus (nonliving plant residues) and humus (amorphous and colloidal
mixtures of organic substances) affects the geochemistry and buffering capacity of
serpentine soils Furthermore biologic reduction of Cr(VI) to Cr(III) via cometabolic
and metabolic pathways coupled with carbon oxidation has been demonstrated for a
number of bacterial strains including aerobes (Gopalan and Veeramani 1994 Park et al
2000) and anaerobes (Turick et al 1996 Tebo and Obraztsova 1998) Additionally
aqueous Fe(II) produced by microbial Fe(III) reduction is a known biogenic pathway for
the reduction of Cr(VI) (Fendorf et al 2000) Fe(III)-Cr(III) (oxy)hydroxides are the
resulting secondary phases (Rai et al 1989 Patterson and Fendorf 1997 Hansel et al
2003) Bacteria oxidizing Cr(III) to Cr(VI) has not been observed
111
Most serpentine soils in temperate regions appear to be either Mollisols or
Alfisols which often contain high concentrations of SOM Alexander et al (1989) and
Bulmer et al (1992) reported that serpentine soils can contain organic carbon
concentrations in excess of 50 mg g-1 Humus has a high cation exchange capacity and
provides much of the pH buffering capacity in these soils (Brady and Weil 1999)
Organic acids such as fulvic and humic acids are known to complex cations such as
Cr3+ Fe3+ and Mn2+ forming stable organometallic complexes (Kaupenjohann and
Wilcke 1995 Brady and Weil 1999) Additionally metal leaching experiments
performed by Schreier et al (1987) established that organic acids produced by soil
organic matter such as oxalic acid and citric acid were extremely effective in releasing
Cr Mg Fe Ni and Co from serpentinite rocks and sediments
CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
Serpentine soils were obtained from California Oregon and New Caledonia and
analyzed to further evaluate the factors influencing the Cr geochemistry in serpentine
soils Analyses for these soils include pH electrical conductivity total organic matter
total N ammonium acetate extractable Ca K Mg and Na soil texture soil color
chemical composition of the soil and the chemical composition of Cr-bearing minerals
The general soil characteristics are described in the following section and the analytical
methods are explained in Appendix 41 Although a majority of these soils are located in
the western United States factors influencing the pedogenesis of these soils vary
considerably allowing a comparative analysis of how the fate of Cr might be influenced
by climate topography biota parent material and time
Sample Locations and General Soil Characteristics
Six serpentine soils were collected from locations shown in Figures 41 and 42 to
evaluate the chemistry of Cr in serpentine soils These soils include 1) a laterite (Oxisol)
collected in New Caledonia South Pacific 2) a laterite (Oxisol) exposed at Eight Dollar
Mountain Selma OR 3) a laterite (Oxisol) near Nickel Mountain Riddle OR 4) a
serpentine soil (Inceptisol) with a CO2-metasomatized serpentinite protolith from
Harvard Mine Jamestown CA 5) a serpentine soil (Mollisol) obtained from a preserved
112
grassland at Jasper Ridge Biological Preserve Stanford CA and 6) a serpentine soil
(Inceptisol) collected in the proximity of Pillikin Mine El Dorado CA Lastly an Oxisol
formed from basalt obtained in Oahu Hawaii is also included for comparison
These seven sites include a range of soil environments from different protoliths
under varying conditions of rainfall temperature and topography Each soil was collect
near the highest and topographically level point for each serpentine soil site Serpentine
grassland soils at Jasper Ridge formed under Mediterranean-like climates (xeric moisture
regime) whereas higher temperatures and less rainfall (ustic moisture regime)
characterize soil formation at the Harvard and Pillikin mines in the Sierra Nevada
Foothills metamorphic belt The vegetation of these three sites is dominantly grasses with
some manzanita These modern soil environments contrast with the Miocene Ni and Cr-
rich laterite soils of Nickel Mountain and Eight Dollar Mountain that were probably
formed under an aquic moisture regime near sea level in a subtropical climate Grasses
manzanita and pine trees are the dominant forms of vegetation at Nickel and Eight
Dollar Mountain An ultramafically-derived laterite formed under tropical conditions
from New Caledonia was collected The vegetation and topography related to the New
Caledonia laterite was not recorded during sampling An Oxisol from Hawaii is also
included in this study to evaluate the differences between a highly weathered soil
originating from an ultramafic rock and one from a mafic rock (basalt) under tropical
conditions The Hawaiian soil was collected from a topographically level area however
the vegetation related to this site was not noted
BedrockProtolith Mineralogy
A broad characterization of the protoliths for the serpentine soils is briefly
summarized due to the bedrock not being collected from all the field sites New
Caledonia is composed of variably serpentinized peridotites predominantly harzburgites
and some dunites (Leblanc 1995) Nickel Mountain and Eight Dollar Mountain are
predominantly serpentinized harzburgite and dunite The bedrock for soils collected at
Harvard Mine is mainly composed of quartz ankerite and mariposite (also referred to as
QAM) caused by the CO2 metasomatism of serpentinite associated with gold
mineralization (Bohlke 1989) The average serpentinite at Jasper Ridge is composed of
113
serpentine (mainly lizardite and antigorite) chlorite talc magnetite and chromite (Oze et
al 2003) however brucite has been reported as a major constituent in related Franciscan
Complex serpentinites (Hostetle et al 1966) Pillikin Mine serpentinites dominantly
consist of talc and serpentine (lizardite antigorite and chrysotile) and these serpentinites
contain podiform chromite deposits (Cater et al 1948) Mafic bedrock for the Hawaiian
Oxisol is tholeittic basalt
Soil Chemistry and Physical Properties
As shown in Figure 47 the colors of the collected soils range from the red New
Caledonia laterite to the greenish gley of the Pillikin Mine Inceptisol The Munsel color
values are also shown in Figure 47 The dark grey color of the Jasper Ridge soil and the
greenish gley of the Pillikin Mine soil are atypical of soils derived from serpentinites due
to the lack of any apparent redness The range of these colors is often attributed to the
presence of Fe-oxides however organic matter and Mn oxides are also dominant factors
controlling color (Brady and Weil 1999)
Chemical properties for the six serpentine soils and the Hawaiian soil are listed in
Table 44 The slightly acidic pH values range from 53 to 69 similar to those shown in
Figure 43 and 46 The New Caledonia laterite is the most acidic soil with a pH of 53
compared to the Harvard Mine carbonate-buffered soil at 69 The pH of the Hawaiian
soil is also slightly acidic with a pH of 604 The electrical conductivities (EC) of the
soils range from 01 to 07 dS m-1 a range indicative of relatively few dissolved inorganic
solutes The highest organic matter content is in the very dark grey Jasper Ridge soil
with a value of 32 and the lowest is in the red New Caledonia laterite (05) The
Hawaiian soil has an organic matter content of 1 Nitrogen concentrations for all the
soils are lt012 with Jasper Ridge having highest N concentration which is consistent
with the soil organic matter content
Ammonium acetate extractable Ca K Mg and Na are also listed in Table 44
Magnesium concentrations are gt6923 mg kg-1 in all of the serpentine soils The lowest
concentration of Mg was measured in the basalt-derived Hawaiian Oxisol with a value of
168 mg kg-1 Calcium concentrations in the serpentine soils range from 167 to 998 mg
kg-1 The 998 mg kg-1 of Ca in the Harvard Mine soil is directly related to the abundance
114
of ankerite (gt20 of the bedrock) Potassium concentrations are low for most of the soils
(lt55 mg kg-1) however Jasper Ridge and Harvard Mine soils have the highest
extractable K with values of 237 and 301 mg kg-1 respectively The extractable K value
for the Harvard Mine soil is due to the K-metasomatism related to gold mineralization
involving the serpentinite soil protolith The extractable concentration of Na is lt44 mg
kg-1 for all the serpentine soils whereas the Hawaiian soil has the highest Na (53 mg kg-
1) due to plagioclase and basaltic glass
Textures for the soils range from clay loam clay loam to sandy clay loam (Table
44) The laterite soil in New Caledonia and Oxisol in Hawaii are extremely different in
which the Hawaiian soil is mostly clay (66 clay) compared to the loam at New
Caledonia (25 clay) Additionally both ancient laterites (Eight Dollar Mountain and
Nickel Mountain) have clay textures Harvard Mine Jasper Ridge and Pillikin Mine
soils are all loamy in character
Major and trace element compositions for the soils are listed in Table 45
Chromium concentrations in the serpentine soils range between 827 and 9528 mg kg-1
The Eight Dollar Mountain laterite has the highest Cr concentration and the New
Caledonia laterite contains the lowest Cr concentration The Cr concentration for the
New Caledonia soil is lower than the values reported in Table 41 The Hawaiian soil has
a Cr concentration well below the analyzed serpentine soils with 388 Cr mg kg-1 but still
greater than the 200 mg kg-1 characteristic of serpentine soils Nickel concentrations in
the serpentine soils range from 1895 to 14100 mg kg-1 with the highest concentration
being identified in the Nickel Mountain laterite Manganese concentrations range
between 257 (New Caledonia) to 3017 (Eight Dollar Mountain) mg kg-1 in the serpentine
soils Element concentrations in the Hawaiian Oxisol contrast from the serpentine soils
with higher Al Cu P Pb and Zn concentrations
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in the soil solids for Jasper Ridge New Caledonia Pillikin Mine Eight Dollar
Mountain and Harvard Mine Although the Cr concentrations in these soils exceed 200
mg kg-1 (Table 45) X-ray absorption near-edge spectra (XANES) shown in Figure 48
illustrate Cr(VI) levels are below the detection limits of this method (~10 Cr mg kg-1) due
to the lack of a pre-edge feature The box in Figure 48 illustrates where a Cr(VI) pre-
115
edge peak should be present Chromium(VI) concentrations in these soils are possibly
present in the parts per billion range
Soil Mineralogy
Backscattered electron images and microprobe analyses were used to identify the
Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49
and corresponding Cr-bearing mineral compositions are listed in Table 46 The New
Caledonia soil exhibits a wide range of particle sizes and a Fe-(oxy)hydroxide was
identified containing 165 Cr2O3 Wt Chromium-spinel has been identified in the New
Caledonia laterite by Becquer et al (2003) Eight Dollar Mountain soils exhibit a coarse
texture and both Cr-magnetite (2387 Cr2O3 Wt ) and chromite (3343 Cr2O3 Wt )
are present Nickel Mountain soils are also coarse and chromite (1315 Cr Wt ) was
identified Soils at Harvard Mine contain numerous chromite grains and the mineral
trevorite (a Ni-bearing spinel) was identified with 3991 Cr2O3 Wt and 2054 NiO Wt
Chromite and magnetite are abundant in soils at Jasper Ridge as well as a fine-
grained mixture of chromite magnetite and an unidentifiable silicate with up to 1878
Cr2O3 Wt (Figure 49) Veins of Cr-chlorite crosscut the spinel-silicate mixture in
Figure 49 and this Cr-chlorite is the only Cr-silicate identified using this method
Chromite (5355 Cr2O3 Wt ) is the dominant Cr-bearing mineral detected in soils at the
Pillikin Mine and these grains contain the highest Cr concentrations compared to the
other minerals in the seven soils The Hawaiian Oxisol possesses ulvoumlspinels (007
Cr2O3 Wt and 2465 TiO2 Wt ) which are evenly distributed throughout the soil
Overall microprobe observations reveal that the Cr-bearing spinels are present in every
soil Additionally Cr-bearing spinels at Jasper Ridge have been shown to undergo
incongruent dissolution progressively enriching the spinel towards a FeCr2O4 end-
member (Oze et al 2003)
DISCUSSION
The physical and chemical characteristics of serpentine soils are directly related to
the alteration and tectonic history of the serpentinite as well as the soil-forming factors
including the parent material climate biota topography and time Identifying Cr-
116
bearing minerals their chemical compositions and abundance in the soils allows a more
definitive explanation for Cr enrichment in these soils as compared to total Cr
concentrations listed in Tables 41 and 45 and sequential extraction experiments reported
in Table 42 Detrital chromite grains from the serpentinite protolith are resistant to
weathering and accumulate in the soil over time Although serpentine soils contain Cr
concentrations typically gt200 mg kg-1 the presence of Cr-spinels and their resistance to
weathering demonstrate that a majority of the Cr is not chemically available to interact
within the soil The role of organic matter and vegetation are important factors related to
the chemistry of serpentine soils however their role for influencing a majority of the Cr
geochemistry is limited due to the nature of the Cr-spinels
Chromium and Serpentine Soils
Total Cr concentrations exceed 200 mg kg-1 for most of the serpentine soils
reported in Table 41 and in the six serpentine soils of this study (Table 45) These Cr
concentrations can be accounted for by assessing the Cr-bearing minerals associated with
these soils and their relative abundance Small grains of Cr-spinels or chromite identified
in the soils (Figure 49 and Table 46) have the potential to appreciably impact Cr
concentrations Table 47 demonstrates the mineral masses and volumes of Cr-bearing
minerals required to cause predetermined Cr concentrations in soil samples A minimal
amount of chromite (lt02 of a soil sample mass or 005 of the volume of serpentine)
can result in Cr concentrations of 1000 mg kg-1 A few grains of chromite could easily
be responsible for the Cr concentrations in Tables 41 and 45 In fact residual grains of
Cr-enriched spinels are observed in every soil analyzed (Figure 49 and Table 46)
Ultimately grains of spinel in the soil do appreciably impact Cr concentrations however
they do not provide a readily available source of Cr based on extraction experiments
performed by Oze et al (2003) Serpentine minerals could also provide minor Cr
concentrations to the soils however soil extractions have consistently shown that a
majority of the Cr is in highly resistant phases capable of resisting dissolution even in
concentrated hydrochloric acid nitric acid and hydroxylamine with acetic acid (Table
42) Chromium end-member minerals including eskolaite Cr-muscovite (fuchsite) Cr-
garnet (uvarovite) Cr-epidote (tawmawite) and Cr-chlinochlore (kaumlmmererite) are also
117
included in Table 47 to demonstrate that relatively low abundances of these minerals are
all that is necessary to increase Cr concentrations in rocks and soils Chromium-
clinochlore is often observed developing adjacent Cr-spinels in serpentinites and is
perhaps a more accessible mineral source of Cr in the serpentinites compared to Cr-
spinels (Phillips et al 1980 Schreyer et al 1981 Kerrich et al 1987 Treloar 1987a
Pan and Fleet 1991 Schandl and Wicks 1993 Christofides et al 1994 Challis et al
1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Oze et al 2003)
As shown in Table 44 and Figures 43 and 46 the chemical properties (pH EC
N etc) of serpentine soils are not correlated to Cr concentrations measured in the soil
solids (Tables 41 Figure 45) due to Cr-spinels accounting for a majority of the Cr
(Figure 49 and 410 Tables 46 and 47) However extraction experiments and soil
solution analyses demonstrate that a limited fraction of the total Cr (lt1) is extractable
If only 1 of the Cr in a serpentine soil is extractable a serpentine soil containing 10000
Cr mg kg-1 could possibly contribute up to100 Cr mg kg-1 still a significant Cr
concentration The chemical properties of the soil such as pH will influence this limited
fraction of the Cr in the soils Acidic pHs may aid in mobilizing Cr(III) however this
will only occur at low pHs (lt4) which exceed the range of most serpentine soils (Figures
43 and 46) pH values greater than 8 like those in some serpentine soils and
serpentinite groundwater may help sustain Cr(VI) concentrations in solution (Figure
43) In addition to pH organic matter in the soils may be a significant factor related to
the mobility of Cr due to the production of organic acids promoting the dissolution of the
Cr-bearing phases but particulate organic matter is known to bind Cr(III) in stable stable
organometallic complexes (Kaupenjohann and Wilcke 1995 Brady and Weil 1999)
Evaluating the valence state of Cr is crucial for understanding the chemistry as
well as the potential toxicity related to serpentine soils The soil solids do not contain
detectable concentrations of Cr(VI) (Figure 48) however Cr(VI) concentrations are
present in soil solutions and groundwater at values lt30 microg L-1 (Gough et al 1989
Becquer et al 2003) A majority of serpentinite Cr-bearing minerals only contain
Cr(III) therefore there must be a reaction or process by which Cr(III) is oxidized in a
water saturated soil High valence Mn oxides such as birnessite are present within
serpentine soils and provide a feasible explanation for Cr(III) oxidation Additionally
118
Mn oxides or other strong oxidants in the soil might accelerate the weathering of Cr-
bearing minerals including chromite and generate the Cr(VI) concentrations observed in
soil solutions and groundwater After Cr(III) is oxidized Cr(VI) is a chemically mobile
species and would be transported away from the serpentine soils before it can be
observed in the solid phase Additionally Cr(VI) may be subsequently reduced by
reacting with organic matter or Fe(II)-bearing minerals such as magnetite which are
present in a majority of serpentine soils Perhaps the best way to evaluate the potential of
serpentine soils being a non-anthropogenic source of Cr contamination is to monitor
Cr(III) and Cr(VI) in the soil solutions over time as demonstrated by Gasser and
Dahlgren (1994)
It is curious that plants do uptake Cr in serpentine soils especially when Cr is
tightly bound in the soil matrix (Table 42) In order for plants to uptake Cr it must be in
a form to be absorbed by plant roots Soil pH in the rhizosphere can be altered by plant
roots by as much as 2 units within two days (Fischer et al 1989) This acidification is
known to increase the solubility of the minerals in the soils and could serve as a
mechanism by which plants can incorporate this relatively insoluable element Another
possibility is that Cr(III) in the soil is oxidized by either organic or inorganic means and
then incorporated into the plant in its more chemically mobile and bioavailable form
Cr(VI) How and why plants incorporate as much Cr as they do needs to be examined
more thoroughly
Chromium Spinel in Soils and Rocks
Compositions of chromite identified in serpentinites worldwide (Figure 44) and
Cr-bearing spinels identified in the six serpentine soils are compared in Figure 10 in order
to determine whether the soil spinels are geochemically similar to those in serpentinites
This comparison is ascertained by examining how Al Fe and Mg Wt values varied
with respect to Cr Wt Trends for magnetitehematite chromite and the
microcrystalline spinel-silicate mixture observed in the soils are noted in each diagram in
Figure 410 Additionally the highest Cr concentrations for soil chromite
magnetitehematite and the microcrystalline spinel-silicate mixture are ~40 Cr Wt
~25 Cr Wt and ~15 Cr Wt respectively Based on Al Wt soil
119
magnetitehematite deviates significantly from the chromite trend with Al Wt values
lt5 Magnetitehematites are enriched with Fe relative to serpentinite and soil chromite
and Mg concentrations in the soil chromite are slightly higher than serpentinite chromite
Soil magnetitehematite and chromite are similar by demonstrating a linear trend when
Al Fe and Mg are summed The microcrystalline spinel-silicate mixture (CSM) does
not follow this trend because of Si present in this mineral mixture Overall soil
chromites are similar to serpentinite chromite and soil magnetiteshematites can be
differentiated from the chromites when Cr Wt is plotted against Al Fe and Mg Wt
CONCLUDING STATEMENT
Serpentine soils are abundant among convergent margins and the chemistry of
these soils is ultimately a function of the ultramaficserpentinite rock mineralogy and
texture as well as all the common soil-forming factors Although ultramafic rocks and
serpentinites produce groundwater that is highly basic most serpentine soil pHs are
slightly acidic Although Proctor and Nagy (1992) argued that it is futile to categorize
serpentine soils based on chemistry and biology (otherwise referred to as the lsquoserpentine
factorrsquo) as proposed by Brooks (1987) Cr concentrations greater than 200 mg kg-1 in
serpentine soils are consistently present Chromium in the soils originates from minerals
in the bedrock where the valence is primarily Cr(III) the exception being Cr(VI) in the
rarely present mineral crocoite (PbCrO4) Increased Cr-concentrations in serpentine soils
are largely the result of chromite or other resistant Cr-bearing minerals causing a
localized enrichment in the soils Although Cr in serpentine soils is not readily released
from the soil matrix serpentine soil vegetation can incorporate Cr concentrations as high
as 700 mg kg-1 (Table 34) Finally the limited extractability of Cr and site dependent
factors such as the presence of organic matter clays Fe(II)-bearing minerals pH and the
poor drainage indicates that the possible formation and sustained presence of Cr(VI) is
not a portentous matter related to serpentine soils
120
ACKNOWLEDGEMENTS
We appreciate technical and field support from Robert Jones Anders Meibom
Rob Dunbar and Guangchao Li Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund GSA Graduate Student Research Grants and the
Office of Teaching and Learning at Stanford University is gratefully acknowledged
121
Table 41 Measured Cr concentrations in serpentine soils worldwide
Location Cr concentration (mg kg-1) Reference
Canyon Creek area OR 920 (White 1971) 66-13200 (Brooks et al 1992)
Ireland 29-5000 (Proctor and Woodell 1971 Jeffrey 1992) Italy 1300-3900 (Gambi 1992) Japan 5200 (Shewry and Peterson 1976) Maryland 1500-6000 (Rabenhorst et al 1982) Mt Tamalpais CA 1700 (White 1971) New Caledonia 6300-80000 (Jaffre 1980) New Caledonia 2-6 (Schwertmann and Latham 1986) New Idria CA 1640 (White 1971) New Zealand 850-3250 (Lee 1992) Newfoundland Canada 309-1447 (Roberts 1992) Nickel Mountain OR 1730 (White 1971) Santa Barbara County CA 2080 (Woodell et al 1975) Southern Africa 045-14 (Morrey et al 1992) Tehama County 023-12 (Gough et al 1989) Transvaal Lowveld South Africa 120-800 (Cole 1992) Unst Shetland 3000-4000 (Proctor 1969) Wenatchee Mnts WA 895 (White 1971) Zimbabwe 110-46000 (Soane and Saunder 1959 Proctor et al 1980)
Goias State Brazil
122
123
Table 42 Summary of Cr Fe and Ni extractions from serpentine soils
Range of Extractable Cr Fe and Ni Location Cr Fe Ni
Method Reference
Italy 2-5 microg g-1 305-640 2300-2700microg g-1 25 CH3COOH (Gambi 1992)
NE Portugal 02-09 microg g-1 120-540 microg g-1 40-290microg g-1 EDTA (Menezes de Sequeira and Pinto da Silva 1992) New Caledonia 22-23 384-581 02-216 DCB (dithionite-citrate bicarbonate) (Schwertmann and Latham 1986) New Caledonia 02-11 microg g-1 02-66microg g-1 Method Unknown (Jaffre 1980)
Norway 357microg g-1 316microg g-1 HClO4HNO3 (concentrated) (Zeien and Bruemmer 1989) Norway 254microg g-1 134microg g-1 Ascorbic and Oxalate Acid pH=325 (Zeien and Bruemmer 1989)
Norway 62 microg g-1 53microg g-1 NH4-EDTA (Zeien and Bruemmer 1989) Norway 22-60 microg g-1 8-25 mg kg-1 1-7 microg g-1 pH-stat reactor EDTA (Kaupenjohann and Wilcke 1995)
Swiss Alps 08 micromol L-1 28micromol L-1 228microg g-1 Lysimeter (Nonfiltered) (Gasser and Dahlgren 1994) Trinity Ophiolite CA 004-045 gkg 27-210 g kg-1 005-245microg g-1 Citrate-Dithionite Extracts (Alexander et al 1989)
Unst lt01 microg g-1 19-77microg g-1 Ammonium Acetate (Shewry and Peterson 1976)
Unst 500-1390microg g-1 1540-3330microg g-1 HNO3-Soluble (Shewry and Peterson 1976) JRBP CA 0 microg g-1 01microg g-1 103microg g-1 BaCl2 Method (Oze et al 2003)
JRBP CA 78 microg g-1 513 microg g-1 150 microg g-1 Hydroxlyamine+Acetic Acid (Oze et al 2003) JRBP CA 31 microg g-1 2039microg g-1 35microg g-1 10M Hydrofluoric Acid (Oze et al 2003)
123
124
Table 43 Serpentine soil plant Cr concentrations
Plant Cr Concentration (microg g-1) Location Reference
Lophostachys villosa 203 Goias State Brazil (Brooks et al 1992) Ruellia geminiflora
56 Goias State Brazil (Brooks et al 1992) Minuartia ophiolitica 83 Italy (Gambi 1992) Sanguisorba minor 194 Italy (Gambi 1992) Plantago lanceolata 135 Italy (Gambi 1992) Cerastium latifolium 110 Italy (Gambi 1992) Funaria flavicans 34-899 Maryland (Shaw 1991) Avena sativa 168-583 New Zealand (Lee 1992) Aerobryopsis longissima 200-700 New Caledonia (Lee et al 1977) Arenaria humifusa 530 Newfoundland Canada (Roberts 1992) Senecio pauperculus 406 Newfoundland Canada (Roberts 1992) Rhalomitrium lanuginosum 246 Newfoundland Canada (Roberts 1992) Santolia semidentata 95 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Asplenium cuneifolium 228 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992) Armeria eriophylla 38 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Arctostaphylos viscida 029-705 Tehama County CA (Gough et al 1989) Calluna vulgaris 417 Unst (Shewry and Peterson 1976) Armeria maritima 149 Unst (Shewry and Peterson 1976) Agrostis stolonifera 120 Unst (Proctor and Woodell 1971) Dicoma niccolifera 77 Zimbabwe (Proctor 1992) Xerophyta equisetoides 68 Zimbabwe (Proctor 1992)
124
125
Table 44 Chemical and textural properties of the collected soils
Ammonium Acetate Extractable ------------------------Hydrometer--------------------- Ca K Mg Na
Sand Silt ClaySample pH EC dSm
Organic Matter ------
LECO Nitrogen ------ -------------------mg kg-1-------------------
------ ------ ------
Texture
New Caledonia 534 01 05 lt001 279 29 692 44 43 32 25 LoamHawaii 604
03 10 002 839 46 168 53 8 26 66 ClayEight Dollar Mnt 651 07 16 006 599 62 3464 25 28 28 44 Clay
Nickel Mnt 665 05 07 001 167 42 3289 17 24 32 44 ClayHarvard Mine 691 03 11 003 998 301 2897 43 41 30 29 LoamJasper Ridge 671 04 32 012 505 237 3291 36 36 31 33 Clay LoamPillikin Mine 642 04 05 002 185 55 3421 29 45 23 32 Sandy Clay Loam
125
126
Table 45 Total major and trace element concentrationsdagger of the collected soils
Element Concentration New Caledonia Hawaii Eight Dollar
Mnt Nickel Mnt Harvard Mine Jasper Ridge Pillikin
Mine Ag mg kg-1 lt05 lt05 lt05 lt05 02 lt05 lt05Al 931 1325 314 095 179 314 179Ba mg kg-1 lt10 180 40 20 30 230 lt10Be mg kg-1 05 16 05 lt05 lt05 08 lt05Ca 003 012 085 018 007 079 002Cd mg kg-1 lt05 lt05 18 lt05 05 lt05 lt05Co mg kg-1 56 183 338 403 85 155 122Cr mg kg-1 827 388 9528 5761 2900 5976 1977Cu mg kg-1 78 248 43 25 51 33 4Fe 638 1296 1635 17 538 727 468K 002 046 004 003 004 039 001
Mg 072 026 753 829 792 1025 gt150Mn mg kg-1 257 2706 3017 2554 920 1664 903Na 001 008 005 001 003 057 lt001Ni mg kg-1 2780 345 5510 14100 1895 2180 2760P mg kg-1 50 260 160 150 130 140 20Pb mg kg-1 6 31 4 3 2 8 6Ti 016 041 008 002 lt001 02 002V mg kg-1 63 132 158 54 65 92 27W mg kg-1 10 10 10 10 lt10 lt10 lt10Zn mg kg-1 43 218 160 132 66 111 47
126
dagger ICP-AES analyses
127
Table 46 Compositionsdagger for soil Cr-containing minerals identified in Figure 49
Wt New Caledonia A
Hawaii B
Eight Dollar Mnt C
Eight Dollar Mnt D
Nickel Mnt E
Harvard Mine F
Jasper Ridge G
Jasper Ridge H
Pillikin Mine I
SiO2 491 008 175 026 002 000 1940 001 000Al2O3
756 185 272 2235 2041 088 559 1575 839NiO 023 005 001 006 011 2054 014 002 007FeO 7766a 6212 6633b 3420b 1784 2445 2629 2314 3105MnO 004 037 181 123 133 031 155 146 216Cr2O3 165 007 2387 3343 4682 3991 1878 5024 5355MgO 048 121 202 653 1315 1062 1668 926 497CaO 001 000 013 008 000 000 015 000 000Na2O 000 000 002 001 000 000 006 000 000TiO2 004 2465 012 006 012 000 000 003 018
TOTAL 9257 9041 9879 9821 9980 9671 8864 9920 10037Mineral Cr-Hydroxide Ulvoumlspinel Cr-Magnetite Chromite Mg-Chromite Trevorite CSM1 Chromite Chromite
127
dagger Electron microprobe analyses This is the form of iron unless otherwise stated a FeO is in the form Fe2O3 b FeO is in the form of Fe3O4 1CSM Microcrystalline chromitesilicate mixture
128
Table 47 Chromium-bearing mineraldagger masses and volumes required to influence Cr concentrations in soils
Cr Concentration Mineral mass (grams) required for 1000 g bulk soil analysis Mineral Mass Percent () (mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite
500 11 07 22 24 48 57 011 007 022 024 048 057 1000 22 15 43 48 95 115 022 015 043 048 095 1152000 43 29 86 96 191 230 043 029 086 096 191 2305000 108 73 216 241 477 574 108 073 216 241 477 574
Cr Concentration Mineral Volume Percent () in Serpentine Mineral Volume Percent () in Quartz
(mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite500 005 004 02 02 03 15 006 004 02 02 04 06
1000 01 007 04 03 07 30 01 007 04 03 07 112000 02 014 07 06 14 59 02 015 07 06 14 225000 06 036 18 16 34 147 06 037 18 16 35 55
128
daggerMineral formulas eskolaite Cr2O3 chromite FeCr2O4 fuchsite KCr2(AlSi3O10)(OH)2 uvarovite Ca3Cr2Si3O12 tawmawite Ca2CrAl2Si3O12(OH)2 kaumlmmererite Mg5CrAlSi3O11(OH)8
Note Molar volumes used in the calculations above are from their Al-enriched counterparts in Robie and Hemingway (1995) The molar volume for kaumlmmererite was estimated using unit cell volumes from XRD unit cell measurements from Robie and Hemingway (1995) Christofides et al (1994) and Phillips et al (1980)
129
Figure Captions
Figure 41 Selected serpentine soil localities as well as soils collected for analyses are
shown on a world map Active plate boundaries are shown Serpentine soil localities not
adjacent the plate boundaries on this map are related to ancient plate boundaries that are
no longer active
Figure 42 Field pictures of soils collected from a) Eight Dollar Mountain b) Nickel
Mountain c) Harvard Mine d) Pillikin Mine and f) Jasper Ridge Biological Preserve
Figure 43 Stability field diagram of Cr under a range of Eh and pH conditions (log a
Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom (1998) Adsorption
isotherms for Cr3+ and CrO42- on hydrous ferric oxide (HFO) are also shown (Leckie et
al 1980 Leckie et al 1984)
Figure 44 Microprobe analyses demonstrating Cr versus Al Wt for Cr-spinel
(chromite) and Cr-silicates including Cr-mica (fuchsitemariposite) Cr-garnet
(uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite) (Whitmore et al
1946 Leo et al 1965 Chen and Lee 1974 Neiva 1978 Cooper 1980 Phillips et al
1980 Grapes 1981 Schreyer et al 1981 Max et al 1983 Nutman et al 1983 Raase
et al 1983 Chao et al 1986 Von Knorring et al 1986 Kerrich et al 1987 Treloar
1987b a Morand 1990 Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen
1999 Proenza et al 1999) Chromium-spinel and Cr-silicate compositions based on
octahedral Cr and octahedral Al are also shown with tie-lines (dotted) denoting ideal
isomorphic substitution between Cr and Al end-members
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above Each dotted line connecting the
points represents a separate soil profile at each site
129
Figure 46 Ranges of pH for serpentine soils and related groundwater worldwide Soil
pH values were obtained from A(Kruckeberg 1992) B (Graham et al 1990) C
(Alexander et al 1989) D (Gough et al 1989) E (Rabenhorst et al 1982) F
(Roberts 1992) G (Bulmer et al 1992) H (Proctor 1992a) I (Jeffrey 1992) J
(Gambi 1992) K (Verger 1992) L (Proctor and Cole 1992) M (Cole 1992b) N
(Morrey et al 1992) O (Proctor 1992b) P (Mizuno and Nosaka 1992) Q (Cole
1992a) and R (Lee 1992) Groundwater pH values from serpentinite are from Barnes
and ONeil (1969) and Barnes et al (1972) Note the range of pH environments possible
for ultramafic rocks and their related soils
Figure 47 Soil pictures with Munsel color values and names from soils collected from
a) New Caledonia b) Hawaii c) Eight Dollar Mountain d) Nickel Mountain e) Harvard
Mine f) Jasper Ridge Biological Preserve and g) Pillikin Mine
Figure 48 Cr XANES spectra for soils analyzed from Jasper Ridge New Caledonia
Pillikin Mine Eight Dollar Mountain and Harvard Mine The pre-edge area indicative of
Cr(VI) peaks is noted in the diagram No definitive Cr(VI) peaks are present in any of
these spectra
Figure 49 Backscattering images of mineral phases within each soil are presented
Electron microprobe analyses are noted by a circle and letter for the Cr-bearing minerals
and the results are listed in Table 46
Figure 410 Chromite magnetite and hematite compositions from the collected soils
(gray circles) as well as analyses for serpentinite chromite (white squares) from Figure
44 are shown in these diagrams with Cr Wt versus Al Fe Mg and Al+Fe+Mg Wt
130
Figure 41
131
Figure 42
132
Figure 43
133
Cr Wt
0 10 20 30 40 50
Al W
t
0
5
10
15
20
25Cr-SpinelCr-MuscoviteCr-EpidoteCr-GarnetCr-Chlorite
Cr-SpinelBased on 4 Oxygens
Cr0 1 2
AlVI
0
1
2
ChromiteFeCr2O4
HercyniteFeAl2O4
Cr-MuscoviteBased on 11 Oxygens
Cr0 1 2
AlVI
0
1
2
MuscoviteKAl2(AlSi3O10)(OH)2
FuchsiteKCr2(AlSi3O10)(OH)2
Cr-GarnetBased on 12 Oxygens
Cr0 1 2
AlVI
0
1
2
GrossularCa3Al2Si3O12
UvaroviteCa3Cr2Si3O12
Cr-EpidoteBased on 125 Oxygens
Cr0 1
AlVI
2
3
ClinozoisiteCa2Al3Si3O12(OH)2
TawmawiteCa2CrAl2Si3O12(OH)2
Cr-ChloriteBased on 14 Oxygens
Cr0 1
AlVI
0
1
2
Kammererite(Mg5Cr)(AlSi3)O10(OH)8
Clinochlore(Mg5Al)(AlSi3)O10(OH)8
Figure 44
134
Maryland (USA)(Rabenhorst et al 1982)
Chromium (mg kg-1)
0 1000 2000 3000 4000 5000 6000 7000
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Tehama County CA (USA)(Gough et al 1989)
Chromium (mg kg-1)
0 2000 4000 6000 8000 10000 12000 14000
Dep
th (m
eter
s)
00
02
04
06
08
Kalamath Mountains CA (USA)(Hotz 1964)
Chromium (mg kg-1)
2000 4000 6000 8000 10000 12000 14000 16000
Dep
th (m
eter
s)
00
05
10
15
20
25
30
35
40
45
50
Newfoundland (Canada)(Roberts 1992)
Chromium (mg kg-1)
800 1000 1200 1400 1600 1800
Dep
th (m
eter
s)
00
02
04
06
08
Harvard Mine CA (USA)(Unpublished Data)
Chromium (mg kg-1)
0 500 1000 1500 2000 2500
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Jasper Ridge CA (USA)(Oze et al 2003)
Chromium (mg kg-1)
1500 2000 2500 3000 3500 4000 4500 5000
Dep
th (m
eter
s)
00
01
Figure 45
135
LocationsZ A B C D E F G H I J K L M N O P Q R S
pH
0
1
2
3
4
5
6
7
8
9
10
11
12
Nor
th A
mer
ica
Can
ada
Brit
ain
Italy
Far E
ast
Japa
n
Zim
babw
e
Sout
h A
fric
a
Wes
tern
Aus
tral
ia
New
Zea
land
Mg-Bicarbonate Waters pH=8
Ca-Hydroxide Waters pH=11-12
Average RainpH=57
Sout
h A
fric
a
Italy
Irela
nd
Cal
iforn
ia
Cal
iforn
ia
Brit
ish
Col
umbi
a
Mar
ylan
dCal
iforn
ia
Figure 46
136
Figure 47
137
-2
-1
0
1
2
3
4
5
5980 5990 6000 6010
Energy (eV)
Nor
mal
ized
Abs
orba
nce
New Caledonia
Pillikin Mine
Eight Dollar Mnt
Jasper Ridge
Harvard Mine
Cr(
VI)
Figure 48
138
Figure 49
139
Cr Wt 0 10 20 30 40 50
Al W
t
0
10
20
30 Soil Serpentinite
Cr Wt 0 10 20 30 40 50
Fe W
t
0
20
40
60
80
Cr Wt 0 10 20 30 40 50
Mg
Wt
0
2
4
6
8
10
Cr Wt 0 10 20 30 40 50
Al +
Fe
+ M
g W
t
0
20
40
60
80
MagnetiteHematite
Chromite
MagnetiteHematite
Chromite
Chromite
MagnetiteHematite
SpinelSilicate
SpinelSilicate
SpinelSilicate
SpinelSilicate
Figure 410
140
- Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip1
-
- Chapter 3 Oxidative Promoted Dissolution of Chromite by
- Manganese Dioxide and Concurrent Production of Ch
-
- Chapter 4 Chromium Geochemistry of Serpentine S
-
- List of References
-
- Chapter 2
- Chapter 3
-
- Chapter 4
-
- Appendix A
- Chapter 2
-
- Chapter 3
-
- Chapter 4
-
- Appendix A
-
- CHAP1pdf
-
- Toxicity and Regulations of Chromium
- Mineralogy of Chromium in Serpentinites
-
- Chapter Summaries
-
- CHAP2Tablepdf
-
- JR305CLAY Depth 0-5cm
- JR3515CLAY Depth 5-15cm
- JR33045CLAY Depth 30-45cm
-
- HCAP2Figspdf
-
- Figure 27
-
- CHAP3Tablepdf
-
- Figure 36
- Figure 37
-
- CHAP4pdf
-
- ABSTRACT
- REVIEW
-
- Chromium Geochemistry
-
- Soil Organic Matter
-
- CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
-
- BedrockProtolith Mineralogy
- Soil Chemistry and Physical Properties
- Soil Mineralogy
- Backscattered electron images and microprobe analyses were used to identify the Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49 and corresponding Cr-bearing mineral compositions are listed in Table 46 The New C
-
- DISCUSSION
-
- Chromium Spinel in Soils and Rocks
-
- CONCLUDING STATEMENT
- ACKNOWLEDGEMENTS
-
- CHAP4Tablepdf
-
- Ulvoumlspinel
-
copy Copyright by Christopher John-Paul Oze 2003
All Rights Reserved
ii
I certify that I have read this dissertation and that in my opinion it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy
____________________________ Dennis K Bird Principal Adviser
I certify that I have read this dissertation and that in my opinion it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy
____________________________ Scott Fendorf Principal Adviser
I certify that I have read this dissertation and that in my opinion it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy
____________________________ Robert Coleman
Approved for the University Committee on Graduate Studies
iii
ABSTRACT
Chromium persists in the environment as Cr(III) a non-hazardous species and a
micronutrient in human nutrition or as Cr(VI) a strong oxidizing agent a toxin to living
cells and a Class A human carcinogen by inhalation Serpentinites or metamorphosed
ultramafic rocks are distributed worldwide and contain Cr concentrations typically
greater than 200 mg kg-1 Additionally serpentine soils a generic term used to describe
any soil derived from serpentinite regardless of its physical or chemical properties
contain elevated Cr concentrations often exceeding their corresponding protolith
Chromium in serpentinites and serpentine soils is dominantly in the form of Cr(III)
however geochemical interactions during serpentinite weathering and soil formation
provide oxidative pathways allowing the formation of Cr(VI) The relationship between
rock and soil and the processes governing the Cr chemistry are difficult to assess due to
the inherent complexity of evaluating and correlating the mineralogy and geochemistry of
both systems Ultimately a multidisciplinary approach combining field analytical and
laboratory studies is necessary to evaluate the Cr geochemistry related to these rocks and
soils and to determine whether or not serpentinites and serpentine soils are sources of
non-anthropogenic Cr contamination
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of
California produces serpentine soils containing high concentrations of Cr as well as other
potentially toxic elements including Ni Co and Mn Chromium concentrations in
serpentine soils from Jasper Ridge Biological Preserve in the Central Coast Range are as
high as 4760 mg kg-1 nearly three times greater than the serpentinite protolith (~1800
mg kg-1) Chromium-containing minerals within the bedrock include chlorite (~03 Cr
Wt ) enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt )
magnetite (82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate
phase containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
iv
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
Despite the resistance of chromite to weathering Cr release and oxidation from
chromite is a potential environmental hazard in sediments and soils and a pathway for
soil development (weathering of primary minerals) related to ultramafic rocks and their
metamorphic derivatives (serpentinites) Birnessite is a common pedologic mineral in
these sediments and soils capable of oxidizing aqueous Cr(III) The interaction between
chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4) and birnessite was investigated with
an interest in the potential generation of Cr(VI) The rate of Cr(VI) in solution increases
with increasing chromite suspension densities and decreasing pH but is independent of
birnessite suspension densities at values greater than ~20 m2 L-1 at 25degC The overall rate
expression of Cr released and oxidized in solution from chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression--rates ranging from 9times10-4 to 44times10-3 microM Cr(VI) h-1 These
experiments therefore demonstrate that serpentine soils are a potential source of non-
anthropogenic Cr(VI)
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
v
chemistry of Cr in serpentine soils including its protolith is reviewed with a focus on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils have widely varying pH contain a variety Fe(III) oxides (magnetite and
hematite) phyllosilicates (serpentine and chlorite) and clays (smectites and
vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1) and
Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Additionally Cr(III) is the only valence state
observed in the serpentine soil solids however Cr(VI) has been identified in New
Caledonia and California serpentine soil solutions at concentrations below 30 microM The
enrichment and range of Cr concentrations in serpentine soils are directly related to the
presence of Cr-spinels specifically chromite and Cr-magnetite These spinels are
resistant to weathering and are preserved in the soil environment however oxidation of
Cr(III) from Cr-spinels by high valent Mn oxides or other strong oxidants is a potential
source of Cr(VI) identified in serpentine soil solutions Due to the chemically and
physically resistant nature of the Cr-spinels Cr-bearing silicates including clay minerals
Cr-chlorite Cr-garnet Cr-mica and Cr-epidote are more viable sources of Cr identified
in vegetation soil extractions soil solutions and related waters
vi
PREFACE
The chemistry and mineralogy of Cr related to serpentinites and serpentine soils
are assessed in this dissertation using a multidisciplinary field analytical experimental
and theoretical approach The four chapters that constitute this dissertation describe my
investigations related to these rocks and soils Additionally the common theme shared
by these chapters is that they all examine the Cr mineralogy in serpentinites and
serpentine soils and the potential of Cr to be mobilized and oxidized in near-surface
environments Chromite a Cr-bearing spinel is the dominant source of Cr in
serpentinite and serpentine soils Although chromite is resistant to weathering the
presence of birnessite a high valent Mn oxide will increase the release of Cr and the
production of Cr(VI) from chromite over a range of pH Chromium-clinochlore or
kaumlmmererite contains Cr concentrations less than 3 Wt however the rate at which Cr-
clinochlore weathers compared to chromite makes this mineral a more probable source of
Cr released into soil solutions These types of concerns are addressed in this thesis Brief
descriptions of the four individual dissertation chapters including their publication status
and a list of the co-authors are presented in the following paragraphs
Chapter 1 presents the main objectives of this dissertation The chemistry
regulations and toxicity of Cr are discussed as well as the general characteristics of
serpentinites and serpentine soils Additionally a brief summary of each chapter is
provided
Chapter 2 focuses on serpentinites and serpentine soils at Jasper Ridge Biological
Preserve Stanford CA The results have been submitted for publication in an article
entitled ldquoChromium geochemistry of serpentinized ultramafic rocks and serpentine soils
in the Franciscan Complex of Californiardquo to the American Journal of Science (March
2003) Scott Fendorf Dennis Bird and Robert Coleman are the second third and fourth
co-authors respectively
Chapter 3 is an experimental study evaluating the release of Cr and the
production of Cr(VI) from chromite in the presence of birnessite over a range of pH The
results for this study will be submitted for publication in an article entitled ldquoOxidative
promoted dissolution of chromite by manganese dioxide and concurrent production of
vii
chromaterdquo to Environmental Science and Technology The co-authors for this paper are
Scott Fendorf and Dennis Bird
Chapter 4 is a review paper that critically evaluates the chemistry of serpentine
soils and serpentinites primarily focusing on Cr geochemistry Additionally analytical
studies of serpentine soils collected from California Oregon and New Caledonia are
presented This study will be presented at the Robert Coleman Symposium and has been
submitted as an article entitled ldquoChromium geochemistry of serpentine soilsrdquo for an
International Geology Review Special Publication Scott Fendorf Dennis Bird and
Robert Coleman are the second third and fourth co-authors respectively
Appendix A is a preliminary investigation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite) a mineral common in serpentinites and present in serpentine
soils This appendix is entitled ldquoEstimation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite)rdquo The results of this chapter will be included in a study that
estimates the thermodynamic properties of Cr-bearing silicates including fuchsite (Cr-
muscovite) tawmawite (Cr-epidote) and uvarovite (Cr-garnet)
viii
ACKNOWLEDGEMENTS
Serpentinites serpentine soils and chromium are the Geologic Trinity of this
dissertationhellipeach topic is intimately related to one another The scope and breadth of
this research required a significant amount of time and funding None of this research
would have been possible without the generous support of several foundations
organizations and Stanford University I am grateful for the financial aid I have
received from the Stanford Fellowship the Stanford University McGee and Shell funds
the National Science Foundation the Geologic Society of America Graduate Research
Fund the Stanford Office of Teaching and Learning grant and the Jasper Ridge
Biological Preserve Melon Grant
I would not be where am I today if it were not for my parents Robert and Sue
Oze and my brother Rob Oze Thank you for the sandbox and all the hose-driven
volcanoeshellipthose Wookies never stood a chance While some families vacation in theme
parks my family frequented gem and mineral conventions collected rocks and shells
from mist-shrouded shores and sought lucky stones including an infamous quartz orb
Sometimes I think I was destined to be a geologist I would also like to thank everyone
in my family and all our family friends who have provided that extra little push and
support along the way These people include Tudor Bennatts Louis Oze Margaret and
John Chambers John Bennatts Nancy Skiles the Izumis the memory of Witchie Poo
hijacking Harleys from Bremerton bikers and the Man in the Sound waiting for that one
last catch Close friends I would like to acknowledge who are not related to the Stanford
crowd include Dave and Leigh Anderson and Rebecca Parker These satellite friends
helped me keep in touch with the world outside of academia and proved to me that I
wasnrsquot meant to have a real job A special thank you goes to Katherine Akako
Izumihellipyou have made this whole trip worthwhile
My PhD advisors Scott Fendorf and Dennis Bird provided an invaluable amount
of help patience and energy over the past five years Scott always made the time to
discuss and refine many of my projects making them relevant to many current
geochemical issues Additionally he helped me hone my soil science skills in the field
and in the lab Out in the field Scott can dig a pretty mean hole Dennis devoted a
significant amount of time and sweat in order to make this dissertation happen His
ix
insightful knowledge of geochemistry and of the geology of the Sierra Nevadas provided
the cornerstone for this dissertation
Robert Coleman played a key role in the production of this dissertation He
always knew what to say at the right time and helped me understand the complexity of
ophiolites and serpentinites through experienced eyes Additionally Keith Loague and
Gary Ernst deserve recognition for their outstanding support as committee members for
this dissertation over the last four years I would also like to thank Jonathan Stebbins and
Gail Mahood for providing me the opportunity to attend Stanford
Bob and Vicki Jones always offered an open ear whether it was science or solving
the worldrsquos problems Rick Hazlett (Pomona College) and John Winter (Whitman
College) both had a significant impact on shaping my future science goals and objectives
during my undergrad years and both encouraged me to pursue a career in geology They
were the catalyst for this whole experiment
I owe a lot to my research group members (both the Bird Cage and Soil
Chemistry Group) and those people who didnrsquot mind me tagging along Before I
introduce these people one person has been my friend and co-conspirator for many years
at Stanford Tim Copehellipyou are a pain in my ass Besides never picking up any of your
crap and breaking all my stuff this whole grad school experience would not have been
the same without you Other friends and acquaintances that will not be forgotten in
chronological order of me walking into my office for the first time on September 11
1998 include Anne Dyreborg Esra Inan Tom Bawden Patrick Redmond Phil
Neuhoff Thrainn Fredrickson Lara Heister Kaye Savage Andrea Fildani Mike Schultz
Trayle Kushan Kevin Theissen Angela Hessler Chris Van de Ven Kono Lemke Colin
Doyle Colleen Hansel Ben Bostick Matthew LaForce Nancy Grumet Scott Kroeker
Anders Meibom Brian Ebel Matt Ginder-Vogel Matt Pollizitto Spud Amy Weislogel
Travis Horton and Jake Waldbauer I would like to end this Acknowledgement section
and begin my dissertation with a haikuhellip
Better than outcrops
Are the people Irsquove met here
Thanks for all the beer
x
TABLE OF CONTENTS
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iv Prefacehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip vii Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip ix Table of Contentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xi List of Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xiv List of Illustrationshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xvi Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 1
Toxicity and Regulations of Chromiumhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 2 Mineralogy of Chromium in Serpentiniteshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 3 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5 Chapter Summarieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5
Chapter 2 Chromium Geochemistry of Serpentinized Ultramafic Rocks and Serpentine Soils in the Franciscan Complex of Californiahelliphelliphelliphelliphelliphellip 8
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 8 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 9 Backgroundhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 10
Geologic Background of the Serpentinite at Jasper Ridgehelliphelliphelliphellip 10 Serpentine Soils Characteristics and Reviewhelliphelliphelliphelliphelliphelliphelliphellip 11 Sources of Chromium ProtolithBedrockhelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 12 Chromium-Bearing Phases in Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 13
Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Location and Sample Collectionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Mineral Identification in Rocks and Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Soil Chemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 15 X-ray Diffraction and Clay-Size Mineral Identificationhelliphelliphelliphelliphellip 16 X-ray Absorption Spectroscopyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17 Sequential Extraction Procedurehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17
Resultshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Serpentinite Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Distribution of Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip21 Serpentine Soil Chemistry and Mineralogy at Site JR3helliphelliphelliphelliphellip 22 Sequential Extractionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 26
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 27 Concluding Remarkshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 33 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 44 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 47
xi
Chapter 3 Oxidative Promoted Dissolution of Chromite by Manganese Dioxide and Concurrent Production of Chromatehelliphelliphelliphelliphelliphellip 63
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 63 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 64 Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 65
Solid Phase Characterization and Synthesishelliphelliphelliphelliphelliphelliphelliphelliphellip 65 Extraction Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 66 Batch Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Solution Analysishelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69
Results helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite Extraction Experimentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite and Birnessite Batch Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 70 Secondary and Solid Phase Examination of Chromite and Birnessitehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 72 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 72
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 73 Chromium(VI) Pathways from Chromite and Birnessite into Solutionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 76 Serpentine Soils and Cr(VI)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 77
Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 79 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 80 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 84 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 87
Chapter 4 Chromium Geochemistry of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip100
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 100 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 101 Reviewhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102
Chromium Geochemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102 Mineralogy of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 103 Sources of Chromium for Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphellip 105 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 106 Soils Solutions and Related Waterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 108 Serpentine Soil Vegetationhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 110 Soil Organic Matterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 111
California Oregon and New Caledonia Serpentine Soilshelliphelliphelliphelliphelliphelliphellip 112 Sample Locations and General Soil Characteristicshelliphelliphelliphelliphelliphellip 112 BedrockProtolith Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 113 Soil Chemistry and Physical Propertieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 114 Soil Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116 Chromium and Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 117
xii
Chromium Spinel in Soils and Rockshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 119 Concluding Statementhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 120 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 121 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 122 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 129 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 131
Appendiceshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 141 Appendix Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 142 Appendix Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 143 Appendix A Estimation of the Thermodynamic Properties of Kaumlmmererite (Cr-Clinochlore)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip145
Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 145 Nomenclature and Crystal Chemistry of Cr-chloritehelliphelliphelliphelliphelliphelliphelliphelliphellip 146 Estimation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 147
Estimation of theStandard Molar Volume Using Unit Cell Measurementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of the Standard Molar Volume Heat Capacity and Entropyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of Standard Molar Enthalpy and Gibbs Free Energyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 150 Critique and Discussion of the Estimation Algorithms and SUPCRT92helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 151
Evaluation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 152 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 155 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 161 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 162
List of References
References to Chapter 1helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 166 References to Chapter 2helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 172 References to Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 179 References to Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 183 References to Appendix Ahelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 195
xiii
LIST OF TABLES Chapter 2 Table 21 Whole rock elemental concentrations of serpentinites chert and sandstone Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in
serpentinite rock at Jasper Ridge Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in
serpentine soil at Jasper Ridge Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25
and 26 Table 24 Bulk soil elemental concentrations for soils and bedrock collected along
the transect in Figure 21b Table 25 Chemical and physical properties of the soil at Site JR3 Table 26 Selected trace metal concentrations for each size fraction versus depth at
Site JR3 Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3 Table 28 X-ray diffraction d-spacings for common silicates Table 29 Results of the Diablo clay serpentine soil and chromite sequential
extraction experiments Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1
kg of sample Chapter 3 Table 31 Chromite and birnessite masses and concentrations used for each batch
experiment Table 32 Serpentine soil and birnessite experimental parameters Table 33 Summary of Cr(VI) rates obtained from batch experiments Table 34 Comparative analyses of batch experiment solutions and chemical
extractions of chromite and birnessite solids Chapter 4 Table 41 Measured Cr concentrations in serpentine soils worldwide Table 42 Summary of Cr Fe and Ni extractions from serpentine soils Table 43 Serpentine soil plant Cr concentrations Table 44 Chemical and textural properties of collected soils Table 45 Total major and trace element concentrations of the collected soils Table 46 Compositions for soil Cr-containing minerals identified in Figure 49 Table 47 Chromium-bearing mineral masses and volumes required to influence Cr
concentrations in soils Appendix A Table A1 Cr and Al end-members in the Cr-clinochlore solid solution
xiv
Table A2 Standard state thermodynamic properties of Cr and Al-bearing minerals Table A3 Unit cell measurements and molar volumes of clinochlore and Cr-
clinochlores Table A4 Entropy volume and heat capacity for kaumlmmererite using the Ransom and
Helgeson Method Table A5 Fictive polyhedral considered in the estimations Table A6 Estimated thermodynamic properties for kaumlmmererite
xv
LIST OF FIGURES Chapter 2 Figure 21a Soils map of Jasper Ridge Biological Preserve CA Figure 21b Serpentinite and serpentine soil sampling sites at Jasper Ridge Biological
Preserve CA Figure 22 Chromium concentrations versus depth for four serpentine soils in
Maryland California and New Foundland Figure 23 Petrographic images of serpentinite at Jasper Ridge Figure 24 Elemental concentrations relative to Cr for chlorite pyroxene magnetite
chromite and a chromite-silicate mixture in the serpentinite at Jasper Ridge
Figure 25 Backscattering electron images of minerals in the serpentinite at Jasper Ridge
Figure 26 Backscattered electron images of Cr-chlorite and Cr-magnetite assemblage in the serpentinite bedrock
Figure 27 Highest and lowest Cr concentrations at each site in the serpentine soil transect
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected for total gt2 mm sand silt and clay size fractions
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soils at Jasper Ridge
Figure 210 X-ray diffraction clay analyses of sample JR3515 Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at specific depths Figure 212 X-ray microprobe elemental maps of Cr and Fe for the serpentine soil and
serpentinite at site JR3 Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments Figure 214 Scanning electron microprobe images of the chromite standard after each
sequential extraction experiment Figure 215 Tie-lines connecting serpentinite chromite and end-member chromite with
soil chromite values Chapter 3 Figure 31 Scanning electron microprobe images of chromite and birnessite standards Figure 32 Chromite extraction experiments evaluating Cr Fe and Al Figure 33 pH and Eh versus time (hours) for each chromite-birnessite chromite and
birnessite batch experiments Figure 34 pH and Eh with Cr stabilities Figure 35 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant chromite and variable birnessite at pH 5 Figure 36 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant birnessite and variable chromite at pH 5
xvi
Figure 37 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for variable pH and constant chromite and birnessite Figure 38 Rate of Cr(VI) production (d[Cr(VI)]dt) as a function of the surface area
concentration of chromite birnessite and pH Figure 39 SEM examination of solids following the termination of the batch
experiments Figure 310 Serpentine soil and birnessite experiments Figure 311 Iron analyses and the solubility of amorphous Fe(OH)3 and goethite Figure 312 Congruent dissolution of chromite based on Fe and Cr Figure 313 Three pathways of Cr release and oxidation Chapter 4 Figure 41 Serpentine soil localities worldwide Figure 42 Field site pictures of serpentine soils Figure 43 Stability field diagram of Cr over a range of Eh and pH conditions Figure 44 Chromium and Al Wt plot and octahedral Cr and octahedral Al plots
for chromite and Cr-silicates including Cr mica (fuchsitemariposite) Cr-garnet (uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite)
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (m) for selected serpentine soils
Figure 46 pH of serpentine soils and related groundwater worldwide Figure 47 Soil pictures with Munsel color values and names from collected soils Figure 48 Cr XANES spectra for collected serpentine soils Figure 49 Backscattering images of mineral phases within each serpentine soil Figure 410 Chromite magnetite and hematite compositions from analyzed soils as
well as analyses for chromite reported from around the world Appendix A Figure A1 Compositional variation of Cr2O3 and Al2O3 Wt in Cr-chlorite Figure A2 Compositional variation of Cr and VIAl in Cr-chlorite based on 14 oxygens Figure A3 Molar volume estimation of kaumlmmererite based on XRD unit-cell
parameters Figure A4 Natural and calculated compositions of co-existing Cr-clinochlore and Cr-
spinel solid solution in the system Mg0-FeO-Cr2O3-Al2O3-SiO2-H2O at 2 kbar pressure
xvii
CHAPTER 1
INTRODUCTION
Serpentinites or metamorphosed ultramafic rocks are distributed worldwide and
are commonly associated with ophiolite complexes (Coleman 1977 Coleman and Jove
1992) These rocks weather to form serpentine soils a generic term used to describe any
soil derived from serpentinite regardless of its physical or chemical properties
Serpentinites and serpentine soils contain Cr-bearing minerals the most prevalent being
chromite and Cr released from these minerals into surface and subsurface environments
is potentially toxic and carcinogenic The main objective of this dissertation is to
evaluate the mineralogy and chemistry of Cr related to serpentinites and serpentine soils
over geologic time using an integrated analytical experimental and theoretical approach
Additionally the potential of serpentinites and serpentine soils as sources of non-
anthropogenic Cr contamination is investigated A discussion of Cr and its significance
to serpentinites and serpentine soils begins with its discovery by a Frenchman
investigating an ore named ldquored leadrdquo from the Urals
Nicolas-Louis Vauquelin discovered Cr in 1797 in the mineral crocoite PbCrO4
(Burns and Burns 1975) The metalelement was named chromium after the Greek word
chroma meaning color due to its strong influence on mineral pigmentation (Burns and
Burns 1975) For example rubies and emeralds derive their color from trace amounts of
Cr in their structures More importantly JT Lowitz in 1798 discovered Cr in chromite
or Cr-spinel the primary source of Cr for industrial purposes (Burns and Burns 1975)
Chromium is a group VIB transition metal with two common oxidation states (Cr(III) and
Cr(VI)) an average crustal abundance of 100 mg kg-1 and an abundance of 513x105 in
the solar system (relative to [H]=1x1012) (Mason and Moore 1982)
The majority of Cr on the earth exists as Cr(III) where it is an immobile cation
that complexes strongly with organic matter and adsorbs on oxides and silicates even at
low pHs (pHgt4) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) Additionally
the chemical mobility of Cr(III) is limited due to the formation of highly stable
(oxy)hydroxides (Rai and Zachara 1988 Rai et al 1989 Ball and Nordstrom 1998)
1
Chromium(III) has the highest octrahedral site preference energy (OSPE) of any trace
element and it isomorphically substitutes into octahedral sites in oxides
(oxy)hydroxides sulfides and silicates Although Cr(III) is known to substitute for both
octahedral Al(III) and Fe(III) the strongest correlation for substitution appears to be for
Al(III) in oxides and silicates as demonstrated in Chapters 2 and 4 and Appendix A
Chromium(VI) is a strongly oxidizing and weakly adsorbing oxyanion in the form of
chromate (HxCrO42-x) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995 Shiraki
1997 Ball and Nordstrom 1998) Consequently Cr(VI) is chemically mobile and
bioavailable over a wide range of pH (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Chromium(VI) is tetrahedrally coordinated and behaves similarly to
sulfate in minerals such as jarosite (Baron and Palmer 1996)
The pathways and chemistry of Cr oxidation and reduction in soils and related
waters are covered in depth by Fendorf (1995) and Ball and Nordstrom (1998) The
reduction of Cr(VI) to Cr(III) may occur by reactions with Fe(II)aq Fe(II)-containing
minerals such as magnetite organic matter H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996) Oxidation of Cr(III)
to Cr(VI) related to serpentinite-related water and serpentine soil solutions is still a matter
of conjecture Chromium(III) oxidation in serpentine soils has been hypothesized to be
the result of Mn oxides such as birnessite as suggested by both Gough et al (1989) and
Becquer et al (2003) This idea is explored in depth in Chapter 3 where birnessite is
shown to influence the release and oxidation of Cr from chromite over a range of pH (3
to 8) at 25oC
Toxicity and Regulations of Chromium
The examination of Cr based on its oxidation state is necessary due to the
juxtaposing behavior and toxicity of Cr(III) and Cr(VI) Chromium(III) is an essential
micronutrient in human nutrition used by the body to process sugar protein and fat
(Vincent 2000) Inhalation of Cr(VI) is known to cause irritation in the nose including
symptoms such as nosebleeds (epistaxis) and ulcers andor perforations in the nasal
septum (Katz and Salem 1993 Kuo et al 1997) Ingesting considerable quantities of
Cr(VI) can result in stomach ulcers convulsions kidney and liver damage and death
2
(Katz and Salem 1993) Skin contact with certain Cr(VI) compounds causes allergic
contact dermatitis (ulcers) and irritant dermatitis (Katz and Salem 1993) Chromate
induced ulcers or chromeholes are typically red swollen crusted painless lesions
(Katz and Salem 1993) The nosology of Cr is beyond the scope of this dissertation
however Lees (1991) is an excellent source for further information Overall the
Environmental Protection Agency (EPA) the Occupational Safety and Health
Administration (OSHA) the World Health Organization (WHO) and the Department of
Health and Human Services (DHHS) have determined that that Cr(VI) compounds can
increase the risk of lung cancer primarily by inhalation as well as other forms of cancer
caused by ingestion
Federal and state regulation standards appropriate for both Cr(III) and Cr(VI) are
summarized and discussed by Proctor et al (1997) Human health action levels for Cr
vary for each state and the ranges for total Cr in soils include the following 200-66000
Cr(III) mg kg-1 (residential) 02-2000 Cr(VI) mg kg-1 (residential) 2500-1000000
Cr(III) mg kg-1 (non-residential) and 64-71000 Cr(VI) mg kg-1 (non-residential)
(Proctor et al 1997) Incidentally Cr concentrations in serpentine soils and serpentinites
(gt200 Cr mg kg-1) exceed many state action levels and would be considered toxic The
EPA has set a limit of Cr(III)+Cr(VI) at 100 microg L-1 in drinking water The OSHA has
set limits of 500 microg of water soluble Cr(III) compounds per cubic meter of workplace air
or 500 microg m-3 1000 microg m-3 for metallic Cr and insoluble Cr compounds and 52 microg m-3
for Cr(VI) compounds for 8-hour work shifts and 40-hour work weeks The limits
proposed above are constantly changing and should not be used for reference without
checking current regulation standards
Mineralogy of Chromium in Serpentinites
Chromium occurs in a variety of minerals either as a major or minor constituent
(Burns and Burns 1975) Chromium-bearing minerals in serpentinites are primarily
oxides and silicates in which the Cr is present as Cr(III) Chromite is a Cr-bearing spinel
abundant in serpentinites and is resistant to low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) The chemistry and
significance of chromite is addressed throughout this dissertation Pyroxenes such as
3
augite and enstatite are primary minerals identified in serpentinites and contain lt8 Cr Wt
(Grapes 1981 Von Knorring et al 1986 Treloar 1987a Challis et al 1995
Sanchez-Vizcaino et al 1995 Deer et al 1996) Chromate minerals such as crocoite
(PbCrO4) are rare in serpentinites and the occurrence of these minerals are limited to Pb
deposits in Australia Tasmania and the Urals (Williams 1974 Crane et al 2001)
Chromium(III) released from the primary Cr-containing phases may
isomorphically substitute into oxides and silicates for Al(III) and Fe(III) due to size
(octahedral radii Cr3+ = 0615 Aring Al3+ = 061 Aring and Fe3+=0643 Aring) and charge
similarities Magnetite formed during serpentinization incorporates Cr into its structure
and serpentine minerals (antigorite lizardite and chrysotile) contain trace concentrations
of Cr (Spangenberg 1943 Page and Coleman 1967 Burkhard 1993) Eskolaite (Cr2O3)
has been identified in Finland (Von Knorring et al 1986) Metasomatic alteration of
ultramafic rocks commonly results in the formation of Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite) uvarovite (Cr-garnet) tawmawite (Cr-epidote) and
kaumlmmererite (Cr-chlorite) (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Neiva 1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981
Max et al 1983 Nutman et al 1983 Raase et al 1983 Chao et al 1986 Von
Knorring et al 1986 Kerrich et al 1987 Treloar 1987b Treloar 1987a Morand 1990
Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993 Christofides et al 1994
Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Proenza et al
1999) Additionally kyanite and andalusite in the proximity of ophiolite complexes
incorporate up 5 Cr Wt (Cooper 1980 Raase et al 1983 Kerrich et al 1987)
Chapter 4 provides an in depth discussion of these Cr-bearing silicates as well as
demonstrating mineral compositions reported in the literature based on octahedral Cr and
Al Appendix 1 is a preliminary investigation devoted to estimating the thermodynamic
properties of kaumlmmererite These properties can then be incorporated into geochemical
modeling programs to begin accessing the stability of this common serpentinite mineral
over a range of temperatures and pressures
4
Chromium in Serpentine Soils
Evaluating Cr in serpentine soils requires an examination of both the mineralogy
and chemistry of the serpentinite and soil Measured Cr concentrations in serpentinites
and serpentine soils range from ~200 mg kg-1 to 6 Wt all of which is Cr(III) (Soane
and Saunder 1959 Proctor and Woodell 1971 Shewry and Peterson 1976 Jaffre 1980
Proctor et al 1980 Rabenhorst et al 1982 Schwertmann and Latham 1986 Brooks
1987 Gough et al 1989 Gambi 1992 Lee 1992 Morrey et al 1992) In Chapters 2
and 4 chromite and Cr-magnetite are identified as the most prevalent sources of Cr
responsible for Cr-enrichment in these soils Chromite is resistant to chemical
weathering and it has been observed to undergo incongruent dissolution conserving Cr in
serpentine soils Chromium is tightly bound in the matrix of the soil less than 1 of the
total Cr can is removed using standard soil extraction methods (10 hydrogen peroxide
(H2O2) solution 01 M potassium permanganate (KMnO4) solution10 M hydrofluoric
(HF) acid 1 M hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and 1
mM barium chloride (BaCl2) solution) due to the stability of chromite Although silicates
contain minor concentrations of Cr compared to chromite they are perhaps the main
contributors of Cr identified in plants (lt200 Cr mg kg-1) and soil solutions (lt50 microg L-1)
During soil-forming processes Cr released to soil solutions by the dissolution of Cr-
bearing minerals are sequestered into a host of structural and surface sites available in
authigenic Fe(III) oxides Fe(III) hydroxides and clay minerals (Schwertmann and
Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990 Gerth 1990 Gasser and
Dahlgren 1994 Ilton and Veblen 1994 Brigatti et al 2000 Ilton et al 2000) The
clay-sized fractions in soils at Jasper Ridge Biological Preserve CA contain Cr
concentrations as high as 1000 Cr mg kg-1
Chapter Summaries
A detailed field and analytical examination of Cr in serpentinites and serpentine
soils at Jasper Ridge is presented in Chapter 2 A majority of the Cr enrichment and
variability with depth in these soils is the result of small grains of chemically resistant
chromite from the serpentinite protolith Chromite identified in the soil undergoes
incongruent dissolution conserving Cr relative to chromite compositions identified in the
5
serpentinite bedrock Sequential extraction experiments using barium chloride (BaCl2)
hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and hydrofluoric (HF)
acid were performed on a chromite standard confirming that Cr-bearing spinels are
resistant to chemical weathering X-ray diffraction analyses of the clay-size fraction in
the serpentine soils identified Cr-clinochlore smectite vermiculite and serpentine with
the smectite being the most abundant mineral in the clay-size fraction Chromium(III)
was the only identified oxidation state of Cr in the serpentine soil solids and in
serpentinites at Jasper Ridge in the solid phase This chapter represents a comprehensive
site assessment identifying the nature and behavior of Cr in serpentine soils with respect
to the bedrock mineralogy and depth
Chromite an abundant Cr-enriched spinel is responsible for a majority of the Cr
enrichment identified in ultramafic rocks serpentinites and serpentine soils The release
and oxidation of Cr from chromite is a pathway to consider when evaluating the potential
of ultramafic rocks and their related sediments and soils as sources of Cr(VI) In
Chapter 3 experiments aimed at quantifying the release and oxidation of Cr from
chromite in the presence of birnessite are described Chromite is extremely resistant to
dissolution (lt8 Cr mg kg-1 is released) when immersed in solutions including 1 mM
barium chloride solution 10 M hydrofluoric acid 10 hydrogen peroxide and 1 M
hydroxylamine hydrochloride with 25 acetic acid Highly oxidizing solutions of 01 M
potassium permanganate increased the rate and release of Cr relative to the other
solutions listed above thus birnessite a high valence Mn oxide present in serpentine
soils was chosen to determine if Mn could accelerate Cr release and oxidation from
chromite The results presented in Chapter 3 demonstrate that birnessite increases the
rate of Cr(VI) observed in solution from chromite This production of Cr(VI) is
dependent on the chromite suspension density where birnessite surface are concentrations
are greater than 20 m2 L-1 between pHs 3 and 8 at ~25degC
A comprehensive review of the chemical characteristics of serpentine soils
including topics such as the mineralogy of serpentine soils soil solutions organic matter
and vegetation is provided within Chapter 4 Additionally chemical and spectroscopic
analyses were performed on serpentine soils from a) New Caledonia b) Eight Dollar
Mountain Selma OR c) Nickel Mountain Riddle OR d) Harvard Mine Jamestown
6
CA e) Jasper Ridge Biological Preserve Stanford CA and f) Pillikin Mine El Dorado
CA Serpentinites and serpentine soils associated with ophiolite complexes cover ~1 of
the earthrsquos exposed surface and they are correlated to modern and ancient convergent
boundaries The physical and chemical characteristics of serpentine soils are directly
related to the alteration and tectonic history of the serpentinite as well as the soil-forming
factors including the parent material climate biota topography and time Serpentine
soils have Cr concentrations exceeding 200 mg kg-1 however the concentration and the
availability of Cr in these soils are not correlated with the five soil-forming factors
(climate organic material parent material topography and time) or the soil chemistry
Although only Cr(III) was identified in the soil solids Cr(VI) has been identified in soil
solutions (lt30 microg L-1) from serpentine soils in New Caledonia and California Chromite
and Cr-magnetite grains are identified in the soils listed above and these minerals are
responsible for a majority of the Cr enrichment in serpentine soils
Kaumlmmererite is a Cr-chlorite associated with serpentinized ultramafic rocks such
as those collected at Jasper Ridge Harvard Mine and Pillikin Mine Although Cr
concentrations are typically 01-2 Wt in kaumlmmererite it is modally abundant in
serpentinites (up to 5 of the volume of the rock) and a potential source of Cr to be
released into soils and groundwater In Appendix A the standard molar thermodynamic
properties of the Cr-clinochlore end-member kaumlmmererite (Mg5Cr(AlSi3)O10(OH)8) is
estimated using established estimation algorithms This initial study provides the
foundation for submitting a future paper presenting the thermodynamic properties for
major Cr-bearing silicates including fuchsite (Cr-muscovite) uvarovite (Cr-garnet) and
tawmawite (Cr-epidote) and their metamorphic significance related to serpentinized
ultramafic rocks
7
CHAPTER 2
CHROMIUM GEOCHEMISTRY IN SERPENTINIZED ULTRAMAFIC ROCKS
AND SERPENTINE SOILS FROM THE FRANCISCAN COMPLEX OF
CALIFORNIA
ABSTRACT
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of California
produces serpentine soils containing high concentrations of Cr as well as other potentially
toxic elements including Ni Co and Mn Chromium concentrations in serpentine soils
from Jasper Ridge Biological Preserve in the Central Coast Range are as high as 4760
mg kg-1 nearly three times greater than the serpentinite protolith (~1800 mg kg-1)
Chromium-containing minerals within the bedrock include chlorite (~03 Cr Wt )
enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt ) magnetite
(82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate phase
containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
8
INTRODUCTION
Serpentine soils are formed from the weathering of ultramafic rocks and their
metamorphic equivalents (serpentinites Brooks 1987) Measured Cr concentrations in
these soils range as high as 80600 mg kg-1 and are enriched compared to non-serpentine
soils (7 to 221 mg kg-1 Proctor 1992 McBride 1994) The distribution and
concentration of Cr in serpentine soils worldwide is highly variable due to the chemical
composition and mineralogy of the parent material and soil forming factors including
climate biota topography precipitation and time (Hotz 1964 Rabenhorst et al 1982
Schwertmann and Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al
1989 Gough et al 1989 Kaupenjohann and Wilcke 1995) Potential Cr-bearing phases
in serpentine soils have been assessed using extraction experiments and these phases
include organic material silicates secondary oxides and spinels (Schwertmann and
Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994 Kaupenjohann and
Wilcke 1995 Becquer et al 2003) The weathering of these Cr-bearing phases over
geologic time and how these sources govern the distribution of Cr in serpentine soils have
not been evaluated in great detail
Chromium is a group VIB transition metal present in the environment as Cr(III) a
non-hazardous species and micronutrient andor as Cr(VI) a toxin to living cells and a
Class A human carcinogen by inhalation (Daugherty 1992 Cohen et al 1993 James et
al 1997) Chromium(III) is chemically immobile due to forming low solubility
(oxy)hydroxides and stable surface complexes whereas Cr(VI) is a weakly adsorbing
(oxy)anion and thus is mobile in most aquatic environments (McBride 1994 Fendorf
1995 Ball and Nordstrom 1998) Chromite and Cr-magnetite have been cited as
primary sources of Cr in serpentine soils (Rabenhorst et al 1982 Schwertmann and
Latham 1986 Alexander et al 1989 Gough et al 1989) Although Cr is primarily
found in the form of Cr(III) in both ultramafic rocks and serpentinites Cr(VI) has been
measured in serpentine soils in California and New Caledonia (Gough et al 1989
Becquer et al 2003) Due to the abundance of Cr and the potential for Cr(VI) formation
serpentine soils and serpentinites are potential sources of non-anthropogenic Cr
contamination
9
The objectives of this study are to determine the distribution and chemistry of Cr
in naturally Cr-enriched soils using an integrated field analytical experimental and
theoretical approach A wide array of techniques and methods including X-ray
absorption spectroscopy and chemical extractions were used to identify the nature of Cr
and its availability in serpentine soils Serpentinites and serpentine soils at Jasper Ridge
Biological Preserve located near Stanford CA were used to evaluate the geochemical
behavior of Cr These soils cover approximately 50000 m2 and contain high
concentrations of the trace elements Cr Ni and Mn The soils are dominantly Mollisols
with grassland vegetation and they are pedalogically similar to those associated with the
Central Coast Range of the Franciscan Complex in the western United States The main
serpentinite body at Jasper Ridge was exposed approximately 400000 years ago allowing
pedogenesis and Cr cycling to occur over a geologically relevant time period (Burgmann
et al 1994) These soils can then be used as a proxy for evaluating the long-term
behavior of Cr
BACKGROUND
Geologic Background of the Serpentinite at Jasper Ridge
Jasper Ridge Biological Preserve is located in the foothills of the Santa Cruz
Mountains in California and is part of the Central Coast Range of the Franciscan
Complex an Early Cretaceous accretionary meacutelange containing blocks of greywacke
greenstone blueschist and serpentinized ophiolite (Figure 21) Formation
metamorphism and emplacement of serpentinite at Jasper Ridge began approximately
130-150 million years ago during Jurassic subduction along the California continental
margin (Page and Tabor 1967) Thin wedges of ultramafic rock (peridotite) were
detached from the down-going slab of oceanic crust and incorporated into the subduction
meacutelange Subsequent serpentinization of the ultramafic rock in the down-going slab
resulted in the formation of serpentinites now exposed at Jasper Ridge
Rare blueschist blocks (1 to 5 meters in width) present in the serpentinite at Jasper
Ridge attest to the degree of metamorphism and the depths from which portions of the
tectonic meacutelange has been elevated Transformation of greenstone in the meacutelange into
blueschist required high pressures in excess of ~5-7 kbars (15-20 kilometers depth) and
10
temperatures of 200-300 degC (Ernst 1971) The low density of the serpentine (245-250 g
cm-3) allowed upward diapiric migration of the serpentinite body through the denser
enclosing tectonic meacutelange (265 g cm-3) Uplift of the present day Coast Range and
activity associated with the San Andreas fault system have shaped the current topography
and exposure of serpentinite at Jasper Ridge Serpentinite is currently present as a narrow
discontinuous band of pervasively sheared rock located along the ridge top (Figure 21)
The serpentinite is in fault contact with the Whiskey Hill Formation (Cretaceous turbidic
sandstone and mudstone) and tectonically interleaved with the Franciscan Complex
Serpentine Soils Characteristics and Review
Relative to other soil types serpentine soils are characterized by higher
concentrations of Cr Ni Co and Fe lower concentrations of plant nutrients such as Ca
K N and P lower CaMg ratios and characteristic flora and physical properties (Brooks
1987) The pH of serpentine soils ranges from ~4 up to 9 (Brooks 1987 Cole 1992
Proctor 1992) Serpentine soil solutions obtained using tension-free lysimetry have Cr
concentrations between 01 and 32 micromol L-1 in which a majority of the Cr is present as
colloidal material (Gasser and Dahlgren 1994) The inhabitability of these soils for most
plants has been attributed to the imbalance of Ca to Mg (Walker 1954) the deficiency of
plant nutrients (Turitzin 1991) and the elevated concentration of heavy metals
(Robinson et al 1935) however linking the lack of plant productivity of these soils to
toxic levels of Cr has not been fully resolved (Brooks 1987 Kruckeberg 1992) The
unique flora and fauna associated with these soils have been well-documented and Cr
concentrations in the plants range as high as 600 mg kg-1 (Brooks 1987 Gough et al
1989)
Chromium concentrations as a function of depth for serpentine soils in Maryland
(Rabenhorst et al 1982) Tehama County CA (Gough et al 1989) the Klamath
Mountains CA (Hotz 1964) and New Foundland Canada (Roberts 1992) are
illustrated in Figure 22 Chromium concentrations with respect to depth at each location
are highly variable demonstrating both maximums and minimums The concentration of
Cr varies significantly at each location due to the chemical composition of the parent
11
material and other soil-forming factors nevertheless Cr concentrations typically decrease
with depth towards the bedrock
A variety of chemical extractions on serpentine soils suggest a broad host of Cr-
containing soil phases including organic material silicates secondary oxides and spinels
(Schwertmann and Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994
Kaupenjohann and Wilcke 1995) The maximum concentration of Cr extracted from a
serpentine soil reported in the literature is 226 Wt Cr from the New Caledonia laterite
using a 03 M dithionite-citrate bicarbonate (DCB) solution (Schwertmann and Latham
1986) Li and Fendorf (2000) performed extractions on the serpentine soils at Jasper
Ridge using organic acids (citric and oxalic acids) in which a 100 microM citric acid solution
was only able to remove 23 Cr microgg-1 (08 Wt Cr) from a soil containing ~3000 Cr mg
kg-1 This suggests a majority of Cr is bound in phases that are not reactive with these
solvents
Sources of Chromium ProtolithBedrock
Chromium primarily occurs within oxide and silicate phases in ultramafic rocks
and serpentinites Chromite is a common primary (igneous) Cr-bearing spinel in mafic
and ultramafic rocks and often survives low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) Isomorphic substitution of
Al3+ Fe3+ and Ti4+ into the octahedral site and Mg2+ Ni2+ Zn2+ and Mn2+ into the
tetrahedral site is common (Sack and Ghiorso 1991) however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987 Pan and Fleet 1991 Burkhard 1993 Christofides et al 1994 Challis et
al 1995 Sanchez-Vizcaino et al 1995 Deer et al 1996 Mathiesen 1999) Pyroxenes
such as augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) are primary minerals in
ultramafic rocks that incorporate varying concentrations of Cr into their structures (Deer
et al 1996)
Serpentinization is associated with the partial oxidation of Fe(II) present in
olivine and pyroxene forming magnetite (Brooks 1987) Chromium(III) derived from
primary chromite and pyroxene may isomorphically substitute into the Fe(III) site of
magnetite due to size (octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge
12
similarities Hydrothermal and CO2 metasomatism of ultramafic rocks may produce Cr-
silicate minerals including fuchsitemariposite (Cr-mica lt17 Cr Wt ) uvarovite (Cr-
garnet lt20 Cr Wt ) tawmawite (Cr-epidote lt10 Cr Wt ) and kaumlmmererite (Cr-
chlorite lt6 Cr Wt Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Phillips et al 1980 Schreyer et al 1981 Max et al 1983 Kerrich et al 1987 Treloar
1987 Pan and Fleet 1991 Christofides et al 1994 Sanchez-Vizcaino et al 1995)
These minerals common in altered ultramafic rocks have varying concentrations of Cr
that isomorphically substitute for octahedral Al in the silicate structure due to similarities
in charge octahedral preference and octahedral radii (Al3+ = 061 Aring) Any of these
silicates and magnetite are potential sources for Cr in soils produced by the weathering of
serpentinite and ultramafic rocks
Chromium-Bearing Phases in Soils
Soil minerals specifically Fe(III) (oxy)hydroxides and clays provide a host of
structural and surface sites that may sequester Cr(III) released to soil solution by
weathering of Cr-bearing phases in serpentinite and ultramafic rocks Isomorphic
substitution of Cr(III) for Fe(III) in oxides (oxy)hydroxides and amorphous compounds
is expected based on crystallographic considerations synthesis experiments and analyses
of selective dissolution of Fe(III) phases (Schwertmann et al 1989 Gerth 1990)
Additionally Cr(III) readily sorbs on surfaces of Fe(III)-containing oxides such as
magnetite and hematite via inner-sphere complexation and forms low-solubility
precipitates at pH values greater than 4 (Charlet and Manceau 1992 Peterson 1996)
Octahedral substitution of Cr(III) for Al(III) in clays (smectites and vermiculites) is
possible based on size and charge similarities Sorption of Cr(III) on edge and interlayer
sites of clay minerals has also been observed by Ilton and Veblen (1994) Soil organic
matter (SOM) is another possible host for soil Cr in which Cr can be chelated into stable
organomineral and organometallic complexes (Kaupenjohann and Wilcke 1995 Brady
and Weil 1999)
13
MATERIALS AND METHODS
Location and Sample Collection
Jasper Ridge Biological Preserve is a microcosm of accreted terranes located in
the foothills of the Santa Cruz Mountains CA Figure 21a is a soils map reported by
Kashiwagi (1985) depicting the alluvial Tertiary sandstone and Franciscan Complex
soils present at Jasper Ridge The Franciscan Complex soils are divided into two
subgroups those derived from greenstone-graywacke and those from serpentine The
serpentine soils dominantly Mollisols and serpentinite bedrock are present as a narrow
discontinuous band located along the ridge top they do not receive sediment from the
adjacent non-serpentine soils Soils and rocks were collected at Jasper Ridge from 1999
to 2002 and sample locations are shown in Figure 21b A 500 meter transect consisting
of 10 sites extends across the length of the serpentine soil body At each site along the
transect soils and bedrock were collected with respect to depth Site 1 (Diablo clay) is
not derived from serpentinite but rather from chert and greenstone Additionally Site 1
is located down slope of the main serpentinite body and has likely received sediment
from the adjacent serpentine soil Rocks were also collected at several exposed outcrops
located in the serpentinite body
Site JR3 (Figure 21b) was chosen for a detailed geochemical study to identify
factors affecting the geochemistry of Cr in serpentine soils Soil at JR3 is a loamy
magnesic thermic Lithic Haploxeroll The thickness of the soil is approximately 50 cm
deep Grass dominantly Lolium multiflorum comprises 95 of the vegetation at this site
and trees and shrubs are not present
Mineral Identification in Rocks and Soils
Mineral identification was accomplished by electron probe microanalysis on an
automated JEOL 733A electron microprobe operated at 15 kV accelerating potential and
15 nA beam current Calibration was conducted using natural geologic standards Beam
width for analysis was 10 microm Raw counts were collected over 20 s and were constant
with time indicating that elemental drift was negligible Detection limits for probe
analyses are ~002 oxide Wt Additionally backscattering electron (BSE) images of
the rocks and soils were obtained using the microprobe
14
Carbon-coated petrographic thin sections of rock samples were used for
microprobe analyses Soil samples were dried in an oven at 60degC for 24 hours mounted
onto a glass slide using epoxy and heated to 100degC until the epoxy cured An initial cut
of each section using a petrographic saw was made followed by a second infusion of
epoxy to ensure total impregnation Finally each section was cut polished and carbon-
coated for electron microprobe measurements
Soil Chemistry
Soils were analyzed for pH electrical conductivity (EC (deciSiemensm))
particle size ammonium acetate extractable cations organic carbon total nitrogen and
elemental composition The pH was determined using a 11 mixture of soil to deionized
water Electrical conductivity (EC) was determined on 2 mL of a soil solution extract
Particle size of the soil was measured using the hydrometer method as described by Gee
and Bauder (1986) These analyses were performed at the Utah State University
Analytical Laboratory
Ammonium acetate was used to measure the concentration of extractable K Ca
Mg and Na within the soil The procedure involved weighing 2 g of soil and adding 25
mL of ammonium acetate (20 M NH4OAc) followed by mixing on an orbital shaker for
30 minutes The solution was filtered using a Whatman 42 filter paper (pore size 25
microm) Potassium Ca Mg and Na concentrations were analyzed using atomic absorption
spectrophotometry (AAS)
Organic carbon was measured using the Walkley-Black titration method (Mebius
1960) Approximately 05 g of sediment was weighed and reacted with 5 mL of 20 M
potassium chromate (K2Cr2O7) followed by the rapid addition of 10 mL concentrated
sulfuric (H2SO4) acid The bottle was cooled for 30 minutes at room temperature and
diluted with 100 mL of deionized water Next 030 mL of 0025 M ortho-
phenanthroline-ferrous solution was added and the solution titrated to the endpoint using
10M ferrous sulfate (FeSO4)
Total nitrogen of soil samples was obtained using an automated combustion
method as described by Sweeny (1989) Approximately 150 mg of air-dried and sieved
15
(40 mesh) soil was placed in a tin foil encapsulating cup A Leco total nitrogen analyzer
with resistance furnace was then used to determine total nitrogen
Major and minor element concentrations for soil and rock samples were obtained
by completely dissolving the samples using a mixture of hot concentrated nitric
perchloric and hydrofluoric acids Elemental concentrations were measured in the
supernatant using inductively coupled atomic emission spectroscopy (ICP-AES)
Standards were plusmn5 mg kg-1 of their known values These analyses were completed at
Chemex Laboratory (Vancouver Canada) Soil samples were dried and sieved (lt2mm)
before analysis Additionally total element concentrations for gt2mm sand silt and clay
size fractions were obtained for each sample collected at Site JR3
X-ray Diffraction and Clay-Size Mineral Identification
X-ray diffraction (XRD) analyses were conducted on the clay-size fraction
(lt2microm) from Site JR3 using a Siemens Diffractometer D5000 at the University of Idaho
The clay-size fraction was isolated using a suspension method described by Vaniman
(2001) Organic matter and calcium carbonate were removed from the clays using a
concentrated hypochlorite solution (Kunze and Dixon 1986) Crystalline and non-
crystalline iron oxides were removed using a citrate-sodium bicarbonate-sodium
dithionite (CBD) solution (Kunze and Dixon 1986) The remaining phyllosilicate
minerals were analyzed using a series of treatments including 1) magnesium-saturation
(2 M MgCl and 2 M MgAOC) and air drying 2) magnesium-saturation and glycolation
(1 glycerol-ethanol solution) 3) potassium-saturation (1 M KCl) and air drying and 4)
potassium (1 M KCl) saturation followed by heating to 500degC (Whittig and Allardice
1986) X-ray diffraction patterns were collected between 2θ values of 20-350deg and 20-
150deg with a generator potential of 30 kV a generator current of 22 mA (using CuKα
radiation) a Ni filter and a scan speed of 1degmin The software package JADE was used
for XRD data analysis and mineral identification
X-ray Absorption Spectroscopy
In-situ fluorescence yield X-ray absorption near-edge structure (XANES)
spectroscopy at the Cr K-edge (edge inflection point is 5989 eV for Cr(0)) was conducted
16
at the Stanford Synchrotron Radiation Laboratory (SSRL) Stanford CA using
beamlines 4-1 and 4-3 A Si(220) monochromator and a Ge-detector were used and
three scans were collected for each XANES analysis Soil and rock samples from Site
JR3 were prepared by removing the gt2mm fraction and crushed to a fine powder
Samples were loaded into a plastic sample holder and covered with Mylar tape
Elemental mapping of three samples were completed at SSRL on Beamline 11-2
using X-ray absorption spectroscopy (XAS) A Si(220) monochromator with a 6 microm
focusing capillary was used and fluorescence intensity was monitored with a Ge detector
Samples were prepared similar to those used for mineral identification using the electron
microprobe however the thickness of the sample was decreased to lt20 microm
Sequential Extraction Procedure
Sequential extraction experiments were completed on 2 g lt2 mm sieved soils
(Diablo clay from Site 1 at 10 cm depth and serpentine soil from Site 2 at 4 cm depth) in
three steps The exchangeable fraction was obtained using a barium chloride (1mM
BaCl2) solution for 1 h at 25 degC (Kunze and Dixon 1986) The Fe-crystalline oxide
fraction was obtained by reacting the solids with a hydroxylamine-hydrochlorideacetic
acid (10 M NH2OHsdotHCl in 25(vv) HOAc) solution for 6 h at 60 degC (Chester and
Hughes 1967) The silicate fraction was obtained by reacting the solids with
concentrated hydrofluoric acid (10 M HF) for 16 h at 25 degC Trace element
concentrations for each extraction were measured on the supernatant using inductively
coupled plasma optical emission spectrophotometry (ICP-OES) All analyses were
completed in triplicate Additionally sequential extraction experiments were completed
on 2g 106-250 mm sieved ultrasonically cleaned commercial grade chromite (MSDS
No 302-1 WHMIS Class D2A) using the same procedures listed above The
stoichiometric composition of chromite obtained using the electron microprobe is
(Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 Chromite was imaged after each sequential
extraction using a JEOL 5600LV environmental SEM equipped with an EDAX thin-
window energy dispersive X-ray detector
17
RESULTS
Serpentinite Mineralogy
The average serpentinite at Jasper Ridge is composed of serpentine (mainly
lizardite and antigorite 70) chlorite (15) talc (10) magnetite (4) and chromite
(1) based on petrographic and microprobe observations of five samples (JR3E JR6E 5
10 and JR3) however brucite has been reported as a major constituent in related
Franciscan Complex serpentinites (Hostetle et al 1966) Other minerals identified
include olivine augite and enstatite Although the serpentinite mineralogy is relatively
simple the modes and textures of the minerals present in the rock vary significantly from
site to site In Table 21 whole rock serpentinite compositions from Sites JR1 JR3E
JR6E 2 5 10 and JR3 are listed The average Cr concentration of the serpentinites is
~1700 mg kg-1 (tab 1) Iron (~7) Mg (~11) Ni (~3000 mg kg-1) and Co (~100 mg
kg-1) concentrations are also elevated in the serpentinite relative to the chert and
sandstone bedrock adjacent the main serpentinite body (Figure 21b and Table 21)
A range of rock and mineral textures are petrographically evident in the
serpentinite at Jasper Ridge (Figure 23) No foliations or lineations are observed in the
serpentinite (Figure 23a) obtained from the bottom of the soil pit excavated at Site JR3
(Figure 21b) Subhedral chromite and magnetite grains are randomly interspersed
amongst the serpentine matrix (Figure 23a) In serpentinite obtained in outcrop at Site
JR5E (Fig 21b) a lsquocrocodile skinrsquo texture is produced by thin irregular veins of fine-
grained subhedral magnetite separating serpentine grains (Figure 23b) In Figure 23c
numerous crosscutting veins of chlorite occur in orthopyroxene and define the foliation of
the rock in outcrop at Site JR3E (Figure 21b) The numerous crosscutting microscopic
veins random foliation patterns (or the lack thereof) and subhedral oxide grains suggest
the Jasper Ridge serpentinite has undergone extensive physical and chemical alteration
typical of most serpentinites
Microprobe analyses of minerals in the serpentinite demonstrate that the main Cr-
bearing phases are chlorite enstatite augite Mg-Al chromite Cr-magnetite and a
chromite-silicate mixture (defined in the following paragraph) Selected compositions
for these minerals are listed in order of their Cr concentrations in Table 22a Chlorite
contains the least Cr whereas the chromite-silicate mixture is the most enriched with
18
respect to Cr Figure 24 shows correlations of Cr content relative to Al Fe Mg and Si in
order to provide insight into the serpentinite mineralogy and to aid in evaluating trends of
Cr substitution
A fine-grained mixture containing spinel (chromite andor Cr-magnetite) and an
unidentifiable Cr-silicate contains the highest Cr concentration in the serpentinite at
Jasper Ridge (Table 22a shown in Figure 25a and 25b) This fine-grained mixture will
be referred to here as the chromite-silicate mixture (CSM) with chemical properties
similar to those in Table 22a and Figure 24 CSM is common around the rims of
protolith chromite grains and a number of the larger (~05 mm) Cr-magnetite grains in
the serpentinite The microcrystalline phase responsible for contributing Si to these
analyses is possibly chlorite due to similarities of false colors generated by backscattering
electron (BSE) images and chlorite commonly rimming CSM Microprobe analyses of
CSM listed in Tables 22 and 23 are considered as mixtures of several mineral phases
due to the fine-grained texture (lt5 microm) Chromium concentrations range from 12 to 16
Cr Wt in this phase CSM is chemically distinct from the chromite and magnetite as
shown in Figure 24
The chemical compositions of protolith chromite exhibit little to no variation in
any of the serpentinites collected at Sites 2 JR3E JR4E JR5E JR6E JR7E and JR3
(Table 22a and Figure 24) Chromite contains low Fe (~9 Wt ) high Mg (~9 Wt
) and high Al (~26 Wt ) compared to Cr-magnetite The average Cr concentration
for chromite is ~12 Cr Wt Additionally chromite grains are irregular and subhedral
suggesting that they have undergone significant alteration during metamorphism
Chromium concentrations in the magnetite vary depending on the proximity and
degree of alteration of the chromite (Table 22a and Figure 24) Magnetite grains with
less than 2 Cr Wt are common in veinlets in serpentinite not formed adjacent
chromite whereas magnetite with 8 to 12 Cr Wt were formed proximal or adjacent
chromite Chromium-enriched magnetite is also referred to as ferrite-chromite in the
literature (Spangenberg 1943 Burkhard 1993)
Chlorite and pyroxenes contain minor concentrations of Cr (lt1 Wt Table
22a) Aluminum Fe and Mg concentrations varied in the pyroxenes however the
majority of chlorites are relatively similar in composition (Figure 24) Chlorites are
19
more abundant in the serpentinite compared to the pyroxenes and both minerals were
subhedral and randomly distributed over the entire sampling area (Figure 21b)
Backscattering electron (BSE) images of what appeared to be a petrographically
opaque single chromite grain in Figure 23a demonstrates chromite has been
metasomatically altered resulting in the formation of Cr-magnetite Cr-chlorite and CSM
(Figure 25a) Represented compositions for these minerals (indicated by letters in Figure
25a) are listed in Table 23 Magnifying the chromite and CSM (Figure 25a)
demonstrates the fine-grained texture of CSM and an alteration rim (transition zone) is
also seen along the chromite edge (Fig 25b) This alteration rim is depleted in Cr with
respect to the chromite as shown in Figure 25b
In Figures 23a and 25c Ni was identified in three phases in the bedrock
including olivine (F 03 Ni Wt ) serpentine (G and H 03 and 0 Ni Wt ) and Ni-
Fe (I sim55 Ni Wt ) alloy (Figure 25c and 25d) their compositions are indicated by
letters and listed in Table 23 Alteration of the olivine resulted in the formation of lt10
microm Ni-Fe alloys shown in Figure 25c These Ni-Fe alloys form a corona surrounding
the remaining olivine and the illustration in Figure 25d highlights the minerals and
chemical changes
Chromite and Cr-magnetite alteration leading to the paragenesis of CSM and Cr-
chlorite is common in the serpentinites at Jasper Ridge The petrographic image in
Figure 26a and its corresponding BSE image in Figure 26b illustrate Cr-magnetite
(originally protolith chromite) partially altering to CSM and Cr-chlorite The chemical
compositions of each mineral (indicated by letters in Figure 26b) are listed in Table 23
CSM appears to marginally and metasomatically replace Cr-magnetite and is enriched by
~2 Cr Wt Chromium-chlorite is present in microcracks in Cr-magnetite and
encompasses the entire Cr-magnetite and CSM grain Serpentine minerals adjacent the
Cr-chlorite contains no measurable Cr
Distribution of Chromium in Serpentine Soils
Serpentine soils and underlying bedrock were collected at ten sites along a 500
meter transect extending the length of the main serpentine body as shown in Figure 21b
20
The range of Cr concentrations at each site (center) and Cr concentrations with respect to
depth (periphery) are shown in Figure 27 and soil analyses are listed in Table 24
Bedrock Cr concentrations for Sites 2 5 and 10 are also shown in these plots (Fig 27)
Chromium enrichment does not appear to be consistently found at a particular depth or
soil horizon for any of the sites Sites 3 9 and 10 demonstrate a Cr maximum at
approximately 006 m however Sites 1 2 5 and 6 exhibit a Cr minimum at roughly the
same depth The highest soil Cr concentration in the soil along the transect is 4760 mg
kg-1 at Site 3 at 006 meters depth The lowest soil Cr concentration is 1725 mg kg-1 at
Site 10 at 0085 meters depth The average bedrock Cr concentration is ~2000 mg kg-1
(Figure 27 Table 24) It is important to note Site 1 consists of Diablo clay not
serpentine soil however this site is located adjacent and down-slope of the main
serpentinite body From these observations it is apparent that there are no systematic
trends in soil Cr concentrations with respect to the position of the site and depth and soil
samples often contain higher concentrations of Cr with respect to their corresponding
bedrock and to the average Cr concentration of ultramafic rocks (1600 mg kg-1 Green
1972)
Trace metals including Ni Mn and Co were also evaluated in the serpentine soil
for Sites 1 through 10 and the results are listed in Table 24 Nickel concentrations range
from 1590 to 3960 mg kg-1 and do not directly correspond to the Cr fluxes with respect
to depth (Table 24) Nickel concentrations are less than Cr from Sites 1 to 4 From Sites
5 to 10 Ni concentrations exceed those of Cr averaging ~3600 mg kg-1 Unlike Cr Ni
concentrations are equivalent or greater in the bedrock (~3300 mg kg-1) as compared to
the soil (~3200 mg kg-1) Soil Mn concentrations at each site are relatively uniform at
~1800 mg kg-1 with little to no variation with depth (Table 24) Cobalt concentrations
do not exceed 238 mg kg-1 (Table 24) and are relatively uniform (~200 mg kg-1) with
respect to location along the transverse and depth at each site
Major elements including Al Fe Ca and Mg are listed for the serpentine soils
along the transect with respect to depth in Table 24 Aluminum Fe Ca and Mg
concentrations in the soils at each site are similar with exception of Site 1 (Diablo clay)
Aluminum concentrations in the Diablo clay are ~46 whereas the serpentine soils
contain less than 2 Al (Table 24) Iron and Mg concentrations are also low lt65 and
21
lt411 respectively in the Diablo clay compared to Fe (gt8) and Mg (gt6)
concentrations in the serpentine soil (Table 24) Calcium concentrations are low (lt1)
for all the serpentine soils and even the Diablo clay (Table 24) With a broad
understanding of the chemical characteristics of the serpentine soil at Jasper Ridge Site
JR3 (Figure 21b) was chosen for a detailed chemical study to examine all the factors
governing the chemistry of Cr in the soil
Serpentine Soil Chemistry and Mineralogy at Site JR3
Soil chemistry
Chemical properties (pH electrical conductivity organic matter nitrogen and
ammonium acetate extractable cations) and textures for four soil samples from Site JR3
are listed in Table 25 The pH of the soil is near neutral (pH from 671 to 698) and
increases slightly with respect to depth Electrical conductivities (EC) in the soil ranges
from 03 to 09 dSM a range indicative of relatively few dissolved salts andor major
dissolved inorganic solutes The organic carbon content of the soil is 32 at the surface
and decreases with depth to 10 near the bedrock Total nitrogen for the samples ranged
from 003 to 064 Ammonium acetate extractable Ca K Mg and Na are listed in
Table 25 The extractable concentration of both Ca and K decrease with depth whereas
there is no apparent trend associated with the concentrations of Mg and Na The
serpentine soil textures are dominantly clay loams however the texture is a sandy clay
loam near the bedrock
Whole soil samples and size fractions including gt2 mm sand (2-002 mm) silt
(002-0002 mm) and clay (lt2 microm) size fractions with respect to depth were chemically
digested and analyzed for total trace metal concentrations shown in Figure 28 and listed
in Table 26 to determine if the abundance of Cr was linked to a particular size fraction
(the percent of sand silt and clay size fractions are reported in Table 25) Additionally
Ni Mn and Co concentrations with respect to depth and size fraction are also presented in
Figure 28 and Table 26 Total and gt2mm size fraction Cr concentrations are similar at
the top and bottom of the soil profile ranging from 3297 to 5976 Cr mg kg-1 Chromium
concentrations exceed detections limits (gt10000 mg kg-1) for the sand-size fraction It is
important to note that the sand-size fraction corresponds to the size of most chromite
22
grains observed in the serpentinite Chromium in the clay fractions are low (~1000 mg
kg-1) compared to the other size fractions however these concentrations are still enriched
compared to non-serpentine soils
In Figure 28 and Table 26 Ni concentrations are enriched in the clay-size
fraction with values ranging from 2840 to 2920 mg kg-1 The silt-size fractions have the
lowest Ni concentrations (~1800 mg kg-1) Manganese concentrations are slightly
elevated relative to total Mn in the sand-size fraction with values ranging from 1728 to
2701 mg kg-1 The clay-size fraction is the least Mn enriched size-fraction with
concentrations at ~1200 mg kg-1 Cobalt concentrations are elevated in the sand-size
fraction with values ranging from 159 to 241 mg kg-1 and concentrations in the gt2mm
silt and clay-size fractions are similar to the total (~150 mg kg-1)
Soil Cr-bearing minerals
Chromium-bearing minerals in the soil identified using the electron microprobe
include chromite Cr-magnetite and CSM Chromium concentrations in soil chromite
Cr-magnetite and CSM are graphically shown in Figure 29 as a function of Al Fe Mg
Si and Al+Fe in Wt Representative analyses of the Cr-bearing soil minerals are listed
in Table 22b Compositions of soil chromite in Figure 29 are chemically distinct from
those identified in the protolith in Figure 24 Chromium concentrations in soil chromites
range from 22 to 37 Cr Wt containing minor concentrations of Al and Mg Values for
chromite in the serpentinite bedrock (Table 22a and Figure 24) and the FeCr2O4 end-
member are also shown in Figure 29 The dashed lines in Figure 29 connect Jasper
Ridge bedrock chromite (Figure 24) and stoichiometric FeCr2O4 where it can be seen
that compositions of soil chromite lie close to these tie lines Overall soil chromite
demonstrates a decrease in Al and Mg and an increase in Fe relative to Cr compared with
protolith chromite Soil Cr-magnetites contain between 7 and 20 Cr Wt and are
enriched in Cr compared to Cr-magnetites identified in the bedrock (Figure 24) Cr-
magnetite however reveals a decrease in Fe relative to Cr Minor concentrations of Si
(lt2 Wt ) in the soil chromite and Cr-magnetite may represent slightly altered rims or
contamination from sample preparation Chromium concentrations in CSM maintain a
consistent Cr concentration of ~11 Cr Wt similar to those found in the bedrock (~14
23
Cr Wt ) Soil CSMs with compositions between 8 and 12 Mg Wt are more than
likely associated with analyses containing more microcrystalline chlorite compared to
soil CSMs with less than 2 Mg Wt Soil CSM appears be less enriched with Si (lt14
Wt ) relative to those identified in the serpentinite (15 to 27 Wt )
The size of the Cr-bearing minerals varied significantly for each sample ranging
from 10 microm up to 2 mm The majority of chromite grains in the soil correspond to the
sand-size fraction and also vary from euhedral to subhedral with no discernable pattern
The sum of cations based on four oxygens for the chromite and Cr-magnetite analyses
range between 297 and 301 indicating these minerals are spinels not (oxy)hydroxides
Compositions obtained for chromite and Cr-magnetite may represent alteration rims
where weathering of the mineral in the subsurface environment has been most active
BSE images of chromite and Cr-magnetite were not able to definitively identify whether
these analyses in fact represent alteration rims
Mineralogy of the Clay-Size Fraction
The clay-size fraction in the serpentine soil at Site JR3 was evaluated with respect
to depth using XRD and specific chemical and heat treatments The d-spacings (Aring) and
their relative intensities for each soil and treatment with respect to depth are listed in
Table 27 Figure 210 shows the XRD analyses for clay sample JR3515 (depth 5-15
cm) and these analyses are almost identical to the other clays examined at Site JR3
Additionally Table 28 from Whittig and Allardice (1986) is provided to interpret the X-
ray diffraction analyses in Table 27 For example a d-spacing of ~14Aring is present in all
the treatments indicating the presence of chlorite in the clay-sized fraction Using an
XRD analysis program the clay-size fraction was determined to be composed of
smectite vermiculite serpentine minerals (lizardite and antigorite) and chlorite
(clinochlore and Cr-clinochlore) Smectite is the most abundant mineral in the clay-size
fraction whereas Cr-chlorite is the least abundant Overall the d-spacings and relative
intensities are similar despite the treatments and changes in depth (Table 27) This
indicates that the mineralogy of the clay-size fraction does not vary significantly in the
soil column
24
Chromium concentrations in the clay-size fractions are ~1010 mg kg-1 (Table
26) Smectite and vermiculite are possible Cr-bearing clay minerals in which isomorphic
substitution of Cr(III) could occur in the dioctahedral sheet Serpentine minerals
(lizardite and antigorite) are not likely hosts of Cr based on crystallographic
considerations and the absence of Cr in analyzed serpentine minerals in the bedrock
(Table 23) Clinochlore may also be a host for Cr in the inner layer octahedral sheet
Chromium-clinochlore is the only Cr-containing mineral identified in the clay-size
fraction using XRD and Cr-clinochlore (Cr-chlorite) is also an abundant mineral
identified in serpentinite protolith (Tables 22 and 23 Figure 26)
XAS of chromium in relation to depth
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in serpentine soils and bedrock at Site JR3 The XANES (X-ray absorption near-edge
structure) spectra demonstrate the absence of Cr(VI) in the soil column at Site JR3
(Figure 211) based on the lack of a pre-edge feature The detection limit for Cr(VI)
using this method is approximately ~20 mg kg-1 Ultimately XAS analyses demonstrate
Cr(VI) is not detected in any of the soil solids
X-ray microprobe analyses of soil and rock samples (0-5cm depth 5-10cm depth
and serpentinite) at Site JR3 were processed to map the relative concentrations of Fe and
Cr in which areas of elemental enrichment are identified by a violet color areas of
depletion by a blue color (Figure 212) Areas of Cr and Fe enrichment in Figure 212 are
localized with well-defined boundaries corresponding to grain edges and these elements
are not correlated in a well-defined trend Minerals assigned to these areas of Cr and Fe
enrichment were made based on petrographic and microprobe observations on the same
thin section For example high Fe and no Cr is magnetite whereas high Fe and high Cr
is either chromite or Cr-magnetite Chromite magnetite and Cr-magnetite appear to be
present in the serpentine soil at a depth between 0 to 5 cm (Figure 212A) In Figure
212b CSM is a possible mineral detected in the soil (depth 5-10 cm) and the presence
of magnetite is clearly seen due to the localized enrichment of Fe and no Cr Iron and
Cr spectra for the serpentinite were much more heterogeneous as compared to the soil
samples and Cr concentrations appear be more scattered (Figure 212c) These X-ray
25
microprobe analyses provide a complimentary method for examining Cr geochemistry
(including Cr(III) and Cr(VI)) in naturally occurring soil and rock solids however
Cr(III) is the only detected valence state in serpentine soils and rocks at Jasper Ridge
Sequential Extractions
Sequential extractions were completed on 1) Diablo clay (Inceptisol) from Site 1
(Figure 21b) at a depth of 10 cm 2) serpentine soil (Mollisol) from Site 2 (Figure 21b) at
a depth of 4 cm and 3) commercial grade chromite (106-250 microm) and these results are
shown in Figure 213 and listed in Table 29 Chromium extracted from the
exchangeable fraction is minimal (lt01 mg kg-1) for all the samples (number 1 in Table
29 Fig 213) Minor Cr concentrations are associated with the Fe-crystalline oxide
fraction in the Diablo clay (19 Cr mg kg-1) and serpentine soil (78 Cr mg kg-1) whereas
chromite only released 05 Cr mg kg-1 (number 2 in Table 29 Figure 213) Increased
concentrations of Ni and Fe are present in the Fe-oxide fraction of the Diablo clay (161
Ni mg kg-1 and 4474 Fe mg kg-1) and serpentine soil (5138 Ni mg kg-1 and 5138 Fe mg
kg-1 Table 29) Chromium extracted from the Diablo clay serpentine soil and chromite
in the silicate fraction is 14 31 and 14 Cr mg kg-1 respectively (number 3 in Table 29
Figure 213) The sum of extracted Cr Fe Ni and Mn concentrations (number 4 in
Table 29) are low with respect to total metals (number 5 in Table 29) In fact only 07
12 and 00 Cr was removed from the Diablo clay serpentine soil and chromite
respectively (number 6 in Table 29)
Chromite was extremely resistant to all the extractions and less than 001 of the
total Cr was removed (Table 29) Scanning electron microprobe (SEM) images of the
chromite after each extraction are shown in Figure 214 Chromium concentrations
released from the chromite (Table 29 and Figure 214a) increased with each sequential
extraction experiment starting with the exchangeable fraction (Figure 214b) progressing
into the Fe-crystalline oxide fraction (Figure 214c) and ending with the silicate fraction
(Figure 214d) The surface of the chromite became smoother after each extraction and
crystal imperfections in defect sights located along the edges and corners progressively
decreased (Figure 214) Chromite does not appear to be an easily accessible source of
Cr based on these extraction experiments
26
DISCUSSION
Serpentinite at Jasper Ridge has undergone significant physical and chemical
alteration since its formation and emplacement off the California continental margin
Despite being subjected to high pressures (~5-7 kbars) and elevated temperatures (200-
300degC) trace element concentrations within the serpentinite remain elevated Jasper
Ridge serpentinites are enriched with Cr Co Fe Mg Mn and Ni compared to chert and
sandstone located adjacent the main serpentinite body (Table 21 and Figure 21b) Over
the past 400000 years Cr-bearing minerals in the serpentinite have been weathering and
accumulating in the soil During this time Cr concentrations in the soil have
progressively increased relative to Cr-concentrations in the serpentinite protolith This Cr
enrichment is directly linked to specific weathering-resistant minerals identified in the
soil
Chromium in Serpentinites
Three minerals in the bedrock at Jasper Ridge contain appreciable concentrations
of Cr Cr-magnetite (82-103 Cr Wt ) chromite (~108 Cr Wt ) and CSM (~133
Cr Wt ) Cr-magnetite and CSM were derived from the alteration of primary igneous
chromite CSM or Cr-spinels with high concentrations of Si have been noted by
(Burkhard 1993) as a Cr-spinel containing a possible microscopic silicate phase Our
analyses support this result in which the microcrystalline silicate is probably Cr-chlorite
due to its common occurrence with CSM (Figure 26) Definitive identification of this
microcrystalline mixture was not possible due to size of the crystals relative to the spot
size of the electron microprobe Chlorite and pyroxene contain minor concentrations of
Cr (~1 Wt ) and these were the only identified Cr-containing silicates in the
serpentinite Minor concentrations of Cr observed in some chlorites are directly related
to the metasomatic alteration of chromite and Cr-magnetite (Figure 26) Other common
metasomatically-derived Cr-bearing silicates such as fuchsitemariposite and uvarovite
are not present possibly due to the lack of aqueous K+ and Ca+2 in the alteration fluids
Overall the chemical compositions of these protolith minerals (Figure 24) demonstrate
only minor chemical variations allowing comparative analysis with Cr-bearing soil
minerals
27
Chromium in Serpentine Soils
Cr-bearing soil minerals identified using the electron microprobe include
chromite Cr-magnetite and CSM and these phases are chemically distinct from those in
the bedrock (Figure 24 and 29) The highest concentration of Cr (41 Wt ) in the soil
is in chromite with a composition approaching the FeCr2O4 end-member This represents
a ~80 increase in the Cr concentration relative to bedrock chromite (see Figure 215)
Cr-magnetite also demonstrates Cr enrichment (up to 20 Cr Wt ) with respect to
magnetite identified in the bedrock (~10 Cr Wt ) (Figures 24 and 29) The textures of
the soil Cr-bearing spinels exhibit sharp irregular edges possibly due to extensive
weathering CSM in the soil is similar to those found in the bedrock with Cr
concentrations ranging from 11 to 14 Cr Wt however Si concentrations in soil CSMs
(8 to 14 Wt ) are noticeably less than those measured in the serpentinite (15 to 20 Wt
) The weathering of CSM might preferentially release Si relative to other elements in
its structure Cr-silicates (chlorite and pyroxenes) were not identified in the soil possibly
due to complete weathering of these phases or they were too fine-grained for microprobe
analyses however Cr-clinochlore was identified in the clay-size fraction using XRD
Elemental maps shown in Figure 212 corroborate electron microprobe observations by
revealing well-defined zones of Cr and Fe enrichment in both the soils and serpentinite
These areas of Cr and Fe enrichment are either chromite Cr-magnetite CSM or
magnetite Overall these maps reveal Cr is not homogeneously distributed in the soil
confirming Cr is in specific mineral phases Additionally the observed abundance of
chromite Cr-magnetite and CSM in the soils indicates that these minerals are probable
sources for a majority of the Cr enrichment identified in the soils
Soil Spinels
Soil chromite and Cr-magnetite appear to have undergone chemical alteration
enriching the minerals with respect to Cr In Figure 215 compositions of bedrock
chromite and end-member FeCr2O4 are shown together with tie-lines like those in figure
8 connecting both minerals Reaction progress in Figure 215 is based on the assumption
that the serpentinite chromite (0 reaction progress) is proceeding toward FeCr2O4
(100 reaction progress) in the serpentine soil Soil chromites from Jasper Ridge are also
28
plotted in Figure 215 where Cr concentrations are assumed to vary linearly between the
bedrock chromite and chromite end-member which defines the reaction progress of the
mineral These tie-lines correlate extremely well to the chemical composition of
chromites identified in the soil For example chromite containing ~30 Cr Wt will
have ~15 Fe Wt ~13 Al Wt and ~5 Mg Wt similar to soil chromite in Figure 29
The retention of Cr in chromite shown in Figure 215 has significant implications towards
the availability and cycling of Cr in serpentine soils
One possible mechanism for the chromite composition trends illustrated in
Figures 29 and 215 is incongruent dissolution where Cr is retained relative to other
elements Sequential extraction experiments (Figure 213 and Table 29) support this
idea in which Fe is more easily released than Cr (1096 Fe mg kg-1 to 19 Cr mg kg-1)
from chromite Unfortunately Mg and Al were not monitored during the experiments
From these sequential extraction experiments and chromite analyses (Figure 215) Al
and Mg have been preferentially released from the chromite structure compared to Fe and
Cr Additionally the alteration of chromite in the serpentinite and the soils are analogous
to one another Chromite in the serpentinite retains a majority of its Cr compared to Mg
and Al when being altered to CSM by fluids at elevated pressures and temperatures
(Figure 26 and Table 22) Under ambient soil conditions chromite is once again
retaining Cr relative to Mg and Al causing what appears to be an increase in Cr
concentration Chromium is not easily released from chromite whether under high
pressures and temperatures or in the soil
Crystal field stabilization energies (CFSE) and octahedral site preference energy
parameters (OSPE) provide a means to explain the incongruent dissolution or Cr
preservation in the Cr-spinels CFSE is a measure of the net energy of stabilization
gained by a metal ions nonbonding d electrons as a result of complex formation This
measure also indirectly describes the energy required to remove an ion from a crystal
structure The octahedral CFSE for Cr(III) is ndash2247 kJmole and the OSPE (the
difference between the tetrahedral and octahedral CFSE) is ndash1578 kJmole (Dunitz and
Orgel 1957 McClure 1957) These values are the most negative for any transition
element and indicating the presence of Cr(III) in an oxide increases the crystalrsquos stability
compared to an analogous crystal lacking Cr(III) Chromium(III) would be among the
29
last ions to removed from the spinel structure compared to other transition elements such
as Fe2+and Mn2+
Soil Chemistry related to Chromium Availability
In Table 25 the chemical properties of the soil at Site JR3 do not appear to be a
factor when evaluating Cr in the soil The pH of the soil did not vary considerably with
depth Additionally the near neutral pH would not have a significant impact accelerating
Cr-bearing mineral dissolution Organic matter within the soil was elevated despite the
low plant productivity of the serpentine soil The role of organic matter could be a
significant factor such as the production of organic acids promoting the dissolution of the
Cr-bearing minerals however Cr concentrations near the surface (where organic matter
was the greatest) are not significantly greater or less than those deeper in the soil profile
The relative concentrations of ammonium acetate extractable Ca K Mg and Na are
characteristic of serpentine soils and did not provide any new insight Finally Cr(III)
was the only identifiable valence state of Cr based on X-ray absorption spectroscopy
(Figure 211) which is what should be expected based on the Cr-bearing minerals and the
pH of the soil
Extraction experiments summarized in Table 29 and Figure 213 demonstrate
~99 of the Cr in a serpentine soil is retained in highly resistant phases at Jasper Ridge
The chromite standard used in these experiments contributed negligible concentrations of
Cr (lt2 mg kg-1 total) for all of the sequential extraction experiments supporting earlier
evidence that Cr-spinels are retaining Cr and are potentially responsible for a majority of
the total Cr enrichment in the serpentine soil At Jasper Ridge Cr extracted from the
serpentine soil (31 mg kg-1) is likely the result of the Cr-silicates compared to the Cr-
spinels Low Cr-containing minerals (possibly chlorite enstatite augite and clay
minerals) appear to be the most probably source of Cr to soil solutions and vegetation of
the serpentine soil Approximately 30 of the serpentine soil is composed of clay-sized
minerals and this size-fraction has ~1000 Cr mg kg-1 The mineral responsible for Cr
enrichment in the clay-size fraction is more than likely Cr-chlorite identified using XRD
Chromium-chlorite in the soils suggests that this mineral is potentially the largest silicate
contributor of Cr in the serpentine soil
30
Interpreting the enrichment and behavior of Cr in serpentine soils such as in
Figure 27 can be accomplished based on the abundance Cr-spinels observed in the soil
Small grains of chromite Cr-magnetite and even CSM in the soil have the potential to
appreciably impact Cr concentrations Table 210 demonstrates the chromite masses
required to change the Cr concentration in a 1 kg sample of soil The highest Cr
concentration at Jasper Ridge was identified at Site 3 with a measured value of 4760 mg
kg-1 which means only ~1 of the soil sample (Table 210) needs to contain chromite
Additionally a minimal amount (lt011 of the sample as shown in Table 210) of
chromite is necessary to change the concentration by 500 mg kg-1 which could easily
cause the sharp Cr increases and decreases noted in Figure 27 This agrees with our
observations of chromite abundance in the serpentine soils Additionally a majority of
the chromite grains identified in the bedrock are between 2 and 002 mm which
correspond directly to the elevated Cr concentrations (gt10000 mg kg-1) in the sand-size
fraction of the serpentine soil (Table 26) Even the ~15000 Cr mg kg-1 values in soils
measured by Hotz (1964) in Figure 22 only require ~2 of the sample of the sample to
be chromite A few grains of chromite possess the ability to significantly impact
measured Cr concentrations in soil samples
Chromium and Nickel in Serpentine Soils
Chromium and Ni have often been correlated to one another due to their common
enrichment in serpentine soils (Hotz 1964 Rabenhorst et al 1982 Schwertmann and
Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al 1989 Kaupenjohann
and Wilcke 1995) A majority of the Ni in the Jasper Ridge serpentine soil originated
from olivine (Figure 25c and 25d Table 23) Chemical alteration of olivine during
serpentinization resulted in the mineral serpentine (Coleman and Jove 1992) in which
minor concentrations of Ni (lt037 NiO Wt ) substituted into its structure Continued
alteration under highly reducing conditions and the incompatibility of Ni in the serpentine
structure resulted in the formation of Ni-Fe alloy (Figure 25c) This type of alteration is
observed in many of the rocks at Jasper Ridge where the Ni-Fe alloys are ~10 microm in size
Olivine in the rocks and soil appear to be less resistant to weathering compared to
chromite therefore Ni is possibly more available to bacteria and vegetation in serpentine
31
soils Sequential extractions (Figure 213 Table 29) support these observations in which
a greater percentage of total Ni was extractable compared to the Cr in the soils Although
Cr and Ni are enriched in serpentine soils the availability and behavior of these two
elements are significantly different
CONCLUDING REMARKS
Chromium is predominantly present in primary chromite Cr-magnetite and a
chromite-silicate mixture (CSM) in both the bedrock and soil despite weathering over
geologic time Sharp variations in soil Cr concentrations with respect to depth are caused
by the presence or absence of chromite Cr-magnetite andor CSM Chromium-spinels in
the soil have undergone incongruent dissolution possibly conserving Cr(III) with respect
to Al3+ and Mg2+ Chromite is highly resistant to weathering and it is more than likely
not contributing bioavailable Cr Although Cr-spinels are a major source of Cr Cr-
silicates in the protolith (chlorite enstatite and augite) and soil (chlorite
montmorillonite and vermiculite) afford the most accessible Cr source identified Based
on our observations and analyses chlorite is the most likely Cr source of these silicate
minerals Finally Cr(VI) was not detected in any of the soil and rock samples
ACKNOWLEDGEMENTS
We thank Jasper Ridge Biological Preserve for allowing us the opportunity to
examine the serpentine soils Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund NSF Grant EAR-9902859 (DKB) GSA Graduate
Student Research Grants Office of Teaching and Learning (SF) at Stanford University is
gratefully acknowledged We appreciate the technical support from the staff of the
Stanford Synchrotron Radiation Laboratory from Bob Jones for assistance in microprobe
analyses of the serpentinites and serpentine soils and from Ben Bostick Travis Horton
and Theresa Barber for field and lab assistance
32
Table 21 Whole rock elemental concentrations of serpentinites (Figure 21b) chert and sandstone
-----------------------Location of Serpentinitesdagger----------------------- Elemental Concentrations JR1 JR3E JR6E Site 2 Site 5 Site 10 JR3
ChertDagger SandstoneDagger
Al 02 08 03 027 22 107 024 079 027 Ca 026 007 005 003 046 013 003 001 lt001 Co mg kg-1 101 99 100 99 155 217 97 1 lt1 Cr mg kg-1 580 2280 1510 1840 1945 2090 1465 10 lt1 Fe 303 71 609 581 76 1106 512 041 024 K 001 031 099 lt001 lt001 001 lt001 029 008
Mg 122 134 154 1336 524 804 112 014 034 Mn mg kg-1 495 800 155 730 1375 1860 460 110 85 Na 181 006 093 lt001 lt001 lt001 lt001 001 001 Ni mg kg-1 2037 1545 3321 3510 3020 3960 2740 20 9
Total digestion ICP-AES analyses daggerSerpentinites are dominantly composed of the minerals serpentine chlorite talc and spinels DaggerSee Figure 21b for sample locations
33
Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in serpentinite rock at Jasper Ridge
Chlorite Enstatite Augite Chromite Cr-Magnetite ChromiteSilicatedagger SiO2 3738 5463 5222 000 024 1053 Al2O3 552 477 516 5059 182 395 NiO 001 01 00 03 00 052 FeODagger 762 614 239 1446 7919 5042 MnO 009 015 009 042 140 182 Cr2O3 024 060 099 1584 1605 1939 MgO 3337 3305 1560 1869 024 605 CaO 023 050 2348 000 000 019 Na2O 000 001 050 000 001 008 K2O 000 000 001 000 000 011 Total 8446 9994 10048 10027 9894 9307
Formula Unit Compositions
Si 370 094 189 000 001 --- IVAl 030 006 011 000 000 --- VIAl 035 004 011 157 008 --- Ni 000 000 000 001 000 ---
Fe+2 063 010 007 030 232 --- Mn 001 000 000 001 004 --- Cr 002 001 003 033 048 --- Mg 493 085 084 073 001 --- Ca 002 001 091 000 000 --- Na 000 000 004 000 000 --- K 000 000 000 000 000 --- O 1400 300 600 400 400 ---
Electron microprobe analyses daggerChromitesilicate is also referred to as CSM DaggerFe+3 is present however all conversions are based on Fe+2
34
Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in serpentine soil at Jasper Ridge
CSM Chromite Cr-Magnetite SiO2 1940 057 267 Al2O3 559 1059 381 NiO 014 001 004 FeOdagger 2629 2259 5244 MnO 155 154 419 Cr2O3 1878 5577 3237 MgO 1668 916 288 CaO 015 001 005 Na2O 006 001 001 K2O 0 008 009 Total 8864 10033 9855
Si --- 002 010 IVAl --- 042 017 VIAl --- 000 000 Ni --- 000 000
Fe+2 --- 063 148 Mn --- 004 013 Cr --- 147 096 Mg --- 046 016 Ca --- 000 000 Na --- 000 000 K --- 000 000 O --- 400 400
Electron microprobe analyses dagger Fe+3 is present however all conversions are based on Fe+2
35
Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25 and 26 A B C D E F G H I J K L Chromite
Chromite CSM Magnetite Chlorite Olivine Serpentine Serpentine Ni-Fe-Metal Cr-Magnetite CSM Chlorite
SiO2 0 0 1053 0 3407 4046 3688 4074 591dagger 021 1143 2855Al2O3
5059 5162 395 034 1114 000 397 085 054 165 980 2293NiO 028 031 052 0 004 037 037 0 5302dagger 0 0 0
FeODagger 1446 1452 5042 9992
1220 901 262 453 1753dagger 8169 3498 806MnO 042 044 182 0 008 010 0 0 0 107 150 00Cr2O3 1584 1576 1939 125 071 001 0 0 0 1369 1905 039MgO 1869 1869 605 003 292 4895 3848 3755 064 1217 2828CaO 0 0 019 0 0 0 0 0 0 0 0Na2O 0 0 0 0 0 0 0 0 0 0 0K2O 0 0 0 0 0 0 0 0 0 0 0H2O --- --- --- --- 1200 --- 1600 1600 --- --- 1200Total 10027 10136 9307
10164 9944 9891 9834 9972
9897 8897 10028
Formula Unit CompositionSi
000 000 --- 000 328 100 184 200 001 --- 269IVAl 000 000 --- 000 072 000 000 000 000 --- 131VIAl 157 159 --- 002 058 000 023 005 007 --- 123Ni 001 001 --- 000 000 001 001 000 000 --- 000
Fe+2 030 029 --- 294 084 019 011 019 239 --- 059Mn 001 001 --- 000 000 000 000 000 003 --- 000Cr 033 032 --- 004 006 000 000 000 041 --- 003Mg 073 073 --- 000 419 180 285 274 004 --- 397Ca 000 000 --- 000 000 000 000 000 000 --- 000Na 000 000 --- 000 000 000 000 000 000 --- 000K 000 000 --- 000 000 000 000 000 000 --- 000O 400 000 --- 000 1400 400 700 700 400 --- 1400
36
Electron microprobe analyses dagger elemental metal Dagger Fe+3 is present however all conversions are based o
36
2321000---
10025
591000054
53021753000000
2321000000000000
n Fe+2
Table 24 Bulk soil elemental concentrations for soils and bedrock collected along the transect
(Fig 21b)
Sites Samples Depth
(meters) Cr
(mg kg-1) Ni
(mg kg-1) Mn
(mg kg-1) Co
(mg kg-1) Al
() Fe
() Ca ()
Mg ()
COI03 0015 2420 1590 1325 118 464 565 065 352 COI36 0045 2370 1630 1600 138 467 55 063 356 COI69 0075 2160 1755 1670 144 471 6 064 365 COI912 0105 2420 1785 1740 151 496 615 067 384
1
COI1215 0135 2570 1985 2340 195 506 649 069 411 COII02 001 3710 2400 1665 169 294 702 100 768 COII25 0035 3600 2290 1625 162 275 673 093 743 COII58 0065 2910 2310 1550 159 25 672 087 771
COII811 0095 3520 2350 1595 158 276 693 093 766 2
COIIBedrock 012 1840 3510 730 99 027 581 003 1336 COIII14 0025 3750 2930 1655 190 178 897 062 762 COIII47 0055 4760 3050 1850 178 178 94 056 805
COIII710 0085 4380 3410 1930 216 216 1044 07 797 3 COIII1013 0115 4320 3380 1955 201 213 1061 063 626 COIV01 0005 3710 2970 1915 214 209 909 053 781 4 COIV15 003 3540 2710 1725 195 181 801 043 792 COV14 0025 2620 2350 1330 150 246 651 063 526 COV47 0055 1775 2710 1520 165 252 766 059 517
COV710 0085 2000 3090 1830 195 225 737 057 583 5 COVBedrock 0115 1945 3020 1375 155 22 76 046 524
COVI14 0025 3040 3720 1855 224 084 1152 015 736 COVI47 0055 2400 3610 1785 218 085 1138 014 68 COVI710 0085 2960 3800 1850 232 093 1232 015 714 6
COVI1013 0115 2800 3860 1800 224 085 1164 014 739 COVII14 0025 2620 3550 1925 229 083 1128 015 718 COVII47 0055 1975 3590 1805 216 084 1085 014 655 7 COVII712 0095 2080 3600 1840 220 07 1105 011 741 COVIII14 0025 2130 3830 1990 238 085 115 017 705 COVIII47 0055 2550 3720 1865 228 085 1099 017 714 8
COVIII712 0095 2240 3910 1860 229 076 1112 014 828 COIX14 0025 2470 3480 1890 227 104 1201 023 579 COIX47 0055 3000 3830 1595 205 084 108 013 842 9 COIX713 01 2720 3710 1845 221 111 1208 017 551 COX14 0025 2810 3740 1760 218 129 114 017 675 COX47 0055 2190 3910 1835 224 131 1206 017 702
COX710 0085 1725 3370 1595 192 11 1061 014 635 10 COXBedrock 0115 2090 3960 1860 217 107 1106 013 804
Total digestion ICP-AES analyses
37
Table 25 Chemical and physical properties of the soil at Site JR3
NH4OAc-extractable ---Hydrometer---Ca K Mg Na
Sand Silt ClaySample Depth (cm) pH EC
dSm
Organic Matter ------
LECO Nitrogen ------ ---------mg kg-1---------
Texture
JR305 25 671 04 32 012 505 2366 3291 355 36 31 33 Clay LoamJR3515
10 683 09 19 064 4205 8176 3343 2392 36 30 34 Clay LoamJR31530 225 688 03 12 003 3725 4903 3750 2305 36 33 31 Clay LoamJR33045 375 698 07 10 003 2873 3665 3388 2549 46 24 30 Sandy Clay Loam
See Methods Section for more details
38
38
Table 26 Selected trace metal concentrations for each size fraction versus depth at Site JR3
Depth (cm) Total gt2mm Fraction Sand Silt Clay Sample
------------------------Chromium (mg kg-1)------------------------ JR305 25 4906 5086 gt10000 4565 988 JR3515 10 5976 4678 8284 4000 1115 JR31530 225 5222 4206 gt10000 4133 980 JR33045 375 3477 3297 gt10000 4571 956
------------------------Nickel (mg kg-1)------------------------------ JR305 25 1935 2050 1660 1305 2920 JR3515 10 2180 2050 1980 1325 2920 JR31530 225 2190 2120 1835 1370 2880 JR33045 375 2090 2140 1740 1285 2840
------------------------Manganese (mg kg-1)-------------------------- JR305 25 1319 1571 1728 1301 1210 JR3515 10 1664 1601 2701 1471 1254 JR31530 225 1639 1626 2217 1599 1231 JR33045 375 1222 1226 1781 1405 1170
-----------------------Cobalt (mg kg-1)-------------------------------- JR305 25 128 146 159 134 131 JR3515 10 155 145 241 147 132 JR31530 225 152 147 194 162 128 JR33045 375 125 124 159 141 122
Total digestion ICP-AES analyses
39
Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3
JR305CLAY Depth 0-5cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1439 100 1833 100 1274 100 1391 100 731 179 1432 431 732 645 1047 105 71 216 941 163 71 41 838 838 474 101 847 141 475 257 726 726 356 125 73 271 365 264 313 83 712 248 355 299
473 143 313 257 364 131 355 193
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1845 1000 1268 1000 1388 1000 852 68 1447 387 724 987 1010 859 733 264 944 143 703 679 836 281 712 247 903 139 471 282 723 576 475 109 850 83 362 244 365 73 733 426 352 321 356 117 712 291 313 154 314 63 475 122
364 170 355 196
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1724 1000 1297 1000 1396 1000 842 45 1397 447 741 957 1175 708 730 318 923 176 715 686 1037 944 710 255 840 118 477 271 837 139 475 97 717 465 367 400 726 375 357 133 468 182 357 386 313 91 362 194 315 286
352 229 JR33045CLAY Depth 30-45cm
Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity
1447 1000 1869 1000 1268 1000 1393 627 842 44 1476 287 737 843 1049 1000 730 207 954 107 715 539 837 190 710 182 915 107 479 157 723 111 475 69 860 61 366 303 364 69 740 332 356 292 356 74 717 256 315 315 313 57 478 111
367 141 357 115
See Methods Section for details
40
Table 28 X-ray diffraction d-spacings for common silicates from Whittig and Allardice (1986)
Diffraction Spacing (Aring) Mineral(s) Indicated Mg-saturated air-dried
14-15 Smectite vermiculite chlorite 99-101 Mica (illite) 71-73 Serpentine
715 Kaolinite chlorite Mg-saturated glycerol-solvated
177-180 Smectite 14-15 Vermiculite chlorite
99-101 Mica (illite) 71-73 Serpentinte
715 Kaolinite chlorite K-saturated air-dried
14-15 Chlorite vermiculite (with hydroxy interlayer) 124-128 Smectite 99-101 Mica vermiculite (contracted) 71-73 Serpentine
715 Kaolinite chlorite K-saturated heated (500ordmC)
14 Chlorite 99-101 Mica vermiculite (contracted) smectite (contracted) 71-73 Serpentine
715 Chlorite
41
Table 29 Results of the Diablo clay serpentine soil and chromite sequential extraction experiments
1 Exchangeable Fraction (1mM BaCl2) 4 Sum of 1 2 amp 3 (mg kg-1) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 0 0 003 Cr 159 388 19 Fe 014 011 001 Fe 24812 25532 1096 Ni 104 103 008 Ni 1854 1868 25 Mn 0 0 0 Mn 778 1172 04
2 Fe-Oxide Fraction (1M NH2OH HCl in 25 (vv) HOAc) 5 Total Cr Fe Ni and Mn (Total Digestion) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 19 78 05 Cr 2290 3310 1767864 Fe 4474 5138 38 Fe 60000 68000 1044325 Ni 1610 1506 01 Ni 1673 2345 00 Mn 231 430 00 Mn 1498 1608 3424547
3 Silicate Fraction (10M HF) 6 Percent Extractable ((45)100) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ------------------------------------- Cr 140 310 14 Cr 07 12 00 Fe 20337 20393 1058 Fe 41 38 01 Ni 234 352 22 Ni 111 80 00 Mn 547 742 04 Mn 52 73 00
ICP-AES analyses
42
Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1 kg of sample
Chromite
FeCr2O4 Cr Concentration
(mg kg-1) Grams Sample 500 11 011 1000 22 022 2000 43 043 5000 108 108
43
Figure Captions
Figure 21a Soils map of Jasper Ridge Biological Preserve (Stanford CA) with
complete descriptions of the soils shown in the legend 21b An enlarged image of the
study area shown in 22a indicates sampling sites of the soils and rocks collected A
dashed line marks the 500 m transect
Figure 22 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above
Figure 23 Petrographic images illustrating the varying serpentinite textures at Sites (A)
JR3 (B) JR5E and (C) JR3E For reference to site locations see Figure 21b
Figure 24 Correlation diagrams of Cr-bearing minerals in the serpentinite at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Si in Wt
Chromite (circle) Cr-magnetite (square) chromitesilicate or CSM (down triangle)
chlorite (hexagon) and pyroxenes (up triangle) analyses are shown on each correlation
diagram
Figure 25 Images of the serpentinite bedrock at Site JR3 demonstrating the mineralogy
and Cr-bearing phases Microprobe analyses are indicated by circled letters and these
values are in Table 23 25a Backscattering electron image demonstrating several Cr-
bearing phases present including chromite Cr-magnetite CSM and Cr-chlorite from
Figure 23a 25b Magnified image shown in Figure 25a demonstrating the fine-grained
texture of CSM Additionally Cr2O3 Wt values are shown for chromite the depletion
zone and CSM 25c Image of Ni-bearing phases present in the bedrock including
olivine serpentine and Ni-Fe alloy from Figure 23a 25d Cartoon illustrating the
mineralogical and chemical changes present in 25c
Figure 26 Images of a Cr-chlorite and Cr-magnetite assemblage in the serpentinite
bedrock 26a Petrographic image of the bedrock at Site 4 (Figure 21b) 26b
44
Backscattered electron image demonstrating the common occurrence of Cr-chlorite
encompassing chromite grains Microprobe analyses are indicated by circled letters and
are located in Table 23
Figure 27 Highest and lowest Cr concentrations (mg kg-1) at each site in the serpentine
soil 500 m transect shown in Figure 21b Samples included soils at various depths (m)
and bedrock from each site The average Cr concentration for the Jasper Ridge
serpentinite is marked on the central figure Chromium concentrations versus depth (m)
at each site (noted in the upper left corner) outline the main diagram
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected
for total gt2 mm sand silt and clay size fractions listed in Table 26
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soil at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Al+Fe in Wt for
chromite (circles) Cr-magnetite (hexagons) and CSM (diamonds) Chromite from the
bedrock is denoted by a larger circle and FeCr2O4 with a square
Figure 210 XRD clay analyses of sample JR3515 (Site JR3 Figure 21b) demonstrate
the influence of each treament Mg-saturated Mg+glycerol K-saturated and K-saturated
+ heat (500oC) The d-spacings and intensities are also listed in Table 27
Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at depths (a) 0-5cm (b) 5-15cm (c) 15-30cm (d) 30-45cm and (e) bedrock The pre-edge
area indicative of Cr(VI) peaks is shown on the right Note that no peaks are present
Figure 212 X-ray microprobe elemental maps of Cr and Fe for (212a) serpentine soil
at a depth of 0-5cm (212b) serpentine soil at a depth of 5-10cm and (212c) serpentinite
bedrock at Jasper Ridge site JR3 (Figure 21b)
45
Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments for the Diablo clay (Site 1 Figure 21b) serpentine soil (Site 2
Figure 21b) and chromite listed in Table 29 The sum refers to the summation of the
three extraction experiments
Figure 214 SEM images of the chromite standard (A) after each sequential extraction
treatment including the (B) exchangeable fraction (C) Fe-crystalline oxide fraction and
(D) silicate fraction The concentration of Cr removed by each treatment is shown
beneath each set of images
Figure 215 Tie-lines from Figure 29 connecting serpentinite chromite and end-
member FeCr2O4 are shown on this single plot Soil analyses from Figure 29 are also
plotted on this diagram For a sum of 100 Wt oxygen must be added to the sum of
Cr Al Fe and Mg Wt
46
Figure 21a
47
Figure 21b
48
Chromium Concentration (mg kg-1)1000 10000
Dep
th (m
eter
s) 01
10
Maryland (Rabenhorst et al 1982)California (Gough et al 1989)California (Hotz 1964)New Foundland (Roberts amp Proctor 1992)
Surface
Figure 22
49
Figure 23
50
Cr (Wt )0 5 10 15 20
Al (
Wt
)
0
5
10
15
20
25
30
Cr (Wt )0 5 10 15 20
Fe (W
t
)
0
20
40
60
80
100
Cr (Wt )0 5 10 15 20
Mg
(Wt
)
0
5
10
15
20
25
Cr (Wt )0 5 10 15 20
Si (W
t
)
0
5
10
15
20
25
30
Cr0 10 20 30 40 50 60 70 80 90 100
Al
0
10
20
30
40
50
60
70
80
90
100
Fe
0
10
20
30
40
50
60
70
80
90
100
Chromite
Chlorite
Magnetite
Chromite-Silicate
Pyroxenes
Chromite
Chlorite
Pyroxenes
Magnetite
Chromite-Silicate
Figure 24
51
Figure 25
52
Figure 26
53
Sites1 2 3 4 5 6 7 8 9 10
Con
cent
ratio
n (m
g kg
-1)
1500
2000
2500
3000
3500
4000
4500
5000Soil Chromium
1
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000002004006008010012014016
2
Cr Concentration (mg kg-1)
1500 2000 2500 3000 3500 4000
Dep
th (m
eter
s)
000002004006008010012014
3
Cr Concentration (mg kg-1)3500 4000 4500 5000
Dep
th (m
eter
s)
000
002
004
006
008
010
012
5
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000002004006008010012
6
Cr Concentration (mg kg-1)2200 2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000002004006008010012
7
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000
002
004
006
008
010
9
Cr Concentration (mg kg-1)
2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000
002
004
006
008
010
012
10
Cr Concentration (mg kg-1)1600 2000 2400 2800 3200
Dep
th (m
eter
s)
000002004006008010012
4
Cr Concentration (mg kg-1)3500 3600 3700 3800
Dep
th (m
eter
s)
000
001
002
003
004
8
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000
002
004
006
008
010
BEDROCK
BEDROCK
BEDROCK
Range of Cr Concentrations in Serpentinite Bedrock
Figure 27
54
Soil Chromium
Depth (cm)0 10 20 30 40 50
Cr m
g kg
-1
02000400060008000
100001200014000
Totalgt2mmSandSiltClay
Soil Nickel
Depth (cm)0 10 20 30 40 50
Ni m
g kg
-1
0
1000
2000
3000
4000
5000
Soil Manganese
Depth (cm)0 10 20 30 40 50
Mn
mg
kg-1
0
1000
2000
3000Totalgt2mmSandSiltClay
Soil Cobalt
Depth (cm)0 10 20 30 40 50
Co
mg
kg-1
0
50
100
150
200
250
300
Totalgt2mmSandSiltClay
Totalgt2mmSandSiltClay
Exceeds detection limits of 10000 mg kg-1 Cr
Figure 28
55
Cr (Wt )0 10 20 30 40 50
Al (W
t
)
0
5
10
15
20
25
30
Cr (Wt )0 10 20 30 40 50
Mg
(Wt
)
0
2
4
6
8
10
12
Cr (Wt )0 10 20 30 40 50
Si (W
t
)
0
2
4
6
8
10
12
14Cr (Wt )
0 10 20 30 40 50
Fe (W
t
)
0
10
20
30
40
50
60
70
Cr (Wt )0 10 20 30 40 50
Al +
Fe
(Wt
)
10
20
30
40
50
60
70
JRBPChromite
FeCr2O4
Chromite
Magnetite
Chromite-Silicate
Figure 29
56
Clay XRD of JR3515 Depth 5-15cm
0
200
400
600
800
1000
1200
0 5 10 15 20 25 30 35 40 45
d-spacing (Aring)
Rel
ativ
e In
tens
ity
Mg-Saturated
Mg+Glycerol
K-saturated
K-saturated+heat (500oC)
Figure 210
57
Energy (kev)
597 598 599 600 601 602 603
Nor
mal
ized
Abs
orba
nce
-4
-3
-2
-1
0
1
2
(a)
(b)
(c)
(d)
(e)
Cr(
VI)
Figure 211
58
Figure 212
59
Diablo Clay
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Serpentine Soil
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Chromite
0
20
40
60
80
100
120
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide
SilicateSum
Figure 213
60
Figure 214
61
Reaction Progress ()0 20 40 60 80 100
Elem
enta
l Wt
0
10
20
30
40
50
ChromiumAluminumIronMagnesium
FeCr2O4Jasper Ridge
Bedrock Chromite
SoilChromite
Figure 215
62
CHAPTER 3
OXIDATIVE PROMOTED DISSOLUTION OF CHROMITE BY MANGANESE
DIOXIDE AND CONCURRENT PRODUCTION OF CHROMATE
ABSTRACT
Chromium release and oxidation from chromite is a potential environmental hazard in
sediments and soils and a pathway for soil development (weathering of primary minerals)
related to ultramafic rocks and their metamorphic derivatives (serpentinites) Birnessite is
a common pedologic mineral in these sediments and soils capable of oxidizing aqueous
Cr(III) In this study the interaction between chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4)
and birnessite is investigated with an interest in the potential generation of Cr(VI)
Specifically the effects of chromite suspension density birnessite suspension density
and pH at a temperature of ~25degC in relation to the reaction kinetics were examined The
rate of Cr(VI) in solution increases with increasing chromite suspension densities and
decreasing pH but is independent of birnessite suspension densities at values greater than
~20 m2 L-1 The overall rate expression of Cr released and oxidized in solution from
chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression with Cr(VI) rates ranging from 9times10-4 to 44times10-3 microM h-1
Additionally these experiments demonstrate that serpentine soils are a source of non-
anthropogenic Cr(VI)
63
INTRODUCTION
Chromite a Cr-enriched spinel is responsible for a majority of the Cr enrichment
identified in ultramafic rocks serpentinites and serpentine soils (Rabenhorst et al 1982
Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989 Gasser and
Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003 Oze et al
2003a Oze et al 2003b) The chemical weathering of chromite is therefore an important
process when evaluating the potential of ultramafic rocks and their related sediments and
soils as sources of Cr Despite the propensity of chromite to resist chemical weathering
the dissolution of chromite and release of Cr will occur over geologic time (Bialowolska
and Salacinski 1984 Oze et al 2003b) Chromium in chromite is in the form of Cr(III)
the non-toxic and chemically immobile valence state however oxidizing conditions in
near surface environments is a potential reaction leading to the enhanced dissolution of
chromite and the formation of Cr(VI) the toxic and chemically mobile valence state
which typically exists as the oxyanion chromate (HxCrO42-x) (Daugherty 1992 Cohen et
al 1993 Fendorf 1995 Deer et al 1996) Consequently chromite may be a non-
anthropogenic source of Cr(VI) in the environment
Naturally occurring high valence Mn oxides present in soils and sediments
derived from ultramafic rocks are strong oxidizers capable of affecting the chemical
mobility of metals through redox reactions (Fendorf and Zasoski 1991 Johnson and
Xyla 1991) These insoluble mixed valent oxides (Mn(IIIIV)) are observed coating
mineral surfaces in these ultramafic sediments and soils where total Mn concentrations
are often gt100 mg kg-1 (Brooks 1987 Schreier et al 1987 Oze et al 2003a Oze et al
2003b Oze et al) Birnessite is a naturally occurring mixed valence oxide known to
oxidize aqueous Cr(III) to the more chemically mobile Cr(VI) (Eary and Rai 1987
Fendorf and Zasoski 1991) Gough et al (1989) and Becquer et al (2003) have
suggested that Mn oxides are responsible for Cr(VI) identified in California and New
Caledonia serpentine soil solutions however the dissolution and oxidation of Cr from a
Cr-bearing mineral due to the presence of birnessite has not been quantitatively defined
In this study we examine the kinetics of Cr release and oxidation when chromite and
birnessite are present in solution
64
The objective of this study is to determine a rate expression describing the
release and oxidation of Cr from chromite in the presence of birnessite Using well-
controlled batch experiments the solution chemistry and geochemistry of Cr related to
chromite and birnessite interaction was evaluated temporally over a range of pH
Conditions were chosen to simulate the slightly acidic conditions characteristic of most
serpentine soils (Rabenhorst et al 1982 Gough et al 1989 Graham et al 1990 Cole
1992 Jeffrey 1992 Lee 1992 Verger 1992 Becquer et al 2003) Ultimately these
experiments allow the examination of one potential pathway leading to Cr(VI) identified
in solutions related to ultramafic rocks and serpentine soils
MATERIALS AND METHODS
Solid Phase Characterization and Synthesis
Chromite used in the experiments was commercial-grade chromite (MSDS No
302-1 WHMIS CLASS D2A) obtained from Barnes Environmental Inc Ontario
Chromite was prepared by cleaning the sand-size material in multiple ultrasonic
isopropanol baths and rinses The chromite was then sieved using a Retsch magnetic
sieve shaker for 30 minutes in order to obtain the 160-250 microm size fraction Impurities in
the 160-250 microm chromite were removed using a slope Franz magnetic separator These
impurities included the minerals olivine augite plagioclase feldspar and quartz Visual
and scanning electron microprobe (SEM) examinations of the remaining 160-250 microm
fraction confirmed its purity Chromite was crushed to a fine powder using an agate
mortar and pestle and then rinsed multiple times in deionized destilled (DDI) water and
001 M HCl solution to remove the ultra-fine material and eliminate highly reactive sites
caused during grinding The chromite powder was inspected before (Figure 31) and
after the experiments using a scanning electron microscope (JEOL 5600LV) the surface
area was 025 m2 g-1 as determined using N2(g) and BET isotherm
An automated JEOL 733A electron microprobe operated at 15 kV accelerating
potential and 15 nA beam current was used to investigate possible zoning and to quantify
the chemical composition of the chromite grains prior to powdering No zoning was
observed and the chemical composition of the chromite was based on the average of 25
electron microprobe analyses The stoichiometric composition of chromite based on 4
65
oxygens is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 X-ray diffraction was also used to
deduce the purity of the chromite powder X-ray diffraction spectra obtained for the
chromite powder matched JCPDS card 03-0873 (unit cell a=8300 b=8300 c=8300
α=90 β=90 and γ=90) and no contaminants were evident
Birnessite was synthesized using the procedure described by Buser et al (1954)
and implemented by Fendorf and Zasoski (1992) A solution of 63 g of potassium
permanganate (KMnO4) in 1 L of DDI water was heated to 90 degC Approximately 100
mL of the heated slurry was poured into a 4 L beaker and reheated to 90degC while being
stirred vigorously The remaining 900 mL of slurry and 66 mL of concentrated
hydrochloric acid were simultaneously and proportionally dispensed into the beaker over
a period of thirty seconds The temperature of the resulting solution was maintained at
90deg C for an additional 10 minutes and then allowed to cool to room temperature The
product was washed with alternate rinses of DDI water and a 001 M HCl solution The
birnessite surface area was 125 m2 g-1 based on BET analyses Solids were inspected
before (Figure 31) and after the experiments using the SEM
Extraction Experiment Procedures
Several chemical extraction experiments were performed on 106-250 microm
chromite grains prior to the chromite-birnessite batch experiments in order to determine
the potential release of Cr Fe and Mn One gram of chromite and 25 mL of solution
were placed into a 25 mL centrifuge tube and the solids and solution were gently agitated
The solutions and parameters used for these extraction experiments include the
following 1) 10 hydrogen peroxide (H2O2) solution at 25degC and pH of ~5 for 6 h 2)
01 M potassium permanganate (KMnO4) solution at 60degC and pH of ~3 for 6 h 3) 10 M
hydrofluoric (HF) acid at 60degC for 6 h 4) 1 M hydroxylamine (NH2OHmiddotHCl) with 25
(vv) acetic acid (HOAc) at 60degC and pH of ~4 for 6 h and 5) 1 mM barium chloride
(BaCl2) solution at 60degC and pH of ~6 for 6 h Solutions were extracted with a syringe
and passed through a 02 microm filter Solids were inspected after the experiments using the
SEM
66
Batch Experiment Procedures
Batch experiments were performed in triplicate using 1 L polypropylene bottles
and chromite and birnessite ratios listed in Table 31 at ~25degC After loading chromite
and birnessite into each bottle 1 L of a pH 3 solution a 10 mM acetate solution (pH 5) a
10 mM MES solution (pH 67) or a pH 8 solution was added to the appropriate
experiment (Table 31) Chromite and birnessite were also run individually in the acetate
buffered solution (Table 31) Aliquots were extracted from the polypropylene bottles
using a syringe and then passed through a 2 microm filter before analyses Less than 40 mL
was extracted over the course of each experiment resulting in a total volume change of
lt4 After the experiments were terminated the remaining solids were resuspended in
20 mL of 1 mM phosphate (5050 KH2PO4K2HPO4 pH 7) 20 mL of 10 mM CaCl2
solution at pH 7 and a 05 M HCl solution These suspensions were agitated for 3 h
followed by centrifugation and filtration (02 microm) before analyses The purpose of the
phosphate and CaCl2 solutions was to desorb reaction produtcts (eg Cr(VI) Cr(III) Fe
and Mn) from the solid phases The HCl treatment provides a means to evaluate
secondary phases such (oxy)hydroxides formed during the experiments Although the
HCl solution may dissolve chromite and birnessite in addition to the (oxy)hydroxides the
short time frame of the extraction (3 h) should limit the dissolution of chromite
Additionally the polypropylene bottles following the termination of five batch
experiments were rinsed with DDI water and 20 mL of 05 M HCl was added to each
bottle The HCl solution was allowed to remain in the bottles for 15 h in order to
determine the concentration of Cr Fe and Mn sorbed onto the interior lining
Serpentine Soil and Birnessite Experiments
Serpentine soils collected from Jasper Ridge Biological Preserve (Stanford CA)
and New Caledonia were used to determine the influence of birnessite in soils known to
contain chromite (Chapters 2 and 4) The Cr geochemistry of these soils as well as the
detrital chromite have been evaluated in Chapters 2 and 4 and by Becquer et al (2003)
Jasper Ridge serpentine soils contain Cr-bearing minerals including chromite Cr-
magnetite and Cr-chlorite however a majority of the Cr enrichment (Table 32) is
directly related to the abundance of chromite A majority of the Cr related to New
67
Caledonia serpentine soils is attributed to Cr (oxy)hydroxides (Chapter 4) however
chromite is present in these soils as well (Becquer et al 2003) Although birnessite has
been noted in these soils the quantity of birnessite could not be determined In order to
evaluate the impact of Mn oxides on Cr within such soils 01 g of birnessite was added to
each soil
The parameters for the experiments are listed in Table 32 Ten grams of the
lt250 microm size-fraction of each soil (JR1 JR2 and NC) and 100 mL of 1 mM acetate
buffered solution (pH 5) were loaded into individual 150 mL Nalgene bottles Aqueous
solutions were evaluated as a function of time (h) Additionally 10 g of each soil 01 g
of birnessite and 100 mL of 1 mM acetate buffered solution (pH 5) were loaded into 150
mL Nalgene bottles and were allowed to react as a function of time (h) These
experiments were rotated at a rate of 2 revolutions min-1 Solutions were removed with a
syringe and passed through a 2 microm filter before analysis Approximately 8 mL of
solution was required for the analyses over the course of each experiment resulting in a
total volume change of 8 Total Cr concentrations (mg kg-1) of the soils listed in Table
32 were obtained by completely dissolving the samples using a mixture of hot (gt200degC)
concentrated nitric perchloric and hydrofluoric acids Elemental compositions were
measured in the supernatant using inductively coupled plasma optical emission
spectroscopy (ICP-OES) The mass of chromite in each soil was calculated using total Cr
concentration for each soil and the calculations provided in Chapters 2 (Table 210) and
Chapter 4 (Table 47) The chromite surface area of each soil was approximated by
comparing the surface area measured for the chromite standard and its observed grain
size from SEM images (Figure 31) to chromites observed in the soils using backscattered
electron images (Chapters 2 and 4) The surface area of the Jasper Ridge chromite is 066
of the chromite standard due to a slightly larger grain size The surface area of the
New Caledonia chromite is approximately 025 of the chromite standard Chromite
suspension densities were calculated by multiplying the chromite mass by the estimated
surface area and then dividing by the volume of the solution
68
Solution Analysis
The pH and Eh of the supernatant was measured with an Orion flat surface
combination pH electrode inserted directly into supernatant Production of soluble
Cr(VI) was measured spectrophotometrically at 540 nm using the s-diphenyl carbazide
method modified from Bartlett and James (1979) Soluble Fe(II) was monitored using
the ferrozine method adopted from Stookey (1970) Total dissolved Cr Fe Al Mn and
Mg were ascertained using ICP-OES Concentrated trace metal grade nitric acid used to
acidify the samples prior to analyses Aluminum and Mg was determined to be present in
the nitric acid at concentrations lt1 microM Magnesium and Al are not plotted in any of the
figures Chromium(III) and Fe(III) was determined by subtracting Cr(VI) and Fe(II)
obtained from colorimetric methods from total dissolved Cr and Fe analyses using ICP-
AES Results are presented as an average of the three runs for each batch experiment
with the standard deviation noted by error bars
RESULTS
Chromite Extraction Experiments
The release of Cr Fe and Mn from chromite using 10 H2O2 01 M KMnO4 10
M HF acid 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 1 mM BaCl2 solutions was
carried out to determine the relative propensity of chromite to resist chemical dissolution
and these results are summarized in Figure 32 The KMnO4 solution with aqueous
Mn(VII) was the most effective in releasing Cr from chromite yielding a concentration
of 62 Cr mg L-1 Iron released using the KMnO4 solution was 03 mg L-1 Although
H2O2 is a direct chemical oxidant the release of Cr from chromite was less as compared
to the other solutions (only ~03 Cr mg L-1 was measured) Concentrated HF was only
able to release ~13 Cr mg L-1 while releasing 106 Fe mg L-1 NH2OHmiddotHCl with 25
(vv) HOAc and CaCl2 solutions only removed lt1 Cr mg L-1 however the NH2OHmiddotHCl
removed 36 Fe mg L-1 which is comparatively higher than the H2O2 KMnO4 and BaCl2
solutions
69
Chromite and Birnessite Batch Experiments
The interaction of chromite and birnessite leading to the release and oxidation of
Cr(III) is a relatively slow process requiring study on the order of hours and days The
pH was maintained constant throughout the duration of the experiments while Eh values
ranged from 320 to 450 mV with no discernable temporal pattern (Figure 33) For pH
values between 3 and 67 Cr3+ CrOH2+ and Cr(OH)2+ are the dominant aqueous Cr(III)
species however at pH 8 conditions are on the stability line between Cr(III) and Cr(VI)
specifically Cr(OH)3 and CrO42-
Higher chromite suspension densities produced higher rates of Cr release to the
aqueous phase and all the Cr in solution was in the form of Cr(VI) (Figure 35) The
rates of oxidation based on Cr(VI) production range from 00005 to 00041 microM h-1
(Figure 35 Table 33) Comparatively Cr released from the chromite control experiment
is maintained at a relatively constant value of ~005 microM and no Cr(VI) is measured in
solution Iron concentrations initially increase to ~ 05 microM and then decrease rapidly to a
relatively constant concentration at 02 microM (Figure 35) Iron concentrations for the
chromite control experiment began at 15 microM between 0 to 50 h and decreased to ~02
microM Iron(II) and Mn concentrations are below detectable limits for all experiments
(Figure 35) A zeroth-order dependence is assumed with respect to birnessite owing to
its large excess surface area relative to chromite
Variable birnessite suspension densities and constant chromite suspension
densities were evaluated at pH 5 (Figure 36) Although suspension densities of
birnessite are 538 and 1075 m2 L-1 (Table 31) the rate of Cr and Cr(VI) measured in
solution for both experiments is 00041 microM h-1 (Figure 36 and Table 33) Iron
concentrations are below 06 microM and maintain concentrations near 02 microM with no
indication of consistently increasing or decreasing with respect to time Iron(II) and Mn
concentrations in solution are below detectable limits for both experiments
In Figure 37 the influence of pH on chromite oxidation by birnessite was
evaluated by conducting experiments at pH values of 3 5 67 and 8 chromite and
birnessite suspension densities are constant at 0075 m2 L-1 and 1075 m2 L-1 respectively
(Table 31) Chromium and Cr(VI) rates increase with decreasing pH ranging from 0
microM h-1 at pH 8 to 00062 microM h-1 at pH 3 (Table 33) Iron is present in solution at
70
concentrations less than 04 microM for all the experiments Generally Fe concentrations are
greater near the start of the experiment and decrease to values lt 05 microM with increasing
reaction time Manganese concentrations are below detectable limits
In order to determine the rate dependence on each prominent reactant (chromite
birnessite and pH) reaction rates were regressed against reactant concentrations (Figure
38) A pseudo first-order rate dependence is noted based on chromite suspension
density the equation regressed for this relationship is listed in Table 33 As the chromite
suspension density increases the rate of Cr(VI) in solution increases In contrast a
zeroth-order rate dependence is evident for birnessite suspension density (dashed line in
Figure 38b) The rate of Cr(VI) generation is expected to decrease with respect to
birnessite suspension density at very low concentrations (dotted line in Figure 38b)
The influence of pH on the rate of Cr(VI) production is well-described by a polynomial
function (Figure 38c Table 33)
Developing an overall rate expression for Cr in solution requires integrating the
factors assessed in Figure 38 and listed in Table 33 All the Cr measured in solution is
in form of Cr(VI) therefore the rate law for Cr or Cr(VI) in solution may be written as
d[Cr(VI))]dt = k sdotChromitea sdot Birnessiteb sdot (aH+)c (31)
and
log d[Cr(VI))]dt = log k + asdotlogChromite+ bsdot log Birnessite+ csdotlog (aH+) (32)
where k is the overall rate constant for Cr release and oxidation Chromite is the
chromite suspension density (Table 31) Birnessite is the birnessite suspension density
(Table 31) and aH+ is the activity of H+ Constants a b and c are the order of
dependency for each reactant The overall rate equation (developed in the Appendix) can
be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069 (33)
71
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 the expression is valid for birnessite suspension densities
greater than 20 m2 L-1 and pH values between 3 to 8
Secondary and Solid Phase Examination of Chromite and Birnessite
The solid products were evaluated to determine the abundance and type of
secondary phases formed during the course of the experiments Solids from the batch
experiment were treated with three solutions 1 mM 5050 KH2PO4K2HPO4 1 mM
CaCl2 and 05 M HCl (Table 34) In the chromite and birnessite reactions Cr was
primarily in solution (gt90 ) The HCl extract was able to remove lt5 additional Cr
while the KH2PO4K2HPO4 and CaCl2 extracts released lt3 additional Cr Aqueous Fe
concentrations were below detectable limits for the final solution extraction Treatment
with the HCl resulted in up to 0127 micromoles of Fe however 0107 micromoles of Fe were
released upon treatment of unreacted chromite with HCl Although not shown in Table
34 Mn values were greater than gt4 micromoles when treated with HCl the
KH2PO4K2HPO4 and CaCl2 removed no detectable Mn
After the termination of each experiment the solids were examined using SEM in
order to identify changes in the surface morphology and secondary phases A large
chromite grain and conglomerate of birnessite adjacent to the chromite are depicted in
Figure 38 The face and edge of the chromite grain is not noticeably altered compared to
grains examined prior the experiments Additionally the particle morphology of
birnessite has not visibly changed Even at higher magnification angular and
conchoidally fractured fine-grained edges are noted (Figure 38bc) Only a slight
rounding of the smaller features of the main chromite grain provides any evidence of
possible chromite dissolution (Figure 38c) Birnessite appears to have limited interaction
with chromite Although chromite and birnessite were allowed to interact over 500 h in
an aqueous solution no secondary phases are observed by SEM
Serpentine Soil and Birnessite Experiments
Two serpentine soils from Jasper Ridge Biological Preserve CA and one
serpentine soil from New Caledonia were used to determine whether or not the overall
72
rate expression (Equation 33) for chromite reaction with birnessite is consistent with
Cr(VI) concentrations produced by serpentine soils containing chromite Three
serpentine soils without birnessite and three soils with the addition of birnessite buffered
at pH 5 were measured for Cr(VI) generation as a function time (Figure 310) The rates
of Cr(VI) measured in solution for both Jasper Ridge soils (JR1 and JR2) are identical at
00013 microM h-1 With the addition of birnessite the rates of Cr(VI) production in solution
both increased to 00044 microM h-1 (JR1 w Bt and JR2 wBt in Figure 310) Based on the
chromite suspension density (Table 32) the rates predicted by Equation 33 range from
00035 to 00041 microM h-1 which are 10 to 20 less than those observed in the
experiments (Figure 310) The rates of Cr(VI) in solution from the New Caledonia
serpentine soil are 00009 microM h-1 and with the addition of birnessite increase up to 00013
microM h-1 (Figure 310) A rate of 00016 microM h-1 was predicted by Equation 33 for the
New Caledonia soil
DISCUSSION
The reaction of birnessite with chromite enhances the rate of Cr release into
solution The rate of Cr(VI) measured in solution (Equation 33) is directly related to the
chromite suspension density and is independent of the birnessite at suspension densities
greater than ~20 m2 L-1 Although the release and oxidation of Cr in solution are
measured the mechanisms and pathways driving this reaction need to be evaluated The
propensity of chromite to resist dissolution the effects of pH the formation of secondary
phases and ultimately the pathways of Cr release and oxidation are addressed in the
following discussion
Extraction experiments performed on chromite (Figure 32) demonstrate it is
resistant to acidic dissolution or competitive ion displacement but strong oxidants such
as high valent Mn may oxidatively dissolve chromite It is evident that the proportions of
Cr Fe and Mn vary for each of the extractions The solutions utilized in the extraction
experiments have varying chemical properties capable of preferentially releasing a
particular species andor element These experiments were also performed over a short
period of time (lt6 h) which may have only released elements from highly reactive sites
73
Finally secondary phases associated with these experiments were possible and would
distort the elemental ratios observed in solution
A majority of the Eh and pH analyses plot in the Cr(III) stability region (Figure
33) however Cr in solution is primarily in the form of Cr(VI) (Figures 35 36 and
37) Eh measurements presented in Figures 32 and 33 do not account for the
microenvironments associated with the surfaces of birnessite Additionally the kinetics
of Cr(VI) reduction are limited in reducing conditions due to electron symmetry
constraints and the absence of a suitable redox couple (Fendorf 1995) Finally highly
oxidizing birnessite in the solution is kinetically and thermodynamically controlling the
oxidation state of Cr relative to the stabilities predicted in Figure 33 (Amacher and
Baker 1982 Fendorf and Zasoski 1992)
A fractional-order dependence on the chromite suspension density (Figure 38)
can be explained by diffusional influences competing oxidation reactions and secondary
reactions involving Cr(III) and Mn(IV) Secondary phases on the surface of chromite
may inhibit the diffusion of Cr from chromite however the clean and relatively unaltered
chromite surfaces do not support this mechanism as a major cause (Figure 38)
Chromium and Fe released from chromite may be incorporated into secondary phases
related to birnessite or chromite Approximately 10 of Cr is not in the aqueous phase
(Table 34) supporting the premise that secondary phases are contributing to the
fractional-order dependence related to the chromite suspension density Also
competition from Fe(II) oxidation on birnessite deposition of Fe(II) on birnessite (or
chromite) and diffusion onwithin birnessite may further impact the rate dependence
The zeroth-order dependence based on the birnessite suspension density (Figure
38) results from its high surface area At very low birnessite suspension densities the
rate of Cr(VI) production should decrease however due to the high surface area of
birnessite it was experimentally prohibitive to determine reaction rates at masses less
than 4 mg Therefore a hypothetical trend-line depicts the expected dependence on
birnessite concentration at very low suspension densities at higher values representative
of most field settings the reaction is independent of the birnessite surface area (ie zero-
order dependence)
74
In Figure 38c the polynomial relationship defines the Cr(VI) rate dependence on
pH As the pH decreases the rate of Cr(VI) increases more rapidly The relationship is
potentially the result of lower pHs progressively increasing the rate of chromite
dissolution However the rate of Cr(VI) in solution might be influenced by higher pHs
increasing the formation of secondary Fe(III) and Cr(III) (oxy)hydroxides on the
chromite andor birnessite surfaces thereby decreasing elemental diffusion from
chromite or decreasing available birnessite sites for Cr(III) oxidation Differentiating
between these two mechanisms is not possible based on the rate equation or extraction
analyses however the polynomial relationship evident in Figure 38c is folded into the
rate constant in order to geochemically account for the influence of pH on Cr(VI)
generation
No conclusive evidence is available to determine the fate of Mn within the
reacting suspensions Aqueous Mn concentrations are consistently maintained below
detectable limits If Mn is released from birnessite then it must be strongly and quickly
incorporated into a nonaqueous phase following its release Due to Cr(III) being oxidized
by Mn(III) andor Mn(IV) Mn(II) must be produced Extractions performed on the
chromite and birnessite solids demonstrate that Mn could only be removed with HCl
however a majority of this Mn is potentially the result of HCl dissolving birnessite On
the basis of electron transfer stoichiometry at least 15 mols of Mn(II) must be produced
for each mol of Cr(VI) generated This amount can be accounted for in the HCl
extraction where gt4 micromoles of Mn were measured Additionally Fendorf et al (1993)
determined that Mn(II) should not significantly influence the extent of Cr(III) oxidation
Iron in solution does not progressively increase in solution over time (Figures 35
36 and 37) The high initial rate of Fe release compared to Cr may be the result highly
reactive sites preferentially releasing Fe Although no Fe is detected in solution from
the extractions performed on the chromite and birnessite solids 0005 to 0035 moles of
Fe were retrieved from the lining of the bottle and 0053 to 0126 moles were released
from the solids using the HCl solution (Table 34) Iron(II) could potentially be oxidized
to Fe(II) in the presence of birnessite and form stable Fe(III)-(oxy)hydroxides (Postma
1985 Myers and Nealson 1988 Villinski et al 2001) A ratio of two Fe for every one
Mn is necessary for the oxidation of Fe(II) and this amount is recoverable based on HCl
75
extractions with gt4 micromoles of Mn measured from the HCl extractions Additionally
dissolved O2 in the batch experiments will rapidly oxidize Fe(II) for all the pHs examined
in these experiments (Stumm and Morgan 1981 Langmuir 1997) The oxidative
pathway for Fe(II) provides an explanation for the lack of Fe(II) in solution for the batch
experiments (Figure 35 36 and 37)
With a detection limit of ~02 microM and equilibrium with respect to Fe(OH)3(am) and
goethite Fe concentrations should either be at or below detection limits at pHs gt 5
Assuming aqueous Fe(III) is present dominantly as the second hydrolysis product at pH
5 the dissolved concentrations approach the solubility of ferrihydrite and are much
greater than expected for goethite (Figure 311) Thus it is probably that the former solid
is generated which is further supported by the chemical extraction experiments (Table
34)
Iron and Cr solution concentrations measured experimentally are not consistent
with congruent dissolution (Figure 312a) Assuming Fe was forming secondary phases
at a constant rate (~00001 Fe h-1) multiplying this rate by the time analysis and then
adding it to the measured aqueous Fe a corrected Fe (Fe) value can be estimated (Figure
312) Additionally the 10 of Cr not in solution can be also be used to calculate a
corrected Cr value (Cr) (Figure 312) When secondary phases are taken into account a
majority of the analyses plot near the congruent dissolution of chromite This provides
evidence that chromite is undergoing congruent dissolution within the batch experiments
but that secondary Fe(III) phases disrupt the solution phase elemental ratios
Chromium(VI) Pathways from Chromite and Birnessite into Solution
Three pathways are evaluated for Cr(VI) measured in solution a) chromite
dissolution b) birnessite dissolution and c) solid-solution reaction (Figure 313) In order
to describe these pathways end-member chromite (FeCr2O4) and birnessite (δ-MnO2) are
used to simplify the analysis It should be recalled however that the composition of the
chromite used in the experiments is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4
Chromite dissolution presents a likely explanation for Cr(VI) observed in solution
based on the following discussion In chromite dissolution model Cr(III) released from
chromite adsorbs onto the surface of birnessite and oxidizes to Cr(VI) After the electron
76
transfer is complete Cr(VI) desorbs from birnessite and remains in solution As this
process continues Cr(VI) increases at a constant rate The rate limiting step with regards
to this specific pathway can be grouped into two steps a) dissolution of chromite (Step
1a) or b) Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) (Figure 313)
Although a dissolution rate for chromite at ambient temperatures and pressures is not
known the chromite extraction experiments (Figure 32) demonstrates that chromite
dissolution is slow Manganese oxides such as birnessite oxidize Cr(III) in solution
rapidly with a measured rate constant of 6300 microM-1 s-1 in fresh water (100 mg L-1 MnO2
100 g L-1 Cr(III) pH~6) (Van der Weijden and Reith 1982) This means that the rate
for Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) is on the order of
seconds therefore chromite dissolution is likely the rate limiting step controlling the
concentration of Cr(VI) in solution for this pathway
Although a zeroth-order dependence related to birnessite concentrations (Figure
37a) is observed the birnessite dissolution pathway in which aqueous Mn4+ released
from birnessite adsorbs onto the surface of chromite oxidizing Cr(III) cannot be entirely
dismissed Manganese dissolution and adsorption could occur quickly not allowing the
Mn to be measured in solution Examination of the solids after the termination of each
experiment demonstrated extremely minimal chromite and birnessite interaction
therefore the solid solution reaction is not a likely pathway leading to the production of
Cr(VI) in solution and is furthermore inconsistent with the noted rate dependencies
Despite the three pathways (Figure 313) the generation of Cr(VI) observed in these
experiments and described by the rate equation (Equation 33) is independent of these
potential pathways
Serpentine Soils and Cr(VI)
The serpentine soil experiments (Figure 310) provide direct evidence that Cr(VI)
is forming within these soils By adding birnessite to the soil the rate of Cr(VI) in
solution increased for all experiments (00013 to 00044 microM Cr(VI) h-1 for Jasper Ridge
and 00009 to 00013 microM Cr(VI) h-1 for New Caledonia) The Cr(VI) rate increase
related to the addition of birnessite is likely the result of low birnessite suspension
densities (lt20 m2 L-1) in the soil Rates predicted for Jasper Ridge and New Caledonia
77
soils using Equation 33 and values in Table 32 are similar to those measured in the
experiments with values of 00041 and 00016 microM Cr(VI) h-1 respectively Overall these
soil experiments demonstrate that a rate equation developed from laboratory batch
experiments predicts the rate of Cr(VI) in serpentine soils reasonably well
According to limits set by the Environmental Protection Agency the California
standard for total Cr in drinking water is 50 microg L-1 (~1 microM) while the federal standard is
100 microg L-1 (~2 microM) The serpentine soil and birnessite experiments produced Cr(VI)
concentrations in excess of 50 microg L-1 after only 140 h or roughly 6 days Serpentine soils
at Jasper Ridge typically remain water saturated for prolonged periods of time (gt1 week)
during the winter season increasing the time Cr(VI) can be produced The degree and
duration of water saturation is dependent on the topography of the serpentinite body
level areas will remain water saturated while sloping areas will allow the water to flow
down gradient Although Cr(VI) was generated in the soil experiments the production of
Cr(VI) in serpentine soils may be more limited The reduction of Cr(VI) to Cr(III) is
favorable by reactions with Fe(II)aq Fe(II)-containing minerals such as magnetite H2S
and organic matter including microbial processes (Ilton and Veblen 1994 Fendorf 1995
Peterson 1996) Both organic matter and magnetite are abundant in Jasper Ridge
serpentine soils (Chapters 2 and 4) however these factors did not deter Cr(VI) from
forming during the serpentine soil experiments Additionally the soils in these
experiments were constantly rotated allowing the mineral surfaces to be continuously
exposed to solution Chromite within an undisturbed soil will not always be in contact
with the solution or proximal to birnessite
Chromium(VI) produced by the interaction of chromite and birnessite is not
limited to serpentine soils Serpentinites the bedrock for serpentine soils are typically
highly sheared due to the diapiric migration of serpentinite masses along zones of
weakness in fault zones Numerous cracks and lineations which are typical of these rocks
provide a means for Cr(VI) to transported away from the serpentine soil potentially into
local groundwater In fact preferential flow will occur along the major shear plains of
the serpentinite Additionally these shear plains are often coated with high valent Mn
oxides providing a reaction pathway for Cr(VI) formation Serpentinite springs such as
those evaluated by Barnes and ONeil (1969) and Barnes et al (1972) at the Cedars
78
ultramafic body in Cazadero CA have Cr concentrations measured at 20 microg L-1 in which
all the Cr is in the form of Cr(VI) (Chapter 4) In conclusion chromite and birnessite in
serpentine soils could contribute a source of Cr(VI) in surface waters and groundwaters
Chromium release and oxidation from chromite is increased by the presence of
birnessite and the interaction between chromite and birnessite is a pathway for Cr(VI)
generation in ultramafic sediments and soils The potential of serpentine soils as a source
of non-anthropogenic Cr contamination can now be evaluated quantitatively based on the
abundance of chromite The overall rate law for Cr(VI) in solution (Equation 39)
provides a means to estimate the Cr(VI) concentration attainable for serpentine soils and
serpentine sediments when the chromite and birnessite suspension densities and the pH of
the soil are known
ACKNOWLEDGEMENTS
We gratefully acknowledge comments and assistance from Robert Jones and Dr
Guangchao Li Funding for this research was provided by the Office of Teaching and
Learning at Stanford University
79
Table 31 Chromite and birnessite batch experiment parameters with chromite and birnessite suspension densities
Experiment Chromitea (grams)
Birnessiteb (grams) pHc
Chromite Surface Aread
(m2 g-1)
Birnessite Surface Aread
(m2 g-1)
Chromite Suspension
Density (m2 L-1)
Birnessite Suspension
Density (m2 L-1)
A 08 043 5 025 125 02 54 B
08 086 5 025 125 02 108C 03 086 5 025 125 008 108D 008 086 5 025 125 002 108E 004 086 5 025 125 001 108F 03 086 3 025 125 008 108G 03 086 67 025 125 008 108H 03 086 8 025 125 008 108
Chromite 08 ---- 5 025 125 02 ----Birnessite ---- 043 5 025 125 ---- 5480
a Chromite chemical composition is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 b Synthetic birnessite (δ-MnO2) c10 mM acetate buffered solution pH = 5 non-buffered pH=3 5 mM MES buffer pH 67 non-buffered pH 8 dMeasured BET analyses
81
Table 32 Serpentine soil and birnessite experiments with soil and birnessite masses used and the estimated chromite suspension density
Samples Soil Mass (g)
Birnessite Mass (g)
Volume (L) of Solutiona
Total Crsoil
b (mg kg-1)
Chromite Mass (g) in
Soil Samplec Estimated Surface
Areac (m2 g-1)
Chromite Suspension Density
(m2 L-1) in Soild Jasper Ridge
(JR1) 10 ---- 01 5976 013 015 02 Jasper Ridge (JR1 w Bt) 10 01 01 5976 013 015 02 Jasper Ridge
(JR2) 10 ---- 01 4978 011 015 016 Jasper Ridge (JR2 w Bt) 10 01 01 4978 011 015 016
New Caledonia (NC) 10 ---- 01 827 0015 006 001
New Caledonia (NC w Bt) 10 01 01 827 0015 006 001
a1mM Acetate Buffer (pH=5) 81 bTotal Cr concentrations were obtained through total digestion and ICP-AES analyses
cEstimated chromite mass and surface area (discussed in the Methods section) was made using chromite observations and calculations in Chapters 2 and 4
dCalculated from the chromite mass in the soil sample estimated surface area of the chromite and the volume of solution
82
Table 33 Summary of Cr(VI) rates obtained from batch experiments (Figure 35 36 and 37) and used in Figure 38
Listed in Table 31
Suspension Densities (m2 L-1)
Chromite (Ch)
Birnessite (Bt)
pH Rate (microM h-1) (d[Cr(VI)]dt)
log Rate (log microM h-1) (log d[Cr(VI)]dt) Expression
Figure 35 02 108 5 00041 -239
008 108 5 00021 -268 002 108 5 00009 -305 001 108 5 00005 -330
1
Figure 36 02 54 5 00041 -239 02 108 5 00041 -239 2
Figure 37 008 108 3 00062 -221 008 108 5 00021 -268 008 108 67 00004 -340 008 108 8 0 ----
3
Expression 1 log d[Cr(VI)]dt = asdot log Chromite +log krsquorsquo log d[Cr(VI)]dt = 06908 log Chromite ndash 1899 Expression 2 log d[Cr(VI)]dt = bsdot log Birnessite + log krsquorsquorsquo
log d[Cr(VI)]dt = 0sdot log Birnessite ndash 23872 log d[Cr(VI)]dt = -23872
Expression 3 d[Cr(VI)]dt = 264E-4sdotpH2 ndash 414E-3sdotpH + 162E-2
82
Table 34 Extractions comparing total solution and extractable Cr and Fe from the solids
------------------Cr----------------- ------------------Fe---------------- microMolesdagger Total
mole Extraction
mole microMoles Total mole
Extraction mole
Chromite Ch 02 (Table 31) Terminated at 30 days Solutiona 0033 554 ---- 0 0 ---- Phosphateb 0000 05 1 0000 03 03 CaCl2
c 0013 388 87 0000 00 00 HCld 0002 53 12 0103 997 997 Bottle HCle 0000 00 0 0000 00 00
A Ch 02 Bt 54 (Table 31) Terminated at 30 days Solution 2977 910 ---- 0 0 ---- Phosphate 0048 16 18 0004 21 21 CaCl2 0055 18 20 0002 09 09 HCl 0161 54 60 0126 758 758 Bottle HCl 0004 01 2 0035 212 212
B Ch 02 Bt 108 (Table 31) Terminated at 30 days Solution 2837 914 ---- 0 0 ---- Phosphate 0039 14 16 0008 49 49 CaCl2 0060 21 25 0000 00 00 HCl 0142 50 58 0127 803 803 Bottle HCl 0004 01 2 0024 148 148
C Ch 08 Bt 108 (Table 31) Terminated at 17 days Solution 0877 894 ---- 0 0 ---- Phosphate 0020 23 22 0002 37 37 CaCl2 0030 34 32 0000 00 00 HCl 0040 46 43 0053 881 881 Bottle HCl 0003 03 3 0005 82 82
daggerMolarity values are normalized to the final solution analysis of each experiment based on volume of solution and mass of the sample in each experiment aSolution refers to the final solution analysis before the experiment was terminated bmicroMoles extracted from the batch experiment solids using a 1 mM phosphate buffer pH 7 3 h cmicroMoles extracted from the batch experiment solids using a 10 mM CaCl2 solution pH 7 3 h dmicroMoles extracted from the batch experiment solids using a 05 N HCl solution 3 h emicroMoles extracted from the batch experiment bottles using a 05 N HCl solution 15 h
83
Figure Captions
Figure 31 Scanning electron microscopic (SEM) images of chromite and birnessite at
varying magnifications prior to the batch experiments
Figure 32 A summary of Cr Fe and Mn released from chromite during extraction
experiments Solutions used for these experiments include 1) 10 H2O2 solution 2) 01
M KMnO4 solution 3) 10 M HF 4) 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 5) 1mM
BaCl2 solution () Fe concentration for the HF extraction is 106 mg L-1
Figure 33 pH and Eh measurements as a function of time (h) for the batch
experiments Chromite (Ch) and birnessite (Bt) suspension densities used for the batch
experiments (Table 31) are listed adjacent their corresponding symbol in the key
Figure 34 Eh and pH batch experiment analyses (Figure 33) plotted on a stability field
diagram for Cr (log a Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom
(1998)
Figure 35 Batch experiments evaluating variable chromite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr and Cr(VI) measured in solution are noted and illustrated with a dotted line and
included in Table 33
Figure 36 Batch experiments evaluating variable birnessite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr(VI) measured in solution are denoted and included in Table 33
Figure 37 Influence of pH on Cr(VI) generation by reaction of constant chromite (Ch
08 m2 L-1) and birnessite (Bt 108 m2 L-1) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key Rates of Cr(VI) measured in solution are included in Table
33
84
Figure 38 Rates of Cr(VI) in solution (Table 33) as a function of a) chromite
suspension density (Figure 35) b) birnessite suspension density (Figure 36) and c) pH
(Figure 37) The line in each diagram represents the best fit for these data points and the
equations for these lines are listed in Table 33 The dotted line deviating from the
dashed line in b) indicates the probable path at low birnessite concentrations
Figure 39 SEM images of the solid phases after the completion of the batch
experiments A (Ch 02 Bt 54) and B (Ch 02 Bt 108) are shown Chromite is the larger
grain with relatively flat surfaces whereas birnessite is the fine-grained rosettes located
adjacent the larger chromite grains
Figure 310 Chromium and Cr(VI) concentrations (microM) in solution as a function of
time (h) from serpentine soils obtained at Jasper Ridge (JR1 and JR2) and New Caledonia
(NC) Soils with the addition of 01 g birnessite are noted in the figure Soil and
birnessite masses used for each experiment are listed in Table 32
Figure 311 Iron analyses as a function of time (Figure 35 and 36) with the solubility
of amorphous Fe(OH)3 and goethite and the dominant Fe3+-OH complex at pH 5 Fe
solubilities are from Langmuir (1997) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key
Figure 312 a) Experimental values of Fe and Cr (Figures 35 and 36) are plotted A
dotted line denotes the ratio of Fe and Cr necessary for chromite congruent dissolution b)
Corrected FeCr ratios as a function of time (h) for data from Figures 35 and 36 taking
into account the concentration of Fe and Cr in secondary phases calculated from Table
34 c) Corrected Fe and Cr as compared to values predicted with congruent dissolution
of chromite
85
Figure 313 Three potential pathways of Cr release and oxidation for chromite-
birnessite interactions are listed
86
Figure 31
87
0
2
4
6
8
10
12
Hydrogen
Peroxid
e
K-Perman
ganate
Hydroflu
oric Acid
Hydroxy
lamine
Barium Chlorid
e
Con
cent
ratio
ns (m
g kg
-1)
Cr
Fe
Mn
Figure 32
88
Time (h)0 100 200 300 400 500
pH
2
3
4
5
6
7
8
9
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
Time (h)0 100 200 300 400 500
Eh (m
V)
100
200
300
400
500
600
Ch Bt ------(m2 L-1)------
Figure 33
89
pH2 3 4 5 6 7 8 9
Eh (m
V)
-05
00
05
10
15
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
HCrO4-
CrO42-
Cr3+
CrOH2+
Cr(OH)3Cr(
OH
) 2+
Upper limit of water stability
Lower limit of water stability
Ch Bt ------(m2 L-1)-----
Figure 34
90
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 108 008 108 002 108 001 108 02 --- --- 54
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
0
1
2
3
4
pH 5 pH 5
pH 5 pH 5 pH 5
Ch Bt ---(m2 L-1)---
Figure 35
91
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 54 02 108
00041 microM h
-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
00
02
04
06
08
10
00041 microM h
-1
pH 5
pH 5 pH 5 pH 5
pH 5 Ch Bt----(m2 L-1)----
Figure 36
92
Time (h)
0 50 100 150 200 250
Cr c
once
ntra
tion
( microM
)
00
02
04
06
08
10
pH 3pH 5pH 67pH 8
Time (h)0 50 100 150 200 250
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250M
n co
ncen
trat
ion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
00004 microM h-1
00021 microM h-1
0006
2 microM h
-1
00021 microM h-100
062 micro
M h-1
00004 microM h-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1
Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Figure 37
93
log Birnessite m2 L-1
05 10 15 20 25 30
log
d[C
r(VI
)]dt
microM
h-1
-40
-35
-30
-25
-20
pH 5
log Chromite m2 L-1-22 -20 -18 -16 -14 -12 -10 -08 -06
log
d[C
r(VI
)]dt
microM
h-1
-34
-32
-30
-28
-26
-24
-22
pH 5
pH1 2 3 4 5 6 7 8 9
d[C
r(VI
)]dt
microM
h-1
0000
0002
0004
0006
0008
0010
a)
c)
b)
Ch 008 m2 L-1 Bt 108 m2 L-1
Figure 38
94
Figure 39
95
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR1JR1 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR2JR2 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
01
02
03
04
05
NCNC w Bt
00044 microM h-1
00013 microM h-1
00044 microM h-1
00013 microM h-1
00013 microM h-1
00009microM h-1
pH 5
pH 5
pH 5
Figure 310
96
Time (h)0 100 200 300 400 500
log ΣF
e (aq)
microM
-10
-8
-6
-4
-2
0
2
4
02 54 02 108 008 108 002 108 001 108
Fe(OH)3(am)
Goethite
pH 5Ch Bt----m2 L-1----
Fe(OH)2(aq)+
Figure 311
97
Time (h)
0 100 200 300 400
FeC
r Rat
io
0
1
2
3
4
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
02 54 02 108 008 108 002 108 001 108
Chrom
ite C
ongr
uent
Dissolu
tion
Chrom
ite C
ongr
uent
Dissolu
tion
a)
b)
c)
Ch Bt----(m2 L-1)----
ChromiteCongruent Dissolution
Figure 312
98
Pathway 1 Chromite Dissolution
1a) Cr dissolution from chromite FeCr2O3(s) + 4H+ rarr 2Cr(OH)2+
aq + Fe2+(Oxyhydroxide) + H2O
1b) Adsorption of CrOH2+ on the surface of birnessite Cr(OH)2+
aq + MnO2(s) rarr Cr(III)-MnO2(s) + OH- 1c) Oxidation of CrOH2+ on the surface of birnessite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
1d) Desorption of Cr(VI) from birnessite Cr(VI)-Mn(II) + 4H2O rarr Mn2+
(aq) + HCrO4-(aq) + 7H+
Pathway 2 Birnessite Dissolution 2a) Mn4+ dissolution from Birnessite MnO2(s) + 4H+ rarr Mn4+
(aq) + 2H2O 2b) Mn4+ adsorption on the surface of chromite Mn4+
(aq) + FeCr2O4(s) rarr Mn(IV)-FeCr2O4 2c) Oxidation of Cr(III) on the surface of chromite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
2d) Desorption of CrO24- from chromite
Cr(VI)-Mn(II) + 4H2O rarr Mn2+(aq) + HCrO4
-(aq) + 7H+
Pathway 3 Solid Solution Reaction 3) Solid solution interaction between chromite and birnessite
Figure 313
99
CHAPTER 4
CHROMIUM GEOCHEMISTRY OF SERPENTINE SOILS
ABSTRACT
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
chemistry of Cr in serpentine soils including its protolith is reviewed focusing on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils are slightly acidic (average pH of ~6) contain a variety Fe(III) oxides
(magnetite and hematite) phyllosilicates (serpentine and chlorite) and clays (smectites
and vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1)
and Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Chromium(III) is the only valence state observed
in the serpentine soil solids however Cr(VI) has been identified in New Caledonia and
California serpentine soil solutions at concentrations below 30 microM The enrichment and
range of Cr concentrations in serpentine soils are directly related to the presence of Cr-
spinels specifically chromite and Cr-magnetite These spinels are resistant to weathering
and are preserved in the soil environment however oxidation of Cr(III) from Cr-spinels
by high valent Mn oxides or other strong oxidants is a potential source of Cr(VI)
identified in serpentine soil solutions Due to the chemically and physically resistant
nature of the Cr-spinels Cr-bearing silicates including clay minerals Cr-chlorite Cr-
garnet Cr-mica and Cr-epidote are more viable sources of Cr identified in vegetation
soil extractions soil solutions and related waters
100
INTRODUCTION
Serpentine soils a generic term used to describe any soil derived from an
ultramafic rock (peridotites and pyroxenites) or serpentinite (metamorphosed ultramafic
rock) are often characterized by 1) low concentrations of plant nutrients such as N P
and K 2) high concentrations of biologically toxic elements including Ni Co and Cd 3)
low ammonium acetate extractable CaMg ratios and 4) unique flora and physical
properties (Brooks 1987) Measured Cr concentrations in serpentine soils range from
~200 mg kg-1 to 6 Wt values enriched compared to non-serpentine soils lt200 mg kg-
1 (Kabata-Pendias and Pendias 1984 Schwertmann and Latham 1986 Brooks 1987)
Chromite and Cr-magnetite are common primary sources of Cr in these soils (Rabenhorst
et al 1982 Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989
Gasser and Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003)
however the plethora of factors relating the chemistry of Cr in serpentine soils and the
weathering of chromite has not been fully addressed using an integrative approach
Overall serpentinites and serpentine soils cover ~1 of the earthrsquos exposed
surface and these rocks and soils are commonly associated with ophiolite complexes
(Coleman 1977) Serpentinites and serpentine soils related to ophiolite complexes are
the primary interest of this paper A map illustrating the worldwide distribution of major
ophiolitic serpentinite bodies and their related soils is shown in Figure 41 A majority of
the sites are near convergent boundaries (Figure 41) Although these rocks and soils
cover a minimal area of the earthrsquos surface they are present in populated areas within the
Circum-Pacific Margin and the Mediterranean Due to the wide-distribution and
occurrence of serpentinites and serpentine soils weathering of these rocks and soils will
differ from location to location due to varying climatic conditions as well as the nature of
the parent material and other factors including topography biota and time (Proctor and
Nagy 1992) As a result the chemistry of these soils as well as the nature of Cr in the
serpentinites and serpentine soils will vary from site to site potentially allowing Cr to be a
source of non-anthropogenic contamination
The objective of this paper is to review the chemical characteristics of serpentine
soils globally evaluate mineralogical and compositional variations between several soils
derived from ultramafic sources and assess the geochemistry of Cr in diverse soil-
101
forming environments Additionally chemical and spectroscopic analyses were
performed on soils derived from ultramafic rocks collected from 1) New Caledonia
South Pacific 2) Eight Dollar Mountain Selma OR 3) Nickel Mountain Riddle OR 4)
Harvard Mine Jamestown CA 5) Jasper Ridge Biological Preserve Stanford CA and 6)
Pillikin Mine El Dorado CA shown geographically in Figure 41 A basalt-derived soil
from Hawaii is included in this study in order to compare a soil formed from a mafic
protolith to those from ultramafic protoliths The various field sites are shown in Figure
42 These soils allow a comparative examination of how soil-forming factors such as
climate topography biota parent material and time affect the chemistry of Cr
REVIEW
Chromium Geochemistry
Chromium is a group VIB transition metal with two common oxidation states
Cr(III) and Cr(VI) differentiated by noticeably different chemical characteristics and
toxicities It should be noted that Cr(II) has been observed in lunar basalt olivines and in
high temperature terrestrial olivines (Sutton et al 1993 Li et al 1995) however this
oxidation state is not observed in soil environments A majority of Cr commonly persists
in ultramafic rocks serpentine soils and related waters as Cr(III) a non-hazardous
species and a micronutrient in human nutrition (Daugherty 1992 Cohen et al 1993
Fendorf 1995 James et al 1997) Chromium(III) has the highest crystal field
stabilization energy for all transition metals strongly favors octahedral coordination and
has 3 d-electrons with a high spin state Chromium(III) is strongly hydrolyzed in
aqueous solutions and the predominant species based on pH redox potential and O2(aq)
are shown in Figure 43 Chromium(III) species are more prevalent under reducing
conditions and its mobility is increased only under strongly acidic conditions At pH
values greater than 9 the hydrolyzed Cr(III) species is negatively charged demonstrating
the amphoteric properties of Cr(III) (Figure 43) Due to the factors described above
Cr(III) forms low solubility compounds and adsorbs strongly on mineral surfaces and
organic matter at pHgt4 (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) For
example the adsorption of Cr(III) on HFO (hydrous ferric oxide) as a function of pH is
shown in Figure 43 illustrating an adsorption edge at a pH of ~4 (Leckie et al 1984
102
Stumm 1992) Chromium(III) oxidation to Cr(VI) has been observed in reactions
involving hydrogen peroxide (H2O2) and high valent Mn oxides (Eary and Rai 1987
Johnson and Xyla 1991 Fendorf et al 1992 Fendorf et al 1993 Rock et al 2001)
Chromium(VI) is a strong oxidizing agent a highly corrosive substance a toxin to
living cells and a Class A human carcinogen by inhalation (Daugherty 1992 Cohen et
al 1993 Fendorf 1995) The hydrolyzed Cr(VI) species are also shown in Figure 43
with respect to pH redox potential and O2(aq) Chromium(VI) is more favorable under
oxidizing conditions and its mobility is enhanced under basic conditions In solution
Cr(VI) is a weakly adsorbing (oxy)anion (HxCrO42-x) on common weathering products
meaning that it forms highly labile complexes that can be readily displaced by ions
including phosphate and sulfate (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Additionally the adsorption of CrO42- on HFO shown in Figure 43
demonstrates the anionic character in which the mol bound of CrO42- increases with
decreasing pH and an adsorption edge is at pH ~72 (Leckie et al 1980 Stumm 1992)
The reduction of Cr(VI) to Cr(III) is known to occur by reactions with Fe(II)aq Fe(II)-
containing minerals such as magnetite H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996)
Mineralogy of Serpentine Soils
Serpentine soils have been identified on every continent excluding Antarctica
(Figure 41) The occurrence and distribution of ultramafic rocks specifically related to
ophiolites are closely associated with convergent plate boundaries (Figure 41 Coleman
1977 Coleman and Jove 1992) Tectonic blocks of ultramafic rock are hypothesized to
be detached from the down-going slab of oceanic crust and incorporated into the
subduction meacutelange Subsequent serpentinization of peridotite forms low density
serpentine group minerals (245-250 g cm-3) resulting in the upward diapiric migration of
the serpentinite masses along zones of structural weakness (fault zones) and through the
denser enclosing subduction meacutelange (~265 g cm-3) For example North America has
numerous exposed ophiolites containing serpentinite located in western California and
Oregon where subduction zones were once active in the Early Cretaceous (Harden
1998)
103
Serpentinization of ultramafic rocks is a significant factor controlling mineral
assemblages and textures in serpentinite Hydration and oxidation of olivine and
pyroxene during serpentinization forms serpentine (lizardite chrysotile and antigorite)
talc chlorite tremolite brucite and magnetite (Coleman 1977 Brooks 1987 Coleman
and Jove 1992 Malpas 1992) Additionally magnetite formation creates favorable
conditions for native metal compounds (such as Fe Co and Ni alloys) to form (Coleman
and Jove 1992) As a serpentinite mass is emplaced the extensive physical and chemical
alteration of the serpentinite results in numerous crosscutting alteration veins and highly
sheared textures Ultimately the mineral assemblages and textures in serpentinites are
complex prior to pedogenesis due to the multiple phases of metamorphosis tectonic
deformation and weathering of the rock
Serpentine soils consist of a mixture of residual minerals from the soil protolith
eolian and illuvial minerals and minerals formed within the soil Additionally highly
weathered serpentine soils enriched in Fe-oxides are also termed laterites soils common
in tropical and subtropical regions Soil protolith minerals may include primary minerals
of ultramafic rocks including olivines pyroxenes and oxides (chromite and magnetite)
as well as a large variety of metamorphic minerals the most common being serpentine
(lizardite chrysotile and antigorite) chlorite talc and actinolite Silica phases occur as
eolian grains concretions andor net type structures or silica webs (Hotz 1964
Coleman 1977 Schwertmann and Latham 1986) Eolian and illuvial aluminosilicate
minerals are important sources of Al in serpentine soils an element of low abundance in
ultramafic rocks
Layer silicates observed in serpentine soils are varied in type and abundance
They include smectites (often as montmorillonite) vermiculite chlorite brucite
serpentine and a variety of intrastratified clays (Veniale and Van der Marel 1963 Hotz
1964 Schellmann 1964 Wildman et al 1968 Rabenhorst et al 1982 Schreier et al
1987 Graham et al 1990 Bulmer et al 1992 Gasser and Dahlgren 1994 Lee et al
2003) Compositional variations crystal structures and parageneses of layer silicates
have been characterized for a number of serpentine soils (Wildman et al 1968
Rabenhorst et al 1982 Schreier et al 1987 Graham et al 1990 Bulmer et al 1992
Lee et al 2003) In addition to the layer silicates goethite hematite maghemite and
104
amorphous Fe (oxy)hydroxides are soil-forming phases in serpentine soils (Coleman and
Jove 1992) Gibbsite is not a common phase in most serpentine soils however limited
isomorphic substitution of Al(III) in goethite and hematite was noted in New Caledonia
(Schwertmann and Latham 1986) Manganese oxides such as birnessite (δ-MnO2) are
also present in serpentine soils and often coat mineral surfaces
Sources of Chromium for Serpentine Soils
Chromium primarily occurs in oxide and silicate phases in ultramafic rocks and
serpentinites Chromium-spinel often referred to as chromite is a common primary
(igneous) spinel in mafic and ultramafic rocks and often resists low-grade metamorphic
processes related to serpentinization (Hoffman and Walker 1978 Malpas 1992)
Isomorphic substitution of Al3+ Fe3+ and Ti4+ into the octahedral sites and Mg2+ Ni2+
Zn2+ and Mn2+ into the tetrahedral sites is common however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987b Pan and Fleet 1991 Sack and Ghiorso 1991 Burkhard 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996 Mathiesen 1999) Serpentinite chromite compositions reported in the literature are
plotted in Figure 44 to demonstrate the wide-range of Cr concentrations possible for Cr-
spinels Isomorphic substitution of octahedral Cr and octahedral Al for these chromite
compositions are also shown in Figure 44 Serpentinite pyroxenes containing Cr include
augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) (Grapes 1981 Von Knorring et al
1986 Treloar 1987a Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996)
Magnetite and serpentine minerals may incorporate varying amounts of Cr into
their structures during serpentinization Chromium(III) from primary Cr-containing
phases may isomorphically substitute into the Fe(III) site of magnetite due to size
(octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge similarities Chromium
concentrations in magnetite have been observed as high as 13 Wt (Oze et al 2003)
Chromate minerals such as crocoite (PbCrO4) are extremely rare in serpentinites and the
occurrence of these minerals are mainly limited to specific localities such as in Australia
Tasmania and the Urals (Williams 1974 Crane et al 2001) Serpentine minerals have
105
been reported to contain as much as 1500 mg kg-1 Cr based on spectrographic analyses
by Page and Coleman (1967) however serpentine minerals typically contain only trace
amounts of Cr
Hydrothermal and CO2 metasomatism of ultramafic rocks during the evolution of
convergent plate margins commonly produce Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite lt19 Cr Wt ) uvarovite (Cr-garnet lt20 Cr Wt )
tawmawite (Cr-epidote lt13 Cr Wt ) and kaumlmmererite (Cr-chlorite lt10 Cr Wt ) as
shown in Figure 44 (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974 Neiva
1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981 Max et al
1983 Nutman et al 1983 Raase et al 1983 Von Knorring et al 1986 Kerrich et al
1987 Treloar 1987b Treloar 1987a Morand 1990 Pan and Fleet 1991 Schandl and
Wicks 1993 Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995
Mathiesen 1999 Proenza et al 1999) Figure 44 illustrates the relationship of Cr and
Al substitution for Cr-silicate compositions based on Cr and Al Wt Additionally Cr-
silicate compositions are plotted based on octahedral Cr and octahedral Al in Figure 44
Compositions for Cr-muscovite Cr-epidote and Cr-garnet plot near the tie lines
connecting Cr and Al end-member minerals Chromium isomorphically substitutes for
octahedral Al in these silicates due to similarities in charge octahedral preference and
octahedral radii (Al3+ = 061 Aring) The numerous isomorphic substitutions possible for Cr
and Al in the chlorite structure results in analyses that do not plot near the tie line
connecting clinochlore and kaumlmmererite Overall any of these minerals are potential
sources for Cr in soils produced by the weathering of serpentinites and their hydrothermal
and CO2 metasomatic derivatives
Chromium in Serpentine Soils
Chromium concentrations vary widely in serpentine soils worldwide (Table 41)
New Caledonia soils appear to be the most Cr-enriched serpentine soil with a Cr
concentration of ~60000 mg kg-1 (Schwertmann and Latham 1986) however Jaffre
(1980) report values ranging from 6300 to 80000 mg kg-1 for New Caledonia soils
Some serpentine soils have measured Cr concentrations as low as 29 and 66 mg kg-1
atypical of most values reported in Table 41 (Proctor and Woodell 1971 Brooks et al
106
1992 Jeffrey 1992) Chromium concentrations as a function of depth for serpentine soil
profiles in Maryland (Rabenhorst et al 1982) Tehama County CA (Gough et al 1989)
the Klamath Mountains CA (Hotz 1964) Newfoundland Canada (Roberts 1992)
Harvard Mine CA (this study) and Jasper Ridge CA (Oze et al 2003) are reported in
Figure 45 The pedogenic process(es) responsible for the varying Cr concentrations as a
function of depth for each soil have not been characterized however Oze et al (2003)
determined that a majority of Cr in Jasper Ridge serpentine soils is the result of
weathering-resistant residual Cr-bearing minerals from the protolith mainly chromite
and Cr-magnetite Additionally Oze et al (2003) demonstrated that soil samples from
Jasper Ridge contain higher concentrations of Cr with respect to their corresponding
bedrock due to the Cr-spinels accumulating over time The observation that Cr
concentrations decrease with depth towards the protolith is characteristic for most of the
profiles in Figure 45
In addition to examining total Cr concentrations chemical extractions performed
on serpentine soils provide additional information with regards to Cr availability and the
minerals and complexes in which Cr may reside in the soils Table 42 summarizes a
number of extraction experiments performed on serpentine soils For a majority of the
soils listed in Table 42 total Cr concentrations for the serpentine soils were not
measured however these extractions do provide a quantitative assessment of where Cr
may be bound in soil (such as silicates organic matter and secondary Fe-bearing phases)
Extractions performed by Schwertmann and Latham (1986) were able to remove ~20000
Cr mg kg-1 from the serpentine soil using a dithionite-citrate bicarbonate (DCB) solution
which preferentially dissolves crystalline Fe-oxides Solutions such as concentrated
nitric and hydrochloric acid were only able to remove a maximum of ~1400 Cr mg kg-1
from Unst serpentine soils (Shewry and Peterson 1976) which preferentially dissolve
Cr-bearing silicates and to a lesser extent the Fe-oxides Where bulk soil Cr values were
measured extraction experiments are only able to remove a small portion (lt1) of the
total Cr present in the serpentine soils (Table 42) (Gasser and Dahlgren 1994 Oze et al
2003) Based on the extractions presented in Table 42 it appears that Cr(III) resides in
minerals and phases resistant to common extraction techniques
107
In serpentine soils Fe(III) oxides and Fe(III) hydroxides provide a host of
structural and surface sites available to sequester Cr(III) released to soil solutions by
dissolution of Cr-bearing minerals during soil-forming processes Substitution of Cr(III)
for Fe(III) in Fe-oxides Fe-(oxy)hydroxide minerals and amorphous compounds is
expected based on crystallographic considerations synthesis experiments and analysis of
extracts from selective dissolution of Fe(III) compounds in serpentine soils
(Schwertmann and Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990
Gerth 1990 Gasser and Dahlgren 1994) Chromium(III) also readily sorbs on surfaces
of Fe(III) oxides and (oxy)hydroxides via inner-sphere surface complexes (Figure 43)
and forms low-solubility precipitates (Charlet and Manceau 1992) The low
concentration of Cr(III) in soil solutions has been attributed to the low solubility of
Fe(III)-Cr(III) amorphous hydroxides (Sass and Rai 1987)
Layer silicates provide structural and surface sites capable of uptaking Cr(III)
Octahedral Cr(III) in clays such as smectites and vermiculites is expected and Cr
concentrations in the clay-sized fraction can achieve concentrations of ~1000 mg kg-1 in
serpentine soils (Oze et al 2003) Sorption of Cr(III) and Cr(VI) on edge and interlayer
sites of phyllosilicates have been evaluated by Ilton and Veblen (1994) Ilton et al (2000)
and Brigatti et al (2000) indicating that layer silicates (montmorillonite chlorite biotite
phlogopite and muscovite) provide sites for both Cr(III) sorption and Cr(VI) reduction
Kaupenjohann and Wilcke (1995) concluded that Cr in serpentine soils from Bavaria is
residually bound mainly in silicate structures including clay minerals based on extraction
experiments
Soil Solutions and Related Water
The addition of water gradually changes the spatial and temporal characterization
of pore fluids in soils including the ionic strength pH and redox potential Leaching of
a soil should progressively enrich the soil with less chemically mobile elements such as
Cr and Fe Elements identified in soil solutions include Mg Ca Fe Ni and Cr and the
ionic strengths of serpentine soil solutions are typically low (lt001) (Anderson et al
1973 Johnston and Proctor 1981 Gasser and Dahlgren 1994) Soil solutions obtained
from serpentine soils using tension-free lysimetry demonstrate Cr concentrations ranging
108
from 01 to 32 micromol L-1 (Gasser and Dahlgren 1994) This study also revealed a
majority of Cr Fe Mg and Ni in extracted soil waters were in the form of colloids
suggesting that nanoparticles are an important factor of Cr translocation in soils
The pH of a soil is a dominant factor controlling the nature of many chemical and
microbial reactions including the geochemistry of Cr (Brady and Weil 1999) Ranges in
pH for serpentine soils worldwide are plotted in Figure 46 The highest soil pH was
measured in a South Africa serpentine soil with a value of 9 (Cole 1992b) whereas the
lowest pH was reported in Ireland with a value of 38 (Jeffrey 1992) This wide range of
pH is a function of the adsorbed H+ and Al3+ the base-forming cations the organic
matter and the mineralogy present within each soil The pH range of serpentine soils
shown in Figure 46 is plotted on the stability diagram in Figure 43 demonstrating that
both Cr(III) and Cr(VI) aqueous species are favorable The Eh range for the serpentine
soils in Figure 3 is based on the soils being in contact with the atmosphere as well as
being a transitional environment Although the activity of H+ varies by four orders of
magnitude the pHs of most serpentine soils are slightly acidic (pH ~6) which is near that
of average rainwater (pH~57 Figures 43 and 46) Additionally serpentine soil pH
values have been observed to increase with depth towards the protolith for a number of
soils (Rabenhorst et al 1982 Gough et al 1989 Gasser and Dahlgren 1994) The
increase in pH with respect to depth may be the result of rainwater being progressively
influenced by the mineral chemistry and hydrolysis reactions as it descends through the
soil towards the protolith
Groundwater from serpentinized ultramafic rocks have pH values ranging from 8
to 12 (Barnes et al 1972) and these waters have been divided into two categories as
shown in Figure 46 Mg bicarbonate (Mg2+-HCO3-) type and Ca hydroxide (Ca2+-OH-)
type (Barnes and ONeil 1969 Barnes et al 1972) The Mg bicarbonate waters (pH~8)
are characteristic of most surface waters and shallow groundwaters from serpentinites
and ultramafic rocks The Ca hydroxide waters (pHasymp10-11) are associated with
incompletely serpentinized peridotites and these fluids are comparatively low in Mg2+
Fe2+ Mn2+ Crtotal and SiO2total The Ca2+ in solution originates from the serpentinization
of Ca-rich orthopyroxene (Barnes et al 1972) The pH of serpentinite groundwaters are
109
plotted on the Cr stability in Figure 43 demonstrating that both Cr(III) and Cr(VI) are
potential valence states in these waters
Chromium predominates as Cr(III) in ultramafic rocks and serpentine soils
however Cr(VI) has been identified in serpentine soil solutions and serpentinite
groundwater in California and New Caledonia at concentrations of lt20 microg L-1 (Gough et
al 1989 Becquer et al 2003) Calcium hydroxide waters (pH~11) collected from the
same spring evaluated by Barnes et al (1972) in the Cedars ultramafic body (Cazadero
CA) contains 12-22 Cr(VI) microg L-1 and no Cr(III) (this study) Additionally water from
chromite-containing ultramafic rocks in the Siniktanneyak Mountain ultramafic complex
of Alaska with pH values ranging from 67 to 93 were measured with low total Cr
concentrations lt2 Cr microg L-1 (Taylor et al 1998) A mechanism for Cr(III) oxidation in
serpentine soils and related water suggested by (Gough et al 1989 Becquer et al 2003)
may be the result of high valence Mn oxides common in serpentine soils High valence
Mn-oxides such as birnessite (δ-MnO2) are naturally occurring oxidants capable of
oxidizing Cr(III) to Cr(VI) at pHlt9 (Eary and Rai 1987 Fendorf et al 1992 Fendorf et
al 1993) It is also important to note that Cr(VI) reduction is hypothesized to occur in
soils due to electron transfer involving organic matter and Fe(II)-bearing minerals such as
magnetite (Peterson 1996 Fendorf et al 2000) Whether toxic concentrations of Cr(VI)
can be present in serpentine soils and related water over geologic time is important to
consider especially when high valence Mn oxides Fe(II)-bearing minerals and organic
matter are capable of significantly influencing the oxidation state of Cr
Serpentine Soil Vegetation
Vegetation associated with serpentine soils has received considerable attention
due to the apparent infertility of these soils The inhabitability of serpentine soils for
most plants has been attributed to the imbalance of available Mg relative to Ca or low
CaMg ratios (Walker 1954) the deficiency of plant nutrients such as K N and P
(Turitzin 1991) elevated heavy metal concentrations such as Ni and Cr (Robinson et al
1935) and poor drainage (Brooks 1987) Linking plant productivity to toxic levels of
Cr in the soil has not been fully resolved (Brooks 1987 Kruckeberg 1992) however a
majority of Cr in the soils is not available for plant uptake due to being in high stable and
110
weathering resistant phases discussed in the previous sections (Table 42) In general Cr
concentrations in plants typically do not exceed 100 mg kg-1 (Brooks 1987)
Table 43 provides a global list of serpentine soil plant species and their measured
Cr concentrations with an emphasis on Cr hyper-accumulators ie plants able
incorporate higher than average concentrations of Cr A majority of these measurements
are based on the dry weight of Cr Aerobryopsis longissima from New Caledonia has the
highest reported Cr concentration with values ranging from 200 to 700 mg kg-1 (Lee et
al 1977) Other noteworthy plants include the Lophostachys villosa from Brazil (Brooks
et al 1992) Arenaria humifusa from Newfoundland (Roberts 1992) the Calluna
vulgaris from Unst (Shewry and Peterson 1976) with Cr concentrations of 203 530 and
417 mg kgndash1 respectively Even mosses such as the Funaria flavicans (Maryland)
incorporate Cr concentrations as high as 90 mg kg-1 (Shaw 1991) By examining
specific parts of the plant Gough et al (1989) determined that Cr concentrations are
highest in the basal burl (600 mg kg-1) of the manzanita bush compared to the roots (30
mg kg-1) stems (20 mg kg-1) and leaves (20 mg kg-1) The study by Gough et al (1989)
demonstrates that Cr is not homogeneously distributed throughout the plant and that
evaluating a leaf versus a stem could result in significantly different Cr concentrations
Soil Organic Matter
Soil organic matter (SOM) including the biomass (living organisms such as
bacteria) detritus (nonliving plant residues) and humus (amorphous and colloidal
mixtures of organic substances) affects the geochemistry and buffering capacity of
serpentine soils Furthermore biologic reduction of Cr(VI) to Cr(III) via cometabolic
and metabolic pathways coupled with carbon oxidation has been demonstrated for a
number of bacterial strains including aerobes (Gopalan and Veeramani 1994 Park et al
2000) and anaerobes (Turick et al 1996 Tebo and Obraztsova 1998) Additionally
aqueous Fe(II) produced by microbial Fe(III) reduction is a known biogenic pathway for
the reduction of Cr(VI) (Fendorf et al 2000) Fe(III)-Cr(III) (oxy)hydroxides are the
resulting secondary phases (Rai et al 1989 Patterson and Fendorf 1997 Hansel et al
2003) Bacteria oxidizing Cr(III) to Cr(VI) has not been observed
111
Most serpentine soils in temperate regions appear to be either Mollisols or
Alfisols which often contain high concentrations of SOM Alexander et al (1989) and
Bulmer et al (1992) reported that serpentine soils can contain organic carbon
concentrations in excess of 50 mg g-1 Humus has a high cation exchange capacity and
provides much of the pH buffering capacity in these soils (Brady and Weil 1999)
Organic acids such as fulvic and humic acids are known to complex cations such as
Cr3+ Fe3+ and Mn2+ forming stable organometallic complexes (Kaupenjohann and
Wilcke 1995 Brady and Weil 1999) Additionally metal leaching experiments
performed by Schreier et al (1987) established that organic acids produced by soil
organic matter such as oxalic acid and citric acid were extremely effective in releasing
Cr Mg Fe Ni and Co from serpentinite rocks and sediments
CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
Serpentine soils were obtained from California Oregon and New Caledonia and
analyzed to further evaluate the factors influencing the Cr geochemistry in serpentine
soils Analyses for these soils include pH electrical conductivity total organic matter
total N ammonium acetate extractable Ca K Mg and Na soil texture soil color
chemical composition of the soil and the chemical composition of Cr-bearing minerals
The general soil characteristics are described in the following section and the analytical
methods are explained in Appendix 41 Although a majority of these soils are located in
the western United States factors influencing the pedogenesis of these soils vary
considerably allowing a comparative analysis of how the fate of Cr might be influenced
by climate topography biota parent material and time
Sample Locations and General Soil Characteristics
Six serpentine soils were collected from locations shown in Figures 41 and 42 to
evaluate the chemistry of Cr in serpentine soils These soils include 1) a laterite (Oxisol)
collected in New Caledonia South Pacific 2) a laterite (Oxisol) exposed at Eight Dollar
Mountain Selma OR 3) a laterite (Oxisol) near Nickel Mountain Riddle OR 4) a
serpentine soil (Inceptisol) with a CO2-metasomatized serpentinite protolith from
Harvard Mine Jamestown CA 5) a serpentine soil (Mollisol) obtained from a preserved
112
grassland at Jasper Ridge Biological Preserve Stanford CA and 6) a serpentine soil
(Inceptisol) collected in the proximity of Pillikin Mine El Dorado CA Lastly an Oxisol
formed from basalt obtained in Oahu Hawaii is also included for comparison
These seven sites include a range of soil environments from different protoliths
under varying conditions of rainfall temperature and topography Each soil was collect
near the highest and topographically level point for each serpentine soil site Serpentine
grassland soils at Jasper Ridge formed under Mediterranean-like climates (xeric moisture
regime) whereas higher temperatures and less rainfall (ustic moisture regime)
characterize soil formation at the Harvard and Pillikin mines in the Sierra Nevada
Foothills metamorphic belt The vegetation of these three sites is dominantly grasses with
some manzanita These modern soil environments contrast with the Miocene Ni and Cr-
rich laterite soils of Nickel Mountain and Eight Dollar Mountain that were probably
formed under an aquic moisture regime near sea level in a subtropical climate Grasses
manzanita and pine trees are the dominant forms of vegetation at Nickel and Eight
Dollar Mountain An ultramafically-derived laterite formed under tropical conditions
from New Caledonia was collected The vegetation and topography related to the New
Caledonia laterite was not recorded during sampling An Oxisol from Hawaii is also
included in this study to evaluate the differences between a highly weathered soil
originating from an ultramafic rock and one from a mafic rock (basalt) under tropical
conditions The Hawaiian soil was collected from a topographically level area however
the vegetation related to this site was not noted
BedrockProtolith Mineralogy
A broad characterization of the protoliths for the serpentine soils is briefly
summarized due to the bedrock not being collected from all the field sites New
Caledonia is composed of variably serpentinized peridotites predominantly harzburgites
and some dunites (Leblanc 1995) Nickel Mountain and Eight Dollar Mountain are
predominantly serpentinized harzburgite and dunite The bedrock for soils collected at
Harvard Mine is mainly composed of quartz ankerite and mariposite (also referred to as
QAM) caused by the CO2 metasomatism of serpentinite associated with gold
mineralization (Bohlke 1989) The average serpentinite at Jasper Ridge is composed of
113
serpentine (mainly lizardite and antigorite) chlorite talc magnetite and chromite (Oze et
al 2003) however brucite has been reported as a major constituent in related Franciscan
Complex serpentinites (Hostetle et al 1966) Pillikin Mine serpentinites dominantly
consist of talc and serpentine (lizardite antigorite and chrysotile) and these serpentinites
contain podiform chromite deposits (Cater et al 1948) Mafic bedrock for the Hawaiian
Oxisol is tholeittic basalt
Soil Chemistry and Physical Properties
As shown in Figure 47 the colors of the collected soils range from the red New
Caledonia laterite to the greenish gley of the Pillikin Mine Inceptisol The Munsel color
values are also shown in Figure 47 The dark grey color of the Jasper Ridge soil and the
greenish gley of the Pillikin Mine soil are atypical of soils derived from serpentinites due
to the lack of any apparent redness The range of these colors is often attributed to the
presence of Fe-oxides however organic matter and Mn oxides are also dominant factors
controlling color (Brady and Weil 1999)
Chemical properties for the six serpentine soils and the Hawaiian soil are listed in
Table 44 The slightly acidic pH values range from 53 to 69 similar to those shown in
Figure 43 and 46 The New Caledonia laterite is the most acidic soil with a pH of 53
compared to the Harvard Mine carbonate-buffered soil at 69 The pH of the Hawaiian
soil is also slightly acidic with a pH of 604 The electrical conductivities (EC) of the
soils range from 01 to 07 dS m-1 a range indicative of relatively few dissolved inorganic
solutes The highest organic matter content is in the very dark grey Jasper Ridge soil
with a value of 32 and the lowest is in the red New Caledonia laterite (05) The
Hawaiian soil has an organic matter content of 1 Nitrogen concentrations for all the
soils are lt012 with Jasper Ridge having highest N concentration which is consistent
with the soil organic matter content
Ammonium acetate extractable Ca K Mg and Na are also listed in Table 44
Magnesium concentrations are gt6923 mg kg-1 in all of the serpentine soils The lowest
concentration of Mg was measured in the basalt-derived Hawaiian Oxisol with a value of
168 mg kg-1 Calcium concentrations in the serpentine soils range from 167 to 998 mg
kg-1 The 998 mg kg-1 of Ca in the Harvard Mine soil is directly related to the abundance
114
of ankerite (gt20 of the bedrock) Potassium concentrations are low for most of the soils
(lt55 mg kg-1) however Jasper Ridge and Harvard Mine soils have the highest
extractable K with values of 237 and 301 mg kg-1 respectively The extractable K value
for the Harvard Mine soil is due to the K-metasomatism related to gold mineralization
involving the serpentinite soil protolith The extractable concentration of Na is lt44 mg
kg-1 for all the serpentine soils whereas the Hawaiian soil has the highest Na (53 mg kg-
1) due to plagioclase and basaltic glass
Textures for the soils range from clay loam clay loam to sandy clay loam (Table
44) The laterite soil in New Caledonia and Oxisol in Hawaii are extremely different in
which the Hawaiian soil is mostly clay (66 clay) compared to the loam at New
Caledonia (25 clay) Additionally both ancient laterites (Eight Dollar Mountain and
Nickel Mountain) have clay textures Harvard Mine Jasper Ridge and Pillikin Mine
soils are all loamy in character
Major and trace element compositions for the soils are listed in Table 45
Chromium concentrations in the serpentine soils range between 827 and 9528 mg kg-1
The Eight Dollar Mountain laterite has the highest Cr concentration and the New
Caledonia laterite contains the lowest Cr concentration The Cr concentration for the
New Caledonia soil is lower than the values reported in Table 41 The Hawaiian soil has
a Cr concentration well below the analyzed serpentine soils with 388 Cr mg kg-1 but still
greater than the 200 mg kg-1 characteristic of serpentine soils Nickel concentrations in
the serpentine soils range from 1895 to 14100 mg kg-1 with the highest concentration
being identified in the Nickel Mountain laterite Manganese concentrations range
between 257 (New Caledonia) to 3017 (Eight Dollar Mountain) mg kg-1 in the serpentine
soils Element concentrations in the Hawaiian Oxisol contrast from the serpentine soils
with higher Al Cu P Pb and Zn concentrations
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in the soil solids for Jasper Ridge New Caledonia Pillikin Mine Eight Dollar
Mountain and Harvard Mine Although the Cr concentrations in these soils exceed 200
mg kg-1 (Table 45) X-ray absorption near-edge spectra (XANES) shown in Figure 48
illustrate Cr(VI) levels are below the detection limits of this method (~10 Cr mg kg-1) due
to the lack of a pre-edge feature The box in Figure 48 illustrates where a Cr(VI) pre-
115
edge peak should be present Chromium(VI) concentrations in these soils are possibly
present in the parts per billion range
Soil Mineralogy
Backscattered electron images and microprobe analyses were used to identify the
Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49
and corresponding Cr-bearing mineral compositions are listed in Table 46 The New
Caledonia soil exhibits a wide range of particle sizes and a Fe-(oxy)hydroxide was
identified containing 165 Cr2O3 Wt Chromium-spinel has been identified in the New
Caledonia laterite by Becquer et al (2003) Eight Dollar Mountain soils exhibit a coarse
texture and both Cr-magnetite (2387 Cr2O3 Wt ) and chromite (3343 Cr2O3 Wt )
are present Nickel Mountain soils are also coarse and chromite (1315 Cr Wt ) was
identified Soils at Harvard Mine contain numerous chromite grains and the mineral
trevorite (a Ni-bearing spinel) was identified with 3991 Cr2O3 Wt and 2054 NiO Wt
Chromite and magnetite are abundant in soils at Jasper Ridge as well as a fine-
grained mixture of chromite magnetite and an unidentifiable silicate with up to 1878
Cr2O3 Wt (Figure 49) Veins of Cr-chlorite crosscut the spinel-silicate mixture in
Figure 49 and this Cr-chlorite is the only Cr-silicate identified using this method
Chromite (5355 Cr2O3 Wt ) is the dominant Cr-bearing mineral detected in soils at the
Pillikin Mine and these grains contain the highest Cr concentrations compared to the
other minerals in the seven soils The Hawaiian Oxisol possesses ulvoumlspinels (007
Cr2O3 Wt and 2465 TiO2 Wt ) which are evenly distributed throughout the soil
Overall microprobe observations reveal that the Cr-bearing spinels are present in every
soil Additionally Cr-bearing spinels at Jasper Ridge have been shown to undergo
incongruent dissolution progressively enriching the spinel towards a FeCr2O4 end-
member (Oze et al 2003)
DISCUSSION
The physical and chemical characteristics of serpentine soils are directly related to
the alteration and tectonic history of the serpentinite as well as the soil-forming factors
including the parent material climate biota topography and time Identifying Cr-
116
bearing minerals their chemical compositions and abundance in the soils allows a more
definitive explanation for Cr enrichment in these soils as compared to total Cr
concentrations listed in Tables 41 and 45 and sequential extraction experiments reported
in Table 42 Detrital chromite grains from the serpentinite protolith are resistant to
weathering and accumulate in the soil over time Although serpentine soils contain Cr
concentrations typically gt200 mg kg-1 the presence of Cr-spinels and their resistance to
weathering demonstrate that a majority of the Cr is not chemically available to interact
within the soil The role of organic matter and vegetation are important factors related to
the chemistry of serpentine soils however their role for influencing a majority of the Cr
geochemistry is limited due to the nature of the Cr-spinels
Chromium and Serpentine Soils
Total Cr concentrations exceed 200 mg kg-1 for most of the serpentine soils
reported in Table 41 and in the six serpentine soils of this study (Table 45) These Cr
concentrations can be accounted for by assessing the Cr-bearing minerals associated with
these soils and their relative abundance Small grains of Cr-spinels or chromite identified
in the soils (Figure 49 and Table 46) have the potential to appreciably impact Cr
concentrations Table 47 demonstrates the mineral masses and volumes of Cr-bearing
minerals required to cause predetermined Cr concentrations in soil samples A minimal
amount of chromite (lt02 of a soil sample mass or 005 of the volume of serpentine)
can result in Cr concentrations of 1000 mg kg-1 A few grains of chromite could easily
be responsible for the Cr concentrations in Tables 41 and 45 In fact residual grains of
Cr-enriched spinels are observed in every soil analyzed (Figure 49 and Table 46)
Ultimately grains of spinel in the soil do appreciably impact Cr concentrations however
they do not provide a readily available source of Cr based on extraction experiments
performed by Oze et al (2003) Serpentine minerals could also provide minor Cr
concentrations to the soils however soil extractions have consistently shown that a
majority of the Cr is in highly resistant phases capable of resisting dissolution even in
concentrated hydrochloric acid nitric acid and hydroxylamine with acetic acid (Table
42) Chromium end-member minerals including eskolaite Cr-muscovite (fuchsite) Cr-
garnet (uvarovite) Cr-epidote (tawmawite) and Cr-chlinochlore (kaumlmmererite) are also
117
included in Table 47 to demonstrate that relatively low abundances of these minerals are
all that is necessary to increase Cr concentrations in rocks and soils Chromium-
clinochlore is often observed developing adjacent Cr-spinels in serpentinites and is
perhaps a more accessible mineral source of Cr in the serpentinites compared to Cr-
spinels (Phillips et al 1980 Schreyer et al 1981 Kerrich et al 1987 Treloar 1987a
Pan and Fleet 1991 Schandl and Wicks 1993 Christofides et al 1994 Challis et al
1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Oze et al 2003)
As shown in Table 44 and Figures 43 and 46 the chemical properties (pH EC
N etc) of serpentine soils are not correlated to Cr concentrations measured in the soil
solids (Tables 41 Figure 45) due to Cr-spinels accounting for a majority of the Cr
(Figure 49 and 410 Tables 46 and 47) However extraction experiments and soil
solution analyses demonstrate that a limited fraction of the total Cr (lt1) is extractable
If only 1 of the Cr in a serpentine soil is extractable a serpentine soil containing 10000
Cr mg kg-1 could possibly contribute up to100 Cr mg kg-1 still a significant Cr
concentration The chemical properties of the soil such as pH will influence this limited
fraction of the Cr in the soils Acidic pHs may aid in mobilizing Cr(III) however this
will only occur at low pHs (lt4) which exceed the range of most serpentine soils (Figures
43 and 46) pH values greater than 8 like those in some serpentine soils and
serpentinite groundwater may help sustain Cr(VI) concentrations in solution (Figure
43) In addition to pH organic matter in the soils may be a significant factor related to
the mobility of Cr due to the production of organic acids promoting the dissolution of the
Cr-bearing phases but particulate organic matter is known to bind Cr(III) in stable stable
organometallic complexes (Kaupenjohann and Wilcke 1995 Brady and Weil 1999)
Evaluating the valence state of Cr is crucial for understanding the chemistry as
well as the potential toxicity related to serpentine soils The soil solids do not contain
detectable concentrations of Cr(VI) (Figure 48) however Cr(VI) concentrations are
present in soil solutions and groundwater at values lt30 microg L-1 (Gough et al 1989
Becquer et al 2003) A majority of serpentinite Cr-bearing minerals only contain
Cr(III) therefore there must be a reaction or process by which Cr(III) is oxidized in a
water saturated soil High valence Mn oxides such as birnessite are present within
serpentine soils and provide a feasible explanation for Cr(III) oxidation Additionally
118
Mn oxides or other strong oxidants in the soil might accelerate the weathering of Cr-
bearing minerals including chromite and generate the Cr(VI) concentrations observed in
soil solutions and groundwater After Cr(III) is oxidized Cr(VI) is a chemically mobile
species and would be transported away from the serpentine soils before it can be
observed in the solid phase Additionally Cr(VI) may be subsequently reduced by
reacting with organic matter or Fe(II)-bearing minerals such as magnetite which are
present in a majority of serpentine soils Perhaps the best way to evaluate the potential of
serpentine soils being a non-anthropogenic source of Cr contamination is to monitor
Cr(III) and Cr(VI) in the soil solutions over time as demonstrated by Gasser and
Dahlgren (1994)
It is curious that plants do uptake Cr in serpentine soils especially when Cr is
tightly bound in the soil matrix (Table 42) In order for plants to uptake Cr it must be in
a form to be absorbed by plant roots Soil pH in the rhizosphere can be altered by plant
roots by as much as 2 units within two days (Fischer et al 1989) This acidification is
known to increase the solubility of the minerals in the soils and could serve as a
mechanism by which plants can incorporate this relatively insoluable element Another
possibility is that Cr(III) in the soil is oxidized by either organic or inorganic means and
then incorporated into the plant in its more chemically mobile and bioavailable form
Cr(VI) How and why plants incorporate as much Cr as they do needs to be examined
more thoroughly
Chromium Spinel in Soils and Rocks
Compositions of chromite identified in serpentinites worldwide (Figure 44) and
Cr-bearing spinels identified in the six serpentine soils are compared in Figure 10 in order
to determine whether the soil spinels are geochemically similar to those in serpentinites
This comparison is ascertained by examining how Al Fe and Mg Wt values varied
with respect to Cr Wt Trends for magnetitehematite chromite and the
microcrystalline spinel-silicate mixture observed in the soils are noted in each diagram in
Figure 410 Additionally the highest Cr concentrations for soil chromite
magnetitehematite and the microcrystalline spinel-silicate mixture are ~40 Cr Wt
~25 Cr Wt and ~15 Cr Wt respectively Based on Al Wt soil
119
magnetitehematite deviates significantly from the chromite trend with Al Wt values
lt5 Magnetitehematites are enriched with Fe relative to serpentinite and soil chromite
and Mg concentrations in the soil chromite are slightly higher than serpentinite chromite
Soil magnetitehematite and chromite are similar by demonstrating a linear trend when
Al Fe and Mg are summed The microcrystalline spinel-silicate mixture (CSM) does
not follow this trend because of Si present in this mineral mixture Overall soil
chromites are similar to serpentinite chromite and soil magnetiteshematites can be
differentiated from the chromites when Cr Wt is plotted against Al Fe and Mg Wt
CONCLUDING STATEMENT
Serpentine soils are abundant among convergent margins and the chemistry of
these soils is ultimately a function of the ultramaficserpentinite rock mineralogy and
texture as well as all the common soil-forming factors Although ultramafic rocks and
serpentinites produce groundwater that is highly basic most serpentine soil pHs are
slightly acidic Although Proctor and Nagy (1992) argued that it is futile to categorize
serpentine soils based on chemistry and biology (otherwise referred to as the lsquoserpentine
factorrsquo) as proposed by Brooks (1987) Cr concentrations greater than 200 mg kg-1 in
serpentine soils are consistently present Chromium in the soils originates from minerals
in the bedrock where the valence is primarily Cr(III) the exception being Cr(VI) in the
rarely present mineral crocoite (PbCrO4) Increased Cr-concentrations in serpentine soils
are largely the result of chromite or other resistant Cr-bearing minerals causing a
localized enrichment in the soils Although Cr in serpentine soils is not readily released
from the soil matrix serpentine soil vegetation can incorporate Cr concentrations as high
as 700 mg kg-1 (Table 34) Finally the limited extractability of Cr and site dependent
factors such as the presence of organic matter clays Fe(II)-bearing minerals pH and the
poor drainage indicates that the possible formation and sustained presence of Cr(VI) is
not a portentous matter related to serpentine soils
120
ACKNOWLEDGEMENTS
We appreciate technical and field support from Robert Jones Anders Meibom
Rob Dunbar and Guangchao Li Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund GSA Graduate Student Research Grants and the
Office of Teaching and Learning at Stanford University is gratefully acknowledged
121
Table 41 Measured Cr concentrations in serpentine soils worldwide
Location Cr concentration (mg kg-1) Reference
Canyon Creek area OR 920 (White 1971) 66-13200 (Brooks et al 1992)
Ireland 29-5000 (Proctor and Woodell 1971 Jeffrey 1992) Italy 1300-3900 (Gambi 1992) Japan 5200 (Shewry and Peterson 1976) Maryland 1500-6000 (Rabenhorst et al 1982) Mt Tamalpais CA 1700 (White 1971) New Caledonia 6300-80000 (Jaffre 1980) New Caledonia 2-6 (Schwertmann and Latham 1986) New Idria CA 1640 (White 1971) New Zealand 850-3250 (Lee 1992) Newfoundland Canada 309-1447 (Roberts 1992) Nickel Mountain OR 1730 (White 1971) Santa Barbara County CA 2080 (Woodell et al 1975) Southern Africa 045-14 (Morrey et al 1992) Tehama County 023-12 (Gough et al 1989) Transvaal Lowveld South Africa 120-800 (Cole 1992) Unst Shetland 3000-4000 (Proctor 1969) Wenatchee Mnts WA 895 (White 1971) Zimbabwe 110-46000 (Soane and Saunder 1959 Proctor et al 1980)
Goias State Brazil
122
123
Table 42 Summary of Cr Fe and Ni extractions from serpentine soils
Range of Extractable Cr Fe and Ni Location Cr Fe Ni
Method Reference
Italy 2-5 microg g-1 305-640 2300-2700microg g-1 25 CH3COOH (Gambi 1992)
NE Portugal 02-09 microg g-1 120-540 microg g-1 40-290microg g-1 EDTA (Menezes de Sequeira and Pinto da Silva 1992) New Caledonia 22-23 384-581 02-216 DCB (dithionite-citrate bicarbonate) (Schwertmann and Latham 1986) New Caledonia 02-11 microg g-1 02-66microg g-1 Method Unknown (Jaffre 1980)
Norway 357microg g-1 316microg g-1 HClO4HNO3 (concentrated) (Zeien and Bruemmer 1989) Norway 254microg g-1 134microg g-1 Ascorbic and Oxalate Acid pH=325 (Zeien and Bruemmer 1989)
Norway 62 microg g-1 53microg g-1 NH4-EDTA (Zeien and Bruemmer 1989) Norway 22-60 microg g-1 8-25 mg kg-1 1-7 microg g-1 pH-stat reactor EDTA (Kaupenjohann and Wilcke 1995)
Swiss Alps 08 micromol L-1 28micromol L-1 228microg g-1 Lysimeter (Nonfiltered) (Gasser and Dahlgren 1994) Trinity Ophiolite CA 004-045 gkg 27-210 g kg-1 005-245microg g-1 Citrate-Dithionite Extracts (Alexander et al 1989)
Unst lt01 microg g-1 19-77microg g-1 Ammonium Acetate (Shewry and Peterson 1976)
Unst 500-1390microg g-1 1540-3330microg g-1 HNO3-Soluble (Shewry and Peterson 1976) JRBP CA 0 microg g-1 01microg g-1 103microg g-1 BaCl2 Method (Oze et al 2003)
JRBP CA 78 microg g-1 513 microg g-1 150 microg g-1 Hydroxlyamine+Acetic Acid (Oze et al 2003) JRBP CA 31 microg g-1 2039microg g-1 35microg g-1 10M Hydrofluoric Acid (Oze et al 2003)
123
124
Table 43 Serpentine soil plant Cr concentrations
Plant Cr Concentration (microg g-1) Location Reference
Lophostachys villosa 203 Goias State Brazil (Brooks et al 1992) Ruellia geminiflora
56 Goias State Brazil (Brooks et al 1992) Minuartia ophiolitica 83 Italy (Gambi 1992) Sanguisorba minor 194 Italy (Gambi 1992) Plantago lanceolata 135 Italy (Gambi 1992) Cerastium latifolium 110 Italy (Gambi 1992) Funaria flavicans 34-899 Maryland (Shaw 1991) Avena sativa 168-583 New Zealand (Lee 1992) Aerobryopsis longissima 200-700 New Caledonia (Lee et al 1977) Arenaria humifusa 530 Newfoundland Canada (Roberts 1992) Senecio pauperculus 406 Newfoundland Canada (Roberts 1992) Rhalomitrium lanuginosum 246 Newfoundland Canada (Roberts 1992) Santolia semidentata 95 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Asplenium cuneifolium 228 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992) Armeria eriophylla 38 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Arctostaphylos viscida 029-705 Tehama County CA (Gough et al 1989) Calluna vulgaris 417 Unst (Shewry and Peterson 1976) Armeria maritima 149 Unst (Shewry and Peterson 1976) Agrostis stolonifera 120 Unst (Proctor and Woodell 1971) Dicoma niccolifera 77 Zimbabwe (Proctor 1992) Xerophyta equisetoides 68 Zimbabwe (Proctor 1992)
124
125
Table 44 Chemical and textural properties of the collected soils
Ammonium Acetate Extractable ------------------------Hydrometer--------------------- Ca K Mg Na
Sand Silt ClaySample pH EC dSm
Organic Matter ------
LECO Nitrogen ------ -------------------mg kg-1-------------------
------ ------ ------
Texture
New Caledonia 534 01 05 lt001 279 29 692 44 43 32 25 LoamHawaii 604
03 10 002 839 46 168 53 8 26 66 ClayEight Dollar Mnt 651 07 16 006 599 62 3464 25 28 28 44 Clay
Nickel Mnt 665 05 07 001 167 42 3289 17 24 32 44 ClayHarvard Mine 691 03 11 003 998 301 2897 43 41 30 29 LoamJasper Ridge 671 04 32 012 505 237 3291 36 36 31 33 Clay LoamPillikin Mine 642 04 05 002 185 55 3421 29 45 23 32 Sandy Clay Loam
125
126
Table 45 Total major and trace element concentrationsdagger of the collected soils
Element Concentration New Caledonia Hawaii Eight Dollar
Mnt Nickel Mnt Harvard Mine Jasper Ridge Pillikin
Mine Ag mg kg-1 lt05 lt05 lt05 lt05 02 lt05 lt05Al 931 1325 314 095 179 314 179Ba mg kg-1 lt10 180 40 20 30 230 lt10Be mg kg-1 05 16 05 lt05 lt05 08 lt05Ca 003 012 085 018 007 079 002Cd mg kg-1 lt05 lt05 18 lt05 05 lt05 lt05Co mg kg-1 56 183 338 403 85 155 122Cr mg kg-1 827 388 9528 5761 2900 5976 1977Cu mg kg-1 78 248 43 25 51 33 4Fe 638 1296 1635 17 538 727 468K 002 046 004 003 004 039 001
Mg 072 026 753 829 792 1025 gt150Mn mg kg-1 257 2706 3017 2554 920 1664 903Na 001 008 005 001 003 057 lt001Ni mg kg-1 2780 345 5510 14100 1895 2180 2760P mg kg-1 50 260 160 150 130 140 20Pb mg kg-1 6 31 4 3 2 8 6Ti 016 041 008 002 lt001 02 002V mg kg-1 63 132 158 54 65 92 27W mg kg-1 10 10 10 10 lt10 lt10 lt10Zn mg kg-1 43 218 160 132 66 111 47
126
dagger ICP-AES analyses
127
Table 46 Compositionsdagger for soil Cr-containing minerals identified in Figure 49
Wt New Caledonia A
Hawaii B
Eight Dollar Mnt C
Eight Dollar Mnt D
Nickel Mnt E
Harvard Mine F
Jasper Ridge G
Jasper Ridge H
Pillikin Mine I
SiO2 491 008 175 026 002 000 1940 001 000Al2O3
756 185 272 2235 2041 088 559 1575 839NiO 023 005 001 006 011 2054 014 002 007FeO 7766a 6212 6633b 3420b 1784 2445 2629 2314 3105MnO 004 037 181 123 133 031 155 146 216Cr2O3 165 007 2387 3343 4682 3991 1878 5024 5355MgO 048 121 202 653 1315 1062 1668 926 497CaO 001 000 013 008 000 000 015 000 000Na2O 000 000 002 001 000 000 006 000 000TiO2 004 2465 012 006 012 000 000 003 018
TOTAL 9257 9041 9879 9821 9980 9671 8864 9920 10037Mineral Cr-Hydroxide Ulvoumlspinel Cr-Magnetite Chromite Mg-Chromite Trevorite CSM1 Chromite Chromite
127
dagger Electron microprobe analyses This is the form of iron unless otherwise stated a FeO is in the form Fe2O3 b FeO is in the form of Fe3O4 1CSM Microcrystalline chromitesilicate mixture
128
Table 47 Chromium-bearing mineraldagger masses and volumes required to influence Cr concentrations in soils
Cr Concentration Mineral mass (grams) required for 1000 g bulk soil analysis Mineral Mass Percent () (mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite
500 11 07 22 24 48 57 011 007 022 024 048 057 1000 22 15 43 48 95 115 022 015 043 048 095 1152000 43 29 86 96 191 230 043 029 086 096 191 2305000 108 73 216 241 477 574 108 073 216 241 477 574
Cr Concentration Mineral Volume Percent () in Serpentine Mineral Volume Percent () in Quartz
(mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite500 005 004 02 02 03 15 006 004 02 02 04 06
1000 01 007 04 03 07 30 01 007 04 03 07 112000 02 014 07 06 14 59 02 015 07 06 14 225000 06 036 18 16 34 147 06 037 18 16 35 55
128
daggerMineral formulas eskolaite Cr2O3 chromite FeCr2O4 fuchsite KCr2(AlSi3O10)(OH)2 uvarovite Ca3Cr2Si3O12 tawmawite Ca2CrAl2Si3O12(OH)2 kaumlmmererite Mg5CrAlSi3O11(OH)8
Note Molar volumes used in the calculations above are from their Al-enriched counterparts in Robie and Hemingway (1995) The molar volume for kaumlmmererite was estimated using unit cell volumes from XRD unit cell measurements from Robie and Hemingway (1995) Christofides et al (1994) and Phillips et al (1980)
129
Figure Captions
Figure 41 Selected serpentine soil localities as well as soils collected for analyses are
shown on a world map Active plate boundaries are shown Serpentine soil localities not
adjacent the plate boundaries on this map are related to ancient plate boundaries that are
no longer active
Figure 42 Field pictures of soils collected from a) Eight Dollar Mountain b) Nickel
Mountain c) Harvard Mine d) Pillikin Mine and f) Jasper Ridge Biological Preserve
Figure 43 Stability field diagram of Cr under a range of Eh and pH conditions (log a
Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom (1998) Adsorption
isotherms for Cr3+ and CrO42- on hydrous ferric oxide (HFO) are also shown (Leckie et
al 1980 Leckie et al 1984)
Figure 44 Microprobe analyses demonstrating Cr versus Al Wt for Cr-spinel
(chromite) and Cr-silicates including Cr-mica (fuchsitemariposite) Cr-garnet
(uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite) (Whitmore et al
1946 Leo et al 1965 Chen and Lee 1974 Neiva 1978 Cooper 1980 Phillips et al
1980 Grapes 1981 Schreyer et al 1981 Max et al 1983 Nutman et al 1983 Raase
et al 1983 Chao et al 1986 Von Knorring et al 1986 Kerrich et al 1987 Treloar
1987b a Morand 1990 Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen
1999 Proenza et al 1999) Chromium-spinel and Cr-silicate compositions based on
octahedral Cr and octahedral Al are also shown with tie-lines (dotted) denoting ideal
isomorphic substitution between Cr and Al end-members
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above Each dotted line connecting the
points represents a separate soil profile at each site
129
Figure 46 Ranges of pH for serpentine soils and related groundwater worldwide Soil
pH values were obtained from A(Kruckeberg 1992) B (Graham et al 1990) C
(Alexander et al 1989) D (Gough et al 1989) E (Rabenhorst et al 1982) F
(Roberts 1992) G (Bulmer et al 1992) H (Proctor 1992a) I (Jeffrey 1992) J
(Gambi 1992) K (Verger 1992) L (Proctor and Cole 1992) M (Cole 1992b) N
(Morrey et al 1992) O (Proctor 1992b) P (Mizuno and Nosaka 1992) Q (Cole
1992a) and R (Lee 1992) Groundwater pH values from serpentinite are from Barnes
and ONeil (1969) and Barnes et al (1972) Note the range of pH environments possible
for ultramafic rocks and their related soils
Figure 47 Soil pictures with Munsel color values and names from soils collected from
a) New Caledonia b) Hawaii c) Eight Dollar Mountain d) Nickel Mountain e) Harvard
Mine f) Jasper Ridge Biological Preserve and g) Pillikin Mine
Figure 48 Cr XANES spectra for soils analyzed from Jasper Ridge New Caledonia
Pillikin Mine Eight Dollar Mountain and Harvard Mine The pre-edge area indicative of
Cr(VI) peaks is noted in the diagram No definitive Cr(VI) peaks are present in any of
these spectra
Figure 49 Backscattering images of mineral phases within each soil are presented
Electron microprobe analyses are noted by a circle and letter for the Cr-bearing minerals
and the results are listed in Table 46
Figure 410 Chromite magnetite and hematite compositions from the collected soils
(gray circles) as well as analyses for serpentinite chromite (white squares) from Figure
44 are shown in these diagrams with Cr Wt versus Al Fe Mg and Al+Fe+Mg Wt
130
Figure 41
131
Figure 42
132
Figure 43
133
Cr Wt
0 10 20 30 40 50
Al W
t
0
5
10
15
20
25Cr-SpinelCr-MuscoviteCr-EpidoteCr-GarnetCr-Chlorite
Cr-SpinelBased on 4 Oxygens
Cr0 1 2
AlVI
0
1
2
ChromiteFeCr2O4
HercyniteFeAl2O4
Cr-MuscoviteBased on 11 Oxygens
Cr0 1 2
AlVI
0
1
2
MuscoviteKAl2(AlSi3O10)(OH)2
FuchsiteKCr2(AlSi3O10)(OH)2
Cr-GarnetBased on 12 Oxygens
Cr0 1 2
AlVI
0
1
2
GrossularCa3Al2Si3O12
UvaroviteCa3Cr2Si3O12
Cr-EpidoteBased on 125 Oxygens
Cr0 1
AlVI
2
3
ClinozoisiteCa2Al3Si3O12(OH)2
TawmawiteCa2CrAl2Si3O12(OH)2
Cr-ChloriteBased on 14 Oxygens
Cr0 1
AlVI
0
1
2
Kammererite(Mg5Cr)(AlSi3)O10(OH)8
Clinochlore(Mg5Al)(AlSi3)O10(OH)8
Figure 44
134
Maryland (USA)(Rabenhorst et al 1982)
Chromium (mg kg-1)
0 1000 2000 3000 4000 5000 6000 7000
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Tehama County CA (USA)(Gough et al 1989)
Chromium (mg kg-1)
0 2000 4000 6000 8000 10000 12000 14000
Dep
th (m
eter
s)
00
02
04
06
08
Kalamath Mountains CA (USA)(Hotz 1964)
Chromium (mg kg-1)
2000 4000 6000 8000 10000 12000 14000 16000
Dep
th (m
eter
s)
00
05
10
15
20
25
30
35
40
45
50
Newfoundland (Canada)(Roberts 1992)
Chromium (mg kg-1)
800 1000 1200 1400 1600 1800
Dep
th (m
eter
s)
00
02
04
06
08
Harvard Mine CA (USA)(Unpublished Data)
Chromium (mg kg-1)
0 500 1000 1500 2000 2500
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Jasper Ridge CA (USA)(Oze et al 2003)
Chromium (mg kg-1)
1500 2000 2500 3000 3500 4000 4500 5000
Dep
th (m
eter
s)
00
01
Figure 45
135
LocationsZ A B C D E F G H I J K L M N O P Q R S
pH
0
1
2
3
4
5
6
7
8
9
10
11
12
Nor
th A
mer
ica
Can
ada
Brit
ain
Italy
Far E
ast
Japa
n
Zim
babw
e
Sout
h A
fric
a
Wes
tern
Aus
tral
ia
New
Zea
land
Mg-Bicarbonate Waters pH=8
Ca-Hydroxide Waters pH=11-12
Average RainpH=57
Sout
h A
fric
a
Italy
Irela
nd
Cal
iforn
ia
Cal
iforn
ia
Brit
ish
Col
umbi
a
Mar
ylan
dCal
iforn
ia
Figure 46
136
Figure 47
137
-2
-1
0
1
2
3
4
5
5980 5990 6000 6010
Energy (eV)
Nor
mal
ized
Abs
orba
nce
New Caledonia
Pillikin Mine
Eight Dollar Mnt
Jasper Ridge
Harvard Mine
Cr(
VI)
Figure 48
138
Figure 49
139
Cr Wt 0 10 20 30 40 50
Al W
t
0
10
20
30 Soil Serpentinite
Cr Wt 0 10 20 30 40 50
Fe W
t
0
20
40
60
80
Cr Wt 0 10 20 30 40 50
Mg
Wt
0
2
4
6
8
10
Cr Wt 0 10 20 30 40 50
Al +
Fe
+ M
g W
t
0
20
40
60
80
MagnetiteHematite
Chromite
MagnetiteHematite
Chromite
Chromite
MagnetiteHematite
SpinelSilicate
SpinelSilicate
SpinelSilicate
SpinelSilicate
Figure 410
140
- Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip1
-
- Chapter 3 Oxidative Promoted Dissolution of Chromite by
- Manganese Dioxide and Concurrent Production of Ch
-
- Chapter 4 Chromium Geochemistry of Serpentine S
-
- List of References
-
- Chapter 2
- Chapter 3
-
- Chapter 4
-
- Appendix A
- Chapter 2
-
- Chapter 3
-
- Chapter 4
-
- Appendix A
-
- CHAP1pdf
-
- Toxicity and Regulations of Chromium
- Mineralogy of Chromium in Serpentinites
-
- Chapter Summaries
-
- CHAP2Tablepdf
-
- JR305CLAY Depth 0-5cm
- JR3515CLAY Depth 5-15cm
- JR33045CLAY Depth 30-45cm
-
- HCAP2Figspdf
-
- Figure 27
-
- CHAP3Tablepdf
-
- Figure 36
- Figure 37
-
- CHAP4pdf
-
- ABSTRACT
- REVIEW
-
- Chromium Geochemistry
-
- Soil Organic Matter
-
- CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
-
- BedrockProtolith Mineralogy
- Soil Chemistry and Physical Properties
- Soil Mineralogy
- Backscattered electron images and microprobe analyses were used to identify the Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49 and corresponding Cr-bearing mineral compositions are listed in Table 46 The New C
-
- DISCUSSION
-
- Chromium Spinel in Soils and Rocks
-
- CONCLUDING STATEMENT
- ACKNOWLEDGEMENTS
-
- CHAP4Tablepdf
-
- Ulvoumlspinel
-
I certify that I have read this dissertation and that in my opinion it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy
____________________________ Dennis K Bird Principal Adviser
I certify that I have read this dissertation and that in my opinion it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy
____________________________ Scott Fendorf Principal Adviser
I certify that I have read this dissertation and that in my opinion it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy
____________________________ Robert Coleman
Approved for the University Committee on Graduate Studies
iii
ABSTRACT
Chromium persists in the environment as Cr(III) a non-hazardous species and a
micronutrient in human nutrition or as Cr(VI) a strong oxidizing agent a toxin to living
cells and a Class A human carcinogen by inhalation Serpentinites or metamorphosed
ultramafic rocks are distributed worldwide and contain Cr concentrations typically
greater than 200 mg kg-1 Additionally serpentine soils a generic term used to describe
any soil derived from serpentinite regardless of its physical or chemical properties
contain elevated Cr concentrations often exceeding their corresponding protolith
Chromium in serpentinites and serpentine soils is dominantly in the form of Cr(III)
however geochemical interactions during serpentinite weathering and soil formation
provide oxidative pathways allowing the formation of Cr(VI) The relationship between
rock and soil and the processes governing the Cr chemistry are difficult to assess due to
the inherent complexity of evaluating and correlating the mineralogy and geochemistry of
both systems Ultimately a multidisciplinary approach combining field analytical and
laboratory studies is necessary to evaluate the Cr geochemistry related to these rocks and
soils and to determine whether or not serpentinites and serpentine soils are sources of
non-anthropogenic Cr contamination
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of
California produces serpentine soils containing high concentrations of Cr as well as other
potentially toxic elements including Ni Co and Mn Chromium concentrations in
serpentine soils from Jasper Ridge Biological Preserve in the Central Coast Range are as
high as 4760 mg kg-1 nearly three times greater than the serpentinite protolith (~1800
mg kg-1) Chromium-containing minerals within the bedrock include chlorite (~03 Cr
Wt ) enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt )
magnetite (82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate
phase containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
iv
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
Despite the resistance of chromite to weathering Cr release and oxidation from
chromite is a potential environmental hazard in sediments and soils and a pathway for
soil development (weathering of primary minerals) related to ultramafic rocks and their
metamorphic derivatives (serpentinites) Birnessite is a common pedologic mineral in
these sediments and soils capable of oxidizing aqueous Cr(III) The interaction between
chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4) and birnessite was investigated with
an interest in the potential generation of Cr(VI) The rate of Cr(VI) in solution increases
with increasing chromite suspension densities and decreasing pH but is independent of
birnessite suspension densities at values greater than ~20 m2 L-1 at 25degC The overall rate
expression of Cr released and oxidized in solution from chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression--rates ranging from 9times10-4 to 44times10-3 microM Cr(VI) h-1 These
experiments therefore demonstrate that serpentine soils are a potential source of non-
anthropogenic Cr(VI)
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
v
chemistry of Cr in serpentine soils including its protolith is reviewed with a focus on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils have widely varying pH contain a variety Fe(III) oxides (magnetite and
hematite) phyllosilicates (serpentine and chlorite) and clays (smectites and
vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1) and
Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Additionally Cr(III) is the only valence state
observed in the serpentine soil solids however Cr(VI) has been identified in New
Caledonia and California serpentine soil solutions at concentrations below 30 microM The
enrichment and range of Cr concentrations in serpentine soils are directly related to the
presence of Cr-spinels specifically chromite and Cr-magnetite These spinels are
resistant to weathering and are preserved in the soil environment however oxidation of
Cr(III) from Cr-spinels by high valent Mn oxides or other strong oxidants is a potential
source of Cr(VI) identified in serpentine soil solutions Due to the chemically and
physically resistant nature of the Cr-spinels Cr-bearing silicates including clay minerals
Cr-chlorite Cr-garnet Cr-mica and Cr-epidote are more viable sources of Cr identified
in vegetation soil extractions soil solutions and related waters
vi
PREFACE
The chemistry and mineralogy of Cr related to serpentinites and serpentine soils
are assessed in this dissertation using a multidisciplinary field analytical experimental
and theoretical approach The four chapters that constitute this dissertation describe my
investigations related to these rocks and soils Additionally the common theme shared
by these chapters is that they all examine the Cr mineralogy in serpentinites and
serpentine soils and the potential of Cr to be mobilized and oxidized in near-surface
environments Chromite a Cr-bearing spinel is the dominant source of Cr in
serpentinite and serpentine soils Although chromite is resistant to weathering the
presence of birnessite a high valent Mn oxide will increase the release of Cr and the
production of Cr(VI) from chromite over a range of pH Chromium-clinochlore or
kaumlmmererite contains Cr concentrations less than 3 Wt however the rate at which Cr-
clinochlore weathers compared to chromite makes this mineral a more probable source of
Cr released into soil solutions These types of concerns are addressed in this thesis Brief
descriptions of the four individual dissertation chapters including their publication status
and a list of the co-authors are presented in the following paragraphs
Chapter 1 presents the main objectives of this dissertation The chemistry
regulations and toxicity of Cr are discussed as well as the general characteristics of
serpentinites and serpentine soils Additionally a brief summary of each chapter is
provided
Chapter 2 focuses on serpentinites and serpentine soils at Jasper Ridge Biological
Preserve Stanford CA The results have been submitted for publication in an article
entitled ldquoChromium geochemistry of serpentinized ultramafic rocks and serpentine soils
in the Franciscan Complex of Californiardquo to the American Journal of Science (March
2003) Scott Fendorf Dennis Bird and Robert Coleman are the second third and fourth
co-authors respectively
Chapter 3 is an experimental study evaluating the release of Cr and the
production of Cr(VI) from chromite in the presence of birnessite over a range of pH The
results for this study will be submitted for publication in an article entitled ldquoOxidative
promoted dissolution of chromite by manganese dioxide and concurrent production of
vii
chromaterdquo to Environmental Science and Technology The co-authors for this paper are
Scott Fendorf and Dennis Bird
Chapter 4 is a review paper that critically evaluates the chemistry of serpentine
soils and serpentinites primarily focusing on Cr geochemistry Additionally analytical
studies of serpentine soils collected from California Oregon and New Caledonia are
presented This study will be presented at the Robert Coleman Symposium and has been
submitted as an article entitled ldquoChromium geochemistry of serpentine soilsrdquo for an
International Geology Review Special Publication Scott Fendorf Dennis Bird and
Robert Coleman are the second third and fourth co-authors respectively
Appendix A is a preliminary investigation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite) a mineral common in serpentinites and present in serpentine
soils This appendix is entitled ldquoEstimation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite)rdquo The results of this chapter will be included in a study that
estimates the thermodynamic properties of Cr-bearing silicates including fuchsite (Cr-
muscovite) tawmawite (Cr-epidote) and uvarovite (Cr-garnet)
viii
ACKNOWLEDGEMENTS
Serpentinites serpentine soils and chromium are the Geologic Trinity of this
dissertationhellipeach topic is intimately related to one another The scope and breadth of
this research required a significant amount of time and funding None of this research
would have been possible without the generous support of several foundations
organizations and Stanford University I am grateful for the financial aid I have
received from the Stanford Fellowship the Stanford University McGee and Shell funds
the National Science Foundation the Geologic Society of America Graduate Research
Fund the Stanford Office of Teaching and Learning grant and the Jasper Ridge
Biological Preserve Melon Grant
I would not be where am I today if it were not for my parents Robert and Sue
Oze and my brother Rob Oze Thank you for the sandbox and all the hose-driven
volcanoeshellipthose Wookies never stood a chance While some families vacation in theme
parks my family frequented gem and mineral conventions collected rocks and shells
from mist-shrouded shores and sought lucky stones including an infamous quartz orb
Sometimes I think I was destined to be a geologist I would also like to thank everyone
in my family and all our family friends who have provided that extra little push and
support along the way These people include Tudor Bennatts Louis Oze Margaret and
John Chambers John Bennatts Nancy Skiles the Izumis the memory of Witchie Poo
hijacking Harleys from Bremerton bikers and the Man in the Sound waiting for that one
last catch Close friends I would like to acknowledge who are not related to the Stanford
crowd include Dave and Leigh Anderson and Rebecca Parker These satellite friends
helped me keep in touch with the world outside of academia and proved to me that I
wasnrsquot meant to have a real job A special thank you goes to Katherine Akako
Izumihellipyou have made this whole trip worthwhile
My PhD advisors Scott Fendorf and Dennis Bird provided an invaluable amount
of help patience and energy over the past five years Scott always made the time to
discuss and refine many of my projects making them relevant to many current
geochemical issues Additionally he helped me hone my soil science skills in the field
and in the lab Out in the field Scott can dig a pretty mean hole Dennis devoted a
significant amount of time and sweat in order to make this dissertation happen His
ix
insightful knowledge of geochemistry and of the geology of the Sierra Nevadas provided
the cornerstone for this dissertation
Robert Coleman played a key role in the production of this dissertation He
always knew what to say at the right time and helped me understand the complexity of
ophiolites and serpentinites through experienced eyes Additionally Keith Loague and
Gary Ernst deserve recognition for their outstanding support as committee members for
this dissertation over the last four years I would also like to thank Jonathan Stebbins and
Gail Mahood for providing me the opportunity to attend Stanford
Bob and Vicki Jones always offered an open ear whether it was science or solving
the worldrsquos problems Rick Hazlett (Pomona College) and John Winter (Whitman
College) both had a significant impact on shaping my future science goals and objectives
during my undergrad years and both encouraged me to pursue a career in geology They
were the catalyst for this whole experiment
I owe a lot to my research group members (both the Bird Cage and Soil
Chemistry Group) and those people who didnrsquot mind me tagging along Before I
introduce these people one person has been my friend and co-conspirator for many years
at Stanford Tim Copehellipyou are a pain in my ass Besides never picking up any of your
crap and breaking all my stuff this whole grad school experience would not have been
the same without you Other friends and acquaintances that will not be forgotten in
chronological order of me walking into my office for the first time on September 11
1998 include Anne Dyreborg Esra Inan Tom Bawden Patrick Redmond Phil
Neuhoff Thrainn Fredrickson Lara Heister Kaye Savage Andrea Fildani Mike Schultz
Trayle Kushan Kevin Theissen Angela Hessler Chris Van de Ven Kono Lemke Colin
Doyle Colleen Hansel Ben Bostick Matthew LaForce Nancy Grumet Scott Kroeker
Anders Meibom Brian Ebel Matt Ginder-Vogel Matt Pollizitto Spud Amy Weislogel
Travis Horton and Jake Waldbauer I would like to end this Acknowledgement section
and begin my dissertation with a haikuhellip
Better than outcrops
Are the people Irsquove met here
Thanks for all the beer
x
TABLE OF CONTENTS
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iv Prefacehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip vii Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip ix Table of Contentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xi List of Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xiv List of Illustrationshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xvi Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 1
Toxicity and Regulations of Chromiumhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 2 Mineralogy of Chromium in Serpentiniteshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 3 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5 Chapter Summarieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5
Chapter 2 Chromium Geochemistry of Serpentinized Ultramafic Rocks and Serpentine Soils in the Franciscan Complex of Californiahelliphelliphelliphelliphelliphellip 8
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 8 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 9 Backgroundhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 10
Geologic Background of the Serpentinite at Jasper Ridgehelliphelliphelliphellip 10 Serpentine Soils Characteristics and Reviewhelliphelliphelliphelliphelliphelliphelliphellip 11 Sources of Chromium ProtolithBedrockhelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 12 Chromium-Bearing Phases in Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 13
Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Location and Sample Collectionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Mineral Identification in Rocks and Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Soil Chemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 15 X-ray Diffraction and Clay-Size Mineral Identificationhelliphelliphelliphelliphellip 16 X-ray Absorption Spectroscopyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17 Sequential Extraction Procedurehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17
Resultshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Serpentinite Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Distribution of Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip21 Serpentine Soil Chemistry and Mineralogy at Site JR3helliphelliphelliphelliphellip 22 Sequential Extractionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 26
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 27 Concluding Remarkshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 33 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 44 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 47
xi
Chapter 3 Oxidative Promoted Dissolution of Chromite by Manganese Dioxide and Concurrent Production of Chromatehelliphelliphelliphelliphelliphellip 63
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 63 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 64 Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 65
Solid Phase Characterization and Synthesishelliphelliphelliphelliphelliphelliphelliphelliphellip 65 Extraction Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 66 Batch Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Solution Analysishelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69
Results helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite Extraction Experimentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite and Birnessite Batch Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 70 Secondary and Solid Phase Examination of Chromite and Birnessitehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 72 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 72
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 73 Chromium(VI) Pathways from Chromite and Birnessite into Solutionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 76 Serpentine Soils and Cr(VI)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 77
Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 79 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 80 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 84 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 87
Chapter 4 Chromium Geochemistry of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip100
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 100 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 101 Reviewhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102
Chromium Geochemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102 Mineralogy of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 103 Sources of Chromium for Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphellip 105 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 106 Soils Solutions and Related Waterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 108 Serpentine Soil Vegetationhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 110 Soil Organic Matterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 111
California Oregon and New Caledonia Serpentine Soilshelliphelliphelliphelliphelliphelliphellip 112 Sample Locations and General Soil Characteristicshelliphelliphelliphelliphelliphellip 112 BedrockProtolith Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 113 Soil Chemistry and Physical Propertieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 114 Soil Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116 Chromium and Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 117
xii
Chromium Spinel in Soils and Rockshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 119 Concluding Statementhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 120 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 121 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 122 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 129 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 131
Appendiceshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 141 Appendix Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 142 Appendix Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 143 Appendix A Estimation of the Thermodynamic Properties of Kaumlmmererite (Cr-Clinochlore)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip145
Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 145 Nomenclature and Crystal Chemistry of Cr-chloritehelliphelliphelliphelliphelliphelliphelliphelliphellip 146 Estimation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 147
Estimation of theStandard Molar Volume Using Unit Cell Measurementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of the Standard Molar Volume Heat Capacity and Entropyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of Standard Molar Enthalpy and Gibbs Free Energyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 150 Critique and Discussion of the Estimation Algorithms and SUPCRT92helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 151
Evaluation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 152 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 155 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 161 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 162
List of References
References to Chapter 1helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 166 References to Chapter 2helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 172 References to Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 179 References to Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 183 References to Appendix Ahelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 195
xiii
LIST OF TABLES Chapter 2 Table 21 Whole rock elemental concentrations of serpentinites chert and sandstone Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in
serpentinite rock at Jasper Ridge Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in
serpentine soil at Jasper Ridge Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25
and 26 Table 24 Bulk soil elemental concentrations for soils and bedrock collected along
the transect in Figure 21b Table 25 Chemical and physical properties of the soil at Site JR3 Table 26 Selected trace metal concentrations for each size fraction versus depth at
Site JR3 Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3 Table 28 X-ray diffraction d-spacings for common silicates Table 29 Results of the Diablo clay serpentine soil and chromite sequential
extraction experiments Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1
kg of sample Chapter 3 Table 31 Chromite and birnessite masses and concentrations used for each batch
experiment Table 32 Serpentine soil and birnessite experimental parameters Table 33 Summary of Cr(VI) rates obtained from batch experiments Table 34 Comparative analyses of batch experiment solutions and chemical
extractions of chromite and birnessite solids Chapter 4 Table 41 Measured Cr concentrations in serpentine soils worldwide Table 42 Summary of Cr Fe and Ni extractions from serpentine soils Table 43 Serpentine soil plant Cr concentrations Table 44 Chemical and textural properties of collected soils Table 45 Total major and trace element concentrations of the collected soils Table 46 Compositions for soil Cr-containing minerals identified in Figure 49 Table 47 Chromium-bearing mineral masses and volumes required to influence Cr
concentrations in soils Appendix A Table A1 Cr and Al end-members in the Cr-clinochlore solid solution
xiv
Table A2 Standard state thermodynamic properties of Cr and Al-bearing minerals Table A3 Unit cell measurements and molar volumes of clinochlore and Cr-
clinochlores Table A4 Entropy volume and heat capacity for kaumlmmererite using the Ransom and
Helgeson Method Table A5 Fictive polyhedral considered in the estimations Table A6 Estimated thermodynamic properties for kaumlmmererite
xv
LIST OF FIGURES Chapter 2 Figure 21a Soils map of Jasper Ridge Biological Preserve CA Figure 21b Serpentinite and serpentine soil sampling sites at Jasper Ridge Biological
Preserve CA Figure 22 Chromium concentrations versus depth for four serpentine soils in
Maryland California and New Foundland Figure 23 Petrographic images of serpentinite at Jasper Ridge Figure 24 Elemental concentrations relative to Cr for chlorite pyroxene magnetite
chromite and a chromite-silicate mixture in the serpentinite at Jasper Ridge
Figure 25 Backscattering electron images of minerals in the serpentinite at Jasper Ridge
Figure 26 Backscattered electron images of Cr-chlorite and Cr-magnetite assemblage in the serpentinite bedrock
Figure 27 Highest and lowest Cr concentrations at each site in the serpentine soil transect
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected for total gt2 mm sand silt and clay size fractions
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soils at Jasper Ridge
Figure 210 X-ray diffraction clay analyses of sample JR3515 Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at specific depths Figure 212 X-ray microprobe elemental maps of Cr and Fe for the serpentine soil and
serpentinite at site JR3 Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments Figure 214 Scanning electron microprobe images of the chromite standard after each
sequential extraction experiment Figure 215 Tie-lines connecting serpentinite chromite and end-member chromite with
soil chromite values Chapter 3 Figure 31 Scanning electron microprobe images of chromite and birnessite standards Figure 32 Chromite extraction experiments evaluating Cr Fe and Al Figure 33 pH and Eh versus time (hours) for each chromite-birnessite chromite and
birnessite batch experiments Figure 34 pH and Eh with Cr stabilities Figure 35 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant chromite and variable birnessite at pH 5 Figure 36 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant birnessite and variable chromite at pH 5
xvi
Figure 37 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for variable pH and constant chromite and birnessite Figure 38 Rate of Cr(VI) production (d[Cr(VI)]dt) as a function of the surface area
concentration of chromite birnessite and pH Figure 39 SEM examination of solids following the termination of the batch
experiments Figure 310 Serpentine soil and birnessite experiments Figure 311 Iron analyses and the solubility of amorphous Fe(OH)3 and goethite Figure 312 Congruent dissolution of chromite based on Fe and Cr Figure 313 Three pathways of Cr release and oxidation Chapter 4 Figure 41 Serpentine soil localities worldwide Figure 42 Field site pictures of serpentine soils Figure 43 Stability field diagram of Cr over a range of Eh and pH conditions Figure 44 Chromium and Al Wt plot and octahedral Cr and octahedral Al plots
for chromite and Cr-silicates including Cr mica (fuchsitemariposite) Cr-garnet (uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite)
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (m) for selected serpentine soils
Figure 46 pH of serpentine soils and related groundwater worldwide Figure 47 Soil pictures with Munsel color values and names from collected soils Figure 48 Cr XANES spectra for collected serpentine soils Figure 49 Backscattering images of mineral phases within each serpentine soil Figure 410 Chromite magnetite and hematite compositions from analyzed soils as
well as analyses for chromite reported from around the world Appendix A Figure A1 Compositional variation of Cr2O3 and Al2O3 Wt in Cr-chlorite Figure A2 Compositional variation of Cr and VIAl in Cr-chlorite based on 14 oxygens Figure A3 Molar volume estimation of kaumlmmererite based on XRD unit-cell
parameters Figure A4 Natural and calculated compositions of co-existing Cr-clinochlore and Cr-
spinel solid solution in the system Mg0-FeO-Cr2O3-Al2O3-SiO2-H2O at 2 kbar pressure
xvii
CHAPTER 1
INTRODUCTION
Serpentinites or metamorphosed ultramafic rocks are distributed worldwide and
are commonly associated with ophiolite complexes (Coleman 1977 Coleman and Jove
1992) These rocks weather to form serpentine soils a generic term used to describe any
soil derived from serpentinite regardless of its physical or chemical properties
Serpentinites and serpentine soils contain Cr-bearing minerals the most prevalent being
chromite and Cr released from these minerals into surface and subsurface environments
is potentially toxic and carcinogenic The main objective of this dissertation is to
evaluate the mineralogy and chemistry of Cr related to serpentinites and serpentine soils
over geologic time using an integrated analytical experimental and theoretical approach
Additionally the potential of serpentinites and serpentine soils as sources of non-
anthropogenic Cr contamination is investigated A discussion of Cr and its significance
to serpentinites and serpentine soils begins with its discovery by a Frenchman
investigating an ore named ldquored leadrdquo from the Urals
Nicolas-Louis Vauquelin discovered Cr in 1797 in the mineral crocoite PbCrO4
(Burns and Burns 1975) The metalelement was named chromium after the Greek word
chroma meaning color due to its strong influence on mineral pigmentation (Burns and
Burns 1975) For example rubies and emeralds derive their color from trace amounts of
Cr in their structures More importantly JT Lowitz in 1798 discovered Cr in chromite
or Cr-spinel the primary source of Cr for industrial purposes (Burns and Burns 1975)
Chromium is a group VIB transition metal with two common oxidation states (Cr(III) and
Cr(VI)) an average crustal abundance of 100 mg kg-1 and an abundance of 513x105 in
the solar system (relative to [H]=1x1012) (Mason and Moore 1982)
The majority of Cr on the earth exists as Cr(III) where it is an immobile cation
that complexes strongly with organic matter and adsorbs on oxides and silicates even at
low pHs (pHgt4) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) Additionally
the chemical mobility of Cr(III) is limited due to the formation of highly stable
(oxy)hydroxides (Rai and Zachara 1988 Rai et al 1989 Ball and Nordstrom 1998)
1
Chromium(III) has the highest octrahedral site preference energy (OSPE) of any trace
element and it isomorphically substitutes into octahedral sites in oxides
(oxy)hydroxides sulfides and silicates Although Cr(III) is known to substitute for both
octahedral Al(III) and Fe(III) the strongest correlation for substitution appears to be for
Al(III) in oxides and silicates as demonstrated in Chapters 2 and 4 and Appendix A
Chromium(VI) is a strongly oxidizing and weakly adsorbing oxyanion in the form of
chromate (HxCrO42-x) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995 Shiraki
1997 Ball and Nordstrom 1998) Consequently Cr(VI) is chemically mobile and
bioavailable over a wide range of pH (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Chromium(VI) is tetrahedrally coordinated and behaves similarly to
sulfate in minerals such as jarosite (Baron and Palmer 1996)
The pathways and chemistry of Cr oxidation and reduction in soils and related
waters are covered in depth by Fendorf (1995) and Ball and Nordstrom (1998) The
reduction of Cr(VI) to Cr(III) may occur by reactions with Fe(II)aq Fe(II)-containing
minerals such as magnetite organic matter H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996) Oxidation of Cr(III)
to Cr(VI) related to serpentinite-related water and serpentine soil solutions is still a matter
of conjecture Chromium(III) oxidation in serpentine soils has been hypothesized to be
the result of Mn oxides such as birnessite as suggested by both Gough et al (1989) and
Becquer et al (2003) This idea is explored in depth in Chapter 3 where birnessite is
shown to influence the release and oxidation of Cr from chromite over a range of pH (3
to 8) at 25oC
Toxicity and Regulations of Chromium
The examination of Cr based on its oxidation state is necessary due to the
juxtaposing behavior and toxicity of Cr(III) and Cr(VI) Chromium(III) is an essential
micronutrient in human nutrition used by the body to process sugar protein and fat
(Vincent 2000) Inhalation of Cr(VI) is known to cause irritation in the nose including
symptoms such as nosebleeds (epistaxis) and ulcers andor perforations in the nasal
septum (Katz and Salem 1993 Kuo et al 1997) Ingesting considerable quantities of
Cr(VI) can result in stomach ulcers convulsions kidney and liver damage and death
2
(Katz and Salem 1993) Skin contact with certain Cr(VI) compounds causes allergic
contact dermatitis (ulcers) and irritant dermatitis (Katz and Salem 1993) Chromate
induced ulcers or chromeholes are typically red swollen crusted painless lesions
(Katz and Salem 1993) The nosology of Cr is beyond the scope of this dissertation
however Lees (1991) is an excellent source for further information Overall the
Environmental Protection Agency (EPA) the Occupational Safety and Health
Administration (OSHA) the World Health Organization (WHO) and the Department of
Health and Human Services (DHHS) have determined that that Cr(VI) compounds can
increase the risk of lung cancer primarily by inhalation as well as other forms of cancer
caused by ingestion
Federal and state regulation standards appropriate for both Cr(III) and Cr(VI) are
summarized and discussed by Proctor et al (1997) Human health action levels for Cr
vary for each state and the ranges for total Cr in soils include the following 200-66000
Cr(III) mg kg-1 (residential) 02-2000 Cr(VI) mg kg-1 (residential) 2500-1000000
Cr(III) mg kg-1 (non-residential) and 64-71000 Cr(VI) mg kg-1 (non-residential)
(Proctor et al 1997) Incidentally Cr concentrations in serpentine soils and serpentinites
(gt200 Cr mg kg-1) exceed many state action levels and would be considered toxic The
EPA has set a limit of Cr(III)+Cr(VI) at 100 microg L-1 in drinking water The OSHA has
set limits of 500 microg of water soluble Cr(III) compounds per cubic meter of workplace air
or 500 microg m-3 1000 microg m-3 for metallic Cr and insoluble Cr compounds and 52 microg m-3
for Cr(VI) compounds for 8-hour work shifts and 40-hour work weeks The limits
proposed above are constantly changing and should not be used for reference without
checking current regulation standards
Mineralogy of Chromium in Serpentinites
Chromium occurs in a variety of minerals either as a major or minor constituent
(Burns and Burns 1975) Chromium-bearing minerals in serpentinites are primarily
oxides and silicates in which the Cr is present as Cr(III) Chromite is a Cr-bearing spinel
abundant in serpentinites and is resistant to low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) The chemistry and
significance of chromite is addressed throughout this dissertation Pyroxenes such as
3
augite and enstatite are primary minerals identified in serpentinites and contain lt8 Cr Wt
(Grapes 1981 Von Knorring et al 1986 Treloar 1987a Challis et al 1995
Sanchez-Vizcaino et al 1995 Deer et al 1996) Chromate minerals such as crocoite
(PbCrO4) are rare in serpentinites and the occurrence of these minerals are limited to Pb
deposits in Australia Tasmania and the Urals (Williams 1974 Crane et al 2001)
Chromium(III) released from the primary Cr-containing phases may
isomorphically substitute into oxides and silicates for Al(III) and Fe(III) due to size
(octahedral radii Cr3+ = 0615 Aring Al3+ = 061 Aring and Fe3+=0643 Aring) and charge
similarities Magnetite formed during serpentinization incorporates Cr into its structure
and serpentine minerals (antigorite lizardite and chrysotile) contain trace concentrations
of Cr (Spangenberg 1943 Page and Coleman 1967 Burkhard 1993) Eskolaite (Cr2O3)
has been identified in Finland (Von Knorring et al 1986) Metasomatic alteration of
ultramafic rocks commonly results in the formation of Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite) uvarovite (Cr-garnet) tawmawite (Cr-epidote) and
kaumlmmererite (Cr-chlorite) (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Neiva 1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981
Max et al 1983 Nutman et al 1983 Raase et al 1983 Chao et al 1986 Von
Knorring et al 1986 Kerrich et al 1987 Treloar 1987b Treloar 1987a Morand 1990
Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993 Christofides et al 1994
Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Proenza et al
1999) Additionally kyanite and andalusite in the proximity of ophiolite complexes
incorporate up 5 Cr Wt (Cooper 1980 Raase et al 1983 Kerrich et al 1987)
Chapter 4 provides an in depth discussion of these Cr-bearing silicates as well as
demonstrating mineral compositions reported in the literature based on octahedral Cr and
Al Appendix 1 is a preliminary investigation devoted to estimating the thermodynamic
properties of kaumlmmererite These properties can then be incorporated into geochemical
modeling programs to begin accessing the stability of this common serpentinite mineral
over a range of temperatures and pressures
4
Chromium in Serpentine Soils
Evaluating Cr in serpentine soils requires an examination of both the mineralogy
and chemistry of the serpentinite and soil Measured Cr concentrations in serpentinites
and serpentine soils range from ~200 mg kg-1 to 6 Wt all of which is Cr(III) (Soane
and Saunder 1959 Proctor and Woodell 1971 Shewry and Peterson 1976 Jaffre 1980
Proctor et al 1980 Rabenhorst et al 1982 Schwertmann and Latham 1986 Brooks
1987 Gough et al 1989 Gambi 1992 Lee 1992 Morrey et al 1992) In Chapters 2
and 4 chromite and Cr-magnetite are identified as the most prevalent sources of Cr
responsible for Cr-enrichment in these soils Chromite is resistant to chemical
weathering and it has been observed to undergo incongruent dissolution conserving Cr in
serpentine soils Chromium is tightly bound in the matrix of the soil less than 1 of the
total Cr can is removed using standard soil extraction methods (10 hydrogen peroxide
(H2O2) solution 01 M potassium permanganate (KMnO4) solution10 M hydrofluoric
(HF) acid 1 M hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and 1
mM barium chloride (BaCl2) solution) due to the stability of chromite Although silicates
contain minor concentrations of Cr compared to chromite they are perhaps the main
contributors of Cr identified in plants (lt200 Cr mg kg-1) and soil solutions (lt50 microg L-1)
During soil-forming processes Cr released to soil solutions by the dissolution of Cr-
bearing minerals are sequestered into a host of structural and surface sites available in
authigenic Fe(III) oxides Fe(III) hydroxides and clay minerals (Schwertmann and
Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990 Gerth 1990 Gasser and
Dahlgren 1994 Ilton and Veblen 1994 Brigatti et al 2000 Ilton et al 2000) The
clay-sized fractions in soils at Jasper Ridge Biological Preserve CA contain Cr
concentrations as high as 1000 Cr mg kg-1
Chapter Summaries
A detailed field and analytical examination of Cr in serpentinites and serpentine
soils at Jasper Ridge is presented in Chapter 2 A majority of the Cr enrichment and
variability with depth in these soils is the result of small grains of chemically resistant
chromite from the serpentinite protolith Chromite identified in the soil undergoes
incongruent dissolution conserving Cr relative to chromite compositions identified in the
5
serpentinite bedrock Sequential extraction experiments using barium chloride (BaCl2)
hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and hydrofluoric (HF)
acid were performed on a chromite standard confirming that Cr-bearing spinels are
resistant to chemical weathering X-ray diffraction analyses of the clay-size fraction in
the serpentine soils identified Cr-clinochlore smectite vermiculite and serpentine with
the smectite being the most abundant mineral in the clay-size fraction Chromium(III)
was the only identified oxidation state of Cr in the serpentine soil solids and in
serpentinites at Jasper Ridge in the solid phase This chapter represents a comprehensive
site assessment identifying the nature and behavior of Cr in serpentine soils with respect
to the bedrock mineralogy and depth
Chromite an abundant Cr-enriched spinel is responsible for a majority of the Cr
enrichment identified in ultramafic rocks serpentinites and serpentine soils The release
and oxidation of Cr from chromite is a pathway to consider when evaluating the potential
of ultramafic rocks and their related sediments and soils as sources of Cr(VI) In
Chapter 3 experiments aimed at quantifying the release and oxidation of Cr from
chromite in the presence of birnessite are described Chromite is extremely resistant to
dissolution (lt8 Cr mg kg-1 is released) when immersed in solutions including 1 mM
barium chloride solution 10 M hydrofluoric acid 10 hydrogen peroxide and 1 M
hydroxylamine hydrochloride with 25 acetic acid Highly oxidizing solutions of 01 M
potassium permanganate increased the rate and release of Cr relative to the other
solutions listed above thus birnessite a high valence Mn oxide present in serpentine
soils was chosen to determine if Mn could accelerate Cr release and oxidation from
chromite The results presented in Chapter 3 demonstrate that birnessite increases the
rate of Cr(VI) observed in solution from chromite This production of Cr(VI) is
dependent on the chromite suspension density where birnessite surface are concentrations
are greater than 20 m2 L-1 between pHs 3 and 8 at ~25degC
A comprehensive review of the chemical characteristics of serpentine soils
including topics such as the mineralogy of serpentine soils soil solutions organic matter
and vegetation is provided within Chapter 4 Additionally chemical and spectroscopic
analyses were performed on serpentine soils from a) New Caledonia b) Eight Dollar
Mountain Selma OR c) Nickel Mountain Riddle OR d) Harvard Mine Jamestown
6
CA e) Jasper Ridge Biological Preserve Stanford CA and f) Pillikin Mine El Dorado
CA Serpentinites and serpentine soils associated with ophiolite complexes cover ~1 of
the earthrsquos exposed surface and they are correlated to modern and ancient convergent
boundaries The physical and chemical characteristics of serpentine soils are directly
related to the alteration and tectonic history of the serpentinite as well as the soil-forming
factors including the parent material climate biota topography and time Serpentine
soils have Cr concentrations exceeding 200 mg kg-1 however the concentration and the
availability of Cr in these soils are not correlated with the five soil-forming factors
(climate organic material parent material topography and time) or the soil chemistry
Although only Cr(III) was identified in the soil solids Cr(VI) has been identified in soil
solutions (lt30 microg L-1) from serpentine soils in New Caledonia and California Chromite
and Cr-magnetite grains are identified in the soils listed above and these minerals are
responsible for a majority of the Cr enrichment in serpentine soils
Kaumlmmererite is a Cr-chlorite associated with serpentinized ultramafic rocks such
as those collected at Jasper Ridge Harvard Mine and Pillikin Mine Although Cr
concentrations are typically 01-2 Wt in kaumlmmererite it is modally abundant in
serpentinites (up to 5 of the volume of the rock) and a potential source of Cr to be
released into soils and groundwater In Appendix A the standard molar thermodynamic
properties of the Cr-clinochlore end-member kaumlmmererite (Mg5Cr(AlSi3)O10(OH)8) is
estimated using established estimation algorithms This initial study provides the
foundation for submitting a future paper presenting the thermodynamic properties for
major Cr-bearing silicates including fuchsite (Cr-muscovite) uvarovite (Cr-garnet) and
tawmawite (Cr-epidote) and their metamorphic significance related to serpentinized
ultramafic rocks
7
CHAPTER 2
CHROMIUM GEOCHEMISTRY IN SERPENTINIZED ULTRAMAFIC ROCKS
AND SERPENTINE SOILS FROM THE FRANCISCAN COMPLEX OF
CALIFORNIA
ABSTRACT
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of California
produces serpentine soils containing high concentrations of Cr as well as other potentially
toxic elements including Ni Co and Mn Chromium concentrations in serpentine soils
from Jasper Ridge Biological Preserve in the Central Coast Range are as high as 4760
mg kg-1 nearly three times greater than the serpentinite protolith (~1800 mg kg-1)
Chromium-containing minerals within the bedrock include chlorite (~03 Cr Wt )
enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt ) magnetite
(82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate phase
containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
8
INTRODUCTION
Serpentine soils are formed from the weathering of ultramafic rocks and their
metamorphic equivalents (serpentinites Brooks 1987) Measured Cr concentrations in
these soils range as high as 80600 mg kg-1 and are enriched compared to non-serpentine
soils (7 to 221 mg kg-1 Proctor 1992 McBride 1994) The distribution and
concentration of Cr in serpentine soils worldwide is highly variable due to the chemical
composition and mineralogy of the parent material and soil forming factors including
climate biota topography precipitation and time (Hotz 1964 Rabenhorst et al 1982
Schwertmann and Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al
1989 Gough et al 1989 Kaupenjohann and Wilcke 1995) Potential Cr-bearing phases
in serpentine soils have been assessed using extraction experiments and these phases
include organic material silicates secondary oxides and spinels (Schwertmann and
Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994 Kaupenjohann and
Wilcke 1995 Becquer et al 2003) The weathering of these Cr-bearing phases over
geologic time and how these sources govern the distribution of Cr in serpentine soils have
not been evaluated in great detail
Chromium is a group VIB transition metal present in the environment as Cr(III) a
non-hazardous species and micronutrient andor as Cr(VI) a toxin to living cells and a
Class A human carcinogen by inhalation (Daugherty 1992 Cohen et al 1993 James et
al 1997) Chromium(III) is chemically immobile due to forming low solubility
(oxy)hydroxides and stable surface complexes whereas Cr(VI) is a weakly adsorbing
(oxy)anion and thus is mobile in most aquatic environments (McBride 1994 Fendorf
1995 Ball and Nordstrom 1998) Chromite and Cr-magnetite have been cited as
primary sources of Cr in serpentine soils (Rabenhorst et al 1982 Schwertmann and
Latham 1986 Alexander et al 1989 Gough et al 1989) Although Cr is primarily
found in the form of Cr(III) in both ultramafic rocks and serpentinites Cr(VI) has been
measured in serpentine soils in California and New Caledonia (Gough et al 1989
Becquer et al 2003) Due to the abundance of Cr and the potential for Cr(VI) formation
serpentine soils and serpentinites are potential sources of non-anthropogenic Cr
contamination
9
The objectives of this study are to determine the distribution and chemistry of Cr
in naturally Cr-enriched soils using an integrated field analytical experimental and
theoretical approach A wide array of techniques and methods including X-ray
absorption spectroscopy and chemical extractions were used to identify the nature of Cr
and its availability in serpentine soils Serpentinites and serpentine soils at Jasper Ridge
Biological Preserve located near Stanford CA were used to evaluate the geochemical
behavior of Cr These soils cover approximately 50000 m2 and contain high
concentrations of the trace elements Cr Ni and Mn The soils are dominantly Mollisols
with grassland vegetation and they are pedalogically similar to those associated with the
Central Coast Range of the Franciscan Complex in the western United States The main
serpentinite body at Jasper Ridge was exposed approximately 400000 years ago allowing
pedogenesis and Cr cycling to occur over a geologically relevant time period (Burgmann
et al 1994) These soils can then be used as a proxy for evaluating the long-term
behavior of Cr
BACKGROUND
Geologic Background of the Serpentinite at Jasper Ridge
Jasper Ridge Biological Preserve is located in the foothills of the Santa Cruz
Mountains in California and is part of the Central Coast Range of the Franciscan
Complex an Early Cretaceous accretionary meacutelange containing blocks of greywacke
greenstone blueschist and serpentinized ophiolite (Figure 21) Formation
metamorphism and emplacement of serpentinite at Jasper Ridge began approximately
130-150 million years ago during Jurassic subduction along the California continental
margin (Page and Tabor 1967) Thin wedges of ultramafic rock (peridotite) were
detached from the down-going slab of oceanic crust and incorporated into the subduction
meacutelange Subsequent serpentinization of the ultramafic rock in the down-going slab
resulted in the formation of serpentinites now exposed at Jasper Ridge
Rare blueschist blocks (1 to 5 meters in width) present in the serpentinite at Jasper
Ridge attest to the degree of metamorphism and the depths from which portions of the
tectonic meacutelange has been elevated Transformation of greenstone in the meacutelange into
blueschist required high pressures in excess of ~5-7 kbars (15-20 kilometers depth) and
10
temperatures of 200-300 degC (Ernst 1971) The low density of the serpentine (245-250 g
cm-3) allowed upward diapiric migration of the serpentinite body through the denser
enclosing tectonic meacutelange (265 g cm-3) Uplift of the present day Coast Range and
activity associated with the San Andreas fault system have shaped the current topography
and exposure of serpentinite at Jasper Ridge Serpentinite is currently present as a narrow
discontinuous band of pervasively sheared rock located along the ridge top (Figure 21)
The serpentinite is in fault contact with the Whiskey Hill Formation (Cretaceous turbidic
sandstone and mudstone) and tectonically interleaved with the Franciscan Complex
Serpentine Soils Characteristics and Review
Relative to other soil types serpentine soils are characterized by higher
concentrations of Cr Ni Co and Fe lower concentrations of plant nutrients such as Ca
K N and P lower CaMg ratios and characteristic flora and physical properties (Brooks
1987) The pH of serpentine soils ranges from ~4 up to 9 (Brooks 1987 Cole 1992
Proctor 1992) Serpentine soil solutions obtained using tension-free lysimetry have Cr
concentrations between 01 and 32 micromol L-1 in which a majority of the Cr is present as
colloidal material (Gasser and Dahlgren 1994) The inhabitability of these soils for most
plants has been attributed to the imbalance of Ca to Mg (Walker 1954) the deficiency of
plant nutrients (Turitzin 1991) and the elevated concentration of heavy metals
(Robinson et al 1935) however linking the lack of plant productivity of these soils to
toxic levels of Cr has not been fully resolved (Brooks 1987 Kruckeberg 1992) The
unique flora and fauna associated with these soils have been well-documented and Cr
concentrations in the plants range as high as 600 mg kg-1 (Brooks 1987 Gough et al
1989)
Chromium concentrations as a function of depth for serpentine soils in Maryland
(Rabenhorst et al 1982) Tehama County CA (Gough et al 1989) the Klamath
Mountains CA (Hotz 1964) and New Foundland Canada (Roberts 1992) are
illustrated in Figure 22 Chromium concentrations with respect to depth at each location
are highly variable demonstrating both maximums and minimums The concentration of
Cr varies significantly at each location due to the chemical composition of the parent
11
material and other soil-forming factors nevertheless Cr concentrations typically decrease
with depth towards the bedrock
A variety of chemical extractions on serpentine soils suggest a broad host of Cr-
containing soil phases including organic material silicates secondary oxides and spinels
(Schwertmann and Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994
Kaupenjohann and Wilcke 1995) The maximum concentration of Cr extracted from a
serpentine soil reported in the literature is 226 Wt Cr from the New Caledonia laterite
using a 03 M dithionite-citrate bicarbonate (DCB) solution (Schwertmann and Latham
1986) Li and Fendorf (2000) performed extractions on the serpentine soils at Jasper
Ridge using organic acids (citric and oxalic acids) in which a 100 microM citric acid solution
was only able to remove 23 Cr microgg-1 (08 Wt Cr) from a soil containing ~3000 Cr mg
kg-1 This suggests a majority of Cr is bound in phases that are not reactive with these
solvents
Sources of Chromium ProtolithBedrock
Chromium primarily occurs within oxide and silicate phases in ultramafic rocks
and serpentinites Chromite is a common primary (igneous) Cr-bearing spinel in mafic
and ultramafic rocks and often survives low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) Isomorphic substitution of
Al3+ Fe3+ and Ti4+ into the octahedral site and Mg2+ Ni2+ Zn2+ and Mn2+ into the
tetrahedral site is common (Sack and Ghiorso 1991) however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987 Pan and Fleet 1991 Burkhard 1993 Christofides et al 1994 Challis et
al 1995 Sanchez-Vizcaino et al 1995 Deer et al 1996 Mathiesen 1999) Pyroxenes
such as augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) are primary minerals in
ultramafic rocks that incorporate varying concentrations of Cr into their structures (Deer
et al 1996)
Serpentinization is associated with the partial oxidation of Fe(II) present in
olivine and pyroxene forming magnetite (Brooks 1987) Chromium(III) derived from
primary chromite and pyroxene may isomorphically substitute into the Fe(III) site of
magnetite due to size (octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge
12
similarities Hydrothermal and CO2 metasomatism of ultramafic rocks may produce Cr-
silicate minerals including fuchsitemariposite (Cr-mica lt17 Cr Wt ) uvarovite (Cr-
garnet lt20 Cr Wt ) tawmawite (Cr-epidote lt10 Cr Wt ) and kaumlmmererite (Cr-
chlorite lt6 Cr Wt Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Phillips et al 1980 Schreyer et al 1981 Max et al 1983 Kerrich et al 1987 Treloar
1987 Pan and Fleet 1991 Christofides et al 1994 Sanchez-Vizcaino et al 1995)
These minerals common in altered ultramafic rocks have varying concentrations of Cr
that isomorphically substitute for octahedral Al in the silicate structure due to similarities
in charge octahedral preference and octahedral radii (Al3+ = 061 Aring) Any of these
silicates and magnetite are potential sources for Cr in soils produced by the weathering of
serpentinite and ultramafic rocks
Chromium-Bearing Phases in Soils
Soil minerals specifically Fe(III) (oxy)hydroxides and clays provide a host of
structural and surface sites that may sequester Cr(III) released to soil solution by
weathering of Cr-bearing phases in serpentinite and ultramafic rocks Isomorphic
substitution of Cr(III) for Fe(III) in oxides (oxy)hydroxides and amorphous compounds
is expected based on crystallographic considerations synthesis experiments and analyses
of selective dissolution of Fe(III) phases (Schwertmann et al 1989 Gerth 1990)
Additionally Cr(III) readily sorbs on surfaces of Fe(III)-containing oxides such as
magnetite and hematite via inner-sphere complexation and forms low-solubility
precipitates at pH values greater than 4 (Charlet and Manceau 1992 Peterson 1996)
Octahedral substitution of Cr(III) for Al(III) in clays (smectites and vermiculites) is
possible based on size and charge similarities Sorption of Cr(III) on edge and interlayer
sites of clay minerals has also been observed by Ilton and Veblen (1994) Soil organic
matter (SOM) is another possible host for soil Cr in which Cr can be chelated into stable
organomineral and organometallic complexes (Kaupenjohann and Wilcke 1995 Brady
and Weil 1999)
13
MATERIALS AND METHODS
Location and Sample Collection
Jasper Ridge Biological Preserve is a microcosm of accreted terranes located in
the foothills of the Santa Cruz Mountains CA Figure 21a is a soils map reported by
Kashiwagi (1985) depicting the alluvial Tertiary sandstone and Franciscan Complex
soils present at Jasper Ridge The Franciscan Complex soils are divided into two
subgroups those derived from greenstone-graywacke and those from serpentine The
serpentine soils dominantly Mollisols and serpentinite bedrock are present as a narrow
discontinuous band located along the ridge top they do not receive sediment from the
adjacent non-serpentine soils Soils and rocks were collected at Jasper Ridge from 1999
to 2002 and sample locations are shown in Figure 21b A 500 meter transect consisting
of 10 sites extends across the length of the serpentine soil body At each site along the
transect soils and bedrock were collected with respect to depth Site 1 (Diablo clay) is
not derived from serpentinite but rather from chert and greenstone Additionally Site 1
is located down slope of the main serpentinite body and has likely received sediment
from the adjacent serpentine soil Rocks were also collected at several exposed outcrops
located in the serpentinite body
Site JR3 (Figure 21b) was chosen for a detailed geochemical study to identify
factors affecting the geochemistry of Cr in serpentine soils Soil at JR3 is a loamy
magnesic thermic Lithic Haploxeroll The thickness of the soil is approximately 50 cm
deep Grass dominantly Lolium multiflorum comprises 95 of the vegetation at this site
and trees and shrubs are not present
Mineral Identification in Rocks and Soils
Mineral identification was accomplished by electron probe microanalysis on an
automated JEOL 733A electron microprobe operated at 15 kV accelerating potential and
15 nA beam current Calibration was conducted using natural geologic standards Beam
width for analysis was 10 microm Raw counts were collected over 20 s and were constant
with time indicating that elemental drift was negligible Detection limits for probe
analyses are ~002 oxide Wt Additionally backscattering electron (BSE) images of
the rocks and soils were obtained using the microprobe
14
Carbon-coated petrographic thin sections of rock samples were used for
microprobe analyses Soil samples were dried in an oven at 60degC for 24 hours mounted
onto a glass slide using epoxy and heated to 100degC until the epoxy cured An initial cut
of each section using a petrographic saw was made followed by a second infusion of
epoxy to ensure total impregnation Finally each section was cut polished and carbon-
coated for electron microprobe measurements
Soil Chemistry
Soils were analyzed for pH electrical conductivity (EC (deciSiemensm))
particle size ammonium acetate extractable cations organic carbon total nitrogen and
elemental composition The pH was determined using a 11 mixture of soil to deionized
water Electrical conductivity (EC) was determined on 2 mL of a soil solution extract
Particle size of the soil was measured using the hydrometer method as described by Gee
and Bauder (1986) These analyses were performed at the Utah State University
Analytical Laboratory
Ammonium acetate was used to measure the concentration of extractable K Ca
Mg and Na within the soil The procedure involved weighing 2 g of soil and adding 25
mL of ammonium acetate (20 M NH4OAc) followed by mixing on an orbital shaker for
30 minutes The solution was filtered using a Whatman 42 filter paper (pore size 25
microm) Potassium Ca Mg and Na concentrations were analyzed using atomic absorption
spectrophotometry (AAS)
Organic carbon was measured using the Walkley-Black titration method (Mebius
1960) Approximately 05 g of sediment was weighed and reacted with 5 mL of 20 M
potassium chromate (K2Cr2O7) followed by the rapid addition of 10 mL concentrated
sulfuric (H2SO4) acid The bottle was cooled for 30 minutes at room temperature and
diluted with 100 mL of deionized water Next 030 mL of 0025 M ortho-
phenanthroline-ferrous solution was added and the solution titrated to the endpoint using
10M ferrous sulfate (FeSO4)
Total nitrogen of soil samples was obtained using an automated combustion
method as described by Sweeny (1989) Approximately 150 mg of air-dried and sieved
15
(40 mesh) soil was placed in a tin foil encapsulating cup A Leco total nitrogen analyzer
with resistance furnace was then used to determine total nitrogen
Major and minor element concentrations for soil and rock samples were obtained
by completely dissolving the samples using a mixture of hot concentrated nitric
perchloric and hydrofluoric acids Elemental concentrations were measured in the
supernatant using inductively coupled atomic emission spectroscopy (ICP-AES)
Standards were plusmn5 mg kg-1 of their known values These analyses were completed at
Chemex Laboratory (Vancouver Canada) Soil samples were dried and sieved (lt2mm)
before analysis Additionally total element concentrations for gt2mm sand silt and clay
size fractions were obtained for each sample collected at Site JR3
X-ray Diffraction and Clay-Size Mineral Identification
X-ray diffraction (XRD) analyses were conducted on the clay-size fraction
(lt2microm) from Site JR3 using a Siemens Diffractometer D5000 at the University of Idaho
The clay-size fraction was isolated using a suspension method described by Vaniman
(2001) Organic matter and calcium carbonate were removed from the clays using a
concentrated hypochlorite solution (Kunze and Dixon 1986) Crystalline and non-
crystalline iron oxides were removed using a citrate-sodium bicarbonate-sodium
dithionite (CBD) solution (Kunze and Dixon 1986) The remaining phyllosilicate
minerals were analyzed using a series of treatments including 1) magnesium-saturation
(2 M MgCl and 2 M MgAOC) and air drying 2) magnesium-saturation and glycolation
(1 glycerol-ethanol solution) 3) potassium-saturation (1 M KCl) and air drying and 4)
potassium (1 M KCl) saturation followed by heating to 500degC (Whittig and Allardice
1986) X-ray diffraction patterns were collected between 2θ values of 20-350deg and 20-
150deg with a generator potential of 30 kV a generator current of 22 mA (using CuKα
radiation) a Ni filter and a scan speed of 1degmin The software package JADE was used
for XRD data analysis and mineral identification
X-ray Absorption Spectroscopy
In-situ fluorescence yield X-ray absorption near-edge structure (XANES)
spectroscopy at the Cr K-edge (edge inflection point is 5989 eV for Cr(0)) was conducted
16
at the Stanford Synchrotron Radiation Laboratory (SSRL) Stanford CA using
beamlines 4-1 and 4-3 A Si(220) monochromator and a Ge-detector were used and
three scans were collected for each XANES analysis Soil and rock samples from Site
JR3 were prepared by removing the gt2mm fraction and crushed to a fine powder
Samples were loaded into a plastic sample holder and covered with Mylar tape
Elemental mapping of three samples were completed at SSRL on Beamline 11-2
using X-ray absorption spectroscopy (XAS) A Si(220) monochromator with a 6 microm
focusing capillary was used and fluorescence intensity was monitored with a Ge detector
Samples were prepared similar to those used for mineral identification using the electron
microprobe however the thickness of the sample was decreased to lt20 microm
Sequential Extraction Procedure
Sequential extraction experiments were completed on 2 g lt2 mm sieved soils
(Diablo clay from Site 1 at 10 cm depth and serpentine soil from Site 2 at 4 cm depth) in
three steps The exchangeable fraction was obtained using a barium chloride (1mM
BaCl2) solution for 1 h at 25 degC (Kunze and Dixon 1986) The Fe-crystalline oxide
fraction was obtained by reacting the solids with a hydroxylamine-hydrochlorideacetic
acid (10 M NH2OHsdotHCl in 25(vv) HOAc) solution for 6 h at 60 degC (Chester and
Hughes 1967) The silicate fraction was obtained by reacting the solids with
concentrated hydrofluoric acid (10 M HF) for 16 h at 25 degC Trace element
concentrations for each extraction were measured on the supernatant using inductively
coupled plasma optical emission spectrophotometry (ICP-OES) All analyses were
completed in triplicate Additionally sequential extraction experiments were completed
on 2g 106-250 mm sieved ultrasonically cleaned commercial grade chromite (MSDS
No 302-1 WHMIS Class D2A) using the same procedures listed above The
stoichiometric composition of chromite obtained using the electron microprobe is
(Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 Chromite was imaged after each sequential
extraction using a JEOL 5600LV environmental SEM equipped with an EDAX thin-
window energy dispersive X-ray detector
17
RESULTS
Serpentinite Mineralogy
The average serpentinite at Jasper Ridge is composed of serpentine (mainly
lizardite and antigorite 70) chlorite (15) talc (10) magnetite (4) and chromite
(1) based on petrographic and microprobe observations of five samples (JR3E JR6E 5
10 and JR3) however brucite has been reported as a major constituent in related
Franciscan Complex serpentinites (Hostetle et al 1966) Other minerals identified
include olivine augite and enstatite Although the serpentinite mineralogy is relatively
simple the modes and textures of the minerals present in the rock vary significantly from
site to site In Table 21 whole rock serpentinite compositions from Sites JR1 JR3E
JR6E 2 5 10 and JR3 are listed The average Cr concentration of the serpentinites is
~1700 mg kg-1 (tab 1) Iron (~7) Mg (~11) Ni (~3000 mg kg-1) and Co (~100 mg
kg-1) concentrations are also elevated in the serpentinite relative to the chert and
sandstone bedrock adjacent the main serpentinite body (Figure 21b and Table 21)
A range of rock and mineral textures are petrographically evident in the
serpentinite at Jasper Ridge (Figure 23) No foliations or lineations are observed in the
serpentinite (Figure 23a) obtained from the bottom of the soil pit excavated at Site JR3
(Figure 21b) Subhedral chromite and magnetite grains are randomly interspersed
amongst the serpentine matrix (Figure 23a) In serpentinite obtained in outcrop at Site
JR5E (Fig 21b) a lsquocrocodile skinrsquo texture is produced by thin irregular veins of fine-
grained subhedral magnetite separating serpentine grains (Figure 23b) In Figure 23c
numerous crosscutting veins of chlorite occur in orthopyroxene and define the foliation of
the rock in outcrop at Site JR3E (Figure 21b) The numerous crosscutting microscopic
veins random foliation patterns (or the lack thereof) and subhedral oxide grains suggest
the Jasper Ridge serpentinite has undergone extensive physical and chemical alteration
typical of most serpentinites
Microprobe analyses of minerals in the serpentinite demonstrate that the main Cr-
bearing phases are chlorite enstatite augite Mg-Al chromite Cr-magnetite and a
chromite-silicate mixture (defined in the following paragraph) Selected compositions
for these minerals are listed in order of their Cr concentrations in Table 22a Chlorite
contains the least Cr whereas the chromite-silicate mixture is the most enriched with
18
respect to Cr Figure 24 shows correlations of Cr content relative to Al Fe Mg and Si in
order to provide insight into the serpentinite mineralogy and to aid in evaluating trends of
Cr substitution
A fine-grained mixture containing spinel (chromite andor Cr-magnetite) and an
unidentifiable Cr-silicate contains the highest Cr concentration in the serpentinite at
Jasper Ridge (Table 22a shown in Figure 25a and 25b) This fine-grained mixture will
be referred to here as the chromite-silicate mixture (CSM) with chemical properties
similar to those in Table 22a and Figure 24 CSM is common around the rims of
protolith chromite grains and a number of the larger (~05 mm) Cr-magnetite grains in
the serpentinite The microcrystalline phase responsible for contributing Si to these
analyses is possibly chlorite due to similarities of false colors generated by backscattering
electron (BSE) images and chlorite commonly rimming CSM Microprobe analyses of
CSM listed in Tables 22 and 23 are considered as mixtures of several mineral phases
due to the fine-grained texture (lt5 microm) Chromium concentrations range from 12 to 16
Cr Wt in this phase CSM is chemically distinct from the chromite and magnetite as
shown in Figure 24
The chemical compositions of protolith chromite exhibit little to no variation in
any of the serpentinites collected at Sites 2 JR3E JR4E JR5E JR6E JR7E and JR3
(Table 22a and Figure 24) Chromite contains low Fe (~9 Wt ) high Mg (~9 Wt
) and high Al (~26 Wt ) compared to Cr-magnetite The average Cr concentration
for chromite is ~12 Cr Wt Additionally chromite grains are irregular and subhedral
suggesting that they have undergone significant alteration during metamorphism
Chromium concentrations in the magnetite vary depending on the proximity and
degree of alteration of the chromite (Table 22a and Figure 24) Magnetite grains with
less than 2 Cr Wt are common in veinlets in serpentinite not formed adjacent
chromite whereas magnetite with 8 to 12 Cr Wt were formed proximal or adjacent
chromite Chromium-enriched magnetite is also referred to as ferrite-chromite in the
literature (Spangenberg 1943 Burkhard 1993)
Chlorite and pyroxenes contain minor concentrations of Cr (lt1 Wt Table
22a) Aluminum Fe and Mg concentrations varied in the pyroxenes however the
majority of chlorites are relatively similar in composition (Figure 24) Chlorites are
19
more abundant in the serpentinite compared to the pyroxenes and both minerals were
subhedral and randomly distributed over the entire sampling area (Figure 21b)
Backscattering electron (BSE) images of what appeared to be a petrographically
opaque single chromite grain in Figure 23a demonstrates chromite has been
metasomatically altered resulting in the formation of Cr-magnetite Cr-chlorite and CSM
(Figure 25a) Represented compositions for these minerals (indicated by letters in Figure
25a) are listed in Table 23 Magnifying the chromite and CSM (Figure 25a)
demonstrates the fine-grained texture of CSM and an alteration rim (transition zone) is
also seen along the chromite edge (Fig 25b) This alteration rim is depleted in Cr with
respect to the chromite as shown in Figure 25b
In Figures 23a and 25c Ni was identified in three phases in the bedrock
including olivine (F 03 Ni Wt ) serpentine (G and H 03 and 0 Ni Wt ) and Ni-
Fe (I sim55 Ni Wt ) alloy (Figure 25c and 25d) their compositions are indicated by
letters and listed in Table 23 Alteration of the olivine resulted in the formation of lt10
microm Ni-Fe alloys shown in Figure 25c These Ni-Fe alloys form a corona surrounding
the remaining olivine and the illustration in Figure 25d highlights the minerals and
chemical changes
Chromite and Cr-magnetite alteration leading to the paragenesis of CSM and Cr-
chlorite is common in the serpentinites at Jasper Ridge The petrographic image in
Figure 26a and its corresponding BSE image in Figure 26b illustrate Cr-magnetite
(originally protolith chromite) partially altering to CSM and Cr-chlorite The chemical
compositions of each mineral (indicated by letters in Figure 26b) are listed in Table 23
CSM appears to marginally and metasomatically replace Cr-magnetite and is enriched by
~2 Cr Wt Chromium-chlorite is present in microcracks in Cr-magnetite and
encompasses the entire Cr-magnetite and CSM grain Serpentine minerals adjacent the
Cr-chlorite contains no measurable Cr
Distribution of Chromium in Serpentine Soils
Serpentine soils and underlying bedrock were collected at ten sites along a 500
meter transect extending the length of the main serpentine body as shown in Figure 21b
20
The range of Cr concentrations at each site (center) and Cr concentrations with respect to
depth (periphery) are shown in Figure 27 and soil analyses are listed in Table 24
Bedrock Cr concentrations for Sites 2 5 and 10 are also shown in these plots (Fig 27)
Chromium enrichment does not appear to be consistently found at a particular depth or
soil horizon for any of the sites Sites 3 9 and 10 demonstrate a Cr maximum at
approximately 006 m however Sites 1 2 5 and 6 exhibit a Cr minimum at roughly the
same depth The highest soil Cr concentration in the soil along the transect is 4760 mg
kg-1 at Site 3 at 006 meters depth The lowest soil Cr concentration is 1725 mg kg-1 at
Site 10 at 0085 meters depth The average bedrock Cr concentration is ~2000 mg kg-1
(Figure 27 Table 24) It is important to note Site 1 consists of Diablo clay not
serpentine soil however this site is located adjacent and down-slope of the main
serpentinite body From these observations it is apparent that there are no systematic
trends in soil Cr concentrations with respect to the position of the site and depth and soil
samples often contain higher concentrations of Cr with respect to their corresponding
bedrock and to the average Cr concentration of ultramafic rocks (1600 mg kg-1 Green
1972)
Trace metals including Ni Mn and Co were also evaluated in the serpentine soil
for Sites 1 through 10 and the results are listed in Table 24 Nickel concentrations range
from 1590 to 3960 mg kg-1 and do not directly correspond to the Cr fluxes with respect
to depth (Table 24) Nickel concentrations are less than Cr from Sites 1 to 4 From Sites
5 to 10 Ni concentrations exceed those of Cr averaging ~3600 mg kg-1 Unlike Cr Ni
concentrations are equivalent or greater in the bedrock (~3300 mg kg-1) as compared to
the soil (~3200 mg kg-1) Soil Mn concentrations at each site are relatively uniform at
~1800 mg kg-1 with little to no variation with depth (Table 24) Cobalt concentrations
do not exceed 238 mg kg-1 (Table 24) and are relatively uniform (~200 mg kg-1) with
respect to location along the transverse and depth at each site
Major elements including Al Fe Ca and Mg are listed for the serpentine soils
along the transect with respect to depth in Table 24 Aluminum Fe Ca and Mg
concentrations in the soils at each site are similar with exception of Site 1 (Diablo clay)
Aluminum concentrations in the Diablo clay are ~46 whereas the serpentine soils
contain less than 2 Al (Table 24) Iron and Mg concentrations are also low lt65 and
21
lt411 respectively in the Diablo clay compared to Fe (gt8) and Mg (gt6)
concentrations in the serpentine soil (Table 24) Calcium concentrations are low (lt1)
for all the serpentine soils and even the Diablo clay (Table 24) With a broad
understanding of the chemical characteristics of the serpentine soil at Jasper Ridge Site
JR3 (Figure 21b) was chosen for a detailed chemical study to examine all the factors
governing the chemistry of Cr in the soil
Serpentine Soil Chemistry and Mineralogy at Site JR3
Soil chemistry
Chemical properties (pH electrical conductivity organic matter nitrogen and
ammonium acetate extractable cations) and textures for four soil samples from Site JR3
are listed in Table 25 The pH of the soil is near neutral (pH from 671 to 698) and
increases slightly with respect to depth Electrical conductivities (EC) in the soil ranges
from 03 to 09 dSM a range indicative of relatively few dissolved salts andor major
dissolved inorganic solutes The organic carbon content of the soil is 32 at the surface
and decreases with depth to 10 near the bedrock Total nitrogen for the samples ranged
from 003 to 064 Ammonium acetate extractable Ca K Mg and Na are listed in
Table 25 The extractable concentration of both Ca and K decrease with depth whereas
there is no apparent trend associated with the concentrations of Mg and Na The
serpentine soil textures are dominantly clay loams however the texture is a sandy clay
loam near the bedrock
Whole soil samples and size fractions including gt2 mm sand (2-002 mm) silt
(002-0002 mm) and clay (lt2 microm) size fractions with respect to depth were chemically
digested and analyzed for total trace metal concentrations shown in Figure 28 and listed
in Table 26 to determine if the abundance of Cr was linked to a particular size fraction
(the percent of sand silt and clay size fractions are reported in Table 25) Additionally
Ni Mn and Co concentrations with respect to depth and size fraction are also presented in
Figure 28 and Table 26 Total and gt2mm size fraction Cr concentrations are similar at
the top and bottom of the soil profile ranging from 3297 to 5976 Cr mg kg-1 Chromium
concentrations exceed detections limits (gt10000 mg kg-1) for the sand-size fraction It is
important to note that the sand-size fraction corresponds to the size of most chromite
22
grains observed in the serpentinite Chromium in the clay fractions are low (~1000 mg
kg-1) compared to the other size fractions however these concentrations are still enriched
compared to non-serpentine soils
In Figure 28 and Table 26 Ni concentrations are enriched in the clay-size
fraction with values ranging from 2840 to 2920 mg kg-1 The silt-size fractions have the
lowest Ni concentrations (~1800 mg kg-1) Manganese concentrations are slightly
elevated relative to total Mn in the sand-size fraction with values ranging from 1728 to
2701 mg kg-1 The clay-size fraction is the least Mn enriched size-fraction with
concentrations at ~1200 mg kg-1 Cobalt concentrations are elevated in the sand-size
fraction with values ranging from 159 to 241 mg kg-1 and concentrations in the gt2mm
silt and clay-size fractions are similar to the total (~150 mg kg-1)
Soil Cr-bearing minerals
Chromium-bearing minerals in the soil identified using the electron microprobe
include chromite Cr-magnetite and CSM Chromium concentrations in soil chromite
Cr-magnetite and CSM are graphically shown in Figure 29 as a function of Al Fe Mg
Si and Al+Fe in Wt Representative analyses of the Cr-bearing soil minerals are listed
in Table 22b Compositions of soil chromite in Figure 29 are chemically distinct from
those identified in the protolith in Figure 24 Chromium concentrations in soil chromites
range from 22 to 37 Cr Wt containing minor concentrations of Al and Mg Values for
chromite in the serpentinite bedrock (Table 22a and Figure 24) and the FeCr2O4 end-
member are also shown in Figure 29 The dashed lines in Figure 29 connect Jasper
Ridge bedrock chromite (Figure 24) and stoichiometric FeCr2O4 where it can be seen
that compositions of soil chromite lie close to these tie lines Overall soil chromite
demonstrates a decrease in Al and Mg and an increase in Fe relative to Cr compared with
protolith chromite Soil Cr-magnetites contain between 7 and 20 Cr Wt and are
enriched in Cr compared to Cr-magnetites identified in the bedrock (Figure 24) Cr-
magnetite however reveals a decrease in Fe relative to Cr Minor concentrations of Si
(lt2 Wt ) in the soil chromite and Cr-magnetite may represent slightly altered rims or
contamination from sample preparation Chromium concentrations in CSM maintain a
consistent Cr concentration of ~11 Cr Wt similar to those found in the bedrock (~14
23
Cr Wt ) Soil CSMs with compositions between 8 and 12 Mg Wt are more than
likely associated with analyses containing more microcrystalline chlorite compared to
soil CSMs with less than 2 Mg Wt Soil CSM appears be less enriched with Si (lt14
Wt ) relative to those identified in the serpentinite (15 to 27 Wt )
The size of the Cr-bearing minerals varied significantly for each sample ranging
from 10 microm up to 2 mm The majority of chromite grains in the soil correspond to the
sand-size fraction and also vary from euhedral to subhedral with no discernable pattern
The sum of cations based on four oxygens for the chromite and Cr-magnetite analyses
range between 297 and 301 indicating these minerals are spinels not (oxy)hydroxides
Compositions obtained for chromite and Cr-magnetite may represent alteration rims
where weathering of the mineral in the subsurface environment has been most active
BSE images of chromite and Cr-magnetite were not able to definitively identify whether
these analyses in fact represent alteration rims
Mineralogy of the Clay-Size Fraction
The clay-size fraction in the serpentine soil at Site JR3 was evaluated with respect
to depth using XRD and specific chemical and heat treatments The d-spacings (Aring) and
their relative intensities for each soil and treatment with respect to depth are listed in
Table 27 Figure 210 shows the XRD analyses for clay sample JR3515 (depth 5-15
cm) and these analyses are almost identical to the other clays examined at Site JR3
Additionally Table 28 from Whittig and Allardice (1986) is provided to interpret the X-
ray diffraction analyses in Table 27 For example a d-spacing of ~14Aring is present in all
the treatments indicating the presence of chlorite in the clay-sized fraction Using an
XRD analysis program the clay-size fraction was determined to be composed of
smectite vermiculite serpentine minerals (lizardite and antigorite) and chlorite
(clinochlore and Cr-clinochlore) Smectite is the most abundant mineral in the clay-size
fraction whereas Cr-chlorite is the least abundant Overall the d-spacings and relative
intensities are similar despite the treatments and changes in depth (Table 27) This
indicates that the mineralogy of the clay-size fraction does not vary significantly in the
soil column
24
Chromium concentrations in the clay-size fractions are ~1010 mg kg-1 (Table
26) Smectite and vermiculite are possible Cr-bearing clay minerals in which isomorphic
substitution of Cr(III) could occur in the dioctahedral sheet Serpentine minerals
(lizardite and antigorite) are not likely hosts of Cr based on crystallographic
considerations and the absence of Cr in analyzed serpentine minerals in the bedrock
(Table 23) Clinochlore may also be a host for Cr in the inner layer octahedral sheet
Chromium-clinochlore is the only Cr-containing mineral identified in the clay-size
fraction using XRD and Cr-clinochlore (Cr-chlorite) is also an abundant mineral
identified in serpentinite protolith (Tables 22 and 23 Figure 26)
XAS of chromium in relation to depth
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in serpentine soils and bedrock at Site JR3 The XANES (X-ray absorption near-edge
structure) spectra demonstrate the absence of Cr(VI) in the soil column at Site JR3
(Figure 211) based on the lack of a pre-edge feature The detection limit for Cr(VI)
using this method is approximately ~20 mg kg-1 Ultimately XAS analyses demonstrate
Cr(VI) is not detected in any of the soil solids
X-ray microprobe analyses of soil and rock samples (0-5cm depth 5-10cm depth
and serpentinite) at Site JR3 were processed to map the relative concentrations of Fe and
Cr in which areas of elemental enrichment are identified by a violet color areas of
depletion by a blue color (Figure 212) Areas of Cr and Fe enrichment in Figure 212 are
localized with well-defined boundaries corresponding to grain edges and these elements
are not correlated in a well-defined trend Minerals assigned to these areas of Cr and Fe
enrichment were made based on petrographic and microprobe observations on the same
thin section For example high Fe and no Cr is magnetite whereas high Fe and high Cr
is either chromite or Cr-magnetite Chromite magnetite and Cr-magnetite appear to be
present in the serpentine soil at a depth between 0 to 5 cm (Figure 212A) In Figure
212b CSM is a possible mineral detected in the soil (depth 5-10 cm) and the presence
of magnetite is clearly seen due to the localized enrichment of Fe and no Cr Iron and
Cr spectra for the serpentinite were much more heterogeneous as compared to the soil
samples and Cr concentrations appear be more scattered (Figure 212c) These X-ray
25
microprobe analyses provide a complimentary method for examining Cr geochemistry
(including Cr(III) and Cr(VI)) in naturally occurring soil and rock solids however
Cr(III) is the only detected valence state in serpentine soils and rocks at Jasper Ridge
Sequential Extractions
Sequential extractions were completed on 1) Diablo clay (Inceptisol) from Site 1
(Figure 21b) at a depth of 10 cm 2) serpentine soil (Mollisol) from Site 2 (Figure 21b) at
a depth of 4 cm and 3) commercial grade chromite (106-250 microm) and these results are
shown in Figure 213 and listed in Table 29 Chromium extracted from the
exchangeable fraction is minimal (lt01 mg kg-1) for all the samples (number 1 in Table
29 Fig 213) Minor Cr concentrations are associated with the Fe-crystalline oxide
fraction in the Diablo clay (19 Cr mg kg-1) and serpentine soil (78 Cr mg kg-1) whereas
chromite only released 05 Cr mg kg-1 (number 2 in Table 29 Figure 213) Increased
concentrations of Ni and Fe are present in the Fe-oxide fraction of the Diablo clay (161
Ni mg kg-1 and 4474 Fe mg kg-1) and serpentine soil (5138 Ni mg kg-1 and 5138 Fe mg
kg-1 Table 29) Chromium extracted from the Diablo clay serpentine soil and chromite
in the silicate fraction is 14 31 and 14 Cr mg kg-1 respectively (number 3 in Table 29
Figure 213) The sum of extracted Cr Fe Ni and Mn concentrations (number 4 in
Table 29) are low with respect to total metals (number 5 in Table 29) In fact only 07
12 and 00 Cr was removed from the Diablo clay serpentine soil and chromite
respectively (number 6 in Table 29)
Chromite was extremely resistant to all the extractions and less than 001 of the
total Cr was removed (Table 29) Scanning electron microprobe (SEM) images of the
chromite after each extraction are shown in Figure 214 Chromium concentrations
released from the chromite (Table 29 and Figure 214a) increased with each sequential
extraction experiment starting with the exchangeable fraction (Figure 214b) progressing
into the Fe-crystalline oxide fraction (Figure 214c) and ending with the silicate fraction
(Figure 214d) The surface of the chromite became smoother after each extraction and
crystal imperfections in defect sights located along the edges and corners progressively
decreased (Figure 214) Chromite does not appear to be an easily accessible source of
Cr based on these extraction experiments
26
DISCUSSION
Serpentinite at Jasper Ridge has undergone significant physical and chemical
alteration since its formation and emplacement off the California continental margin
Despite being subjected to high pressures (~5-7 kbars) and elevated temperatures (200-
300degC) trace element concentrations within the serpentinite remain elevated Jasper
Ridge serpentinites are enriched with Cr Co Fe Mg Mn and Ni compared to chert and
sandstone located adjacent the main serpentinite body (Table 21 and Figure 21b) Over
the past 400000 years Cr-bearing minerals in the serpentinite have been weathering and
accumulating in the soil During this time Cr concentrations in the soil have
progressively increased relative to Cr-concentrations in the serpentinite protolith This Cr
enrichment is directly linked to specific weathering-resistant minerals identified in the
soil
Chromium in Serpentinites
Three minerals in the bedrock at Jasper Ridge contain appreciable concentrations
of Cr Cr-magnetite (82-103 Cr Wt ) chromite (~108 Cr Wt ) and CSM (~133
Cr Wt ) Cr-magnetite and CSM were derived from the alteration of primary igneous
chromite CSM or Cr-spinels with high concentrations of Si have been noted by
(Burkhard 1993) as a Cr-spinel containing a possible microscopic silicate phase Our
analyses support this result in which the microcrystalline silicate is probably Cr-chlorite
due to its common occurrence with CSM (Figure 26) Definitive identification of this
microcrystalline mixture was not possible due to size of the crystals relative to the spot
size of the electron microprobe Chlorite and pyroxene contain minor concentrations of
Cr (~1 Wt ) and these were the only identified Cr-containing silicates in the
serpentinite Minor concentrations of Cr observed in some chlorites are directly related
to the metasomatic alteration of chromite and Cr-magnetite (Figure 26) Other common
metasomatically-derived Cr-bearing silicates such as fuchsitemariposite and uvarovite
are not present possibly due to the lack of aqueous K+ and Ca+2 in the alteration fluids
Overall the chemical compositions of these protolith minerals (Figure 24) demonstrate
only minor chemical variations allowing comparative analysis with Cr-bearing soil
minerals
27
Chromium in Serpentine Soils
Cr-bearing soil minerals identified using the electron microprobe include
chromite Cr-magnetite and CSM and these phases are chemically distinct from those in
the bedrock (Figure 24 and 29) The highest concentration of Cr (41 Wt ) in the soil
is in chromite with a composition approaching the FeCr2O4 end-member This represents
a ~80 increase in the Cr concentration relative to bedrock chromite (see Figure 215)
Cr-magnetite also demonstrates Cr enrichment (up to 20 Cr Wt ) with respect to
magnetite identified in the bedrock (~10 Cr Wt ) (Figures 24 and 29) The textures of
the soil Cr-bearing spinels exhibit sharp irregular edges possibly due to extensive
weathering CSM in the soil is similar to those found in the bedrock with Cr
concentrations ranging from 11 to 14 Cr Wt however Si concentrations in soil CSMs
(8 to 14 Wt ) are noticeably less than those measured in the serpentinite (15 to 20 Wt
) The weathering of CSM might preferentially release Si relative to other elements in
its structure Cr-silicates (chlorite and pyroxenes) were not identified in the soil possibly
due to complete weathering of these phases or they were too fine-grained for microprobe
analyses however Cr-clinochlore was identified in the clay-size fraction using XRD
Elemental maps shown in Figure 212 corroborate electron microprobe observations by
revealing well-defined zones of Cr and Fe enrichment in both the soils and serpentinite
These areas of Cr and Fe enrichment are either chromite Cr-magnetite CSM or
magnetite Overall these maps reveal Cr is not homogeneously distributed in the soil
confirming Cr is in specific mineral phases Additionally the observed abundance of
chromite Cr-magnetite and CSM in the soils indicates that these minerals are probable
sources for a majority of the Cr enrichment identified in the soils
Soil Spinels
Soil chromite and Cr-magnetite appear to have undergone chemical alteration
enriching the minerals with respect to Cr In Figure 215 compositions of bedrock
chromite and end-member FeCr2O4 are shown together with tie-lines like those in figure
8 connecting both minerals Reaction progress in Figure 215 is based on the assumption
that the serpentinite chromite (0 reaction progress) is proceeding toward FeCr2O4
(100 reaction progress) in the serpentine soil Soil chromites from Jasper Ridge are also
28
plotted in Figure 215 where Cr concentrations are assumed to vary linearly between the
bedrock chromite and chromite end-member which defines the reaction progress of the
mineral These tie-lines correlate extremely well to the chemical composition of
chromites identified in the soil For example chromite containing ~30 Cr Wt will
have ~15 Fe Wt ~13 Al Wt and ~5 Mg Wt similar to soil chromite in Figure 29
The retention of Cr in chromite shown in Figure 215 has significant implications towards
the availability and cycling of Cr in serpentine soils
One possible mechanism for the chromite composition trends illustrated in
Figures 29 and 215 is incongruent dissolution where Cr is retained relative to other
elements Sequential extraction experiments (Figure 213 and Table 29) support this
idea in which Fe is more easily released than Cr (1096 Fe mg kg-1 to 19 Cr mg kg-1)
from chromite Unfortunately Mg and Al were not monitored during the experiments
From these sequential extraction experiments and chromite analyses (Figure 215) Al
and Mg have been preferentially released from the chromite structure compared to Fe and
Cr Additionally the alteration of chromite in the serpentinite and the soils are analogous
to one another Chromite in the serpentinite retains a majority of its Cr compared to Mg
and Al when being altered to CSM by fluids at elevated pressures and temperatures
(Figure 26 and Table 22) Under ambient soil conditions chromite is once again
retaining Cr relative to Mg and Al causing what appears to be an increase in Cr
concentration Chromium is not easily released from chromite whether under high
pressures and temperatures or in the soil
Crystal field stabilization energies (CFSE) and octahedral site preference energy
parameters (OSPE) provide a means to explain the incongruent dissolution or Cr
preservation in the Cr-spinels CFSE is a measure of the net energy of stabilization
gained by a metal ions nonbonding d electrons as a result of complex formation This
measure also indirectly describes the energy required to remove an ion from a crystal
structure The octahedral CFSE for Cr(III) is ndash2247 kJmole and the OSPE (the
difference between the tetrahedral and octahedral CFSE) is ndash1578 kJmole (Dunitz and
Orgel 1957 McClure 1957) These values are the most negative for any transition
element and indicating the presence of Cr(III) in an oxide increases the crystalrsquos stability
compared to an analogous crystal lacking Cr(III) Chromium(III) would be among the
29
last ions to removed from the spinel structure compared to other transition elements such
as Fe2+and Mn2+
Soil Chemistry related to Chromium Availability
In Table 25 the chemical properties of the soil at Site JR3 do not appear to be a
factor when evaluating Cr in the soil The pH of the soil did not vary considerably with
depth Additionally the near neutral pH would not have a significant impact accelerating
Cr-bearing mineral dissolution Organic matter within the soil was elevated despite the
low plant productivity of the serpentine soil The role of organic matter could be a
significant factor such as the production of organic acids promoting the dissolution of the
Cr-bearing minerals however Cr concentrations near the surface (where organic matter
was the greatest) are not significantly greater or less than those deeper in the soil profile
The relative concentrations of ammonium acetate extractable Ca K Mg and Na are
characteristic of serpentine soils and did not provide any new insight Finally Cr(III)
was the only identifiable valence state of Cr based on X-ray absorption spectroscopy
(Figure 211) which is what should be expected based on the Cr-bearing minerals and the
pH of the soil
Extraction experiments summarized in Table 29 and Figure 213 demonstrate
~99 of the Cr in a serpentine soil is retained in highly resistant phases at Jasper Ridge
The chromite standard used in these experiments contributed negligible concentrations of
Cr (lt2 mg kg-1 total) for all of the sequential extraction experiments supporting earlier
evidence that Cr-spinels are retaining Cr and are potentially responsible for a majority of
the total Cr enrichment in the serpentine soil At Jasper Ridge Cr extracted from the
serpentine soil (31 mg kg-1) is likely the result of the Cr-silicates compared to the Cr-
spinels Low Cr-containing minerals (possibly chlorite enstatite augite and clay
minerals) appear to be the most probably source of Cr to soil solutions and vegetation of
the serpentine soil Approximately 30 of the serpentine soil is composed of clay-sized
minerals and this size-fraction has ~1000 Cr mg kg-1 The mineral responsible for Cr
enrichment in the clay-size fraction is more than likely Cr-chlorite identified using XRD
Chromium-chlorite in the soils suggests that this mineral is potentially the largest silicate
contributor of Cr in the serpentine soil
30
Interpreting the enrichment and behavior of Cr in serpentine soils such as in
Figure 27 can be accomplished based on the abundance Cr-spinels observed in the soil
Small grains of chromite Cr-magnetite and even CSM in the soil have the potential to
appreciably impact Cr concentrations Table 210 demonstrates the chromite masses
required to change the Cr concentration in a 1 kg sample of soil The highest Cr
concentration at Jasper Ridge was identified at Site 3 with a measured value of 4760 mg
kg-1 which means only ~1 of the soil sample (Table 210) needs to contain chromite
Additionally a minimal amount (lt011 of the sample as shown in Table 210) of
chromite is necessary to change the concentration by 500 mg kg-1 which could easily
cause the sharp Cr increases and decreases noted in Figure 27 This agrees with our
observations of chromite abundance in the serpentine soils Additionally a majority of
the chromite grains identified in the bedrock are between 2 and 002 mm which
correspond directly to the elevated Cr concentrations (gt10000 mg kg-1) in the sand-size
fraction of the serpentine soil (Table 26) Even the ~15000 Cr mg kg-1 values in soils
measured by Hotz (1964) in Figure 22 only require ~2 of the sample of the sample to
be chromite A few grains of chromite possess the ability to significantly impact
measured Cr concentrations in soil samples
Chromium and Nickel in Serpentine Soils
Chromium and Ni have often been correlated to one another due to their common
enrichment in serpentine soils (Hotz 1964 Rabenhorst et al 1982 Schwertmann and
Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al 1989 Kaupenjohann
and Wilcke 1995) A majority of the Ni in the Jasper Ridge serpentine soil originated
from olivine (Figure 25c and 25d Table 23) Chemical alteration of olivine during
serpentinization resulted in the mineral serpentine (Coleman and Jove 1992) in which
minor concentrations of Ni (lt037 NiO Wt ) substituted into its structure Continued
alteration under highly reducing conditions and the incompatibility of Ni in the serpentine
structure resulted in the formation of Ni-Fe alloy (Figure 25c) This type of alteration is
observed in many of the rocks at Jasper Ridge where the Ni-Fe alloys are ~10 microm in size
Olivine in the rocks and soil appear to be less resistant to weathering compared to
chromite therefore Ni is possibly more available to bacteria and vegetation in serpentine
31
soils Sequential extractions (Figure 213 Table 29) support these observations in which
a greater percentage of total Ni was extractable compared to the Cr in the soils Although
Cr and Ni are enriched in serpentine soils the availability and behavior of these two
elements are significantly different
CONCLUDING REMARKS
Chromium is predominantly present in primary chromite Cr-magnetite and a
chromite-silicate mixture (CSM) in both the bedrock and soil despite weathering over
geologic time Sharp variations in soil Cr concentrations with respect to depth are caused
by the presence or absence of chromite Cr-magnetite andor CSM Chromium-spinels in
the soil have undergone incongruent dissolution possibly conserving Cr(III) with respect
to Al3+ and Mg2+ Chromite is highly resistant to weathering and it is more than likely
not contributing bioavailable Cr Although Cr-spinels are a major source of Cr Cr-
silicates in the protolith (chlorite enstatite and augite) and soil (chlorite
montmorillonite and vermiculite) afford the most accessible Cr source identified Based
on our observations and analyses chlorite is the most likely Cr source of these silicate
minerals Finally Cr(VI) was not detected in any of the soil and rock samples
ACKNOWLEDGEMENTS
We thank Jasper Ridge Biological Preserve for allowing us the opportunity to
examine the serpentine soils Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund NSF Grant EAR-9902859 (DKB) GSA Graduate
Student Research Grants Office of Teaching and Learning (SF) at Stanford University is
gratefully acknowledged We appreciate the technical support from the staff of the
Stanford Synchrotron Radiation Laboratory from Bob Jones for assistance in microprobe
analyses of the serpentinites and serpentine soils and from Ben Bostick Travis Horton
and Theresa Barber for field and lab assistance
32
Table 21 Whole rock elemental concentrations of serpentinites (Figure 21b) chert and sandstone
-----------------------Location of Serpentinitesdagger----------------------- Elemental Concentrations JR1 JR3E JR6E Site 2 Site 5 Site 10 JR3
ChertDagger SandstoneDagger
Al 02 08 03 027 22 107 024 079 027 Ca 026 007 005 003 046 013 003 001 lt001 Co mg kg-1 101 99 100 99 155 217 97 1 lt1 Cr mg kg-1 580 2280 1510 1840 1945 2090 1465 10 lt1 Fe 303 71 609 581 76 1106 512 041 024 K 001 031 099 lt001 lt001 001 lt001 029 008
Mg 122 134 154 1336 524 804 112 014 034 Mn mg kg-1 495 800 155 730 1375 1860 460 110 85 Na 181 006 093 lt001 lt001 lt001 lt001 001 001 Ni mg kg-1 2037 1545 3321 3510 3020 3960 2740 20 9
Total digestion ICP-AES analyses daggerSerpentinites are dominantly composed of the minerals serpentine chlorite talc and spinels DaggerSee Figure 21b for sample locations
33
Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in serpentinite rock at Jasper Ridge
Chlorite Enstatite Augite Chromite Cr-Magnetite ChromiteSilicatedagger SiO2 3738 5463 5222 000 024 1053 Al2O3 552 477 516 5059 182 395 NiO 001 01 00 03 00 052 FeODagger 762 614 239 1446 7919 5042 MnO 009 015 009 042 140 182 Cr2O3 024 060 099 1584 1605 1939 MgO 3337 3305 1560 1869 024 605 CaO 023 050 2348 000 000 019 Na2O 000 001 050 000 001 008 K2O 000 000 001 000 000 011 Total 8446 9994 10048 10027 9894 9307
Formula Unit Compositions
Si 370 094 189 000 001 --- IVAl 030 006 011 000 000 --- VIAl 035 004 011 157 008 --- Ni 000 000 000 001 000 ---
Fe+2 063 010 007 030 232 --- Mn 001 000 000 001 004 --- Cr 002 001 003 033 048 --- Mg 493 085 084 073 001 --- Ca 002 001 091 000 000 --- Na 000 000 004 000 000 --- K 000 000 000 000 000 --- O 1400 300 600 400 400 ---
Electron microprobe analyses daggerChromitesilicate is also referred to as CSM DaggerFe+3 is present however all conversions are based on Fe+2
34
Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in serpentine soil at Jasper Ridge
CSM Chromite Cr-Magnetite SiO2 1940 057 267 Al2O3 559 1059 381 NiO 014 001 004 FeOdagger 2629 2259 5244 MnO 155 154 419 Cr2O3 1878 5577 3237 MgO 1668 916 288 CaO 015 001 005 Na2O 006 001 001 K2O 0 008 009 Total 8864 10033 9855
Si --- 002 010 IVAl --- 042 017 VIAl --- 000 000 Ni --- 000 000
Fe+2 --- 063 148 Mn --- 004 013 Cr --- 147 096 Mg --- 046 016 Ca --- 000 000 Na --- 000 000 K --- 000 000 O --- 400 400
Electron microprobe analyses dagger Fe+3 is present however all conversions are based on Fe+2
35
Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25 and 26 A B C D E F G H I J K L Chromite
Chromite CSM Magnetite Chlorite Olivine Serpentine Serpentine Ni-Fe-Metal Cr-Magnetite CSM Chlorite
SiO2 0 0 1053 0 3407 4046 3688 4074 591dagger 021 1143 2855Al2O3
5059 5162 395 034 1114 000 397 085 054 165 980 2293NiO 028 031 052 0 004 037 037 0 5302dagger 0 0 0
FeODagger 1446 1452 5042 9992
1220 901 262 453 1753dagger 8169 3498 806MnO 042 044 182 0 008 010 0 0 0 107 150 00Cr2O3 1584 1576 1939 125 071 001 0 0 0 1369 1905 039MgO 1869 1869 605 003 292 4895 3848 3755 064 1217 2828CaO 0 0 019 0 0 0 0 0 0 0 0Na2O 0 0 0 0 0 0 0 0 0 0 0K2O 0 0 0 0 0 0 0 0 0 0 0H2O --- --- --- --- 1200 --- 1600 1600 --- --- 1200Total 10027 10136 9307
10164 9944 9891 9834 9972
9897 8897 10028
Formula Unit CompositionSi
000 000 --- 000 328 100 184 200 001 --- 269IVAl 000 000 --- 000 072 000 000 000 000 --- 131VIAl 157 159 --- 002 058 000 023 005 007 --- 123Ni 001 001 --- 000 000 001 001 000 000 --- 000
Fe+2 030 029 --- 294 084 019 011 019 239 --- 059Mn 001 001 --- 000 000 000 000 000 003 --- 000Cr 033 032 --- 004 006 000 000 000 041 --- 003Mg 073 073 --- 000 419 180 285 274 004 --- 397Ca 000 000 --- 000 000 000 000 000 000 --- 000Na 000 000 --- 000 000 000 000 000 000 --- 000K 000 000 --- 000 000 000 000 000 000 --- 000O 400 000 --- 000 1400 400 700 700 400 --- 1400
36
Electron microprobe analyses dagger elemental metal Dagger Fe+3 is present however all conversions are based o
36
2321000---
10025
591000054
53021753000000
2321000000000000
n Fe+2
Table 24 Bulk soil elemental concentrations for soils and bedrock collected along the transect
(Fig 21b)
Sites Samples Depth
(meters) Cr
(mg kg-1) Ni
(mg kg-1) Mn
(mg kg-1) Co
(mg kg-1) Al
() Fe
() Ca ()
Mg ()
COI03 0015 2420 1590 1325 118 464 565 065 352 COI36 0045 2370 1630 1600 138 467 55 063 356 COI69 0075 2160 1755 1670 144 471 6 064 365 COI912 0105 2420 1785 1740 151 496 615 067 384
1
COI1215 0135 2570 1985 2340 195 506 649 069 411 COII02 001 3710 2400 1665 169 294 702 100 768 COII25 0035 3600 2290 1625 162 275 673 093 743 COII58 0065 2910 2310 1550 159 25 672 087 771
COII811 0095 3520 2350 1595 158 276 693 093 766 2
COIIBedrock 012 1840 3510 730 99 027 581 003 1336 COIII14 0025 3750 2930 1655 190 178 897 062 762 COIII47 0055 4760 3050 1850 178 178 94 056 805
COIII710 0085 4380 3410 1930 216 216 1044 07 797 3 COIII1013 0115 4320 3380 1955 201 213 1061 063 626 COIV01 0005 3710 2970 1915 214 209 909 053 781 4 COIV15 003 3540 2710 1725 195 181 801 043 792 COV14 0025 2620 2350 1330 150 246 651 063 526 COV47 0055 1775 2710 1520 165 252 766 059 517
COV710 0085 2000 3090 1830 195 225 737 057 583 5 COVBedrock 0115 1945 3020 1375 155 22 76 046 524
COVI14 0025 3040 3720 1855 224 084 1152 015 736 COVI47 0055 2400 3610 1785 218 085 1138 014 68 COVI710 0085 2960 3800 1850 232 093 1232 015 714 6
COVI1013 0115 2800 3860 1800 224 085 1164 014 739 COVII14 0025 2620 3550 1925 229 083 1128 015 718 COVII47 0055 1975 3590 1805 216 084 1085 014 655 7 COVII712 0095 2080 3600 1840 220 07 1105 011 741 COVIII14 0025 2130 3830 1990 238 085 115 017 705 COVIII47 0055 2550 3720 1865 228 085 1099 017 714 8
COVIII712 0095 2240 3910 1860 229 076 1112 014 828 COIX14 0025 2470 3480 1890 227 104 1201 023 579 COIX47 0055 3000 3830 1595 205 084 108 013 842 9 COIX713 01 2720 3710 1845 221 111 1208 017 551 COX14 0025 2810 3740 1760 218 129 114 017 675 COX47 0055 2190 3910 1835 224 131 1206 017 702
COX710 0085 1725 3370 1595 192 11 1061 014 635 10 COXBedrock 0115 2090 3960 1860 217 107 1106 013 804
Total digestion ICP-AES analyses
37
Table 25 Chemical and physical properties of the soil at Site JR3
NH4OAc-extractable ---Hydrometer---Ca K Mg Na
Sand Silt ClaySample Depth (cm) pH EC
dSm
Organic Matter ------
LECO Nitrogen ------ ---------mg kg-1---------
Texture
JR305 25 671 04 32 012 505 2366 3291 355 36 31 33 Clay LoamJR3515
10 683 09 19 064 4205 8176 3343 2392 36 30 34 Clay LoamJR31530 225 688 03 12 003 3725 4903 3750 2305 36 33 31 Clay LoamJR33045 375 698 07 10 003 2873 3665 3388 2549 46 24 30 Sandy Clay Loam
See Methods Section for more details
38
38
Table 26 Selected trace metal concentrations for each size fraction versus depth at Site JR3
Depth (cm) Total gt2mm Fraction Sand Silt Clay Sample
------------------------Chromium (mg kg-1)------------------------ JR305 25 4906 5086 gt10000 4565 988 JR3515 10 5976 4678 8284 4000 1115 JR31530 225 5222 4206 gt10000 4133 980 JR33045 375 3477 3297 gt10000 4571 956
------------------------Nickel (mg kg-1)------------------------------ JR305 25 1935 2050 1660 1305 2920 JR3515 10 2180 2050 1980 1325 2920 JR31530 225 2190 2120 1835 1370 2880 JR33045 375 2090 2140 1740 1285 2840
------------------------Manganese (mg kg-1)-------------------------- JR305 25 1319 1571 1728 1301 1210 JR3515 10 1664 1601 2701 1471 1254 JR31530 225 1639 1626 2217 1599 1231 JR33045 375 1222 1226 1781 1405 1170
-----------------------Cobalt (mg kg-1)-------------------------------- JR305 25 128 146 159 134 131 JR3515 10 155 145 241 147 132 JR31530 225 152 147 194 162 128 JR33045 375 125 124 159 141 122
Total digestion ICP-AES analyses
39
Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3
JR305CLAY Depth 0-5cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1439 100 1833 100 1274 100 1391 100 731 179 1432 431 732 645 1047 105 71 216 941 163 71 41 838 838 474 101 847 141 475 257 726 726 356 125 73 271 365 264 313 83 712 248 355 299
473 143 313 257 364 131 355 193
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1845 1000 1268 1000 1388 1000 852 68 1447 387 724 987 1010 859 733 264 944 143 703 679 836 281 712 247 903 139 471 282 723 576 475 109 850 83 362 244 365 73 733 426 352 321 356 117 712 291 313 154 314 63 475 122
364 170 355 196
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1724 1000 1297 1000 1396 1000 842 45 1397 447 741 957 1175 708 730 318 923 176 715 686 1037 944 710 255 840 118 477 271 837 139 475 97 717 465 367 400 726 375 357 133 468 182 357 386 313 91 362 194 315 286
352 229 JR33045CLAY Depth 30-45cm
Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity
1447 1000 1869 1000 1268 1000 1393 627 842 44 1476 287 737 843 1049 1000 730 207 954 107 715 539 837 190 710 182 915 107 479 157 723 111 475 69 860 61 366 303 364 69 740 332 356 292 356 74 717 256 315 315 313 57 478 111
367 141 357 115
See Methods Section for details
40
Table 28 X-ray diffraction d-spacings for common silicates from Whittig and Allardice (1986)
Diffraction Spacing (Aring) Mineral(s) Indicated Mg-saturated air-dried
14-15 Smectite vermiculite chlorite 99-101 Mica (illite) 71-73 Serpentine
715 Kaolinite chlorite Mg-saturated glycerol-solvated
177-180 Smectite 14-15 Vermiculite chlorite
99-101 Mica (illite) 71-73 Serpentinte
715 Kaolinite chlorite K-saturated air-dried
14-15 Chlorite vermiculite (with hydroxy interlayer) 124-128 Smectite 99-101 Mica vermiculite (contracted) 71-73 Serpentine
715 Kaolinite chlorite K-saturated heated (500ordmC)
14 Chlorite 99-101 Mica vermiculite (contracted) smectite (contracted) 71-73 Serpentine
715 Chlorite
41
Table 29 Results of the Diablo clay serpentine soil and chromite sequential extraction experiments
1 Exchangeable Fraction (1mM BaCl2) 4 Sum of 1 2 amp 3 (mg kg-1) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 0 0 003 Cr 159 388 19 Fe 014 011 001 Fe 24812 25532 1096 Ni 104 103 008 Ni 1854 1868 25 Mn 0 0 0 Mn 778 1172 04
2 Fe-Oxide Fraction (1M NH2OH HCl in 25 (vv) HOAc) 5 Total Cr Fe Ni and Mn (Total Digestion) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 19 78 05 Cr 2290 3310 1767864 Fe 4474 5138 38 Fe 60000 68000 1044325 Ni 1610 1506 01 Ni 1673 2345 00 Mn 231 430 00 Mn 1498 1608 3424547
3 Silicate Fraction (10M HF) 6 Percent Extractable ((45)100) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ------------------------------------- Cr 140 310 14 Cr 07 12 00 Fe 20337 20393 1058 Fe 41 38 01 Ni 234 352 22 Ni 111 80 00 Mn 547 742 04 Mn 52 73 00
ICP-AES analyses
42
Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1 kg of sample
Chromite
FeCr2O4 Cr Concentration
(mg kg-1) Grams Sample 500 11 011 1000 22 022 2000 43 043 5000 108 108
43
Figure Captions
Figure 21a Soils map of Jasper Ridge Biological Preserve (Stanford CA) with
complete descriptions of the soils shown in the legend 21b An enlarged image of the
study area shown in 22a indicates sampling sites of the soils and rocks collected A
dashed line marks the 500 m transect
Figure 22 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above
Figure 23 Petrographic images illustrating the varying serpentinite textures at Sites (A)
JR3 (B) JR5E and (C) JR3E For reference to site locations see Figure 21b
Figure 24 Correlation diagrams of Cr-bearing minerals in the serpentinite at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Si in Wt
Chromite (circle) Cr-magnetite (square) chromitesilicate or CSM (down triangle)
chlorite (hexagon) and pyroxenes (up triangle) analyses are shown on each correlation
diagram
Figure 25 Images of the serpentinite bedrock at Site JR3 demonstrating the mineralogy
and Cr-bearing phases Microprobe analyses are indicated by circled letters and these
values are in Table 23 25a Backscattering electron image demonstrating several Cr-
bearing phases present including chromite Cr-magnetite CSM and Cr-chlorite from
Figure 23a 25b Magnified image shown in Figure 25a demonstrating the fine-grained
texture of CSM Additionally Cr2O3 Wt values are shown for chromite the depletion
zone and CSM 25c Image of Ni-bearing phases present in the bedrock including
olivine serpentine and Ni-Fe alloy from Figure 23a 25d Cartoon illustrating the
mineralogical and chemical changes present in 25c
Figure 26 Images of a Cr-chlorite and Cr-magnetite assemblage in the serpentinite
bedrock 26a Petrographic image of the bedrock at Site 4 (Figure 21b) 26b
44
Backscattered electron image demonstrating the common occurrence of Cr-chlorite
encompassing chromite grains Microprobe analyses are indicated by circled letters and
are located in Table 23
Figure 27 Highest and lowest Cr concentrations (mg kg-1) at each site in the serpentine
soil 500 m transect shown in Figure 21b Samples included soils at various depths (m)
and bedrock from each site The average Cr concentration for the Jasper Ridge
serpentinite is marked on the central figure Chromium concentrations versus depth (m)
at each site (noted in the upper left corner) outline the main diagram
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected
for total gt2 mm sand silt and clay size fractions listed in Table 26
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soil at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Al+Fe in Wt for
chromite (circles) Cr-magnetite (hexagons) and CSM (diamonds) Chromite from the
bedrock is denoted by a larger circle and FeCr2O4 with a square
Figure 210 XRD clay analyses of sample JR3515 (Site JR3 Figure 21b) demonstrate
the influence of each treament Mg-saturated Mg+glycerol K-saturated and K-saturated
+ heat (500oC) The d-spacings and intensities are also listed in Table 27
Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at depths (a) 0-5cm (b) 5-15cm (c) 15-30cm (d) 30-45cm and (e) bedrock The pre-edge
area indicative of Cr(VI) peaks is shown on the right Note that no peaks are present
Figure 212 X-ray microprobe elemental maps of Cr and Fe for (212a) serpentine soil
at a depth of 0-5cm (212b) serpentine soil at a depth of 5-10cm and (212c) serpentinite
bedrock at Jasper Ridge site JR3 (Figure 21b)
45
Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments for the Diablo clay (Site 1 Figure 21b) serpentine soil (Site 2
Figure 21b) and chromite listed in Table 29 The sum refers to the summation of the
three extraction experiments
Figure 214 SEM images of the chromite standard (A) after each sequential extraction
treatment including the (B) exchangeable fraction (C) Fe-crystalline oxide fraction and
(D) silicate fraction The concentration of Cr removed by each treatment is shown
beneath each set of images
Figure 215 Tie-lines from Figure 29 connecting serpentinite chromite and end-
member FeCr2O4 are shown on this single plot Soil analyses from Figure 29 are also
plotted on this diagram For a sum of 100 Wt oxygen must be added to the sum of
Cr Al Fe and Mg Wt
46
Figure 21a
47
Figure 21b
48
Chromium Concentration (mg kg-1)1000 10000
Dep
th (m
eter
s) 01
10
Maryland (Rabenhorst et al 1982)California (Gough et al 1989)California (Hotz 1964)New Foundland (Roberts amp Proctor 1992)
Surface
Figure 22
49
Figure 23
50
Cr (Wt )0 5 10 15 20
Al (
Wt
)
0
5
10
15
20
25
30
Cr (Wt )0 5 10 15 20
Fe (W
t
)
0
20
40
60
80
100
Cr (Wt )0 5 10 15 20
Mg
(Wt
)
0
5
10
15
20
25
Cr (Wt )0 5 10 15 20
Si (W
t
)
0
5
10
15
20
25
30
Cr0 10 20 30 40 50 60 70 80 90 100
Al
0
10
20
30
40
50
60
70
80
90
100
Fe
0
10
20
30
40
50
60
70
80
90
100
Chromite
Chlorite
Magnetite
Chromite-Silicate
Pyroxenes
Chromite
Chlorite
Pyroxenes
Magnetite
Chromite-Silicate
Figure 24
51
Figure 25
52
Figure 26
53
Sites1 2 3 4 5 6 7 8 9 10
Con
cent
ratio
n (m
g kg
-1)
1500
2000
2500
3000
3500
4000
4500
5000Soil Chromium
1
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000002004006008010012014016
2
Cr Concentration (mg kg-1)
1500 2000 2500 3000 3500 4000
Dep
th (m
eter
s)
000002004006008010012014
3
Cr Concentration (mg kg-1)3500 4000 4500 5000
Dep
th (m
eter
s)
000
002
004
006
008
010
012
5
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000002004006008010012
6
Cr Concentration (mg kg-1)2200 2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000002004006008010012
7
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000
002
004
006
008
010
9
Cr Concentration (mg kg-1)
2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000
002
004
006
008
010
012
10
Cr Concentration (mg kg-1)1600 2000 2400 2800 3200
Dep
th (m
eter
s)
000002004006008010012
4
Cr Concentration (mg kg-1)3500 3600 3700 3800
Dep
th (m
eter
s)
000
001
002
003
004
8
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000
002
004
006
008
010
BEDROCK
BEDROCK
BEDROCK
Range of Cr Concentrations in Serpentinite Bedrock
Figure 27
54
Soil Chromium
Depth (cm)0 10 20 30 40 50
Cr m
g kg
-1
02000400060008000
100001200014000
Totalgt2mmSandSiltClay
Soil Nickel
Depth (cm)0 10 20 30 40 50
Ni m
g kg
-1
0
1000
2000
3000
4000
5000
Soil Manganese
Depth (cm)0 10 20 30 40 50
Mn
mg
kg-1
0
1000
2000
3000Totalgt2mmSandSiltClay
Soil Cobalt
Depth (cm)0 10 20 30 40 50
Co
mg
kg-1
0
50
100
150
200
250
300
Totalgt2mmSandSiltClay
Totalgt2mmSandSiltClay
Exceeds detection limits of 10000 mg kg-1 Cr
Figure 28
55
Cr (Wt )0 10 20 30 40 50
Al (W
t
)
0
5
10
15
20
25
30
Cr (Wt )0 10 20 30 40 50
Mg
(Wt
)
0
2
4
6
8
10
12
Cr (Wt )0 10 20 30 40 50
Si (W
t
)
0
2
4
6
8
10
12
14Cr (Wt )
0 10 20 30 40 50
Fe (W
t
)
0
10
20
30
40
50
60
70
Cr (Wt )0 10 20 30 40 50
Al +
Fe
(Wt
)
10
20
30
40
50
60
70
JRBPChromite
FeCr2O4
Chromite
Magnetite
Chromite-Silicate
Figure 29
56
Clay XRD of JR3515 Depth 5-15cm
0
200
400
600
800
1000
1200
0 5 10 15 20 25 30 35 40 45
d-spacing (Aring)
Rel
ativ
e In
tens
ity
Mg-Saturated
Mg+Glycerol
K-saturated
K-saturated+heat (500oC)
Figure 210
57
Energy (kev)
597 598 599 600 601 602 603
Nor
mal
ized
Abs
orba
nce
-4
-3
-2
-1
0
1
2
(a)
(b)
(c)
(d)
(e)
Cr(
VI)
Figure 211
58
Figure 212
59
Diablo Clay
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Serpentine Soil
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Chromite
0
20
40
60
80
100
120
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide
SilicateSum
Figure 213
60
Figure 214
61
Reaction Progress ()0 20 40 60 80 100
Elem
enta
l Wt
0
10
20
30
40
50
ChromiumAluminumIronMagnesium
FeCr2O4Jasper Ridge
Bedrock Chromite
SoilChromite
Figure 215
62
CHAPTER 3
OXIDATIVE PROMOTED DISSOLUTION OF CHROMITE BY MANGANESE
DIOXIDE AND CONCURRENT PRODUCTION OF CHROMATE
ABSTRACT
Chromium release and oxidation from chromite is a potential environmental hazard in
sediments and soils and a pathway for soil development (weathering of primary minerals)
related to ultramafic rocks and their metamorphic derivatives (serpentinites) Birnessite is
a common pedologic mineral in these sediments and soils capable of oxidizing aqueous
Cr(III) In this study the interaction between chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4)
and birnessite is investigated with an interest in the potential generation of Cr(VI)
Specifically the effects of chromite suspension density birnessite suspension density
and pH at a temperature of ~25degC in relation to the reaction kinetics were examined The
rate of Cr(VI) in solution increases with increasing chromite suspension densities and
decreasing pH but is independent of birnessite suspension densities at values greater than
~20 m2 L-1 The overall rate expression of Cr released and oxidized in solution from
chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression with Cr(VI) rates ranging from 9times10-4 to 44times10-3 microM h-1
Additionally these experiments demonstrate that serpentine soils are a source of non-
anthropogenic Cr(VI)
63
INTRODUCTION
Chromite a Cr-enriched spinel is responsible for a majority of the Cr enrichment
identified in ultramafic rocks serpentinites and serpentine soils (Rabenhorst et al 1982
Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989 Gasser and
Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003 Oze et al
2003a Oze et al 2003b) The chemical weathering of chromite is therefore an important
process when evaluating the potential of ultramafic rocks and their related sediments and
soils as sources of Cr Despite the propensity of chromite to resist chemical weathering
the dissolution of chromite and release of Cr will occur over geologic time (Bialowolska
and Salacinski 1984 Oze et al 2003b) Chromium in chromite is in the form of Cr(III)
the non-toxic and chemically immobile valence state however oxidizing conditions in
near surface environments is a potential reaction leading to the enhanced dissolution of
chromite and the formation of Cr(VI) the toxic and chemically mobile valence state
which typically exists as the oxyanion chromate (HxCrO42-x) (Daugherty 1992 Cohen et
al 1993 Fendorf 1995 Deer et al 1996) Consequently chromite may be a non-
anthropogenic source of Cr(VI) in the environment
Naturally occurring high valence Mn oxides present in soils and sediments
derived from ultramafic rocks are strong oxidizers capable of affecting the chemical
mobility of metals through redox reactions (Fendorf and Zasoski 1991 Johnson and
Xyla 1991) These insoluble mixed valent oxides (Mn(IIIIV)) are observed coating
mineral surfaces in these ultramafic sediments and soils where total Mn concentrations
are often gt100 mg kg-1 (Brooks 1987 Schreier et al 1987 Oze et al 2003a Oze et al
2003b Oze et al) Birnessite is a naturally occurring mixed valence oxide known to
oxidize aqueous Cr(III) to the more chemically mobile Cr(VI) (Eary and Rai 1987
Fendorf and Zasoski 1991) Gough et al (1989) and Becquer et al (2003) have
suggested that Mn oxides are responsible for Cr(VI) identified in California and New
Caledonia serpentine soil solutions however the dissolution and oxidation of Cr from a
Cr-bearing mineral due to the presence of birnessite has not been quantitatively defined
In this study we examine the kinetics of Cr release and oxidation when chromite and
birnessite are present in solution
64
The objective of this study is to determine a rate expression describing the
release and oxidation of Cr from chromite in the presence of birnessite Using well-
controlled batch experiments the solution chemistry and geochemistry of Cr related to
chromite and birnessite interaction was evaluated temporally over a range of pH
Conditions were chosen to simulate the slightly acidic conditions characteristic of most
serpentine soils (Rabenhorst et al 1982 Gough et al 1989 Graham et al 1990 Cole
1992 Jeffrey 1992 Lee 1992 Verger 1992 Becquer et al 2003) Ultimately these
experiments allow the examination of one potential pathway leading to Cr(VI) identified
in solutions related to ultramafic rocks and serpentine soils
MATERIALS AND METHODS
Solid Phase Characterization and Synthesis
Chromite used in the experiments was commercial-grade chromite (MSDS No
302-1 WHMIS CLASS D2A) obtained from Barnes Environmental Inc Ontario
Chromite was prepared by cleaning the sand-size material in multiple ultrasonic
isopropanol baths and rinses The chromite was then sieved using a Retsch magnetic
sieve shaker for 30 minutes in order to obtain the 160-250 microm size fraction Impurities in
the 160-250 microm chromite were removed using a slope Franz magnetic separator These
impurities included the minerals olivine augite plagioclase feldspar and quartz Visual
and scanning electron microprobe (SEM) examinations of the remaining 160-250 microm
fraction confirmed its purity Chromite was crushed to a fine powder using an agate
mortar and pestle and then rinsed multiple times in deionized destilled (DDI) water and
001 M HCl solution to remove the ultra-fine material and eliminate highly reactive sites
caused during grinding The chromite powder was inspected before (Figure 31) and
after the experiments using a scanning electron microscope (JEOL 5600LV) the surface
area was 025 m2 g-1 as determined using N2(g) and BET isotherm
An automated JEOL 733A electron microprobe operated at 15 kV accelerating
potential and 15 nA beam current was used to investigate possible zoning and to quantify
the chemical composition of the chromite grains prior to powdering No zoning was
observed and the chemical composition of the chromite was based on the average of 25
electron microprobe analyses The stoichiometric composition of chromite based on 4
65
oxygens is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 X-ray diffraction was also used to
deduce the purity of the chromite powder X-ray diffraction spectra obtained for the
chromite powder matched JCPDS card 03-0873 (unit cell a=8300 b=8300 c=8300
α=90 β=90 and γ=90) and no contaminants were evident
Birnessite was synthesized using the procedure described by Buser et al (1954)
and implemented by Fendorf and Zasoski (1992) A solution of 63 g of potassium
permanganate (KMnO4) in 1 L of DDI water was heated to 90 degC Approximately 100
mL of the heated slurry was poured into a 4 L beaker and reheated to 90degC while being
stirred vigorously The remaining 900 mL of slurry and 66 mL of concentrated
hydrochloric acid were simultaneously and proportionally dispensed into the beaker over
a period of thirty seconds The temperature of the resulting solution was maintained at
90deg C for an additional 10 minutes and then allowed to cool to room temperature The
product was washed with alternate rinses of DDI water and a 001 M HCl solution The
birnessite surface area was 125 m2 g-1 based on BET analyses Solids were inspected
before (Figure 31) and after the experiments using the SEM
Extraction Experiment Procedures
Several chemical extraction experiments were performed on 106-250 microm
chromite grains prior to the chromite-birnessite batch experiments in order to determine
the potential release of Cr Fe and Mn One gram of chromite and 25 mL of solution
were placed into a 25 mL centrifuge tube and the solids and solution were gently agitated
The solutions and parameters used for these extraction experiments include the
following 1) 10 hydrogen peroxide (H2O2) solution at 25degC and pH of ~5 for 6 h 2)
01 M potassium permanganate (KMnO4) solution at 60degC and pH of ~3 for 6 h 3) 10 M
hydrofluoric (HF) acid at 60degC for 6 h 4) 1 M hydroxylamine (NH2OHmiddotHCl) with 25
(vv) acetic acid (HOAc) at 60degC and pH of ~4 for 6 h and 5) 1 mM barium chloride
(BaCl2) solution at 60degC and pH of ~6 for 6 h Solutions were extracted with a syringe
and passed through a 02 microm filter Solids were inspected after the experiments using the
SEM
66
Batch Experiment Procedures
Batch experiments were performed in triplicate using 1 L polypropylene bottles
and chromite and birnessite ratios listed in Table 31 at ~25degC After loading chromite
and birnessite into each bottle 1 L of a pH 3 solution a 10 mM acetate solution (pH 5) a
10 mM MES solution (pH 67) or a pH 8 solution was added to the appropriate
experiment (Table 31) Chromite and birnessite were also run individually in the acetate
buffered solution (Table 31) Aliquots were extracted from the polypropylene bottles
using a syringe and then passed through a 2 microm filter before analyses Less than 40 mL
was extracted over the course of each experiment resulting in a total volume change of
lt4 After the experiments were terminated the remaining solids were resuspended in
20 mL of 1 mM phosphate (5050 KH2PO4K2HPO4 pH 7) 20 mL of 10 mM CaCl2
solution at pH 7 and a 05 M HCl solution These suspensions were agitated for 3 h
followed by centrifugation and filtration (02 microm) before analyses The purpose of the
phosphate and CaCl2 solutions was to desorb reaction produtcts (eg Cr(VI) Cr(III) Fe
and Mn) from the solid phases The HCl treatment provides a means to evaluate
secondary phases such (oxy)hydroxides formed during the experiments Although the
HCl solution may dissolve chromite and birnessite in addition to the (oxy)hydroxides the
short time frame of the extraction (3 h) should limit the dissolution of chromite
Additionally the polypropylene bottles following the termination of five batch
experiments were rinsed with DDI water and 20 mL of 05 M HCl was added to each
bottle The HCl solution was allowed to remain in the bottles for 15 h in order to
determine the concentration of Cr Fe and Mn sorbed onto the interior lining
Serpentine Soil and Birnessite Experiments
Serpentine soils collected from Jasper Ridge Biological Preserve (Stanford CA)
and New Caledonia were used to determine the influence of birnessite in soils known to
contain chromite (Chapters 2 and 4) The Cr geochemistry of these soils as well as the
detrital chromite have been evaluated in Chapters 2 and 4 and by Becquer et al (2003)
Jasper Ridge serpentine soils contain Cr-bearing minerals including chromite Cr-
magnetite and Cr-chlorite however a majority of the Cr enrichment (Table 32) is
directly related to the abundance of chromite A majority of the Cr related to New
67
Caledonia serpentine soils is attributed to Cr (oxy)hydroxides (Chapter 4) however
chromite is present in these soils as well (Becquer et al 2003) Although birnessite has
been noted in these soils the quantity of birnessite could not be determined In order to
evaluate the impact of Mn oxides on Cr within such soils 01 g of birnessite was added to
each soil
The parameters for the experiments are listed in Table 32 Ten grams of the
lt250 microm size-fraction of each soil (JR1 JR2 and NC) and 100 mL of 1 mM acetate
buffered solution (pH 5) were loaded into individual 150 mL Nalgene bottles Aqueous
solutions were evaluated as a function of time (h) Additionally 10 g of each soil 01 g
of birnessite and 100 mL of 1 mM acetate buffered solution (pH 5) were loaded into 150
mL Nalgene bottles and were allowed to react as a function of time (h) These
experiments were rotated at a rate of 2 revolutions min-1 Solutions were removed with a
syringe and passed through a 2 microm filter before analysis Approximately 8 mL of
solution was required for the analyses over the course of each experiment resulting in a
total volume change of 8 Total Cr concentrations (mg kg-1) of the soils listed in Table
32 were obtained by completely dissolving the samples using a mixture of hot (gt200degC)
concentrated nitric perchloric and hydrofluoric acids Elemental compositions were
measured in the supernatant using inductively coupled plasma optical emission
spectroscopy (ICP-OES) The mass of chromite in each soil was calculated using total Cr
concentration for each soil and the calculations provided in Chapters 2 (Table 210) and
Chapter 4 (Table 47) The chromite surface area of each soil was approximated by
comparing the surface area measured for the chromite standard and its observed grain
size from SEM images (Figure 31) to chromites observed in the soils using backscattered
electron images (Chapters 2 and 4) The surface area of the Jasper Ridge chromite is 066
of the chromite standard due to a slightly larger grain size The surface area of the
New Caledonia chromite is approximately 025 of the chromite standard Chromite
suspension densities were calculated by multiplying the chromite mass by the estimated
surface area and then dividing by the volume of the solution
68
Solution Analysis
The pH and Eh of the supernatant was measured with an Orion flat surface
combination pH electrode inserted directly into supernatant Production of soluble
Cr(VI) was measured spectrophotometrically at 540 nm using the s-diphenyl carbazide
method modified from Bartlett and James (1979) Soluble Fe(II) was monitored using
the ferrozine method adopted from Stookey (1970) Total dissolved Cr Fe Al Mn and
Mg were ascertained using ICP-OES Concentrated trace metal grade nitric acid used to
acidify the samples prior to analyses Aluminum and Mg was determined to be present in
the nitric acid at concentrations lt1 microM Magnesium and Al are not plotted in any of the
figures Chromium(III) and Fe(III) was determined by subtracting Cr(VI) and Fe(II)
obtained from colorimetric methods from total dissolved Cr and Fe analyses using ICP-
AES Results are presented as an average of the three runs for each batch experiment
with the standard deviation noted by error bars
RESULTS
Chromite Extraction Experiments
The release of Cr Fe and Mn from chromite using 10 H2O2 01 M KMnO4 10
M HF acid 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 1 mM BaCl2 solutions was
carried out to determine the relative propensity of chromite to resist chemical dissolution
and these results are summarized in Figure 32 The KMnO4 solution with aqueous
Mn(VII) was the most effective in releasing Cr from chromite yielding a concentration
of 62 Cr mg L-1 Iron released using the KMnO4 solution was 03 mg L-1 Although
H2O2 is a direct chemical oxidant the release of Cr from chromite was less as compared
to the other solutions (only ~03 Cr mg L-1 was measured) Concentrated HF was only
able to release ~13 Cr mg L-1 while releasing 106 Fe mg L-1 NH2OHmiddotHCl with 25
(vv) HOAc and CaCl2 solutions only removed lt1 Cr mg L-1 however the NH2OHmiddotHCl
removed 36 Fe mg L-1 which is comparatively higher than the H2O2 KMnO4 and BaCl2
solutions
69
Chromite and Birnessite Batch Experiments
The interaction of chromite and birnessite leading to the release and oxidation of
Cr(III) is a relatively slow process requiring study on the order of hours and days The
pH was maintained constant throughout the duration of the experiments while Eh values
ranged from 320 to 450 mV with no discernable temporal pattern (Figure 33) For pH
values between 3 and 67 Cr3+ CrOH2+ and Cr(OH)2+ are the dominant aqueous Cr(III)
species however at pH 8 conditions are on the stability line between Cr(III) and Cr(VI)
specifically Cr(OH)3 and CrO42-
Higher chromite suspension densities produced higher rates of Cr release to the
aqueous phase and all the Cr in solution was in the form of Cr(VI) (Figure 35) The
rates of oxidation based on Cr(VI) production range from 00005 to 00041 microM h-1
(Figure 35 Table 33) Comparatively Cr released from the chromite control experiment
is maintained at a relatively constant value of ~005 microM and no Cr(VI) is measured in
solution Iron concentrations initially increase to ~ 05 microM and then decrease rapidly to a
relatively constant concentration at 02 microM (Figure 35) Iron concentrations for the
chromite control experiment began at 15 microM between 0 to 50 h and decreased to ~02
microM Iron(II) and Mn concentrations are below detectable limits for all experiments
(Figure 35) A zeroth-order dependence is assumed with respect to birnessite owing to
its large excess surface area relative to chromite
Variable birnessite suspension densities and constant chromite suspension
densities were evaluated at pH 5 (Figure 36) Although suspension densities of
birnessite are 538 and 1075 m2 L-1 (Table 31) the rate of Cr and Cr(VI) measured in
solution for both experiments is 00041 microM h-1 (Figure 36 and Table 33) Iron
concentrations are below 06 microM and maintain concentrations near 02 microM with no
indication of consistently increasing or decreasing with respect to time Iron(II) and Mn
concentrations in solution are below detectable limits for both experiments
In Figure 37 the influence of pH on chromite oxidation by birnessite was
evaluated by conducting experiments at pH values of 3 5 67 and 8 chromite and
birnessite suspension densities are constant at 0075 m2 L-1 and 1075 m2 L-1 respectively
(Table 31) Chromium and Cr(VI) rates increase with decreasing pH ranging from 0
microM h-1 at pH 8 to 00062 microM h-1 at pH 3 (Table 33) Iron is present in solution at
70
concentrations less than 04 microM for all the experiments Generally Fe concentrations are
greater near the start of the experiment and decrease to values lt 05 microM with increasing
reaction time Manganese concentrations are below detectable limits
In order to determine the rate dependence on each prominent reactant (chromite
birnessite and pH) reaction rates were regressed against reactant concentrations (Figure
38) A pseudo first-order rate dependence is noted based on chromite suspension
density the equation regressed for this relationship is listed in Table 33 As the chromite
suspension density increases the rate of Cr(VI) in solution increases In contrast a
zeroth-order rate dependence is evident for birnessite suspension density (dashed line in
Figure 38b) The rate of Cr(VI) generation is expected to decrease with respect to
birnessite suspension density at very low concentrations (dotted line in Figure 38b)
The influence of pH on the rate of Cr(VI) production is well-described by a polynomial
function (Figure 38c Table 33)
Developing an overall rate expression for Cr in solution requires integrating the
factors assessed in Figure 38 and listed in Table 33 All the Cr measured in solution is
in form of Cr(VI) therefore the rate law for Cr or Cr(VI) in solution may be written as
d[Cr(VI))]dt = k sdotChromitea sdot Birnessiteb sdot (aH+)c (31)
and
log d[Cr(VI))]dt = log k + asdotlogChromite+ bsdot log Birnessite+ csdotlog (aH+) (32)
where k is the overall rate constant for Cr release and oxidation Chromite is the
chromite suspension density (Table 31) Birnessite is the birnessite suspension density
(Table 31) and aH+ is the activity of H+ Constants a b and c are the order of
dependency for each reactant The overall rate equation (developed in the Appendix) can
be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069 (33)
71
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 the expression is valid for birnessite suspension densities
greater than 20 m2 L-1 and pH values between 3 to 8
Secondary and Solid Phase Examination of Chromite and Birnessite
The solid products were evaluated to determine the abundance and type of
secondary phases formed during the course of the experiments Solids from the batch
experiment were treated with three solutions 1 mM 5050 KH2PO4K2HPO4 1 mM
CaCl2 and 05 M HCl (Table 34) In the chromite and birnessite reactions Cr was
primarily in solution (gt90 ) The HCl extract was able to remove lt5 additional Cr
while the KH2PO4K2HPO4 and CaCl2 extracts released lt3 additional Cr Aqueous Fe
concentrations were below detectable limits for the final solution extraction Treatment
with the HCl resulted in up to 0127 micromoles of Fe however 0107 micromoles of Fe were
released upon treatment of unreacted chromite with HCl Although not shown in Table
34 Mn values were greater than gt4 micromoles when treated with HCl the
KH2PO4K2HPO4 and CaCl2 removed no detectable Mn
After the termination of each experiment the solids were examined using SEM in
order to identify changes in the surface morphology and secondary phases A large
chromite grain and conglomerate of birnessite adjacent to the chromite are depicted in
Figure 38 The face and edge of the chromite grain is not noticeably altered compared to
grains examined prior the experiments Additionally the particle morphology of
birnessite has not visibly changed Even at higher magnification angular and
conchoidally fractured fine-grained edges are noted (Figure 38bc) Only a slight
rounding of the smaller features of the main chromite grain provides any evidence of
possible chromite dissolution (Figure 38c) Birnessite appears to have limited interaction
with chromite Although chromite and birnessite were allowed to interact over 500 h in
an aqueous solution no secondary phases are observed by SEM
Serpentine Soil and Birnessite Experiments
Two serpentine soils from Jasper Ridge Biological Preserve CA and one
serpentine soil from New Caledonia were used to determine whether or not the overall
72
rate expression (Equation 33) for chromite reaction with birnessite is consistent with
Cr(VI) concentrations produced by serpentine soils containing chromite Three
serpentine soils without birnessite and three soils with the addition of birnessite buffered
at pH 5 were measured for Cr(VI) generation as a function time (Figure 310) The rates
of Cr(VI) measured in solution for both Jasper Ridge soils (JR1 and JR2) are identical at
00013 microM h-1 With the addition of birnessite the rates of Cr(VI) production in solution
both increased to 00044 microM h-1 (JR1 w Bt and JR2 wBt in Figure 310) Based on the
chromite suspension density (Table 32) the rates predicted by Equation 33 range from
00035 to 00041 microM h-1 which are 10 to 20 less than those observed in the
experiments (Figure 310) The rates of Cr(VI) in solution from the New Caledonia
serpentine soil are 00009 microM h-1 and with the addition of birnessite increase up to 00013
microM h-1 (Figure 310) A rate of 00016 microM h-1 was predicted by Equation 33 for the
New Caledonia soil
DISCUSSION
The reaction of birnessite with chromite enhances the rate of Cr release into
solution The rate of Cr(VI) measured in solution (Equation 33) is directly related to the
chromite suspension density and is independent of the birnessite at suspension densities
greater than ~20 m2 L-1 Although the release and oxidation of Cr in solution are
measured the mechanisms and pathways driving this reaction need to be evaluated The
propensity of chromite to resist dissolution the effects of pH the formation of secondary
phases and ultimately the pathways of Cr release and oxidation are addressed in the
following discussion
Extraction experiments performed on chromite (Figure 32) demonstrate it is
resistant to acidic dissolution or competitive ion displacement but strong oxidants such
as high valent Mn may oxidatively dissolve chromite It is evident that the proportions of
Cr Fe and Mn vary for each of the extractions The solutions utilized in the extraction
experiments have varying chemical properties capable of preferentially releasing a
particular species andor element These experiments were also performed over a short
period of time (lt6 h) which may have only released elements from highly reactive sites
73
Finally secondary phases associated with these experiments were possible and would
distort the elemental ratios observed in solution
A majority of the Eh and pH analyses plot in the Cr(III) stability region (Figure
33) however Cr in solution is primarily in the form of Cr(VI) (Figures 35 36 and
37) Eh measurements presented in Figures 32 and 33 do not account for the
microenvironments associated with the surfaces of birnessite Additionally the kinetics
of Cr(VI) reduction are limited in reducing conditions due to electron symmetry
constraints and the absence of a suitable redox couple (Fendorf 1995) Finally highly
oxidizing birnessite in the solution is kinetically and thermodynamically controlling the
oxidation state of Cr relative to the stabilities predicted in Figure 33 (Amacher and
Baker 1982 Fendorf and Zasoski 1992)
A fractional-order dependence on the chromite suspension density (Figure 38)
can be explained by diffusional influences competing oxidation reactions and secondary
reactions involving Cr(III) and Mn(IV) Secondary phases on the surface of chromite
may inhibit the diffusion of Cr from chromite however the clean and relatively unaltered
chromite surfaces do not support this mechanism as a major cause (Figure 38)
Chromium and Fe released from chromite may be incorporated into secondary phases
related to birnessite or chromite Approximately 10 of Cr is not in the aqueous phase
(Table 34) supporting the premise that secondary phases are contributing to the
fractional-order dependence related to the chromite suspension density Also
competition from Fe(II) oxidation on birnessite deposition of Fe(II) on birnessite (or
chromite) and diffusion onwithin birnessite may further impact the rate dependence
The zeroth-order dependence based on the birnessite suspension density (Figure
38) results from its high surface area At very low birnessite suspension densities the
rate of Cr(VI) production should decrease however due to the high surface area of
birnessite it was experimentally prohibitive to determine reaction rates at masses less
than 4 mg Therefore a hypothetical trend-line depicts the expected dependence on
birnessite concentration at very low suspension densities at higher values representative
of most field settings the reaction is independent of the birnessite surface area (ie zero-
order dependence)
74
In Figure 38c the polynomial relationship defines the Cr(VI) rate dependence on
pH As the pH decreases the rate of Cr(VI) increases more rapidly The relationship is
potentially the result of lower pHs progressively increasing the rate of chromite
dissolution However the rate of Cr(VI) in solution might be influenced by higher pHs
increasing the formation of secondary Fe(III) and Cr(III) (oxy)hydroxides on the
chromite andor birnessite surfaces thereby decreasing elemental diffusion from
chromite or decreasing available birnessite sites for Cr(III) oxidation Differentiating
between these two mechanisms is not possible based on the rate equation or extraction
analyses however the polynomial relationship evident in Figure 38c is folded into the
rate constant in order to geochemically account for the influence of pH on Cr(VI)
generation
No conclusive evidence is available to determine the fate of Mn within the
reacting suspensions Aqueous Mn concentrations are consistently maintained below
detectable limits If Mn is released from birnessite then it must be strongly and quickly
incorporated into a nonaqueous phase following its release Due to Cr(III) being oxidized
by Mn(III) andor Mn(IV) Mn(II) must be produced Extractions performed on the
chromite and birnessite solids demonstrate that Mn could only be removed with HCl
however a majority of this Mn is potentially the result of HCl dissolving birnessite On
the basis of electron transfer stoichiometry at least 15 mols of Mn(II) must be produced
for each mol of Cr(VI) generated This amount can be accounted for in the HCl
extraction where gt4 micromoles of Mn were measured Additionally Fendorf et al (1993)
determined that Mn(II) should not significantly influence the extent of Cr(III) oxidation
Iron in solution does not progressively increase in solution over time (Figures 35
36 and 37) The high initial rate of Fe release compared to Cr may be the result highly
reactive sites preferentially releasing Fe Although no Fe is detected in solution from
the extractions performed on the chromite and birnessite solids 0005 to 0035 moles of
Fe were retrieved from the lining of the bottle and 0053 to 0126 moles were released
from the solids using the HCl solution (Table 34) Iron(II) could potentially be oxidized
to Fe(II) in the presence of birnessite and form stable Fe(III)-(oxy)hydroxides (Postma
1985 Myers and Nealson 1988 Villinski et al 2001) A ratio of two Fe for every one
Mn is necessary for the oxidation of Fe(II) and this amount is recoverable based on HCl
75
extractions with gt4 micromoles of Mn measured from the HCl extractions Additionally
dissolved O2 in the batch experiments will rapidly oxidize Fe(II) for all the pHs examined
in these experiments (Stumm and Morgan 1981 Langmuir 1997) The oxidative
pathway for Fe(II) provides an explanation for the lack of Fe(II) in solution for the batch
experiments (Figure 35 36 and 37)
With a detection limit of ~02 microM and equilibrium with respect to Fe(OH)3(am) and
goethite Fe concentrations should either be at or below detection limits at pHs gt 5
Assuming aqueous Fe(III) is present dominantly as the second hydrolysis product at pH
5 the dissolved concentrations approach the solubility of ferrihydrite and are much
greater than expected for goethite (Figure 311) Thus it is probably that the former solid
is generated which is further supported by the chemical extraction experiments (Table
34)
Iron and Cr solution concentrations measured experimentally are not consistent
with congruent dissolution (Figure 312a) Assuming Fe was forming secondary phases
at a constant rate (~00001 Fe h-1) multiplying this rate by the time analysis and then
adding it to the measured aqueous Fe a corrected Fe (Fe) value can be estimated (Figure
312) Additionally the 10 of Cr not in solution can be also be used to calculate a
corrected Cr value (Cr) (Figure 312) When secondary phases are taken into account a
majority of the analyses plot near the congruent dissolution of chromite This provides
evidence that chromite is undergoing congruent dissolution within the batch experiments
but that secondary Fe(III) phases disrupt the solution phase elemental ratios
Chromium(VI) Pathways from Chromite and Birnessite into Solution
Three pathways are evaluated for Cr(VI) measured in solution a) chromite
dissolution b) birnessite dissolution and c) solid-solution reaction (Figure 313) In order
to describe these pathways end-member chromite (FeCr2O4) and birnessite (δ-MnO2) are
used to simplify the analysis It should be recalled however that the composition of the
chromite used in the experiments is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4
Chromite dissolution presents a likely explanation for Cr(VI) observed in solution
based on the following discussion In chromite dissolution model Cr(III) released from
chromite adsorbs onto the surface of birnessite and oxidizes to Cr(VI) After the electron
76
transfer is complete Cr(VI) desorbs from birnessite and remains in solution As this
process continues Cr(VI) increases at a constant rate The rate limiting step with regards
to this specific pathway can be grouped into two steps a) dissolution of chromite (Step
1a) or b) Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) (Figure 313)
Although a dissolution rate for chromite at ambient temperatures and pressures is not
known the chromite extraction experiments (Figure 32) demonstrates that chromite
dissolution is slow Manganese oxides such as birnessite oxidize Cr(III) in solution
rapidly with a measured rate constant of 6300 microM-1 s-1 in fresh water (100 mg L-1 MnO2
100 g L-1 Cr(III) pH~6) (Van der Weijden and Reith 1982) This means that the rate
for Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) is on the order of
seconds therefore chromite dissolution is likely the rate limiting step controlling the
concentration of Cr(VI) in solution for this pathway
Although a zeroth-order dependence related to birnessite concentrations (Figure
37a) is observed the birnessite dissolution pathway in which aqueous Mn4+ released
from birnessite adsorbs onto the surface of chromite oxidizing Cr(III) cannot be entirely
dismissed Manganese dissolution and adsorption could occur quickly not allowing the
Mn to be measured in solution Examination of the solids after the termination of each
experiment demonstrated extremely minimal chromite and birnessite interaction
therefore the solid solution reaction is not a likely pathway leading to the production of
Cr(VI) in solution and is furthermore inconsistent with the noted rate dependencies
Despite the three pathways (Figure 313) the generation of Cr(VI) observed in these
experiments and described by the rate equation (Equation 33) is independent of these
potential pathways
Serpentine Soils and Cr(VI)
The serpentine soil experiments (Figure 310) provide direct evidence that Cr(VI)
is forming within these soils By adding birnessite to the soil the rate of Cr(VI) in
solution increased for all experiments (00013 to 00044 microM Cr(VI) h-1 for Jasper Ridge
and 00009 to 00013 microM Cr(VI) h-1 for New Caledonia) The Cr(VI) rate increase
related to the addition of birnessite is likely the result of low birnessite suspension
densities (lt20 m2 L-1) in the soil Rates predicted for Jasper Ridge and New Caledonia
77
soils using Equation 33 and values in Table 32 are similar to those measured in the
experiments with values of 00041 and 00016 microM Cr(VI) h-1 respectively Overall these
soil experiments demonstrate that a rate equation developed from laboratory batch
experiments predicts the rate of Cr(VI) in serpentine soils reasonably well
According to limits set by the Environmental Protection Agency the California
standard for total Cr in drinking water is 50 microg L-1 (~1 microM) while the federal standard is
100 microg L-1 (~2 microM) The serpentine soil and birnessite experiments produced Cr(VI)
concentrations in excess of 50 microg L-1 after only 140 h or roughly 6 days Serpentine soils
at Jasper Ridge typically remain water saturated for prolonged periods of time (gt1 week)
during the winter season increasing the time Cr(VI) can be produced The degree and
duration of water saturation is dependent on the topography of the serpentinite body
level areas will remain water saturated while sloping areas will allow the water to flow
down gradient Although Cr(VI) was generated in the soil experiments the production of
Cr(VI) in serpentine soils may be more limited The reduction of Cr(VI) to Cr(III) is
favorable by reactions with Fe(II)aq Fe(II)-containing minerals such as magnetite H2S
and organic matter including microbial processes (Ilton and Veblen 1994 Fendorf 1995
Peterson 1996) Both organic matter and magnetite are abundant in Jasper Ridge
serpentine soils (Chapters 2 and 4) however these factors did not deter Cr(VI) from
forming during the serpentine soil experiments Additionally the soils in these
experiments were constantly rotated allowing the mineral surfaces to be continuously
exposed to solution Chromite within an undisturbed soil will not always be in contact
with the solution or proximal to birnessite
Chromium(VI) produced by the interaction of chromite and birnessite is not
limited to serpentine soils Serpentinites the bedrock for serpentine soils are typically
highly sheared due to the diapiric migration of serpentinite masses along zones of
weakness in fault zones Numerous cracks and lineations which are typical of these rocks
provide a means for Cr(VI) to transported away from the serpentine soil potentially into
local groundwater In fact preferential flow will occur along the major shear plains of
the serpentinite Additionally these shear plains are often coated with high valent Mn
oxides providing a reaction pathway for Cr(VI) formation Serpentinite springs such as
those evaluated by Barnes and ONeil (1969) and Barnes et al (1972) at the Cedars
78
ultramafic body in Cazadero CA have Cr concentrations measured at 20 microg L-1 in which
all the Cr is in the form of Cr(VI) (Chapter 4) In conclusion chromite and birnessite in
serpentine soils could contribute a source of Cr(VI) in surface waters and groundwaters
Chromium release and oxidation from chromite is increased by the presence of
birnessite and the interaction between chromite and birnessite is a pathway for Cr(VI)
generation in ultramafic sediments and soils The potential of serpentine soils as a source
of non-anthropogenic Cr contamination can now be evaluated quantitatively based on the
abundance of chromite The overall rate law for Cr(VI) in solution (Equation 39)
provides a means to estimate the Cr(VI) concentration attainable for serpentine soils and
serpentine sediments when the chromite and birnessite suspension densities and the pH of
the soil are known
ACKNOWLEDGEMENTS
We gratefully acknowledge comments and assistance from Robert Jones and Dr
Guangchao Li Funding for this research was provided by the Office of Teaching and
Learning at Stanford University
79
Table 31 Chromite and birnessite batch experiment parameters with chromite and birnessite suspension densities
Experiment Chromitea (grams)
Birnessiteb (grams) pHc
Chromite Surface Aread
(m2 g-1)
Birnessite Surface Aread
(m2 g-1)
Chromite Suspension
Density (m2 L-1)
Birnessite Suspension
Density (m2 L-1)
A 08 043 5 025 125 02 54 B
08 086 5 025 125 02 108C 03 086 5 025 125 008 108D 008 086 5 025 125 002 108E 004 086 5 025 125 001 108F 03 086 3 025 125 008 108G 03 086 67 025 125 008 108H 03 086 8 025 125 008 108
Chromite 08 ---- 5 025 125 02 ----Birnessite ---- 043 5 025 125 ---- 5480
a Chromite chemical composition is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 b Synthetic birnessite (δ-MnO2) c10 mM acetate buffered solution pH = 5 non-buffered pH=3 5 mM MES buffer pH 67 non-buffered pH 8 dMeasured BET analyses
81
Table 32 Serpentine soil and birnessite experiments with soil and birnessite masses used and the estimated chromite suspension density
Samples Soil Mass (g)
Birnessite Mass (g)
Volume (L) of Solutiona
Total Crsoil
b (mg kg-1)
Chromite Mass (g) in
Soil Samplec Estimated Surface
Areac (m2 g-1)
Chromite Suspension Density
(m2 L-1) in Soild Jasper Ridge
(JR1) 10 ---- 01 5976 013 015 02 Jasper Ridge (JR1 w Bt) 10 01 01 5976 013 015 02 Jasper Ridge
(JR2) 10 ---- 01 4978 011 015 016 Jasper Ridge (JR2 w Bt) 10 01 01 4978 011 015 016
New Caledonia (NC) 10 ---- 01 827 0015 006 001
New Caledonia (NC w Bt) 10 01 01 827 0015 006 001
a1mM Acetate Buffer (pH=5) 81 bTotal Cr concentrations were obtained through total digestion and ICP-AES analyses
cEstimated chromite mass and surface area (discussed in the Methods section) was made using chromite observations and calculations in Chapters 2 and 4
dCalculated from the chromite mass in the soil sample estimated surface area of the chromite and the volume of solution
82
Table 33 Summary of Cr(VI) rates obtained from batch experiments (Figure 35 36 and 37) and used in Figure 38
Listed in Table 31
Suspension Densities (m2 L-1)
Chromite (Ch)
Birnessite (Bt)
pH Rate (microM h-1) (d[Cr(VI)]dt)
log Rate (log microM h-1) (log d[Cr(VI)]dt) Expression
Figure 35 02 108 5 00041 -239
008 108 5 00021 -268 002 108 5 00009 -305 001 108 5 00005 -330
1
Figure 36 02 54 5 00041 -239 02 108 5 00041 -239 2
Figure 37 008 108 3 00062 -221 008 108 5 00021 -268 008 108 67 00004 -340 008 108 8 0 ----
3
Expression 1 log d[Cr(VI)]dt = asdot log Chromite +log krsquorsquo log d[Cr(VI)]dt = 06908 log Chromite ndash 1899 Expression 2 log d[Cr(VI)]dt = bsdot log Birnessite + log krsquorsquorsquo
log d[Cr(VI)]dt = 0sdot log Birnessite ndash 23872 log d[Cr(VI)]dt = -23872
Expression 3 d[Cr(VI)]dt = 264E-4sdotpH2 ndash 414E-3sdotpH + 162E-2
82
Table 34 Extractions comparing total solution and extractable Cr and Fe from the solids
------------------Cr----------------- ------------------Fe---------------- microMolesdagger Total
mole Extraction
mole microMoles Total mole
Extraction mole
Chromite Ch 02 (Table 31) Terminated at 30 days Solutiona 0033 554 ---- 0 0 ---- Phosphateb 0000 05 1 0000 03 03 CaCl2
c 0013 388 87 0000 00 00 HCld 0002 53 12 0103 997 997 Bottle HCle 0000 00 0 0000 00 00
A Ch 02 Bt 54 (Table 31) Terminated at 30 days Solution 2977 910 ---- 0 0 ---- Phosphate 0048 16 18 0004 21 21 CaCl2 0055 18 20 0002 09 09 HCl 0161 54 60 0126 758 758 Bottle HCl 0004 01 2 0035 212 212
B Ch 02 Bt 108 (Table 31) Terminated at 30 days Solution 2837 914 ---- 0 0 ---- Phosphate 0039 14 16 0008 49 49 CaCl2 0060 21 25 0000 00 00 HCl 0142 50 58 0127 803 803 Bottle HCl 0004 01 2 0024 148 148
C Ch 08 Bt 108 (Table 31) Terminated at 17 days Solution 0877 894 ---- 0 0 ---- Phosphate 0020 23 22 0002 37 37 CaCl2 0030 34 32 0000 00 00 HCl 0040 46 43 0053 881 881 Bottle HCl 0003 03 3 0005 82 82
daggerMolarity values are normalized to the final solution analysis of each experiment based on volume of solution and mass of the sample in each experiment aSolution refers to the final solution analysis before the experiment was terminated bmicroMoles extracted from the batch experiment solids using a 1 mM phosphate buffer pH 7 3 h cmicroMoles extracted from the batch experiment solids using a 10 mM CaCl2 solution pH 7 3 h dmicroMoles extracted from the batch experiment solids using a 05 N HCl solution 3 h emicroMoles extracted from the batch experiment bottles using a 05 N HCl solution 15 h
83
Figure Captions
Figure 31 Scanning electron microscopic (SEM) images of chromite and birnessite at
varying magnifications prior to the batch experiments
Figure 32 A summary of Cr Fe and Mn released from chromite during extraction
experiments Solutions used for these experiments include 1) 10 H2O2 solution 2) 01
M KMnO4 solution 3) 10 M HF 4) 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 5) 1mM
BaCl2 solution () Fe concentration for the HF extraction is 106 mg L-1
Figure 33 pH and Eh measurements as a function of time (h) for the batch
experiments Chromite (Ch) and birnessite (Bt) suspension densities used for the batch
experiments (Table 31) are listed adjacent their corresponding symbol in the key
Figure 34 Eh and pH batch experiment analyses (Figure 33) plotted on a stability field
diagram for Cr (log a Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom
(1998)
Figure 35 Batch experiments evaluating variable chromite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr and Cr(VI) measured in solution are noted and illustrated with a dotted line and
included in Table 33
Figure 36 Batch experiments evaluating variable birnessite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr(VI) measured in solution are denoted and included in Table 33
Figure 37 Influence of pH on Cr(VI) generation by reaction of constant chromite (Ch
08 m2 L-1) and birnessite (Bt 108 m2 L-1) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key Rates of Cr(VI) measured in solution are included in Table
33
84
Figure 38 Rates of Cr(VI) in solution (Table 33) as a function of a) chromite
suspension density (Figure 35) b) birnessite suspension density (Figure 36) and c) pH
(Figure 37) The line in each diagram represents the best fit for these data points and the
equations for these lines are listed in Table 33 The dotted line deviating from the
dashed line in b) indicates the probable path at low birnessite concentrations
Figure 39 SEM images of the solid phases after the completion of the batch
experiments A (Ch 02 Bt 54) and B (Ch 02 Bt 108) are shown Chromite is the larger
grain with relatively flat surfaces whereas birnessite is the fine-grained rosettes located
adjacent the larger chromite grains
Figure 310 Chromium and Cr(VI) concentrations (microM) in solution as a function of
time (h) from serpentine soils obtained at Jasper Ridge (JR1 and JR2) and New Caledonia
(NC) Soils with the addition of 01 g birnessite are noted in the figure Soil and
birnessite masses used for each experiment are listed in Table 32
Figure 311 Iron analyses as a function of time (Figure 35 and 36) with the solubility
of amorphous Fe(OH)3 and goethite and the dominant Fe3+-OH complex at pH 5 Fe
solubilities are from Langmuir (1997) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key
Figure 312 a) Experimental values of Fe and Cr (Figures 35 and 36) are plotted A
dotted line denotes the ratio of Fe and Cr necessary for chromite congruent dissolution b)
Corrected FeCr ratios as a function of time (h) for data from Figures 35 and 36 taking
into account the concentration of Fe and Cr in secondary phases calculated from Table
34 c) Corrected Fe and Cr as compared to values predicted with congruent dissolution
of chromite
85
Figure 313 Three potential pathways of Cr release and oxidation for chromite-
birnessite interactions are listed
86
Figure 31
87
0
2
4
6
8
10
12
Hydrogen
Peroxid
e
K-Perman
ganate
Hydroflu
oric Acid
Hydroxy
lamine
Barium Chlorid
e
Con
cent
ratio
ns (m
g kg
-1)
Cr
Fe
Mn
Figure 32
88
Time (h)0 100 200 300 400 500
pH
2
3
4
5
6
7
8
9
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
Time (h)0 100 200 300 400 500
Eh (m
V)
100
200
300
400
500
600
Ch Bt ------(m2 L-1)------
Figure 33
89
pH2 3 4 5 6 7 8 9
Eh (m
V)
-05
00
05
10
15
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
HCrO4-
CrO42-
Cr3+
CrOH2+
Cr(OH)3Cr(
OH
) 2+
Upper limit of water stability
Lower limit of water stability
Ch Bt ------(m2 L-1)-----
Figure 34
90
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 108 008 108 002 108 001 108 02 --- --- 54
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
0
1
2
3
4
pH 5 pH 5
pH 5 pH 5 pH 5
Ch Bt ---(m2 L-1)---
Figure 35
91
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 54 02 108
00041 microM h
-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
00
02
04
06
08
10
00041 microM h
-1
pH 5
pH 5 pH 5 pH 5
pH 5 Ch Bt----(m2 L-1)----
Figure 36
92
Time (h)
0 50 100 150 200 250
Cr c
once
ntra
tion
( microM
)
00
02
04
06
08
10
pH 3pH 5pH 67pH 8
Time (h)0 50 100 150 200 250
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250M
n co
ncen
trat
ion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
00004 microM h-1
00021 microM h-1
0006
2 microM h
-1
00021 microM h-100
062 micro
M h-1
00004 microM h-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1
Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Figure 37
93
log Birnessite m2 L-1
05 10 15 20 25 30
log
d[C
r(VI
)]dt
microM
h-1
-40
-35
-30
-25
-20
pH 5
log Chromite m2 L-1-22 -20 -18 -16 -14 -12 -10 -08 -06
log
d[C
r(VI
)]dt
microM
h-1
-34
-32
-30
-28
-26
-24
-22
pH 5
pH1 2 3 4 5 6 7 8 9
d[C
r(VI
)]dt
microM
h-1
0000
0002
0004
0006
0008
0010
a)
c)
b)
Ch 008 m2 L-1 Bt 108 m2 L-1
Figure 38
94
Figure 39
95
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR1JR1 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR2JR2 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
01
02
03
04
05
NCNC w Bt
00044 microM h-1
00013 microM h-1
00044 microM h-1
00013 microM h-1
00013 microM h-1
00009microM h-1
pH 5
pH 5
pH 5
Figure 310
96
Time (h)0 100 200 300 400 500
log ΣF
e (aq)
microM
-10
-8
-6
-4
-2
0
2
4
02 54 02 108 008 108 002 108 001 108
Fe(OH)3(am)
Goethite
pH 5Ch Bt----m2 L-1----
Fe(OH)2(aq)+
Figure 311
97
Time (h)
0 100 200 300 400
FeC
r Rat
io
0
1
2
3
4
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
02 54 02 108 008 108 002 108 001 108
Chrom
ite C
ongr
uent
Dissolu
tion
Chrom
ite C
ongr
uent
Dissolu
tion
a)
b)
c)
Ch Bt----(m2 L-1)----
ChromiteCongruent Dissolution
Figure 312
98
Pathway 1 Chromite Dissolution
1a) Cr dissolution from chromite FeCr2O3(s) + 4H+ rarr 2Cr(OH)2+
aq + Fe2+(Oxyhydroxide) + H2O
1b) Adsorption of CrOH2+ on the surface of birnessite Cr(OH)2+
aq + MnO2(s) rarr Cr(III)-MnO2(s) + OH- 1c) Oxidation of CrOH2+ on the surface of birnessite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
1d) Desorption of Cr(VI) from birnessite Cr(VI)-Mn(II) + 4H2O rarr Mn2+
(aq) + HCrO4-(aq) + 7H+
Pathway 2 Birnessite Dissolution 2a) Mn4+ dissolution from Birnessite MnO2(s) + 4H+ rarr Mn4+
(aq) + 2H2O 2b) Mn4+ adsorption on the surface of chromite Mn4+
(aq) + FeCr2O4(s) rarr Mn(IV)-FeCr2O4 2c) Oxidation of Cr(III) on the surface of chromite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
2d) Desorption of CrO24- from chromite
Cr(VI)-Mn(II) + 4H2O rarr Mn2+(aq) + HCrO4
-(aq) + 7H+
Pathway 3 Solid Solution Reaction 3) Solid solution interaction between chromite and birnessite
Figure 313
99
CHAPTER 4
CHROMIUM GEOCHEMISTRY OF SERPENTINE SOILS
ABSTRACT
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
chemistry of Cr in serpentine soils including its protolith is reviewed focusing on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils are slightly acidic (average pH of ~6) contain a variety Fe(III) oxides
(magnetite and hematite) phyllosilicates (serpentine and chlorite) and clays (smectites
and vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1)
and Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Chromium(III) is the only valence state observed
in the serpentine soil solids however Cr(VI) has been identified in New Caledonia and
California serpentine soil solutions at concentrations below 30 microM The enrichment and
range of Cr concentrations in serpentine soils are directly related to the presence of Cr-
spinels specifically chromite and Cr-magnetite These spinels are resistant to weathering
and are preserved in the soil environment however oxidation of Cr(III) from Cr-spinels
by high valent Mn oxides or other strong oxidants is a potential source of Cr(VI)
identified in serpentine soil solutions Due to the chemically and physically resistant
nature of the Cr-spinels Cr-bearing silicates including clay minerals Cr-chlorite Cr-
garnet Cr-mica and Cr-epidote are more viable sources of Cr identified in vegetation
soil extractions soil solutions and related waters
100
INTRODUCTION
Serpentine soils a generic term used to describe any soil derived from an
ultramafic rock (peridotites and pyroxenites) or serpentinite (metamorphosed ultramafic
rock) are often characterized by 1) low concentrations of plant nutrients such as N P
and K 2) high concentrations of biologically toxic elements including Ni Co and Cd 3)
low ammonium acetate extractable CaMg ratios and 4) unique flora and physical
properties (Brooks 1987) Measured Cr concentrations in serpentine soils range from
~200 mg kg-1 to 6 Wt values enriched compared to non-serpentine soils lt200 mg kg-
1 (Kabata-Pendias and Pendias 1984 Schwertmann and Latham 1986 Brooks 1987)
Chromite and Cr-magnetite are common primary sources of Cr in these soils (Rabenhorst
et al 1982 Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989
Gasser and Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003)
however the plethora of factors relating the chemistry of Cr in serpentine soils and the
weathering of chromite has not been fully addressed using an integrative approach
Overall serpentinites and serpentine soils cover ~1 of the earthrsquos exposed
surface and these rocks and soils are commonly associated with ophiolite complexes
(Coleman 1977) Serpentinites and serpentine soils related to ophiolite complexes are
the primary interest of this paper A map illustrating the worldwide distribution of major
ophiolitic serpentinite bodies and their related soils is shown in Figure 41 A majority of
the sites are near convergent boundaries (Figure 41) Although these rocks and soils
cover a minimal area of the earthrsquos surface they are present in populated areas within the
Circum-Pacific Margin and the Mediterranean Due to the wide-distribution and
occurrence of serpentinites and serpentine soils weathering of these rocks and soils will
differ from location to location due to varying climatic conditions as well as the nature of
the parent material and other factors including topography biota and time (Proctor and
Nagy 1992) As a result the chemistry of these soils as well as the nature of Cr in the
serpentinites and serpentine soils will vary from site to site potentially allowing Cr to be a
source of non-anthropogenic contamination
The objective of this paper is to review the chemical characteristics of serpentine
soils globally evaluate mineralogical and compositional variations between several soils
derived from ultramafic sources and assess the geochemistry of Cr in diverse soil-
101
forming environments Additionally chemical and spectroscopic analyses were
performed on soils derived from ultramafic rocks collected from 1) New Caledonia
South Pacific 2) Eight Dollar Mountain Selma OR 3) Nickel Mountain Riddle OR 4)
Harvard Mine Jamestown CA 5) Jasper Ridge Biological Preserve Stanford CA and 6)
Pillikin Mine El Dorado CA shown geographically in Figure 41 A basalt-derived soil
from Hawaii is included in this study in order to compare a soil formed from a mafic
protolith to those from ultramafic protoliths The various field sites are shown in Figure
42 These soils allow a comparative examination of how soil-forming factors such as
climate topography biota parent material and time affect the chemistry of Cr
REVIEW
Chromium Geochemistry
Chromium is a group VIB transition metal with two common oxidation states
Cr(III) and Cr(VI) differentiated by noticeably different chemical characteristics and
toxicities It should be noted that Cr(II) has been observed in lunar basalt olivines and in
high temperature terrestrial olivines (Sutton et al 1993 Li et al 1995) however this
oxidation state is not observed in soil environments A majority of Cr commonly persists
in ultramafic rocks serpentine soils and related waters as Cr(III) a non-hazardous
species and a micronutrient in human nutrition (Daugherty 1992 Cohen et al 1993
Fendorf 1995 James et al 1997) Chromium(III) has the highest crystal field
stabilization energy for all transition metals strongly favors octahedral coordination and
has 3 d-electrons with a high spin state Chromium(III) is strongly hydrolyzed in
aqueous solutions and the predominant species based on pH redox potential and O2(aq)
are shown in Figure 43 Chromium(III) species are more prevalent under reducing
conditions and its mobility is increased only under strongly acidic conditions At pH
values greater than 9 the hydrolyzed Cr(III) species is negatively charged demonstrating
the amphoteric properties of Cr(III) (Figure 43) Due to the factors described above
Cr(III) forms low solubility compounds and adsorbs strongly on mineral surfaces and
organic matter at pHgt4 (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) For
example the adsorption of Cr(III) on HFO (hydrous ferric oxide) as a function of pH is
shown in Figure 43 illustrating an adsorption edge at a pH of ~4 (Leckie et al 1984
102
Stumm 1992) Chromium(III) oxidation to Cr(VI) has been observed in reactions
involving hydrogen peroxide (H2O2) and high valent Mn oxides (Eary and Rai 1987
Johnson and Xyla 1991 Fendorf et al 1992 Fendorf et al 1993 Rock et al 2001)
Chromium(VI) is a strong oxidizing agent a highly corrosive substance a toxin to
living cells and a Class A human carcinogen by inhalation (Daugherty 1992 Cohen et
al 1993 Fendorf 1995) The hydrolyzed Cr(VI) species are also shown in Figure 43
with respect to pH redox potential and O2(aq) Chromium(VI) is more favorable under
oxidizing conditions and its mobility is enhanced under basic conditions In solution
Cr(VI) is a weakly adsorbing (oxy)anion (HxCrO42-x) on common weathering products
meaning that it forms highly labile complexes that can be readily displaced by ions
including phosphate and sulfate (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Additionally the adsorption of CrO42- on HFO shown in Figure 43
demonstrates the anionic character in which the mol bound of CrO42- increases with
decreasing pH and an adsorption edge is at pH ~72 (Leckie et al 1980 Stumm 1992)
The reduction of Cr(VI) to Cr(III) is known to occur by reactions with Fe(II)aq Fe(II)-
containing minerals such as magnetite H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996)
Mineralogy of Serpentine Soils
Serpentine soils have been identified on every continent excluding Antarctica
(Figure 41) The occurrence and distribution of ultramafic rocks specifically related to
ophiolites are closely associated with convergent plate boundaries (Figure 41 Coleman
1977 Coleman and Jove 1992) Tectonic blocks of ultramafic rock are hypothesized to
be detached from the down-going slab of oceanic crust and incorporated into the
subduction meacutelange Subsequent serpentinization of peridotite forms low density
serpentine group minerals (245-250 g cm-3) resulting in the upward diapiric migration of
the serpentinite masses along zones of structural weakness (fault zones) and through the
denser enclosing subduction meacutelange (~265 g cm-3) For example North America has
numerous exposed ophiolites containing serpentinite located in western California and
Oregon where subduction zones were once active in the Early Cretaceous (Harden
1998)
103
Serpentinization of ultramafic rocks is a significant factor controlling mineral
assemblages and textures in serpentinite Hydration and oxidation of olivine and
pyroxene during serpentinization forms serpentine (lizardite chrysotile and antigorite)
talc chlorite tremolite brucite and magnetite (Coleman 1977 Brooks 1987 Coleman
and Jove 1992 Malpas 1992) Additionally magnetite formation creates favorable
conditions for native metal compounds (such as Fe Co and Ni alloys) to form (Coleman
and Jove 1992) As a serpentinite mass is emplaced the extensive physical and chemical
alteration of the serpentinite results in numerous crosscutting alteration veins and highly
sheared textures Ultimately the mineral assemblages and textures in serpentinites are
complex prior to pedogenesis due to the multiple phases of metamorphosis tectonic
deformation and weathering of the rock
Serpentine soils consist of a mixture of residual minerals from the soil protolith
eolian and illuvial minerals and minerals formed within the soil Additionally highly
weathered serpentine soils enriched in Fe-oxides are also termed laterites soils common
in tropical and subtropical regions Soil protolith minerals may include primary minerals
of ultramafic rocks including olivines pyroxenes and oxides (chromite and magnetite)
as well as a large variety of metamorphic minerals the most common being serpentine
(lizardite chrysotile and antigorite) chlorite talc and actinolite Silica phases occur as
eolian grains concretions andor net type structures or silica webs (Hotz 1964
Coleman 1977 Schwertmann and Latham 1986) Eolian and illuvial aluminosilicate
minerals are important sources of Al in serpentine soils an element of low abundance in
ultramafic rocks
Layer silicates observed in serpentine soils are varied in type and abundance
They include smectites (often as montmorillonite) vermiculite chlorite brucite
serpentine and a variety of intrastratified clays (Veniale and Van der Marel 1963 Hotz
1964 Schellmann 1964 Wildman et al 1968 Rabenhorst et al 1982 Schreier et al
1987 Graham et al 1990 Bulmer et al 1992 Gasser and Dahlgren 1994 Lee et al
2003) Compositional variations crystal structures and parageneses of layer silicates
have been characterized for a number of serpentine soils (Wildman et al 1968
Rabenhorst et al 1982 Schreier et al 1987 Graham et al 1990 Bulmer et al 1992
Lee et al 2003) In addition to the layer silicates goethite hematite maghemite and
104
amorphous Fe (oxy)hydroxides are soil-forming phases in serpentine soils (Coleman and
Jove 1992) Gibbsite is not a common phase in most serpentine soils however limited
isomorphic substitution of Al(III) in goethite and hematite was noted in New Caledonia
(Schwertmann and Latham 1986) Manganese oxides such as birnessite (δ-MnO2) are
also present in serpentine soils and often coat mineral surfaces
Sources of Chromium for Serpentine Soils
Chromium primarily occurs in oxide and silicate phases in ultramafic rocks and
serpentinites Chromium-spinel often referred to as chromite is a common primary
(igneous) spinel in mafic and ultramafic rocks and often resists low-grade metamorphic
processes related to serpentinization (Hoffman and Walker 1978 Malpas 1992)
Isomorphic substitution of Al3+ Fe3+ and Ti4+ into the octahedral sites and Mg2+ Ni2+
Zn2+ and Mn2+ into the tetrahedral sites is common however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987b Pan and Fleet 1991 Sack and Ghiorso 1991 Burkhard 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996 Mathiesen 1999) Serpentinite chromite compositions reported in the literature are
plotted in Figure 44 to demonstrate the wide-range of Cr concentrations possible for Cr-
spinels Isomorphic substitution of octahedral Cr and octahedral Al for these chromite
compositions are also shown in Figure 44 Serpentinite pyroxenes containing Cr include
augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) (Grapes 1981 Von Knorring et al
1986 Treloar 1987a Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996)
Magnetite and serpentine minerals may incorporate varying amounts of Cr into
their structures during serpentinization Chromium(III) from primary Cr-containing
phases may isomorphically substitute into the Fe(III) site of magnetite due to size
(octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge similarities Chromium
concentrations in magnetite have been observed as high as 13 Wt (Oze et al 2003)
Chromate minerals such as crocoite (PbCrO4) are extremely rare in serpentinites and the
occurrence of these minerals are mainly limited to specific localities such as in Australia
Tasmania and the Urals (Williams 1974 Crane et al 2001) Serpentine minerals have
105
been reported to contain as much as 1500 mg kg-1 Cr based on spectrographic analyses
by Page and Coleman (1967) however serpentine minerals typically contain only trace
amounts of Cr
Hydrothermal and CO2 metasomatism of ultramafic rocks during the evolution of
convergent plate margins commonly produce Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite lt19 Cr Wt ) uvarovite (Cr-garnet lt20 Cr Wt )
tawmawite (Cr-epidote lt13 Cr Wt ) and kaumlmmererite (Cr-chlorite lt10 Cr Wt ) as
shown in Figure 44 (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974 Neiva
1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981 Max et al
1983 Nutman et al 1983 Raase et al 1983 Von Knorring et al 1986 Kerrich et al
1987 Treloar 1987b Treloar 1987a Morand 1990 Pan and Fleet 1991 Schandl and
Wicks 1993 Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995
Mathiesen 1999 Proenza et al 1999) Figure 44 illustrates the relationship of Cr and
Al substitution for Cr-silicate compositions based on Cr and Al Wt Additionally Cr-
silicate compositions are plotted based on octahedral Cr and octahedral Al in Figure 44
Compositions for Cr-muscovite Cr-epidote and Cr-garnet plot near the tie lines
connecting Cr and Al end-member minerals Chromium isomorphically substitutes for
octahedral Al in these silicates due to similarities in charge octahedral preference and
octahedral radii (Al3+ = 061 Aring) The numerous isomorphic substitutions possible for Cr
and Al in the chlorite structure results in analyses that do not plot near the tie line
connecting clinochlore and kaumlmmererite Overall any of these minerals are potential
sources for Cr in soils produced by the weathering of serpentinites and their hydrothermal
and CO2 metasomatic derivatives
Chromium in Serpentine Soils
Chromium concentrations vary widely in serpentine soils worldwide (Table 41)
New Caledonia soils appear to be the most Cr-enriched serpentine soil with a Cr
concentration of ~60000 mg kg-1 (Schwertmann and Latham 1986) however Jaffre
(1980) report values ranging from 6300 to 80000 mg kg-1 for New Caledonia soils
Some serpentine soils have measured Cr concentrations as low as 29 and 66 mg kg-1
atypical of most values reported in Table 41 (Proctor and Woodell 1971 Brooks et al
106
1992 Jeffrey 1992) Chromium concentrations as a function of depth for serpentine soil
profiles in Maryland (Rabenhorst et al 1982) Tehama County CA (Gough et al 1989)
the Klamath Mountains CA (Hotz 1964) Newfoundland Canada (Roberts 1992)
Harvard Mine CA (this study) and Jasper Ridge CA (Oze et al 2003) are reported in
Figure 45 The pedogenic process(es) responsible for the varying Cr concentrations as a
function of depth for each soil have not been characterized however Oze et al (2003)
determined that a majority of Cr in Jasper Ridge serpentine soils is the result of
weathering-resistant residual Cr-bearing minerals from the protolith mainly chromite
and Cr-magnetite Additionally Oze et al (2003) demonstrated that soil samples from
Jasper Ridge contain higher concentrations of Cr with respect to their corresponding
bedrock due to the Cr-spinels accumulating over time The observation that Cr
concentrations decrease with depth towards the protolith is characteristic for most of the
profiles in Figure 45
In addition to examining total Cr concentrations chemical extractions performed
on serpentine soils provide additional information with regards to Cr availability and the
minerals and complexes in which Cr may reside in the soils Table 42 summarizes a
number of extraction experiments performed on serpentine soils For a majority of the
soils listed in Table 42 total Cr concentrations for the serpentine soils were not
measured however these extractions do provide a quantitative assessment of where Cr
may be bound in soil (such as silicates organic matter and secondary Fe-bearing phases)
Extractions performed by Schwertmann and Latham (1986) were able to remove ~20000
Cr mg kg-1 from the serpentine soil using a dithionite-citrate bicarbonate (DCB) solution
which preferentially dissolves crystalline Fe-oxides Solutions such as concentrated
nitric and hydrochloric acid were only able to remove a maximum of ~1400 Cr mg kg-1
from Unst serpentine soils (Shewry and Peterson 1976) which preferentially dissolve
Cr-bearing silicates and to a lesser extent the Fe-oxides Where bulk soil Cr values were
measured extraction experiments are only able to remove a small portion (lt1) of the
total Cr present in the serpentine soils (Table 42) (Gasser and Dahlgren 1994 Oze et al
2003) Based on the extractions presented in Table 42 it appears that Cr(III) resides in
minerals and phases resistant to common extraction techniques
107
In serpentine soils Fe(III) oxides and Fe(III) hydroxides provide a host of
structural and surface sites available to sequester Cr(III) released to soil solutions by
dissolution of Cr-bearing minerals during soil-forming processes Substitution of Cr(III)
for Fe(III) in Fe-oxides Fe-(oxy)hydroxide minerals and amorphous compounds is
expected based on crystallographic considerations synthesis experiments and analysis of
extracts from selective dissolution of Fe(III) compounds in serpentine soils
(Schwertmann and Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990
Gerth 1990 Gasser and Dahlgren 1994) Chromium(III) also readily sorbs on surfaces
of Fe(III) oxides and (oxy)hydroxides via inner-sphere surface complexes (Figure 43)
and forms low-solubility precipitates (Charlet and Manceau 1992) The low
concentration of Cr(III) in soil solutions has been attributed to the low solubility of
Fe(III)-Cr(III) amorphous hydroxides (Sass and Rai 1987)
Layer silicates provide structural and surface sites capable of uptaking Cr(III)
Octahedral Cr(III) in clays such as smectites and vermiculites is expected and Cr
concentrations in the clay-sized fraction can achieve concentrations of ~1000 mg kg-1 in
serpentine soils (Oze et al 2003) Sorption of Cr(III) and Cr(VI) on edge and interlayer
sites of phyllosilicates have been evaluated by Ilton and Veblen (1994) Ilton et al (2000)
and Brigatti et al (2000) indicating that layer silicates (montmorillonite chlorite biotite
phlogopite and muscovite) provide sites for both Cr(III) sorption and Cr(VI) reduction
Kaupenjohann and Wilcke (1995) concluded that Cr in serpentine soils from Bavaria is
residually bound mainly in silicate structures including clay minerals based on extraction
experiments
Soil Solutions and Related Water
The addition of water gradually changes the spatial and temporal characterization
of pore fluids in soils including the ionic strength pH and redox potential Leaching of
a soil should progressively enrich the soil with less chemically mobile elements such as
Cr and Fe Elements identified in soil solutions include Mg Ca Fe Ni and Cr and the
ionic strengths of serpentine soil solutions are typically low (lt001) (Anderson et al
1973 Johnston and Proctor 1981 Gasser and Dahlgren 1994) Soil solutions obtained
from serpentine soils using tension-free lysimetry demonstrate Cr concentrations ranging
108
from 01 to 32 micromol L-1 (Gasser and Dahlgren 1994) This study also revealed a
majority of Cr Fe Mg and Ni in extracted soil waters were in the form of colloids
suggesting that nanoparticles are an important factor of Cr translocation in soils
The pH of a soil is a dominant factor controlling the nature of many chemical and
microbial reactions including the geochemistry of Cr (Brady and Weil 1999) Ranges in
pH for serpentine soils worldwide are plotted in Figure 46 The highest soil pH was
measured in a South Africa serpentine soil with a value of 9 (Cole 1992b) whereas the
lowest pH was reported in Ireland with a value of 38 (Jeffrey 1992) This wide range of
pH is a function of the adsorbed H+ and Al3+ the base-forming cations the organic
matter and the mineralogy present within each soil The pH range of serpentine soils
shown in Figure 46 is plotted on the stability diagram in Figure 43 demonstrating that
both Cr(III) and Cr(VI) aqueous species are favorable The Eh range for the serpentine
soils in Figure 3 is based on the soils being in contact with the atmosphere as well as
being a transitional environment Although the activity of H+ varies by four orders of
magnitude the pHs of most serpentine soils are slightly acidic (pH ~6) which is near that
of average rainwater (pH~57 Figures 43 and 46) Additionally serpentine soil pH
values have been observed to increase with depth towards the protolith for a number of
soils (Rabenhorst et al 1982 Gough et al 1989 Gasser and Dahlgren 1994) The
increase in pH with respect to depth may be the result of rainwater being progressively
influenced by the mineral chemistry and hydrolysis reactions as it descends through the
soil towards the protolith
Groundwater from serpentinized ultramafic rocks have pH values ranging from 8
to 12 (Barnes et al 1972) and these waters have been divided into two categories as
shown in Figure 46 Mg bicarbonate (Mg2+-HCO3-) type and Ca hydroxide (Ca2+-OH-)
type (Barnes and ONeil 1969 Barnes et al 1972) The Mg bicarbonate waters (pH~8)
are characteristic of most surface waters and shallow groundwaters from serpentinites
and ultramafic rocks The Ca hydroxide waters (pHasymp10-11) are associated with
incompletely serpentinized peridotites and these fluids are comparatively low in Mg2+
Fe2+ Mn2+ Crtotal and SiO2total The Ca2+ in solution originates from the serpentinization
of Ca-rich orthopyroxene (Barnes et al 1972) The pH of serpentinite groundwaters are
109
plotted on the Cr stability in Figure 43 demonstrating that both Cr(III) and Cr(VI) are
potential valence states in these waters
Chromium predominates as Cr(III) in ultramafic rocks and serpentine soils
however Cr(VI) has been identified in serpentine soil solutions and serpentinite
groundwater in California and New Caledonia at concentrations of lt20 microg L-1 (Gough et
al 1989 Becquer et al 2003) Calcium hydroxide waters (pH~11) collected from the
same spring evaluated by Barnes et al (1972) in the Cedars ultramafic body (Cazadero
CA) contains 12-22 Cr(VI) microg L-1 and no Cr(III) (this study) Additionally water from
chromite-containing ultramafic rocks in the Siniktanneyak Mountain ultramafic complex
of Alaska with pH values ranging from 67 to 93 were measured with low total Cr
concentrations lt2 Cr microg L-1 (Taylor et al 1998) A mechanism for Cr(III) oxidation in
serpentine soils and related water suggested by (Gough et al 1989 Becquer et al 2003)
may be the result of high valence Mn oxides common in serpentine soils High valence
Mn-oxides such as birnessite (δ-MnO2) are naturally occurring oxidants capable of
oxidizing Cr(III) to Cr(VI) at pHlt9 (Eary and Rai 1987 Fendorf et al 1992 Fendorf et
al 1993) It is also important to note that Cr(VI) reduction is hypothesized to occur in
soils due to electron transfer involving organic matter and Fe(II)-bearing minerals such as
magnetite (Peterson 1996 Fendorf et al 2000) Whether toxic concentrations of Cr(VI)
can be present in serpentine soils and related water over geologic time is important to
consider especially when high valence Mn oxides Fe(II)-bearing minerals and organic
matter are capable of significantly influencing the oxidation state of Cr
Serpentine Soil Vegetation
Vegetation associated with serpentine soils has received considerable attention
due to the apparent infertility of these soils The inhabitability of serpentine soils for
most plants has been attributed to the imbalance of available Mg relative to Ca or low
CaMg ratios (Walker 1954) the deficiency of plant nutrients such as K N and P
(Turitzin 1991) elevated heavy metal concentrations such as Ni and Cr (Robinson et al
1935) and poor drainage (Brooks 1987) Linking plant productivity to toxic levels of
Cr in the soil has not been fully resolved (Brooks 1987 Kruckeberg 1992) however a
majority of Cr in the soils is not available for plant uptake due to being in high stable and
110
weathering resistant phases discussed in the previous sections (Table 42) In general Cr
concentrations in plants typically do not exceed 100 mg kg-1 (Brooks 1987)
Table 43 provides a global list of serpentine soil plant species and their measured
Cr concentrations with an emphasis on Cr hyper-accumulators ie plants able
incorporate higher than average concentrations of Cr A majority of these measurements
are based on the dry weight of Cr Aerobryopsis longissima from New Caledonia has the
highest reported Cr concentration with values ranging from 200 to 700 mg kg-1 (Lee et
al 1977) Other noteworthy plants include the Lophostachys villosa from Brazil (Brooks
et al 1992) Arenaria humifusa from Newfoundland (Roberts 1992) the Calluna
vulgaris from Unst (Shewry and Peterson 1976) with Cr concentrations of 203 530 and
417 mg kgndash1 respectively Even mosses such as the Funaria flavicans (Maryland)
incorporate Cr concentrations as high as 90 mg kg-1 (Shaw 1991) By examining
specific parts of the plant Gough et al (1989) determined that Cr concentrations are
highest in the basal burl (600 mg kg-1) of the manzanita bush compared to the roots (30
mg kg-1) stems (20 mg kg-1) and leaves (20 mg kg-1) The study by Gough et al (1989)
demonstrates that Cr is not homogeneously distributed throughout the plant and that
evaluating a leaf versus a stem could result in significantly different Cr concentrations
Soil Organic Matter
Soil organic matter (SOM) including the biomass (living organisms such as
bacteria) detritus (nonliving plant residues) and humus (amorphous and colloidal
mixtures of organic substances) affects the geochemistry and buffering capacity of
serpentine soils Furthermore biologic reduction of Cr(VI) to Cr(III) via cometabolic
and metabolic pathways coupled with carbon oxidation has been demonstrated for a
number of bacterial strains including aerobes (Gopalan and Veeramani 1994 Park et al
2000) and anaerobes (Turick et al 1996 Tebo and Obraztsova 1998) Additionally
aqueous Fe(II) produced by microbial Fe(III) reduction is a known biogenic pathway for
the reduction of Cr(VI) (Fendorf et al 2000) Fe(III)-Cr(III) (oxy)hydroxides are the
resulting secondary phases (Rai et al 1989 Patterson and Fendorf 1997 Hansel et al
2003) Bacteria oxidizing Cr(III) to Cr(VI) has not been observed
111
Most serpentine soils in temperate regions appear to be either Mollisols or
Alfisols which often contain high concentrations of SOM Alexander et al (1989) and
Bulmer et al (1992) reported that serpentine soils can contain organic carbon
concentrations in excess of 50 mg g-1 Humus has a high cation exchange capacity and
provides much of the pH buffering capacity in these soils (Brady and Weil 1999)
Organic acids such as fulvic and humic acids are known to complex cations such as
Cr3+ Fe3+ and Mn2+ forming stable organometallic complexes (Kaupenjohann and
Wilcke 1995 Brady and Weil 1999) Additionally metal leaching experiments
performed by Schreier et al (1987) established that organic acids produced by soil
organic matter such as oxalic acid and citric acid were extremely effective in releasing
Cr Mg Fe Ni and Co from serpentinite rocks and sediments
CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
Serpentine soils were obtained from California Oregon and New Caledonia and
analyzed to further evaluate the factors influencing the Cr geochemistry in serpentine
soils Analyses for these soils include pH electrical conductivity total organic matter
total N ammonium acetate extractable Ca K Mg and Na soil texture soil color
chemical composition of the soil and the chemical composition of Cr-bearing minerals
The general soil characteristics are described in the following section and the analytical
methods are explained in Appendix 41 Although a majority of these soils are located in
the western United States factors influencing the pedogenesis of these soils vary
considerably allowing a comparative analysis of how the fate of Cr might be influenced
by climate topography biota parent material and time
Sample Locations and General Soil Characteristics
Six serpentine soils were collected from locations shown in Figures 41 and 42 to
evaluate the chemistry of Cr in serpentine soils These soils include 1) a laterite (Oxisol)
collected in New Caledonia South Pacific 2) a laterite (Oxisol) exposed at Eight Dollar
Mountain Selma OR 3) a laterite (Oxisol) near Nickel Mountain Riddle OR 4) a
serpentine soil (Inceptisol) with a CO2-metasomatized serpentinite protolith from
Harvard Mine Jamestown CA 5) a serpentine soil (Mollisol) obtained from a preserved
112
grassland at Jasper Ridge Biological Preserve Stanford CA and 6) a serpentine soil
(Inceptisol) collected in the proximity of Pillikin Mine El Dorado CA Lastly an Oxisol
formed from basalt obtained in Oahu Hawaii is also included for comparison
These seven sites include a range of soil environments from different protoliths
under varying conditions of rainfall temperature and topography Each soil was collect
near the highest and topographically level point for each serpentine soil site Serpentine
grassland soils at Jasper Ridge formed under Mediterranean-like climates (xeric moisture
regime) whereas higher temperatures and less rainfall (ustic moisture regime)
characterize soil formation at the Harvard and Pillikin mines in the Sierra Nevada
Foothills metamorphic belt The vegetation of these three sites is dominantly grasses with
some manzanita These modern soil environments contrast with the Miocene Ni and Cr-
rich laterite soils of Nickel Mountain and Eight Dollar Mountain that were probably
formed under an aquic moisture regime near sea level in a subtropical climate Grasses
manzanita and pine trees are the dominant forms of vegetation at Nickel and Eight
Dollar Mountain An ultramafically-derived laterite formed under tropical conditions
from New Caledonia was collected The vegetation and topography related to the New
Caledonia laterite was not recorded during sampling An Oxisol from Hawaii is also
included in this study to evaluate the differences between a highly weathered soil
originating from an ultramafic rock and one from a mafic rock (basalt) under tropical
conditions The Hawaiian soil was collected from a topographically level area however
the vegetation related to this site was not noted
BedrockProtolith Mineralogy
A broad characterization of the protoliths for the serpentine soils is briefly
summarized due to the bedrock not being collected from all the field sites New
Caledonia is composed of variably serpentinized peridotites predominantly harzburgites
and some dunites (Leblanc 1995) Nickel Mountain and Eight Dollar Mountain are
predominantly serpentinized harzburgite and dunite The bedrock for soils collected at
Harvard Mine is mainly composed of quartz ankerite and mariposite (also referred to as
QAM) caused by the CO2 metasomatism of serpentinite associated with gold
mineralization (Bohlke 1989) The average serpentinite at Jasper Ridge is composed of
113
serpentine (mainly lizardite and antigorite) chlorite talc magnetite and chromite (Oze et
al 2003) however brucite has been reported as a major constituent in related Franciscan
Complex serpentinites (Hostetle et al 1966) Pillikin Mine serpentinites dominantly
consist of talc and serpentine (lizardite antigorite and chrysotile) and these serpentinites
contain podiform chromite deposits (Cater et al 1948) Mafic bedrock for the Hawaiian
Oxisol is tholeittic basalt
Soil Chemistry and Physical Properties
As shown in Figure 47 the colors of the collected soils range from the red New
Caledonia laterite to the greenish gley of the Pillikin Mine Inceptisol The Munsel color
values are also shown in Figure 47 The dark grey color of the Jasper Ridge soil and the
greenish gley of the Pillikin Mine soil are atypical of soils derived from serpentinites due
to the lack of any apparent redness The range of these colors is often attributed to the
presence of Fe-oxides however organic matter and Mn oxides are also dominant factors
controlling color (Brady and Weil 1999)
Chemical properties for the six serpentine soils and the Hawaiian soil are listed in
Table 44 The slightly acidic pH values range from 53 to 69 similar to those shown in
Figure 43 and 46 The New Caledonia laterite is the most acidic soil with a pH of 53
compared to the Harvard Mine carbonate-buffered soil at 69 The pH of the Hawaiian
soil is also slightly acidic with a pH of 604 The electrical conductivities (EC) of the
soils range from 01 to 07 dS m-1 a range indicative of relatively few dissolved inorganic
solutes The highest organic matter content is in the very dark grey Jasper Ridge soil
with a value of 32 and the lowest is in the red New Caledonia laterite (05) The
Hawaiian soil has an organic matter content of 1 Nitrogen concentrations for all the
soils are lt012 with Jasper Ridge having highest N concentration which is consistent
with the soil organic matter content
Ammonium acetate extractable Ca K Mg and Na are also listed in Table 44
Magnesium concentrations are gt6923 mg kg-1 in all of the serpentine soils The lowest
concentration of Mg was measured in the basalt-derived Hawaiian Oxisol with a value of
168 mg kg-1 Calcium concentrations in the serpentine soils range from 167 to 998 mg
kg-1 The 998 mg kg-1 of Ca in the Harvard Mine soil is directly related to the abundance
114
of ankerite (gt20 of the bedrock) Potassium concentrations are low for most of the soils
(lt55 mg kg-1) however Jasper Ridge and Harvard Mine soils have the highest
extractable K with values of 237 and 301 mg kg-1 respectively The extractable K value
for the Harvard Mine soil is due to the K-metasomatism related to gold mineralization
involving the serpentinite soil protolith The extractable concentration of Na is lt44 mg
kg-1 for all the serpentine soils whereas the Hawaiian soil has the highest Na (53 mg kg-
1) due to plagioclase and basaltic glass
Textures for the soils range from clay loam clay loam to sandy clay loam (Table
44) The laterite soil in New Caledonia and Oxisol in Hawaii are extremely different in
which the Hawaiian soil is mostly clay (66 clay) compared to the loam at New
Caledonia (25 clay) Additionally both ancient laterites (Eight Dollar Mountain and
Nickel Mountain) have clay textures Harvard Mine Jasper Ridge and Pillikin Mine
soils are all loamy in character
Major and trace element compositions for the soils are listed in Table 45
Chromium concentrations in the serpentine soils range between 827 and 9528 mg kg-1
The Eight Dollar Mountain laterite has the highest Cr concentration and the New
Caledonia laterite contains the lowest Cr concentration The Cr concentration for the
New Caledonia soil is lower than the values reported in Table 41 The Hawaiian soil has
a Cr concentration well below the analyzed serpentine soils with 388 Cr mg kg-1 but still
greater than the 200 mg kg-1 characteristic of serpentine soils Nickel concentrations in
the serpentine soils range from 1895 to 14100 mg kg-1 with the highest concentration
being identified in the Nickel Mountain laterite Manganese concentrations range
between 257 (New Caledonia) to 3017 (Eight Dollar Mountain) mg kg-1 in the serpentine
soils Element concentrations in the Hawaiian Oxisol contrast from the serpentine soils
with higher Al Cu P Pb and Zn concentrations
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in the soil solids for Jasper Ridge New Caledonia Pillikin Mine Eight Dollar
Mountain and Harvard Mine Although the Cr concentrations in these soils exceed 200
mg kg-1 (Table 45) X-ray absorption near-edge spectra (XANES) shown in Figure 48
illustrate Cr(VI) levels are below the detection limits of this method (~10 Cr mg kg-1) due
to the lack of a pre-edge feature The box in Figure 48 illustrates where a Cr(VI) pre-
115
edge peak should be present Chromium(VI) concentrations in these soils are possibly
present in the parts per billion range
Soil Mineralogy
Backscattered electron images and microprobe analyses were used to identify the
Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49
and corresponding Cr-bearing mineral compositions are listed in Table 46 The New
Caledonia soil exhibits a wide range of particle sizes and a Fe-(oxy)hydroxide was
identified containing 165 Cr2O3 Wt Chromium-spinel has been identified in the New
Caledonia laterite by Becquer et al (2003) Eight Dollar Mountain soils exhibit a coarse
texture and both Cr-magnetite (2387 Cr2O3 Wt ) and chromite (3343 Cr2O3 Wt )
are present Nickel Mountain soils are also coarse and chromite (1315 Cr Wt ) was
identified Soils at Harvard Mine contain numerous chromite grains and the mineral
trevorite (a Ni-bearing spinel) was identified with 3991 Cr2O3 Wt and 2054 NiO Wt
Chromite and magnetite are abundant in soils at Jasper Ridge as well as a fine-
grained mixture of chromite magnetite and an unidentifiable silicate with up to 1878
Cr2O3 Wt (Figure 49) Veins of Cr-chlorite crosscut the spinel-silicate mixture in
Figure 49 and this Cr-chlorite is the only Cr-silicate identified using this method
Chromite (5355 Cr2O3 Wt ) is the dominant Cr-bearing mineral detected in soils at the
Pillikin Mine and these grains contain the highest Cr concentrations compared to the
other minerals in the seven soils The Hawaiian Oxisol possesses ulvoumlspinels (007
Cr2O3 Wt and 2465 TiO2 Wt ) which are evenly distributed throughout the soil
Overall microprobe observations reveal that the Cr-bearing spinels are present in every
soil Additionally Cr-bearing spinels at Jasper Ridge have been shown to undergo
incongruent dissolution progressively enriching the spinel towards a FeCr2O4 end-
member (Oze et al 2003)
DISCUSSION
The physical and chemical characteristics of serpentine soils are directly related to
the alteration and tectonic history of the serpentinite as well as the soil-forming factors
including the parent material climate biota topography and time Identifying Cr-
116
bearing minerals their chemical compositions and abundance in the soils allows a more
definitive explanation for Cr enrichment in these soils as compared to total Cr
concentrations listed in Tables 41 and 45 and sequential extraction experiments reported
in Table 42 Detrital chromite grains from the serpentinite protolith are resistant to
weathering and accumulate in the soil over time Although serpentine soils contain Cr
concentrations typically gt200 mg kg-1 the presence of Cr-spinels and their resistance to
weathering demonstrate that a majority of the Cr is not chemically available to interact
within the soil The role of organic matter and vegetation are important factors related to
the chemistry of serpentine soils however their role for influencing a majority of the Cr
geochemistry is limited due to the nature of the Cr-spinels
Chromium and Serpentine Soils
Total Cr concentrations exceed 200 mg kg-1 for most of the serpentine soils
reported in Table 41 and in the six serpentine soils of this study (Table 45) These Cr
concentrations can be accounted for by assessing the Cr-bearing minerals associated with
these soils and their relative abundance Small grains of Cr-spinels or chromite identified
in the soils (Figure 49 and Table 46) have the potential to appreciably impact Cr
concentrations Table 47 demonstrates the mineral masses and volumes of Cr-bearing
minerals required to cause predetermined Cr concentrations in soil samples A minimal
amount of chromite (lt02 of a soil sample mass or 005 of the volume of serpentine)
can result in Cr concentrations of 1000 mg kg-1 A few grains of chromite could easily
be responsible for the Cr concentrations in Tables 41 and 45 In fact residual grains of
Cr-enriched spinels are observed in every soil analyzed (Figure 49 and Table 46)
Ultimately grains of spinel in the soil do appreciably impact Cr concentrations however
they do not provide a readily available source of Cr based on extraction experiments
performed by Oze et al (2003) Serpentine minerals could also provide minor Cr
concentrations to the soils however soil extractions have consistently shown that a
majority of the Cr is in highly resistant phases capable of resisting dissolution even in
concentrated hydrochloric acid nitric acid and hydroxylamine with acetic acid (Table
42) Chromium end-member minerals including eskolaite Cr-muscovite (fuchsite) Cr-
garnet (uvarovite) Cr-epidote (tawmawite) and Cr-chlinochlore (kaumlmmererite) are also
117
included in Table 47 to demonstrate that relatively low abundances of these minerals are
all that is necessary to increase Cr concentrations in rocks and soils Chromium-
clinochlore is often observed developing adjacent Cr-spinels in serpentinites and is
perhaps a more accessible mineral source of Cr in the serpentinites compared to Cr-
spinels (Phillips et al 1980 Schreyer et al 1981 Kerrich et al 1987 Treloar 1987a
Pan and Fleet 1991 Schandl and Wicks 1993 Christofides et al 1994 Challis et al
1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Oze et al 2003)
As shown in Table 44 and Figures 43 and 46 the chemical properties (pH EC
N etc) of serpentine soils are not correlated to Cr concentrations measured in the soil
solids (Tables 41 Figure 45) due to Cr-spinels accounting for a majority of the Cr
(Figure 49 and 410 Tables 46 and 47) However extraction experiments and soil
solution analyses demonstrate that a limited fraction of the total Cr (lt1) is extractable
If only 1 of the Cr in a serpentine soil is extractable a serpentine soil containing 10000
Cr mg kg-1 could possibly contribute up to100 Cr mg kg-1 still a significant Cr
concentration The chemical properties of the soil such as pH will influence this limited
fraction of the Cr in the soils Acidic pHs may aid in mobilizing Cr(III) however this
will only occur at low pHs (lt4) which exceed the range of most serpentine soils (Figures
43 and 46) pH values greater than 8 like those in some serpentine soils and
serpentinite groundwater may help sustain Cr(VI) concentrations in solution (Figure
43) In addition to pH organic matter in the soils may be a significant factor related to
the mobility of Cr due to the production of organic acids promoting the dissolution of the
Cr-bearing phases but particulate organic matter is known to bind Cr(III) in stable stable
organometallic complexes (Kaupenjohann and Wilcke 1995 Brady and Weil 1999)
Evaluating the valence state of Cr is crucial for understanding the chemistry as
well as the potential toxicity related to serpentine soils The soil solids do not contain
detectable concentrations of Cr(VI) (Figure 48) however Cr(VI) concentrations are
present in soil solutions and groundwater at values lt30 microg L-1 (Gough et al 1989
Becquer et al 2003) A majority of serpentinite Cr-bearing minerals only contain
Cr(III) therefore there must be a reaction or process by which Cr(III) is oxidized in a
water saturated soil High valence Mn oxides such as birnessite are present within
serpentine soils and provide a feasible explanation for Cr(III) oxidation Additionally
118
Mn oxides or other strong oxidants in the soil might accelerate the weathering of Cr-
bearing minerals including chromite and generate the Cr(VI) concentrations observed in
soil solutions and groundwater After Cr(III) is oxidized Cr(VI) is a chemically mobile
species and would be transported away from the serpentine soils before it can be
observed in the solid phase Additionally Cr(VI) may be subsequently reduced by
reacting with organic matter or Fe(II)-bearing minerals such as magnetite which are
present in a majority of serpentine soils Perhaps the best way to evaluate the potential of
serpentine soils being a non-anthropogenic source of Cr contamination is to monitor
Cr(III) and Cr(VI) in the soil solutions over time as demonstrated by Gasser and
Dahlgren (1994)
It is curious that plants do uptake Cr in serpentine soils especially when Cr is
tightly bound in the soil matrix (Table 42) In order for plants to uptake Cr it must be in
a form to be absorbed by plant roots Soil pH in the rhizosphere can be altered by plant
roots by as much as 2 units within two days (Fischer et al 1989) This acidification is
known to increase the solubility of the minerals in the soils and could serve as a
mechanism by which plants can incorporate this relatively insoluable element Another
possibility is that Cr(III) in the soil is oxidized by either organic or inorganic means and
then incorporated into the plant in its more chemically mobile and bioavailable form
Cr(VI) How and why plants incorporate as much Cr as they do needs to be examined
more thoroughly
Chromium Spinel in Soils and Rocks
Compositions of chromite identified in serpentinites worldwide (Figure 44) and
Cr-bearing spinels identified in the six serpentine soils are compared in Figure 10 in order
to determine whether the soil spinels are geochemically similar to those in serpentinites
This comparison is ascertained by examining how Al Fe and Mg Wt values varied
with respect to Cr Wt Trends for magnetitehematite chromite and the
microcrystalline spinel-silicate mixture observed in the soils are noted in each diagram in
Figure 410 Additionally the highest Cr concentrations for soil chromite
magnetitehematite and the microcrystalline spinel-silicate mixture are ~40 Cr Wt
~25 Cr Wt and ~15 Cr Wt respectively Based on Al Wt soil
119
magnetitehematite deviates significantly from the chromite trend with Al Wt values
lt5 Magnetitehematites are enriched with Fe relative to serpentinite and soil chromite
and Mg concentrations in the soil chromite are slightly higher than serpentinite chromite
Soil magnetitehematite and chromite are similar by demonstrating a linear trend when
Al Fe and Mg are summed The microcrystalline spinel-silicate mixture (CSM) does
not follow this trend because of Si present in this mineral mixture Overall soil
chromites are similar to serpentinite chromite and soil magnetiteshematites can be
differentiated from the chromites when Cr Wt is plotted against Al Fe and Mg Wt
CONCLUDING STATEMENT
Serpentine soils are abundant among convergent margins and the chemistry of
these soils is ultimately a function of the ultramaficserpentinite rock mineralogy and
texture as well as all the common soil-forming factors Although ultramafic rocks and
serpentinites produce groundwater that is highly basic most serpentine soil pHs are
slightly acidic Although Proctor and Nagy (1992) argued that it is futile to categorize
serpentine soils based on chemistry and biology (otherwise referred to as the lsquoserpentine
factorrsquo) as proposed by Brooks (1987) Cr concentrations greater than 200 mg kg-1 in
serpentine soils are consistently present Chromium in the soils originates from minerals
in the bedrock where the valence is primarily Cr(III) the exception being Cr(VI) in the
rarely present mineral crocoite (PbCrO4) Increased Cr-concentrations in serpentine soils
are largely the result of chromite or other resistant Cr-bearing minerals causing a
localized enrichment in the soils Although Cr in serpentine soils is not readily released
from the soil matrix serpentine soil vegetation can incorporate Cr concentrations as high
as 700 mg kg-1 (Table 34) Finally the limited extractability of Cr and site dependent
factors such as the presence of organic matter clays Fe(II)-bearing minerals pH and the
poor drainage indicates that the possible formation and sustained presence of Cr(VI) is
not a portentous matter related to serpentine soils
120
ACKNOWLEDGEMENTS
We appreciate technical and field support from Robert Jones Anders Meibom
Rob Dunbar and Guangchao Li Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund GSA Graduate Student Research Grants and the
Office of Teaching and Learning at Stanford University is gratefully acknowledged
121
Table 41 Measured Cr concentrations in serpentine soils worldwide
Location Cr concentration (mg kg-1) Reference
Canyon Creek area OR 920 (White 1971) 66-13200 (Brooks et al 1992)
Ireland 29-5000 (Proctor and Woodell 1971 Jeffrey 1992) Italy 1300-3900 (Gambi 1992) Japan 5200 (Shewry and Peterson 1976) Maryland 1500-6000 (Rabenhorst et al 1982) Mt Tamalpais CA 1700 (White 1971) New Caledonia 6300-80000 (Jaffre 1980) New Caledonia 2-6 (Schwertmann and Latham 1986) New Idria CA 1640 (White 1971) New Zealand 850-3250 (Lee 1992) Newfoundland Canada 309-1447 (Roberts 1992) Nickel Mountain OR 1730 (White 1971) Santa Barbara County CA 2080 (Woodell et al 1975) Southern Africa 045-14 (Morrey et al 1992) Tehama County 023-12 (Gough et al 1989) Transvaal Lowveld South Africa 120-800 (Cole 1992) Unst Shetland 3000-4000 (Proctor 1969) Wenatchee Mnts WA 895 (White 1971) Zimbabwe 110-46000 (Soane and Saunder 1959 Proctor et al 1980)
Goias State Brazil
122
123
Table 42 Summary of Cr Fe and Ni extractions from serpentine soils
Range of Extractable Cr Fe and Ni Location Cr Fe Ni
Method Reference
Italy 2-5 microg g-1 305-640 2300-2700microg g-1 25 CH3COOH (Gambi 1992)
NE Portugal 02-09 microg g-1 120-540 microg g-1 40-290microg g-1 EDTA (Menezes de Sequeira and Pinto da Silva 1992) New Caledonia 22-23 384-581 02-216 DCB (dithionite-citrate bicarbonate) (Schwertmann and Latham 1986) New Caledonia 02-11 microg g-1 02-66microg g-1 Method Unknown (Jaffre 1980)
Norway 357microg g-1 316microg g-1 HClO4HNO3 (concentrated) (Zeien and Bruemmer 1989) Norway 254microg g-1 134microg g-1 Ascorbic and Oxalate Acid pH=325 (Zeien and Bruemmer 1989)
Norway 62 microg g-1 53microg g-1 NH4-EDTA (Zeien and Bruemmer 1989) Norway 22-60 microg g-1 8-25 mg kg-1 1-7 microg g-1 pH-stat reactor EDTA (Kaupenjohann and Wilcke 1995)
Swiss Alps 08 micromol L-1 28micromol L-1 228microg g-1 Lysimeter (Nonfiltered) (Gasser and Dahlgren 1994) Trinity Ophiolite CA 004-045 gkg 27-210 g kg-1 005-245microg g-1 Citrate-Dithionite Extracts (Alexander et al 1989)
Unst lt01 microg g-1 19-77microg g-1 Ammonium Acetate (Shewry and Peterson 1976)
Unst 500-1390microg g-1 1540-3330microg g-1 HNO3-Soluble (Shewry and Peterson 1976) JRBP CA 0 microg g-1 01microg g-1 103microg g-1 BaCl2 Method (Oze et al 2003)
JRBP CA 78 microg g-1 513 microg g-1 150 microg g-1 Hydroxlyamine+Acetic Acid (Oze et al 2003) JRBP CA 31 microg g-1 2039microg g-1 35microg g-1 10M Hydrofluoric Acid (Oze et al 2003)
123
124
Table 43 Serpentine soil plant Cr concentrations
Plant Cr Concentration (microg g-1) Location Reference
Lophostachys villosa 203 Goias State Brazil (Brooks et al 1992) Ruellia geminiflora
56 Goias State Brazil (Brooks et al 1992) Minuartia ophiolitica 83 Italy (Gambi 1992) Sanguisorba minor 194 Italy (Gambi 1992) Plantago lanceolata 135 Italy (Gambi 1992) Cerastium latifolium 110 Italy (Gambi 1992) Funaria flavicans 34-899 Maryland (Shaw 1991) Avena sativa 168-583 New Zealand (Lee 1992) Aerobryopsis longissima 200-700 New Caledonia (Lee et al 1977) Arenaria humifusa 530 Newfoundland Canada (Roberts 1992) Senecio pauperculus 406 Newfoundland Canada (Roberts 1992) Rhalomitrium lanuginosum 246 Newfoundland Canada (Roberts 1992) Santolia semidentata 95 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Asplenium cuneifolium 228 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992) Armeria eriophylla 38 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Arctostaphylos viscida 029-705 Tehama County CA (Gough et al 1989) Calluna vulgaris 417 Unst (Shewry and Peterson 1976) Armeria maritima 149 Unst (Shewry and Peterson 1976) Agrostis stolonifera 120 Unst (Proctor and Woodell 1971) Dicoma niccolifera 77 Zimbabwe (Proctor 1992) Xerophyta equisetoides 68 Zimbabwe (Proctor 1992)
124
125
Table 44 Chemical and textural properties of the collected soils
Ammonium Acetate Extractable ------------------------Hydrometer--------------------- Ca K Mg Na
Sand Silt ClaySample pH EC dSm
Organic Matter ------
LECO Nitrogen ------ -------------------mg kg-1-------------------
------ ------ ------
Texture
New Caledonia 534 01 05 lt001 279 29 692 44 43 32 25 LoamHawaii 604
03 10 002 839 46 168 53 8 26 66 ClayEight Dollar Mnt 651 07 16 006 599 62 3464 25 28 28 44 Clay
Nickel Mnt 665 05 07 001 167 42 3289 17 24 32 44 ClayHarvard Mine 691 03 11 003 998 301 2897 43 41 30 29 LoamJasper Ridge 671 04 32 012 505 237 3291 36 36 31 33 Clay LoamPillikin Mine 642 04 05 002 185 55 3421 29 45 23 32 Sandy Clay Loam
125
126
Table 45 Total major and trace element concentrationsdagger of the collected soils
Element Concentration New Caledonia Hawaii Eight Dollar
Mnt Nickel Mnt Harvard Mine Jasper Ridge Pillikin
Mine Ag mg kg-1 lt05 lt05 lt05 lt05 02 lt05 lt05Al 931 1325 314 095 179 314 179Ba mg kg-1 lt10 180 40 20 30 230 lt10Be mg kg-1 05 16 05 lt05 lt05 08 lt05Ca 003 012 085 018 007 079 002Cd mg kg-1 lt05 lt05 18 lt05 05 lt05 lt05Co mg kg-1 56 183 338 403 85 155 122Cr mg kg-1 827 388 9528 5761 2900 5976 1977Cu mg kg-1 78 248 43 25 51 33 4Fe 638 1296 1635 17 538 727 468K 002 046 004 003 004 039 001
Mg 072 026 753 829 792 1025 gt150Mn mg kg-1 257 2706 3017 2554 920 1664 903Na 001 008 005 001 003 057 lt001Ni mg kg-1 2780 345 5510 14100 1895 2180 2760P mg kg-1 50 260 160 150 130 140 20Pb mg kg-1 6 31 4 3 2 8 6Ti 016 041 008 002 lt001 02 002V mg kg-1 63 132 158 54 65 92 27W mg kg-1 10 10 10 10 lt10 lt10 lt10Zn mg kg-1 43 218 160 132 66 111 47
126
dagger ICP-AES analyses
127
Table 46 Compositionsdagger for soil Cr-containing minerals identified in Figure 49
Wt New Caledonia A
Hawaii B
Eight Dollar Mnt C
Eight Dollar Mnt D
Nickel Mnt E
Harvard Mine F
Jasper Ridge G
Jasper Ridge H
Pillikin Mine I
SiO2 491 008 175 026 002 000 1940 001 000Al2O3
756 185 272 2235 2041 088 559 1575 839NiO 023 005 001 006 011 2054 014 002 007FeO 7766a 6212 6633b 3420b 1784 2445 2629 2314 3105MnO 004 037 181 123 133 031 155 146 216Cr2O3 165 007 2387 3343 4682 3991 1878 5024 5355MgO 048 121 202 653 1315 1062 1668 926 497CaO 001 000 013 008 000 000 015 000 000Na2O 000 000 002 001 000 000 006 000 000TiO2 004 2465 012 006 012 000 000 003 018
TOTAL 9257 9041 9879 9821 9980 9671 8864 9920 10037Mineral Cr-Hydroxide Ulvoumlspinel Cr-Magnetite Chromite Mg-Chromite Trevorite CSM1 Chromite Chromite
127
dagger Electron microprobe analyses This is the form of iron unless otherwise stated a FeO is in the form Fe2O3 b FeO is in the form of Fe3O4 1CSM Microcrystalline chromitesilicate mixture
128
Table 47 Chromium-bearing mineraldagger masses and volumes required to influence Cr concentrations in soils
Cr Concentration Mineral mass (grams) required for 1000 g bulk soil analysis Mineral Mass Percent () (mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite
500 11 07 22 24 48 57 011 007 022 024 048 057 1000 22 15 43 48 95 115 022 015 043 048 095 1152000 43 29 86 96 191 230 043 029 086 096 191 2305000 108 73 216 241 477 574 108 073 216 241 477 574
Cr Concentration Mineral Volume Percent () in Serpentine Mineral Volume Percent () in Quartz
(mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite500 005 004 02 02 03 15 006 004 02 02 04 06
1000 01 007 04 03 07 30 01 007 04 03 07 112000 02 014 07 06 14 59 02 015 07 06 14 225000 06 036 18 16 34 147 06 037 18 16 35 55
128
daggerMineral formulas eskolaite Cr2O3 chromite FeCr2O4 fuchsite KCr2(AlSi3O10)(OH)2 uvarovite Ca3Cr2Si3O12 tawmawite Ca2CrAl2Si3O12(OH)2 kaumlmmererite Mg5CrAlSi3O11(OH)8
Note Molar volumes used in the calculations above are from their Al-enriched counterparts in Robie and Hemingway (1995) The molar volume for kaumlmmererite was estimated using unit cell volumes from XRD unit cell measurements from Robie and Hemingway (1995) Christofides et al (1994) and Phillips et al (1980)
129
Figure Captions
Figure 41 Selected serpentine soil localities as well as soils collected for analyses are
shown on a world map Active plate boundaries are shown Serpentine soil localities not
adjacent the plate boundaries on this map are related to ancient plate boundaries that are
no longer active
Figure 42 Field pictures of soils collected from a) Eight Dollar Mountain b) Nickel
Mountain c) Harvard Mine d) Pillikin Mine and f) Jasper Ridge Biological Preserve
Figure 43 Stability field diagram of Cr under a range of Eh and pH conditions (log a
Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom (1998) Adsorption
isotherms for Cr3+ and CrO42- on hydrous ferric oxide (HFO) are also shown (Leckie et
al 1980 Leckie et al 1984)
Figure 44 Microprobe analyses demonstrating Cr versus Al Wt for Cr-spinel
(chromite) and Cr-silicates including Cr-mica (fuchsitemariposite) Cr-garnet
(uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite) (Whitmore et al
1946 Leo et al 1965 Chen and Lee 1974 Neiva 1978 Cooper 1980 Phillips et al
1980 Grapes 1981 Schreyer et al 1981 Max et al 1983 Nutman et al 1983 Raase
et al 1983 Chao et al 1986 Von Knorring et al 1986 Kerrich et al 1987 Treloar
1987b a Morand 1990 Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen
1999 Proenza et al 1999) Chromium-spinel and Cr-silicate compositions based on
octahedral Cr and octahedral Al are also shown with tie-lines (dotted) denoting ideal
isomorphic substitution between Cr and Al end-members
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above Each dotted line connecting the
points represents a separate soil profile at each site
129
Figure 46 Ranges of pH for serpentine soils and related groundwater worldwide Soil
pH values were obtained from A(Kruckeberg 1992) B (Graham et al 1990) C
(Alexander et al 1989) D (Gough et al 1989) E (Rabenhorst et al 1982) F
(Roberts 1992) G (Bulmer et al 1992) H (Proctor 1992a) I (Jeffrey 1992) J
(Gambi 1992) K (Verger 1992) L (Proctor and Cole 1992) M (Cole 1992b) N
(Morrey et al 1992) O (Proctor 1992b) P (Mizuno and Nosaka 1992) Q (Cole
1992a) and R (Lee 1992) Groundwater pH values from serpentinite are from Barnes
and ONeil (1969) and Barnes et al (1972) Note the range of pH environments possible
for ultramafic rocks and their related soils
Figure 47 Soil pictures with Munsel color values and names from soils collected from
a) New Caledonia b) Hawaii c) Eight Dollar Mountain d) Nickel Mountain e) Harvard
Mine f) Jasper Ridge Biological Preserve and g) Pillikin Mine
Figure 48 Cr XANES spectra for soils analyzed from Jasper Ridge New Caledonia
Pillikin Mine Eight Dollar Mountain and Harvard Mine The pre-edge area indicative of
Cr(VI) peaks is noted in the diagram No definitive Cr(VI) peaks are present in any of
these spectra
Figure 49 Backscattering images of mineral phases within each soil are presented
Electron microprobe analyses are noted by a circle and letter for the Cr-bearing minerals
and the results are listed in Table 46
Figure 410 Chromite magnetite and hematite compositions from the collected soils
(gray circles) as well as analyses for serpentinite chromite (white squares) from Figure
44 are shown in these diagrams with Cr Wt versus Al Fe Mg and Al+Fe+Mg Wt
130
Figure 41
131
Figure 42
132
Figure 43
133
Cr Wt
0 10 20 30 40 50
Al W
t
0
5
10
15
20
25Cr-SpinelCr-MuscoviteCr-EpidoteCr-GarnetCr-Chlorite
Cr-SpinelBased on 4 Oxygens
Cr0 1 2
AlVI
0
1
2
ChromiteFeCr2O4
HercyniteFeAl2O4
Cr-MuscoviteBased on 11 Oxygens
Cr0 1 2
AlVI
0
1
2
MuscoviteKAl2(AlSi3O10)(OH)2
FuchsiteKCr2(AlSi3O10)(OH)2
Cr-GarnetBased on 12 Oxygens
Cr0 1 2
AlVI
0
1
2
GrossularCa3Al2Si3O12
UvaroviteCa3Cr2Si3O12
Cr-EpidoteBased on 125 Oxygens
Cr0 1
AlVI
2
3
ClinozoisiteCa2Al3Si3O12(OH)2
TawmawiteCa2CrAl2Si3O12(OH)2
Cr-ChloriteBased on 14 Oxygens
Cr0 1
AlVI
0
1
2
Kammererite(Mg5Cr)(AlSi3)O10(OH)8
Clinochlore(Mg5Al)(AlSi3)O10(OH)8
Figure 44
134
Maryland (USA)(Rabenhorst et al 1982)
Chromium (mg kg-1)
0 1000 2000 3000 4000 5000 6000 7000
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Tehama County CA (USA)(Gough et al 1989)
Chromium (mg kg-1)
0 2000 4000 6000 8000 10000 12000 14000
Dep
th (m
eter
s)
00
02
04
06
08
Kalamath Mountains CA (USA)(Hotz 1964)
Chromium (mg kg-1)
2000 4000 6000 8000 10000 12000 14000 16000
Dep
th (m
eter
s)
00
05
10
15
20
25
30
35
40
45
50
Newfoundland (Canada)(Roberts 1992)
Chromium (mg kg-1)
800 1000 1200 1400 1600 1800
Dep
th (m
eter
s)
00
02
04
06
08
Harvard Mine CA (USA)(Unpublished Data)
Chromium (mg kg-1)
0 500 1000 1500 2000 2500
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Jasper Ridge CA (USA)(Oze et al 2003)
Chromium (mg kg-1)
1500 2000 2500 3000 3500 4000 4500 5000
Dep
th (m
eter
s)
00
01
Figure 45
135
LocationsZ A B C D E F G H I J K L M N O P Q R S
pH
0
1
2
3
4
5
6
7
8
9
10
11
12
Nor
th A
mer
ica
Can
ada
Brit
ain
Italy
Far E
ast
Japa
n
Zim
babw
e
Sout
h A
fric
a
Wes
tern
Aus
tral
ia
New
Zea
land
Mg-Bicarbonate Waters pH=8
Ca-Hydroxide Waters pH=11-12
Average RainpH=57
Sout
h A
fric
a
Italy
Irela
nd
Cal
iforn
ia
Cal
iforn
ia
Brit
ish
Col
umbi
a
Mar
ylan
dCal
iforn
ia
Figure 46
136
Figure 47
137
-2
-1
0
1
2
3
4
5
5980 5990 6000 6010
Energy (eV)
Nor
mal
ized
Abs
orba
nce
New Caledonia
Pillikin Mine
Eight Dollar Mnt
Jasper Ridge
Harvard Mine
Cr(
VI)
Figure 48
138
Figure 49
139
Cr Wt 0 10 20 30 40 50
Al W
t
0
10
20
30 Soil Serpentinite
Cr Wt 0 10 20 30 40 50
Fe W
t
0
20
40
60
80
Cr Wt 0 10 20 30 40 50
Mg
Wt
0
2
4
6
8
10
Cr Wt 0 10 20 30 40 50
Al +
Fe
+ M
g W
t
0
20
40
60
80
MagnetiteHematite
Chromite
MagnetiteHematite
Chromite
Chromite
MagnetiteHematite
SpinelSilicate
SpinelSilicate
SpinelSilicate
SpinelSilicate
Figure 410
140
- Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip1
-
- Chapter 3 Oxidative Promoted Dissolution of Chromite by
- Manganese Dioxide and Concurrent Production of Ch
-
- Chapter 4 Chromium Geochemistry of Serpentine S
-
- List of References
-
- Chapter 2
- Chapter 3
-
- Chapter 4
-
- Appendix A
- Chapter 2
-
- Chapter 3
-
- Chapter 4
-
- Appendix A
-
- CHAP1pdf
-
- Toxicity and Regulations of Chromium
- Mineralogy of Chromium in Serpentinites
-
- Chapter Summaries
-
- CHAP2Tablepdf
-
- JR305CLAY Depth 0-5cm
- JR3515CLAY Depth 5-15cm
- JR33045CLAY Depth 30-45cm
-
- HCAP2Figspdf
-
- Figure 27
-
- CHAP3Tablepdf
-
- Figure 36
- Figure 37
-
- CHAP4pdf
-
- ABSTRACT
- REVIEW
-
- Chromium Geochemistry
-
- Soil Organic Matter
-
- CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
-
- BedrockProtolith Mineralogy
- Soil Chemistry and Physical Properties
- Soil Mineralogy
- Backscattered electron images and microprobe analyses were used to identify the Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49 and corresponding Cr-bearing mineral compositions are listed in Table 46 The New C
-
- DISCUSSION
-
- Chromium Spinel in Soils and Rocks
-
- CONCLUDING STATEMENT
- ACKNOWLEDGEMENTS
-
- CHAP4Tablepdf
-
- Ulvoumlspinel
-
ABSTRACT
Chromium persists in the environment as Cr(III) a non-hazardous species and a
micronutrient in human nutrition or as Cr(VI) a strong oxidizing agent a toxin to living
cells and a Class A human carcinogen by inhalation Serpentinites or metamorphosed
ultramafic rocks are distributed worldwide and contain Cr concentrations typically
greater than 200 mg kg-1 Additionally serpentine soils a generic term used to describe
any soil derived from serpentinite regardless of its physical or chemical properties
contain elevated Cr concentrations often exceeding their corresponding protolith
Chromium in serpentinites and serpentine soils is dominantly in the form of Cr(III)
however geochemical interactions during serpentinite weathering and soil formation
provide oxidative pathways allowing the formation of Cr(VI) The relationship between
rock and soil and the processes governing the Cr chemistry are difficult to assess due to
the inherent complexity of evaluating and correlating the mineralogy and geochemistry of
both systems Ultimately a multidisciplinary approach combining field analytical and
laboratory studies is necessary to evaluate the Cr geochemistry related to these rocks and
soils and to determine whether or not serpentinites and serpentine soils are sources of
non-anthropogenic Cr contamination
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of
California produces serpentine soils containing high concentrations of Cr as well as other
potentially toxic elements including Ni Co and Mn Chromium concentrations in
serpentine soils from Jasper Ridge Biological Preserve in the Central Coast Range are as
high as 4760 mg kg-1 nearly three times greater than the serpentinite protolith (~1800
mg kg-1) Chromium-containing minerals within the bedrock include chlorite (~03 Cr
Wt ) enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt )
magnetite (82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate
phase containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
iv
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
Despite the resistance of chromite to weathering Cr release and oxidation from
chromite is a potential environmental hazard in sediments and soils and a pathway for
soil development (weathering of primary minerals) related to ultramafic rocks and their
metamorphic derivatives (serpentinites) Birnessite is a common pedologic mineral in
these sediments and soils capable of oxidizing aqueous Cr(III) The interaction between
chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4) and birnessite was investigated with
an interest in the potential generation of Cr(VI) The rate of Cr(VI) in solution increases
with increasing chromite suspension densities and decreasing pH but is independent of
birnessite suspension densities at values greater than ~20 m2 L-1 at 25degC The overall rate
expression of Cr released and oxidized in solution from chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression--rates ranging from 9times10-4 to 44times10-3 microM Cr(VI) h-1 These
experiments therefore demonstrate that serpentine soils are a potential source of non-
anthropogenic Cr(VI)
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
v
chemistry of Cr in serpentine soils including its protolith is reviewed with a focus on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils have widely varying pH contain a variety Fe(III) oxides (magnetite and
hematite) phyllosilicates (serpentine and chlorite) and clays (smectites and
vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1) and
Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Additionally Cr(III) is the only valence state
observed in the serpentine soil solids however Cr(VI) has been identified in New
Caledonia and California serpentine soil solutions at concentrations below 30 microM The
enrichment and range of Cr concentrations in serpentine soils are directly related to the
presence of Cr-spinels specifically chromite and Cr-magnetite These spinels are
resistant to weathering and are preserved in the soil environment however oxidation of
Cr(III) from Cr-spinels by high valent Mn oxides or other strong oxidants is a potential
source of Cr(VI) identified in serpentine soil solutions Due to the chemically and
physically resistant nature of the Cr-spinels Cr-bearing silicates including clay minerals
Cr-chlorite Cr-garnet Cr-mica and Cr-epidote are more viable sources of Cr identified
in vegetation soil extractions soil solutions and related waters
vi
PREFACE
The chemistry and mineralogy of Cr related to serpentinites and serpentine soils
are assessed in this dissertation using a multidisciplinary field analytical experimental
and theoretical approach The four chapters that constitute this dissertation describe my
investigations related to these rocks and soils Additionally the common theme shared
by these chapters is that they all examine the Cr mineralogy in serpentinites and
serpentine soils and the potential of Cr to be mobilized and oxidized in near-surface
environments Chromite a Cr-bearing spinel is the dominant source of Cr in
serpentinite and serpentine soils Although chromite is resistant to weathering the
presence of birnessite a high valent Mn oxide will increase the release of Cr and the
production of Cr(VI) from chromite over a range of pH Chromium-clinochlore or
kaumlmmererite contains Cr concentrations less than 3 Wt however the rate at which Cr-
clinochlore weathers compared to chromite makes this mineral a more probable source of
Cr released into soil solutions These types of concerns are addressed in this thesis Brief
descriptions of the four individual dissertation chapters including their publication status
and a list of the co-authors are presented in the following paragraphs
Chapter 1 presents the main objectives of this dissertation The chemistry
regulations and toxicity of Cr are discussed as well as the general characteristics of
serpentinites and serpentine soils Additionally a brief summary of each chapter is
provided
Chapter 2 focuses on serpentinites and serpentine soils at Jasper Ridge Biological
Preserve Stanford CA The results have been submitted for publication in an article
entitled ldquoChromium geochemistry of serpentinized ultramafic rocks and serpentine soils
in the Franciscan Complex of Californiardquo to the American Journal of Science (March
2003) Scott Fendorf Dennis Bird and Robert Coleman are the second third and fourth
co-authors respectively
Chapter 3 is an experimental study evaluating the release of Cr and the
production of Cr(VI) from chromite in the presence of birnessite over a range of pH The
results for this study will be submitted for publication in an article entitled ldquoOxidative
promoted dissolution of chromite by manganese dioxide and concurrent production of
vii
chromaterdquo to Environmental Science and Technology The co-authors for this paper are
Scott Fendorf and Dennis Bird
Chapter 4 is a review paper that critically evaluates the chemistry of serpentine
soils and serpentinites primarily focusing on Cr geochemistry Additionally analytical
studies of serpentine soils collected from California Oregon and New Caledonia are
presented This study will be presented at the Robert Coleman Symposium and has been
submitted as an article entitled ldquoChromium geochemistry of serpentine soilsrdquo for an
International Geology Review Special Publication Scott Fendorf Dennis Bird and
Robert Coleman are the second third and fourth co-authors respectively
Appendix A is a preliminary investigation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite) a mineral common in serpentinites and present in serpentine
soils This appendix is entitled ldquoEstimation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite)rdquo The results of this chapter will be included in a study that
estimates the thermodynamic properties of Cr-bearing silicates including fuchsite (Cr-
muscovite) tawmawite (Cr-epidote) and uvarovite (Cr-garnet)
viii
ACKNOWLEDGEMENTS
Serpentinites serpentine soils and chromium are the Geologic Trinity of this
dissertationhellipeach topic is intimately related to one another The scope and breadth of
this research required a significant amount of time and funding None of this research
would have been possible without the generous support of several foundations
organizations and Stanford University I am grateful for the financial aid I have
received from the Stanford Fellowship the Stanford University McGee and Shell funds
the National Science Foundation the Geologic Society of America Graduate Research
Fund the Stanford Office of Teaching and Learning grant and the Jasper Ridge
Biological Preserve Melon Grant
I would not be where am I today if it were not for my parents Robert and Sue
Oze and my brother Rob Oze Thank you for the sandbox and all the hose-driven
volcanoeshellipthose Wookies never stood a chance While some families vacation in theme
parks my family frequented gem and mineral conventions collected rocks and shells
from mist-shrouded shores and sought lucky stones including an infamous quartz orb
Sometimes I think I was destined to be a geologist I would also like to thank everyone
in my family and all our family friends who have provided that extra little push and
support along the way These people include Tudor Bennatts Louis Oze Margaret and
John Chambers John Bennatts Nancy Skiles the Izumis the memory of Witchie Poo
hijacking Harleys from Bremerton bikers and the Man in the Sound waiting for that one
last catch Close friends I would like to acknowledge who are not related to the Stanford
crowd include Dave and Leigh Anderson and Rebecca Parker These satellite friends
helped me keep in touch with the world outside of academia and proved to me that I
wasnrsquot meant to have a real job A special thank you goes to Katherine Akako
Izumihellipyou have made this whole trip worthwhile
My PhD advisors Scott Fendorf and Dennis Bird provided an invaluable amount
of help patience and energy over the past five years Scott always made the time to
discuss and refine many of my projects making them relevant to many current
geochemical issues Additionally he helped me hone my soil science skills in the field
and in the lab Out in the field Scott can dig a pretty mean hole Dennis devoted a
significant amount of time and sweat in order to make this dissertation happen His
ix
insightful knowledge of geochemistry and of the geology of the Sierra Nevadas provided
the cornerstone for this dissertation
Robert Coleman played a key role in the production of this dissertation He
always knew what to say at the right time and helped me understand the complexity of
ophiolites and serpentinites through experienced eyes Additionally Keith Loague and
Gary Ernst deserve recognition for their outstanding support as committee members for
this dissertation over the last four years I would also like to thank Jonathan Stebbins and
Gail Mahood for providing me the opportunity to attend Stanford
Bob and Vicki Jones always offered an open ear whether it was science or solving
the worldrsquos problems Rick Hazlett (Pomona College) and John Winter (Whitman
College) both had a significant impact on shaping my future science goals and objectives
during my undergrad years and both encouraged me to pursue a career in geology They
were the catalyst for this whole experiment
I owe a lot to my research group members (both the Bird Cage and Soil
Chemistry Group) and those people who didnrsquot mind me tagging along Before I
introduce these people one person has been my friend and co-conspirator for many years
at Stanford Tim Copehellipyou are a pain in my ass Besides never picking up any of your
crap and breaking all my stuff this whole grad school experience would not have been
the same without you Other friends and acquaintances that will not be forgotten in
chronological order of me walking into my office for the first time on September 11
1998 include Anne Dyreborg Esra Inan Tom Bawden Patrick Redmond Phil
Neuhoff Thrainn Fredrickson Lara Heister Kaye Savage Andrea Fildani Mike Schultz
Trayle Kushan Kevin Theissen Angela Hessler Chris Van de Ven Kono Lemke Colin
Doyle Colleen Hansel Ben Bostick Matthew LaForce Nancy Grumet Scott Kroeker
Anders Meibom Brian Ebel Matt Ginder-Vogel Matt Pollizitto Spud Amy Weislogel
Travis Horton and Jake Waldbauer I would like to end this Acknowledgement section
and begin my dissertation with a haikuhellip
Better than outcrops
Are the people Irsquove met here
Thanks for all the beer
x
TABLE OF CONTENTS
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iv Prefacehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip vii Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip ix Table of Contentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xi List of Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xiv List of Illustrationshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xvi Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 1
Toxicity and Regulations of Chromiumhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 2 Mineralogy of Chromium in Serpentiniteshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 3 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5 Chapter Summarieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5
Chapter 2 Chromium Geochemistry of Serpentinized Ultramafic Rocks and Serpentine Soils in the Franciscan Complex of Californiahelliphelliphelliphelliphelliphellip 8
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 8 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 9 Backgroundhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 10
Geologic Background of the Serpentinite at Jasper Ridgehelliphelliphelliphellip 10 Serpentine Soils Characteristics and Reviewhelliphelliphelliphelliphelliphelliphelliphellip 11 Sources of Chromium ProtolithBedrockhelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 12 Chromium-Bearing Phases in Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 13
Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Location and Sample Collectionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Mineral Identification in Rocks and Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Soil Chemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 15 X-ray Diffraction and Clay-Size Mineral Identificationhelliphelliphelliphelliphellip 16 X-ray Absorption Spectroscopyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17 Sequential Extraction Procedurehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17
Resultshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Serpentinite Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Distribution of Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip21 Serpentine Soil Chemistry and Mineralogy at Site JR3helliphelliphelliphelliphellip 22 Sequential Extractionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 26
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 27 Concluding Remarkshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 33 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 44 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 47
xi
Chapter 3 Oxidative Promoted Dissolution of Chromite by Manganese Dioxide and Concurrent Production of Chromatehelliphelliphelliphelliphelliphellip 63
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 63 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 64 Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 65
Solid Phase Characterization and Synthesishelliphelliphelliphelliphelliphelliphelliphelliphellip 65 Extraction Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 66 Batch Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Solution Analysishelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69
Results helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite Extraction Experimentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite and Birnessite Batch Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 70 Secondary and Solid Phase Examination of Chromite and Birnessitehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 72 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 72
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 73 Chromium(VI) Pathways from Chromite and Birnessite into Solutionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 76 Serpentine Soils and Cr(VI)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 77
Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 79 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 80 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 84 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 87
Chapter 4 Chromium Geochemistry of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip100
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 100 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 101 Reviewhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102
Chromium Geochemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102 Mineralogy of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 103 Sources of Chromium for Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphellip 105 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 106 Soils Solutions and Related Waterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 108 Serpentine Soil Vegetationhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 110 Soil Organic Matterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 111
California Oregon and New Caledonia Serpentine Soilshelliphelliphelliphelliphelliphelliphellip 112 Sample Locations and General Soil Characteristicshelliphelliphelliphelliphelliphellip 112 BedrockProtolith Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 113 Soil Chemistry and Physical Propertieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 114 Soil Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116 Chromium and Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 117
xii
Chromium Spinel in Soils and Rockshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 119 Concluding Statementhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 120 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 121 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 122 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 129 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 131
Appendiceshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 141 Appendix Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 142 Appendix Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 143 Appendix A Estimation of the Thermodynamic Properties of Kaumlmmererite (Cr-Clinochlore)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip145
Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 145 Nomenclature and Crystal Chemistry of Cr-chloritehelliphelliphelliphelliphelliphelliphelliphelliphellip 146 Estimation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 147
Estimation of theStandard Molar Volume Using Unit Cell Measurementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of the Standard Molar Volume Heat Capacity and Entropyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of Standard Molar Enthalpy and Gibbs Free Energyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 150 Critique and Discussion of the Estimation Algorithms and SUPCRT92helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 151
Evaluation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 152 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 155 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 161 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 162
List of References
References to Chapter 1helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 166 References to Chapter 2helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 172 References to Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 179 References to Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 183 References to Appendix Ahelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 195
xiii
LIST OF TABLES Chapter 2 Table 21 Whole rock elemental concentrations of serpentinites chert and sandstone Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in
serpentinite rock at Jasper Ridge Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in
serpentine soil at Jasper Ridge Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25
and 26 Table 24 Bulk soil elemental concentrations for soils and bedrock collected along
the transect in Figure 21b Table 25 Chemical and physical properties of the soil at Site JR3 Table 26 Selected trace metal concentrations for each size fraction versus depth at
Site JR3 Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3 Table 28 X-ray diffraction d-spacings for common silicates Table 29 Results of the Diablo clay serpentine soil and chromite sequential
extraction experiments Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1
kg of sample Chapter 3 Table 31 Chromite and birnessite masses and concentrations used for each batch
experiment Table 32 Serpentine soil and birnessite experimental parameters Table 33 Summary of Cr(VI) rates obtained from batch experiments Table 34 Comparative analyses of batch experiment solutions and chemical
extractions of chromite and birnessite solids Chapter 4 Table 41 Measured Cr concentrations in serpentine soils worldwide Table 42 Summary of Cr Fe and Ni extractions from serpentine soils Table 43 Serpentine soil plant Cr concentrations Table 44 Chemical and textural properties of collected soils Table 45 Total major and trace element concentrations of the collected soils Table 46 Compositions for soil Cr-containing minerals identified in Figure 49 Table 47 Chromium-bearing mineral masses and volumes required to influence Cr
concentrations in soils Appendix A Table A1 Cr and Al end-members in the Cr-clinochlore solid solution
xiv
Table A2 Standard state thermodynamic properties of Cr and Al-bearing minerals Table A3 Unit cell measurements and molar volumes of clinochlore and Cr-
clinochlores Table A4 Entropy volume and heat capacity for kaumlmmererite using the Ransom and
Helgeson Method Table A5 Fictive polyhedral considered in the estimations Table A6 Estimated thermodynamic properties for kaumlmmererite
xv
LIST OF FIGURES Chapter 2 Figure 21a Soils map of Jasper Ridge Biological Preserve CA Figure 21b Serpentinite and serpentine soil sampling sites at Jasper Ridge Biological
Preserve CA Figure 22 Chromium concentrations versus depth for four serpentine soils in
Maryland California and New Foundland Figure 23 Petrographic images of serpentinite at Jasper Ridge Figure 24 Elemental concentrations relative to Cr for chlorite pyroxene magnetite
chromite and a chromite-silicate mixture in the serpentinite at Jasper Ridge
Figure 25 Backscattering electron images of minerals in the serpentinite at Jasper Ridge
Figure 26 Backscattered electron images of Cr-chlorite and Cr-magnetite assemblage in the serpentinite bedrock
Figure 27 Highest and lowest Cr concentrations at each site in the serpentine soil transect
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected for total gt2 mm sand silt and clay size fractions
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soils at Jasper Ridge
Figure 210 X-ray diffraction clay analyses of sample JR3515 Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at specific depths Figure 212 X-ray microprobe elemental maps of Cr and Fe for the serpentine soil and
serpentinite at site JR3 Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments Figure 214 Scanning electron microprobe images of the chromite standard after each
sequential extraction experiment Figure 215 Tie-lines connecting serpentinite chromite and end-member chromite with
soil chromite values Chapter 3 Figure 31 Scanning electron microprobe images of chromite and birnessite standards Figure 32 Chromite extraction experiments evaluating Cr Fe and Al Figure 33 pH and Eh versus time (hours) for each chromite-birnessite chromite and
birnessite batch experiments Figure 34 pH and Eh with Cr stabilities Figure 35 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant chromite and variable birnessite at pH 5 Figure 36 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant birnessite and variable chromite at pH 5
xvi
Figure 37 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for variable pH and constant chromite and birnessite Figure 38 Rate of Cr(VI) production (d[Cr(VI)]dt) as a function of the surface area
concentration of chromite birnessite and pH Figure 39 SEM examination of solids following the termination of the batch
experiments Figure 310 Serpentine soil and birnessite experiments Figure 311 Iron analyses and the solubility of amorphous Fe(OH)3 and goethite Figure 312 Congruent dissolution of chromite based on Fe and Cr Figure 313 Three pathways of Cr release and oxidation Chapter 4 Figure 41 Serpentine soil localities worldwide Figure 42 Field site pictures of serpentine soils Figure 43 Stability field diagram of Cr over a range of Eh and pH conditions Figure 44 Chromium and Al Wt plot and octahedral Cr and octahedral Al plots
for chromite and Cr-silicates including Cr mica (fuchsitemariposite) Cr-garnet (uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite)
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (m) for selected serpentine soils
Figure 46 pH of serpentine soils and related groundwater worldwide Figure 47 Soil pictures with Munsel color values and names from collected soils Figure 48 Cr XANES spectra for collected serpentine soils Figure 49 Backscattering images of mineral phases within each serpentine soil Figure 410 Chromite magnetite and hematite compositions from analyzed soils as
well as analyses for chromite reported from around the world Appendix A Figure A1 Compositional variation of Cr2O3 and Al2O3 Wt in Cr-chlorite Figure A2 Compositional variation of Cr and VIAl in Cr-chlorite based on 14 oxygens Figure A3 Molar volume estimation of kaumlmmererite based on XRD unit-cell
parameters Figure A4 Natural and calculated compositions of co-existing Cr-clinochlore and Cr-
spinel solid solution in the system Mg0-FeO-Cr2O3-Al2O3-SiO2-H2O at 2 kbar pressure
xvii
CHAPTER 1
INTRODUCTION
Serpentinites or metamorphosed ultramafic rocks are distributed worldwide and
are commonly associated with ophiolite complexes (Coleman 1977 Coleman and Jove
1992) These rocks weather to form serpentine soils a generic term used to describe any
soil derived from serpentinite regardless of its physical or chemical properties
Serpentinites and serpentine soils contain Cr-bearing minerals the most prevalent being
chromite and Cr released from these minerals into surface and subsurface environments
is potentially toxic and carcinogenic The main objective of this dissertation is to
evaluate the mineralogy and chemistry of Cr related to serpentinites and serpentine soils
over geologic time using an integrated analytical experimental and theoretical approach
Additionally the potential of serpentinites and serpentine soils as sources of non-
anthropogenic Cr contamination is investigated A discussion of Cr and its significance
to serpentinites and serpentine soils begins with its discovery by a Frenchman
investigating an ore named ldquored leadrdquo from the Urals
Nicolas-Louis Vauquelin discovered Cr in 1797 in the mineral crocoite PbCrO4
(Burns and Burns 1975) The metalelement was named chromium after the Greek word
chroma meaning color due to its strong influence on mineral pigmentation (Burns and
Burns 1975) For example rubies and emeralds derive their color from trace amounts of
Cr in their structures More importantly JT Lowitz in 1798 discovered Cr in chromite
or Cr-spinel the primary source of Cr for industrial purposes (Burns and Burns 1975)
Chromium is a group VIB transition metal with two common oxidation states (Cr(III) and
Cr(VI)) an average crustal abundance of 100 mg kg-1 and an abundance of 513x105 in
the solar system (relative to [H]=1x1012) (Mason and Moore 1982)
The majority of Cr on the earth exists as Cr(III) where it is an immobile cation
that complexes strongly with organic matter and adsorbs on oxides and silicates even at
low pHs (pHgt4) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) Additionally
the chemical mobility of Cr(III) is limited due to the formation of highly stable
(oxy)hydroxides (Rai and Zachara 1988 Rai et al 1989 Ball and Nordstrom 1998)
1
Chromium(III) has the highest octrahedral site preference energy (OSPE) of any trace
element and it isomorphically substitutes into octahedral sites in oxides
(oxy)hydroxides sulfides and silicates Although Cr(III) is known to substitute for both
octahedral Al(III) and Fe(III) the strongest correlation for substitution appears to be for
Al(III) in oxides and silicates as demonstrated in Chapters 2 and 4 and Appendix A
Chromium(VI) is a strongly oxidizing and weakly adsorbing oxyanion in the form of
chromate (HxCrO42-x) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995 Shiraki
1997 Ball and Nordstrom 1998) Consequently Cr(VI) is chemically mobile and
bioavailable over a wide range of pH (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Chromium(VI) is tetrahedrally coordinated and behaves similarly to
sulfate in minerals such as jarosite (Baron and Palmer 1996)
The pathways and chemistry of Cr oxidation and reduction in soils and related
waters are covered in depth by Fendorf (1995) and Ball and Nordstrom (1998) The
reduction of Cr(VI) to Cr(III) may occur by reactions with Fe(II)aq Fe(II)-containing
minerals such as magnetite organic matter H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996) Oxidation of Cr(III)
to Cr(VI) related to serpentinite-related water and serpentine soil solutions is still a matter
of conjecture Chromium(III) oxidation in serpentine soils has been hypothesized to be
the result of Mn oxides such as birnessite as suggested by both Gough et al (1989) and
Becquer et al (2003) This idea is explored in depth in Chapter 3 where birnessite is
shown to influence the release and oxidation of Cr from chromite over a range of pH (3
to 8) at 25oC
Toxicity and Regulations of Chromium
The examination of Cr based on its oxidation state is necessary due to the
juxtaposing behavior and toxicity of Cr(III) and Cr(VI) Chromium(III) is an essential
micronutrient in human nutrition used by the body to process sugar protein and fat
(Vincent 2000) Inhalation of Cr(VI) is known to cause irritation in the nose including
symptoms such as nosebleeds (epistaxis) and ulcers andor perforations in the nasal
septum (Katz and Salem 1993 Kuo et al 1997) Ingesting considerable quantities of
Cr(VI) can result in stomach ulcers convulsions kidney and liver damage and death
2
(Katz and Salem 1993) Skin contact with certain Cr(VI) compounds causes allergic
contact dermatitis (ulcers) and irritant dermatitis (Katz and Salem 1993) Chromate
induced ulcers or chromeholes are typically red swollen crusted painless lesions
(Katz and Salem 1993) The nosology of Cr is beyond the scope of this dissertation
however Lees (1991) is an excellent source for further information Overall the
Environmental Protection Agency (EPA) the Occupational Safety and Health
Administration (OSHA) the World Health Organization (WHO) and the Department of
Health and Human Services (DHHS) have determined that that Cr(VI) compounds can
increase the risk of lung cancer primarily by inhalation as well as other forms of cancer
caused by ingestion
Federal and state regulation standards appropriate for both Cr(III) and Cr(VI) are
summarized and discussed by Proctor et al (1997) Human health action levels for Cr
vary for each state and the ranges for total Cr in soils include the following 200-66000
Cr(III) mg kg-1 (residential) 02-2000 Cr(VI) mg kg-1 (residential) 2500-1000000
Cr(III) mg kg-1 (non-residential) and 64-71000 Cr(VI) mg kg-1 (non-residential)
(Proctor et al 1997) Incidentally Cr concentrations in serpentine soils and serpentinites
(gt200 Cr mg kg-1) exceed many state action levels and would be considered toxic The
EPA has set a limit of Cr(III)+Cr(VI) at 100 microg L-1 in drinking water The OSHA has
set limits of 500 microg of water soluble Cr(III) compounds per cubic meter of workplace air
or 500 microg m-3 1000 microg m-3 for metallic Cr and insoluble Cr compounds and 52 microg m-3
for Cr(VI) compounds for 8-hour work shifts and 40-hour work weeks The limits
proposed above are constantly changing and should not be used for reference without
checking current regulation standards
Mineralogy of Chromium in Serpentinites
Chromium occurs in a variety of minerals either as a major or minor constituent
(Burns and Burns 1975) Chromium-bearing minerals in serpentinites are primarily
oxides and silicates in which the Cr is present as Cr(III) Chromite is a Cr-bearing spinel
abundant in serpentinites and is resistant to low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) The chemistry and
significance of chromite is addressed throughout this dissertation Pyroxenes such as
3
augite and enstatite are primary minerals identified in serpentinites and contain lt8 Cr Wt
(Grapes 1981 Von Knorring et al 1986 Treloar 1987a Challis et al 1995
Sanchez-Vizcaino et al 1995 Deer et al 1996) Chromate minerals such as crocoite
(PbCrO4) are rare in serpentinites and the occurrence of these minerals are limited to Pb
deposits in Australia Tasmania and the Urals (Williams 1974 Crane et al 2001)
Chromium(III) released from the primary Cr-containing phases may
isomorphically substitute into oxides and silicates for Al(III) and Fe(III) due to size
(octahedral radii Cr3+ = 0615 Aring Al3+ = 061 Aring and Fe3+=0643 Aring) and charge
similarities Magnetite formed during serpentinization incorporates Cr into its structure
and serpentine minerals (antigorite lizardite and chrysotile) contain trace concentrations
of Cr (Spangenberg 1943 Page and Coleman 1967 Burkhard 1993) Eskolaite (Cr2O3)
has been identified in Finland (Von Knorring et al 1986) Metasomatic alteration of
ultramafic rocks commonly results in the formation of Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite) uvarovite (Cr-garnet) tawmawite (Cr-epidote) and
kaumlmmererite (Cr-chlorite) (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Neiva 1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981
Max et al 1983 Nutman et al 1983 Raase et al 1983 Chao et al 1986 Von
Knorring et al 1986 Kerrich et al 1987 Treloar 1987b Treloar 1987a Morand 1990
Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993 Christofides et al 1994
Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Proenza et al
1999) Additionally kyanite and andalusite in the proximity of ophiolite complexes
incorporate up 5 Cr Wt (Cooper 1980 Raase et al 1983 Kerrich et al 1987)
Chapter 4 provides an in depth discussion of these Cr-bearing silicates as well as
demonstrating mineral compositions reported in the literature based on octahedral Cr and
Al Appendix 1 is a preliminary investigation devoted to estimating the thermodynamic
properties of kaumlmmererite These properties can then be incorporated into geochemical
modeling programs to begin accessing the stability of this common serpentinite mineral
over a range of temperatures and pressures
4
Chromium in Serpentine Soils
Evaluating Cr in serpentine soils requires an examination of both the mineralogy
and chemistry of the serpentinite and soil Measured Cr concentrations in serpentinites
and serpentine soils range from ~200 mg kg-1 to 6 Wt all of which is Cr(III) (Soane
and Saunder 1959 Proctor and Woodell 1971 Shewry and Peterson 1976 Jaffre 1980
Proctor et al 1980 Rabenhorst et al 1982 Schwertmann and Latham 1986 Brooks
1987 Gough et al 1989 Gambi 1992 Lee 1992 Morrey et al 1992) In Chapters 2
and 4 chromite and Cr-magnetite are identified as the most prevalent sources of Cr
responsible for Cr-enrichment in these soils Chromite is resistant to chemical
weathering and it has been observed to undergo incongruent dissolution conserving Cr in
serpentine soils Chromium is tightly bound in the matrix of the soil less than 1 of the
total Cr can is removed using standard soil extraction methods (10 hydrogen peroxide
(H2O2) solution 01 M potassium permanganate (KMnO4) solution10 M hydrofluoric
(HF) acid 1 M hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and 1
mM barium chloride (BaCl2) solution) due to the stability of chromite Although silicates
contain minor concentrations of Cr compared to chromite they are perhaps the main
contributors of Cr identified in plants (lt200 Cr mg kg-1) and soil solutions (lt50 microg L-1)
During soil-forming processes Cr released to soil solutions by the dissolution of Cr-
bearing minerals are sequestered into a host of structural and surface sites available in
authigenic Fe(III) oxides Fe(III) hydroxides and clay minerals (Schwertmann and
Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990 Gerth 1990 Gasser and
Dahlgren 1994 Ilton and Veblen 1994 Brigatti et al 2000 Ilton et al 2000) The
clay-sized fractions in soils at Jasper Ridge Biological Preserve CA contain Cr
concentrations as high as 1000 Cr mg kg-1
Chapter Summaries
A detailed field and analytical examination of Cr in serpentinites and serpentine
soils at Jasper Ridge is presented in Chapter 2 A majority of the Cr enrichment and
variability with depth in these soils is the result of small grains of chemically resistant
chromite from the serpentinite protolith Chromite identified in the soil undergoes
incongruent dissolution conserving Cr relative to chromite compositions identified in the
5
serpentinite bedrock Sequential extraction experiments using barium chloride (BaCl2)
hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and hydrofluoric (HF)
acid were performed on a chromite standard confirming that Cr-bearing spinels are
resistant to chemical weathering X-ray diffraction analyses of the clay-size fraction in
the serpentine soils identified Cr-clinochlore smectite vermiculite and serpentine with
the smectite being the most abundant mineral in the clay-size fraction Chromium(III)
was the only identified oxidation state of Cr in the serpentine soil solids and in
serpentinites at Jasper Ridge in the solid phase This chapter represents a comprehensive
site assessment identifying the nature and behavior of Cr in serpentine soils with respect
to the bedrock mineralogy and depth
Chromite an abundant Cr-enriched spinel is responsible for a majority of the Cr
enrichment identified in ultramafic rocks serpentinites and serpentine soils The release
and oxidation of Cr from chromite is a pathway to consider when evaluating the potential
of ultramafic rocks and their related sediments and soils as sources of Cr(VI) In
Chapter 3 experiments aimed at quantifying the release and oxidation of Cr from
chromite in the presence of birnessite are described Chromite is extremely resistant to
dissolution (lt8 Cr mg kg-1 is released) when immersed in solutions including 1 mM
barium chloride solution 10 M hydrofluoric acid 10 hydrogen peroxide and 1 M
hydroxylamine hydrochloride with 25 acetic acid Highly oxidizing solutions of 01 M
potassium permanganate increased the rate and release of Cr relative to the other
solutions listed above thus birnessite a high valence Mn oxide present in serpentine
soils was chosen to determine if Mn could accelerate Cr release and oxidation from
chromite The results presented in Chapter 3 demonstrate that birnessite increases the
rate of Cr(VI) observed in solution from chromite This production of Cr(VI) is
dependent on the chromite suspension density where birnessite surface are concentrations
are greater than 20 m2 L-1 between pHs 3 and 8 at ~25degC
A comprehensive review of the chemical characteristics of serpentine soils
including topics such as the mineralogy of serpentine soils soil solutions organic matter
and vegetation is provided within Chapter 4 Additionally chemical and spectroscopic
analyses were performed on serpentine soils from a) New Caledonia b) Eight Dollar
Mountain Selma OR c) Nickel Mountain Riddle OR d) Harvard Mine Jamestown
6
CA e) Jasper Ridge Biological Preserve Stanford CA and f) Pillikin Mine El Dorado
CA Serpentinites and serpentine soils associated with ophiolite complexes cover ~1 of
the earthrsquos exposed surface and they are correlated to modern and ancient convergent
boundaries The physical and chemical characteristics of serpentine soils are directly
related to the alteration and tectonic history of the serpentinite as well as the soil-forming
factors including the parent material climate biota topography and time Serpentine
soils have Cr concentrations exceeding 200 mg kg-1 however the concentration and the
availability of Cr in these soils are not correlated with the five soil-forming factors
(climate organic material parent material topography and time) or the soil chemistry
Although only Cr(III) was identified in the soil solids Cr(VI) has been identified in soil
solutions (lt30 microg L-1) from serpentine soils in New Caledonia and California Chromite
and Cr-magnetite grains are identified in the soils listed above and these minerals are
responsible for a majority of the Cr enrichment in serpentine soils
Kaumlmmererite is a Cr-chlorite associated with serpentinized ultramafic rocks such
as those collected at Jasper Ridge Harvard Mine and Pillikin Mine Although Cr
concentrations are typically 01-2 Wt in kaumlmmererite it is modally abundant in
serpentinites (up to 5 of the volume of the rock) and a potential source of Cr to be
released into soils and groundwater In Appendix A the standard molar thermodynamic
properties of the Cr-clinochlore end-member kaumlmmererite (Mg5Cr(AlSi3)O10(OH)8) is
estimated using established estimation algorithms This initial study provides the
foundation for submitting a future paper presenting the thermodynamic properties for
major Cr-bearing silicates including fuchsite (Cr-muscovite) uvarovite (Cr-garnet) and
tawmawite (Cr-epidote) and their metamorphic significance related to serpentinized
ultramafic rocks
7
CHAPTER 2
CHROMIUM GEOCHEMISTRY IN SERPENTINIZED ULTRAMAFIC ROCKS
AND SERPENTINE SOILS FROM THE FRANCISCAN COMPLEX OF
CALIFORNIA
ABSTRACT
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of California
produces serpentine soils containing high concentrations of Cr as well as other potentially
toxic elements including Ni Co and Mn Chromium concentrations in serpentine soils
from Jasper Ridge Biological Preserve in the Central Coast Range are as high as 4760
mg kg-1 nearly three times greater than the serpentinite protolith (~1800 mg kg-1)
Chromium-containing minerals within the bedrock include chlorite (~03 Cr Wt )
enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt ) magnetite
(82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate phase
containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
8
INTRODUCTION
Serpentine soils are formed from the weathering of ultramafic rocks and their
metamorphic equivalents (serpentinites Brooks 1987) Measured Cr concentrations in
these soils range as high as 80600 mg kg-1 and are enriched compared to non-serpentine
soils (7 to 221 mg kg-1 Proctor 1992 McBride 1994) The distribution and
concentration of Cr in serpentine soils worldwide is highly variable due to the chemical
composition and mineralogy of the parent material and soil forming factors including
climate biota topography precipitation and time (Hotz 1964 Rabenhorst et al 1982
Schwertmann and Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al
1989 Gough et al 1989 Kaupenjohann and Wilcke 1995) Potential Cr-bearing phases
in serpentine soils have been assessed using extraction experiments and these phases
include organic material silicates secondary oxides and spinels (Schwertmann and
Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994 Kaupenjohann and
Wilcke 1995 Becquer et al 2003) The weathering of these Cr-bearing phases over
geologic time and how these sources govern the distribution of Cr in serpentine soils have
not been evaluated in great detail
Chromium is a group VIB transition metal present in the environment as Cr(III) a
non-hazardous species and micronutrient andor as Cr(VI) a toxin to living cells and a
Class A human carcinogen by inhalation (Daugherty 1992 Cohen et al 1993 James et
al 1997) Chromium(III) is chemically immobile due to forming low solubility
(oxy)hydroxides and stable surface complexes whereas Cr(VI) is a weakly adsorbing
(oxy)anion and thus is mobile in most aquatic environments (McBride 1994 Fendorf
1995 Ball and Nordstrom 1998) Chromite and Cr-magnetite have been cited as
primary sources of Cr in serpentine soils (Rabenhorst et al 1982 Schwertmann and
Latham 1986 Alexander et al 1989 Gough et al 1989) Although Cr is primarily
found in the form of Cr(III) in both ultramafic rocks and serpentinites Cr(VI) has been
measured in serpentine soils in California and New Caledonia (Gough et al 1989
Becquer et al 2003) Due to the abundance of Cr and the potential for Cr(VI) formation
serpentine soils and serpentinites are potential sources of non-anthropogenic Cr
contamination
9
The objectives of this study are to determine the distribution and chemistry of Cr
in naturally Cr-enriched soils using an integrated field analytical experimental and
theoretical approach A wide array of techniques and methods including X-ray
absorption spectroscopy and chemical extractions were used to identify the nature of Cr
and its availability in serpentine soils Serpentinites and serpentine soils at Jasper Ridge
Biological Preserve located near Stanford CA were used to evaluate the geochemical
behavior of Cr These soils cover approximately 50000 m2 and contain high
concentrations of the trace elements Cr Ni and Mn The soils are dominantly Mollisols
with grassland vegetation and they are pedalogically similar to those associated with the
Central Coast Range of the Franciscan Complex in the western United States The main
serpentinite body at Jasper Ridge was exposed approximately 400000 years ago allowing
pedogenesis and Cr cycling to occur over a geologically relevant time period (Burgmann
et al 1994) These soils can then be used as a proxy for evaluating the long-term
behavior of Cr
BACKGROUND
Geologic Background of the Serpentinite at Jasper Ridge
Jasper Ridge Biological Preserve is located in the foothills of the Santa Cruz
Mountains in California and is part of the Central Coast Range of the Franciscan
Complex an Early Cretaceous accretionary meacutelange containing blocks of greywacke
greenstone blueschist and serpentinized ophiolite (Figure 21) Formation
metamorphism and emplacement of serpentinite at Jasper Ridge began approximately
130-150 million years ago during Jurassic subduction along the California continental
margin (Page and Tabor 1967) Thin wedges of ultramafic rock (peridotite) were
detached from the down-going slab of oceanic crust and incorporated into the subduction
meacutelange Subsequent serpentinization of the ultramafic rock in the down-going slab
resulted in the formation of serpentinites now exposed at Jasper Ridge
Rare blueschist blocks (1 to 5 meters in width) present in the serpentinite at Jasper
Ridge attest to the degree of metamorphism and the depths from which portions of the
tectonic meacutelange has been elevated Transformation of greenstone in the meacutelange into
blueschist required high pressures in excess of ~5-7 kbars (15-20 kilometers depth) and
10
temperatures of 200-300 degC (Ernst 1971) The low density of the serpentine (245-250 g
cm-3) allowed upward diapiric migration of the serpentinite body through the denser
enclosing tectonic meacutelange (265 g cm-3) Uplift of the present day Coast Range and
activity associated with the San Andreas fault system have shaped the current topography
and exposure of serpentinite at Jasper Ridge Serpentinite is currently present as a narrow
discontinuous band of pervasively sheared rock located along the ridge top (Figure 21)
The serpentinite is in fault contact with the Whiskey Hill Formation (Cretaceous turbidic
sandstone and mudstone) and tectonically interleaved with the Franciscan Complex
Serpentine Soils Characteristics and Review
Relative to other soil types serpentine soils are characterized by higher
concentrations of Cr Ni Co and Fe lower concentrations of plant nutrients such as Ca
K N and P lower CaMg ratios and characteristic flora and physical properties (Brooks
1987) The pH of serpentine soils ranges from ~4 up to 9 (Brooks 1987 Cole 1992
Proctor 1992) Serpentine soil solutions obtained using tension-free lysimetry have Cr
concentrations between 01 and 32 micromol L-1 in which a majority of the Cr is present as
colloidal material (Gasser and Dahlgren 1994) The inhabitability of these soils for most
plants has been attributed to the imbalance of Ca to Mg (Walker 1954) the deficiency of
plant nutrients (Turitzin 1991) and the elevated concentration of heavy metals
(Robinson et al 1935) however linking the lack of plant productivity of these soils to
toxic levels of Cr has not been fully resolved (Brooks 1987 Kruckeberg 1992) The
unique flora and fauna associated with these soils have been well-documented and Cr
concentrations in the plants range as high as 600 mg kg-1 (Brooks 1987 Gough et al
1989)
Chromium concentrations as a function of depth for serpentine soils in Maryland
(Rabenhorst et al 1982) Tehama County CA (Gough et al 1989) the Klamath
Mountains CA (Hotz 1964) and New Foundland Canada (Roberts 1992) are
illustrated in Figure 22 Chromium concentrations with respect to depth at each location
are highly variable demonstrating both maximums and minimums The concentration of
Cr varies significantly at each location due to the chemical composition of the parent
11
material and other soil-forming factors nevertheless Cr concentrations typically decrease
with depth towards the bedrock
A variety of chemical extractions on serpentine soils suggest a broad host of Cr-
containing soil phases including organic material silicates secondary oxides and spinels
(Schwertmann and Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994
Kaupenjohann and Wilcke 1995) The maximum concentration of Cr extracted from a
serpentine soil reported in the literature is 226 Wt Cr from the New Caledonia laterite
using a 03 M dithionite-citrate bicarbonate (DCB) solution (Schwertmann and Latham
1986) Li and Fendorf (2000) performed extractions on the serpentine soils at Jasper
Ridge using organic acids (citric and oxalic acids) in which a 100 microM citric acid solution
was only able to remove 23 Cr microgg-1 (08 Wt Cr) from a soil containing ~3000 Cr mg
kg-1 This suggests a majority of Cr is bound in phases that are not reactive with these
solvents
Sources of Chromium ProtolithBedrock
Chromium primarily occurs within oxide and silicate phases in ultramafic rocks
and serpentinites Chromite is a common primary (igneous) Cr-bearing spinel in mafic
and ultramafic rocks and often survives low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) Isomorphic substitution of
Al3+ Fe3+ and Ti4+ into the octahedral site and Mg2+ Ni2+ Zn2+ and Mn2+ into the
tetrahedral site is common (Sack and Ghiorso 1991) however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987 Pan and Fleet 1991 Burkhard 1993 Christofides et al 1994 Challis et
al 1995 Sanchez-Vizcaino et al 1995 Deer et al 1996 Mathiesen 1999) Pyroxenes
such as augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) are primary minerals in
ultramafic rocks that incorporate varying concentrations of Cr into their structures (Deer
et al 1996)
Serpentinization is associated with the partial oxidation of Fe(II) present in
olivine and pyroxene forming magnetite (Brooks 1987) Chromium(III) derived from
primary chromite and pyroxene may isomorphically substitute into the Fe(III) site of
magnetite due to size (octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge
12
similarities Hydrothermal and CO2 metasomatism of ultramafic rocks may produce Cr-
silicate minerals including fuchsitemariposite (Cr-mica lt17 Cr Wt ) uvarovite (Cr-
garnet lt20 Cr Wt ) tawmawite (Cr-epidote lt10 Cr Wt ) and kaumlmmererite (Cr-
chlorite lt6 Cr Wt Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Phillips et al 1980 Schreyer et al 1981 Max et al 1983 Kerrich et al 1987 Treloar
1987 Pan and Fleet 1991 Christofides et al 1994 Sanchez-Vizcaino et al 1995)
These minerals common in altered ultramafic rocks have varying concentrations of Cr
that isomorphically substitute for octahedral Al in the silicate structure due to similarities
in charge octahedral preference and octahedral radii (Al3+ = 061 Aring) Any of these
silicates and magnetite are potential sources for Cr in soils produced by the weathering of
serpentinite and ultramafic rocks
Chromium-Bearing Phases in Soils
Soil minerals specifically Fe(III) (oxy)hydroxides and clays provide a host of
structural and surface sites that may sequester Cr(III) released to soil solution by
weathering of Cr-bearing phases in serpentinite and ultramafic rocks Isomorphic
substitution of Cr(III) for Fe(III) in oxides (oxy)hydroxides and amorphous compounds
is expected based on crystallographic considerations synthesis experiments and analyses
of selective dissolution of Fe(III) phases (Schwertmann et al 1989 Gerth 1990)
Additionally Cr(III) readily sorbs on surfaces of Fe(III)-containing oxides such as
magnetite and hematite via inner-sphere complexation and forms low-solubility
precipitates at pH values greater than 4 (Charlet and Manceau 1992 Peterson 1996)
Octahedral substitution of Cr(III) for Al(III) in clays (smectites and vermiculites) is
possible based on size and charge similarities Sorption of Cr(III) on edge and interlayer
sites of clay minerals has also been observed by Ilton and Veblen (1994) Soil organic
matter (SOM) is another possible host for soil Cr in which Cr can be chelated into stable
organomineral and organometallic complexes (Kaupenjohann and Wilcke 1995 Brady
and Weil 1999)
13
MATERIALS AND METHODS
Location and Sample Collection
Jasper Ridge Biological Preserve is a microcosm of accreted terranes located in
the foothills of the Santa Cruz Mountains CA Figure 21a is a soils map reported by
Kashiwagi (1985) depicting the alluvial Tertiary sandstone and Franciscan Complex
soils present at Jasper Ridge The Franciscan Complex soils are divided into two
subgroups those derived from greenstone-graywacke and those from serpentine The
serpentine soils dominantly Mollisols and serpentinite bedrock are present as a narrow
discontinuous band located along the ridge top they do not receive sediment from the
adjacent non-serpentine soils Soils and rocks were collected at Jasper Ridge from 1999
to 2002 and sample locations are shown in Figure 21b A 500 meter transect consisting
of 10 sites extends across the length of the serpentine soil body At each site along the
transect soils and bedrock were collected with respect to depth Site 1 (Diablo clay) is
not derived from serpentinite but rather from chert and greenstone Additionally Site 1
is located down slope of the main serpentinite body and has likely received sediment
from the adjacent serpentine soil Rocks were also collected at several exposed outcrops
located in the serpentinite body
Site JR3 (Figure 21b) was chosen for a detailed geochemical study to identify
factors affecting the geochemistry of Cr in serpentine soils Soil at JR3 is a loamy
magnesic thermic Lithic Haploxeroll The thickness of the soil is approximately 50 cm
deep Grass dominantly Lolium multiflorum comprises 95 of the vegetation at this site
and trees and shrubs are not present
Mineral Identification in Rocks and Soils
Mineral identification was accomplished by electron probe microanalysis on an
automated JEOL 733A electron microprobe operated at 15 kV accelerating potential and
15 nA beam current Calibration was conducted using natural geologic standards Beam
width for analysis was 10 microm Raw counts were collected over 20 s and were constant
with time indicating that elemental drift was negligible Detection limits for probe
analyses are ~002 oxide Wt Additionally backscattering electron (BSE) images of
the rocks and soils were obtained using the microprobe
14
Carbon-coated petrographic thin sections of rock samples were used for
microprobe analyses Soil samples were dried in an oven at 60degC for 24 hours mounted
onto a glass slide using epoxy and heated to 100degC until the epoxy cured An initial cut
of each section using a petrographic saw was made followed by a second infusion of
epoxy to ensure total impregnation Finally each section was cut polished and carbon-
coated for electron microprobe measurements
Soil Chemistry
Soils were analyzed for pH electrical conductivity (EC (deciSiemensm))
particle size ammonium acetate extractable cations organic carbon total nitrogen and
elemental composition The pH was determined using a 11 mixture of soil to deionized
water Electrical conductivity (EC) was determined on 2 mL of a soil solution extract
Particle size of the soil was measured using the hydrometer method as described by Gee
and Bauder (1986) These analyses were performed at the Utah State University
Analytical Laboratory
Ammonium acetate was used to measure the concentration of extractable K Ca
Mg and Na within the soil The procedure involved weighing 2 g of soil and adding 25
mL of ammonium acetate (20 M NH4OAc) followed by mixing on an orbital shaker for
30 minutes The solution was filtered using a Whatman 42 filter paper (pore size 25
microm) Potassium Ca Mg and Na concentrations were analyzed using atomic absorption
spectrophotometry (AAS)
Organic carbon was measured using the Walkley-Black titration method (Mebius
1960) Approximately 05 g of sediment was weighed and reacted with 5 mL of 20 M
potassium chromate (K2Cr2O7) followed by the rapid addition of 10 mL concentrated
sulfuric (H2SO4) acid The bottle was cooled for 30 minutes at room temperature and
diluted with 100 mL of deionized water Next 030 mL of 0025 M ortho-
phenanthroline-ferrous solution was added and the solution titrated to the endpoint using
10M ferrous sulfate (FeSO4)
Total nitrogen of soil samples was obtained using an automated combustion
method as described by Sweeny (1989) Approximately 150 mg of air-dried and sieved
15
(40 mesh) soil was placed in a tin foil encapsulating cup A Leco total nitrogen analyzer
with resistance furnace was then used to determine total nitrogen
Major and minor element concentrations for soil and rock samples were obtained
by completely dissolving the samples using a mixture of hot concentrated nitric
perchloric and hydrofluoric acids Elemental concentrations were measured in the
supernatant using inductively coupled atomic emission spectroscopy (ICP-AES)
Standards were plusmn5 mg kg-1 of their known values These analyses were completed at
Chemex Laboratory (Vancouver Canada) Soil samples were dried and sieved (lt2mm)
before analysis Additionally total element concentrations for gt2mm sand silt and clay
size fractions were obtained for each sample collected at Site JR3
X-ray Diffraction and Clay-Size Mineral Identification
X-ray diffraction (XRD) analyses were conducted on the clay-size fraction
(lt2microm) from Site JR3 using a Siemens Diffractometer D5000 at the University of Idaho
The clay-size fraction was isolated using a suspension method described by Vaniman
(2001) Organic matter and calcium carbonate were removed from the clays using a
concentrated hypochlorite solution (Kunze and Dixon 1986) Crystalline and non-
crystalline iron oxides were removed using a citrate-sodium bicarbonate-sodium
dithionite (CBD) solution (Kunze and Dixon 1986) The remaining phyllosilicate
minerals were analyzed using a series of treatments including 1) magnesium-saturation
(2 M MgCl and 2 M MgAOC) and air drying 2) magnesium-saturation and glycolation
(1 glycerol-ethanol solution) 3) potassium-saturation (1 M KCl) and air drying and 4)
potassium (1 M KCl) saturation followed by heating to 500degC (Whittig and Allardice
1986) X-ray diffraction patterns were collected between 2θ values of 20-350deg and 20-
150deg with a generator potential of 30 kV a generator current of 22 mA (using CuKα
radiation) a Ni filter and a scan speed of 1degmin The software package JADE was used
for XRD data analysis and mineral identification
X-ray Absorption Spectroscopy
In-situ fluorescence yield X-ray absorption near-edge structure (XANES)
spectroscopy at the Cr K-edge (edge inflection point is 5989 eV for Cr(0)) was conducted
16
at the Stanford Synchrotron Radiation Laboratory (SSRL) Stanford CA using
beamlines 4-1 and 4-3 A Si(220) monochromator and a Ge-detector were used and
three scans were collected for each XANES analysis Soil and rock samples from Site
JR3 were prepared by removing the gt2mm fraction and crushed to a fine powder
Samples were loaded into a plastic sample holder and covered with Mylar tape
Elemental mapping of three samples were completed at SSRL on Beamline 11-2
using X-ray absorption spectroscopy (XAS) A Si(220) monochromator with a 6 microm
focusing capillary was used and fluorescence intensity was monitored with a Ge detector
Samples were prepared similar to those used for mineral identification using the electron
microprobe however the thickness of the sample was decreased to lt20 microm
Sequential Extraction Procedure
Sequential extraction experiments were completed on 2 g lt2 mm sieved soils
(Diablo clay from Site 1 at 10 cm depth and serpentine soil from Site 2 at 4 cm depth) in
three steps The exchangeable fraction was obtained using a barium chloride (1mM
BaCl2) solution for 1 h at 25 degC (Kunze and Dixon 1986) The Fe-crystalline oxide
fraction was obtained by reacting the solids with a hydroxylamine-hydrochlorideacetic
acid (10 M NH2OHsdotHCl in 25(vv) HOAc) solution for 6 h at 60 degC (Chester and
Hughes 1967) The silicate fraction was obtained by reacting the solids with
concentrated hydrofluoric acid (10 M HF) for 16 h at 25 degC Trace element
concentrations for each extraction were measured on the supernatant using inductively
coupled plasma optical emission spectrophotometry (ICP-OES) All analyses were
completed in triplicate Additionally sequential extraction experiments were completed
on 2g 106-250 mm sieved ultrasonically cleaned commercial grade chromite (MSDS
No 302-1 WHMIS Class D2A) using the same procedures listed above The
stoichiometric composition of chromite obtained using the electron microprobe is
(Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 Chromite was imaged after each sequential
extraction using a JEOL 5600LV environmental SEM equipped with an EDAX thin-
window energy dispersive X-ray detector
17
RESULTS
Serpentinite Mineralogy
The average serpentinite at Jasper Ridge is composed of serpentine (mainly
lizardite and antigorite 70) chlorite (15) talc (10) magnetite (4) and chromite
(1) based on petrographic and microprobe observations of five samples (JR3E JR6E 5
10 and JR3) however brucite has been reported as a major constituent in related
Franciscan Complex serpentinites (Hostetle et al 1966) Other minerals identified
include olivine augite and enstatite Although the serpentinite mineralogy is relatively
simple the modes and textures of the minerals present in the rock vary significantly from
site to site In Table 21 whole rock serpentinite compositions from Sites JR1 JR3E
JR6E 2 5 10 and JR3 are listed The average Cr concentration of the serpentinites is
~1700 mg kg-1 (tab 1) Iron (~7) Mg (~11) Ni (~3000 mg kg-1) and Co (~100 mg
kg-1) concentrations are also elevated in the serpentinite relative to the chert and
sandstone bedrock adjacent the main serpentinite body (Figure 21b and Table 21)
A range of rock and mineral textures are petrographically evident in the
serpentinite at Jasper Ridge (Figure 23) No foliations or lineations are observed in the
serpentinite (Figure 23a) obtained from the bottom of the soil pit excavated at Site JR3
(Figure 21b) Subhedral chromite and magnetite grains are randomly interspersed
amongst the serpentine matrix (Figure 23a) In serpentinite obtained in outcrop at Site
JR5E (Fig 21b) a lsquocrocodile skinrsquo texture is produced by thin irregular veins of fine-
grained subhedral magnetite separating serpentine grains (Figure 23b) In Figure 23c
numerous crosscutting veins of chlorite occur in orthopyroxene and define the foliation of
the rock in outcrop at Site JR3E (Figure 21b) The numerous crosscutting microscopic
veins random foliation patterns (or the lack thereof) and subhedral oxide grains suggest
the Jasper Ridge serpentinite has undergone extensive physical and chemical alteration
typical of most serpentinites
Microprobe analyses of minerals in the serpentinite demonstrate that the main Cr-
bearing phases are chlorite enstatite augite Mg-Al chromite Cr-magnetite and a
chromite-silicate mixture (defined in the following paragraph) Selected compositions
for these minerals are listed in order of their Cr concentrations in Table 22a Chlorite
contains the least Cr whereas the chromite-silicate mixture is the most enriched with
18
respect to Cr Figure 24 shows correlations of Cr content relative to Al Fe Mg and Si in
order to provide insight into the serpentinite mineralogy and to aid in evaluating trends of
Cr substitution
A fine-grained mixture containing spinel (chromite andor Cr-magnetite) and an
unidentifiable Cr-silicate contains the highest Cr concentration in the serpentinite at
Jasper Ridge (Table 22a shown in Figure 25a and 25b) This fine-grained mixture will
be referred to here as the chromite-silicate mixture (CSM) with chemical properties
similar to those in Table 22a and Figure 24 CSM is common around the rims of
protolith chromite grains and a number of the larger (~05 mm) Cr-magnetite grains in
the serpentinite The microcrystalline phase responsible for contributing Si to these
analyses is possibly chlorite due to similarities of false colors generated by backscattering
electron (BSE) images and chlorite commonly rimming CSM Microprobe analyses of
CSM listed in Tables 22 and 23 are considered as mixtures of several mineral phases
due to the fine-grained texture (lt5 microm) Chromium concentrations range from 12 to 16
Cr Wt in this phase CSM is chemically distinct from the chromite and magnetite as
shown in Figure 24
The chemical compositions of protolith chromite exhibit little to no variation in
any of the serpentinites collected at Sites 2 JR3E JR4E JR5E JR6E JR7E and JR3
(Table 22a and Figure 24) Chromite contains low Fe (~9 Wt ) high Mg (~9 Wt
) and high Al (~26 Wt ) compared to Cr-magnetite The average Cr concentration
for chromite is ~12 Cr Wt Additionally chromite grains are irregular and subhedral
suggesting that they have undergone significant alteration during metamorphism
Chromium concentrations in the magnetite vary depending on the proximity and
degree of alteration of the chromite (Table 22a and Figure 24) Magnetite grains with
less than 2 Cr Wt are common in veinlets in serpentinite not formed adjacent
chromite whereas magnetite with 8 to 12 Cr Wt were formed proximal or adjacent
chromite Chromium-enriched magnetite is also referred to as ferrite-chromite in the
literature (Spangenberg 1943 Burkhard 1993)
Chlorite and pyroxenes contain minor concentrations of Cr (lt1 Wt Table
22a) Aluminum Fe and Mg concentrations varied in the pyroxenes however the
majority of chlorites are relatively similar in composition (Figure 24) Chlorites are
19
more abundant in the serpentinite compared to the pyroxenes and both minerals were
subhedral and randomly distributed over the entire sampling area (Figure 21b)
Backscattering electron (BSE) images of what appeared to be a petrographically
opaque single chromite grain in Figure 23a demonstrates chromite has been
metasomatically altered resulting in the formation of Cr-magnetite Cr-chlorite and CSM
(Figure 25a) Represented compositions for these minerals (indicated by letters in Figure
25a) are listed in Table 23 Magnifying the chromite and CSM (Figure 25a)
demonstrates the fine-grained texture of CSM and an alteration rim (transition zone) is
also seen along the chromite edge (Fig 25b) This alteration rim is depleted in Cr with
respect to the chromite as shown in Figure 25b
In Figures 23a and 25c Ni was identified in three phases in the bedrock
including olivine (F 03 Ni Wt ) serpentine (G and H 03 and 0 Ni Wt ) and Ni-
Fe (I sim55 Ni Wt ) alloy (Figure 25c and 25d) their compositions are indicated by
letters and listed in Table 23 Alteration of the olivine resulted in the formation of lt10
microm Ni-Fe alloys shown in Figure 25c These Ni-Fe alloys form a corona surrounding
the remaining olivine and the illustration in Figure 25d highlights the minerals and
chemical changes
Chromite and Cr-magnetite alteration leading to the paragenesis of CSM and Cr-
chlorite is common in the serpentinites at Jasper Ridge The petrographic image in
Figure 26a and its corresponding BSE image in Figure 26b illustrate Cr-magnetite
(originally protolith chromite) partially altering to CSM and Cr-chlorite The chemical
compositions of each mineral (indicated by letters in Figure 26b) are listed in Table 23
CSM appears to marginally and metasomatically replace Cr-magnetite and is enriched by
~2 Cr Wt Chromium-chlorite is present in microcracks in Cr-magnetite and
encompasses the entire Cr-magnetite and CSM grain Serpentine minerals adjacent the
Cr-chlorite contains no measurable Cr
Distribution of Chromium in Serpentine Soils
Serpentine soils and underlying bedrock were collected at ten sites along a 500
meter transect extending the length of the main serpentine body as shown in Figure 21b
20
The range of Cr concentrations at each site (center) and Cr concentrations with respect to
depth (periphery) are shown in Figure 27 and soil analyses are listed in Table 24
Bedrock Cr concentrations for Sites 2 5 and 10 are also shown in these plots (Fig 27)
Chromium enrichment does not appear to be consistently found at a particular depth or
soil horizon for any of the sites Sites 3 9 and 10 demonstrate a Cr maximum at
approximately 006 m however Sites 1 2 5 and 6 exhibit a Cr minimum at roughly the
same depth The highest soil Cr concentration in the soil along the transect is 4760 mg
kg-1 at Site 3 at 006 meters depth The lowest soil Cr concentration is 1725 mg kg-1 at
Site 10 at 0085 meters depth The average bedrock Cr concentration is ~2000 mg kg-1
(Figure 27 Table 24) It is important to note Site 1 consists of Diablo clay not
serpentine soil however this site is located adjacent and down-slope of the main
serpentinite body From these observations it is apparent that there are no systematic
trends in soil Cr concentrations with respect to the position of the site and depth and soil
samples often contain higher concentrations of Cr with respect to their corresponding
bedrock and to the average Cr concentration of ultramafic rocks (1600 mg kg-1 Green
1972)
Trace metals including Ni Mn and Co were also evaluated in the serpentine soil
for Sites 1 through 10 and the results are listed in Table 24 Nickel concentrations range
from 1590 to 3960 mg kg-1 and do not directly correspond to the Cr fluxes with respect
to depth (Table 24) Nickel concentrations are less than Cr from Sites 1 to 4 From Sites
5 to 10 Ni concentrations exceed those of Cr averaging ~3600 mg kg-1 Unlike Cr Ni
concentrations are equivalent or greater in the bedrock (~3300 mg kg-1) as compared to
the soil (~3200 mg kg-1) Soil Mn concentrations at each site are relatively uniform at
~1800 mg kg-1 with little to no variation with depth (Table 24) Cobalt concentrations
do not exceed 238 mg kg-1 (Table 24) and are relatively uniform (~200 mg kg-1) with
respect to location along the transverse and depth at each site
Major elements including Al Fe Ca and Mg are listed for the serpentine soils
along the transect with respect to depth in Table 24 Aluminum Fe Ca and Mg
concentrations in the soils at each site are similar with exception of Site 1 (Diablo clay)
Aluminum concentrations in the Diablo clay are ~46 whereas the serpentine soils
contain less than 2 Al (Table 24) Iron and Mg concentrations are also low lt65 and
21
lt411 respectively in the Diablo clay compared to Fe (gt8) and Mg (gt6)
concentrations in the serpentine soil (Table 24) Calcium concentrations are low (lt1)
for all the serpentine soils and even the Diablo clay (Table 24) With a broad
understanding of the chemical characteristics of the serpentine soil at Jasper Ridge Site
JR3 (Figure 21b) was chosen for a detailed chemical study to examine all the factors
governing the chemistry of Cr in the soil
Serpentine Soil Chemistry and Mineralogy at Site JR3
Soil chemistry
Chemical properties (pH electrical conductivity organic matter nitrogen and
ammonium acetate extractable cations) and textures for four soil samples from Site JR3
are listed in Table 25 The pH of the soil is near neutral (pH from 671 to 698) and
increases slightly with respect to depth Electrical conductivities (EC) in the soil ranges
from 03 to 09 dSM a range indicative of relatively few dissolved salts andor major
dissolved inorganic solutes The organic carbon content of the soil is 32 at the surface
and decreases with depth to 10 near the bedrock Total nitrogen for the samples ranged
from 003 to 064 Ammonium acetate extractable Ca K Mg and Na are listed in
Table 25 The extractable concentration of both Ca and K decrease with depth whereas
there is no apparent trend associated with the concentrations of Mg and Na The
serpentine soil textures are dominantly clay loams however the texture is a sandy clay
loam near the bedrock
Whole soil samples and size fractions including gt2 mm sand (2-002 mm) silt
(002-0002 mm) and clay (lt2 microm) size fractions with respect to depth were chemically
digested and analyzed for total trace metal concentrations shown in Figure 28 and listed
in Table 26 to determine if the abundance of Cr was linked to a particular size fraction
(the percent of sand silt and clay size fractions are reported in Table 25) Additionally
Ni Mn and Co concentrations with respect to depth and size fraction are also presented in
Figure 28 and Table 26 Total and gt2mm size fraction Cr concentrations are similar at
the top and bottom of the soil profile ranging from 3297 to 5976 Cr mg kg-1 Chromium
concentrations exceed detections limits (gt10000 mg kg-1) for the sand-size fraction It is
important to note that the sand-size fraction corresponds to the size of most chromite
22
grains observed in the serpentinite Chromium in the clay fractions are low (~1000 mg
kg-1) compared to the other size fractions however these concentrations are still enriched
compared to non-serpentine soils
In Figure 28 and Table 26 Ni concentrations are enriched in the clay-size
fraction with values ranging from 2840 to 2920 mg kg-1 The silt-size fractions have the
lowest Ni concentrations (~1800 mg kg-1) Manganese concentrations are slightly
elevated relative to total Mn in the sand-size fraction with values ranging from 1728 to
2701 mg kg-1 The clay-size fraction is the least Mn enriched size-fraction with
concentrations at ~1200 mg kg-1 Cobalt concentrations are elevated in the sand-size
fraction with values ranging from 159 to 241 mg kg-1 and concentrations in the gt2mm
silt and clay-size fractions are similar to the total (~150 mg kg-1)
Soil Cr-bearing minerals
Chromium-bearing minerals in the soil identified using the electron microprobe
include chromite Cr-magnetite and CSM Chromium concentrations in soil chromite
Cr-magnetite and CSM are graphically shown in Figure 29 as a function of Al Fe Mg
Si and Al+Fe in Wt Representative analyses of the Cr-bearing soil minerals are listed
in Table 22b Compositions of soil chromite in Figure 29 are chemically distinct from
those identified in the protolith in Figure 24 Chromium concentrations in soil chromites
range from 22 to 37 Cr Wt containing minor concentrations of Al and Mg Values for
chromite in the serpentinite bedrock (Table 22a and Figure 24) and the FeCr2O4 end-
member are also shown in Figure 29 The dashed lines in Figure 29 connect Jasper
Ridge bedrock chromite (Figure 24) and stoichiometric FeCr2O4 where it can be seen
that compositions of soil chromite lie close to these tie lines Overall soil chromite
demonstrates a decrease in Al and Mg and an increase in Fe relative to Cr compared with
protolith chromite Soil Cr-magnetites contain between 7 and 20 Cr Wt and are
enriched in Cr compared to Cr-magnetites identified in the bedrock (Figure 24) Cr-
magnetite however reveals a decrease in Fe relative to Cr Minor concentrations of Si
(lt2 Wt ) in the soil chromite and Cr-magnetite may represent slightly altered rims or
contamination from sample preparation Chromium concentrations in CSM maintain a
consistent Cr concentration of ~11 Cr Wt similar to those found in the bedrock (~14
23
Cr Wt ) Soil CSMs with compositions between 8 and 12 Mg Wt are more than
likely associated with analyses containing more microcrystalline chlorite compared to
soil CSMs with less than 2 Mg Wt Soil CSM appears be less enriched with Si (lt14
Wt ) relative to those identified in the serpentinite (15 to 27 Wt )
The size of the Cr-bearing minerals varied significantly for each sample ranging
from 10 microm up to 2 mm The majority of chromite grains in the soil correspond to the
sand-size fraction and also vary from euhedral to subhedral with no discernable pattern
The sum of cations based on four oxygens for the chromite and Cr-magnetite analyses
range between 297 and 301 indicating these minerals are spinels not (oxy)hydroxides
Compositions obtained for chromite and Cr-magnetite may represent alteration rims
where weathering of the mineral in the subsurface environment has been most active
BSE images of chromite and Cr-magnetite were not able to definitively identify whether
these analyses in fact represent alteration rims
Mineralogy of the Clay-Size Fraction
The clay-size fraction in the serpentine soil at Site JR3 was evaluated with respect
to depth using XRD and specific chemical and heat treatments The d-spacings (Aring) and
their relative intensities for each soil and treatment with respect to depth are listed in
Table 27 Figure 210 shows the XRD analyses for clay sample JR3515 (depth 5-15
cm) and these analyses are almost identical to the other clays examined at Site JR3
Additionally Table 28 from Whittig and Allardice (1986) is provided to interpret the X-
ray diffraction analyses in Table 27 For example a d-spacing of ~14Aring is present in all
the treatments indicating the presence of chlorite in the clay-sized fraction Using an
XRD analysis program the clay-size fraction was determined to be composed of
smectite vermiculite serpentine minerals (lizardite and antigorite) and chlorite
(clinochlore and Cr-clinochlore) Smectite is the most abundant mineral in the clay-size
fraction whereas Cr-chlorite is the least abundant Overall the d-spacings and relative
intensities are similar despite the treatments and changes in depth (Table 27) This
indicates that the mineralogy of the clay-size fraction does not vary significantly in the
soil column
24
Chromium concentrations in the clay-size fractions are ~1010 mg kg-1 (Table
26) Smectite and vermiculite are possible Cr-bearing clay minerals in which isomorphic
substitution of Cr(III) could occur in the dioctahedral sheet Serpentine minerals
(lizardite and antigorite) are not likely hosts of Cr based on crystallographic
considerations and the absence of Cr in analyzed serpentine minerals in the bedrock
(Table 23) Clinochlore may also be a host for Cr in the inner layer octahedral sheet
Chromium-clinochlore is the only Cr-containing mineral identified in the clay-size
fraction using XRD and Cr-clinochlore (Cr-chlorite) is also an abundant mineral
identified in serpentinite protolith (Tables 22 and 23 Figure 26)
XAS of chromium in relation to depth
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in serpentine soils and bedrock at Site JR3 The XANES (X-ray absorption near-edge
structure) spectra demonstrate the absence of Cr(VI) in the soil column at Site JR3
(Figure 211) based on the lack of a pre-edge feature The detection limit for Cr(VI)
using this method is approximately ~20 mg kg-1 Ultimately XAS analyses demonstrate
Cr(VI) is not detected in any of the soil solids
X-ray microprobe analyses of soil and rock samples (0-5cm depth 5-10cm depth
and serpentinite) at Site JR3 were processed to map the relative concentrations of Fe and
Cr in which areas of elemental enrichment are identified by a violet color areas of
depletion by a blue color (Figure 212) Areas of Cr and Fe enrichment in Figure 212 are
localized with well-defined boundaries corresponding to grain edges and these elements
are not correlated in a well-defined trend Minerals assigned to these areas of Cr and Fe
enrichment were made based on petrographic and microprobe observations on the same
thin section For example high Fe and no Cr is magnetite whereas high Fe and high Cr
is either chromite or Cr-magnetite Chromite magnetite and Cr-magnetite appear to be
present in the serpentine soil at a depth between 0 to 5 cm (Figure 212A) In Figure
212b CSM is a possible mineral detected in the soil (depth 5-10 cm) and the presence
of magnetite is clearly seen due to the localized enrichment of Fe and no Cr Iron and
Cr spectra for the serpentinite were much more heterogeneous as compared to the soil
samples and Cr concentrations appear be more scattered (Figure 212c) These X-ray
25
microprobe analyses provide a complimentary method for examining Cr geochemistry
(including Cr(III) and Cr(VI)) in naturally occurring soil and rock solids however
Cr(III) is the only detected valence state in serpentine soils and rocks at Jasper Ridge
Sequential Extractions
Sequential extractions were completed on 1) Diablo clay (Inceptisol) from Site 1
(Figure 21b) at a depth of 10 cm 2) serpentine soil (Mollisol) from Site 2 (Figure 21b) at
a depth of 4 cm and 3) commercial grade chromite (106-250 microm) and these results are
shown in Figure 213 and listed in Table 29 Chromium extracted from the
exchangeable fraction is minimal (lt01 mg kg-1) for all the samples (number 1 in Table
29 Fig 213) Minor Cr concentrations are associated with the Fe-crystalline oxide
fraction in the Diablo clay (19 Cr mg kg-1) and serpentine soil (78 Cr mg kg-1) whereas
chromite only released 05 Cr mg kg-1 (number 2 in Table 29 Figure 213) Increased
concentrations of Ni and Fe are present in the Fe-oxide fraction of the Diablo clay (161
Ni mg kg-1 and 4474 Fe mg kg-1) and serpentine soil (5138 Ni mg kg-1 and 5138 Fe mg
kg-1 Table 29) Chromium extracted from the Diablo clay serpentine soil and chromite
in the silicate fraction is 14 31 and 14 Cr mg kg-1 respectively (number 3 in Table 29
Figure 213) The sum of extracted Cr Fe Ni and Mn concentrations (number 4 in
Table 29) are low with respect to total metals (number 5 in Table 29) In fact only 07
12 and 00 Cr was removed from the Diablo clay serpentine soil and chromite
respectively (number 6 in Table 29)
Chromite was extremely resistant to all the extractions and less than 001 of the
total Cr was removed (Table 29) Scanning electron microprobe (SEM) images of the
chromite after each extraction are shown in Figure 214 Chromium concentrations
released from the chromite (Table 29 and Figure 214a) increased with each sequential
extraction experiment starting with the exchangeable fraction (Figure 214b) progressing
into the Fe-crystalline oxide fraction (Figure 214c) and ending with the silicate fraction
(Figure 214d) The surface of the chromite became smoother after each extraction and
crystal imperfections in defect sights located along the edges and corners progressively
decreased (Figure 214) Chromite does not appear to be an easily accessible source of
Cr based on these extraction experiments
26
DISCUSSION
Serpentinite at Jasper Ridge has undergone significant physical and chemical
alteration since its formation and emplacement off the California continental margin
Despite being subjected to high pressures (~5-7 kbars) and elevated temperatures (200-
300degC) trace element concentrations within the serpentinite remain elevated Jasper
Ridge serpentinites are enriched with Cr Co Fe Mg Mn and Ni compared to chert and
sandstone located adjacent the main serpentinite body (Table 21 and Figure 21b) Over
the past 400000 years Cr-bearing minerals in the serpentinite have been weathering and
accumulating in the soil During this time Cr concentrations in the soil have
progressively increased relative to Cr-concentrations in the serpentinite protolith This Cr
enrichment is directly linked to specific weathering-resistant minerals identified in the
soil
Chromium in Serpentinites
Three minerals in the bedrock at Jasper Ridge contain appreciable concentrations
of Cr Cr-magnetite (82-103 Cr Wt ) chromite (~108 Cr Wt ) and CSM (~133
Cr Wt ) Cr-magnetite and CSM were derived from the alteration of primary igneous
chromite CSM or Cr-spinels with high concentrations of Si have been noted by
(Burkhard 1993) as a Cr-spinel containing a possible microscopic silicate phase Our
analyses support this result in which the microcrystalline silicate is probably Cr-chlorite
due to its common occurrence with CSM (Figure 26) Definitive identification of this
microcrystalline mixture was not possible due to size of the crystals relative to the spot
size of the electron microprobe Chlorite and pyroxene contain minor concentrations of
Cr (~1 Wt ) and these were the only identified Cr-containing silicates in the
serpentinite Minor concentrations of Cr observed in some chlorites are directly related
to the metasomatic alteration of chromite and Cr-magnetite (Figure 26) Other common
metasomatically-derived Cr-bearing silicates such as fuchsitemariposite and uvarovite
are not present possibly due to the lack of aqueous K+ and Ca+2 in the alteration fluids
Overall the chemical compositions of these protolith minerals (Figure 24) demonstrate
only minor chemical variations allowing comparative analysis with Cr-bearing soil
minerals
27
Chromium in Serpentine Soils
Cr-bearing soil minerals identified using the electron microprobe include
chromite Cr-magnetite and CSM and these phases are chemically distinct from those in
the bedrock (Figure 24 and 29) The highest concentration of Cr (41 Wt ) in the soil
is in chromite with a composition approaching the FeCr2O4 end-member This represents
a ~80 increase in the Cr concentration relative to bedrock chromite (see Figure 215)
Cr-magnetite also demonstrates Cr enrichment (up to 20 Cr Wt ) with respect to
magnetite identified in the bedrock (~10 Cr Wt ) (Figures 24 and 29) The textures of
the soil Cr-bearing spinels exhibit sharp irregular edges possibly due to extensive
weathering CSM in the soil is similar to those found in the bedrock with Cr
concentrations ranging from 11 to 14 Cr Wt however Si concentrations in soil CSMs
(8 to 14 Wt ) are noticeably less than those measured in the serpentinite (15 to 20 Wt
) The weathering of CSM might preferentially release Si relative to other elements in
its structure Cr-silicates (chlorite and pyroxenes) were not identified in the soil possibly
due to complete weathering of these phases or they were too fine-grained for microprobe
analyses however Cr-clinochlore was identified in the clay-size fraction using XRD
Elemental maps shown in Figure 212 corroborate electron microprobe observations by
revealing well-defined zones of Cr and Fe enrichment in both the soils and serpentinite
These areas of Cr and Fe enrichment are either chromite Cr-magnetite CSM or
magnetite Overall these maps reveal Cr is not homogeneously distributed in the soil
confirming Cr is in specific mineral phases Additionally the observed abundance of
chromite Cr-magnetite and CSM in the soils indicates that these minerals are probable
sources for a majority of the Cr enrichment identified in the soils
Soil Spinels
Soil chromite and Cr-magnetite appear to have undergone chemical alteration
enriching the minerals with respect to Cr In Figure 215 compositions of bedrock
chromite and end-member FeCr2O4 are shown together with tie-lines like those in figure
8 connecting both minerals Reaction progress in Figure 215 is based on the assumption
that the serpentinite chromite (0 reaction progress) is proceeding toward FeCr2O4
(100 reaction progress) in the serpentine soil Soil chromites from Jasper Ridge are also
28
plotted in Figure 215 where Cr concentrations are assumed to vary linearly between the
bedrock chromite and chromite end-member which defines the reaction progress of the
mineral These tie-lines correlate extremely well to the chemical composition of
chromites identified in the soil For example chromite containing ~30 Cr Wt will
have ~15 Fe Wt ~13 Al Wt and ~5 Mg Wt similar to soil chromite in Figure 29
The retention of Cr in chromite shown in Figure 215 has significant implications towards
the availability and cycling of Cr in serpentine soils
One possible mechanism for the chromite composition trends illustrated in
Figures 29 and 215 is incongruent dissolution where Cr is retained relative to other
elements Sequential extraction experiments (Figure 213 and Table 29) support this
idea in which Fe is more easily released than Cr (1096 Fe mg kg-1 to 19 Cr mg kg-1)
from chromite Unfortunately Mg and Al were not monitored during the experiments
From these sequential extraction experiments and chromite analyses (Figure 215) Al
and Mg have been preferentially released from the chromite structure compared to Fe and
Cr Additionally the alteration of chromite in the serpentinite and the soils are analogous
to one another Chromite in the serpentinite retains a majority of its Cr compared to Mg
and Al when being altered to CSM by fluids at elevated pressures and temperatures
(Figure 26 and Table 22) Under ambient soil conditions chromite is once again
retaining Cr relative to Mg and Al causing what appears to be an increase in Cr
concentration Chromium is not easily released from chromite whether under high
pressures and temperatures or in the soil
Crystal field stabilization energies (CFSE) and octahedral site preference energy
parameters (OSPE) provide a means to explain the incongruent dissolution or Cr
preservation in the Cr-spinels CFSE is a measure of the net energy of stabilization
gained by a metal ions nonbonding d electrons as a result of complex formation This
measure also indirectly describes the energy required to remove an ion from a crystal
structure The octahedral CFSE for Cr(III) is ndash2247 kJmole and the OSPE (the
difference between the tetrahedral and octahedral CFSE) is ndash1578 kJmole (Dunitz and
Orgel 1957 McClure 1957) These values are the most negative for any transition
element and indicating the presence of Cr(III) in an oxide increases the crystalrsquos stability
compared to an analogous crystal lacking Cr(III) Chromium(III) would be among the
29
last ions to removed from the spinel structure compared to other transition elements such
as Fe2+and Mn2+
Soil Chemistry related to Chromium Availability
In Table 25 the chemical properties of the soil at Site JR3 do not appear to be a
factor when evaluating Cr in the soil The pH of the soil did not vary considerably with
depth Additionally the near neutral pH would not have a significant impact accelerating
Cr-bearing mineral dissolution Organic matter within the soil was elevated despite the
low plant productivity of the serpentine soil The role of organic matter could be a
significant factor such as the production of organic acids promoting the dissolution of the
Cr-bearing minerals however Cr concentrations near the surface (where organic matter
was the greatest) are not significantly greater or less than those deeper in the soil profile
The relative concentrations of ammonium acetate extractable Ca K Mg and Na are
characteristic of serpentine soils and did not provide any new insight Finally Cr(III)
was the only identifiable valence state of Cr based on X-ray absorption spectroscopy
(Figure 211) which is what should be expected based on the Cr-bearing minerals and the
pH of the soil
Extraction experiments summarized in Table 29 and Figure 213 demonstrate
~99 of the Cr in a serpentine soil is retained in highly resistant phases at Jasper Ridge
The chromite standard used in these experiments contributed negligible concentrations of
Cr (lt2 mg kg-1 total) for all of the sequential extraction experiments supporting earlier
evidence that Cr-spinels are retaining Cr and are potentially responsible for a majority of
the total Cr enrichment in the serpentine soil At Jasper Ridge Cr extracted from the
serpentine soil (31 mg kg-1) is likely the result of the Cr-silicates compared to the Cr-
spinels Low Cr-containing minerals (possibly chlorite enstatite augite and clay
minerals) appear to be the most probably source of Cr to soil solutions and vegetation of
the serpentine soil Approximately 30 of the serpentine soil is composed of clay-sized
minerals and this size-fraction has ~1000 Cr mg kg-1 The mineral responsible for Cr
enrichment in the clay-size fraction is more than likely Cr-chlorite identified using XRD
Chromium-chlorite in the soils suggests that this mineral is potentially the largest silicate
contributor of Cr in the serpentine soil
30
Interpreting the enrichment and behavior of Cr in serpentine soils such as in
Figure 27 can be accomplished based on the abundance Cr-spinels observed in the soil
Small grains of chromite Cr-magnetite and even CSM in the soil have the potential to
appreciably impact Cr concentrations Table 210 demonstrates the chromite masses
required to change the Cr concentration in a 1 kg sample of soil The highest Cr
concentration at Jasper Ridge was identified at Site 3 with a measured value of 4760 mg
kg-1 which means only ~1 of the soil sample (Table 210) needs to contain chromite
Additionally a minimal amount (lt011 of the sample as shown in Table 210) of
chromite is necessary to change the concentration by 500 mg kg-1 which could easily
cause the sharp Cr increases and decreases noted in Figure 27 This agrees with our
observations of chromite abundance in the serpentine soils Additionally a majority of
the chromite grains identified in the bedrock are between 2 and 002 mm which
correspond directly to the elevated Cr concentrations (gt10000 mg kg-1) in the sand-size
fraction of the serpentine soil (Table 26) Even the ~15000 Cr mg kg-1 values in soils
measured by Hotz (1964) in Figure 22 only require ~2 of the sample of the sample to
be chromite A few grains of chromite possess the ability to significantly impact
measured Cr concentrations in soil samples
Chromium and Nickel in Serpentine Soils
Chromium and Ni have often been correlated to one another due to their common
enrichment in serpentine soils (Hotz 1964 Rabenhorst et al 1982 Schwertmann and
Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al 1989 Kaupenjohann
and Wilcke 1995) A majority of the Ni in the Jasper Ridge serpentine soil originated
from olivine (Figure 25c and 25d Table 23) Chemical alteration of olivine during
serpentinization resulted in the mineral serpentine (Coleman and Jove 1992) in which
minor concentrations of Ni (lt037 NiO Wt ) substituted into its structure Continued
alteration under highly reducing conditions and the incompatibility of Ni in the serpentine
structure resulted in the formation of Ni-Fe alloy (Figure 25c) This type of alteration is
observed in many of the rocks at Jasper Ridge where the Ni-Fe alloys are ~10 microm in size
Olivine in the rocks and soil appear to be less resistant to weathering compared to
chromite therefore Ni is possibly more available to bacteria and vegetation in serpentine
31
soils Sequential extractions (Figure 213 Table 29) support these observations in which
a greater percentage of total Ni was extractable compared to the Cr in the soils Although
Cr and Ni are enriched in serpentine soils the availability and behavior of these two
elements are significantly different
CONCLUDING REMARKS
Chromium is predominantly present in primary chromite Cr-magnetite and a
chromite-silicate mixture (CSM) in both the bedrock and soil despite weathering over
geologic time Sharp variations in soil Cr concentrations with respect to depth are caused
by the presence or absence of chromite Cr-magnetite andor CSM Chromium-spinels in
the soil have undergone incongruent dissolution possibly conserving Cr(III) with respect
to Al3+ and Mg2+ Chromite is highly resistant to weathering and it is more than likely
not contributing bioavailable Cr Although Cr-spinels are a major source of Cr Cr-
silicates in the protolith (chlorite enstatite and augite) and soil (chlorite
montmorillonite and vermiculite) afford the most accessible Cr source identified Based
on our observations and analyses chlorite is the most likely Cr source of these silicate
minerals Finally Cr(VI) was not detected in any of the soil and rock samples
ACKNOWLEDGEMENTS
We thank Jasper Ridge Biological Preserve for allowing us the opportunity to
examine the serpentine soils Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund NSF Grant EAR-9902859 (DKB) GSA Graduate
Student Research Grants Office of Teaching and Learning (SF) at Stanford University is
gratefully acknowledged We appreciate the technical support from the staff of the
Stanford Synchrotron Radiation Laboratory from Bob Jones for assistance in microprobe
analyses of the serpentinites and serpentine soils and from Ben Bostick Travis Horton
and Theresa Barber for field and lab assistance
32
Table 21 Whole rock elemental concentrations of serpentinites (Figure 21b) chert and sandstone
-----------------------Location of Serpentinitesdagger----------------------- Elemental Concentrations JR1 JR3E JR6E Site 2 Site 5 Site 10 JR3
ChertDagger SandstoneDagger
Al 02 08 03 027 22 107 024 079 027 Ca 026 007 005 003 046 013 003 001 lt001 Co mg kg-1 101 99 100 99 155 217 97 1 lt1 Cr mg kg-1 580 2280 1510 1840 1945 2090 1465 10 lt1 Fe 303 71 609 581 76 1106 512 041 024 K 001 031 099 lt001 lt001 001 lt001 029 008
Mg 122 134 154 1336 524 804 112 014 034 Mn mg kg-1 495 800 155 730 1375 1860 460 110 85 Na 181 006 093 lt001 lt001 lt001 lt001 001 001 Ni mg kg-1 2037 1545 3321 3510 3020 3960 2740 20 9
Total digestion ICP-AES analyses daggerSerpentinites are dominantly composed of the minerals serpentine chlorite talc and spinels DaggerSee Figure 21b for sample locations
33
Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in serpentinite rock at Jasper Ridge
Chlorite Enstatite Augite Chromite Cr-Magnetite ChromiteSilicatedagger SiO2 3738 5463 5222 000 024 1053 Al2O3 552 477 516 5059 182 395 NiO 001 01 00 03 00 052 FeODagger 762 614 239 1446 7919 5042 MnO 009 015 009 042 140 182 Cr2O3 024 060 099 1584 1605 1939 MgO 3337 3305 1560 1869 024 605 CaO 023 050 2348 000 000 019 Na2O 000 001 050 000 001 008 K2O 000 000 001 000 000 011 Total 8446 9994 10048 10027 9894 9307
Formula Unit Compositions
Si 370 094 189 000 001 --- IVAl 030 006 011 000 000 --- VIAl 035 004 011 157 008 --- Ni 000 000 000 001 000 ---
Fe+2 063 010 007 030 232 --- Mn 001 000 000 001 004 --- Cr 002 001 003 033 048 --- Mg 493 085 084 073 001 --- Ca 002 001 091 000 000 --- Na 000 000 004 000 000 --- K 000 000 000 000 000 --- O 1400 300 600 400 400 ---
Electron microprobe analyses daggerChromitesilicate is also referred to as CSM DaggerFe+3 is present however all conversions are based on Fe+2
34
Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in serpentine soil at Jasper Ridge
CSM Chromite Cr-Magnetite SiO2 1940 057 267 Al2O3 559 1059 381 NiO 014 001 004 FeOdagger 2629 2259 5244 MnO 155 154 419 Cr2O3 1878 5577 3237 MgO 1668 916 288 CaO 015 001 005 Na2O 006 001 001 K2O 0 008 009 Total 8864 10033 9855
Si --- 002 010 IVAl --- 042 017 VIAl --- 000 000 Ni --- 000 000
Fe+2 --- 063 148 Mn --- 004 013 Cr --- 147 096 Mg --- 046 016 Ca --- 000 000 Na --- 000 000 K --- 000 000 O --- 400 400
Electron microprobe analyses dagger Fe+3 is present however all conversions are based on Fe+2
35
Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25 and 26 A B C D E F G H I J K L Chromite
Chromite CSM Magnetite Chlorite Olivine Serpentine Serpentine Ni-Fe-Metal Cr-Magnetite CSM Chlorite
SiO2 0 0 1053 0 3407 4046 3688 4074 591dagger 021 1143 2855Al2O3
5059 5162 395 034 1114 000 397 085 054 165 980 2293NiO 028 031 052 0 004 037 037 0 5302dagger 0 0 0
FeODagger 1446 1452 5042 9992
1220 901 262 453 1753dagger 8169 3498 806MnO 042 044 182 0 008 010 0 0 0 107 150 00Cr2O3 1584 1576 1939 125 071 001 0 0 0 1369 1905 039MgO 1869 1869 605 003 292 4895 3848 3755 064 1217 2828CaO 0 0 019 0 0 0 0 0 0 0 0Na2O 0 0 0 0 0 0 0 0 0 0 0K2O 0 0 0 0 0 0 0 0 0 0 0H2O --- --- --- --- 1200 --- 1600 1600 --- --- 1200Total 10027 10136 9307
10164 9944 9891 9834 9972
9897 8897 10028
Formula Unit CompositionSi
000 000 --- 000 328 100 184 200 001 --- 269IVAl 000 000 --- 000 072 000 000 000 000 --- 131VIAl 157 159 --- 002 058 000 023 005 007 --- 123Ni 001 001 --- 000 000 001 001 000 000 --- 000
Fe+2 030 029 --- 294 084 019 011 019 239 --- 059Mn 001 001 --- 000 000 000 000 000 003 --- 000Cr 033 032 --- 004 006 000 000 000 041 --- 003Mg 073 073 --- 000 419 180 285 274 004 --- 397Ca 000 000 --- 000 000 000 000 000 000 --- 000Na 000 000 --- 000 000 000 000 000 000 --- 000K 000 000 --- 000 000 000 000 000 000 --- 000O 400 000 --- 000 1400 400 700 700 400 --- 1400
36
Electron microprobe analyses dagger elemental metal Dagger Fe+3 is present however all conversions are based o
36
2321000---
10025
591000054
53021753000000
2321000000000000
n Fe+2
Table 24 Bulk soil elemental concentrations for soils and bedrock collected along the transect
(Fig 21b)
Sites Samples Depth
(meters) Cr
(mg kg-1) Ni
(mg kg-1) Mn
(mg kg-1) Co
(mg kg-1) Al
() Fe
() Ca ()
Mg ()
COI03 0015 2420 1590 1325 118 464 565 065 352 COI36 0045 2370 1630 1600 138 467 55 063 356 COI69 0075 2160 1755 1670 144 471 6 064 365 COI912 0105 2420 1785 1740 151 496 615 067 384
1
COI1215 0135 2570 1985 2340 195 506 649 069 411 COII02 001 3710 2400 1665 169 294 702 100 768 COII25 0035 3600 2290 1625 162 275 673 093 743 COII58 0065 2910 2310 1550 159 25 672 087 771
COII811 0095 3520 2350 1595 158 276 693 093 766 2
COIIBedrock 012 1840 3510 730 99 027 581 003 1336 COIII14 0025 3750 2930 1655 190 178 897 062 762 COIII47 0055 4760 3050 1850 178 178 94 056 805
COIII710 0085 4380 3410 1930 216 216 1044 07 797 3 COIII1013 0115 4320 3380 1955 201 213 1061 063 626 COIV01 0005 3710 2970 1915 214 209 909 053 781 4 COIV15 003 3540 2710 1725 195 181 801 043 792 COV14 0025 2620 2350 1330 150 246 651 063 526 COV47 0055 1775 2710 1520 165 252 766 059 517
COV710 0085 2000 3090 1830 195 225 737 057 583 5 COVBedrock 0115 1945 3020 1375 155 22 76 046 524
COVI14 0025 3040 3720 1855 224 084 1152 015 736 COVI47 0055 2400 3610 1785 218 085 1138 014 68 COVI710 0085 2960 3800 1850 232 093 1232 015 714 6
COVI1013 0115 2800 3860 1800 224 085 1164 014 739 COVII14 0025 2620 3550 1925 229 083 1128 015 718 COVII47 0055 1975 3590 1805 216 084 1085 014 655 7 COVII712 0095 2080 3600 1840 220 07 1105 011 741 COVIII14 0025 2130 3830 1990 238 085 115 017 705 COVIII47 0055 2550 3720 1865 228 085 1099 017 714 8
COVIII712 0095 2240 3910 1860 229 076 1112 014 828 COIX14 0025 2470 3480 1890 227 104 1201 023 579 COIX47 0055 3000 3830 1595 205 084 108 013 842 9 COIX713 01 2720 3710 1845 221 111 1208 017 551 COX14 0025 2810 3740 1760 218 129 114 017 675 COX47 0055 2190 3910 1835 224 131 1206 017 702
COX710 0085 1725 3370 1595 192 11 1061 014 635 10 COXBedrock 0115 2090 3960 1860 217 107 1106 013 804
Total digestion ICP-AES analyses
37
Table 25 Chemical and physical properties of the soil at Site JR3
NH4OAc-extractable ---Hydrometer---Ca K Mg Na
Sand Silt ClaySample Depth (cm) pH EC
dSm
Organic Matter ------
LECO Nitrogen ------ ---------mg kg-1---------
Texture
JR305 25 671 04 32 012 505 2366 3291 355 36 31 33 Clay LoamJR3515
10 683 09 19 064 4205 8176 3343 2392 36 30 34 Clay LoamJR31530 225 688 03 12 003 3725 4903 3750 2305 36 33 31 Clay LoamJR33045 375 698 07 10 003 2873 3665 3388 2549 46 24 30 Sandy Clay Loam
See Methods Section for more details
38
38
Table 26 Selected trace metal concentrations for each size fraction versus depth at Site JR3
Depth (cm) Total gt2mm Fraction Sand Silt Clay Sample
------------------------Chromium (mg kg-1)------------------------ JR305 25 4906 5086 gt10000 4565 988 JR3515 10 5976 4678 8284 4000 1115 JR31530 225 5222 4206 gt10000 4133 980 JR33045 375 3477 3297 gt10000 4571 956
------------------------Nickel (mg kg-1)------------------------------ JR305 25 1935 2050 1660 1305 2920 JR3515 10 2180 2050 1980 1325 2920 JR31530 225 2190 2120 1835 1370 2880 JR33045 375 2090 2140 1740 1285 2840
------------------------Manganese (mg kg-1)-------------------------- JR305 25 1319 1571 1728 1301 1210 JR3515 10 1664 1601 2701 1471 1254 JR31530 225 1639 1626 2217 1599 1231 JR33045 375 1222 1226 1781 1405 1170
-----------------------Cobalt (mg kg-1)-------------------------------- JR305 25 128 146 159 134 131 JR3515 10 155 145 241 147 132 JR31530 225 152 147 194 162 128 JR33045 375 125 124 159 141 122
Total digestion ICP-AES analyses
39
Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3
JR305CLAY Depth 0-5cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1439 100 1833 100 1274 100 1391 100 731 179 1432 431 732 645 1047 105 71 216 941 163 71 41 838 838 474 101 847 141 475 257 726 726 356 125 73 271 365 264 313 83 712 248 355 299
473 143 313 257 364 131 355 193
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1845 1000 1268 1000 1388 1000 852 68 1447 387 724 987 1010 859 733 264 944 143 703 679 836 281 712 247 903 139 471 282 723 576 475 109 850 83 362 244 365 73 733 426 352 321 356 117 712 291 313 154 314 63 475 122
364 170 355 196
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1724 1000 1297 1000 1396 1000 842 45 1397 447 741 957 1175 708 730 318 923 176 715 686 1037 944 710 255 840 118 477 271 837 139 475 97 717 465 367 400 726 375 357 133 468 182 357 386 313 91 362 194 315 286
352 229 JR33045CLAY Depth 30-45cm
Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity
1447 1000 1869 1000 1268 1000 1393 627 842 44 1476 287 737 843 1049 1000 730 207 954 107 715 539 837 190 710 182 915 107 479 157 723 111 475 69 860 61 366 303 364 69 740 332 356 292 356 74 717 256 315 315 313 57 478 111
367 141 357 115
See Methods Section for details
40
Table 28 X-ray diffraction d-spacings for common silicates from Whittig and Allardice (1986)
Diffraction Spacing (Aring) Mineral(s) Indicated Mg-saturated air-dried
14-15 Smectite vermiculite chlorite 99-101 Mica (illite) 71-73 Serpentine
715 Kaolinite chlorite Mg-saturated glycerol-solvated
177-180 Smectite 14-15 Vermiculite chlorite
99-101 Mica (illite) 71-73 Serpentinte
715 Kaolinite chlorite K-saturated air-dried
14-15 Chlorite vermiculite (with hydroxy interlayer) 124-128 Smectite 99-101 Mica vermiculite (contracted) 71-73 Serpentine
715 Kaolinite chlorite K-saturated heated (500ordmC)
14 Chlorite 99-101 Mica vermiculite (contracted) smectite (contracted) 71-73 Serpentine
715 Chlorite
41
Table 29 Results of the Diablo clay serpentine soil and chromite sequential extraction experiments
1 Exchangeable Fraction (1mM BaCl2) 4 Sum of 1 2 amp 3 (mg kg-1) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 0 0 003 Cr 159 388 19 Fe 014 011 001 Fe 24812 25532 1096 Ni 104 103 008 Ni 1854 1868 25 Mn 0 0 0 Mn 778 1172 04
2 Fe-Oxide Fraction (1M NH2OH HCl in 25 (vv) HOAc) 5 Total Cr Fe Ni and Mn (Total Digestion) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 19 78 05 Cr 2290 3310 1767864 Fe 4474 5138 38 Fe 60000 68000 1044325 Ni 1610 1506 01 Ni 1673 2345 00 Mn 231 430 00 Mn 1498 1608 3424547
3 Silicate Fraction (10M HF) 6 Percent Extractable ((45)100) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ------------------------------------- Cr 140 310 14 Cr 07 12 00 Fe 20337 20393 1058 Fe 41 38 01 Ni 234 352 22 Ni 111 80 00 Mn 547 742 04 Mn 52 73 00
ICP-AES analyses
42
Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1 kg of sample
Chromite
FeCr2O4 Cr Concentration
(mg kg-1) Grams Sample 500 11 011 1000 22 022 2000 43 043 5000 108 108
43
Figure Captions
Figure 21a Soils map of Jasper Ridge Biological Preserve (Stanford CA) with
complete descriptions of the soils shown in the legend 21b An enlarged image of the
study area shown in 22a indicates sampling sites of the soils and rocks collected A
dashed line marks the 500 m transect
Figure 22 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above
Figure 23 Petrographic images illustrating the varying serpentinite textures at Sites (A)
JR3 (B) JR5E and (C) JR3E For reference to site locations see Figure 21b
Figure 24 Correlation diagrams of Cr-bearing minerals in the serpentinite at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Si in Wt
Chromite (circle) Cr-magnetite (square) chromitesilicate or CSM (down triangle)
chlorite (hexagon) and pyroxenes (up triangle) analyses are shown on each correlation
diagram
Figure 25 Images of the serpentinite bedrock at Site JR3 demonstrating the mineralogy
and Cr-bearing phases Microprobe analyses are indicated by circled letters and these
values are in Table 23 25a Backscattering electron image demonstrating several Cr-
bearing phases present including chromite Cr-magnetite CSM and Cr-chlorite from
Figure 23a 25b Magnified image shown in Figure 25a demonstrating the fine-grained
texture of CSM Additionally Cr2O3 Wt values are shown for chromite the depletion
zone and CSM 25c Image of Ni-bearing phases present in the bedrock including
olivine serpentine and Ni-Fe alloy from Figure 23a 25d Cartoon illustrating the
mineralogical and chemical changes present in 25c
Figure 26 Images of a Cr-chlorite and Cr-magnetite assemblage in the serpentinite
bedrock 26a Petrographic image of the bedrock at Site 4 (Figure 21b) 26b
44
Backscattered electron image demonstrating the common occurrence of Cr-chlorite
encompassing chromite grains Microprobe analyses are indicated by circled letters and
are located in Table 23
Figure 27 Highest and lowest Cr concentrations (mg kg-1) at each site in the serpentine
soil 500 m transect shown in Figure 21b Samples included soils at various depths (m)
and bedrock from each site The average Cr concentration for the Jasper Ridge
serpentinite is marked on the central figure Chromium concentrations versus depth (m)
at each site (noted in the upper left corner) outline the main diagram
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected
for total gt2 mm sand silt and clay size fractions listed in Table 26
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soil at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Al+Fe in Wt for
chromite (circles) Cr-magnetite (hexagons) and CSM (diamonds) Chromite from the
bedrock is denoted by a larger circle and FeCr2O4 with a square
Figure 210 XRD clay analyses of sample JR3515 (Site JR3 Figure 21b) demonstrate
the influence of each treament Mg-saturated Mg+glycerol K-saturated and K-saturated
+ heat (500oC) The d-spacings and intensities are also listed in Table 27
Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at depths (a) 0-5cm (b) 5-15cm (c) 15-30cm (d) 30-45cm and (e) bedrock The pre-edge
area indicative of Cr(VI) peaks is shown on the right Note that no peaks are present
Figure 212 X-ray microprobe elemental maps of Cr and Fe for (212a) serpentine soil
at a depth of 0-5cm (212b) serpentine soil at a depth of 5-10cm and (212c) serpentinite
bedrock at Jasper Ridge site JR3 (Figure 21b)
45
Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments for the Diablo clay (Site 1 Figure 21b) serpentine soil (Site 2
Figure 21b) and chromite listed in Table 29 The sum refers to the summation of the
three extraction experiments
Figure 214 SEM images of the chromite standard (A) after each sequential extraction
treatment including the (B) exchangeable fraction (C) Fe-crystalline oxide fraction and
(D) silicate fraction The concentration of Cr removed by each treatment is shown
beneath each set of images
Figure 215 Tie-lines from Figure 29 connecting serpentinite chromite and end-
member FeCr2O4 are shown on this single plot Soil analyses from Figure 29 are also
plotted on this diagram For a sum of 100 Wt oxygen must be added to the sum of
Cr Al Fe and Mg Wt
46
Figure 21a
47
Figure 21b
48
Chromium Concentration (mg kg-1)1000 10000
Dep
th (m
eter
s) 01
10
Maryland (Rabenhorst et al 1982)California (Gough et al 1989)California (Hotz 1964)New Foundland (Roberts amp Proctor 1992)
Surface
Figure 22
49
Figure 23
50
Cr (Wt )0 5 10 15 20
Al (
Wt
)
0
5
10
15
20
25
30
Cr (Wt )0 5 10 15 20
Fe (W
t
)
0
20
40
60
80
100
Cr (Wt )0 5 10 15 20
Mg
(Wt
)
0
5
10
15
20
25
Cr (Wt )0 5 10 15 20
Si (W
t
)
0
5
10
15
20
25
30
Cr0 10 20 30 40 50 60 70 80 90 100
Al
0
10
20
30
40
50
60
70
80
90
100
Fe
0
10
20
30
40
50
60
70
80
90
100
Chromite
Chlorite
Magnetite
Chromite-Silicate
Pyroxenes
Chromite
Chlorite
Pyroxenes
Magnetite
Chromite-Silicate
Figure 24
51
Figure 25
52
Figure 26
53
Sites1 2 3 4 5 6 7 8 9 10
Con
cent
ratio
n (m
g kg
-1)
1500
2000
2500
3000
3500
4000
4500
5000Soil Chromium
1
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000002004006008010012014016
2
Cr Concentration (mg kg-1)
1500 2000 2500 3000 3500 4000
Dep
th (m
eter
s)
000002004006008010012014
3
Cr Concentration (mg kg-1)3500 4000 4500 5000
Dep
th (m
eter
s)
000
002
004
006
008
010
012
5
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000002004006008010012
6
Cr Concentration (mg kg-1)2200 2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000002004006008010012
7
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000
002
004
006
008
010
9
Cr Concentration (mg kg-1)
2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000
002
004
006
008
010
012
10
Cr Concentration (mg kg-1)1600 2000 2400 2800 3200
Dep
th (m
eter
s)
000002004006008010012
4
Cr Concentration (mg kg-1)3500 3600 3700 3800
Dep
th (m
eter
s)
000
001
002
003
004
8
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000
002
004
006
008
010
BEDROCK
BEDROCK
BEDROCK
Range of Cr Concentrations in Serpentinite Bedrock
Figure 27
54
Soil Chromium
Depth (cm)0 10 20 30 40 50
Cr m
g kg
-1
02000400060008000
100001200014000
Totalgt2mmSandSiltClay
Soil Nickel
Depth (cm)0 10 20 30 40 50
Ni m
g kg
-1
0
1000
2000
3000
4000
5000
Soil Manganese
Depth (cm)0 10 20 30 40 50
Mn
mg
kg-1
0
1000
2000
3000Totalgt2mmSandSiltClay
Soil Cobalt
Depth (cm)0 10 20 30 40 50
Co
mg
kg-1
0
50
100
150
200
250
300
Totalgt2mmSandSiltClay
Totalgt2mmSandSiltClay
Exceeds detection limits of 10000 mg kg-1 Cr
Figure 28
55
Cr (Wt )0 10 20 30 40 50
Al (W
t
)
0
5
10
15
20
25
30
Cr (Wt )0 10 20 30 40 50
Mg
(Wt
)
0
2
4
6
8
10
12
Cr (Wt )0 10 20 30 40 50
Si (W
t
)
0
2
4
6
8
10
12
14Cr (Wt )
0 10 20 30 40 50
Fe (W
t
)
0
10
20
30
40
50
60
70
Cr (Wt )0 10 20 30 40 50
Al +
Fe
(Wt
)
10
20
30
40
50
60
70
JRBPChromite
FeCr2O4
Chromite
Magnetite
Chromite-Silicate
Figure 29
56
Clay XRD of JR3515 Depth 5-15cm
0
200
400
600
800
1000
1200
0 5 10 15 20 25 30 35 40 45
d-spacing (Aring)
Rel
ativ
e In
tens
ity
Mg-Saturated
Mg+Glycerol
K-saturated
K-saturated+heat (500oC)
Figure 210
57
Energy (kev)
597 598 599 600 601 602 603
Nor
mal
ized
Abs
orba
nce
-4
-3
-2
-1
0
1
2
(a)
(b)
(c)
(d)
(e)
Cr(
VI)
Figure 211
58
Figure 212
59
Diablo Clay
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Serpentine Soil
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Chromite
0
20
40
60
80
100
120
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide
SilicateSum
Figure 213
60
Figure 214
61
Reaction Progress ()0 20 40 60 80 100
Elem
enta
l Wt
0
10
20
30
40
50
ChromiumAluminumIronMagnesium
FeCr2O4Jasper Ridge
Bedrock Chromite
SoilChromite
Figure 215
62
CHAPTER 3
OXIDATIVE PROMOTED DISSOLUTION OF CHROMITE BY MANGANESE
DIOXIDE AND CONCURRENT PRODUCTION OF CHROMATE
ABSTRACT
Chromium release and oxidation from chromite is a potential environmental hazard in
sediments and soils and a pathway for soil development (weathering of primary minerals)
related to ultramafic rocks and their metamorphic derivatives (serpentinites) Birnessite is
a common pedologic mineral in these sediments and soils capable of oxidizing aqueous
Cr(III) In this study the interaction between chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4)
and birnessite is investigated with an interest in the potential generation of Cr(VI)
Specifically the effects of chromite suspension density birnessite suspension density
and pH at a temperature of ~25degC in relation to the reaction kinetics were examined The
rate of Cr(VI) in solution increases with increasing chromite suspension densities and
decreasing pH but is independent of birnessite suspension densities at values greater than
~20 m2 L-1 The overall rate expression of Cr released and oxidized in solution from
chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression with Cr(VI) rates ranging from 9times10-4 to 44times10-3 microM h-1
Additionally these experiments demonstrate that serpentine soils are a source of non-
anthropogenic Cr(VI)
63
INTRODUCTION
Chromite a Cr-enriched spinel is responsible for a majority of the Cr enrichment
identified in ultramafic rocks serpentinites and serpentine soils (Rabenhorst et al 1982
Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989 Gasser and
Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003 Oze et al
2003a Oze et al 2003b) The chemical weathering of chromite is therefore an important
process when evaluating the potential of ultramafic rocks and their related sediments and
soils as sources of Cr Despite the propensity of chromite to resist chemical weathering
the dissolution of chromite and release of Cr will occur over geologic time (Bialowolska
and Salacinski 1984 Oze et al 2003b) Chromium in chromite is in the form of Cr(III)
the non-toxic and chemically immobile valence state however oxidizing conditions in
near surface environments is a potential reaction leading to the enhanced dissolution of
chromite and the formation of Cr(VI) the toxic and chemically mobile valence state
which typically exists as the oxyanion chromate (HxCrO42-x) (Daugherty 1992 Cohen et
al 1993 Fendorf 1995 Deer et al 1996) Consequently chromite may be a non-
anthropogenic source of Cr(VI) in the environment
Naturally occurring high valence Mn oxides present in soils and sediments
derived from ultramafic rocks are strong oxidizers capable of affecting the chemical
mobility of metals through redox reactions (Fendorf and Zasoski 1991 Johnson and
Xyla 1991) These insoluble mixed valent oxides (Mn(IIIIV)) are observed coating
mineral surfaces in these ultramafic sediments and soils where total Mn concentrations
are often gt100 mg kg-1 (Brooks 1987 Schreier et al 1987 Oze et al 2003a Oze et al
2003b Oze et al) Birnessite is a naturally occurring mixed valence oxide known to
oxidize aqueous Cr(III) to the more chemically mobile Cr(VI) (Eary and Rai 1987
Fendorf and Zasoski 1991) Gough et al (1989) and Becquer et al (2003) have
suggested that Mn oxides are responsible for Cr(VI) identified in California and New
Caledonia serpentine soil solutions however the dissolution and oxidation of Cr from a
Cr-bearing mineral due to the presence of birnessite has not been quantitatively defined
In this study we examine the kinetics of Cr release and oxidation when chromite and
birnessite are present in solution
64
The objective of this study is to determine a rate expression describing the
release and oxidation of Cr from chromite in the presence of birnessite Using well-
controlled batch experiments the solution chemistry and geochemistry of Cr related to
chromite and birnessite interaction was evaluated temporally over a range of pH
Conditions were chosen to simulate the slightly acidic conditions characteristic of most
serpentine soils (Rabenhorst et al 1982 Gough et al 1989 Graham et al 1990 Cole
1992 Jeffrey 1992 Lee 1992 Verger 1992 Becquer et al 2003) Ultimately these
experiments allow the examination of one potential pathway leading to Cr(VI) identified
in solutions related to ultramafic rocks and serpentine soils
MATERIALS AND METHODS
Solid Phase Characterization and Synthesis
Chromite used in the experiments was commercial-grade chromite (MSDS No
302-1 WHMIS CLASS D2A) obtained from Barnes Environmental Inc Ontario
Chromite was prepared by cleaning the sand-size material in multiple ultrasonic
isopropanol baths and rinses The chromite was then sieved using a Retsch magnetic
sieve shaker for 30 minutes in order to obtain the 160-250 microm size fraction Impurities in
the 160-250 microm chromite were removed using a slope Franz magnetic separator These
impurities included the minerals olivine augite plagioclase feldspar and quartz Visual
and scanning electron microprobe (SEM) examinations of the remaining 160-250 microm
fraction confirmed its purity Chromite was crushed to a fine powder using an agate
mortar and pestle and then rinsed multiple times in deionized destilled (DDI) water and
001 M HCl solution to remove the ultra-fine material and eliminate highly reactive sites
caused during grinding The chromite powder was inspected before (Figure 31) and
after the experiments using a scanning electron microscope (JEOL 5600LV) the surface
area was 025 m2 g-1 as determined using N2(g) and BET isotherm
An automated JEOL 733A electron microprobe operated at 15 kV accelerating
potential and 15 nA beam current was used to investigate possible zoning and to quantify
the chemical composition of the chromite grains prior to powdering No zoning was
observed and the chemical composition of the chromite was based on the average of 25
electron microprobe analyses The stoichiometric composition of chromite based on 4
65
oxygens is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 X-ray diffraction was also used to
deduce the purity of the chromite powder X-ray diffraction spectra obtained for the
chromite powder matched JCPDS card 03-0873 (unit cell a=8300 b=8300 c=8300
α=90 β=90 and γ=90) and no contaminants were evident
Birnessite was synthesized using the procedure described by Buser et al (1954)
and implemented by Fendorf and Zasoski (1992) A solution of 63 g of potassium
permanganate (KMnO4) in 1 L of DDI water was heated to 90 degC Approximately 100
mL of the heated slurry was poured into a 4 L beaker and reheated to 90degC while being
stirred vigorously The remaining 900 mL of slurry and 66 mL of concentrated
hydrochloric acid were simultaneously and proportionally dispensed into the beaker over
a period of thirty seconds The temperature of the resulting solution was maintained at
90deg C for an additional 10 minutes and then allowed to cool to room temperature The
product was washed with alternate rinses of DDI water and a 001 M HCl solution The
birnessite surface area was 125 m2 g-1 based on BET analyses Solids were inspected
before (Figure 31) and after the experiments using the SEM
Extraction Experiment Procedures
Several chemical extraction experiments were performed on 106-250 microm
chromite grains prior to the chromite-birnessite batch experiments in order to determine
the potential release of Cr Fe and Mn One gram of chromite and 25 mL of solution
were placed into a 25 mL centrifuge tube and the solids and solution were gently agitated
The solutions and parameters used for these extraction experiments include the
following 1) 10 hydrogen peroxide (H2O2) solution at 25degC and pH of ~5 for 6 h 2)
01 M potassium permanganate (KMnO4) solution at 60degC and pH of ~3 for 6 h 3) 10 M
hydrofluoric (HF) acid at 60degC for 6 h 4) 1 M hydroxylamine (NH2OHmiddotHCl) with 25
(vv) acetic acid (HOAc) at 60degC and pH of ~4 for 6 h and 5) 1 mM barium chloride
(BaCl2) solution at 60degC and pH of ~6 for 6 h Solutions were extracted with a syringe
and passed through a 02 microm filter Solids were inspected after the experiments using the
SEM
66
Batch Experiment Procedures
Batch experiments were performed in triplicate using 1 L polypropylene bottles
and chromite and birnessite ratios listed in Table 31 at ~25degC After loading chromite
and birnessite into each bottle 1 L of a pH 3 solution a 10 mM acetate solution (pH 5) a
10 mM MES solution (pH 67) or a pH 8 solution was added to the appropriate
experiment (Table 31) Chromite and birnessite were also run individually in the acetate
buffered solution (Table 31) Aliquots were extracted from the polypropylene bottles
using a syringe and then passed through a 2 microm filter before analyses Less than 40 mL
was extracted over the course of each experiment resulting in a total volume change of
lt4 After the experiments were terminated the remaining solids were resuspended in
20 mL of 1 mM phosphate (5050 KH2PO4K2HPO4 pH 7) 20 mL of 10 mM CaCl2
solution at pH 7 and a 05 M HCl solution These suspensions were agitated for 3 h
followed by centrifugation and filtration (02 microm) before analyses The purpose of the
phosphate and CaCl2 solutions was to desorb reaction produtcts (eg Cr(VI) Cr(III) Fe
and Mn) from the solid phases The HCl treatment provides a means to evaluate
secondary phases such (oxy)hydroxides formed during the experiments Although the
HCl solution may dissolve chromite and birnessite in addition to the (oxy)hydroxides the
short time frame of the extraction (3 h) should limit the dissolution of chromite
Additionally the polypropylene bottles following the termination of five batch
experiments were rinsed with DDI water and 20 mL of 05 M HCl was added to each
bottle The HCl solution was allowed to remain in the bottles for 15 h in order to
determine the concentration of Cr Fe and Mn sorbed onto the interior lining
Serpentine Soil and Birnessite Experiments
Serpentine soils collected from Jasper Ridge Biological Preserve (Stanford CA)
and New Caledonia were used to determine the influence of birnessite in soils known to
contain chromite (Chapters 2 and 4) The Cr geochemistry of these soils as well as the
detrital chromite have been evaluated in Chapters 2 and 4 and by Becquer et al (2003)
Jasper Ridge serpentine soils contain Cr-bearing minerals including chromite Cr-
magnetite and Cr-chlorite however a majority of the Cr enrichment (Table 32) is
directly related to the abundance of chromite A majority of the Cr related to New
67
Caledonia serpentine soils is attributed to Cr (oxy)hydroxides (Chapter 4) however
chromite is present in these soils as well (Becquer et al 2003) Although birnessite has
been noted in these soils the quantity of birnessite could not be determined In order to
evaluate the impact of Mn oxides on Cr within such soils 01 g of birnessite was added to
each soil
The parameters for the experiments are listed in Table 32 Ten grams of the
lt250 microm size-fraction of each soil (JR1 JR2 and NC) and 100 mL of 1 mM acetate
buffered solution (pH 5) were loaded into individual 150 mL Nalgene bottles Aqueous
solutions were evaluated as a function of time (h) Additionally 10 g of each soil 01 g
of birnessite and 100 mL of 1 mM acetate buffered solution (pH 5) were loaded into 150
mL Nalgene bottles and were allowed to react as a function of time (h) These
experiments were rotated at a rate of 2 revolutions min-1 Solutions were removed with a
syringe and passed through a 2 microm filter before analysis Approximately 8 mL of
solution was required for the analyses over the course of each experiment resulting in a
total volume change of 8 Total Cr concentrations (mg kg-1) of the soils listed in Table
32 were obtained by completely dissolving the samples using a mixture of hot (gt200degC)
concentrated nitric perchloric and hydrofluoric acids Elemental compositions were
measured in the supernatant using inductively coupled plasma optical emission
spectroscopy (ICP-OES) The mass of chromite in each soil was calculated using total Cr
concentration for each soil and the calculations provided in Chapters 2 (Table 210) and
Chapter 4 (Table 47) The chromite surface area of each soil was approximated by
comparing the surface area measured for the chromite standard and its observed grain
size from SEM images (Figure 31) to chromites observed in the soils using backscattered
electron images (Chapters 2 and 4) The surface area of the Jasper Ridge chromite is 066
of the chromite standard due to a slightly larger grain size The surface area of the
New Caledonia chromite is approximately 025 of the chromite standard Chromite
suspension densities were calculated by multiplying the chromite mass by the estimated
surface area and then dividing by the volume of the solution
68
Solution Analysis
The pH and Eh of the supernatant was measured with an Orion flat surface
combination pH electrode inserted directly into supernatant Production of soluble
Cr(VI) was measured spectrophotometrically at 540 nm using the s-diphenyl carbazide
method modified from Bartlett and James (1979) Soluble Fe(II) was monitored using
the ferrozine method adopted from Stookey (1970) Total dissolved Cr Fe Al Mn and
Mg were ascertained using ICP-OES Concentrated trace metal grade nitric acid used to
acidify the samples prior to analyses Aluminum and Mg was determined to be present in
the nitric acid at concentrations lt1 microM Magnesium and Al are not plotted in any of the
figures Chromium(III) and Fe(III) was determined by subtracting Cr(VI) and Fe(II)
obtained from colorimetric methods from total dissolved Cr and Fe analyses using ICP-
AES Results are presented as an average of the three runs for each batch experiment
with the standard deviation noted by error bars
RESULTS
Chromite Extraction Experiments
The release of Cr Fe and Mn from chromite using 10 H2O2 01 M KMnO4 10
M HF acid 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 1 mM BaCl2 solutions was
carried out to determine the relative propensity of chromite to resist chemical dissolution
and these results are summarized in Figure 32 The KMnO4 solution with aqueous
Mn(VII) was the most effective in releasing Cr from chromite yielding a concentration
of 62 Cr mg L-1 Iron released using the KMnO4 solution was 03 mg L-1 Although
H2O2 is a direct chemical oxidant the release of Cr from chromite was less as compared
to the other solutions (only ~03 Cr mg L-1 was measured) Concentrated HF was only
able to release ~13 Cr mg L-1 while releasing 106 Fe mg L-1 NH2OHmiddotHCl with 25
(vv) HOAc and CaCl2 solutions only removed lt1 Cr mg L-1 however the NH2OHmiddotHCl
removed 36 Fe mg L-1 which is comparatively higher than the H2O2 KMnO4 and BaCl2
solutions
69
Chromite and Birnessite Batch Experiments
The interaction of chromite and birnessite leading to the release and oxidation of
Cr(III) is a relatively slow process requiring study on the order of hours and days The
pH was maintained constant throughout the duration of the experiments while Eh values
ranged from 320 to 450 mV with no discernable temporal pattern (Figure 33) For pH
values between 3 and 67 Cr3+ CrOH2+ and Cr(OH)2+ are the dominant aqueous Cr(III)
species however at pH 8 conditions are on the stability line between Cr(III) and Cr(VI)
specifically Cr(OH)3 and CrO42-
Higher chromite suspension densities produced higher rates of Cr release to the
aqueous phase and all the Cr in solution was in the form of Cr(VI) (Figure 35) The
rates of oxidation based on Cr(VI) production range from 00005 to 00041 microM h-1
(Figure 35 Table 33) Comparatively Cr released from the chromite control experiment
is maintained at a relatively constant value of ~005 microM and no Cr(VI) is measured in
solution Iron concentrations initially increase to ~ 05 microM and then decrease rapidly to a
relatively constant concentration at 02 microM (Figure 35) Iron concentrations for the
chromite control experiment began at 15 microM between 0 to 50 h and decreased to ~02
microM Iron(II) and Mn concentrations are below detectable limits for all experiments
(Figure 35) A zeroth-order dependence is assumed with respect to birnessite owing to
its large excess surface area relative to chromite
Variable birnessite suspension densities and constant chromite suspension
densities were evaluated at pH 5 (Figure 36) Although suspension densities of
birnessite are 538 and 1075 m2 L-1 (Table 31) the rate of Cr and Cr(VI) measured in
solution for both experiments is 00041 microM h-1 (Figure 36 and Table 33) Iron
concentrations are below 06 microM and maintain concentrations near 02 microM with no
indication of consistently increasing or decreasing with respect to time Iron(II) and Mn
concentrations in solution are below detectable limits for both experiments
In Figure 37 the influence of pH on chromite oxidation by birnessite was
evaluated by conducting experiments at pH values of 3 5 67 and 8 chromite and
birnessite suspension densities are constant at 0075 m2 L-1 and 1075 m2 L-1 respectively
(Table 31) Chromium and Cr(VI) rates increase with decreasing pH ranging from 0
microM h-1 at pH 8 to 00062 microM h-1 at pH 3 (Table 33) Iron is present in solution at
70
concentrations less than 04 microM for all the experiments Generally Fe concentrations are
greater near the start of the experiment and decrease to values lt 05 microM with increasing
reaction time Manganese concentrations are below detectable limits
In order to determine the rate dependence on each prominent reactant (chromite
birnessite and pH) reaction rates were regressed against reactant concentrations (Figure
38) A pseudo first-order rate dependence is noted based on chromite suspension
density the equation regressed for this relationship is listed in Table 33 As the chromite
suspension density increases the rate of Cr(VI) in solution increases In contrast a
zeroth-order rate dependence is evident for birnessite suspension density (dashed line in
Figure 38b) The rate of Cr(VI) generation is expected to decrease with respect to
birnessite suspension density at very low concentrations (dotted line in Figure 38b)
The influence of pH on the rate of Cr(VI) production is well-described by a polynomial
function (Figure 38c Table 33)
Developing an overall rate expression for Cr in solution requires integrating the
factors assessed in Figure 38 and listed in Table 33 All the Cr measured in solution is
in form of Cr(VI) therefore the rate law for Cr or Cr(VI) in solution may be written as
d[Cr(VI))]dt = k sdotChromitea sdot Birnessiteb sdot (aH+)c (31)
and
log d[Cr(VI))]dt = log k + asdotlogChromite+ bsdot log Birnessite+ csdotlog (aH+) (32)
where k is the overall rate constant for Cr release and oxidation Chromite is the
chromite suspension density (Table 31) Birnessite is the birnessite suspension density
(Table 31) and aH+ is the activity of H+ Constants a b and c are the order of
dependency for each reactant The overall rate equation (developed in the Appendix) can
be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069 (33)
71
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 the expression is valid for birnessite suspension densities
greater than 20 m2 L-1 and pH values between 3 to 8
Secondary and Solid Phase Examination of Chromite and Birnessite
The solid products were evaluated to determine the abundance and type of
secondary phases formed during the course of the experiments Solids from the batch
experiment were treated with three solutions 1 mM 5050 KH2PO4K2HPO4 1 mM
CaCl2 and 05 M HCl (Table 34) In the chromite and birnessite reactions Cr was
primarily in solution (gt90 ) The HCl extract was able to remove lt5 additional Cr
while the KH2PO4K2HPO4 and CaCl2 extracts released lt3 additional Cr Aqueous Fe
concentrations were below detectable limits for the final solution extraction Treatment
with the HCl resulted in up to 0127 micromoles of Fe however 0107 micromoles of Fe were
released upon treatment of unreacted chromite with HCl Although not shown in Table
34 Mn values were greater than gt4 micromoles when treated with HCl the
KH2PO4K2HPO4 and CaCl2 removed no detectable Mn
After the termination of each experiment the solids were examined using SEM in
order to identify changes in the surface morphology and secondary phases A large
chromite grain and conglomerate of birnessite adjacent to the chromite are depicted in
Figure 38 The face and edge of the chromite grain is not noticeably altered compared to
grains examined prior the experiments Additionally the particle morphology of
birnessite has not visibly changed Even at higher magnification angular and
conchoidally fractured fine-grained edges are noted (Figure 38bc) Only a slight
rounding of the smaller features of the main chromite grain provides any evidence of
possible chromite dissolution (Figure 38c) Birnessite appears to have limited interaction
with chromite Although chromite and birnessite were allowed to interact over 500 h in
an aqueous solution no secondary phases are observed by SEM
Serpentine Soil and Birnessite Experiments
Two serpentine soils from Jasper Ridge Biological Preserve CA and one
serpentine soil from New Caledonia were used to determine whether or not the overall
72
rate expression (Equation 33) for chromite reaction with birnessite is consistent with
Cr(VI) concentrations produced by serpentine soils containing chromite Three
serpentine soils without birnessite and three soils with the addition of birnessite buffered
at pH 5 were measured for Cr(VI) generation as a function time (Figure 310) The rates
of Cr(VI) measured in solution for both Jasper Ridge soils (JR1 and JR2) are identical at
00013 microM h-1 With the addition of birnessite the rates of Cr(VI) production in solution
both increased to 00044 microM h-1 (JR1 w Bt and JR2 wBt in Figure 310) Based on the
chromite suspension density (Table 32) the rates predicted by Equation 33 range from
00035 to 00041 microM h-1 which are 10 to 20 less than those observed in the
experiments (Figure 310) The rates of Cr(VI) in solution from the New Caledonia
serpentine soil are 00009 microM h-1 and with the addition of birnessite increase up to 00013
microM h-1 (Figure 310) A rate of 00016 microM h-1 was predicted by Equation 33 for the
New Caledonia soil
DISCUSSION
The reaction of birnessite with chromite enhances the rate of Cr release into
solution The rate of Cr(VI) measured in solution (Equation 33) is directly related to the
chromite suspension density and is independent of the birnessite at suspension densities
greater than ~20 m2 L-1 Although the release and oxidation of Cr in solution are
measured the mechanisms and pathways driving this reaction need to be evaluated The
propensity of chromite to resist dissolution the effects of pH the formation of secondary
phases and ultimately the pathways of Cr release and oxidation are addressed in the
following discussion
Extraction experiments performed on chromite (Figure 32) demonstrate it is
resistant to acidic dissolution or competitive ion displacement but strong oxidants such
as high valent Mn may oxidatively dissolve chromite It is evident that the proportions of
Cr Fe and Mn vary for each of the extractions The solutions utilized in the extraction
experiments have varying chemical properties capable of preferentially releasing a
particular species andor element These experiments were also performed over a short
period of time (lt6 h) which may have only released elements from highly reactive sites
73
Finally secondary phases associated with these experiments were possible and would
distort the elemental ratios observed in solution
A majority of the Eh and pH analyses plot in the Cr(III) stability region (Figure
33) however Cr in solution is primarily in the form of Cr(VI) (Figures 35 36 and
37) Eh measurements presented in Figures 32 and 33 do not account for the
microenvironments associated with the surfaces of birnessite Additionally the kinetics
of Cr(VI) reduction are limited in reducing conditions due to electron symmetry
constraints and the absence of a suitable redox couple (Fendorf 1995) Finally highly
oxidizing birnessite in the solution is kinetically and thermodynamically controlling the
oxidation state of Cr relative to the stabilities predicted in Figure 33 (Amacher and
Baker 1982 Fendorf and Zasoski 1992)
A fractional-order dependence on the chromite suspension density (Figure 38)
can be explained by diffusional influences competing oxidation reactions and secondary
reactions involving Cr(III) and Mn(IV) Secondary phases on the surface of chromite
may inhibit the diffusion of Cr from chromite however the clean and relatively unaltered
chromite surfaces do not support this mechanism as a major cause (Figure 38)
Chromium and Fe released from chromite may be incorporated into secondary phases
related to birnessite or chromite Approximately 10 of Cr is not in the aqueous phase
(Table 34) supporting the premise that secondary phases are contributing to the
fractional-order dependence related to the chromite suspension density Also
competition from Fe(II) oxidation on birnessite deposition of Fe(II) on birnessite (or
chromite) and diffusion onwithin birnessite may further impact the rate dependence
The zeroth-order dependence based on the birnessite suspension density (Figure
38) results from its high surface area At very low birnessite suspension densities the
rate of Cr(VI) production should decrease however due to the high surface area of
birnessite it was experimentally prohibitive to determine reaction rates at masses less
than 4 mg Therefore a hypothetical trend-line depicts the expected dependence on
birnessite concentration at very low suspension densities at higher values representative
of most field settings the reaction is independent of the birnessite surface area (ie zero-
order dependence)
74
In Figure 38c the polynomial relationship defines the Cr(VI) rate dependence on
pH As the pH decreases the rate of Cr(VI) increases more rapidly The relationship is
potentially the result of lower pHs progressively increasing the rate of chromite
dissolution However the rate of Cr(VI) in solution might be influenced by higher pHs
increasing the formation of secondary Fe(III) and Cr(III) (oxy)hydroxides on the
chromite andor birnessite surfaces thereby decreasing elemental diffusion from
chromite or decreasing available birnessite sites for Cr(III) oxidation Differentiating
between these two mechanisms is not possible based on the rate equation or extraction
analyses however the polynomial relationship evident in Figure 38c is folded into the
rate constant in order to geochemically account for the influence of pH on Cr(VI)
generation
No conclusive evidence is available to determine the fate of Mn within the
reacting suspensions Aqueous Mn concentrations are consistently maintained below
detectable limits If Mn is released from birnessite then it must be strongly and quickly
incorporated into a nonaqueous phase following its release Due to Cr(III) being oxidized
by Mn(III) andor Mn(IV) Mn(II) must be produced Extractions performed on the
chromite and birnessite solids demonstrate that Mn could only be removed with HCl
however a majority of this Mn is potentially the result of HCl dissolving birnessite On
the basis of electron transfer stoichiometry at least 15 mols of Mn(II) must be produced
for each mol of Cr(VI) generated This amount can be accounted for in the HCl
extraction where gt4 micromoles of Mn were measured Additionally Fendorf et al (1993)
determined that Mn(II) should not significantly influence the extent of Cr(III) oxidation
Iron in solution does not progressively increase in solution over time (Figures 35
36 and 37) The high initial rate of Fe release compared to Cr may be the result highly
reactive sites preferentially releasing Fe Although no Fe is detected in solution from
the extractions performed on the chromite and birnessite solids 0005 to 0035 moles of
Fe were retrieved from the lining of the bottle and 0053 to 0126 moles were released
from the solids using the HCl solution (Table 34) Iron(II) could potentially be oxidized
to Fe(II) in the presence of birnessite and form stable Fe(III)-(oxy)hydroxides (Postma
1985 Myers and Nealson 1988 Villinski et al 2001) A ratio of two Fe for every one
Mn is necessary for the oxidation of Fe(II) and this amount is recoverable based on HCl
75
extractions with gt4 micromoles of Mn measured from the HCl extractions Additionally
dissolved O2 in the batch experiments will rapidly oxidize Fe(II) for all the pHs examined
in these experiments (Stumm and Morgan 1981 Langmuir 1997) The oxidative
pathway for Fe(II) provides an explanation for the lack of Fe(II) in solution for the batch
experiments (Figure 35 36 and 37)
With a detection limit of ~02 microM and equilibrium with respect to Fe(OH)3(am) and
goethite Fe concentrations should either be at or below detection limits at pHs gt 5
Assuming aqueous Fe(III) is present dominantly as the second hydrolysis product at pH
5 the dissolved concentrations approach the solubility of ferrihydrite and are much
greater than expected for goethite (Figure 311) Thus it is probably that the former solid
is generated which is further supported by the chemical extraction experiments (Table
34)
Iron and Cr solution concentrations measured experimentally are not consistent
with congruent dissolution (Figure 312a) Assuming Fe was forming secondary phases
at a constant rate (~00001 Fe h-1) multiplying this rate by the time analysis and then
adding it to the measured aqueous Fe a corrected Fe (Fe) value can be estimated (Figure
312) Additionally the 10 of Cr not in solution can be also be used to calculate a
corrected Cr value (Cr) (Figure 312) When secondary phases are taken into account a
majority of the analyses plot near the congruent dissolution of chromite This provides
evidence that chromite is undergoing congruent dissolution within the batch experiments
but that secondary Fe(III) phases disrupt the solution phase elemental ratios
Chromium(VI) Pathways from Chromite and Birnessite into Solution
Three pathways are evaluated for Cr(VI) measured in solution a) chromite
dissolution b) birnessite dissolution and c) solid-solution reaction (Figure 313) In order
to describe these pathways end-member chromite (FeCr2O4) and birnessite (δ-MnO2) are
used to simplify the analysis It should be recalled however that the composition of the
chromite used in the experiments is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4
Chromite dissolution presents a likely explanation for Cr(VI) observed in solution
based on the following discussion In chromite dissolution model Cr(III) released from
chromite adsorbs onto the surface of birnessite and oxidizes to Cr(VI) After the electron
76
transfer is complete Cr(VI) desorbs from birnessite and remains in solution As this
process continues Cr(VI) increases at a constant rate The rate limiting step with regards
to this specific pathway can be grouped into two steps a) dissolution of chromite (Step
1a) or b) Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) (Figure 313)
Although a dissolution rate for chromite at ambient temperatures and pressures is not
known the chromite extraction experiments (Figure 32) demonstrates that chromite
dissolution is slow Manganese oxides such as birnessite oxidize Cr(III) in solution
rapidly with a measured rate constant of 6300 microM-1 s-1 in fresh water (100 mg L-1 MnO2
100 g L-1 Cr(III) pH~6) (Van der Weijden and Reith 1982) This means that the rate
for Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) is on the order of
seconds therefore chromite dissolution is likely the rate limiting step controlling the
concentration of Cr(VI) in solution for this pathway
Although a zeroth-order dependence related to birnessite concentrations (Figure
37a) is observed the birnessite dissolution pathway in which aqueous Mn4+ released
from birnessite adsorbs onto the surface of chromite oxidizing Cr(III) cannot be entirely
dismissed Manganese dissolution and adsorption could occur quickly not allowing the
Mn to be measured in solution Examination of the solids after the termination of each
experiment demonstrated extremely minimal chromite and birnessite interaction
therefore the solid solution reaction is not a likely pathway leading to the production of
Cr(VI) in solution and is furthermore inconsistent with the noted rate dependencies
Despite the three pathways (Figure 313) the generation of Cr(VI) observed in these
experiments and described by the rate equation (Equation 33) is independent of these
potential pathways
Serpentine Soils and Cr(VI)
The serpentine soil experiments (Figure 310) provide direct evidence that Cr(VI)
is forming within these soils By adding birnessite to the soil the rate of Cr(VI) in
solution increased for all experiments (00013 to 00044 microM Cr(VI) h-1 for Jasper Ridge
and 00009 to 00013 microM Cr(VI) h-1 for New Caledonia) The Cr(VI) rate increase
related to the addition of birnessite is likely the result of low birnessite suspension
densities (lt20 m2 L-1) in the soil Rates predicted for Jasper Ridge and New Caledonia
77
soils using Equation 33 and values in Table 32 are similar to those measured in the
experiments with values of 00041 and 00016 microM Cr(VI) h-1 respectively Overall these
soil experiments demonstrate that a rate equation developed from laboratory batch
experiments predicts the rate of Cr(VI) in serpentine soils reasonably well
According to limits set by the Environmental Protection Agency the California
standard for total Cr in drinking water is 50 microg L-1 (~1 microM) while the federal standard is
100 microg L-1 (~2 microM) The serpentine soil and birnessite experiments produced Cr(VI)
concentrations in excess of 50 microg L-1 after only 140 h or roughly 6 days Serpentine soils
at Jasper Ridge typically remain water saturated for prolonged periods of time (gt1 week)
during the winter season increasing the time Cr(VI) can be produced The degree and
duration of water saturation is dependent on the topography of the serpentinite body
level areas will remain water saturated while sloping areas will allow the water to flow
down gradient Although Cr(VI) was generated in the soil experiments the production of
Cr(VI) in serpentine soils may be more limited The reduction of Cr(VI) to Cr(III) is
favorable by reactions with Fe(II)aq Fe(II)-containing minerals such as magnetite H2S
and organic matter including microbial processes (Ilton and Veblen 1994 Fendorf 1995
Peterson 1996) Both organic matter and magnetite are abundant in Jasper Ridge
serpentine soils (Chapters 2 and 4) however these factors did not deter Cr(VI) from
forming during the serpentine soil experiments Additionally the soils in these
experiments were constantly rotated allowing the mineral surfaces to be continuously
exposed to solution Chromite within an undisturbed soil will not always be in contact
with the solution or proximal to birnessite
Chromium(VI) produced by the interaction of chromite and birnessite is not
limited to serpentine soils Serpentinites the bedrock for serpentine soils are typically
highly sheared due to the diapiric migration of serpentinite masses along zones of
weakness in fault zones Numerous cracks and lineations which are typical of these rocks
provide a means for Cr(VI) to transported away from the serpentine soil potentially into
local groundwater In fact preferential flow will occur along the major shear plains of
the serpentinite Additionally these shear plains are often coated with high valent Mn
oxides providing a reaction pathway for Cr(VI) formation Serpentinite springs such as
those evaluated by Barnes and ONeil (1969) and Barnes et al (1972) at the Cedars
78
ultramafic body in Cazadero CA have Cr concentrations measured at 20 microg L-1 in which
all the Cr is in the form of Cr(VI) (Chapter 4) In conclusion chromite and birnessite in
serpentine soils could contribute a source of Cr(VI) in surface waters and groundwaters
Chromium release and oxidation from chromite is increased by the presence of
birnessite and the interaction between chromite and birnessite is a pathway for Cr(VI)
generation in ultramafic sediments and soils The potential of serpentine soils as a source
of non-anthropogenic Cr contamination can now be evaluated quantitatively based on the
abundance of chromite The overall rate law for Cr(VI) in solution (Equation 39)
provides a means to estimate the Cr(VI) concentration attainable for serpentine soils and
serpentine sediments when the chromite and birnessite suspension densities and the pH of
the soil are known
ACKNOWLEDGEMENTS
We gratefully acknowledge comments and assistance from Robert Jones and Dr
Guangchao Li Funding for this research was provided by the Office of Teaching and
Learning at Stanford University
79
Table 31 Chromite and birnessite batch experiment parameters with chromite and birnessite suspension densities
Experiment Chromitea (grams)
Birnessiteb (grams) pHc
Chromite Surface Aread
(m2 g-1)
Birnessite Surface Aread
(m2 g-1)
Chromite Suspension
Density (m2 L-1)
Birnessite Suspension
Density (m2 L-1)
A 08 043 5 025 125 02 54 B
08 086 5 025 125 02 108C 03 086 5 025 125 008 108D 008 086 5 025 125 002 108E 004 086 5 025 125 001 108F 03 086 3 025 125 008 108G 03 086 67 025 125 008 108H 03 086 8 025 125 008 108
Chromite 08 ---- 5 025 125 02 ----Birnessite ---- 043 5 025 125 ---- 5480
a Chromite chemical composition is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 b Synthetic birnessite (δ-MnO2) c10 mM acetate buffered solution pH = 5 non-buffered pH=3 5 mM MES buffer pH 67 non-buffered pH 8 dMeasured BET analyses
81
Table 32 Serpentine soil and birnessite experiments with soil and birnessite masses used and the estimated chromite suspension density
Samples Soil Mass (g)
Birnessite Mass (g)
Volume (L) of Solutiona
Total Crsoil
b (mg kg-1)
Chromite Mass (g) in
Soil Samplec Estimated Surface
Areac (m2 g-1)
Chromite Suspension Density
(m2 L-1) in Soild Jasper Ridge
(JR1) 10 ---- 01 5976 013 015 02 Jasper Ridge (JR1 w Bt) 10 01 01 5976 013 015 02 Jasper Ridge
(JR2) 10 ---- 01 4978 011 015 016 Jasper Ridge (JR2 w Bt) 10 01 01 4978 011 015 016
New Caledonia (NC) 10 ---- 01 827 0015 006 001
New Caledonia (NC w Bt) 10 01 01 827 0015 006 001
a1mM Acetate Buffer (pH=5) 81 bTotal Cr concentrations were obtained through total digestion and ICP-AES analyses
cEstimated chromite mass and surface area (discussed in the Methods section) was made using chromite observations and calculations in Chapters 2 and 4
dCalculated from the chromite mass in the soil sample estimated surface area of the chromite and the volume of solution
82
Table 33 Summary of Cr(VI) rates obtained from batch experiments (Figure 35 36 and 37) and used in Figure 38
Listed in Table 31
Suspension Densities (m2 L-1)
Chromite (Ch)
Birnessite (Bt)
pH Rate (microM h-1) (d[Cr(VI)]dt)
log Rate (log microM h-1) (log d[Cr(VI)]dt) Expression
Figure 35 02 108 5 00041 -239
008 108 5 00021 -268 002 108 5 00009 -305 001 108 5 00005 -330
1
Figure 36 02 54 5 00041 -239 02 108 5 00041 -239 2
Figure 37 008 108 3 00062 -221 008 108 5 00021 -268 008 108 67 00004 -340 008 108 8 0 ----
3
Expression 1 log d[Cr(VI)]dt = asdot log Chromite +log krsquorsquo log d[Cr(VI)]dt = 06908 log Chromite ndash 1899 Expression 2 log d[Cr(VI)]dt = bsdot log Birnessite + log krsquorsquorsquo
log d[Cr(VI)]dt = 0sdot log Birnessite ndash 23872 log d[Cr(VI)]dt = -23872
Expression 3 d[Cr(VI)]dt = 264E-4sdotpH2 ndash 414E-3sdotpH + 162E-2
82
Table 34 Extractions comparing total solution and extractable Cr and Fe from the solids
------------------Cr----------------- ------------------Fe---------------- microMolesdagger Total
mole Extraction
mole microMoles Total mole
Extraction mole
Chromite Ch 02 (Table 31) Terminated at 30 days Solutiona 0033 554 ---- 0 0 ---- Phosphateb 0000 05 1 0000 03 03 CaCl2
c 0013 388 87 0000 00 00 HCld 0002 53 12 0103 997 997 Bottle HCle 0000 00 0 0000 00 00
A Ch 02 Bt 54 (Table 31) Terminated at 30 days Solution 2977 910 ---- 0 0 ---- Phosphate 0048 16 18 0004 21 21 CaCl2 0055 18 20 0002 09 09 HCl 0161 54 60 0126 758 758 Bottle HCl 0004 01 2 0035 212 212
B Ch 02 Bt 108 (Table 31) Terminated at 30 days Solution 2837 914 ---- 0 0 ---- Phosphate 0039 14 16 0008 49 49 CaCl2 0060 21 25 0000 00 00 HCl 0142 50 58 0127 803 803 Bottle HCl 0004 01 2 0024 148 148
C Ch 08 Bt 108 (Table 31) Terminated at 17 days Solution 0877 894 ---- 0 0 ---- Phosphate 0020 23 22 0002 37 37 CaCl2 0030 34 32 0000 00 00 HCl 0040 46 43 0053 881 881 Bottle HCl 0003 03 3 0005 82 82
daggerMolarity values are normalized to the final solution analysis of each experiment based on volume of solution and mass of the sample in each experiment aSolution refers to the final solution analysis before the experiment was terminated bmicroMoles extracted from the batch experiment solids using a 1 mM phosphate buffer pH 7 3 h cmicroMoles extracted from the batch experiment solids using a 10 mM CaCl2 solution pH 7 3 h dmicroMoles extracted from the batch experiment solids using a 05 N HCl solution 3 h emicroMoles extracted from the batch experiment bottles using a 05 N HCl solution 15 h
83
Figure Captions
Figure 31 Scanning electron microscopic (SEM) images of chromite and birnessite at
varying magnifications prior to the batch experiments
Figure 32 A summary of Cr Fe and Mn released from chromite during extraction
experiments Solutions used for these experiments include 1) 10 H2O2 solution 2) 01
M KMnO4 solution 3) 10 M HF 4) 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 5) 1mM
BaCl2 solution () Fe concentration for the HF extraction is 106 mg L-1
Figure 33 pH and Eh measurements as a function of time (h) for the batch
experiments Chromite (Ch) and birnessite (Bt) suspension densities used for the batch
experiments (Table 31) are listed adjacent their corresponding symbol in the key
Figure 34 Eh and pH batch experiment analyses (Figure 33) plotted on a stability field
diagram for Cr (log a Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom
(1998)
Figure 35 Batch experiments evaluating variable chromite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr and Cr(VI) measured in solution are noted and illustrated with a dotted line and
included in Table 33
Figure 36 Batch experiments evaluating variable birnessite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr(VI) measured in solution are denoted and included in Table 33
Figure 37 Influence of pH on Cr(VI) generation by reaction of constant chromite (Ch
08 m2 L-1) and birnessite (Bt 108 m2 L-1) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key Rates of Cr(VI) measured in solution are included in Table
33
84
Figure 38 Rates of Cr(VI) in solution (Table 33) as a function of a) chromite
suspension density (Figure 35) b) birnessite suspension density (Figure 36) and c) pH
(Figure 37) The line in each diagram represents the best fit for these data points and the
equations for these lines are listed in Table 33 The dotted line deviating from the
dashed line in b) indicates the probable path at low birnessite concentrations
Figure 39 SEM images of the solid phases after the completion of the batch
experiments A (Ch 02 Bt 54) and B (Ch 02 Bt 108) are shown Chromite is the larger
grain with relatively flat surfaces whereas birnessite is the fine-grained rosettes located
adjacent the larger chromite grains
Figure 310 Chromium and Cr(VI) concentrations (microM) in solution as a function of
time (h) from serpentine soils obtained at Jasper Ridge (JR1 and JR2) and New Caledonia
(NC) Soils with the addition of 01 g birnessite are noted in the figure Soil and
birnessite masses used for each experiment are listed in Table 32
Figure 311 Iron analyses as a function of time (Figure 35 and 36) with the solubility
of amorphous Fe(OH)3 and goethite and the dominant Fe3+-OH complex at pH 5 Fe
solubilities are from Langmuir (1997) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key
Figure 312 a) Experimental values of Fe and Cr (Figures 35 and 36) are plotted A
dotted line denotes the ratio of Fe and Cr necessary for chromite congruent dissolution b)
Corrected FeCr ratios as a function of time (h) for data from Figures 35 and 36 taking
into account the concentration of Fe and Cr in secondary phases calculated from Table
34 c) Corrected Fe and Cr as compared to values predicted with congruent dissolution
of chromite
85
Figure 313 Three potential pathways of Cr release and oxidation for chromite-
birnessite interactions are listed
86
Figure 31
87
0
2
4
6
8
10
12
Hydrogen
Peroxid
e
K-Perman
ganate
Hydroflu
oric Acid
Hydroxy
lamine
Barium Chlorid
e
Con
cent
ratio
ns (m
g kg
-1)
Cr
Fe
Mn
Figure 32
88
Time (h)0 100 200 300 400 500
pH
2
3
4
5
6
7
8
9
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
Time (h)0 100 200 300 400 500
Eh (m
V)
100
200
300
400
500
600
Ch Bt ------(m2 L-1)------
Figure 33
89
pH2 3 4 5 6 7 8 9
Eh (m
V)
-05
00
05
10
15
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
HCrO4-
CrO42-
Cr3+
CrOH2+
Cr(OH)3Cr(
OH
) 2+
Upper limit of water stability
Lower limit of water stability
Ch Bt ------(m2 L-1)-----
Figure 34
90
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 108 008 108 002 108 001 108 02 --- --- 54
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
0
1
2
3
4
pH 5 pH 5
pH 5 pH 5 pH 5
Ch Bt ---(m2 L-1)---
Figure 35
91
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 54 02 108
00041 microM h
-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
00
02
04
06
08
10
00041 microM h
-1
pH 5
pH 5 pH 5 pH 5
pH 5 Ch Bt----(m2 L-1)----
Figure 36
92
Time (h)
0 50 100 150 200 250
Cr c
once
ntra
tion
( microM
)
00
02
04
06
08
10
pH 3pH 5pH 67pH 8
Time (h)0 50 100 150 200 250
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250M
n co
ncen
trat
ion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
00004 microM h-1
00021 microM h-1
0006
2 microM h
-1
00021 microM h-100
062 micro
M h-1
00004 microM h-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1
Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Figure 37
93
log Birnessite m2 L-1
05 10 15 20 25 30
log
d[C
r(VI
)]dt
microM
h-1
-40
-35
-30
-25
-20
pH 5
log Chromite m2 L-1-22 -20 -18 -16 -14 -12 -10 -08 -06
log
d[C
r(VI
)]dt
microM
h-1
-34
-32
-30
-28
-26
-24
-22
pH 5
pH1 2 3 4 5 6 7 8 9
d[C
r(VI
)]dt
microM
h-1
0000
0002
0004
0006
0008
0010
a)
c)
b)
Ch 008 m2 L-1 Bt 108 m2 L-1
Figure 38
94
Figure 39
95
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR1JR1 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR2JR2 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
01
02
03
04
05
NCNC w Bt
00044 microM h-1
00013 microM h-1
00044 microM h-1
00013 microM h-1
00013 microM h-1
00009microM h-1
pH 5
pH 5
pH 5
Figure 310
96
Time (h)0 100 200 300 400 500
log ΣF
e (aq)
microM
-10
-8
-6
-4
-2
0
2
4
02 54 02 108 008 108 002 108 001 108
Fe(OH)3(am)
Goethite
pH 5Ch Bt----m2 L-1----
Fe(OH)2(aq)+
Figure 311
97
Time (h)
0 100 200 300 400
FeC
r Rat
io
0
1
2
3
4
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
02 54 02 108 008 108 002 108 001 108
Chrom
ite C
ongr
uent
Dissolu
tion
Chrom
ite C
ongr
uent
Dissolu
tion
a)
b)
c)
Ch Bt----(m2 L-1)----
ChromiteCongruent Dissolution
Figure 312
98
Pathway 1 Chromite Dissolution
1a) Cr dissolution from chromite FeCr2O3(s) + 4H+ rarr 2Cr(OH)2+
aq + Fe2+(Oxyhydroxide) + H2O
1b) Adsorption of CrOH2+ on the surface of birnessite Cr(OH)2+
aq + MnO2(s) rarr Cr(III)-MnO2(s) + OH- 1c) Oxidation of CrOH2+ on the surface of birnessite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
1d) Desorption of Cr(VI) from birnessite Cr(VI)-Mn(II) + 4H2O rarr Mn2+
(aq) + HCrO4-(aq) + 7H+
Pathway 2 Birnessite Dissolution 2a) Mn4+ dissolution from Birnessite MnO2(s) + 4H+ rarr Mn4+
(aq) + 2H2O 2b) Mn4+ adsorption on the surface of chromite Mn4+
(aq) + FeCr2O4(s) rarr Mn(IV)-FeCr2O4 2c) Oxidation of Cr(III) on the surface of chromite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
2d) Desorption of CrO24- from chromite
Cr(VI)-Mn(II) + 4H2O rarr Mn2+(aq) + HCrO4
-(aq) + 7H+
Pathway 3 Solid Solution Reaction 3) Solid solution interaction between chromite and birnessite
Figure 313
99
CHAPTER 4
CHROMIUM GEOCHEMISTRY OF SERPENTINE SOILS
ABSTRACT
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
chemistry of Cr in serpentine soils including its protolith is reviewed focusing on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils are slightly acidic (average pH of ~6) contain a variety Fe(III) oxides
(magnetite and hematite) phyllosilicates (serpentine and chlorite) and clays (smectites
and vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1)
and Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Chromium(III) is the only valence state observed
in the serpentine soil solids however Cr(VI) has been identified in New Caledonia and
California serpentine soil solutions at concentrations below 30 microM The enrichment and
range of Cr concentrations in serpentine soils are directly related to the presence of Cr-
spinels specifically chromite and Cr-magnetite These spinels are resistant to weathering
and are preserved in the soil environment however oxidation of Cr(III) from Cr-spinels
by high valent Mn oxides or other strong oxidants is a potential source of Cr(VI)
identified in serpentine soil solutions Due to the chemically and physically resistant
nature of the Cr-spinels Cr-bearing silicates including clay minerals Cr-chlorite Cr-
garnet Cr-mica and Cr-epidote are more viable sources of Cr identified in vegetation
soil extractions soil solutions and related waters
100
INTRODUCTION
Serpentine soils a generic term used to describe any soil derived from an
ultramafic rock (peridotites and pyroxenites) or serpentinite (metamorphosed ultramafic
rock) are often characterized by 1) low concentrations of plant nutrients such as N P
and K 2) high concentrations of biologically toxic elements including Ni Co and Cd 3)
low ammonium acetate extractable CaMg ratios and 4) unique flora and physical
properties (Brooks 1987) Measured Cr concentrations in serpentine soils range from
~200 mg kg-1 to 6 Wt values enriched compared to non-serpentine soils lt200 mg kg-
1 (Kabata-Pendias and Pendias 1984 Schwertmann and Latham 1986 Brooks 1987)
Chromite and Cr-magnetite are common primary sources of Cr in these soils (Rabenhorst
et al 1982 Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989
Gasser and Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003)
however the plethora of factors relating the chemistry of Cr in serpentine soils and the
weathering of chromite has not been fully addressed using an integrative approach
Overall serpentinites and serpentine soils cover ~1 of the earthrsquos exposed
surface and these rocks and soils are commonly associated with ophiolite complexes
(Coleman 1977) Serpentinites and serpentine soils related to ophiolite complexes are
the primary interest of this paper A map illustrating the worldwide distribution of major
ophiolitic serpentinite bodies and their related soils is shown in Figure 41 A majority of
the sites are near convergent boundaries (Figure 41) Although these rocks and soils
cover a minimal area of the earthrsquos surface they are present in populated areas within the
Circum-Pacific Margin and the Mediterranean Due to the wide-distribution and
occurrence of serpentinites and serpentine soils weathering of these rocks and soils will
differ from location to location due to varying climatic conditions as well as the nature of
the parent material and other factors including topography biota and time (Proctor and
Nagy 1992) As a result the chemistry of these soils as well as the nature of Cr in the
serpentinites and serpentine soils will vary from site to site potentially allowing Cr to be a
source of non-anthropogenic contamination
The objective of this paper is to review the chemical characteristics of serpentine
soils globally evaluate mineralogical and compositional variations between several soils
derived from ultramafic sources and assess the geochemistry of Cr in diverse soil-
101
forming environments Additionally chemical and spectroscopic analyses were
performed on soils derived from ultramafic rocks collected from 1) New Caledonia
South Pacific 2) Eight Dollar Mountain Selma OR 3) Nickel Mountain Riddle OR 4)
Harvard Mine Jamestown CA 5) Jasper Ridge Biological Preserve Stanford CA and 6)
Pillikin Mine El Dorado CA shown geographically in Figure 41 A basalt-derived soil
from Hawaii is included in this study in order to compare a soil formed from a mafic
protolith to those from ultramafic protoliths The various field sites are shown in Figure
42 These soils allow a comparative examination of how soil-forming factors such as
climate topography biota parent material and time affect the chemistry of Cr
REVIEW
Chromium Geochemistry
Chromium is a group VIB transition metal with two common oxidation states
Cr(III) and Cr(VI) differentiated by noticeably different chemical characteristics and
toxicities It should be noted that Cr(II) has been observed in lunar basalt olivines and in
high temperature terrestrial olivines (Sutton et al 1993 Li et al 1995) however this
oxidation state is not observed in soil environments A majority of Cr commonly persists
in ultramafic rocks serpentine soils and related waters as Cr(III) a non-hazardous
species and a micronutrient in human nutrition (Daugherty 1992 Cohen et al 1993
Fendorf 1995 James et al 1997) Chromium(III) has the highest crystal field
stabilization energy for all transition metals strongly favors octahedral coordination and
has 3 d-electrons with a high spin state Chromium(III) is strongly hydrolyzed in
aqueous solutions and the predominant species based on pH redox potential and O2(aq)
are shown in Figure 43 Chromium(III) species are more prevalent under reducing
conditions and its mobility is increased only under strongly acidic conditions At pH
values greater than 9 the hydrolyzed Cr(III) species is negatively charged demonstrating
the amphoteric properties of Cr(III) (Figure 43) Due to the factors described above
Cr(III) forms low solubility compounds and adsorbs strongly on mineral surfaces and
organic matter at pHgt4 (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) For
example the adsorption of Cr(III) on HFO (hydrous ferric oxide) as a function of pH is
shown in Figure 43 illustrating an adsorption edge at a pH of ~4 (Leckie et al 1984
102
Stumm 1992) Chromium(III) oxidation to Cr(VI) has been observed in reactions
involving hydrogen peroxide (H2O2) and high valent Mn oxides (Eary and Rai 1987
Johnson and Xyla 1991 Fendorf et al 1992 Fendorf et al 1993 Rock et al 2001)
Chromium(VI) is a strong oxidizing agent a highly corrosive substance a toxin to
living cells and a Class A human carcinogen by inhalation (Daugherty 1992 Cohen et
al 1993 Fendorf 1995) The hydrolyzed Cr(VI) species are also shown in Figure 43
with respect to pH redox potential and O2(aq) Chromium(VI) is more favorable under
oxidizing conditions and its mobility is enhanced under basic conditions In solution
Cr(VI) is a weakly adsorbing (oxy)anion (HxCrO42-x) on common weathering products
meaning that it forms highly labile complexes that can be readily displaced by ions
including phosphate and sulfate (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Additionally the adsorption of CrO42- on HFO shown in Figure 43
demonstrates the anionic character in which the mol bound of CrO42- increases with
decreasing pH and an adsorption edge is at pH ~72 (Leckie et al 1980 Stumm 1992)
The reduction of Cr(VI) to Cr(III) is known to occur by reactions with Fe(II)aq Fe(II)-
containing minerals such as magnetite H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996)
Mineralogy of Serpentine Soils
Serpentine soils have been identified on every continent excluding Antarctica
(Figure 41) The occurrence and distribution of ultramafic rocks specifically related to
ophiolites are closely associated with convergent plate boundaries (Figure 41 Coleman
1977 Coleman and Jove 1992) Tectonic blocks of ultramafic rock are hypothesized to
be detached from the down-going slab of oceanic crust and incorporated into the
subduction meacutelange Subsequent serpentinization of peridotite forms low density
serpentine group minerals (245-250 g cm-3) resulting in the upward diapiric migration of
the serpentinite masses along zones of structural weakness (fault zones) and through the
denser enclosing subduction meacutelange (~265 g cm-3) For example North America has
numerous exposed ophiolites containing serpentinite located in western California and
Oregon where subduction zones were once active in the Early Cretaceous (Harden
1998)
103
Serpentinization of ultramafic rocks is a significant factor controlling mineral
assemblages and textures in serpentinite Hydration and oxidation of olivine and
pyroxene during serpentinization forms serpentine (lizardite chrysotile and antigorite)
talc chlorite tremolite brucite and magnetite (Coleman 1977 Brooks 1987 Coleman
and Jove 1992 Malpas 1992) Additionally magnetite formation creates favorable
conditions for native metal compounds (such as Fe Co and Ni alloys) to form (Coleman
and Jove 1992) As a serpentinite mass is emplaced the extensive physical and chemical
alteration of the serpentinite results in numerous crosscutting alteration veins and highly
sheared textures Ultimately the mineral assemblages and textures in serpentinites are
complex prior to pedogenesis due to the multiple phases of metamorphosis tectonic
deformation and weathering of the rock
Serpentine soils consist of a mixture of residual minerals from the soil protolith
eolian and illuvial minerals and minerals formed within the soil Additionally highly
weathered serpentine soils enriched in Fe-oxides are also termed laterites soils common
in tropical and subtropical regions Soil protolith minerals may include primary minerals
of ultramafic rocks including olivines pyroxenes and oxides (chromite and magnetite)
as well as a large variety of metamorphic minerals the most common being serpentine
(lizardite chrysotile and antigorite) chlorite talc and actinolite Silica phases occur as
eolian grains concretions andor net type structures or silica webs (Hotz 1964
Coleman 1977 Schwertmann and Latham 1986) Eolian and illuvial aluminosilicate
minerals are important sources of Al in serpentine soils an element of low abundance in
ultramafic rocks
Layer silicates observed in serpentine soils are varied in type and abundance
They include smectites (often as montmorillonite) vermiculite chlorite brucite
serpentine and a variety of intrastratified clays (Veniale and Van der Marel 1963 Hotz
1964 Schellmann 1964 Wildman et al 1968 Rabenhorst et al 1982 Schreier et al
1987 Graham et al 1990 Bulmer et al 1992 Gasser and Dahlgren 1994 Lee et al
2003) Compositional variations crystal structures and parageneses of layer silicates
have been characterized for a number of serpentine soils (Wildman et al 1968
Rabenhorst et al 1982 Schreier et al 1987 Graham et al 1990 Bulmer et al 1992
Lee et al 2003) In addition to the layer silicates goethite hematite maghemite and
104
amorphous Fe (oxy)hydroxides are soil-forming phases in serpentine soils (Coleman and
Jove 1992) Gibbsite is not a common phase in most serpentine soils however limited
isomorphic substitution of Al(III) in goethite and hematite was noted in New Caledonia
(Schwertmann and Latham 1986) Manganese oxides such as birnessite (δ-MnO2) are
also present in serpentine soils and often coat mineral surfaces
Sources of Chromium for Serpentine Soils
Chromium primarily occurs in oxide and silicate phases in ultramafic rocks and
serpentinites Chromium-spinel often referred to as chromite is a common primary
(igneous) spinel in mafic and ultramafic rocks and often resists low-grade metamorphic
processes related to serpentinization (Hoffman and Walker 1978 Malpas 1992)
Isomorphic substitution of Al3+ Fe3+ and Ti4+ into the octahedral sites and Mg2+ Ni2+
Zn2+ and Mn2+ into the tetrahedral sites is common however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987b Pan and Fleet 1991 Sack and Ghiorso 1991 Burkhard 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996 Mathiesen 1999) Serpentinite chromite compositions reported in the literature are
plotted in Figure 44 to demonstrate the wide-range of Cr concentrations possible for Cr-
spinels Isomorphic substitution of octahedral Cr and octahedral Al for these chromite
compositions are also shown in Figure 44 Serpentinite pyroxenes containing Cr include
augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) (Grapes 1981 Von Knorring et al
1986 Treloar 1987a Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996)
Magnetite and serpentine minerals may incorporate varying amounts of Cr into
their structures during serpentinization Chromium(III) from primary Cr-containing
phases may isomorphically substitute into the Fe(III) site of magnetite due to size
(octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge similarities Chromium
concentrations in magnetite have been observed as high as 13 Wt (Oze et al 2003)
Chromate minerals such as crocoite (PbCrO4) are extremely rare in serpentinites and the
occurrence of these minerals are mainly limited to specific localities such as in Australia
Tasmania and the Urals (Williams 1974 Crane et al 2001) Serpentine minerals have
105
been reported to contain as much as 1500 mg kg-1 Cr based on spectrographic analyses
by Page and Coleman (1967) however serpentine minerals typically contain only trace
amounts of Cr
Hydrothermal and CO2 metasomatism of ultramafic rocks during the evolution of
convergent plate margins commonly produce Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite lt19 Cr Wt ) uvarovite (Cr-garnet lt20 Cr Wt )
tawmawite (Cr-epidote lt13 Cr Wt ) and kaumlmmererite (Cr-chlorite lt10 Cr Wt ) as
shown in Figure 44 (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974 Neiva
1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981 Max et al
1983 Nutman et al 1983 Raase et al 1983 Von Knorring et al 1986 Kerrich et al
1987 Treloar 1987b Treloar 1987a Morand 1990 Pan and Fleet 1991 Schandl and
Wicks 1993 Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995
Mathiesen 1999 Proenza et al 1999) Figure 44 illustrates the relationship of Cr and
Al substitution for Cr-silicate compositions based on Cr and Al Wt Additionally Cr-
silicate compositions are plotted based on octahedral Cr and octahedral Al in Figure 44
Compositions for Cr-muscovite Cr-epidote and Cr-garnet plot near the tie lines
connecting Cr and Al end-member minerals Chromium isomorphically substitutes for
octahedral Al in these silicates due to similarities in charge octahedral preference and
octahedral radii (Al3+ = 061 Aring) The numerous isomorphic substitutions possible for Cr
and Al in the chlorite structure results in analyses that do not plot near the tie line
connecting clinochlore and kaumlmmererite Overall any of these minerals are potential
sources for Cr in soils produced by the weathering of serpentinites and their hydrothermal
and CO2 metasomatic derivatives
Chromium in Serpentine Soils
Chromium concentrations vary widely in serpentine soils worldwide (Table 41)
New Caledonia soils appear to be the most Cr-enriched serpentine soil with a Cr
concentration of ~60000 mg kg-1 (Schwertmann and Latham 1986) however Jaffre
(1980) report values ranging from 6300 to 80000 mg kg-1 for New Caledonia soils
Some serpentine soils have measured Cr concentrations as low as 29 and 66 mg kg-1
atypical of most values reported in Table 41 (Proctor and Woodell 1971 Brooks et al
106
1992 Jeffrey 1992) Chromium concentrations as a function of depth for serpentine soil
profiles in Maryland (Rabenhorst et al 1982) Tehama County CA (Gough et al 1989)
the Klamath Mountains CA (Hotz 1964) Newfoundland Canada (Roberts 1992)
Harvard Mine CA (this study) and Jasper Ridge CA (Oze et al 2003) are reported in
Figure 45 The pedogenic process(es) responsible for the varying Cr concentrations as a
function of depth for each soil have not been characterized however Oze et al (2003)
determined that a majority of Cr in Jasper Ridge serpentine soils is the result of
weathering-resistant residual Cr-bearing minerals from the protolith mainly chromite
and Cr-magnetite Additionally Oze et al (2003) demonstrated that soil samples from
Jasper Ridge contain higher concentrations of Cr with respect to their corresponding
bedrock due to the Cr-spinels accumulating over time The observation that Cr
concentrations decrease with depth towards the protolith is characteristic for most of the
profiles in Figure 45
In addition to examining total Cr concentrations chemical extractions performed
on serpentine soils provide additional information with regards to Cr availability and the
minerals and complexes in which Cr may reside in the soils Table 42 summarizes a
number of extraction experiments performed on serpentine soils For a majority of the
soils listed in Table 42 total Cr concentrations for the serpentine soils were not
measured however these extractions do provide a quantitative assessment of where Cr
may be bound in soil (such as silicates organic matter and secondary Fe-bearing phases)
Extractions performed by Schwertmann and Latham (1986) were able to remove ~20000
Cr mg kg-1 from the serpentine soil using a dithionite-citrate bicarbonate (DCB) solution
which preferentially dissolves crystalline Fe-oxides Solutions such as concentrated
nitric and hydrochloric acid were only able to remove a maximum of ~1400 Cr mg kg-1
from Unst serpentine soils (Shewry and Peterson 1976) which preferentially dissolve
Cr-bearing silicates and to a lesser extent the Fe-oxides Where bulk soil Cr values were
measured extraction experiments are only able to remove a small portion (lt1) of the
total Cr present in the serpentine soils (Table 42) (Gasser and Dahlgren 1994 Oze et al
2003) Based on the extractions presented in Table 42 it appears that Cr(III) resides in
minerals and phases resistant to common extraction techniques
107
In serpentine soils Fe(III) oxides and Fe(III) hydroxides provide a host of
structural and surface sites available to sequester Cr(III) released to soil solutions by
dissolution of Cr-bearing minerals during soil-forming processes Substitution of Cr(III)
for Fe(III) in Fe-oxides Fe-(oxy)hydroxide minerals and amorphous compounds is
expected based on crystallographic considerations synthesis experiments and analysis of
extracts from selective dissolution of Fe(III) compounds in serpentine soils
(Schwertmann and Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990
Gerth 1990 Gasser and Dahlgren 1994) Chromium(III) also readily sorbs on surfaces
of Fe(III) oxides and (oxy)hydroxides via inner-sphere surface complexes (Figure 43)
and forms low-solubility precipitates (Charlet and Manceau 1992) The low
concentration of Cr(III) in soil solutions has been attributed to the low solubility of
Fe(III)-Cr(III) amorphous hydroxides (Sass and Rai 1987)
Layer silicates provide structural and surface sites capable of uptaking Cr(III)
Octahedral Cr(III) in clays such as smectites and vermiculites is expected and Cr
concentrations in the clay-sized fraction can achieve concentrations of ~1000 mg kg-1 in
serpentine soils (Oze et al 2003) Sorption of Cr(III) and Cr(VI) on edge and interlayer
sites of phyllosilicates have been evaluated by Ilton and Veblen (1994) Ilton et al (2000)
and Brigatti et al (2000) indicating that layer silicates (montmorillonite chlorite biotite
phlogopite and muscovite) provide sites for both Cr(III) sorption and Cr(VI) reduction
Kaupenjohann and Wilcke (1995) concluded that Cr in serpentine soils from Bavaria is
residually bound mainly in silicate structures including clay minerals based on extraction
experiments
Soil Solutions and Related Water
The addition of water gradually changes the spatial and temporal characterization
of pore fluids in soils including the ionic strength pH and redox potential Leaching of
a soil should progressively enrich the soil with less chemically mobile elements such as
Cr and Fe Elements identified in soil solutions include Mg Ca Fe Ni and Cr and the
ionic strengths of serpentine soil solutions are typically low (lt001) (Anderson et al
1973 Johnston and Proctor 1981 Gasser and Dahlgren 1994) Soil solutions obtained
from serpentine soils using tension-free lysimetry demonstrate Cr concentrations ranging
108
from 01 to 32 micromol L-1 (Gasser and Dahlgren 1994) This study also revealed a
majority of Cr Fe Mg and Ni in extracted soil waters were in the form of colloids
suggesting that nanoparticles are an important factor of Cr translocation in soils
The pH of a soil is a dominant factor controlling the nature of many chemical and
microbial reactions including the geochemistry of Cr (Brady and Weil 1999) Ranges in
pH for serpentine soils worldwide are plotted in Figure 46 The highest soil pH was
measured in a South Africa serpentine soil with a value of 9 (Cole 1992b) whereas the
lowest pH was reported in Ireland with a value of 38 (Jeffrey 1992) This wide range of
pH is a function of the adsorbed H+ and Al3+ the base-forming cations the organic
matter and the mineralogy present within each soil The pH range of serpentine soils
shown in Figure 46 is plotted on the stability diagram in Figure 43 demonstrating that
both Cr(III) and Cr(VI) aqueous species are favorable The Eh range for the serpentine
soils in Figure 3 is based on the soils being in contact with the atmosphere as well as
being a transitional environment Although the activity of H+ varies by four orders of
magnitude the pHs of most serpentine soils are slightly acidic (pH ~6) which is near that
of average rainwater (pH~57 Figures 43 and 46) Additionally serpentine soil pH
values have been observed to increase with depth towards the protolith for a number of
soils (Rabenhorst et al 1982 Gough et al 1989 Gasser and Dahlgren 1994) The
increase in pH with respect to depth may be the result of rainwater being progressively
influenced by the mineral chemistry and hydrolysis reactions as it descends through the
soil towards the protolith
Groundwater from serpentinized ultramafic rocks have pH values ranging from 8
to 12 (Barnes et al 1972) and these waters have been divided into two categories as
shown in Figure 46 Mg bicarbonate (Mg2+-HCO3-) type and Ca hydroxide (Ca2+-OH-)
type (Barnes and ONeil 1969 Barnes et al 1972) The Mg bicarbonate waters (pH~8)
are characteristic of most surface waters and shallow groundwaters from serpentinites
and ultramafic rocks The Ca hydroxide waters (pHasymp10-11) are associated with
incompletely serpentinized peridotites and these fluids are comparatively low in Mg2+
Fe2+ Mn2+ Crtotal and SiO2total The Ca2+ in solution originates from the serpentinization
of Ca-rich orthopyroxene (Barnes et al 1972) The pH of serpentinite groundwaters are
109
plotted on the Cr stability in Figure 43 demonstrating that both Cr(III) and Cr(VI) are
potential valence states in these waters
Chromium predominates as Cr(III) in ultramafic rocks and serpentine soils
however Cr(VI) has been identified in serpentine soil solutions and serpentinite
groundwater in California and New Caledonia at concentrations of lt20 microg L-1 (Gough et
al 1989 Becquer et al 2003) Calcium hydroxide waters (pH~11) collected from the
same spring evaluated by Barnes et al (1972) in the Cedars ultramafic body (Cazadero
CA) contains 12-22 Cr(VI) microg L-1 and no Cr(III) (this study) Additionally water from
chromite-containing ultramafic rocks in the Siniktanneyak Mountain ultramafic complex
of Alaska with pH values ranging from 67 to 93 were measured with low total Cr
concentrations lt2 Cr microg L-1 (Taylor et al 1998) A mechanism for Cr(III) oxidation in
serpentine soils and related water suggested by (Gough et al 1989 Becquer et al 2003)
may be the result of high valence Mn oxides common in serpentine soils High valence
Mn-oxides such as birnessite (δ-MnO2) are naturally occurring oxidants capable of
oxidizing Cr(III) to Cr(VI) at pHlt9 (Eary and Rai 1987 Fendorf et al 1992 Fendorf et
al 1993) It is also important to note that Cr(VI) reduction is hypothesized to occur in
soils due to electron transfer involving organic matter and Fe(II)-bearing minerals such as
magnetite (Peterson 1996 Fendorf et al 2000) Whether toxic concentrations of Cr(VI)
can be present in serpentine soils and related water over geologic time is important to
consider especially when high valence Mn oxides Fe(II)-bearing minerals and organic
matter are capable of significantly influencing the oxidation state of Cr
Serpentine Soil Vegetation
Vegetation associated with serpentine soils has received considerable attention
due to the apparent infertility of these soils The inhabitability of serpentine soils for
most plants has been attributed to the imbalance of available Mg relative to Ca or low
CaMg ratios (Walker 1954) the deficiency of plant nutrients such as K N and P
(Turitzin 1991) elevated heavy metal concentrations such as Ni and Cr (Robinson et al
1935) and poor drainage (Brooks 1987) Linking plant productivity to toxic levels of
Cr in the soil has not been fully resolved (Brooks 1987 Kruckeberg 1992) however a
majority of Cr in the soils is not available for plant uptake due to being in high stable and
110
weathering resistant phases discussed in the previous sections (Table 42) In general Cr
concentrations in plants typically do not exceed 100 mg kg-1 (Brooks 1987)
Table 43 provides a global list of serpentine soil plant species and their measured
Cr concentrations with an emphasis on Cr hyper-accumulators ie plants able
incorporate higher than average concentrations of Cr A majority of these measurements
are based on the dry weight of Cr Aerobryopsis longissima from New Caledonia has the
highest reported Cr concentration with values ranging from 200 to 700 mg kg-1 (Lee et
al 1977) Other noteworthy plants include the Lophostachys villosa from Brazil (Brooks
et al 1992) Arenaria humifusa from Newfoundland (Roberts 1992) the Calluna
vulgaris from Unst (Shewry and Peterson 1976) with Cr concentrations of 203 530 and
417 mg kgndash1 respectively Even mosses such as the Funaria flavicans (Maryland)
incorporate Cr concentrations as high as 90 mg kg-1 (Shaw 1991) By examining
specific parts of the plant Gough et al (1989) determined that Cr concentrations are
highest in the basal burl (600 mg kg-1) of the manzanita bush compared to the roots (30
mg kg-1) stems (20 mg kg-1) and leaves (20 mg kg-1) The study by Gough et al (1989)
demonstrates that Cr is not homogeneously distributed throughout the plant and that
evaluating a leaf versus a stem could result in significantly different Cr concentrations
Soil Organic Matter
Soil organic matter (SOM) including the biomass (living organisms such as
bacteria) detritus (nonliving plant residues) and humus (amorphous and colloidal
mixtures of organic substances) affects the geochemistry and buffering capacity of
serpentine soils Furthermore biologic reduction of Cr(VI) to Cr(III) via cometabolic
and metabolic pathways coupled with carbon oxidation has been demonstrated for a
number of bacterial strains including aerobes (Gopalan and Veeramani 1994 Park et al
2000) and anaerobes (Turick et al 1996 Tebo and Obraztsova 1998) Additionally
aqueous Fe(II) produced by microbial Fe(III) reduction is a known biogenic pathway for
the reduction of Cr(VI) (Fendorf et al 2000) Fe(III)-Cr(III) (oxy)hydroxides are the
resulting secondary phases (Rai et al 1989 Patterson and Fendorf 1997 Hansel et al
2003) Bacteria oxidizing Cr(III) to Cr(VI) has not been observed
111
Most serpentine soils in temperate regions appear to be either Mollisols or
Alfisols which often contain high concentrations of SOM Alexander et al (1989) and
Bulmer et al (1992) reported that serpentine soils can contain organic carbon
concentrations in excess of 50 mg g-1 Humus has a high cation exchange capacity and
provides much of the pH buffering capacity in these soils (Brady and Weil 1999)
Organic acids such as fulvic and humic acids are known to complex cations such as
Cr3+ Fe3+ and Mn2+ forming stable organometallic complexes (Kaupenjohann and
Wilcke 1995 Brady and Weil 1999) Additionally metal leaching experiments
performed by Schreier et al (1987) established that organic acids produced by soil
organic matter such as oxalic acid and citric acid were extremely effective in releasing
Cr Mg Fe Ni and Co from serpentinite rocks and sediments
CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
Serpentine soils were obtained from California Oregon and New Caledonia and
analyzed to further evaluate the factors influencing the Cr geochemistry in serpentine
soils Analyses for these soils include pH electrical conductivity total organic matter
total N ammonium acetate extractable Ca K Mg and Na soil texture soil color
chemical composition of the soil and the chemical composition of Cr-bearing minerals
The general soil characteristics are described in the following section and the analytical
methods are explained in Appendix 41 Although a majority of these soils are located in
the western United States factors influencing the pedogenesis of these soils vary
considerably allowing a comparative analysis of how the fate of Cr might be influenced
by climate topography biota parent material and time
Sample Locations and General Soil Characteristics
Six serpentine soils were collected from locations shown in Figures 41 and 42 to
evaluate the chemistry of Cr in serpentine soils These soils include 1) a laterite (Oxisol)
collected in New Caledonia South Pacific 2) a laterite (Oxisol) exposed at Eight Dollar
Mountain Selma OR 3) a laterite (Oxisol) near Nickel Mountain Riddle OR 4) a
serpentine soil (Inceptisol) with a CO2-metasomatized serpentinite protolith from
Harvard Mine Jamestown CA 5) a serpentine soil (Mollisol) obtained from a preserved
112
grassland at Jasper Ridge Biological Preserve Stanford CA and 6) a serpentine soil
(Inceptisol) collected in the proximity of Pillikin Mine El Dorado CA Lastly an Oxisol
formed from basalt obtained in Oahu Hawaii is also included for comparison
These seven sites include a range of soil environments from different protoliths
under varying conditions of rainfall temperature and topography Each soil was collect
near the highest and topographically level point for each serpentine soil site Serpentine
grassland soils at Jasper Ridge formed under Mediterranean-like climates (xeric moisture
regime) whereas higher temperatures and less rainfall (ustic moisture regime)
characterize soil formation at the Harvard and Pillikin mines in the Sierra Nevada
Foothills metamorphic belt The vegetation of these three sites is dominantly grasses with
some manzanita These modern soil environments contrast with the Miocene Ni and Cr-
rich laterite soils of Nickel Mountain and Eight Dollar Mountain that were probably
formed under an aquic moisture regime near sea level in a subtropical climate Grasses
manzanita and pine trees are the dominant forms of vegetation at Nickel and Eight
Dollar Mountain An ultramafically-derived laterite formed under tropical conditions
from New Caledonia was collected The vegetation and topography related to the New
Caledonia laterite was not recorded during sampling An Oxisol from Hawaii is also
included in this study to evaluate the differences between a highly weathered soil
originating from an ultramafic rock and one from a mafic rock (basalt) under tropical
conditions The Hawaiian soil was collected from a topographically level area however
the vegetation related to this site was not noted
BedrockProtolith Mineralogy
A broad characterization of the protoliths for the serpentine soils is briefly
summarized due to the bedrock not being collected from all the field sites New
Caledonia is composed of variably serpentinized peridotites predominantly harzburgites
and some dunites (Leblanc 1995) Nickel Mountain and Eight Dollar Mountain are
predominantly serpentinized harzburgite and dunite The bedrock for soils collected at
Harvard Mine is mainly composed of quartz ankerite and mariposite (also referred to as
QAM) caused by the CO2 metasomatism of serpentinite associated with gold
mineralization (Bohlke 1989) The average serpentinite at Jasper Ridge is composed of
113
serpentine (mainly lizardite and antigorite) chlorite talc magnetite and chromite (Oze et
al 2003) however brucite has been reported as a major constituent in related Franciscan
Complex serpentinites (Hostetle et al 1966) Pillikin Mine serpentinites dominantly
consist of talc and serpentine (lizardite antigorite and chrysotile) and these serpentinites
contain podiform chromite deposits (Cater et al 1948) Mafic bedrock for the Hawaiian
Oxisol is tholeittic basalt
Soil Chemistry and Physical Properties
As shown in Figure 47 the colors of the collected soils range from the red New
Caledonia laterite to the greenish gley of the Pillikin Mine Inceptisol The Munsel color
values are also shown in Figure 47 The dark grey color of the Jasper Ridge soil and the
greenish gley of the Pillikin Mine soil are atypical of soils derived from serpentinites due
to the lack of any apparent redness The range of these colors is often attributed to the
presence of Fe-oxides however organic matter and Mn oxides are also dominant factors
controlling color (Brady and Weil 1999)
Chemical properties for the six serpentine soils and the Hawaiian soil are listed in
Table 44 The slightly acidic pH values range from 53 to 69 similar to those shown in
Figure 43 and 46 The New Caledonia laterite is the most acidic soil with a pH of 53
compared to the Harvard Mine carbonate-buffered soil at 69 The pH of the Hawaiian
soil is also slightly acidic with a pH of 604 The electrical conductivities (EC) of the
soils range from 01 to 07 dS m-1 a range indicative of relatively few dissolved inorganic
solutes The highest organic matter content is in the very dark grey Jasper Ridge soil
with a value of 32 and the lowest is in the red New Caledonia laterite (05) The
Hawaiian soil has an organic matter content of 1 Nitrogen concentrations for all the
soils are lt012 with Jasper Ridge having highest N concentration which is consistent
with the soil organic matter content
Ammonium acetate extractable Ca K Mg and Na are also listed in Table 44
Magnesium concentrations are gt6923 mg kg-1 in all of the serpentine soils The lowest
concentration of Mg was measured in the basalt-derived Hawaiian Oxisol with a value of
168 mg kg-1 Calcium concentrations in the serpentine soils range from 167 to 998 mg
kg-1 The 998 mg kg-1 of Ca in the Harvard Mine soil is directly related to the abundance
114
of ankerite (gt20 of the bedrock) Potassium concentrations are low for most of the soils
(lt55 mg kg-1) however Jasper Ridge and Harvard Mine soils have the highest
extractable K with values of 237 and 301 mg kg-1 respectively The extractable K value
for the Harvard Mine soil is due to the K-metasomatism related to gold mineralization
involving the serpentinite soil protolith The extractable concentration of Na is lt44 mg
kg-1 for all the serpentine soils whereas the Hawaiian soil has the highest Na (53 mg kg-
1) due to plagioclase and basaltic glass
Textures for the soils range from clay loam clay loam to sandy clay loam (Table
44) The laterite soil in New Caledonia and Oxisol in Hawaii are extremely different in
which the Hawaiian soil is mostly clay (66 clay) compared to the loam at New
Caledonia (25 clay) Additionally both ancient laterites (Eight Dollar Mountain and
Nickel Mountain) have clay textures Harvard Mine Jasper Ridge and Pillikin Mine
soils are all loamy in character
Major and trace element compositions for the soils are listed in Table 45
Chromium concentrations in the serpentine soils range between 827 and 9528 mg kg-1
The Eight Dollar Mountain laterite has the highest Cr concentration and the New
Caledonia laterite contains the lowest Cr concentration The Cr concentration for the
New Caledonia soil is lower than the values reported in Table 41 The Hawaiian soil has
a Cr concentration well below the analyzed serpentine soils with 388 Cr mg kg-1 but still
greater than the 200 mg kg-1 characteristic of serpentine soils Nickel concentrations in
the serpentine soils range from 1895 to 14100 mg kg-1 with the highest concentration
being identified in the Nickel Mountain laterite Manganese concentrations range
between 257 (New Caledonia) to 3017 (Eight Dollar Mountain) mg kg-1 in the serpentine
soils Element concentrations in the Hawaiian Oxisol contrast from the serpentine soils
with higher Al Cu P Pb and Zn concentrations
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in the soil solids for Jasper Ridge New Caledonia Pillikin Mine Eight Dollar
Mountain and Harvard Mine Although the Cr concentrations in these soils exceed 200
mg kg-1 (Table 45) X-ray absorption near-edge spectra (XANES) shown in Figure 48
illustrate Cr(VI) levels are below the detection limits of this method (~10 Cr mg kg-1) due
to the lack of a pre-edge feature The box in Figure 48 illustrates where a Cr(VI) pre-
115
edge peak should be present Chromium(VI) concentrations in these soils are possibly
present in the parts per billion range
Soil Mineralogy
Backscattered electron images and microprobe analyses were used to identify the
Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49
and corresponding Cr-bearing mineral compositions are listed in Table 46 The New
Caledonia soil exhibits a wide range of particle sizes and a Fe-(oxy)hydroxide was
identified containing 165 Cr2O3 Wt Chromium-spinel has been identified in the New
Caledonia laterite by Becquer et al (2003) Eight Dollar Mountain soils exhibit a coarse
texture and both Cr-magnetite (2387 Cr2O3 Wt ) and chromite (3343 Cr2O3 Wt )
are present Nickel Mountain soils are also coarse and chromite (1315 Cr Wt ) was
identified Soils at Harvard Mine contain numerous chromite grains and the mineral
trevorite (a Ni-bearing spinel) was identified with 3991 Cr2O3 Wt and 2054 NiO Wt
Chromite and magnetite are abundant in soils at Jasper Ridge as well as a fine-
grained mixture of chromite magnetite and an unidentifiable silicate with up to 1878
Cr2O3 Wt (Figure 49) Veins of Cr-chlorite crosscut the spinel-silicate mixture in
Figure 49 and this Cr-chlorite is the only Cr-silicate identified using this method
Chromite (5355 Cr2O3 Wt ) is the dominant Cr-bearing mineral detected in soils at the
Pillikin Mine and these grains contain the highest Cr concentrations compared to the
other minerals in the seven soils The Hawaiian Oxisol possesses ulvoumlspinels (007
Cr2O3 Wt and 2465 TiO2 Wt ) which are evenly distributed throughout the soil
Overall microprobe observations reveal that the Cr-bearing spinels are present in every
soil Additionally Cr-bearing spinels at Jasper Ridge have been shown to undergo
incongruent dissolution progressively enriching the spinel towards a FeCr2O4 end-
member (Oze et al 2003)
DISCUSSION
The physical and chemical characteristics of serpentine soils are directly related to
the alteration and tectonic history of the serpentinite as well as the soil-forming factors
including the parent material climate biota topography and time Identifying Cr-
116
bearing minerals their chemical compositions and abundance in the soils allows a more
definitive explanation for Cr enrichment in these soils as compared to total Cr
concentrations listed in Tables 41 and 45 and sequential extraction experiments reported
in Table 42 Detrital chromite grains from the serpentinite protolith are resistant to
weathering and accumulate in the soil over time Although serpentine soils contain Cr
concentrations typically gt200 mg kg-1 the presence of Cr-spinels and their resistance to
weathering demonstrate that a majority of the Cr is not chemically available to interact
within the soil The role of organic matter and vegetation are important factors related to
the chemistry of serpentine soils however their role for influencing a majority of the Cr
geochemistry is limited due to the nature of the Cr-spinels
Chromium and Serpentine Soils
Total Cr concentrations exceed 200 mg kg-1 for most of the serpentine soils
reported in Table 41 and in the six serpentine soils of this study (Table 45) These Cr
concentrations can be accounted for by assessing the Cr-bearing minerals associated with
these soils and their relative abundance Small grains of Cr-spinels or chromite identified
in the soils (Figure 49 and Table 46) have the potential to appreciably impact Cr
concentrations Table 47 demonstrates the mineral masses and volumes of Cr-bearing
minerals required to cause predetermined Cr concentrations in soil samples A minimal
amount of chromite (lt02 of a soil sample mass or 005 of the volume of serpentine)
can result in Cr concentrations of 1000 mg kg-1 A few grains of chromite could easily
be responsible for the Cr concentrations in Tables 41 and 45 In fact residual grains of
Cr-enriched spinels are observed in every soil analyzed (Figure 49 and Table 46)
Ultimately grains of spinel in the soil do appreciably impact Cr concentrations however
they do not provide a readily available source of Cr based on extraction experiments
performed by Oze et al (2003) Serpentine minerals could also provide minor Cr
concentrations to the soils however soil extractions have consistently shown that a
majority of the Cr is in highly resistant phases capable of resisting dissolution even in
concentrated hydrochloric acid nitric acid and hydroxylamine with acetic acid (Table
42) Chromium end-member minerals including eskolaite Cr-muscovite (fuchsite) Cr-
garnet (uvarovite) Cr-epidote (tawmawite) and Cr-chlinochlore (kaumlmmererite) are also
117
included in Table 47 to demonstrate that relatively low abundances of these minerals are
all that is necessary to increase Cr concentrations in rocks and soils Chromium-
clinochlore is often observed developing adjacent Cr-spinels in serpentinites and is
perhaps a more accessible mineral source of Cr in the serpentinites compared to Cr-
spinels (Phillips et al 1980 Schreyer et al 1981 Kerrich et al 1987 Treloar 1987a
Pan and Fleet 1991 Schandl and Wicks 1993 Christofides et al 1994 Challis et al
1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Oze et al 2003)
As shown in Table 44 and Figures 43 and 46 the chemical properties (pH EC
N etc) of serpentine soils are not correlated to Cr concentrations measured in the soil
solids (Tables 41 Figure 45) due to Cr-spinels accounting for a majority of the Cr
(Figure 49 and 410 Tables 46 and 47) However extraction experiments and soil
solution analyses demonstrate that a limited fraction of the total Cr (lt1) is extractable
If only 1 of the Cr in a serpentine soil is extractable a serpentine soil containing 10000
Cr mg kg-1 could possibly contribute up to100 Cr mg kg-1 still a significant Cr
concentration The chemical properties of the soil such as pH will influence this limited
fraction of the Cr in the soils Acidic pHs may aid in mobilizing Cr(III) however this
will only occur at low pHs (lt4) which exceed the range of most serpentine soils (Figures
43 and 46) pH values greater than 8 like those in some serpentine soils and
serpentinite groundwater may help sustain Cr(VI) concentrations in solution (Figure
43) In addition to pH organic matter in the soils may be a significant factor related to
the mobility of Cr due to the production of organic acids promoting the dissolution of the
Cr-bearing phases but particulate organic matter is known to bind Cr(III) in stable stable
organometallic complexes (Kaupenjohann and Wilcke 1995 Brady and Weil 1999)
Evaluating the valence state of Cr is crucial for understanding the chemistry as
well as the potential toxicity related to serpentine soils The soil solids do not contain
detectable concentrations of Cr(VI) (Figure 48) however Cr(VI) concentrations are
present in soil solutions and groundwater at values lt30 microg L-1 (Gough et al 1989
Becquer et al 2003) A majority of serpentinite Cr-bearing minerals only contain
Cr(III) therefore there must be a reaction or process by which Cr(III) is oxidized in a
water saturated soil High valence Mn oxides such as birnessite are present within
serpentine soils and provide a feasible explanation for Cr(III) oxidation Additionally
118
Mn oxides or other strong oxidants in the soil might accelerate the weathering of Cr-
bearing minerals including chromite and generate the Cr(VI) concentrations observed in
soil solutions and groundwater After Cr(III) is oxidized Cr(VI) is a chemically mobile
species and would be transported away from the serpentine soils before it can be
observed in the solid phase Additionally Cr(VI) may be subsequently reduced by
reacting with organic matter or Fe(II)-bearing minerals such as magnetite which are
present in a majority of serpentine soils Perhaps the best way to evaluate the potential of
serpentine soils being a non-anthropogenic source of Cr contamination is to monitor
Cr(III) and Cr(VI) in the soil solutions over time as demonstrated by Gasser and
Dahlgren (1994)
It is curious that plants do uptake Cr in serpentine soils especially when Cr is
tightly bound in the soil matrix (Table 42) In order for plants to uptake Cr it must be in
a form to be absorbed by plant roots Soil pH in the rhizosphere can be altered by plant
roots by as much as 2 units within two days (Fischer et al 1989) This acidification is
known to increase the solubility of the minerals in the soils and could serve as a
mechanism by which plants can incorporate this relatively insoluable element Another
possibility is that Cr(III) in the soil is oxidized by either organic or inorganic means and
then incorporated into the plant in its more chemically mobile and bioavailable form
Cr(VI) How and why plants incorporate as much Cr as they do needs to be examined
more thoroughly
Chromium Spinel in Soils and Rocks
Compositions of chromite identified in serpentinites worldwide (Figure 44) and
Cr-bearing spinels identified in the six serpentine soils are compared in Figure 10 in order
to determine whether the soil spinels are geochemically similar to those in serpentinites
This comparison is ascertained by examining how Al Fe and Mg Wt values varied
with respect to Cr Wt Trends for magnetitehematite chromite and the
microcrystalline spinel-silicate mixture observed in the soils are noted in each diagram in
Figure 410 Additionally the highest Cr concentrations for soil chromite
magnetitehematite and the microcrystalline spinel-silicate mixture are ~40 Cr Wt
~25 Cr Wt and ~15 Cr Wt respectively Based on Al Wt soil
119
magnetitehematite deviates significantly from the chromite trend with Al Wt values
lt5 Magnetitehematites are enriched with Fe relative to serpentinite and soil chromite
and Mg concentrations in the soil chromite are slightly higher than serpentinite chromite
Soil magnetitehematite and chromite are similar by demonstrating a linear trend when
Al Fe and Mg are summed The microcrystalline spinel-silicate mixture (CSM) does
not follow this trend because of Si present in this mineral mixture Overall soil
chromites are similar to serpentinite chromite and soil magnetiteshematites can be
differentiated from the chromites when Cr Wt is plotted against Al Fe and Mg Wt
CONCLUDING STATEMENT
Serpentine soils are abundant among convergent margins and the chemistry of
these soils is ultimately a function of the ultramaficserpentinite rock mineralogy and
texture as well as all the common soil-forming factors Although ultramafic rocks and
serpentinites produce groundwater that is highly basic most serpentine soil pHs are
slightly acidic Although Proctor and Nagy (1992) argued that it is futile to categorize
serpentine soils based on chemistry and biology (otherwise referred to as the lsquoserpentine
factorrsquo) as proposed by Brooks (1987) Cr concentrations greater than 200 mg kg-1 in
serpentine soils are consistently present Chromium in the soils originates from minerals
in the bedrock where the valence is primarily Cr(III) the exception being Cr(VI) in the
rarely present mineral crocoite (PbCrO4) Increased Cr-concentrations in serpentine soils
are largely the result of chromite or other resistant Cr-bearing minerals causing a
localized enrichment in the soils Although Cr in serpentine soils is not readily released
from the soil matrix serpentine soil vegetation can incorporate Cr concentrations as high
as 700 mg kg-1 (Table 34) Finally the limited extractability of Cr and site dependent
factors such as the presence of organic matter clays Fe(II)-bearing minerals pH and the
poor drainage indicates that the possible formation and sustained presence of Cr(VI) is
not a portentous matter related to serpentine soils
120
ACKNOWLEDGEMENTS
We appreciate technical and field support from Robert Jones Anders Meibom
Rob Dunbar and Guangchao Li Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund GSA Graduate Student Research Grants and the
Office of Teaching and Learning at Stanford University is gratefully acknowledged
121
Table 41 Measured Cr concentrations in serpentine soils worldwide
Location Cr concentration (mg kg-1) Reference
Canyon Creek area OR 920 (White 1971) 66-13200 (Brooks et al 1992)
Ireland 29-5000 (Proctor and Woodell 1971 Jeffrey 1992) Italy 1300-3900 (Gambi 1992) Japan 5200 (Shewry and Peterson 1976) Maryland 1500-6000 (Rabenhorst et al 1982) Mt Tamalpais CA 1700 (White 1971) New Caledonia 6300-80000 (Jaffre 1980) New Caledonia 2-6 (Schwertmann and Latham 1986) New Idria CA 1640 (White 1971) New Zealand 850-3250 (Lee 1992) Newfoundland Canada 309-1447 (Roberts 1992) Nickel Mountain OR 1730 (White 1971) Santa Barbara County CA 2080 (Woodell et al 1975) Southern Africa 045-14 (Morrey et al 1992) Tehama County 023-12 (Gough et al 1989) Transvaal Lowveld South Africa 120-800 (Cole 1992) Unst Shetland 3000-4000 (Proctor 1969) Wenatchee Mnts WA 895 (White 1971) Zimbabwe 110-46000 (Soane and Saunder 1959 Proctor et al 1980)
Goias State Brazil
122
123
Table 42 Summary of Cr Fe and Ni extractions from serpentine soils
Range of Extractable Cr Fe and Ni Location Cr Fe Ni
Method Reference
Italy 2-5 microg g-1 305-640 2300-2700microg g-1 25 CH3COOH (Gambi 1992)
NE Portugal 02-09 microg g-1 120-540 microg g-1 40-290microg g-1 EDTA (Menezes de Sequeira and Pinto da Silva 1992) New Caledonia 22-23 384-581 02-216 DCB (dithionite-citrate bicarbonate) (Schwertmann and Latham 1986) New Caledonia 02-11 microg g-1 02-66microg g-1 Method Unknown (Jaffre 1980)
Norway 357microg g-1 316microg g-1 HClO4HNO3 (concentrated) (Zeien and Bruemmer 1989) Norway 254microg g-1 134microg g-1 Ascorbic and Oxalate Acid pH=325 (Zeien and Bruemmer 1989)
Norway 62 microg g-1 53microg g-1 NH4-EDTA (Zeien and Bruemmer 1989) Norway 22-60 microg g-1 8-25 mg kg-1 1-7 microg g-1 pH-stat reactor EDTA (Kaupenjohann and Wilcke 1995)
Swiss Alps 08 micromol L-1 28micromol L-1 228microg g-1 Lysimeter (Nonfiltered) (Gasser and Dahlgren 1994) Trinity Ophiolite CA 004-045 gkg 27-210 g kg-1 005-245microg g-1 Citrate-Dithionite Extracts (Alexander et al 1989)
Unst lt01 microg g-1 19-77microg g-1 Ammonium Acetate (Shewry and Peterson 1976)
Unst 500-1390microg g-1 1540-3330microg g-1 HNO3-Soluble (Shewry and Peterson 1976) JRBP CA 0 microg g-1 01microg g-1 103microg g-1 BaCl2 Method (Oze et al 2003)
JRBP CA 78 microg g-1 513 microg g-1 150 microg g-1 Hydroxlyamine+Acetic Acid (Oze et al 2003) JRBP CA 31 microg g-1 2039microg g-1 35microg g-1 10M Hydrofluoric Acid (Oze et al 2003)
123
124
Table 43 Serpentine soil plant Cr concentrations
Plant Cr Concentration (microg g-1) Location Reference
Lophostachys villosa 203 Goias State Brazil (Brooks et al 1992) Ruellia geminiflora
56 Goias State Brazil (Brooks et al 1992) Minuartia ophiolitica 83 Italy (Gambi 1992) Sanguisorba minor 194 Italy (Gambi 1992) Plantago lanceolata 135 Italy (Gambi 1992) Cerastium latifolium 110 Italy (Gambi 1992) Funaria flavicans 34-899 Maryland (Shaw 1991) Avena sativa 168-583 New Zealand (Lee 1992) Aerobryopsis longissima 200-700 New Caledonia (Lee et al 1977) Arenaria humifusa 530 Newfoundland Canada (Roberts 1992) Senecio pauperculus 406 Newfoundland Canada (Roberts 1992) Rhalomitrium lanuginosum 246 Newfoundland Canada (Roberts 1992) Santolia semidentata 95 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Asplenium cuneifolium 228 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992) Armeria eriophylla 38 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Arctostaphylos viscida 029-705 Tehama County CA (Gough et al 1989) Calluna vulgaris 417 Unst (Shewry and Peterson 1976) Armeria maritima 149 Unst (Shewry and Peterson 1976) Agrostis stolonifera 120 Unst (Proctor and Woodell 1971) Dicoma niccolifera 77 Zimbabwe (Proctor 1992) Xerophyta equisetoides 68 Zimbabwe (Proctor 1992)
124
125
Table 44 Chemical and textural properties of the collected soils
Ammonium Acetate Extractable ------------------------Hydrometer--------------------- Ca K Mg Na
Sand Silt ClaySample pH EC dSm
Organic Matter ------
LECO Nitrogen ------ -------------------mg kg-1-------------------
------ ------ ------
Texture
New Caledonia 534 01 05 lt001 279 29 692 44 43 32 25 LoamHawaii 604
03 10 002 839 46 168 53 8 26 66 ClayEight Dollar Mnt 651 07 16 006 599 62 3464 25 28 28 44 Clay
Nickel Mnt 665 05 07 001 167 42 3289 17 24 32 44 ClayHarvard Mine 691 03 11 003 998 301 2897 43 41 30 29 LoamJasper Ridge 671 04 32 012 505 237 3291 36 36 31 33 Clay LoamPillikin Mine 642 04 05 002 185 55 3421 29 45 23 32 Sandy Clay Loam
125
126
Table 45 Total major and trace element concentrationsdagger of the collected soils
Element Concentration New Caledonia Hawaii Eight Dollar
Mnt Nickel Mnt Harvard Mine Jasper Ridge Pillikin
Mine Ag mg kg-1 lt05 lt05 lt05 lt05 02 lt05 lt05Al 931 1325 314 095 179 314 179Ba mg kg-1 lt10 180 40 20 30 230 lt10Be mg kg-1 05 16 05 lt05 lt05 08 lt05Ca 003 012 085 018 007 079 002Cd mg kg-1 lt05 lt05 18 lt05 05 lt05 lt05Co mg kg-1 56 183 338 403 85 155 122Cr mg kg-1 827 388 9528 5761 2900 5976 1977Cu mg kg-1 78 248 43 25 51 33 4Fe 638 1296 1635 17 538 727 468K 002 046 004 003 004 039 001
Mg 072 026 753 829 792 1025 gt150Mn mg kg-1 257 2706 3017 2554 920 1664 903Na 001 008 005 001 003 057 lt001Ni mg kg-1 2780 345 5510 14100 1895 2180 2760P mg kg-1 50 260 160 150 130 140 20Pb mg kg-1 6 31 4 3 2 8 6Ti 016 041 008 002 lt001 02 002V mg kg-1 63 132 158 54 65 92 27W mg kg-1 10 10 10 10 lt10 lt10 lt10Zn mg kg-1 43 218 160 132 66 111 47
126
dagger ICP-AES analyses
127
Table 46 Compositionsdagger for soil Cr-containing minerals identified in Figure 49
Wt New Caledonia A
Hawaii B
Eight Dollar Mnt C
Eight Dollar Mnt D
Nickel Mnt E
Harvard Mine F
Jasper Ridge G
Jasper Ridge H
Pillikin Mine I
SiO2 491 008 175 026 002 000 1940 001 000Al2O3
756 185 272 2235 2041 088 559 1575 839NiO 023 005 001 006 011 2054 014 002 007FeO 7766a 6212 6633b 3420b 1784 2445 2629 2314 3105MnO 004 037 181 123 133 031 155 146 216Cr2O3 165 007 2387 3343 4682 3991 1878 5024 5355MgO 048 121 202 653 1315 1062 1668 926 497CaO 001 000 013 008 000 000 015 000 000Na2O 000 000 002 001 000 000 006 000 000TiO2 004 2465 012 006 012 000 000 003 018
TOTAL 9257 9041 9879 9821 9980 9671 8864 9920 10037Mineral Cr-Hydroxide Ulvoumlspinel Cr-Magnetite Chromite Mg-Chromite Trevorite CSM1 Chromite Chromite
127
dagger Electron microprobe analyses This is the form of iron unless otherwise stated a FeO is in the form Fe2O3 b FeO is in the form of Fe3O4 1CSM Microcrystalline chromitesilicate mixture
128
Table 47 Chromium-bearing mineraldagger masses and volumes required to influence Cr concentrations in soils
Cr Concentration Mineral mass (grams) required for 1000 g bulk soil analysis Mineral Mass Percent () (mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite
500 11 07 22 24 48 57 011 007 022 024 048 057 1000 22 15 43 48 95 115 022 015 043 048 095 1152000 43 29 86 96 191 230 043 029 086 096 191 2305000 108 73 216 241 477 574 108 073 216 241 477 574
Cr Concentration Mineral Volume Percent () in Serpentine Mineral Volume Percent () in Quartz
(mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite500 005 004 02 02 03 15 006 004 02 02 04 06
1000 01 007 04 03 07 30 01 007 04 03 07 112000 02 014 07 06 14 59 02 015 07 06 14 225000 06 036 18 16 34 147 06 037 18 16 35 55
128
daggerMineral formulas eskolaite Cr2O3 chromite FeCr2O4 fuchsite KCr2(AlSi3O10)(OH)2 uvarovite Ca3Cr2Si3O12 tawmawite Ca2CrAl2Si3O12(OH)2 kaumlmmererite Mg5CrAlSi3O11(OH)8
Note Molar volumes used in the calculations above are from their Al-enriched counterparts in Robie and Hemingway (1995) The molar volume for kaumlmmererite was estimated using unit cell volumes from XRD unit cell measurements from Robie and Hemingway (1995) Christofides et al (1994) and Phillips et al (1980)
129
Figure Captions
Figure 41 Selected serpentine soil localities as well as soils collected for analyses are
shown on a world map Active plate boundaries are shown Serpentine soil localities not
adjacent the plate boundaries on this map are related to ancient plate boundaries that are
no longer active
Figure 42 Field pictures of soils collected from a) Eight Dollar Mountain b) Nickel
Mountain c) Harvard Mine d) Pillikin Mine and f) Jasper Ridge Biological Preserve
Figure 43 Stability field diagram of Cr under a range of Eh and pH conditions (log a
Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom (1998) Adsorption
isotherms for Cr3+ and CrO42- on hydrous ferric oxide (HFO) are also shown (Leckie et
al 1980 Leckie et al 1984)
Figure 44 Microprobe analyses demonstrating Cr versus Al Wt for Cr-spinel
(chromite) and Cr-silicates including Cr-mica (fuchsitemariposite) Cr-garnet
(uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite) (Whitmore et al
1946 Leo et al 1965 Chen and Lee 1974 Neiva 1978 Cooper 1980 Phillips et al
1980 Grapes 1981 Schreyer et al 1981 Max et al 1983 Nutman et al 1983 Raase
et al 1983 Chao et al 1986 Von Knorring et al 1986 Kerrich et al 1987 Treloar
1987b a Morand 1990 Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen
1999 Proenza et al 1999) Chromium-spinel and Cr-silicate compositions based on
octahedral Cr and octahedral Al are also shown with tie-lines (dotted) denoting ideal
isomorphic substitution between Cr and Al end-members
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above Each dotted line connecting the
points represents a separate soil profile at each site
129
Figure 46 Ranges of pH for serpentine soils and related groundwater worldwide Soil
pH values were obtained from A(Kruckeberg 1992) B (Graham et al 1990) C
(Alexander et al 1989) D (Gough et al 1989) E (Rabenhorst et al 1982) F
(Roberts 1992) G (Bulmer et al 1992) H (Proctor 1992a) I (Jeffrey 1992) J
(Gambi 1992) K (Verger 1992) L (Proctor and Cole 1992) M (Cole 1992b) N
(Morrey et al 1992) O (Proctor 1992b) P (Mizuno and Nosaka 1992) Q (Cole
1992a) and R (Lee 1992) Groundwater pH values from serpentinite are from Barnes
and ONeil (1969) and Barnes et al (1972) Note the range of pH environments possible
for ultramafic rocks and their related soils
Figure 47 Soil pictures with Munsel color values and names from soils collected from
a) New Caledonia b) Hawaii c) Eight Dollar Mountain d) Nickel Mountain e) Harvard
Mine f) Jasper Ridge Biological Preserve and g) Pillikin Mine
Figure 48 Cr XANES spectra for soils analyzed from Jasper Ridge New Caledonia
Pillikin Mine Eight Dollar Mountain and Harvard Mine The pre-edge area indicative of
Cr(VI) peaks is noted in the diagram No definitive Cr(VI) peaks are present in any of
these spectra
Figure 49 Backscattering images of mineral phases within each soil are presented
Electron microprobe analyses are noted by a circle and letter for the Cr-bearing minerals
and the results are listed in Table 46
Figure 410 Chromite magnetite and hematite compositions from the collected soils
(gray circles) as well as analyses for serpentinite chromite (white squares) from Figure
44 are shown in these diagrams with Cr Wt versus Al Fe Mg and Al+Fe+Mg Wt
130
Figure 41
131
Figure 42
132
Figure 43
133
Cr Wt
0 10 20 30 40 50
Al W
t
0
5
10
15
20
25Cr-SpinelCr-MuscoviteCr-EpidoteCr-GarnetCr-Chlorite
Cr-SpinelBased on 4 Oxygens
Cr0 1 2
AlVI
0
1
2
ChromiteFeCr2O4
HercyniteFeAl2O4
Cr-MuscoviteBased on 11 Oxygens
Cr0 1 2
AlVI
0
1
2
MuscoviteKAl2(AlSi3O10)(OH)2
FuchsiteKCr2(AlSi3O10)(OH)2
Cr-GarnetBased on 12 Oxygens
Cr0 1 2
AlVI
0
1
2
GrossularCa3Al2Si3O12
UvaroviteCa3Cr2Si3O12
Cr-EpidoteBased on 125 Oxygens
Cr0 1
AlVI
2
3
ClinozoisiteCa2Al3Si3O12(OH)2
TawmawiteCa2CrAl2Si3O12(OH)2
Cr-ChloriteBased on 14 Oxygens
Cr0 1
AlVI
0
1
2
Kammererite(Mg5Cr)(AlSi3)O10(OH)8
Clinochlore(Mg5Al)(AlSi3)O10(OH)8
Figure 44
134
Maryland (USA)(Rabenhorst et al 1982)
Chromium (mg kg-1)
0 1000 2000 3000 4000 5000 6000 7000
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Tehama County CA (USA)(Gough et al 1989)
Chromium (mg kg-1)
0 2000 4000 6000 8000 10000 12000 14000
Dep
th (m
eter
s)
00
02
04
06
08
Kalamath Mountains CA (USA)(Hotz 1964)
Chromium (mg kg-1)
2000 4000 6000 8000 10000 12000 14000 16000
Dep
th (m
eter
s)
00
05
10
15
20
25
30
35
40
45
50
Newfoundland (Canada)(Roberts 1992)
Chromium (mg kg-1)
800 1000 1200 1400 1600 1800
Dep
th (m
eter
s)
00
02
04
06
08
Harvard Mine CA (USA)(Unpublished Data)
Chromium (mg kg-1)
0 500 1000 1500 2000 2500
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Jasper Ridge CA (USA)(Oze et al 2003)
Chromium (mg kg-1)
1500 2000 2500 3000 3500 4000 4500 5000
Dep
th (m
eter
s)
00
01
Figure 45
135
LocationsZ A B C D E F G H I J K L M N O P Q R S
pH
0
1
2
3
4
5
6
7
8
9
10
11
12
Nor
th A
mer
ica
Can
ada
Brit
ain
Italy
Far E
ast
Japa
n
Zim
babw
e
Sout
h A
fric
a
Wes
tern
Aus
tral
ia
New
Zea
land
Mg-Bicarbonate Waters pH=8
Ca-Hydroxide Waters pH=11-12
Average RainpH=57
Sout
h A
fric
a
Italy
Irela
nd
Cal
iforn
ia
Cal
iforn
ia
Brit
ish
Col
umbi
a
Mar
ylan
dCal
iforn
ia
Figure 46
136
Figure 47
137
-2
-1
0
1
2
3
4
5
5980 5990 6000 6010
Energy (eV)
Nor
mal
ized
Abs
orba
nce
New Caledonia
Pillikin Mine
Eight Dollar Mnt
Jasper Ridge
Harvard Mine
Cr(
VI)
Figure 48
138
Figure 49
139
Cr Wt 0 10 20 30 40 50
Al W
t
0
10
20
30 Soil Serpentinite
Cr Wt 0 10 20 30 40 50
Fe W
t
0
20
40
60
80
Cr Wt 0 10 20 30 40 50
Mg
Wt
0
2
4
6
8
10
Cr Wt 0 10 20 30 40 50
Al +
Fe
+ M
g W
t
0
20
40
60
80
MagnetiteHematite
Chromite
MagnetiteHematite
Chromite
Chromite
MagnetiteHematite
SpinelSilicate
SpinelSilicate
SpinelSilicate
SpinelSilicate
Figure 410
140
- Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip1
-
- Chapter 3 Oxidative Promoted Dissolution of Chromite by
- Manganese Dioxide and Concurrent Production of Ch
-
- Chapter 4 Chromium Geochemistry of Serpentine S
-
- List of References
-
- Chapter 2
- Chapter 3
-
- Chapter 4
-
- Appendix A
- Chapter 2
-
- Chapter 3
-
- Chapter 4
-
- Appendix A
-
- CHAP1pdf
-
- Toxicity and Regulations of Chromium
- Mineralogy of Chromium in Serpentinites
-
- Chapter Summaries
-
- CHAP2Tablepdf
-
- JR305CLAY Depth 0-5cm
- JR3515CLAY Depth 5-15cm
- JR33045CLAY Depth 30-45cm
-
- HCAP2Figspdf
-
- Figure 27
-
- CHAP3Tablepdf
-
- Figure 36
- Figure 37
-
- CHAP4pdf
-
- ABSTRACT
- REVIEW
-
- Chromium Geochemistry
-
- Soil Organic Matter
-
- CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
-
- BedrockProtolith Mineralogy
- Soil Chemistry and Physical Properties
- Soil Mineralogy
- Backscattered electron images and microprobe analyses were used to identify the Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49 and corresponding Cr-bearing mineral compositions are listed in Table 46 The New C
-
- DISCUSSION
-
- Chromium Spinel in Soils and Rocks
-
- CONCLUDING STATEMENT
- ACKNOWLEDGEMENTS
-
- CHAP4Tablepdf
-
- Ulvoumlspinel
-
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
Despite the resistance of chromite to weathering Cr release and oxidation from
chromite is a potential environmental hazard in sediments and soils and a pathway for
soil development (weathering of primary minerals) related to ultramafic rocks and their
metamorphic derivatives (serpentinites) Birnessite is a common pedologic mineral in
these sediments and soils capable of oxidizing aqueous Cr(III) The interaction between
chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4) and birnessite was investigated with
an interest in the potential generation of Cr(VI) The rate of Cr(VI) in solution increases
with increasing chromite suspension densities and decreasing pH but is independent of
birnessite suspension densities at values greater than ~20 m2 L-1 at 25degC The overall rate
expression of Cr released and oxidized in solution from chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression--rates ranging from 9times10-4 to 44times10-3 microM Cr(VI) h-1 These
experiments therefore demonstrate that serpentine soils are a potential source of non-
anthropogenic Cr(VI)
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
v
chemistry of Cr in serpentine soils including its protolith is reviewed with a focus on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils have widely varying pH contain a variety Fe(III) oxides (magnetite and
hematite) phyllosilicates (serpentine and chlorite) and clays (smectites and
vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1) and
Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Additionally Cr(III) is the only valence state
observed in the serpentine soil solids however Cr(VI) has been identified in New
Caledonia and California serpentine soil solutions at concentrations below 30 microM The
enrichment and range of Cr concentrations in serpentine soils are directly related to the
presence of Cr-spinels specifically chromite and Cr-magnetite These spinels are
resistant to weathering and are preserved in the soil environment however oxidation of
Cr(III) from Cr-spinels by high valent Mn oxides or other strong oxidants is a potential
source of Cr(VI) identified in serpentine soil solutions Due to the chemically and
physically resistant nature of the Cr-spinels Cr-bearing silicates including clay minerals
Cr-chlorite Cr-garnet Cr-mica and Cr-epidote are more viable sources of Cr identified
in vegetation soil extractions soil solutions and related waters
vi
PREFACE
The chemistry and mineralogy of Cr related to serpentinites and serpentine soils
are assessed in this dissertation using a multidisciplinary field analytical experimental
and theoretical approach The four chapters that constitute this dissertation describe my
investigations related to these rocks and soils Additionally the common theme shared
by these chapters is that they all examine the Cr mineralogy in serpentinites and
serpentine soils and the potential of Cr to be mobilized and oxidized in near-surface
environments Chromite a Cr-bearing spinel is the dominant source of Cr in
serpentinite and serpentine soils Although chromite is resistant to weathering the
presence of birnessite a high valent Mn oxide will increase the release of Cr and the
production of Cr(VI) from chromite over a range of pH Chromium-clinochlore or
kaumlmmererite contains Cr concentrations less than 3 Wt however the rate at which Cr-
clinochlore weathers compared to chromite makes this mineral a more probable source of
Cr released into soil solutions These types of concerns are addressed in this thesis Brief
descriptions of the four individual dissertation chapters including their publication status
and a list of the co-authors are presented in the following paragraphs
Chapter 1 presents the main objectives of this dissertation The chemistry
regulations and toxicity of Cr are discussed as well as the general characteristics of
serpentinites and serpentine soils Additionally a brief summary of each chapter is
provided
Chapter 2 focuses on serpentinites and serpentine soils at Jasper Ridge Biological
Preserve Stanford CA The results have been submitted for publication in an article
entitled ldquoChromium geochemistry of serpentinized ultramafic rocks and serpentine soils
in the Franciscan Complex of Californiardquo to the American Journal of Science (March
2003) Scott Fendorf Dennis Bird and Robert Coleman are the second third and fourth
co-authors respectively
Chapter 3 is an experimental study evaluating the release of Cr and the
production of Cr(VI) from chromite in the presence of birnessite over a range of pH The
results for this study will be submitted for publication in an article entitled ldquoOxidative
promoted dissolution of chromite by manganese dioxide and concurrent production of
vii
chromaterdquo to Environmental Science and Technology The co-authors for this paper are
Scott Fendorf and Dennis Bird
Chapter 4 is a review paper that critically evaluates the chemistry of serpentine
soils and serpentinites primarily focusing on Cr geochemistry Additionally analytical
studies of serpentine soils collected from California Oregon and New Caledonia are
presented This study will be presented at the Robert Coleman Symposium and has been
submitted as an article entitled ldquoChromium geochemistry of serpentine soilsrdquo for an
International Geology Review Special Publication Scott Fendorf Dennis Bird and
Robert Coleman are the second third and fourth co-authors respectively
Appendix A is a preliminary investigation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite) a mineral common in serpentinites and present in serpentine
soils This appendix is entitled ldquoEstimation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite)rdquo The results of this chapter will be included in a study that
estimates the thermodynamic properties of Cr-bearing silicates including fuchsite (Cr-
muscovite) tawmawite (Cr-epidote) and uvarovite (Cr-garnet)
viii
ACKNOWLEDGEMENTS
Serpentinites serpentine soils and chromium are the Geologic Trinity of this
dissertationhellipeach topic is intimately related to one another The scope and breadth of
this research required a significant amount of time and funding None of this research
would have been possible without the generous support of several foundations
organizations and Stanford University I am grateful for the financial aid I have
received from the Stanford Fellowship the Stanford University McGee and Shell funds
the National Science Foundation the Geologic Society of America Graduate Research
Fund the Stanford Office of Teaching and Learning grant and the Jasper Ridge
Biological Preserve Melon Grant
I would not be where am I today if it were not for my parents Robert and Sue
Oze and my brother Rob Oze Thank you for the sandbox and all the hose-driven
volcanoeshellipthose Wookies never stood a chance While some families vacation in theme
parks my family frequented gem and mineral conventions collected rocks and shells
from mist-shrouded shores and sought lucky stones including an infamous quartz orb
Sometimes I think I was destined to be a geologist I would also like to thank everyone
in my family and all our family friends who have provided that extra little push and
support along the way These people include Tudor Bennatts Louis Oze Margaret and
John Chambers John Bennatts Nancy Skiles the Izumis the memory of Witchie Poo
hijacking Harleys from Bremerton bikers and the Man in the Sound waiting for that one
last catch Close friends I would like to acknowledge who are not related to the Stanford
crowd include Dave and Leigh Anderson and Rebecca Parker These satellite friends
helped me keep in touch with the world outside of academia and proved to me that I
wasnrsquot meant to have a real job A special thank you goes to Katherine Akako
Izumihellipyou have made this whole trip worthwhile
My PhD advisors Scott Fendorf and Dennis Bird provided an invaluable amount
of help patience and energy over the past five years Scott always made the time to
discuss and refine many of my projects making them relevant to many current
geochemical issues Additionally he helped me hone my soil science skills in the field
and in the lab Out in the field Scott can dig a pretty mean hole Dennis devoted a
significant amount of time and sweat in order to make this dissertation happen His
ix
insightful knowledge of geochemistry and of the geology of the Sierra Nevadas provided
the cornerstone for this dissertation
Robert Coleman played a key role in the production of this dissertation He
always knew what to say at the right time and helped me understand the complexity of
ophiolites and serpentinites through experienced eyes Additionally Keith Loague and
Gary Ernst deserve recognition for their outstanding support as committee members for
this dissertation over the last four years I would also like to thank Jonathan Stebbins and
Gail Mahood for providing me the opportunity to attend Stanford
Bob and Vicki Jones always offered an open ear whether it was science or solving
the worldrsquos problems Rick Hazlett (Pomona College) and John Winter (Whitman
College) both had a significant impact on shaping my future science goals and objectives
during my undergrad years and both encouraged me to pursue a career in geology They
were the catalyst for this whole experiment
I owe a lot to my research group members (both the Bird Cage and Soil
Chemistry Group) and those people who didnrsquot mind me tagging along Before I
introduce these people one person has been my friend and co-conspirator for many years
at Stanford Tim Copehellipyou are a pain in my ass Besides never picking up any of your
crap and breaking all my stuff this whole grad school experience would not have been
the same without you Other friends and acquaintances that will not be forgotten in
chronological order of me walking into my office for the first time on September 11
1998 include Anne Dyreborg Esra Inan Tom Bawden Patrick Redmond Phil
Neuhoff Thrainn Fredrickson Lara Heister Kaye Savage Andrea Fildani Mike Schultz
Trayle Kushan Kevin Theissen Angela Hessler Chris Van de Ven Kono Lemke Colin
Doyle Colleen Hansel Ben Bostick Matthew LaForce Nancy Grumet Scott Kroeker
Anders Meibom Brian Ebel Matt Ginder-Vogel Matt Pollizitto Spud Amy Weislogel
Travis Horton and Jake Waldbauer I would like to end this Acknowledgement section
and begin my dissertation with a haikuhellip
Better than outcrops
Are the people Irsquove met here
Thanks for all the beer
x
TABLE OF CONTENTS
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iv Prefacehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip vii Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip ix Table of Contentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xi List of Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xiv List of Illustrationshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xvi Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 1
Toxicity and Regulations of Chromiumhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 2 Mineralogy of Chromium in Serpentiniteshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 3 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5 Chapter Summarieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5
Chapter 2 Chromium Geochemistry of Serpentinized Ultramafic Rocks and Serpentine Soils in the Franciscan Complex of Californiahelliphelliphelliphelliphelliphellip 8
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 8 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 9 Backgroundhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 10
Geologic Background of the Serpentinite at Jasper Ridgehelliphelliphelliphellip 10 Serpentine Soils Characteristics and Reviewhelliphelliphelliphelliphelliphelliphelliphellip 11 Sources of Chromium ProtolithBedrockhelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 12 Chromium-Bearing Phases in Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 13
Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Location and Sample Collectionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Mineral Identification in Rocks and Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Soil Chemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 15 X-ray Diffraction and Clay-Size Mineral Identificationhelliphelliphelliphelliphellip 16 X-ray Absorption Spectroscopyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17 Sequential Extraction Procedurehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17
Resultshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Serpentinite Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Distribution of Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip21 Serpentine Soil Chemistry and Mineralogy at Site JR3helliphelliphelliphelliphellip 22 Sequential Extractionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 26
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 27 Concluding Remarkshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 33 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 44 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 47
xi
Chapter 3 Oxidative Promoted Dissolution of Chromite by Manganese Dioxide and Concurrent Production of Chromatehelliphelliphelliphelliphelliphellip 63
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 63 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 64 Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 65
Solid Phase Characterization and Synthesishelliphelliphelliphelliphelliphelliphelliphelliphellip 65 Extraction Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 66 Batch Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Solution Analysishelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69
Results helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite Extraction Experimentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite and Birnessite Batch Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 70 Secondary and Solid Phase Examination of Chromite and Birnessitehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 72 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 72
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 73 Chromium(VI) Pathways from Chromite and Birnessite into Solutionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 76 Serpentine Soils and Cr(VI)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 77
Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 79 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 80 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 84 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 87
Chapter 4 Chromium Geochemistry of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip100
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 100 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 101 Reviewhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102
Chromium Geochemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102 Mineralogy of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 103 Sources of Chromium for Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphellip 105 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 106 Soils Solutions and Related Waterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 108 Serpentine Soil Vegetationhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 110 Soil Organic Matterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 111
California Oregon and New Caledonia Serpentine Soilshelliphelliphelliphelliphelliphelliphellip 112 Sample Locations and General Soil Characteristicshelliphelliphelliphelliphelliphellip 112 BedrockProtolith Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 113 Soil Chemistry and Physical Propertieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 114 Soil Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116 Chromium and Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 117
xii
Chromium Spinel in Soils and Rockshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 119 Concluding Statementhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 120 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 121 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 122 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 129 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 131
Appendiceshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 141 Appendix Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 142 Appendix Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 143 Appendix A Estimation of the Thermodynamic Properties of Kaumlmmererite (Cr-Clinochlore)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip145
Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 145 Nomenclature and Crystal Chemistry of Cr-chloritehelliphelliphelliphelliphelliphelliphelliphelliphellip 146 Estimation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 147
Estimation of theStandard Molar Volume Using Unit Cell Measurementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of the Standard Molar Volume Heat Capacity and Entropyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of Standard Molar Enthalpy and Gibbs Free Energyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 150 Critique and Discussion of the Estimation Algorithms and SUPCRT92helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 151
Evaluation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 152 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 155 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 161 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 162
List of References
References to Chapter 1helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 166 References to Chapter 2helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 172 References to Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 179 References to Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 183 References to Appendix Ahelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 195
xiii
LIST OF TABLES Chapter 2 Table 21 Whole rock elemental concentrations of serpentinites chert and sandstone Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in
serpentinite rock at Jasper Ridge Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in
serpentine soil at Jasper Ridge Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25
and 26 Table 24 Bulk soil elemental concentrations for soils and bedrock collected along
the transect in Figure 21b Table 25 Chemical and physical properties of the soil at Site JR3 Table 26 Selected trace metal concentrations for each size fraction versus depth at
Site JR3 Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3 Table 28 X-ray diffraction d-spacings for common silicates Table 29 Results of the Diablo clay serpentine soil and chromite sequential
extraction experiments Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1
kg of sample Chapter 3 Table 31 Chromite and birnessite masses and concentrations used for each batch
experiment Table 32 Serpentine soil and birnessite experimental parameters Table 33 Summary of Cr(VI) rates obtained from batch experiments Table 34 Comparative analyses of batch experiment solutions and chemical
extractions of chromite and birnessite solids Chapter 4 Table 41 Measured Cr concentrations in serpentine soils worldwide Table 42 Summary of Cr Fe and Ni extractions from serpentine soils Table 43 Serpentine soil plant Cr concentrations Table 44 Chemical and textural properties of collected soils Table 45 Total major and trace element concentrations of the collected soils Table 46 Compositions for soil Cr-containing minerals identified in Figure 49 Table 47 Chromium-bearing mineral masses and volumes required to influence Cr
concentrations in soils Appendix A Table A1 Cr and Al end-members in the Cr-clinochlore solid solution
xiv
Table A2 Standard state thermodynamic properties of Cr and Al-bearing minerals Table A3 Unit cell measurements and molar volumes of clinochlore and Cr-
clinochlores Table A4 Entropy volume and heat capacity for kaumlmmererite using the Ransom and
Helgeson Method Table A5 Fictive polyhedral considered in the estimations Table A6 Estimated thermodynamic properties for kaumlmmererite
xv
LIST OF FIGURES Chapter 2 Figure 21a Soils map of Jasper Ridge Biological Preserve CA Figure 21b Serpentinite and serpentine soil sampling sites at Jasper Ridge Biological
Preserve CA Figure 22 Chromium concentrations versus depth for four serpentine soils in
Maryland California and New Foundland Figure 23 Petrographic images of serpentinite at Jasper Ridge Figure 24 Elemental concentrations relative to Cr for chlorite pyroxene magnetite
chromite and a chromite-silicate mixture in the serpentinite at Jasper Ridge
Figure 25 Backscattering electron images of minerals in the serpentinite at Jasper Ridge
Figure 26 Backscattered electron images of Cr-chlorite and Cr-magnetite assemblage in the serpentinite bedrock
Figure 27 Highest and lowest Cr concentrations at each site in the serpentine soil transect
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected for total gt2 mm sand silt and clay size fractions
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soils at Jasper Ridge
Figure 210 X-ray diffraction clay analyses of sample JR3515 Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at specific depths Figure 212 X-ray microprobe elemental maps of Cr and Fe for the serpentine soil and
serpentinite at site JR3 Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments Figure 214 Scanning electron microprobe images of the chromite standard after each
sequential extraction experiment Figure 215 Tie-lines connecting serpentinite chromite and end-member chromite with
soil chromite values Chapter 3 Figure 31 Scanning electron microprobe images of chromite and birnessite standards Figure 32 Chromite extraction experiments evaluating Cr Fe and Al Figure 33 pH and Eh versus time (hours) for each chromite-birnessite chromite and
birnessite batch experiments Figure 34 pH and Eh with Cr stabilities Figure 35 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant chromite and variable birnessite at pH 5 Figure 36 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant birnessite and variable chromite at pH 5
xvi
Figure 37 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for variable pH and constant chromite and birnessite Figure 38 Rate of Cr(VI) production (d[Cr(VI)]dt) as a function of the surface area
concentration of chromite birnessite and pH Figure 39 SEM examination of solids following the termination of the batch
experiments Figure 310 Serpentine soil and birnessite experiments Figure 311 Iron analyses and the solubility of amorphous Fe(OH)3 and goethite Figure 312 Congruent dissolution of chromite based on Fe and Cr Figure 313 Three pathways of Cr release and oxidation Chapter 4 Figure 41 Serpentine soil localities worldwide Figure 42 Field site pictures of serpentine soils Figure 43 Stability field diagram of Cr over a range of Eh and pH conditions Figure 44 Chromium and Al Wt plot and octahedral Cr and octahedral Al plots
for chromite and Cr-silicates including Cr mica (fuchsitemariposite) Cr-garnet (uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite)
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (m) for selected serpentine soils
Figure 46 pH of serpentine soils and related groundwater worldwide Figure 47 Soil pictures with Munsel color values and names from collected soils Figure 48 Cr XANES spectra for collected serpentine soils Figure 49 Backscattering images of mineral phases within each serpentine soil Figure 410 Chromite magnetite and hematite compositions from analyzed soils as
well as analyses for chromite reported from around the world Appendix A Figure A1 Compositional variation of Cr2O3 and Al2O3 Wt in Cr-chlorite Figure A2 Compositional variation of Cr and VIAl in Cr-chlorite based on 14 oxygens Figure A3 Molar volume estimation of kaumlmmererite based on XRD unit-cell
parameters Figure A4 Natural and calculated compositions of co-existing Cr-clinochlore and Cr-
spinel solid solution in the system Mg0-FeO-Cr2O3-Al2O3-SiO2-H2O at 2 kbar pressure
xvii
CHAPTER 1
INTRODUCTION
Serpentinites or metamorphosed ultramafic rocks are distributed worldwide and
are commonly associated with ophiolite complexes (Coleman 1977 Coleman and Jove
1992) These rocks weather to form serpentine soils a generic term used to describe any
soil derived from serpentinite regardless of its physical or chemical properties
Serpentinites and serpentine soils contain Cr-bearing minerals the most prevalent being
chromite and Cr released from these minerals into surface and subsurface environments
is potentially toxic and carcinogenic The main objective of this dissertation is to
evaluate the mineralogy and chemistry of Cr related to serpentinites and serpentine soils
over geologic time using an integrated analytical experimental and theoretical approach
Additionally the potential of serpentinites and serpentine soils as sources of non-
anthropogenic Cr contamination is investigated A discussion of Cr and its significance
to serpentinites and serpentine soils begins with its discovery by a Frenchman
investigating an ore named ldquored leadrdquo from the Urals
Nicolas-Louis Vauquelin discovered Cr in 1797 in the mineral crocoite PbCrO4
(Burns and Burns 1975) The metalelement was named chromium after the Greek word
chroma meaning color due to its strong influence on mineral pigmentation (Burns and
Burns 1975) For example rubies and emeralds derive their color from trace amounts of
Cr in their structures More importantly JT Lowitz in 1798 discovered Cr in chromite
or Cr-spinel the primary source of Cr for industrial purposes (Burns and Burns 1975)
Chromium is a group VIB transition metal with two common oxidation states (Cr(III) and
Cr(VI)) an average crustal abundance of 100 mg kg-1 and an abundance of 513x105 in
the solar system (relative to [H]=1x1012) (Mason and Moore 1982)
The majority of Cr on the earth exists as Cr(III) where it is an immobile cation
that complexes strongly with organic matter and adsorbs on oxides and silicates even at
low pHs (pHgt4) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) Additionally
the chemical mobility of Cr(III) is limited due to the formation of highly stable
(oxy)hydroxides (Rai and Zachara 1988 Rai et al 1989 Ball and Nordstrom 1998)
1
Chromium(III) has the highest octrahedral site preference energy (OSPE) of any trace
element and it isomorphically substitutes into octahedral sites in oxides
(oxy)hydroxides sulfides and silicates Although Cr(III) is known to substitute for both
octahedral Al(III) and Fe(III) the strongest correlation for substitution appears to be for
Al(III) in oxides and silicates as demonstrated in Chapters 2 and 4 and Appendix A
Chromium(VI) is a strongly oxidizing and weakly adsorbing oxyanion in the form of
chromate (HxCrO42-x) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995 Shiraki
1997 Ball and Nordstrom 1998) Consequently Cr(VI) is chemically mobile and
bioavailable over a wide range of pH (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Chromium(VI) is tetrahedrally coordinated and behaves similarly to
sulfate in minerals such as jarosite (Baron and Palmer 1996)
The pathways and chemistry of Cr oxidation and reduction in soils and related
waters are covered in depth by Fendorf (1995) and Ball and Nordstrom (1998) The
reduction of Cr(VI) to Cr(III) may occur by reactions with Fe(II)aq Fe(II)-containing
minerals such as magnetite organic matter H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996) Oxidation of Cr(III)
to Cr(VI) related to serpentinite-related water and serpentine soil solutions is still a matter
of conjecture Chromium(III) oxidation in serpentine soils has been hypothesized to be
the result of Mn oxides such as birnessite as suggested by both Gough et al (1989) and
Becquer et al (2003) This idea is explored in depth in Chapter 3 where birnessite is
shown to influence the release and oxidation of Cr from chromite over a range of pH (3
to 8) at 25oC
Toxicity and Regulations of Chromium
The examination of Cr based on its oxidation state is necessary due to the
juxtaposing behavior and toxicity of Cr(III) and Cr(VI) Chromium(III) is an essential
micronutrient in human nutrition used by the body to process sugar protein and fat
(Vincent 2000) Inhalation of Cr(VI) is known to cause irritation in the nose including
symptoms such as nosebleeds (epistaxis) and ulcers andor perforations in the nasal
septum (Katz and Salem 1993 Kuo et al 1997) Ingesting considerable quantities of
Cr(VI) can result in stomach ulcers convulsions kidney and liver damage and death
2
(Katz and Salem 1993) Skin contact with certain Cr(VI) compounds causes allergic
contact dermatitis (ulcers) and irritant dermatitis (Katz and Salem 1993) Chromate
induced ulcers or chromeholes are typically red swollen crusted painless lesions
(Katz and Salem 1993) The nosology of Cr is beyond the scope of this dissertation
however Lees (1991) is an excellent source for further information Overall the
Environmental Protection Agency (EPA) the Occupational Safety and Health
Administration (OSHA) the World Health Organization (WHO) and the Department of
Health and Human Services (DHHS) have determined that that Cr(VI) compounds can
increase the risk of lung cancer primarily by inhalation as well as other forms of cancer
caused by ingestion
Federal and state regulation standards appropriate for both Cr(III) and Cr(VI) are
summarized and discussed by Proctor et al (1997) Human health action levels for Cr
vary for each state and the ranges for total Cr in soils include the following 200-66000
Cr(III) mg kg-1 (residential) 02-2000 Cr(VI) mg kg-1 (residential) 2500-1000000
Cr(III) mg kg-1 (non-residential) and 64-71000 Cr(VI) mg kg-1 (non-residential)
(Proctor et al 1997) Incidentally Cr concentrations in serpentine soils and serpentinites
(gt200 Cr mg kg-1) exceed many state action levels and would be considered toxic The
EPA has set a limit of Cr(III)+Cr(VI) at 100 microg L-1 in drinking water The OSHA has
set limits of 500 microg of water soluble Cr(III) compounds per cubic meter of workplace air
or 500 microg m-3 1000 microg m-3 for metallic Cr and insoluble Cr compounds and 52 microg m-3
for Cr(VI) compounds for 8-hour work shifts and 40-hour work weeks The limits
proposed above are constantly changing and should not be used for reference without
checking current regulation standards
Mineralogy of Chromium in Serpentinites
Chromium occurs in a variety of minerals either as a major or minor constituent
(Burns and Burns 1975) Chromium-bearing minerals in serpentinites are primarily
oxides and silicates in which the Cr is present as Cr(III) Chromite is a Cr-bearing spinel
abundant in serpentinites and is resistant to low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) The chemistry and
significance of chromite is addressed throughout this dissertation Pyroxenes such as
3
augite and enstatite are primary minerals identified in serpentinites and contain lt8 Cr Wt
(Grapes 1981 Von Knorring et al 1986 Treloar 1987a Challis et al 1995
Sanchez-Vizcaino et al 1995 Deer et al 1996) Chromate minerals such as crocoite
(PbCrO4) are rare in serpentinites and the occurrence of these minerals are limited to Pb
deposits in Australia Tasmania and the Urals (Williams 1974 Crane et al 2001)
Chromium(III) released from the primary Cr-containing phases may
isomorphically substitute into oxides and silicates for Al(III) and Fe(III) due to size
(octahedral radii Cr3+ = 0615 Aring Al3+ = 061 Aring and Fe3+=0643 Aring) and charge
similarities Magnetite formed during serpentinization incorporates Cr into its structure
and serpentine minerals (antigorite lizardite and chrysotile) contain trace concentrations
of Cr (Spangenberg 1943 Page and Coleman 1967 Burkhard 1993) Eskolaite (Cr2O3)
has been identified in Finland (Von Knorring et al 1986) Metasomatic alteration of
ultramafic rocks commonly results in the formation of Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite) uvarovite (Cr-garnet) tawmawite (Cr-epidote) and
kaumlmmererite (Cr-chlorite) (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Neiva 1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981
Max et al 1983 Nutman et al 1983 Raase et al 1983 Chao et al 1986 Von
Knorring et al 1986 Kerrich et al 1987 Treloar 1987b Treloar 1987a Morand 1990
Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993 Christofides et al 1994
Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Proenza et al
1999) Additionally kyanite and andalusite in the proximity of ophiolite complexes
incorporate up 5 Cr Wt (Cooper 1980 Raase et al 1983 Kerrich et al 1987)
Chapter 4 provides an in depth discussion of these Cr-bearing silicates as well as
demonstrating mineral compositions reported in the literature based on octahedral Cr and
Al Appendix 1 is a preliminary investigation devoted to estimating the thermodynamic
properties of kaumlmmererite These properties can then be incorporated into geochemical
modeling programs to begin accessing the stability of this common serpentinite mineral
over a range of temperatures and pressures
4
Chromium in Serpentine Soils
Evaluating Cr in serpentine soils requires an examination of both the mineralogy
and chemistry of the serpentinite and soil Measured Cr concentrations in serpentinites
and serpentine soils range from ~200 mg kg-1 to 6 Wt all of which is Cr(III) (Soane
and Saunder 1959 Proctor and Woodell 1971 Shewry and Peterson 1976 Jaffre 1980
Proctor et al 1980 Rabenhorst et al 1982 Schwertmann and Latham 1986 Brooks
1987 Gough et al 1989 Gambi 1992 Lee 1992 Morrey et al 1992) In Chapters 2
and 4 chromite and Cr-magnetite are identified as the most prevalent sources of Cr
responsible for Cr-enrichment in these soils Chromite is resistant to chemical
weathering and it has been observed to undergo incongruent dissolution conserving Cr in
serpentine soils Chromium is tightly bound in the matrix of the soil less than 1 of the
total Cr can is removed using standard soil extraction methods (10 hydrogen peroxide
(H2O2) solution 01 M potassium permanganate (KMnO4) solution10 M hydrofluoric
(HF) acid 1 M hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and 1
mM barium chloride (BaCl2) solution) due to the stability of chromite Although silicates
contain minor concentrations of Cr compared to chromite they are perhaps the main
contributors of Cr identified in plants (lt200 Cr mg kg-1) and soil solutions (lt50 microg L-1)
During soil-forming processes Cr released to soil solutions by the dissolution of Cr-
bearing minerals are sequestered into a host of structural and surface sites available in
authigenic Fe(III) oxides Fe(III) hydroxides and clay minerals (Schwertmann and
Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990 Gerth 1990 Gasser and
Dahlgren 1994 Ilton and Veblen 1994 Brigatti et al 2000 Ilton et al 2000) The
clay-sized fractions in soils at Jasper Ridge Biological Preserve CA contain Cr
concentrations as high as 1000 Cr mg kg-1
Chapter Summaries
A detailed field and analytical examination of Cr in serpentinites and serpentine
soils at Jasper Ridge is presented in Chapter 2 A majority of the Cr enrichment and
variability with depth in these soils is the result of small grains of chemically resistant
chromite from the serpentinite protolith Chromite identified in the soil undergoes
incongruent dissolution conserving Cr relative to chromite compositions identified in the
5
serpentinite bedrock Sequential extraction experiments using barium chloride (BaCl2)
hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and hydrofluoric (HF)
acid were performed on a chromite standard confirming that Cr-bearing spinels are
resistant to chemical weathering X-ray diffraction analyses of the clay-size fraction in
the serpentine soils identified Cr-clinochlore smectite vermiculite and serpentine with
the smectite being the most abundant mineral in the clay-size fraction Chromium(III)
was the only identified oxidation state of Cr in the serpentine soil solids and in
serpentinites at Jasper Ridge in the solid phase This chapter represents a comprehensive
site assessment identifying the nature and behavior of Cr in serpentine soils with respect
to the bedrock mineralogy and depth
Chromite an abundant Cr-enriched spinel is responsible for a majority of the Cr
enrichment identified in ultramafic rocks serpentinites and serpentine soils The release
and oxidation of Cr from chromite is a pathway to consider when evaluating the potential
of ultramafic rocks and their related sediments and soils as sources of Cr(VI) In
Chapter 3 experiments aimed at quantifying the release and oxidation of Cr from
chromite in the presence of birnessite are described Chromite is extremely resistant to
dissolution (lt8 Cr mg kg-1 is released) when immersed in solutions including 1 mM
barium chloride solution 10 M hydrofluoric acid 10 hydrogen peroxide and 1 M
hydroxylamine hydrochloride with 25 acetic acid Highly oxidizing solutions of 01 M
potassium permanganate increased the rate and release of Cr relative to the other
solutions listed above thus birnessite a high valence Mn oxide present in serpentine
soils was chosen to determine if Mn could accelerate Cr release and oxidation from
chromite The results presented in Chapter 3 demonstrate that birnessite increases the
rate of Cr(VI) observed in solution from chromite This production of Cr(VI) is
dependent on the chromite suspension density where birnessite surface are concentrations
are greater than 20 m2 L-1 between pHs 3 and 8 at ~25degC
A comprehensive review of the chemical characteristics of serpentine soils
including topics such as the mineralogy of serpentine soils soil solutions organic matter
and vegetation is provided within Chapter 4 Additionally chemical and spectroscopic
analyses were performed on serpentine soils from a) New Caledonia b) Eight Dollar
Mountain Selma OR c) Nickel Mountain Riddle OR d) Harvard Mine Jamestown
6
CA e) Jasper Ridge Biological Preserve Stanford CA and f) Pillikin Mine El Dorado
CA Serpentinites and serpentine soils associated with ophiolite complexes cover ~1 of
the earthrsquos exposed surface and they are correlated to modern and ancient convergent
boundaries The physical and chemical characteristics of serpentine soils are directly
related to the alteration and tectonic history of the serpentinite as well as the soil-forming
factors including the parent material climate biota topography and time Serpentine
soils have Cr concentrations exceeding 200 mg kg-1 however the concentration and the
availability of Cr in these soils are not correlated with the five soil-forming factors
(climate organic material parent material topography and time) or the soil chemistry
Although only Cr(III) was identified in the soil solids Cr(VI) has been identified in soil
solutions (lt30 microg L-1) from serpentine soils in New Caledonia and California Chromite
and Cr-magnetite grains are identified in the soils listed above and these minerals are
responsible for a majority of the Cr enrichment in serpentine soils
Kaumlmmererite is a Cr-chlorite associated with serpentinized ultramafic rocks such
as those collected at Jasper Ridge Harvard Mine and Pillikin Mine Although Cr
concentrations are typically 01-2 Wt in kaumlmmererite it is modally abundant in
serpentinites (up to 5 of the volume of the rock) and a potential source of Cr to be
released into soils and groundwater In Appendix A the standard molar thermodynamic
properties of the Cr-clinochlore end-member kaumlmmererite (Mg5Cr(AlSi3)O10(OH)8) is
estimated using established estimation algorithms This initial study provides the
foundation for submitting a future paper presenting the thermodynamic properties for
major Cr-bearing silicates including fuchsite (Cr-muscovite) uvarovite (Cr-garnet) and
tawmawite (Cr-epidote) and their metamorphic significance related to serpentinized
ultramafic rocks
7
CHAPTER 2
CHROMIUM GEOCHEMISTRY IN SERPENTINIZED ULTRAMAFIC ROCKS
AND SERPENTINE SOILS FROM THE FRANCISCAN COMPLEX OF
CALIFORNIA
ABSTRACT
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of California
produces serpentine soils containing high concentrations of Cr as well as other potentially
toxic elements including Ni Co and Mn Chromium concentrations in serpentine soils
from Jasper Ridge Biological Preserve in the Central Coast Range are as high as 4760
mg kg-1 nearly three times greater than the serpentinite protolith (~1800 mg kg-1)
Chromium-containing minerals within the bedrock include chlorite (~03 Cr Wt )
enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt ) magnetite
(82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate phase
containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
8
INTRODUCTION
Serpentine soils are formed from the weathering of ultramafic rocks and their
metamorphic equivalents (serpentinites Brooks 1987) Measured Cr concentrations in
these soils range as high as 80600 mg kg-1 and are enriched compared to non-serpentine
soils (7 to 221 mg kg-1 Proctor 1992 McBride 1994) The distribution and
concentration of Cr in serpentine soils worldwide is highly variable due to the chemical
composition and mineralogy of the parent material and soil forming factors including
climate biota topography precipitation and time (Hotz 1964 Rabenhorst et al 1982
Schwertmann and Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al
1989 Gough et al 1989 Kaupenjohann and Wilcke 1995) Potential Cr-bearing phases
in serpentine soils have been assessed using extraction experiments and these phases
include organic material silicates secondary oxides and spinels (Schwertmann and
Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994 Kaupenjohann and
Wilcke 1995 Becquer et al 2003) The weathering of these Cr-bearing phases over
geologic time and how these sources govern the distribution of Cr in serpentine soils have
not been evaluated in great detail
Chromium is a group VIB transition metal present in the environment as Cr(III) a
non-hazardous species and micronutrient andor as Cr(VI) a toxin to living cells and a
Class A human carcinogen by inhalation (Daugherty 1992 Cohen et al 1993 James et
al 1997) Chromium(III) is chemically immobile due to forming low solubility
(oxy)hydroxides and stable surface complexes whereas Cr(VI) is a weakly adsorbing
(oxy)anion and thus is mobile in most aquatic environments (McBride 1994 Fendorf
1995 Ball and Nordstrom 1998) Chromite and Cr-magnetite have been cited as
primary sources of Cr in serpentine soils (Rabenhorst et al 1982 Schwertmann and
Latham 1986 Alexander et al 1989 Gough et al 1989) Although Cr is primarily
found in the form of Cr(III) in both ultramafic rocks and serpentinites Cr(VI) has been
measured in serpentine soils in California and New Caledonia (Gough et al 1989
Becquer et al 2003) Due to the abundance of Cr and the potential for Cr(VI) formation
serpentine soils and serpentinites are potential sources of non-anthropogenic Cr
contamination
9
The objectives of this study are to determine the distribution and chemistry of Cr
in naturally Cr-enriched soils using an integrated field analytical experimental and
theoretical approach A wide array of techniques and methods including X-ray
absorption spectroscopy and chemical extractions were used to identify the nature of Cr
and its availability in serpentine soils Serpentinites and serpentine soils at Jasper Ridge
Biological Preserve located near Stanford CA were used to evaluate the geochemical
behavior of Cr These soils cover approximately 50000 m2 and contain high
concentrations of the trace elements Cr Ni and Mn The soils are dominantly Mollisols
with grassland vegetation and they are pedalogically similar to those associated with the
Central Coast Range of the Franciscan Complex in the western United States The main
serpentinite body at Jasper Ridge was exposed approximately 400000 years ago allowing
pedogenesis and Cr cycling to occur over a geologically relevant time period (Burgmann
et al 1994) These soils can then be used as a proxy for evaluating the long-term
behavior of Cr
BACKGROUND
Geologic Background of the Serpentinite at Jasper Ridge
Jasper Ridge Biological Preserve is located in the foothills of the Santa Cruz
Mountains in California and is part of the Central Coast Range of the Franciscan
Complex an Early Cretaceous accretionary meacutelange containing blocks of greywacke
greenstone blueschist and serpentinized ophiolite (Figure 21) Formation
metamorphism and emplacement of serpentinite at Jasper Ridge began approximately
130-150 million years ago during Jurassic subduction along the California continental
margin (Page and Tabor 1967) Thin wedges of ultramafic rock (peridotite) were
detached from the down-going slab of oceanic crust and incorporated into the subduction
meacutelange Subsequent serpentinization of the ultramafic rock in the down-going slab
resulted in the formation of serpentinites now exposed at Jasper Ridge
Rare blueschist blocks (1 to 5 meters in width) present in the serpentinite at Jasper
Ridge attest to the degree of metamorphism and the depths from which portions of the
tectonic meacutelange has been elevated Transformation of greenstone in the meacutelange into
blueschist required high pressures in excess of ~5-7 kbars (15-20 kilometers depth) and
10
temperatures of 200-300 degC (Ernst 1971) The low density of the serpentine (245-250 g
cm-3) allowed upward diapiric migration of the serpentinite body through the denser
enclosing tectonic meacutelange (265 g cm-3) Uplift of the present day Coast Range and
activity associated with the San Andreas fault system have shaped the current topography
and exposure of serpentinite at Jasper Ridge Serpentinite is currently present as a narrow
discontinuous band of pervasively sheared rock located along the ridge top (Figure 21)
The serpentinite is in fault contact with the Whiskey Hill Formation (Cretaceous turbidic
sandstone and mudstone) and tectonically interleaved with the Franciscan Complex
Serpentine Soils Characteristics and Review
Relative to other soil types serpentine soils are characterized by higher
concentrations of Cr Ni Co and Fe lower concentrations of plant nutrients such as Ca
K N and P lower CaMg ratios and characteristic flora and physical properties (Brooks
1987) The pH of serpentine soils ranges from ~4 up to 9 (Brooks 1987 Cole 1992
Proctor 1992) Serpentine soil solutions obtained using tension-free lysimetry have Cr
concentrations between 01 and 32 micromol L-1 in which a majority of the Cr is present as
colloidal material (Gasser and Dahlgren 1994) The inhabitability of these soils for most
plants has been attributed to the imbalance of Ca to Mg (Walker 1954) the deficiency of
plant nutrients (Turitzin 1991) and the elevated concentration of heavy metals
(Robinson et al 1935) however linking the lack of plant productivity of these soils to
toxic levels of Cr has not been fully resolved (Brooks 1987 Kruckeberg 1992) The
unique flora and fauna associated with these soils have been well-documented and Cr
concentrations in the plants range as high as 600 mg kg-1 (Brooks 1987 Gough et al
1989)
Chromium concentrations as a function of depth for serpentine soils in Maryland
(Rabenhorst et al 1982) Tehama County CA (Gough et al 1989) the Klamath
Mountains CA (Hotz 1964) and New Foundland Canada (Roberts 1992) are
illustrated in Figure 22 Chromium concentrations with respect to depth at each location
are highly variable demonstrating both maximums and minimums The concentration of
Cr varies significantly at each location due to the chemical composition of the parent
11
material and other soil-forming factors nevertheless Cr concentrations typically decrease
with depth towards the bedrock
A variety of chemical extractions on serpentine soils suggest a broad host of Cr-
containing soil phases including organic material silicates secondary oxides and spinels
(Schwertmann and Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994
Kaupenjohann and Wilcke 1995) The maximum concentration of Cr extracted from a
serpentine soil reported in the literature is 226 Wt Cr from the New Caledonia laterite
using a 03 M dithionite-citrate bicarbonate (DCB) solution (Schwertmann and Latham
1986) Li and Fendorf (2000) performed extractions on the serpentine soils at Jasper
Ridge using organic acids (citric and oxalic acids) in which a 100 microM citric acid solution
was only able to remove 23 Cr microgg-1 (08 Wt Cr) from a soil containing ~3000 Cr mg
kg-1 This suggests a majority of Cr is bound in phases that are not reactive with these
solvents
Sources of Chromium ProtolithBedrock
Chromium primarily occurs within oxide and silicate phases in ultramafic rocks
and serpentinites Chromite is a common primary (igneous) Cr-bearing spinel in mafic
and ultramafic rocks and often survives low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) Isomorphic substitution of
Al3+ Fe3+ and Ti4+ into the octahedral site and Mg2+ Ni2+ Zn2+ and Mn2+ into the
tetrahedral site is common (Sack and Ghiorso 1991) however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987 Pan and Fleet 1991 Burkhard 1993 Christofides et al 1994 Challis et
al 1995 Sanchez-Vizcaino et al 1995 Deer et al 1996 Mathiesen 1999) Pyroxenes
such as augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) are primary minerals in
ultramafic rocks that incorporate varying concentrations of Cr into their structures (Deer
et al 1996)
Serpentinization is associated with the partial oxidation of Fe(II) present in
olivine and pyroxene forming magnetite (Brooks 1987) Chromium(III) derived from
primary chromite and pyroxene may isomorphically substitute into the Fe(III) site of
magnetite due to size (octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge
12
similarities Hydrothermal and CO2 metasomatism of ultramafic rocks may produce Cr-
silicate minerals including fuchsitemariposite (Cr-mica lt17 Cr Wt ) uvarovite (Cr-
garnet lt20 Cr Wt ) tawmawite (Cr-epidote lt10 Cr Wt ) and kaumlmmererite (Cr-
chlorite lt6 Cr Wt Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Phillips et al 1980 Schreyer et al 1981 Max et al 1983 Kerrich et al 1987 Treloar
1987 Pan and Fleet 1991 Christofides et al 1994 Sanchez-Vizcaino et al 1995)
These minerals common in altered ultramafic rocks have varying concentrations of Cr
that isomorphically substitute for octahedral Al in the silicate structure due to similarities
in charge octahedral preference and octahedral radii (Al3+ = 061 Aring) Any of these
silicates and magnetite are potential sources for Cr in soils produced by the weathering of
serpentinite and ultramafic rocks
Chromium-Bearing Phases in Soils
Soil minerals specifically Fe(III) (oxy)hydroxides and clays provide a host of
structural and surface sites that may sequester Cr(III) released to soil solution by
weathering of Cr-bearing phases in serpentinite and ultramafic rocks Isomorphic
substitution of Cr(III) for Fe(III) in oxides (oxy)hydroxides and amorphous compounds
is expected based on crystallographic considerations synthesis experiments and analyses
of selective dissolution of Fe(III) phases (Schwertmann et al 1989 Gerth 1990)
Additionally Cr(III) readily sorbs on surfaces of Fe(III)-containing oxides such as
magnetite and hematite via inner-sphere complexation and forms low-solubility
precipitates at pH values greater than 4 (Charlet and Manceau 1992 Peterson 1996)
Octahedral substitution of Cr(III) for Al(III) in clays (smectites and vermiculites) is
possible based on size and charge similarities Sorption of Cr(III) on edge and interlayer
sites of clay minerals has also been observed by Ilton and Veblen (1994) Soil organic
matter (SOM) is another possible host for soil Cr in which Cr can be chelated into stable
organomineral and organometallic complexes (Kaupenjohann and Wilcke 1995 Brady
and Weil 1999)
13
MATERIALS AND METHODS
Location and Sample Collection
Jasper Ridge Biological Preserve is a microcosm of accreted terranes located in
the foothills of the Santa Cruz Mountains CA Figure 21a is a soils map reported by
Kashiwagi (1985) depicting the alluvial Tertiary sandstone and Franciscan Complex
soils present at Jasper Ridge The Franciscan Complex soils are divided into two
subgroups those derived from greenstone-graywacke and those from serpentine The
serpentine soils dominantly Mollisols and serpentinite bedrock are present as a narrow
discontinuous band located along the ridge top they do not receive sediment from the
adjacent non-serpentine soils Soils and rocks were collected at Jasper Ridge from 1999
to 2002 and sample locations are shown in Figure 21b A 500 meter transect consisting
of 10 sites extends across the length of the serpentine soil body At each site along the
transect soils and bedrock were collected with respect to depth Site 1 (Diablo clay) is
not derived from serpentinite but rather from chert and greenstone Additionally Site 1
is located down slope of the main serpentinite body and has likely received sediment
from the adjacent serpentine soil Rocks were also collected at several exposed outcrops
located in the serpentinite body
Site JR3 (Figure 21b) was chosen for a detailed geochemical study to identify
factors affecting the geochemistry of Cr in serpentine soils Soil at JR3 is a loamy
magnesic thermic Lithic Haploxeroll The thickness of the soil is approximately 50 cm
deep Grass dominantly Lolium multiflorum comprises 95 of the vegetation at this site
and trees and shrubs are not present
Mineral Identification in Rocks and Soils
Mineral identification was accomplished by electron probe microanalysis on an
automated JEOL 733A electron microprobe operated at 15 kV accelerating potential and
15 nA beam current Calibration was conducted using natural geologic standards Beam
width for analysis was 10 microm Raw counts were collected over 20 s and were constant
with time indicating that elemental drift was negligible Detection limits for probe
analyses are ~002 oxide Wt Additionally backscattering electron (BSE) images of
the rocks and soils were obtained using the microprobe
14
Carbon-coated petrographic thin sections of rock samples were used for
microprobe analyses Soil samples were dried in an oven at 60degC for 24 hours mounted
onto a glass slide using epoxy and heated to 100degC until the epoxy cured An initial cut
of each section using a petrographic saw was made followed by a second infusion of
epoxy to ensure total impregnation Finally each section was cut polished and carbon-
coated for electron microprobe measurements
Soil Chemistry
Soils were analyzed for pH electrical conductivity (EC (deciSiemensm))
particle size ammonium acetate extractable cations organic carbon total nitrogen and
elemental composition The pH was determined using a 11 mixture of soil to deionized
water Electrical conductivity (EC) was determined on 2 mL of a soil solution extract
Particle size of the soil was measured using the hydrometer method as described by Gee
and Bauder (1986) These analyses were performed at the Utah State University
Analytical Laboratory
Ammonium acetate was used to measure the concentration of extractable K Ca
Mg and Na within the soil The procedure involved weighing 2 g of soil and adding 25
mL of ammonium acetate (20 M NH4OAc) followed by mixing on an orbital shaker for
30 minutes The solution was filtered using a Whatman 42 filter paper (pore size 25
microm) Potassium Ca Mg and Na concentrations were analyzed using atomic absorption
spectrophotometry (AAS)
Organic carbon was measured using the Walkley-Black titration method (Mebius
1960) Approximately 05 g of sediment was weighed and reacted with 5 mL of 20 M
potassium chromate (K2Cr2O7) followed by the rapid addition of 10 mL concentrated
sulfuric (H2SO4) acid The bottle was cooled for 30 minutes at room temperature and
diluted with 100 mL of deionized water Next 030 mL of 0025 M ortho-
phenanthroline-ferrous solution was added and the solution titrated to the endpoint using
10M ferrous sulfate (FeSO4)
Total nitrogen of soil samples was obtained using an automated combustion
method as described by Sweeny (1989) Approximately 150 mg of air-dried and sieved
15
(40 mesh) soil was placed in a tin foil encapsulating cup A Leco total nitrogen analyzer
with resistance furnace was then used to determine total nitrogen
Major and minor element concentrations for soil and rock samples were obtained
by completely dissolving the samples using a mixture of hot concentrated nitric
perchloric and hydrofluoric acids Elemental concentrations were measured in the
supernatant using inductively coupled atomic emission spectroscopy (ICP-AES)
Standards were plusmn5 mg kg-1 of their known values These analyses were completed at
Chemex Laboratory (Vancouver Canada) Soil samples were dried and sieved (lt2mm)
before analysis Additionally total element concentrations for gt2mm sand silt and clay
size fractions were obtained for each sample collected at Site JR3
X-ray Diffraction and Clay-Size Mineral Identification
X-ray diffraction (XRD) analyses were conducted on the clay-size fraction
(lt2microm) from Site JR3 using a Siemens Diffractometer D5000 at the University of Idaho
The clay-size fraction was isolated using a suspension method described by Vaniman
(2001) Organic matter and calcium carbonate were removed from the clays using a
concentrated hypochlorite solution (Kunze and Dixon 1986) Crystalline and non-
crystalline iron oxides were removed using a citrate-sodium bicarbonate-sodium
dithionite (CBD) solution (Kunze and Dixon 1986) The remaining phyllosilicate
minerals were analyzed using a series of treatments including 1) magnesium-saturation
(2 M MgCl and 2 M MgAOC) and air drying 2) magnesium-saturation and glycolation
(1 glycerol-ethanol solution) 3) potassium-saturation (1 M KCl) and air drying and 4)
potassium (1 M KCl) saturation followed by heating to 500degC (Whittig and Allardice
1986) X-ray diffraction patterns were collected between 2θ values of 20-350deg and 20-
150deg with a generator potential of 30 kV a generator current of 22 mA (using CuKα
radiation) a Ni filter and a scan speed of 1degmin The software package JADE was used
for XRD data analysis and mineral identification
X-ray Absorption Spectroscopy
In-situ fluorescence yield X-ray absorption near-edge structure (XANES)
spectroscopy at the Cr K-edge (edge inflection point is 5989 eV for Cr(0)) was conducted
16
at the Stanford Synchrotron Radiation Laboratory (SSRL) Stanford CA using
beamlines 4-1 and 4-3 A Si(220) monochromator and a Ge-detector were used and
three scans were collected for each XANES analysis Soil and rock samples from Site
JR3 were prepared by removing the gt2mm fraction and crushed to a fine powder
Samples were loaded into a plastic sample holder and covered with Mylar tape
Elemental mapping of three samples were completed at SSRL on Beamline 11-2
using X-ray absorption spectroscopy (XAS) A Si(220) monochromator with a 6 microm
focusing capillary was used and fluorescence intensity was monitored with a Ge detector
Samples were prepared similar to those used for mineral identification using the electron
microprobe however the thickness of the sample was decreased to lt20 microm
Sequential Extraction Procedure
Sequential extraction experiments were completed on 2 g lt2 mm sieved soils
(Diablo clay from Site 1 at 10 cm depth and serpentine soil from Site 2 at 4 cm depth) in
three steps The exchangeable fraction was obtained using a barium chloride (1mM
BaCl2) solution for 1 h at 25 degC (Kunze and Dixon 1986) The Fe-crystalline oxide
fraction was obtained by reacting the solids with a hydroxylamine-hydrochlorideacetic
acid (10 M NH2OHsdotHCl in 25(vv) HOAc) solution for 6 h at 60 degC (Chester and
Hughes 1967) The silicate fraction was obtained by reacting the solids with
concentrated hydrofluoric acid (10 M HF) for 16 h at 25 degC Trace element
concentrations for each extraction were measured on the supernatant using inductively
coupled plasma optical emission spectrophotometry (ICP-OES) All analyses were
completed in triplicate Additionally sequential extraction experiments were completed
on 2g 106-250 mm sieved ultrasonically cleaned commercial grade chromite (MSDS
No 302-1 WHMIS Class D2A) using the same procedures listed above The
stoichiometric composition of chromite obtained using the electron microprobe is
(Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 Chromite was imaged after each sequential
extraction using a JEOL 5600LV environmental SEM equipped with an EDAX thin-
window energy dispersive X-ray detector
17
RESULTS
Serpentinite Mineralogy
The average serpentinite at Jasper Ridge is composed of serpentine (mainly
lizardite and antigorite 70) chlorite (15) talc (10) magnetite (4) and chromite
(1) based on petrographic and microprobe observations of five samples (JR3E JR6E 5
10 and JR3) however brucite has been reported as a major constituent in related
Franciscan Complex serpentinites (Hostetle et al 1966) Other minerals identified
include olivine augite and enstatite Although the serpentinite mineralogy is relatively
simple the modes and textures of the minerals present in the rock vary significantly from
site to site In Table 21 whole rock serpentinite compositions from Sites JR1 JR3E
JR6E 2 5 10 and JR3 are listed The average Cr concentration of the serpentinites is
~1700 mg kg-1 (tab 1) Iron (~7) Mg (~11) Ni (~3000 mg kg-1) and Co (~100 mg
kg-1) concentrations are also elevated in the serpentinite relative to the chert and
sandstone bedrock adjacent the main serpentinite body (Figure 21b and Table 21)
A range of rock and mineral textures are petrographically evident in the
serpentinite at Jasper Ridge (Figure 23) No foliations or lineations are observed in the
serpentinite (Figure 23a) obtained from the bottom of the soil pit excavated at Site JR3
(Figure 21b) Subhedral chromite and magnetite grains are randomly interspersed
amongst the serpentine matrix (Figure 23a) In serpentinite obtained in outcrop at Site
JR5E (Fig 21b) a lsquocrocodile skinrsquo texture is produced by thin irregular veins of fine-
grained subhedral magnetite separating serpentine grains (Figure 23b) In Figure 23c
numerous crosscutting veins of chlorite occur in orthopyroxene and define the foliation of
the rock in outcrop at Site JR3E (Figure 21b) The numerous crosscutting microscopic
veins random foliation patterns (or the lack thereof) and subhedral oxide grains suggest
the Jasper Ridge serpentinite has undergone extensive physical and chemical alteration
typical of most serpentinites
Microprobe analyses of minerals in the serpentinite demonstrate that the main Cr-
bearing phases are chlorite enstatite augite Mg-Al chromite Cr-magnetite and a
chromite-silicate mixture (defined in the following paragraph) Selected compositions
for these minerals are listed in order of their Cr concentrations in Table 22a Chlorite
contains the least Cr whereas the chromite-silicate mixture is the most enriched with
18
respect to Cr Figure 24 shows correlations of Cr content relative to Al Fe Mg and Si in
order to provide insight into the serpentinite mineralogy and to aid in evaluating trends of
Cr substitution
A fine-grained mixture containing spinel (chromite andor Cr-magnetite) and an
unidentifiable Cr-silicate contains the highest Cr concentration in the serpentinite at
Jasper Ridge (Table 22a shown in Figure 25a and 25b) This fine-grained mixture will
be referred to here as the chromite-silicate mixture (CSM) with chemical properties
similar to those in Table 22a and Figure 24 CSM is common around the rims of
protolith chromite grains and a number of the larger (~05 mm) Cr-magnetite grains in
the serpentinite The microcrystalline phase responsible for contributing Si to these
analyses is possibly chlorite due to similarities of false colors generated by backscattering
electron (BSE) images and chlorite commonly rimming CSM Microprobe analyses of
CSM listed in Tables 22 and 23 are considered as mixtures of several mineral phases
due to the fine-grained texture (lt5 microm) Chromium concentrations range from 12 to 16
Cr Wt in this phase CSM is chemically distinct from the chromite and magnetite as
shown in Figure 24
The chemical compositions of protolith chromite exhibit little to no variation in
any of the serpentinites collected at Sites 2 JR3E JR4E JR5E JR6E JR7E and JR3
(Table 22a and Figure 24) Chromite contains low Fe (~9 Wt ) high Mg (~9 Wt
) and high Al (~26 Wt ) compared to Cr-magnetite The average Cr concentration
for chromite is ~12 Cr Wt Additionally chromite grains are irregular and subhedral
suggesting that they have undergone significant alteration during metamorphism
Chromium concentrations in the magnetite vary depending on the proximity and
degree of alteration of the chromite (Table 22a and Figure 24) Magnetite grains with
less than 2 Cr Wt are common in veinlets in serpentinite not formed adjacent
chromite whereas magnetite with 8 to 12 Cr Wt were formed proximal or adjacent
chromite Chromium-enriched magnetite is also referred to as ferrite-chromite in the
literature (Spangenberg 1943 Burkhard 1993)
Chlorite and pyroxenes contain minor concentrations of Cr (lt1 Wt Table
22a) Aluminum Fe and Mg concentrations varied in the pyroxenes however the
majority of chlorites are relatively similar in composition (Figure 24) Chlorites are
19
more abundant in the serpentinite compared to the pyroxenes and both minerals were
subhedral and randomly distributed over the entire sampling area (Figure 21b)
Backscattering electron (BSE) images of what appeared to be a petrographically
opaque single chromite grain in Figure 23a demonstrates chromite has been
metasomatically altered resulting in the formation of Cr-magnetite Cr-chlorite and CSM
(Figure 25a) Represented compositions for these minerals (indicated by letters in Figure
25a) are listed in Table 23 Magnifying the chromite and CSM (Figure 25a)
demonstrates the fine-grained texture of CSM and an alteration rim (transition zone) is
also seen along the chromite edge (Fig 25b) This alteration rim is depleted in Cr with
respect to the chromite as shown in Figure 25b
In Figures 23a and 25c Ni was identified in three phases in the bedrock
including olivine (F 03 Ni Wt ) serpentine (G and H 03 and 0 Ni Wt ) and Ni-
Fe (I sim55 Ni Wt ) alloy (Figure 25c and 25d) their compositions are indicated by
letters and listed in Table 23 Alteration of the olivine resulted in the formation of lt10
microm Ni-Fe alloys shown in Figure 25c These Ni-Fe alloys form a corona surrounding
the remaining olivine and the illustration in Figure 25d highlights the minerals and
chemical changes
Chromite and Cr-magnetite alteration leading to the paragenesis of CSM and Cr-
chlorite is common in the serpentinites at Jasper Ridge The petrographic image in
Figure 26a and its corresponding BSE image in Figure 26b illustrate Cr-magnetite
(originally protolith chromite) partially altering to CSM and Cr-chlorite The chemical
compositions of each mineral (indicated by letters in Figure 26b) are listed in Table 23
CSM appears to marginally and metasomatically replace Cr-magnetite and is enriched by
~2 Cr Wt Chromium-chlorite is present in microcracks in Cr-magnetite and
encompasses the entire Cr-magnetite and CSM grain Serpentine minerals adjacent the
Cr-chlorite contains no measurable Cr
Distribution of Chromium in Serpentine Soils
Serpentine soils and underlying bedrock were collected at ten sites along a 500
meter transect extending the length of the main serpentine body as shown in Figure 21b
20
The range of Cr concentrations at each site (center) and Cr concentrations with respect to
depth (periphery) are shown in Figure 27 and soil analyses are listed in Table 24
Bedrock Cr concentrations for Sites 2 5 and 10 are also shown in these plots (Fig 27)
Chromium enrichment does not appear to be consistently found at a particular depth or
soil horizon for any of the sites Sites 3 9 and 10 demonstrate a Cr maximum at
approximately 006 m however Sites 1 2 5 and 6 exhibit a Cr minimum at roughly the
same depth The highest soil Cr concentration in the soil along the transect is 4760 mg
kg-1 at Site 3 at 006 meters depth The lowest soil Cr concentration is 1725 mg kg-1 at
Site 10 at 0085 meters depth The average bedrock Cr concentration is ~2000 mg kg-1
(Figure 27 Table 24) It is important to note Site 1 consists of Diablo clay not
serpentine soil however this site is located adjacent and down-slope of the main
serpentinite body From these observations it is apparent that there are no systematic
trends in soil Cr concentrations with respect to the position of the site and depth and soil
samples often contain higher concentrations of Cr with respect to their corresponding
bedrock and to the average Cr concentration of ultramafic rocks (1600 mg kg-1 Green
1972)
Trace metals including Ni Mn and Co were also evaluated in the serpentine soil
for Sites 1 through 10 and the results are listed in Table 24 Nickel concentrations range
from 1590 to 3960 mg kg-1 and do not directly correspond to the Cr fluxes with respect
to depth (Table 24) Nickel concentrations are less than Cr from Sites 1 to 4 From Sites
5 to 10 Ni concentrations exceed those of Cr averaging ~3600 mg kg-1 Unlike Cr Ni
concentrations are equivalent or greater in the bedrock (~3300 mg kg-1) as compared to
the soil (~3200 mg kg-1) Soil Mn concentrations at each site are relatively uniform at
~1800 mg kg-1 with little to no variation with depth (Table 24) Cobalt concentrations
do not exceed 238 mg kg-1 (Table 24) and are relatively uniform (~200 mg kg-1) with
respect to location along the transverse and depth at each site
Major elements including Al Fe Ca and Mg are listed for the serpentine soils
along the transect with respect to depth in Table 24 Aluminum Fe Ca and Mg
concentrations in the soils at each site are similar with exception of Site 1 (Diablo clay)
Aluminum concentrations in the Diablo clay are ~46 whereas the serpentine soils
contain less than 2 Al (Table 24) Iron and Mg concentrations are also low lt65 and
21
lt411 respectively in the Diablo clay compared to Fe (gt8) and Mg (gt6)
concentrations in the serpentine soil (Table 24) Calcium concentrations are low (lt1)
for all the serpentine soils and even the Diablo clay (Table 24) With a broad
understanding of the chemical characteristics of the serpentine soil at Jasper Ridge Site
JR3 (Figure 21b) was chosen for a detailed chemical study to examine all the factors
governing the chemistry of Cr in the soil
Serpentine Soil Chemistry and Mineralogy at Site JR3
Soil chemistry
Chemical properties (pH electrical conductivity organic matter nitrogen and
ammonium acetate extractable cations) and textures for four soil samples from Site JR3
are listed in Table 25 The pH of the soil is near neutral (pH from 671 to 698) and
increases slightly with respect to depth Electrical conductivities (EC) in the soil ranges
from 03 to 09 dSM a range indicative of relatively few dissolved salts andor major
dissolved inorganic solutes The organic carbon content of the soil is 32 at the surface
and decreases with depth to 10 near the bedrock Total nitrogen for the samples ranged
from 003 to 064 Ammonium acetate extractable Ca K Mg and Na are listed in
Table 25 The extractable concentration of both Ca and K decrease with depth whereas
there is no apparent trend associated with the concentrations of Mg and Na The
serpentine soil textures are dominantly clay loams however the texture is a sandy clay
loam near the bedrock
Whole soil samples and size fractions including gt2 mm sand (2-002 mm) silt
(002-0002 mm) and clay (lt2 microm) size fractions with respect to depth were chemically
digested and analyzed for total trace metal concentrations shown in Figure 28 and listed
in Table 26 to determine if the abundance of Cr was linked to a particular size fraction
(the percent of sand silt and clay size fractions are reported in Table 25) Additionally
Ni Mn and Co concentrations with respect to depth and size fraction are also presented in
Figure 28 and Table 26 Total and gt2mm size fraction Cr concentrations are similar at
the top and bottom of the soil profile ranging from 3297 to 5976 Cr mg kg-1 Chromium
concentrations exceed detections limits (gt10000 mg kg-1) for the sand-size fraction It is
important to note that the sand-size fraction corresponds to the size of most chromite
22
grains observed in the serpentinite Chromium in the clay fractions are low (~1000 mg
kg-1) compared to the other size fractions however these concentrations are still enriched
compared to non-serpentine soils
In Figure 28 and Table 26 Ni concentrations are enriched in the clay-size
fraction with values ranging from 2840 to 2920 mg kg-1 The silt-size fractions have the
lowest Ni concentrations (~1800 mg kg-1) Manganese concentrations are slightly
elevated relative to total Mn in the sand-size fraction with values ranging from 1728 to
2701 mg kg-1 The clay-size fraction is the least Mn enriched size-fraction with
concentrations at ~1200 mg kg-1 Cobalt concentrations are elevated in the sand-size
fraction with values ranging from 159 to 241 mg kg-1 and concentrations in the gt2mm
silt and clay-size fractions are similar to the total (~150 mg kg-1)
Soil Cr-bearing minerals
Chromium-bearing minerals in the soil identified using the electron microprobe
include chromite Cr-magnetite and CSM Chromium concentrations in soil chromite
Cr-magnetite and CSM are graphically shown in Figure 29 as a function of Al Fe Mg
Si and Al+Fe in Wt Representative analyses of the Cr-bearing soil minerals are listed
in Table 22b Compositions of soil chromite in Figure 29 are chemically distinct from
those identified in the protolith in Figure 24 Chromium concentrations in soil chromites
range from 22 to 37 Cr Wt containing minor concentrations of Al and Mg Values for
chromite in the serpentinite bedrock (Table 22a and Figure 24) and the FeCr2O4 end-
member are also shown in Figure 29 The dashed lines in Figure 29 connect Jasper
Ridge bedrock chromite (Figure 24) and stoichiometric FeCr2O4 where it can be seen
that compositions of soil chromite lie close to these tie lines Overall soil chromite
demonstrates a decrease in Al and Mg and an increase in Fe relative to Cr compared with
protolith chromite Soil Cr-magnetites contain between 7 and 20 Cr Wt and are
enriched in Cr compared to Cr-magnetites identified in the bedrock (Figure 24) Cr-
magnetite however reveals a decrease in Fe relative to Cr Minor concentrations of Si
(lt2 Wt ) in the soil chromite and Cr-magnetite may represent slightly altered rims or
contamination from sample preparation Chromium concentrations in CSM maintain a
consistent Cr concentration of ~11 Cr Wt similar to those found in the bedrock (~14
23
Cr Wt ) Soil CSMs with compositions between 8 and 12 Mg Wt are more than
likely associated with analyses containing more microcrystalline chlorite compared to
soil CSMs with less than 2 Mg Wt Soil CSM appears be less enriched with Si (lt14
Wt ) relative to those identified in the serpentinite (15 to 27 Wt )
The size of the Cr-bearing minerals varied significantly for each sample ranging
from 10 microm up to 2 mm The majority of chromite grains in the soil correspond to the
sand-size fraction and also vary from euhedral to subhedral with no discernable pattern
The sum of cations based on four oxygens for the chromite and Cr-magnetite analyses
range between 297 and 301 indicating these minerals are spinels not (oxy)hydroxides
Compositions obtained for chromite and Cr-magnetite may represent alteration rims
where weathering of the mineral in the subsurface environment has been most active
BSE images of chromite and Cr-magnetite were not able to definitively identify whether
these analyses in fact represent alteration rims
Mineralogy of the Clay-Size Fraction
The clay-size fraction in the serpentine soil at Site JR3 was evaluated with respect
to depth using XRD and specific chemical and heat treatments The d-spacings (Aring) and
their relative intensities for each soil and treatment with respect to depth are listed in
Table 27 Figure 210 shows the XRD analyses for clay sample JR3515 (depth 5-15
cm) and these analyses are almost identical to the other clays examined at Site JR3
Additionally Table 28 from Whittig and Allardice (1986) is provided to interpret the X-
ray diffraction analyses in Table 27 For example a d-spacing of ~14Aring is present in all
the treatments indicating the presence of chlorite in the clay-sized fraction Using an
XRD analysis program the clay-size fraction was determined to be composed of
smectite vermiculite serpentine minerals (lizardite and antigorite) and chlorite
(clinochlore and Cr-clinochlore) Smectite is the most abundant mineral in the clay-size
fraction whereas Cr-chlorite is the least abundant Overall the d-spacings and relative
intensities are similar despite the treatments and changes in depth (Table 27) This
indicates that the mineralogy of the clay-size fraction does not vary significantly in the
soil column
24
Chromium concentrations in the clay-size fractions are ~1010 mg kg-1 (Table
26) Smectite and vermiculite are possible Cr-bearing clay minerals in which isomorphic
substitution of Cr(III) could occur in the dioctahedral sheet Serpentine minerals
(lizardite and antigorite) are not likely hosts of Cr based on crystallographic
considerations and the absence of Cr in analyzed serpentine minerals in the bedrock
(Table 23) Clinochlore may also be a host for Cr in the inner layer octahedral sheet
Chromium-clinochlore is the only Cr-containing mineral identified in the clay-size
fraction using XRD and Cr-clinochlore (Cr-chlorite) is also an abundant mineral
identified in serpentinite protolith (Tables 22 and 23 Figure 26)
XAS of chromium in relation to depth
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in serpentine soils and bedrock at Site JR3 The XANES (X-ray absorption near-edge
structure) spectra demonstrate the absence of Cr(VI) in the soil column at Site JR3
(Figure 211) based on the lack of a pre-edge feature The detection limit for Cr(VI)
using this method is approximately ~20 mg kg-1 Ultimately XAS analyses demonstrate
Cr(VI) is not detected in any of the soil solids
X-ray microprobe analyses of soil and rock samples (0-5cm depth 5-10cm depth
and serpentinite) at Site JR3 were processed to map the relative concentrations of Fe and
Cr in which areas of elemental enrichment are identified by a violet color areas of
depletion by a blue color (Figure 212) Areas of Cr and Fe enrichment in Figure 212 are
localized with well-defined boundaries corresponding to grain edges and these elements
are not correlated in a well-defined trend Minerals assigned to these areas of Cr and Fe
enrichment were made based on petrographic and microprobe observations on the same
thin section For example high Fe and no Cr is magnetite whereas high Fe and high Cr
is either chromite or Cr-magnetite Chromite magnetite and Cr-magnetite appear to be
present in the serpentine soil at a depth between 0 to 5 cm (Figure 212A) In Figure
212b CSM is a possible mineral detected in the soil (depth 5-10 cm) and the presence
of magnetite is clearly seen due to the localized enrichment of Fe and no Cr Iron and
Cr spectra for the serpentinite were much more heterogeneous as compared to the soil
samples and Cr concentrations appear be more scattered (Figure 212c) These X-ray
25
microprobe analyses provide a complimentary method for examining Cr geochemistry
(including Cr(III) and Cr(VI)) in naturally occurring soil and rock solids however
Cr(III) is the only detected valence state in serpentine soils and rocks at Jasper Ridge
Sequential Extractions
Sequential extractions were completed on 1) Diablo clay (Inceptisol) from Site 1
(Figure 21b) at a depth of 10 cm 2) serpentine soil (Mollisol) from Site 2 (Figure 21b) at
a depth of 4 cm and 3) commercial grade chromite (106-250 microm) and these results are
shown in Figure 213 and listed in Table 29 Chromium extracted from the
exchangeable fraction is minimal (lt01 mg kg-1) for all the samples (number 1 in Table
29 Fig 213) Minor Cr concentrations are associated with the Fe-crystalline oxide
fraction in the Diablo clay (19 Cr mg kg-1) and serpentine soil (78 Cr mg kg-1) whereas
chromite only released 05 Cr mg kg-1 (number 2 in Table 29 Figure 213) Increased
concentrations of Ni and Fe are present in the Fe-oxide fraction of the Diablo clay (161
Ni mg kg-1 and 4474 Fe mg kg-1) and serpentine soil (5138 Ni mg kg-1 and 5138 Fe mg
kg-1 Table 29) Chromium extracted from the Diablo clay serpentine soil and chromite
in the silicate fraction is 14 31 and 14 Cr mg kg-1 respectively (number 3 in Table 29
Figure 213) The sum of extracted Cr Fe Ni and Mn concentrations (number 4 in
Table 29) are low with respect to total metals (number 5 in Table 29) In fact only 07
12 and 00 Cr was removed from the Diablo clay serpentine soil and chromite
respectively (number 6 in Table 29)
Chromite was extremely resistant to all the extractions and less than 001 of the
total Cr was removed (Table 29) Scanning electron microprobe (SEM) images of the
chromite after each extraction are shown in Figure 214 Chromium concentrations
released from the chromite (Table 29 and Figure 214a) increased with each sequential
extraction experiment starting with the exchangeable fraction (Figure 214b) progressing
into the Fe-crystalline oxide fraction (Figure 214c) and ending with the silicate fraction
(Figure 214d) The surface of the chromite became smoother after each extraction and
crystal imperfections in defect sights located along the edges and corners progressively
decreased (Figure 214) Chromite does not appear to be an easily accessible source of
Cr based on these extraction experiments
26
DISCUSSION
Serpentinite at Jasper Ridge has undergone significant physical and chemical
alteration since its formation and emplacement off the California continental margin
Despite being subjected to high pressures (~5-7 kbars) and elevated temperatures (200-
300degC) trace element concentrations within the serpentinite remain elevated Jasper
Ridge serpentinites are enriched with Cr Co Fe Mg Mn and Ni compared to chert and
sandstone located adjacent the main serpentinite body (Table 21 and Figure 21b) Over
the past 400000 years Cr-bearing minerals in the serpentinite have been weathering and
accumulating in the soil During this time Cr concentrations in the soil have
progressively increased relative to Cr-concentrations in the serpentinite protolith This Cr
enrichment is directly linked to specific weathering-resistant minerals identified in the
soil
Chromium in Serpentinites
Three minerals in the bedrock at Jasper Ridge contain appreciable concentrations
of Cr Cr-magnetite (82-103 Cr Wt ) chromite (~108 Cr Wt ) and CSM (~133
Cr Wt ) Cr-magnetite and CSM were derived from the alteration of primary igneous
chromite CSM or Cr-spinels with high concentrations of Si have been noted by
(Burkhard 1993) as a Cr-spinel containing a possible microscopic silicate phase Our
analyses support this result in which the microcrystalline silicate is probably Cr-chlorite
due to its common occurrence with CSM (Figure 26) Definitive identification of this
microcrystalline mixture was not possible due to size of the crystals relative to the spot
size of the electron microprobe Chlorite and pyroxene contain minor concentrations of
Cr (~1 Wt ) and these were the only identified Cr-containing silicates in the
serpentinite Minor concentrations of Cr observed in some chlorites are directly related
to the metasomatic alteration of chromite and Cr-magnetite (Figure 26) Other common
metasomatically-derived Cr-bearing silicates such as fuchsitemariposite and uvarovite
are not present possibly due to the lack of aqueous K+ and Ca+2 in the alteration fluids
Overall the chemical compositions of these protolith minerals (Figure 24) demonstrate
only minor chemical variations allowing comparative analysis with Cr-bearing soil
minerals
27
Chromium in Serpentine Soils
Cr-bearing soil minerals identified using the electron microprobe include
chromite Cr-magnetite and CSM and these phases are chemically distinct from those in
the bedrock (Figure 24 and 29) The highest concentration of Cr (41 Wt ) in the soil
is in chromite with a composition approaching the FeCr2O4 end-member This represents
a ~80 increase in the Cr concentration relative to bedrock chromite (see Figure 215)
Cr-magnetite also demonstrates Cr enrichment (up to 20 Cr Wt ) with respect to
magnetite identified in the bedrock (~10 Cr Wt ) (Figures 24 and 29) The textures of
the soil Cr-bearing spinels exhibit sharp irregular edges possibly due to extensive
weathering CSM in the soil is similar to those found in the bedrock with Cr
concentrations ranging from 11 to 14 Cr Wt however Si concentrations in soil CSMs
(8 to 14 Wt ) are noticeably less than those measured in the serpentinite (15 to 20 Wt
) The weathering of CSM might preferentially release Si relative to other elements in
its structure Cr-silicates (chlorite and pyroxenes) were not identified in the soil possibly
due to complete weathering of these phases or they were too fine-grained for microprobe
analyses however Cr-clinochlore was identified in the clay-size fraction using XRD
Elemental maps shown in Figure 212 corroborate electron microprobe observations by
revealing well-defined zones of Cr and Fe enrichment in both the soils and serpentinite
These areas of Cr and Fe enrichment are either chromite Cr-magnetite CSM or
magnetite Overall these maps reveal Cr is not homogeneously distributed in the soil
confirming Cr is in specific mineral phases Additionally the observed abundance of
chromite Cr-magnetite and CSM in the soils indicates that these minerals are probable
sources for a majority of the Cr enrichment identified in the soils
Soil Spinels
Soil chromite and Cr-magnetite appear to have undergone chemical alteration
enriching the minerals with respect to Cr In Figure 215 compositions of bedrock
chromite and end-member FeCr2O4 are shown together with tie-lines like those in figure
8 connecting both minerals Reaction progress in Figure 215 is based on the assumption
that the serpentinite chromite (0 reaction progress) is proceeding toward FeCr2O4
(100 reaction progress) in the serpentine soil Soil chromites from Jasper Ridge are also
28
plotted in Figure 215 where Cr concentrations are assumed to vary linearly between the
bedrock chromite and chromite end-member which defines the reaction progress of the
mineral These tie-lines correlate extremely well to the chemical composition of
chromites identified in the soil For example chromite containing ~30 Cr Wt will
have ~15 Fe Wt ~13 Al Wt and ~5 Mg Wt similar to soil chromite in Figure 29
The retention of Cr in chromite shown in Figure 215 has significant implications towards
the availability and cycling of Cr in serpentine soils
One possible mechanism for the chromite composition trends illustrated in
Figures 29 and 215 is incongruent dissolution where Cr is retained relative to other
elements Sequential extraction experiments (Figure 213 and Table 29) support this
idea in which Fe is more easily released than Cr (1096 Fe mg kg-1 to 19 Cr mg kg-1)
from chromite Unfortunately Mg and Al were not monitored during the experiments
From these sequential extraction experiments and chromite analyses (Figure 215) Al
and Mg have been preferentially released from the chromite structure compared to Fe and
Cr Additionally the alteration of chromite in the serpentinite and the soils are analogous
to one another Chromite in the serpentinite retains a majority of its Cr compared to Mg
and Al when being altered to CSM by fluids at elevated pressures and temperatures
(Figure 26 and Table 22) Under ambient soil conditions chromite is once again
retaining Cr relative to Mg and Al causing what appears to be an increase in Cr
concentration Chromium is not easily released from chromite whether under high
pressures and temperatures or in the soil
Crystal field stabilization energies (CFSE) and octahedral site preference energy
parameters (OSPE) provide a means to explain the incongruent dissolution or Cr
preservation in the Cr-spinels CFSE is a measure of the net energy of stabilization
gained by a metal ions nonbonding d electrons as a result of complex formation This
measure also indirectly describes the energy required to remove an ion from a crystal
structure The octahedral CFSE for Cr(III) is ndash2247 kJmole and the OSPE (the
difference between the tetrahedral and octahedral CFSE) is ndash1578 kJmole (Dunitz and
Orgel 1957 McClure 1957) These values are the most negative for any transition
element and indicating the presence of Cr(III) in an oxide increases the crystalrsquos stability
compared to an analogous crystal lacking Cr(III) Chromium(III) would be among the
29
last ions to removed from the spinel structure compared to other transition elements such
as Fe2+and Mn2+
Soil Chemistry related to Chromium Availability
In Table 25 the chemical properties of the soil at Site JR3 do not appear to be a
factor when evaluating Cr in the soil The pH of the soil did not vary considerably with
depth Additionally the near neutral pH would not have a significant impact accelerating
Cr-bearing mineral dissolution Organic matter within the soil was elevated despite the
low plant productivity of the serpentine soil The role of organic matter could be a
significant factor such as the production of organic acids promoting the dissolution of the
Cr-bearing minerals however Cr concentrations near the surface (where organic matter
was the greatest) are not significantly greater or less than those deeper in the soil profile
The relative concentrations of ammonium acetate extractable Ca K Mg and Na are
characteristic of serpentine soils and did not provide any new insight Finally Cr(III)
was the only identifiable valence state of Cr based on X-ray absorption spectroscopy
(Figure 211) which is what should be expected based on the Cr-bearing minerals and the
pH of the soil
Extraction experiments summarized in Table 29 and Figure 213 demonstrate
~99 of the Cr in a serpentine soil is retained in highly resistant phases at Jasper Ridge
The chromite standard used in these experiments contributed negligible concentrations of
Cr (lt2 mg kg-1 total) for all of the sequential extraction experiments supporting earlier
evidence that Cr-spinels are retaining Cr and are potentially responsible for a majority of
the total Cr enrichment in the serpentine soil At Jasper Ridge Cr extracted from the
serpentine soil (31 mg kg-1) is likely the result of the Cr-silicates compared to the Cr-
spinels Low Cr-containing minerals (possibly chlorite enstatite augite and clay
minerals) appear to be the most probably source of Cr to soil solutions and vegetation of
the serpentine soil Approximately 30 of the serpentine soil is composed of clay-sized
minerals and this size-fraction has ~1000 Cr mg kg-1 The mineral responsible for Cr
enrichment in the clay-size fraction is more than likely Cr-chlorite identified using XRD
Chromium-chlorite in the soils suggests that this mineral is potentially the largest silicate
contributor of Cr in the serpentine soil
30
Interpreting the enrichment and behavior of Cr in serpentine soils such as in
Figure 27 can be accomplished based on the abundance Cr-spinels observed in the soil
Small grains of chromite Cr-magnetite and even CSM in the soil have the potential to
appreciably impact Cr concentrations Table 210 demonstrates the chromite masses
required to change the Cr concentration in a 1 kg sample of soil The highest Cr
concentration at Jasper Ridge was identified at Site 3 with a measured value of 4760 mg
kg-1 which means only ~1 of the soil sample (Table 210) needs to contain chromite
Additionally a minimal amount (lt011 of the sample as shown in Table 210) of
chromite is necessary to change the concentration by 500 mg kg-1 which could easily
cause the sharp Cr increases and decreases noted in Figure 27 This agrees with our
observations of chromite abundance in the serpentine soils Additionally a majority of
the chromite grains identified in the bedrock are between 2 and 002 mm which
correspond directly to the elevated Cr concentrations (gt10000 mg kg-1) in the sand-size
fraction of the serpentine soil (Table 26) Even the ~15000 Cr mg kg-1 values in soils
measured by Hotz (1964) in Figure 22 only require ~2 of the sample of the sample to
be chromite A few grains of chromite possess the ability to significantly impact
measured Cr concentrations in soil samples
Chromium and Nickel in Serpentine Soils
Chromium and Ni have often been correlated to one another due to their common
enrichment in serpentine soils (Hotz 1964 Rabenhorst et al 1982 Schwertmann and
Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al 1989 Kaupenjohann
and Wilcke 1995) A majority of the Ni in the Jasper Ridge serpentine soil originated
from olivine (Figure 25c and 25d Table 23) Chemical alteration of olivine during
serpentinization resulted in the mineral serpentine (Coleman and Jove 1992) in which
minor concentrations of Ni (lt037 NiO Wt ) substituted into its structure Continued
alteration under highly reducing conditions and the incompatibility of Ni in the serpentine
structure resulted in the formation of Ni-Fe alloy (Figure 25c) This type of alteration is
observed in many of the rocks at Jasper Ridge where the Ni-Fe alloys are ~10 microm in size
Olivine in the rocks and soil appear to be less resistant to weathering compared to
chromite therefore Ni is possibly more available to bacteria and vegetation in serpentine
31
soils Sequential extractions (Figure 213 Table 29) support these observations in which
a greater percentage of total Ni was extractable compared to the Cr in the soils Although
Cr and Ni are enriched in serpentine soils the availability and behavior of these two
elements are significantly different
CONCLUDING REMARKS
Chromium is predominantly present in primary chromite Cr-magnetite and a
chromite-silicate mixture (CSM) in both the bedrock and soil despite weathering over
geologic time Sharp variations in soil Cr concentrations with respect to depth are caused
by the presence or absence of chromite Cr-magnetite andor CSM Chromium-spinels in
the soil have undergone incongruent dissolution possibly conserving Cr(III) with respect
to Al3+ and Mg2+ Chromite is highly resistant to weathering and it is more than likely
not contributing bioavailable Cr Although Cr-spinels are a major source of Cr Cr-
silicates in the protolith (chlorite enstatite and augite) and soil (chlorite
montmorillonite and vermiculite) afford the most accessible Cr source identified Based
on our observations and analyses chlorite is the most likely Cr source of these silicate
minerals Finally Cr(VI) was not detected in any of the soil and rock samples
ACKNOWLEDGEMENTS
We thank Jasper Ridge Biological Preserve for allowing us the opportunity to
examine the serpentine soils Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund NSF Grant EAR-9902859 (DKB) GSA Graduate
Student Research Grants Office of Teaching and Learning (SF) at Stanford University is
gratefully acknowledged We appreciate the technical support from the staff of the
Stanford Synchrotron Radiation Laboratory from Bob Jones for assistance in microprobe
analyses of the serpentinites and serpentine soils and from Ben Bostick Travis Horton
and Theresa Barber for field and lab assistance
32
Table 21 Whole rock elemental concentrations of serpentinites (Figure 21b) chert and sandstone
-----------------------Location of Serpentinitesdagger----------------------- Elemental Concentrations JR1 JR3E JR6E Site 2 Site 5 Site 10 JR3
ChertDagger SandstoneDagger
Al 02 08 03 027 22 107 024 079 027 Ca 026 007 005 003 046 013 003 001 lt001 Co mg kg-1 101 99 100 99 155 217 97 1 lt1 Cr mg kg-1 580 2280 1510 1840 1945 2090 1465 10 lt1 Fe 303 71 609 581 76 1106 512 041 024 K 001 031 099 lt001 lt001 001 lt001 029 008
Mg 122 134 154 1336 524 804 112 014 034 Mn mg kg-1 495 800 155 730 1375 1860 460 110 85 Na 181 006 093 lt001 lt001 lt001 lt001 001 001 Ni mg kg-1 2037 1545 3321 3510 3020 3960 2740 20 9
Total digestion ICP-AES analyses daggerSerpentinites are dominantly composed of the minerals serpentine chlorite talc and spinels DaggerSee Figure 21b for sample locations
33
Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in serpentinite rock at Jasper Ridge
Chlorite Enstatite Augite Chromite Cr-Magnetite ChromiteSilicatedagger SiO2 3738 5463 5222 000 024 1053 Al2O3 552 477 516 5059 182 395 NiO 001 01 00 03 00 052 FeODagger 762 614 239 1446 7919 5042 MnO 009 015 009 042 140 182 Cr2O3 024 060 099 1584 1605 1939 MgO 3337 3305 1560 1869 024 605 CaO 023 050 2348 000 000 019 Na2O 000 001 050 000 001 008 K2O 000 000 001 000 000 011 Total 8446 9994 10048 10027 9894 9307
Formula Unit Compositions
Si 370 094 189 000 001 --- IVAl 030 006 011 000 000 --- VIAl 035 004 011 157 008 --- Ni 000 000 000 001 000 ---
Fe+2 063 010 007 030 232 --- Mn 001 000 000 001 004 --- Cr 002 001 003 033 048 --- Mg 493 085 084 073 001 --- Ca 002 001 091 000 000 --- Na 000 000 004 000 000 --- K 000 000 000 000 000 --- O 1400 300 600 400 400 ---
Electron microprobe analyses daggerChromitesilicate is also referred to as CSM DaggerFe+3 is present however all conversions are based on Fe+2
34
Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in serpentine soil at Jasper Ridge
CSM Chromite Cr-Magnetite SiO2 1940 057 267 Al2O3 559 1059 381 NiO 014 001 004 FeOdagger 2629 2259 5244 MnO 155 154 419 Cr2O3 1878 5577 3237 MgO 1668 916 288 CaO 015 001 005 Na2O 006 001 001 K2O 0 008 009 Total 8864 10033 9855
Si --- 002 010 IVAl --- 042 017 VIAl --- 000 000 Ni --- 000 000
Fe+2 --- 063 148 Mn --- 004 013 Cr --- 147 096 Mg --- 046 016 Ca --- 000 000 Na --- 000 000 K --- 000 000 O --- 400 400
Electron microprobe analyses dagger Fe+3 is present however all conversions are based on Fe+2
35
Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25 and 26 A B C D E F G H I J K L Chromite
Chromite CSM Magnetite Chlorite Olivine Serpentine Serpentine Ni-Fe-Metal Cr-Magnetite CSM Chlorite
SiO2 0 0 1053 0 3407 4046 3688 4074 591dagger 021 1143 2855Al2O3
5059 5162 395 034 1114 000 397 085 054 165 980 2293NiO 028 031 052 0 004 037 037 0 5302dagger 0 0 0
FeODagger 1446 1452 5042 9992
1220 901 262 453 1753dagger 8169 3498 806MnO 042 044 182 0 008 010 0 0 0 107 150 00Cr2O3 1584 1576 1939 125 071 001 0 0 0 1369 1905 039MgO 1869 1869 605 003 292 4895 3848 3755 064 1217 2828CaO 0 0 019 0 0 0 0 0 0 0 0Na2O 0 0 0 0 0 0 0 0 0 0 0K2O 0 0 0 0 0 0 0 0 0 0 0H2O --- --- --- --- 1200 --- 1600 1600 --- --- 1200Total 10027 10136 9307
10164 9944 9891 9834 9972
9897 8897 10028
Formula Unit CompositionSi
000 000 --- 000 328 100 184 200 001 --- 269IVAl 000 000 --- 000 072 000 000 000 000 --- 131VIAl 157 159 --- 002 058 000 023 005 007 --- 123Ni 001 001 --- 000 000 001 001 000 000 --- 000
Fe+2 030 029 --- 294 084 019 011 019 239 --- 059Mn 001 001 --- 000 000 000 000 000 003 --- 000Cr 033 032 --- 004 006 000 000 000 041 --- 003Mg 073 073 --- 000 419 180 285 274 004 --- 397Ca 000 000 --- 000 000 000 000 000 000 --- 000Na 000 000 --- 000 000 000 000 000 000 --- 000K 000 000 --- 000 000 000 000 000 000 --- 000O 400 000 --- 000 1400 400 700 700 400 --- 1400
36
Electron microprobe analyses dagger elemental metal Dagger Fe+3 is present however all conversions are based o
36
2321000---
10025
591000054
53021753000000
2321000000000000
n Fe+2
Table 24 Bulk soil elemental concentrations for soils and bedrock collected along the transect
(Fig 21b)
Sites Samples Depth
(meters) Cr
(mg kg-1) Ni
(mg kg-1) Mn
(mg kg-1) Co
(mg kg-1) Al
() Fe
() Ca ()
Mg ()
COI03 0015 2420 1590 1325 118 464 565 065 352 COI36 0045 2370 1630 1600 138 467 55 063 356 COI69 0075 2160 1755 1670 144 471 6 064 365 COI912 0105 2420 1785 1740 151 496 615 067 384
1
COI1215 0135 2570 1985 2340 195 506 649 069 411 COII02 001 3710 2400 1665 169 294 702 100 768 COII25 0035 3600 2290 1625 162 275 673 093 743 COII58 0065 2910 2310 1550 159 25 672 087 771
COII811 0095 3520 2350 1595 158 276 693 093 766 2
COIIBedrock 012 1840 3510 730 99 027 581 003 1336 COIII14 0025 3750 2930 1655 190 178 897 062 762 COIII47 0055 4760 3050 1850 178 178 94 056 805
COIII710 0085 4380 3410 1930 216 216 1044 07 797 3 COIII1013 0115 4320 3380 1955 201 213 1061 063 626 COIV01 0005 3710 2970 1915 214 209 909 053 781 4 COIV15 003 3540 2710 1725 195 181 801 043 792 COV14 0025 2620 2350 1330 150 246 651 063 526 COV47 0055 1775 2710 1520 165 252 766 059 517
COV710 0085 2000 3090 1830 195 225 737 057 583 5 COVBedrock 0115 1945 3020 1375 155 22 76 046 524
COVI14 0025 3040 3720 1855 224 084 1152 015 736 COVI47 0055 2400 3610 1785 218 085 1138 014 68 COVI710 0085 2960 3800 1850 232 093 1232 015 714 6
COVI1013 0115 2800 3860 1800 224 085 1164 014 739 COVII14 0025 2620 3550 1925 229 083 1128 015 718 COVII47 0055 1975 3590 1805 216 084 1085 014 655 7 COVII712 0095 2080 3600 1840 220 07 1105 011 741 COVIII14 0025 2130 3830 1990 238 085 115 017 705 COVIII47 0055 2550 3720 1865 228 085 1099 017 714 8
COVIII712 0095 2240 3910 1860 229 076 1112 014 828 COIX14 0025 2470 3480 1890 227 104 1201 023 579 COIX47 0055 3000 3830 1595 205 084 108 013 842 9 COIX713 01 2720 3710 1845 221 111 1208 017 551 COX14 0025 2810 3740 1760 218 129 114 017 675 COX47 0055 2190 3910 1835 224 131 1206 017 702
COX710 0085 1725 3370 1595 192 11 1061 014 635 10 COXBedrock 0115 2090 3960 1860 217 107 1106 013 804
Total digestion ICP-AES analyses
37
Table 25 Chemical and physical properties of the soil at Site JR3
NH4OAc-extractable ---Hydrometer---Ca K Mg Na
Sand Silt ClaySample Depth (cm) pH EC
dSm
Organic Matter ------
LECO Nitrogen ------ ---------mg kg-1---------
Texture
JR305 25 671 04 32 012 505 2366 3291 355 36 31 33 Clay LoamJR3515
10 683 09 19 064 4205 8176 3343 2392 36 30 34 Clay LoamJR31530 225 688 03 12 003 3725 4903 3750 2305 36 33 31 Clay LoamJR33045 375 698 07 10 003 2873 3665 3388 2549 46 24 30 Sandy Clay Loam
See Methods Section for more details
38
38
Table 26 Selected trace metal concentrations for each size fraction versus depth at Site JR3
Depth (cm) Total gt2mm Fraction Sand Silt Clay Sample
------------------------Chromium (mg kg-1)------------------------ JR305 25 4906 5086 gt10000 4565 988 JR3515 10 5976 4678 8284 4000 1115 JR31530 225 5222 4206 gt10000 4133 980 JR33045 375 3477 3297 gt10000 4571 956
------------------------Nickel (mg kg-1)------------------------------ JR305 25 1935 2050 1660 1305 2920 JR3515 10 2180 2050 1980 1325 2920 JR31530 225 2190 2120 1835 1370 2880 JR33045 375 2090 2140 1740 1285 2840
------------------------Manganese (mg kg-1)-------------------------- JR305 25 1319 1571 1728 1301 1210 JR3515 10 1664 1601 2701 1471 1254 JR31530 225 1639 1626 2217 1599 1231 JR33045 375 1222 1226 1781 1405 1170
-----------------------Cobalt (mg kg-1)-------------------------------- JR305 25 128 146 159 134 131 JR3515 10 155 145 241 147 132 JR31530 225 152 147 194 162 128 JR33045 375 125 124 159 141 122
Total digestion ICP-AES analyses
39
Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3
JR305CLAY Depth 0-5cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1439 100 1833 100 1274 100 1391 100 731 179 1432 431 732 645 1047 105 71 216 941 163 71 41 838 838 474 101 847 141 475 257 726 726 356 125 73 271 365 264 313 83 712 248 355 299
473 143 313 257 364 131 355 193
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1845 1000 1268 1000 1388 1000 852 68 1447 387 724 987 1010 859 733 264 944 143 703 679 836 281 712 247 903 139 471 282 723 576 475 109 850 83 362 244 365 73 733 426 352 321 356 117 712 291 313 154 314 63 475 122
364 170 355 196
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1724 1000 1297 1000 1396 1000 842 45 1397 447 741 957 1175 708 730 318 923 176 715 686 1037 944 710 255 840 118 477 271 837 139 475 97 717 465 367 400 726 375 357 133 468 182 357 386 313 91 362 194 315 286
352 229 JR33045CLAY Depth 30-45cm
Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity
1447 1000 1869 1000 1268 1000 1393 627 842 44 1476 287 737 843 1049 1000 730 207 954 107 715 539 837 190 710 182 915 107 479 157 723 111 475 69 860 61 366 303 364 69 740 332 356 292 356 74 717 256 315 315 313 57 478 111
367 141 357 115
See Methods Section for details
40
Table 28 X-ray diffraction d-spacings for common silicates from Whittig and Allardice (1986)
Diffraction Spacing (Aring) Mineral(s) Indicated Mg-saturated air-dried
14-15 Smectite vermiculite chlorite 99-101 Mica (illite) 71-73 Serpentine
715 Kaolinite chlorite Mg-saturated glycerol-solvated
177-180 Smectite 14-15 Vermiculite chlorite
99-101 Mica (illite) 71-73 Serpentinte
715 Kaolinite chlorite K-saturated air-dried
14-15 Chlorite vermiculite (with hydroxy interlayer) 124-128 Smectite 99-101 Mica vermiculite (contracted) 71-73 Serpentine
715 Kaolinite chlorite K-saturated heated (500ordmC)
14 Chlorite 99-101 Mica vermiculite (contracted) smectite (contracted) 71-73 Serpentine
715 Chlorite
41
Table 29 Results of the Diablo clay serpentine soil and chromite sequential extraction experiments
1 Exchangeable Fraction (1mM BaCl2) 4 Sum of 1 2 amp 3 (mg kg-1) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 0 0 003 Cr 159 388 19 Fe 014 011 001 Fe 24812 25532 1096 Ni 104 103 008 Ni 1854 1868 25 Mn 0 0 0 Mn 778 1172 04
2 Fe-Oxide Fraction (1M NH2OH HCl in 25 (vv) HOAc) 5 Total Cr Fe Ni and Mn (Total Digestion) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 19 78 05 Cr 2290 3310 1767864 Fe 4474 5138 38 Fe 60000 68000 1044325 Ni 1610 1506 01 Ni 1673 2345 00 Mn 231 430 00 Mn 1498 1608 3424547
3 Silicate Fraction (10M HF) 6 Percent Extractable ((45)100) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ------------------------------------- Cr 140 310 14 Cr 07 12 00 Fe 20337 20393 1058 Fe 41 38 01 Ni 234 352 22 Ni 111 80 00 Mn 547 742 04 Mn 52 73 00
ICP-AES analyses
42
Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1 kg of sample
Chromite
FeCr2O4 Cr Concentration
(mg kg-1) Grams Sample 500 11 011 1000 22 022 2000 43 043 5000 108 108
43
Figure Captions
Figure 21a Soils map of Jasper Ridge Biological Preserve (Stanford CA) with
complete descriptions of the soils shown in the legend 21b An enlarged image of the
study area shown in 22a indicates sampling sites of the soils and rocks collected A
dashed line marks the 500 m transect
Figure 22 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above
Figure 23 Petrographic images illustrating the varying serpentinite textures at Sites (A)
JR3 (B) JR5E and (C) JR3E For reference to site locations see Figure 21b
Figure 24 Correlation diagrams of Cr-bearing minerals in the serpentinite at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Si in Wt
Chromite (circle) Cr-magnetite (square) chromitesilicate or CSM (down triangle)
chlorite (hexagon) and pyroxenes (up triangle) analyses are shown on each correlation
diagram
Figure 25 Images of the serpentinite bedrock at Site JR3 demonstrating the mineralogy
and Cr-bearing phases Microprobe analyses are indicated by circled letters and these
values are in Table 23 25a Backscattering electron image demonstrating several Cr-
bearing phases present including chromite Cr-magnetite CSM and Cr-chlorite from
Figure 23a 25b Magnified image shown in Figure 25a demonstrating the fine-grained
texture of CSM Additionally Cr2O3 Wt values are shown for chromite the depletion
zone and CSM 25c Image of Ni-bearing phases present in the bedrock including
olivine serpentine and Ni-Fe alloy from Figure 23a 25d Cartoon illustrating the
mineralogical and chemical changes present in 25c
Figure 26 Images of a Cr-chlorite and Cr-magnetite assemblage in the serpentinite
bedrock 26a Petrographic image of the bedrock at Site 4 (Figure 21b) 26b
44
Backscattered electron image demonstrating the common occurrence of Cr-chlorite
encompassing chromite grains Microprobe analyses are indicated by circled letters and
are located in Table 23
Figure 27 Highest and lowest Cr concentrations (mg kg-1) at each site in the serpentine
soil 500 m transect shown in Figure 21b Samples included soils at various depths (m)
and bedrock from each site The average Cr concentration for the Jasper Ridge
serpentinite is marked on the central figure Chromium concentrations versus depth (m)
at each site (noted in the upper left corner) outline the main diagram
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected
for total gt2 mm sand silt and clay size fractions listed in Table 26
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soil at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Al+Fe in Wt for
chromite (circles) Cr-magnetite (hexagons) and CSM (diamonds) Chromite from the
bedrock is denoted by a larger circle and FeCr2O4 with a square
Figure 210 XRD clay analyses of sample JR3515 (Site JR3 Figure 21b) demonstrate
the influence of each treament Mg-saturated Mg+glycerol K-saturated and K-saturated
+ heat (500oC) The d-spacings and intensities are also listed in Table 27
Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at depths (a) 0-5cm (b) 5-15cm (c) 15-30cm (d) 30-45cm and (e) bedrock The pre-edge
area indicative of Cr(VI) peaks is shown on the right Note that no peaks are present
Figure 212 X-ray microprobe elemental maps of Cr and Fe for (212a) serpentine soil
at a depth of 0-5cm (212b) serpentine soil at a depth of 5-10cm and (212c) serpentinite
bedrock at Jasper Ridge site JR3 (Figure 21b)
45
Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments for the Diablo clay (Site 1 Figure 21b) serpentine soil (Site 2
Figure 21b) and chromite listed in Table 29 The sum refers to the summation of the
three extraction experiments
Figure 214 SEM images of the chromite standard (A) after each sequential extraction
treatment including the (B) exchangeable fraction (C) Fe-crystalline oxide fraction and
(D) silicate fraction The concentration of Cr removed by each treatment is shown
beneath each set of images
Figure 215 Tie-lines from Figure 29 connecting serpentinite chromite and end-
member FeCr2O4 are shown on this single plot Soil analyses from Figure 29 are also
plotted on this diagram For a sum of 100 Wt oxygen must be added to the sum of
Cr Al Fe and Mg Wt
46
Figure 21a
47
Figure 21b
48
Chromium Concentration (mg kg-1)1000 10000
Dep
th (m
eter
s) 01
10
Maryland (Rabenhorst et al 1982)California (Gough et al 1989)California (Hotz 1964)New Foundland (Roberts amp Proctor 1992)
Surface
Figure 22
49
Figure 23
50
Cr (Wt )0 5 10 15 20
Al (
Wt
)
0
5
10
15
20
25
30
Cr (Wt )0 5 10 15 20
Fe (W
t
)
0
20
40
60
80
100
Cr (Wt )0 5 10 15 20
Mg
(Wt
)
0
5
10
15
20
25
Cr (Wt )0 5 10 15 20
Si (W
t
)
0
5
10
15
20
25
30
Cr0 10 20 30 40 50 60 70 80 90 100
Al
0
10
20
30
40
50
60
70
80
90
100
Fe
0
10
20
30
40
50
60
70
80
90
100
Chromite
Chlorite
Magnetite
Chromite-Silicate
Pyroxenes
Chromite
Chlorite
Pyroxenes
Magnetite
Chromite-Silicate
Figure 24
51
Figure 25
52
Figure 26
53
Sites1 2 3 4 5 6 7 8 9 10
Con
cent
ratio
n (m
g kg
-1)
1500
2000
2500
3000
3500
4000
4500
5000Soil Chromium
1
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000002004006008010012014016
2
Cr Concentration (mg kg-1)
1500 2000 2500 3000 3500 4000
Dep
th (m
eter
s)
000002004006008010012014
3
Cr Concentration (mg kg-1)3500 4000 4500 5000
Dep
th (m
eter
s)
000
002
004
006
008
010
012
5
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000002004006008010012
6
Cr Concentration (mg kg-1)2200 2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000002004006008010012
7
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000
002
004
006
008
010
9
Cr Concentration (mg kg-1)
2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000
002
004
006
008
010
012
10
Cr Concentration (mg kg-1)1600 2000 2400 2800 3200
Dep
th (m
eter
s)
000002004006008010012
4
Cr Concentration (mg kg-1)3500 3600 3700 3800
Dep
th (m
eter
s)
000
001
002
003
004
8
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000
002
004
006
008
010
BEDROCK
BEDROCK
BEDROCK
Range of Cr Concentrations in Serpentinite Bedrock
Figure 27
54
Soil Chromium
Depth (cm)0 10 20 30 40 50
Cr m
g kg
-1
02000400060008000
100001200014000
Totalgt2mmSandSiltClay
Soil Nickel
Depth (cm)0 10 20 30 40 50
Ni m
g kg
-1
0
1000
2000
3000
4000
5000
Soil Manganese
Depth (cm)0 10 20 30 40 50
Mn
mg
kg-1
0
1000
2000
3000Totalgt2mmSandSiltClay
Soil Cobalt
Depth (cm)0 10 20 30 40 50
Co
mg
kg-1
0
50
100
150
200
250
300
Totalgt2mmSandSiltClay
Totalgt2mmSandSiltClay
Exceeds detection limits of 10000 mg kg-1 Cr
Figure 28
55
Cr (Wt )0 10 20 30 40 50
Al (W
t
)
0
5
10
15
20
25
30
Cr (Wt )0 10 20 30 40 50
Mg
(Wt
)
0
2
4
6
8
10
12
Cr (Wt )0 10 20 30 40 50
Si (W
t
)
0
2
4
6
8
10
12
14Cr (Wt )
0 10 20 30 40 50
Fe (W
t
)
0
10
20
30
40
50
60
70
Cr (Wt )0 10 20 30 40 50
Al +
Fe
(Wt
)
10
20
30
40
50
60
70
JRBPChromite
FeCr2O4
Chromite
Magnetite
Chromite-Silicate
Figure 29
56
Clay XRD of JR3515 Depth 5-15cm
0
200
400
600
800
1000
1200
0 5 10 15 20 25 30 35 40 45
d-spacing (Aring)
Rel
ativ
e In
tens
ity
Mg-Saturated
Mg+Glycerol
K-saturated
K-saturated+heat (500oC)
Figure 210
57
Energy (kev)
597 598 599 600 601 602 603
Nor
mal
ized
Abs
orba
nce
-4
-3
-2
-1
0
1
2
(a)
(b)
(c)
(d)
(e)
Cr(
VI)
Figure 211
58
Figure 212
59
Diablo Clay
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Serpentine Soil
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Chromite
0
20
40
60
80
100
120
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide
SilicateSum
Figure 213
60
Figure 214
61
Reaction Progress ()0 20 40 60 80 100
Elem
enta
l Wt
0
10
20
30
40
50
ChromiumAluminumIronMagnesium
FeCr2O4Jasper Ridge
Bedrock Chromite
SoilChromite
Figure 215
62
CHAPTER 3
OXIDATIVE PROMOTED DISSOLUTION OF CHROMITE BY MANGANESE
DIOXIDE AND CONCURRENT PRODUCTION OF CHROMATE
ABSTRACT
Chromium release and oxidation from chromite is a potential environmental hazard in
sediments and soils and a pathway for soil development (weathering of primary minerals)
related to ultramafic rocks and their metamorphic derivatives (serpentinites) Birnessite is
a common pedologic mineral in these sediments and soils capable of oxidizing aqueous
Cr(III) In this study the interaction between chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4)
and birnessite is investigated with an interest in the potential generation of Cr(VI)
Specifically the effects of chromite suspension density birnessite suspension density
and pH at a temperature of ~25degC in relation to the reaction kinetics were examined The
rate of Cr(VI) in solution increases with increasing chromite suspension densities and
decreasing pH but is independent of birnessite suspension densities at values greater than
~20 m2 L-1 The overall rate expression of Cr released and oxidized in solution from
chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression with Cr(VI) rates ranging from 9times10-4 to 44times10-3 microM h-1
Additionally these experiments demonstrate that serpentine soils are a source of non-
anthropogenic Cr(VI)
63
INTRODUCTION
Chromite a Cr-enriched spinel is responsible for a majority of the Cr enrichment
identified in ultramafic rocks serpentinites and serpentine soils (Rabenhorst et al 1982
Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989 Gasser and
Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003 Oze et al
2003a Oze et al 2003b) The chemical weathering of chromite is therefore an important
process when evaluating the potential of ultramafic rocks and their related sediments and
soils as sources of Cr Despite the propensity of chromite to resist chemical weathering
the dissolution of chromite and release of Cr will occur over geologic time (Bialowolska
and Salacinski 1984 Oze et al 2003b) Chromium in chromite is in the form of Cr(III)
the non-toxic and chemically immobile valence state however oxidizing conditions in
near surface environments is a potential reaction leading to the enhanced dissolution of
chromite and the formation of Cr(VI) the toxic and chemically mobile valence state
which typically exists as the oxyanion chromate (HxCrO42-x) (Daugherty 1992 Cohen et
al 1993 Fendorf 1995 Deer et al 1996) Consequently chromite may be a non-
anthropogenic source of Cr(VI) in the environment
Naturally occurring high valence Mn oxides present in soils and sediments
derived from ultramafic rocks are strong oxidizers capable of affecting the chemical
mobility of metals through redox reactions (Fendorf and Zasoski 1991 Johnson and
Xyla 1991) These insoluble mixed valent oxides (Mn(IIIIV)) are observed coating
mineral surfaces in these ultramafic sediments and soils where total Mn concentrations
are often gt100 mg kg-1 (Brooks 1987 Schreier et al 1987 Oze et al 2003a Oze et al
2003b Oze et al) Birnessite is a naturally occurring mixed valence oxide known to
oxidize aqueous Cr(III) to the more chemically mobile Cr(VI) (Eary and Rai 1987
Fendorf and Zasoski 1991) Gough et al (1989) and Becquer et al (2003) have
suggested that Mn oxides are responsible for Cr(VI) identified in California and New
Caledonia serpentine soil solutions however the dissolution and oxidation of Cr from a
Cr-bearing mineral due to the presence of birnessite has not been quantitatively defined
In this study we examine the kinetics of Cr release and oxidation when chromite and
birnessite are present in solution
64
The objective of this study is to determine a rate expression describing the
release and oxidation of Cr from chromite in the presence of birnessite Using well-
controlled batch experiments the solution chemistry and geochemistry of Cr related to
chromite and birnessite interaction was evaluated temporally over a range of pH
Conditions were chosen to simulate the slightly acidic conditions characteristic of most
serpentine soils (Rabenhorst et al 1982 Gough et al 1989 Graham et al 1990 Cole
1992 Jeffrey 1992 Lee 1992 Verger 1992 Becquer et al 2003) Ultimately these
experiments allow the examination of one potential pathway leading to Cr(VI) identified
in solutions related to ultramafic rocks and serpentine soils
MATERIALS AND METHODS
Solid Phase Characterization and Synthesis
Chromite used in the experiments was commercial-grade chromite (MSDS No
302-1 WHMIS CLASS D2A) obtained from Barnes Environmental Inc Ontario
Chromite was prepared by cleaning the sand-size material in multiple ultrasonic
isopropanol baths and rinses The chromite was then sieved using a Retsch magnetic
sieve shaker for 30 minutes in order to obtain the 160-250 microm size fraction Impurities in
the 160-250 microm chromite were removed using a slope Franz magnetic separator These
impurities included the minerals olivine augite plagioclase feldspar and quartz Visual
and scanning electron microprobe (SEM) examinations of the remaining 160-250 microm
fraction confirmed its purity Chromite was crushed to a fine powder using an agate
mortar and pestle and then rinsed multiple times in deionized destilled (DDI) water and
001 M HCl solution to remove the ultra-fine material and eliminate highly reactive sites
caused during grinding The chromite powder was inspected before (Figure 31) and
after the experiments using a scanning electron microscope (JEOL 5600LV) the surface
area was 025 m2 g-1 as determined using N2(g) and BET isotherm
An automated JEOL 733A electron microprobe operated at 15 kV accelerating
potential and 15 nA beam current was used to investigate possible zoning and to quantify
the chemical composition of the chromite grains prior to powdering No zoning was
observed and the chemical composition of the chromite was based on the average of 25
electron microprobe analyses The stoichiometric composition of chromite based on 4
65
oxygens is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 X-ray diffraction was also used to
deduce the purity of the chromite powder X-ray diffraction spectra obtained for the
chromite powder matched JCPDS card 03-0873 (unit cell a=8300 b=8300 c=8300
α=90 β=90 and γ=90) and no contaminants were evident
Birnessite was synthesized using the procedure described by Buser et al (1954)
and implemented by Fendorf and Zasoski (1992) A solution of 63 g of potassium
permanganate (KMnO4) in 1 L of DDI water was heated to 90 degC Approximately 100
mL of the heated slurry was poured into a 4 L beaker and reheated to 90degC while being
stirred vigorously The remaining 900 mL of slurry and 66 mL of concentrated
hydrochloric acid were simultaneously and proportionally dispensed into the beaker over
a period of thirty seconds The temperature of the resulting solution was maintained at
90deg C for an additional 10 minutes and then allowed to cool to room temperature The
product was washed with alternate rinses of DDI water and a 001 M HCl solution The
birnessite surface area was 125 m2 g-1 based on BET analyses Solids were inspected
before (Figure 31) and after the experiments using the SEM
Extraction Experiment Procedures
Several chemical extraction experiments were performed on 106-250 microm
chromite grains prior to the chromite-birnessite batch experiments in order to determine
the potential release of Cr Fe and Mn One gram of chromite and 25 mL of solution
were placed into a 25 mL centrifuge tube and the solids and solution were gently agitated
The solutions and parameters used for these extraction experiments include the
following 1) 10 hydrogen peroxide (H2O2) solution at 25degC and pH of ~5 for 6 h 2)
01 M potassium permanganate (KMnO4) solution at 60degC and pH of ~3 for 6 h 3) 10 M
hydrofluoric (HF) acid at 60degC for 6 h 4) 1 M hydroxylamine (NH2OHmiddotHCl) with 25
(vv) acetic acid (HOAc) at 60degC and pH of ~4 for 6 h and 5) 1 mM barium chloride
(BaCl2) solution at 60degC and pH of ~6 for 6 h Solutions were extracted with a syringe
and passed through a 02 microm filter Solids were inspected after the experiments using the
SEM
66
Batch Experiment Procedures
Batch experiments were performed in triplicate using 1 L polypropylene bottles
and chromite and birnessite ratios listed in Table 31 at ~25degC After loading chromite
and birnessite into each bottle 1 L of a pH 3 solution a 10 mM acetate solution (pH 5) a
10 mM MES solution (pH 67) or a pH 8 solution was added to the appropriate
experiment (Table 31) Chromite and birnessite were also run individually in the acetate
buffered solution (Table 31) Aliquots were extracted from the polypropylene bottles
using a syringe and then passed through a 2 microm filter before analyses Less than 40 mL
was extracted over the course of each experiment resulting in a total volume change of
lt4 After the experiments were terminated the remaining solids were resuspended in
20 mL of 1 mM phosphate (5050 KH2PO4K2HPO4 pH 7) 20 mL of 10 mM CaCl2
solution at pH 7 and a 05 M HCl solution These suspensions were agitated for 3 h
followed by centrifugation and filtration (02 microm) before analyses The purpose of the
phosphate and CaCl2 solutions was to desorb reaction produtcts (eg Cr(VI) Cr(III) Fe
and Mn) from the solid phases The HCl treatment provides a means to evaluate
secondary phases such (oxy)hydroxides formed during the experiments Although the
HCl solution may dissolve chromite and birnessite in addition to the (oxy)hydroxides the
short time frame of the extraction (3 h) should limit the dissolution of chromite
Additionally the polypropylene bottles following the termination of five batch
experiments were rinsed with DDI water and 20 mL of 05 M HCl was added to each
bottle The HCl solution was allowed to remain in the bottles for 15 h in order to
determine the concentration of Cr Fe and Mn sorbed onto the interior lining
Serpentine Soil and Birnessite Experiments
Serpentine soils collected from Jasper Ridge Biological Preserve (Stanford CA)
and New Caledonia were used to determine the influence of birnessite in soils known to
contain chromite (Chapters 2 and 4) The Cr geochemistry of these soils as well as the
detrital chromite have been evaluated in Chapters 2 and 4 and by Becquer et al (2003)
Jasper Ridge serpentine soils contain Cr-bearing minerals including chromite Cr-
magnetite and Cr-chlorite however a majority of the Cr enrichment (Table 32) is
directly related to the abundance of chromite A majority of the Cr related to New
67
Caledonia serpentine soils is attributed to Cr (oxy)hydroxides (Chapter 4) however
chromite is present in these soils as well (Becquer et al 2003) Although birnessite has
been noted in these soils the quantity of birnessite could not be determined In order to
evaluate the impact of Mn oxides on Cr within such soils 01 g of birnessite was added to
each soil
The parameters for the experiments are listed in Table 32 Ten grams of the
lt250 microm size-fraction of each soil (JR1 JR2 and NC) and 100 mL of 1 mM acetate
buffered solution (pH 5) were loaded into individual 150 mL Nalgene bottles Aqueous
solutions were evaluated as a function of time (h) Additionally 10 g of each soil 01 g
of birnessite and 100 mL of 1 mM acetate buffered solution (pH 5) were loaded into 150
mL Nalgene bottles and were allowed to react as a function of time (h) These
experiments were rotated at a rate of 2 revolutions min-1 Solutions were removed with a
syringe and passed through a 2 microm filter before analysis Approximately 8 mL of
solution was required for the analyses over the course of each experiment resulting in a
total volume change of 8 Total Cr concentrations (mg kg-1) of the soils listed in Table
32 were obtained by completely dissolving the samples using a mixture of hot (gt200degC)
concentrated nitric perchloric and hydrofluoric acids Elemental compositions were
measured in the supernatant using inductively coupled plasma optical emission
spectroscopy (ICP-OES) The mass of chromite in each soil was calculated using total Cr
concentration for each soil and the calculations provided in Chapters 2 (Table 210) and
Chapter 4 (Table 47) The chromite surface area of each soil was approximated by
comparing the surface area measured for the chromite standard and its observed grain
size from SEM images (Figure 31) to chromites observed in the soils using backscattered
electron images (Chapters 2 and 4) The surface area of the Jasper Ridge chromite is 066
of the chromite standard due to a slightly larger grain size The surface area of the
New Caledonia chromite is approximately 025 of the chromite standard Chromite
suspension densities were calculated by multiplying the chromite mass by the estimated
surface area and then dividing by the volume of the solution
68
Solution Analysis
The pH and Eh of the supernatant was measured with an Orion flat surface
combination pH electrode inserted directly into supernatant Production of soluble
Cr(VI) was measured spectrophotometrically at 540 nm using the s-diphenyl carbazide
method modified from Bartlett and James (1979) Soluble Fe(II) was monitored using
the ferrozine method adopted from Stookey (1970) Total dissolved Cr Fe Al Mn and
Mg were ascertained using ICP-OES Concentrated trace metal grade nitric acid used to
acidify the samples prior to analyses Aluminum and Mg was determined to be present in
the nitric acid at concentrations lt1 microM Magnesium and Al are not plotted in any of the
figures Chromium(III) and Fe(III) was determined by subtracting Cr(VI) and Fe(II)
obtained from colorimetric methods from total dissolved Cr and Fe analyses using ICP-
AES Results are presented as an average of the three runs for each batch experiment
with the standard deviation noted by error bars
RESULTS
Chromite Extraction Experiments
The release of Cr Fe and Mn from chromite using 10 H2O2 01 M KMnO4 10
M HF acid 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 1 mM BaCl2 solutions was
carried out to determine the relative propensity of chromite to resist chemical dissolution
and these results are summarized in Figure 32 The KMnO4 solution with aqueous
Mn(VII) was the most effective in releasing Cr from chromite yielding a concentration
of 62 Cr mg L-1 Iron released using the KMnO4 solution was 03 mg L-1 Although
H2O2 is a direct chemical oxidant the release of Cr from chromite was less as compared
to the other solutions (only ~03 Cr mg L-1 was measured) Concentrated HF was only
able to release ~13 Cr mg L-1 while releasing 106 Fe mg L-1 NH2OHmiddotHCl with 25
(vv) HOAc and CaCl2 solutions only removed lt1 Cr mg L-1 however the NH2OHmiddotHCl
removed 36 Fe mg L-1 which is comparatively higher than the H2O2 KMnO4 and BaCl2
solutions
69
Chromite and Birnessite Batch Experiments
The interaction of chromite and birnessite leading to the release and oxidation of
Cr(III) is a relatively slow process requiring study on the order of hours and days The
pH was maintained constant throughout the duration of the experiments while Eh values
ranged from 320 to 450 mV with no discernable temporal pattern (Figure 33) For pH
values between 3 and 67 Cr3+ CrOH2+ and Cr(OH)2+ are the dominant aqueous Cr(III)
species however at pH 8 conditions are on the stability line between Cr(III) and Cr(VI)
specifically Cr(OH)3 and CrO42-
Higher chromite suspension densities produced higher rates of Cr release to the
aqueous phase and all the Cr in solution was in the form of Cr(VI) (Figure 35) The
rates of oxidation based on Cr(VI) production range from 00005 to 00041 microM h-1
(Figure 35 Table 33) Comparatively Cr released from the chromite control experiment
is maintained at a relatively constant value of ~005 microM and no Cr(VI) is measured in
solution Iron concentrations initially increase to ~ 05 microM and then decrease rapidly to a
relatively constant concentration at 02 microM (Figure 35) Iron concentrations for the
chromite control experiment began at 15 microM between 0 to 50 h and decreased to ~02
microM Iron(II) and Mn concentrations are below detectable limits for all experiments
(Figure 35) A zeroth-order dependence is assumed with respect to birnessite owing to
its large excess surface area relative to chromite
Variable birnessite suspension densities and constant chromite suspension
densities were evaluated at pH 5 (Figure 36) Although suspension densities of
birnessite are 538 and 1075 m2 L-1 (Table 31) the rate of Cr and Cr(VI) measured in
solution for both experiments is 00041 microM h-1 (Figure 36 and Table 33) Iron
concentrations are below 06 microM and maintain concentrations near 02 microM with no
indication of consistently increasing or decreasing with respect to time Iron(II) and Mn
concentrations in solution are below detectable limits for both experiments
In Figure 37 the influence of pH on chromite oxidation by birnessite was
evaluated by conducting experiments at pH values of 3 5 67 and 8 chromite and
birnessite suspension densities are constant at 0075 m2 L-1 and 1075 m2 L-1 respectively
(Table 31) Chromium and Cr(VI) rates increase with decreasing pH ranging from 0
microM h-1 at pH 8 to 00062 microM h-1 at pH 3 (Table 33) Iron is present in solution at
70
concentrations less than 04 microM for all the experiments Generally Fe concentrations are
greater near the start of the experiment and decrease to values lt 05 microM with increasing
reaction time Manganese concentrations are below detectable limits
In order to determine the rate dependence on each prominent reactant (chromite
birnessite and pH) reaction rates were regressed against reactant concentrations (Figure
38) A pseudo first-order rate dependence is noted based on chromite suspension
density the equation regressed for this relationship is listed in Table 33 As the chromite
suspension density increases the rate of Cr(VI) in solution increases In contrast a
zeroth-order rate dependence is evident for birnessite suspension density (dashed line in
Figure 38b) The rate of Cr(VI) generation is expected to decrease with respect to
birnessite suspension density at very low concentrations (dotted line in Figure 38b)
The influence of pH on the rate of Cr(VI) production is well-described by a polynomial
function (Figure 38c Table 33)
Developing an overall rate expression for Cr in solution requires integrating the
factors assessed in Figure 38 and listed in Table 33 All the Cr measured in solution is
in form of Cr(VI) therefore the rate law for Cr or Cr(VI) in solution may be written as
d[Cr(VI))]dt = k sdotChromitea sdot Birnessiteb sdot (aH+)c (31)
and
log d[Cr(VI))]dt = log k + asdotlogChromite+ bsdot log Birnessite+ csdotlog (aH+) (32)
where k is the overall rate constant for Cr release and oxidation Chromite is the
chromite suspension density (Table 31) Birnessite is the birnessite suspension density
(Table 31) and aH+ is the activity of H+ Constants a b and c are the order of
dependency for each reactant The overall rate equation (developed in the Appendix) can
be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069 (33)
71
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 the expression is valid for birnessite suspension densities
greater than 20 m2 L-1 and pH values between 3 to 8
Secondary and Solid Phase Examination of Chromite and Birnessite
The solid products were evaluated to determine the abundance and type of
secondary phases formed during the course of the experiments Solids from the batch
experiment were treated with three solutions 1 mM 5050 KH2PO4K2HPO4 1 mM
CaCl2 and 05 M HCl (Table 34) In the chromite and birnessite reactions Cr was
primarily in solution (gt90 ) The HCl extract was able to remove lt5 additional Cr
while the KH2PO4K2HPO4 and CaCl2 extracts released lt3 additional Cr Aqueous Fe
concentrations were below detectable limits for the final solution extraction Treatment
with the HCl resulted in up to 0127 micromoles of Fe however 0107 micromoles of Fe were
released upon treatment of unreacted chromite with HCl Although not shown in Table
34 Mn values were greater than gt4 micromoles when treated with HCl the
KH2PO4K2HPO4 and CaCl2 removed no detectable Mn
After the termination of each experiment the solids were examined using SEM in
order to identify changes in the surface morphology and secondary phases A large
chromite grain and conglomerate of birnessite adjacent to the chromite are depicted in
Figure 38 The face and edge of the chromite grain is not noticeably altered compared to
grains examined prior the experiments Additionally the particle morphology of
birnessite has not visibly changed Even at higher magnification angular and
conchoidally fractured fine-grained edges are noted (Figure 38bc) Only a slight
rounding of the smaller features of the main chromite grain provides any evidence of
possible chromite dissolution (Figure 38c) Birnessite appears to have limited interaction
with chromite Although chromite and birnessite were allowed to interact over 500 h in
an aqueous solution no secondary phases are observed by SEM
Serpentine Soil and Birnessite Experiments
Two serpentine soils from Jasper Ridge Biological Preserve CA and one
serpentine soil from New Caledonia were used to determine whether or not the overall
72
rate expression (Equation 33) for chromite reaction with birnessite is consistent with
Cr(VI) concentrations produced by serpentine soils containing chromite Three
serpentine soils without birnessite and three soils with the addition of birnessite buffered
at pH 5 were measured for Cr(VI) generation as a function time (Figure 310) The rates
of Cr(VI) measured in solution for both Jasper Ridge soils (JR1 and JR2) are identical at
00013 microM h-1 With the addition of birnessite the rates of Cr(VI) production in solution
both increased to 00044 microM h-1 (JR1 w Bt and JR2 wBt in Figure 310) Based on the
chromite suspension density (Table 32) the rates predicted by Equation 33 range from
00035 to 00041 microM h-1 which are 10 to 20 less than those observed in the
experiments (Figure 310) The rates of Cr(VI) in solution from the New Caledonia
serpentine soil are 00009 microM h-1 and with the addition of birnessite increase up to 00013
microM h-1 (Figure 310) A rate of 00016 microM h-1 was predicted by Equation 33 for the
New Caledonia soil
DISCUSSION
The reaction of birnessite with chromite enhances the rate of Cr release into
solution The rate of Cr(VI) measured in solution (Equation 33) is directly related to the
chromite suspension density and is independent of the birnessite at suspension densities
greater than ~20 m2 L-1 Although the release and oxidation of Cr in solution are
measured the mechanisms and pathways driving this reaction need to be evaluated The
propensity of chromite to resist dissolution the effects of pH the formation of secondary
phases and ultimately the pathways of Cr release and oxidation are addressed in the
following discussion
Extraction experiments performed on chromite (Figure 32) demonstrate it is
resistant to acidic dissolution or competitive ion displacement but strong oxidants such
as high valent Mn may oxidatively dissolve chromite It is evident that the proportions of
Cr Fe and Mn vary for each of the extractions The solutions utilized in the extraction
experiments have varying chemical properties capable of preferentially releasing a
particular species andor element These experiments were also performed over a short
period of time (lt6 h) which may have only released elements from highly reactive sites
73
Finally secondary phases associated with these experiments were possible and would
distort the elemental ratios observed in solution
A majority of the Eh and pH analyses plot in the Cr(III) stability region (Figure
33) however Cr in solution is primarily in the form of Cr(VI) (Figures 35 36 and
37) Eh measurements presented in Figures 32 and 33 do not account for the
microenvironments associated with the surfaces of birnessite Additionally the kinetics
of Cr(VI) reduction are limited in reducing conditions due to electron symmetry
constraints and the absence of a suitable redox couple (Fendorf 1995) Finally highly
oxidizing birnessite in the solution is kinetically and thermodynamically controlling the
oxidation state of Cr relative to the stabilities predicted in Figure 33 (Amacher and
Baker 1982 Fendorf and Zasoski 1992)
A fractional-order dependence on the chromite suspension density (Figure 38)
can be explained by diffusional influences competing oxidation reactions and secondary
reactions involving Cr(III) and Mn(IV) Secondary phases on the surface of chromite
may inhibit the diffusion of Cr from chromite however the clean and relatively unaltered
chromite surfaces do not support this mechanism as a major cause (Figure 38)
Chromium and Fe released from chromite may be incorporated into secondary phases
related to birnessite or chromite Approximately 10 of Cr is not in the aqueous phase
(Table 34) supporting the premise that secondary phases are contributing to the
fractional-order dependence related to the chromite suspension density Also
competition from Fe(II) oxidation on birnessite deposition of Fe(II) on birnessite (or
chromite) and diffusion onwithin birnessite may further impact the rate dependence
The zeroth-order dependence based on the birnessite suspension density (Figure
38) results from its high surface area At very low birnessite suspension densities the
rate of Cr(VI) production should decrease however due to the high surface area of
birnessite it was experimentally prohibitive to determine reaction rates at masses less
than 4 mg Therefore a hypothetical trend-line depicts the expected dependence on
birnessite concentration at very low suspension densities at higher values representative
of most field settings the reaction is independent of the birnessite surface area (ie zero-
order dependence)
74
In Figure 38c the polynomial relationship defines the Cr(VI) rate dependence on
pH As the pH decreases the rate of Cr(VI) increases more rapidly The relationship is
potentially the result of lower pHs progressively increasing the rate of chromite
dissolution However the rate of Cr(VI) in solution might be influenced by higher pHs
increasing the formation of secondary Fe(III) and Cr(III) (oxy)hydroxides on the
chromite andor birnessite surfaces thereby decreasing elemental diffusion from
chromite or decreasing available birnessite sites for Cr(III) oxidation Differentiating
between these two mechanisms is not possible based on the rate equation or extraction
analyses however the polynomial relationship evident in Figure 38c is folded into the
rate constant in order to geochemically account for the influence of pH on Cr(VI)
generation
No conclusive evidence is available to determine the fate of Mn within the
reacting suspensions Aqueous Mn concentrations are consistently maintained below
detectable limits If Mn is released from birnessite then it must be strongly and quickly
incorporated into a nonaqueous phase following its release Due to Cr(III) being oxidized
by Mn(III) andor Mn(IV) Mn(II) must be produced Extractions performed on the
chromite and birnessite solids demonstrate that Mn could only be removed with HCl
however a majority of this Mn is potentially the result of HCl dissolving birnessite On
the basis of electron transfer stoichiometry at least 15 mols of Mn(II) must be produced
for each mol of Cr(VI) generated This amount can be accounted for in the HCl
extraction where gt4 micromoles of Mn were measured Additionally Fendorf et al (1993)
determined that Mn(II) should not significantly influence the extent of Cr(III) oxidation
Iron in solution does not progressively increase in solution over time (Figures 35
36 and 37) The high initial rate of Fe release compared to Cr may be the result highly
reactive sites preferentially releasing Fe Although no Fe is detected in solution from
the extractions performed on the chromite and birnessite solids 0005 to 0035 moles of
Fe were retrieved from the lining of the bottle and 0053 to 0126 moles were released
from the solids using the HCl solution (Table 34) Iron(II) could potentially be oxidized
to Fe(II) in the presence of birnessite and form stable Fe(III)-(oxy)hydroxides (Postma
1985 Myers and Nealson 1988 Villinski et al 2001) A ratio of two Fe for every one
Mn is necessary for the oxidation of Fe(II) and this amount is recoverable based on HCl
75
extractions with gt4 micromoles of Mn measured from the HCl extractions Additionally
dissolved O2 in the batch experiments will rapidly oxidize Fe(II) for all the pHs examined
in these experiments (Stumm and Morgan 1981 Langmuir 1997) The oxidative
pathway for Fe(II) provides an explanation for the lack of Fe(II) in solution for the batch
experiments (Figure 35 36 and 37)
With a detection limit of ~02 microM and equilibrium with respect to Fe(OH)3(am) and
goethite Fe concentrations should either be at or below detection limits at pHs gt 5
Assuming aqueous Fe(III) is present dominantly as the second hydrolysis product at pH
5 the dissolved concentrations approach the solubility of ferrihydrite and are much
greater than expected for goethite (Figure 311) Thus it is probably that the former solid
is generated which is further supported by the chemical extraction experiments (Table
34)
Iron and Cr solution concentrations measured experimentally are not consistent
with congruent dissolution (Figure 312a) Assuming Fe was forming secondary phases
at a constant rate (~00001 Fe h-1) multiplying this rate by the time analysis and then
adding it to the measured aqueous Fe a corrected Fe (Fe) value can be estimated (Figure
312) Additionally the 10 of Cr not in solution can be also be used to calculate a
corrected Cr value (Cr) (Figure 312) When secondary phases are taken into account a
majority of the analyses plot near the congruent dissolution of chromite This provides
evidence that chromite is undergoing congruent dissolution within the batch experiments
but that secondary Fe(III) phases disrupt the solution phase elemental ratios
Chromium(VI) Pathways from Chromite and Birnessite into Solution
Three pathways are evaluated for Cr(VI) measured in solution a) chromite
dissolution b) birnessite dissolution and c) solid-solution reaction (Figure 313) In order
to describe these pathways end-member chromite (FeCr2O4) and birnessite (δ-MnO2) are
used to simplify the analysis It should be recalled however that the composition of the
chromite used in the experiments is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4
Chromite dissolution presents a likely explanation for Cr(VI) observed in solution
based on the following discussion In chromite dissolution model Cr(III) released from
chromite adsorbs onto the surface of birnessite and oxidizes to Cr(VI) After the electron
76
transfer is complete Cr(VI) desorbs from birnessite and remains in solution As this
process continues Cr(VI) increases at a constant rate The rate limiting step with regards
to this specific pathway can be grouped into two steps a) dissolution of chromite (Step
1a) or b) Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) (Figure 313)
Although a dissolution rate for chromite at ambient temperatures and pressures is not
known the chromite extraction experiments (Figure 32) demonstrates that chromite
dissolution is slow Manganese oxides such as birnessite oxidize Cr(III) in solution
rapidly with a measured rate constant of 6300 microM-1 s-1 in fresh water (100 mg L-1 MnO2
100 g L-1 Cr(III) pH~6) (Van der Weijden and Reith 1982) This means that the rate
for Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) is on the order of
seconds therefore chromite dissolution is likely the rate limiting step controlling the
concentration of Cr(VI) in solution for this pathway
Although a zeroth-order dependence related to birnessite concentrations (Figure
37a) is observed the birnessite dissolution pathway in which aqueous Mn4+ released
from birnessite adsorbs onto the surface of chromite oxidizing Cr(III) cannot be entirely
dismissed Manganese dissolution and adsorption could occur quickly not allowing the
Mn to be measured in solution Examination of the solids after the termination of each
experiment demonstrated extremely minimal chromite and birnessite interaction
therefore the solid solution reaction is not a likely pathway leading to the production of
Cr(VI) in solution and is furthermore inconsistent with the noted rate dependencies
Despite the three pathways (Figure 313) the generation of Cr(VI) observed in these
experiments and described by the rate equation (Equation 33) is independent of these
potential pathways
Serpentine Soils and Cr(VI)
The serpentine soil experiments (Figure 310) provide direct evidence that Cr(VI)
is forming within these soils By adding birnessite to the soil the rate of Cr(VI) in
solution increased for all experiments (00013 to 00044 microM Cr(VI) h-1 for Jasper Ridge
and 00009 to 00013 microM Cr(VI) h-1 for New Caledonia) The Cr(VI) rate increase
related to the addition of birnessite is likely the result of low birnessite suspension
densities (lt20 m2 L-1) in the soil Rates predicted for Jasper Ridge and New Caledonia
77
soils using Equation 33 and values in Table 32 are similar to those measured in the
experiments with values of 00041 and 00016 microM Cr(VI) h-1 respectively Overall these
soil experiments demonstrate that a rate equation developed from laboratory batch
experiments predicts the rate of Cr(VI) in serpentine soils reasonably well
According to limits set by the Environmental Protection Agency the California
standard for total Cr in drinking water is 50 microg L-1 (~1 microM) while the federal standard is
100 microg L-1 (~2 microM) The serpentine soil and birnessite experiments produced Cr(VI)
concentrations in excess of 50 microg L-1 after only 140 h or roughly 6 days Serpentine soils
at Jasper Ridge typically remain water saturated for prolonged periods of time (gt1 week)
during the winter season increasing the time Cr(VI) can be produced The degree and
duration of water saturation is dependent on the topography of the serpentinite body
level areas will remain water saturated while sloping areas will allow the water to flow
down gradient Although Cr(VI) was generated in the soil experiments the production of
Cr(VI) in serpentine soils may be more limited The reduction of Cr(VI) to Cr(III) is
favorable by reactions with Fe(II)aq Fe(II)-containing minerals such as magnetite H2S
and organic matter including microbial processes (Ilton and Veblen 1994 Fendorf 1995
Peterson 1996) Both organic matter and magnetite are abundant in Jasper Ridge
serpentine soils (Chapters 2 and 4) however these factors did not deter Cr(VI) from
forming during the serpentine soil experiments Additionally the soils in these
experiments were constantly rotated allowing the mineral surfaces to be continuously
exposed to solution Chromite within an undisturbed soil will not always be in contact
with the solution or proximal to birnessite
Chromium(VI) produced by the interaction of chromite and birnessite is not
limited to serpentine soils Serpentinites the bedrock for serpentine soils are typically
highly sheared due to the diapiric migration of serpentinite masses along zones of
weakness in fault zones Numerous cracks and lineations which are typical of these rocks
provide a means for Cr(VI) to transported away from the serpentine soil potentially into
local groundwater In fact preferential flow will occur along the major shear plains of
the serpentinite Additionally these shear plains are often coated with high valent Mn
oxides providing a reaction pathway for Cr(VI) formation Serpentinite springs such as
those evaluated by Barnes and ONeil (1969) and Barnes et al (1972) at the Cedars
78
ultramafic body in Cazadero CA have Cr concentrations measured at 20 microg L-1 in which
all the Cr is in the form of Cr(VI) (Chapter 4) In conclusion chromite and birnessite in
serpentine soils could contribute a source of Cr(VI) in surface waters and groundwaters
Chromium release and oxidation from chromite is increased by the presence of
birnessite and the interaction between chromite and birnessite is a pathway for Cr(VI)
generation in ultramafic sediments and soils The potential of serpentine soils as a source
of non-anthropogenic Cr contamination can now be evaluated quantitatively based on the
abundance of chromite The overall rate law for Cr(VI) in solution (Equation 39)
provides a means to estimate the Cr(VI) concentration attainable for serpentine soils and
serpentine sediments when the chromite and birnessite suspension densities and the pH of
the soil are known
ACKNOWLEDGEMENTS
We gratefully acknowledge comments and assistance from Robert Jones and Dr
Guangchao Li Funding for this research was provided by the Office of Teaching and
Learning at Stanford University
79
Table 31 Chromite and birnessite batch experiment parameters with chromite and birnessite suspension densities
Experiment Chromitea (grams)
Birnessiteb (grams) pHc
Chromite Surface Aread
(m2 g-1)
Birnessite Surface Aread
(m2 g-1)
Chromite Suspension
Density (m2 L-1)
Birnessite Suspension
Density (m2 L-1)
A 08 043 5 025 125 02 54 B
08 086 5 025 125 02 108C 03 086 5 025 125 008 108D 008 086 5 025 125 002 108E 004 086 5 025 125 001 108F 03 086 3 025 125 008 108G 03 086 67 025 125 008 108H 03 086 8 025 125 008 108
Chromite 08 ---- 5 025 125 02 ----Birnessite ---- 043 5 025 125 ---- 5480
a Chromite chemical composition is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 b Synthetic birnessite (δ-MnO2) c10 mM acetate buffered solution pH = 5 non-buffered pH=3 5 mM MES buffer pH 67 non-buffered pH 8 dMeasured BET analyses
81
Table 32 Serpentine soil and birnessite experiments with soil and birnessite masses used and the estimated chromite suspension density
Samples Soil Mass (g)
Birnessite Mass (g)
Volume (L) of Solutiona
Total Crsoil
b (mg kg-1)
Chromite Mass (g) in
Soil Samplec Estimated Surface
Areac (m2 g-1)
Chromite Suspension Density
(m2 L-1) in Soild Jasper Ridge
(JR1) 10 ---- 01 5976 013 015 02 Jasper Ridge (JR1 w Bt) 10 01 01 5976 013 015 02 Jasper Ridge
(JR2) 10 ---- 01 4978 011 015 016 Jasper Ridge (JR2 w Bt) 10 01 01 4978 011 015 016
New Caledonia (NC) 10 ---- 01 827 0015 006 001
New Caledonia (NC w Bt) 10 01 01 827 0015 006 001
a1mM Acetate Buffer (pH=5) 81 bTotal Cr concentrations were obtained through total digestion and ICP-AES analyses
cEstimated chromite mass and surface area (discussed in the Methods section) was made using chromite observations and calculations in Chapters 2 and 4
dCalculated from the chromite mass in the soil sample estimated surface area of the chromite and the volume of solution
82
Table 33 Summary of Cr(VI) rates obtained from batch experiments (Figure 35 36 and 37) and used in Figure 38
Listed in Table 31
Suspension Densities (m2 L-1)
Chromite (Ch)
Birnessite (Bt)
pH Rate (microM h-1) (d[Cr(VI)]dt)
log Rate (log microM h-1) (log d[Cr(VI)]dt) Expression
Figure 35 02 108 5 00041 -239
008 108 5 00021 -268 002 108 5 00009 -305 001 108 5 00005 -330
1
Figure 36 02 54 5 00041 -239 02 108 5 00041 -239 2
Figure 37 008 108 3 00062 -221 008 108 5 00021 -268 008 108 67 00004 -340 008 108 8 0 ----
3
Expression 1 log d[Cr(VI)]dt = asdot log Chromite +log krsquorsquo log d[Cr(VI)]dt = 06908 log Chromite ndash 1899 Expression 2 log d[Cr(VI)]dt = bsdot log Birnessite + log krsquorsquorsquo
log d[Cr(VI)]dt = 0sdot log Birnessite ndash 23872 log d[Cr(VI)]dt = -23872
Expression 3 d[Cr(VI)]dt = 264E-4sdotpH2 ndash 414E-3sdotpH + 162E-2
82
Table 34 Extractions comparing total solution and extractable Cr and Fe from the solids
------------------Cr----------------- ------------------Fe---------------- microMolesdagger Total
mole Extraction
mole microMoles Total mole
Extraction mole
Chromite Ch 02 (Table 31) Terminated at 30 days Solutiona 0033 554 ---- 0 0 ---- Phosphateb 0000 05 1 0000 03 03 CaCl2
c 0013 388 87 0000 00 00 HCld 0002 53 12 0103 997 997 Bottle HCle 0000 00 0 0000 00 00
A Ch 02 Bt 54 (Table 31) Terminated at 30 days Solution 2977 910 ---- 0 0 ---- Phosphate 0048 16 18 0004 21 21 CaCl2 0055 18 20 0002 09 09 HCl 0161 54 60 0126 758 758 Bottle HCl 0004 01 2 0035 212 212
B Ch 02 Bt 108 (Table 31) Terminated at 30 days Solution 2837 914 ---- 0 0 ---- Phosphate 0039 14 16 0008 49 49 CaCl2 0060 21 25 0000 00 00 HCl 0142 50 58 0127 803 803 Bottle HCl 0004 01 2 0024 148 148
C Ch 08 Bt 108 (Table 31) Terminated at 17 days Solution 0877 894 ---- 0 0 ---- Phosphate 0020 23 22 0002 37 37 CaCl2 0030 34 32 0000 00 00 HCl 0040 46 43 0053 881 881 Bottle HCl 0003 03 3 0005 82 82
daggerMolarity values are normalized to the final solution analysis of each experiment based on volume of solution and mass of the sample in each experiment aSolution refers to the final solution analysis before the experiment was terminated bmicroMoles extracted from the batch experiment solids using a 1 mM phosphate buffer pH 7 3 h cmicroMoles extracted from the batch experiment solids using a 10 mM CaCl2 solution pH 7 3 h dmicroMoles extracted from the batch experiment solids using a 05 N HCl solution 3 h emicroMoles extracted from the batch experiment bottles using a 05 N HCl solution 15 h
83
Figure Captions
Figure 31 Scanning electron microscopic (SEM) images of chromite and birnessite at
varying magnifications prior to the batch experiments
Figure 32 A summary of Cr Fe and Mn released from chromite during extraction
experiments Solutions used for these experiments include 1) 10 H2O2 solution 2) 01
M KMnO4 solution 3) 10 M HF 4) 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 5) 1mM
BaCl2 solution () Fe concentration for the HF extraction is 106 mg L-1
Figure 33 pH and Eh measurements as a function of time (h) for the batch
experiments Chromite (Ch) and birnessite (Bt) suspension densities used for the batch
experiments (Table 31) are listed adjacent their corresponding symbol in the key
Figure 34 Eh and pH batch experiment analyses (Figure 33) plotted on a stability field
diagram for Cr (log a Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom
(1998)
Figure 35 Batch experiments evaluating variable chromite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr and Cr(VI) measured in solution are noted and illustrated with a dotted line and
included in Table 33
Figure 36 Batch experiments evaluating variable birnessite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr(VI) measured in solution are denoted and included in Table 33
Figure 37 Influence of pH on Cr(VI) generation by reaction of constant chromite (Ch
08 m2 L-1) and birnessite (Bt 108 m2 L-1) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key Rates of Cr(VI) measured in solution are included in Table
33
84
Figure 38 Rates of Cr(VI) in solution (Table 33) as a function of a) chromite
suspension density (Figure 35) b) birnessite suspension density (Figure 36) and c) pH
(Figure 37) The line in each diagram represents the best fit for these data points and the
equations for these lines are listed in Table 33 The dotted line deviating from the
dashed line in b) indicates the probable path at low birnessite concentrations
Figure 39 SEM images of the solid phases after the completion of the batch
experiments A (Ch 02 Bt 54) and B (Ch 02 Bt 108) are shown Chromite is the larger
grain with relatively flat surfaces whereas birnessite is the fine-grained rosettes located
adjacent the larger chromite grains
Figure 310 Chromium and Cr(VI) concentrations (microM) in solution as a function of
time (h) from serpentine soils obtained at Jasper Ridge (JR1 and JR2) and New Caledonia
(NC) Soils with the addition of 01 g birnessite are noted in the figure Soil and
birnessite masses used for each experiment are listed in Table 32
Figure 311 Iron analyses as a function of time (Figure 35 and 36) with the solubility
of amorphous Fe(OH)3 and goethite and the dominant Fe3+-OH complex at pH 5 Fe
solubilities are from Langmuir (1997) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key
Figure 312 a) Experimental values of Fe and Cr (Figures 35 and 36) are plotted A
dotted line denotes the ratio of Fe and Cr necessary for chromite congruent dissolution b)
Corrected FeCr ratios as a function of time (h) for data from Figures 35 and 36 taking
into account the concentration of Fe and Cr in secondary phases calculated from Table
34 c) Corrected Fe and Cr as compared to values predicted with congruent dissolution
of chromite
85
Figure 313 Three potential pathways of Cr release and oxidation for chromite-
birnessite interactions are listed
86
Figure 31
87
0
2
4
6
8
10
12
Hydrogen
Peroxid
e
K-Perman
ganate
Hydroflu
oric Acid
Hydroxy
lamine
Barium Chlorid
e
Con
cent
ratio
ns (m
g kg
-1)
Cr
Fe
Mn
Figure 32
88
Time (h)0 100 200 300 400 500
pH
2
3
4
5
6
7
8
9
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
Time (h)0 100 200 300 400 500
Eh (m
V)
100
200
300
400
500
600
Ch Bt ------(m2 L-1)------
Figure 33
89
pH2 3 4 5 6 7 8 9
Eh (m
V)
-05
00
05
10
15
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
HCrO4-
CrO42-
Cr3+
CrOH2+
Cr(OH)3Cr(
OH
) 2+
Upper limit of water stability
Lower limit of water stability
Ch Bt ------(m2 L-1)-----
Figure 34
90
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 108 008 108 002 108 001 108 02 --- --- 54
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
0
1
2
3
4
pH 5 pH 5
pH 5 pH 5 pH 5
Ch Bt ---(m2 L-1)---
Figure 35
91
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 54 02 108
00041 microM h
-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
00
02
04
06
08
10
00041 microM h
-1
pH 5
pH 5 pH 5 pH 5
pH 5 Ch Bt----(m2 L-1)----
Figure 36
92
Time (h)
0 50 100 150 200 250
Cr c
once
ntra
tion
( microM
)
00
02
04
06
08
10
pH 3pH 5pH 67pH 8
Time (h)0 50 100 150 200 250
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250M
n co
ncen
trat
ion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
00004 microM h-1
00021 microM h-1
0006
2 microM h
-1
00021 microM h-100
062 micro
M h-1
00004 microM h-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1
Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Figure 37
93
log Birnessite m2 L-1
05 10 15 20 25 30
log
d[C
r(VI
)]dt
microM
h-1
-40
-35
-30
-25
-20
pH 5
log Chromite m2 L-1-22 -20 -18 -16 -14 -12 -10 -08 -06
log
d[C
r(VI
)]dt
microM
h-1
-34
-32
-30
-28
-26
-24
-22
pH 5
pH1 2 3 4 5 6 7 8 9
d[C
r(VI
)]dt
microM
h-1
0000
0002
0004
0006
0008
0010
a)
c)
b)
Ch 008 m2 L-1 Bt 108 m2 L-1
Figure 38
94
Figure 39
95
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR1JR1 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR2JR2 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
01
02
03
04
05
NCNC w Bt
00044 microM h-1
00013 microM h-1
00044 microM h-1
00013 microM h-1
00013 microM h-1
00009microM h-1
pH 5
pH 5
pH 5
Figure 310
96
Time (h)0 100 200 300 400 500
log ΣF
e (aq)
microM
-10
-8
-6
-4
-2
0
2
4
02 54 02 108 008 108 002 108 001 108
Fe(OH)3(am)
Goethite
pH 5Ch Bt----m2 L-1----
Fe(OH)2(aq)+
Figure 311
97
Time (h)
0 100 200 300 400
FeC
r Rat
io
0
1
2
3
4
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
02 54 02 108 008 108 002 108 001 108
Chrom
ite C
ongr
uent
Dissolu
tion
Chrom
ite C
ongr
uent
Dissolu
tion
a)
b)
c)
Ch Bt----(m2 L-1)----
ChromiteCongruent Dissolution
Figure 312
98
Pathway 1 Chromite Dissolution
1a) Cr dissolution from chromite FeCr2O3(s) + 4H+ rarr 2Cr(OH)2+
aq + Fe2+(Oxyhydroxide) + H2O
1b) Adsorption of CrOH2+ on the surface of birnessite Cr(OH)2+
aq + MnO2(s) rarr Cr(III)-MnO2(s) + OH- 1c) Oxidation of CrOH2+ on the surface of birnessite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
1d) Desorption of Cr(VI) from birnessite Cr(VI)-Mn(II) + 4H2O rarr Mn2+
(aq) + HCrO4-(aq) + 7H+
Pathway 2 Birnessite Dissolution 2a) Mn4+ dissolution from Birnessite MnO2(s) + 4H+ rarr Mn4+
(aq) + 2H2O 2b) Mn4+ adsorption on the surface of chromite Mn4+
(aq) + FeCr2O4(s) rarr Mn(IV)-FeCr2O4 2c) Oxidation of Cr(III) on the surface of chromite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
2d) Desorption of CrO24- from chromite
Cr(VI)-Mn(II) + 4H2O rarr Mn2+(aq) + HCrO4
-(aq) + 7H+
Pathway 3 Solid Solution Reaction 3) Solid solution interaction between chromite and birnessite
Figure 313
99
CHAPTER 4
CHROMIUM GEOCHEMISTRY OF SERPENTINE SOILS
ABSTRACT
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
chemistry of Cr in serpentine soils including its protolith is reviewed focusing on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils are slightly acidic (average pH of ~6) contain a variety Fe(III) oxides
(magnetite and hematite) phyllosilicates (serpentine and chlorite) and clays (smectites
and vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1)
and Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Chromium(III) is the only valence state observed
in the serpentine soil solids however Cr(VI) has been identified in New Caledonia and
California serpentine soil solutions at concentrations below 30 microM The enrichment and
range of Cr concentrations in serpentine soils are directly related to the presence of Cr-
spinels specifically chromite and Cr-magnetite These spinels are resistant to weathering
and are preserved in the soil environment however oxidation of Cr(III) from Cr-spinels
by high valent Mn oxides or other strong oxidants is a potential source of Cr(VI)
identified in serpentine soil solutions Due to the chemically and physically resistant
nature of the Cr-spinels Cr-bearing silicates including clay minerals Cr-chlorite Cr-
garnet Cr-mica and Cr-epidote are more viable sources of Cr identified in vegetation
soil extractions soil solutions and related waters
100
INTRODUCTION
Serpentine soils a generic term used to describe any soil derived from an
ultramafic rock (peridotites and pyroxenites) or serpentinite (metamorphosed ultramafic
rock) are often characterized by 1) low concentrations of plant nutrients such as N P
and K 2) high concentrations of biologically toxic elements including Ni Co and Cd 3)
low ammonium acetate extractable CaMg ratios and 4) unique flora and physical
properties (Brooks 1987) Measured Cr concentrations in serpentine soils range from
~200 mg kg-1 to 6 Wt values enriched compared to non-serpentine soils lt200 mg kg-
1 (Kabata-Pendias and Pendias 1984 Schwertmann and Latham 1986 Brooks 1987)
Chromite and Cr-magnetite are common primary sources of Cr in these soils (Rabenhorst
et al 1982 Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989
Gasser and Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003)
however the plethora of factors relating the chemistry of Cr in serpentine soils and the
weathering of chromite has not been fully addressed using an integrative approach
Overall serpentinites and serpentine soils cover ~1 of the earthrsquos exposed
surface and these rocks and soils are commonly associated with ophiolite complexes
(Coleman 1977) Serpentinites and serpentine soils related to ophiolite complexes are
the primary interest of this paper A map illustrating the worldwide distribution of major
ophiolitic serpentinite bodies and their related soils is shown in Figure 41 A majority of
the sites are near convergent boundaries (Figure 41) Although these rocks and soils
cover a minimal area of the earthrsquos surface they are present in populated areas within the
Circum-Pacific Margin and the Mediterranean Due to the wide-distribution and
occurrence of serpentinites and serpentine soils weathering of these rocks and soils will
differ from location to location due to varying climatic conditions as well as the nature of
the parent material and other factors including topography biota and time (Proctor and
Nagy 1992) As a result the chemistry of these soils as well as the nature of Cr in the
serpentinites and serpentine soils will vary from site to site potentially allowing Cr to be a
source of non-anthropogenic contamination
The objective of this paper is to review the chemical characteristics of serpentine
soils globally evaluate mineralogical and compositional variations between several soils
derived from ultramafic sources and assess the geochemistry of Cr in diverse soil-
101
forming environments Additionally chemical and spectroscopic analyses were
performed on soils derived from ultramafic rocks collected from 1) New Caledonia
South Pacific 2) Eight Dollar Mountain Selma OR 3) Nickel Mountain Riddle OR 4)
Harvard Mine Jamestown CA 5) Jasper Ridge Biological Preserve Stanford CA and 6)
Pillikin Mine El Dorado CA shown geographically in Figure 41 A basalt-derived soil
from Hawaii is included in this study in order to compare a soil formed from a mafic
protolith to those from ultramafic protoliths The various field sites are shown in Figure
42 These soils allow a comparative examination of how soil-forming factors such as
climate topography biota parent material and time affect the chemistry of Cr
REVIEW
Chromium Geochemistry
Chromium is a group VIB transition metal with two common oxidation states
Cr(III) and Cr(VI) differentiated by noticeably different chemical characteristics and
toxicities It should be noted that Cr(II) has been observed in lunar basalt olivines and in
high temperature terrestrial olivines (Sutton et al 1993 Li et al 1995) however this
oxidation state is not observed in soil environments A majority of Cr commonly persists
in ultramafic rocks serpentine soils and related waters as Cr(III) a non-hazardous
species and a micronutrient in human nutrition (Daugherty 1992 Cohen et al 1993
Fendorf 1995 James et al 1997) Chromium(III) has the highest crystal field
stabilization energy for all transition metals strongly favors octahedral coordination and
has 3 d-electrons with a high spin state Chromium(III) is strongly hydrolyzed in
aqueous solutions and the predominant species based on pH redox potential and O2(aq)
are shown in Figure 43 Chromium(III) species are more prevalent under reducing
conditions and its mobility is increased only under strongly acidic conditions At pH
values greater than 9 the hydrolyzed Cr(III) species is negatively charged demonstrating
the amphoteric properties of Cr(III) (Figure 43) Due to the factors described above
Cr(III) forms low solubility compounds and adsorbs strongly on mineral surfaces and
organic matter at pHgt4 (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) For
example the adsorption of Cr(III) on HFO (hydrous ferric oxide) as a function of pH is
shown in Figure 43 illustrating an adsorption edge at a pH of ~4 (Leckie et al 1984
102
Stumm 1992) Chromium(III) oxidation to Cr(VI) has been observed in reactions
involving hydrogen peroxide (H2O2) and high valent Mn oxides (Eary and Rai 1987
Johnson and Xyla 1991 Fendorf et al 1992 Fendorf et al 1993 Rock et al 2001)
Chromium(VI) is a strong oxidizing agent a highly corrosive substance a toxin to
living cells and a Class A human carcinogen by inhalation (Daugherty 1992 Cohen et
al 1993 Fendorf 1995) The hydrolyzed Cr(VI) species are also shown in Figure 43
with respect to pH redox potential and O2(aq) Chromium(VI) is more favorable under
oxidizing conditions and its mobility is enhanced under basic conditions In solution
Cr(VI) is a weakly adsorbing (oxy)anion (HxCrO42-x) on common weathering products
meaning that it forms highly labile complexes that can be readily displaced by ions
including phosphate and sulfate (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Additionally the adsorption of CrO42- on HFO shown in Figure 43
demonstrates the anionic character in which the mol bound of CrO42- increases with
decreasing pH and an adsorption edge is at pH ~72 (Leckie et al 1980 Stumm 1992)
The reduction of Cr(VI) to Cr(III) is known to occur by reactions with Fe(II)aq Fe(II)-
containing minerals such as magnetite H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996)
Mineralogy of Serpentine Soils
Serpentine soils have been identified on every continent excluding Antarctica
(Figure 41) The occurrence and distribution of ultramafic rocks specifically related to
ophiolites are closely associated with convergent plate boundaries (Figure 41 Coleman
1977 Coleman and Jove 1992) Tectonic blocks of ultramafic rock are hypothesized to
be detached from the down-going slab of oceanic crust and incorporated into the
subduction meacutelange Subsequent serpentinization of peridotite forms low density
serpentine group minerals (245-250 g cm-3) resulting in the upward diapiric migration of
the serpentinite masses along zones of structural weakness (fault zones) and through the
denser enclosing subduction meacutelange (~265 g cm-3) For example North America has
numerous exposed ophiolites containing serpentinite located in western California and
Oregon where subduction zones were once active in the Early Cretaceous (Harden
1998)
103
Serpentinization of ultramafic rocks is a significant factor controlling mineral
assemblages and textures in serpentinite Hydration and oxidation of olivine and
pyroxene during serpentinization forms serpentine (lizardite chrysotile and antigorite)
talc chlorite tremolite brucite and magnetite (Coleman 1977 Brooks 1987 Coleman
and Jove 1992 Malpas 1992) Additionally magnetite formation creates favorable
conditions for native metal compounds (such as Fe Co and Ni alloys) to form (Coleman
and Jove 1992) As a serpentinite mass is emplaced the extensive physical and chemical
alteration of the serpentinite results in numerous crosscutting alteration veins and highly
sheared textures Ultimately the mineral assemblages and textures in serpentinites are
complex prior to pedogenesis due to the multiple phases of metamorphosis tectonic
deformation and weathering of the rock
Serpentine soils consist of a mixture of residual minerals from the soil protolith
eolian and illuvial minerals and minerals formed within the soil Additionally highly
weathered serpentine soils enriched in Fe-oxides are also termed laterites soils common
in tropical and subtropical regions Soil protolith minerals may include primary minerals
of ultramafic rocks including olivines pyroxenes and oxides (chromite and magnetite)
as well as a large variety of metamorphic minerals the most common being serpentine
(lizardite chrysotile and antigorite) chlorite talc and actinolite Silica phases occur as
eolian grains concretions andor net type structures or silica webs (Hotz 1964
Coleman 1977 Schwertmann and Latham 1986) Eolian and illuvial aluminosilicate
minerals are important sources of Al in serpentine soils an element of low abundance in
ultramafic rocks
Layer silicates observed in serpentine soils are varied in type and abundance
They include smectites (often as montmorillonite) vermiculite chlorite brucite
serpentine and a variety of intrastratified clays (Veniale and Van der Marel 1963 Hotz
1964 Schellmann 1964 Wildman et al 1968 Rabenhorst et al 1982 Schreier et al
1987 Graham et al 1990 Bulmer et al 1992 Gasser and Dahlgren 1994 Lee et al
2003) Compositional variations crystal structures and parageneses of layer silicates
have been characterized for a number of serpentine soils (Wildman et al 1968
Rabenhorst et al 1982 Schreier et al 1987 Graham et al 1990 Bulmer et al 1992
Lee et al 2003) In addition to the layer silicates goethite hematite maghemite and
104
amorphous Fe (oxy)hydroxides are soil-forming phases in serpentine soils (Coleman and
Jove 1992) Gibbsite is not a common phase in most serpentine soils however limited
isomorphic substitution of Al(III) in goethite and hematite was noted in New Caledonia
(Schwertmann and Latham 1986) Manganese oxides such as birnessite (δ-MnO2) are
also present in serpentine soils and often coat mineral surfaces
Sources of Chromium for Serpentine Soils
Chromium primarily occurs in oxide and silicate phases in ultramafic rocks and
serpentinites Chromium-spinel often referred to as chromite is a common primary
(igneous) spinel in mafic and ultramafic rocks and often resists low-grade metamorphic
processes related to serpentinization (Hoffman and Walker 1978 Malpas 1992)
Isomorphic substitution of Al3+ Fe3+ and Ti4+ into the octahedral sites and Mg2+ Ni2+
Zn2+ and Mn2+ into the tetrahedral sites is common however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987b Pan and Fleet 1991 Sack and Ghiorso 1991 Burkhard 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996 Mathiesen 1999) Serpentinite chromite compositions reported in the literature are
plotted in Figure 44 to demonstrate the wide-range of Cr concentrations possible for Cr-
spinels Isomorphic substitution of octahedral Cr and octahedral Al for these chromite
compositions are also shown in Figure 44 Serpentinite pyroxenes containing Cr include
augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) (Grapes 1981 Von Knorring et al
1986 Treloar 1987a Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996)
Magnetite and serpentine minerals may incorporate varying amounts of Cr into
their structures during serpentinization Chromium(III) from primary Cr-containing
phases may isomorphically substitute into the Fe(III) site of magnetite due to size
(octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge similarities Chromium
concentrations in magnetite have been observed as high as 13 Wt (Oze et al 2003)
Chromate minerals such as crocoite (PbCrO4) are extremely rare in serpentinites and the
occurrence of these minerals are mainly limited to specific localities such as in Australia
Tasmania and the Urals (Williams 1974 Crane et al 2001) Serpentine minerals have
105
been reported to contain as much as 1500 mg kg-1 Cr based on spectrographic analyses
by Page and Coleman (1967) however serpentine minerals typically contain only trace
amounts of Cr
Hydrothermal and CO2 metasomatism of ultramafic rocks during the evolution of
convergent plate margins commonly produce Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite lt19 Cr Wt ) uvarovite (Cr-garnet lt20 Cr Wt )
tawmawite (Cr-epidote lt13 Cr Wt ) and kaumlmmererite (Cr-chlorite lt10 Cr Wt ) as
shown in Figure 44 (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974 Neiva
1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981 Max et al
1983 Nutman et al 1983 Raase et al 1983 Von Knorring et al 1986 Kerrich et al
1987 Treloar 1987b Treloar 1987a Morand 1990 Pan and Fleet 1991 Schandl and
Wicks 1993 Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995
Mathiesen 1999 Proenza et al 1999) Figure 44 illustrates the relationship of Cr and
Al substitution for Cr-silicate compositions based on Cr and Al Wt Additionally Cr-
silicate compositions are plotted based on octahedral Cr and octahedral Al in Figure 44
Compositions for Cr-muscovite Cr-epidote and Cr-garnet plot near the tie lines
connecting Cr and Al end-member minerals Chromium isomorphically substitutes for
octahedral Al in these silicates due to similarities in charge octahedral preference and
octahedral radii (Al3+ = 061 Aring) The numerous isomorphic substitutions possible for Cr
and Al in the chlorite structure results in analyses that do not plot near the tie line
connecting clinochlore and kaumlmmererite Overall any of these minerals are potential
sources for Cr in soils produced by the weathering of serpentinites and their hydrothermal
and CO2 metasomatic derivatives
Chromium in Serpentine Soils
Chromium concentrations vary widely in serpentine soils worldwide (Table 41)
New Caledonia soils appear to be the most Cr-enriched serpentine soil with a Cr
concentration of ~60000 mg kg-1 (Schwertmann and Latham 1986) however Jaffre
(1980) report values ranging from 6300 to 80000 mg kg-1 for New Caledonia soils
Some serpentine soils have measured Cr concentrations as low as 29 and 66 mg kg-1
atypical of most values reported in Table 41 (Proctor and Woodell 1971 Brooks et al
106
1992 Jeffrey 1992) Chromium concentrations as a function of depth for serpentine soil
profiles in Maryland (Rabenhorst et al 1982) Tehama County CA (Gough et al 1989)
the Klamath Mountains CA (Hotz 1964) Newfoundland Canada (Roberts 1992)
Harvard Mine CA (this study) and Jasper Ridge CA (Oze et al 2003) are reported in
Figure 45 The pedogenic process(es) responsible for the varying Cr concentrations as a
function of depth for each soil have not been characterized however Oze et al (2003)
determined that a majority of Cr in Jasper Ridge serpentine soils is the result of
weathering-resistant residual Cr-bearing minerals from the protolith mainly chromite
and Cr-magnetite Additionally Oze et al (2003) demonstrated that soil samples from
Jasper Ridge contain higher concentrations of Cr with respect to their corresponding
bedrock due to the Cr-spinels accumulating over time The observation that Cr
concentrations decrease with depth towards the protolith is characteristic for most of the
profiles in Figure 45
In addition to examining total Cr concentrations chemical extractions performed
on serpentine soils provide additional information with regards to Cr availability and the
minerals and complexes in which Cr may reside in the soils Table 42 summarizes a
number of extraction experiments performed on serpentine soils For a majority of the
soils listed in Table 42 total Cr concentrations for the serpentine soils were not
measured however these extractions do provide a quantitative assessment of where Cr
may be bound in soil (such as silicates organic matter and secondary Fe-bearing phases)
Extractions performed by Schwertmann and Latham (1986) were able to remove ~20000
Cr mg kg-1 from the serpentine soil using a dithionite-citrate bicarbonate (DCB) solution
which preferentially dissolves crystalline Fe-oxides Solutions such as concentrated
nitric and hydrochloric acid were only able to remove a maximum of ~1400 Cr mg kg-1
from Unst serpentine soils (Shewry and Peterson 1976) which preferentially dissolve
Cr-bearing silicates and to a lesser extent the Fe-oxides Where bulk soil Cr values were
measured extraction experiments are only able to remove a small portion (lt1) of the
total Cr present in the serpentine soils (Table 42) (Gasser and Dahlgren 1994 Oze et al
2003) Based on the extractions presented in Table 42 it appears that Cr(III) resides in
minerals and phases resistant to common extraction techniques
107
In serpentine soils Fe(III) oxides and Fe(III) hydroxides provide a host of
structural and surface sites available to sequester Cr(III) released to soil solutions by
dissolution of Cr-bearing minerals during soil-forming processes Substitution of Cr(III)
for Fe(III) in Fe-oxides Fe-(oxy)hydroxide minerals and amorphous compounds is
expected based on crystallographic considerations synthesis experiments and analysis of
extracts from selective dissolution of Fe(III) compounds in serpentine soils
(Schwertmann and Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990
Gerth 1990 Gasser and Dahlgren 1994) Chromium(III) also readily sorbs on surfaces
of Fe(III) oxides and (oxy)hydroxides via inner-sphere surface complexes (Figure 43)
and forms low-solubility precipitates (Charlet and Manceau 1992) The low
concentration of Cr(III) in soil solutions has been attributed to the low solubility of
Fe(III)-Cr(III) amorphous hydroxides (Sass and Rai 1987)
Layer silicates provide structural and surface sites capable of uptaking Cr(III)
Octahedral Cr(III) in clays such as smectites and vermiculites is expected and Cr
concentrations in the clay-sized fraction can achieve concentrations of ~1000 mg kg-1 in
serpentine soils (Oze et al 2003) Sorption of Cr(III) and Cr(VI) on edge and interlayer
sites of phyllosilicates have been evaluated by Ilton and Veblen (1994) Ilton et al (2000)
and Brigatti et al (2000) indicating that layer silicates (montmorillonite chlorite biotite
phlogopite and muscovite) provide sites for both Cr(III) sorption and Cr(VI) reduction
Kaupenjohann and Wilcke (1995) concluded that Cr in serpentine soils from Bavaria is
residually bound mainly in silicate structures including clay minerals based on extraction
experiments
Soil Solutions and Related Water
The addition of water gradually changes the spatial and temporal characterization
of pore fluids in soils including the ionic strength pH and redox potential Leaching of
a soil should progressively enrich the soil with less chemically mobile elements such as
Cr and Fe Elements identified in soil solutions include Mg Ca Fe Ni and Cr and the
ionic strengths of serpentine soil solutions are typically low (lt001) (Anderson et al
1973 Johnston and Proctor 1981 Gasser and Dahlgren 1994) Soil solutions obtained
from serpentine soils using tension-free lysimetry demonstrate Cr concentrations ranging
108
from 01 to 32 micromol L-1 (Gasser and Dahlgren 1994) This study also revealed a
majority of Cr Fe Mg and Ni in extracted soil waters were in the form of colloids
suggesting that nanoparticles are an important factor of Cr translocation in soils
The pH of a soil is a dominant factor controlling the nature of many chemical and
microbial reactions including the geochemistry of Cr (Brady and Weil 1999) Ranges in
pH for serpentine soils worldwide are plotted in Figure 46 The highest soil pH was
measured in a South Africa serpentine soil with a value of 9 (Cole 1992b) whereas the
lowest pH was reported in Ireland with a value of 38 (Jeffrey 1992) This wide range of
pH is a function of the adsorbed H+ and Al3+ the base-forming cations the organic
matter and the mineralogy present within each soil The pH range of serpentine soils
shown in Figure 46 is plotted on the stability diagram in Figure 43 demonstrating that
both Cr(III) and Cr(VI) aqueous species are favorable The Eh range for the serpentine
soils in Figure 3 is based on the soils being in contact with the atmosphere as well as
being a transitional environment Although the activity of H+ varies by four orders of
magnitude the pHs of most serpentine soils are slightly acidic (pH ~6) which is near that
of average rainwater (pH~57 Figures 43 and 46) Additionally serpentine soil pH
values have been observed to increase with depth towards the protolith for a number of
soils (Rabenhorst et al 1982 Gough et al 1989 Gasser and Dahlgren 1994) The
increase in pH with respect to depth may be the result of rainwater being progressively
influenced by the mineral chemistry and hydrolysis reactions as it descends through the
soil towards the protolith
Groundwater from serpentinized ultramafic rocks have pH values ranging from 8
to 12 (Barnes et al 1972) and these waters have been divided into two categories as
shown in Figure 46 Mg bicarbonate (Mg2+-HCO3-) type and Ca hydroxide (Ca2+-OH-)
type (Barnes and ONeil 1969 Barnes et al 1972) The Mg bicarbonate waters (pH~8)
are characteristic of most surface waters and shallow groundwaters from serpentinites
and ultramafic rocks The Ca hydroxide waters (pHasymp10-11) are associated with
incompletely serpentinized peridotites and these fluids are comparatively low in Mg2+
Fe2+ Mn2+ Crtotal and SiO2total The Ca2+ in solution originates from the serpentinization
of Ca-rich orthopyroxene (Barnes et al 1972) The pH of serpentinite groundwaters are
109
plotted on the Cr stability in Figure 43 demonstrating that both Cr(III) and Cr(VI) are
potential valence states in these waters
Chromium predominates as Cr(III) in ultramafic rocks and serpentine soils
however Cr(VI) has been identified in serpentine soil solutions and serpentinite
groundwater in California and New Caledonia at concentrations of lt20 microg L-1 (Gough et
al 1989 Becquer et al 2003) Calcium hydroxide waters (pH~11) collected from the
same spring evaluated by Barnes et al (1972) in the Cedars ultramafic body (Cazadero
CA) contains 12-22 Cr(VI) microg L-1 and no Cr(III) (this study) Additionally water from
chromite-containing ultramafic rocks in the Siniktanneyak Mountain ultramafic complex
of Alaska with pH values ranging from 67 to 93 were measured with low total Cr
concentrations lt2 Cr microg L-1 (Taylor et al 1998) A mechanism for Cr(III) oxidation in
serpentine soils and related water suggested by (Gough et al 1989 Becquer et al 2003)
may be the result of high valence Mn oxides common in serpentine soils High valence
Mn-oxides such as birnessite (δ-MnO2) are naturally occurring oxidants capable of
oxidizing Cr(III) to Cr(VI) at pHlt9 (Eary and Rai 1987 Fendorf et al 1992 Fendorf et
al 1993) It is also important to note that Cr(VI) reduction is hypothesized to occur in
soils due to electron transfer involving organic matter and Fe(II)-bearing minerals such as
magnetite (Peterson 1996 Fendorf et al 2000) Whether toxic concentrations of Cr(VI)
can be present in serpentine soils and related water over geologic time is important to
consider especially when high valence Mn oxides Fe(II)-bearing minerals and organic
matter are capable of significantly influencing the oxidation state of Cr
Serpentine Soil Vegetation
Vegetation associated with serpentine soils has received considerable attention
due to the apparent infertility of these soils The inhabitability of serpentine soils for
most plants has been attributed to the imbalance of available Mg relative to Ca or low
CaMg ratios (Walker 1954) the deficiency of plant nutrients such as K N and P
(Turitzin 1991) elevated heavy metal concentrations such as Ni and Cr (Robinson et al
1935) and poor drainage (Brooks 1987) Linking plant productivity to toxic levels of
Cr in the soil has not been fully resolved (Brooks 1987 Kruckeberg 1992) however a
majority of Cr in the soils is not available for plant uptake due to being in high stable and
110
weathering resistant phases discussed in the previous sections (Table 42) In general Cr
concentrations in plants typically do not exceed 100 mg kg-1 (Brooks 1987)
Table 43 provides a global list of serpentine soil plant species and their measured
Cr concentrations with an emphasis on Cr hyper-accumulators ie plants able
incorporate higher than average concentrations of Cr A majority of these measurements
are based on the dry weight of Cr Aerobryopsis longissima from New Caledonia has the
highest reported Cr concentration with values ranging from 200 to 700 mg kg-1 (Lee et
al 1977) Other noteworthy plants include the Lophostachys villosa from Brazil (Brooks
et al 1992) Arenaria humifusa from Newfoundland (Roberts 1992) the Calluna
vulgaris from Unst (Shewry and Peterson 1976) with Cr concentrations of 203 530 and
417 mg kgndash1 respectively Even mosses such as the Funaria flavicans (Maryland)
incorporate Cr concentrations as high as 90 mg kg-1 (Shaw 1991) By examining
specific parts of the plant Gough et al (1989) determined that Cr concentrations are
highest in the basal burl (600 mg kg-1) of the manzanita bush compared to the roots (30
mg kg-1) stems (20 mg kg-1) and leaves (20 mg kg-1) The study by Gough et al (1989)
demonstrates that Cr is not homogeneously distributed throughout the plant and that
evaluating a leaf versus a stem could result in significantly different Cr concentrations
Soil Organic Matter
Soil organic matter (SOM) including the biomass (living organisms such as
bacteria) detritus (nonliving plant residues) and humus (amorphous and colloidal
mixtures of organic substances) affects the geochemistry and buffering capacity of
serpentine soils Furthermore biologic reduction of Cr(VI) to Cr(III) via cometabolic
and metabolic pathways coupled with carbon oxidation has been demonstrated for a
number of bacterial strains including aerobes (Gopalan and Veeramani 1994 Park et al
2000) and anaerobes (Turick et al 1996 Tebo and Obraztsova 1998) Additionally
aqueous Fe(II) produced by microbial Fe(III) reduction is a known biogenic pathway for
the reduction of Cr(VI) (Fendorf et al 2000) Fe(III)-Cr(III) (oxy)hydroxides are the
resulting secondary phases (Rai et al 1989 Patterson and Fendorf 1997 Hansel et al
2003) Bacteria oxidizing Cr(III) to Cr(VI) has not been observed
111
Most serpentine soils in temperate regions appear to be either Mollisols or
Alfisols which often contain high concentrations of SOM Alexander et al (1989) and
Bulmer et al (1992) reported that serpentine soils can contain organic carbon
concentrations in excess of 50 mg g-1 Humus has a high cation exchange capacity and
provides much of the pH buffering capacity in these soils (Brady and Weil 1999)
Organic acids such as fulvic and humic acids are known to complex cations such as
Cr3+ Fe3+ and Mn2+ forming stable organometallic complexes (Kaupenjohann and
Wilcke 1995 Brady and Weil 1999) Additionally metal leaching experiments
performed by Schreier et al (1987) established that organic acids produced by soil
organic matter such as oxalic acid and citric acid were extremely effective in releasing
Cr Mg Fe Ni and Co from serpentinite rocks and sediments
CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
Serpentine soils were obtained from California Oregon and New Caledonia and
analyzed to further evaluate the factors influencing the Cr geochemistry in serpentine
soils Analyses for these soils include pH electrical conductivity total organic matter
total N ammonium acetate extractable Ca K Mg and Na soil texture soil color
chemical composition of the soil and the chemical composition of Cr-bearing minerals
The general soil characteristics are described in the following section and the analytical
methods are explained in Appendix 41 Although a majority of these soils are located in
the western United States factors influencing the pedogenesis of these soils vary
considerably allowing a comparative analysis of how the fate of Cr might be influenced
by climate topography biota parent material and time
Sample Locations and General Soil Characteristics
Six serpentine soils were collected from locations shown in Figures 41 and 42 to
evaluate the chemistry of Cr in serpentine soils These soils include 1) a laterite (Oxisol)
collected in New Caledonia South Pacific 2) a laterite (Oxisol) exposed at Eight Dollar
Mountain Selma OR 3) a laterite (Oxisol) near Nickel Mountain Riddle OR 4) a
serpentine soil (Inceptisol) with a CO2-metasomatized serpentinite protolith from
Harvard Mine Jamestown CA 5) a serpentine soil (Mollisol) obtained from a preserved
112
grassland at Jasper Ridge Biological Preserve Stanford CA and 6) a serpentine soil
(Inceptisol) collected in the proximity of Pillikin Mine El Dorado CA Lastly an Oxisol
formed from basalt obtained in Oahu Hawaii is also included for comparison
These seven sites include a range of soil environments from different protoliths
under varying conditions of rainfall temperature and topography Each soil was collect
near the highest and topographically level point for each serpentine soil site Serpentine
grassland soils at Jasper Ridge formed under Mediterranean-like climates (xeric moisture
regime) whereas higher temperatures and less rainfall (ustic moisture regime)
characterize soil formation at the Harvard and Pillikin mines in the Sierra Nevada
Foothills metamorphic belt The vegetation of these three sites is dominantly grasses with
some manzanita These modern soil environments contrast with the Miocene Ni and Cr-
rich laterite soils of Nickel Mountain and Eight Dollar Mountain that were probably
formed under an aquic moisture regime near sea level in a subtropical climate Grasses
manzanita and pine trees are the dominant forms of vegetation at Nickel and Eight
Dollar Mountain An ultramafically-derived laterite formed under tropical conditions
from New Caledonia was collected The vegetation and topography related to the New
Caledonia laterite was not recorded during sampling An Oxisol from Hawaii is also
included in this study to evaluate the differences between a highly weathered soil
originating from an ultramafic rock and one from a mafic rock (basalt) under tropical
conditions The Hawaiian soil was collected from a topographically level area however
the vegetation related to this site was not noted
BedrockProtolith Mineralogy
A broad characterization of the protoliths for the serpentine soils is briefly
summarized due to the bedrock not being collected from all the field sites New
Caledonia is composed of variably serpentinized peridotites predominantly harzburgites
and some dunites (Leblanc 1995) Nickel Mountain and Eight Dollar Mountain are
predominantly serpentinized harzburgite and dunite The bedrock for soils collected at
Harvard Mine is mainly composed of quartz ankerite and mariposite (also referred to as
QAM) caused by the CO2 metasomatism of serpentinite associated with gold
mineralization (Bohlke 1989) The average serpentinite at Jasper Ridge is composed of
113
serpentine (mainly lizardite and antigorite) chlorite talc magnetite and chromite (Oze et
al 2003) however brucite has been reported as a major constituent in related Franciscan
Complex serpentinites (Hostetle et al 1966) Pillikin Mine serpentinites dominantly
consist of talc and serpentine (lizardite antigorite and chrysotile) and these serpentinites
contain podiform chromite deposits (Cater et al 1948) Mafic bedrock for the Hawaiian
Oxisol is tholeittic basalt
Soil Chemistry and Physical Properties
As shown in Figure 47 the colors of the collected soils range from the red New
Caledonia laterite to the greenish gley of the Pillikin Mine Inceptisol The Munsel color
values are also shown in Figure 47 The dark grey color of the Jasper Ridge soil and the
greenish gley of the Pillikin Mine soil are atypical of soils derived from serpentinites due
to the lack of any apparent redness The range of these colors is often attributed to the
presence of Fe-oxides however organic matter and Mn oxides are also dominant factors
controlling color (Brady and Weil 1999)
Chemical properties for the six serpentine soils and the Hawaiian soil are listed in
Table 44 The slightly acidic pH values range from 53 to 69 similar to those shown in
Figure 43 and 46 The New Caledonia laterite is the most acidic soil with a pH of 53
compared to the Harvard Mine carbonate-buffered soil at 69 The pH of the Hawaiian
soil is also slightly acidic with a pH of 604 The electrical conductivities (EC) of the
soils range from 01 to 07 dS m-1 a range indicative of relatively few dissolved inorganic
solutes The highest organic matter content is in the very dark grey Jasper Ridge soil
with a value of 32 and the lowest is in the red New Caledonia laterite (05) The
Hawaiian soil has an organic matter content of 1 Nitrogen concentrations for all the
soils are lt012 with Jasper Ridge having highest N concentration which is consistent
with the soil organic matter content
Ammonium acetate extractable Ca K Mg and Na are also listed in Table 44
Magnesium concentrations are gt6923 mg kg-1 in all of the serpentine soils The lowest
concentration of Mg was measured in the basalt-derived Hawaiian Oxisol with a value of
168 mg kg-1 Calcium concentrations in the serpentine soils range from 167 to 998 mg
kg-1 The 998 mg kg-1 of Ca in the Harvard Mine soil is directly related to the abundance
114
of ankerite (gt20 of the bedrock) Potassium concentrations are low for most of the soils
(lt55 mg kg-1) however Jasper Ridge and Harvard Mine soils have the highest
extractable K with values of 237 and 301 mg kg-1 respectively The extractable K value
for the Harvard Mine soil is due to the K-metasomatism related to gold mineralization
involving the serpentinite soil protolith The extractable concentration of Na is lt44 mg
kg-1 for all the serpentine soils whereas the Hawaiian soil has the highest Na (53 mg kg-
1) due to plagioclase and basaltic glass
Textures for the soils range from clay loam clay loam to sandy clay loam (Table
44) The laterite soil in New Caledonia and Oxisol in Hawaii are extremely different in
which the Hawaiian soil is mostly clay (66 clay) compared to the loam at New
Caledonia (25 clay) Additionally both ancient laterites (Eight Dollar Mountain and
Nickel Mountain) have clay textures Harvard Mine Jasper Ridge and Pillikin Mine
soils are all loamy in character
Major and trace element compositions for the soils are listed in Table 45
Chromium concentrations in the serpentine soils range between 827 and 9528 mg kg-1
The Eight Dollar Mountain laterite has the highest Cr concentration and the New
Caledonia laterite contains the lowest Cr concentration The Cr concentration for the
New Caledonia soil is lower than the values reported in Table 41 The Hawaiian soil has
a Cr concentration well below the analyzed serpentine soils with 388 Cr mg kg-1 but still
greater than the 200 mg kg-1 characteristic of serpentine soils Nickel concentrations in
the serpentine soils range from 1895 to 14100 mg kg-1 with the highest concentration
being identified in the Nickel Mountain laterite Manganese concentrations range
between 257 (New Caledonia) to 3017 (Eight Dollar Mountain) mg kg-1 in the serpentine
soils Element concentrations in the Hawaiian Oxisol contrast from the serpentine soils
with higher Al Cu P Pb and Zn concentrations
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in the soil solids for Jasper Ridge New Caledonia Pillikin Mine Eight Dollar
Mountain and Harvard Mine Although the Cr concentrations in these soils exceed 200
mg kg-1 (Table 45) X-ray absorption near-edge spectra (XANES) shown in Figure 48
illustrate Cr(VI) levels are below the detection limits of this method (~10 Cr mg kg-1) due
to the lack of a pre-edge feature The box in Figure 48 illustrates where a Cr(VI) pre-
115
edge peak should be present Chromium(VI) concentrations in these soils are possibly
present in the parts per billion range
Soil Mineralogy
Backscattered electron images and microprobe analyses were used to identify the
Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49
and corresponding Cr-bearing mineral compositions are listed in Table 46 The New
Caledonia soil exhibits a wide range of particle sizes and a Fe-(oxy)hydroxide was
identified containing 165 Cr2O3 Wt Chromium-spinel has been identified in the New
Caledonia laterite by Becquer et al (2003) Eight Dollar Mountain soils exhibit a coarse
texture and both Cr-magnetite (2387 Cr2O3 Wt ) and chromite (3343 Cr2O3 Wt )
are present Nickel Mountain soils are also coarse and chromite (1315 Cr Wt ) was
identified Soils at Harvard Mine contain numerous chromite grains and the mineral
trevorite (a Ni-bearing spinel) was identified with 3991 Cr2O3 Wt and 2054 NiO Wt
Chromite and magnetite are abundant in soils at Jasper Ridge as well as a fine-
grained mixture of chromite magnetite and an unidentifiable silicate with up to 1878
Cr2O3 Wt (Figure 49) Veins of Cr-chlorite crosscut the spinel-silicate mixture in
Figure 49 and this Cr-chlorite is the only Cr-silicate identified using this method
Chromite (5355 Cr2O3 Wt ) is the dominant Cr-bearing mineral detected in soils at the
Pillikin Mine and these grains contain the highest Cr concentrations compared to the
other minerals in the seven soils The Hawaiian Oxisol possesses ulvoumlspinels (007
Cr2O3 Wt and 2465 TiO2 Wt ) which are evenly distributed throughout the soil
Overall microprobe observations reveal that the Cr-bearing spinels are present in every
soil Additionally Cr-bearing spinels at Jasper Ridge have been shown to undergo
incongruent dissolution progressively enriching the spinel towards a FeCr2O4 end-
member (Oze et al 2003)
DISCUSSION
The physical and chemical characteristics of serpentine soils are directly related to
the alteration and tectonic history of the serpentinite as well as the soil-forming factors
including the parent material climate biota topography and time Identifying Cr-
116
bearing minerals their chemical compositions and abundance in the soils allows a more
definitive explanation for Cr enrichment in these soils as compared to total Cr
concentrations listed in Tables 41 and 45 and sequential extraction experiments reported
in Table 42 Detrital chromite grains from the serpentinite protolith are resistant to
weathering and accumulate in the soil over time Although serpentine soils contain Cr
concentrations typically gt200 mg kg-1 the presence of Cr-spinels and their resistance to
weathering demonstrate that a majority of the Cr is not chemically available to interact
within the soil The role of organic matter and vegetation are important factors related to
the chemistry of serpentine soils however their role for influencing a majority of the Cr
geochemistry is limited due to the nature of the Cr-spinels
Chromium and Serpentine Soils
Total Cr concentrations exceed 200 mg kg-1 for most of the serpentine soils
reported in Table 41 and in the six serpentine soils of this study (Table 45) These Cr
concentrations can be accounted for by assessing the Cr-bearing minerals associated with
these soils and their relative abundance Small grains of Cr-spinels or chromite identified
in the soils (Figure 49 and Table 46) have the potential to appreciably impact Cr
concentrations Table 47 demonstrates the mineral masses and volumes of Cr-bearing
minerals required to cause predetermined Cr concentrations in soil samples A minimal
amount of chromite (lt02 of a soil sample mass or 005 of the volume of serpentine)
can result in Cr concentrations of 1000 mg kg-1 A few grains of chromite could easily
be responsible for the Cr concentrations in Tables 41 and 45 In fact residual grains of
Cr-enriched spinels are observed in every soil analyzed (Figure 49 and Table 46)
Ultimately grains of spinel in the soil do appreciably impact Cr concentrations however
they do not provide a readily available source of Cr based on extraction experiments
performed by Oze et al (2003) Serpentine minerals could also provide minor Cr
concentrations to the soils however soil extractions have consistently shown that a
majority of the Cr is in highly resistant phases capable of resisting dissolution even in
concentrated hydrochloric acid nitric acid and hydroxylamine with acetic acid (Table
42) Chromium end-member minerals including eskolaite Cr-muscovite (fuchsite) Cr-
garnet (uvarovite) Cr-epidote (tawmawite) and Cr-chlinochlore (kaumlmmererite) are also
117
included in Table 47 to demonstrate that relatively low abundances of these minerals are
all that is necessary to increase Cr concentrations in rocks and soils Chromium-
clinochlore is often observed developing adjacent Cr-spinels in serpentinites and is
perhaps a more accessible mineral source of Cr in the serpentinites compared to Cr-
spinels (Phillips et al 1980 Schreyer et al 1981 Kerrich et al 1987 Treloar 1987a
Pan and Fleet 1991 Schandl and Wicks 1993 Christofides et al 1994 Challis et al
1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Oze et al 2003)
As shown in Table 44 and Figures 43 and 46 the chemical properties (pH EC
N etc) of serpentine soils are not correlated to Cr concentrations measured in the soil
solids (Tables 41 Figure 45) due to Cr-spinels accounting for a majority of the Cr
(Figure 49 and 410 Tables 46 and 47) However extraction experiments and soil
solution analyses demonstrate that a limited fraction of the total Cr (lt1) is extractable
If only 1 of the Cr in a serpentine soil is extractable a serpentine soil containing 10000
Cr mg kg-1 could possibly contribute up to100 Cr mg kg-1 still a significant Cr
concentration The chemical properties of the soil such as pH will influence this limited
fraction of the Cr in the soils Acidic pHs may aid in mobilizing Cr(III) however this
will only occur at low pHs (lt4) which exceed the range of most serpentine soils (Figures
43 and 46) pH values greater than 8 like those in some serpentine soils and
serpentinite groundwater may help sustain Cr(VI) concentrations in solution (Figure
43) In addition to pH organic matter in the soils may be a significant factor related to
the mobility of Cr due to the production of organic acids promoting the dissolution of the
Cr-bearing phases but particulate organic matter is known to bind Cr(III) in stable stable
organometallic complexes (Kaupenjohann and Wilcke 1995 Brady and Weil 1999)
Evaluating the valence state of Cr is crucial for understanding the chemistry as
well as the potential toxicity related to serpentine soils The soil solids do not contain
detectable concentrations of Cr(VI) (Figure 48) however Cr(VI) concentrations are
present in soil solutions and groundwater at values lt30 microg L-1 (Gough et al 1989
Becquer et al 2003) A majority of serpentinite Cr-bearing minerals only contain
Cr(III) therefore there must be a reaction or process by which Cr(III) is oxidized in a
water saturated soil High valence Mn oxides such as birnessite are present within
serpentine soils and provide a feasible explanation for Cr(III) oxidation Additionally
118
Mn oxides or other strong oxidants in the soil might accelerate the weathering of Cr-
bearing minerals including chromite and generate the Cr(VI) concentrations observed in
soil solutions and groundwater After Cr(III) is oxidized Cr(VI) is a chemically mobile
species and would be transported away from the serpentine soils before it can be
observed in the solid phase Additionally Cr(VI) may be subsequently reduced by
reacting with organic matter or Fe(II)-bearing minerals such as magnetite which are
present in a majority of serpentine soils Perhaps the best way to evaluate the potential of
serpentine soils being a non-anthropogenic source of Cr contamination is to monitor
Cr(III) and Cr(VI) in the soil solutions over time as demonstrated by Gasser and
Dahlgren (1994)
It is curious that plants do uptake Cr in serpentine soils especially when Cr is
tightly bound in the soil matrix (Table 42) In order for plants to uptake Cr it must be in
a form to be absorbed by plant roots Soil pH in the rhizosphere can be altered by plant
roots by as much as 2 units within two days (Fischer et al 1989) This acidification is
known to increase the solubility of the minerals in the soils and could serve as a
mechanism by which plants can incorporate this relatively insoluable element Another
possibility is that Cr(III) in the soil is oxidized by either organic or inorganic means and
then incorporated into the plant in its more chemically mobile and bioavailable form
Cr(VI) How and why plants incorporate as much Cr as they do needs to be examined
more thoroughly
Chromium Spinel in Soils and Rocks
Compositions of chromite identified in serpentinites worldwide (Figure 44) and
Cr-bearing spinels identified in the six serpentine soils are compared in Figure 10 in order
to determine whether the soil spinels are geochemically similar to those in serpentinites
This comparison is ascertained by examining how Al Fe and Mg Wt values varied
with respect to Cr Wt Trends for magnetitehematite chromite and the
microcrystalline spinel-silicate mixture observed in the soils are noted in each diagram in
Figure 410 Additionally the highest Cr concentrations for soil chromite
magnetitehematite and the microcrystalline spinel-silicate mixture are ~40 Cr Wt
~25 Cr Wt and ~15 Cr Wt respectively Based on Al Wt soil
119
magnetitehematite deviates significantly from the chromite trend with Al Wt values
lt5 Magnetitehematites are enriched with Fe relative to serpentinite and soil chromite
and Mg concentrations in the soil chromite are slightly higher than serpentinite chromite
Soil magnetitehematite and chromite are similar by demonstrating a linear trend when
Al Fe and Mg are summed The microcrystalline spinel-silicate mixture (CSM) does
not follow this trend because of Si present in this mineral mixture Overall soil
chromites are similar to serpentinite chromite and soil magnetiteshematites can be
differentiated from the chromites when Cr Wt is plotted against Al Fe and Mg Wt
CONCLUDING STATEMENT
Serpentine soils are abundant among convergent margins and the chemistry of
these soils is ultimately a function of the ultramaficserpentinite rock mineralogy and
texture as well as all the common soil-forming factors Although ultramafic rocks and
serpentinites produce groundwater that is highly basic most serpentine soil pHs are
slightly acidic Although Proctor and Nagy (1992) argued that it is futile to categorize
serpentine soils based on chemistry and biology (otherwise referred to as the lsquoserpentine
factorrsquo) as proposed by Brooks (1987) Cr concentrations greater than 200 mg kg-1 in
serpentine soils are consistently present Chromium in the soils originates from minerals
in the bedrock where the valence is primarily Cr(III) the exception being Cr(VI) in the
rarely present mineral crocoite (PbCrO4) Increased Cr-concentrations in serpentine soils
are largely the result of chromite or other resistant Cr-bearing minerals causing a
localized enrichment in the soils Although Cr in serpentine soils is not readily released
from the soil matrix serpentine soil vegetation can incorporate Cr concentrations as high
as 700 mg kg-1 (Table 34) Finally the limited extractability of Cr and site dependent
factors such as the presence of organic matter clays Fe(II)-bearing minerals pH and the
poor drainage indicates that the possible formation and sustained presence of Cr(VI) is
not a portentous matter related to serpentine soils
120
ACKNOWLEDGEMENTS
We appreciate technical and field support from Robert Jones Anders Meibom
Rob Dunbar and Guangchao Li Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund GSA Graduate Student Research Grants and the
Office of Teaching and Learning at Stanford University is gratefully acknowledged
121
Table 41 Measured Cr concentrations in serpentine soils worldwide
Location Cr concentration (mg kg-1) Reference
Canyon Creek area OR 920 (White 1971) 66-13200 (Brooks et al 1992)
Ireland 29-5000 (Proctor and Woodell 1971 Jeffrey 1992) Italy 1300-3900 (Gambi 1992) Japan 5200 (Shewry and Peterson 1976) Maryland 1500-6000 (Rabenhorst et al 1982) Mt Tamalpais CA 1700 (White 1971) New Caledonia 6300-80000 (Jaffre 1980) New Caledonia 2-6 (Schwertmann and Latham 1986) New Idria CA 1640 (White 1971) New Zealand 850-3250 (Lee 1992) Newfoundland Canada 309-1447 (Roberts 1992) Nickel Mountain OR 1730 (White 1971) Santa Barbara County CA 2080 (Woodell et al 1975) Southern Africa 045-14 (Morrey et al 1992) Tehama County 023-12 (Gough et al 1989) Transvaal Lowveld South Africa 120-800 (Cole 1992) Unst Shetland 3000-4000 (Proctor 1969) Wenatchee Mnts WA 895 (White 1971) Zimbabwe 110-46000 (Soane and Saunder 1959 Proctor et al 1980)
Goias State Brazil
122
123
Table 42 Summary of Cr Fe and Ni extractions from serpentine soils
Range of Extractable Cr Fe and Ni Location Cr Fe Ni
Method Reference
Italy 2-5 microg g-1 305-640 2300-2700microg g-1 25 CH3COOH (Gambi 1992)
NE Portugal 02-09 microg g-1 120-540 microg g-1 40-290microg g-1 EDTA (Menezes de Sequeira and Pinto da Silva 1992) New Caledonia 22-23 384-581 02-216 DCB (dithionite-citrate bicarbonate) (Schwertmann and Latham 1986) New Caledonia 02-11 microg g-1 02-66microg g-1 Method Unknown (Jaffre 1980)
Norway 357microg g-1 316microg g-1 HClO4HNO3 (concentrated) (Zeien and Bruemmer 1989) Norway 254microg g-1 134microg g-1 Ascorbic and Oxalate Acid pH=325 (Zeien and Bruemmer 1989)
Norway 62 microg g-1 53microg g-1 NH4-EDTA (Zeien and Bruemmer 1989) Norway 22-60 microg g-1 8-25 mg kg-1 1-7 microg g-1 pH-stat reactor EDTA (Kaupenjohann and Wilcke 1995)
Swiss Alps 08 micromol L-1 28micromol L-1 228microg g-1 Lysimeter (Nonfiltered) (Gasser and Dahlgren 1994) Trinity Ophiolite CA 004-045 gkg 27-210 g kg-1 005-245microg g-1 Citrate-Dithionite Extracts (Alexander et al 1989)
Unst lt01 microg g-1 19-77microg g-1 Ammonium Acetate (Shewry and Peterson 1976)
Unst 500-1390microg g-1 1540-3330microg g-1 HNO3-Soluble (Shewry and Peterson 1976) JRBP CA 0 microg g-1 01microg g-1 103microg g-1 BaCl2 Method (Oze et al 2003)
JRBP CA 78 microg g-1 513 microg g-1 150 microg g-1 Hydroxlyamine+Acetic Acid (Oze et al 2003) JRBP CA 31 microg g-1 2039microg g-1 35microg g-1 10M Hydrofluoric Acid (Oze et al 2003)
123
124
Table 43 Serpentine soil plant Cr concentrations
Plant Cr Concentration (microg g-1) Location Reference
Lophostachys villosa 203 Goias State Brazil (Brooks et al 1992) Ruellia geminiflora
56 Goias State Brazil (Brooks et al 1992) Minuartia ophiolitica 83 Italy (Gambi 1992) Sanguisorba minor 194 Italy (Gambi 1992) Plantago lanceolata 135 Italy (Gambi 1992) Cerastium latifolium 110 Italy (Gambi 1992) Funaria flavicans 34-899 Maryland (Shaw 1991) Avena sativa 168-583 New Zealand (Lee 1992) Aerobryopsis longissima 200-700 New Caledonia (Lee et al 1977) Arenaria humifusa 530 Newfoundland Canada (Roberts 1992) Senecio pauperculus 406 Newfoundland Canada (Roberts 1992) Rhalomitrium lanuginosum 246 Newfoundland Canada (Roberts 1992) Santolia semidentata 95 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Asplenium cuneifolium 228 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992) Armeria eriophylla 38 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Arctostaphylos viscida 029-705 Tehama County CA (Gough et al 1989) Calluna vulgaris 417 Unst (Shewry and Peterson 1976) Armeria maritima 149 Unst (Shewry and Peterson 1976) Agrostis stolonifera 120 Unst (Proctor and Woodell 1971) Dicoma niccolifera 77 Zimbabwe (Proctor 1992) Xerophyta equisetoides 68 Zimbabwe (Proctor 1992)
124
125
Table 44 Chemical and textural properties of the collected soils
Ammonium Acetate Extractable ------------------------Hydrometer--------------------- Ca K Mg Na
Sand Silt ClaySample pH EC dSm
Organic Matter ------
LECO Nitrogen ------ -------------------mg kg-1-------------------
------ ------ ------
Texture
New Caledonia 534 01 05 lt001 279 29 692 44 43 32 25 LoamHawaii 604
03 10 002 839 46 168 53 8 26 66 ClayEight Dollar Mnt 651 07 16 006 599 62 3464 25 28 28 44 Clay
Nickel Mnt 665 05 07 001 167 42 3289 17 24 32 44 ClayHarvard Mine 691 03 11 003 998 301 2897 43 41 30 29 LoamJasper Ridge 671 04 32 012 505 237 3291 36 36 31 33 Clay LoamPillikin Mine 642 04 05 002 185 55 3421 29 45 23 32 Sandy Clay Loam
125
126
Table 45 Total major and trace element concentrationsdagger of the collected soils
Element Concentration New Caledonia Hawaii Eight Dollar
Mnt Nickel Mnt Harvard Mine Jasper Ridge Pillikin
Mine Ag mg kg-1 lt05 lt05 lt05 lt05 02 lt05 lt05Al 931 1325 314 095 179 314 179Ba mg kg-1 lt10 180 40 20 30 230 lt10Be mg kg-1 05 16 05 lt05 lt05 08 lt05Ca 003 012 085 018 007 079 002Cd mg kg-1 lt05 lt05 18 lt05 05 lt05 lt05Co mg kg-1 56 183 338 403 85 155 122Cr mg kg-1 827 388 9528 5761 2900 5976 1977Cu mg kg-1 78 248 43 25 51 33 4Fe 638 1296 1635 17 538 727 468K 002 046 004 003 004 039 001
Mg 072 026 753 829 792 1025 gt150Mn mg kg-1 257 2706 3017 2554 920 1664 903Na 001 008 005 001 003 057 lt001Ni mg kg-1 2780 345 5510 14100 1895 2180 2760P mg kg-1 50 260 160 150 130 140 20Pb mg kg-1 6 31 4 3 2 8 6Ti 016 041 008 002 lt001 02 002V mg kg-1 63 132 158 54 65 92 27W mg kg-1 10 10 10 10 lt10 lt10 lt10Zn mg kg-1 43 218 160 132 66 111 47
126
dagger ICP-AES analyses
127
Table 46 Compositionsdagger for soil Cr-containing minerals identified in Figure 49
Wt New Caledonia A
Hawaii B
Eight Dollar Mnt C
Eight Dollar Mnt D
Nickel Mnt E
Harvard Mine F
Jasper Ridge G
Jasper Ridge H
Pillikin Mine I
SiO2 491 008 175 026 002 000 1940 001 000Al2O3
756 185 272 2235 2041 088 559 1575 839NiO 023 005 001 006 011 2054 014 002 007FeO 7766a 6212 6633b 3420b 1784 2445 2629 2314 3105MnO 004 037 181 123 133 031 155 146 216Cr2O3 165 007 2387 3343 4682 3991 1878 5024 5355MgO 048 121 202 653 1315 1062 1668 926 497CaO 001 000 013 008 000 000 015 000 000Na2O 000 000 002 001 000 000 006 000 000TiO2 004 2465 012 006 012 000 000 003 018
TOTAL 9257 9041 9879 9821 9980 9671 8864 9920 10037Mineral Cr-Hydroxide Ulvoumlspinel Cr-Magnetite Chromite Mg-Chromite Trevorite CSM1 Chromite Chromite
127
dagger Electron microprobe analyses This is the form of iron unless otherwise stated a FeO is in the form Fe2O3 b FeO is in the form of Fe3O4 1CSM Microcrystalline chromitesilicate mixture
128
Table 47 Chromium-bearing mineraldagger masses and volumes required to influence Cr concentrations in soils
Cr Concentration Mineral mass (grams) required for 1000 g bulk soil analysis Mineral Mass Percent () (mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite
500 11 07 22 24 48 57 011 007 022 024 048 057 1000 22 15 43 48 95 115 022 015 043 048 095 1152000 43 29 86 96 191 230 043 029 086 096 191 2305000 108 73 216 241 477 574 108 073 216 241 477 574
Cr Concentration Mineral Volume Percent () in Serpentine Mineral Volume Percent () in Quartz
(mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite500 005 004 02 02 03 15 006 004 02 02 04 06
1000 01 007 04 03 07 30 01 007 04 03 07 112000 02 014 07 06 14 59 02 015 07 06 14 225000 06 036 18 16 34 147 06 037 18 16 35 55
128
daggerMineral formulas eskolaite Cr2O3 chromite FeCr2O4 fuchsite KCr2(AlSi3O10)(OH)2 uvarovite Ca3Cr2Si3O12 tawmawite Ca2CrAl2Si3O12(OH)2 kaumlmmererite Mg5CrAlSi3O11(OH)8
Note Molar volumes used in the calculations above are from their Al-enriched counterparts in Robie and Hemingway (1995) The molar volume for kaumlmmererite was estimated using unit cell volumes from XRD unit cell measurements from Robie and Hemingway (1995) Christofides et al (1994) and Phillips et al (1980)
129
Figure Captions
Figure 41 Selected serpentine soil localities as well as soils collected for analyses are
shown on a world map Active plate boundaries are shown Serpentine soil localities not
adjacent the plate boundaries on this map are related to ancient plate boundaries that are
no longer active
Figure 42 Field pictures of soils collected from a) Eight Dollar Mountain b) Nickel
Mountain c) Harvard Mine d) Pillikin Mine and f) Jasper Ridge Biological Preserve
Figure 43 Stability field diagram of Cr under a range of Eh and pH conditions (log a
Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom (1998) Adsorption
isotherms for Cr3+ and CrO42- on hydrous ferric oxide (HFO) are also shown (Leckie et
al 1980 Leckie et al 1984)
Figure 44 Microprobe analyses demonstrating Cr versus Al Wt for Cr-spinel
(chromite) and Cr-silicates including Cr-mica (fuchsitemariposite) Cr-garnet
(uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite) (Whitmore et al
1946 Leo et al 1965 Chen and Lee 1974 Neiva 1978 Cooper 1980 Phillips et al
1980 Grapes 1981 Schreyer et al 1981 Max et al 1983 Nutman et al 1983 Raase
et al 1983 Chao et al 1986 Von Knorring et al 1986 Kerrich et al 1987 Treloar
1987b a Morand 1990 Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen
1999 Proenza et al 1999) Chromium-spinel and Cr-silicate compositions based on
octahedral Cr and octahedral Al are also shown with tie-lines (dotted) denoting ideal
isomorphic substitution between Cr and Al end-members
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above Each dotted line connecting the
points represents a separate soil profile at each site
129
Figure 46 Ranges of pH for serpentine soils and related groundwater worldwide Soil
pH values were obtained from A(Kruckeberg 1992) B (Graham et al 1990) C
(Alexander et al 1989) D (Gough et al 1989) E (Rabenhorst et al 1982) F
(Roberts 1992) G (Bulmer et al 1992) H (Proctor 1992a) I (Jeffrey 1992) J
(Gambi 1992) K (Verger 1992) L (Proctor and Cole 1992) M (Cole 1992b) N
(Morrey et al 1992) O (Proctor 1992b) P (Mizuno and Nosaka 1992) Q (Cole
1992a) and R (Lee 1992) Groundwater pH values from serpentinite are from Barnes
and ONeil (1969) and Barnes et al (1972) Note the range of pH environments possible
for ultramafic rocks and their related soils
Figure 47 Soil pictures with Munsel color values and names from soils collected from
a) New Caledonia b) Hawaii c) Eight Dollar Mountain d) Nickel Mountain e) Harvard
Mine f) Jasper Ridge Biological Preserve and g) Pillikin Mine
Figure 48 Cr XANES spectra for soils analyzed from Jasper Ridge New Caledonia
Pillikin Mine Eight Dollar Mountain and Harvard Mine The pre-edge area indicative of
Cr(VI) peaks is noted in the diagram No definitive Cr(VI) peaks are present in any of
these spectra
Figure 49 Backscattering images of mineral phases within each soil are presented
Electron microprobe analyses are noted by a circle and letter for the Cr-bearing minerals
and the results are listed in Table 46
Figure 410 Chromite magnetite and hematite compositions from the collected soils
(gray circles) as well as analyses for serpentinite chromite (white squares) from Figure
44 are shown in these diagrams with Cr Wt versus Al Fe Mg and Al+Fe+Mg Wt
130
Figure 41
131
Figure 42
132
Figure 43
133
Cr Wt
0 10 20 30 40 50
Al W
t
0
5
10
15
20
25Cr-SpinelCr-MuscoviteCr-EpidoteCr-GarnetCr-Chlorite
Cr-SpinelBased on 4 Oxygens
Cr0 1 2
AlVI
0
1
2
ChromiteFeCr2O4
HercyniteFeAl2O4
Cr-MuscoviteBased on 11 Oxygens
Cr0 1 2
AlVI
0
1
2
MuscoviteKAl2(AlSi3O10)(OH)2
FuchsiteKCr2(AlSi3O10)(OH)2
Cr-GarnetBased on 12 Oxygens
Cr0 1 2
AlVI
0
1
2
GrossularCa3Al2Si3O12
UvaroviteCa3Cr2Si3O12
Cr-EpidoteBased on 125 Oxygens
Cr0 1
AlVI
2
3
ClinozoisiteCa2Al3Si3O12(OH)2
TawmawiteCa2CrAl2Si3O12(OH)2
Cr-ChloriteBased on 14 Oxygens
Cr0 1
AlVI
0
1
2
Kammererite(Mg5Cr)(AlSi3)O10(OH)8
Clinochlore(Mg5Al)(AlSi3)O10(OH)8
Figure 44
134
Maryland (USA)(Rabenhorst et al 1982)
Chromium (mg kg-1)
0 1000 2000 3000 4000 5000 6000 7000
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Tehama County CA (USA)(Gough et al 1989)
Chromium (mg kg-1)
0 2000 4000 6000 8000 10000 12000 14000
Dep
th (m
eter
s)
00
02
04
06
08
Kalamath Mountains CA (USA)(Hotz 1964)
Chromium (mg kg-1)
2000 4000 6000 8000 10000 12000 14000 16000
Dep
th (m
eter
s)
00
05
10
15
20
25
30
35
40
45
50
Newfoundland (Canada)(Roberts 1992)
Chromium (mg kg-1)
800 1000 1200 1400 1600 1800
Dep
th (m
eter
s)
00
02
04
06
08
Harvard Mine CA (USA)(Unpublished Data)
Chromium (mg kg-1)
0 500 1000 1500 2000 2500
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Jasper Ridge CA (USA)(Oze et al 2003)
Chromium (mg kg-1)
1500 2000 2500 3000 3500 4000 4500 5000
Dep
th (m
eter
s)
00
01
Figure 45
135
LocationsZ A B C D E F G H I J K L M N O P Q R S
pH
0
1
2
3
4
5
6
7
8
9
10
11
12
Nor
th A
mer
ica
Can
ada
Brit
ain
Italy
Far E
ast
Japa
n
Zim
babw
e
Sout
h A
fric
a
Wes
tern
Aus
tral
ia
New
Zea
land
Mg-Bicarbonate Waters pH=8
Ca-Hydroxide Waters pH=11-12
Average RainpH=57
Sout
h A
fric
a
Italy
Irela
nd
Cal
iforn
ia
Cal
iforn
ia
Brit
ish
Col
umbi
a
Mar
ylan
dCal
iforn
ia
Figure 46
136
Figure 47
137
-2
-1
0
1
2
3
4
5
5980 5990 6000 6010
Energy (eV)
Nor
mal
ized
Abs
orba
nce
New Caledonia
Pillikin Mine
Eight Dollar Mnt
Jasper Ridge
Harvard Mine
Cr(
VI)
Figure 48
138
Figure 49
139
Cr Wt 0 10 20 30 40 50
Al W
t
0
10
20
30 Soil Serpentinite
Cr Wt 0 10 20 30 40 50
Fe W
t
0
20
40
60
80
Cr Wt 0 10 20 30 40 50
Mg
Wt
0
2
4
6
8
10
Cr Wt 0 10 20 30 40 50
Al +
Fe
+ M
g W
t
0
20
40
60
80
MagnetiteHematite
Chromite
MagnetiteHematite
Chromite
Chromite
MagnetiteHematite
SpinelSilicate
SpinelSilicate
SpinelSilicate
SpinelSilicate
Figure 410
140
- Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip1
-
- Chapter 3 Oxidative Promoted Dissolution of Chromite by
- Manganese Dioxide and Concurrent Production of Ch
-
- Chapter 4 Chromium Geochemistry of Serpentine S
-
- List of References
-
- Chapter 2
- Chapter 3
-
- Chapter 4
-
- Appendix A
- Chapter 2
-
- Chapter 3
-
- Chapter 4
-
- Appendix A
-
- CHAP1pdf
-
- Toxicity and Regulations of Chromium
- Mineralogy of Chromium in Serpentinites
-
- Chapter Summaries
-
- CHAP2Tablepdf
-
- JR305CLAY Depth 0-5cm
- JR3515CLAY Depth 5-15cm
- JR33045CLAY Depth 30-45cm
-
- HCAP2Figspdf
-
- Figure 27
-
- CHAP3Tablepdf
-
- Figure 36
- Figure 37
-
- CHAP4pdf
-
- ABSTRACT
- REVIEW
-
- Chromium Geochemistry
-
- Soil Organic Matter
-
- CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
-
- BedrockProtolith Mineralogy
- Soil Chemistry and Physical Properties
- Soil Mineralogy
- Backscattered electron images and microprobe analyses were used to identify the Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49 and corresponding Cr-bearing mineral compositions are listed in Table 46 The New C
-
- DISCUSSION
-
- Chromium Spinel in Soils and Rocks
-
- CONCLUDING STATEMENT
- ACKNOWLEDGEMENTS
-
- CHAP4Tablepdf
-
- Ulvoumlspinel
-
chemistry of Cr in serpentine soils including its protolith is reviewed with a focus on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils have widely varying pH contain a variety Fe(III) oxides (magnetite and
hematite) phyllosilicates (serpentine and chlorite) and clays (smectites and
vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1) and
Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Additionally Cr(III) is the only valence state
observed in the serpentine soil solids however Cr(VI) has been identified in New
Caledonia and California serpentine soil solutions at concentrations below 30 microM The
enrichment and range of Cr concentrations in serpentine soils are directly related to the
presence of Cr-spinels specifically chromite and Cr-magnetite These spinels are
resistant to weathering and are preserved in the soil environment however oxidation of
Cr(III) from Cr-spinels by high valent Mn oxides or other strong oxidants is a potential
source of Cr(VI) identified in serpentine soil solutions Due to the chemically and
physically resistant nature of the Cr-spinels Cr-bearing silicates including clay minerals
Cr-chlorite Cr-garnet Cr-mica and Cr-epidote are more viable sources of Cr identified
in vegetation soil extractions soil solutions and related waters
vi
PREFACE
The chemistry and mineralogy of Cr related to serpentinites and serpentine soils
are assessed in this dissertation using a multidisciplinary field analytical experimental
and theoretical approach The four chapters that constitute this dissertation describe my
investigations related to these rocks and soils Additionally the common theme shared
by these chapters is that they all examine the Cr mineralogy in serpentinites and
serpentine soils and the potential of Cr to be mobilized and oxidized in near-surface
environments Chromite a Cr-bearing spinel is the dominant source of Cr in
serpentinite and serpentine soils Although chromite is resistant to weathering the
presence of birnessite a high valent Mn oxide will increase the release of Cr and the
production of Cr(VI) from chromite over a range of pH Chromium-clinochlore or
kaumlmmererite contains Cr concentrations less than 3 Wt however the rate at which Cr-
clinochlore weathers compared to chromite makes this mineral a more probable source of
Cr released into soil solutions These types of concerns are addressed in this thesis Brief
descriptions of the four individual dissertation chapters including their publication status
and a list of the co-authors are presented in the following paragraphs
Chapter 1 presents the main objectives of this dissertation The chemistry
regulations and toxicity of Cr are discussed as well as the general characteristics of
serpentinites and serpentine soils Additionally a brief summary of each chapter is
provided
Chapter 2 focuses on serpentinites and serpentine soils at Jasper Ridge Biological
Preserve Stanford CA The results have been submitted for publication in an article
entitled ldquoChromium geochemistry of serpentinized ultramafic rocks and serpentine soils
in the Franciscan Complex of Californiardquo to the American Journal of Science (March
2003) Scott Fendorf Dennis Bird and Robert Coleman are the second third and fourth
co-authors respectively
Chapter 3 is an experimental study evaluating the release of Cr and the
production of Cr(VI) from chromite in the presence of birnessite over a range of pH The
results for this study will be submitted for publication in an article entitled ldquoOxidative
promoted dissolution of chromite by manganese dioxide and concurrent production of
vii
chromaterdquo to Environmental Science and Technology The co-authors for this paper are
Scott Fendorf and Dennis Bird
Chapter 4 is a review paper that critically evaluates the chemistry of serpentine
soils and serpentinites primarily focusing on Cr geochemistry Additionally analytical
studies of serpentine soils collected from California Oregon and New Caledonia are
presented This study will be presented at the Robert Coleman Symposium and has been
submitted as an article entitled ldquoChromium geochemistry of serpentine soilsrdquo for an
International Geology Review Special Publication Scott Fendorf Dennis Bird and
Robert Coleman are the second third and fourth co-authors respectively
Appendix A is a preliminary investigation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite) a mineral common in serpentinites and present in serpentine
soils This appendix is entitled ldquoEstimation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite)rdquo The results of this chapter will be included in a study that
estimates the thermodynamic properties of Cr-bearing silicates including fuchsite (Cr-
muscovite) tawmawite (Cr-epidote) and uvarovite (Cr-garnet)
viii
ACKNOWLEDGEMENTS
Serpentinites serpentine soils and chromium are the Geologic Trinity of this
dissertationhellipeach topic is intimately related to one another The scope and breadth of
this research required a significant amount of time and funding None of this research
would have been possible without the generous support of several foundations
organizations and Stanford University I am grateful for the financial aid I have
received from the Stanford Fellowship the Stanford University McGee and Shell funds
the National Science Foundation the Geologic Society of America Graduate Research
Fund the Stanford Office of Teaching and Learning grant and the Jasper Ridge
Biological Preserve Melon Grant
I would not be where am I today if it were not for my parents Robert and Sue
Oze and my brother Rob Oze Thank you for the sandbox and all the hose-driven
volcanoeshellipthose Wookies never stood a chance While some families vacation in theme
parks my family frequented gem and mineral conventions collected rocks and shells
from mist-shrouded shores and sought lucky stones including an infamous quartz orb
Sometimes I think I was destined to be a geologist I would also like to thank everyone
in my family and all our family friends who have provided that extra little push and
support along the way These people include Tudor Bennatts Louis Oze Margaret and
John Chambers John Bennatts Nancy Skiles the Izumis the memory of Witchie Poo
hijacking Harleys from Bremerton bikers and the Man in the Sound waiting for that one
last catch Close friends I would like to acknowledge who are not related to the Stanford
crowd include Dave and Leigh Anderson and Rebecca Parker These satellite friends
helped me keep in touch with the world outside of academia and proved to me that I
wasnrsquot meant to have a real job A special thank you goes to Katherine Akako
Izumihellipyou have made this whole trip worthwhile
My PhD advisors Scott Fendorf and Dennis Bird provided an invaluable amount
of help patience and energy over the past five years Scott always made the time to
discuss and refine many of my projects making them relevant to many current
geochemical issues Additionally he helped me hone my soil science skills in the field
and in the lab Out in the field Scott can dig a pretty mean hole Dennis devoted a
significant amount of time and sweat in order to make this dissertation happen His
ix
insightful knowledge of geochemistry and of the geology of the Sierra Nevadas provided
the cornerstone for this dissertation
Robert Coleman played a key role in the production of this dissertation He
always knew what to say at the right time and helped me understand the complexity of
ophiolites and serpentinites through experienced eyes Additionally Keith Loague and
Gary Ernst deserve recognition for their outstanding support as committee members for
this dissertation over the last four years I would also like to thank Jonathan Stebbins and
Gail Mahood for providing me the opportunity to attend Stanford
Bob and Vicki Jones always offered an open ear whether it was science or solving
the worldrsquos problems Rick Hazlett (Pomona College) and John Winter (Whitman
College) both had a significant impact on shaping my future science goals and objectives
during my undergrad years and both encouraged me to pursue a career in geology They
were the catalyst for this whole experiment
I owe a lot to my research group members (both the Bird Cage and Soil
Chemistry Group) and those people who didnrsquot mind me tagging along Before I
introduce these people one person has been my friend and co-conspirator for many years
at Stanford Tim Copehellipyou are a pain in my ass Besides never picking up any of your
crap and breaking all my stuff this whole grad school experience would not have been
the same without you Other friends and acquaintances that will not be forgotten in
chronological order of me walking into my office for the first time on September 11
1998 include Anne Dyreborg Esra Inan Tom Bawden Patrick Redmond Phil
Neuhoff Thrainn Fredrickson Lara Heister Kaye Savage Andrea Fildani Mike Schultz
Trayle Kushan Kevin Theissen Angela Hessler Chris Van de Ven Kono Lemke Colin
Doyle Colleen Hansel Ben Bostick Matthew LaForce Nancy Grumet Scott Kroeker
Anders Meibom Brian Ebel Matt Ginder-Vogel Matt Pollizitto Spud Amy Weislogel
Travis Horton and Jake Waldbauer I would like to end this Acknowledgement section
and begin my dissertation with a haikuhellip
Better than outcrops
Are the people Irsquove met here
Thanks for all the beer
x
TABLE OF CONTENTS
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iv Prefacehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip vii Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip ix Table of Contentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xi List of Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xiv List of Illustrationshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xvi Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 1
Toxicity and Regulations of Chromiumhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 2 Mineralogy of Chromium in Serpentiniteshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 3 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5 Chapter Summarieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5
Chapter 2 Chromium Geochemistry of Serpentinized Ultramafic Rocks and Serpentine Soils in the Franciscan Complex of Californiahelliphelliphelliphelliphelliphellip 8
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 8 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 9 Backgroundhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 10
Geologic Background of the Serpentinite at Jasper Ridgehelliphelliphelliphellip 10 Serpentine Soils Characteristics and Reviewhelliphelliphelliphelliphelliphelliphelliphellip 11 Sources of Chromium ProtolithBedrockhelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 12 Chromium-Bearing Phases in Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 13
Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Location and Sample Collectionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Mineral Identification in Rocks and Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Soil Chemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 15 X-ray Diffraction and Clay-Size Mineral Identificationhelliphelliphelliphelliphellip 16 X-ray Absorption Spectroscopyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17 Sequential Extraction Procedurehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17
Resultshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Serpentinite Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Distribution of Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip21 Serpentine Soil Chemistry and Mineralogy at Site JR3helliphelliphelliphelliphellip 22 Sequential Extractionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 26
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 27 Concluding Remarkshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 33 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 44 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 47
xi
Chapter 3 Oxidative Promoted Dissolution of Chromite by Manganese Dioxide and Concurrent Production of Chromatehelliphelliphelliphelliphelliphellip 63
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 63 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 64 Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 65
Solid Phase Characterization and Synthesishelliphelliphelliphelliphelliphelliphelliphelliphellip 65 Extraction Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 66 Batch Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Solution Analysishelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69
Results helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite Extraction Experimentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite and Birnessite Batch Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 70 Secondary and Solid Phase Examination of Chromite and Birnessitehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 72 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 72
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 73 Chromium(VI) Pathways from Chromite and Birnessite into Solutionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 76 Serpentine Soils and Cr(VI)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 77
Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 79 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 80 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 84 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 87
Chapter 4 Chromium Geochemistry of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip100
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 100 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 101 Reviewhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102
Chromium Geochemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102 Mineralogy of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 103 Sources of Chromium for Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphellip 105 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 106 Soils Solutions and Related Waterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 108 Serpentine Soil Vegetationhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 110 Soil Organic Matterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 111
California Oregon and New Caledonia Serpentine Soilshelliphelliphelliphelliphelliphelliphellip 112 Sample Locations and General Soil Characteristicshelliphelliphelliphelliphelliphellip 112 BedrockProtolith Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 113 Soil Chemistry and Physical Propertieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 114 Soil Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116 Chromium and Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 117
xii
Chromium Spinel in Soils and Rockshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 119 Concluding Statementhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 120 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 121 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 122 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 129 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 131
Appendiceshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 141 Appendix Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 142 Appendix Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 143 Appendix A Estimation of the Thermodynamic Properties of Kaumlmmererite (Cr-Clinochlore)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip145
Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 145 Nomenclature and Crystal Chemistry of Cr-chloritehelliphelliphelliphelliphelliphelliphelliphelliphellip 146 Estimation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 147
Estimation of theStandard Molar Volume Using Unit Cell Measurementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of the Standard Molar Volume Heat Capacity and Entropyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of Standard Molar Enthalpy and Gibbs Free Energyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 150 Critique and Discussion of the Estimation Algorithms and SUPCRT92helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 151
Evaluation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 152 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 155 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 161 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 162
List of References
References to Chapter 1helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 166 References to Chapter 2helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 172 References to Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 179 References to Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 183 References to Appendix Ahelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 195
xiii
LIST OF TABLES Chapter 2 Table 21 Whole rock elemental concentrations of serpentinites chert and sandstone Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in
serpentinite rock at Jasper Ridge Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in
serpentine soil at Jasper Ridge Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25
and 26 Table 24 Bulk soil elemental concentrations for soils and bedrock collected along
the transect in Figure 21b Table 25 Chemical and physical properties of the soil at Site JR3 Table 26 Selected trace metal concentrations for each size fraction versus depth at
Site JR3 Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3 Table 28 X-ray diffraction d-spacings for common silicates Table 29 Results of the Diablo clay serpentine soil and chromite sequential
extraction experiments Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1
kg of sample Chapter 3 Table 31 Chromite and birnessite masses and concentrations used for each batch
experiment Table 32 Serpentine soil and birnessite experimental parameters Table 33 Summary of Cr(VI) rates obtained from batch experiments Table 34 Comparative analyses of batch experiment solutions and chemical
extractions of chromite and birnessite solids Chapter 4 Table 41 Measured Cr concentrations in serpentine soils worldwide Table 42 Summary of Cr Fe and Ni extractions from serpentine soils Table 43 Serpentine soil plant Cr concentrations Table 44 Chemical and textural properties of collected soils Table 45 Total major and trace element concentrations of the collected soils Table 46 Compositions for soil Cr-containing minerals identified in Figure 49 Table 47 Chromium-bearing mineral masses and volumes required to influence Cr
concentrations in soils Appendix A Table A1 Cr and Al end-members in the Cr-clinochlore solid solution
xiv
Table A2 Standard state thermodynamic properties of Cr and Al-bearing minerals Table A3 Unit cell measurements and molar volumes of clinochlore and Cr-
clinochlores Table A4 Entropy volume and heat capacity for kaumlmmererite using the Ransom and
Helgeson Method Table A5 Fictive polyhedral considered in the estimations Table A6 Estimated thermodynamic properties for kaumlmmererite
xv
LIST OF FIGURES Chapter 2 Figure 21a Soils map of Jasper Ridge Biological Preserve CA Figure 21b Serpentinite and serpentine soil sampling sites at Jasper Ridge Biological
Preserve CA Figure 22 Chromium concentrations versus depth for four serpentine soils in
Maryland California and New Foundland Figure 23 Petrographic images of serpentinite at Jasper Ridge Figure 24 Elemental concentrations relative to Cr for chlorite pyroxene magnetite
chromite and a chromite-silicate mixture in the serpentinite at Jasper Ridge
Figure 25 Backscattering electron images of minerals in the serpentinite at Jasper Ridge
Figure 26 Backscattered electron images of Cr-chlorite and Cr-magnetite assemblage in the serpentinite bedrock
Figure 27 Highest and lowest Cr concentrations at each site in the serpentine soil transect
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected for total gt2 mm sand silt and clay size fractions
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soils at Jasper Ridge
Figure 210 X-ray diffraction clay analyses of sample JR3515 Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at specific depths Figure 212 X-ray microprobe elemental maps of Cr and Fe for the serpentine soil and
serpentinite at site JR3 Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments Figure 214 Scanning electron microprobe images of the chromite standard after each
sequential extraction experiment Figure 215 Tie-lines connecting serpentinite chromite and end-member chromite with
soil chromite values Chapter 3 Figure 31 Scanning electron microprobe images of chromite and birnessite standards Figure 32 Chromite extraction experiments evaluating Cr Fe and Al Figure 33 pH and Eh versus time (hours) for each chromite-birnessite chromite and
birnessite batch experiments Figure 34 pH and Eh with Cr stabilities Figure 35 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant chromite and variable birnessite at pH 5 Figure 36 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant birnessite and variable chromite at pH 5
xvi
Figure 37 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for variable pH and constant chromite and birnessite Figure 38 Rate of Cr(VI) production (d[Cr(VI)]dt) as a function of the surface area
concentration of chromite birnessite and pH Figure 39 SEM examination of solids following the termination of the batch
experiments Figure 310 Serpentine soil and birnessite experiments Figure 311 Iron analyses and the solubility of amorphous Fe(OH)3 and goethite Figure 312 Congruent dissolution of chromite based on Fe and Cr Figure 313 Three pathways of Cr release and oxidation Chapter 4 Figure 41 Serpentine soil localities worldwide Figure 42 Field site pictures of serpentine soils Figure 43 Stability field diagram of Cr over a range of Eh and pH conditions Figure 44 Chromium and Al Wt plot and octahedral Cr and octahedral Al plots
for chromite and Cr-silicates including Cr mica (fuchsitemariposite) Cr-garnet (uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite)
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (m) for selected serpentine soils
Figure 46 pH of serpentine soils and related groundwater worldwide Figure 47 Soil pictures with Munsel color values and names from collected soils Figure 48 Cr XANES spectra for collected serpentine soils Figure 49 Backscattering images of mineral phases within each serpentine soil Figure 410 Chromite magnetite and hematite compositions from analyzed soils as
well as analyses for chromite reported from around the world Appendix A Figure A1 Compositional variation of Cr2O3 and Al2O3 Wt in Cr-chlorite Figure A2 Compositional variation of Cr and VIAl in Cr-chlorite based on 14 oxygens Figure A3 Molar volume estimation of kaumlmmererite based on XRD unit-cell
parameters Figure A4 Natural and calculated compositions of co-existing Cr-clinochlore and Cr-
spinel solid solution in the system Mg0-FeO-Cr2O3-Al2O3-SiO2-H2O at 2 kbar pressure
xvii
CHAPTER 1
INTRODUCTION
Serpentinites or metamorphosed ultramafic rocks are distributed worldwide and
are commonly associated with ophiolite complexes (Coleman 1977 Coleman and Jove
1992) These rocks weather to form serpentine soils a generic term used to describe any
soil derived from serpentinite regardless of its physical or chemical properties
Serpentinites and serpentine soils contain Cr-bearing minerals the most prevalent being
chromite and Cr released from these minerals into surface and subsurface environments
is potentially toxic and carcinogenic The main objective of this dissertation is to
evaluate the mineralogy and chemistry of Cr related to serpentinites and serpentine soils
over geologic time using an integrated analytical experimental and theoretical approach
Additionally the potential of serpentinites and serpentine soils as sources of non-
anthropogenic Cr contamination is investigated A discussion of Cr and its significance
to serpentinites and serpentine soils begins with its discovery by a Frenchman
investigating an ore named ldquored leadrdquo from the Urals
Nicolas-Louis Vauquelin discovered Cr in 1797 in the mineral crocoite PbCrO4
(Burns and Burns 1975) The metalelement was named chromium after the Greek word
chroma meaning color due to its strong influence on mineral pigmentation (Burns and
Burns 1975) For example rubies and emeralds derive their color from trace amounts of
Cr in their structures More importantly JT Lowitz in 1798 discovered Cr in chromite
or Cr-spinel the primary source of Cr for industrial purposes (Burns and Burns 1975)
Chromium is a group VIB transition metal with two common oxidation states (Cr(III) and
Cr(VI)) an average crustal abundance of 100 mg kg-1 and an abundance of 513x105 in
the solar system (relative to [H]=1x1012) (Mason and Moore 1982)
The majority of Cr on the earth exists as Cr(III) where it is an immobile cation
that complexes strongly with organic matter and adsorbs on oxides and silicates even at
low pHs (pHgt4) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) Additionally
the chemical mobility of Cr(III) is limited due to the formation of highly stable
(oxy)hydroxides (Rai and Zachara 1988 Rai et al 1989 Ball and Nordstrom 1998)
1
Chromium(III) has the highest octrahedral site preference energy (OSPE) of any trace
element and it isomorphically substitutes into octahedral sites in oxides
(oxy)hydroxides sulfides and silicates Although Cr(III) is known to substitute for both
octahedral Al(III) and Fe(III) the strongest correlation for substitution appears to be for
Al(III) in oxides and silicates as demonstrated in Chapters 2 and 4 and Appendix A
Chromium(VI) is a strongly oxidizing and weakly adsorbing oxyanion in the form of
chromate (HxCrO42-x) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995 Shiraki
1997 Ball and Nordstrom 1998) Consequently Cr(VI) is chemically mobile and
bioavailable over a wide range of pH (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Chromium(VI) is tetrahedrally coordinated and behaves similarly to
sulfate in minerals such as jarosite (Baron and Palmer 1996)
The pathways and chemistry of Cr oxidation and reduction in soils and related
waters are covered in depth by Fendorf (1995) and Ball and Nordstrom (1998) The
reduction of Cr(VI) to Cr(III) may occur by reactions with Fe(II)aq Fe(II)-containing
minerals such as magnetite organic matter H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996) Oxidation of Cr(III)
to Cr(VI) related to serpentinite-related water and serpentine soil solutions is still a matter
of conjecture Chromium(III) oxidation in serpentine soils has been hypothesized to be
the result of Mn oxides such as birnessite as suggested by both Gough et al (1989) and
Becquer et al (2003) This idea is explored in depth in Chapter 3 where birnessite is
shown to influence the release and oxidation of Cr from chromite over a range of pH (3
to 8) at 25oC
Toxicity and Regulations of Chromium
The examination of Cr based on its oxidation state is necessary due to the
juxtaposing behavior and toxicity of Cr(III) and Cr(VI) Chromium(III) is an essential
micronutrient in human nutrition used by the body to process sugar protein and fat
(Vincent 2000) Inhalation of Cr(VI) is known to cause irritation in the nose including
symptoms such as nosebleeds (epistaxis) and ulcers andor perforations in the nasal
septum (Katz and Salem 1993 Kuo et al 1997) Ingesting considerable quantities of
Cr(VI) can result in stomach ulcers convulsions kidney and liver damage and death
2
(Katz and Salem 1993) Skin contact with certain Cr(VI) compounds causes allergic
contact dermatitis (ulcers) and irritant dermatitis (Katz and Salem 1993) Chromate
induced ulcers or chromeholes are typically red swollen crusted painless lesions
(Katz and Salem 1993) The nosology of Cr is beyond the scope of this dissertation
however Lees (1991) is an excellent source for further information Overall the
Environmental Protection Agency (EPA) the Occupational Safety and Health
Administration (OSHA) the World Health Organization (WHO) and the Department of
Health and Human Services (DHHS) have determined that that Cr(VI) compounds can
increase the risk of lung cancer primarily by inhalation as well as other forms of cancer
caused by ingestion
Federal and state regulation standards appropriate for both Cr(III) and Cr(VI) are
summarized and discussed by Proctor et al (1997) Human health action levels for Cr
vary for each state and the ranges for total Cr in soils include the following 200-66000
Cr(III) mg kg-1 (residential) 02-2000 Cr(VI) mg kg-1 (residential) 2500-1000000
Cr(III) mg kg-1 (non-residential) and 64-71000 Cr(VI) mg kg-1 (non-residential)
(Proctor et al 1997) Incidentally Cr concentrations in serpentine soils and serpentinites
(gt200 Cr mg kg-1) exceed many state action levels and would be considered toxic The
EPA has set a limit of Cr(III)+Cr(VI) at 100 microg L-1 in drinking water The OSHA has
set limits of 500 microg of water soluble Cr(III) compounds per cubic meter of workplace air
or 500 microg m-3 1000 microg m-3 for metallic Cr and insoluble Cr compounds and 52 microg m-3
for Cr(VI) compounds for 8-hour work shifts and 40-hour work weeks The limits
proposed above are constantly changing and should not be used for reference without
checking current regulation standards
Mineralogy of Chromium in Serpentinites
Chromium occurs in a variety of minerals either as a major or minor constituent
(Burns and Burns 1975) Chromium-bearing minerals in serpentinites are primarily
oxides and silicates in which the Cr is present as Cr(III) Chromite is a Cr-bearing spinel
abundant in serpentinites and is resistant to low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) The chemistry and
significance of chromite is addressed throughout this dissertation Pyroxenes such as
3
augite and enstatite are primary minerals identified in serpentinites and contain lt8 Cr Wt
(Grapes 1981 Von Knorring et al 1986 Treloar 1987a Challis et al 1995
Sanchez-Vizcaino et al 1995 Deer et al 1996) Chromate minerals such as crocoite
(PbCrO4) are rare in serpentinites and the occurrence of these minerals are limited to Pb
deposits in Australia Tasmania and the Urals (Williams 1974 Crane et al 2001)
Chromium(III) released from the primary Cr-containing phases may
isomorphically substitute into oxides and silicates for Al(III) and Fe(III) due to size
(octahedral radii Cr3+ = 0615 Aring Al3+ = 061 Aring and Fe3+=0643 Aring) and charge
similarities Magnetite formed during serpentinization incorporates Cr into its structure
and serpentine minerals (antigorite lizardite and chrysotile) contain trace concentrations
of Cr (Spangenberg 1943 Page and Coleman 1967 Burkhard 1993) Eskolaite (Cr2O3)
has been identified in Finland (Von Knorring et al 1986) Metasomatic alteration of
ultramafic rocks commonly results in the formation of Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite) uvarovite (Cr-garnet) tawmawite (Cr-epidote) and
kaumlmmererite (Cr-chlorite) (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Neiva 1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981
Max et al 1983 Nutman et al 1983 Raase et al 1983 Chao et al 1986 Von
Knorring et al 1986 Kerrich et al 1987 Treloar 1987b Treloar 1987a Morand 1990
Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993 Christofides et al 1994
Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Proenza et al
1999) Additionally kyanite and andalusite in the proximity of ophiolite complexes
incorporate up 5 Cr Wt (Cooper 1980 Raase et al 1983 Kerrich et al 1987)
Chapter 4 provides an in depth discussion of these Cr-bearing silicates as well as
demonstrating mineral compositions reported in the literature based on octahedral Cr and
Al Appendix 1 is a preliminary investigation devoted to estimating the thermodynamic
properties of kaumlmmererite These properties can then be incorporated into geochemical
modeling programs to begin accessing the stability of this common serpentinite mineral
over a range of temperatures and pressures
4
Chromium in Serpentine Soils
Evaluating Cr in serpentine soils requires an examination of both the mineralogy
and chemistry of the serpentinite and soil Measured Cr concentrations in serpentinites
and serpentine soils range from ~200 mg kg-1 to 6 Wt all of which is Cr(III) (Soane
and Saunder 1959 Proctor and Woodell 1971 Shewry and Peterson 1976 Jaffre 1980
Proctor et al 1980 Rabenhorst et al 1982 Schwertmann and Latham 1986 Brooks
1987 Gough et al 1989 Gambi 1992 Lee 1992 Morrey et al 1992) In Chapters 2
and 4 chromite and Cr-magnetite are identified as the most prevalent sources of Cr
responsible for Cr-enrichment in these soils Chromite is resistant to chemical
weathering and it has been observed to undergo incongruent dissolution conserving Cr in
serpentine soils Chromium is tightly bound in the matrix of the soil less than 1 of the
total Cr can is removed using standard soil extraction methods (10 hydrogen peroxide
(H2O2) solution 01 M potassium permanganate (KMnO4) solution10 M hydrofluoric
(HF) acid 1 M hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and 1
mM barium chloride (BaCl2) solution) due to the stability of chromite Although silicates
contain minor concentrations of Cr compared to chromite they are perhaps the main
contributors of Cr identified in plants (lt200 Cr mg kg-1) and soil solutions (lt50 microg L-1)
During soil-forming processes Cr released to soil solutions by the dissolution of Cr-
bearing minerals are sequestered into a host of structural and surface sites available in
authigenic Fe(III) oxides Fe(III) hydroxides and clay minerals (Schwertmann and
Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990 Gerth 1990 Gasser and
Dahlgren 1994 Ilton and Veblen 1994 Brigatti et al 2000 Ilton et al 2000) The
clay-sized fractions in soils at Jasper Ridge Biological Preserve CA contain Cr
concentrations as high as 1000 Cr mg kg-1
Chapter Summaries
A detailed field and analytical examination of Cr in serpentinites and serpentine
soils at Jasper Ridge is presented in Chapter 2 A majority of the Cr enrichment and
variability with depth in these soils is the result of small grains of chemically resistant
chromite from the serpentinite protolith Chromite identified in the soil undergoes
incongruent dissolution conserving Cr relative to chromite compositions identified in the
5
serpentinite bedrock Sequential extraction experiments using barium chloride (BaCl2)
hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and hydrofluoric (HF)
acid were performed on a chromite standard confirming that Cr-bearing spinels are
resistant to chemical weathering X-ray diffraction analyses of the clay-size fraction in
the serpentine soils identified Cr-clinochlore smectite vermiculite and serpentine with
the smectite being the most abundant mineral in the clay-size fraction Chromium(III)
was the only identified oxidation state of Cr in the serpentine soil solids and in
serpentinites at Jasper Ridge in the solid phase This chapter represents a comprehensive
site assessment identifying the nature and behavior of Cr in serpentine soils with respect
to the bedrock mineralogy and depth
Chromite an abundant Cr-enriched spinel is responsible for a majority of the Cr
enrichment identified in ultramafic rocks serpentinites and serpentine soils The release
and oxidation of Cr from chromite is a pathway to consider when evaluating the potential
of ultramafic rocks and their related sediments and soils as sources of Cr(VI) In
Chapter 3 experiments aimed at quantifying the release and oxidation of Cr from
chromite in the presence of birnessite are described Chromite is extremely resistant to
dissolution (lt8 Cr mg kg-1 is released) when immersed in solutions including 1 mM
barium chloride solution 10 M hydrofluoric acid 10 hydrogen peroxide and 1 M
hydroxylamine hydrochloride with 25 acetic acid Highly oxidizing solutions of 01 M
potassium permanganate increased the rate and release of Cr relative to the other
solutions listed above thus birnessite a high valence Mn oxide present in serpentine
soils was chosen to determine if Mn could accelerate Cr release and oxidation from
chromite The results presented in Chapter 3 demonstrate that birnessite increases the
rate of Cr(VI) observed in solution from chromite This production of Cr(VI) is
dependent on the chromite suspension density where birnessite surface are concentrations
are greater than 20 m2 L-1 between pHs 3 and 8 at ~25degC
A comprehensive review of the chemical characteristics of serpentine soils
including topics such as the mineralogy of serpentine soils soil solutions organic matter
and vegetation is provided within Chapter 4 Additionally chemical and spectroscopic
analyses were performed on serpentine soils from a) New Caledonia b) Eight Dollar
Mountain Selma OR c) Nickel Mountain Riddle OR d) Harvard Mine Jamestown
6
CA e) Jasper Ridge Biological Preserve Stanford CA and f) Pillikin Mine El Dorado
CA Serpentinites and serpentine soils associated with ophiolite complexes cover ~1 of
the earthrsquos exposed surface and they are correlated to modern and ancient convergent
boundaries The physical and chemical characteristics of serpentine soils are directly
related to the alteration and tectonic history of the serpentinite as well as the soil-forming
factors including the parent material climate biota topography and time Serpentine
soils have Cr concentrations exceeding 200 mg kg-1 however the concentration and the
availability of Cr in these soils are not correlated with the five soil-forming factors
(climate organic material parent material topography and time) or the soil chemistry
Although only Cr(III) was identified in the soil solids Cr(VI) has been identified in soil
solutions (lt30 microg L-1) from serpentine soils in New Caledonia and California Chromite
and Cr-magnetite grains are identified in the soils listed above and these minerals are
responsible for a majority of the Cr enrichment in serpentine soils
Kaumlmmererite is a Cr-chlorite associated with serpentinized ultramafic rocks such
as those collected at Jasper Ridge Harvard Mine and Pillikin Mine Although Cr
concentrations are typically 01-2 Wt in kaumlmmererite it is modally abundant in
serpentinites (up to 5 of the volume of the rock) and a potential source of Cr to be
released into soils and groundwater In Appendix A the standard molar thermodynamic
properties of the Cr-clinochlore end-member kaumlmmererite (Mg5Cr(AlSi3)O10(OH)8) is
estimated using established estimation algorithms This initial study provides the
foundation for submitting a future paper presenting the thermodynamic properties for
major Cr-bearing silicates including fuchsite (Cr-muscovite) uvarovite (Cr-garnet) and
tawmawite (Cr-epidote) and their metamorphic significance related to serpentinized
ultramafic rocks
7
CHAPTER 2
CHROMIUM GEOCHEMISTRY IN SERPENTINIZED ULTRAMAFIC ROCKS
AND SERPENTINE SOILS FROM THE FRANCISCAN COMPLEX OF
CALIFORNIA
ABSTRACT
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of California
produces serpentine soils containing high concentrations of Cr as well as other potentially
toxic elements including Ni Co and Mn Chromium concentrations in serpentine soils
from Jasper Ridge Biological Preserve in the Central Coast Range are as high as 4760
mg kg-1 nearly three times greater than the serpentinite protolith (~1800 mg kg-1)
Chromium-containing minerals within the bedrock include chlorite (~03 Cr Wt )
enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt ) magnetite
(82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate phase
containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
8
INTRODUCTION
Serpentine soils are formed from the weathering of ultramafic rocks and their
metamorphic equivalents (serpentinites Brooks 1987) Measured Cr concentrations in
these soils range as high as 80600 mg kg-1 and are enriched compared to non-serpentine
soils (7 to 221 mg kg-1 Proctor 1992 McBride 1994) The distribution and
concentration of Cr in serpentine soils worldwide is highly variable due to the chemical
composition and mineralogy of the parent material and soil forming factors including
climate biota topography precipitation and time (Hotz 1964 Rabenhorst et al 1982
Schwertmann and Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al
1989 Gough et al 1989 Kaupenjohann and Wilcke 1995) Potential Cr-bearing phases
in serpentine soils have been assessed using extraction experiments and these phases
include organic material silicates secondary oxides and spinels (Schwertmann and
Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994 Kaupenjohann and
Wilcke 1995 Becquer et al 2003) The weathering of these Cr-bearing phases over
geologic time and how these sources govern the distribution of Cr in serpentine soils have
not been evaluated in great detail
Chromium is a group VIB transition metal present in the environment as Cr(III) a
non-hazardous species and micronutrient andor as Cr(VI) a toxin to living cells and a
Class A human carcinogen by inhalation (Daugherty 1992 Cohen et al 1993 James et
al 1997) Chromium(III) is chemically immobile due to forming low solubility
(oxy)hydroxides and stable surface complexes whereas Cr(VI) is a weakly adsorbing
(oxy)anion and thus is mobile in most aquatic environments (McBride 1994 Fendorf
1995 Ball and Nordstrom 1998) Chromite and Cr-magnetite have been cited as
primary sources of Cr in serpentine soils (Rabenhorst et al 1982 Schwertmann and
Latham 1986 Alexander et al 1989 Gough et al 1989) Although Cr is primarily
found in the form of Cr(III) in both ultramafic rocks and serpentinites Cr(VI) has been
measured in serpentine soils in California and New Caledonia (Gough et al 1989
Becquer et al 2003) Due to the abundance of Cr and the potential for Cr(VI) formation
serpentine soils and serpentinites are potential sources of non-anthropogenic Cr
contamination
9
The objectives of this study are to determine the distribution and chemistry of Cr
in naturally Cr-enriched soils using an integrated field analytical experimental and
theoretical approach A wide array of techniques and methods including X-ray
absorption spectroscopy and chemical extractions were used to identify the nature of Cr
and its availability in serpentine soils Serpentinites and serpentine soils at Jasper Ridge
Biological Preserve located near Stanford CA were used to evaluate the geochemical
behavior of Cr These soils cover approximately 50000 m2 and contain high
concentrations of the trace elements Cr Ni and Mn The soils are dominantly Mollisols
with grassland vegetation and they are pedalogically similar to those associated with the
Central Coast Range of the Franciscan Complex in the western United States The main
serpentinite body at Jasper Ridge was exposed approximately 400000 years ago allowing
pedogenesis and Cr cycling to occur over a geologically relevant time period (Burgmann
et al 1994) These soils can then be used as a proxy for evaluating the long-term
behavior of Cr
BACKGROUND
Geologic Background of the Serpentinite at Jasper Ridge
Jasper Ridge Biological Preserve is located in the foothills of the Santa Cruz
Mountains in California and is part of the Central Coast Range of the Franciscan
Complex an Early Cretaceous accretionary meacutelange containing blocks of greywacke
greenstone blueschist and serpentinized ophiolite (Figure 21) Formation
metamorphism and emplacement of serpentinite at Jasper Ridge began approximately
130-150 million years ago during Jurassic subduction along the California continental
margin (Page and Tabor 1967) Thin wedges of ultramafic rock (peridotite) were
detached from the down-going slab of oceanic crust and incorporated into the subduction
meacutelange Subsequent serpentinization of the ultramafic rock in the down-going slab
resulted in the formation of serpentinites now exposed at Jasper Ridge
Rare blueschist blocks (1 to 5 meters in width) present in the serpentinite at Jasper
Ridge attest to the degree of metamorphism and the depths from which portions of the
tectonic meacutelange has been elevated Transformation of greenstone in the meacutelange into
blueschist required high pressures in excess of ~5-7 kbars (15-20 kilometers depth) and
10
temperatures of 200-300 degC (Ernst 1971) The low density of the serpentine (245-250 g
cm-3) allowed upward diapiric migration of the serpentinite body through the denser
enclosing tectonic meacutelange (265 g cm-3) Uplift of the present day Coast Range and
activity associated with the San Andreas fault system have shaped the current topography
and exposure of serpentinite at Jasper Ridge Serpentinite is currently present as a narrow
discontinuous band of pervasively sheared rock located along the ridge top (Figure 21)
The serpentinite is in fault contact with the Whiskey Hill Formation (Cretaceous turbidic
sandstone and mudstone) and tectonically interleaved with the Franciscan Complex
Serpentine Soils Characteristics and Review
Relative to other soil types serpentine soils are characterized by higher
concentrations of Cr Ni Co and Fe lower concentrations of plant nutrients such as Ca
K N and P lower CaMg ratios and characteristic flora and physical properties (Brooks
1987) The pH of serpentine soils ranges from ~4 up to 9 (Brooks 1987 Cole 1992
Proctor 1992) Serpentine soil solutions obtained using tension-free lysimetry have Cr
concentrations between 01 and 32 micromol L-1 in which a majority of the Cr is present as
colloidal material (Gasser and Dahlgren 1994) The inhabitability of these soils for most
plants has been attributed to the imbalance of Ca to Mg (Walker 1954) the deficiency of
plant nutrients (Turitzin 1991) and the elevated concentration of heavy metals
(Robinson et al 1935) however linking the lack of plant productivity of these soils to
toxic levels of Cr has not been fully resolved (Brooks 1987 Kruckeberg 1992) The
unique flora and fauna associated with these soils have been well-documented and Cr
concentrations in the plants range as high as 600 mg kg-1 (Brooks 1987 Gough et al
1989)
Chromium concentrations as a function of depth for serpentine soils in Maryland
(Rabenhorst et al 1982) Tehama County CA (Gough et al 1989) the Klamath
Mountains CA (Hotz 1964) and New Foundland Canada (Roberts 1992) are
illustrated in Figure 22 Chromium concentrations with respect to depth at each location
are highly variable demonstrating both maximums and minimums The concentration of
Cr varies significantly at each location due to the chemical composition of the parent
11
material and other soil-forming factors nevertheless Cr concentrations typically decrease
with depth towards the bedrock
A variety of chemical extractions on serpentine soils suggest a broad host of Cr-
containing soil phases including organic material silicates secondary oxides and spinels
(Schwertmann and Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994
Kaupenjohann and Wilcke 1995) The maximum concentration of Cr extracted from a
serpentine soil reported in the literature is 226 Wt Cr from the New Caledonia laterite
using a 03 M dithionite-citrate bicarbonate (DCB) solution (Schwertmann and Latham
1986) Li and Fendorf (2000) performed extractions on the serpentine soils at Jasper
Ridge using organic acids (citric and oxalic acids) in which a 100 microM citric acid solution
was only able to remove 23 Cr microgg-1 (08 Wt Cr) from a soil containing ~3000 Cr mg
kg-1 This suggests a majority of Cr is bound in phases that are not reactive with these
solvents
Sources of Chromium ProtolithBedrock
Chromium primarily occurs within oxide and silicate phases in ultramafic rocks
and serpentinites Chromite is a common primary (igneous) Cr-bearing spinel in mafic
and ultramafic rocks and often survives low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) Isomorphic substitution of
Al3+ Fe3+ and Ti4+ into the octahedral site and Mg2+ Ni2+ Zn2+ and Mn2+ into the
tetrahedral site is common (Sack and Ghiorso 1991) however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987 Pan and Fleet 1991 Burkhard 1993 Christofides et al 1994 Challis et
al 1995 Sanchez-Vizcaino et al 1995 Deer et al 1996 Mathiesen 1999) Pyroxenes
such as augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) are primary minerals in
ultramafic rocks that incorporate varying concentrations of Cr into their structures (Deer
et al 1996)
Serpentinization is associated with the partial oxidation of Fe(II) present in
olivine and pyroxene forming magnetite (Brooks 1987) Chromium(III) derived from
primary chromite and pyroxene may isomorphically substitute into the Fe(III) site of
magnetite due to size (octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge
12
similarities Hydrothermal and CO2 metasomatism of ultramafic rocks may produce Cr-
silicate minerals including fuchsitemariposite (Cr-mica lt17 Cr Wt ) uvarovite (Cr-
garnet lt20 Cr Wt ) tawmawite (Cr-epidote lt10 Cr Wt ) and kaumlmmererite (Cr-
chlorite lt6 Cr Wt Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Phillips et al 1980 Schreyer et al 1981 Max et al 1983 Kerrich et al 1987 Treloar
1987 Pan and Fleet 1991 Christofides et al 1994 Sanchez-Vizcaino et al 1995)
These minerals common in altered ultramafic rocks have varying concentrations of Cr
that isomorphically substitute for octahedral Al in the silicate structure due to similarities
in charge octahedral preference and octahedral radii (Al3+ = 061 Aring) Any of these
silicates and magnetite are potential sources for Cr in soils produced by the weathering of
serpentinite and ultramafic rocks
Chromium-Bearing Phases in Soils
Soil minerals specifically Fe(III) (oxy)hydroxides and clays provide a host of
structural and surface sites that may sequester Cr(III) released to soil solution by
weathering of Cr-bearing phases in serpentinite and ultramafic rocks Isomorphic
substitution of Cr(III) for Fe(III) in oxides (oxy)hydroxides and amorphous compounds
is expected based on crystallographic considerations synthesis experiments and analyses
of selective dissolution of Fe(III) phases (Schwertmann et al 1989 Gerth 1990)
Additionally Cr(III) readily sorbs on surfaces of Fe(III)-containing oxides such as
magnetite and hematite via inner-sphere complexation and forms low-solubility
precipitates at pH values greater than 4 (Charlet and Manceau 1992 Peterson 1996)
Octahedral substitution of Cr(III) for Al(III) in clays (smectites and vermiculites) is
possible based on size and charge similarities Sorption of Cr(III) on edge and interlayer
sites of clay minerals has also been observed by Ilton and Veblen (1994) Soil organic
matter (SOM) is another possible host for soil Cr in which Cr can be chelated into stable
organomineral and organometallic complexes (Kaupenjohann and Wilcke 1995 Brady
and Weil 1999)
13
MATERIALS AND METHODS
Location and Sample Collection
Jasper Ridge Biological Preserve is a microcosm of accreted terranes located in
the foothills of the Santa Cruz Mountains CA Figure 21a is a soils map reported by
Kashiwagi (1985) depicting the alluvial Tertiary sandstone and Franciscan Complex
soils present at Jasper Ridge The Franciscan Complex soils are divided into two
subgroups those derived from greenstone-graywacke and those from serpentine The
serpentine soils dominantly Mollisols and serpentinite bedrock are present as a narrow
discontinuous band located along the ridge top they do not receive sediment from the
adjacent non-serpentine soils Soils and rocks were collected at Jasper Ridge from 1999
to 2002 and sample locations are shown in Figure 21b A 500 meter transect consisting
of 10 sites extends across the length of the serpentine soil body At each site along the
transect soils and bedrock were collected with respect to depth Site 1 (Diablo clay) is
not derived from serpentinite but rather from chert and greenstone Additionally Site 1
is located down slope of the main serpentinite body and has likely received sediment
from the adjacent serpentine soil Rocks were also collected at several exposed outcrops
located in the serpentinite body
Site JR3 (Figure 21b) was chosen for a detailed geochemical study to identify
factors affecting the geochemistry of Cr in serpentine soils Soil at JR3 is a loamy
magnesic thermic Lithic Haploxeroll The thickness of the soil is approximately 50 cm
deep Grass dominantly Lolium multiflorum comprises 95 of the vegetation at this site
and trees and shrubs are not present
Mineral Identification in Rocks and Soils
Mineral identification was accomplished by electron probe microanalysis on an
automated JEOL 733A electron microprobe operated at 15 kV accelerating potential and
15 nA beam current Calibration was conducted using natural geologic standards Beam
width for analysis was 10 microm Raw counts were collected over 20 s and were constant
with time indicating that elemental drift was negligible Detection limits for probe
analyses are ~002 oxide Wt Additionally backscattering electron (BSE) images of
the rocks and soils were obtained using the microprobe
14
Carbon-coated petrographic thin sections of rock samples were used for
microprobe analyses Soil samples were dried in an oven at 60degC for 24 hours mounted
onto a glass slide using epoxy and heated to 100degC until the epoxy cured An initial cut
of each section using a petrographic saw was made followed by a second infusion of
epoxy to ensure total impregnation Finally each section was cut polished and carbon-
coated for electron microprobe measurements
Soil Chemistry
Soils were analyzed for pH electrical conductivity (EC (deciSiemensm))
particle size ammonium acetate extractable cations organic carbon total nitrogen and
elemental composition The pH was determined using a 11 mixture of soil to deionized
water Electrical conductivity (EC) was determined on 2 mL of a soil solution extract
Particle size of the soil was measured using the hydrometer method as described by Gee
and Bauder (1986) These analyses were performed at the Utah State University
Analytical Laboratory
Ammonium acetate was used to measure the concentration of extractable K Ca
Mg and Na within the soil The procedure involved weighing 2 g of soil and adding 25
mL of ammonium acetate (20 M NH4OAc) followed by mixing on an orbital shaker for
30 minutes The solution was filtered using a Whatman 42 filter paper (pore size 25
microm) Potassium Ca Mg and Na concentrations were analyzed using atomic absorption
spectrophotometry (AAS)
Organic carbon was measured using the Walkley-Black titration method (Mebius
1960) Approximately 05 g of sediment was weighed and reacted with 5 mL of 20 M
potassium chromate (K2Cr2O7) followed by the rapid addition of 10 mL concentrated
sulfuric (H2SO4) acid The bottle was cooled for 30 minutes at room temperature and
diluted with 100 mL of deionized water Next 030 mL of 0025 M ortho-
phenanthroline-ferrous solution was added and the solution titrated to the endpoint using
10M ferrous sulfate (FeSO4)
Total nitrogen of soil samples was obtained using an automated combustion
method as described by Sweeny (1989) Approximately 150 mg of air-dried and sieved
15
(40 mesh) soil was placed in a tin foil encapsulating cup A Leco total nitrogen analyzer
with resistance furnace was then used to determine total nitrogen
Major and minor element concentrations for soil and rock samples were obtained
by completely dissolving the samples using a mixture of hot concentrated nitric
perchloric and hydrofluoric acids Elemental concentrations were measured in the
supernatant using inductively coupled atomic emission spectroscopy (ICP-AES)
Standards were plusmn5 mg kg-1 of their known values These analyses were completed at
Chemex Laboratory (Vancouver Canada) Soil samples were dried and sieved (lt2mm)
before analysis Additionally total element concentrations for gt2mm sand silt and clay
size fractions were obtained for each sample collected at Site JR3
X-ray Diffraction and Clay-Size Mineral Identification
X-ray diffraction (XRD) analyses were conducted on the clay-size fraction
(lt2microm) from Site JR3 using a Siemens Diffractometer D5000 at the University of Idaho
The clay-size fraction was isolated using a suspension method described by Vaniman
(2001) Organic matter and calcium carbonate were removed from the clays using a
concentrated hypochlorite solution (Kunze and Dixon 1986) Crystalline and non-
crystalline iron oxides were removed using a citrate-sodium bicarbonate-sodium
dithionite (CBD) solution (Kunze and Dixon 1986) The remaining phyllosilicate
minerals were analyzed using a series of treatments including 1) magnesium-saturation
(2 M MgCl and 2 M MgAOC) and air drying 2) magnesium-saturation and glycolation
(1 glycerol-ethanol solution) 3) potassium-saturation (1 M KCl) and air drying and 4)
potassium (1 M KCl) saturation followed by heating to 500degC (Whittig and Allardice
1986) X-ray diffraction patterns were collected between 2θ values of 20-350deg and 20-
150deg with a generator potential of 30 kV a generator current of 22 mA (using CuKα
radiation) a Ni filter and a scan speed of 1degmin The software package JADE was used
for XRD data analysis and mineral identification
X-ray Absorption Spectroscopy
In-situ fluorescence yield X-ray absorption near-edge structure (XANES)
spectroscopy at the Cr K-edge (edge inflection point is 5989 eV for Cr(0)) was conducted
16
at the Stanford Synchrotron Radiation Laboratory (SSRL) Stanford CA using
beamlines 4-1 and 4-3 A Si(220) monochromator and a Ge-detector were used and
three scans were collected for each XANES analysis Soil and rock samples from Site
JR3 were prepared by removing the gt2mm fraction and crushed to a fine powder
Samples were loaded into a plastic sample holder and covered with Mylar tape
Elemental mapping of three samples were completed at SSRL on Beamline 11-2
using X-ray absorption spectroscopy (XAS) A Si(220) monochromator with a 6 microm
focusing capillary was used and fluorescence intensity was monitored with a Ge detector
Samples were prepared similar to those used for mineral identification using the electron
microprobe however the thickness of the sample was decreased to lt20 microm
Sequential Extraction Procedure
Sequential extraction experiments were completed on 2 g lt2 mm sieved soils
(Diablo clay from Site 1 at 10 cm depth and serpentine soil from Site 2 at 4 cm depth) in
three steps The exchangeable fraction was obtained using a barium chloride (1mM
BaCl2) solution for 1 h at 25 degC (Kunze and Dixon 1986) The Fe-crystalline oxide
fraction was obtained by reacting the solids with a hydroxylamine-hydrochlorideacetic
acid (10 M NH2OHsdotHCl in 25(vv) HOAc) solution for 6 h at 60 degC (Chester and
Hughes 1967) The silicate fraction was obtained by reacting the solids with
concentrated hydrofluoric acid (10 M HF) for 16 h at 25 degC Trace element
concentrations for each extraction were measured on the supernatant using inductively
coupled plasma optical emission spectrophotometry (ICP-OES) All analyses were
completed in triplicate Additionally sequential extraction experiments were completed
on 2g 106-250 mm sieved ultrasonically cleaned commercial grade chromite (MSDS
No 302-1 WHMIS Class D2A) using the same procedures listed above The
stoichiometric composition of chromite obtained using the electron microprobe is
(Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 Chromite was imaged after each sequential
extraction using a JEOL 5600LV environmental SEM equipped with an EDAX thin-
window energy dispersive X-ray detector
17
RESULTS
Serpentinite Mineralogy
The average serpentinite at Jasper Ridge is composed of serpentine (mainly
lizardite and antigorite 70) chlorite (15) talc (10) magnetite (4) and chromite
(1) based on petrographic and microprobe observations of five samples (JR3E JR6E 5
10 and JR3) however brucite has been reported as a major constituent in related
Franciscan Complex serpentinites (Hostetle et al 1966) Other minerals identified
include olivine augite and enstatite Although the serpentinite mineralogy is relatively
simple the modes and textures of the minerals present in the rock vary significantly from
site to site In Table 21 whole rock serpentinite compositions from Sites JR1 JR3E
JR6E 2 5 10 and JR3 are listed The average Cr concentration of the serpentinites is
~1700 mg kg-1 (tab 1) Iron (~7) Mg (~11) Ni (~3000 mg kg-1) and Co (~100 mg
kg-1) concentrations are also elevated in the serpentinite relative to the chert and
sandstone bedrock adjacent the main serpentinite body (Figure 21b and Table 21)
A range of rock and mineral textures are petrographically evident in the
serpentinite at Jasper Ridge (Figure 23) No foliations or lineations are observed in the
serpentinite (Figure 23a) obtained from the bottom of the soil pit excavated at Site JR3
(Figure 21b) Subhedral chromite and magnetite grains are randomly interspersed
amongst the serpentine matrix (Figure 23a) In serpentinite obtained in outcrop at Site
JR5E (Fig 21b) a lsquocrocodile skinrsquo texture is produced by thin irregular veins of fine-
grained subhedral magnetite separating serpentine grains (Figure 23b) In Figure 23c
numerous crosscutting veins of chlorite occur in orthopyroxene and define the foliation of
the rock in outcrop at Site JR3E (Figure 21b) The numerous crosscutting microscopic
veins random foliation patterns (or the lack thereof) and subhedral oxide grains suggest
the Jasper Ridge serpentinite has undergone extensive physical and chemical alteration
typical of most serpentinites
Microprobe analyses of minerals in the serpentinite demonstrate that the main Cr-
bearing phases are chlorite enstatite augite Mg-Al chromite Cr-magnetite and a
chromite-silicate mixture (defined in the following paragraph) Selected compositions
for these minerals are listed in order of their Cr concentrations in Table 22a Chlorite
contains the least Cr whereas the chromite-silicate mixture is the most enriched with
18
respect to Cr Figure 24 shows correlations of Cr content relative to Al Fe Mg and Si in
order to provide insight into the serpentinite mineralogy and to aid in evaluating trends of
Cr substitution
A fine-grained mixture containing spinel (chromite andor Cr-magnetite) and an
unidentifiable Cr-silicate contains the highest Cr concentration in the serpentinite at
Jasper Ridge (Table 22a shown in Figure 25a and 25b) This fine-grained mixture will
be referred to here as the chromite-silicate mixture (CSM) with chemical properties
similar to those in Table 22a and Figure 24 CSM is common around the rims of
protolith chromite grains and a number of the larger (~05 mm) Cr-magnetite grains in
the serpentinite The microcrystalline phase responsible for contributing Si to these
analyses is possibly chlorite due to similarities of false colors generated by backscattering
electron (BSE) images and chlorite commonly rimming CSM Microprobe analyses of
CSM listed in Tables 22 and 23 are considered as mixtures of several mineral phases
due to the fine-grained texture (lt5 microm) Chromium concentrations range from 12 to 16
Cr Wt in this phase CSM is chemically distinct from the chromite and magnetite as
shown in Figure 24
The chemical compositions of protolith chromite exhibit little to no variation in
any of the serpentinites collected at Sites 2 JR3E JR4E JR5E JR6E JR7E and JR3
(Table 22a and Figure 24) Chromite contains low Fe (~9 Wt ) high Mg (~9 Wt
) and high Al (~26 Wt ) compared to Cr-magnetite The average Cr concentration
for chromite is ~12 Cr Wt Additionally chromite grains are irregular and subhedral
suggesting that they have undergone significant alteration during metamorphism
Chromium concentrations in the magnetite vary depending on the proximity and
degree of alteration of the chromite (Table 22a and Figure 24) Magnetite grains with
less than 2 Cr Wt are common in veinlets in serpentinite not formed adjacent
chromite whereas magnetite with 8 to 12 Cr Wt were formed proximal or adjacent
chromite Chromium-enriched magnetite is also referred to as ferrite-chromite in the
literature (Spangenberg 1943 Burkhard 1993)
Chlorite and pyroxenes contain minor concentrations of Cr (lt1 Wt Table
22a) Aluminum Fe and Mg concentrations varied in the pyroxenes however the
majority of chlorites are relatively similar in composition (Figure 24) Chlorites are
19
more abundant in the serpentinite compared to the pyroxenes and both minerals were
subhedral and randomly distributed over the entire sampling area (Figure 21b)
Backscattering electron (BSE) images of what appeared to be a petrographically
opaque single chromite grain in Figure 23a demonstrates chromite has been
metasomatically altered resulting in the formation of Cr-magnetite Cr-chlorite and CSM
(Figure 25a) Represented compositions for these minerals (indicated by letters in Figure
25a) are listed in Table 23 Magnifying the chromite and CSM (Figure 25a)
demonstrates the fine-grained texture of CSM and an alteration rim (transition zone) is
also seen along the chromite edge (Fig 25b) This alteration rim is depleted in Cr with
respect to the chromite as shown in Figure 25b
In Figures 23a and 25c Ni was identified in three phases in the bedrock
including olivine (F 03 Ni Wt ) serpentine (G and H 03 and 0 Ni Wt ) and Ni-
Fe (I sim55 Ni Wt ) alloy (Figure 25c and 25d) their compositions are indicated by
letters and listed in Table 23 Alteration of the olivine resulted in the formation of lt10
microm Ni-Fe alloys shown in Figure 25c These Ni-Fe alloys form a corona surrounding
the remaining olivine and the illustration in Figure 25d highlights the minerals and
chemical changes
Chromite and Cr-magnetite alteration leading to the paragenesis of CSM and Cr-
chlorite is common in the serpentinites at Jasper Ridge The petrographic image in
Figure 26a and its corresponding BSE image in Figure 26b illustrate Cr-magnetite
(originally protolith chromite) partially altering to CSM and Cr-chlorite The chemical
compositions of each mineral (indicated by letters in Figure 26b) are listed in Table 23
CSM appears to marginally and metasomatically replace Cr-magnetite and is enriched by
~2 Cr Wt Chromium-chlorite is present in microcracks in Cr-magnetite and
encompasses the entire Cr-magnetite and CSM grain Serpentine minerals adjacent the
Cr-chlorite contains no measurable Cr
Distribution of Chromium in Serpentine Soils
Serpentine soils and underlying bedrock were collected at ten sites along a 500
meter transect extending the length of the main serpentine body as shown in Figure 21b
20
The range of Cr concentrations at each site (center) and Cr concentrations with respect to
depth (periphery) are shown in Figure 27 and soil analyses are listed in Table 24
Bedrock Cr concentrations for Sites 2 5 and 10 are also shown in these plots (Fig 27)
Chromium enrichment does not appear to be consistently found at a particular depth or
soil horizon for any of the sites Sites 3 9 and 10 demonstrate a Cr maximum at
approximately 006 m however Sites 1 2 5 and 6 exhibit a Cr minimum at roughly the
same depth The highest soil Cr concentration in the soil along the transect is 4760 mg
kg-1 at Site 3 at 006 meters depth The lowest soil Cr concentration is 1725 mg kg-1 at
Site 10 at 0085 meters depth The average bedrock Cr concentration is ~2000 mg kg-1
(Figure 27 Table 24) It is important to note Site 1 consists of Diablo clay not
serpentine soil however this site is located adjacent and down-slope of the main
serpentinite body From these observations it is apparent that there are no systematic
trends in soil Cr concentrations with respect to the position of the site and depth and soil
samples often contain higher concentrations of Cr with respect to their corresponding
bedrock and to the average Cr concentration of ultramafic rocks (1600 mg kg-1 Green
1972)
Trace metals including Ni Mn and Co were also evaluated in the serpentine soil
for Sites 1 through 10 and the results are listed in Table 24 Nickel concentrations range
from 1590 to 3960 mg kg-1 and do not directly correspond to the Cr fluxes with respect
to depth (Table 24) Nickel concentrations are less than Cr from Sites 1 to 4 From Sites
5 to 10 Ni concentrations exceed those of Cr averaging ~3600 mg kg-1 Unlike Cr Ni
concentrations are equivalent or greater in the bedrock (~3300 mg kg-1) as compared to
the soil (~3200 mg kg-1) Soil Mn concentrations at each site are relatively uniform at
~1800 mg kg-1 with little to no variation with depth (Table 24) Cobalt concentrations
do not exceed 238 mg kg-1 (Table 24) and are relatively uniform (~200 mg kg-1) with
respect to location along the transverse and depth at each site
Major elements including Al Fe Ca and Mg are listed for the serpentine soils
along the transect with respect to depth in Table 24 Aluminum Fe Ca and Mg
concentrations in the soils at each site are similar with exception of Site 1 (Diablo clay)
Aluminum concentrations in the Diablo clay are ~46 whereas the serpentine soils
contain less than 2 Al (Table 24) Iron and Mg concentrations are also low lt65 and
21
lt411 respectively in the Diablo clay compared to Fe (gt8) and Mg (gt6)
concentrations in the serpentine soil (Table 24) Calcium concentrations are low (lt1)
for all the serpentine soils and even the Diablo clay (Table 24) With a broad
understanding of the chemical characteristics of the serpentine soil at Jasper Ridge Site
JR3 (Figure 21b) was chosen for a detailed chemical study to examine all the factors
governing the chemistry of Cr in the soil
Serpentine Soil Chemistry and Mineralogy at Site JR3
Soil chemistry
Chemical properties (pH electrical conductivity organic matter nitrogen and
ammonium acetate extractable cations) and textures for four soil samples from Site JR3
are listed in Table 25 The pH of the soil is near neutral (pH from 671 to 698) and
increases slightly with respect to depth Electrical conductivities (EC) in the soil ranges
from 03 to 09 dSM a range indicative of relatively few dissolved salts andor major
dissolved inorganic solutes The organic carbon content of the soil is 32 at the surface
and decreases with depth to 10 near the bedrock Total nitrogen for the samples ranged
from 003 to 064 Ammonium acetate extractable Ca K Mg and Na are listed in
Table 25 The extractable concentration of both Ca and K decrease with depth whereas
there is no apparent trend associated with the concentrations of Mg and Na The
serpentine soil textures are dominantly clay loams however the texture is a sandy clay
loam near the bedrock
Whole soil samples and size fractions including gt2 mm sand (2-002 mm) silt
(002-0002 mm) and clay (lt2 microm) size fractions with respect to depth were chemically
digested and analyzed for total trace metal concentrations shown in Figure 28 and listed
in Table 26 to determine if the abundance of Cr was linked to a particular size fraction
(the percent of sand silt and clay size fractions are reported in Table 25) Additionally
Ni Mn and Co concentrations with respect to depth and size fraction are also presented in
Figure 28 and Table 26 Total and gt2mm size fraction Cr concentrations are similar at
the top and bottom of the soil profile ranging from 3297 to 5976 Cr mg kg-1 Chromium
concentrations exceed detections limits (gt10000 mg kg-1) for the sand-size fraction It is
important to note that the sand-size fraction corresponds to the size of most chromite
22
grains observed in the serpentinite Chromium in the clay fractions are low (~1000 mg
kg-1) compared to the other size fractions however these concentrations are still enriched
compared to non-serpentine soils
In Figure 28 and Table 26 Ni concentrations are enriched in the clay-size
fraction with values ranging from 2840 to 2920 mg kg-1 The silt-size fractions have the
lowest Ni concentrations (~1800 mg kg-1) Manganese concentrations are slightly
elevated relative to total Mn in the sand-size fraction with values ranging from 1728 to
2701 mg kg-1 The clay-size fraction is the least Mn enriched size-fraction with
concentrations at ~1200 mg kg-1 Cobalt concentrations are elevated in the sand-size
fraction with values ranging from 159 to 241 mg kg-1 and concentrations in the gt2mm
silt and clay-size fractions are similar to the total (~150 mg kg-1)
Soil Cr-bearing minerals
Chromium-bearing minerals in the soil identified using the electron microprobe
include chromite Cr-magnetite and CSM Chromium concentrations in soil chromite
Cr-magnetite and CSM are graphically shown in Figure 29 as a function of Al Fe Mg
Si and Al+Fe in Wt Representative analyses of the Cr-bearing soil minerals are listed
in Table 22b Compositions of soil chromite in Figure 29 are chemically distinct from
those identified in the protolith in Figure 24 Chromium concentrations in soil chromites
range from 22 to 37 Cr Wt containing minor concentrations of Al and Mg Values for
chromite in the serpentinite bedrock (Table 22a and Figure 24) and the FeCr2O4 end-
member are also shown in Figure 29 The dashed lines in Figure 29 connect Jasper
Ridge bedrock chromite (Figure 24) and stoichiometric FeCr2O4 where it can be seen
that compositions of soil chromite lie close to these tie lines Overall soil chromite
demonstrates a decrease in Al and Mg and an increase in Fe relative to Cr compared with
protolith chromite Soil Cr-magnetites contain between 7 and 20 Cr Wt and are
enriched in Cr compared to Cr-magnetites identified in the bedrock (Figure 24) Cr-
magnetite however reveals a decrease in Fe relative to Cr Minor concentrations of Si
(lt2 Wt ) in the soil chromite and Cr-magnetite may represent slightly altered rims or
contamination from sample preparation Chromium concentrations in CSM maintain a
consistent Cr concentration of ~11 Cr Wt similar to those found in the bedrock (~14
23
Cr Wt ) Soil CSMs with compositions between 8 and 12 Mg Wt are more than
likely associated with analyses containing more microcrystalline chlorite compared to
soil CSMs with less than 2 Mg Wt Soil CSM appears be less enriched with Si (lt14
Wt ) relative to those identified in the serpentinite (15 to 27 Wt )
The size of the Cr-bearing minerals varied significantly for each sample ranging
from 10 microm up to 2 mm The majority of chromite grains in the soil correspond to the
sand-size fraction and also vary from euhedral to subhedral with no discernable pattern
The sum of cations based on four oxygens for the chromite and Cr-magnetite analyses
range between 297 and 301 indicating these minerals are spinels not (oxy)hydroxides
Compositions obtained for chromite and Cr-magnetite may represent alteration rims
where weathering of the mineral in the subsurface environment has been most active
BSE images of chromite and Cr-magnetite were not able to definitively identify whether
these analyses in fact represent alteration rims
Mineralogy of the Clay-Size Fraction
The clay-size fraction in the serpentine soil at Site JR3 was evaluated with respect
to depth using XRD and specific chemical and heat treatments The d-spacings (Aring) and
their relative intensities for each soil and treatment with respect to depth are listed in
Table 27 Figure 210 shows the XRD analyses for clay sample JR3515 (depth 5-15
cm) and these analyses are almost identical to the other clays examined at Site JR3
Additionally Table 28 from Whittig and Allardice (1986) is provided to interpret the X-
ray diffraction analyses in Table 27 For example a d-spacing of ~14Aring is present in all
the treatments indicating the presence of chlorite in the clay-sized fraction Using an
XRD analysis program the clay-size fraction was determined to be composed of
smectite vermiculite serpentine minerals (lizardite and antigorite) and chlorite
(clinochlore and Cr-clinochlore) Smectite is the most abundant mineral in the clay-size
fraction whereas Cr-chlorite is the least abundant Overall the d-spacings and relative
intensities are similar despite the treatments and changes in depth (Table 27) This
indicates that the mineralogy of the clay-size fraction does not vary significantly in the
soil column
24
Chromium concentrations in the clay-size fractions are ~1010 mg kg-1 (Table
26) Smectite and vermiculite are possible Cr-bearing clay minerals in which isomorphic
substitution of Cr(III) could occur in the dioctahedral sheet Serpentine minerals
(lizardite and antigorite) are not likely hosts of Cr based on crystallographic
considerations and the absence of Cr in analyzed serpentine minerals in the bedrock
(Table 23) Clinochlore may also be a host for Cr in the inner layer octahedral sheet
Chromium-clinochlore is the only Cr-containing mineral identified in the clay-size
fraction using XRD and Cr-clinochlore (Cr-chlorite) is also an abundant mineral
identified in serpentinite protolith (Tables 22 and 23 Figure 26)
XAS of chromium in relation to depth
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in serpentine soils and bedrock at Site JR3 The XANES (X-ray absorption near-edge
structure) spectra demonstrate the absence of Cr(VI) in the soil column at Site JR3
(Figure 211) based on the lack of a pre-edge feature The detection limit for Cr(VI)
using this method is approximately ~20 mg kg-1 Ultimately XAS analyses demonstrate
Cr(VI) is not detected in any of the soil solids
X-ray microprobe analyses of soil and rock samples (0-5cm depth 5-10cm depth
and serpentinite) at Site JR3 were processed to map the relative concentrations of Fe and
Cr in which areas of elemental enrichment are identified by a violet color areas of
depletion by a blue color (Figure 212) Areas of Cr and Fe enrichment in Figure 212 are
localized with well-defined boundaries corresponding to grain edges and these elements
are not correlated in a well-defined trend Minerals assigned to these areas of Cr and Fe
enrichment were made based on petrographic and microprobe observations on the same
thin section For example high Fe and no Cr is magnetite whereas high Fe and high Cr
is either chromite or Cr-magnetite Chromite magnetite and Cr-magnetite appear to be
present in the serpentine soil at a depth between 0 to 5 cm (Figure 212A) In Figure
212b CSM is a possible mineral detected in the soil (depth 5-10 cm) and the presence
of magnetite is clearly seen due to the localized enrichment of Fe and no Cr Iron and
Cr spectra for the serpentinite were much more heterogeneous as compared to the soil
samples and Cr concentrations appear be more scattered (Figure 212c) These X-ray
25
microprobe analyses provide a complimentary method for examining Cr geochemistry
(including Cr(III) and Cr(VI)) in naturally occurring soil and rock solids however
Cr(III) is the only detected valence state in serpentine soils and rocks at Jasper Ridge
Sequential Extractions
Sequential extractions were completed on 1) Diablo clay (Inceptisol) from Site 1
(Figure 21b) at a depth of 10 cm 2) serpentine soil (Mollisol) from Site 2 (Figure 21b) at
a depth of 4 cm and 3) commercial grade chromite (106-250 microm) and these results are
shown in Figure 213 and listed in Table 29 Chromium extracted from the
exchangeable fraction is minimal (lt01 mg kg-1) for all the samples (number 1 in Table
29 Fig 213) Minor Cr concentrations are associated with the Fe-crystalline oxide
fraction in the Diablo clay (19 Cr mg kg-1) and serpentine soil (78 Cr mg kg-1) whereas
chromite only released 05 Cr mg kg-1 (number 2 in Table 29 Figure 213) Increased
concentrations of Ni and Fe are present in the Fe-oxide fraction of the Diablo clay (161
Ni mg kg-1 and 4474 Fe mg kg-1) and serpentine soil (5138 Ni mg kg-1 and 5138 Fe mg
kg-1 Table 29) Chromium extracted from the Diablo clay serpentine soil and chromite
in the silicate fraction is 14 31 and 14 Cr mg kg-1 respectively (number 3 in Table 29
Figure 213) The sum of extracted Cr Fe Ni and Mn concentrations (number 4 in
Table 29) are low with respect to total metals (number 5 in Table 29) In fact only 07
12 and 00 Cr was removed from the Diablo clay serpentine soil and chromite
respectively (number 6 in Table 29)
Chromite was extremely resistant to all the extractions and less than 001 of the
total Cr was removed (Table 29) Scanning electron microprobe (SEM) images of the
chromite after each extraction are shown in Figure 214 Chromium concentrations
released from the chromite (Table 29 and Figure 214a) increased with each sequential
extraction experiment starting with the exchangeable fraction (Figure 214b) progressing
into the Fe-crystalline oxide fraction (Figure 214c) and ending with the silicate fraction
(Figure 214d) The surface of the chromite became smoother after each extraction and
crystal imperfections in defect sights located along the edges and corners progressively
decreased (Figure 214) Chromite does not appear to be an easily accessible source of
Cr based on these extraction experiments
26
DISCUSSION
Serpentinite at Jasper Ridge has undergone significant physical and chemical
alteration since its formation and emplacement off the California continental margin
Despite being subjected to high pressures (~5-7 kbars) and elevated temperatures (200-
300degC) trace element concentrations within the serpentinite remain elevated Jasper
Ridge serpentinites are enriched with Cr Co Fe Mg Mn and Ni compared to chert and
sandstone located adjacent the main serpentinite body (Table 21 and Figure 21b) Over
the past 400000 years Cr-bearing minerals in the serpentinite have been weathering and
accumulating in the soil During this time Cr concentrations in the soil have
progressively increased relative to Cr-concentrations in the serpentinite protolith This Cr
enrichment is directly linked to specific weathering-resistant minerals identified in the
soil
Chromium in Serpentinites
Three minerals in the bedrock at Jasper Ridge contain appreciable concentrations
of Cr Cr-magnetite (82-103 Cr Wt ) chromite (~108 Cr Wt ) and CSM (~133
Cr Wt ) Cr-magnetite and CSM were derived from the alteration of primary igneous
chromite CSM or Cr-spinels with high concentrations of Si have been noted by
(Burkhard 1993) as a Cr-spinel containing a possible microscopic silicate phase Our
analyses support this result in which the microcrystalline silicate is probably Cr-chlorite
due to its common occurrence with CSM (Figure 26) Definitive identification of this
microcrystalline mixture was not possible due to size of the crystals relative to the spot
size of the electron microprobe Chlorite and pyroxene contain minor concentrations of
Cr (~1 Wt ) and these were the only identified Cr-containing silicates in the
serpentinite Minor concentrations of Cr observed in some chlorites are directly related
to the metasomatic alteration of chromite and Cr-magnetite (Figure 26) Other common
metasomatically-derived Cr-bearing silicates such as fuchsitemariposite and uvarovite
are not present possibly due to the lack of aqueous K+ and Ca+2 in the alteration fluids
Overall the chemical compositions of these protolith minerals (Figure 24) demonstrate
only minor chemical variations allowing comparative analysis with Cr-bearing soil
minerals
27
Chromium in Serpentine Soils
Cr-bearing soil minerals identified using the electron microprobe include
chromite Cr-magnetite and CSM and these phases are chemically distinct from those in
the bedrock (Figure 24 and 29) The highest concentration of Cr (41 Wt ) in the soil
is in chromite with a composition approaching the FeCr2O4 end-member This represents
a ~80 increase in the Cr concentration relative to bedrock chromite (see Figure 215)
Cr-magnetite also demonstrates Cr enrichment (up to 20 Cr Wt ) with respect to
magnetite identified in the bedrock (~10 Cr Wt ) (Figures 24 and 29) The textures of
the soil Cr-bearing spinels exhibit sharp irregular edges possibly due to extensive
weathering CSM in the soil is similar to those found in the bedrock with Cr
concentrations ranging from 11 to 14 Cr Wt however Si concentrations in soil CSMs
(8 to 14 Wt ) are noticeably less than those measured in the serpentinite (15 to 20 Wt
) The weathering of CSM might preferentially release Si relative to other elements in
its structure Cr-silicates (chlorite and pyroxenes) were not identified in the soil possibly
due to complete weathering of these phases or they were too fine-grained for microprobe
analyses however Cr-clinochlore was identified in the clay-size fraction using XRD
Elemental maps shown in Figure 212 corroborate electron microprobe observations by
revealing well-defined zones of Cr and Fe enrichment in both the soils and serpentinite
These areas of Cr and Fe enrichment are either chromite Cr-magnetite CSM or
magnetite Overall these maps reveal Cr is not homogeneously distributed in the soil
confirming Cr is in specific mineral phases Additionally the observed abundance of
chromite Cr-magnetite and CSM in the soils indicates that these minerals are probable
sources for a majority of the Cr enrichment identified in the soils
Soil Spinels
Soil chromite and Cr-magnetite appear to have undergone chemical alteration
enriching the minerals with respect to Cr In Figure 215 compositions of bedrock
chromite and end-member FeCr2O4 are shown together with tie-lines like those in figure
8 connecting both minerals Reaction progress in Figure 215 is based on the assumption
that the serpentinite chromite (0 reaction progress) is proceeding toward FeCr2O4
(100 reaction progress) in the serpentine soil Soil chromites from Jasper Ridge are also
28
plotted in Figure 215 where Cr concentrations are assumed to vary linearly between the
bedrock chromite and chromite end-member which defines the reaction progress of the
mineral These tie-lines correlate extremely well to the chemical composition of
chromites identified in the soil For example chromite containing ~30 Cr Wt will
have ~15 Fe Wt ~13 Al Wt and ~5 Mg Wt similar to soil chromite in Figure 29
The retention of Cr in chromite shown in Figure 215 has significant implications towards
the availability and cycling of Cr in serpentine soils
One possible mechanism for the chromite composition trends illustrated in
Figures 29 and 215 is incongruent dissolution where Cr is retained relative to other
elements Sequential extraction experiments (Figure 213 and Table 29) support this
idea in which Fe is more easily released than Cr (1096 Fe mg kg-1 to 19 Cr mg kg-1)
from chromite Unfortunately Mg and Al were not monitored during the experiments
From these sequential extraction experiments and chromite analyses (Figure 215) Al
and Mg have been preferentially released from the chromite structure compared to Fe and
Cr Additionally the alteration of chromite in the serpentinite and the soils are analogous
to one another Chromite in the serpentinite retains a majority of its Cr compared to Mg
and Al when being altered to CSM by fluids at elevated pressures and temperatures
(Figure 26 and Table 22) Under ambient soil conditions chromite is once again
retaining Cr relative to Mg and Al causing what appears to be an increase in Cr
concentration Chromium is not easily released from chromite whether under high
pressures and temperatures or in the soil
Crystal field stabilization energies (CFSE) and octahedral site preference energy
parameters (OSPE) provide a means to explain the incongruent dissolution or Cr
preservation in the Cr-spinels CFSE is a measure of the net energy of stabilization
gained by a metal ions nonbonding d electrons as a result of complex formation This
measure also indirectly describes the energy required to remove an ion from a crystal
structure The octahedral CFSE for Cr(III) is ndash2247 kJmole and the OSPE (the
difference between the tetrahedral and octahedral CFSE) is ndash1578 kJmole (Dunitz and
Orgel 1957 McClure 1957) These values are the most negative for any transition
element and indicating the presence of Cr(III) in an oxide increases the crystalrsquos stability
compared to an analogous crystal lacking Cr(III) Chromium(III) would be among the
29
last ions to removed from the spinel structure compared to other transition elements such
as Fe2+and Mn2+
Soil Chemistry related to Chromium Availability
In Table 25 the chemical properties of the soil at Site JR3 do not appear to be a
factor when evaluating Cr in the soil The pH of the soil did not vary considerably with
depth Additionally the near neutral pH would not have a significant impact accelerating
Cr-bearing mineral dissolution Organic matter within the soil was elevated despite the
low plant productivity of the serpentine soil The role of organic matter could be a
significant factor such as the production of organic acids promoting the dissolution of the
Cr-bearing minerals however Cr concentrations near the surface (where organic matter
was the greatest) are not significantly greater or less than those deeper in the soil profile
The relative concentrations of ammonium acetate extractable Ca K Mg and Na are
characteristic of serpentine soils and did not provide any new insight Finally Cr(III)
was the only identifiable valence state of Cr based on X-ray absorption spectroscopy
(Figure 211) which is what should be expected based on the Cr-bearing minerals and the
pH of the soil
Extraction experiments summarized in Table 29 and Figure 213 demonstrate
~99 of the Cr in a serpentine soil is retained in highly resistant phases at Jasper Ridge
The chromite standard used in these experiments contributed negligible concentrations of
Cr (lt2 mg kg-1 total) for all of the sequential extraction experiments supporting earlier
evidence that Cr-spinels are retaining Cr and are potentially responsible for a majority of
the total Cr enrichment in the serpentine soil At Jasper Ridge Cr extracted from the
serpentine soil (31 mg kg-1) is likely the result of the Cr-silicates compared to the Cr-
spinels Low Cr-containing minerals (possibly chlorite enstatite augite and clay
minerals) appear to be the most probably source of Cr to soil solutions and vegetation of
the serpentine soil Approximately 30 of the serpentine soil is composed of clay-sized
minerals and this size-fraction has ~1000 Cr mg kg-1 The mineral responsible for Cr
enrichment in the clay-size fraction is more than likely Cr-chlorite identified using XRD
Chromium-chlorite in the soils suggests that this mineral is potentially the largest silicate
contributor of Cr in the serpentine soil
30
Interpreting the enrichment and behavior of Cr in serpentine soils such as in
Figure 27 can be accomplished based on the abundance Cr-spinels observed in the soil
Small grains of chromite Cr-magnetite and even CSM in the soil have the potential to
appreciably impact Cr concentrations Table 210 demonstrates the chromite masses
required to change the Cr concentration in a 1 kg sample of soil The highest Cr
concentration at Jasper Ridge was identified at Site 3 with a measured value of 4760 mg
kg-1 which means only ~1 of the soil sample (Table 210) needs to contain chromite
Additionally a minimal amount (lt011 of the sample as shown in Table 210) of
chromite is necessary to change the concentration by 500 mg kg-1 which could easily
cause the sharp Cr increases and decreases noted in Figure 27 This agrees with our
observations of chromite abundance in the serpentine soils Additionally a majority of
the chromite grains identified in the bedrock are between 2 and 002 mm which
correspond directly to the elevated Cr concentrations (gt10000 mg kg-1) in the sand-size
fraction of the serpentine soil (Table 26) Even the ~15000 Cr mg kg-1 values in soils
measured by Hotz (1964) in Figure 22 only require ~2 of the sample of the sample to
be chromite A few grains of chromite possess the ability to significantly impact
measured Cr concentrations in soil samples
Chromium and Nickel in Serpentine Soils
Chromium and Ni have often been correlated to one another due to their common
enrichment in serpentine soils (Hotz 1964 Rabenhorst et al 1982 Schwertmann and
Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al 1989 Kaupenjohann
and Wilcke 1995) A majority of the Ni in the Jasper Ridge serpentine soil originated
from olivine (Figure 25c and 25d Table 23) Chemical alteration of olivine during
serpentinization resulted in the mineral serpentine (Coleman and Jove 1992) in which
minor concentrations of Ni (lt037 NiO Wt ) substituted into its structure Continued
alteration under highly reducing conditions and the incompatibility of Ni in the serpentine
structure resulted in the formation of Ni-Fe alloy (Figure 25c) This type of alteration is
observed in many of the rocks at Jasper Ridge where the Ni-Fe alloys are ~10 microm in size
Olivine in the rocks and soil appear to be less resistant to weathering compared to
chromite therefore Ni is possibly more available to bacteria and vegetation in serpentine
31
soils Sequential extractions (Figure 213 Table 29) support these observations in which
a greater percentage of total Ni was extractable compared to the Cr in the soils Although
Cr and Ni are enriched in serpentine soils the availability and behavior of these two
elements are significantly different
CONCLUDING REMARKS
Chromium is predominantly present in primary chromite Cr-magnetite and a
chromite-silicate mixture (CSM) in both the bedrock and soil despite weathering over
geologic time Sharp variations in soil Cr concentrations with respect to depth are caused
by the presence or absence of chromite Cr-magnetite andor CSM Chromium-spinels in
the soil have undergone incongruent dissolution possibly conserving Cr(III) with respect
to Al3+ and Mg2+ Chromite is highly resistant to weathering and it is more than likely
not contributing bioavailable Cr Although Cr-spinels are a major source of Cr Cr-
silicates in the protolith (chlorite enstatite and augite) and soil (chlorite
montmorillonite and vermiculite) afford the most accessible Cr source identified Based
on our observations and analyses chlorite is the most likely Cr source of these silicate
minerals Finally Cr(VI) was not detected in any of the soil and rock samples
ACKNOWLEDGEMENTS
We thank Jasper Ridge Biological Preserve for allowing us the opportunity to
examine the serpentine soils Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund NSF Grant EAR-9902859 (DKB) GSA Graduate
Student Research Grants Office of Teaching and Learning (SF) at Stanford University is
gratefully acknowledged We appreciate the technical support from the staff of the
Stanford Synchrotron Radiation Laboratory from Bob Jones for assistance in microprobe
analyses of the serpentinites and serpentine soils and from Ben Bostick Travis Horton
and Theresa Barber for field and lab assistance
32
Table 21 Whole rock elemental concentrations of serpentinites (Figure 21b) chert and sandstone
-----------------------Location of Serpentinitesdagger----------------------- Elemental Concentrations JR1 JR3E JR6E Site 2 Site 5 Site 10 JR3
ChertDagger SandstoneDagger
Al 02 08 03 027 22 107 024 079 027 Ca 026 007 005 003 046 013 003 001 lt001 Co mg kg-1 101 99 100 99 155 217 97 1 lt1 Cr mg kg-1 580 2280 1510 1840 1945 2090 1465 10 lt1 Fe 303 71 609 581 76 1106 512 041 024 K 001 031 099 lt001 lt001 001 lt001 029 008
Mg 122 134 154 1336 524 804 112 014 034 Mn mg kg-1 495 800 155 730 1375 1860 460 110 85 Na 181 006 093 lt001 lt001 lt001 lt001 001 001 Ni mg kg-1 2037 1545 3321 3510 3020 3960 2740 20 9
Total digestion ICP-AES analyses daggerSerpentinites are dominantly composed of the minerals serpentine chlorite talc and spinels DaggerSee Figure 21b for sample locations
33
Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in serpentinite rock at Jasper Ridge
Chlorite Enstatite Augite Chromite Cr-Magnetite ChromiteSilicatedagger SiO2 3738 5463 5222 000 024 1053 Al2O3 552 477 516 5059 182 395 NiO 001 01 00 03 00 052 FeODagger 762 614 239 1446 7919 5042 MnO 009 015 009 042 140 182 Cr2O3 024 060 099 1584 1605 1939 MgO 3337 3305 1560 1869 024 605 CaO 023 050 2348 000 000 019 Na2O 000 001 050 000 001 008 K2O 000 000 001 000 000 011 Total 8446 9994 10048 10027 9894 9307
Formula Unit Compositions
Si 370 094 189 000 001 --- IVAl 030 006 011 000 000 --- VIAl 035 004 011 157 008 --- Ni 000 000 000 001 000 ---
Fe+2 063 010 007 030 232 --- Mn 001 000 000 001 004 --- Cr 002 001 003 033 048 --- Mg 493 085 084 073 001 --- Ca 002 001 091 000 000 --- Na 000 000 004 000 000 --- K 000 000 000 000 000 --- O 1400 300 600 400 400 ---
Electron microprobe analyses daggerChromitesilicate is also referred to as CSM DaggerFe+3 is present however all conversions are based on Fe+2
34
Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in serpentine soil at Jasper Ridge
CSM Chromite Cr-Magnetite SiO2 1940 057 267 Al2O3 559 1059 381 NiO 014 001 004 FeOdagger 2629 2259 5244 MnO 155 154 419 Cr2O3 1878 5577 3237 MgO 1668 916 288 CaO 015 001 005 Na2O 006 001 001 K2O 0 008 009 Total 8864 10033 9855
Si --- 002 010 IVAl --- 042 017 VIAl --- 000 000 Ni --- 000 000
Fe+2 --- 063 148 Mn --- 004 013 Cr --- 147 096 Mg --- 046 016 Ca --- 000 000 Na --- 000 000 K --- 000 000 O --- 400 400
Electron microprobe analyses dagger Fe+3 is present however all conversions are based on Fe+2
35
Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25 and 26 A B C D E F G H I J K L Chromite
Chromite CSM Magnetite Chlorite Olivine Serpentine Serpentine Ni-Fe-Metal Cr-Magnetite CSM Chlorite
SiO2 0 0 1053 0 3407 4046 3688 4074 591dagger 021 1143 2855Al2O3
5059 5162 395 034 1114 000 397 085 054 165 980 2293NiO 028 031 052 0 004 037 037 0 5302dagger 0 0 0
FeODagger 1446 1452 5042 9992
1220 901 262 453 1753dagger 8169 3498 806MnO 042 044 182 0 008 010 0 0 0 107 150 00Cr2O3 1584 1576 1939 125 071 001 0 0 0 1369 1905 039MgO 1869 1869 605 003 292 4895 3848 3755 064 1217 2828CaO 0 0 019 0 0 0 0 0 0 0 0Na2O 0 0 0 0 0 0 0 0 0 0 0K2O 0 0 0 0 0 0 0 0 0 0 0H2O --- --- --- --- 1200 --- 1600 1600 --- --- 1200Total 10027 10136 9307
10164 9944 9891 9834 9972
9897 8897 10028
Formula Unit CompositionSi
000 000 --- 000 328 100 184 200 001 --- 269IVAl 000 000 --- 000 072 000 000 000 000 --- 131VIAl 157 159 --- 002 058 000 023 005 007 --- 123Ni 001 001 --- 000 000 001 001 000 000 --- 000
Fe+2 030 029 --- 294 084 019 011 019 239 --- 059Mn 001 001 --- 000 000 000 000 000 003 --- 000Cr 033 032 --- 004 006 000 000 000 041 --- 003Mg 073 073 --- 000 419 180 285 274 004 --- 397Ca 000 000 --- 000 000 000 000 000 000 --- 000Na 000 000 --- 000 000 000 000 000 000 --- 000K 000 000 --- 000 000 000 000 000 000 --- 000O 400 000 --- 000 1400 400 700 700 400 --- 1400
36
Electron microprobe analyses dagger elemental metal Dagger Fe+3 is present however all conversions are based o
36
2321000---
10025
591000054
53021753000000
2321000000000000
n Fe+2
Table 24 Bulk soil elemental concentrations for soils and bedrock collected along the transect
(Fig 21b)
Sites Samples Depth
(meters) Cr
(mg kg-1) Ni
(mg kg-1) Mn
(mg kg-1) Co
(mg kg-1) Al
() Fe
() Ca ()
Mg ()
COI03 0015 2420 1590 1325 118 464 565 065 352 COI36 0045 2370 1630 1600 138 467 55 063 356 COI69 0075 2160 1755 1670 144 471 6 064 365 COI912 0105 2420 1785 1740 151 496 615 067 384
1
COI1215 0135 2570 1985 2340 195 506 649 069 411 COII02 001 3710 2400 1665 169 294 702 100 768 COII25 0035 3600 2290 1625 162 275 673 093 743 COII58 0065 2910 2310 1550 159 25 672 087 771
COII811 0095 3520 2350 1595 158 276 693 093 766 2
COIIBedrock 012 1840 3510 730 99 027 581 003 1336 COIII14 0025 3750 2930 1655 190 178 897 062 762 COIII47 0055 4760 3050 1850 178 178 94 056 805
COIII710 0085 4380 3410 1930 216 216 1044 07 797 3 COIII1013 0115 4320 3380 1955 201 213 1061 063 626 COIV01 0005 3710 2970 1915 214 209 909 053 781 4 COIV15 003 3540 2710 1725 195 181 801 043 792 COV14 0025 2620 2350 1330 150 246 651 063 526 COV47 0055 1775 2710 1520 165 252 766 059 517
COV710 0085 2000 3090 1830 195 225 737 057 583 5 COVBedrock 0115 1945 3020 1375 155 22 76 046 524
COVI14 0025 3040 3720 1855 224 084 1152 015 736 COVI47 0055 2400 3610 1785 218 085 1138 014 68 COVI710 0085 2960 3800 1850 232 093 1232 015 714 6
COVI1013 0115 2800 3860 1800 224 085 1164 014 739 COVII14 0025 2620 3550 1925 229 083 1128 015 718 COVII47 0055 1975 3590 1805 216 084 1085 014 655 7 COVII712 0095 2080 3600 1840 220 07 1105 011 741 COVIII14 0025 2130 3830 1990 238 085 115 017 705 COVIII47 0055 2550 3720 1865 228 085 1099 017 714 8
COVIII712 0095 2240 3910 1860 229 076 1112 014 828 COIX14 0025 2470 3480 1890 227 104 1201 023 579 COIX47 0055 3000 3830 1595 205 084 108 013 842 9 COIX713 01 2720 3710 1845 221 111 1208 017 551 COX14 0025 2810 3740 1760 218 129 114 017 675 COX47 0055 2190 3910 1835 224 131 1206 017 702
COX710 0085 1725 3370 1595 192 11 1061 014 635 10 COXBedrock 0115 2090 3960 1860 217 107 1106 013 804
Total digestion ICP-AES analyses
37
Table 25 Chemical and physical properties of the soil at Site JR3
NH4OAc-extractable ---Hydrometer---Ca K Mg Na
Sand Silt ClaySample Depth (cm) pH EC
dSm
Organic Matter ------
LECO Nitrogen ------ ---------mg kg-1---------
Texture
JR305 25 671 04 32 012 505 2366 3291 355 36 31 33 Clay LoamJR3515
10 683 09 19 064 4205 8176 3343 2392 36 30 34 Clay LoamJR31530 225 688 03 12 003 3725 4903 3750 2305 36 33 31 Clay LoamJR33045 375 698 07 10 003 2873 3665 3388 2549 46 24 30 Sandy Clay Loam
See Methods Section for more details
38
38
Table 26 Selected trace metal concentrations for each size fraction versus depth at Site JR3
Depth (cm) Total gt2mm Fraction Sand Silt Clay Sample
------------------------Chromium (mg kg-1)------------------------ JR305 25 4906 5086 gt10000 4565 988 JR3515 10 5976 4678 8284 4000 1115 JR31530 225 5222 4206 gt10000 4133 980 JR33045 375 3477 3297 gt10000 4571 956
------------------------Nickel (mg kg-1)------------------------------ JR305 25 1935 2050 1660 1305 2920 JR3515 10 2180 2050 1980 1325 2920 JR31530 225 2190 2120 1835 1370 2880 JR33045 375 2090 2140 1740 1285 2840
------------------------Manganese (mg kg-1)-------------------------- JR305 25 1319 1571 1728 1301 1210 JR3515 10 1664 1601 2701 1471 1254 JR31530 225 1639 1626 2217 1599 1231 JR33045 375 1222 1226 1781 1405 1170
-----------------------Cobalt (mg kg-1)-------------------------------- JR305 25 128 146 159 134 131 JR3515 10 155 145 241 147 132 JR31530 225 152 147 194 162 128 JR33045 375 125 124 159 141 122
Total digestion ICP-AES analyses
39
Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3
JR305CLAY Depth 0-5cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1439 100 1833 100 1274 100 1391 100 731 179 1432 431 732 645 1047 105 71 216 941 163 71 41 838 838 474 101 847 141 475 257 726 726 356 125 73 271 365 264 313 83 712 248 355 299
473 143 313 257 364 131 355 193
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1845 1000 1268 1000 1388 1000 852 68 1447 387 724 987 1010 859 733 264 944 143 703 679 836 281 712 247 903 139 471 282 723 576 475 109 850 83 362 244 365 73 733 426 352 321 356 117 712 291 313 154 314 63 475 122
364 170 355 196
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1724 1000 1297 1000 1396 1000 842 45 1397 447 741 957 1175 708 730 318 923 176 715 686 1037 944 710 255 840 118 477 271 837 139 475 97 717 465 367 400 726 375 357 133 468 182 357 386 313 91 362 194 315 286
352 229 JR33045CLAY Depth 30-45cm
Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity
1447 1000 1869 1000 1268 1000 1393 627 842 44 1476 287 737 843 1049 1000 730 207 954 107 715 539 837 190 710 182 915 107 479 157 723 111 475 69 860 61 366 303 364 69 740 332 356 292 356 74 717 256 315 315 313 57 478 111
367 141 357 115
See Methods Section for details
40
Table 28 X-ray diffraction d-spacings for common silicates from Whittig and Allardice (1986)
Diffraction Spacing (Aring) Mineral(s) Indicated Mg-saturated air-dried
14-15 Smectite vermiculite chlorite 99-101 Mica (illite) 71-73 Serpentine
715 Kaolinite chlorite Mg-saturated glycerol-solvated
177-180 Smectite 14-15 Vermiculite chlorite
99-101 Mica (illite) 71-73 Serpentinte
715 Kaolinite chlorite K-saturated air-dried
14-15 Chlorite vermiculite (with hydroxy interlayer) 124-128 Smectite 99-101 Mica vermiculite (contracted) 71-73 Serpentine
715 Kaolinite chlorite K-saturated heated (500ordmC)
14 Chlorite 99-101 Mica vermiculite (contracted) smectite (contracted) 71-73 Serpentine
715 Chlorite
41
Table 29 Results of the Diablo clay serpentine soil and chromite sequential extraction experiments
1 Exchangeable Fraction (1mM BaCl2) 4 Sum of 1 2 amp 3 (mg kg-1) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 0 0 003 Cr 159 388 19 Fe 014 011 001 Fe 24812 25532 1096 Ni 104 103 008 Ni 1854 1868 25 Mn 0 0 0 Mn 778 1172 04
2 Fe-Oxide Fraction (1M NH2OH HCl in 25 (vv) HOAc) 5 Total Cr Fe Ni and Mn (Total Digestion) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 19 78 05 Cr 2290 3310 1767864 Fe 4474 5138 38 Fe 60000 68000 1044325 Ni 1610 1506 01 Ni 1673 2345 00 Mn 231 430 00 Mn 1498 1608 3424547
3 Silicate Fraction (10M HF) 6 Percent Extractable ((45)100) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ------------------------------------- Cr 140 310 14 Cr 07 12 00 Fe 20337 20393 1058 Fe 41 38 01 Ni 234 352 22 Ni 111 80 00 Mn 547 742 04 Mn 52 73 00
ICP-AES analyses
42
Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1 kg of sample
Chromite
FeCr2O4 Cr Concentration
(mg kg-1) Grams Sample 500 11 011 1000 22 022 2000 43 043 5000 108 108
43
Figure Captions
Figure 21a Soils map of Jasper Ridge Biological Preserve (Stanford CA) with
complete descriptions of the soils shown in the legend 21b An enlarged image of the
study area shown in 22a indicates sampling sites of the soils and rocks collected A
dashed line marks the 500 m transect
Figure 22 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above
Figure 23 Petrographic images illustrating the varying serpentinite textures at Sites (A)
JR3 (B) JR5E and (C) JR3E For reference to site locations see Figure 21b
Figure 24 Correlation diagrams of Cr-bearing minerals in the serpentinite at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Si in Wt
Chromite (circle) Cr-magnetite (square) chromitesilicate or CSM (down triangle)
chlorite (hexagon) and pyroxenes (up triangle) analyses are shown on each correlation
diagram
Figure 25 Images of the serpentinite bedrock at Site JR3 demonstrating the mineralogy
and Cr-bearing phases Microprobe analyses are indicated by circled letters and these
values are in Table 23 25a Backscattering electron image demonstrating several Cr-
bearing phases present including chromite Cr-magnetite CSM and Cr-chlorite from
Figure 23a 25b Magnified image shown in Figure 25a demonstrating the fine-grained
texture of CSM Additionally Cr2O3 Wt values are shown for chromite the depletion
zone and CSM 25c Image of Ni-bearing phases present in the bedrock including
olivine serpentine and Ni-Fe alloy from Figure 23a 25d Cartoon illustrating the
mineralogical and chemical changes present in 25c
Figure 26 Images of a Cr-chlorite and Cr-magnetite assemblage in the serpentinite
bedrock 26a Petrographic image of the bedrock at Site 4 (Figure 21b) 26b
44
Backscattered electron image demonstrating the common occurrence of Cr-chlorite
encompassing chromite grains Microprobe analyses are indicated by circled letters and
are located in Table 23
Figure 27 Highest and lowest Cr concentrations (mg kg-1) at each site in the serpentine
soil 500 m transect shown in Figure 21b Samples included soils at various depths (m)
and bedrock from each site The average Cr concentration for the Jasper Ridge
serpentinite is marked on the central figure Chromium concentrations versus depth (m)
at each site (noted in the upper left corner) outline the main diagram
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected
for total gt2 mm sand silt and clay size fractions listed in Table 26
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soil at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Al+Fe in Wt for
chromite (circles) Cr-magnetite (hexagons) and CSM (diamonds) Chromite from the
bedrock is denoted by a larger circle and FeCr2O4 with a square
Figure 210 XRD clay analyses of sample JR3515 (Site JR3 Figure 21b) demonstrate
the influence of each treament Mg-saturated Mg+glycerol K-saturated and K-saturated
+ heat (500oC) The d-spacings and intensities are also listed in Table 27
Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at depths (a) 0-5cm (b) 5-15cm (c) 15-30cm (d) 30-45cm and (e) bedrock The pre-edge
area indicative of Cr(VI) peaks is shown on the right Note that no peaks are present
Figure 212 X-ray microprobe elemental maps of Cr and Fe for (212a) serpentine soil
at a depth of 0-5cm (212b) serpentine soil at a depth of 5-10cm and (212c) serpentinite
bedrock at Jasper Ridge site JR3 (Figure 21b)
45
Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments for the Diablo clay (Site 1 Figure 21b) serpentine soil (Site 2
Figure 21b) and chromite listed in Table 29 The sum refers to the summation of the
three extraction experiments
Figure 214 SEM images of the chromite standard (A) after each sequential extraction
treatment including the (B) exchangeable fraction (C) Fe-crystalline oxide fraction and
(D) silicate fraction The concentration of Cr removed by each treatment is shown
beneath each set of images
Figure 215 Tie-lines from Figure 29 connecting serpentinite chromite and end-
member FeCr2O4 are shown on this single plot Soil analyses from Figure 29 are also
plotted on this diagram For a sum of 100 Wt oxygen must be added to the sum of
Cr Al Fe and Mg Wt
46
Figure 21a
47
Figure 21b
48
Chromium Concentration (mg kg-1)1000 10000
Dep
th (m
eter
s) 01
10
Maryland (Rabenhorst et al 1982)California (Gough et al 1989)California (Hotz 1964)New Foundland (Roberts amp Proctor 1992)
Surface
Figure 22
49
Figure 23
50
Cr (Wt )0 5 10 15 20
Al (
Wt
)
0
5
10
15
20
25
30
Cr (Wt )0 5 10 15 20
Fe (W
t
)
0
20
40
60
80
100
Cr (Wt )0 5 10 15 20
Mg
(Wt
)
0
5
10
15
20
25
Cr (Wt )0 5 10 15 20
Si (W
t
)
0
5
10
15
20
25
30
Cr0 10 20 30 40 50 60 70 80 90 100
Al
0
10
20
30
40
50
60
70
80
90
100
Fe
0
10
20
30
40
50
60
70
80
90
100
Chromite
Chlorite
Magnetite
Chromite-Silicate
Pyroxenes
Chromite
Chlorite
Pyroxenes
Magnetite
Chromite-Silicate
Figure 24
51
Figure 25
52
Figure 26
53
Sites1 2 3 4 5 6 7 8 9 10
Con
cent
ratio
n (m
g kg
-1)
1500
2000
2500
3000
3500
4000
4500
5000Soil Chromium
1
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000002004006008010012014016
2
Cr Concentration (mg kg-1)
1500 2000 2500 3000 3500 4000
Dep
th (m
eter
s)
000002004006008010012014
3
Cr Concentration (mg kg-1)3500 4000 4500 5000
Dep
th (m
eter
s)
000
002
004
006
008
010
012
5
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000002004006008010012
6
Cr Concentration (mg kg-1)2200 2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000002004006008010012
7
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000
002
004
006
008
010
9
Cr Concentration (mg kg-1)
2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000
002
004
006
008
010
012
10
Cr Concentration (mg kg-1)1600 2000 2400 2800 3200
Dep
th (m
eter
s)
000002004006008010012
4
Cr Concentration (mg kg-1)3500 3600 3700 3800
Dep
th (m
eter
s)
000
001
002
003
004
8
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000
002
004
006
008
010
BEDROCK
BEDROCK
BEDROCK
Range of Cr Concentrations in Serpentinite Bedrock
Figure 27
54
Soil Chromium
Depth (cm)0 10 20 30 40 50
Cr m
g kg
-1
02000400060008000
100001200014000
Totalgt2mmSandSiltClay
Soil Nickel
Depth (cm)0 10 20 30 40 50
Ni m
g kg
-1
0
1000
2000
3000
4000
5000
Soil Manganese
Depth (cm)0 10 20 30 40 50
Mn
mg
kg-1
0
1000
2000
3000Totalgt2mmSandSiltClay
Soil Cobalt
Depth (cm)0 10 20 30 40 50
Co
mg
kg-1
0
50
100
150
200
250
300
Totalgt2mmSandSiltClay
Totalgt2mmSandSiltClay
Exceeds detection limits of 10000 mg kg-1 Cr
Figure 28
55
Cr (Wt )0 10 20 30 40 50
Al (W
t
)
0
5
10
15
20
25
30
Cr (Wt )0 10 20 30 40 50
Mg
(Wt
)
0
2
4
6
8
10
12
Cr (Wt )0 10 20 30 40 50
Si (W
t
)
0
2
4
6
8
10
12
14Cr (Wt )
0 10 20 30 40 50
Fe (W
t
)
0
10
20
30
40
50
60
70
Cr (Wt )0 10 20 30 40 50
Al +
Fe
(Wt
)
10
20
30
40
50
60
70
JRBPChromite
FeCr2O4
Chromite
Magnetite
Chromite-Silicate
Figure 29
56
Clay XRD of JR3515 Depth 5-15cm
0
200
400
600
800
1000
1200
0 5 10 15 20 25 30 35 40 45
d-spacing (Aring)
Rel
ativ
e In
tens
ity
Mg-Saturated
Mg+Glycerol
K-saturated
K-saturated+heat (500oC)
Figure 210
57
Energy (kev)
597 598 599 600 601 602 603
Nor
mal
ized
Abs
orba
nce
-4
-3
-2
-1
0
1
2
(a)
(b)
(c)
(d)
(e)
Cr(
VI)
Figure 211
58
Figure 212
59
Diablo Clay
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Serpentine Soil
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Chromite
0
20
40
60
80
100
120
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide
SilicateSum
Figure 213
60
Figure 214
61
Reaction Progress ()0 20 40 60 80 100
Elem
enta
l Wt
0
10
20
30
40
50
ChromiumAluminumIronMagnesium
FeCr2O4Jasper Ridge
Bedrock Chromite
SoilChromite
Figure 215
62
CHAPTER 3
OXIDATIVE PROMOTED DISSOLUTION OF CHROMITE BY MANGANESE
DIOXIDE AND CONCURRENT PRODUCTION OF CHROMATE
ABSTRACT
Chromium release and oxidation from chromite is a potential environmental hazard in
sediments and soils and a pathway for soil development (weathering of primary minerals)
related to ultramafic rocks and their metamorphic derivatives (serpentinites) Birnessite is
a common pedologic mineral in these sediments and soils capable of oxidizing aqueous
Cr(III) In this study the interaction between chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4)
and birnessite is investigated with an interest in the potential generation of Cr(VI)
Specifically the effects of chromite suspension density birnessite suspension density
and pH at a temperature of ~25degC in relation to the reaction kinetics were examined The
rate of Cr(VI) in solution increases with increasing chromite suspension densities and
decreasing pH but is independent of birnessite suspension densities at values greater than
~20 m2 L-1 The overall rate expression of Cr released and oxidized in solution from
chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression with Cr(VI) rates ranging from 9times10-4 to 44times10-3 microM h-1
Additionally these experiments demonstrate that serpentine soils are a source of non-
anthropogenic Cr(VI)
63
INTRODUCTION
Chromite a Cr-enriched spinel is responsible for a majority of the Cr enrichment
identified in ultramafic rocks serpentinites and serpentine soils (Rabenhorst et al 1982
Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989 Gasser and
Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003 Oze et al
2003a Oze et al 2003b) The chemical weathering of chromite is therefore an important
process when evaluating the potential of ultramafic rocks and their related sediments and
soils as sources of Cr Despite the propensity of chromite to resist chemical weathering
the dissolution of chromite and release of Cr will occur over geologic time (Bialowolska
and Salacinski 1984 Oze et al 2003b) Chromium in chromite is in the form of Cr(III)
the non-toxic and chemically immobile valence state however oxidizing conditions in
near surface environments is a potential reaction leading to the enhanced dissolution of
chromite and the formation of Cr(VI) the toxic and chemically mobile valence state
which typically exists as the oxyanion chromate (HxCrO42-x) (Daugherty 1992 Cohen et
al 1993 Fendorf 1995 Deer et al 1996) Consequently chromite may be a non-
anthropogenic source of Cr(VI) in the environment
Naturally occurring high valence Mn oxides present in soils and sediments
derived from ultramafic rocks are strong oxidizers capable of affecting the chemical
mobility of metals through redox reactions (Fendorf and Zasoski 1991 Johnson and
Xyla 1991) These insoluble mixed valent oxides (Mn(IIIIV)) are observed coating
mineral surfaces in these ultramafic sediments and soils where total Mn concentrations
are often gt100 mg kg-1 (Brooks 1987 Schreier et al 1987 Oze et al 2003a Oze et al
2003b Oze et al) Birnessite is a naturally occurring mixed valence oxide known to
oxidize aqueous Cr(III) to the more chemically mobile Cr(VI) (Eary and Rai 1987
Fendorf and Zasoski 1991) Gough et al (1989) and Becquer et al (2003) have
suggested that Mn oxides are responsible for Cr(VI) identified in California and New
Caledonia serpentine soil solutions however the dissolution and oxidation of Cr from a
Cr-bearing mineral due to the presence of birnessite has not been quantitatively defined
In this study we examine the kinetics of Cr release and oxidation when chromite and
birnessite are present in solution
64
The objective of this study is to determine a rate expression describing the
release and oxidation of Cr from chromite in the presence of birnessite Using well-
controlled batch experiments the solution chemistry and geochemistry of Cr related to
chromite and birnessite interaction was evaluated temporally over a range of pH
Conditions were chosen to simulate the slightly acidic conditions characteristic of most
serpentine soils (Rabenhorst et al 1982 Gough et al 1989 Graham et al 1990 Cole
1992 Jeffrey 1992 Lee 1992 Verger 1992 Becquer et al 2003) Ultimately these
experiments allow the examination of one potential pathway leading to Cr(VI) identified
in solutions related to ultramafic rocks and serpentine soils
MATERIALS AND METHODS
Solid Phase Characterization and Synthesis
Chromite used in the experiments was commercial-grade chromite (MSDS No
302-1 WHMIS CLASS D2A) obtained from Barnes Environmental Inc Ontario
Chromite was prepared by cleaning the sand-size material in multiple ultrasonic
isopropanol baths and rinses The chromite was then sieved using a Retsch magnetic
sieve shaker for 30 minutes in order to obtain the 160-250 microm size fraction Impurities in
the 160-250 microm chromite were removed using a slope Franz magnetic separator These
impurities included the minerals olivine augite plagioclase feldspar and quartz Visual
and scanning electron microprobe (SEM) examinations of the remaining 160-250 microm
fraction confirmed its purity Chromite was crushed to a fine powder using an agate
mortar and pestle and then rinsed multiple times in deionized destilled (DDI) water and
001 M HCl solution to remove the ultra-fine material and eliminate highly reactive sites
caused during grinding The chromite powder was inspected before (Figure 31) and
after the experiments using a scanning electron microscope (JEOL 5600LV) the surface
area was 025 m2 g-1 as determined using N2(g) and BET isotherm
An automated JEOL 733A electron microprobe operated at 15 kV accelerating
potential and 15 nA beam current was used to investigate possible zoning and to quantify
the chemical composition of the chromite grains prior to powdering No zoning was
observed and the chemical composition of the chromite was based on the average of 25
electron microprobe analyses The stoichiometric composition of chromite based on 4
65
oxygens is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 X-ray diffraction was also used to
deduce the purity of the chromite powder X-ray diffraction spectra obtained for the
chromite powder matched JCPDS card 03-0873 (unit cell a=8300 b=8300 c=8300
α=90 β=90 and γ=90) and no contaminants were evident
Birnessite was synthesized using the procedure described by Buser et al (1954)
and implemented by Fendorf and Zasoski (1992) A solution of 63 g of potassium
permanganate (KMnO4) in 1 L of DDI water was heated to 90 degC Approximately 100
mL of the heated slurry was poured into a 4 L beaker and reheated to 90degC while being
stirred vigorously The remaining 900 mL of slurry and 66 mL of concentrated
hydrochloric acid were simultaneously and proportionally dispensed into the beaker over
a period of thirty seconds The temperature of the resulting solution was maintained at
90deg C for an additional 10 minutes and then allowed to cool to room temperature The
product was washed with alternate rinses of DDI water and a 001 M HCl solution The
birnessite surface area was 125 m2 g-1 based on BET analyses Solids were inspected
before (Figure 31) and after the experiments using the SEM
Extraction Experiment Procedures
Several chemical extraction experiments were performed on 106-250 microm
chromite grains prior to the chromite-birnessite batch experiments in order to determine
the potential release of Cr Fe and Mn One gram of chromite and 25 mL of solution
were placed into a 25 mL centrifuge tube and the solids and solution were gently agitated
The solutions and parameters used for these extraction experiments include the
following 1) 10 hydrogen peroxide (H2O2) solution at 25degC and pH of ~5 for 6 h 2)
01 M potassium permanganate (KMnO4) solution at 60degC and pH of ~3 for 6 h 3) 10 M
hydrofluoric (HF) acid at 60degC for 6 h 4) 1 M hydroxylamine (NH2OHmiddotHCl) with 25
(vv) acetic acid (HOAc) at 60degC and pH of ~4 for 6 h and 5) 1 mM barium chloride
(BaCl2) solution at 60degC and pH of ~6 for 6 h Solutions were extracted with a syringe
and passed through a 02 microm filter Solids were inspected after the experiments using the
SEM
66
Batch Experiment Procedures
Batch experiments were performed in triplicate using 1 L polypropylene bottles
and chromite and birnessite ratios listed in Table 31 at ~25degC After loading chromite
and birnessite into each bottle 1 L of a pH 3 solution a 10 mM acetate solution (pH 5) a
10 mM MES solution (pH 67) or a pH 8 solution was added to the appropriate
experiment (Table 31) Chromite and birnessite were also run individually in the acetate
buffered solution (Table 31) Aliquots were extracted from the polypropylene bottles
using a syringe and then passed through a 2 microm filter before analyses Less than 40 mL
was extracted over the course of each experiment resulting in a total volume change of
lt4 After the experiments were terminated the remaining solids were resuspended in
20 mL of 1 mM phosphate (5050 KH2PO4K2HPO4 pH 7) 20 mL of 10 mM CaCl2
solution at pH 7 and a 05 M HCl solution These suspensions were agitated for 3 h
followed by centrifugation and filtration (02 microm) before analyses The purpose of the
phosphate and CaCl2 solutions was to desorb reaction produtcts (eg Cr(VI) Cr(III) Fe
and Mn) from the solid phases The HCl treatment provides a means to evaluate
secondary phases such (oxy)hydroxides formed during the experiments Although the
HCl solution may dissolve chromite and birnessite in addition to the (oxy)hydroxides the
short time frame of the extraction (3 h) should limit the dissolution of chromite
Additionally the polypropylene bottles following the termination of five batch
experiments were rinsed with DDI water and 20 mL of 05 M HCl was added to each
bottle The HCl solution was allowed to remain in the bottles for 15 h in order to
determine the concentration of Cr Fe and Mn sorbed onto the interior lining
Serpentine Soil and Birnessite Experiments
Serpentine soils collected from Jasper Ridge Biological Preserve (Stanford CA)
and New Caledonia were used to determine the influence of birnessite in soils known to
contain chromite (Chapters 2 and 4) The Cr geochemistry of these soils as well as the
detrital chromite have been evaluated in Chapters 2 and 4 and by Becquer et al (2003)
Jasper Ridge serpentine soils contain Cr-bearing minerals including chromite Cr-
magnetite and Cr-chlorite however a majority of the Cr enrichment (Table 32) is
directly related to the abundance of chromite A majority of the Cr related to New
67
Caledonia serpentine soils is attributed to Cr (oxy)hydroxides (Chapter 4) however
chromite is present in these soils as well (Becquer et al 2003) Although birnessite has
been noted in these soils the quantity of birnessite could not be determined In order to
evaluate the impact of Mn oxides on Cr within such soils 01 g of birnessite was added to
each soil
The parameters for the experiments are listed in Table 32 Ten grams of the
lt250 microm size-fraction of each soil (JR1 JR2 and NC) and 100 mL of 1 mM acetate
buffered solution (pH 5) were loaded into individual 150 mL Nalgene bottles Aqueous
solutions were evaluated as a function of time (h) Additionally 10 g of each soil 01 g
of birnessite and 100 mL of 1 mM acetate buffered solution (pH 5) were loaded into 150
mL Nalgene bottles and were allowed to react as a function of time (h) These
experiments were rotated at a rate of 2 revolutions min-1 Solutions were removed with a
syringe and passed through a 2 microm filter before analysis Approximately 8 mL of
solution was required for the analyses over the course of each experiment resulting in a
total volume change of 8 Total Cr concentrations (mg kg-1) of the soils listed in Table
32 were obtained by completely dissolving the samples using a mixture of hot (gt200degC)
concentrated nitric perchloric and hydrofluoric acids Elemental compositions were
measured in the supernatant using inductively coupled plasma optical emission
spectroscopy (ICP-OES) The mass of chromite in each soil was calculated using total Cr
concentration for each soil and the calculations provided in Chapters 2 (Table 210) and
Chapter 4 (Table 47) The chromite surface area of each soil was approximated by
comparing the surface area measured for the chromite standard and its observed grain
size from SEM images (Figure 31) to chromites observed in the soils using backscattered
electron images (Chapters 2 and 4) The surface area of the Jasper Ridge chromite is 066
of the chromite standard due to a slightly larger grain size The surface area of the
New Caledonia chromite is approximately 025 of the chromite standard Chromite
suspension densities were calculated by multiplying the chromite mass by the estimated
surface area and then dividing by the volume of the solution
68
Solution Analysis
The pH and Eh of the supernatant was measured with an Orion flat surface
combination pH electrode inserted directly into supernatant Production of soluble
Cr(VI) was measured spectrophotometrically at 540 nm using the s-diphenyl carbazide
method modified from Bartlett and James (1979) Soluble Fe(II) was monitored using
the ferrozine method adopted from Stookey (1970) Total dissolved Cr Fe Al Mn and
Mg were ascertained using ICP-OES Concentrated trace metal grade nitric acid used to
acidify the samples prior to analyses Aluminum and Mg was determined to be present in
the nitric acid at concentrations lt1 microM Magnesium and Al are not plotted in any of the
figures Chromium(III) and Fe(III) was determined by subtracting Cr(VI) and Fe(II)
obtained from colorimetric methods from total dissolved Cr and Fe analyses using ICP-
AES Results are presented as an average of the three runs for each batch experiment
with the standard deviation noted by error bars
RESULTS
Chromite Extraction Experiments
The release of Cr Fe and Mn from chromite using 10 H2O2 01 M KMnO4 10
M HF acid 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 1 mM BaCl2 solutions was
carried out to determine the relative propensity of chromite to resist chemical dissolution
and these results are summarized in Figure 32 The KMnO4 solution with aqueous
Mn(VII) was the most effective in releasing Cr from chromite yielding a concentration
of 62 Cr mg L-1 Iron released using the KMnO4 solution was 03 mg L-1 Although
H2O2 is a direct chemical oxidant the release of Cr from chromite was less as compared
to the other solutions (only ~03 Cr mg L-1 was measured) Concentrated HF was only
able to release ~13 Cr mg L-1 while releasing 106 Fe mg L-1 NH2OHmiddotHCl with 25
(vv) HOAc and CaCl2 solutions only removed lt1 Cr mg L-1 however the NH2OHmiddotHCl
removed 36 Fe mg L-1 which is comparatively higher than the H2O2 KMnO4 and BaCl2
solutions
69
Chromite and Birnessite Batch Experiments
The interaction of chromite and birnessite leading to the release and oxidation of
Cr(III) is a relatively slow process requiring study on the order of hours and days The
pH was maintained constant throughout the duration of the experiments while Eh values
ranged from 320 to 450 mV with no discernable temporal pattern (Figure 33) For pH
values between 3 and 67 Cr3+ CrOH2+ and Cr(OH)2+ are the dominant aqueous Cr(III)
species however at pH 8 conditions are on the stability line between Cr(III) and Cr(VI)
specifically Cr(OH)3 and CrO42-
Higher chromite suspension densities produced higher rates of Cr release to the
aqueous phase and all the Cr in solution was in the form of Cr(VI) (Figure 35) The
rates of oxidation based on Cr(VI) production range from 00005 to 00041 microM h-1
(Figure 35 Table 33) Comparatively Cr released from the chromite control experiment
is maintained at a relatively constant value of ~005 microM and no Cr(VI) is measured in
solution Iron concentrations initially increase to ~ 05 microM and then decrease rapidly to a
relatively constant concentration at 02 microM (Figure 35) Iron concentrations for the
chromite control experiment began at 15 microM between 0 to 50 h and decreased to ~02
microM Iron(II) and Mn concentrations are below detectable limits for all experiments
(Figure 35) A zeroth-order dependence is assumed with respect to birnessite owing to
its large excess surface area relative to chromite
Variable birnessite suspension densities and constant chromite suspension
densities were evaluated at pH 5 (Figure 36) Although suspension densities of
birnessite are 538 and 1075 m2 L-1 (Table 31) the rate of Cr and Cr(VI) measured in
solution for both experiments is 00041 microM h-1 (Figure 36 and Table 33) Iron
concentrations are below 06 microM and maintain concentrations near 02 microM with no
indication of consistently increasing or decreasing with respect to time Iron(II) and Mn
concentrations in solution are below detectable limits for both experiments
In Figure 37 the influence of pH on chromite oxidation by birnessite was
evaluated by conducting experiments at pH values of 3 5 67 and 8 chromite and
birnessite suspension densities are constant at 0075 m2 L-1 and 1075 m2 L-1 respectively
(Table 31) Chromium and Cr(VI) rates increase with decreasing pH ranging from 0
microM h-1 at pH 8 to 00062 microM h-1 at pH 3 (Table 33) Iron is present in solution at
70
concentrations less than 04 microM for all the experiments Generally Fe concentrations are
greater near the start of the experiment and decrease to values lt 05 microM with increasing
reaction time Manganese concentrations are below detectable limits
In order to determine the rate dependence on each prominent reactant (chromite
birnessite and pH) reaction rates were regressed against reactant concentrations (Figure
38) A pseudo first-order rate dependence is noted based on chromite suspension
density the equation regressed for this relationship is listed in Table 33 As the chromite
suspension density increases the rate of Cr(VI) in solution increases In contrast a
zeroth-order rate dependence is evident for birnessite suspension density (dashed line in
Figure 38b) The rate of Cr(VI) generation is expected to decrease with respect to
birnessite suspension density at very low concentrations (dotted line in Figure 38b)
The influence of pH on the rate of Cr(VI) production is well-described by a polynomial
function (Figure 38c Table 33)
Developing an overall rate expression for Cr in solution requires integrating the
factors assessed in Figure 38 and listed in Table 33 All the Cr measured in solution is
in form of Cr(VI) therefore the rate law for Cr or Cr(VI) in solution may be written as
d[Cr(VI))]dt = k sdotChromitea sdot Birnessiteb sdot (aH+)c (31)
and
log d[Cr(VI))]dt = log k + asdotlogChromite+ bsdot log Birnessite+ csdotlog (aH+) (32)
where k is the overall rate constant for Cr release and oxidation Chromite is the
chromite suspension density (Table 31) Birnessite is the birnessite suspension density
(Table 31) and aH+ is the activity of H+ Constants a b and c are the order of
dependency for each reactant The overall rate equation (developed in the Appendix) can
be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069 (33)
71
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 the expression is valid for birnessite suspension densities
greater than 20 m2 L-1 and pH values between 3 to 8
Secondary and Solid Phase Examination of Chromite and Birnessite
The solid products were evaluated to determine the abundance and type of
secondary phases formed during the course of the experiments Solids from the batch
experiment were treated with three solutions 1 mM 5050 KH2PO4K2HPO4 1 mM
CaCl2 and 05 M HCl (Table 34) In the chromite and birnessite reactions Cr was
primarily in solution (gt90 ) The HCl extract was able to remove lt5 additional Cr
while the KH2PO4K2HPO4 and CaCl2 extracts released lt3 additional Cr Aqueous Fe
concentrations were below detectable limits for the final solution extraction Treatment
with the HCl resulted in up to 0127 micromoles of Fe however 0107 micromoles of Fe were
released upon treatment of unreacted chromite with HCl Although not shown in Table
34 Mn values were greater than gt4 micromoles when treated with HCl the
KH2PO4K2HPO4 and CaCl2 removed no detectable Mn
After the termination of each experiment the solids were examined using SEM in
order to identify changes in the surface morphology and secondary phases A large
chromite grain and conglomerate of birnessite adjacent to the chromite are depicted in
Figure 38 The face and edge of the chromite grain is not noticeably altered compared to
grains examined prior the experiments Additionally the particle morphology of
birnessite has not visibly changed Even at higher magnification angular and
conchoidally fractured fine-grained edges are noted (Figure 38bc) Only a slight
rounding of the smaller features of the main chromite grain provides any evidence of
possible chromite dissolution (Figure 38c) Birnessite appears to have limited interaction
with chromite Although chromite and birnessite were allowed to interact over 500 h in
an aqueous solution no secondary phases are observed by SEM
Serpentine Soil and Birnessite Experiments
Two serpentine soils from Jasper Ridge Biological Preserve CA and one
serpentine soil from New Caledonia were used to determine whether or not the overall
72
rate expression (Equation 33) for chromite reaction with birnessite is consistent with
Cr(VI) concentrations produced by serpentine soils containing chromite Three
serpentine soils without birnessite and three soils with the addition of birnessite buffered
at pH 5 were measured for Cr(VI) generation as a function time (Figure 310) The rates
of Cr(VI) measured in solution for both Jasper Ridge soils (JR1 and JR2) are identical at
00013 microM h-1 With the addition of birnessite the rates of Cr(VI) production in solution
both increased to 00044 microM h-1 (JR1 w Bt and JR2 wBt in Figure 310) Based on the
chromite suspension density (Table 32) the rates predicted by Equation 33 range from
00035 to 00041 microM h-1 which are 10 to 20 less than those observed in the
experiments (Figure 310) The rates of Cr(VI) in solution from the New Caledonia
serpentine soil are 00009 microM h-1 and with the addition of birnessite increase up to 00013
microM h-1 (Figure 310) A rate of 00016 microM h-1 was predicted by Equation 33 for the
New Caledonia soil
DISCUSSION
The reaction of birnessite with chromite enhances the rate of Cr release into
solution The rate of Cr(VI) measured in solution (Equation 33) is directly related to the
chromite suspension density and is independent of the birnessite at suspension densities
greater than ~20 m2 L-1 Although the release and oxidation of Cr in solution are
measured the mechanisms and pathways driving this reaction need to be evaluated The
propensity of chromite to resist dissolution the effects of pH the formation of secondary
phases and ultimately the pathways of Cr release and oxidation are addressed in the
following discussion
Extraction experiments performed on chromite (Figure 32) demonstrate it is
resistant to acidic dissolution or competitive ion displacement but strong oxidants such
as high valent Mn may oxidatively dissolve chromite It is evident that the proportions of
Cr Fe and Mn vary for each of the extractions The solutions utilized in the extraction
experiments have varying chemical properties capable of preferentially releasing a
particular species andor element These experiments were also performed over a short
period of time (lt6 h) which may have only released elements from highly reactive sites
73
Finally secondary phases associated with these experiments were possible and would
distort the elemental ratios observed in solution
A majority of the Eh and pH analyses plot in the Cr(III) stability region (Figure
33) however Cr in solution is primarily in the form of Cr(VI) (Figures 35 36 and
37) Eh measurements presented in Figures 32 and 33 do not account for the
microenvironments associated with the surfaces of birnessite Additionally the kinetics
of Cr(VI) reduction are limited in reducing conditions due to electron symmetry
constraints and the absence of a suitable redox couple (Fendorf 1995) Finally highly
oxidizing birnessite in the solution is kinetically and thermodynamically controlling the
oxidation state of Cr relative to the stabilities predicted in Figure 33 (Amacher and
Baker 1982 Fendorf and Zasoski 1992)
A fractional-order dependence on the chromite suspension density (Figure 38)
can be explained by diffusional influences competing oxidation reactions and secondary
reactions involving Cr(III) and Mn(IV) Secondary phases on the surface of chromite
may inhibit the diffusion of Cr from chromite however the clean and relatively unaltered
chromite surfaces do not support this mechanism as a major cause (Figure 38)
Chromium and Fe released from chromite may be incorporated into secondary phases
related to birnessite or chromite Approximately 10 of Cr is not in the aqueous phase
(Table 34) supporting the premise that secondary phases are contributing to the
fractional-order dependence related to the chromite suspension density Also
competition from Fe(II) oxidation on birnessite deposition of Fe(II) on birnessite (or
chromite) and diffusion onwithin birnessite may further impact the rate dependence
The zeroth-order dependence based on the birnessite suspension density (Figure
38) results from its high surface area At very low birnessite suspension densities the
rate of Cr(VI) production should decrease however due to the high surface area of
birnessite it was experimentally prohibitive to determine reaction rates at masses less
than 4 mg Therefore a hypothetical trend-line depicts the expected dependence on
birnessite concentration at very low suspension densities at higher values representative
of most field settings the reaction is independent of the birnessite surface area (ie zero-
order dependence)
74
In Figure 38c the polynomial relationship defines the Cr(VI) rate dependence on
pH As the pH decreases the rate of Cr(VI) increases more rapidly The relationship is
potentially the result of lower pHs progressively increasing the rate of chromite
dissolution However the rate of Cr(VI) in solution might be influenced by higher pHs
increasing the formation of secondary Fe(III) and Cr(III) (oxy)hydroxides on the
chromite andor birnessite surfaces thereby decreasing elemental diffusion from
chromite or decreasing available birnessite sites for Cr(III) oxidation Differentiating
between these two mechanisms is not possible based on the rate equation or extraction
analyses however the polynomial relationship evident in Figure 38c is folded into the
rate constant in order to geochemically account for the influence of pH on Cr(VI)
generation
No conclusive evidence is available to determine the fate of Mn within the
reacting suspensions Aqueous Mn concentrations are consistently maintained below
detectable limits If Mn is released from birnessite then it must be strongly and quickly
incorporated into a nonaqueous phase following its release Due to Cr(III) being oxidized
by Mn(III) andor Mn(IV) Mn(II) must be produced Extractions performed on the
chromite and birnessite solids demonstrate that Mn could only be removed with HCl
however a majority of this Mn is potentially the result of HCl dissolving birnessite On
the basis of electron transfer stoichiometry at least 15 mols of Mn(II) must be produced
for each mol of Cr(VI) generated This amount can be accounted for in the HCl
extraction where gt4 micromoles of Mn were measured Additionally Fendorf et al (1993)
determined that Mn(II) should not significantly influence the extent of Cr(III) oxidation
Iron in solution does not progressively increase in solution over time (Figures 35
36 and 37) The high initial rate of Fe release compared to Cr may be the result highly
reactive sites preferentially releasing Fe Although no Fe is detected in solution from
the extractions performed on the chromite and birnessite solids 0005 to 0035 moles of
Fe were retrieved from the lining of the bottle and 0053 to 0126 moles were released
from the solids using the HCl solution (Table 34) Iron(II) could potentially be oxidized
to Fe(II) in the presence of birnessite and form stable Fe(III)-(oxy)hydroxides (Postma
1985 Myers and Nealson 1988 Villinski et al 2001) A ratio of two Fe for every one
Mn is necessary for the oxidation of Fe(II) and this amount is recoverable based on HCl
75
extractions with gt4 micromoles of Mn measured from the HCl extractions Additionally
dissolved O2 in the batch experiments will rapidly oxidize Fe(II) for all the pHs examined
in these experiments (Stumm and Morgan 1981 Langmuir 1997) The oxidative
pathway for Fe(II) provides an explanation for the lack of Fe(II) in solution for the batch
experiments (Figure 35 36 and 37)
With a detection limit of ~02 microM and equilibrium with respect to Fe(OH)3(am) and
goethite Fe concentrations should either be at or below detection limits at pHs gt 5
Assuming aqueous Fe(III) is present dominantly as the second hydrolysis product at pH
5 the dissolved concentrations approach the solubility of ferrihydrite and are much
greater than expected for goethite (Figure 311) Thus it is probably that the former solid
is generated which is further supported by the chemical extraction experiments (Table
34)
Iron and Cr solution concentrations measured experimentally are not consistent
with congruent dissolution (Figure 312a) Assuming Fe was forming secondary phases
at a constant rate (~00001 Fe h-1) multiplying this rate by the time analysis and then
adding it to the measured aqueous Fe a corrected Fe (Fe) value can be estimated (Figure
312) Additionally the 10 of Cr not in solution can be also be used to calculate a
corrected Cr value (Cr) (Figure 312) When secondary phases are taken into account a
majority of the analyses plot near the congruent dissolution of chromite This provides
evidence that chromite is undergoing congruent dissolution within the batch experiments
but that secondary Fe(III) phases disrupt the solution phase elemental ratios
Chromium(VI) Pathways from Chromite and Birnessite into Solution
Three pathways are evaluated for Cr(VI) measured in solution a) chromite
dissolution b) birnessite dissolution and c) solid-solution reaction (Figure 313) In order
to describe these pathways end-member chromite (FeCr2O4) and birnessite (δ-MnO2) are
used to simplify the analysis It should be recalled however that the composition of the
chromite used in the experiments is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4
Chromite dissolution presents a likely explanation for Cr(VI) observed in solution
based on the following discussion In chromite dissolution model Cr(III) released from
chromite adsorbs onto the surface of birnessite and oxidizes to Cr(VI) After the electron
76
transfer is complete Cr(VI) desorbs from birnessite and remains in solution As this
process continues Cr(VI) increases at a constant rate The rate limiting step with regards
to this specific pathway can be grouped into two steps a) dissolution of chromite (Step
1a) or b) Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) (Figure 313)
Although a dissolution rate for chromite at ambient temperatures and pressures is not
known the chromite extraction experiments (Figure 32) demonstrates that chromite
dissolution is slow Manganese oxides such as birnessite oxidize Cr(III) in solution
rapidly with a measured rate constant of 6300 microM-1 s-1 in fresh water (100 mg L-1 MnO2
100 g L-1 Cr(III) pH~6) (Van der Weijden and Reith 1982) This means that the rate
for Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) is on the order of
seconds therefore chromite dissolution is likely the rate limiting step controlling the
concentration of Cr(VI) in solution for this pathway
Although a zeroth-order dependence related to birnessite concentrations (Figure
37a) is observed the birnessite dissolution pathway in which aqueous Mn4+ released
from birnessite adsorbs onto the surface of chromite oxidizing Cr(III) cannot be entirely
dismissed Manganese dissolution and adsorption could occur quickly not allowing the
Mn to be measured in solution Examination of the solids after the termination of each
experiment demonstrated extremely minimal chromite and birnessite interaction
therefore the solid solution reaction is not a likely pathway leading to the production of
Cr(VI) in solution and is furthermore inconsistent with the noted rate dependencies
Despite the three pathways (Figure 313) the generation of Cr(VI) observed in these
experiments and described by the rate equation (Equation 33) is independent of these
potential pathways
Serpentine Soils and Cr(VI)
The serpentine soil experiments (Figure 310) provide direct evidence that Cr(VI)
is forming within these soils By adding birnessite to the soil the rate of Cr(VI) in
solution increased for all experiments (00013 to 00044 microM Cr(VI) h-1 for Jasper Ridge
and 00009 to 00013 microM Cr(VI) h-1 for New Caledonia) The Cr(VI) rate increase
related to the addition of birnessite is likely the result of low birnessite suspension
densities (lt20 m2 L-1) in the soil Rates predicted for Jasper Ridge and New Caledonia
77
soils using Equation 33 and values in Table 32 are similar to those measured in the
experiments with values of 00041 and 00016 microM Cr(VI) h-1 respectively Overall these
soil experiments demonstrate that a rate equation developed from laboratory batch
experiments predicts the rate of Cr(VI) in serpentine soils reasonably well
According to limits set by the Environmental Protection Agency the California
standard for total Cr in drinking water is 50 microg L-1 (~1 microM) while the federal standard is
100 microg L-1 (~2 microM) The serpentine soil and birnessite experiments produced Cr(VI)
concentrations in excess of 50 microg L-1 after only 140 h or roughly 6 days Serpentine soils
at Jasper Ridge typically remain water saturated for prolonged periods of time (gt1 week)
during the winter season increasing the time Cr(VI) can be produced The degree and
duration of water saturation is dependent on the topography of the serpentinite body
level areas will remain water saturated while sloping areas will allow the water to flow
down gradient Although Cr(VI) was generated in the soil experiments the production of
Cr(VI) in serpentine soils may be more limited The reduction of Cr(VI) to Cr(III) is
favorable by reactions with Fe(II)aq Fe(II)-containing minerals such as magnetite H2S
and organic matter including microbial processes (Ilton and Veblen 1994 Fendorf 1995
Peterson 1996) Both organic matter and magnetite are abundant in Jasper Ridge
serpentine soils (Chapters 2 and 4) however these factors did not deter Cr(VI) from
forming during the serpentine soil experiments Additionally the soils in these
experiments were constantly rotated allowing the mineral surfaces to be continuously
exposed to solution Chromite within an undisturbed soil will not always be in contact
with the solution or proximal to birnessite
Chromium(VI) produced by the interaction of chromite and birnessite is not
limited to serpentine soils Serpentinites the bedrock for serpentine soils are typically
highly sheared due to the diapiric migration of serpentinite masses along zones of
weakness in fault zones Numerous cracks and lineations which are typical of these rocks
provide a means for Cr(VI) to transported away from the serpentine soil potentially into
local groundwater In fact preferential flow will occur along the major shear plains of
the serpentinite Additionally these shear plains are often coated with high valent Mn
oxides providing a reaction pathway for Cr(VI) formation Serpentinite springs such as
those evaluated by Barnes and ONeil (1969) and Barnes et al (1972) at the Cedars
78
ultramafic body in Cazadero CA have Cr concentrations measured at 20 microg L-1 in which
all the Cr is in the form of Cr(VI) (Chapter 4) In conclusion chromite and birnessite in
serpentine soils could contribute a source of Cr(VI) in surface waters and groundwaters
Chromium release and oxidation from chromite is increased by the presence of
birnessite and the interaction between chromite and birnessite is a pathway for Cr(VI)
generation in ultramafic sediments and soils The potential of serpentine soils as a source
of non-anthropogenic Cr contamination can now be evaluated quantitatively based on the
abundance of chromite The overall rate law for Cr(VI) in solution (Equation 39)
provides a means to estimate the Cr(VI) concentration attainable for serpentine soils and
serpentine sediments when the chromite and birnessite suspension densities and the pH of
the soil are known
ACKNOWLEDGEMENTS
We gratefully acknowledge comments and assistance from Robert Jones and Dr
Guangchao Li Funding for this research was provided by the Office of Teaching and
Learning at Stanford University
79
Table 31 Chromite and birnessite batch experiment parameters with chromite and birnessite suspension densities
Experiment Chromitea (grams)
Birnessiteb (grams) pHc
Chromite Surface Aread
(m2 g-1)
Birnessite Surface Aread
(m2 g-1)
Chromite Suspension
Density (m2 L-1)
Birnessite Suspension
Density (m2 L-1)
A 08 043 5 025 125 02 54 B
08 086 5 025 125 02 108C 03 086 5 025 125 008 108D 008 086 5 025 125 002 108E 004 086 5 025 125 001 108F 03 086 3 025 125 008 108G 03 086 67 025 125 008 108H 03 086 8 025 125 008 108
Chromite 08 ---- 5 025 125 02 ----Birnessite ---- 043 5 025 125 ---- 5480
a Chromite chemical composition is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 b Synthetic birnessite (δ-MnO2) c10 mM acetate buffered solution pH = 5 non-buffered pH=3 5 mM MES buffer pH 67 non-buffered pH 8 dMeasured BET analyses
81
Table 32 Serpentine soil and birnessite experiments with soil and birnessite masses used and the estimated chromite suspension density
Samples Soil Mass (g)
Birnessite Mass (g)
Volume (L) of Solutiona
Total Crsoil
b (mg kg-1)
Chromite Mass (g) in
Soil Samplec Estimated Surface
Areac (m2 g-1)
Chromite Suspension Density
(m2 L-1) in Soild Jasper Ridge
(JR1) 10 ---- 01 5976 013 015 02 Jasper Ridge (JR1 w Bt) 10 01 01 5976 013 015 02 Jasper Ridge
(JR2) 10 ---- 01 4978 011 015 016 Jasper Ridge (JR2 w Bt) 10 01 01 4978 011 015 016
New Caledonia (NC) 10 ---- 01 827 0015 006 001
New Caledonia (NC w Bt) 10 01 01 827 0015 006 001
a1mM Acetate Buffer (pH=5) 81 bTotal Cr concentrations were obtained through total digestion and ICP-AES analyses
cEstimated chromite mass and surface area (discussed in the Methods section) was made using chromite observations and calculations in Chapters 2 and 4
dCalculated from the chromite mass in the soil sample estimated surface area of the chromite and the volume of solution
82
Table 33 Summary of Cr(VI) rates obtained from batch experiments (Figure 35 36 and 37) and used in Figure 38
Listed in Table 31
Suspension Densities (m2 L-1)
Chromite (Ch)
Birnessite (Bt)
pH Rate (microM h-1) (d[Cr(VI)]dt)
log Rate (log microM h-1) (log d[Cr(VI)]dt) Expression
Figure 35 02 108 5 00041 -239
008 108 5 00021 -268 002 108 5 00009 -305 001 108 5 00005 -330
1
Figure 36 02 54 5 00041 -239 02 108 5 00041 -239 2
Figure 37 008 108 3 00062 -221 008 108 5 00021 -268 008 108 67 00004 -340 008 108 8 0 ----
3
Expression 1 log d[Cr(VI)]dt = asdot log Chromite +log krsquorsquo log d[Cr(VI)]dt = 06908 log Chromite ndash 1899 Expression 2 log d[Cr(VI)]dt = bsdot log Birnessite + log krsquorsquorsquo
log d[Cr(VI)]dt = 0sdot log Birnessite ndash 23872 log d[Cr(VI)]dt = -23872
Expression 3 d[Cr(VI)]dt = 264E-4sdotpH2 ndash 414E-3sdotpH + 162E-2
82
Table 34 Extractions comparing total solution and extractable Cr and Fe from the solids
------------------Cr----------------- ------------------Fe---------------- microMolesdagger Total
mole Extraction
mole microMoles Total mole
Extraction mole
Chromite Ch 02 (Table 31) Terminated at 30 days Solutiona 0033 554 ---- 0 0 ---- Phosphateb 0000 05 1 0000 03 03 CaCl2
c 0013 388 87 0000 00 00 HCld 0002 53 12 0103 997 997 Bottle HCle 0000 00 0 0000 00 00
A Ch 02 Bt 54 (Table 31) Terminated at 30 days Solution 2977 910 ---- 0 0 ---- Phosphate 0048 16 18 0004 21 21 CaCl2 0055 18 20 0002 09 09 HCl 0161 54 60 0126 758 758 Bottle HCl 0004 01 2 0035 212 212
B Ch 02 Bt 108 (Table 31) Terminated at 30 days Solution 2837 914 ---- 0 0 ---- Phosphate 0039 14 16 0008 49 49 CaCl2 0060 21 25 0000 00 00 HCl 0142 50 58 0127 803 803 Bottle HCl 0004 01 2 0024 148 148
C Ch 08 Bt 108 (Table 31) Terminated at 17 days Solution 0877 894 ---- 0 0 ---- Phosphate 0020 23 22 0002 37 37 CaCl2 0030 34 32 0000 00 00 HCl 0040 46 43 0053 881 881 Bottle HCl 0003 03 3 0005 82 82
daggerMolarity values are normalized to the final solution analysis of each experiment based on volume of solution and mass of the sample in each experiment aSolution refers to the final solution analysis before the experiment was terminated bmicroMoles extracted from the batch experiment solids using a 1 mM phosphate buffer pH 7 3 h cmicroMoles extracted from the batch experiment solids using a 10 mM CaCl2 solution pH 7 3 h dmicroMoles extracted from the batch experiment solids using a 05 N HCl solution 3 h emicroMoles extracted from the batch experiment bottles using a 05 N HCl solution 15 h
83
Figure Captions
Figure 31 Scanning electron microscopic (SEM) images of chromite and birnessite at
varying magnifications prior to the batch experiments
Figure 32 A summary of Cr Fe and Mn released from chromite during extraction
experiments Solutions used for these experiments include 1) 10 H2O2 solution 2) 01
M KMnO4 solution 3) 10 M HF 4) 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 5) 1mM
BaCl2 solution () Fe concentration for the HF extraction is 106 mg L-1
Figure 33 pH and Eh measurements as a function of time (h) for the batch
experiments Chromite (Ch) and birnessite (Bt) suspension densities used for the batch
experiments (Table 31) are listed adjacent their corresponding symbol in the key
Figure 34 Eh and pH batch experiment analyses (Figure 33) plotted on a stability field
diagram for Cr (log a Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom
(1998)
Figure 35 Batch experiments evaluating variable chromite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr and Cr(VI) measured in solution are noted and illustrated with a dotted line and
included in Table 33
Figure 36 Batch experiments evaluating variable birnessite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr(VI) measured in solution are denoted and included in Table 33
Figure 37 Influence of pH on Cr(VI) generation by reaction of constant chromite (Ch
08 m2 L-1) and birnessite (Bt 108 m2 L-1) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key Rates of Cr(VI) measured in solution are included in Table
33
84
Figure 38 Rates of Cr(VI) in solution (Table 33) as a function of a) chromite
suspension density (Figure 35) b) birnessite suspension density (Figure 36) and c) pH
(Figure 37) The line in each diagram represents the best fit for these data points and the
equations for these lines are listed in Table 33 The dotted line deviating from the
dashed line in b) indicates the probable path at low birnessite concentrations
Figure 39 SEM images of the solid phases after the completion of the batch
experiments A (Ch 02 Bt 54) and B (Ch 02 Bt 108) are shown Chromite is the larger
grain with relatively flat surfaces whereas birnessite is the fine-grained rosettes located
adjacent the larger chromite grains
Figure 310 Chromium and Cr(VI) concentrations (microM) in solution as a function of
time (h) from serpentine soils obtained at Jasper Ridge (JR1 and JR2) and New Caledonia
(NC) Soils with the addition of 01 g birnessite are noted in the figure Soil and
birnessite masses used for each experiment are listed in Table 32
Figure 311 Iron analyses as a function of time (Figure 35 and 36) with the solubility
of amorphous Fe(OH)3 and goethite and the dominant Fe3+-OH complex at pH 5 Fe
solubilities are from Langmuir (1997) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key
Figure 312 a) Experimental values of Fe and Cr (Figures 35 and 36) are plotted A
dotted line denotes the ratio of Fe and Cr necessary for chromite congruent dissolution b)
Corrected FeCr ratios as a function of time (h) for data from Figures 35 and 36 taking
into account the concentration of Fe and Cr in secondary phases calculated from Table
34 c) Corrected Fe and Cr as compared to values predicted with congruent dissolution
of chromite
85
Figure 313 Three potential pathways of Cr release and oxidation for chromite-
birnessite interactions are listed
86
Figure 31
87
0
2
4
6
8
10
12
Hydrogen
Peroxid
e
K-Perman
ganate
Hydroflu
oric Acid
Hydroxy
lamine
Barium Chlorid
e
Con
cent
ratio
ns (m
g kg
-1)
Cr
Fe
Mn
Figure 32
88
Time (h)0 100 200 300 400 500
pH
2
3
4
5
6
7
8
9
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
Time (h)0 100 200 300 400 500
Eh (m
V)
100
200
300
400
500
600
Ch Bt ------(m2 L-1)------
Figure 33
89
pH2 3 4 5 6 7 8 9
Eh (m
V)
-05
00
05
10
15
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
HCrO4-
CrO42-
Cr3+
CrOH2+
Cr(OH)3Cr(
OH
) 2+
Upper limit of water stability
Lower limit of water stability
Ch Bt ------(m2 L-1)-----
Figure 34
90
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 108 008 108 002 108 001 108 02 --- --- 54
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
0
1
2
3
4
pH 5 pH 5
pH 5 pH 5 pH 5
Ch Bt ---(m2 L-1)---
Figure 35
91
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 54 02 108
00041 microM h
-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
00
02
04
06
08
10
00041 microM h
-1
pH 5
pH 5 pH 5 pH 5
pH 5 Ch Bt----(m2 L-1)----
Figure 36
92
Time (h)
0 50 100 150 200 250
Cr c
once
ntra
tion
( microM
)
00
02
04
06
08
10
pH 3pH 5pH 67pH 8
Time (h)0 50 100 150 200 250
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250M
n co
ncen
trat
ion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
00004 microM h-1
00021 microM h-1
0006
2 microM h
-1
00021 microM h-100
062 micro
M h-1
00004 microM h-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1
Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Figure 37
93
log Birnessite m2 L-1
05 10 15 20 25 30
log
d[C
r(VI
)]dt
microM
h-1
-40
-35
-30
-25
-20
pH 5
log Chromite m2 L-1-22 -20 -18 -16 -14 -12 -10 -08 -06
log
d[C
r(VI
)]dt
microM
h-1
-34
-32
-30
-28
-26
-24
-22
pH 5
pH1 2 3 4 5 6 7 8 9
d[C
r(VI
)]dt
microM
h-1
0000
0002
0004
0006
0008
0010
a)
c)
b)
Ch 008 m2 L-1 Bt 108 m2 L-1
Figure 38
94
Figure 39
95
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR1JR1 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR2JR2 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
01
02
03
04
05
NCNC w Bt
00044 microM h-1
00013 microM h-1
00044 microM h-1
00013 microM h-1
00013 microM h-1
00009microM h-1
pH 5
pH 5
pH 5
Figure 310
96
Time (h)0 100 200 300 400 500
log ΣF
e (aq)
microM
-10
-8
-6
-4
-2
0
2
4
02 54 02 108 008 108 002 108 001 108
Fe(OH)3(am)
Goethite
pH 5Ch Bt----m2 L-1----
Fe(OH)2(aq)+
Figure 311
97
Time (h)
0 100 200 300 400
FeC
r Rat
io
0
1
2
3
4
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
02 54 02 108 008 108 002 108 001 108
Chrom
ite C
ongr
uent
Dissolu
tion
Chrom
ite C
ongr
uent
Dissolu
tion
a)
b)
c)
Ch Bt----(m2 L-1)----
ChromiteCongruent Dissolution
Figure 312
98
Pathway 1 Chromite Dissolution
1a) Cr dissolution from chromite FeCr2O3(s) + 4H+ rarr 2Cr(OH)2+
aq + Fe2+(Oxyhydroxide) + H2O
1b) Adsorption of CrOH2+ on the surface of birnessite Cr(OH)2+
aq + MnO2(s) rarr Cr(III)-MnO2(s) + OH- 1c) Oxidation of CrOH2+ on the surface of birnessite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
1d) Desorption of Cr(VI) from birnessite Cr(VI)-Mn(II) + 4H2O rarr Mn2+
(aq) + HCrO4-(aq) + 7H+
Pathway 2 Birnessite Dissolution 2a) Mn4+ dissolution from Birnessite MnO2(s) + 4H+ rarr Mn4+
(aq) + 2H2O 2b) Mn4+ adsorption on the surface of chromite Mn4+
(aq) + FeCr2O4(s) rarr Mn(IV)-FeCr2O4 2c) Oxidation of Cr(III) on the surface of chromite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
2d) Desorption of CrO24- from chromite
Cr(VI)-Mn(II) + 4H2O rarr Mn2+(aq) + HCrO4
-(aq) + 7H+
Pathway 3 Solid Solution Reaction 3) Solid solution interaction between chromite and birnessite
Figure 313
99
CHAPTER 4
CHROMIUM GEOCHEMISTRY OF SERPENTINE SOILS
ABSTRACT
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
chemistry of Cr in serpentine soils including its protolith is reviewed focusing on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils are slightly acidic (average pH of ~6) contain a variety Fe(III) oxides
(magnetite and hematite) phyllosilicates (serpentine and chlorite) and clays (smectites
and vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1)
and Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Chromium(III) is the only valence state observed
in the serpentine soil solids however Cr(VI) has been identified in New Caledonia and
California serpentine soil solutions at concentrations below 30 microM The enrichment and
range of Cr concentrations in serpentine soils are directly related to the presence of Cr-
spinels specifically chromite and Cr-magnetite These spinels are resistant to weathering
and are preserved in the soil environment however oxidation of Cr(III) from Cr-spinels
by high valent Mn oxides or other strong oxidants is a potential source of Cr(VI)
identified in serpentine soil solutions Due to the chemically and physically resistant
nature of the Cr-spinels Cr-bearing silicates including clay minerals Cr-chlorite Cr-
garnet Cr-mica and Cr-epidote are more viable sources of Cr identified in vegetation
soil extractions soil solutions and related waters
100
INTRODUCTION
Serpentine soils a generic term used to describe any soil derived from an
ultramafic rock (peridotites and pyroxenites) or serpentinite (metamorphosed ultramafic
rock) are often characterized by 1) low concentrations of plant nutrients such as N P
and K 2) high concentrations of biologically toxic elements including Ni Co and Cd 3)
low ammonium acetate extractable CaMg ratios and 4) unique flora and physical
properties (Brooks 1987) Measured Cr concentrations in serpentine soils range from
~200 mg kg-1 to 6 Wt values enriched compared to non-serpentine soils lt200 mg kg-
1 (Kabata-Pendias and Pendias 1984 Schwertmann and Latham 1986 Brooks 1987)
Chromite and Cr-magnetite are common primary sources of Cr in these soils (Rabenhorst
et al 1982 Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989
Gasser and Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003)
however the plethora of factors relating the chemistry of Cr in serpentine soils and the
weathering of chromite has not been fully addressed using an integrative approach
Overall serpentinites and serpentine soils cover ~1 of the earthrsquos exposed
surface and these rocks and soils are commonly associated with ophiolite complexes
(Coleman 1977) Serpentinites and serpentine soils related to ophiolite complexes are
the primary interest of this paper A map illustrating the worldwide distribution of major
ophiolitic serpentinite bodies and their related soils is shown in Figure 41 A majority of
the sites are near convergent boundaries (Figure 41) Although these rocks and soils
cover a minimal area of the earthrsquos surface they are present in populated areas within the
Circum-Pacific Margin and the Mediterranean Due to the wide-distribution and
occurrence of serpentinites and serpentine soils weathering of these rocks and soils will
differ from location to location due to varying climatic conditions as well as the nature of
the parent material and other factors including topography biota and time (Proctor and
Nagy 1992) As a result the chemistry of these soils as well as the nature of Cr in the
serpentinites and serpentine soils will vary from site to site potentially allowing Cr to be a
source of non-anthropogenic contamination
The objective of this paper is to review the chemical characteristics of serpentine
soils globally evaluate mineralogical and compositional variations between several soils
derived from ultramafic sources and assess the geochemistry of Cr in diverse soil-
101
forming environments Additionally chemical and spectroscopic analyses were
performed on soils derived from ultramafic rocks collected from 1) New Caledonia
South Pacific 2) Eight Dollar Mountain Selma OR 3) Nickel Mountain Riddle OR 4)
Harvard Mine Jamestown CA 5) Jasper Ridge Biological Preserve Stanford CA and 6)
Pillikin Mine El Dorado CA shown geographically in Figure 41 A basalt-derived soil
from Hawaii is included in this study in order to compare a soil formed from a mafic
protolith to those from ultramafic protoliths The various field sites are shown in Figure
42 These soils allow a comparative examination of how soil-forming factors such as
climate topography biota parent material and time affect the chemistry of Cr
REVIEW
Chromium Geochemistry
Chromium is a group VIB transition metal with two common oxidation states
Cr(III) and Cr(VI) differentiated by noticeably different chemical characteristics and
toxicities It should be noted that Cr(II) has been observed in lunar basalt olivines and in
high temperature terrestrial olivines (Sutton et al 1993 Li et al 1995) however this
oxidation state is not observed in soil environments A majority of Cr commonly persists
in ultramafic rocks serpentine soils and related waters as Cr(III) a non-hazardous
species and a micronutrient in human nutrition (Daugherty 1992 Cohen et al 1993
Fendorf 1995 James et al 1997) Chromium(III) has the highest crystal field
stabilization energy for all transition metals strongly favors octahedral coordination and
has 3 d-electrons with a high spin state Chromium(III) is strongly hydrolyzed in
aqueous solutions and the predominant species based on pH redox potential and O2(aq)
are shown in Figure 43 Chromium(III) species are more prevalent under reducing
conditions and its mobility is increased only under strongly acidic conditions At pH
values greater than 9 the hydrolyzed Cr(III) species is negatively charged demonstrating
the amphoteric properties of Cr(III) (Figure 43) Due to the factors described above
Cr(III) forms low solubility compounds and adsorbs strongly on mineral surfaces and
organic matter at pHgt4 (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) For
example the adsorption of Cr(III) on HFO (hydrous ferric oxide) as a function of pH is
shown in Figure 43 illustrating an adsorption edge at a pH of ~4 (Leckie et al 1984
102
Stumm 1992) Chromium(III) oxidation to Cr(VI) has been observed in reactions
involving hydrogen peroxide (H2O2) and high valent Mn oxides (Eary and Rai 1987
Johnson and Xyla 1991 Fendorf et al 1992 Fendorf et al 1993 Rock et al 2001)
Chromium(VI) is a strong oxidizing agent a highly corrosive substance a toxin to
living cells and a Class A human carcinogen by inhalation (Daugherty 1992 Cohen et
al 1993 Fendorf 1995) The hydrolyzed Cr(VI) species are also shown in Figure 43
with respect to pH redox potential and O2(aq) Chromium(VI) is more favorable under
oxidizing conditions and its mobility is enhanced under basic conditions In solution
Cr(VI) is a weakly adsorbing (oxy)anion (HxCrO42-x) on common weathering products
meaning that it forms highly labile complexes that can be readily displaced by ions
including phosphate and sulfate (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Additionally the adsorption of CrO42- on HFO shown in Figure 43
demonstrates the anionic character in which the mol bound of CrO42- increases with
decreasing pH and an adsorption edge is at pH ~72 (Leckie et al 1980 Stumm 1992)
The reduction of Cr(VI) to Cr(III) is known to occur by reactions with Fe(II)aq Fe(II)-
containing minerals such as magnetite H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996)
Mineralogy of Serpentine Soils
Serpentine soils have been identified on every continent excluding Antarctica
(Figure 41) The occurrence and distribution of ultramafic rocks specifically related to
ophiolites are closely associated with convergent plate boundaries (Figure 41 Coleman
1977 Coleman and Jove 1992) Tectonic blocks of ultramafic rock are hypothesized to
be detached from the down-going slab of oceanic crust and incorporated into the
subduction meacutelange Subsequent serpentinization of peridotite forms low density
serpentine group minerals (245-250 g cm-3) resulting in the upward diapiric migration of
the serpentinite masses along zones of structural weakness (fault zones) and through the
denser enclosing subduction meacutelange (~265 g cm-3) For example North America has
numerous exposed ophiolites containing serpentinite located in western California and
Oregon where subduction zones were once active in the Early Cretaceous (Harden
1998)
103
Serpentinization of ultramafic rocks is a significant factor controlling mineral
assemblages and textures in serpentinite Hydration and oxidation of olivine and
pyroxene during serpentinization forms serpentine (lizardite chrysotile and antigorite)
talc chlorite tremolite brucite and magnetite (Coleman 1977 Brooks 1987 Coleman
and Jove 1992 Malpas 1992) Additionally magnetite formation creates favorable
conditions for native metal compounds (such as Fe Co and Ni alloys) to form (Coleman
and Jove 1992) As a serpentinite mass is emplaced the extensive physical and chemical
alteration of the serpentinite results in numerous crosscutting alteration veins and highly
sheared textures Ultimately the mineral assemblages and textures in serpentinites are
complex prior to pedogenesis due to the multiple phases of metamorphosis tectonic
deformation and weathering of the rock
Serpentine soils consist of a mixture of residual minerals from the soil protolith
eolian and illuvial minerals and minerals formed within the soil Additionally highly
weathered serpentine soils enriched in Fe-oxides are also termed laterites soils common
in tropical and subtropical regions Soil protolith minerals may include primary minerals
of ultramafic rocks including olivines pyroxenes and oxides (chromite and magnetite)
as well as a large variety of metamorphic minerals the most common being serpentine
(lizardite chrysotile and antigorite) chlorite talc and actinolite Silica phases occur as
eolian grains concretions andor net type structures or silica webs (Hotz 1964
Coleman 1977 Schwertmann and Latham 1986) Eolian and illuvial aluminosilicate
minerals are important sources of Al in serpentine soils an element of low abundance in
ultramafic rocks
Layer silicates observed in serpentine soils are varied in type and abundance
They include smectites (often as montmorillonite) vermiculite chlorite brucite
serpentine and a variety of intrastratified clays (Veniale and Van der Marel 1963 Hotz
1964 Schellmann 1964 Wildman et al 1968 Rabenhorst et al 1982 Schreier et al
1987 Graham et al 1990 Bulmer et al 1992 Gasser and Dahlgren 1994 Lee et al
2003) Compositional variations crystal structures and parageneses of layer silicates
have been characterized for a number of serpentine soils (Wildman et al 1968
Rabenhorst et al 1982 Schreier et al 1987 Graham et al 1990 Bulmer et al 1992
Lee et al 2003) In addition to the layer silicates goethite hematite maghemite and
104
amorphous Fe (oxy)hydroxides are soil-forming phases in serpentine soils (Coleman and
Jove 1992) Gibbsite is not a common phase in most serpentine soils however limited
isomorphic substitution of Al(III) in goethite and hematite was noted in New Caledonia
(Schwertmann and Latham 1986) Manganese oxides such as birnessite (δ-MnO2) are
also present in serpentine soils and often coat mineral surfaces
Sources of Chromium for Serpentine Soils
Chromium primarily occurs in oxide and silicate phases in ultramafic rocks and
serpentinites Chromium-spinel often referred to as chromite is a common primary
(igneous) spinel in mafic and ultramafic rocks and often resists low-grade metamorphic
processes related to serpentinization (Hoffman and Walker 1978 Malpas 1992)
Isomorphic substitution of Al3+ Fe3+ and Ti4+ into the octahedral sites and Mg2+ Ni2+
Zn2+ and Mn2+ into the tetrahedral sites is common however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987b Pan and Fleet 1991 Sack and Ghiorso 1991 Burkhard 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996 Mathiesen 1999) Serpentinite chromite compositions reported in the literature are
plotted in Figure 44 to demonstrate the wide-range of Cr concentrations possible for Cr-
spinels Isomorphic substitution of octahedral Cr and octahedral Al for these chromite
compositions are also shown in Figure 44 Serpentinite pyroxenes containing Cr include
augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) (Grapes 1981 Von Knorring et al
1986 Treloar 1987a Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996)
Magnetite and serpentine minerals may incorporate varying amounts of Cr into
their structures during serpentinization Chromium(III) from primary Cr-containing
phases may isomorphically substitute into the Fe(III) site of magnetite due to size
(octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge similarities Chromium
concentrations in magnetite have been observed as high as 13 Wt (Oze et al 2003)
Chromate minerals such as crocoite (PbCrO4) are extremely rare in serpentinites and the
occurrence of these minerals are mainly limited to specific localities such as in Australia
Tasmania and the Urals (Williams 1974 Crane et al 2001) Serpentine minerals have
105
been reported to contain as much as 1500 mg kg-1 Cr based on spectrographic analyses
by Page and Coleman (1967) however serpentine minerals typically contain only trace
amounts of Cr
Hydrothermal and CO2 metasomatism of ultramafic rocks during the evolution of
convergent plate margins commonly produce Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite lt19 Cr Wt ) uvarovite (Cr-garnet lt20 Cr Wt )
tawmawite (Cr-epidote lt13 Cr Wt ) and kaumlmmererite (Cr-chlorite lt10 Cr Wt ) as
shown in Figure 44 (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974 Neiva
1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981 Max et al
1983 Nutman et al 1983 Raase et al 1983 Von Knorring et al 1986 Kerrich et al
1987 Treloar 1987b Treloar 1987a Morand 1990 Pan and Fleet 1991 Schandl and
Wicks 1993 Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995
Mathiesen 1999 Proenza et al 1999) Figure 44 illustrates the relationship of Cr and
Al substitution for Cr-silicate compositions based on Cr and Al Wt Additionally Cr-
silicate compositions are plotted based on octahedral Cr and octahedral Al in Figure 44
Compositions for Cr-muscovite Cr-epidote and Cr-garnet plot near the tie lines
connecting Cr and Al end-member minerals Chromium isomorphically substitutes for
octahedral Al in these silicates due to similarities in charge octahedral preference and
octahedral radii (Al3+ = 061 Aring) The numerous isomorphic substitutions possible for Cr
and Al in the chlorite structure results in analyses that do not plot near the tie line
connecting clinochlore and kaumlmmererite Overall any of these minerals are potential
sources for Cr in soils produced by the weathering of serpentinites and their hydrothermal
and CO2 metasomatic derivatives
Chromium in Serpentine Soils
Chromium concentrations vary widely in serpentine soils worldwide (Table 41)
New Caledonia soils appear to be the most Cr-enriched serpentine soil with a Cr
concentration of ~60000 mg kg-1 (Schwertmann and Latham 1986) however Jaffre
(1980) report values ranging from 6300 to 80000 mg kg-1 for New Caledonia soils
Some serpentine soils have measured Cr concentrations as low as 29 and 66 mg kg-1
atypical of most values reported in Table 41 (Proctor and Woodell 1971 Brooks et al
106
1992 Jeffrey 1992) Chromium concentrations as a function of depth for serpentine soil
profiles in Maryland (Rabenhorst et al 1982) Tehama County CA (Gough et al 1989)
the Klamath Mountains CA (Hotz 1964) Newfoundland Canada (Roberts 1992)
Harvard Mine CA (this study) and Jasper Ridge CA (Oze et al 2003) are reported in
Figure 45 The pedogenic process(es) responsible for the varying Cr concentrations as a
function of depth for each soil have not been characterized however Oze et al (2003)
determined that a majority of Cr in Jasper Ridge serpentine soils is the result of
weathering-resistant residual Cr-bearing minerals from the protolith mainly chromite
and Cr-magnetite Additionally Oze et al (2003) demonstrated that soil samples from
Jasper Ridge contain higher concentrations of Cr with respect to their corresponding
bedrock due to the Cr-spinels accumulating over time The observation that Cr
concentrations decrease with depth towards the protolith is characteristic for most of the
profiles in Figure 45
In addition to examining total Cr concentrations chemical extractions performed
on serpentine soils provide additional information with regards to Cr availability and the
minerals and complexes in which Cr may reside in the soils Table 42 summarizes a
number of extraction experiments performed on serpentine soils For a majority of the
soils listed in Table 42 total Cr concentrations for the serpentine soils were not
measured however these extractions do provide a quantitative assessment of where Cr
may be bound in soil (such as silicates organic matter and secondary Fe-bearing phases)
Extractions performed by Schwertmann and Latham (1986) were able to remove ~20000
Cr mg kg-1 from the serpentine soil using a dithionite-citrate bicarbonate (DCB) solution
which preferentially dissolves crystalline Fe-oxides Solutions such as concentrated
nitric and hydrochloric acid were only able to remove a maximum of ~1400 Cr mg kg-1
from Unst serpentine soils (Shewry and Peterson 1976) which preferentially dissolve
Cr-bearing silicates and to a lesser extent the Fe-oxides Where bulk soil Cr values were
measured extraction experiments are only able to remove a small portion (lt1) of the
total Cr present in the serpentine soils (Table 42) (Gasser and Dahlgren 1994 Oze et al
2003) Based on the extractions presented in Table 42 it appears that Cr(III) resides in
minerals and phases resistant to common extraction techniques
107
In serpentine soils Fe(III) oxides and Fe(III) hydroxides provide a host of
structural and surface sites available to sequester Cr(III) released to soil solutions by
dissolution of Cr-bearing minerals during soil-forming processes Substitution of Cr(III)
for Fe(III) in Fe-oxides Fe-(oxy)hydroxide minerals and amorphous compounds is
expected based on crystallographic considerations synthesis experiments and analysis of
extracts from selective dissolution of Fe(III) compounds in serpentine soils
(Schwertmann and Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990
Gerth 1990 Gasser and Dahlgren 1994) Chromium(III) also readily sorbs on surfaces
of Fe(III) oxides and (oxy)hydroxides via inner-sphere surface complexes (Figure 43)
and forms low-solubility precipitates (Charlet and Manceau 1992) The low
concentration of Cr(III) in soil solutions has been attributed to the low solubility of
Fe(III)-Cr(III) amorphous hydroxides (Sass and Rai 1987)
Layer silicates provide structural and surface sites capable of uptaking Cr(III)
Octahedral Cr(III) in clays such as smectites and vermiculites is expected and Cr
concentrations in the clay-sized fraction can achieve concentrations of ~1000 mg kg-1 in
serpentine soils (Oze et al 2003) Sorption of Cr(III) and Cr(VI) on edge and interlayer
sites of phyllosilicates have been evaluated by Ilton and Veblen (1994) Ilton et al (2000)
and Brigatti et al (2000) indicating that layer silicates (montmorillonite chlorite biotite
phlogopite and muscovite) provide sites for both Cr(III) sorption and Cr(VI) reduction
Kaupenjohann and Wilcke (1995) concluded that Cr in serpentine soils from Bavaria is
residually bound mainly in silicate structures including clay minerals based on extraction
experiments
Soil Solutions and Related Water
The addition of water gradually changes the spatial and temporal characterization
of pore fluids in soils including the ionic strength pH and redox potential Leaching of
a soil should progressively enrich the soil with less chemically mobile elements such as
Cr and Fe Elements identified in soil solutions include Mg Ca Fe Ni and Cr and the
ionic strengths of serpentine soil solutions are typically low (lt001) (Anderson et al
1973 Johnston and Proctor 1981 Gasser and Dahlgren 1994) Soil solutions obtained
from serpentine soils using tension-free lysimetry demonstrate Cr concentrations ranging
108
from 01 to 32 micromol L-1 (Gasser and Dahlgren 1994) This study also revealed a
majority of Cr Fe Mg and Ni in extracted soil waters were in the form of colloids
suggesting that nanoparticles are an important factor of Cr translocation in soils
The pH of a soil is a dominant factor controlling the nature of many chemical and
microbial reactions including the geochemistry of Cr (Brady and Weil 1999) Ranges in
pH for serpentine soils worldwide are plotted in Figure 46 The highest soil pH was
measured in a South Africa serpentine soil with a value of 9 (Cole 1992b) whereas the
lowest pH was reported in Ireland with a value of 38 (Jeffrey 1992) This wide range of
pH is a function of the adsorbed H+ and Al3+ the base-forming cations the organic
matter and the mineralogy present within each soil The pH range of serpentine soils
shown in Figure 46 is plotted on the stability diagram in Figure 43 demonstrating that
both Cr(III) and Cr(VI) aqueous species are favorable The Eh range for the serpentine
soils in Figure 3 is based on the soils being in contact with the atmosphere as well as
being a transitional environment Although the activity of H+ varies by four orders of
magnitude the pHs of most serpentine soils are slightly acidic (pH ~6) which is near that
of average rainwater (pH~57 Figures 43 and 46) Additionally serpentine soil pH
values have been observed to increase with depth towards the protolith for a number of
soils (Rabenhorst et al 1982 Gough et al 1989 Gasser and Dahlgren 1994) The
increase in pH with respect to depth may be the result of rainwater being progressively
influenced by the mineral chemistry and hydrolysis reactions as it descends through the
soil towards the protolith
Groundwater from serpentinized ultramafic rocks have pH values ranging from 8
to 12 (Barnes et al 1972) and these waters have been divided into two categories as
shown in Figure 46 Mg bicarbonate (Mg2+-HCO3-) type and Ca hydroxide (Ca2+-OH-)
type (Barnes and ONeil 1969 Barnes et al 1972) The Mg bicarbonate waters (pH~8)
are characteristic of most surface waters and shallow groundwaters from serpentinites
and ultramafic rocks The Ca hydroxide waters (pHasymp10-11) are associated with
incompletely serpentinized peridotites and these fluids are comparatively low in Mg2+
Fe2+ Mn2+ Crtotal and SiO2total The Ca2+ in solution originates from the serpentinization
of Ca-rich orthopyroxene (Barnes et al 1972) The pH of serpentinite groundwaters are
109
plotted on the Cr stability in Figure 43 demonstrating that both Cr(III) and Cr(VI) are
potential valence states in these waters
Chromium predominates as Cr(III) in ultramafic rocks and serpentine soils
however Cr(VI) has been identified in serpentine soil solutions and serpentinite
groundwater in California and New Caledonia at concentrations of lt20 microg L-1 (Gough et
al 1989 Becquer et al 2003) Calcium hydroxide waters (pH~11) collected from the
same spring evaluated by Barnes et al (1972) in the Cedars ultramafic body (Cazadero
CA) contains 12-22 Cr(VI) microg L-1 and no Cr(III) (this study) Additionally water from
chromite-containing ultramafic rocks in the Siniktanneyak Mountain ultramafic complex
of Alaska with pH values ranging from 67 to 93 were measured with low total Cr
concentrations lt2 Cr microg L-1 (Taylor et al 1998) A mechanism for Cr(III) oxidation in
serpentine soils and related water suggested by (Gough et al 1989 Becquer et al 2003)
may be the result of high valence Mn oxides common in serpentine soils High valence
Mn-oxides such as birnessite (δ-MnO2) are naturally occurring oxidants capable of
oxidizing Cr(III) to Cr(VI) at pHlt9 (Eary and Rai 1987 Fendorf et al 1992 Fendorf et
al 1993) It is also important to note that Cr(VI) reduction is hypothesized to occur in
soils due to electron transfer involving organic matter and Fe(II)-bearing minerals such as
magnetite (Peterson 1996 Fendorf et al 2000) Whether toxic concentrations of Cr(VI)
can be present in serpentine soils and related water over geologic time is important to
consider especially when high valence Mn oxides Fe(II)-bearing minerals and organic
matter are capable of significantly influencing the oxidation state of Cr
Serpentine Soil Vegetation
Vegetation associated with serpentine soils has received considerable attention
due to the apparent infertility of these soils The inhabitability of serpentine soils for
most plants has been attributed to the imbalance of available Mg relative to Ca or low
CaMg ratios (Walker 1954) the deficiency of plant nutrients such as K N and P
(Turitzin 1991) elevated heavy metal concentrations such as Ni and Cr (Robinson et al
1935) and poor drainage (Brooks 1987) Linking plant productivity to toxic levels of
Cr in the soil has not been fully resolved (Brooks 1987 Kruckeberg 1992) however a
majority of Cr in the soils is not available for plant uptake due to being in high stable and
110
weathering resistant phases discussed in the previous sections (Table 42) In general Cr
concentrations in plants typically do not exceed 100 mg kg-1 (Brooks 1987)
Table 43 provides a global list of serpentine soil plant species and their measured
Cr concentrations with an emphasis on Cr hyper-accumulators ie plants able
incorporate higher than average concentrations of Cr A majority of these measurements
are based on the dry weight of Cr Aerobryopsis longissima from New Caledonia has the
highest reported Cr concentration with values ranging from 200 to 700 mg kg-1 (Lee et
al 1977) Other noteworthy plants include the Lophostachys villosa from Brazil (Brooks
et al 1992) Arenaria humifusa from Newfoundland (Roberts 1992) the Calluna
vulgaris from Unst (Shewry and Peterson 1976) with Cr concentrations of 203 530 and
417 mg kgndash1 respectively Even mosses such as the Funaria flavicans (Maryland)
incorporate Cr concentrations as high as 90 mg kg-1 (Shaw 1991) By examining
specific parts of the plant Gough et al (1989) determined that Cr concentrations are
highest in the basal burl (600 mg kg-1) of the manzanita bush compared to the roots (30
mg kg-1) stems (20 mg kg-1) and leaves (20 mg kg-1) The study by Gough et al (1989)
demonstrates that Cr is not homogeneously distributed throughout the plant and that
evaluating a leaf versus a stem could result in significantly different Cr concentrations
Soil Organic Matter
Soil organic matter (SOM) including the biomass (living organisms such as
bacteria) detritus (nonliving plant residues) and humus (amorphous and colloidal
mixtures of organic substances) affects the geochemistry and buffering capacity of
serpentine soils Furthermore biologic reduction of Cr(VI) to Cr(III) via cometabolic
and metabolic pathways coupled with carbon oxidation has been demonstrated for a
number of bacterial strains including aerobes (Gopalan and Veeramani 1994 Park et al
2000) and anaerobes (Turick et al 1996 Tebo and Obraztsova 1998) Additionally
aqueous Fe(II) produced by microbial Fe(III) reduction is a known biogenic pathway for
the reduction of Cr(VI) (Fendorf et al 2000) Fe(III)-Cr(III) (oxy)hydroxides are the
resulting secondary phases (Rai et al 1989 Patterson and Fendorf 1997 Hansel et al
2003) Bacteria oxidizing Cr(III) to Cr(VI) has not been observed
111
Most serpentine soils in temperate regions appear to be either Mollisols or
Alfisols which often contain high concentrations of SOM Alexander et al (1989) and
Bulmer et al (1992) reported that serpentine soils can contain organic carbon
concentrations in excess of 50 mg g-1 Humus has a high cation exchange capacity and
provides much of the pH buffering capacity in these soils (Brady and Weil 1999)
Organic acids such as fulvic and humic acids are known to complex cations such as
Cr3+ Fe3+ and Mn2+ forming stable organometallic complexes (Kaupenjohann and
Wilcke 1995 Brady and Weil 1999) Additionally metal leaching experiments
performed by Schreier et al (1987) established that organic acids produced by soil
organic matter such as oxalic acid and citric acid were extremely effective in releasing
Cr Mg Fe Ni and Co from serpentinite rocks and sediments
CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
Serpentine soils were obtained from California Oregon and New Caledonia and
analyzed to further evaluate the factors influencing the Cr geochemistry in serpentine
soils Analyses for these soils include pH electrical conductivity total organic matter
total N ammonium acetate extractable Ca K Mg and Na soil texture soil color
chemical composition of the soil and the chemical composition of Cr-bearing minerals
The general soil characteristics are described in the following section and the analytical
methods are explained in Appendix 41 Although a majority of these soils are located in
the western United States factors influencing the pedogenesis of these soils vary
considerably allowing a comparative analysis of how the fate of Cr might be influenced
by climate topography biota parent material and time
Sample Locations and General Soil Characteristics
Six serpentine soils were collected from locations shown in Figures 41 and 42 to
evaluate the chemistry of Cr in serpentine soils These soils include 1) a laterite (Oxisol)
collected in New Caledonia South Pacific 2) a laterite (Oxisol) exposed at Eight Dollar
Mountain Selma OR 3) a laterite (Oxisol) near Nickel Mountain Riddle OR 4) a
serpentine soil (Inceptisol) with a CO2-metasomatized serpentinite protolith from
Harvard Mine Jamestown CA 5) a serpentine soil (Mollisol) obtained from a preserved
112
grassland at Jasper Ridge Biological Preserve Stanford CA and 6) a serpentine soil
(Inceptisol) collected in the proximity of Pillikin Mine El Dorado CA Lastly an Oxisol
formed from basalt obtained in Oahu Hawaii is also included for comparison
These seven sites include a range of soil environments from different protoliths
under varying conditions of rainfall temperature and topography Each soil was collect
near the highest and topographically level point for each serpentine soil site Serpentine
grassland soils at Jasper Ridge formed under Mediterranean-like climates (xeric moisture
regime) whereas higher temperatures and less rainfall (ustic moisture regime)
characterize soil formation at the Harvard and Pillikin mines in the Sierra Nevada
Foothills metamorphic belt The vegetation of these three sites is dominantly grasses with
some manzanita These modern soil environments contrast with the Miocene Ni and Cr-
rich laterite soils of Nickel Mountain and Eight Dollar Mountain that were probably
formed under an aquic moisture regime near sea level in a subtropical climate Grasses
manzanita and pine trees are the dominant forms of vegetation at Nickel and Eight
Dollar Mountain An ultramafically-derived laterite formed under tropical conditions
from New Caledonia was collected The vegetation and topography related to the New
Caledonia laterite was not recorded during sampling An Oxisol from Hawaii is also
included in this study to evaluate the differences between a highly weathered soil
originating from an ultramafic rock and one from a mafic rock (basalt) under tropical
conditions The Hawaiian soil was collected from a topographically level area however
the vegetation related to this site was not noted
BedrockProtolith Mineralogy
A broad characterization of the protoliths for the serpentine soils is briefly
summarized due to the bedrock not being collected from all the field sites New
Caledonia is composed of variably serpentinized peridotites predominantly harzburgites
and some dunites (Leblanc 1995) Nickel Mountain and Eight Dollar Mountain are
predominantly serpentinized harzburgite and dunite The bedrock for soils collected at
Harvard Mine is mainly composed of quartz ankerite and mariposite (also referred to as
QAM) caused by the CO2 metasomatism of serpentinite associated with gold
mineralization (Bohlke 1989) The average serpentinite at Jasper Ridge is composed of
113
serpentine (mainly lizardite and antigorite) chlorite talc magnetite and chromite (Oze et
al 2003) however brucite has been reported as a major constituent in related Franciscan
Complex serpentinites (Hostetle et al 1966) Pillikin Mine serpentinites dominantly
consist of talc and serpentine (lizardite antigorite and chrysotile) and these serpentinites
contain podiform chromite deposits (Cater et al 1948) Mafic bedrock for the Hawaiian
Oxisol is tholeittic basalt
Soil Chemistry and Physical Properties
As shown in Figure 47 the colors of the collected soils range from the red New
Caledonia laterite to the greenish gley of the Pillikin Mine Inceptisol The Munsel color
values are also shown in Figure 47 The dark grey color of the Jasper Ridge soil and the
greenish gley of the Pillikin Mine soil are atypical of soils derived from serpentinites due
to the lack of any apparent redness The range of these colors is often attributed to the
presence of Fe-oxides however organic matter and Mn oxides are also dominant factors
controlling color (Brady and Weil 1999)
Chemical properties for the six serpentine soils and the Hawaiian soil are listed in
Table 44 The slightly acidic pH values range from 53 to 69 similar to those shown in
Figure 43 and 46 The New Caledonia laterite is the most acidic soil with a pH of 53
compared to the Harvard Mine carbonate-buffered soil at 69 The pH of the Hawaiian
soil is also slightly acidic with a pH of 604 The electrical conductivities (EC) of the
soils range from 01 to 07 dS m-1 a range indicative of relatively few dissolved inorganic
solutes The highest organic matter content is in the very dark grey Jasper Ridge soil
with a value of 32 and the lowest is in the red New Caledonia laterite (05) The
Hawaiian soil has an organic matter content of 1 Nitrogen concentrations for all the
soils are lt012 with Jasper Ridge having highest N concentration which is consistent
with the soil organic matter content
Ammonium acetate extractable Ca K Mg and Na are also listed in Table 44
Magnesium concentrations are gt6923 mg kg-1 in all of the serpentine soils The lowest
concentration of Mg was measured in the basalt-derived Hawaiian Oxisol with a value of
168 mg kg-1 Calcium concentrations in the serpentine soils range from 167 to 998 mg
kg-1 The 998 mg kg-1 of Ca in the Harvard Mine soil is directly related to the abundance
114
of ankerite (gt20 of the bedrock) Potassium concentrations are low for most of the soils
(lt55 mg kg-1) however Jasper Ridge and Harvard Mine soils have the highest
extractable K with values of 237 and 301 mg kg-1 respectively The extractable K value
for the Harvard Mine soil is due to the K-metasomatism related to gold mineralization
involving the serpentinite soil protolith The extractable concentration of Na is lt44 mg
kg-1 for all the serpentine soils whereas the Hawaiian soil has the highest Na (53 mg kg-
1) due to plagioclase and basaltic glass
Textures for the soils range from clay loam clay loam to sandy clay loam (Table
44) The laterite soil in New Caledonia and Oxisol in Hawaii are extremely different in
which the Hawaiian soil is mostly clay (66 clay) compared to the loam at New
Caledonia (25 clay) Additionally both ancient laterites (Eight Dollar Mountain and
Nickel Mountain) have clay textures Harvard Mine Jasper Ridge and Pillikin Mine
soils are all loamy in character
Major and trace element compositions for the soils are listed in Table 45
Chromium concentrations in the serpentine soils range between 827 and 9528 mg kg-1
The Eight Dollar Mountain laterite has the highest Cr concentration and the New
Caledonia laterite contains the lowest Cr concentration The Cr concentration for the
New Caledonia soil is lower than the values reported in Table 41 The Hawaiian soil has
a Cr concentration well below the analyzed serpentine soils with 388 Cr mg kg-1 but still
greater than the 200 mg kg-1 characteristic of serpentine soils Nickel concentrations in
the serpentine soils range from 1895 to 14100 mg kg-1 with the highest concentration
being identified in the Nickel Mountain laterite Manganese concentrations range
between 257 (New Caledonia) to 3017 (Eight Dollar Mountain) mg kg-1 in the serpentine
soils Element concentrations in the Hawaiian Oxisol contrast from the serpentine soils
with higher Al Cu P Pb and Zn concentrations
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in the soil solids for Jasper Ridge New Caledonia Pillikin Mine Eight Dollar
Mountain and Harvard Mine Although the Cr concentrations in these soils exceed 200
mg kg-1 (Table 45) X-ray absorption near-edge spectra (XANES) shown in Figure 48
illustrate Cr(VI) levels are below the detection limits of this method (~10 Cr mg kg-1) due
to the lack of a pre-edge feature The box in Figure 48 illustrates where a Cr(VI) pre-
115
edge peak should be present Chromium(VI) concentrations in these soils are possibly
present in the parts per billion range
Soil Mineralogy
Backscattered electron images and microprobe analyses were used to identify the
Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49
and corresponding Cr-bearing mineral compositions are listed in Table 46 The New
Caledonia soil exhibits a wide range of particle sizes and a Fe-(oxy)hydroxide was
identified containing 165 Cr2O3 Wt Chromium-spinel has been identified in the New
Caledonia laterite by Becquer et al (2003) Eight Dollar Mountain soils exhibit a coarse
texture and both Cr-magnetite (2387 Cr2O3 Wt ) and chromite (3343 Cr2O3 Wt )
are present Nickel Mountain soils are also coarse and chromite (1315 Cr Wt ) was
identified Soils at Harvard Mine contain numerous chromite grains and the mineral
trevorite (a Ni-bearing spinel) was identified with 3991 Cr2O3 Wt and 2054 NiO Wt
Chromite and magnetite are abundant in soils at Jasper Ridge as well as a fine-
grained mixture of chromite magnetite and an unidentifiable silicate with up to 1878
Cr2O3 Wt (Figure 49) Veins of Cr-chlorite crosscut the spinel-silicate mixture in
Figure 49 and this Cr-chlorite is the only Cr-silicate identified using this method
Chromite (5355 Cr2O3 Wt ) is the dominant Cr-bearing mineral detected in soils at the
Pillikin Mine and these grains contain the highest Cr concentrations compared to the
other minerals in the seven soils The Hawaiian Oxisol possesses ulvoumlspinels (007
Cr2O3 Wt and 2465 TiO2 Wt ) which are evenly distributed throughout the soil
Overall microprobe observations reveal that the Cr-bearing spinels are present in every
soil Additionally Cr-bearing spinels at Jasper Ridge have been shown to undergo
incongruent dissolution progressively enriching the spinel towards a FeCr2O4 end-
member (Oze et al 2003)
DISCUSSION
The physical and chemical characteristics of serpentine soils are directly related to
the alteration and tectonic history of the serpentinite as well as the soil-forming factors
including the parent material climate biota topography and time Identifying Cr-
116
bearing minerals their chemical compositions and abundance in the soils allows a more
definitive explanation for Cr enrichment in these soils as compared to total Cr
concentrations listed in Tables 41 and 45 and sequential extraction experiments reported
in Table 42 Detrital chromite grains from the serpentinite protolith are resistant to
weathering and accumulate in the soil over time Although serpentine soils contain Cr
concentrations typically gt200 mg kg-1 the presence of Cr-spinels and their resistance to
weathering demonstrate that a majority of the Cr is not chemically available to interact
within the soil The role of organic matter and vegetation are important factors related to
the chemistry of serpentine soils however their role for influencing a majority of the Cr
geochemistry is limited due to the nature of the Cr-spinels
Chromium and Serpentine Soils
Total Cr concentrations exceed 200 mg kg-1 for most of the serpentine soils
reported in Table 41 and in the six serpentine soils of this study (Table 45) These Cr
concentrations can be accounted for by assessing the Cr-bearing minerals associated with
these soils and their relative abundance Small grains of Cr-spinels or chromite identified
in the soils (Figure 49 and Table 46) have the potential to appreciably impact Cr
concentrations Table 47 demonstrates the mineral masses and volumes of Cr-bearing
minerals required to cause predetermined Cr concentrations in soil samples A minimal
amount of chromite (lt02 of a soil sample mass or 005 of the volume of serpentine)
can result in Cr concentrations of 1000 mg kg-1 A few grains of chromite could easily
be responsible for the Cr concentrations in Tables 41 and 45 In fact residual grains of
Cr-enriched spinels are observed in every soil analyzed (Figure 49 and Table 46)
Ultimately grains of spinel in the soil do appreciably impact Cr concentrations however
they do not provide a readily available source of Cr based on extraction experiments
performed by Oze et al (2003) Serpentine minerals could also provide minor Cr
concentrations to the soils however soil extractions have consistently shown that a
majority of the Cr is in highly resistant phases capable of resisting dissolution even in
concentrated hydrochloric acid nitric acid and hydroxylamine with acetic acid (Table
42) Chromium end-member minerals including eskolaite Cr-muscovite (fuchsite) Cr-
garnet (uvarovite) Cr-epidote (tawmawite) and Cr-chlinochlore (kaumlmmererite) are also
117
included in Table 47 to demonstrate that relatively low abundances of these minerals are
all that is necessary to increase Cr concentrations in rocks and soils Chromium-
clinochlore is often observed developing adjacent Cr-spinels in serpentinites and is
perhaps a more accessible mineral source of Cr in the serpentinites compared to Cr-
spinels (Phillips et al 1980 Schreyer et al 1981 Kerrich et al 1987 Treloar 1987a
Pan and Fleet 1991 Schandl and Wicks 1993 Christofides et al 1994 Challis et al
1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Oze et al 2003)
As shown in Table 44 and Figures 43 and 46 the chemical properties (pH EC
N etc) of serpentine soils are not correlated to Cr concentrations measured in the soil
solids (Tables 41 Figure 45) due to Cr-spinels accounting for a majority of the Cr
(Figure 49 and 410 Tables 46 and 47) However extraction experiments and soil
solution analyses demonstrate that a limited fraction of the total Cr (lt1) is extractable
If only 1 of the Cr in a serpentine soil is extractable a serpentine soil containing 10000
Cr mg kg-1 could possibly contribute up to100 Cr mg kg-1 still a significant Cr
concentration The chemical properties of the soil such as pH will influence this limited
fraction of the Cr in the soils Acidic pHs may aid in mobilizing Cr(III) however this
will only occur at low pHs (lt4) which exceed the range of most serpentine soils (Figures
43 and 46) pH values greater than 8 like those in some serpentine soils and
serpentinite groundwater may help sustain Cr(VI) concentrations in solution (Figure
43) In addition to pH organic matter in the soils may be a significant factor related to
the mobility of Cr due to the production of organic acids promoting the dissolution of the
Cr-bearing phases but particulate organic matter is known to bind Cr(III) in stable stable
organometallic complexes (Kaupenjohann and Wilcke 1995 Brady and Weil 1999)
Evaluating the valence state of Cr is crucial for understanding the chemistry as
well as the potential toxicity related to serpentine soils The soil solids do not contain
detectable concentrations of Cr(VI) (Figure 48) however Cr(VI) concentrations are
present in soil solutions and groundwater at values lt30 microg L-1 (Gough et al 1989
Becquer et al 2003) A majority of serpentinite Cr-bearing minerals only contain
Cr(III) therefore there must be a reaction or process by which Cr(III) is oxidized in a
water saturated soil High valence Mn oxides such as birnessite are present within
serpentine soils and provide a feasible explanation for Cr(III) oxidation Additionally
118
Mn oxides or other strong oxidants in the soil might accelerate the weathering of Cr-
bearing minerals including chromite and generate the Cr(VI) concentrations observed in
soil solutions and groundwater After Cr(III) is oxidized Cr(VI) is a chemically mobile
species and would be transported away from the serpentine soils before it can be
observed in the solid phase Additionally Cr(VI) may be subsequently reduced by
reacting with organic matter or Fe(II)-bearing minerals such as magnetite which are
present in a majority of serpentine soils Perhaps the best way to evaluate the potential of
serpentine soils being a non-anthropogenic source of Cr contamination is to monitor
Cr(III) and Cr(VI) in the soil solutions over time as demonstrated by Gasser and
Dahlgren (1994)
It is curious that plants do uptake Cr in serpentine soils especially when Cr is
tightly bound in the soil matrix (Table 42) In order for plants to uptake Cr it must be in
a form to be absorbed by plant roots Soil pH in the rhizosphere can be altered by plant
roots by as much as 2 units within two days (Fischer et al 1989) This acidification is
known to increase the solubility of the minerals in the soils and could serve as a
mechanism by which plants can incorporate this relatively insoluable element Another
possibility is that Cr(III) in the soil is oxidized by either organic or inorganic means and
then incorporated into the plant in its more chemically mobile and bioavailable form
Cr(VI) How and why plants incorporate as much Cr as they do needs to be examined
more thoroughly
Chromium Spinel in Soils and Rocks
Compositions of chromite identified in serpentinites worldwide (Figure 44) and
Cr-bearing spinels identified in the six serpentine soils are compared in Figure 10 in order
to determine whether the soil spinels are geochemically similar to those in serpentinites
This comparison is ascertained by examining how Al Fe and Mg Wt values varied
with respect to Cr Wt Trends for magnetitehematite chromite and the
microcrystalline spinel-silicate mixture observed in the soils are noted in each diagram in
Figure 410 Additionally the highest Cr concentrations for soil chromite
magnetitehematite and the microcrystalline spinel-silicate mixture are ~40 Cr Wt
~25 Cr Wt and ~15 Cr Wt respectively Based on Al Wt soil
119
magnetitehematite deviates significantly from the chromite trend with Al Wt values
lt5 Magnetitehematites are enriched with Fe relative to serpentinite and soil chromite
and Mg concentrations in the soil chromite are slightly higher than serpentinite chromite
Soil magnetitehematite and chromite are similar by demonstrating a linear trend when
Al Fe and Mg are summed The microcrystalline spinel-silicate mixture (CSM) does
not follow this trend because of Si present in this mineral mixture Overall soil
chromites are similar to serpentinite chromite and soil magnetiteshematites can be
differentiated from the chromites when Cr Wt is plotted against Al Fe and Mg Wt
CONCLUDING STATEMENT
Serpentine soils are abundant among convergent margins and the chemistry of
these soils is ultimately a function of the ultramaficserpentinite rock mineralogy and
texture as well as all the common soil-forming factors Although ultramafic rocks and
serpentinites produce groundwater that is highly basic most serpentine soil pHs are
slightly acidic Although Proctor and Nagy (1992) argued that it is futile to categorize
serpentine soils based on chemistry and biology (otherwise referred to as the lsquoserpentine
factorrsquo) as proposed by Brooks (1987) Cr concentrations greater than 200 mg kg-1 in
serpentine soils are consistently present Chromium in the soils originates from minerals
in the bedrock where the valence is primarily Cr(III) the exception being Cr(VI) in the
rarely present mineral crocoite (PbCrO4) Increased Cr-concentrations in serpentine soils
are largely the result of chromite or other resistant Cr-bearing minerals causing a
localized enrichment in the soils Although Cr in serpentine soils is not readily released
from the soil matrix serpentine soil vegetation can incorporate Cr concentrations as high
as 700 mg kg-1 (Table 34) Finally the limited extractability of Cr and site dependent
factors such as the presence of organic matter clays Fe(II)-bearing minerals pH and the
poor drainage indicates that the possible formation and sustained presence of Cr(VI) is
not a portentous matter related to serpentine soils
120
ACKNOWLEDGEMENTS
We appreciate technical and field support from Robert Jones Anders Meibom
Rob Dunbar and Guangchao Li Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund GSA Graduate Student Research Grants and the
Office of Teaching and Learning at Stanford University is gratefully acknowledged
121
Table 41 Measured Cr concentrations in serpentine soils worldwide
Location Cr concentration (mg kg-1) Reference
Canyon Creek area OR 920 (White 1971) 66-13200 (Brooks et al 1992)
Ireland 29-5000 (Proctor and Woodell 1971 Jeffrey 1992) Italy 1300-3900 (Gambi 1992) Japan 5200 (Shewry and Peterson 1976) Maryland 1500-6000 (Rabenhorst et al 1982) Mt Tamalpais CA 1700 (White 1971) New Caledonia 6300-80000 (Jaffre 1980) New Caledonia 2-6 (Schwertmann and Latham 1986) New Idria CA 1640 (White 1971) New Zealand 850-3250 (Lee 1992) Newfoundland Canada 309-1447 (Roberts 1992) Nickel Mountain OR 1730 (White 1971) Santa Barbara County CA 2080 (Woodell et al 1975) Southern Africa 045-14 (Morrey et al 1992) Tehama County 023-12 (Gough et al 1989) Transvaal Lowveld South Africa 120-800 (Cole 1992) Unst Shetland 3000-4000 (Proctor 1969) Wenatchee Mnts WA 895 (White 1971) Zimbabwe 110-46000 (Soane and Saunder 1959 Proctor et al 1980)
Goias State Brazil
122
123
Table 42 Summary of Cr Fe and Ni extractions from serpentine soils
Range of Extractable Cr Fe and Ni Location Cr Fe Ni
Method Reference
Italy 2-5 microg g-1 305-640 2300-2700microg g-1 25 CH3COOH (Gambi 1992)
NE Portugal 02-09 microg g-1 120-540 microg g-1 40-290microg g-1 EDTA (Menezes de Sequeira and Pinto da Silva 1992) New Caledonia 22-23 384-581 02-216 DCB (dithionite-citrate bicarbonate) (Schwertmann and Latham 1986) New Caledonia 02-11 microg g-1 02-66microg g-1 Method Unknown (Jaffre 1980)
Norway 357microg g-1 316microg g-1 HClO4HNO3 (concentrated) (Zeien and Bruemmer 1989) Norway 254microg g-1 134microg g-1 Ascorbic and Oxalate Acid pH=325 (Zeien and Bruemmer 1989)
Norway 62 microg g-1 53microg g-1 NH4-EDTA (Zeien and Bruemmer 1989) Norway 22-60 microg g-1 8-25 mg kg-1 1-7 microg g-1 pH-stat reactor EDTA (Kaupenjohann and Wilcke 1995)
Swiss Alps 08 micromol L-1 28micromol L-1 228microg g-1 Lysimeter (Nonfiltered) (Gasser and Dahlgren 1994) Trinity Ophiolite CA 004-045 gkg 27-210 g kg-1 005-245microg g-1 Citrate-Dithionite Extracts (Alexander et al 1989)
Unst lt01 microg g-1 19-77microg g-1 Ammonium Acetate (Shewry and Peterson 1976)
Unst 500-1390microg g-1 1540-3330microg g-1 HNO3-Soluble (Shewry and Peterson 1976) JRBP CA 0 microg g-1 01microg g-1 103microg g-1 BaCl2 Method (Oze et al 2003)
JRBP CA 78 microg g-1 513 microg g-1 150 microg g-1 Hydroxlyamine+Acetic Acid (Oze et al 2003) JRBP CA 31 microg g-1 2039microg g-1 35microg g-1 10M Hydrofluoric Acid (Oze et al 2003)
123
124
Table 43 Serpentine soil plant Cr concentrations
Plant Cr Concentration (microg g-1) Location Reference
Lophostachys villosa 203 Goias State Brazil (Brooks et al 1992) Ruellia geminiflora
56 Goias State Brazil (Brooks et al 1992) Minuartia ophiolitica 83 Italy (Gambi 1992) Sanguisorba minor 194 Italy (Gambi 1992) Plantago lanceolata 135 Italy (Gambi 1992) Cerastium latifolium 110 Italy (Gambi 1992) Funaria flavicans 34-899 Maryland (Shaw 1991) Avena sativa 168-583 New Zealand (Lee 1992) Aerobryopsis longissima 200-700 New Caledonia (Lee et al 1977) Arenaria humifusa 530 Newfoundland Canada (Roberts 1992) Senecio pauperculus 406 Newfoundland Canada (Roberts 1992) Rhalomitrium lanuginosum 246 Newfoundland Canada (Roberts 1992) Santolia semidentata 95 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Asplenium cuneifolium 228 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992) Armeria eriophylla 38 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Arctostaphylos viscida 029-705 Tehama County CA (Gough et al 1989) Calluna vulgaris 417 Unst (Shewry and Peterson 1976) Armeria maritima 149 Unst (Shewry and Peterson 1976) Agrostis stolonifera 120 Unst (Proctor and Woodell 1971) Dicoma niccolifera 77 Zimbabwe (Proctor 1992) Xerophyta equisetoides 68 Zimbabwe (Proctor 1992)
124
125
Table 44 Chemical and textural properties of the collected soils
Ammonium Acetate Extractable ------------------------Hydrometer--------------------- Ca K Mg Na
Sand Silt ClaySample pH EC dSm
Organic Matter ------
LECO Nitrogen ------ -------------------mg kg-1-------------------
------ ------ ------
Texture
New Caledonia 534 01 05 lt001 279 29 692 44 43 32 25 LoamHawaii 604
03 10 002 839 46 168 53 8 26 66 ClayEight Dollar Mnt 651 07 16 006 599 62 3464 25 28 28 44 Clay
Nickel Mnt 665 05 07 001 167 42 3289 17 24 32 44 ClayHarvard Mine 691 03 11 003 998 301 2897 43 41 30 29 LoamJasper Ridge 671 04 32 012 505 237 3291 36 36 31 33 Clay LoamPillikin Mine 642 04 05 002 185 55 3421 29 45 23 32 Sandy Clay Loam
125
126
Table 45 Total major and trace element concentrationsdagger of the collected soils
Element Concentration New Caledonia Hawaii Eight Dollar
Mnt Nickel Mnt Harvard Mine Jasper Ridge Pillikin
Mine Ag mg kg-1 lt05 lt05 lt05 lt05 02 lt05 lt05Al 931 1325 314 095 179 314 179Ba mg kg-1 lt10 180 40 20 30 230 lt10Be mg kg-1 05 16 05 lt05 lt05 08 lt05Ca 003 012 085 018 007 079 002Cd mg kg-1 lt05 lt05 18 lt05 05 lt05 lt05Co mg kg-1 56 183 338 403 85 155 122Cr mg kg-1 827 388 9528 5761 2900 5976 1977Cu mg kg-1 78 248 43 25 51 33 4Fe 638 1296 1635 17 538 727 468K 002 046 004 003 004 039 001
Mg 072 026 753 829 792 1025 gt150Mn mg kg-1 257 2706 3017 2554 920 1664 903Na 001 008 005 001 003 057 lt001Ni mg kg-1 2780 345 5510 14100 1895 2180 2760P mg kg-1 50 260 160 150 130 140 20Pb mg kg-1 6 31 4 3 2 8 6Ti 016 041 008 002 lt001 02 002V mg kg-1 63 132 158 54 65 92 27W mg kg-1 10 10 10 10 lt10 lt10 lt10Zn mg kg-1 43 218 160 132 66 111 47
126
dagger ICP-AES analyses
127
Table 46 Compositionsdagger for soil Cr-containing minerals identified in Figure 49
Wt New Caledonia A
Hawaii B
Eight Dollar Mnt C
Eight Dollar Mnt D
Nickel Mnt E
Harvard Mine F
Jasper Ridge G
Jasper Ridge H
Pillikin Mine I
SiO2 491 008 175 026 002 000 1940 001 000Al2O3
756 185 272 2235 2041 088 559 1575 839NiO 023 005 001 006 011 2054 014 002 007FeO 7766a 6212 6633b 3420b 1784 2445 2629 2314 3105MnO 004 037 181 123 133 031 155 146 216Cr2O3 165 007 2387 3343 4682 3991 1878 5024 5355MgO 048 121 202 653 1315 1062 1668 926 497CaO 001 000 013 008 000 000 015 000 000Na2O 000 000 002 001 000 000 006 000 000TiO2 004 2465 012 006 012 000 000 003 018
TOTAL 9257 9041 9879 9821 9980 9671 8864 9920 10037Mineral Cr-Hydroxide Ulvoumlspinel Cr-Magnetite Chromite Mg-Chromite Trevorite CSM1 Chromite Chromite
127
dagger Electron microprobe analyses This is the form of iron unless otherwise stated a FeO is in the form Fe2O3 b FeO is in the form of Fe3O4 1CSM Microcrystalline chromitesilicate mixture
128
Table 47 Chromium-bearing mineraldagger masses and volumes required to influence Cr concentrations in soils
Cr Concentration Mineral mass (grams) required for 1000 g bulk soil analysis Mineral Mass Percent () (mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite
500 11 07 22 24 48 57 011 007 022 024 048 057 1000 22 15 43 48 95 115 022 015 043 048 095 1152000 43 29 86 96 191 230 043 029 086 096 191 2305000 108 73 216 241 477 574 108 073 216 241 477 574
Cr Concentration Mineral Volume Percent () in Serpentine Mineral Volume Percent () in Quartz
(mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite500 005 004 02 02 03 15 006 004 02 02 04 06
1000 01 007 04 03 07 30 01 007 04 03 07 112000 02 014 07 06 14 59 02 015 07 06 14 225000 06 036 18 16 34 147 06 037 18 16 35 55
128
daggerMineral formulas eskolaite Cr2O3 chromite FeCr2O4 fuchsite KCr2(AlSi3O10)(OH)2 uvarovite Ca3Cr2Si3O12 tawmawite Ca2CrAl2Si3O12(OH)2 kaumlmmererite Mg5CrAlSi3O11(OH)8
Note Molar volumes used in the calculations above are from their Al-enriched counterparts in Robie and Hemingway (1995) The molar volume for kaumlmmererite was estimated using unit cell volumes from XRD unit cell measurements from Robie and Hemingway (1995) Christofides et al (1994) and Phillips et al (1980)
129
Figure Captions
Figure 41 Selected serpentine soil localities as well as soils collected for analyses are
shown on a world map Active plate boundaries are shown Serpentine soil localities not
adjacent the plate boundaries on this map are related to ancient plate boundaries that are
no longer active
Figure 42 Field pictures of soils collected from a) Eight Dollar Mountain b) Nickel
Mountain c) Harvard Mine d) Pillikin Mine and f) Jasper Ridge Biological Preserve
Figure 43 Stability field diagram of Cr under a range of Eh and pH conditions (log a
Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom (1998) Adsorption
isotherms for Cr3+ and CrO42- on hydrous ferric oxide (HFO) are also shown (Leckie et
al 1980 Leckie et al 1984)
Figure 44 Microprobe analyses demonstrating Cr versus Al Wt for Cr-spinel
(chromite) and Cr-silicates including Cr-mica (fuchsitemariposite) Cr-garnet
(uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite) (Whitmore et al
1946 Leo et al 1965 Chen and Lee 1974 Neiva 1978 Cooper 1980 Phillips et al
1980 Grapes 1981 Schreyer et al 1981 Max et al 1983 Nutman et al 1983 Raase
et al 1983 Chao et al 1986 Von Knorring et al 1986 Kerrich et al 1987 Treloar
1987b a Morand 1990 Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen
1999 Proenza et al 1999) Chromium-spinel and Cr-silicate compositions based on
octahedral Cr and octahedral Al are also shown with tie-lines (dotted) denoting ideal
isomorphic substitution between Cr and Al end-members
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above Each dotted line connecting the
points represents a separate soil profile at each site
129
Figure 46 Ranges of pH for serpentine soils and related groundwater worldwide Soil
pH values were obtained from A(Kruckeberg 1992) B (Graham et al 1990) C
(Alexander et al 1989) D (Gough et al 1989) E (Rabenhorst et al 1982) F
(Roberts 1992) G (Bulmer et al 1992) H (Proctor 1992a) I (Jeffrey 1992) J
(Gambi 1992) K (Verger 1992) L (Proctor and Cole 1992) M (Cole 1992b) N
(Morrey et al 1992) O (Proctor 1992b) P (Mizuno and Nosaka 1992) Q (Cole
1992a) and R (Lee 1992) Groundwater pH values from serpentinite are from Barnes
and ONeil (1969) and Barnes et al (1972) Note the range of pH environments possible
for ultramafic rocks and their related soils
Figure 47 Soil pictures with Munsel color values and names from soils collected from
a) New Caledonia b) Hawaii c) Eight Dollar Mountain d) Nickel Mountain e) Harvard
Mine f) Jasper Ridge Biological Preserve and g) Pillikin Mine
Figure 48 Cr XANES spectra for soils analyzed from Jasper Ridge New Caledonia
Pillikin Mine Eight Dollar Mountain and Harvard Mine The pre-edge area indicative of
Cr(VI) peaks is noted in the diagram No definitive Cr(VI) peaks are present in any of
these spectra
Figure 49 Backscattering images of mineral phases within each soil are presented
Electron microprobe analyses are noted by a circle and letter for the Cr-bearing minerals
and the results are listed in Table 46
Figure 410 Chromite magnetite and hematite compositions from the collected soils
(gray circles) as well as analyses for serpentinite chromite (white squares) from Figure
44 are shown in these diagrams with Cr Wt versus Al Fe Mg and Al+Fe+Mg Wt
130
Figure 41
131
Figure 42
132
Figure 43
133
Cr Wt
0 10 20 30 40 50
Al W
t
0
5
10
15
20
25Cr-SpinelCr-MuscoviteCr-EpidoteCr-GarnetCr-Chlorite
Cr-SpinelBased on 4 Oxygens
Cr0 1 2
AlVI
0
1
2
ChromiteFeCr2O4
HercyniteFeAl2O4
Cr-MuscoviteBased on 11 Oxygens
Cr0 1 2
AlVI
0
1
2
MuscoviteKAl2(AlSi3O10)(OH)2
FuchsiteKCr2(AlSi3O10)(OH)2
Cr-GarnetBased on 12 Oxygens
Cr0 1 2
AlVI
0
1
2
GrossularCa3Al2Si3O12
UvaroviteCa3Cr2Si3O12
Cr-EpidoteBased on 125 Oxygens
Cr0 1
AlVI
2
3
ClinozoisiteCa2Al3Si3O12(OH)2
TawmawiteCa2CrAl2Si3O12(OH)2
Cr-ChloriteBased on 14 Oxygens
Cr0 1
AlVI
0
1
2
Kammererite(Mg5Cr)(AlSi3)O10(OH)8
Clinochlore(Mg5Al)(AlSi3)O10(OH)8
Figure 44
134
Maryland (USA)(Rabenhorst et al 1982)
Chromium (mg kg-1)
0 1000 2000 3000 4000 5000 6000 7000
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Tehama County CA (USA)(Gough et al 1989)
Chromium (mg kg-1)
0 2000 4000 6000 8000 10000 12000 14000
Dep
th (m
eter
s)
00
02
04
06
08
Kalamath Mountains CA (USA)(Hotz 1964)
Chromium (mg kg-1)
2000 4000 6000 8000 10000 12000 14000 16000
Dep
th (m
eter
s)
00
05
10
15
20
25
30
35
40
45
50
Newfoundland (Canada)(Roberts 1992)
Chromium (mg kg-1)
800 1000 1200 1400 1600 1800
Dep
th (m
eter
s)
00
02
04
06
08
Harvard Mine CA (USA)(Unpublished Data)
Chromium (mg kg-1)
0 500 1000 1500 2000 2500
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Jasper Ridge CA (USA)(Oze et al 2003)
Chromium (mg kg-1)
1500 2000 2500 3000 3500 4000 4500 5000
Dep
th (m
eter
s)
00
01
Figure 45
135
LocationsZ A B C D E F G H I J K L M N O P Q R S
pH
0
1
2
3
4
5
6
7
8
9
10
11
12
Nor
th A
mer
ica
Can
ada
Brit
ain
Italy
Far E
ast
Japa
n
Zim
babw
e
Sout
h A
fric
a
Wes
tern
Aus
tral
ia
New
Zea
land
Mg-Bicarbonate Waters pH=8
Ca-Hydroxide Waters pH=11-12
Average RainpH=57
Sout
h A
fric
a
Italy
Irela
nd
Cal
iforn
ia
Cal
iforn
ia
Brit
ish
Col
umbi
a
Mar
ylan
dCal
iforn
ia
Figure 46
136
Figure 47
137
-2
-1
0
1
2
3
4
5
5980 5990 6000 6010
Energy (eV)
Nor
mal
ized
Abs
orba
nce
New Caledonia
Pillikin Mine
Eight Dollar Mnt
Jasper Ridge
Harvard Mine
Cr(
VI)
Figure 48
138
Figure 49
139
Cr Wt 0 10 20 30 40 50
Al W
t
0
10
20
30 Soil Serpentinite
Cr Wt 0 10 20 30 40 50
Fe W
t
0
20
40
60
80
Cr Wt 0 10 20 30 40 50
Mg
Wt
0
2
4
6
8
10
Cr Wt 0 10 20 30 40 50
Al +
Fe
+ M
g W
t
0
20
40
60
80
MagnetiteHematite
Chromite
MagnetiteHematite
Chromite
Chromite
MagnetiteHematite
SpinelSilicate
SpinelSilicate
SpinelSilicate
SpinelSilicate
Figure 410
140
- Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip1
-
- Chapter 3 Oxidative Promoted Dissolution of Chromite by
- Manganese Dioxide and Concurrent Production of Ch
-
- Chapter 4 Chromium Geochemistry of Serpentine S
-
- List of References
-
- Chapter 2
- Chapter 3
-
- Chapter 4
-
- Appendix A
- Chapter 2
-
- Chapter 3
-
- Chapter 4
-
- Appendix A
-
- CHAP1pdf
-
- Toxicity and Regulations of Chromium
- Mineralogy of Chromium in Serpentinites
-
- Chapter Summaries
-
- CHAP2Tablepdf
-
- JR305CLAY Depth 0-5cm
- JR3515CLAY Depth 5-15cm
- JR33045CLAY Depth 30-45cm
-
- HCAP2Figspdf
-
- Figure 27
-
- CHAP3Tablepdf
-
- Figure 36
- Figure 37
-
- CHAP4pdf
-
- ABSTRACT
- REVIEW
-
- Chromium Geochemistry
-
- Soil Organic Matter
-
- CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
-
- BedrockProtolith Mineralogy
- Soil Chemistry and Physical Properties
- Soil Mineralogy
- Backscattered electron images and microprobe analyses were used to identify the Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49 and corresponding Cr-bearing mineral compositions are listed in Table 46 The New C
-
- DISCUSSION
-
- Chromium Spinel in Soils and Rocks
-
- CONCLUDING STATEMENT
- ACKNOWLEDGEMENTS
-
- CHAP4Tablepdf
-
- Ulvoumlspinel
-
PREFACE
The chemistry and mineralogy of Cr related to serpentinites and serpentine soils
are assessed in this dissertation using a multidisciplinary field analytical experimental
and theoretical approach The four chapters that constitute this dissertation describe my
investigations related to these rocks and soils Additionally the common theme shared
by these chapters is that they all examine the Cr mineralogy in serpentinites and
serpentine soils and the potential of Cr to be mobilized and oxidized in near-surface
environments Chromite a Cr-bearing spinel is the dominant source of Cr in
serpentinite and serpentine soils Although chromite is resistant to weathering the
presence of birnessite a high valent Mn oxide will increase the release of Cr and the
production of Cr(VI) from chromite over a range of pH Chromium-clinochlore or
kaumlmmererite contains Cr concentrations less than 3 Wt however the rate at which Cr-
clinochlore weathers compared to chromite makes this mineral a more probable source of
Cr released into soil solutions These types of concerns are addressed in this thesis Brief
descriptions of the four individual dissertation chapters including their publication status
and a list of the co-authors are presented in the following paragraphs
Chapter 1 presents the main objectives of this dissertation The chemistry
regulations and toxicity of Cr are discussed as well as the general characteristics of
serpentinites and serpentine soils Additionally a brief summary of each chapter is
provided
Chapter 2 focuses on serpentinites and serpentine soils at Jasper Ridge Biological
Preserve Stanford CA The results have been submitted for publication in an article
entitled ldquoChromium geochemistry of serpentinized ultramafic rocks and serpentine soils
in the Franciscan Complex of Californiardquo to the American Journal of Science (March
2003) Scott Fendorf Dennis Bird and Robert Coleman are the second third and fourth
co-authors respectively
Chapter 3 is an experimental study evaluating the release of Cr and the
production of Cr(VI) from chromite in the presence of birnessite over a range of pH The
results for this study will be submitted for publication in an article entitled ldquoOxidative
promoted dissolution of chromite by manganese dioxide and concurrent production of
vii
chromaterdquo to Environmental Science and Technology The co-authors for this paper are
Scott Fendorf and Dennis Bird
Chapter 4 is a review paper that critically evaluates the chemistry of serpentine
soils and serpentinites primarily focusing on Cr geochemistry Additionally analytical
studies of serpentine soils collected from California Oregon and New Caledonia are
presented This study will be presented at the Robert Coleman Symposium and has been
submitted as an article entitled ldquoChromium geochemistry of serpentine soilsrdquo for an
International Geology Review Special Publication Scott Fendorf Dennis Bird and
Robert Coleman are the second third and fourth co-authors respectively
Appendix A is a preliminary investigation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite) a mineral common in serpentinites and present in serpentine
soils This appendix is entitled ldquoEstimation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite)rdquo The results of this chapter will be included in a study that
estimates the thermodynamic properties of Cr-bearing silicates including fuchsite (Cr-
muscovite) tawmawite (Cr-epidote) and uvarovite (Cr-garnet)
viii
ACKNOWLEDGEMENTS
Serpentinites serpentine soils and chromium are the Geologic Trinity of this
dissertationhellipeach topic is intimately related to one another The scope and breadth of
this research required a significant amount of time and funding None of this research
would have been possible without the generous support of several foundations
organizations and Stanford University I am grateful for the financial aid I have
received from the Stanford Fellowship the Stanford University McGee and Shell funds
the National Science Foundation the Geologic Society of America Graduate Research
Fund the Stanford Office of Teaching and Learning grant and the Jasper Ridge
Biological Preserve Melon Grant
I would not be where am I today if it were not for my parents Robert and Sue
Oze and my brother Rob Oze Thank you for the sandbox and all the hose-driven
volcanoeshellipthose Wookies never stood a chance While some families vacation in theme
parks my family frequented gem and mineral conventions collected rocks and shells
from mist-shrouded shores and sought lucky stones including an infamous quartz orb
Sometimes I think I was destined to be a geologist I would also like to thank everyone
in my family and all our family friends who have provided that extra little push and
support along the way These people include Tudor Bennatts Louis Oze Margaret and
John Chambers John Bennatts Nancy Skiles the Izumis the memory of Witchie Poo
hijacking Harleys from Bremerton bikers and the Man in the Sound waiting for that one
last catch Close friends I would like to acknowledge who are not related to the Stanford
crowd include Dave and Leigh Anderson and Rebecca Parker These satellite friends
helped me keep in touch with the world outside of academia and proved to me that I
wasnrsquot meant to have a real job A special thank you goes to Katherine Akako
Izumihellipyou have made this whole trip worthwhile
My PhD advisors Scott Fendorf and Dennis Bird provided an invaluable amount
of help patience and energy over the past five years Scott always made the time to
discuss and refine many of my projects making them relevant to many current
geochemical issues Additionally he helped me hone my soil science skills in the field
and in the lab Out in the field Scott can dig a pretty mean hole Dennis devoted a
significant amount of time and sweat in order to make this dissertation happen His
ix
insightful knowledge of geochemistry and of the geology of the Sierra Nevadas provided
the cornerstone for this dissertation
Robert Coleman played a key role in the production of this dissertation He
always knew what to say at the right time and helped me understand the complexity of
ophiolites and serpentinites through experienced eyes Additionally Keith Loague and
Gary Ernst deserve recognition for their outstanding support as committee members for
this dissertation over the last four years I would also like to thank Jonathan Stebbins and
Gail Mahood for providing me the opportunity to attend Stanford
Bob and Vicki Jones always offered an open ear whether it was science or solving
the worldrsquos problems Rick Hazlett (Pomona College) and John Winter (Whitman
College) both had a significant impact on shaping my future science goals and objectives
during my undergrad years and both encouraged me to pursue a career in geology They
were the catalyst for this whole experiment
I owe a lot to my research group members (both the Bird Cage and Soil
Chemistry Group) and those people who didnrsquot mind me tagging along Before I
introduce these people one person has been my friend and co-conspirator for many years
at Stanford Tim Copehellipyou are a pain in my ass Besides never picking up any of your
crap and breaking all my stuff this whole grad school experience would not have been
the same without you Other friends and acquaintances that will not be forgotten in
chronological order of me walking into my office for the first time on September 11
1998 include Anne Dyreborg Esra Inan Tom Bawden Patrick Redmond Phil
Neuhoff Thrainn Fredrickson Lara Heister Kaye Savage Andrea Fildani Mike Schultz
Trayle Kushan Kevin Theissen Angela Hessler Chris Van de Ven Kono Lemke Colin
Doyle Colleen Hansel Ben Bostick Matthew LaForce Nancy Grumet Scott Kroeker
Anders Meibom Brian Ebel Matt Ginder-Vogel Matt Pollizitto Spud Amy Weislogel
Travis Horton and Jake Waldbauer I would like to end this Acknowledgement section
and begin my dissertation with a haikuhellip
Better than outcrops
Are the people Irsquove met here
Thanks for all the beer
x
TABLE OF CONTENTS
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iv Prefacehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip vii Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip ix Table of Contentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xi List of Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xiv List of Illustrationshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xvi Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 1
Toxicity and Regulations of Chromiumhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 2 Mineralogy of Chromium in Serpentiniteshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 3 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5 Chapter Summarieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5
Chapter 2 Chromium Geochemistry of Serpentinized Ultramafic Rocks and Serpentine Soils in the Franciscan Complex of Californiahelliphelliphelliphelliphelliphellip 8
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 8 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 9 Backgroundhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 10
Geologic Background of the Serpentinite at Jasper Ridgehelliphelliphelliphellip 10 Serpentine Soils Characteristics and Reviewhelliphelliphelliphelliphelliphelliphelliphellip 11 Sources of Chromium ProtolithBedrockhelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 12 Chromium-Bearing Phases in Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 13
Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Location and Sample Collectionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Mineral Identification in Rocks and Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Soil Chemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 15 X-ray Diffraction and Clay-Size Mineral Identificationhelliphelliphelliphelliphellip 16 X-ray Absorption Spectroscopyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17 Sequential Extraction Procedurehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17
Resultshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Serpentinite Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Distribution of Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip21 Serpentine Soil Chemistry and Mineralogy at Site JR3helliphelliphelliphelliphellip 22 Sequential Extractionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 26
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 27 Concluding Remarkshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 33 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 44 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 47
xi
Chapter 3 Oxidative Promoted Dissolution of Chromite by Manganese Dioxide and Concurrent Production of Chromatehelliphelliphelliphelliphelliphellip 63
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 63 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 64 Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 65
Solid Phase Characterization and Synthesishelliphelliphelliphelliphelliphelliphelliphelliphellip 65 Extraction Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 66 Batch Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Solution Analysishelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69
Results helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite Extraction Experimentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite and Birnessite Batch Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 70 Secondary and Solid Phase Examination of Chromite and Birnessitehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 72 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 72
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 73 Chromium(VI) Pathways from Chromite and Birnessite into Solutionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 76 Serpentine Soils and Cr(VI)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 77
Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 79 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 80 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 84 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 87
Chapter 4 Chromium Geochemistry of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip100
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 100 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 101 Reviewhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102
Chromium Geochemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102 Mineralogy of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 103 Sources of Chromium for Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphellip 105 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 106 Soils Solutions and Related Waterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 108 Serpentine Soil Vegetationhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 110 Soil Organic Matterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 111
California Oregon and New Caledonia Serpentine Soilshelliphelliphelliphelliphelliphelliphellip 112 Sample Locations and General Soil Characteristicshelliphelliphelliphelliphelliphellip 112 BedrockProtolith Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 113 Soil Chemistry and Physical Propertieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 114 Soil Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116 Chromium and Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 117
xii
Chromium Spinel in Soils and Rockshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 119 Concluding Statementhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 120 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 121 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 122 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 129 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 131
Appendiceshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 141 Appendix Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 142 Appendix Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 143 Appendix A Estimation of the Thermodynamic Properties of Kaumlmmererite (Cr-Clinochlore)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip145
Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 145 Nomenclature and Crystal Chemistry of Cr-chloritehelliphelliphelliphelliphelliphelliphelliphelliphellip 146 Estimation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 147
Estimation of theStandard Molar Volume Using Unit Cell Measurementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of the Standard Molar Volume Heat Capacity and Entropyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of Standard Molar Enthalpy and Gibbs Free Energyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 150 Critique and Discussion of the Estimation Algorithms and SUPCRT92helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 151
Evaluation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 152 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 155 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 161 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 162
List of References
References to Chapter 1helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 166 References to Chapter 2helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 172 References to Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 179 References to Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 183 References to Appendix Ahelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 195
xiii
LIST OF TABLES Chapter 2 Table 21 Whole rock elemental concentrations of serpentinites chert and sandstone Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in
serpentinite rock at Jasper Ridge Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in
serpentine soil at Jasper Ridge Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25
and 26 Table 24 Bulk soil elemental concentrations for soils and bedrock collected along
the transect in Figure 21b Table 25 Chemical and physical properties of the soil at Site JR3 Table 26 Selected trace metal concentrations for each size fraction versus depth at
Site JR3 Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3 Table 28 X-ray diffraction d-spacings for common silicates Table 29 Results of the Diablo clay serpentine soil and chromite sequential
extraction experiments Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1
kg of sample Chapter 3 Table 31 Chromite and birnessite masses and concentrations used for each batch
experiment Table 32 Serpentine soil and birnessite experimental parameters Table 33 Summary of Cr(VI) rates obtained from batch experiments Table 34 Comparative analyses of batch experiment solutions and chemical
extractions of chromite and birnessite solids Chapter 4 Table 41 Measured Cr concentrations in serpentine soils worldwide Table 42 Summary of Cr Fe and Ni extractions from serpentine soils Table 43 Serpentine soil plant Cr concentrations Table 44 Chemical and textural properties of collected soils Table 45 Total major and trace element concentrations of the collected soils Table 46 Compositions for soil Cr-containing minerals identified in Figure 49 Table 47 Chromium-bearing mineral masses and volumes required to influence Cr
concentrations in soils Appendix A Table A1 Cr and Al end-members in the Cr-clinochlore solid solution
xiv
Table A2 Standard state thermodynamic properties of Cr and Al-bearing minerals Table A3 Unit cell measurements and molar volumes of clinochlore and Cr-
clinochlores Table A4 Entropy volume and heat capacity for kaumlmmererite using the Ransom and
Helgeson Method Table A5 Fictive polyhedral considered in the estimations Table A6 Estimated thermodynamic properties for kaumlmmererite
xv
LIST OF FIGURES Chapter 2 Figure 21a Soils map of Jasper Ridge Biological Preserve CA Figure 21b Serpentinite and serpentine soil sampling sites at Jasper Ridge Biological
Preserve CA Figure 22 Chromium concentrations versus depth for four serpentine soils in
Maryland California and New Foundland Figure 23 Petrographic images of serpentinite at Jasper Ridge Figure 24 Elemental concentrations relative to Cr for chlorite pyroxene magnetite
chromite and a chromite-silicate mixture in the serpentinite at Jasper Ridge
Figure 25 Backscattering electron images of minerals in the serpentinite at Jasper Ridge
Figure 26 Backscattered electron images of Cr-chlorite and Cr-magnetite assemblage in the serpentinite bedrock
Figure 27 Highest and lowest Cr concentrations at each site in the serpentine soil transect
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected for total gt2 mm sand silt and clay size fractions
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soils at Jasper Ridge
Figure 210 X-ray diffraction clay analyses of sample JR3515 Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at specific depths Figure 212 X-ray microprobe elemental maps of Cr and Fe for the serpentine soil and
serpentinite at site JR3 Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments Figure 214 Scanning electron microprobe images of the chromite standard after each
sequential extraction experiment Figure 215 Tie-lines connecting serpentinite chromite and end-member chromite with
soil chromite values Chapter 3 Figure 31 Scanning electron microprobe images of chromite and birnessite standards Figure 32 Chromite extraction experiments evaluating Cr Fe and Al Figure 33 pH and Eh versus time (hours) for each chromite-birnessite chromite and
birnessite batch experiments Figure 34 pH and Eh with Cr stabilities Figure 35 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant chromite and variable birnessite at pH 5 Figure 36 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant birnessite and variable chromite at pH 5
xvi
Figure 37 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for variable pH and constant chromite and birnessite Figure 38 Rate of Cr(VI) production (d[Cr(VI)]dt) as a function of the surface area
concentration of chromite birnessite and pH Figure 39 SEM examination of solids following the termination of the batch
experiments Figure 310 Serpentine soil and birnessite experiments Figure 311 Iron analyses and the solubility of amorphous Fe(OH)3 and goethite Figure 312 Congruent dissolution of chromite based on Fe and Cr Figure 313 Three pathways of Cr release and oxidation Chapter 4 Figure 41 Serpentine soil localities worldwide Figure 42 Field site pictures of serpentine soils Figure 43 Stability field diagram of Cr over a range of Eh and pH conditions Figure 44 Chromium and Al Wt plot and octahedral Cr and octahedral Al plots
for chromite and Cr-silicates including Cr mica (fuchsitemariposite) Cr-garnet (uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite)
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (m) for selected serpentine soils
Figure 46 pH of serpentine soils and related groundwater worldwide Figure 47 Soil pictures with Munsel color values and names from collected soils Figure 48 Cr XANES spectra for collected serpentine soils Figure 49 Backscattering images of mineral phases within each serpentine soil Figure 410 Chromite magnetite and hematite compositions from analyzed soils as
well as analyses for chromite reported from around the world Appendix A Figure A1 Compositional variation of Cr2O3 and Al2O3 Wt in Cr-chlorite Figure A2 Compositional variation of Cr and VIAl in Cr-chlorite based on 14 oxygens Figure A3 Molar volume estimation of kaumlmmererite based on XRD unit-cell
parameters Figure A4 Natural and calculated compositions of co-existing Cr-clinochlore and Cr-
spinel solid solution in the system Mg0-FeO-Cr2O3-Al2O3-SiO2-H2O at 2 kbar pressure
xvii
CHAPTER 1
INTRODUCTION
Serpentinites or metamorphosed ultramafic rocks are distributed worldwide and
are commonly associated with ophiolite complexes (Coleman 1977 Coleman and Jove
1992) These rocks weather to form serpentine soils a generic term used to describe any
soil derived from serpentinite regardless of its physical or chemical properties
Serpentinites and serpentine soils contain Cr-bearing minerals the most prevalent being
chromite and Cr released from these minerals into surface and subsurface environments
is potentially toxic and carcinogenic The main objective of this dissertation is to
evaluate the mineralogy and chemistry of Cr related to serpentinites and serpentine soils
over geologic time using an integrated analytical experimental and theoretical approach
Additionally the potential of serpentinites and serpentine soils as sources of non-
anthropogenic Cr contamination is investigated A discussion of Cr and its significance
to serpentinites and serpentine soils begins with its discovery by a Frenchman
investigating an ore named ldquored leadrdquo from the Urals
Nicolas-Louis Vauquelin discovered Cr in 1797 in the mineral crocoite PbCrO4
(Burns and Burns 1975) The metalelement was named chromium after the Greek word
chroma meaning color due to its strong influence on mineral pigmentation (Burns and
Burns 1975) For example rubies and emeralds derive their color from trace amounts of
Cr in their structures More importantly JT Lowitz in 1798 discovered Cr in chromite
or Cr-spinel the primary source of Cr for industrial purposes (Burns and Burns 1975)
Chromium is a group VIB transition metal with two common oxidation states (Cr(III) and
Cr(VI)) an average crustal abundance of 100 mg kg-1 and an abundance of 513x105 in
the solar system (relative to [H]=1x1012) (Mason and Moore 1982)
The majority of Cr on the earth exists as Cr(III) where it is an immobile cation
that complexes strongly with organic matter and adsorbs on oxides and silicates even at
low pHs (pHgt4) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) Additionally
the chemical mobility of Cr(III) is limited due to the formation of highly stable
(oxy)hydroxides (Rai and Zachara 1988 Rai et al 1989 Ball and Nordstrom 1998)
1
Chromium(III) has the highest octrahedral site preference energy (OSPE) of any trace
element and it isomorphically substitutes into octahedral sites in oxides
(oxy)hydroxides sulfides and silicates Although Cr(III) is known to substitute for both
octahedral Al(III) and Fe(III) the strongest correlation for substitution appears to be for
Al(III) in oxides and silicates as demonstrated in Chapters 2 and 4 and Appendix A
Chromium(VI) is a strongly oxidizing and weakly adsorbing oxyanion in the form of
chromate (HxCrO42-x) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995 Shiraki
1997 Ball and Nordstrom 1998) Consequently Cr(VI) is chemically mobile and
bioavailable over a wide range of pH (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Chromium(VI) is tetrahedrally coordinated and behaves similarly to
sulfate in minerals such as jarosite (Baron and Palmer 1996)
The pathways and chemistry of Cr oxidation and reduction in soils and related
waters are covered in depth by Fendorf (1995) and Ball and Nordstrom (1998) The
reduction of Cr(VI) to Cr(III) may occur by reactions with Fe(II)aq Fe(II)-containing
minerals such as magnetite organic matter H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996) Oxidation of Cr(III)
to Cr(VI) related to serpentinite-related water and serpentine soil solutions is still a matter
of conjecture Chromium(III) oxidation in serpentine soils has been hypothesized to be
the result of Mn oxides such as birnessite as suggested by both Gough et al (1989) and
Becquer et al (2003) This idea is explored in depth in Chapter 3 where birnessite is
shown to influence the release and oxidation of Cr from chromite over a range of pH (3
to 8) at 25oC
Toxicity and Regulations of Chromium
The examination of Cr based on its oxidation state is necessary due to the
juxtaposing behavior and toxicity of Cr(III) and Cr(VI) Chromium(III) is an essential
micronutrient in human nutrition used by the body to process sugar protein and fat
(Vincent 2000) Inhalation of Cr(VI) is known to cause irritation in the nose including
symptoms such as nosebleeds (epistaxis) and ulcers andor perforations in the nasal
septum (Katz and Salem 1993 Kuo et al 1997) Ingesting considerable quantities of
Cr(VI) can result in stomach ulcers convulsions kidney and liver damage and death
2
(Katz and Salem 1993) Skin contact with certain Cr(VI) compounds causes allergic
contact dermatitis (ulcers) and irritant dermatitis (Katz and Salem 1993) Chromate
induced ulcers or chromeholes are typically red swollen crusted painless lesions
(Katz and Salem 1993) The nosology of Cr is beyond the scope of this dissertation
however Lees (1991) is an excellent source for further information Overall the
Environmental Protection Agency (EPA) the Occupational Safety and Health
Administration (OSHA) the World Health Organization (WHO) and the Department of
Health and Human Services (DHHS) have determined that that Cr(VI) compounds can
increase the risk of lung cancer primarily by inhalation as well as other forms of cancer
caused by ingestion
Federal and state regulation standards appropriate for both Cr(III) and Cr(VI) are
summarized and discussed by Proctor et al (1997) Human health action levels for Cr
vary for each state and the ranges for total Cr in soils include the following 200-66000
Cr(III) mg kg-1 (residential) 02-2000 Cr(VI) mg kg-1 (residential) 2500-1000000
Cr(III) mg kg-1 (non-residential) and 64-71000 Cr(VI) mg kg-1 (non-residential)
(Proctor et al 1997) Incidentally Cr concentrations in serpentine soils and serpentinites
(gt200 Cr mg kg-1) exceed many state action levels and would be considered toxic The
EPA has set a limit of Cr(III)+Cr(VI) at 100 microg L-1 in drinking water The OSHA has
set limits of 500 microg of water soluble Cr(III) compounds per cubic meter of workplace air
or 500 microg m-3 1000 microg m-3 for metallic Cr and insoluble Cr compounds and 52 microg m-3
for Cr(VI) compounds for 8-hour work shifts and 40-hour work weeks The limits
proposed above are constantly changing and should not be used for reference without
checking current regulation standards
Mineralogy of Chromium in Serpentinites
Chromium occurs in a variety of minerals either as a major or minor constituent
(Burns and Burns 1975) Chromium-bearing minerals in serpentinites are primarily
oxides and silicates in which the Cr is present as Cr(III) Chromite is a Cr-bearing spinel
abundant in serpentinites and is resistant to low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) The chemistry and
significance of chromite is addressed throughout this dissertation Pyroxenes such as
3
augite and enstatite are primary minerals identified in serpentinites and contain lt8 Cr Wt
(Grapes 1981 Von Knorring et al 1986 Treloar 1987a Challis et al 1995
Sanchez-Vizcaino et al 1995 Deer et al 1996) Chromate minerals such as crocoite
(PbCrO4) are rare in serpentinites and the occurrence of these minerals are limited to Pb
deposits in Australia Tasmania and the Urals (Williams 1974 Crane et al 2001)
Chromium(III) released from the primary Cr-containing phases may
isomorphically substitute into oxides and silicates for Al(III) and Fe(III) due to size
(octahedral radii Cr3+ = 0615 Aring Al3+ = 061 Aring and Fe3+=0643 Aring) and charge
similarities Magnetite formed during serpentinization incorporates Cr into its structure
and serpentine minerals (antigorite lizardite and chrysotile) contain trace concentrations
of Cr (Spangenberg 1943 Page and Coleman 1967 Burkhard 1993) Eskolaite (Cr2O3)
has been identified in Finland (Von Knorring et al 1986) Metasomatic alteration of
ultramafic rocks commonly results in the formation of Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite) uvarovite (Cr-garnet) tawmawite (Cr-epidote) and
kaumlmmererite (Cr-chlorite) (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Neiva 1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981
Max et al 1983 Nutman et al 1983 Raase et al 1983 Chao et al 1986 Von
Knorring et al 1986 Kerrich et al 1987 Treloar 1987b Treloar 1987a Morand 1990
Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993 Christofides et al 1994
Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Proenza et al
1999) Additionally kyanite and andalusite in the proximity of ophiolite complexes
incorporate up 5 Cr Wt (Cooper 1980 Raase et al 1983 Kerrich et al 1987)
Chapter 4 provides an in depth discussion of these Cr-bearing silicates as well as
demonstrating mineral compositions reported in the literature based on octahedral Cr and
Al Appendix 1 is a preliminary investigation devoted to estimating the thermodynamic
properties of kaumlmmererite These properties can then be incorporated into geochemical
modeling programs to begin accessing the stability of this common serpentinite mineral
over a range of temperatures and pressures
4
Chromium in Serpentine Soils
Evaluating Cr in serpentine soils requires an examination of both the mineralogy
and chemistry of the serpentinite and soil Measured Cr concentrations in serpentinites
and serpentine soils range from ~200 mg kg-1 to 6 Wt all of which is Cr(III) (Soane
and Saunder 1959 Proctor and Woodell 1971 Shewry and Peterson 1976 Jaffre 1980
Proctor et al 1980 Rabenhorst et al 1982 Schwertmann and Latham 1986 Brooks
1987 Gough et al 1989 Gambi 1992 Lee 1992 Morrey et al 1992) In Chapters 2
and 4 chromite and Cr-magnetite are identified as the most prevalent sources of Cr
responsible for Cr-enrichment in these soils Chromite is resistant to chemical
weathering and it has been observed to undergo incongruent dissolution conserving Cr in
serpentine soils Chromium is tightly bound in the matrix of the soil less than 1 of the
total Cr can is removed using standard soil extraction methods (10 hydrogen peroxide
(H2O2) solution 01 M potassium permanganate (KMnO4) solution10 M hydrofluoric
(HF) acid 1 M hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and 1
mM barium chloride (BaCl2) solution) due to the stability of chromite Although silicates
contain minor concentrations of Cr compared to chromite they are perhaps the main
contributors of Cr identified in plants (lt200 Cr mg kg-1) and soil solutions (lt50 microg L-1)
During soil-forming processes Cr released to soil solutions by the dissolution of Cr-
bearing minerals are sequestered into a host of structural and surface sites available in
authigenic Fe(III) oxides Fe(III) hydroxides and clay minerals (Schwertmann and
Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990 Gerth 1990 Gasser and
Dahlgren 1994 Ilton and Veblen 1994 Brigatti et al 2000 Ilton et al 2000) The
clay-sized fractions in soils at Jasper Ridge Biological Preserve CA contain Cr
concentrations as high as 1000 Cr mg kg-1
Chapter Summaries
A detailed field and analytical examination of Cr in serpentinites and serpentine
soils at Jasper Ridge is presented in Chapter 2 A majority of the Cr enrichment and
variability with depth in these soils is the result of small grains of chemically resistant
chromite from the serpentinite protolith Chromite identified in the soil undergoes
incongruent dissolution conserving Cr relative to chromite compositions identified in the
5
serpentinite bedrock Sequential extraction experiments using barium chloride (BaCl2)
hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and hydrofluoric (HF)
acid were performed on a chromite standard confirming that Cr-bearing spinels are
resistant to chemical weathering X-ray diffraction analyses of the clay-size fraction in
the serpentine soils identified Cr-clinochlore smectite vermiculite and serpentine with
the smectite being the most abundant mineral in the clay-size fraction Chromium(III)
was the only identified oxidation state of Cr in the serpentine soil solids and in
serpentinites at Jasper Ridge in the solid phase This chapter represents a comprehensive
site assessment identifying the nature and behavior of Cr in serpentine soils with respect
to the bedrock mineralogy and depth
Chromite an abundant Cr-enriched spinel is responsible for a majority of the Cr
enrichment identified in ultramafic rocks serpentinites and serpentine soils The release
and oxidation of Cr from chromite is a pathway to consider when evaluating the potential
of ultramafic rocks and their related sediments and soils as sources of Cr(VI) In
Chapter 3 experiments aimed at quantifying the release and oxidation of Cr from
chromite in the presence of birnessite are described Chromite is extremely resistant to
dissolution (lt8 Cr mg kg-1 is released) when immersed in solutions including 1 mM
barium chloride solution 10 M hydrofluoric acid 10 hydrogen peroxide and 1 M
hydroxylamine hydrochloride with 25 acetic acid Highly oxidizing solutions of 01 M
potassium permanganate increased the rate and release of Cr relative to the other
solutions listed above thus birnessite a high valence Mn oxide present in serpentine
soils was chosen to determine if Mn could accelerate Cr release and oxidation from
chromite The results presented in Chapter 3 demonstrate that birnessite increases the
rate of Cr(VI) observed in solution from chromite This production of Cr(VI) is
dependent on the chromite suspension density where birnessite surface are concentrations
are greater than 20 m2 L-1 between pHs 3 and 8 at ~25degC
A comprehensive review of the chemical characteristics of serpentine soils
including topics such as the mineralogy of serpentine soils soil solutions organic matter
and vegetation is provided within Chapter 4 Additionally chemical and spectroscopic
analyses were performed on serpentine soils from a) New Caledonia b) Eight Dollar
Mountain Selma OR c) Nickel Mountain Riddle OR d) Harvard Mine Jamestown
6
CA e) Jasper Ridge Biological Preserve Stanford CA and f) Pillikin Mine El Dorado
CA Serpentinites and serpentine soils associated with ophiolite complexes cover ~1 of
the earthrsquos exposed surface and they are correlated to modern and ancient convergent
boundaries The physical and chemical characteristics of serpentine soils are directly
related to the alteration and tectonic history of the serpentinite as well as the soil-forming
factors including the parent material climate biota topography and time Serpentine
soils have Cr concentrations exceeding 200 mg kg-1 however the concentration and the
availability of Cr in these soils are not correlated with the five soil-forming factors
(climate organic material parent material topography and time) or the soil chemistry
Although only Cr(III) was identified in the soil solids Cr(VI) has been identified in soil
solutions (lt30 microg L-1) from serpentine soils in New Caledonia and California Chromite
and Cr-magnetite grains are identified in the soils listed above and these minerals are
responsible for a majority of the Cr enrichment in serpentine soils
Kaumlmmererite is a Cr-chlorite associated with serpentinized ultramafic rocks such
as those collected at Jasper Ridge Harvard Mine and Pillikin Mine Although Cr
concentrations are typically 01-2 Wt in kaumlmmererite it is modally abundant in
serpentinites (up to 5 of the volume of the rock) and a potential source of Cr to be
released into soils and groundwater In Appendix A the standard molar thermodynamic
properties of the Cr-clinochlore end-member kaumlmmererite (Mg5Cr(AlSi3)O10(OH)8) is
estimated using established estimation algorithms This initial study provides the
foundation for submitting a future paper presenting the thermodynamic properties for
major Cr-bearing silicates including fuchsite (Cr-muscovite) uvarovite (Cr-garnet) and
tawmawite (Cr-epidote) and their metamorphic significance related to serpentinized
ultramafic rocks
7
CHAPTER 2
CHROMIUM GEOCHEMISTRY IN SERPENTINIZED ULTRAMAFIC ROCKS
AND SERPENTINE SOILS FROM THE FRANCISCAN COMPLEX OF
CALIFORNIA
ABSTRACT
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of California
produces serpentine soils containing high concentrations of Cr as well as other potentially
toxic elements including Ni Co and Mn Chromium concentrations in serpentine soils
from Jasper Ridge Biological Preserve in the Central Coast Range are as high as 4760
mg kg-1 nearly three times greater than the serpentinite protolith (~1800 mg kg-1)
Chromium-containing minerals within the bedrock include chlorite (~03 Cr Wt )
enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt ) magnetite
(82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate phase
containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
8
INTRODUCTION
Serpentine soils are formed from the weathering of ultramafic rocks and their
metamorphic equivalents (serpentinites Brooks 1987) Measured Cr concentrations in
these soils range as high as 80600 mg kg-1 and are enriched compared to non-serpentine
soils (7 to 221 mg kg-1 Proctor 1992 McBride 1994) The distribution and
concentration of Cr in serpentine soils worldwide is highly variable due to the chemical
composition and mineralogy of the parent material and soil forming factors including
climate biota topography precipitation and time (Hotz 1964 Rabenhorst et al 1982
Schwertmann and Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al
1989 Gough et al 1989 Kaupenjohann and Wilcke 1995) Potential Cr-bearing phases
in serpentine soils have been assessed using extraction experiments and these phases
include organic material silicates secondary oxides and spinels (Schwertmann and
Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994 Kaupenjohann and
Wilcke 1995 Becquer et al 2003) The weathering of these Cr-bearing phases over
geologic time and how these sources govern the distribution of Cr in serpentine soils have
not been evaluated in great detail
Chromium is a group VIB transition metal present in the environment as Cr(III) a
non-hazardous species and micronutrient andor as Cr(VI) a toxin to living cells and a
Class A human carcinogen by inhalation (Daugherty 1992 Cohen et al 1993 James et
al 1997) Chromium(III) is chemically immobile due to forming low solubility
(oxy)hydroxides and stable surface complexes whereas Cr(VI) is a weakly adsorbing
(oxy)anion and thus is mobile in most aquatic environments (McBride 1994 Fendorf
1995 Ball and Nordstrom 1998) Chromite and Cr-magnetite have been cited as
primary sources of Cr in serpentine soils (Rabenhorst et al 1982 Schwertmann and
Latham 1986 Alexander et al 1989 Gough et al 1989) Although Cr is primarily
found in the form of Cr(III) in both ultramafic rocks and serpentinites Cr(VI) has been
measured in serpentine soils in California and New Caledonia (Gough et al 1989
Becquer et al 2003) Due to the abundance of Cr and the potential for Cr(VI) formation
serpentine soils and serpentinites are potential sources of non-anthropogenic Cr
contamination
9
The objectives of this study are to determine the distribution and chemistry of Cr
in naturally Cr-enriched soils using an integrated field analytical experimental and
theoretical approach A wide array of techniques and methods including X-ray
absorption spectroscopy and chemical extractions were used to identify the nature of Cr
and its availability in serpentine soils Serpentinites and serpentine soils at Jasper Ridge
Biological Preserve located near Stanford CA were used to evaluate the geochemical
behavior of Cr These soils cover approximately 50000 m2 and contain high
concentrations of the trace elements Cr Ni and Mn The soils are dominantly Mollisols
with grassland vegetation and they are pedalogically similar to those associated with the
Central Coast Range of the Franciscan Complex in the western United States The main
serpentinite body at Jasper Ridge was exposed approximately 400000 years ago allowing
pedogenesis and Cr cycling to occur over a geologically relevant time period (Burgmann
et al 1994) These soils can then be used as a proxy for evaluating the long-term
behavior of Cr
BACKGROUND
Geologic Background of the Serpentinite at Jasper Ridge
Jasper Ridge Biological Preserve is located in the foothills of the Santa Cruz
Mountains in California and is part of the Central Coast Range of the Franciscan
Complex an Early Cretaceous accretionary meacutelange containing blocks of greywacke
greenstone blueschist and serpentinized ophiolite (Figure 21) Formation
metamorphism and emplacement of serpentinite at Jasper Ridge began approximately
130-150 million years ago during Jurassic subduction along the California continental
margin (Page and Tabor 1967) Thin wedges of ultramafic rock (peridotite) were
detached from the down-going slab of oceanic crust and incorporated into the subduction
meacutelange Subsequent serpentinization of the ultramafic rock in the down-going slab
resulted in the formation of serpentinites now exposed at Jasper Ridge
Rare blueschist blocks (1 to 5 meters in width) present in the serpentinite at Jasper
Ridge attest to the degree of metamorphism and the depths from which portions of the
tectonic meacutelange has been elevated Transformation of greenstone in the meacutelange into
blueschist required high pressures in excess of ~5-7 kbars (15-20 kilometers depth) and
10
temperatures of 200-300 degC (Ernst 1971) The low density of the serpentine (245-250 g
cm-3) allowed upward diapiric migration of the serpentinite body through the denser
enclosing tectonic meacutelange (265 g cm-3) Uplift of the present day Coast Range and
activity associated with the San Andreas fault system have shaped the current topography
and exposure of serpentinite at Jasper Ridge Serpentinite is currently present as a narrow
discontinuous band of pervasively sheared rock located along the ridge top (Figure 21)
The serpentinite is in fault contact with the Whiskey Hill Formation (Cretaceous turbidic
sandstone and mudstone) and tectonically interleaved with the Franciscan Complex
Serpentine Soils Characteristics and Review
Relative to other soil types serpentine soils are characterized by higher
concentrations of Cr Ni Co and Fe lower concentrations of plant nutrients such as Ca
K N and P lower CaMg ratios and characteristic flora and physical properties (Brooks
1987) The pH of serpentine soils ranges from ~4 up to 9 (Brooks 1987 Cole 1992
Proctor 1992) Serpentine soil solutions obtained using tension-free lysimetry have Cr
concentrations between 01 and 32 micromol L-1 in which a majority of the Cr is present as
colloidal material (Gasser and Dahlgren 1994) The inhabitability of these soils for most
plants has been attributed to the imbalance of Ca to Mg (Walker 1954) the deficiency of
plant nutrients (Turitzin 1991) and the elevated concentration of heavy metals
(Robinson et al 1935) however linking the lack of plant productivity of these soils to
toxic levels of Cr has not been fully resolved (Brooks 1987 Kruckeberg 1992) The
unique flora and fauna associated with these soils have been well-documented and Cr
concentrations in the plants range as high as 600 mg kg-1 (Brooks 1987 Gough et al
1989)
Chromium concentrations as a function of depth for serpentine soils in Maryland
(Rabenhorst et al 1982) Tehama County CA (Gough et al 1989) the Klamath
Mountains CA (Hotz 1964) and New Foundland Canada (Roberts 1992) are
illustrated in Figure 22 Chromium concentrations with respect to depth at each location
are highly variable demonstrating both maximums and minimums The concentration of
Cr varies significantly at each location due to the chemical composition of the parent
11
material and other soil-forming factors nevertheless Cr concentrations typically decrease
with depth towards the bedrock
A variety of chemical extractions on serpentine soils suggest a broad host of Cr-
containing soil phases including organic material silicates secondary oxides and spinels
(Schwertmann and Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994
Kaupenjohann and Wilcke 1995) The maximum concentration of Cr extracted from a
serpentine soil reported in the literature is 226 Wt Cr from the New Caledonia laterite
using a 03 M dithionite-citrate bicarbonate (DCB) solution (Schwertmann and Latham
1986) Li and Fendorf (2000) performed extractions on the serpentine soils at Jasper
Ridge using organic acids (citric and oxalic acids) in which a 100 microM citric acid solution
was only able to remove 23 Cr microgg-1 (08 Wt Cr) from a soil containing ~3000 Cr mg
kg-1 This suggests a majority of Cr is bound in phases that are not reactive with these
solvents
Sources of Chromium ProtolithBedrock
Chromium primarily occurs within oxide and silicate phases in ultramafic rocks
and serpentinites Chromite is a common primary (igneous) Cr-bearing spinel in mafic
and ultramafic rocks and often survives low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) Isomorphic substitution of
Al3+ Fe3+ and Ti4+ into the octahedral site and Mg2+ Ni2+ Zn2+ and Mn2+ into the
tetrahedral site is common (Sack and Ghiorso 1991) however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987 Pan and Fleet 1991 Burkhard 1993 Christofides et al 1994 Challis et
al 1995 Sanchez-Vizcaino et al 1995 Deer et al 1996 Mathiesen 1999) Pyroxenes
such as augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) are primary minerals in
ultramafic rocks that incorporate varying concentrations of Cr into their structures (Deer
et al 1996)
Serpentinization is associated with the partial oxidation of Fe(II) present in
olivine and pyroxene forming magnetite (Brooks 1987) Chromium(III) derived from
primary chromite and pyroxene may isomorphically substitute into the Fe(III) site of
magnetite due to size (octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge
12
similarities Hydrothermal and CO2 metasomatism of ultramafic rocks may produce Cr-
silicate minerals including fuchsitemariposite (Cr-mica lt17 Cr Wt ) uvarovite (Cr-
garnet lt20 Cr Wt ) tawmawite (Cr-epidote lt10 Cr Wt ) and kaumlmmererite (Cr-
chlorite lt6 Cr Wt Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Phillips et al 1980 Schreyer et al 1981 Max et al 1983 Kerrich et al 1987 Treloar
1987 Pan and Fleet 1991 Christofides et al 1994 Sanchez-Vizcaino et al 1995)
These minerals common in altered ultramafic rocks have varying concentrations of Cr
that isomorphically substitute for octahedral Al in the silicate structure due to similarities
in charge octahedral preference and octahedral radii (Al3+ = 061 Aring) Any of these
silicates and magnetite are potential sources for Cr in soils produced by the weathering of
serpentinite and ultramafic rocks
Chromium-Bearing Phases in Soils
Soil minerals specifically Fe(III) (oxy)hydroxides and clays provide a host of
structural and surface sites that may sequester Cr(III) released to soil solution by
weathering of Cr-bearing phases in serpentinite and ultramafic rocks Isomorphic
substitution of Cr(III) for Fe(III) in oxides (oxy)hydroxides and amorphous compounds
is expected based on crystallographic considerations synthesis experiments and analyses
of selective dissolution of Fe(III) phases (Schwertmann et al 1989 Gerth 1990)
Additionally Cr(III) readily sorbs on surfaces of Fe(III)-containing oxides such as
magnetite and hematite via inner-sphere complexation and forms low-solubility
precipitates at pH values greater than 4 (Charlet and Manceau 1992 Peterson 1996)
Octahedral substitution of Cr(III) for Al(III) in clays (smectites and vermiculites) is
possible based on size and charge similarities Sorption of Cr(III) on edge and interlayer
sites of clay minerals has also been observed by Ilton and Veblen (1994) Soil organic
matter (SOM) is another possible host for soil Cr in which Cr can be chelated into stable
organomineral and organometallic complexes (Kaupenjohann and Wilcke 1995 Brady
and Weil 1999)
13
MATERIALS AND METHODS
Location and Sample Collection
Jasper Ridge Biological Preserve is a microcosm of accreted terranes located in
the foothills of the Santa Cruz Mountains CA Figure 21a is a soils map reported by
Kashiwagi (1985) depicting the alluvial Tertiary sandstone and Franciscan Complex
soils present at Jasper Ridge The Franciscan Complex soils are divided into two
subgroups those derived from greenstone-graywacke and those from serpentine The
serpentine soils dominantly Mollisols and serpentinite bedrock are present as a narrow
discontinuous band located along the ridge top they do not receive sediment from the
adjacent non-serpentine soils Soils and rocks were collected at Jasper Ridge from 1999
to 2002 and sample locations are shown in Figure 21b A 500 meter transect consisting
of 10 sites extends across the length of the serpentine soil body At each site along the
transect soils and bedrock were collected with respect to depth Site 1 (Diablo clay) is
not derived from serpentinite but rather from chert and greenstone Additionally Site 1
is located down slope of the main serpentinite body and has likely received sediment
from the adjacent serpentine soil Rocks were also collected at several exposed outcrops
located in the serpentinite body
Site JR3 (Figure 21b) was chosen for a detailed geochemical study to identify
factors affecting the geochemistry of Cr in serpentine soils Soil at JR3 is a loamy
magnesic thermic Lithic Haploxeroll The thickness of the soil is approximately 50 cm
deep Grass dominantly Lolium multiflorum comprises 95 of the vegetation at this site
and trees and shrubs are not present
Mineral Identification in Rocks and Soils
Mineral identification was accomplished by electron probe microanalysis on an
automated JEOL 733A electron microprobe operated at 15 kV accelerating potential and
15 nA beam current Calibration was conducted using natural geologic standards Beam
width for analysis was 10 microm Raw counts were collected over 20 s and were constant
with time indicating that elemental drift was negligible Detection limits for probe
analyses are ~002 oxide Wt Additionally backscattering electron (BSE) images of
the rocks and soils were obtained using the microprobe
14
Carbon-coated petrographic thin sections of rock samples were used for
microprobe analyses Soil samples were dried in an oven at 60degC for 24 hours mounted
onto a glass slide using epoxy and heated to 100degC until the epoxy cured An initial cut
of each section using a petrographic saw was made followed by a second infusion of
epoxy to ensure total impregnation Finally each section was cut polished and carbon-
coated for electron microprobe measurements
Soil Chemistry
Soils were analyzed for pH electrical conductivity (EC (deciSiemensm))
particle size ammonium acetate extractable cations organic carbon total nitrogen and
elemental composition The pH was determined using a 11 mixture of soil to deionized
water Electrical conductivity (EC) was determined on 2 mL of a soil solution extract
Particle size of the soil was measured using the hydrometer method as described by Gee
and Bauder (1986) These analyses were performed at the Utah State University
Analytical Laboratory
Ammonium acetate was used to measure the concentration of extractable K Ca
Mg and Na within the soil The procedure involved weighing 2 g of soil and adding 25
mL of ammonium acetate (20 M NH4OAc) followed by mixing on an orbital shaker for
30 minutes The solution was filtered using a Whatman 42 filter paper (pore size 25
microm) Potassium Ca Mg and Na concentrations were analyzed using atomic absorption
spectrophotometry (AAS)
Organic carbon was measured using the Walkley-Black titration method (Mebius
1960) Approximately 05 g of sediment was weighed and reacted with 5 mL of 20 M
potassium chromate (K2Cr2O7) followed by the rapid addition of 10 mL concentrated
sulfuric (H2SO4) acid The bottle was cooled for 30 minutes at room temperature and
diluted with 100 mL of deionized water Next 030 mL of 0025 M ortho-
phenanthroline-ferrous solution was added and the solution titrated to the endpoint using
10M ferrous sulfate (FeSO4)
Total nitrogen of soil samples was obtained using an automated combustion
method as described by Sweeny (1989) Approximately 150 mg of air-dried and sieved
15
(40 mesh) soil was placed in a tin foil encapsulating cup A Leco total nitrogen analyzer
with resistance furnace was then used to determine total nitrogen
Major and minor element concentrations for soil and rock samples were obtained
by completely dissolving the samples using a mixture of hot concentrated nitric
perchloric and hydrofluoric acids Elemental concentrations were measured in the
supernatant using inductively coupled atomic emission spectroscopy (ICP-AES)
Standards were plusmn5 mg kg-1 of their known values These analyses were completed at
Chemex Laboratory (Vancouver Canada) Soil samples were dried and sieved (lt2mm)
before analysis Additionally total element concentrations for gt2mm sand silt and clay
size fractions were obtained for each sample collected at Site JR3
X-ray Diffraction and Clay-Size Mineral Identification
X-ray diffraction (XRD) analyses were conducted on the clay-size fraction
(lt2microm) from Site JR3 using a Siemens Diffractometer D5000 at the University of Idaho
The clay-size fraction was isolated using a suspension method described by Vaniman
(2001) Organic matter and calcium carbonate were removed from the clays using a
concentrated hypochlorite solution (Kunze and Dixon 1986) Crystalline and non-
crystalline iron oxides were removed using a citrate-sodium bicarbonate-sodium
dithionite (CBD) solution (Kunze and Dixon 1986) The remaining phyllosilicate
minerals were analyzed using a series of treatments including 1) magnesium-saturation
(2 M MgCl and 2 M MgAOC) and air drying 2) magnesium-saturation and glycolation
(1 glycerol-ethanol solution) 3) potassium-saturation (1 M KCl) and air drying and 4)
potassium (1 M KCl) saturation followed by heating to 500degC (Whittig and Allardice
1986) X-ray diffraction patterns were collected between 2θ values of 20-350deg and 20-
150deg with a generator potential of 30 kV a generator current of 22 mA (using CuKα
radiation) a Ni filter and a scan speed of 1degmin The software package JADE was used
for XRD data analysis and mineral identification
X-ray Absorption Spectroscopy
In-situ fluorescence yield X-ray absorption near-edge structure (XANES)
spectroscopy at the Cr K-edge (edge inflection point is 5989 eV for Cr(0)) was conducted
16
at the Stanford Synchrotron Radiation Laboratory (SSRL) Stanford CA using
beamlines 4-1 and 4-3 A Si(220) monochromator and a Ge-detector were used and
three scans were collected for each XANES analysis Soil and rock samples from Site
JR3 were prepared by removing the gt2mm fraction and crushed to a fine powder
Samples were loaded into a plastic sample holder and covered with Mylar tape
Elemental mapping of three samples were completed at SSRL on Beamline 11-2
using X-ray absorption spectroscopy (XAS) A Si(220) monochromator with a 6 microm
focusing capillary was used and fluorescence intensity was monitored with a Ge detector
Samples were prepared similar to those used for mineral identification using the electron
microprobe however the thickness of the sample was decreased to lt20 microm
Sequential Extraction Procedure
Sequential extraction experiments were completed on 2 g lt2 mm sieved soils
(Diablo clay from Site 1 at 10 cm depth and serpentine soil from Site 2 at 4 cm depth) in
three steps The exchangeable fraction was obtained using a barium chloride (1mM
BaCl2) solution for 1 h at 25 degC (Kunze and Dixon 1986) The Fe-crystalline oxide
fraction was obtained by reacting the solids with a hydroxylamine-hydrochlorideacetic
acid (10 M NH2OHsdotHCl in 25(vv) HOAc) solution for 6 h at 60 degC (Chester and
Hughes 1967) The silicate fraction was obtained by reacting the solids with
concentrated hydrofluoric acid (10 M HF) for 16 h at 25 degC Trace element
concentrations for each extraction were measured on the supernatant using inductively
coupled plasma optical emission spectrophotometry (ICP-OES) All analyses were
completed in triplicate Additionally sequential extraction experiments were completed
on 2g 106-250 mm sieved ultrasonically cleaned commercial grade chromite (MSDS
No 302-1 WHMIS Class D2A) using the same procedures listed above The
stoichiometric composition of chromite obtained using the electron microprobe is
(Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 Chromite was imaged after each sequential
extraction using a JEOL 5600LV environmental SEM equipped with an EDAX thin-
window energy dispersive X-ray detector
17
RESULTS
Serpentinite Mineralogy
The average serpentinite at Jasper Ridge is composed of serpentine (mainly
lizardite and antigorite 70) chlorite (15) talc (10) magnetite (4) and chromite
(1) based on petrographic and microprobe observations of five samples (JR3E JR6E 5
10 and JR3) however brucite has been reported as a major constituent in related
Franciscan Complex serpentinites (Hostetle et al 1966) Other minerals identified
include olivine augite and enstatite Although the serpentinite mineralogy is relatively
simple the modes and textures of the minerals present in the rock vary significantly from
site to site In Table 21 whole rock serpentinite compositions from Sites JR1 JR3E
JR6E 2 5 10 and JR3 are listed The average Cr concentration of the serpentinites is
~1700 mg kg-1 (tab 1) Iron (~7) Mg (~11) Ni (~3000 mg kg-1) and Co (~100 mg
kg-1) concentrations are also elevated in the serpentinite relative to the chert and
sandstone bedrock adjacent the main serpentinite body (Figure 21b and Table 21)
A range of rock and mineral textures are petrographically evident in the
serpentinite at Jasper Ridge (Figure 23) No foliations or lineations are observed in the
serpentinite (Figure 23a) obtained from the bottom of the soil pit excavated at Site JR3
(Figure 21b) Subhedral chromite and magnetite grains are randomly interspersed
amongst the serpentine matrix (Figure 23a) In serpentinite obtained in outcrop at Site
JR5E (Fig 21b) a lsquocrocodile skinrsquo texture is produced by thin irregular veins of fine-
grained subhedral magnetite separating serpentine grains (Figure 23b) In Figure 23c
numerous crosscutting veins of chlorite occur in orthopyroxene and define the foliation of
the rock in outcrop at Site JR3E (Figure 21b) The numerous crosscutting microscopic
veins random foliation patterns (or the lack thereof) and subhedral oxide grains suggest
the Jasper Ridge serpentinite has undergone extensive physical and chemical alteration
typical of most serpentinites
Microprobe analyses of minerals in the serpentinite demonstrate that the main Cr-
bearing phases are chlorite enstatite augite Mg-Al chromite Cr-magnetite and a
chromite-silicate mixture (defined in the following paragraph) Selected compositions
for these minerals are listed in order of their Cr concentrations in Table 22a Chlorite
contains the least Cr whereas the chromite-silicate mixture is the most enriched with
18
respect to Cr Figure 24 shows correlations of Cr content relative to Al Fe Mg and Si in
order to provide insight into the serpentinite mineralogy and to aid in evaluating trends of
Cr substitution
A fine-grained mixture containing spinel (chromite andor Cr-magnetite) and an
unidentifiable Cr-silicate contains the highest Cr concentration in the serpentinite at
Jasper Ridge (Table 22a shown in Figure 25a and 25b) This fine-grained mixture will
be referred to here as the chromite-silicate mixture (CSM) with chemical properties
similar to those in Table 22a and Figure 24 CSM is common around the rims of
protolith chromite grains and a number of the larger (~05 mm) Cr-magnetite grains in
the serpentinite The microcrystalline phase responsible for contributing Si to these
analyses is possibly chlorite due to similarities of false colors generated by backscattering
electron (BSE) images and chlorite commonly rimming CSM Microprobe analyses of
CSM listed in Tables 22 and 23 are considered as mixtures of several mineral phases
due to the fine-grained texture (lt5 microm) Chromium concentrations range from 12 to 16
Cr Wt in this phase CSM is chemically distinct from the chromite and magnetite as
shown in Figure 24
The chemical compositions of protolith chromite exhibit little to no variation in
any of the serpentinites collected at Sites 2 JR3E JR4E JR5E JR6E JR7E and JR3
(Table 22a and Figure 24) Chromite contains low Fe (~9 Wt ) high Mg (~9 Wt
) and high Al (~26 Wt ) compared to Cr-magnetite The average Cr concentration
for chromite is ~12 Cr Wt Additionally chromite grains are irregular and subhedral
suggesting that they have undergone significant alteration during metamorphism
Chromium concentrations in the magnetite vary depending on the proximity and
degree of alteration of the chromite (Table 22a and Figure 24) Magnetite grains with
less than 2 Cr Wt are common in veinlets in serpentinite not formed adjacent
chromite whereas magnetite with 8 to 12 Cr Wt were formed proximal or adjacent
chromite Chromium-enriched magnetite is also referred to as ferrite-chromite in the
literature (Spangenberg 1943 Burkhard 1993)
Chlorite and pyroxenes contain minor concentrations of Cr (lt1 Wt Table
22a) Aluminum Fe and Mg concentrations varied in the pyroxenes however the
majority of chlorites are relatively similar in composition (Figure 24) Chlorites are
19
more abundant in the serpentinite compared to the pyroxenes and both minerals were
subhedral and randomly distributed over the entire sampling area (Figure 21b)
Backscattering electron (BSE) images of what appeared to be a petrographically
opaque single chromite grain in Figure 23a demonstrates chromite has been
metasomatically altered resulting in the formation of Cr-magnetite Cr-chlorite and CSM
(Figure 25a) Represented compositions for these minerals (indicated by letters in Figure
25a) are listed in Table 23 Magnifying the chromite and CSM (Figure 25a)
demonstrates the fine-grained texture of CSM and an alteration rim (transition zone) is
also seen along the chromite edge (Fig 25b) This alteration rim is depleted in Cr with
respect to the chromite as shown in Figure 25b
In Figures 23a and 25c Ni was identified in three phases in the bedrock
including olivine (F 03 Ni Wt ) serpentine (G and H 03 and 0 Ni Wt ) and Ni-
Fe (I sim55 Ni Wt ) alloy (Figure 25c and 25d) their compositions are indicated by
letters and listed in Table 23 Alteration of the olivine resulted in the formation of lt10
microm Ni-Fe alloys shown in Figure 25c These Ni-Fe alloys form a corona surrounding
the remaining olivine and the illustration in Figure 25d highlights the minerals and
chemical changes
Chromite and Cr-magnetite alteration leading to the paragenesis of CSM and Cr-
chlorite is common in the serpentinites at Jasper Ridge The petrographic image in
Figure 26a and its corresponding BSE image in Figure 26b illustrate Cr-magnetite
(originally protolith chromite) partially altering to CSM and Cr-chlorite The chemical
compositions of each mineral (indicated by letters in Figure 26b) are listed in Table 23
CSM appears to marginally and metasomatically replace Cr-magnetite and is enriched by
~2 Cr Wt Chromium-chlorite is present in microcracks in Cr-magnetite and
encompasses the entire Cr-magnetite and CSM grain Serpentine minerals adjacent the
Cr-chlorite contains no measurable Cr
Distribution of Chromium in Serpentine Soils
Serpentine soils and underlying bedrock were collected at ten sites along a 500
meter transect extending the length of the main serpentine body as shown in Figure 21b
20
The range of Cr concentrations at each site (center) and Cr concentrations with respect to
depth (periphery) are shown in Figure 27 and soil analyses are listed in Table 24
Bedrock Cr concentrations for Sites 2 5 and 10 are also shown in these plots (Fig 27)
Chromium enrichment does not appear to be consistently found at a particular depth or
soil horizon for any of the sites Sites 3 9 and 10 demonstrate a Cr maximum at
approximately 006 m however Sites 1 2 5 and 6 exhibit a Cr minimum at roughly the
same depth The highest soil Cr concentration in the soil along the transect is 4760 mg
kg-1 at Site 3 at 006 meters depth The lowest soil Cr concentration is 1725 mg kg-1 at
Site 10 at 0085 meters depth The average bedrock Cr concentration is ~2000 mg kg-1
(Figure 27 Table 24) It is important to note Site 1 consists of Diablo clay not
serpentine soil however this site is located adjacent and down-slope of the main
serpentinite body From these observations it is apparent that there are no systematic
trends in soil Cr concentrations with respect to the position of the site and depth and soil
samples often contain higher concentrations of Cr with respect to their corresponding
bedrock and to the average Cr concentration of ultramafic rocks (1600 mg kg-1 Green
1972)
Trace metals including Ni Mn and Co were also evaluated in the serpentine soil
for Sites 1 through 10 and the results are listed in Table 24 Nickel concentrations range
from 1590 to 3960 mg kg-1 and do not directly correspond to the Cr fluxes with respect
to depth (Table 24) Nickel concentrations are less than Cr from Sites 1 to 4 From Sites
5 to 10 Ni concentrations exceed those of Cr averaging ~3600 mg kg-1 Unlike Cr Ni
concentrations are equivalent or greater in the bedrock (~3300 mg kg-1) as compared to
the soil (~3200 mg kg-1) Soil Mn concentrations at each site are relatively uniform at
~1800 mg kg-1 with little to no variation with depth (Table 24) Cobalt concentrations
do not exceed 238 mg kg-1 (Table 24) and are relatively uniform (~200 mg kg-1) with
respect to location along the transverse and depth at each site
Major elements including Al Fe Ca and Mg are listed for the serpentine soils
along the transect with respect to depth in Table 24 Aluminum Fe Ca and Mg
concentrations in the soils at each site are similar with exception of Site 1 (Diablo clay)
Aluminum concentrations in the Diablo clay are ~46 whereas the serpentine soils
contain less than 2 Al (Table 24) Iron and Mg concentrations are also low lt65 and
21
lt411 respectively in the Diablo clay compared to Fe (gt8) and Mg (gt6)
concentrations in the serpentine soil (Table 24) Calcium concentrations are low (lt1)
for all the serpentine soils and even the Diablo clay (Table 24) With a broad
understanding of the chemical characteristics of the serpentine soil at Jasper Ridge Site
JR3 (Figure 21b) was chosen for a detailed chemical study to examine all the factors
governing the chemistry of Cr in the soil
Serpentine Soil Chemistry and Mineralogy at Site JR3
Soil chemistry
Chemical properties (pH electrical conductivity organic matter nitrogen and
ammonium acetate extractable cations) and textures for four soil samples from Site JR3
are listed in Table 25 The pH of the soil is near neutral (pH from 671 to 698) and
increases slightly with respect to depth Electrical conductivities (EC) in the soil ranges
from 03 to 09 dSM a range indicative of relatively few dissolved salts andor major
dissolved inorganic solutes The organic carbon content of the soil is 32 at the surface
and decreases with depth to 10 near the bedrock Total nitrogen for the samples ranged
from 003 to 064 Ammonium acetate extractable Ca K Mg and Na are listed in
Table 25 The extractable concentration of both Ca and K decrease with depth whereas
there is no apparent trend associated with the concentrations of Mg and Na The
serpentine soil textures are dominantly clay loams however the texture is a sandy clay
loam near the bedrock
Whole soil samples and size fractions including gt2 mm sand (2-002 mm) silt
(002-0002 mm) and clay (lt2 microm) size fractions with respect to depth were chemically
digested and analyzed for total trace metal concentrations shown in Figure 28 and listed
in Table 26 to determine if the abundance of Cr was linked to a particular size fraction
(the percent of sand silt and clay size fractions are reported in Table 25) Additionally
Ni Mn and Co concentrations with respect to depth and size fraction are also presented in
Figure 28 and Table 26 Total and gt2mm size fraction Cr concentrations are similar at
the top and bottom of the soil profile ranging from 3297 to 5976 Cr mg kg-1 Chromium
concentrations exceed detections limits (gt10000 mg kg-1) for the sand-size fraction It is
important to note that the sand-size fraction corresponds to the size of most chromite
22
grains observed in the serpentinite Chromium in the clay fractions are low (~1000 mg
kg-1) compared to the other size fractions however these concentrations are still enriched
compared to non-serpentine soils
In Figure 28 and Table 26 Ni concentrations are enriched in the clay-size
fraction with values ranging from 2840 to 2920 mg kg-1 The silt-size fractions have the
lowest Ni concentrations (~1800 mg kg-1) Manganese concentrations are slightly
elevated relative to total Mn in the sand-size fraction with values ranging from 1728 to
2701 mg kg-1 The clay-size fraction is the least Mn enriched size-fraction with
concentrations at ~1200 mg kg-1 Cobalt concentrations are elevated in the sand-size
fraction with values ranging from 159 to 241 mg kg-1 and concentrations in the gt2mm
silt and clay-size fractions are similar to the total (~150 mg kg-1)
Soil Cr-bearing minerals
Chromium-bearing minerals in the soil identified using the electron microprobe
include chromite Cr-magnetite and CSM Chromium concentrations in soil chromite
Cr-magnetite and CSM are graphically shown in Figure 29 as a function of Al Fe Mg
Si and Al+Fe in Wt Representative analyses of the Cr-bearing soil minerals are listed
in Table 22b Compositions of soil chromite in Figure 29 are chemically distinct from
those identified in the protolith in Figure 24 Chromium concentrations in soil chromites
range from 22 to 37 Cr Wt containing minor concentrations of Al and Mg Values for
chromite in the serpentinite bedrock (Table 22a and Figure 24) and the FeCr2O4 end-
member are also shown in Figure 29 The dashed lines in Figure 29 connect Jasper
Ridge bedrock chromite (Figure 24) and stoichiometric FeCr2O4 where it can be seen
that compositions of soil chromite lie close to these tie lines Overall soil chromite
demonstrates a decrease in Al and Mg and an increase in Fe relative to Cr compared with
protolith chromite Soil Cr-magnetites contain between 7 and 20 Cr Wt and are
enriched in Cr compared to Cr-magnetites identified in the bedrock (Figure 24) Cr-
magnetite however reveals a decrease in Fe relative to Cr Minor concentrations of Si
(lt2 Wt ) in the soil chromite and Cr-magnetite may represent slightly altered rims or
contamination from sample preparation Chromium concentrations in CSM maintain a
consistent Cr concentration of ~11 Cr Wt similar to those found in the bedrock (~14
23
Cr Wt ) Soil CSMs with compositions between 8 and 12 Mg Wt are more than
likely associated with analyses containing more microcrystalline chlorite compared to
soil CSMs with less than 2 Mg Wt Soil CSM appears be less enriched with Si (lt14
Wt ) relative to those identified in the serpentinite (15 to 27 Wt )
The size of the Cr-bearing minerals varied significantly for each sample ranging
from 10 microm up to 2 mm The majority of chromite grains in the soil correspond to the
sand-size fraction and also vary from euhedral to subhedral with no discernable pattern
The sum of cations based on four oxygens for the chromite and Cr-magnetite analyses
range between 297 and 301 indicating these minerals are spinels not (oxy)hydroxides
Compositions obtained for chromite and Cr-magnetite may represent alteration rims
where weathering of the mineral in the subsurface environment has been most active
BSE images of chromite and Cr-magnetite were not able to definitively identify whether
these analyses in fact represent alteration rims
Mineralogy of the Clay-Size Fraction
The clay-size fraction in the serpentine soil at Site JR3 was evaluated with respect
to depth using XRD and specific chemical and heat treatments The d-spacings (Aring) and
their relative intensities for each soil and treatment with respect to depth are listed in
Table 27 Figure 210 shows the XRD analyses for clay sample JR3515 (depth 5-15
cm) and these analyses are almost identical to the other clays examined at Site JR3
Additionally Table 28 from Whittig and Allardice (1986) is provided to interpret the X-
ray diffraction analyses in Table 27 For example a d-spacing of ~14Aring is present in all
the treatments indicating the presence of chlorite in the clay-sized fraction Using an
XRD analysis program the clay-size fraction was determined to be composed of
smectite vermiculite serpentine minerals (lizardite and antigorite) and chlorite
(clinochlore and Cr-clinochlore) Smectite is the most abundant mineral in the clay-size
fraction whereas Cr-chlorite is the least abundant Overall the d-spacings and relative
intensities are similar despite the treatments and changes in depth (Table 27) This
indicates that the mineralogy of the clay-size fraction does not vary significantly in the
soil column
24
Chromium concentrations in the clay-size fractions are ~1010 mg kg-1 (Table
26) Smectite and vermiculite are possible Cr-bearing clay minerals in which isomorphic
substitution of Cr(III) could occur in the dioctahedral sheet Serpentine minerals
(lizardite and antigorite) are not likely hosts of Cr based on crystallographic
considerations and the absence of Cr in analyzed serpentine minerals in the bedrock
(Table 23) Clinochlore may also be a host for Cr in the inner layer octahedral sheet
Chromium-clinochlore is the only Cr-containing mineral identified in the clay-size
fraction using XRD and Cr-clinochlore (Cr-chlorite) is also an abundant mineral
identified in serpentinite protolith (Tables 22 and 23 Figure 26)
XAS of chromium in relation to depth
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in serpentine soils and bedrock at Site JR3 The XANES (X-ray absorption near-edge
structure) spectra demonstrate the absence of Cr(VI) in the soil column at Site JR3
(Figure 211) based on the lack of a pre-edge feature The detection limit for Cr(VI)
using this method is approximately ~20 mg kg-1 Ultimately XAS analyses demonstrate
Cr(VI) is not detected in any of the soil solids
X-ray microprobe analyses of soil and rock samples (0-5cm depth 5-10cm depth
and serpentinite) at Site JR3 were processed to map the relative concentrations of Fe and
Cr in which areas of elemental enrichment are identified by a violet color areas of
depletion by a blue color (Figure 212) Areas of Cr and Fe enrichment in Figure 212 are
localized with well-defined boundaries corresponding to grain edges and these elements
are not correlated in a well-defined trend Minerals assigned to these areas of Cr and Fe
enrichment were made based on petrographic and microprobe observations on the same
thin section For example high Fe and no Cr is magnetite whereas high Fe and high Cr
is either chromite or Cr-magnetite Chromite magnetite and Cr-magnetite appear to be
present in the serpentine soil at a depth between 0 to 5 cm (Figure 212A) In Figure
212b CSM is a possible mineral detected in the soil (depth 5-10 cm) and the presence
of magnetite is clearly seen due to the localized enrichment of Fe and no Cr Iron and
Cr spectra for the serpentinite were much more heterogeneous as compared to the soil
samples and Cr concentrations appear be more scattered (Figure 212c) These X-ray
25
microprobe analyses provide a complimentary method for examining Cr geochemistry
(including Cr(III) and Cr(VI)) in naturally occurring soil and rock solids however
Cr(III) is the only detected valence state in serpentine soils and rocks at Jasper Ridge
Sequential Extractions
Sequential extractions were completed on 1) Diablo clay (Inceptisol) from Site 1
(Figure 21b) at a depth of 10 cm 2) serpentine soil (Mollisol) from Site 2 (Figure 21b) at
a depth of 4 cm and 3) commercial grade chromite (106-250 microm) and these results are
shown in Figure 213 and listed in Table 29 Chromium extracted from the
exchangeable fraction is minimal (lt01 mg kg-1) for all the samples (number 1 in Table
29 Fig 213) Minor Cr concentrations are associated with the Fe-crystalline oxide
fraction in the Diablo clay (19 Cr mg kg-1) and serpentine soil (78 Cr mg kg-1) whereas
chromite only released 05 Cr mg kg-1 (number 2 in Table 29 Figure 213) Increased
concentrations of Ni and Fe are present in the Fe-oxide fraction of the Diablo clay (161
Ni mg kg-1 and 4474 Fe mg kg-1) and serpentine soil (5138 Ni mg kg-1 and 5138 Fe mg
kg-1 Table 29) Chromium extracted from the Diablo clay serpentine soil and chromite
in the silicate fraction is 14 31 and 14 Cr mg kg-1 respectively (number 3 in Table 29
Figure 213) The sum of extracted Cr Fe Ni and Mn concentrations (number 4 in
Table 29) are low with respect to total metals (number 5 in Table 29) In fact only 07
12 and 00 Cr was removed from the Diablo clay serpentine soil and chromite
respectively (number 6 in Table 29)
Chromite was extremely resistant to all the extractions and less than 001 of the
total Cr was removed (Table 29) Scanning electron microprobe (SEM) images of the
chromite after each extraction are shown in Figure 214 Chromium concentrations
released from the chromite (Table 29 and Figure 214a) increased with each sequential
extraction experiment starting with the exchangeable fraction (Figure 214b) progressing
into the Fe-crystalline oxide fraction (Figure 214c) and ending with the silicate fraction
(Figure 214d) The surface of the chromite became smoother after each extraction and
crystal imperfections in defect sights located along the edges and corners progressively
decreased (Figure 214) Chromite does not appear to be an easily accessible source of
Cr based on these extraction experiments
26
DISCUSSION
Serpentinite at Jasper Ridge has undergone significant physical and chemical
alteration since its formation and emplacement off the California continental margin
Despite being subjected to high pressures (~5-7 kbars) and elevated temperatures (200-
300degC) trace element concentrations within the serpentinite remain elevated Jasper
Ridge serpentinites are enriched with Cr Co Fe Mg Mn and Ni compared to chert and
sandstone located adjacent the main serpentinite body (Table 21 and Figure 21b) Over
the past 400000 years Cr-bearing minerals in the serpentinite have been weathering and
accumulating in the soil During this time Cr concentrations in the soil have
progressively increased relative to Cr-concentrations in the serpentinite protolith This Cr
enrichment is directly linked to specific weathering-resistant minerals identified in the
soil
Chromium in Serpentinites
Three minerals in the bedrock at Jasper Ridge contain appreciable concentrations
of Cr Cr-magnetite (82-103 Cr Wt ) chromite (~108 Cr Wt ) and CSM (~133
Cr Wt ) Cr-magnetite and CSM were derived from the alteration of primary igneous
chromite CSM or Cr-spinels with high concentrations of Si have been noted by
(Burkhard 1993) as a Cr-spinel containing a possible microscopic silicate phase Our
analyses support this result in which the microcrystalline silicate is probably Cr-chlorite
due to its common occurrence with CSM (Figure 26) Definitive identification of this
microcrystalline mixture was not possible due to size of the crystals relative to the spot
size of the electron microprobe Chlorite and pyroxene contain minor concentrations of
Cr (~1 Wt ) and these were the only identified Cr-containing silicates in the
serpentinite Minor concentrations of Cr observed in some chlorites are directly related
to the metasomatic alteration of chromite and Cr-magnetite (Figure 26) Other common
metasomatically-derived Cr-bearing silicates such as fuchsitemariposite and uvarovite
are not present possibly due to the lack of aqueous K+ and Ca+2 in the alteration fluids
Overall the chemical compositions of these protolith minerals (Figure 24) demonstrate
only minor chemical variations allowing comparative analysis with Cr-bearing soil
minerals
27
Chromium in Serpentine Soils
Cr-bearing soil minerals identified using the electron microprobe include
chromite Cr-magnetite and CSM and these phases are chemically distinct from those in
the bedrock (Figure 24 and 29) The highest concentration of Cr (41 Wt ) in the soil
is in chromite with a composition approaching the FeCr2O4 end-member This represents
a ~80 increase in the Cr concentration relative to bedrock chromite (see Figure 215)
Cr-magnetite also demonstrates Cr enrichment (up to 20 Cr Wt ) with respect to
magnetite identified in the bedrock (~10 Cr Wt ) (Figures 24 and 29) The textures of
the soil Cr-bearing spinels exhibit sharp irregular edges possibly due to extensive
weathering CSM in the soil is similar to those found in the bedrock with Cr
concentrations ranging from 11 to 14 Cr Wt however Si concentrations in soil CSMs
(8 to 14 Wt ) are noticeably less than those measured in the serpentinite (15 to 20 Wt
) The weathering of CSM might preferentially release Si relative to other elements in
its structure Cr-silicates (chlorite and pyroxenes) were not identified in the soil possibly
due to complete weathering of these phases or they were too fine-grained for microprobe
analyses however Cr-clinochlore was identified in the clay-size fraction using XRD
Elemental maps shown in Figure 212 corroborate electron microprobe observations by
revealing well-defined zones of Cr and Fe enrichment in both the soils and serpentinite
These areas of Cr and Fe enrichment are either chromite Cr-magnetite CSM or
magnetite Overall these maps reveal Cr is not homogeneously distributed in the soil
confirming Cr is in specific mineral phases Additionally the observed abundance of
chromite Cr-magnetite and CSM in the soils indicates that these minerals are probable
sources for a majority of the Cr enrichment identified in the soils
Soil Spinels
Soil chromite and Cr-magnetite appear to have undergone chemical alteration
enriching the minerals with respect to Cr In Figure 215 compositions of bedrock
chromite and end-member FeCr2O4 are shown together with tie-lines like those in figure
8 connecting both minerals Reaction progress in Figure 215 is based on the assumption
that the serpentinite chromite (0 reaction progress) is proceeding toward FeCr2O4
(100 reaction progress) in the serpentine soil Soil chromites from Jasper Ridge are also
28
plotted in Figure 215 where Cr concentrations are assumed to vary linearly between the
bedrock chromite and chromite end-member which defines the reaction progress of the
mineral These tie-lines correlate extremely well to the chemical composition of
chromites identified in the soil For example chromite containing ~30 Cr Wt will
have ~15 Fe Wt ~13 Al Wt and ~5 Mg Wt similar to soil chromite in Figure 29
The retention of Cr in chromite shown in Figure 215 has significant implications towards
the availability and cycling of Cr in serpentine soils
One possible mechanism for the chromite composition trends illustrated in
Figures 29 and 215 is incongruent dissolution where Cr is retained relative to other
elements Sequential extraction experiments (Figure 213 and Table 29) support this
idea in which Fe is more easily released than Cr (1096 Fe mg kg-1 to 19 Cr mg kg-1)
from chromite Unfortunately Mg and Al were not monitored during the experiments
From these sequential extraction experiments and chromite analyses (Figure 215) Al
and Mg have been preferentially released from the chromite structure compared to Fe and
Cr Additionally the alteration of chromite in the serpentinite and the soils are analogous
to one another Chromite in the serpentinite retains a majority of its Cr compared to Mg
and Al when being altered to CSM by fluids at elevated pressures and temperatures
(Figure 26 and Table 22) Under ambient soil conditions chromite is once again
retaining Cr relative to Mg and Al causing what appears to be an increase in Cr
concentration Chromium is not easily released from chromite whether under high
pressures and temperatures or in the soil
Crystal field stabilization energies (CFSE) and octahedral site preference energy
parameters (OSPE) provide a means to explain the incongruent dissolution or Cr
preservation in the Cr-spinels CFSE is a measure of the net energy of stabilization
gained by a metal ions nonbonding d electrons as a result of complex formation This
measure also indirectly describes the energy required to remove an ion from a crystal
structure The octahedral CFSE for Cr(III) is ndash2247 kJmole and the OSPE (the
difference between the tetrahedral and octahedral CFSE) is ndash1578 kJmole (Dunitz and
Orgel 1957 McClure 1957) These values are the most negative for any transition
element and indicating the presence of Cr(III) in an oxide increases the crystalrsquos stability
compared to an analogous crystal lacking Cr(III) Chromium(III) would be among the
29
last ions to removed from the spinel structure compared to other transition elements such
as Fe2+and Mn2+
Soil Chemistry related to Chromium Availability
In Table 25 the chemical properties of the soil at Site JR3 do not appear to be a
factor when evaluating Cr in the soil The pH of the soil did not vary considerably with
depth Additionally the near neutral pH would not have a significant impact accelerating
Cr-bearing mineral dissolution Organic matter within the soil was elevated despite the
low plant productivity of the serpentine soil The role of organic matter could be a
significant factor such as the production of organic acids promoting the dissolution of the
Cr-bearing minerals however Cr concentrations near the surface (where organic matter
was the greatest) are not significantly greater or less than those deeper in the soil profile
The relative concentrations of ammonium acetate extractable Ca K Mg and Na are
characteristic of serpentine soils and did not provide any new insight Finally Cr(III)
was the only identifiable valence state of Cr based on X-ray absorption spectroscopy
(Figure 211) which is what should be expected based on the Cr-bearing minerals and the
pH of the soil
Extraction experiments summarized in Table 29 and Figure 213 demonstrate
~99 of the Cr in a serpentine soil is retained in highly resistant phases at Jasper Ridge
The chromite standard used in these experiments contributed negligible concentrations of
Cr (lt2 mg kg-1 total) for all of the sequential extraction experiments supporting earlier
evidence that Cr-spinels are retaining Cr and are potentially responsible for a majority of
the total Cr enrichment in the serpentine soil At Jasper Ridge Cr extracted from the
serpentine soil (31 mg kg-1) is likely the result of the Cr-silicates compared to the Cr-
spinels Low Cr-containing minerals (possibly chlorite enstatite augite and clay
minerals) appear to be the most probably source of Cr to soil solutions and vegetation of
the serpentine soil Approximately 30 of the serpentine soil is composed of clay-sized
minerals and this size-fraction has ~1000 Cr mg kg-1 The mineral responsible for Cr
enrichment in the clay-size fraction is more than likely Cr-chlorite identified using XRD
Chromium-chlorite in the soils suggests that this mineral is potentially the largest silicate
contributor of Cr in the serpentine soil
30
Interpreting the enrichment and behavior of Cr in serpentine soils such as in
Figure 27 can be accomplished based on the abundance Cr-spinels observed in the soil
Small grains of chromite Cr-magnetite and even CSM in the soil have the potential to
appreciably impact Cr concentrations Table 210 demonstrates the chromite masses
required to change the Cr concentration in a 1 kg sample of soil The highest Cr
concentration at Jasper Ridge was identified at Site 3 with a measured value of 4760 mg
kg-1 which means only ~1 of the soil sample (Table 210) needs to contain chromite
Additionally a minimal amount (lt011 of the sample as shown in Table 210) of
chromite is necessary to change the concentration by 500 mg kg-1 which could easily
cause the sharp Cr increases and decreases noted in Figure 27 This agrees with our
observations of chromite abundance in the serpentine soils Additionally a majority of
the chromite grains identified in the bedrock are between 2 and 002 mm which
correspond directly to the elevated Cr concentrations (gt10000 mg kg-1) in the sand-size
fraction of the serpentine soil (Table 26) Even the ~15000 Cr mg kg-1 values in soils
measured by Hotz (1964) in Figure 22 only require ~2 of the sample of the sample to
be chromite A few grains of chromite possess the ability to significantly impact
measured Cr concentrations in soil samples
Chromium and Nickel in Serpentine Soils
Chromium and Ni have often been correlated to one another due to their common
enrichment in serpentine soils (Hotz 1964 Rabenhorst et al 1982 Schwertmann and
Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al 1989 Kaupenjohann
and Wilcke 1995) A majority of the Ni in the Jasper Ridge serpentine soil originated
from olivine (Figure 25c and 25d Table 23) Chemical alteration of olivine during
serpentinization resulted in the mineral serpentine (Coleman and Jove 1992) in which
minor concentrations of Ni (lt037 NiO Wt ) substituted into its structure Continued
alteration under highly reducing conditions and the incompatibility of Ni in the serpentine
structure resulted in the formation of Ni-Fe alloy (Figure 25c) This type of alteration is
observed in many of the rocks at Jasper Ridge where the Ni-Fe alloys are ~10 microm in size
Olivine in the rocks and soil appear to be less resistant to weathering compared to
chromite therefore Ni is possibly more available to bacteria and vegetation in serpentine
31
soils Sequential extractions (Figure 213 Table 29) support these observations in which
a greater percentage of total Ni was extractable compared to the Cr in the soils Although
Cr and Ni are enriched in serpentine soils the availability and behavior of these two
elements are significantly different
CONCLUDING REMARKS
Chromium is predominantly present in primary chromite Cr-magnetite and a
chromite-silicate mixture (CSM) in both the bedrock and soil despite weathering over
geologic time Sharp variations in soil Cr concentrations with respect to depth are caused
by the presence or absence of chromite Cr-magnetite andor CSM Chromium-spinels in
the soil have undergone incongruent dissolution possibly conserving Cr(III) with respect
to Al3+ and Mg2+ Chromite is highly resistant to weathering and it is more than likely
not contributing bioavailable Cr Although Cr-spinels are a major source of Cr Cr-
silicates in the protolith (chlorite enstatite and augite) and soil (chlorite
montmorillonite and vermiculite) afford the most accessible Cr source identified Based
on our observations and analyses chlorite is the most likely Cr source of these silicate
minerals Finally Cr(VI) was not detected in any of the soil and rock samples
ACKNOWLEDGEMENTS
We thank Jasper Ridge Biological Preserve for allowing us the opportunity to
examine the serpentine soils Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund NSF Grant EAR-9902859 (DKB) GSA Graduate
Student Research Grants Office of Teaching and Learning (SF) at Stanford University is
gratefully acknowledged We appreciate the technical support from the staff of the
Stanford Synchrotron Radiation Laboratory from Bob Jones for assistance in microprobe
analyses of the serpentinites and serpentine soils and from Ben Bostick Travis Horton
and Theresa Barber for field and lab assistance
32
Table 21 Whole rock elemental concentrations of serpentinites (Figure 21b) chert and sandstone
-----------------------Location of Serpentinitesdagger----------------------- Elemental Concentrations JR1 JR3E JR6E Site 2 Site 5 Site 10 JR3
ChertDagger SandstoneDagger
Al 02 08 03 027 22 107 024 079 027 Ca 026 007 005 003 046 013 003 001 lt001 Co mg kg-1 101 99 100 99 155 217 97 1 lt1 Cr mg kg-1 580 2280 1510 1840 1945 2090 1465 10 lt1 Fe 303 71 609 581 76 1106 512 041 024 K 001 031 099 lt001 lt001 001 lt001 029 008
Mg 122 134 154 1336 524 804 112 014 034 Mn mg kg-1 495 800 155 730 1375 1860 460 110 85 Na 181 006 093 lt001 lt001 lt001 lt001 001 001 Ni mg kg-1 2037 1545 3321 3510 3020 3960 2740 20 9
Total digestion ICP-AES analyses daggerSerpentinites are dominantly composed of the minerals serpentine chlorite talc and spinels DaggerSee Figure 21b for sample locations
33
Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in serpentinite rock at Jasper Ridge
Chlorite Enstatite Augite Chromite Cr-Magnetite ChromiteSilicatedagger SiO2 3738 5463 5222 000 024 1053 Al2O3 552 477 516 5059 182 395 NiO 001 01 00 03 00 052 FeODagger 762 614 239 1446 7919 5042 MnO 009 015 009 042 140 182 Cr2O3 024 060 099 1584 1605 1939 MgO 3337 3305 1560 1869 024 605 CaO 023 050 2348 000 000 019 Na2O 000 001 050 000 001 008 K2O 000 000 001 000 000 011 Total 8446 9994 10048 10027 9894 9307
Formula Unit Compositions
Si 370 094 189 000 001 --- IVAl 030 006 011 000 000 --- VIAl 035 004 011 157 008 --- Ni 000 000 000 001 000 ---
Fe+2 063 010 007 030 232 --- Mn 001 000 000 001 004 --- Cr 002 001 003 033 048 --- Mg 493 085 084 073 001 --- Ca 002 001 091 000 000 --- Na 000 000 004 000 000 --- K 000 000 000 000 000 --- O 1400 300 600 400 400 ---
Electron microprobe analyses daggerChromitesilicate is also referred to as CSM DaggerFe+3 is present however all conversions are based on Fe+2
34
Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in serpentine soil at Jasper Ridge
CSM Chromite Cr-Magnetite SiO2 1940 057 267 Al2O3 559 1059 381 NiO 014 001 004 FeOdagger 2629 2259 5244 MnO 155 154 419 Cr2O3 1878 5577 3237 MgO 1668 916 288 CaO 015 001 005 Na2O 006 001 001 K2O 0 008 009 Total 8864 10033 9855
Si --- 002 010 IVAl --- 042 017 VIAl --- 000 000 Ni --- 000 000
Fe+2 --- 063 148 Mn --- 004 013 Cr --- 147 096 Mg --- 046 016 Ca --- 000 000 Na --- 000 000 K --- 000 000 O --- 400 400
Electron microprobe analyses dagger Fe+3 is present however all conversions are based on Fe+2
35
Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25 and 26 A B C D E F G H I J K L Chromite
Chromite CSM Magnetite Chlorite Olivine Serpentine Serpentine Ni-Fe-Metal Cr-Magnetite CSM Chlorite
SiO2 0 0 1053 0 3407 4046 3688 4074 591dagger 021 1143 2855Al2O3
5059 5162 395 034 1114 000 397 085 054 165 980 2293NiO 028 031 052 0 004 037 037 0 5302dagger 0 0 0
FeODagger 1446 1452 5042 9992
1220 901 262 453 1753dagger 8169 3498 806MnO 042 044 182 0 008 010 0 0 0 107 150 00Cr2O3 1584 1576 1939 125 071 001 0 0 0 1369 1905 039MgO 1869 1869 605 003 292 4895 3848 3755 064 1217 2828CaO 0 0 019 0 0 0 0 0 0 0 0Na2O 0 0 0 0 0 0 0 0 0 0 0K2O 0 0 0 0 0 0 0 0 0 0 0H2O --- --- --- --- 1200 --- 1600 1600 --- --- 1200Total 10027 10136 9307
10164 9944 9891 9834 9972
9897 8897 10028
Formula Unit CompositionSi
000 000 --- 000 328 100 184 200 001 --- 269IVAl 000 000 --- 000 072 000 000 000 000 --- 131VIAl 157 159 --- 002 058 000 023 005 007 --- 123Ni 001 001 --- 000 000 001 001 000 000 --- 000
Fe+2 030 029 --- 294 084 019 011 019 239 --- 059Mn 001 001 --- 000 000 000 000 000 003 --- 000Cr 033 032 --- 004 006 000 000 000 041 --- 003Mg 073 073 --- 000 419 180 285 274 004 --- 397Ca 000 000 --- 000 000 000 000 000 000 --- 000Na 000 000 --- 000 000 000 000 000 000 --- 000K 000 000 --- 000 000 000 000 000 000 --- 000O 400 000 --- 000 1400 400 700 700 400 --- 1400
36
Electron microprobe analyses dagger elemental metal Dagger Fe+3 is present however all conversions are based o
36
2321000---
10025
591000054
53021753000000
2321000000000000
n Fe+2
Table 24 Bulk soil elemental concentrations for soils and bedrock collected along the transect
(Fig 21b)
Sites Samples Depth
(meters) Cr
(mg kg-1) Ni
(mg kg-1) Mn
(mg kg-1) Co
(mg kg-1) Al
() Fe
() Ca ()
Mg ()
COI03 0015 2420 1590 1325 118 464 565 065 352 COI36 0045 2370 1630 1600 138 467 55 063 356 COI69 0075 2160 1755 1670 144 471 6 064 365 COI912 0105 2420 1785 1740 151 496 615 067 384
1
COI1215 0135 2570 1985 2340 195 506 649 069 411 COII02 001 3710 2400 1665 169 294 702 100 768 COII25 0035 3600 2290 1625 162 275 673 093 743 COII58 0065 2910 2310 1550 159 25 672 087 771
COII811 0095 3520 2350 1595 158 276 693 093 766 2
COIIBedrock 012 1840 3510 730 99 027 581 003 1336 COIII14 0025 3750 2930 1655 190 178 897 062 762 COIII47 0055 4760 3050 1850 178 178 94 056 805
COIII710 0085 4380 3410 1930 216 216 1044 07 797 3 COIII1013 0115 4320 3380 1955 201 213 1061 063 626 COIV01 0005 3710 2970 1915 214 209 909 053 781 4 COIV15 003 3540 2710 1725 195 181 801 043 792 COV14 0025 2620 2350 1330 150 246 651 063 526 COV47 0055 1775 2710 1520 165 252 766 059 517
COV710 0085 2000 3090 1830 195 225 737 057 583 5 COVBedrock 0115 1945 3020 1375 155 22 76 046 524
COVI14 0025 3040 3720 1855 224 084 1152 015 736 COVI47 0055 2400 3610 1785 218 085 1138 014 68 COVI710 0085 2960 3800 1850 232 093 1232 015 714 6
COVI1013 0115 2800 3860 1800 224 085 1164 014 739 COVII14 0025 2620 3550 1925 229 083 1128 015 718 COVII47 0055 1975 3590 1805 216 084 1085 014 655 7 COVII712 0095 2080 3600 1840 220 07 1105 011 741 COVIII14 0025 2130 3830 1990 238 085 115 017 705 COVIII47 0055 2550 3720 1865 228 085 1099 017 714 8
COVIII712 0095 2240 3910 1860 229 076 1112 014 828 COIX14 0025 2470 3480 1890 227 104 1201 023 579 COIX47 0055 3000 3830 1595 205 084 108 013 842 9 COIX713 01 2720 3710 1845 221 111 1208 017 551 COX14 0025 2810 3740 1760 218 129 114 017 675 COX47 0055 2190 3910 1835 224 131 1206 017 702
COX710 0085 1725 3370 1595 192 11 1061 014 635 10 COXBedrock 0115 2090 3960 1860 217 107 1106 013 804
Total digestion ICP-AES analyses
37
Table 25 Chemical and physical properties of the soil at Site JR3
NH4OAc-extractable ---Hydrometer---Ca K Mg Na
Sand Silt ClaySample Depth (cm) pH EC
dSm
Organic Matter ------
LECO Nitrogen ------ ---------mg kg-1---------
Texture
JR305 25 671 04 32 012 505 2366 3291 355 36 31 33 Clay LoamJR3515
10 683 09 19 064 4205 8176 3343 2392 36 30 34 Clay LoamJR31530 225 688 03 12 003 3725 4903 3750 2305 36 33 31 Clay LoamJR33045 375 698 07 10 003 2873 3665 3388 2549 46 24 30 Sandy Clay Loam
See Methods Section for more details
38
38
Table 26 Selected trace metal concentrations for each size fraction versus depth at Site JR3
Depth (cm) Total gt2mm Fraction Sand Silt Clay Sample
------------------------Chromium (mg kg-1)------------------------ JR305 25 4906 5086 gt10000 4565 988 JR3515 10 5976 4678 8284 4000 1115 JR31530 225 5222 4206 gt10000 4133 980 JR33045 375 3477 3297 gt10000 4571 956
------------------------Nickel (mg kg-1)------------------------------ JR305 25 1935 2050 1660 1305 2920 JR3515 10 2180 2050 1980 1325 2920 JR31530 225 2190 2120 1835 1370 2880 JR33045 375 2090 2140 1740 1285 2840
------------------------Manganese (mg kg-1)-------------------------- JR305 25 1319 1571 1728 1301 1210 JR3515 10 1664 1601 2701 1471 1254 JR31530 225 1639 1626 2217 1599 1231 JR33045 375 1222 1226 1781 1405 1170
-----------------------Cobalt (mg kg-1)-------------------------------- JR305 25 128 146 159 134 131 JR3515 10 155 145 241 147 132 JR31530 225 152 147 194 162 128 JR33045 375 125 124 159 141 122
Total digestion ICP-AES analyses
39
Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3
JR305CLAY Depth 0-5cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1439 100 1833 100 1274 100 1391 100 731 179 1432 431 732 645 1047 105 71 216 941 163 71 41 838 838 474 101 847 141 475 257 726 726 356 125 73 271 365 264 313 83 712 248 355 299
473 143 313 257 364 131 355 193
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1845 1000 1268 1000 1388 1000 852 68 1447 387 724 987 1010 859 733 264 944 143 703 679 836 281 712 247 903 139 471 282 723 576 475 109 850 83 362 244 365 73 733 426 352 321 356 117 712 291 313 154 314 63 475 122
364 170 355 196
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1724 1000 1297 1000 1396 1000 842 45 1397 447 741 957 1175 708 730 318 923 176 715 686 1037 944 710 255 840 118 477 271 837 139 475 97 717 465 367 400 726 375 357 133 468 182 357 386 313 91 362 194 315 286
352 229 JR33045CLAY Depth 30-45cm
Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity
1447 1000 1869 1000 1268 1000 1393 627 842 44 1476 287 737 843 1049 1000 730 207 954 107 715 539 837 190 710 182 915 107 479 157 723 111 475 69 860 61 366 303 364 69 740 332 356 292 356 74 717 256 315 315 313 57 478 111
367 141 357 115
See Methods Section for details
40
Table 28 X-ray diffraction d-spacings for common silicates from Whittig and Allardice (1986)
Diffraction Spacing (Aring) Mineral(s) Indicated Mg-saturated air-dried
14-15 Smectite vermiculite chlorite 99-101 Mica (illite) 71-73 Serpentine
715 Kaolinite chlorite Mg-saturated glycerol-solvated
177-180 Smectite 14-15 Vermiculite chlorite
99-101 Mica (illite) 71-73 Serpentinte
715 Kaolinite chlorite K-saturated air-dried
14-15 Chlorite vermiculite (with hydroxy interlayer) 124-128 Smectite 99-101 Mica vermiculite (contracted) 71-73 Serpentine
715 Kaolinite chlorite K-saturated heated (500ordmC)
14 Chlorite 99-101 Mica vermiculite (contracted) smectite (contracted) 71-73 Serpentine
715 Chlorite
41
Table 29 Results of the Diablo clay serpentine soil and chromite sequential extraction experiments
1 Exchangeable Fraction (1mM BaCl2) 4 Sum of 1 2 amp 3 (mg kg-1) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 0 0 003 Cr 159 388 19 Fe 014 011 001 Fe 24812 25532 1096 Ni 104 103 008 Ni 1854 1868 25 Mn 0 0 0 Mn 778 1172 04
2 Fe-Oxide Fraction (1M NH2OH HCl in 25 (vv) HOAc) 5 Total Cr Fe Ni and Mn (Total Digestion) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 19 78 05 Cr 2290 3310 1767864 Fe 4474 5138 38 Fe 60000 68000 1044325 Ni 1610 1506 01 Ni 1673 2345 00 Mn 231 430 00 Mn 1498 1608 3424547
3 Silicate Fraction (10M HF) 6 Percent Extractable ((45)100) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ------------------------------------- Cr 140 310 14 Cr 07 12 00 Fe 20337 20393 1058 Fe 41 38 01 Ni 234 352 22 Ni 111 80 00 Mn 547 742 04 Mn 52 73 00
ICP-AES analyses
42
Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1 kg of sample
Chromite
FeCr2O4 Cr Concentration
(mg kg-1) Grams Sample 500 11 011 1000 22 022 2000 43 043 5000 108 108
43
Figure Captions
Figure 21a Soils map of Jasper Ridge Biological Preserve (Stanford CA) with
complete descriptions of the soils shown in the legend 21b An enlarged image of the
study area shown in 22a indicates sampling sites of the soils and rocks collected A
dashed line marks the 500 m transect
Figure 22 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above
Figure 23 Petrographic images illustrating the varying serpentinite textures at Sites (A)
JR3 (B) JR5E and (C) JR3E For reference to site locations see Figure 21b
Figure 24 Correlation diagrams of Cr-bearing minerals in the serpentinite at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Si in Wt
Chromite (circle) Cr-magnetite (square) chromitesilicate or CSM (down triangle)
chlorite (hexagon) and pyroxenes (up triangle) analyses are shown on each correlation
diagram
Figure 25 Images of the serpentinite bedrock at Site JR3 demonstrating the mineralogy
and Cr-bearing phases Microprobe analyses are indicated by circled letters and these
values are in Table 23 25a Backscattering electron image demonstrating several Cr-
bearing phases present including chromite Cr-magnetite CSM and Cr-chlorite from
Figure 23a 25b Magnified image shown in Figure 25a demonstrating the fine-grained
texture of CSM Additionally Cr2O3 Wt values are shown for chromite the depletion
zone and CSM 25c Image of Ni-bearing phases present in the bedrock including
olivine serpentine and Ni-Fe alloy from Figure 23a 25d Cartoon illustrating the
mineralogical and chemical changes present in 25c
Figure 26 Images of a Cr-chlorite and Cr-magnetite assemblage in the serpentinite
bedrock 26a Petrographic image of the bedrock at Site 4 (Figure 21b) 26b
44
Backscattered electron image demonstrating the common occurrence of Cr-chlorite
encompassing chromite grains Microprobe analyses are indicated by circled letters and
are located in Table 23
Figure 27 Highest and lowest Cr concentrations (mg kg-1) at each site in the serpentine
soil 500 m transect shown in Figure 21b Samples included soils at various depths (m)
and bedrock from each site The average Cr concentration for the Jasper Ridge
serpentinite is marked on the central figure Chromium concentrations versus depth (m)
at each site (noted in the upper left corner) outline the main diagram
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected
for total gt2 mm sand silt and clay size fractions listed in Table 26
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soil at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Al+Fe in Wt for
chromite (circles) Cr-magnetite (hexagons) and CSM (diamonds) Chromite from the
bedrock is denoted by a larger circle and FeCr2O4 with a square
Figure 210 XRD clay analyses of sample JR3515 (Site JR3 Figure 21b) demonstrate
the influence of each treament Mg-saturated Mg+glycerol K-saturated and K-saturated
+ heat (500oC) The d-spacings and intensities are also listed in Table 27
Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at depths (a) 0-5cm (b) 5-15cm (c) 15-30cm (d) 30-45cm and (e) bedrock The pre-edge
area indicative of Cr(VI) peaks is shown on the right Note that no peaks are present
Figure 212 X-ray microprobe elemental maps of Cr and Fe for (212a) serpentine soil
at a depth of 0-5cm (212b) serpentine soil at a depth of 5-10cm and (212c) serpentinite
bedrock at Jasper Ridge site JR3 (Figure 21b)
45
Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments for the Diablo clay (Site 1 Figure 21b) serpentine soil (Site 2
Figure 21b) and chromite listed in Table 29 The sum refers to the summation of the
three extraction experiments
Figure 214 SEM images of the chromite standard (A) after each sequential extraction
treatment including the (B) exchangeable fraction (C) Fe-crystalline oxide fraction and
(D) silicate fraction The concentration of Cr removed by each treatment is shown
beneath each set of images
Figure 215 Tie-lines from Figure 29 connecting serpentinite chromite and end-
member FeCr2O4 are shown on this single plot Soil analyses from Figure 29 are also
plotted on this diagram For a sum of 100 Wt oxygen must be added to the sum of
Cr Al Fe and Mg Wt
46
Figure 21a
47
Figure 21b
48
Chromium Concentration (mg kg-1)1000 10000
Dep
th (m
eter
s) 01
10
Maryland (Rabenhorst et al 1982)California (Gough et al 1989)California (Hotz 1964)New Foundland (Roberts amp Proctor 1992)
Surface
Figure 22
49
Figure 23
50
Cr (Wt )0 5 10 15 20
Al (
Wt
)
0
5
10
15
20
25
30
Cr (Wt )0 5 10 15 20
Fe (W
t
)
0
20
40
60
80
100
Cr (Wt )0 5 10 15 20
Mg
(Wt
)
0
5
10
15
20
25
Cr (Wt )0 5 10 15 20
Si (W
t
)
0
5
10
15
20
25
30
Cr0 10 20 30 40 50 60 70 80 90 100
Al
0
10
20
30
40
50
60
70
80
90
100
Fe
0
10
20
30
40
50
60
70
80
90
100
Chromite
Chlorite
Magnetite
Chromite-Silicate
Pyroxenes
Chromite
Chlorite
Pyroxenes
Magnetite
Chromite-Silicate
Figure 24
51
Figure 25
52
Figure 26
53
Sites1 2 3 4 5 6 7 8 9 10
Con
cent
ratio
n (m
g kg
-1)
1500
2000
2500
3000
3500
4000
4500
5000Soil Chromium
1
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000002004006008010012014016
2
Cr Concentration (mg kg-1)
1500 2000 2500 3000 3500 4000
Dep
th (m
eter
s)
000002004006008010012014
3
Cr Concentration (mg kg-1)3500 4000 4500 5000
Dep
th (m
eter
s)
000
002
004
006
008
010
012
5
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000002004006008010012
6
Cr Concentration (mg kg-1)2200 2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000002004006008010012
7
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000
002
004
006
008
010
9
Cr Concentration (mg kg-1)
2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000
002
004
006
008
010
012
10
Cr Concentration (mg kg-1)1600 2000 2400 2800 3200
Dep
th (m
eter
s)
000002004006008010012
4
Cr Concentration (mg kg-1)3500 3600 3700 3800
Dep
th (m
eter
s)
000
001
002
003
004
8
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000
002
004
006
008
010
BEDROCK
BEDROCK
BEDROCK
Range of Cr Concentrations in Serpentinite Bedrock
Figure 27
54
Soil Chromium
Depth (cm)0 10 20 30 40 50
Cr m
g kg
-1
02000400060008000
100001200014000
Totalgt2mmSandSiltClay
Soil Nickel
Depth (cm)0 10 20 30 40 50
Ni m
g kg
-1
0
1000
2000
3000
4000
5000
Soil Manganese
Depth (cm)0 10 20 30 40 50
Mn
mg
kg-1
0
1000
2000
3000Totalgt2mmSandSiltClay
Soil Cobalt
Depth (cm)0 10 20 30 40 50
Co
mg
kg-1
0
50
100
150
200
250
300
Totalgt2mmSandSiltClay
Totalgt2mmSandSiltClay
Exceeds detection limits of 10000 mg kg-1 Cr
Figure 28
55
Cr (Wt )0 10 20 30 40 50
Al (W
t
)
0
5
10
15
20
25
30
Cr (Wt )0 10 20 30 40 50
Mg
(Wt
)
0
2
4
6
8
10
12
Cr (Wt )0 10 20 30 40 50
Si (W
t
)
0
2
4
6
8
10
12
14Cr (Wt )
0 10 20 30 40 50
Fe (W
t
)
0
10
20
30
40
50
60
70
Cr (Wt )0 10 20 30 40 50
Al +
Fe
(Wt
)
10
20
30
40
50
60
70
JRBPChromite
FeCr2O4
Chromite
Magnetite
Chromite-Silicate
Figure 29
56
Clay XRD of JR3515 Depth 5-15cm
0
200
400
600
800
1000
1200
0 5 10 15 20 25 30 35 40 45
d-spacing (Aring)
Rel
ativ
e In
tens
ity
Mg-Saturated
Mg+Glycerol
K-saturated
K-saturated+heat (500oC)
Figure 210
57
Energy (kev)
597 598 599 600 601 602 603
Nor
mal
ized
Abs
orba
nce
-4
-3
-2
-1
0
1
2
(a)
(b)
(c)
(d)
(e)
Cr(
VI)
Figure 211
58
Figure 212
59
Diablo Clay
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Serpentine Soil
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Chromite
0
20
40
60
80
100
120
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide
SilicateSum
Figure 213
60
Figure 214
61
Reaction Progress ()0 20 40 60 80 100
Elem
enta
l Wt
0
10
20
30
40
50
ChromiumAluminumIronMagnesium
FeCr2O4Jasper Ridge
Bedrock Chromite
SoilChromite
Figure 215
62
CHAPTER 3
OXIDATIVE PROMOTED DISSOLUTION OF CHROMITE BY MANGANESE
DIOXIDE AND CONCURRENT PRODUCTION OF CHROMATE
ABSTRACT
Chromium release and oxidation from chromite is a potential environmental hazard in
sediments and soils and a pathway for soil development (weathering of primary minerals)
related to ultramafic rocks and their metamorphic derivatives (serpentinites) Birnessite is
a common pedologic mineral in these sediments and soils capable of oxidizing aqueous
Cr(III) In this study the interaction between chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4)
and birnessite is investigated with an interest in the potential generation of Cr(VI)
Specifically the effects of chromite suspension density birnessite suspension density
and pH at a temperature of ~25degC in relation to the reaction kinetics were examined The
rate of Cr(VI) in solution increases with increasing chromite suspension densities and
decreasing pH but is independent of birnessite suspension densities at values greater than
~20 m2 L-1 The overall rate expression of Cr released and oxidized in solution from
chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression with Cr(VI) rates ranging from 9times10-4 to 44times10-3 microM h-1
Additionally these experiments demonstrate that serpentine soils are a source of non-
anthropogenic Cr(VI)
63
INTRODUCTION
Chromite a Cr-enriched spinel is responsible for a majority of the Cr enrichment
identified in ultramafic rocks serpentinites and serpentine soils (Rabenhorst et al 1982
Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989 Gasser and
Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003 Oze et al
2003a Oze et al 2003b) The chemical weathering of chromite is therefore an important
process when evaluating the potential of ultramafic rocks and their related sediments and
soils as sources of Cr Despite the propensity of chromite to resist chemical weathering
the dissolution of chromite and release of Cr will occur over geologic time (Bialowolska
and Salacinski 1984 Oze et al 2003b) Chromium in chromite is in the form of Cr(III)
the non-toxic and chemically immobile valence state however oxidizing conditions in
near surface environments is a potential reaction leading to the enhanced dissolution of
chromite and the formation of Cr(VI) the toxic and chemically mobile valence state
which typically exists as the oxyanion chromate (HxCrO42-x) (Daugherty 1992 Cohen et
al 1993 Fendorf 1995 Deer et al 1996) Consequently chromite may be a non-
anthropogenic source of Cr(VI) in the environment
Naturally occurring high valence Mn oxides present in soils and sediments
derived from ultramafic rocks are strong oxidizers capable of affecting the chemical
mobility of metals through redox reactions (Fendorf and Zasoski 1991 Johnson and
Xyla 1991) These insoluble mixed valent oxides (Mn(IIIIV)) are observed coating
mineral surfaces in these ultramafic sediments and soils where total Mn concentrations
are often gt100 mg kg-1 (Brooks 1987 Schreier et al 1987 Oze et al 2003a Oze et al
2003b Oze et al) Birnessite is a naturally occurring mixed valence oxide known to
oxidize aqueous Cr(III) to the more chemically mobile Cr(VI) (Eary and Rai 1987
Fendorf and Zasoski 1991) Gough et al (1989) and Becquer et al (2003) have
suggested that Mn oxides are responsible for Cr(VI) identified in California and New
Caledonia serpentine soil solutions however the dissolution and oxidation of Cr from a
Cr-bearing mineral due to the presence of birnessite has not been quantitatively defined
In this study we examine the kinetics of Cr release and oxidation when chromite and
birnessite are present in solution
64
The objective of this study is to determine a rate expression describing the
release and oxidation of Cr from chromite in the presence of birnessite Using well-
controlled batch experiments the solution chemistry and geochemistry of Cr related to
chromite and birnessite interaction was evaluated temporally over a range of pH
Conditions were chosen to simulate the slightly acidic conditions characteristic of most
serpentine soils (Rabenhorst et al 1982 Gough et al 1989 Graham et al 1990 Cole
1992 Jeffrey 1992 Lee 1992 Verger 1992 Becquer et al 2003) Ultimately these
experiments allow the examination of one potential pathway leading to Cr(VI) identified
in solutions related to ultramafic rocks and serpentine soils
MATERIALS AND METHODS
Solid Phase Characterization and Synthesis
Chromite used in the experiments was commercial-grade chromite (MSDS No
302-1 WHMIS CLASS D2A) obtained from Barnes Environmental Inc Ontario
Chromite was prepared by cleaning the sand-size material in multiple ultrasonic
isopropanol baths and rinses The chromite was then sieved using a Retsch magnetic
sieve shaker for 30 minutes in order to obtain the 160-250 microm size fraction Impurities in
the 160-250 microm chromite were removed using a slope Franz magnetic separator These
impurities included the minerals olivine augite plagioclase feldspar and quartz Visual
and scanning electron microprobe (SEM) examinations of the remaining 160-250 microm
fraction confirmed its purity Chromite was crushed to a fine powder using an agate
mortar and pestle and then rinsed multiple times in deionized destilled (DDI) water and
001 M HCl solution to remove the ultra-fine material and eliminate highly reactive sites
caused during grinding The chromite powder was inspected before (Figure 31) and
after the experiments using a scanning electron microscope (JEOL 5600LV) the surface
area was 025 m2 g-1 as determined using N2(g) and BET isotherm
An automated JEOL 733A electron microprobe operated at 15 kV accelerating
potential and 15 nA beam current was used to investigate possible zoning and to quantify
the chemical composition of the chromite grains prior to powdering No zoning was
observed and the chemical composition of the chromite was based on the average of 25
electron microprobe analyses The stoichiometric composition of chromite based on 4
65
oxygens is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 X-ray diffraction was also used to
deduce the purity of the chromite powder X-ray diffraction spectra obtained for the
chromite powder matched JCPDS card 03-0873 (unit cell a=8300 b=8300 c=8300
α=90 β=90 and γ=90) and no contaminants were evident
Birnessite was synthesized using the procedure described by Buser et al (1954)
and implemented by Fendorf and Zasoski (1992) A solution of 63 g of potassium
permanganate (KMnO4) in 1 L of DDI water was heated to 90 degC Approximately 100
mL of the heated slurry was poured into a 4 L beaker and reheated to 90degC while being
stirred vigorously The remaining 900 mL of slurry and 66 mL of concentrated
hydrochloric acid were simultaneously and proportionally dispensed into the beaker over
a period of thirty seconds The temperature of the resulting solution was maintained at
90deg C for an additional 10 minutes and then allowed to cool to room temperature The
product was washed with alternate rinses of DDI water and a 001 M HCl solution The
birnessite surface area was 125 m2 g-1 based on BET analyses Solids were inspected
before (Figure 31) and after the experiments using the SEM
Extraction Experiment Procedures
Several chemical extraction experiments were performed on 106-250 microm
chromite grains prior to the chromite-birnessite batch experiments in order to determine
the potential release of Cr Fe and Mn One gram of chromite and 25 mL of solution
were placed into a 25 mL centrifuge tube and the solids and solution were gently agitated
The solutions and parameters used for these extraction experiments include the
following 1) 10 hydrogen peroxide (H2O2) solution at 25degC and pH of ~5 for 6 h 2)
01 M potassium permanganate (KMnO4) solution at 60degC and pH of ~3 for 6 h 3) 10 M
hydrofluoric (HF) acid at 60degC for 6 h 4) 1 M hydroxylamine (NH2OHmiddotHCl) with 25
(vv) acetic acid (HOAc) at 60degC and pH of ~4 for 6 h and 5) 1 mM barium chloride
(BaCl2) solution at 60degC and pH of ~6 for 6 h Solutions were extracted with a syringe
and passed through a 02 microm filter Solids were inspected after the experiments using the
SEM
66
Batch Experiment Procedures
Batch experiments were performed in triplicate using 1 L polypropylene bottles
and chromite and birnessite ratios listed in Table 31 at ~25degC After loading chromite
and birnessite into each bottle 1 L of a pH 3 solution a 10 mM acetate solution (pH 5) a
10 mM MES solution (pH 67) or a pH 8 solution was added to the appropriate
experiment (Table 31) Chromite and birnessite were also run individually in the acetate
buffered solution (Table 31) Aliquots were extracted from the polypropylene bottles
using a syringe and then passed through a 2 microm filter before analyses Less than 40 mL
was extracted over the course of each experiment resulting in a total volume change of
lt4 After the experiments were terminated the remaining solids were resuspended in
20 mL of 1 mM phosphate (5050 KH2PO4K2HPO4 pH 7) 20 mL of 10 mM CaCl2
solution at pH 7 and a 05 M HCl solution These suspensions were agitated for 3 h
followed by centrifugation and filtration (02 microm) before analyses The purpose of the
phosphate and CaCl2 solutions was to desorb reaction produtcts (eg Cr(VI) Cr(III) Fe
and Mn) from the solid phases The HCl treatment provides a means to evaluate
secondary phases such (oxy)hydroxides formed during the experiments Although the
HCl solution may dissolve chromite and birnessite in addition to the (oxy)hydroxides the
short time frame of the extraction (3 h) should limit the dissolution of chromite
Additionally the polypropylene bottles following the termination of five batch
experiments were rinsed with DDI water and 20 mL of 05 M HCl was added to each
bottle The HCl solution was allowed to remain in the bottles for 15 h in order to
determine the concentration of Cr Fe and Mn sorbed onto the interior lining
Serpentine Soil and Birnessite Experiments
Serpentine soils collected from Jasper Ridge Biological Preserve (Stanford CA)
and New Caledonia were used to determine the influence of birnessite in soils known to
contain chromite (Chapters 2 and 4) The Cr geochemistry of these soils as well as the
detrital chromite have been evaluated in Chapters 2 and 4 and by Becquer et al (2003)
Jasper Ridge serpentine soils contain Cr-bearing minerals including chromite Cr-
magnetite and Cr-chlorite however a majority of the Cr enrichment (Table 32) is
directly related to the abundance of chromite A majority of the Cr related to New
67
Caledonia serpentine soils is attributed to Cr (oxy)hydroxides (Chapter 4) however
chromite is present in these soils as well (Becquer et al 2003) Although birnessite has
been noted in these soils the quantity of birnessite could not be determined In order to
evaluate the impact of Mn oxides on Cr within such soils 01 g of birnessite was added to
each soil
The parameters for the experiments are listed in Table 32 Ten grams of the
lt250 microm size-fraction of each soil (JR1 JR2 and NC) and 100 mL of 1 mM acetate
buffered solution (pH 5) were loaded into individual 150 mL Nalgene bottles Aqueous
solutions were evaluated as a function of time (h) Additionally 10 g of each soil 01 g
of birnessite and 100 mL of 1 mM acetate buffered solution (pH 5) were loaded into 150
mL Nalgene bottles and were allowed to react as a function of time (h) These
experiments were rotated at a rate of 2 revolutions min-1 Solutions were removed with a
syringe and passed through a 2 microm filter before analysis Approximately 8 mL of
solution was required for the analyses over the course of each experiment resulting in a
total volume change of 8 Total Cr concentrations (mg kg-1) of the soils listed in Table
32 were obtained by completely dissolving the samples using a mixture of hot (gt200degC)
concentrated nitric perchloric and hydrofluoric acids Elemental compositions were
measured in the supernatant using inductively coupled plasma optical emission
spectroscopy (ICP-OES) The mass of chromite in each soil was calculated using total Cr
concentration for each soil and the calculations provided in Chapters 2 (Table 210) and
Chapter 4 (Table 47) The chromite surface area of each soil was approximated by
comparing the surface area measured for the chromite standard and its observed grain
size from SEM images (Figure 31) to chromites observed in the soils using backscattered
electron images (Chapters 2 and 4) The surface area of the Jasper Ridge chromite is 066
of the chromite standard due to a slightly larger grain size The surface area of the
New Caledonia chromite is approximately 025 of the chromite standard Chromite
suspension densities were calculated by multiplying the chromite mass by the estimated
surface area and then dividing by the volume of the solution
68
Solution Analysis
The pH and Eh of the supernatant was measured with an Orion flat surface
combination pH electrode inserted directly into supernatant Production of soluble
Cr(VI) was measured spectrophotometrically at 540 nm using the s-diphenyl carbazide
method modified from Bartlett and James (1979) Soluble Fe(II) was monitored using
the ferrozine method adopted from Stookey (1970) Total dissolved Cr Fe Al Mn and
Mg were ascertained using ICP-OES Concentrated trace metal grade nitric acid used to
acidify the samples prior to analyses Aluminum and Mg was determined to be present in
the nitric acid at concentrations lt1 microM Magnesium and Al are not plotted in any of the
figures Chromium(III) and Fe(III) was determined by subtracting Cr(VI) and Fe(II)
obtained from colorimetric methods from total dissolved Cr and Fe analyses using ICP-
AES Results are presented as an average of the three runs for each batch experiment
with the standard deviation noted by error bars
RESULTS
Chromite Extraction Experiments
The release of Cr Fe and Mn from chromite using 10 H2O2 01 M KMnO4 10
M HF acid 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 1 mM BaCl2 solutions was
carried out to determine the relative propensity of chromite to resist chemical dissolution
and these results are summarized in Figure 32 The KMnO4 solution with aqueous
Mn(VII) was the most effective in releasing Cr from chromite yielding a concentration
of 62 Cr mg L-1 Iron released using the KMnO4 solution was 03 mg L-1 Although
H2O2 is a direct chemical oxidant the release of Cr from chromite was less as compared
to the other solutions (only ~03 Cr mg L-1 was measured) Concentrated HF was only
able to release ~13 Cr mg L-1 while releasing 106 Fe mg L-1 NH2OHmiddotHCl with 25
(vv) HOAc and CaCl2 solutions only removed lt1 Cr mg L-1 however the NH2OHmiddotHCl
removed 36 Fe mg L-1 which is comparatively higher than the H2O2 KMnO4 and BaCl2
solutions
69
Chromite and Birnessite Batch Experiments
The interaction of chromite and birnessite leading to the release and oxidation of
Cr(III) is a relatively slow process requiring study on the order of hours and days The
pH was maintained constant throughout the duration of the experiments while Eh values
ranged from 320 to 450 mV with no discernable temporal pattern (Figure 33) For pH
values between 3 and 67 Cr3+ CrOH2+ and Cr(OH)2+ are the dominant aqueous Cr(III)
species however at pH 8 conditions are on the stability line between Cr(III) and Cr(VI)
specifically Cr(OH)3 and CrO42-
Higher chromite suspension densities produced higher rates of Cr release to the
aqueous phase and all the Cr in solution was in the form of Cr(VI) (Figure 35) The
rates of oxidation based on Cr(VI) production range from 00005 to 00041 microM h-1
(Figure 35 Table 33) Comparatively Cr released from the chromite control experiment
is maintained at a relatively constant value of ~005 microM and no Cr(VI) is measured in
solution Iron concentrations initially increase to ~ 05 microM and then decrease rapidly to a
relatively constant concentration at 02 microM (Figure 35) Iron concentrations for the
chromite control experiment began at 15 microM between 0 to 50 h and decreased to ~02
microM Iron(II) and Mn concentrations are below detectable limits for all experiments
(Figure 35) A zeroth-order dependence is assumed with respect to birnessite owing to
its large excess surface area relative to chromite
Variable birnessite suspension densities and constant chromite suspension
densities were evaluated at pH 5 (Figure 36) Although suspension densities of
birnessite are 538 and 1075 m2 L-1 (Table 31) the rate of Cr and Cr(VI) measured in
solution for both experiments is 00041 microM h-1 (Figure 36 and Table 33) Iron
concentrations are below 06 microM and maintain concentrations near 02 microM with no
indication of consistently increasing or decreasing with respect to time Iron(II) and Mn
concentrations in solution are below detectable limits for both experiments
In Figure 37 the influence of pH on chromite oxidation by birnessite was
evaluated by conducting experiments at pH values of 3 5 67 and 8 chromite and
birnessite suspension densities are constant at 0075 m2 L-1 and 1075 m2 L-1 respectively
(Table 31) Chromium and Cr(VI) rates increase with decreasing pH ranging from 0
microM h-1 at pH 8 to 00062 microM h-1 at pH 3 (Table 33) Iron is present in solution at
70
concentrations less than 04 microM for all the experiments Generally Fe concentrations are
greater near the start of the experiment and decrease to values lt 05 microM with increasing
reaction time Manganese concentrations are below detectable limits
In order to determine the rate dependence on each prominent reactant (chromite
birnessite and pH) reaction rates were regressed against reactant concentrations (Figure
38) A pseudo first-order rate dependence is noted based on chromite suspension
density the equation regressed for this relationship is listed in Table 33 As the chromite
suspension density increases the rate of Cr(VI) in solution increases In contrast a
zeroth-order rate dependence is evident for birnessite suspension density (dashed line in
Figure 38b) The rate of Cr(VI) generation is expected to decrease with respect to
birnessite suspension density at very low concentrations (dotted line in Figure 38b)
The influence of pH on the rate of Cr(VI) production is well-described by a polynomial
function (Figure 38c Table 33)
Developing an overall rate expression for Cr in solution requires integrating the
factors assessed in Figure 38 and listed in Table 33 All the Cr measured in solution is
in form of Cr(VI) therefore the rate law for Cr or Cr(VI) in solution may be written as
d[Cr(VI))]dt = k sdotChromitea sdot Birnessiteb sdot (aH+)c (31)
and
log d[Cr(VI))]dt = log k + asdotlogChromite+ bsdot log Birnessite+ csdotlog (aH+) (32)
where k is the overall rate constant for Cr release and oxidation Chromite is the
chromite suspension density (Table 31) Birnessite is the birnessite suspension density
(Table 31) and aH+ is the activity of H+ Constants a b and c are the order of
dependency for each reactant The overall rate equation (developed in the Appendix) can
be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069 (33)
71
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 the expression is valid for birnessite suspension densities
greater than 20 m2 L-1 and pH values between 3 to 8
Secondary and Solid Phase Examination of Chromite and Birnessite
The solid products were evaluated to determine the abundance and type of
secondary phases formed during the course of the experiments Solids from the batch
experiment were treated with three solutions 1 mM 5050 KH2PO4K2HPO4 1 mM
CaCl2 and 05 M HCl (Table 34) In the chromite and birnessite reactions Cr was
primarily in solution (gt90 ) The HCl extract was able to remove lt5 additional Cr
while the KH2PO4K2HPO4 and CaCl2 extracts released lt3 additional Cr Aqueous Fe
concentrations were below detectable limits for the final solution extraction Treatment
with the HCl resulted in up to 0127 micromoles of Fe however 0107 micromoles of Fe were
released upon treatment of unreacted chromite with HCl Although not shown in Table
34 Mn values were greater than gt4 micromoles when treated with HCl the
KH2PO4K2HPO4 and CaCl2 removed no detectable Mn
After the termination of each experiment the solids were examined using SEM in
order to identify changes in the surface morphology and secondary phases A large
chromite grain and conglomerate of birnessite adjacent to the chromite are depicted in
Figure 38 The face and edge of the chromite grain is not noticeably altered compared to
grains examined prior the experiments Additionally the particle morphology of
birnessite has not visibly changed Even at higher magnification angular and
conchoidally fractured fine-grained edges are noted (Figure 38bc) Only a slight
rounding of the smaller features of the main chromite grain provides any evidence of
possible chromite dissolution (Figure 38c) Birnessite appears to have limited interaction
with chromite Although chromite and birnessite were allowed to interact over 500 h in
an aqueous solution no secondary phases are observed by SEM
Serpentine Soil and Birnessite Experiments
Two serpentine soils from Jasper Ridge Biological Preserve CA and one
serpentine soil from New Caledonia were used to determine whether or not the overall
72
rate expression (Equation 33) for chromite reaction with birnessite is consistent with
Cr(VI) concentrations produced by serpentine soils containing chromite Three
serpentine soils without birnessite and three soils with the addition of birnessite buffered
at pH 5 were measured for Cr(VI) generation as a function time (Figure 310) The rates
of Cr(VI) measured in solution for both Jasper Ridge soils (JR1 and JR2) are identical at
00013 microM h-1 With the addition of birnessite the rates of Cr(VI) production in solution
both increased to 00044 microM h-1 (JR1 w Bt and JR2 wBt in Figure 310) Based on the
chromite suspension density (Table 32) the rates predicted by Equation 33 range from
00035 to 00041 microM h-1 which are 10 to 20 less than those observed in the
experiments (Figure 310) The rates of Cr(VI) in solution from the New Caledonia
serpentine soil are 00009 microM h-1 and with the addition of birnessite increase up to 00013
microM h-1 (Figure 310) A rate of 00016 microM h-1 was predicted by Equation 33 for the
New Caledonia soil
DISCUSSION
The reaction of birnessite with chromite enhances the rate of Cr release into
solution The rate of Cr(VI) measured in solution (Equation 33) is directly related to the
chromite suspension density and is independent of the birnessite at suspension densities
greater than ~20 m2 L-1 Although the release and oxidation of Cr in solution are
measured the mechanisms and pathways driving this reaction need to be evaluated The
propensity of chromite to resist dissolution the effects of pH the formation of secondary
phases and ultimately the pathways of Cr release and oxidation are addressed in the
following discussion
Extraction experiments performed on chromite (Figure 32) demonstrate it is
resistant to acidic dissolution or competitive ion displacement but strong oxidants such
as high valent Mn may oxidatively dissolve chromite It is evident that the proportions of
Cr Fe and Mn vary for each of the extractions The solutions utilized in the extraction
experiments have varying chemical properties capable of preferentially releasing a
particular species andor element These experiments were also performed over a short
period of time (lt6 h) which may have only released elements from highly reactive sites
73
Finally secondary phases associated with these experiments were possible and would
distort the elemental ratios observed in solution
A majority of the Eh and pH analyses plot in the Cr(III) stability region (Figure
33) however Cr in solution is primarily in the form of Cr(VI) (Figures 35 36 and
37) Eh measurements presented in Figures 32 and 33 do not account for the
microenvironments associated with the surfaces of birnessite Additionally the kinetics
of Cr(VI) reduction are limited in reducing conditions due to electron symmetry
constraints and the absence of a suitable redox couple (Fendorf 1995) Finally highly
oxidizing birnessite in the solution is kinetically and thermodynamically controlling the
oxidation state of Cr relative to the stabilities predicted in Figure 33 (Amacher and
Baker 1982 Fendorf and Zasoski 1992)
A fractional-order dependence on the chromite suspension density (Figure 38)
can be explained by diffusional influences competing oxidation reactions and secondary
reactions involving Cr(III) and Mn(IV) Secondary phases on the surface of chromite
may inhibit the diffusion of Cr from chromite however the clean and relatively unaltered
chromite surfaces do not support this mechanism as a major cause (Figure 38)
Chromium and Fe released from chromite may be incorporated into secondary phases
related to birnessite or chromite Approximately 10 of Cr is not in the aqueous phase
(Table 34) supporting the premise that secondary phases are contributing to the
fractional-order dependence related to the chromite suspension density Also
competition from Fe(II) oxidation on birnessite deposition of Fe(II) on birnessite (or
chromite) and diffusion onwithin birnessite may further impact the rate dependence
The zeroth-order dependence based on the birnessite suspension density (Figure
38) results from its high surface area At very low birnessite suspension densities the
rate of Cr(VI) production should decrease however due to the high surface area of
birnessite it was experimentally prohibitive to determine reaction rates at masses less
than 4 mg Therefore a hypothetical trend-line depicts the expected dependence on
birnessite concentration at very low suspension densities at higher values representative
of most field settings the reaction is independent of the birnessite surface area (ie zero-
order dependence)
74
In Figure 38c the polynomial relationship defines the Cr(VI) rate dependence on
pH As the pH decreases the rate of Cr(VI) increases more rapidly The relationship is
potentially the result of lower pHs progressively increasing the rate of chromite
dissolution However the rate of Cr(VI) in solution might be influenced by higher pHs
increasing the formation of secondary Fe(III) and Cr(III) (oxy)hydroxides on the
chromite andor birnessite surfaces thereby decreasing elemental diffusion from
chromite or decreasing available birnessite sites for Cr(III) oxidation Differentiating
between these two mechanisms is not possible based on the rate equation or extraction
analyses however the polynomial relationship evident in Figure 38c is folded into the
rate constant in order to geochemically account for the influence of pH on Cr(VI)
generation
No conclusive evidence is available to determine the fate of Mn within the
reacting suspensions Aqueous Mn concentrations are consistently maintained below
detectable limits If Mn is released from birnessite then it must be strongly and quickly
incorporated into a nonaqueous phase following its release Due to Cr(III) being oxidized
by Mn(III) andor Mn(IV) Mn(II) must be produced Extractions performed on the
chromite and birnessite solids demonstrate that Mn could only be removed with HCl
however a majority of this Mn is potentially the result of HCl dissolving birnessite On
the basis of electron transfer stoichiometry at least 15 mols of Mn(II) must be produced
for each mol of Cr(VI) generated This amount can be accounted for in the HCl
extraction where gt4 micromoles of Mn were measured Additionally Fendorf et al (1993)
determined that Mn(II) should not significantly influence the extent of Cr(III) oxidation
Iron in solution does not progressively increase in solution over time (Figures 35
36 and 37) The high initial rate of Fe release compared to Cr may be the result highly
reactive sites preferentially releasing Fe Although no Fe is detected in solution from
the extractions performed on the chromite and birnessite solids 0005 to 0035 moles of
Fe were retrieved from the lining of the bottle and 0053 to 0126 moles were released
from the solids using the HCl solution (Table 34) Iron(II) could potentially be oxidized
to Fe(II) in the presence of birnessite and form stable Fe(III)-(oxy)hydroxides (Postma
1985 Myers and Nealson 1988 Villinski et al 2001) A ratio of two Fe for every one
Mn is necessary for the oxidation of Fe(II) and this amount is recoverable based on HCl
75
extractions with gt4 micromoles of Mn measured from the HCl extractions Additionally
dissolved O2 in the batch experiments will rapidly oxidize Fe(II) for all the pHs examined
in these experiments (Stumm and Morgan 1981 Langmuir 1997) The oxidative
pathway for Fe(II) provides an explanation for the lack of Fe(II) in solution for the batch
experiments (Figure 35 36 and 37)
With a detection limit of ~02 microM and equilibrium with respect to Fe(OH)3(am) and
goethite Fe concentrations should either be at or below detection limits at pHs gt 5
Assuming aqueous Fe(III) is present dominantly as the second hydrolysis product at pH
5 the dissolved concentrations approach the solubility of ferrihydrite and are much
greater than expected for goethite (Figure 311) Thus it is probably that the former solid
is generated which is further supported by the chemical extraction experiments (Table
34)
Iron and Cr solution concentrations measured experimentally are not consistent
with congruent dissolution (Figure 312a) Assuming Fe was forming secondary phases
at a constant rate (~00001 Fe h-1) multiplying this rate by the time analysis and then
adding it to the measured aqueous Fe a corrected Fe (Fe) value can be estimated (Figure
312) Additionally the 10 of Cr not in solution can be also be used to calculate a
corrected Cr value (Cr) (Figure 312) When secondary phases are taken into account a
majority of the analyses plot near the congruent dissolution of chromite This provides
evidence that chromite is undergoing congruent dissolution within the batch experiments
but that secondary Fe(III) phases disrupt the solution phase elemental ratios
Chromium(VI) Pathways from Chromite and Birnessite into Solution
Three pathways are evaluated for Cr(VI) measured in solution a) chromite
dissolution b) birnessite dissolution and c) solid-solution reaction (Figure 313) In order
to describe these pathways end-member chromite (FeCr2O4) and birnessite (δ-MnO2) are
used to simplify the analysis It should be recalled however that the composition of the
chromite used in the experiments is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4
Chromite dissolution presents a likely explanation for Cr(VI) observed in solution
based on the following discussion In chromite dissolution model Cr(III) released from
chromite adsorbs onto the surface of birnessite and oxidizes to Cr(VI) After the electron
76
transfer is complete Cr(VI) desorbs from birnessite and remains in solution As this
process continues Cr(VI) increases at a constant rate The rate limiting step with regards
to this specific pathway can be grouped into two steps a) dissolution of chromite (Step
1a) or b) Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) (Figure 313)
Although a dissolution rate for chromite at ambient temperatures and pressures is not
known the chromite extraction experiments (Figure 32) demonstrates that chromite
dissolution is slow Manganese oxides such as birnessite oxidize Cr(III) in solution
rapidly with a measured rate constant of 6300 microM-1 s-1 in fresh water (100 mg L-1 MnO2
100 g L-1 Cr(III) pH~6) (Van der Weijden and Reith 1982) This means that the rate
for Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) is on the order of
seconds therefore chromite dissolution is likely the rate limiting step controlling the
concentration of Cr(VI) in solution for this pathway
Although a zeroth-order dependence related to birnessite concentrations (Figure
37a) is observed the birnessite dissolution pathway in which aqueous Mn4+ released
from birnessite adsorbs onto the surface of chromite oxidizing Cr(III) cannot be entirely
dismissed Manganese dissolution and adsorption could occur quickly not allowing the
Mn to be measured in solution Examination of the solids after the termination of each
experiment demonstrated extremely minimal chromite and birnessite interaction
therefore the solid solution reaction is not a likely pathway leading to the production of
Cr(VI) in solution and is furthermore inconsistent with the noted rate dependencies
Despite the three pathways (Figure 313) the generation of Cr(VI) observed in these
experiments and described by the rate equation (Equation 33) is independent of these
potential pathways
Serpentine Soils and Cr(VI)
The serpentine soil experiments (Figure 310) provide direct evidence that Cr(VI)
is forming within these soils By adding birnessite to the soil the rate of Cr(VI) in
solution increased for all experiments (00013 to 00044 microM Cr(VI) h-1 for Jasper Ridge
and 00009 to 00013 microM Cr(VI) h-1 for New Caledonia) The Cr(VI) rate increase
related to the addition of birnessite is likely the result of low birnessite suspension
densities (lt20 m2 L-1) in the soil Rates predicted for Jasper Ridge and New Caledonia
77
soils using Equation 33 and values in Table 32 are similar to those measured in the
experiments with values of 00041 and 00016 microM Cr(VI) h-1 respectively Overall these
soil experiments demonstrate that a rate equation developed from laboratory batch
experiments predicts the rate of Cr(VI) in serpentine soils reasonably well
According to limits set by the Environmental Protection Agency the California
standard for total Cr in drinking water is 50 microg L-1 (~1 microM) while the federal standard is
100 microg L-1 (~2 microM) The serpentine soil and birnessite experiments produced Cr(VI)
concentrations in excess of 50 microg L-1 after only 140 h or roughly 6 days Serpentine soils
at Jasper Ridge typically remain water saturated for prolonged periods of time (gt1 week)
during the winter season increasing the time Cr(VI) can be produced The degree and
duration of water saturation is dependent on the topography of the serpentinite body
level areas will remain water saturated while sloping areas will allow the water to flow
down gradient Although Cr(VI) was generated in the soil experiments the production of
Cr(VI) in serpentine soils may be more limited The reduction of Cr(VI) to Cr(III) is
favorable by reactions with Fe(II)aq Fe(II)-containing minerals such as magnetite H2S
and organic matter including microbial processes (Ilton and Veblen 1994 Fendorf 1995
Peterson 1996) Both organic matter and magnetite are abundant in Jasper Ridge
serpentine soils (Chapters 2 and 4) however these factors did not deter Cr(VI) from
forming during the serpentine soil experiments Additionally the soils in these
experiments were constantly rotated allowing the mineral surfaces to be continuously
exposed to solution Chromite within an undisturbed soil will not always be in contact
with the solution or proximal to birnessite
Chromium(VI) produced by the interaction of chromite and birnessite is not
limited to serpentine soils Serpentinites the bedrock for serpentine soils are typically
highly sheared due to the diapiric migration of serpentinite masses along zones of
weakness in fault zones Numerous cracks and lineations which are typical of these rocks
provide a means for Cr(VI) to transported away from the serpentine soil potentially into
local groundwater In fact preferential flow will occur along the major shear plains of
the serpentinite Additionally these shear plains are often coated with high valent Mn
oxides providing a reaction pathway for Cr(VI) formation Serpentinite springs such as
those evaluated by Barnes and ONeil (1969) and Barnes et al (1972) at the Cedars
78
ultramafic body in Cazadero CA have Cr concentrations measured at 20 microg L-1 in which
all the Cr is in the form of Cr(VI) (Chapter 4) In conclusion chromite and birnessite in
serpentine soils could contribute a source of Cr(VI) in surface waters and groundwaters
Chromium release and oxidation from chromite is increased by the presence of
birnessite and the interaction between chromite and birnessite is a pathway for Cr(VI)
generation in ultramafic sediments and soils The potential of serpentine soils as a source
of non-anthropogenic Cr contamination can now be evaluated quantitatively based on the
abundance of chromite The overall rate law for Cr(VI) in solution (Equation 39)
provides a means to estimate the Cr(VI) concentration attainable for serpentine soils and
serpentine sediments when the chromite and birnessite suspension densities and the pH of
the soil are known
ACKNOWLEDGEMENTS
We gratefully acknowledge comments and assistance from Robert Jones and Dr
Guangchao Li Funding for this research was provided by the Office of Teaching and
Learning at Stanford University
79
Table 31 Chromite and birnessite batch experiment parameters with chromite and birnessite suspension densities
Experiment Chromitea (grams)
Birnessiteb (grams) pHc
Chromite Surface Aread
(m2 g-1)
Birnessite Surface Aread
(m2 g-1)
Chromite Suspension
Density (m2 L-1)
Birnessite Suspension
Density (m2 L-1)
A 08 043 5 025 125 02 54 B
08 086 5 025 125 02 108C 03 086 5 025 125 008 108D 008 086 5 025 125 002 108E 004 086 5 025 125 001 108F 03 086 3 025 125 008 108G 03 086 67 025 125 008 108H 03 086 8 025 125 008 108
Chromite 08 ---- 5 025 125 02 ----Birnessite ---- 043 5 025 125 ---- 5480
a Chromite chemical composition is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 b Synthetic birnessite (δ-MnO2) c10 mM acetate buffered solution pH = 5 non-buffered pH=3 5 mM MES buffer pH 67 non-buffered pH 8 dMeasured BET analyses
81
Table 32 Serpentine soil and birnessite experiments with soil and birnessite masses used and the estimated chromite suspension density
Samples Soil Mass (g)
Birnessite Mass (g)
Volume (L) of Solutiona
Total Crsoil
b (mg kg-1)
Chromite Mass (g) in
Soil Samplec Estimated Surface
Areac (m2 g-1)
Chromite Suspension Density
(m2 L-1) in Soild Jasper Ridge
(JR1) 10 ---- 01 5976 013 015 02 Jasper Ridge (JR1 w Bt) 10 01 01 5976 013 015 02 Jasper Ridge
(JR2) 10 ---- 01 4978 011 015 016 Jasper Ridge (JR2 w Bt) 10 01 01 4978 011 015 016
New Caledonia (NC) 10 ---- 01 827 0015 006 001
New Caledonia (NC w Bt) 10 01 01 827 0015 006 001
a1mM Acetate Buffer (pH=5) 81 bTotal Cr concentrations were obtained through total digestion and ICP-AES analyses
cEstimated chromite mass and surface area (discussed in the Methods section) was made using chromite observations and calculations in Chapters 2 and 4
dCalculated from the chromite mass in the soil sample estimated surface area of the chromite and the volume of solution
82
Table 33 Summary of Cr(VI) rates obtained from batch experiments (Figure 35 36 and 37) and used in Figure 38
Listed in Table 31
Suspension Densities (m2 L-1)
Chromite (Ch)
Birnessite (Bt)
pH Rate (microM h-1) (d[Cr(VI)]dt)
log Rate (log microM h-1) (log d[Cr(VI)]dt) Expression
Figure 35 02 108 5 00041 -239
008 108 5 00021 -268 002 108 5 00009 -305 001 108 5 00005 -330
1
Figure 36 02 54 5 00041 -239 02 108 5 00041 -239 2
Figure 37 008 108 3 00062 -221 008 108 5 00021 -268 008 108 67 00004 -340 008 108 8 0 ----
3
Expression 1 log d[Cr(VI)]dt = asdot log Chromite +log krsquorsquo log d[Cr(VI)]dt = 06908 log Chromite ndash 1899 Expression 2 log d[Cr(VI)]dt = bsdot log Birnessite + log krsquorsquorsquo
log d[Cr(VI)]dt = 0sdot log Birnessite ndash 23872 log d[Cr(VI)]dt = -23872
Expression 3 d[Cr(VI)]dt = 264E-4sdotpH2 ndash 414E-3sdotpH + 162E-2
82
Table 34 Extractions comparing total solution and extractable Cr and Fe from the solids
------------------Cr----------------- ------------------Fe---------------- microMolesdagger Total
mole Extraction
mole microMoles Total mole
Extraction mole
Chromite Ch 02 (Table 31) Terminated at 30 days Solutiona 0033 554 ---- 0 0 ---- Phosphateb 0000 05 1 0000 03 03 CaCl2
c 0013 388 87 0000 00 00 HCld 0002 53 12 0103 997 997 Bottle HCle 0000 00 0 0000 00 00
A Ch 02 Bt 54 (Table 31) Terminated at 30 days Solution 2977 910 ---- 0 0 ---- Phosphate 0048 16 18 0004 21 21 CaCl2 0055 18 20 0002 09 09 HCl 0161 54 60 0126 758 758 Bottle HCl 0004 01 2 0035 212 212
B Ch 02 Bt 108 (Table 31) Terminated at 30 days Solution 2837 914 ---- 0 0 ---- Phosphate 0039 14 16 0008 49 49 CaCl2 0060 21 25 0000 00 00 HCl 0142 50 58 0127 803 803 Bottle HCl 0004 01 2 0024 148 148
C Ch 08 Bt 108 (Table 31) Terminated at 17 days Solution 0877 894 ---- 0 0 ---- Phosphate 0020 23 22 0002 37 37 CaCl2 0030 34 32 0000 00 00 HCl 0040 46 43 0053 881 881 Bottle HCl 0003 03 3 0005 82 82
daggerMolarity values are normalized to the final solution analysis of each experiment based on volume of solution and mass of the sample in each experiment aSolution refers to the final solution analysis before the experiment was terminated bmicroMoles extracted from the batch experiment solids using a 1 mM phosphate buffer pH 7 3 h cmicroMoles extracted from the batch experiment solids using a 10 mM CaCl2 solution pH 7 3 h dmicroMoles extracted from the batch experiment solids using a 05 N HCl solution 3 h emicroMoles extracted from the batch experiment bottles using a 05 N HCl solution 15 h
83
Figure Captions
Figure 31 Scanning electron microscopic (SEM) images of chromite and birnessite at
varying magnifications prior to the batch experiments
Figure 32 A summary of Cr Fe and Mn released from chromite during extraction
experiments Solutions used for these experiments include 1) 10 H2O2 solution 2) 01
M KMnO4 solution 3) 10 M HF 4) 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 5) 1mM
BaCl2 solution () Fe concentration for the HF extraction is 106 mg L-1
Figure 33 pH and Eh measurements as a function of time (h) for the batch
experiments Chromite (Ch) and birnessite (Bt) suspension densities used for the batch
experiments (Table 31) are listed adjacent their corresponding symbol in the key
Figure 34 Eh and pH batch experiment analyses (Figure 33) plotted on a stability field
diagram for Cr (log a Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom
(1998)
Figure 35 Batch experiments evaluating variable chromite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr and Cr(VI) measured in solution are noted and illustrated with a dotted line and
included in Table 33
Figure 36 Batch experiments evaluating variable birnessite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr(VI) measured in solution are denoted and included in Table 33
Figure 37 Influence of pH on Cr(VI) generation by reaction of constant chromite (Ch
08 m2 L-1) and birnessite (Bt 108 m2 L-1) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key Rates of Cr(VI) measured in solution are included in Table
33
84
Figure 38 Rates of Cr(VI) in solution (Table 33) as a function of a) chromite
suspension density (Figure 35) b) birnessite suspension density (Figure 36) and c) pH
(Figure 37) The line in each diagram represents the best fit for these data points and the
equations for these lines are listed in Table 33 The dotted line deviating from the
dashed line in b) indicates the probable path at low birnessite concentrations
Figure 39 SEM images of the solid phases after the completion of the batch
experiments A (Ch 02 Bt 54) and B (Ch 02 Bt 108) are shown Chromite is the larger
grain with relatively flat surfaces whereas birnessite is the fine-grained rosettes located
adjacent the larger chromite grains
Figure 310 Chromium and Cr(VI) concentrations (microM) in solution as a function of
time (h) from serpentine soils obtained at Jasper Ridge (JR1 and JR2) and New Caledonia
(NC) Soils with the addition of 01 g birnessite are noted in the figure Soil and
birnessite masses used for each experiment are listed in Table 32
Figure 311 Iron analyses as a function of time (Figure 35 and 36) with the solubility
of amorphous Fe(OH)3 and goethite and the dominant Fe3+-OH complex at pH 5 Fe
solubilities are from Langmuir (1997) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key
Figure 312 a) Experimental values of Fe and Cr (Figures 35 and 36) are plotted A
dotted line denotes the ratio of Fe and Cr necessary for chromite congruent dissolution b)
Corrected FeCr ratios as a function of time (h) for data from Figures 35 and 36 taking
into account the concentration of Fe and Cr in secondary phases calculated from Table
34 c) Corrected Fe and Cr as compared to values predicted with congruent dissolution
of chromite
85
Figure 313 Three potential pathways of Cr release and oxidation for chromite-
birnessite interactions are listed
86
Figure 31
87
0
2
4
6
8
10
12
Hydrogen
Peroxid
e
K-Perman
ganate
Hydroflu
oric Acid
Hydroxy
lamine
Barium Chlorid
e
Con
cent
ratio
ns (m
g kg
-1)
Cr
Fe
Mn
Figure 32
88
Time (h)0 100 200 300 400 500
pH
2
3
4
5
6
7
8
9
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
Time (h)0 100 200 300 400 500
Eh (m
V)
100
200
300
400
500
600
Ch Bt ------(m2 L-1)------
Figure 33
89
pH2 3 4 5 6 7 8 9
Eh (m
V)
-05
00
05
10
15
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
HCrO4-
CrO42-
Cr3+
CrOH2+
Cr(OH)3Cr(
OH
) 2+
Upper limit of water stability
Lower limit of water stability
Ch Bt ------(m2 L-1)-----
Figure 34
90
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 108 008 108 002 108 001 108 02 --- --- 54
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
0
1
2
3
4
pH 5 pH 5
pH 5 pH 5 pH 5
Ch Bt ---(m2 L-1)---
Figure 35
91
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 54 02 108
00041 microM h
-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
00
02
04
06
08
10
00041 microM h
-1
pH 5
pH 5 pH 5 pH 5
pH 5 Ch Bt----(m2 L-1)----
Figure 36
92
Time (h)
0 50 100 150 200 250
Cr c
once
ntra
tion
( microM
)
00
02
04
06
08
10
pH 3pH 5pH 67pH 8
Time (h)0 50 100 150 200 250
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250M
n co
ncen
trat
ion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
00004 microM h-1
00021 microM h-1
0006
2 microM h
-1
00021 microM h-100
062 micro
M h-1
00004 microM h-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1
Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Figure 37
93
log Birnessite m2 L-1
05 10 15 20 25 30
log
d[C
r(VI
)]dt
microM
h-1
-40
-35
-30
-25
-20
pH 5
log Chromite m2 L-1-22 -20 -18 -16 -14 -12 -10 -08 -06
log
d[C
r(VI
)]dt
microM
h-1
-34
-32
-30
-28
-26
-24
-22
pH 5
pH1 2 3 4 5 6 7 8 9
d[C
r(VI
)]dt
microM
h-1
0000
0002
0004
0006
0008
0010
a)
c)
b)
Ch 008 m2 L-1 Bt 108 m2 L-1
Figure 38
94
Figure 39
95
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR1JR1 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR2JR2 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
01
02
03
04
05
NCNC w Bt
00044 microM h-1
00013 microM h-1
00044 microM h-1
00013 microM h-1
00013 microM h-1
00009microM h-1
pH 5
pH 5
pH 5
Figure 310
96
Time (h)0 100 200 300 400 500
log ΣF
e (aq)
microM
-10
-8
-6
-4
-2
0
2
4
02 54 02 108 008 108 002 108 001 108
Fe(OH)3(am)
Goethite
pH 5Ch Bt----m2 L-1----
Fe(OH)2(aq)+
Figure 311
97
Time (h)
0 100 200 300 400
FeC
r Rat
io
0
1
2
3
4
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
02 54 02 108 008 108 002 108 001 108
Chrom
ite C
ongr
uent
Dissolu
tion
Chrom
ite C
ongr
uent
Dissolu
tion
a)
b)
c)
Ch Bt----(m2 L-1)----
ChromiteCongruent Dissolution
Figure 312
98
Pathway 1 Chromite Dissolution
1a) Cr dissolution from chromite FeCr2O3(s) + 4H+ rarr 2Cr(OH)2+
aq + Fe2+(Oxyhydroxide) + H2O
1b) Adsorption of CrOH2+ on the surface of birnessite Cr(OH)2+
aq + MnO2(s) rarr Cr(III)-MnO2(s) + OH- 1c) Oxidation of CrOH2+ on the surface of birnessite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
1d) Desorption of Cr(VI) from birnessite Cr(VI)-Mn(II) + 4H2O rarr Mn2+
(aq) + HCrO4-(aq) + 7H+
Pathway 2 Birnessite Dissolution 2a) Mn4+ dissolution from Birnessite MnO2(s) + 4H+ rarr Mn4+
(aq) + 2H2O 2b) Mn4+ adsorption on the surface of chromite Mn4+
(aq) + FeCr2O4(s) rarr Mn(IV)-FeCr2O4 2c) Oxidation of Cr(III) on the surface of chromite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
2d) Desorption of CrO24- from chromite
Cr(VI)-Mn(II) + 4H2O rarr Mn2+(aq) + HCrO4
-(aq) + 7H+
Pathway 3 Solid Solution Reaction 3) Solid solution interaction between chromite and birnessite
Figure 313
99
CHAPTER 4
CHROMIUM GEOCHEMISTRY OF SERPENTINE SOILS
ABSTRACT
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
chemistry of Cr in serpentine soils including its protolith is reviewed focusing on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils are slightly acidic (average pH of ~6) contain a variety Fe(III) oxides
(magnetite and hematite) phyllosilicates (serpentine and chlorite) and clays (smectites
and vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1)
and Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Chromium(III) is the only valence state observed
in the serpentine soil solids however Cr(VI) has been identified in New Caledonia and
California serpentine soil solutions at concentrations below 30 microM The enrichment and
range of Cr concentrations in serpentine soils are directly related to the presence of Cr-
spinels specifically chromite and Cr-magnetite These spinels are resistant to weathering
and are preserved in the soil environment however oxidation of Cr(III) from Cr-spinels
by high valent Mn oxides or other strong oxidants is a potential source of Cr(VI)
identified in serpentine soil solutions Due to the chemically and physically resistant
nature of the Cr-spinels Cr-bearing silicates including clay minerals Cr-chlorite Cr-
garnet Cr-mica and Cr-epidote are more viable sources of Cr identified in vegetation
soil extractions soil solutions and related waters
100
INTRODUCTION
Serpentine soils a generic term used to describe any soil derived from an
ultramafic rock (peridotites and pyroxenites) or serpentinite (metamorphosed ultramafic
rock) are often characterized by 1) low concentrations of plant nutrients such as N P
and K 2) high concentrations of biologically toxic elements including Ni Co and Cd 3)
low ammonium acetate extractable CaMg ratios and 4) unique flora and physical
properties (Brooks 1987) Measured Cr concentrations in serpentine soils range from
~200 mg kg-1 to 6 Wt values enriched compared to non-serpentine soils lt200 mg kg-
1 (Kabata-Pendias and Pendias 1984 Schwertmann and Latham 1986 Brooks 1987)
Chromite and Cr-magnetite are common primary sources of Cr in these soils (Rabenhorst
et al 1982 Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989
Gasser and Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003)
however the plethora of factors relating the chemistry of Cr in serpentine soils and the
weathering of chromite has not been fully addressed using an integrative approach
Overall serpentinites and serpentine soils cover ~1 of the earthrsquos exposed
surface and these rocks and soils are commonly associated with ophiolite complexes
(Coleman 1977) Serpentinites and serpentine soils related to ophiolite complexes are
the primary interest of this paper A map illustrating the worldwide distribution of major
ophiolitic serpentinite bodies and their related soils is shown in Figure 41 A majority of
the sites are near convergent boundaries (Figure 41) Although these rocks and soils
cover a minimal area of the earthrsquos surface they are present in populated areas within the
Circum-Pacific Margin and the Mediterranean Due to the wide-distribution and
occurrence of serpentinites and serpentine soils weathering of these rocks and soils will
differ from location to location due to varying climatic conditions as well as the nature of
the parent material and other factors including topography biota and time (Proctor and
Nagy 1992) As a result the chemistry of these soils as well as the nature of Cr in the
serpentinites and serpentine soils will vary from site to site potentially allowing Cr to be a
source of non-anthropogenic contamination
The objective of this paper is to review the chemical characteristics of serpentine
soils globally evaluate mineralogical and compositional variations between several soils
derived from ultramafic sources and assess the geochemistry of Cr in diverse soil-
101
forming environments Additionally chemical and spectroscopic analyses were
performed on soils derived from ultramafic rocks collected from 1) New Caledonia
South Pacific 2) Eight Dollar Mountain Selma OR 3) Nickel Mountain Riddle OR 4)
Harvard Mine Jamestown CA 5) Jasper Ridge Biological Preserve Stanford CA and 6)
Pillikin Mine El Dorado CA shown geographically in Figure 41 A basalt-derived soil
from Hawaii is included in this study in order to compare a soil formed from a mafic
protolith to those from ultramafic protoliths The various field sites are shown in Figure
42 These soils allow a comparative examination of how soil-forming factors such as
climate topography biota parent material and time affect the chemistry of Cr
REVIEW
Chromium Geochemistry
Chromium is a group VIB transition metal with two common oxidation states
Cr(III) and Cr(VI) differentiated by noticeably different chemical characteristics and
toxicities It should be noted that Cr(II) has been observed in lunar basalt olivines and in
high temperature terrestrial olivines (Sutton et al 1993 Li et al 1995) however this
oxidation state is not observed in soil environments A majority of Cr commonly persists
in ultramafic rocks serpentine soils and related waters as Cr(III) a non-hazardous
species and a micronutrient in human nutrition (Daugherty 1992 Cohen et al 1993
Fendorf 1995 James et al 1997) Chromium(III) has the highest crystal field
stabilization energy for all transition metals strongly favors octahedral coordination and
has 3 d-electrons with a high spin state Chromium(III) is strongly hydrolyzed in
aqueous solutions and the predominant species based on pH redox potential and O2(aq)
are shown in Figure 43 Chromium(III) species are more prevalent under reducing
conditions and its mobility is increased only under strongly acidic conditions At pH
values greater than 9 the hydrolyzed Cr(III) species is negatively charged demonstrating
the amphoteric properties of Cr(III) (Figure 43) Due to the factors described above
Cr(III) forms low solubility compounds and adsorbs strongly on mineral surfaces and
organic matter at pHgt4 (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) For
example the adsorption of Cr(III) on HFO (hydrous ferric oxide) as a function of pH is
shown in Figure 43 illustrating an adsorption edge at a pH of ~4 (Leckie et al 1984
102
Stumm 1992) Chromium(III) oxidation to Cr(VI) has been observed in reactions
involving hydrogen peroxide (H2O2) and high valent Mn oxides (Eary and Rai 1987
Johnson and Xyla 1991 Fendorf et al 1992 Fendorf et al 1993 Rock et al 2001)
Chromium(VI) is a strong oxidizing agent a highly corrosive substance a toxin to
living cells and a Class A human carcinogen by inhalation (Daugherty 1992 Cohen et
al 1993 Fendorf 1995) The hydrolyzed Cr(VI) species are also shown in Figure 43
with respect to pH redox potential and O2(aq) Chromium(VI) is more favorable under
oxidizing conditions and its mobility is enhanced under basic conditions In solution
Cr(VI) is a weakly adsorbing (oxy)anion (HxCrO42-x) on common weathering products
meaning that it forms highly labile complexes that can be readily displaced by ions
including phosphate and sulfate (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Additionally the adsorption of CrO42- on HFO shown in Figure 43
demonstrates the anionic character in which the mol bound of CrO42- increases with
decreasing pH and an adsorption edge is at pH ~72 (Leckie et al 1980 Stumm 1992)
The reduction of Cr(VI) to Cr(III) is known to occur by reactions with Fe(II)aq Fe(II)-
containing minerals such as magnetite H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996)
Mineralogy of Serpentine Soils
Serpentine soils have been identified on every continent excluding Antarctica
(Figure 41) The occurrence and distribution of ultramafic rocks specifically related to
ophiolites are closely associated with convergent plate boundaries (Figure 41 Coleman
1977 Coleman and Jove 1992) Tectonic blocks of ultramafic rock are hypothesized to
be detached from the down-going slab of oceanic crust and incorporated into the
subduction meacutelange Subsequent serpentinization of peridotite forms low density
serpentine group minerals (245-250 g cm-3) resulting in the upward diapiric migration of
the serpentinite masses along zones of structural weakness (fault zones) and through the
denser enclosing subduction meacutelange (~265 g cm-3) For example North America has
numerous exposed ophiolites containing serpentinite located in western California and
Oregon where subduction zones were once active in the Early Cretaceous (Harden
1998)
103
Serpentinization of ultramafic rocks is a significant factor controlling mineral
assemblages and textures in serpentinite Hydration and oxidation of olivine and
pyroxene during serpentinization forms serpentine (lizardite chrysotile and antigorite)
talc chlorite tremolite brucite and magnetite (Coleman 1977 Brooks 1987 Coleman
and Jove 1992 Malpas 1992) Additionally magnetite formation creates favorable
conditions for native metal compounds (such as Fe Co and Ni alloys) to form (Coleman
and Jove 1992) As a serpentinite mass is emplaced the extensive physical and chemical
alteration of the serpentinite results in numerous crosscutting alteration veins and highly
sheared textures Ultimately the mineral assemblages and textures in serpentinites are
complex prior to pedogenesis due to the multiple phases of metamorphosis tectonic
deformation and weathering of the rock
Serpentine soils consist of a mixture of residual minerals from the soil protolith
eolian and illuvial minerals and minerals formed within the soil Additionally highly
weathered serpentine soils enriched in Fe-oxides are also termed laterites soils common
in tropical and subtropical regions Soil protolith minerals may include primary minerals
of ultramafic rocks including olivines pyroxenes and oxides (chromite and magnetite)
as well as a large variety of metamorphic minerals the most common being serpentine
(lizardite chrysotile and antigorite) chlorite talc and actinolite Silica phases occur as
eolian grains concretions andor net type structures or silica webs (Hotz 1964
Coleman 1977 Schwertmann and Latham 1986) Eolian and illuvial aluminosilicate
minerals are important sources of Al in serpentine soils an element of low abundance in
ultramafic rocks
Layer silicates observed in serpentine soils are varied in type and abundance
They include smectites (often as montmorillonite) vermiculite chlorite brucite
serpentine and a variety of intrastratified clays (Veniale and Van der Marel 1963 Hotz
1964 Schellmann 1964 Wildman et al 1968 Rabenhorst et al 1982 Schreier et al
1987 Graham et al 1990 Bulmer et al 1992 Gasser and Dahlgren 1994 Lee et al
2003) Compositional variations crystal structures and parageneses of layer silicates
have been characterized for a number of serpentine soils (Wildman et al 1968
Rabenhorst et al 1982 Schreier et al 1987 Graham et al 1990 Bulmer et al 1992
Lee et al 2003) In addition to the layer silicates goethite hematite maghemite and
104
amorphous Fe (oxy)hydroxides are soil-forming phases in serpentine soils (Coleman and
Jove 1992) Gibbsite is not a common phase in most serpentine soils however limited
isomorphic substitution of Al(III) in goethite and hematite was noted in New Caledonia
(Schwertmann and Latham 1986) Manganese oxides such as birnessite (δ-MnO2) are
also present in serpentine soils and often coat mineral surfaces
Sources of Chromium for Serpentine Soils
Chromium primarily occurs in oxide and silicate phases in ultramafic rocks and
serpentinites Chromium-spinel often referred to as chromite is a common primary
(igneous) spinel in mafic and ultramafic rocks and often resists low-grade metamorphic
processes related to serpentinization (Hoffman and Walker 1978 Malpas 1992)
Isomorphic substitution of Al3+ Fe3+ and Ti4+ into the octahedral sites and Mg2+ Ni2+
Zn2+ and Mn2+ into the tetrahedral sites is common however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987b Pan and Fleet 1991 Sack and Ghiorso 1991 Burkhard 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996 Mathiesen 1999) Serpentinite chromite compositions reported in the literature are
plotted in Figure 44 to demonstrate the wide-range of Cr concentrations possible for Cr-
spinels Isomorphic substitution of octahedral Cr and octahedral Al for these chromite
compositions are also shown in Figure 44 Serpentinite pyroxenes containing Cr include
augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) (Grapes 1981 Von Knorring et al
1986 Treloar 1987a Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996)
Magnetite and serpentine minerals may incorporate varying amounts of Cr into
their structures during serpentinization Chromium(III) from primary Cr-containing
phases may isomorphically substitute into the Fe(III) site of magnetite due to size
(octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge similarities Chromium
concentrations in magnetite have been observed as high as 13 Wt (Oze et al 2003)
Chromate minerals such as crocoite (PbCrO4) are extremely rare in serpentinites and the
occurrence of these minerals are mainly limited to specific localities such as in Australia
Tasmania and the Urals (Williams 1974 Crane et al 2001) Serpentine minerals have
105
been reported to contain as much as 1500 mg kg-1 Cr based on spectrographic analyses
by Page and Coleman (1967) however serpentine minerals typically contain only trace
amounts of Cr
Hydrothermal and CO2 metasomatism of ultramafic rocks during the evolution of
convergent plate margins commonly produce Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite lt19 Cr Wt ) uvarovite (Cr-garnet lt20 Cr Wt )
tawmawite (Cr-epidote lt13 Cr Wt ) and kaumlmmererite (Cr-chlorite lt10 Cr Wt ) as
shown in Figure 44 (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974 Neiva
1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981 Max et al
1983 Nutman et al 1983 Raase et al 1983 Von Knorring et al 1986 Kerrich et al
1987 Treloar 1987b Treloar 1987a Morand 1990 Pan and Fleet 1991 Schandl and
Wicks 1993 Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995
Mathiesen 1999 Proenza et al 1999) Figure 44 illustrates the relationship of Cr and
Al substitution for Cr-silicate compositions based on Cr and Al Wt Additionally Cr-
silicate compositions are plotted based on octahedral Cr and octahedral Al in Figure 44
Compositions for Cr-muscovite Cr-epidote and Cr-garnet plot near the tie lines
connecting Cr and Al end-member minerals Chromium isomorphically substitutes for
octahedral Al in these silicates due to similarities in charge octahedral preference and
octahedral radii (Al3+ = 061 Aring) The numerous isomorphic substitutions possible for Cr
and Al in the chlorite structure results in analyses that do not plot near the tie line
connecting clinochlore and kaumlmmererite Overall any of these minerals are potential
sources for Cr in soils produced by the weathering of serpentinites and their hydrothermal
and CO2 metasomatic derivatives
Chromium in Serpentine Soils
Chromium concentrations vary widely in serpentine soils worldwide (Table 41)
New Caledonia soils appear to be the most Cr-enriched serpentine soil with a Cr
concentration of ~60000 mg kg-1 (Schwertmann and Latham 1986) however Jaffre
(1980) report values ranging from 6300 to 80000 mg kg-1 for New Caledonia soils
Some serpentine soils have measured Cr concentrations as low as 29 and 66 mg kg-1
atypical of most values reported in Table 41 (Proctor and Woodell 1971 Brooks et al
106
1992 Jeffrey 1992) Chromium concentrations as a function of depth for serpentine soil
profiles in Maryland (Rabenhorst et al 1982) Tehama County CA (Gough et al 1989)
the Klamath Mountains CA (Hotz 1964) Newfoundland Canada (Roberts 1992)
Harvard Mine CA (this study) and Jasper Ridge CA (Oze et al 2003) are reported in
Figure 45 The pedogenic process(es) responsible for the varying Cr concentrations as a
function of depth for each soil have not been characterized however Oze et al (2003)
determined that a majority of Cr in Jasper Ridge serpentine soils is the result of
weathering-resistant residual Cr-bearing minerals from the protolith mainly chromite
and Cr-magnetite Additionally Oze et al (2003) demonstrated that soil samples from
Jasper Ridge contain higher concentrations of Cr with respect to their corresponding
bedrock due to the Cr-spinels accumulating over time The observation that Cr
concentrations decrease with depth towards the protolith is characteristic for most of the
profiles in Figure 45
In addition to examining total Cr concentrations chemical extractions performed
on serpentine soils provide additional information with regards to Cr availability and the
minerals and complexes in which Cr may reside in the soils Table 42 summarizes a
number of extraction experiments performed on serpentine soils For a majority of the
soils listed in Table 42 total Cr concentrations for the serpentine soils were not
measured however these extractions do provide a quantitative assessment of where Cr
may be bound in soil (such as silicates organic matter and secondary Fe-bearing phases)
Extractions performed by Schwertmann and Latham (1986) were able to remove ~20000
Cr mg kg-1 from the serpentine soil using a dithionite-citrate bicarbonate (DCB) solution
which preferentially dissolves crystalline Fe-oxides Solutions such as concentrated
nitric and hydrochloric acid were only able to remove a maximum of ~1400 Cr mg kg-1
from Unst serpentine soils (Shewry and Peterson 1976) which preferentially dissolve
Cr-bearing silicates and to a lesser extent the Fe-oxides Where bulk soil Cr values were
measured extraction experiments are only able to remove a small portion (lt1) of the
total Cr present in the serpentine soils (Table 42) (Gasser and Dahlgren 1994 Oze et al
2003) Based on the extractions presented in Table 42 it appears that Cr(III) resides in
minerals and phases resistant to common extraction techniques
107
In serpentine soils Fe(III) oxides and Fe(III) hydroxides provide a host of
structural and surface sites available to sequester Cr(III) released to soil solutions by
dissolution of Cr-bearing minerals during soil-forming processes Substitution of Cr(III)
for Fe(III) in Fe-oxides Fe-(oxy)hydroxide minerals and amorphous compounds is
expected based on crystallographic considerations synthesis experiments and analysis of
extracts from selective dissolution of Fe(III) compounds in serpentine soils
(Schwertmann and Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990
Gerth 1990 Gasser and Dahlgren 1994) Chromium(III) also readily sorbs on surfaces
of Fe(III) oxides and (oxy)hydroxides via inner-sphere surface complexes (Figure 43)
and forms low-solubility precipitates (Charlet and Manceau 1992) The low
concentration of Cr(III) in soil solutions has been attributed to the low solubility of
Fe(III)-Cr(III) amorphous hydroxides (Sass and Rai 1987)
Layer silicates provide structural and surface sites capable of uptaking Cr(III)
Octahedral Cr(III) in clays such as smectites and vermiculites is expected and Cr
concentrations in the clay-sized fraction can achieve concentrations of ~1000 mg kg-1 in
serpentine soils (Oze et al 2003) Sorption of Cr(III) and Cr(VI) on edge and interlayer
sites of phyllosilicates have been evaluated by Ilton and Veblen (1994) Ilton et al (2000)
and Brigatti et al (2000) indicating that layer silicates (montmorillonite chlorite biotite
phlogopite and muscovite) provide sites for both Cr(III) sorption and Cr(VI) reduction
Kaupenjohann and Wilcke (1995) concluded that Cr in serpentine soils from Bavaria is
residually bound mainly in silicate structures including clay minerals based on extraction
experiments
Soil Solutions and Related Water
The addition of water gradually changes the spatial and temporal characterization
of pore fluids in soils including the ionic strength pH and redox potential Leaching of
a soil should progressively enrich the soil with less chemically mobile elements such as
Cr and Fe Elements identified in soil solutions include Mg Ca Fe Ni and Cr and the
ionic strengths of serpentine soil solutions are typically low (lt001) (Anderson et al
1973 Johnston and Proctor 1981 Gasser and Dahlgren 1994) Soil solutions obtained
from serpentine soils using tension-free lysimetry demonstrate Cr concentrations ranging
108
from 01 to 32 micromol L-1 (Gasser and Dahlgren 1994) This study also revealed a
majority of Cr Fe Mg and Ni in extracted soil waters were in the form of colloids
suggesting that nanoparticles are an important factor of Cr translocation in soils
The pH of a soil is a dominant factor controlling the nature of many chemical and
microbial reactions including the geochemistry of Cr (Brady and Weil 1999) Ranges in
pH for serpentine soils worldwide are plotted in Figure 46 The highest soil pH was
measured in a South Africa serpentine soil with a value of 9 (Cole 1992b) whereas the
lowest pH was reported in Ireland with a value of 38 (Jeffrey 1992) This wide range of
pH is a function of the adsorbed H+ and Al3+ the base-forming cations the organic
matter and the mineralogy present within each soil The pH range of serpentine soils
shown in Figure 46 is plotted on the stability diagram in Figure 43 demonstrating that
both Cr(III) and Cr(VI) aqueous species are favorable The Eh range for the serpentine
soils in Figure 3 is based on the soils being in contact with the atmosphere as well as
being a transitional environment Although the activity of H+ varies by four orders of
magnitude the pHs of most serpentine soils are slightly acidic (pH ~6) which is near that
of average rainwater (pH~57 Figures 43 and 46) Additionally serpentine soil pH
values have been observed to increase with depth towards the protolith for a number of
soils (Rabenhorst et al 1982 Gough et al 1989 Gasser and Dahlgren 1994) The
increase in pH with respect to depth may be the result of rainwater being progressively
influenced by the mineral chemistry and hydrolysis reactions as it descends through the
soil towards the protolith
Groundwater from serpentinized ultramafic rocks have pH values ranging from 8
to 12 (Barnes et al 1972) and these waters have been divided into two categories as
shown in Figure 46 Mg bicarbonate (Mg2+-HCO3-) type and Ca hydroxide (Ca2+-OH-)
type (Barnes and ONeil 1969 Barnes et al 1972) The Mg bicarbonate waters (pH~8)
are characteristic of most surface waters and shallow groundwaters from serpentinites
and ultramafic rocks The Ca hydroxide waters (pHasymp10-11) are associated with
incompletely serpentinized peridotites and these fluids are comparatively low in Mg2+
Fe2+ Mn2+ Crtotal and SiO2total The Ca2+ in solution originates from the serpentinization
of Ca-rich orthopyroxene (Barnes et al 1972) The pH of serpentinite groundwaters are
109
plotted on the Cr stability in Figure 43 demonstrating that both Cr(III) and Cr(VI) are
potential valence states in these waters
Chromium predominates as Cr(III) in ultramafic rocks and serpentine soils
however Cr(VI) has been identified in serpentine soil solutions and serpentinite
groundwater in California and New Caledonia at concentrations of lt20 microg L-1 (Gough et
al 1989 Becquer et al 2003) Calcium hydroxide waters (pH~11) collected from the
same spring evaluated by Barnes et al (1972) in the Cedars ultramafic body (Cazadero
CA) contains 12-22 Cr(VI) microg L-1 and no Cr(III) (this study) Additionally water from
chromite-containing ultramafic rocks in the Siniktanneyak Mountain ultramafic complex
of Alaska with pH values ranging from 67 to 93 were measured with low total Cr
concentrations lt2 Cr microg L-1 (Taylor et al 1998) A mechanism for Cr(III) oxidation in
serpentine soils and related water suggested by (Gough et al 1989 Becquer et al 2003)
may be the result of high valence Mn oxides common in serpentine soils High valence
Mn-oxides such as birnessite (δ-MnO2) are naturally occurring oxidants capable of
oxidizing Cr(III) to Cr(VI) at pHlt9 (Eary and Rai 1987 Fendorf et al 1992 Fendorf et
al 1993) It is also important to note that Cr(VI) reduction is hypothesized to occur in
soils due to electron transfer involving organic matter and Fe(II)-bearing minerals such as
magnetite (Peterson 1996 Fendorf et al 2000) Whether toxic concentrations of Cr(VI)
can be present in serpentine soils and related water over geologic time is important to
consider especially when high valence Mn oxides Fe(II)-bearing minerals and organic
matter are capable of significantly influencing the oxidation state of Cr
Serpentine Soil Vegetation
Vegetation associated with serpentine soils has received considerable attention
due to the apparent infertility of these soils The inhabitability of serpentine soils for
most plants has been attributed to the imbalance of available Mg relative to Ca or low
CaMg ratios (Walker 1954) the deficiency of plant nutrients such as K N and P
(Turitzin 1991) elevated heavy metal concentrations such as Ni and Cr (Robinson et al
1935) and poor drainage (Brooks 1987) Linking plant productivity to toxic levels of
Cr in the soil has not been fully resolved (Brooks 1987 Kruckeberg 1992) however a
majority of Cr in the soils is not available for plant uptake due to being in high stable and
110
weathering resistant phases discussed in the previous sections (Table 42) In general Cr
concentrations in plants typically do not exceed 100 mg kg-1 (Brooks 1987)
Table 43 provides a global list of serpentine soil plant species and their measured
Cr concentrations with an emphasis on Cr hyper-accumulators ie plants able
incorporate higher than average concentrations of Cr A majority of these measurements
are based on the dry weight of Cr Aerobryopsis longissima from New Caledonia has the
highest reported Cr concentration with values ranging from 200 to 700 mg kg-1 (Lee et
al 1977) Other noteworthy plants include the Lophostachys villosa from Brazil (Brooks
et al 1992) Arenaria humifusa from Newfoundland (Roberts 1992) the Calluna
vulgaris from Unst (Shewry and Peterson 1976) with Cr concentrations of 203 530 and
417 mg kgndash1 respectively Even mosses such as the Funaria flavicans (Maryland)
incorporate Cr concentrations as high as 90 mg kg-1 (Shaw 1991) By examining
specific parts of the plant Gough et al (1989) determined that Cr concentrations are
highest in the basal burl (600 mg kg-1) of the manzanita bush compared to the roots (30
mg kg-1) stems (20 mg kg-1) and leaves (20 mg kg-1) The study by Gough et al (1989)
demonstrates that Cr is not homogeneously distributed throughout the plant and that
evaluating a leaf versus a stem could result in significantly different Cr concentrations
Soil Organic Matter
Soil organic matter (SOM) including the biomass (living organisms such as
bacteria) detritus (nonliving plant residues) and humus (amorphous and colloidal
mixtures of organic substances) affects the geochemistry and buffering capacity of
serpentine soils Furthermore biologic reduction of Cr(VI) to Cr(III) via cometabolic
and metabolic pathways coupled with carbon oxidation has been demonstrated for a
number of bacterial strains including aerobes (Gopalan and Veeramani 1994 Park et al
2000) and anaerobes (Turick et al 1996 Tebo and Obraztsova 1998) Additionally
aqueous Fe(II) produced by microbial Fe(III) reduction is a known biogenic pathway for
the reduction of Cr(VI) (Fendorf et al 2000) Fe(III)-Cr(III) (oxy)hydroxides are the
resulting secondary phases (Rai et al 1989 Patterson and Fendorf 1997 Hansel et al
2003) Bacteria oxidizing Cr(III) to Cr(VI) has not been observed
111
Most serpentine soils in temperate regions appear to be either Mollisols or
Alfisols which often contain high concentrations of SOM Alexander et al (1989) and
Bulmer et al (1992) reported that serpentine soils can contain organic carbon
concentrations in excess of 50 mg g-1 Humus has a high cation exchange capacity and
provides much of the pH buffering capacity in these soils (Brady and Weil 1999)
Organic acids such as fulvic and humic acids are known to complex cations such as
Cr3+ Fe3+ and Mn2+ forming stable organometallic complexes (Kaupenjohann and
Wilcke 1995 Brady and Weil 1999) Additionally metal leaching experiments
performed by Schreier et al (1987) established that organic acids produced by soil
organic matter such as oxalic acid and citric acid were extremely effective in releasing
Cr Mg Fe Ni and Co from serpentinite rocks and sediments
CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
Serpentine soils were obtained from California Oregon and New Caledonia and
analyzed to further evaluate the factors influencing the Cr geochemistry in serpentine
soils Analyses for these soils include pH electrical conductivity total organic matter
total N ammonium acetate extractable Ca K Mg and Na soil texture soil color
chemical composition of the soil and the chemical composition of Cr-bearing minerals
The general soil characteristics are described in the following section and the analytical
methods are explained in Appendix 41 Although a majority of these soils are located in
the western United States factors influencing the pedogenesis of these soils vary
considerably allowing a comparative analysis of how the fate of Cr might be influenced
by climate topography biota parent material and time
Sample Locations and General Soil Characteristics
Six serpentine soils were collected from locations shown in Figures 41 and 42 to
evaluate the chemistry of Cr in serpentine soils These soils include 1) a laterite (Oxisol)
collected in New Caledonia South Pacific 2) a laterite (Oxisol) exposed at Eight Dollar
Mountain Selma OR 3) a laterite (Oxisol) near Nickel Mountain Riddle OR 4) a
serpentine soil (Inceptisol) with a CO2-metasomatized serpentinite protolith from
Harvard Mine Jamestown CA 5) a serpentine soil (Mollisol) obtained from a preserved
112
grassland at Jasper Ridge Biological Preserve Stanford CA and 6) a serpentine soil
(Inceptisol) collected in the proximity of Pillikin Mine El Dorado CA Lastly an Oxisol
formed from basalt obtained in Oahu Hawaii is also included for comparison
These seven sites include a range of soil environments from different protoliths
under varying conditions of rainfall temperature and topography Each soil was collect
near the highest and topographically level point for each serpentine soil site Serpentine
grassland soils at Jasper Ridge formed under Mediterranean-like climates (xeric moisture
regime) whereas higher temperatures and less rainfall (ustic moisture regime)
characterize soil formation at the Harvard and Pillikin mines in the Sierra Nevada
Foothills metamorphic belt The vegetation of these three sites is dominantly grasses with
some manzanita These modern soil environments contrast with the Miocene Ni and Cr-
rich laterite soils of Nickel Mountain and Eight Dollar Mountain that were probably
formed under an aquic moisture regime near sea level in a subtropical climate Grasses
manzanita and pine trees are the dominant forms of vegetation at Nickel and Eight
Dollar Mountain An ultramafically-derived laterite formed under tropical conditions
from New Caledonia was collected The vegetation and topography related to the New
Caledonia laterite was not recorded during sampling An Oxisol from Hawaii is also
included in this study to evaluate the differences between a highly weathered soil
originating from an ultramafic rock and one from a mafic rock (basalt) under tropical
conditions The Hawaiian soil was collected from a topographically level area however
the vegetation related to this site was not noted
BedrockProtolith Mineralogy
A broad characterization of the protoliths for the serpentine soils is briefly
summarized due to the bedrock not being collected from all the field sites New
Caledonia is composed of variably serpentinized peridotites predominantly harzburgites
and some dunites (Leblanc 1995) Nickel Mountain and Eight Dollar Mountain are
predominantly serpentinized harzburgite and dunite The bedrock for soils collected at
Harvard Mine is mainly composed of quartz ankerite and mariposite (also referred to as
QAM) caused by the CO2 metasomatism of serpentinite associated with gold
mineralization (Bohlke 1989) The average serpentinite at Jasper Ridge is composed of
113
serpentine (mainly lizardite and antigorite) chlorite talc magnetite and chromite (Oze et
al 2003) however brucite has been reported as a major constituent in related Franciscan
Complex serpentinites (Hostetle et al 1966) Pillikin Mine serpentinites dominantly
consist of talc and serpentine (lizardite antigorite and chrysotile) and these serpentinites
contain podiform chromite deposits (Cater et al 1948) Mafic bedrock for the Hawaiian
Oxisol is tholeittic basalt
Soil Chemistry and Physical Properties
As shown in Figure 47 the colors of the collected soils range from the red New
Caledonia laterite to the greenish gley of the Pillikin Mine Inceptisol The Munsel color
values are also shown in Figure 47 The dark grey color of the Jasper Ridge soil and the
greenish gley of the Pillikin Mine soil are atypical of soils derived from serpentinites due
to the lack of any apparent redness The range of these colors is often attributed to the
presence of Fe-oxides however organic matter and Mn oxides are also dominant factors
controlling color (Brady and Weil 1999)
Chemical properties for the six serpentine soils and the Hawaiian soil are listed in
Table 44 The slightly acidic pH values range from 53 to 69 similar to those shown in
Figure 43 and 46 The New Caledonia laterite is the most acidic soil with a pH of 53
compared to the Harvard Mine carbonate-buffered soil at 69 The pH of the Hawaiian
soil is also slightly acidic with a pH of 604 The electrical conductivities (EC) of the
soils range from 01 to 07 dS m-1 a range indicative of relatively few dissolved inorganic
solutes The highest organic matter content is in the very dark grey Jasper Ridge soil
with a value of 32 and the lowest is in the red New Caledonia laterite (05) The
Hawaiian soil has an organic matter content of 1 Nitrogen concentrations for all the
soils are lt012 with Jasper Ridge having highest N concentration which is consistent
with the soil organic matter content
Ammonium acetate extractable Ca K Mg and Na are also listed in Table 44
Magnesium concentrations are gt6923 mg kg-1 in all of the serpentine soils The lowest
concentration of Mg was measured in the basalt-derived Hawaiian Oxisol with a value of
168 mg kg-1 Calcium concentrations in the serpentine soils range from 167 to 998 mg
kg-1 The 998 mg kg-1 of Ca in the Harvard Mine soil is directly related to the abundance
114
of ankerite (gt20 of the bedrock) Potassium concentrations are low for most of the soils
(lt55 mg kg-1) however Jasper Ridge and Harvard Mine soils have the highest
extractable K with values of 237 and 301 mg kg-1 respectively The extractable K value
for the Harvard Mine soil is due to the K-metasomatism related to gold mineralization
involving the serpentinite soil protolith The extractable concentration of Na is lt44 mg
kg-1 for all the serpentine soils whereas the Hawaiian soil has the highest Na (53 mg kg-
1) due to plagioclase and basaltic glass
Textures for the soils range from clay loam clay loam to sandy clay loam (Table
44) The laterite soil in New Caledonia and Oxisol in Hawaii are extremely different in
which the Hawaiian soil is mostly clay (66 clay) compared to the loam at New
Caledonia (25 clay) Additionally both ancient laterites (Eight Dollar Mountain and
Nickel Mountain) have clay textures Harvard Mine Jasper Ridge and Pillikin Mine
soils are all loamy in character
Major and trace element compositions for the soils are listed in Table 45
Chromium concentrations in the serpentine soils range between 827 and 9528 mg kg-1
The Eight Dollar Mountain laterite has the highest Cr concentration and the New
Caledonia laterite contains the lowest Cr concentration The Cr concentration for the
New Caledonia soil is lower than the values reported in Table 41 The Hawaiian soil has
a Cr concentration well below the analyzed serpentine soils with 388 Cr mg kg-1 but still
greater than the 200 mg kg-1 characteristic of serpentine soils Nickel concentrations in
the serpentine soils range from 1895 to 14100 mg kg-1 with the highest concentration
being identified in the Nickel Mountain laterite Manganese concentrations range
between 257 (New Caledonia) to 3017 (Eight Dollar Mountain) mg kg-1 in the serpentine
soils Element concentrations in the Hawaiian Oxisol contrast from the serpentine soils
with higher Al Cu P Pb and Zn concentrations
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in the soil solids for Jasper Ridge New Caledonia Pillikin Mine Eight Dollar
Mountain and Harvard Mine Although the Cr concentrations in these soils exceed 200
mg kg-1 (Table 45) X-ray absorption near-edge spectra (XANES) shown in Figure 48
illustrate Cr(VI) levels are below the detection limits of this method (~10 Cr mg kg-1) due
to the lack of a pre-edge feature The box in Figure 48 illustrates where a Cr(VI) pre-
115
edge peak should be present Chromium(VI) concentrations in these soils are possibly
present in the parts per billion range
Soil Mineralogy
Backscattered electron images and microprobe analyses were used to identify the
Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49
and corresponding Cr-bearing mineral compositions are listed in Table 46 The New
Caledonia soil exhibits a wide range of particle sizes and a Fe-(oxy)hydroxide was
identified containing 165 Cr2O3 Wt Chromium-spinel has been identified in the New
Caledonia laterite by Becquer et al (2003) Eight Dollar Mountain soils exhibit a coarse
texture and both Cr-magnetite (2387 Cr2O3 Wt ) and chromite (3343 Cr2O3 Wt )
are present Nickel Mountain soils are also coarse and chromite (1315 Cr Wt ) was
identified Soils at Harvard Mine contain numerous chromite grains and the mineral
trevorite (a Ni-bearing spinel) was identified with 3991 Cr2O3 Wt and 2054 NiO Wt
Chromite and magnetite are abundant in soils at Jasper Ridge as well as a fine-
grained mixture of chromite magnetite and an unidentifiable silicate with up to 1878
Cr2O3 Wt (Figure 49) Veins of Cr-chlorite crosscut the spinel-silicate mixture in
Figure 49 and this Cr-chlorite is the only Cr-silicate identified using this method
Chromite (5355 Cr2O3 Wt ) is the dominant Cr-bearing mineral detected in soils at the
Pillikin Mine and these grains contain the highest Cr concentrations compared to the
other minerals in the seven soils The Hawaiian Oxisol possesses ulvoumlspinels (007
Cr2O3 Wt and 2465 TiO2 Wt ) which are evenly distributed throughout the soil
Overall microprobe observations reveal that the Cr-bearing spinels are present in every
soil Additionally Cr-bearing spinels at Jasper Ridge have been shown to undergo
incongruent dissolution progressively enriching the spinel towards a FeCr2O4 end-
member (Oze et al 2003)
DISCUSSION
The physical and chemical characteristics of serpentine soils are directly related to
the alteration and tectonic history of the serpentinite as well as the soil-forming factors
including the parent material climate biota topography and time Identifying Cr-
116
bearing minerals their chemical compositions and abundance in the soils allows a more
definitive explanation for Cr enrichment in these soils as compared to total Cr
concentrations listed in Tables 41 and 45 and sequential extraction experiments reported
in Table 42 Detrital chromite grains from the serpentinite protolith are resistant to
weathering and accumulate in the soil over time Although serpentine soils contain Cr
concentrations typically gt200 mg kg-1 the presence of Cr-spinels and their resistance to
weathering demonstrate that a majority of the Cr is not chemically available to interact
within the soil The role of organic matter and vegetation are important factors related to
the chemistry of serpentine soils however their role for influencing a majority of the Cr
geochemistry is limited due to the nature of the Cr-spinels
Chromium and Serpentine Soils
Total Cr concentrations exceed 200 mg kg-1 for most of the serpentine soils
reported in Table 41 and in the six serpentine soils of this study (Table 45) These Cr
concentrations can be accounted for by assessing the Cr-bearing minerals associated with
these soils and their relative abundance Small grains of Cr-spinels or chromite identified
in the soils (Figure 49 and Table 46) have the potential to appreciably impact Cr
concentrations Table 47 demonstrates the mineral masses and volumes of Cr-bearing
minerals required to cause predetermined Cr concentrations in soil samples A minimal
amount of chromite (lt02 of a soil sample mass or 005 of the volume of serpentine)
can result in Cr concentrations of 1000 mg kg-1 A few grains of chromite could easily
be responsible for the Cr concentrations in Tables 41 and 45 In fact residual grains of
Cr-enriched spinels are observed in every soil analyzed (Figure 49 and Table 46)
Ultimately grains of spinel in the soil do appreciably impact Cr concentrations however
they do not provide a readily available source of Cr based on extraction experiments
performed by Oze et al (2003) Serpentine minerals could also provide minor Cr
concentrations to the soils however soil extractions have consistently shown that a
majority of the Cr is in highly resistant phases capable of resisting dissolution even in
concentrated hydrochloric acid nitric acid and hydroxylamine with acetic acid (Table
42) Chromium end-member minerals including eskolaite Cr-muscovite (fuchsite) Cr-
garnet (uvarovite) Cr-epidote (tawmawite) and Cr-chlinochlore (kaumlmmererite) are also
117
included in Table 47 to demonstrate that relatively low abundances of these minerals are
all that is necessary to increase Cr concentrations in rocks and soils Chromium-
clinochlore is often observed developing adjacent Cr-spinels in serpentinites and is
perhaps a more accessible mineral source of Cr in the serpentinites compared to Cr-
spinels (Phillips et al 1980 Schreyer et al 1981 Kerrich et al 1987 Treloar 1987a
Pan and Fleet 1991 Schandl and Wicks 1993 Christofides et al 1994 Challis et al
1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Oze et al 2003)
As shown in Table 44 and Figures 43 and 46 the chemical properties (pH EC
N etc) of serpentine soils are not correlated to Cr concentrations measured in the soil
solids (Tables 41 Figure 45) due to Cr-spinels accounting for a majority of the Cr
(Figure 49 and 410 Tables 46 and 47) However extraction experiments and soil
solution analyses demonstrate that a limited fraction of the total Cr (lt1) is extractable
If only 1 of the Cr in a serpentine soil is extractable a serpentine soil containing 10000
Cr mg kg-1 could possibly contribute up to100 Cr mg kg-1 still a significant Cr
concentration The chemical properties of the soil such as pH will influence this limited
fraction of the Cr in the soils Acidic pHs may aid in mobilizing Cr(III) however this
will only occur at low pHs (lt4) which exceed the range of most serpentine soils (Figures
43 and 46) pH values greater than 8 like those in some serpentine soils and
serpentinite groundwater may help sustain Cr(VI) concentrations in solution (Figure
43) In addition to pH organic matter in the soils may be a significant factor related to
the mobility of Cr due to the production of organic acids promoting the dissolution of the
Cr-bearing phases but particulate organic matter is known to bind Cr(III) in stable stable
organometallic complexes (Kaupenjohann and Wilcke 1995 Brady and Weil 1999)
Evaluating the valence state of Cr is crucial for understanding the chemistry as
well as the potential toxicity related to serpentine soils The soil solids do not contain
detectable concentrations of Cr(VI) (Figure 48) however Cr(VI) concentrations are
present in soil solutions and groundwater at values lt30 microg L-1 (Gough et al 1989
Becquer et al 2003) A majority of serpentinite Cr-bearing minerals only contain
Cr(III) therefore there must be a reaction or process by which Cr(III) is oxidized in a
water saturated soil High valence Mn oxides such as birnessite are present within
serpentine soils and provide a feasible explanation for Cr(III) oxidation Additionally
118
Mn oxides or other strong oxidants in the soil might accelerate the weathering of Cr-
bearing minerals including chromite and generate the Cr(VI) concentrations observed in
soil solutions and groundwater After Cr(III) is oxidized Cr(VI) is a chemically mobile
species and would be transported away from the serpentine soils before it can be
observed in the solid phase Additionally Cr(VI) may be subsequently reduced by
reacting with organic matter or Fe(II)-bearing minerals such as magnetite which are
present in a majority of serpentine soils Perhaps the best way to evaluate the potential of
serpentine soils being a non-anthropogenic source of Cr contamination is to monitor
Cr(III) and Cr(VI) in the soil solutions over time as demonstrated by Gasser and
Dahlgren (1994)
It is curious that plants do uptake Cr in serpentine soils especially when Cr is
tightly bound in the soil matrix (Table 42) In order for plants to uptake Cr it must be in
a form to be absorbed by plant roots Soil pH in the rhizosphere can be altered by plant
roots by as much as 2 units within two days (Fischer et al 1989) This acidification is
known to increase the solubility of the minerals in the soils and could serve as a
mechanism by which plants can incorporate this relatively insoluable element Another
possibility is that Cr(III) in the soil is oxidized by either organic or inorganic means and
then incorporated into the plant in its more chemically mobile and bioavailable form
Cr(VI) How and why plants incorporate as much Cr as they do needs to be examined
more thoroughly
Chromium Spinel in Soils and Rocks
Compositions of chromite identified in serpentinites worldwide (Figure 44) and
Cr-bearing spinels identified in the six serpentine soils are compared in Figure 10 in order
to determine whether the soil spinels are geochemically similar to those in serpentinites
This comparison is ascertained by examining how Al Fe and Mg Wt values varied
with respect to Cr Wt Trends for magnetitehematite chromite and the
microcrystalline spinel-silicate mixture observed in the soils are noted in each diagram in
Figure 410 Additionally the highest Cr concentrations for soil chromite
magnetitehematite and the microcrystalline spinel-silicate mixture are ~40 Cr Wt
~25 Cr Wt and ~15 Cr Wt respectively Based on Al Wt soil
119
magnetitehematite deviates significantly from the chromite trend with Al Wt values
lt5 Magnetitehematites are enriched with Fe relative to serpentinite and soil chromite
and Mg concentrations in the soil chromite are slightly higher than serpentinite chromite
Soil magnetitehematite and chromite are similar by demonstrating a linear trend when
Al Fe and Mg are summed The microcrystalline spinel-silicate mixture (CSM) does
not follow this trend because of Si present in this mineral mixture Overall soil
chromites are similar to serpentinite chromite and soil magnetiteshematites can be
differentiated from the chromites when Cr Wt is plotted against Al Fe and Mg Wt
CONCLUDING STATEMENT
Serpentine soils are abundant among convergent margins and the chemistry of
these soils is ultimately a function of the ultramaficserpentinite rock mineralogy and
texture as well as all the common soil-forming factors Although ultramafic rocks and
serpentinites produce groundwater that is highly basic most serpentine soil pHs are
slightly acidic Although Proctor and Nagy (1992) argued that it is futile to categorize
serpentine soils based on chemistry and biology (otherwise referred to as the lsquoserpentine
factorrsquo) as proposed by Brooks (1987) Cr concentrations greater than 200 mg kg-1 in
serpentine soils are consistently present Chromium in the soils originates from minerals
in the bedrock where the valence is primarily Cr(III) the exception being Cr(VI) in the
rarely present mineral crocoite (PbCrO4) Increased Cr-concentrations in serpentine soils
are largely the result of chromite or other resistant Cr-bearing minerals causing a
localized enrichment in the soils Although Cr in serpentine soils is not readily released
from the soil matrix serpentine soil vegetation can incorporate Cr concentrations as high
as 700 mg kg-1 (Table 34) Finally the limited extractability of Cr and site dependent
factors such as the presence of organic matter clays Fe(II)-bearing minerals pH and the
poor drainage indicates that the possible formation and sustained presence of Cr(VI) is
not a portentous matter related to serpentine soils
120
ACKNOWLEDGEMENTS
We appreciate technical and field support from Robert Jones Anders Meibom
Rob Dunbar and Guangchao Li Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund GSA Graduate Student Research Grants and the
Office of Teaching and Learning at Stanford University is gratefully acknowledged
121
Table 41 Measured Cr concentrations in serpentine soils worldwide
Location Cr concentration (mg kg-1) Reference
Canyon Creek area OR 920 (White 1971) 66-13200 (Brooks et al 1992)
Ireland 29-5000 (Proctor and Woodell 1971 Jeffrey 1992) Italy 1300-3900 (Gambi 1992) Japan 5200 (Shewry and Peterson 1976) Maryland 1500-6000 (Rabenhorst et al 1982) Mt Tamalpais CA 1700 (White 1971) New Caledonia 6300-80000 (Jaffre 1980) New Caledonia 2-6 (Schwertmann and Latham 1986) New Idria CA 1640 (White 1971) New Zealand 850-3250 (Lee 1992) Newfoundland Canada 309-1447 (Roberts 1992) Nickel Mountain OR 1730 (White 1971) Santa Barbara County CA 2080 (Woodell et al 1975) Southern Africa 045-14 (Morrey et al 1992) Tehama County 023-12 (Gough et al 1989) Transvaal Lowveld South Africa 120-800 (Cole 1992) Unst Shetland 3000-4000 (Proctor 1969) Wenatchee Mnts WA 895 (White 1971) Zimbabwe 110-46000 (Soane and Saunder 1959 Proctor et al 1980)
Goias State Brazil
122
123
Table 42 Summary of Cr Fe and Ni extractions from serpentine soils
Range of Extractable Cr Fe and Ni Location Cr Fe Ni
Method Reference
Italy 2-5 microg g-1 305-640 2300-2700microg g-1 25 CH3COOH (Gambi 1992)
NE Portugal 02-09 microg g-1 120-540 microg g-1 40-290microg g-1 EDTA (Menezes de Sequeira and Pinto da Silva 1992) New Caledonia 22-23 384-581 02-216 DCB (dithionite-citrate bicarbonate) (Schwertmann and Latham 1986) New Caledonia 02-11 microg g-1 02-66microg g-1 Method Unknown (Jaffre 1980)
Norway 357microg g-1 316microg g-1 HClO4HNO3 (concentrated) (Zeien and Bruemmer 1989) Norway 254microg g-1 134microg g-1 Ascorbic and Oxalate Acid pH=325 (Zeien and Bruemmer 1989)
Norway 62 microg g-1 53microg g-1 NH4-EDTA (Zeien and Bruemmer 1989) Norway 22-60 microg g-1 8-25 mg kg-1 1-7 microg g-1 pH-stat reactor EDTA (Kaupenjohann and Wilcke 1995)
Swiss Alps 08 micromol L-1 28micromol L-1 228microg g-1 Lysimeter (Nonfiltered) (Gasser and Dahlgren 1994) Trinity Ophiolite CA 004-045 gkg 27-210 g kg-1 005-245microg g-1 Citrate-Dithionite Extracts (Alexander et al 1989)
Unst lt01 microg g-1 19-77microg g-1 Ammonium Acetate (Shewry and Peterson 1976)
Unst 500-1390microg g-1 1540-3330microg g-1 HNO3-Soluble (Shewry and Peterson 1976) JRBP CA 0 microg g-1 01microg g-1 103microg g-1 BaCl2 Method (Oze et al 2003)
JRBP CA 78 microg g-1 513 microg g-1 150 microg g-1 Hydroxlyamine+Acetic Acid (Oze et al 2003) JRBP CA 31 microg g-1 2039microg g-1 35microg g-1 10M Hydrofluoric Acid (Oze et al 2003)
123
124
Table 43 Serpentine soil plant Cr concentrations
Plant Cr Concentration (microg g-1) Location Reference
Lophostachys villosa 203 Goias State Brazil (Brooks et al 1992) Ruellia geminiflora
56 Goias State Brazil (Brooks et al 1992) Minuartia ophiolitica 83 Italy (Gambi 1992) Sanguisorba minor 194 Italy (Gambi 1992) Plantago lanceolata 135 Italy (Gambi 1992) Cerastium latifolium 110 Italy (Gambi 1992) Funaria flavicans 34-899 Maryland (Shaw 1991) Avena sativa 168-583 New Zealand (Lee 1992) Aerobryopsis longissima 200-700 New Caledonia (Lee et al 1977) Arenaria humifusa 530 Newfoundland Canada (Roberts 1992) Senecio pauperculus 406 Newfoundland Canada (Roberts 1992) Rhalomitrium lanuginosum 246 Newfoundland Canada (Roberts 1992) Santolia semidentata 95 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Asplenium cuneifolium 228 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992) Armeria eriophylla 38 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Arctostaphylos viscida 029-705 Tehama County CA (Gough et al 1989) Calluna vulgaris 417 Unst (Shewry and Peterson 1976) Armeria maritima 149 Unst (Shewry and Peterson 1976) Agrostis stolonifera 120 Unst (Proctor and Woodell 1971) Dicoma niccolifera 77 Zimbabwe (Proctor 1992) Xerophyta equisetoides 68 Zimbabwe (Proctor 1992)
124
125
Table 44 Chemical and textural properties of the collected soils
Ammonium Acetate Extractable ------------------------Hydrometer--------------------- Ca K Mg Na
Sand Silt ClaySample pH EC dSm
Organic Matter ------
LECO Nitrogen ------ -------------------mg kg-1-------------------
------ ------ ------
Texture
New Caledonia 534 01 05 lt001 279 29 692 44 43 32 25 LoamHawaii 604
03 10 002 839 46 168 53 8 26 66 ClayEight Dollar Mnt 651 07 16 006 599 62 3464 25 28 28 44 Clay
Nickel Mnt 665 05 07 001 167 42 3289 17 24 32 44 ClayHarvard Mine 691 03 11 003 998 301 2897 43 41 30 29 LoamJasper Ridge 671 04 32 012 505 237 3291 36 36 31 33 Clay LoamPillikin Mine 642 04 05 002 185 55 3421 29 45 23 32 Sandy Clay Loam
125
126
Table 45 Total major and trace element concentrationsdagger of the collected soils
Element Concentration New Caledonia Hawaii Eight Dollar
Mnt Nickel Mnt Harvard Mine Jasper Ridge Pillikin
Mine Ag mg kg-1 lt05 lt05 lt05 lt05 02 lt05 lt05Al 931 1325 314 095 179 314 179Ba mg kg-1 lt10 180 40 20 30 230 lt10Be mg kg-1 05 16 05 lt05 lt05 08 lt05Ca 003 012 085 018 007 079 002Cd mg kg-1 lt05 lt05 18 lt05 05 lt05 lt05Co mg kg-1 56 183 338 403 85 155 122Cr mg kg-1 827 388 9528 5761 2900 5976 1977Cu mg kg-1 78 248 43 25 51 33 4Fe 638 1296 1635 17 538 727 468K 002 046 004 003 004 039 001
Mg 072 026 753 829 792 1025 gt150Mn mg kg-1 257 2706 3017 2554 920 1664 903Na 001 008 005 001 003 057 lt001Ni mg kg-1 2780 345 5510 14100 1895 2180 2760P mg kg-1 50 260 160 150 130 140 20Pb mg kg-1 6 31 4 3 2 8 6Ti 016 041 008 002 lt001 02 002V mg kg-1 63 132 158 54 65 92 27W mg kg-1 10 10 10 10 lt10 lt10 lt10Zn mg kg-1 43 218 160 132 66 111 47
126
dagger ICP-AES analyses
127
Table 46 Compositionsdagger for soil Cr-containing minerals identified in Figure 49
Wt New Caledonia A
Hawaii B
Eight Dollar Mnt C
Eight Dollar Mnt D
Nickel Mnt E
Harvard Mine F
Jasper Ridge G
Jasper Ridge H
Pillikin Mine I
SiO2 491 008 175 026 002 000 1940 001 000Al2O3
756 185 272 2235 2041 088 559 1575 839NiO 023 005 001 006 011 2054 014 002 007FeO 7766a 6212 6633b 3420b 1784 2445 2629 2314 3105MnO 004 037 181 123 133 031 155 146 216Cr2O3 165 007 2387 3343 4682 3991 1878 5024 5355MgO 048 121 202 653 1315 1062 1668 926 497CaO 001 000 013 008 000 000 015 000 000Na2O 000 000 002 001 000 000 006 000 000TiO2 004 2465 012 006 012 000 000 003 018
TOTAL 9257 9041 9879 9821 9980 9671 8864 9920 10037Mineral Cr-Hydroxide Ulvoumlspinel Cr-Magnetite Chromite Mg-Chromite Trevorite CSM1 Chromite Chromite
127
dagger Electron microprobe analyses This is the form of iron unless otherwise stated a FeO is in the form Fe2O3 b FeO is in the form of Fe3O4 1CSM Microcrystalline chromitesilicate mixture
128
Table 47 Chromium-bearing mineraldagger masses and volumes required to influence Cr concentrations in soils
Cr Concentration Mineral mass (grams) required for 1000 g bulk soil analysis Mineral Mass Percent () (mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite
500 11 07 22 24 48 57 011 007 022 024 048 057 1000 22 15 43 48 95 115 022 015 043 048 095 1152000 43 29 86 96 191 230 043 029 086 096 191 2305000 108 73 216 241 477 574 108 073 216 241 477 574
Cr Concentration Mineral Volume Percent () in Serpentine Mineral Volume Percent () in Quartz
(mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite500 005 004 02 02 03 15 006 004 02 02 04 06
1000 01 007 04 03 07 30 01 007 04 03 07 112000 02 014 07 06 14 59 02 015 07 06 14 225000 06 036 18 16 34 147 06 037 18 16 35 55
128
daggerMineral formulas eskolaite Cr2O3 chromite FeCr2O4 fuchsite KCr2(AlSi3O10)(OH)2 uvarovite Ca3Cr2Si3O12 tawmawite Ca2CrAl2Si3O12(OH)2 kaumlmmererite Mg5CrAlSi3O11(OH)8
Note Molar volumes used in the calculations above are from their Al-enriched counterparts in Robie and Hemingway (1995) The molar volume for kaumlmmererite was estimated using unit cell volumes from XRD unit cell measurements from Robie and Hemingway (1995) Christofides et al (1994) and Phillips et al (1980)
129
Figure Captions
Figure 41 Selected serpentine soil localities as well as soils collected for analyses are
shown on a world map Active plate boundaries are shown Serpentine soil localities not
adjacent the plate boundaries on this map are related to ancient plate boundaries that are
no longer active
Figure 42 Field pictures of soils collected from a) Eight Dollar Mountain b) Nickel
Mountain c) Harvard Mine d) Pillikin Mine and f) Jasper Ridge Biological Preserve
Figure 43 Stability field diagram of Cr under a range of Eh and pH conditions (log a
Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom (1998) Adsorption
isotherms for Cr3+ and CrO42- on hydrous ferric oxide (HFO) are also shown (Leckie et
al 1980 Leckie et al 1984)
Figure 44 Microprobe analyses demonstrating Cr versus Al Wt for Cr-spinel
(chromite) and Cr-silicates including Cr-mica (fuchsitemariposite) Cr-garnet
(uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite) (Whitmore et al
1946 Leo et al 1965 Chen and Lee 1974 Neiva 1978 Cooper 1980 Phillips et al
1980 Grapes 1981 Schreyer et al 1981 Max et al 1983 Nutman et al 1983 Raase
et al 1983 Chao et al 1986 Von Knorring et al 1986 Kerrich et al 1987 Treloar
1987b a Morand 1990 Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen
1999 Proenza et al 1999) Chromium-spinel and Cr-silicate compositions based on
octahedral Cr and octahedral Al are also shown with tie-lines (dotted) denoting ideal
isomorphic substitution between Cr and Al end-members
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above Each dotted line connecting the
points represents a separate soil profile at each site
129
Figure 46 Ranges of pH for serpentine soils and related groundwater worldwide Soil
pH values were obtained from A(Kruckeberg 1992) B (Graham et al 1990) C
(Alexander et al 1989) D (Gough et al 1989) E (Rabenhorst et al 1982) F
(Roberts 1992) G (Bulmer et al 1992) H (Proctor 1992a) I (Jeffrey 1992) J
(Gambi 1992) K (Verger 1992) L (Proctor and Cole 1992) M (Cole 1992b) N
(Morrey et al 1992) O (Proctor 1992b) P (Mizuno and Nosaka 1992) Q (Cole
1992a) and R (Lee 1992) Groundwater pH values from serpentinite are from Barnes
and ONeil (1969) and Barnes et al (1972) Note the range of pH environments possible
for ultramafic rocks and their related soils
Figure 47 Soil pictures with Munsel color values and names from soils collected from
a) New Caledonia b) Hawaii c) Eight Dollar Mountain d) Nickel Mountain e) Harvard
Mine f) Jasper Ridge Biological Preserve and g) Pillikin Mine
Figure 48 Cr XANES spectra for soils analyzed from Jasper Ridge New Caledonia
Pillikin Mine Eight Dollar Mountain and Harvard Mine The pre-edge area indicative of
Cr(VI) peaks is noted in the diagram No definitive Cr(VI) peaks are present in any of
these spectra
Figure 49 Backscattering images of mineral phases within each soil are presented
Electron microprobe analyses are noted by a circle and letter for the Cr-bearing minerals
and the results are listed in Table 46
Figure 410 Chromite magnetite and hematite compositions from the collected soils
(gray circles) as well as analyses for serpentinite chromite (white squares) from Figure
44 are shown in these diagrams with Cr Wt versus Al Fe Mg and Al+Fe+Mg Wt
130
Figure 41
131
Figure 42
132
Figure 43
133
Cr Wt
0 10 20 30 40 50
Al W
t
0
5
10
15
20
25Cr-SpinelCr-MuscoviteCr-EpidoteCr-GarnetCr-Chlorite
Cr-SpinelBased on 4 Oxygens
Cr0 1 2
AlVI
0
1
2
ChromiteFeCr2O4
HercyniteFeAl2O4
Cr-MuscoviteBased on 11 Oxygens
Cr0 1 2
AlVI
0
1
2
MuscoviteKAl2(AlSi3O10)(OH)2
FuchsiteKCr2(AlSi3O10)(OH)2
Cr-GarnetBased on 12 Oxygens
Cr0 1 2
AlVI
0
1
2
GrossularCa3Al2Si3O12
UvaroviteCa3Cr2Si3O12
Cr-EpidoteBased on 125 Oxygens
Cr0 1
AlVI
2
3
ClinozoisiteCa2Al3Si3O12(OH)2
TawmawiteCa2CrAl2Si3O12(OH)2
Cr-ChloriteBased on 14 Oxygens
Cr0 1
AlVI
0
1
2
Kammererite(Mg5Cr)(AlSi3)O10(OH)8
Clinochlore(Mg5Al)(AlSi3)O10(OH)8
Figure 44
134
Maryland (USA)(Rabenhorst et al 1982)
Chromium (mg kg-1)
0 1000 2000 3000 4000 5000 6000 7000
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Tehama County CA (USA)(Gough et al 1989)
Chromium (mg kg-1)
0 2000 4000 6000 8000 10000 12000 14000
Dep
th (m
eter
s)
00
02
04
06
08
Kalamath Mountains CA (USA)(Hotz 1964)
Chromium (mg kg-1)
2000 4000 6000 8000 10000 12000 14000 16000
Dep
th (m
eter
s)
00
05
10
15
20
25
30
35
40
45
50
Newfoundland (Canada)(Roberts 1992)
Chromium (mg kg-1)
800 1000 1200 1400 1600 1800
Dep
th (m
eter
s)
00
02
04
06
08
Harvard Mine CA (USA)(Unpublished Data)
Chromium (mg kg-1)
0 500 1000 1500 2000 2500
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Jasper Ridge CA (USA)(Oze et al 2003)
Chromium (mg kg-1)
1500 2000 2500 3000 3500 4000 4500 5000
Dep
th (m
eter
s)
00
01
Figure 45
135
LocationsZ A B C D E F G H I J K L M N O P Q R S
pH
0
1
2
3
4
5
6
7
8
9
10
11
12
Nor
th A
mer
ica
Can
ada
Brit
ain
Italy
Far E
ast
Japa
n
Zim
babw
e
Sout
h A
fric
a
Wes
tern
Aus
tral
ia
New
Zea
land
Mg-Bicarbonate Waters pH=8
Ca-Hydroxide Waters pH=11-12
Average RainpH=57
Sout
h A
fric
a
Italy
Irela
nd
Cal
iforn
ia
Cal
iforn
ia
Brit
ish
Col
umbi
a
Mar
ylan
dCal
iforn
ia
Figure 46
136
Figure 47
137
-2
-1
0
1
2
3
4
5
5980 5990 6000 6010
Energy (eV)
Nor
mal
ized
Abs
orba
nce
New Caledonia
Pillikin Mine
Eight Dollar Mnt
Jasper Ridge
Harvard Mine
Cr(
VI)
Figure 48
138
Figure 49
139
Cr Wt 0 10 20 30 40 50
Al W
t
0
10
20
30 Soil Serpentinite
Cr Wt 0 10 20 30 40 50
Fe W
t
0
20
40
60
80
Cr Wt 0 10 20 30 40 50
Mg
Wt
0
2
4
6
8
10
Cr Wt 0 10 20 30 40 50
Al +
Fe
+ M
g W
t
0
20
40
60
80
MagnetiteHematite
Chromite
MagnetiteHematite
Chromite
Chromite
MagnetiteHematite
SpinelSilicate
SpinelSilicate
SpinelSilicate
SpinelSilicate
Figure 410
140
- Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip1
-
- Chapter 3 Oxidative Promoted Dissolution of Chromite by
- Manganese Dioxide and Concurrent Production of Ch
-
- Chapter 4 Chromium Geochemistry of Serpentine S
-
- List of References
-
- Chapter 2
- Chapter 3
-
- Chapter 4
-
- Appendix A
- Chapter 2
-
- Chapter 3
-
- Chapter 4
-
- Appendix A
-
- CHAP1pdf
-
- Toxicity and Regulations of Chromium
- Mineralogy of Chromium in Serpentinites
-
- Chapter Summaries
-
- CHAP2Tablepdf
-
- JR305CLAY Depth 0-5cm
- JR3515CLAY Depth 5-15cm
- JR33045CLAY Depth 30-45cm
-
- HCAP2Figspdf
-
- Figure 27
-
- CHAP3Tablepdf
-
- Figure 36
- Figure 37
-
- CHAP4pdf
-
- ABSTRACT
- REVIEW
-
- Chromium Geochemistry
-
- Soil Organic Matter
-
- CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
-
- BedrockProtolith Mineralogy
- Soil Chemistry and Physical Properties
- Soil Mineralogy
- Backscattered electron images and microprobe analyses were used to identify the Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49 and corresponding Cr-bearing mineral compositions are listed in Table 46 The New C
-
- DISCUSSION
-
- Chromium Spinel in Soils and Rocks
-
- CONCLUDING STATEMENT
- ACKNOWLEDGEMENTS
-
- CHAP4Tablepdf
-
- Ulvoumlspinel
-
chromaterdquo to Environmental Science and Technology The co-authors for this paper are
Scott Fendorf and Dennis Bird
Chapter 4 is a review paper that critically evaluates the chemistry of serpentine
soils and serpentinites primarily focusing on Cr geochemistry Additionally analytical
studies of serpentine soils collected from California Oregon and New Caledonia are
presented This study will be presented at the Robert Coleman Symposium and has been
submitted as an article entitled ldquoChromium geochemistry of serpentine soilsrdquo for an
International Geology Review Special Publication Scott Fendorf Dennis Bird and
Robert Coleman are the second third and fourth co-authors respectively
Appendix A is a preliminary investigation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite) a mineral common in serpentinites and present in serpentine
soils This appendix is entitled ldquoEstimation of the thermodynamic properties of Cr-
clinochlore (kaumlmmererite)rdquo The results of this chapter will be included in a study that
estimates the thermodynamic properties of Cr-bearing silicates including fuchsite (Cr-
muscovite) tawmawite (Cr-epidote) and uvarovite (Cr-garnet)
viii
ACKNOWLEDGEMENTS
Serpentinites serpentine soils and chromium are the Geologic Trinity of this
dissertationhellipeach topic is intimately related to one another The scope and breadth of
this research required a significant amount of time and funding None of this research
would have been possible without the generous support of several foundations
organizations and Stanford University I am grateful for the financial aid I have
received from the Stanford Fellowship the Stanford University McGee and Shell funds
the National Science Foundation the Geologic Society of America Graduate Research
Fund the Stanford Office of Teaching and Learning grant and the Jasper Ridge
Biological Preserve Melon Grant
I would not be where am I today if it were not for my parents Robert and Sue
Oze and my brother Rob Oze Thank you for the sandbox and all the hose-driven
volcanoeshellipthose Wookies never stood a chance While some families vacation in theme
parks my family frequented gem and mineral conventions collected rocks and shells
from mist-shrouded shores and sought lucky stones including an infamous quartz orb
Sometimes I think I was destined to be a geologist I would also like to thank everyone
in my family and all our family friends who have provided that extra little push and
support along the way These people include Tudor Bennatts Louis Oze Margaret and
John Chambers John Bennatts Nancy Skiles the Izumis the memory of Witchie Poo
hijacking Harleys from Bremerton bikers and the Man in the Sound waiting for that one
last catch Close friends I would like to acknowledge who are not related to the Stanford
crowd include Dave and Leigh Anderson and Rebecca Parker These satellite friends
helped me keep in touch with the world outside of academia and proved to me that I
wasnrsquot meant to have a real job A special thank you goes to Katherine Akako
Izumihellipyou have made this whole trip worthwhile
My PhD advisors Scott Fendorf and Dennis Bird provided an invaluable amount
of help patience and energy over the past five years Scott always made the time to
discuss and refine many of my projects making them relevant to many current
geochemical issues Additionally he helped me hone my soil science skills in the field
and in the lab Out in the field Scott can dig a pretty mean hole Dennis devoted a
significant amount of time and sweat in order to make this dissertation happen His
ix
insightful knowledge of geochemistry and of the geology of the Sierra Nevadas provided
the cornerstone for this dissertation
Robert Coleman played a key role in the production of this dissertation He
always knew what to say at the right time and helped me understand the complexity of
ophiolites and serpentinites through experienced eyes Additionally Keith Loague and
Gary Ernst deserve recognition for their outstanding support as committee members for
this dissertation over the last four years I would also like to thank Jonathan Stebbins and
Gail Mahood for providing me the opportunity to attend Stanford
Bob and Vicki Jones always offered an open ear whether it was science or solving
the worldrsquos problems Rick Hazlett (Pomona College) and John Winter (Whitman
College) both had a significant impact on shaping my future science goals and objectives
during my undergrad years and both encouraged me to pursue a career in geology They
were the catalyst for this whole experiment
I owe a lot to my research group members (both the Bird Cage and Soil
Chemistry Group) and those people who didnrsquot mind me tagging along Before I
introduce these people one person has been my friend and co-conspirator for many years
at Stanford Tim Copehellipyou are a pain in my ass Besides never picking up any of your
crap and breaking all my stuff this whole grad school experience would not have been
the same without you Other friends and acquaintances that will not be forgotten in
chronological order of me walking into my office for the first time on September 11
1998 include Anne Dyreborg Esra Inan Tom Bawden Patrick Redmond Phil
Neuhoff Thrainn Fredrickson Lara Heister Kaye Savage Andrea Fildani Mike Schultz
Trayle Kushan Kevin Theissen Angela Hessler Chris Van de Ven Kono Lemke Colin
Doyle Colleen Hansel Ben Bostick Matthew LaForce Nancy Grumet Scott Kroeker
Anders Meibom Brian Ebel Matt Ginder-Vogel Matt Pollizitto Spud Amy Weislogel
Travis Horton and Jake Waldbauer I would like to end this Acknowledgement section
and begin my dissertation with a haikuhellip
Better than outcrops
Are the people Irsquove met here
Thanks for all the beer
x
TABLE OF CONTENTS
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iv Prefacehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip vii Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip ix Table of Contentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xi List of Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xiv List of Illustrationshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip xvi Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 1
Toxicity and Regulations of Chromiumhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 2 Mineralogy of Chromium in Serpentiniteshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 3 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5 Chapter Summarieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 5
Chapter 2 Chromium Geochemistry of Serpentinized Ultramafic Rocks and Serpentine Soils in the Franciscan Complex of Californiahelliphelliphelliphelliphelliphellip 8
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 8 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 9 Backgroundhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 10
Geologic Background of the Serpentinite at Jasper Ridgehelliphelliphelliphellip 10 Serpentine Soils Characteristics and Reviewhelliphelliphelliphelliphelliphelliphelliphellip 11 Sources of Chromium ProtolithBedrockhelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 12 Chromium-Bearing Phases in Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 13
Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Location and Sample Collectionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Mineral Identification in Rocks and Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 14 Soil Chemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 15 X-ray Diffraction and Clay-Size Mineral Identificationhelliphelliphelliphelliphellip 16 X-ray Absorption Spectroscopyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17 Sequential Extraction Procedurehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 17
Resultshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Serpentinite Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 18 Distribution of Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip21 Serpentine Soil Chemistry and Mineralogy at Site JR3helliphelliphelliphelliphellip 22 Sequential Extractionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 26
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 27 Concluding Remarkshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 32 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 33 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 44 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 47
xi
Chapter 3 Oxidative Promoted Dissolution of Chromite by Manganese Dioxide and Concurrent Production of Chromatehelliphelliphelliphelliphelliphellip 63
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 63 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 64 Materials and Methodshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 65
Solid Phase Characterization and Synthesishelliphelliphelliphelliphelliphelliphelliphelliphellip 65 Extraction Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 66 Batch Experiment Procedureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 67 Solution Analysishelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69
Results helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite Extraction Experimentshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 69 Chromite and Birnessite Batch Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 70 Secondary and Solid Phase Examination of Chromite and Birnessitehelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 72 Serpentine Soil and Birnessite Experimentshelliphelliphelliphelliphelliphelliphelliphelliphellip 72
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 73 Chromium(VI) Pathways from Chromite and Birnessite into Solutionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 76 Serpentine Soils and Cr(VI)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 77
Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 79 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 80 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 84 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 87
Chapter 4 Chromium Geochemistry of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphellip100
Abstracthelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 100 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 101 Reviewhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102
Chromium Geochemistryhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 102 Mineralogy of Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 103 Sources of Chromium for Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphellip 105 Chromium in Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 106 Soils Solutions and Related Waterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 108 Serpentine Soil Vegetationhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 110 Soil Organic Matterhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 111
California Oregon and New Caledonia Serpentine Soilshelliphelliphelliphelliphelliphelliphellip 112 Sample Locations and General Soil Characteristicshelliphelliphelliphelliphelliphellip 112 BedrockProtolith Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 113 Soil Chemistry and Physical Propertieshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 114 Soil Mineralogyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116
Discussionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 116 Chromium and Serpentine Soilshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 117
xii
Chromium Spinel in Soils and Rockshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 119 Concluding Statementhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 120 Acknowledgementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 121 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 122 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 129 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 131
Appendiceshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 141 Appendix Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 142 Appendix Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 143 Appendix A Estimation of the Thermodynamic Properties of Kaumlmmererite (Cr-Clinochlore)helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip145
Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 145 Nomenclature and Crystal Chemistry of Cr-chloritehelliphelliphelliphelliphelliphelliphelliphelliphellip 146 Estimation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 147
Estimation of theStandard Molar Volume Using Unit Cell Measurementshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of the Standard Molar Volume Heat Capacity and Entropyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 148 Estimation of Standard Molar Enthalpy and Gibbs Free Energyhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 150 Critique and Discussion of the Estimation Algorithms and SUPCRT92helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 151
Evaluation of the Thermodynamic Properties of Kaumlmmereritehelliphelliphelliphelliphellip 152 Tableshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 155 Figure Captionshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 161 Figureshelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 162
List of References
References to Chapter 1helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 166 References to Chapter 2helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 172 References to Chapter 3helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 179 References to Chapter 4helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 183 References to Appendix Ahelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip 195
xiii
LIST OF TABLES Chapter 2 Table 21 Whole rock elemental concentrations of serpentinites chert and sandstone Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in
serpentinite rock at Jasper Ridge Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in
serpentine soil at Jasper Ridge Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25
and 26 Table 24 Bulk soil elemental concentrations for soils and bedrock collected along
the transect in Figure 21b Table 25 Chemical and physical properties of the soil at Site JR3 Table 26 Selected trace metal concentrations for each size fraction versus depth at
Site JR3 Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3 Table 28 X-ray diffraction d-spacings for common silicates Table 29 Results of the Diablo clay serpentine soil and chromite sequential
extraction experiments Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1
kg of sample Chapter 3 Table 31 Chromite and birnessite masses and concentrations used for each batch
experiment Table 32 Serpentine soil and birnessite experimental parameters Table 33 Summary of Cr(VI) rates obtained from batch experiments Table 34 Comparative analyses of batch experiment solutions and chemical
extractions of chromite and birnessite solids Chapter 4 Table 41 Measured Cr concentrations in serpentine soils worldwide Table 42 Summary of Cr Fe and Ni extractions from serpentine soils Table 43 Serpentine soil plant Cr concentrations Table 44 Chemical and textural properties of collected soils Table 45 Total major and trace element concentrations of the collected soils Table 46 Compositions for soil Cr-containing minerals identified in Figure 49 Table 47 Chromium-bearing mineral masses and volumes required to influence Cr
concentrations in soils Appendix A Table A1 Cr and Al end-members in the Cr-clinochlore solid solution
xiv
Table A2 Standard state thermodynamic properties of Cr and Al-bearing minerals Table A3 Unit cell measurements and molar volumes of clinochlore and Cr-
clinochlores Table A4 Entropy volume and heat capacity for kaumlmmererite using the Ransom and
Helgeson Method Table A5 Fictive polyhedral considered in the estimations Table A6 Estimated thermodynamic properties for kaumlmmererite
xv
LIST OF FIGURES Chapter 2 Figure 21a Soils map of Jasper Ridge Biological Preserve CA Figure 21b Serpentinite and serpentine soil sampling sites at Jasper Ridge Biological
Preserve CA Figure 22 Chromium concentrations versus depth for four serpentine soils in
Maryland California and New Foundland Figure 23 Petrographic images of serpentinite at Jasper Ridge Figure 24 Elemental concentrations relative to Cr for chlorite pyroxene magnetite
chromite and a chromite-silicate mixture in the serpentinite at Jasper Ridge
Figure 25 Backscattering electron images of minerals in the serpentinite at Jasper Ridge
Figure 26 Backscattered electron images of Cr-chlorite and Cr-magnetite assemblage in the serpentinite bedrock
Figure 27 Highest and lowest Cr concentrations at each site in the serpentine soil transect
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected for total gt2 mm sand silt and clay size fractions
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soils at Jasper Ridge
Figure 210 X-ray diffraction clay analyses of sample JR3515 Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at specific depths Figure 212 X-ray microprobe elemental maps of Cr and Fe for the serpentine soil and
serpentinite at site JR3 Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments Figure 214 Scanning electron microprobe images of the chromite standard after each
sequential extraction experiment Figure 215 Tie-lines connecting serpentinite chromite and end-member chromite with
soil chromite values Chapter 3 Figure 31 Scanning electron microprobe images of chromite and birnessite standards Figure 32 Chromite extraction experiments evaluating Cr Fe and Al Figure 33 pH and Eh versus time (hours) for each chromite-birnessite chromite and
birnessite batch experiments Figure 34 pH and Eh with Cr stabilities Figure 35 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant chromite and variable birnessite at pH 5 Figure 36 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for constant birnessite and variable chromite at pH 5
xvi
Figure 37 Concentrations of Cr Cr(VI) Fe Fe(II) Mn and Mg as a function of time
for variable pH and constant chromite and birnessite Figure 38 Rate of Cr(VI) production (d[Cr(VI)]dt) as a function of the surface area
concentration of chromite birnessite and pH Figure 39 SEM examination of solids following the termination of the batch
experiments Figure 310 Serpentine soil and birnessite experiments Figure 311 Iron analyses and the solubility of amorphous Fe(OH)3 and goethite Figure 312 Congruent dissolution of chromite based on Fe and Cr Figure 313 Three pathways of Cr release and oxidation Chapter 4 Figure 41 Serpentine soil localities worldwide Figure 42 Field site pictures of serpentine soils Figure 43 Stability field diagram of Cr over a range of Eh and pH conditions Figure 44 Chromium and Al Wt plot and octahedral Cr and octahedral Al plots
for chromite and Cr-silicates including Cr mica (fuchsitemariposite) Cr-garnet (uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite)
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (m) for selected serpentine soils
Figure 46 pH of serpentine soils and related groundwater worldwide Figure 47 Soil pictures with Munsel color values and names from collected soils Figure 48 Cr XANES spectra for collected serpentine soils Figure 49 Backscattering images of mineral phases within each serpentine soil Figure 410 Chromite magnetite and hematite compositions from analyzed soils as
well as analyses for chromite reported from around the world Appendix A Figure A1 Compositional variation of Cr2O3 and Al2O3 Wt in Cr-chlorite Figure A2 Compositional variation of Cr and VIAl in Cr-chlorite based on 14 oxygens Figure A3 Molar volume estimation of kaumlmmererite based on XRD unit-cell
parameters Figure A4 Natural and calculated compositions of co-existing Cr-clinochlore and Cr-
spinel solid solution in the system Mg0-FeO-Cr2O3-Al2O3-SiO2-H2O at 2 kbar pressure
xvii
CHAPTER 1
INTRODUCTION
Serpentinites or metamorphosed ultramafic rocks are distributed worldwide and
are commonly associated with ophiolite complexes (Coleman 1977 Coleman and Jove
1992) These rocks weather to form serpentine soils a generic term used to describe any
soil derived from serpentinite regardless of its physical or chemical properties
Serpentinites and serpentine soils contain Cr-bearing minerals the most prevalent being
chromite and Cr released from these minerals into surface and subsurface environments
is potentially toxic and carcinogenic The main objective of this dissertation is to
evaluate the mineralogy and chemistry of Cr related to serpentinites and serpentine soils
over geologic time using an integrated analytical experimental and theoretical approach
Additionally the potential of serpentinites and serpentine soils as sources of non-
anthropogenic Cr contamination is investigated A discussion of Cr and its significance
to serpentinites and serpentine soils begins with its discovery by a Frenchman
investigating an ore named ldquored leadrdquo from the Urals
Nicolas-Louis Vauquelin discovered Cr in 1797 in the mineral crocoite PbCrO4
(Burns and Burns 1975) The metalelement was named chromium after the Greek word
chroma meaning color due to its strong influence on mineral pigmentation (Burns and
Burns 1975) For example rubies and emeralds derive their color from trace amounts of
Cr in their structures More importantly JT Lowitz in 1798 discovered Cr in chromite
or Cr-spinel the primary source of Cr for industrial purposes (Burns and Burns 1975)
Chromium is a group VIB transition metal with two common oxidation states (Cr(III) and
Cr(VI)) an average crustal abundance of 100 mg kg-1 and an abundance of 513x105 in
the solar system (relative to [H]=1x1012) (Mason and Moore 1982)
The majority of Cr on the earth exists as Cr(III) where it is an immobile cation
that complexes strongly with organic matter and adsorbs on oxides and silicates even at
low pHs (pHgt4) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) Additionally
the chemical mobility of Cr(III) is limited due to the formation of highly stable
(oxy)hydroxides (Rai and Zachara 1988 Rai et al 1989 Ball and Nordstrom 1998)
1
Chromium(III) has the highest octrahedral site preference energy (OSPE) of any trace
element and it isomorphically substitutes into octahedral sites in oxides
(oxy)hydroxides sulfides and silicates Although Cr(III) is known to substitute for both
octahedral Al(III) and Fe(III) the strongest correlation for substitution appears to be for
Al(III) in oxides and silicates as demonstrated in Chapters 2 and 4 and Appendix A
Chromium(VI) is a strongly oxidizing and weakly adsorbing oxyanion in the form of
chromate (HxCrO42-x) (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995 Shiraki
1997 Ball and Nordstrom 1998) Consequently Cr(VI) is chemically mobile and
bioavailable over a wide range of pH (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Chromium(VI) is tetrahedrally coordinated and behaves similarly to
sulfate in minerals such as jarosite (Baron and Palmer 1996)
The pathways and chemistry of Cr oxidation and reduction in soils and related
waters are covered in depth by Fendorf (1995) and Ball and Nordstrom (1998) The
reduction of Cr(VI) to Cr(III) may occur by reactions with Fe(II)aq Fe(II)-containing
minerals such as magnetite organic matter H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996) Oxidation of Cr(III)
to Cr(VI) related to serpentinite-related water and serpentine soil solutions is still a matter
of conjecture Chromium(III) oxidation in serpentine soils has been hypothesized to be
the result of Mn oxides such as birnessite as suggested by both Gough et al (1989) and
Becquer et al (2003) This idea is explored in depth in Chapter 3 where birnessite is
shown to influence the release and oxidation of Cr from chromite over a range of pH (3
to 8) at 25oC
Toxicity and Regulations of Chromium
The examination of Cr based on its oxidation state is necessary due to the
juxtaposing behavior and toxicity of Cr(III) and Cr(VI) Chromium(III) is an essential
micronutrient in human nutrition used by the body to process sugar protein and fat
(Vincent 2000) Inhalation of Cr(VI) is known to cause irritation in the nose including
symptoms such as nosebleeds (epistaxis) and ulcers andor perforations in the nasal
septum (Katz and Salem 1993 Kuo et al 1997) Ingesting considerable quantities of
Cr(VI) can result in stomach ulcers convulsions kidney and liver damage and death
2
(Katz and Salem 1993) Skin contact with certain Cr(VI) compounds causes allergic
contact dermatitis (ulcers) and irritant dermatitis (Katz and Salem 1993) Chromate
induced ulcers or chromeholes are typically red swollen crusted painless lesions
(Katz and Salem 1993) The nosology of Cr is beyond the scope of this dissertation
however Lees (1991) is an excellent source for further information Overall the
Environmental Protection Agency (EPA) the Occupational Safety and Health
Administration (OSHA) the World Health Organization (WHO) and the Department of
Health and Human Services (DHHS) have determined that that Cr(VI) compounds can
increase the risk of lung cancer primarily by inhalation as well as other forms of cancer
caused by ingestion
Federal and state regulation standards appropriate for both Cr(III) and Cr(VI) are
summarized and discussed by Proctor et al (1997) Human health action levels for Cr
vary for each state and the ranges for total Cr in soils include the following 200-66000
Cr(III) mg kg-1 (residential) 02-2000 Cr(VI) mg kg-1 (residential) 2500-1000000
Cr(III) mg kg-1 (non-residential) and 64-71000 Cr(VI) mg kg-1 (non-residential)
(Proctor et al 1997) Incidentally Cr concentrations in serpentine soils and serpentinites
(gt200 Cr mg kg-1) exceed many state action levels and would be considered toxic The
EPA has set a limit of Cr(III)+Cr(VI) at 100 microg L-1 in drinking water The OSHA has
set limits of 500 microg of water soluble Cr(III) compounds per cubic meter of workplace air
or 500 microg m-3 1000 microg m-3 for metallic Cr and insoluble Cr compounds and 52 microg m-3
for Cr(VI) compounds for 8-hour work shifts and 40-hour work weeks The limits
proposed above are constantly changing and should not be used for reference without
checking current regulation standards
Mineralogy of Chromium in Serpentinites
Chromium occurs in a variety of minerals either as a major or minor constituent
(Burns and Burns 1975) Chromium-bearing minerals in serpentinites are primarily
oxides and silicates in which the Cr is present as Cr(III) Chromite is a Cr-bearing spinel
abundant in serpentinites and is resistant to low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) The chemistry and
significance of chromite is addressed throughout this dissertation Pyroxenes such as
3
augite and enstatite are primary minerals identified in serpentinites and contain lt8 Cr Wt
(Grapes 1981 Von Knorring et al 1986 Treloar 1987a Challis et al 1995
Sanchez-Vizcaino et al 1995 Deer et al 1996) Chromate minerals such as crocoite
(PbCrO4) are rare in serpentinites and the occurrence of these minerals are limited to Pb
deposits in Australia Tasmania and the Urals (Williams 1974 Crane et al 2001)
Chromium(III) released from the primary Cr-containing phases may
isomorphically substitute into oxides and silicates for Al(III) and Fe(III) due to size
(octahedral radii Cr3+ = 0615 Aring Al3+ = 061 Aring and Fe3+=0643 Aring) and charge
similarities Magnetite formed during serpentinization incorporates Cr into its structure
and serpentine minerals (antigorite lizardite and chrysotile) contain trace concentrations
of Cr (Spangenberg 1943 Page and Coleman 1967 Burkhard 1993) Eskolaite (Cr2O3)
has been identified in Finland (Von Knorring et al 1986) Metasomatic alteration of
ultramafic rocks commonly results in the formation of Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite) uvarovite (Cr-garnet) tawmawite (Cr-epidote) and
kaumlmmererite (Cr-chlorite) (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Neiva 1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981
Max et al 1983 Nutman et al 1983 Raase et al 1983 Chao et al 1986 Von
Knorring et al 1986 Kerrich et al 1987 Treloar 1987b Treloar 1987a Morand 1990
Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993 Christofides et al 1994
Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Proenza et al
1999) Additionally kyanite and andalusite in the proximity of ophiolite complexes
incorporate up 5 Cr Wt (Cooper 1980 Raase et al 1983 Kerrich et al 1987)
Chapter 4 provides an in depth discussion of these Cr-bearing silicates as well as
demonstrating mineral compositions reported in the literature based on octahedral Cr and
Al Appendix 1 is a preliminary investigation devoted to estimating the thermodynamic
properties of kaumlmmererite These properties can then be incorporated into geochemical
modeling programs to begin accessing the stability of this common serpentinite mineral
over a range of temperatures and pressures
4
Chromium in Serpentine Soils
Evaluating Cr in serpentine soils requires an examination of both the mineralogy
and chemistry of the serpentinite and soil Measured Cr concentrations in serpentinites
and serpentine soils range from ~200 mg kg-1 to 6 Wt all of which is Cr(III) (Soane
and Saunder 1959 Proctor and Woodell 1971 Shewry and Peterson 1976 Jaffre 1980
Proctor et al 1980 Rabenhorst et al 1982 Schwertmann and Latham 1986 Brooks
1987 Gough et al 1989 Gambi 1992 Lee 1992 Morrey et al 1992) In Chapters 2
and 4 chromite and Cr-magnetite are identified as the most prevalent sources of Cr
responsible for Cr-enrichment in these soils Chromite is resistant to chemical
weathering and it has been observed to undergo incongruent dissolution conserving Cr in
serpentine soils Chromium is tightly bound in the matrix of the soil less than 1 of the
total Cr can is removed using standard soil extraction methods (10 hydrogen peroxide
(H2O2) solution 01 M potassium permanganate (KMnO4) solution10 M hydrofluoric
(HF) acid 1 M hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and 1
mM barium chloride (BaCl2) solution) due to the stability of chromite Although silicates
contain minor concentrations of Cr compared to chromite they are perhaps the main
contributors of Cr identified in plants (lt200 Cr mg kg-1) and soil solutions (lt50 microg L-1)
During soil-forming processes Cr released to soil solutions by the dissolution of Cr-
bearing minerals are sequestered into a host of structural and surface sites available in
authigenic Fe(III) oxides Fe(III) hydroxides and clay minerals (Schwertmann and
Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990 Gerth 1990 Gasser and
Dahlgren 1994 Ilton and Veblen 1994 Brigatti et al 2000 Ilton et al 2000) The
clay-sized fractions in soils at Jasper Ridge Biological Preserve CA contain Cr
concentrations as high as 1000 Cr mg kg-1
Chapter Summaries
A detailed field and analytical examination of Cr in serpentinites and serpentine
soils at Jasper Ridge is presented in Chapter 2 A majority of the Cr enrichment and
variability with depth in these soils is the result of small grains of chemically resistant
chromite from the serpentinite protolith Chromite identified in the soil undergoes
incongruent dissolution conserving Cr relative to chromite compositions identified in the
5
serpentinite bedrock Sequential extraction experiments using barium chloride (BaCl2)
hydroxylamine (NH2OHmiddotHCl) with 25 (vv) acetic acid (HOAc) and hydrofluoric (HF)
acid were performed on a chromite standard confirming that Cr-bearing spinels are
resistant to chemical weathering X-ray diffraction analyses of the clay-size fraction in
the serpentine soils identified Cr-clinochlore smectite vermiculite and serpentine with
the smectite being the most abundant mineral in the clay-size fraction Chromium(III)
was the only identified oxidation state of Cr in the serpentine soil solids and in
serpentinites at Jasper Ridge in the solid phase This chapter represents a comprehensive
site assessment identifying the nature and behavior of Cr in serpentine soils with respect
to the bedrock mineralogy and depth
Chromite an abundant Cr-enriched spinel is responsible for a majority of the Cr
enrichment identified in ultramafic rocks serpentinites and serpentine soils The release
and oxidation of Cr from chromite is a pathway to consider when evaluating the potential
of ultramafic rocks and their related sediments and soils as sources of Cr(VI) In
Chapter 3 experiments aimed at quantifying the release and oxidation of Cr from
chromite in the presence of birnessite are described Chromite is extremely resistant to
dissolution (lt8 Cr mg kg-1 is released) when immersed in solutions including 1 mM
barium chloride solution 10 M hydrofluoric acid 10 hydrogen peroxide and 1 M
hydroxylamine hydrochloride with 25 acetic acid Highly oxidizing solutions of 01 M
potassium permanganate increased the rate and release of Cr relative to the other
solutions listed above thus birnessite a high valence Mn oxide present in serpentine
soils was chosen to determine if Mn could accelerate Cr release and oxidation from
chromite The results presented in Chapter 3 demonstrate that birnessite increases the
rate of Cr(VI) observed in solution from chromite This production of Cr(VI) is
dependent on the chromite suspension density where birnessite surface are concentrations
are greater than 20 m2 L-1 between pHs 3 and 8 at ~25degC
A comprehensive review of the chemical characteristics of serpentine soils
including topics such as the mineralogy of serpentine soils soil solutions organic matter
and vegetation is provided within Chapter 4 Additionally chemical and spectroscopic
analyses were performed on serpentine soils from a) New Caledonia b) Eight Dollar
Mountain Selma OR c) Nickel Mountain Riddle OR d) Harvard Mine Jamestown
6
CA e) Jasper Ridge Biological Preserve Stanford CA and f) Pillikin Mine El Dorado
CA Serpentinites and serpentine soils associated with ophiolite complexes cover ~1 of
the earthrsquos exposed surface and they are correlated to modern and ancient convergent
boundaries The physical and chemical characteristics of serpentine soils are directly
related to the alteration and tectonic history of the serpentinite as well as the soil-forming
factors including the parent material climate biota topography and time Serpentine
soils have Cr concentrations exceeding 200 mg kg-1 however the concentration and the
availability of Cr in these soils are not correlated with the five soil-forming factors
(climate organic material parent material topography and time) or the soil chemistry
Although only Cr(III) was identified in the soil solids Cr(VI) has been identified in soil
solutions (lt30 microg L-1) from serpentine soils in New Caledonia and California Chromite
and Cr-magnetite grains are identified in the soils listed above and these minerals are
responsible for a majority of the Cr enrichment in serpentine soils
Kaumlmmererite is a Cr-chlorite associated with serpentinized ultramafic rocks such
as those collected at Jasper Ridge Harvard Mine and Pillikin Mine Although Cr
concentrations are typically 01-2 Wt in kaumlmmererite it is modally abundant in
serpentinites (up to 5 of the volume of the rock) and a potential source of Cr to be
released into soils and groundwater In Appendix A the standard molar thermodynamic
properties of the Cr-clinochlore end-member kaumlmmererite (Mg5Cr(AlSi3)O10(OH)8) is
estimated using established estimation algorithms This initial study provides the
foundation for submitting a future paper presenting the thermodynamic properties for
major Cr-bearing silicates including fuchsite (Cr-muscovite) uvarovite (Cr-garnet) and
tawmawite (Cr-epidote) and their metamorphic significance related to serpentinized
ultramafic rocks
7
CHAPTER 2
CHROMIUM GEOCHEMISTRY IN SERPENTINIZED ULTRAMAFIC ROCKS
AND SERPENTINE SOILS FROM THE FRANCISCAN COMPLEX OF
CALIFORNIA
ABSTRACT
Weathering of ultramafic rocks and serpentinites in the Franciscan Complex of California
produces serpentine soils containing high concentrations of Cr as well as other potentially
toxic elements including Ni Co and Mn Chromium concentrations in serpentine soils
from Jasper Ridge Biological Preserve in the Central Coast Range are as high as 4760
mg kg-1 nearly three times greater than the serpentinite protolith (~1800 mg kg-1)
Chromium-containing minerals within the bedrock include chlorite (~03 Cr Wt )
enstatite (~04 Cr Wt ) augite (~07 Cr Wt ) chromite (~108 Cr Wt ) magnetite
(82-103 Cr Wt ) and an ultra fine-grained mixture of spinel and a silicate phase
containing ~133 Cr Wt Chromium concentrations in serpentine soil profiles
fluctuate between 1725 to 4760 mg kg-1 and do not correspond to variations in soil pH
organic matter or electrical conductivity The enrichment and variability of soil Cr is
directly related to the modal abundance and weathering of chromite Cr-magnetite and
the spinel-silicate mixture By comparison Cr silicates account for lt10 of the total soil
Cr Chemical analyses and X-ray microprobe maps demonstrate that Cr-spinels in these
soils undergo incongruent dissolution progressively enriching the spinel toward a Cr-
enriched end-member (FeCr2O4) Chromium occurs in the trivalent state in both the rock
and soil samples The apparent resistance of Cr-spinels to weathering evident from
extraction experiments suggests that these minerals are not large inputs for Cr in soil
solutions and vegetation associated with serpentine soils Chromium-bearing igneous
and metamorphic silicates in the protolith and Cr-bearing clay minerals in the soil are
more likely sources of chemically mobile and bioavailable Cr
8
INTRODUCTION
Serpentine soils are formed from the weathering of ultramafic rocks and their
metamorphic equivalents (serpentinites Brooks 1987) Measured Cr concentrations in
these soils range as high as 80600 mg kg-1 and are enriched compared to non-serpentine
soils (7 to 221 mg kg-1 Proctor 1992 McBride 1994) The distribution and
concentration of Cr in serpentine soils worldwide is highly variable due to the chemical
composition and mineralogy of the parent material and soil forming factors including
climate biota topography precipitation and time (Hotz 1964 Rabenhorst et al 1982
Schwertmann and Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al
1989 Gough et al 1989 Kaupenjohann and Wilcke 1995) Potential Cr-bearing phases
in serpentine soils have been assessed using extraction experiments and these phases
include organic material silicates secondary oxides and spinels (Schwertmann and
Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994 Kaupenjohann and
Wilcke 1995 Becquer et al 2003) The weathering of these Cr-bearing phases over
geologic time and how these sources govern the distribution of Cr in serpentine soils have
not been evaluated in great detail
Chromium is a group VIB transition metal present in the environment as Cr(III) a
non-hazardous species and micronutrient andor as Cr(VI) a toxin to living cells and a
Class A human carcinogen by inhalation (Daugherty 1992 Cohen et al 1993 James et
al 1997) Chromium(III) is chemically immobile due to forming low solubility
(oxy)hydroxides and stable surface complexes whereas Cr(VI) is a weakly adsorbing
(oxy)anion and thus is mobile in most aquatic environments (McBride 1994 Fendorf
1995 Ball and Nordstrom 1998) Chromite and Cr-magnetite have been cited as
primary sources of Cr in serpentine soils (Rabenhorst et al 1982 Schwertmann and
Latham 1986 Alexander et al 1989 Gough et al 1989) Although Cr is primarily
found in the form of Cr(III) in both ultramafic rocks and serpentinites Cr(VI) has been
measured in serpentine soils in California and New Caledonia (Gough et al 1989
Becquer et al 2003) Due to the abundance of Cr and the potential for Cr(VI) formation
serpentine soils and serpentinites are potential sources of non-anthropogenic Cr
contamination
9
The objectives of this study are to determine the distribution and chemistry of Cr
in naturally Cr-enriched soils using an integrated field analytical experimental and
theoretical approach A wide array of techniques and methods including X-ray
absorption spectroscopy and chemical extractions were used to identify the nature of Cr
and its availability in serpentine soils Serpentinites and serpentine soils at Jasper Ridge
Biological Preserve located near Stanford CA were used to evaluate the geochemical
behavior of Cr These soils cover approximately 50000 m2 and contain high
concentrations of the trace elements Cr Ni and Mn The soils are dominantly Mollisols
with grassland vegetation and they are pedalogically similar to those associated with the
Central Coast Range of the Franciscan Complex in the western United States The main
serpentinite body at Jasper Ridge was exposed approximately 400000 years ago allowing
pedogenesis and Cr cycling to occur over a geologically relevant time period (Burgmann
et al 1994) These soils can then be used as a proxy for evaluating the long-term
behavior of Cr
BACKGROUND
Geologic Background of the Serpentinite at Jasper Ridge
Jasper Ridge Biological Preserve is located in the foothills of the Santa Cruz
Mountains in California and is part of the Central Coast Range of the Franciscan
Complex an Early Cretaceous accretionary meacutelange containing blocks of greywacke
greenstone blueschist and serpentinized ophiolite (Figure 21) Formation
metamorphism and emplacement of serpentinite at Jasper Ridge began approximately
130-150 million years ago during Jurassic subduction along the California continental
margin (Page and Tabor 1967) Thin wedges of ultramafic rock (peridotite) were
detached from the down-going slab of oceanic crust and incorporated into the subduction
meacutelange Subsequent serpentinization of the ultramafic rock in the down-going slab
resulted in the formation of serpentinites now exposed at Jasper Ridge
Rare blueschist blocks (1 to 5 meters in width) present in the serpentinite at Jasper
Ridge attest to the degree of metamorphism and the depths from which portions of the
tectonic meacutelange has been elevated Transformation of greenstone in the meacutelange into
blueschist required high pressures in excess of ~5-7 kbars (15-20 kilometers depth) and
10
temperatures of 200-300 degC (Ernst 1971) The low density of the serpentine (245-250 g
cm-3) allowed upward diapiric migration of the serpentinite body through the denser
enclosing tectonic meacutelange (265 g cm-3) Uplift of the present day Coast Range and
activity associated with the San Andreas fault system have shaped the current topography
and exposure of serpentinite at Jasper Ridge Serpentinite is currently present as a narrow
discontinuous band of pervasively sheared rock located along the ridge top (Figure 21)
The serpentinite is in fault contact with the Whiskey Hill Formation (Cretaceous turbidic
sandstone and mudstone) and tectonically interleaved with the Franciscan Complex
Serpentine Soils Characteristics and Review
Relative to other soil types serpentine soils are characterized by higher
concentrations of Cr Ni Co and Fe lower concentrations of plant nutrients such as Ca
K N and P lower CaMg ratios and characteristic flora and physical properties (Brooks
1987) The pH of serpentine soils ranges from ~4 up to 9 (Brooks 1987 Cole 1992
Proctor 1992) Serpentine soil solutions obtained using tension-free lysimetry have Cr
concentrations between 01 and 32 micromol L-1 in which a majority of the Cr is present as
colloidal material (Gasser and Dahlgren 1994) The inhabitability of these soils for most
plants has been attributed to the imbalance of Ca to Mg (Walker 1954) the deficiency of
plant nutrients (Turitzin 1991) and the elevated concentration of heavy metals
(Robinson et al 1935) however linking the lack of plant productivity of these soils to
toxic levels of Cr has not been fully resolved (Brooks 1987 Kruckeberg 1992) The
unique flora and fauna associated with these soils have been well-documented and Cr
concentrations in the plants range as high as 600 mg kg-1 (Brooks 1987 Gough et al
1989)
Chromium concentrations as a function of depth for serpentine soils in Maryland
(Rabenhorst et al 1982) Tehama County CA (Gough et al 1989) the Klamath
Mountains CA (Hotz 1964) and New Foundland Canada (Roberts 1992) are
illustrated in Figure 22 Chromium concentrations with respect to depth at each location
are highly variable demonstrating both maximums and minimums The concentration of
Cr varies significantly at each location due to the chemical composition of the parent
11
material and other soil-forming factors nevertheless Cr concentrations typically decrease
with depth towards the bedrock
A variety of chemical extractions on serpentine soils suggest a broad host of Cr-
containing soil phases including organic material silicates secondary oxides and spinels
(Schwertmann and Latham 1986 Alexander et al 1989 Gasser and Dahlgren 1994
Kaupenjohann and Wilcke 1995) The maximum concentration of Cr extracted from a
serpentine soil reported in the literature is 226 Wt Cr from the New Caledonia laterite
using a 03 M dithionite-citrate bicarbonate (DCB) solution (Schwertmann and Latham
1986) Li and Fendorf (2000) performed extractions on the serpentine soils at Jasper
Ridge using organic acids (citric and oxalic acids) in which a 100 microM citric acid solution
was only able to remove 23 Cr microgg-1 (08 Wt Cr) from a soil containing ~3000 Cr mg
kg-1 This suggests a majority of Cr is bound in phases that are not reactive with these
solvents
Sources of Chromium ProtolithBedrock
Chromium primarily occurs within oxide and silicate phases in ultramafic rocks
and serpentinites Chromite is a common primary (igneous) Cr-bearing spinel in mafic
and ultramafic rocks and often survives low-grade metamorphic processes related to
serpentinization (Hoffman and Walker 1978 Malpas 1992) Isomorphic substitution of
Al3+ Fe3+ and Ti4+ into the octahedral site and Mg2+ Ni2+ Zn2+ and Mn2+ into the
tetrahedral site is common (Sack and Ghiorso 1991) however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987 Pan and Fleet 1991 Burkhard 1993 Christofides et al 1994 Challis et
al 1995 Sanchez-Vizcaino et al 1995 Deer et al 1996 Mathiesen 1999) Pyroxenes
such as augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) are primary minerals in
ultramafic rocks that incorporate varying concentrations of Cr into their structures (Deer
et al 1996)
Serpentinization is associated with the partial oxidation of Fe(II) present in
olivine and pyroxene forming magnetite (Brooks 1987) Chromium(III) derived from
primary chromite and pyroxene may isomorphically substitute into the Fe(III) site of
magnetite due to size (octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge
12
similarities Hydrothermal and CO2 metasomatism of ultramafic rocks may produce Cr-
silicate minerals including fuchsitemariposite (Cr-mica lt17 Cr Wt ) uvarovite (Cr-
garnet lt20 Cr Wt ) tawmawite (Cr-epidote lt10 Cr Wt ) and kaumlmmererite (Cr-
chlorite lt6 Cr Wt Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974
Phillips et al 1980 Schreyer et al 1981 Max et al 1983 Kerrich et al 1987 Treloar
1987 Pan and Fleet 1991 Christofides et al 1994 Sanchez-Vizcaino et al 1995)
These minerals common in altered ultramafic rocks have varying concentrations of Cr
that isomorphically substitute for octahedral Al in the silicate structure due to similarities
in charge octahedral preference and octahedral radii (Al3+ = 061 Aring) Any of these
silicates and magnetite are potential sources for Cr in soils produced by the weathering of
serpentinite and ultramafic rocks
Chromium-Bearing Phases in Soils
Soil minerals specifically Fe(III) (oxy)hydroxides and clays provide a host of
structural and surface sites that may sequester Cr(III) released to soil solution by
weathering of Cr-bearing phases in serpentinite and ultramafic rocks Isomorphic
substitution of Cr(III) for Fe(III) in oxides (oxy)hydroxides and amorphous compounds
is expected based on crystallographic considerations synthesis experiments and analyses
of selective dissolution of Fe(III) phases (Schwertmann et al 1989 Gerth 1990)
Additionally Cr(III) readily sorbs on surfaces of Fe(III)-containing oxides such as
magnetite and hematite via inner-sphere complexation and forms low-solubility
precipitates at pH values greater than 4 (Charlet and Manceau 1992 Peterson 1996)
Octahedral substitution of Cr(III) for Al(III) in clays (smectites and vermiculites) is
possible based on size and charge similarities Sorption of Cr(III) on edge and interlayer
sites of clay minerals has also been observed by Ilton and Veblen (1994) Soil organic
matter (SOM) is another possible host for soil Cr in which Cr can be chelated into stable
organomineral and organometallic complexes (Kaupenjohann and Wilcke 1995 Brady
and Weil 1999)
13
MATERIALS AND METHODS
Location and Sample Collection
Jasper Ridge Biological Preserve is a microcosm of accreted terranes located in
the foothills of the Santa Cruz Mountains CA Figure 21a is a soils map reported by
Kashiwagi (1985) depicting the alluvial Tertiary sandstone and Franciscan Complex
soils present at Jasper Ridge The Franciscan Complex soils are divided into two
subgroups those derived from greenstone-graywacke and those from serpentine The
serpentine soils dominantly Mollisols and serpentinite bedrock are present as a narrow
discontinuous band located along the ridge top they do not receive sediment from the
adjacent non-serpentine soils Soils and rocks were collected at Jasper Ridge from 1999
to 2002 and sample locations are shown in Figure 21b A 500 meter transect consisting
of 10 sites extends across the length of the serpentine soil body At each site along the
transect soils and bedrock were collected with respect to depth Site 1 (Diablo clay) is
not derived from serpentinite but rather from chert and greenstone Additionally Site 1
is located down slope of the main serpentinite body and has likely received sediment
from the adjacent serpentine soil Rocks were also collected at several exposed outcrops
located in the serpentinite body
Site JR3 (Figure 21b) was chosen for a detailed geochemical study to identify
factors affecting the geochemistry of Cr in serpentine soils Soil at JR3 is a loamy
magnesic thermic Lithic Haploxeroll The thickness of the soil is approximately 50 cm
deep Grass dominantly Lolium multiflorum comprises 95 of the vegetation at this site
and trees and shrubs are not present
Mineral Identification in Rocks and Soils
Mineral identification was accomplished by electron probe microanalysis on an
automated JEOL 733A electron microprobe operated at 15 kV accelerating potential and
15 nA beam current Calibration was conducted using natural geologic standards Beam
width for analysis was 10 microm Raw counts were collected over 20 s and were constant
with time indicating that elemental drift was negligible Detection limits for probe
analyses are ~002 oxide Wt Additionally backscattering electron (BSE) images of
the rocks and soils were obtained using the microprobe
14
Carbon-coated petrographic thin sections of rock samples were used for
microprobe analyses Soil samples were dried in an oven at 60degC for 24 hours mounted
onto a glass slide using epoxy and heated to 100degC until the epoxy cured An initial cut
of each section using a petrographic saw was made followed by a second infusion of
epoxy to ensure total impregnation Finally each section was cut polished and carbon-
coated for electron microprobe measurements
Soil Chemistry
Soils were analyzed for pH electrical conductivity (EC (deciSiemensm))
particle size ammonium acetate extractable cations organic carbon total nitrogen and
elemental composition The pH was determined using a 11 mixture of soil to deionized
water Electrical conductivity (EC) was determined on 2 mL of a soil solution extract
Particle size of the soil was measured using the hydrometer method as described by Gee
and Bauder (1986) These analyses were performed at the Utah State University
Analytical Laboratory
Ammonium acetate was used to measure the concentration of extractable K Ca
Mg and Na within the soil The procedure involved weighing 2 g of soil and adding 25
mL of ammonium acetate (20 M NH4OAc) followed by mixing on an orbital shaker for
30 minutes The solution was filtered using a Whatman 42 filter paper (pore size 25
microm) Potassium Ca Mg and Na concentrations were analyzed using atomic absorption
spectrophotometry (AAS)
Organic carbon was measured using the Walkley-Black titration method (Mebius
1960) Approximately 05 g of sediment was weighed and reacted with 5 mL of 20 M
potassium chromate (K2Cr2O7) followed by the rapid addition of 10 mL concentrated
sulfuric (H2SO4) acid The bottle was cooled for 30 minutes at room temperature and
diluted with 100 mL of deionized water Next 030 mL of 0025 M ortho-
phenanthroline-ferrous solution was added and the solution titrated to the endpoint using
10M ferrous sulfate (FeSO4)
Total nitrogen of soil samples was obtained using an automated combustion
method as described by Sweeny (1989) Approximately 150 mg of air-dried and sieved
15
(40 mesh) soil was placed in a tin foil encapsulating cup A Leco total nitrogen analyzer
with resistance furnace was then used to determine total nitrogen
Major and minor element concentrations for soil and rock samples were obtained
by completely dissolving the samples using a mixture of hot concentrated nitric
perchloric and hydrofluoric acids Elemental concentrations were measured in the
supernatant using inductively coupled atomic emission spectroscopy (ICP-AES)
Standards were plusmn5 mg kg-1 of their known values These analyses were completed at
Chemex Laboratory (Vancouver Canada) Soil samples were dried and sieved (lt2mm)
before analysis Additionally total element concentrations for gt2mm sand silt and clay
size fractions were obtained for each sample collected at Site JR3
X-ray Diffraction and Clay-Size Mineral Identification
X-ray diffraction (XRD) analyses were conducted on the clay-size fraction
(lt2microm) from Site JR3 using a Siemens Diffractometer D5000 at the University of Idaho
The clay-size fraction was isolated using a suspension method described by Vaniman
(2001) Organic matter and calcium carbonate were removed from the clays using a
concentrated hypochlorite solution (Kunze and Dixon 1986) Crystalline and non-
crystalline iron oxides were removed using a citrate-sodium bicarbonate-sodium
dithionite (CBD) solution (Kunze and Dixon 1986) The remaining phyllosilicate
minerals were analyzed using a series of treatments including 1) magnesium-saturation
(2 M MgCl and 2 M MgAOC) and air drying 2) magnesium-saturation and glycolation
(1 glycerol-ethanol solution) 3) potassium-saturation (1 M KCl) and air drying and 4)
potassium (1 M KCl) saturation followed by heating to 500degC (Whittig and Allardice
1986) X-ray diffraction patterns were collected between 2θ values of 20-350deg and 20-
150deg with a generator potential of 30 kV a generator current of 22 mA (using CuKα
radiation) a Ni filter and a scan speed of 1degmin The software package JADE was used
for XRD data analysis and mineral identification
X-ray Absorption Spectroscopy
In-situ fluorescence yield X-ray absorption near-edge structure (XANES)
spectroscopy at the Cr K-edge (edge inflection point is 5989 eV for Cr(0)) was conducted
16
at the Stanford Synchrotron Radiation Laboratory (SSRL) Stanford CA using
beamlines 4-1 and 4-3 A Si(220) monochromator and a Ge-detector were used and
three scans were collected for each XANES analysis Soil and rock samples from Site
JR3 were prepared by removing the gt2mm fraction and crushed to a fine powder
Samples were loaded into a plastic sample holder and covered with Mylar tape
Elemental mapping of three samples were completed at SSRL on Beamline 11-2
using X-ray absorption spectroscopy (XAS) A Si(220) monochromator with a 6 microm
focusing capillary was used and fluorescence intensity was monitored with a Ge detector
Samples were prepared similar to those used for mineral identification using the electron
microprobe however the thickness of the sample was decreased to lt20 microm
Sequential Extraction Procedure
Sequential extraction experiments were completed on 2 g lt2 mm sieved soils
(Diablo clay from Site 1 at 10 cm depth and serpentine soil from Site 2 at 4 cm depth) in
three steps The exchangeable fraction was obtained using a barium chloride (1mM
BaCl2) solution for 1 h at 25 degC (Kunze and Dixon 1986) The Fe-crystalline oxide
fraction was obtained by reacting the solids with a hydroxylamine-hydrochlorideacetic
acid (10 M NH2OHsdotHCl in 25(vv) HOAc) solution for 6 h at 60 degC (Chester and
Hughes 1967) The silicate fraction was obtained by reacting the solids with
concentrated hydrofluoric acid (10 M HF) for 16 h at 25 degC Trace element
concentrations for each extraction were measured on the supernatant using inductively
coupled plasma optical emission spectrophotometry (ICP-OES) All analyses were
completed in triplicate Additionally sequential extraction experiments were completed
on 2g 106-250 mm sieved ultrasonically cleaned commercial grade chromite (MSDS
No 302-1 WHMIS Class D2A) using the same procedures listed above The
stoichiometric composition of chromite obtained using the electron microprobe is
(Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 Chromite was imaged after each sequential
extraction using a JEOL 5600LV environmental SEM equipped with an EDAX thin-
window energy dispersive X-ray detector
17
RESULTS
Serpentinite Mineralogy
The average serpentinite at Jasper Ridge is composed of serpentine (mainly
lizardite and antigorite 70) chlorite (15) talc (10) magnetite (4) and chromite
(1) based on petrographic and microprobe observations of five samples (JR3E JR6E 5
10 and JR3) however brucite has been reported as a major constituent in related
Franciscan Complex serpentinites (Hostetle et al 1966) Other minerals identified
include olivine augite and enstatite Although the serpentinite mineralogy is relatively
simple the modes and textures of the minerals present in the rock vary significantly from
site to site In Table 21 whole rock serpentinite compositions from Sites JR1 JR3E
JR6E 2 5 10 and JR3 are listed The average Cr concentration of the serpentinites is
~1700 mg kg-1 (tab 1) Iron (~7) Mg (~11) Ni (~3000 mg kg-1) and Co (~100 mg
kg-1) concentrations are also elevated in the serpentinite relative to the chert and
sandstone bedrock adjacent the main serpentinite body (Figure 21b and Table 21)
A range of rock and mineral textures are petrographically evident in the
serpentinite at Jasper Ridge (Figure 23) No foliations or lineations are observed in the
serpentinite (Figure 23a) obtained from the bottom of the soil pit excavated at Site JR3
(Figure 21b) Subhedral chromite and magnetite grains are randomly interspersed
amongst the serpentine matrix (Figure 23a) In serpentinite obtained in outcrop at Site
JR5E (Fig 21b) a lsquocrocodile skinrsquo texture is produced by thin irregular veins of fine-
grained subhedral magnetite separating serpentine grains (Figure 23b) In Figure 23c
numerous crosscutting veins of chlorite occur in orthopyroxene and define the foliation of
the rock in outcrop at Site JR3E (Figure 21b) The numerous crosscutting microscopic
veins random foliation patterns (or the lack thereof) and subhedral oxide grains suggest
the Jasper Ridge serpentinite has undergone extensive physical and chemical alteration
typical of most serpentinites
Microprobe analyses of minerals in the serpentinite demonstrate that the main Cr-
bearing phases are chlorite enstatite augite Mg-Al chromite Cr-magnetite and a
chromite-silicate mixture (defined in the following paragraph) Selected compositions
for these minerals are listed in order of their Cr concentrations in Table 22a Chlorite
contains the least Cr whereas the chromite-silicate mixture is the most enriched with
18
respect to Cr Figure 24 shows correlations of Cr content relative to Al Fe Mg and Si in
order to provide insight into the serpentinite mineralogy and to aid in evaluating trends of
Cr substitution
A fine-grained mixture containing spinel (chromite andor Cr-magnetite) and an
unidentifiable Cr-silicate contains the highest Cr concentration in the serpentinite at
Jasper Ridge (Table 22a shown in Figure 25a and 25b) This fine-grained mixture will
be referred to here as the chromite-silicate mixture (CSM) with chemical properties
similar to those in Table 22a and Figure 24 CSM is common around the rims of
protolith chromite grains and a number of the larger (~05 mm) Cr-magnetite grains in
the serpentinite The microcrystalline phase responsible for contributing Si to these
analyses is possibly chlorite due to similarities of false colors generated by backscattering
electron (BSE) images and chlorite commonly rimming CSM Microprobe analyses of
CSM listed in Tables 22 and 23 are considered as mixtures of several mineral phases
due to the fine-grained texture (lt5 microm) Chromium concentrations range from 12 to 16
Cr Wt in this phase CSM is chemically distinct from the chromite and magnetite as
shown in Figure 24
The chemical compositions of protolith chromite exhibit little to no variation in
any of the serpentinites collected at Sites 2 JR3E JR4E JR5E JR6E JR7E and JR3
(Table 22a and Figure 24) Chromite contains low Fe (~9 Wt ) high Mg (~9 Wt
) and high Al (~26 Wt ) compared to Cr-magnetite The average Cr concentration
for chromite is ~12 Cr Wt Additionally chromite grains are irregular and subhedral
suggesting that they have undergone significant alteration during metamorphism
Chromium concentrations in the magnetite vary depending on the proximity and
degree of alteration of the chromite (Table 22a and Figure 24) Magnetite grains with
less than 2 Cr Wt are common in veinlets in serpentinite not formed adjacent
chromite whereas magnetite with 8 to 12 Cr Wt were formed proximal or adjacent
chromite Chromium-enriched magnetite is also referred to as ferrite-chromite in the
literature (Spangenberg 1943 Burkhard 1993)
Chlorite and pyroxenes contain minor concentrations of Cr (lt1 Wt Table
22a) Aluminum Fe and Mg concentrations varied in the pyroxenes however the
majority of chlorites are relatively similar in composition (Figure 24) Chlorites are
19
more abundant in the serpentinite compared to the pyroxenes and both minerals were
subhedral and randomly distributed over the entire sampling area (Figure 21b)
Backscattering electron (BSE) images of what appeared to be a petrographically
opaque single chromite grain in Figure 23a demonstrates chromite has been
metasomatically altered resulting in the formation of Cr-magnetite Cr-chlorite and CSM
(Figure 25a) Represented compositions for these minerals (indicated by letters in Figure
25a) are listed in Table 23 Magnifying the chromite and CSM (Figure 25a)
demonstrates the fine-grained texture of CSM and an alteration rim (transition zone) is
also seen along the chromite edge (Fig 25b) This alteration rim is depleted in Cr with
respect to the chromite as shown in Figure 25b
In Figures 23a and 25c Ni was identified in three phases in the bedrock
including olivine (F 03 Ni Wt ) serpentine (G and H 03 and 0 Ni Wt ) and Ni-
Fe (I sim55 Ni Wt ) alloy (Figure 25c and 25d) their compositions are indicated by
letters and listed in Table 23 Alteration of the olivine resulted in the formation of lt10
microm Ni-Fe alloys shown in Figure 25c These Ni-Fe alloys form a corona surrounding
the remaining olivine and the illustration in Figure 25d highlights the minerals and
chemical changes
Chromite and Cr-magnetite alteration leading to the paragenesis of CSM and Cr-
chlorite is common in the serpentinites at Jasper Ridge The petrographic image in
Figure 26a and its corresponding BSE image in Figure 26b illustrate Cr-magnetite
(originally protolith chromite) partially altering to CSM and Cr-chlorite The chemical
compositions of each mineral (indicated by letters in Figure 26b) are listed in Table 23
CSM appears to marginally and metasomatically replace Cr-magnetite and is enriched by
~2 Cr Wt Chromium-chlorite is present in microcracks in Cr-magnetite and
encompasses the entire Cr-magnetite and CSM grain Serpentine minerals adjacent the
Cr-chlorite contains no measurable Cr
Distribution of Chromium in Serpentine Soils
Serpentine soils and underlying bedrock were collected at ten sites along a 500
meter transect extending the length of the main serpentine body as shown in Figure 21b
20
The range of Cr concentrations at each site (center) and Cr concentrations with respect to
depth (periphery) are shown in Figure 27 and soil analyses are listed in Table 24
Bedrock Cr concentrations for Sites 2 5 and 10 are also shown in these plots (Fig 27)
Chromium enrichment does not appear to be consistently found at a particular depth or
soil horizon for any of the sites Sites 3 9 and 10 demonstrate a Cr maximum at
approximately 006 m however Sites 1 2 5 and 6 exhibit a Cr minimum at roughly the
same depth The highest soil Cr concentration in the soil along the transect is 4760 mg
kg-1 at Site 3 at 006 meters depth The lowest soil Cr concentration is 1725 mg kg-1 at
Site 10 at 0085 meters depth The average bedrock Cr concentration is ~2000 mg kg-1
(Figure 27 Table 24) It is important to note Site 1 consists of Diablo clay not
serpentine soil however this site is located adjacent and down-slope of the main
serpentinite body From these observations it is apparent that there are no systematic
trends in soil Cr concentrations with respect to the position of the site and depth and soil
samples often contain higher concentrations of Cr with respect to their corresponding
bedrock and to the average Cr concentration of ultramafic rocks (1600 mg kg-1 Green
1972)
Trace metals including Ni Mn and Co were also evaluated in the serpentine soil
for Sites 1 through 10 and the results are listed in Table 24 Nickel concentrations range
from 1590 to 3960 mg kg-1 and do not directly correspond to the Cr fluxes with respect
to depth (Table 24) Nickel concentrations are less than Cr from Sites 1 to 4 From Sites
5 to 10 Ni concentrations exceed those of Cr averaging ~3600 mg kg-1 Unlike Cr Ni
concentrations are equivalent or greater in the bedrock (~3300 mg kg-1) as compared to
the soil (~3200 mg kg-1) Soil Mn concentrations at each site are relatively uniform at
~1800 mg kg-1 with little to no variation with depth (Table 24) Cobalt concentrations
do not exceed 238 mg kg-1 (Table 24) and are relatively uniform (~200 mg kg-1) with
respect to location along the transverse and depth at each site
Major elements including Al Fe Ca and Mg are listed for the serpentine soils
along the transect with respect to depth in Table 24 Aluminum Fe Ca and Mg
concentrations in the soils at each site are similar with exception of Site 1 (Diablo clay)
Aluminum concentrations in the Diablo clay are ~46 whereas the serpentine soils
contain less than 2 Al (Table 24) Iron and Mg concentrations are also low lt65 and
21
lt411 respectively in the Diablo clay compared to Fe (gt8) and Mg (gt6)
concentrations in the serpentine soil (Table 24) Calcium concentrations are low (lt1)
for all the serpentine soils and even the Diablo clay (Table 24) With a broad
understanding of the chemical characteristics of the serpentine soil at Jasper Ridge Site
JR3 (Figure 21b) was chosen for a detailed chemical study to examine all the factors
governing the chemistry of Cr in the soil
Serpentine Soil Chemistry and Mineralogy at Site JR3
Soil chemistry
Chemical properties (pH electrical conductivity organic matter nitrogen and
ammonium acetate extractable cations) and textures for four soil samples from Site JR3
are listed in Table 25 The pH of the soil is near neutral (pH from 671 to 698) and
increases slightly with respect to depth Electrical conductivities (EC) in the soil ranges
from 03 to 09 dSM a range indicative of relatively few dissolved salts andor major
dissolved inorganic solutes The organic carbon content of the soil is 32 at the surface
and decreases with depth to 10 near the bedrock Total nitrogen for the samples ranged
from 003 to 064 Ammonium acetate extractable Ca K Mg and Na are listed in
Table 25 The extractable concentration of both Ca and K decrease with depth whereas
there is no apparent trend associated with the concentrations of Mg and Na The
serpentine soil textures are dominantly clay loams however the texture is a sandy clay
loam near the bedrock
Whole soil samples and size fractions including gt2 mm sand (2-002 mm) silt
(002-0002 mm) and clay (lt2 microm) size fractions with respect to depth were chemically
digested and analyzed for total trace metal concentrations shown in Figure 28 and listed
in Table 26 to determine if the abundance of Cr was linked to a particular size fraction
(the percent of sand silt and clay size fractions are reported in Table 25) Additionally
Ni Mn and Co concentrations with respect to depth and size fraction are also presented in
Figure 28 and Table 26 Total and gt2mm size fraction Cr concentrations are similar at
the top and bottom of the soil profile ranging from 3297 to 5976 Cr mg kg-1 Chromium
concentrations exceed detections limits (gt10000 mg kg-1) for the sand-size fraction It is
important to note that the sand-size fraction corresponds to the size of most chromite
22
grains observed in the serpentinite Chromium in the clay fractions are low (~1000 mg
kg-1) compared to the other size fractions however these concentrations are still enriched
compared to non-serpentine soils
In Figure 28 and Table 26 Ni concentrations are enriched in the clay-size
fraction with values ranging from 2840 to 2920 mg kg-1 The silt-size fractions have the
lowest Ni concentrations (~1800 mg kg-1) Manganese concentrations are slightly
elevated relative to total Mn in the sand-size fraction with values ranging from 1728 to
2701 mg kg-1 The clay-size fraction is the least Mn enriched size-fraction with
concentrations at ~1200 mg kg-1 Cobalt concentrations are elevated in the sand-size
fraction with values ranging from 159 to 241 mg kg-1 and concentrations in the gt2mm
silt and clay-size fractions are similar to the total (~150 mg kg-1)
Soil Cr-bearing minerals
Chromium-bearing minerals in the soil identified using the electron microprobe
include chromite Cr-magnetite and CSM Chromium concentrations in soil chromite
Cr-magnetite and CSM are graphically shown in Figure 29 as a function of Al Fe Mg
Si and Al+Fe in Wt Representative analyses of the Cr-bearing soil minerals are listed
in Table 22b Compositions of soil chromite in Figure 29 are chemically distinct from
those identified in the protolith in Figure 24 Chromium concentrations in soil chromites
range from 22 to 37 Cr Wt containing minor concentrations of Al and Mg Values for
chromite in the serpentinite bedrock (Table 22a and Figure 24) and the FeCr2O4 end-
member are also shown in Figure 29 The dashed lines in Figure 29 connect Jasper
Ridge bedrock chromite (Figure 24) and stoichiometric FeCr2O4 where it can be seen
that compositions of soil chromite lie close to these tie lines Overall soil chromite
demonstrates a decrease in Al and Mg and an increase in Fe relative to Cr compared with
protolith chromite Soil Cr-magnetites contain between 7 and 20 Cr Wt and are
enriched in Cr compared to Cr-magnetites identified in the bedrock (Figure 24) Cr-
magnetite however reveals a decrease in Fe relative to Cr Minor concentrations of Si
(lt2 Wt ) in the soil chromite and Cr-magnetite may represent slightly altered rims or
contamination from sample preparation Chromium concentrations in CSM maintain a
consistent Cr concentration of ~11 Cr Wt similar to those found in the bedrock (~14
23
Cr Wt ) Soil CSMs with compositions between 8 and 12 Mg Wt are more than
likely associated with analyses containing more microcrystalline chlorite compared to
soil CSMs with less than 2 Mg Wt Soil CSM appears be less enriched with Si (lt14
Wt ) relative to those identified in the serpentinite (15 to 27 Wt )
The size of the Cr-bearing minerals varied significantly for each sample ranging
from 10 microm up to 2 mm The majority of chromite grains in the soil correspond to the
sand-size fraction and also vary from euhedral to subhedral with no discernable pattern
The sum of cations based on four oxygens for the chromite and Cr-magnetite analyses
range between 297 and 301 indicating these minerals are spinels not (oxy)hydroxides
Compositions obtained for chromite and Cr-magnetite may represent alteration rims
where weathering of the mineral in the subsurface environment has been most active
BSE images of chromite and Cr-magnetite were not able to definitively identify whether
these analyses in fact represent alteration rims
Mineralogy of the Clay-Size Fraction
The clay-size fraction in the serpentine soil at Site JR3 was evaluated with respect
to depth using XRD and specific chemical and heat treatments The d-spacings (Aring) and
their relative intensities for each soil and treatment with respect to depth are listed in
Table 27 Figure 210 shows the XRD analyses for clay sample JR3515 (depth 5-15
cm) and these analyses are almost identical to the other clays examined at Site JR3
Additionally Table 28 from Whittig and Allardice (1986) is provided to interpret the X-
ray diffraction analyses in Table 27 For example a d-spacing of ~14Aring is present in all
the treatments indicating the presence of chlorite in the clay-sized fraction Using an
XRD analysis program the clay-size fraction was determined to be composed of
smectite vermiculite serpentine minerals (lizardite and antigorite) and chlorite
(clinochlore and Cr-clinochlore) Smectite is the most abundant mineral in the clay-size
fraction whereas Cr-chlorite is the least abundant Overall the d-spacings and relative
intensities are similar despite the treatments and changes in depth (Table 27) This
indicates that the mineralogy of the clay-size fraction does not vary significantly in the
soil column
24
Chromium concentrations in the clay-size fractions are ~1010 mg kg-1 (Table
26) Smectite and vermiculite are possible Cr-bearing clay minerals in which isomorphic
substitution of Cr(III) could occur in the dioctahedral sheet Serpentine minerals
(lizardite and antigorite) are not likely hosts of Cr based on crystallographic
considerations and the absence of Cr in analyzed serpentine minerals in the bedrock
(Table 23) Clinochlore may also be a host for Cr in the inner layer octahedral sheet
Chromium-clinochlore is the only Cr-containing mineral identified in the clay-size
fraction using XRD and Cr-clinochlore (Cr-chlorite) is also an abundant mineral
identified in serpentinite protolith (Tables 22 and 23 Figure 26)
XAS of chromium in relation to depth
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in serpentine soils and bedrock at Site JR3 The XANES (X-ray absorption near-edge
structure) spectra demonstrate the absence of Cr(VI) in the soil column at Site JR3
(Figure 211) based on the lack of a pre-edge feature The detection limit for Cr(VI)
using this method is approximately ~20 mg kg-1 Ultimately XAS analyses demonstrate
Cr(VI) is not detected in any of the soil solids
X-ray microprobe analyses of soil and rock samples (0-5cm depth 5-10cm depth
and serpentinite) at Site JR3 were processed to map the relative concentrations of Fe and
Cr in which areas of elemental enrichment are identified by a violet color areas of
depletion by a blue color (Figure 212) Areas of Cr and Fe enrichment in Figure 212 are
localized with well-defined boundaries corresponding to grain edges and these elements
are not correlated in a well-defined trend Minerals assigned to these areas of Cr and Fe
enrichment were made based on petrographic and microprobe observations on the same
thin section For example high Fe and no Cr is magnetite whereas high Fe and high Cr
is either chromite or Cr-magnetite Chromite magnetite and Cr-magnetite appear to be
present in the serpentine soil at a depth between 0 to 5 cm (Figure 212A) In Figure
212b CSM is a possible mineral detected in the soil (depth 5-10 cm) and the presence
of magnetite is clearly seen due to the localized enrichment of Fe and no Cr Iron and
Cr spectra for the serpentinite were much more heterogeneous as compared to the soil
samples and Cr concentrations appear be more scattered (Figure 212c) These X-ray
25
microprobe analyses provide a complimentary method for examining Cr geochemistry
(including Cr(III) and Cr(VI)) in naturally occurring soil and rock solids however
Cr(III) is the only detected valence state in serpentine soils and rocks at Jasper Ridge
Sequential Extractions
Sequential extractions were completed on 1) Diablo clay (Inceptisol) from Site 1
(Figure 21b) at a depth of 10 cm 2) serpentine soil (Mollisol) from Site 2 (Figure 21b) at
a depth of 4 cm and 3) commercial grade chromite (106-250 microm) and these results are
shown in Figure 213 and listed in Table 29 Chromium extracted from the
exchangeable fraction is minimal (lt01 mg kg-1) for all the samples (number 1 in Table
29 Fig 213) Minor Cr concentrations are associated with the Fe-crystalline oxide
fraction in the Diablo clay (19 Cr mg kg-1) and serpentine soil (78 Cr mg kg-1) whereas
chromite only released 05 Cr mg kg-1 (number 2 in Table 29 Figure 213) Increased
concentrations of Ni and Fe are present in the Fe-oxide fraction of the Diablo clay (161
Ni mg kg-1 and 4474 Fe mg kg-1) and serpentine soil (5138 Ni mg kg-1 and 5138 Fe mg
kg-1 Table 29) Chromium extracted from the Diablo clay serpentine soil and chromite
in the silicate fraction is 14 31 and 14 Cr mg kg-1 respectively (number 3 in Table 29
Figure 213) The sum of extracted Cr Fe Ni and Mn concentrations (number 4 in
Table 29) are low with respect to total metals (number 5 in Table 29) In fact only 07
12 and 00 Cr was removed from the Diablo clay serpentine soil and chromite
respectively (number 6 in Table 29)
Chromite was extremely resistant to all the extractions and less than 001 of the
total Cr was removed (Table 29) Scanning electron microprobe (SEM) images of the
chromite after each extraction are shown in Figure 214 Chromium concentrations
released from the chromite (Table 29 and Figure 214a) increased with each sequential
extraction experiment starting with the exchangeable fraction (Figure 214b) progressing
into the Fe-crystalline oxide fraction (Figure 214c) and ending with the silicate fraction
(Figure 214d) The surface of the chromite became smoother after each extraction and
crystal imperfections in defect sights located along the edges and corners progressively
decreased (Figure 214) Chromite does not appear to be an easily accessible source of
Cr based on these extraction experiments
26
DISCUSSION
Serpentinite at Jasper Ridge has undergone significant physical and chemical
alteration since its formation and emplacement off the California continental margin
Despite being subjected to high pressures (~5-7 kbars) and elevated temperatures (200-
300degC) trace element concentrations within the serpentinite remain elevated Jasper
Ridge serpentinites are enriched with Cr Co Fe Mg Mn and Ni compared to chert and
sandstone located adjacent the main serpentinite body (Table 21 and Figure 21b) Over
the past 400000 years Cr-bearing minerals in the serpentinite have been weathering and
accumulating in the soil During this time Cr concentrations in the soil have
progressively increased relative to Cr-concentrations in the serpentinite protolith This Cr
enrichment is directly linked to specific weathering-resistant minerals identified in the
soil
Chromium in Serpentinites
Three minerals in the bedrock at Jasper Ridge contain appreciable concentrations
of Cr Cr-magnetite (82-103 Cr Wt ) chromite (~108 Cr Wt ) and CSM (~133
Cr Wt ) Cr-magnetite and CSM were derived from the alteration of primary igneous
chromite CSM or Cr-spinels with high concentrations of Si have been noted by
(Burkhard 1993) as a Cr-spinel containing a possible microscopic silicate phase Our
analyses support this result in which the microcrystalline silicate is probably Cr-chlorite
due to its common occurrence with CSM (Figure 26) Definitive identification of this
microcrystalline mixture was not possible due to size of the crystals relative to the spot
size of the electron microprobe Chlorite and pyroxene contain minor concentrations of
Cr (~1 Wt ) and these were the only identified Cr-containing silicates in the
serpentinite Minor concentrations of Cr observed in some chlorites are directly related
to the metasomatic alteration of chromite and Cr-magnetite (Figure 26) Other common
metasomatically-derived Cr-bearing silicates such as fuchsitemariposite and uvarovite
are not present possibly due to the lack of aqueous K+ and Ca+2 in the alteration fluids
Overall the chemical compositions of these protolith minerals (Figure 24) demonstrate
only minor chemical variations allowing comparative analysis with Cr-bearing soil
minerals
27
Chromium in Serpentine Soils
Cr-bearing soil minerals identified using the electron microprobe include
chromite Cr-magnetite and CSM and these phases are chemically distinct from those in
the bedrock (Figure 24 and 29) The highest concentration of Cr (41 Wt ) in the soil
is in chromite with a composition approaching the FeCr2O4 end-member This represents
a ~80 increase in the Cr concentration relative to bedrock chromite (see Figure 215)
Cr-magnetite also demonstrates Cr enrichment (up to 20 Cr Wt ) with respect to
magnetite identified in the bedrock (~10 Cr Wt ) (Figures 24 and 29) The textures of
the soil Cr-bearing spinels exhibit sharp irregular edges possibly due to extensive
weathering CSM in the soil is similar to those found in the bedrock with Cr
concentrations ranging from 11 to 14 Cr Wt however Si concentrations in soil CSMs
(8 to 14 Wt ) are noticeably less than those measured in the serpentinite (15 to 20 Wt
) The weathering of CSM might preferentially release Si relative to other elements in
its structure Cr-silicates (chlorite and pyroxenes) were not identified in the soil possibly
due to complete weathering of these phases or they were too fine-grained for microprobe
analyses however Cr-clinochlore was identified in the clay-size fraction using XRD
Elemental maps shown in Figure 212 corroborate electron microprobe observations by
revealing well-defined zones of Cr and Fe enrichment in both the soils and serpentinite
These areas of Cr and Fe enrichment are either chromite Cr-magnetite CSM or
magnetite Overall these maps reveal Cr is not homogeneously distributed in the soil
confirming Cr is in specific mineral phases Additionally the observed abundance of
chromite Cr-magnetite and CSM in the soils indicates that these minerals are probable
sources for a majority of the Cr enrichment identified in the soils
Soil Spinels
Soil chromite and Cr-magnetite appear to have undergone chemical alteration
enriching the minerals with respect to Cr In Figure 215 compositions of bedrock
chromite and end-member FeCr2O4 are shown together with tie-lines like those in figure
8 connecting both minerals Reaction progress in Figure 215 is based on the assumption
that the serpentinite chromite (0 reaction progress) is proceeding toward FeCr2O4
(100 reaction progress) in the serpentine soil Soil chromites from Jasper Ridge are also
28
plotted in Figure 215 where Cr concentrations are assumed to vary linearly between the
bedrock chromite and chromite end-member which defines the reaction progress of the
mineral These tie-lines correlate extremely well to the chemical composition of
chromites identified in the soil For example chromite containing ~30 Cr Wt will
have ~15 Fe Wt ~13 Al Wt and ~5 Mg Wt similar to soil chromite in Figure 29
The retention of Cr in chromite shown in Figure 215 has significant implications towards
the availability and cycling of Cr in serpentine soils
One possible mechanism for the chromite composition trends illustrated in
Figures 29 and 215 is incongruent dissolution where Cr is retained relative to other
elements Sequential extraction experiments (Figure 213 and Table 29) support this
idea in which Fe is more easily released than Cr (1096 Fe mg kg-1 to 19 Cr mg kg-1)
from chromite Unfortunately Mg and Al were not monitored during the experiments
From these sequential extraction experiments and chromite analyses (Figure 215) Al
and Mg have been preferentially released from the chromite structure compared to Fe and
Cr Additionally the alteration of chromite in the serpentinite and the soils are analogous
to one another Chromite in the serpentinite retains a majority of its Cr compared to Mg
and Al when being altered to CSM by fluids at elevated pressures and temperatures
(Figure 26 and Table 22) Under ambient soil conditions chromite is once again
retaining Cr relative to Mg and Al causing what appears to be an increase in Cr
concentration Chromium is not easily released from chromite whether under high
pressures and temperatures or in the soil
Crystal field stabilization energies (CFSE) and octahedral site preference energy
parameters (OSPE) provide a means to explain the incongruent dissolution or Cr
preservation in the Cr-spinels CFSE is a measure of the net energy of stabilization
gained by a metal ions nonbonding d electrons as a result of complex formation This
measure also indirectly describes the energy required to remove an ion from a crystal
structure The octahedral CFSE for Cr(III) is ndash2247 kJmole and the OSPE (the
difference between the tetrahedral and octahedral CFSE) is ndash1578 kJmole (Dunitz and
Orgel 1957 McClure 1957) These values are the most negative for any transition
element and indicating the presence of Cr(III) in an oxide increases the crystalrsquos stability
compared to an analogous crystal lacking Cr(III) Chromium(III) would be among the
29
last ions to removed from the spinel structure compared to other transition elements such
as Fe2+and Mn2+
Soil Chemistry related to Chromium Availability
In Table 25 the chemical properties of the soil at Site JR3 do not appear to be a
factor when evaluating Cr in the soil The pH of the soil did not vary considerably with
depth Additionally the near neutral pH would not have a significant impact accelerating
Cr-bearing mineral dissolution Organic matter within the soil was elevated despite the
low plant productivity of the serpentine soil The role of organic matter could be a
significant factor such as the production of organic acids promoting the dissolution of the
Cr-bearing minerals however Cr concentrations near the surface (where organic matter
was the greatest) are not significantly greater or less than those deeper in the soil profile
The relative concentrations of ammonium acetate extractable Ca K Mg and Na are
characteristic of serpentine soils and did not provide any new insight Finally Cr(III)
was the only identifiable valence state of Cr based on X-ray absorption spectroscopy
(Figure 211) which is what should be expected based on the Cr-bearing minerals and the
pH of the soil
Extraction experiments summarized in Table 29 and Figure 213 demonstrate
~99 of the Cr in a serpentine soil is retained in highly resistant phases at Jasper Ridge
The chromite standard used in these experiments contributed negligible concentrations of
Cr (lt2 mg kg-1 total) for all of the sequential extraction experiments supporting earlier
evidence that Cr-spinels are retaining Cr and are potentially responsible for a majority of
the total Cr enrichment in the serpentine soil At Jasper Ridge Cr extracted from the
serpentine soil (31 mg kg-1) is likely the result of the Cr-silicates compared to the Cr-
spinels Low Cr-containing minerals (possibly chlorite enstatite augite and clay
minerals) appear to be the most probably source of Cr to soil solutions and vegetation of
the serpentine soil Approximately 30 of the serpentine soil is composed of clay-sized
minerals and this size-fraction has ~1000 Cr mg kg-1 The mineral responsible for Cr
enrichment in the clay-size fraction is more than likely Cr-chlorite identified using XRD
Chromium-chlorite in the soils suggests that this mineral is potentially the largest silicate
contributor of Cr in the serpentine soil
30
Interpreting the enrichment and behavior of Cr in serpentine soils such as in
Figure 27 can be accomplished based on the abundance Cr-spinels observed in the soil
Small grains of chromite Cr-magnetite and even CSM in the soil have the potential to
appreciably impact Cr concentrations Table 210 demonstrates the chromite masses
required to change the Cr concentration in a 1 kg sample of soil The highest Cr
concentration at Jasper Ridge was identified at Site 3 with a measured value of 4760 mg
kg-1 which means only ~1 of the soil sample (Table 210) needs to contain chromite
Additionally a minimal amount (lt011 of the sample as shown in Table 210) of
chromite is necessary to change the concentration by 500 mg kg-1 which could easily
cause the sharp Cr increases and decreases noted in Figure 27 This agrees with our
observations of chromite abundance in the serpentine soils Additionally a majority of
the chromite grains identified in the bedrock are between 2 and 002 mm which
correspond directly to the elevated Cr concentrations (gt10000 mg kg-1) in the sand-size
fraction of the serpentine soil (Table 26) Even the ~15000 Cr mg kg-1 values in soils
measured by Hotz (1964) in Figure 22 only require ~2 of the sample of the sample to
be chromite A few grains of chromite possess the ability to significantly impact
measured Cr concentrations in soil samples
Chromium and Nickel in Serpentine Soils
Chromium and Ni have often been correlated to one another due to their common
enrichment in serpentine soils (Hotz 1964 Rabenhorst et al 1982 Schwertmann and
Latham 1986 Brooks 1987 Schreier et al 1987 Alexander et al 1989 Kaupenjohann
and Wilcke 1995) A majority of the Ni in the Jasper Ridge serpentine soil originated
from olivine (Figure 25c and 25d Table 23) Chemical alteration of olivine during
serpentinization resulted in the mineral serpentine (Coleman and Jove 1992) in which
minor concentrations of Ni (lt037 NiO Wt ) substituted into its structure Continued
alteration under highly reducing conditions and the incompatibility of Ni in the serpentine
structure resulted in the formation of Ni-Fe alloy (Figure 25c) This type of alteration is
observed in many of the rocks at Jasper Ridge where the Ni-Fe alloys are ~10 microm in size
Olivine in the rocks and soil appear to be less resistant to weathering compared to
chromite therefore Ni is possibly more available to bacteria and vegetation in serpentine
31
soils Sequential extractions (Figure 213 Table 29) support these observations in which
a greater percentage of total Ni was extractable compared to the Cr in the soils Although
Cr and Ni are enriched in serpentine soils the availability and behavior of these two
elements are significantly different
CONCLUDING REMARKS
Chromium is predominantly present in primary chromite Cr-magnetite and a
chromite-silicate mixture (CSM) in both the bedrock and soil despite weathering over
geologic time Sharp variations in soil Cr concentrations with respect to depth are caused
by the presence or absence of chromite Cr-magnetite andor CSM Chromium-spinels in
the soil have undergone incongruent dissolution possibly conserving Cr(III) with respect
to Al3+ and Mg2+ Chromite is highly resistant to weathering and it is more than likely
not contributing bioavailable Cr Although Cr-spinels are a major source of Cr Cr-
silicates in the protolith (chlorite enstatite and augite) and soil (chlorite
montmorillonite and vermiculite) afford the most accessible Cr source identified Based
on our observations and analyses chlorite is the most likely Cr source of these silicate
minerals Finally Cr(VI) was not detected in any of the soil and rock samples
ACKNOWLEDGEMENTS
We thank Jasper Ridge Biological Preserve for allowing us the opportunity to
examine the serpentine soils Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund NSF Grant EAR-9902859 (DKB) GSA Graduate
Student Research Grants Office of Teaching and Learning (SF) at Stanford University is
gratefully acknowledged We appreciate the technical support from the staff of the
Stanford Synchrotron Radiation Laboratory from Bob Jones for assistance in microprobe
analyses of the serpentinites and serpentine soils and from Ben Bostick Travis Horton
and Theresa Barber for field and lab assistance
32
Table 21 Whole rock elemental concentrations of serpentinites (Figure 21b) chert and sandstone
-----------------------Location of Serpentinitesdagger----------------------- Elemental Concentrations JR1 JR3E JR6E Site 2 Site 5 Site 10 JR3
ChertDagger SandstoneDagger
Al 02 08 03 027 22 107 024 079 027 Ca 026 007 005 003 046 013 003 001 lt001 Co mg kg-1 101 99 100 99 155 217 97 1 lt1 Cr mg kg-1 580 2280 1510 1840 1945 2090 1465 10 lt1 Fe 303 71 609 581 76 1106 512 041 024 K 001 031 099 lt001 lt001 001 lt001 029 008
Mg 122 134 154 1336 524 804 112 014 034 Mn mg kg-1 495 800 155 730 1375 1860 460 110 85 Na 181 006 093 lt001 lt001 lt001 lt001 001 001 Ni mg kg-1 2037 1545 3321 3510 3020 3960 2740 20 9
Total digestion ICP-AES analyses daggerSerpentinites are dominantly composed of the minerals serpentine chlorite talc and spinels DaggerSee Figure 21b for sample locations
33
Table 22a Selected Cr-bearing minerals and compositions in oxide Wt in serpentinite rock at Jasper Ridge
Chlorite Enstatite Augite Chromite Cr-Magnetite ChromiteSilicatedagger SiO2 3738 5463 5222 000 024 1053 Al2O3 552 477 516 5059 182 395 NiO 001 01 00 03 00 052 FeODagger 762 614 239 1446 7919 5042 MnO 009 015 009 042 140 182 Cr2O3 024 060 099 1584 1605 1939 MgO 3337 3305 1560 1869 024 605 CaO 023 050 2348 000 000 019 Na2O 000 001 050 000 001 008 K2O 000 000 001 000 000 011 Total 8446 9994 10048 10027 9894 9307
Formula Unit Compositions
Si 370 094 189 000 001 --- IVAl 030 006 011 000 000 --- VIAl 035 004 011 157 008 --- Ni 000 000 000 001 000 ---
Fe+2 063 010 007 030 232 --- Mn 001 000 000 001 004 --- Cr 002 001 003 033 048 --- Mg 493 085 084 073 001 --- Ca 002 001 091 000 000 --- Na 000 000 004 000 000 --- K 000 000 000 000 000 --- O 1400 300 600 400 400 ---
Electron microprobe analyses daggerChromitesilicate is also referred to as CSM DaggerFe+3 is present however all conversions are based on Fe+2
34
Table 22b Selected Cr-bearing minerals and compositions in oxide Wt in serpentine soil at Jasper Ridge
CSM Chromite Cr-Magnetite SiO2 1940 057 267 Al2O3 559 1059 381 NiO 014 001 004 FeOdagger 2629 2259 5244 MnO 155 154 419 Cr2O3 1878 5577 3237 MgO 1668 916 288 CaO 015 001 005 Na2O 006 001 001 K2O 0 008 009 Total 8864 10033 9855
Si --- 002 010 IVAl --- 042 017 VIAl --- 000 000 Ni --- 000 000
Fe+2 --- 063 148 Mn --- 004 013 Cr --- 147 096 Mg --- 046 016 Ca --- 000 000 Na --- 000 000 K --- 000 000 O --- 400 400
Electron microprobe analyses dagger Fe+3 is present however all conversions are based on Fe+2
35
Table 23 Mineral compositions in oxide Wt of minerals identified in Figures 25 and 26 A B C D E F G H I J K L Chromite
Chromite CSM Magnetite Chlorite Olivine Serpentine Serpentine Ni-Fe-Metal Cr-Magnetite CSM Chlorite
SiO2 0 0 1053 0 3407 4046 3688 4074 591dagger 021 1143 2855Al2O3
5059 5162 395 034 1114 000 397 085 054 165 980 2293NiO 028 031 052 0 004 037 037 0 5302dagger 0 0 0
FeODagger 1446 1452 5042 9992
1220 901 262 453 1753dagger 8169 3498 806MnO 042 044 182 0 008 010 0 0 0 107 150 00Cr2O3 1584 1576 1939 125 071 001 0 0 0 1369 1905 039MgO 1869 1869 605 003 292 4895 3848 3755 064 1217 2828CaO 0 0 019 0 0 0 0 0 0 0 0Na2O 0 0 0 0 0 0 0 0 0 0 0K2O 0 0 0 0 0 0 0 0 0 0 0H2O --- --- --- --- 1200 --- 1600 1600 --- --- 1200Total 10027 10136 9307
10164 9944 9891 9834 9972
9897 8897 10028
Formula Unit CompositionSi
000 000 --- 000 328 100 184 200 001 --- 269IVAl 000 000 --- 000 072 000 000 000 000 --- 131VIAl 157 159 --- 002 058 000 023 005 007 --- 123Ni 001 001 --- 000 000 001 001 000 000 --- 000
Fe+2 030 029 --- 294 084 019 011 019 239 --- 059Mn 001 001 --- 000 000 000 000 000 003 --- 000Cr 033 032 --- 004 006 000 000 000 041 --- 003Mg 073 073 --- 000 419 180 285 274 004 --- 397Ca 000 000 --- 000 000 000 000 000 000 --- 000Na 000 000 --- 000 000 000 000 000 000 --- 000K 000 000 --- 000 000 000 000 000 000 --- 000O 400 000 --- 000 1400 400 700 700 400 --- 1400
36
Electron microprobe analyses dagger elemental metal Dagger Fe+3 is present however all conversions are based o
36
2321000---
10025
591000054
53021753000000
2321000000000000
n Fe+2
Table 24 Bulk soil elemental concentrations for soils and bedrock collected along the transect
(Fig 21b)
Sites Samples Depth
(meters) Cr
(mg kg-1) Ni
(mg kg-1) Mn
(mg kg-1) Co
(mg kg-1) Al
() Fe
() Ca ()
Mg ()
COI03 0015 2420 1590 1325 118 464 565 065 352 COI36 0045 2370 1630 1600 138 467 55 063 356 COI69 0075 2160 1755 1670 144 471 6 064 365 COI912 0105 2420 1785 1740 151 496 615 067 384
1
COI1215 0135 2570 1985 2340 195 506 649 069 411 COII02 001 3710 2400 1665 169 294 702 100 768 COII25 0035 3600 2290 1625 162 275 673 093 743 COII58 0065 2910 2310 1550 159 25 672 087 771
COII811 0095 3520 2350 1595 158 276 693 093 766 2
COIIBedrock 012 1840 3510 730 99 027 581 003 1336 COIII14 0025 3750 2930 1655 190 178 897 062 762 COIII47 0055 4760 3050 1850 178 178 94 056 805
COIII710 0085 4380 3410 1930 216 216 1044 07 797 3 COIII1013 0115 4320 3380 1955 201 213 1061 063 626 COIV01 0005 3710 2970 1915 214 209 909 053 781 4 COIV15 003 3540 2710 1725 195 181 801 043 792 COV14 0025 2620 2350 1330 150 246 651 063 526 COV47 0055 1775 2710 1520 165 252 766 059 517
COV710 0085 2000 3090 1830 195 225 737 057 583 5 COVBedrock 0115 1945 3020 1375 155 22 76 046 524
COVI14 0025 3040 3720 1855 224 084 1152 015 736 COVI47 0055 2400 3610 1785 218 085 1138 014 68 COVI710 0085 2960 3800 1850 232 093 1232 015 714 6
COVI1013 0115 2800 3860 1800 224 085 1164 014 739 COVII14 0025 2620 3550 1925 229 083 1128 015 718 COVII47 0055 1975 3590 1805 216 084 1085 014 655 7 COVII712 0095 2080 3600 1840 220 07 1105 011 741 COVIII14 0025 2130 3830 1990 238 085 115 017 705 COVIII47 0055 2550 3720 1865 228 085 1099 017 714 8
COVIII712 0095 2240 3910 1860 229 076 1112 014 828 COIX14 0025 2470 3480 1890 227 104 1201 023 579 COIX47 0055 3000 3830 1595 205 084 108 013 842 9 COIX713 01 2720 3710 1845 221 111 1208 017 551 COX14 0025 2810 3740 1760 218 129 114 017 675 COX47 0055 2190 3910 1835 224 131 1206 017 702
COX710 0085 1725 3370 1595 192 11 1061 014 635 10 COXBedrock 0115 2090 3960 1860 217 107 1106 013 804
Total digestion ICP-AES analyses
37
Table 25 Chemical and physical properties of the soil at Site JR3
NH4OAc-extractable ---Hydrometer---Ca K Mg Na
Sand Silt ClaySample Depth (cm) pH EC
dSm
Organic Matter ------
LECO Nitrogen ------ ---------mg kg-1---------
Texture
JR305 25 671 04 32 012 505 2366 3291 355 36 31 33 Clay LoamJR3515
10 683 09 19 064 4205 8176 3343 2392 36 30 34 Clay LoamJR31530 225 688 03 12 003 3725 4903 3750 2305 36 33 31 Clay LoamJR33045 375 698 07 10 003 2873 3665 3388 2549 46 24 30 Sandy Clay Loam
See Methods Section for more details
38
38
Table 26 Selected trace metal concentrations for each size fraction versus depth at Site JR3
Depth (cm) Total gt2mm Fraction Sand Silt Clay Sample
------------------------Chromium (mg kg-1)------------------------ JR305 25 4906 5086 gt10000 4565 988 JR3515 10 5976 4678 8284 4000 1115 JR31530 225 5222 4206 gt10000 4133 980 JR33045 375 3477 3297 gt10000 4571 956
------------------------Nickel (mg kg-1)------------------------------ JR305 25 1935 2050 1660 1305 2920 JR3515 10 2180 2050 1980 1325 2920 JR31530 225 2190 2120 1835 1370 2880 JR33045 375 2090 2140 1740 1285 2840
------------------------Manganese (mg kg-1)-------------------------- JR305 25 1319 1571 1728 1301 1210 JR3515 10 1664 1601 2701 1471 1254 JR31530 225 1639 1626 2217 1599 1231 JR33045 375 1222 1226 1781 1405 1170
-----------------------Cobalt (mg kg-1)-------------------------------- JR305 25 128 146 159 134 131 JR3515 10 155 145 241 147 132 JR31530 225 152 147 194 162 128 JR33045 375 125 124 159 141 122
Total digestion ICP-AES analyses
39
Table 27 X-ray diffraction analyses of the clay-sized fraction from site JR3
JR305CLAY Depth 0-5cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1439 100 1833 100 1274 100 1391 100 731 179 1432 431 732 645 1047 105 71 216 941 163 71 41 838 838 474 101 847 141 475 257 726 726 356 125 73 271 365 264 313 83 712 248 355 299
473 143 313 257 364 131 355 193
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1845 1000 1268 1000 1388 1000 852 68 1447 387 724 987 1010 859 733 264 944 143 703 679 836 281 712 247 903 139 471 282 723 576 475 109 850 83 362 244 365 73 733 426 352 321 356 117 712 291 313 154 314 63 475 122
364 170 355 196
JR3515CLAY Depth 5-15cm Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat
d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity 1447 1000 1724 1000 1297 1000 1396 1000 842 45 1397 447 741 957 1175 708 730 318 923 176 715 686 1037 944 710 255 840 118 477 271 837 139 475 97 717 465 367 400 726 375 357 133 468 182 357 386 313 91 362 194 315 286
352 229 JR33045CLAY Depth 30-45cm
Mg-Saturated air-dried Magnesium+Glycerol K-saturated air-dried K-saturated+Heat d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity d-spacing(Aring) Intensity
1447 1000 1869 1000 1268 1000 1393 627 842 44 1476 287 737 843 1049 1000 730 207 954 107 715 539 837 190 710 182 915 107 479 157 723 111 475 69 860 61 366 303 364 69 740 332 356 292 356 74 717 256 315 315 313 57 478 111
367 141 357 115
See Methods Section for details
40
Table 28 X-ray diffraction d-spacings for common silicates from Whittig and Allardice (1986)
Diffraction Spacing (Aring) Mineral(s) Indicated Mg-saturated air-dried
14-15 Smectite vermiculite chlorite 99-101 Mica (illite) 71-73 Serpentine
715 Kaolinite chlorite Mg-saturated glycerol-solvated
177-180 Smectite 14-15 Vermiculite chlorite
99-101 Mica (illite) 71-73 Serpentinte
715 Kaolinite chlorite K-saturated air-dried
14-15 Chlorite vermiculite (with hydroxy interlayer) 124-128 Smectite 99-101 Mica vermiculite (contracted) 71-73 Serpentine
715 Kaolinite chlorite K-saturated heated (500ordmC)
14 Chlorite 99-101 Mica vermiculite (contracted) smectite (contracted) 71-73 Serpentine
715 Chlorite
41
Table 29 Results of the Diablo clay serpentine soil and chromite sequential extraction experiments
1 Exchangeable Fraction (1mM BaCl2) 4 Sum of 1 2 amp 3 (mg kg-1) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 0 0 003 Cr 159 388 19 Fe 014 011 001 Fe 24812 25532 1096 Ni 104 103 008 Ni 1854 1868 25 Mn 0 0 0 Mn 778 1172 04
2 Fe-Oxide Fraction (1M NH2OH HCl in 25 (vv) HOAc) 5 Total Cr Fe Ni and Mn (Total Digestion) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ----------------mg kg-1----------------- Cr 19 78 05 Cr 2290 3310 1767864 Fe 4474 5138 38 Fe 60000 68000 1044325 Ni 1610 1506 01 Ni 1673 2345 00 Mn 231 430 00 Mn 1498 1608 3424547
3 Silicate Fraction (10M HF) 6 Percent Extractable ((45)100) Diablo Clay Serpentine Soil Chromite Diablo Clay Serpentine Soil Chromite
Element ----------------mg kg-1----------------- Element ------------------------------------- Cr 140 310 14 Cr 07 12 00 Fe 20337 20393 1058 Fe 41 38 01 Ni 234 352 22 Ni 111 80 00 Mn 547 742 04 Mn 52 73 00
ICP-AES analyses
42
Table 210 Chromite masses required for determined Cr concentrations (mg kg-1) in 1 kg of sample
Chromite
FeCr2O4 Cr Concentration
(mg kg-1) Grams Sample 500 11 011 1000 22 022 2000 43 043 5000 108 108
43
Figure Captions
Figure 21a Soils map of Jasper Ridge Biological Preserve (Stanford CA) with
complete descriptions of the soils shown in the legend 21b An enlarged image of the
study area shown in 22a indicates sampling sites of the soils and rocks collected A
dashed line marks the 500 m transect
Figure 22 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above
Figure 23 Petrographic images illustrating the varying serpentinite textures at Sites (A)
JR3 (B) JR5E and (C) JR3E For reference to site locations see Figure 21b
Figure 24 Correlation diagrams of Cr-bearing minerals in the serpentinite at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Si in Wt
Chromite (circle) Cr-magnetite (square) chromitesilicate or CSM (down triangle)
chlorite (hexagon) and pyroxenes (up triangle) analyses are shown on each correlation
diagram
Figure 25 Images of the serpentinite bedrock at Site JR3 demonstrating the mineralogy
and Cr-bearing phases Microprobe analyses are indicated by circled letters and these
values are in Table 23 25a Backscattering electron image demonstrating several Cr-
bearing phases present including chromite Cr-magnetite CSM and Cr-chlorite from
Figure 23a 25b Magnified image shown in Figure 25a demonstrating the fine-grained
texture of CSM Additionally Cr2O3 Wt values are shown for chromite the depletion
zone and CSM 25c Image of Ni-bearing phases present in the bedrock including
olivine serpentine and Ni-Fe alloy from Figure 23a 25d Cartoon illustrating the
mineralogical and chemical changes present in 25c
Figure 26 Images of a Cr-chlorite and Cr-magnetite assemblage in the serpentinite
bedrock 26a Petrographic image of the bedrock at Site 4 (Figure 21b) 26b
44
Backscattered electron image demonstrating the common occurrence of Cr-chlorite
encompassing chromite grains Microprobe analyses are indicated by circled letters and
are located in Table 23
Figure 27 Highest and lowest Cr concentrations (mg kg-1) at each site in the serpentine
soil 500 m transect shown in Figure 21b Samples included soils at various depths (m)
and bedrock from each site The average Cr concentration for the Jasper Ridge
serpentinite is marked on the central figure Chromium concentrations versus depth (m)
at each site (noted in the upper left corner) outline the main diagram
Figure 28 A graphical representation of Cr Ni Mn and Co concentrations collected
for total gt2 mm sand silt and clay size fractions listed in Table 26
Figure 29 Correlation diagrams of Cr-bearing minerals in the serpentine soil at Jasper
Ridge comparing microprobe analyses of Cr versus Al Fe Mg and Al+Fe in Wt for
chromite (circles) Cr-magnetite (hexagons) and CSM (diamonds) Chromite from the
bedrock is denoted by a larger circle and FeCr2O4 with a square
Figure 210 XRD clay analyses of sample JR3515 (Site JR3 Figure 21b) demonstrate
the influence of each treament Mg-saturated Mg+glycerol K-saturated and K-saturated
+ heat (500oC) The d-spacings and intensities are also listed in Table 27
Figure 211 Cr K XANES spectra for soils analyzed from Jasper Ridge serpentine soil
at depths (a) 0-5cm (b) 5-15cm (c) 15-30cm (d) 30-45cm and (e) bedrock The pre-edge
area indicative of Cr(VI) peaks is shown on the right Note that no peaks are present
Figure 212 X-ray microprobe elemental maps of Cr and Fe for (212a) serpentine soil
at a depth of 0-5cm (212b) serpentine soil at a depth of 5-10cm and (212c) serpentinite
bedrock at Jasper Ridge site JR3 (Figure 21b)
45
Figure 213 A graphical representation of the exchangeable Fe oxide and silicate
extraction experiments for the Diablo clay (Site 1 Figure 21b) serpentine soil (Site 2
Figure 21b) and chromite listed in Table 29 The sum refers to the summation of the
three extraction experiments
Figure 214 SEM images of the chromite standard (A) after each sequential extraction
treatment including the (B) exchangeable fraction (C) Fe-crystalline oxide fraction and
(D) silicate fraction The concentration of Cr removed by each treatment is shown
beneath each set of images
Figure 215 Tie-lines from Figure 29 connecting serpentinite chromite and end-
member FeCr2O4 are shown on this single plot Soil analyses from Figure 29 are also
plotted on this diagram For a sum of 100 Wt oxygen must be added to the sum of
Cr Al Fe and Mg Wt
46
Figure 21a
47
Figure 21b
48
Chromium Concentration (mg kg-1)1000 10000
Dep
th (m
eter
s) 01
10
Maryland (Rabenhorst et al 1982)California (Gough et al 1989)California (Hotz 1964)New Foundland (Roberts amp Proctor 1992)
Surface
Figure 22
49
Figure 23
50
Cr (Wt )0 5 10 15 20
Al (
Wt
)
0
5
10
15
20
25
30
Cr (Wt )0 5 10 15 20
Fe (W
t
)
0
20
40
60
80
100
Cr (Wt )0 5 10 15 20
Mg
(Wt
)
0
5
10
15
20
25
Cr (Wt )0 5 10 15 20
Si (W
t
)
0
5
10
15
20
25
30
Cr0 10 20 30 40 50 60 70 80 90 100
Al
0
10
20
30
40
50
60
70
80
90
100
Fe
0
10
20
30
40
50
60
70
80
90
100
Chromite
Chlorite
Magnetite
Chromite-Silicate
Pyroxenes
Chromite
Chlorite
Pyroxenes
Magnetite
Chromite-Silicate
Figure 24
51
Figure 25
52
Figure 26
53
Sites1 2 3 4 5 6 7 8 9 10
Con
cent
ratio
n (m
g kg
-1)
1500
2000
2500
3000
3500
4000
4500
5000Soil Chromium
1
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000002004006008010012014016
2
Cr Concentration (mg kg-1)
1500 2000 2500 3000 3500 4000
Dep
th (m
eter
s)
000002004006008010012014
3
Cr Concentration (mg kg-1)3500 4000 4500 5000
Dep
th (m
eter
s)
000
002
004
006
008
010
012
5
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000002004006008010012
6
Cr Concentration (mg kg-1)2200 2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000002004006008010012
7
Cr Concentration (mg kg-1)1600 2000 2400 2800
Dep
th (m
eter
s)
000
002
004
006
008
010
9
Cr Concentration (mg kg-1)
2400 2600 2800 3000 3200
Dep
th (m
eter
s)
000
002
004
006
008
010
012
10
Cr Concentration (mg kg-1)1600 2000 2400 2800 3200
Dep
th (m
eter
s)
000002004006008010012
4
Cr Concentration (mg kg-1)3500 3600 3700 3800
Dep
th (m
eter
s)
000
001
002
003
004
8
Cr Concentration (mg kg-1)
2100 2200 2300 2400 2500 2600
Dep
th (m
eter
s)
000
002
004
006
008
010
BEDROCK
BEDROCK
BEDROCK
Range of Cr Concentrations in Serpentinite Bedrock
Figure 27
54
Soil Chromium
Depth (cm)0 10 20 30 40 50
Cr m
g kg
-1
02000400060008000
100001200014000
Totalgt2mmSandSiltClay
Soil Nickel
Depth (cm)0 10 20 30 40 50
Ni m
g kg
-1
0
1000
2000
3000
4000
5000
Soil Manganese
Depth (cm)0 10 20 30 40 50
Mn
mg
kg-1
0
1000
2000
3000Totalgt2mmSandSiltClay
Soil Cobalt
Depth (cm)0 10 20 30 40 50
Co
mg
kg-1
0
50
100
150
200
250
300
Totalgt2mmSandSiltClay
Totalgt2mmSandSiltClay
Exceeds detection limits of 10000 mg kg-1 Cr
Figure 28
55
Cr (Wt )0 10 20 30 40 50
Al (W
t
)
0
5
10
15
20
25
30
Cr (Wt )0 10 20 30 40 50
Mg
(Wt
)
0
2
4
6
8
10
12
Cr (Wt )0 10 20 30 40 50
Si (W
t
)
0
2
4
6
8
10
12
14Cr (Wt )
0 10 20 30 40 50
Fe (W
t
)
0
10
20
30
40
50
60
70
Cr (Wt )0 10 20 30 40 50
Al +
Fe
(Wt
)
10
20
30
40
50
60
70
JRBPChromite
FeCr2O4
Chromite
Magnetite
Chromite-Silicate
Figure 29
56
Clay XRD of JR3515 Depth 5-15cm
0
200
400
600
800
1000
1200
0 5 10 15 20 25 30 35 40 45
d-spacing (Aring)
Rel
ativ
e In
tens
ity
Mg-Saturated
Mg+Glycerol
K-saturated
K-saturated+heat (500oC)
Figure 210
57
Energy (kev)
597 598 599 600 601 602 603
Nor
mal
ized
Abs
orba
nce
-4
-3
-2
-1
0
1
2
(a)
(b)
(c)
(d)
(e)
Cr(
VI)
Figure 211
58
Figure 212
59
Diablo Clay
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Serpentine Soil
0
500
1000
1500
2000
2500
3000
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide SilicateSum
Chromite
0
20
40
60
80
100
120
Cr Fe Ni Mn
Conc
entra
tion
(mg
kg-1
)
ExchangeableFe Oxide
SilicateSum
Figure 213
60
Figure 214
61
Reaction Progress ()0 20 40 60 80 100
Elem
enta
l Wt
0
10
20
30
40
50
ChromiumAluminumIronMagnesium
FeCr2O4Jasper Ridge
Bedrock Chromite
SoilChromite
Figure 215
62
CHAPTER 3
OXIDATIVE PROMOTED DISSOLUTION OF CHROMITE BY MANGANESE
DIOXIDE AND CONCURRENT PRODUCTION OF CHROMATE
ABSTRACT
Chromium release and oxidation from chromite is a potential environmental hazard in
sediments and soils and a pathway for soil development (weathering of primary minerals)
related to ultramafic rocks and their metamorphic derivatives (serpentinites) Birnessite is
a common pedologic mineral in these sediments and soils capable of oxidizing aqueous
Cr(III) In this study the interaction between chromite ((Fe046Mg052Mn002)(Cr061Al029Fe010)2O4)
and birnessite is investigated with an interest in the potential generation of Cr(VI)
Specifically the effects of chromite suspension density birnessite suspension density
and pH at a temperature of ~25degC in relation to the reaction kinetics were examined The
rate of Cr(VI) in solution increases with increasing chromite suspension densities and
decreasing pH but is independent of birnessite suspension densities at values greater than
~20 m2 L-1 The overall rate expression of Cr released and oxidized in solution from
chromite can be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097 microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 birnessite suspension densities are greater than 20 m2 L-1
and pH values are between 3 to 8 Experiments involving serpentine soils containing
chromite and birnessite produce Cr(VI) rates consistent with those predicted by the
overall rate expression with Cr(VI) rates ranging from 9times10-4 to 44times10-3 microM h-1
Additionally these experiments demonstrate that serpentine soils are a source of non-
anthropogenic Cr(VI)
63
INTRODUCTION
Chromite a Cr-enriched spinel is responsible for a majority of the Cr enrichment
identified in ultramafic rocks serpentinites and serpentine soils (Rabenhorst et al 1982
Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989 Gasser and
Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003 Oze et al
2003a Oze et al 2003b) The chemical weathering of chromite is therefore an important
process when evaluating the potential of ultramafic rocks and their related sediments and
soils as sources of Cr Despite the propensity of chromite to resist chemical weathering
the dissolution of chromite and release of Cr will occur over geologic time (Bialowolska
and Salacinski 1984 Oze et al 2003b) Chromium in chromite is in the form of Cr(III)
the non-toxic and chemically immobile valence state however oxidizing conditions in
near surface environments is a potential reaction leading to the enhanced dissolution of
chromite and the formation of Cr(VI) the toxic and chemically mobile valence state
which typically exists as the oxyanion chromate (HxCrO42-x) (Daugherty 1992 Cohen et
al 1993 Fendorf 1995 Deer et al 1996) Consequently chromite may be a non-
anthropogenic source of Cr(VI) in the environment
Naturally occurring high valence Mn oxides present in soils and sediments
derived from ultramafic rocks are strong oxidizers capable of affecting the chemical
mobility of metals through redox reactions (Fendorf and Zasoski 1991 Johnson and
Xyla 1991) These insoluble mixed valent oxides (Mn(IIIIV)) are observed coating
mineral surfaces in these ultramafic sediments and soils where total Mn concentrations
are often gt100 mg kg-1 (Brooks 1987 Schreier et al 1987 Oze et al 2003a Oze et al
2003b Oze et al) Birnessite is a naturally occurring mixed valence oxide known to
oxidize aqueous Cr(III) to the more chemically mobile Cr(VI) (Eary and Rai 1987
Fendorf and Zasoski 1991) Gough et al (1989) and Becquer et al (2003) have
suggested that Mn oxides are responsible for Cr(VI) identified in California and New
Caledonia serpentine soil solutions however the dissolution and oxidation of Cr from a
Cr-bearing mineral due to the presence of birnessite has not been quantitatively defined
In this study we examine the kinetics of Cr release and oxidation when chromite and
birnessite are present in solution
64
The objective of this study is to determine a rate expression describing the
release and oxidation of Cr from chromite in the presence of birnessite Using well-
controlled batch experiments the solution chemistry and geochemistry of Cr related to
chromite and birnessite interaction was evaluated temporally over a range of pH
Conditions were chosen to simulate the slightly acidic conditions characteristic of most
serpentine soils (Rabenhorst et al 1982 Gough et al 1989 Graham et al 1990 Cole
1992 Jeffrey 1992 Lee 1992 Verger 1992 Becquer et al 2003) Ultimately these
experiments allow the examination of one potential pathway leading to Cr(VI) identified
in solutions related to ultramafic rocks and serpentine soils
MATERIALS AND METHODS
Solid Phase Characterization and Synthesis
Chromite used in the experiments was commercial-grade chromite (MSDS No
302-1 WHMIS CLASS D2A) obtained from Barnes Environmental Inc Ontario
Chromite was prepared by cleaning the sand-size material in multiple ultrasonic
isopropanol baths and rinses The chromite was then sieved using a Retsch magnetic
sieve shaker for 30 minutes in order to obtain the 160-250 microm size fraction Impurities in
the 160-250 microm chromite were removed using a slope Franz magnetic separator These
impurities included the minerals olivine augite plagioclase feldspar and quartz Visual
and scanning electron microprobe (SEM) examinations of the remaining 160-250 microm
fraction confirmed its purity Chromite was crushed to a fine powder using an agate
mortar and pestle and then rinsed multiple times in deionized destilled (DDI) water and
001 M HCl solution to remove the ultra-fine material and eliminate highly reactive sites
caused during grinding The chromite powder was inspected before (Figure 31) and
after the experiments using a scanning electron microscope (JEOL 5600LV) the surface
area was 025 m2 g-1 as determined using N2(g) and BET isotherm
An automated JEOL 733A electron microprobe operated at 15 kV accelerating
potential and 15 nA beam current was used to investigate possible zoning and to quantify
the chemical composition of the chromite grains prior to powdering No zoning was
observed and the chemical composition of the chromite was based on the average of 25
electron microprobe analyses The stoichiometric composition of chromite based on 4
65
oxygens is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 X-ray diffraction was also used to
deduce the purity of the chromite powder X-ray diffraction spectra obtained for the
chromite powder matched JCPDS card 03-0873 (unit cell a=8300 b=8300 c=8300
α=90 β=90 and γ=90) and no contaminants were evident
Birnessite was synthesized using the procedure described by Buser et al (1954)
and implemented by Fendorf and Zasoski (1992) A solution of 63 g of potassium
permanganate (KMnO4) in 1 L of DDI water was heated to 90 degC Approximately 100
mL of the heated slurry was poured into a 4 L beaker and reheated to 90degC while being
stirred vigorously The remaining 900 mL of slurry and 66 mL of concentrated
hydrochloric acid were simultaneously and proportionally dispensed into the beaker over
a period of thirty seconds The temperature of the resulting solution was maintained at
90deg C for an additional 10 minutes and then allowed to cool to room temperature The
product was washed with alternate rinses of DDI water and a 001 M HCl solution The
birnessite surface area was 125 m2 g-1 based on BET analyses Solids were inspected
before (Figure 31) and after the experiments using the SEM
Extraction Experiment Procedures
Several chemical extraction experiments were performed on 106-250 microm
chromite grains prior to the chromite-birnessite batch experiments in order to determine
the potential release of Cr Fe and Mn One gram of chromite and 25 mL of solution
were placed into a 25 mL centrifuge tube and the solids and solution were gently agitated
The solutions and parameters used for these extraction experiments include the
following 1) 10 hydrogen peroxide (H2O2) solution at 25degC and pH of ~5 for 6 h 2)
01 M potassium permanganate (KMnO4) solution at 60degC and pH of ~3 for 6 h 3) 10 M
hydrofluoric (HF) acid at 60degC for 6 h 4) 1 M hydroxylamine (NH2OHmiddotHCl) with 25
(vv) acetic acid (HOAc) at 60degC and pH of ~4 for 6 h and 5) 1 mM barium chloride
(BaCl2) solution at 60degC and pH of ~6 for 6 h Solutions were extracted with a syringe
and passed through a 02 microm filter Solids were inspected after the experiments using the
SEM
66
Batch Experiment Procedures
Batch experiments were performed in triplicate using 1 L polypropylene bottles
and chromite and birnessite ratios listed in Table 31 at ~25degC After loading chromite
and birnessite into each bottle 1 L of a pH 3 solution a 10 mM acetate solution (pH 5) a
10 mM MES solution (pH 67) or a pH 8 solution was added to the appropriate
experiment (Table 31) Chromite and birnessite were also run individually in the acetate
buffered solution (Table 31) Aliquots were extracted from the polypropylene bottles
using a syringe and then passed through a 2 microm filter before analyses Less than 40 mL
was extracted over the course of each experiment resulting in a total volume change of
lt4 After the experiments were terminated the remaining solids were resuspended in
20 mL of 1 mM phosphate (5050 KH2PO4K2HPO4 pH 7) 20 mL of 10 mM CaCl2
solution at pH 7 and a 05 M HCl solution These suspensions were agitated for 3 h
followed by centrifugation and filtration (02 microm) before analyses The purpose of the
phosphate and CaCl2 solutions was to desorb reaction produtcts (eg Cr(VI) Cr(III) Fe
and Mn) from the solid phases The HCl treatment provides a means to evaluate
secondary phases such (oxy)hydroxides formed during the experiments Although the
HCl solution may dissolve chromite and birnessite in addition to the (oxy)hydroxides the
short time frame of the extraction (3 h) should limit the dissolution of chromite
Additionally the polypropylene bottles following the termination of five batch
experiments were rinsed with DDI water and 20 mL of 05 M HCl was added to each
bottle The HCl solution was allowed to remain in the bottles for 15 h in order to
determine the concentration of Cr Fe and Mn sorbed onto the interior lining
Serpentine Soil and Birnessite Experiments
Serpentine soils collected from Jasper Ridge Biological Preserve (Stanford CA)
and New Caledonia were used to determine the influence of birnessite in soils known to
contain chromite (Chapters 2 and 4) The Cr geochemistry of these soils as well as the
detrital chromite have been evaluated in Chapters 2 and 4 and by Becquer et al (2003)
Jasper Ridge serpentine soils contain Cr-bearing minerals including chromite Cr-
magnetite and Cr-chlorite however a majority of the Cr enrichment (Table 32) is
directly related to the abundance of chromite A majority of the Cr related to New
67
Caledonia serpentine soils is attributed to Cr (oxy)hydroxides (Chapter 4) however
chromite is present in these soils as well (Becquer et al 2003) Although birnessite has
been noted in these soils the quantity of birnessite could not be determined In order to
evaluate the impact of Mn oxides on Cr within such soils 01 g of birnessite was added to
each soil
The parameters for the experiments are listed in Table 32 Ten grams of the
lt250 microm size-fraction of each soil (JR1 JR2 and NC) and 100 mL of 1 mM acetate
buffered solution (pH 5) were loaded into individual 150 mL Nalgene bottles Aqueous
solutions were evaluated as a function of time (h) Additionally 10 g of each soil 01 g
of birnessite and 100 mL of 1 mM acetate buffered solution (pH 5) were loaded into 150
mL Nalgene bottles and were allowed to react as a function of time (h) These
experiments were rotated at a rate of 2 revolutions min-1 Solutions were removed with a
syringe and passed through a 2 microm filter before analysis Approximately 8 mL of
solution was required for the analyses over the course of each experiment resulting in a
total volume change of 8 Total Cr concentrations (mg kg-1) of the soils listed in Table
32 were obtained by completely dissolving the samples using a mixture of hot (gt200degC)
concentrated nitric perchloric and hydrofluoric acids Elemental compositions were
measured in the supernatant using inductively coupled plasma optical emission
spectroscopy (ICP-OES) The mass of chromite in each soil was calculated using total Cr
concentration for each soil and the calculations provided in Chapters 2 (Table 210) and
Chapter 4 (Table 47) The chromite surface area of each soil was approximated by
comparing the surface area measured for the chromite standard and its observed grain
size from SEM images (Figure 31) to chromites observed in the soils using backscattered
electron images (Chapters 2 and 4) The surface area of the Jasper Ridge chromite is 066
of the chromite standard due to a slightly larger grain size The surface area of the
New Caledonia chromite is approximately 025 of the chromite standard Chromite
suspension densities were calculated by multiplying the chromite mass by the estimated
surface area and then dividing by the volume of the solution
68
Solution Analysis
The pH and Eh of the supernatant was measured with an Orion flat surface
combination pH electrode inserted directly into supernatant Production of soluble
Cr(VI) was measured spectrophotometrically at 540 nm using the s-diphenyl carbazide
method modified from Bartlett and James (1979) Soluble Fe(II) was monitored using
the ferrozine method adopted from Stookey (1970) Total dissolved Cr Fe Al Mn and
Mg were ascertained using ICP-OES Concentrated trace metal grade nitric acid used to
acidify the samples prior to analyses Aluminum and Mg was determined to be present in
the nitric acid at concentrations lt1 microM Magnesium and Al are not plotted in any of the
figures Chromium(III) and Fe(III) was determined by subtracting Cr(VI) and Fe(II)
obtained from colorimetric methods from total dissolved Cr and Fe analyses using ICP-
AES Results are presented as an average of the three runs for each batch experiment
with the standard deviation noted by error bars
RESULTS
Chromite Extraction Experiments
The release of Cr Fe and Mn from chromite using 10 H2O2 01 M KMnO4 10
M HF acid 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 1 mM BaCl2 solutions was
carried out to determine the relative propensity of chromite to resist chemical dissolution
and these results are summarized in Figure 32 The KMnO4 solution with aqueous
Mn(VII) was the most effective in releasing Cr from chromite yielding a concentration
of 62 Cr mg L-1 Iron released using the KMnO4 solution was 03 mg L-1 Although
H2O2 is a direct chemical oxidant the release of Cr from chromite was less as compared
to the other solutions (only ~03 Cr mg L-1 was measured) Concentrated HF was only
able to release ~13 Cr mg L-1 while releasing 106 Fe mg L-1 NH2OHmiddotHCl with 25
(vv) HOAc and CaCl2 solutions only removed lt1 Cr mg L-1 however the NH2OHmiddotHCl
removed 36 Fe mg L-1 which is comparatively higher than the H2O2 KMnO4 and BaCl2
solutions
69
Chromite and Birnessite Batch Experiments
The interaction of chromite and birnessite leading to the release and oxidation of
Cr(III) is a relatively slow process requiring study on the order of hours and days The
pH was maintained constant throughout the duration of the experiments while Eh values
ranged from 320 to 450 mV with no discernable temporal pattern (Figure 33) For pH
values between 3 and 67 Cr3+ CrOH2+ and Cr(OH)2+ are the dominant aqueous Cr(III)
species however at pH 8 conditions are on the stability line between Cr(III) and Cr(VI)
specifically Cr(OH)3 and CrO42-
Higher chromite suspension densities produced higher rates of Cr release to the
aqueous phase and all the Cr in solution was in the form of Cr(VI) (Figure 35) The
rates of oxidation based on Cr(VI) production range from 00005 to 00041 microM h-1
(Figure 35 Table 33) Comparatively Cr released from the chromite control experiment
is maintained at a relatively constant value of ~005 microM and no Cr(VI) is measured in
solution Iron concentrations initially increase to ~ 05 microM and then decrease rapidly to a
relatively constant concentration at 02 microM (Figure 35) Iron concentrations for the
chromite control experiment began at 15 microM between 0 to 50 h and decreased to ~02
microM Iron(II) and Mn concentrations are below detectable limits for all experiments
(Figure 35) A zeroth-order dependence is assumed with respect to birnessite owing to
its large excess surface area relative to chromite
Variable birnessite suspension densities and constant chromite suspension
densities were evaluated at pH 5 (Figure 36) Although suspension densities of
birnessite are 538 and 1075 m2 L-1 (Table 31) the rate of Cr and Cr(VI) measured in
solution for both experiments is 00041 microM h-1 (Figure 36 and Table 33) Iron
concentrations are below 06 microM and maintain concentrations near 02 microM with no
indication of consistently increasing or decreasing with respect to time Iron(II) and Mn
concentrations in solution are below detectable limits for both experiments
In Figure 37 the influence of pH on chromite oxidation by birnessite was
evaluated by conducting experiments at pH values of 3 5 67 and 8 chromite and
birnessite suspension densities are constant at 0075 m2 L-1 and 1075 m2 L-1 respectively
(Table 31) Chromium and Cr(VI) rates increase with decreasing pH ranging from 0
microM h-1 at pH 8 to 00062 microM h-1 at pH 3 (Table 33) Iron is present in solution at
70
concentrations less than 04 microM for all the experiments Generally Fe concentrations are
greater near the start of the experiment and decrease to values lt 05 microM with increasing
reaction time Manganese concentrations are below detectable limits
In order to determine the rate dependence on each prominent reactant (chromite
birnessite and pH) reaction rates were regressed against reactant concentrations (Figure
38) A pseudo first-order rate dependence is noted based on chromite suspension
density the equation regressed for this relationship is listed in Table 33 As the chromite
suspension density increases the rate of Cr(VI) in solution increases In contrast a
zeroth-order rate dependence is evident for birnessite suspension density (dashed line in
Figure 38b) The rate of Cr(VI) generation is expected to decrease with respect to
birnessite suspension density at very low concentrations (dotted line in Figure 38b)
The influence of pH on the rate of Cr(VI) production is well-described by a polynomial
function (Figure 38c Table 33)
Developing an overall rate expression for Cr in solution requires integrating the
factors assessed in Figure 38 and listed in Table 33 All the Cr measured in solution is
in form of Cr(VI) therefore the rate law for Cr or Cr(VI) in solution may be written as
d[Cr(VI))]dt = k sdotChromitea sdot Birnessiteb sdot (aH+)c (31)
and
log d[Cr(VI))]dt = log k + asdotlogChromite+ bsdot log Birnessite+ csdotlog (aH+) (32)
where k is the overall rate constant for Cr release and oxidation Chromite is the
chromite suspension density (Table 31) Birnessite is the birnessite suspension density
(Table 31) and aH+ is the activity of H+ Constants a b and c are the order of
dependency for each reactant The overall rate equation (developed in the Appendix) can
be expressed as
d[Cr(VI))]dt = krsquo sdotChromite069 (33)
71
where krsquo is 00016sdotpH2 ndash 0025sdotpH + 0097microM L m-2 h-1 Chromite is the chromite
suspension density in m2 L-1 the expression is valid for birnessite suspension densities
greater than 20 m2 L-1 and pH values between 3 to 8
Secondary and Solid Phase Examination of Chromite and Birnessite
The solid products were evaluated to determine the abundance and type of
secondary phases formed during the course of the experiments Solids from the batch
experiment were treated with three solutions 1 mM 5050 KH2PO4K2HPO4 1 mM
CaCl2 and 05 M HCl (Table 34) In the chromite and birnessite reactions Cr was
primarily in solution (gt90 ) The HCl extract was able to remove lt5 additional Cr
while the KH2PO4K2HPO4 and CaCl2 extracts released lt3 additional Cr Aqueous Fe
concentrations were below detectable limits for the final solution extraction Treatment
with the HCl resulted in up to 0127 micromoles of Fe however 0107 micromoles of Fe were
released upon treatment of unreacted chromite with HCl Although not shown in Table
34 Mn values were greater than gt4 micromoles when treated with HCl the
KH2PO4K2HPO4 and CaCl2 removed no detectable Mn
After the termination of each experiment the solids were examined using SEM in
order to identify changes in the surface morphology and secondary phases A large
chromite grain and conglomerate of birnessite adjacent to the chromite are depicted in
Figure 38 The face and edge of the chromite grain is not noticeably altered compared to
grains examined prior the experiments Additionally the particle morphology of
birnessite has not visibly changed Even at higher magnification angular and
conchoidally fractured fine-grained edges are noted (Figure 38bc) Only a slight
rounding of the smaller features of the main chromite grain provides any evidence of
possible chromite dissolution (Figure 38c) Birnessite appears to have limited interaction
with chromite Although chromite and birnessite were allowed to interact over 500 h in
an aqueous solution no secondary phases are observed by SEM
Serpentine Soil and Birnessite Experiments
Two serpentine soils from Jasper Ridge Biological Preserve CA and one
serpentine soil from New Caledonia were used to determine whether or not the overall
72
rate expression (Equation 33) for chromite reaction with birnessite is consistent with
Cr(VI) concentrations produced by serpentine soils containing chromite Three
serpentine soils without birnessite and three soils with the addition of birnessite buffered
at pH 5 were measured for Cr(VI) generation as a function time (Figure 310) The rates
of Cr(VI) measured in solution for both Jasper Ridge soils (JR1 and JR2) are identical at
00013 microM h-1 With the addition of birnessite the rates of Cr(VI) production in solution
both increased to 00044 microM h-1 (JR1 w Bt and JR2 wBt in Figure 310) Based on the
chromite suspension density (Table 32) the rates predicted by Equation 33 range from
00035 to 00041 microM h-1 which are 10 to 20 less than those observed in the
experiments (Figure 310) The rates of Cr(VI) in solution from the New Caledonia
serpentine soil are 00009 microM h-1 and with the addition of birnessite increase up to 00013
microM h-1 (Figure 310) A rate of 00016 microM h-1 was predicted by Equation 33 for the
New Caledonia soil
DISCUSSION
The reaction of birnessite with chromite enhances the rate of Cr release into
solution The rate of Cr(VI) measured in solution (Equation 33) is directly related to the
chromite suspension density and is independent of the birnessite at suspension densities
greater than ~20 m2 L-1 Although the release and oxidation of Cr in solution are
measured the mechanisms and pathways driving this reaction need to be evaluated The
propensity of chromite to resist dissolution the effects of pH the formation of secondary
phases and ultimately the pathways of Cr release and oxidation are addressed in the
following discussion
Extraction experiments performed on chromite (Figure 32) demonstrate it is
resistant to acidic dissolution or competitive ion displacement but strong oxidants such
as high valent Mn may oxidatively dissolve chromite It is evident that the proportions of
Cr Fe and Mn vary for each of the extractions The solutions utilized in the extraction
experiments have varying chemical properties capable of preferentially releasing a
particular species andor element These experiments were also performed over a short
period of time (lt6 h) which may have only released elements from highly reactive sites
73
Finally secondary phases associated with these experiments were possible and would
distort the elemental ratios observed in solution
A majority of the Eh and pH analyses plot in the Cr(III) stability region (Figure
33) however Cr in solution is primarily in the form of Cr(VI) (Figures 35 36 and
37) Eh measurements presented in Figures 32 and 33 do not account for the
microenvironments associated with the surfaces of birnessite Additionally the kinetics
of Cr(VI) reduction are limited in reducing conditions due to electron symmetry
constraints and the absence of a suitable redox couple (Fendorf 1995) Finally highly
oxidizing birnessite in the solution is kinetically and thermodynamically controlling the
oxidation state of Cr relative to the stabilities predicted in Figure 33 (Amacher and
Baker 1982 Fendorf and Zasoski 1992)
A fractional-order dependence on the chromite suspension density (Figure 38)
can be explained by diffusional influences competing oxidation reactions and secondary
reactions involving Cr(III) and Mn(IV) Secondary phases on the surface of chromite
may inhibit the diffusion of Cr from chromite however the clean and relatively unaltered
chromite surfaces do not support this mechanism as a major cause (Figure 38)
Chromium and Fe released from chromite may be incorporated into secondary phases
related to birnessite or chromite Approximately 10 of Cr is not in the aqueous phase
(Table 34) supporting the premise that secondary phases are contributing to the
fractional-order dependence related to the chromite suspension density Also
competition from Fe(II) oxidation on birnessite deposition of Fe(II) on birnessite (or
chromite) and diffusion onwithin birnessite may further impact the rate dependence
The zeroth-order dependence based on the birnessite suspension density (Figure
38) results from its high surface area At very low birnessite suspension densities the
rate of Cr(VI) production should decrease however due to the high surface area of
birnessite it was experimentally prohibitive to determine reaction rates at masses less
than 4 mg Therefore a hypothetical trend-line depicts the expected dependence on
birnessite concentration at very low suspension densities at higher values representative
of most field settings the reaction is independent of the birnessite surface area (ie zero-
order dependence)
74
In Figure 38c the polynomial relationship defines the Cr(VI) rate dependence on
pH As the pH decreases the rate of Cr(VI) increases more rapidly The relationship is
potentially the result of lower pHs progressively increasing the rate of chromite
dissolution However the rate of Cr(VI) in solution might be influenced by higher pHs
increasing the formation of secondary Fe(III) and Cr(III) (oxy)hydroxides on the
chromite andor birnessite surfaces thereby decreasing elemental diffusion from
chromite or decreasing available birnessite sites for Cr(III) oxidation Differentiating
between these two mechanisms is not possible based on the rate equation or extraction
analyses however the polynomial relationship evident in Figure 38c is folded into the
rate constant in order to geochemically account for the influence of pH on Cr(VI)
generation
No conclusive evidence is available to determine the fate of Mn within the
reacting suspensions Aqueous Mn concentrations are consistently maintained below
detectable limits If Mn is released from birnessite then it must be strongly and quickly
incorporated into a nonaqueous phase following its release Due to Cr(III) being oxidized
by Mn(III) andor Mn(IV) Mn(II) must be produced Extractions performed on the
chromite and birnessite solids demonstrate that Mn could only be removed with HCl
however a majority of this Mn is potentially the result of HCl dissolving birnessite On
the basis of electron transfer stoichiometry at least 15 mols of Mn(II) must be produced
for each mol of Cr(VI) generated This amount can be accounted for in the HCl
extraction where gt4 micromoles of Mn were measured Additionally Fendorf et al (1993)
determined that Mn(II) should not significantly influence the extent of Cr(III) oxidation
Iron in solution does not progressively increase in solution over time (Figures 35
36 and 37) The high initial rate of Fe release compared to Cr may be the result highly
reactive sites preferentially releasing Fe Although no Fe is detected in solution from
the extractions performed on the chromite and birnessite solids 0005 to 0035 moles of
Fe were retrieved from the lining of the bottle and 0053 to 0126 moles were released
from the solids using the HCl solution (Table 34) Iron(II) could potentially be oxidized
to Fe(II) in the presence of birnessite and form stable Fe(III)-(oxy)hydroxides (Postma
1985 Myers and Nealson 1988 Villinski et al 2001) A ratio of two Fe for every one
Mn is necessary for the oxidation of Fe(II) and this amount is recoverable based on HCl
75
extractions with gt4 micromoles of Mn measured from the HCl extractions Additionally
dissolved O2 in the batch experiments will rapidly oxidize Fe(II) for all the pHs examined
in these experiments (Stumm and Morgan 1981 Langmuir 1997) The oxidative
pathway for Fe(II) provides an explanation for the lack of Fe(II) in solution for the batch
experiments (Figure 35 36 and 37)
With a detection limit of ~02 microM and equilibrium with respect to Fe(OH)3(am) and
goethite Fe concentrations should either be at or below detection limits at pHs gt 5
Assuming aqueous Fe(III) is present dominantly as the second hydrolysis product at pH
5 the dissolved concentrations approach the solubility of ferrihydrite and are much
greater than expected for goethite (Figure 311) Thus it is probably that the former solid
is generated which is further supported by the chemical extraction experiments (Table
34)
Iron and Cr solution concentrations measured experimentally are not consistent
with congruent dissolution (Figure 312a) Assuming Fe was forming secondary phases
at a constant rate (~00001 Fe h-1) multiplying this rate by the time analysis and then
adding it to the measured aqueous Fe a corrected Fe (Fe) value can be estimated (Figure
312) Additionally the 10 of Cr not in solution can be also be used to calculate a
corrected Cr value (Cr) (Figure 312) When secondary phases are taken into account a
majority of the analyses plot near the congruent dissolution of chromite This provides
evidence that chromite is undergoing congruent dissolution within the batch experiments
but that secondary Fe(III) phases disrupt the solution phase elemental ratios
Chromium(VI) Pathways from Chromite and Birnessite into Solution
Three pathways are evaluated for Cr(VI) measured in solution a) chromite
dissolution b) birnessite dissolution and c) solid-solution reaction (Figure 313) In order
to describe these pathways end-member chromite (FeCr2O4) and birnessite (δ-MnO2) are
used to simplify the analysis It should be recalled however that the composition of the
chromite used in the experiments is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4
Chromite dissolution presents a likely explanation for Cr(VI) observed in solution
based on the following discussion In chromite dissolution model Cr(III) released from
chromite adsorbs onto the surface of birnessite and oxidizes to Cr(VI) After the electron
76
transfer is complete Cr(VI) desorbs from birnessite and remains in solution As this
process continues Cr(VI) increases at a constant rate The rate limiting step with regards
to this specific pathway can be grouped into two steps a) dissolution of chromite (Step
1a) or b) Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) (Figure 313)
Although a dissolution rate for chromite at ambient temperatures and pressures is not
known the chromite extraction experiments (Figure 32) demonstrates that chromite
dissolution is slow Manganese oxides such as birnessite oxidize Cr(III) in solution
rapidly with a measured rate constant of 6300 microM-1 s-1 in fresh water (100 mg L-1 MnO2
100 g L-1 Cr(III) pH~6) (Van der Weijden and Reith 1982) This means that the rate
for Cr(III) adsorption oxidation and desorption (Steps 1b to 1d) is on the order of
seconds therefore chromite dissolution is likely the rate limiting step controlling the
concentration of Cr(VI) in solution for this pathway
Although a zeroth-order dependence related to birnessite concentrations (Figure
37a) is observed the birnessite dissolution pathway in which aqueous Mn4+ released
from birnessite adsorbs onto the surface of chromite oxidizing Cr(III) cannot be entirely
dismissed Manganese dissolution and adsorption could occur quickly not allowing the
Mn to be measured in solution Examination of the solids after the termination of each
experiment demonstrated extremely minimal chromite and birnessite interaction
therefore the solid solution reaction is not a likely pathway leading to the production of
Cr(VI) in solution and is furthermore inconsistent with the noted rate dependencies
Despite the three pathways (Figure 313) the generation of Cr(VI) observed in these
experiments and described by the rate equation (Equation 33) is independent of these
potential pathways
Serpentine Soils and Cr(VI)
The serpentine soil experiments (Figure 310) provide direct evidence that Cr(VI)
is forming within these soils By adding birnessite to the soil the rate of Cr(VI) in
solution increased for all experiments (00013 to 00044 microM Cr(VI) h-1 for Jasper Ridge
and 00009 to 00013 microM Cr(VI) h-1 for New Caledonia) The Cr(VI) rate increase
related to the addition of birnessite is likely the result of low birnessite suspension
densities (lt20 m2 L-1) in the soil Rates predicted for Jasper Ridge and New Caledonia
77
soils using Equation 33 and values in Table 32 are similar to those measured in the
experiments with values of 00041 and 00016 microM Cr(VI) h-1 respectively Overall these
soil experiments demonstrate that a rate equation developed from laboratory batch
experiments predicts the rate of Cr(VI) in serpentine soils reasonably well
According to limits set by the Environmental Protection Agency the California
standard for total Cr in drinking water is 50 microg L-1 (~1 microM) while the federal standard is
100 microg L-1 (~2 microM) The serpentine soil and birnessite experiments produced Cr(VI)
concentrations in excess of 50 microg L-1 after only 140 h or roughly 6 days Serpentine soils
at Jasper Ridge typically remain water saturated for prolonged periods of time (gt1 week)
during the winter season increasing the time Cr(VI) can be produced The degree and
duration of water saturation is dependent on the topography of the serpentinite body
level areas will remain water saturated while sloping areas will allow the water to flow
down gradient Although Cr(VI) was generated in the soil experiments the production of
Cr(VI) in serpentine soils may be more limited The reduction of Cr(VI) to Cr(III) is
favorable by reactions with Fe(II)aq Fe(II)-containing minerals such as magnetite H2S
and organic matter including microbial processes (Ilton and Veblen 1994 Fendorf 1995
Peterson 1996) Both organic matter and magnetite are abundant in Jasper Ridge
serpentine soils (Chapters 2 and 4) however these factors did not deter Cr(VI) from
forming during the serpentine soil experiments Additionally the soils in these
experiments were constantly rotated allowing the mineral surfaces to be continuously
exposed to solution Chromite within an undisturbed soil will not always be in contact
with the solution or proximal to birnessite
Chromium(VI) produced by the interaction of chromite and birnessite is not
limited to serpentine soils Serpentinites the bedrock for serpentine soils are typically
highly sheared due to the diapiric migration of serpentinite masses along zones of
weakness in fault zones Numerous cracks and lineations which are typical of these rocks
provide a means for Cr(VI) to transported away from the serpentine soil potentially into
local groundwater In fact preferential flow will occur along the major shear plains of
the serpentinite Additionally these shear plains are often coated with high valent Mn
oxides providing a reaction pathway for Cr(VI) formation Serpentinite springs such as
those evaluated by Barnes and ONeil (1969) and Barnes et al (1972) at the Cedars
78
ultramafic body in Cazadero CA have Cr concentrations measured at 20 microg L-1 in which
all the Cr is in the form of Cr(VI) (Chapter 4) In conclusion chromite and birnessite in
serpentine soils could contribute a source of Cr(VI) in surface waters and groundwaters
Chromium release and oxidation from chromite is increased by the presence of
birnessite and the interaction between chromite and birnessite is a pathway for Cr(VI)
generation in ultramafic sediments and soils The potential of serpentine soils as a source
of non-anthropogenic Cr contamination can now be evaluated quantitatively based on the
abundance of chromite The overall rate law for Cr(VI) in solution (Equation 39)
provides a means to estimate the Cr(VI) concentration attainable for serpentine soils and
serpentine sediments when the chromite and birnessite suspension densities and the pH of
the soil are known
ACKNOWLEDGEMENTS
We gratefully acknowledge comments and assistance from Robert Jones and Dr
Guangchao Li Funding for this research was provided by the Office of Teaching and
Learning at Stanford University
79
Table 31 Chromite and birnessite batch experiment parameters with chromite and birnessite suspension densities
Experiment Chromitea (grams)
Birnessiteb (grams) pHc
Chromite Surface Aread
(m2 g-1)
Birnessite Surface Aread
(m2 g-1)
Chromite Suspension
Density (m2 L-1)
Birnessite Suspension
Density (m2 L-1)
A 08 043 5 025 125 02 54 B
08 086 5 025 125 02 108C 03 086 5 025 125 008 108D 008 086 5 025 125 002 108E 004 086 5 025 125 001 108F 03 086 3 025 125 008 108G 03 086 67 025 125 008 108H 03 086 8 025 125 008 108
Chromite 08 ---- 5 025 125 02 ----Birnessite ---- 043 5 025 125 ---- 5480
a Chromite chemical composition is (Fe046Mg052Mn002)(Cr061Al029Fe010)2O4 b Synthetic birnessite (δ-MnO2) c10 mM acetate buffered solution pH = 5 non-buffered pH=3 5 mM MES buffer pH 67 non-buffered pH 8 dMeasured BET analyses
81
Table 32 Serpentine soil and birnessite experiments with soil and birnessite masses used and the estimated chromite suspension density
Samples Soil Mass (g)
Birnessite Mass (g)
Volume (L) of Solutiona
Total Crsoil
b (mg kg-1)
Chromite Mass (g) in
Soil Samplec Estimated Surface
Areac (m2 g-1)
Chromite Suspension Density
(m2 L-1) in Soild Jasper Ridge
(JR1) 10 ---- 01 5976 013 015 02 Jasper Ridge (JR1 w Bt) 10 01 01 5976 013 015 02 Jasper Ridge
(JR2) 10 ---- 01 4978 011 015 016 Jasper Ridge (JR2 w Bt) 10 01 01 4978 011 015 016
New Caledonia (NC) 10 ---- 01 827 0015 006 001
New Caledonia (NC w Bt) 10 01 01 827 0015 006 001
a1mM Acetate Buffer (pH=5) 81 bTotal Cr concentrations were obtained through total digestion and ICP-AES analyses
cEstimated chromite mass and surface area (discussed in the Methods section) was made using chromite observations and calculations in Chapters 2 and 4
dCalculated from the chromite mass in the soil sample estimated surface area of the chromite and the volume of solution
82
Table 33 Summary of Cr(VI) rates obtained from batch experiments (Figure 35 36 and 37) and used in Figure 38
Listed in Table 31
Suspension Densities (m2 L-1)
Chromite (Ch)
Birnessite (Bt)
pH Rate (microM h-1) (d[Cr(VI)]dt)
log Rate (log microM h-1) (log d[Cr(VI)]dt) Expression
Figure 35 02 108 5 00041 -239
008 108 5 00021 -268 002 108 5 00009 -305 001 108 5 00005 -330
1
Figure 36 02 54 5 00041 -239 02 108 5 00041 -239 2
Figure 37 008 108 3 00062 -221 008 108 5 00021 -268 008 108 67 00004 -340 008 108 8 0 ----
3
Expression 1 log d[Cr(VI)]dt = asdot log Chromite +log krsquorsquo log d[Cr(VI)]dt = 06908 log Chromite ndash 1899 Expression 2 log d[Cr(VI)]dt = bsdot log Birnessite + log krsquorsquorsquo
log d[Cr(VI)]dt = 0sdot log Birnessite ndash 23872 log d[Cr(VI)]dt = -23872
Expression 3 d[Cr(VI)]dt = 264E-4sdotpH2 ndash 414E-3sdotpH + 162E-2
82
Table 34 Extractions comparing total solution and extractable Cr and Fe from the solids
------------------Cr----------------- ------------------Fe---------------- microMolesdagger Total
mole Extraction
mole microMoles Total mole
Extraction mole
Chromite Ch 02 (Table 31) Terminated at 30 days Solutiona 0033 554 ---- 0 0 ---- Phosphateb 0000 05 1 0000 03 03 CaCl2
c 0013 388 87 0000 00 00 HCld 0002 53 12 0103 997 997 Bottle HCle 0000 00 0 0000 00 00
A Ch 02 Bt 54 (Table 31) Terminated at 30 days Solution 2977 910 ---- 0 0 ---- Phosphate 0048 16 18 0004 21 21 CaCl2 0055 18 20 0002 09 09 HCl 0161 54 60 0126 758 758 Bottle HCl 0004 01 2 0035 212 212
B Ch 02 Bt 108 (Table 31) Terminated at 30 days Solution 2837 914 ---- 0 0 ---- Phosphate 0039 14 16 0008 49 49 CaCl2 0060 21 25 0000 00 00 HCl 0142 50 58 0127 803 803 Bottle HCl 0004 01 2 0024 148 148
C Ch 08 Bt 108 (Table 31) Terminated at 17 days Solution 0877 894 ---- 0 0 ---- Phosphate 0020 23 22 0002 37 37 CaCl2 0030 34 32 0000 00 00 HCl 0040 46 43 0053 881 881 Bottle HCl 0003 03 3 0005 82 82
daggerMolarity values are normalized to the final solution analysis of each experiment based on volume of solution and mass of the sample in each experiment aSolution refers to the final solution analysis before the experiment was terminated bmicroMoles extracted from the batch experiment solids using a 1 mM phosphate buffer pH 7 3 h cmicroMoles extracted from the batch experiment solids using a 10 mM CaCl2 solution pH 7 3 h dmicroMoles extracted from the batch experiment solids using a 05 N HCl solution 3 h emicroMoles extracted from the batch experiment bottles using a 05 N HCl solution 15 h
83
Figure Captions
Figure 31 Scanning electron microscopic (SEM) images of chromite and birnessite at
varying magnifications prior to the batch experiments
Figure 32 A summary of Cr Fe and Mn released from chromite during extraction
experiments Solutions used for these experiments include 1) 10 H2O2 solution 2) 01
M KMnO4 solution 3) 10 M HF 4) 1 M NH2OHmiddotHCl with 25 (vv) HOAc and 5) 1mM
BaCl2 solution () Fe concentration for the HF extraction is 106 mg L-1
Figure 33 pH and Eh measurements as a function of time (h) for the batch
experiments Chromite (Ch) and birnessite (Bt) suspension densities used for the batch
experiments (Table 31) are listed adjacent their corresponding symbol in the key
Figure 34 Eh and pH batch experiment analyses (Figure 33) plotted on a stability field
diagram for Cr (log a Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom
(1998)
Figure 35 Batch experiments evaluating variable chromite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr and Cr(VI) measured in solution are noted and illustrated with a dotted line and
included in Table 33
Figure 36 Batch experiments evaluating variable birnessite suspension densities at pH
5 Chromite (Ch) and birnessite (Bt) suspension densities are listed in the key Rates of
Cr(VI) measured in solution are denoted and included in Table 33
Figure 37 Influence of pH on Cr(VI) generation by reaction of constant chromite (Ch
08 m2 L-1) and birnessite (Bt 108 m2 L-1) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key Rates of Cr(VI) measured in solution are included in Table
33
84
Figure 38 Rates of Cr(VI) in solution (Table 33) as a function of a) chromite
suspension density (Figure 35) b) birnessite suspension density (Figure 36) and c) pH
(Figure 37) The line in each diagram represents the best fit for these data points and the
equations for these lines are listed in Table 33 The dotted line deviating from the
dashed line in b) indicates the probable path at low birnessite concentrations
Figure 39 SEM images of the solid phases after the completion of the batch
experiments A (Ch 02 Bt 54) and B (Ch 02 Bt 108) are shown Chromite is the larger
grain with relatively flat surfaces whereas birnessite is the fine-grained rosettes located
adjacent the larger chromite grains
Figure 310 Chromium and Cr(VI) concentrations (microM) in solution as a function of
time (h) from serpentine soils obtained at Jasper Ridge (JR1 and JR2) and New Caledonia
(NC) Soils with the addition of 01 g birnessite are noted in the figure Soil and
birnessite masses used for each experiment are listed in Table 32
Figure 311 Iron analyses as a function of time (Figure 35 and 36) with the solubility
of amorphous Fe(OH)3 and goethite and the dominant Fe3+-OH complex at pH 5 Fe
solubilities are from Langmuir (1997) Chromite (Ch) and birnessite (Bt) suspension
densities are listed in the key
Figure 312 a) Experimental values of Fe and Cr (Figures 35 and 36) are plotted A
dotted line denotes the ratio of Fe and Cr necessary for chromite congruent dissolution b)
Corrected FeCr ratios as a function of time (h) for data from Figures 35 and 36 taking
into account the concentration of Fe and Cr in secondary phases calculated from Table
34 c) Corrected Fe and Cr as compared to values predicted with congruent dissolution
of chromite
85
Figure 313 Three potential pathways of Cr release and oxidation for chromite-
birnessite interactions are listed
86
Figure 31
87
0
2
4
6
8
10
12
Hydrogen
Peroxid
e
K-Perman
ganate
Hydroflu
oric Acid
Hydroxy
lamine
Barium Chlorid
e
Con
cent
ratio
ns (m
g kg
-1)
Cr
Fe
Mn
Figure 32
88
Time (h)0 100 200 300 400 500
pH
2
3
4
5
6
7
8
9
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
Time (h)0 100 200 300 400 500
Eh (m
V)
100
200
300
400
500
600
Ch Bt ------(m2 L-1)------
Figure 33
89
pH2 3 4 5 6 7 8 9
Eh (m
V)
-05
00
05
10
15
02 54 02 108 008 108 002 108 001 108 008 108 008 108 008 108 02 --- --- 54
HCrO4-
CrO42-
Cr3+
CrOH2+
Cr(OH)3Cr(
OH
) 2+
Upper limit of water stability
Lower limit of water stability
Ch Bt ------(m2 L-1)-----
Figure 34
90
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 108 008 108 002 108 001 108 02 --- --- 54
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
00041 microM h
-1
00021 microM h-1
00009 microM h-1
00005 microM h-1
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
0
1
2
3
4
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
0
1
2
3
4
pH 5 pH 5
pH 5 pH 5 pH 5
Ch Bt ---(m2 L-1)---
Figure 35
91
Time (h)
0 100 200 300 400 500
Cr c
once
ntra
tion
( microM
)
00
05
10
15
20
25
02 54 02 108
00041 microM h
-1
Time (h)
0 100 200 300 400 500
Cr(
VI) c
once
ntra
tion
( microM
)
00
05
10
15
20
25
Time (h)
0 100 200 300 400 500
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
Time (h)
0 100 200 300 400 500
Mn
conc
entr
atio
n ( micro
M)
00
02
04
06
08
10
00041 microM h
-1
pH 5
pH 5 pH 5 pH 5
pH 5 Ch Bt----(m2 L-1)----
Figure 36
92
Time (h)
0 50 100 150 200 250
Cr c
once
ntra
tion
( microM
)
00
02
04
06
08
10
pH 3pH 5pH 67pH 8
Time (h)0 50 100 150 200 250
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe c
once
ntra
tion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250M
n co
ncen
trat
ion
( microM
)
00
02
04
06
08
10
Time (h)
0 50 100 150 200 250
Fe(II
) con
cent
ratio
n ( micro
M)
00
02
04
06
08
10
00004 microM h-1
00021 microM h-1
0006
2 microM h
-1
00021 microM h-100
062 micro
M h-1
00004 microM h-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Ch 08 m2 L-1 Bt 108 m2 L-1 Ch 08 m2 L-1
Bt 108 m2 L-1 Ch 08 m2 L-1 Bt 108 m2 L-1
Figure 37
93
log Birnessite m2 L-1
05 10 15 20 25 30
log
d[C
r(VI
)]dt
microM
h-1
-40
-35
-30
-25
-20
pH 5
log Chromite m2 L-1-22 -20 -18 -16 -14 -12 -10 -08 -06
log
d[C
r(VI
)]dt
microM
h-1
-34
-32
-30
-28
-26
-24
-22
pH 5
pH1 2 3 4 5 6 7 8 9
d[C
r(VI
)]dt
microM
h-1
0000
0002
0004
0006
0008
0010
a)
c)
b)
Ch 008 m2 L-1 Bt 108 m2 L-1
Figure 38
94
Figure 39
95
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR1JR1 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
02
04
06
08
10
JR2JR2 w Bt
Time (h)
0 50 100 150
Cr(
VI) c
once
ntra
tion
( microM
)
00
01
02
03
04
05
NCNC w Bt
00044 microM h-1
00013 microM h-1
00044 microM h-1
00013 microM h-1
00013 microM h-1
00009microM h-1
pH 5
pH 5
pH 5
Figure 310
96
Time (h)0 100 200 300 400 500
log ΣF
e (aq)
microM
-10
-8
-6
-4
-2
0
2
4
02 54 02 108 008 108 002 108 001 108
Fe(OH)3(am)
Goethite
pH 5Ch Bt----m2 L-1----
Fe(OH)2(aq)+
Figure 311
97
Time (h)
0 100 200 300 400
FeC
r Rat
io
0
1
2
3
4
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
Cr
00 05 10 15 20
Fe
00
02
04
06
08
10
02 54 02 108 008 108 002 108 001 108
Chrom
ite C
ongr
uent
Dissolu
tion
Chrom
ite C
ongr
uent
Dissolu
tion
a)
b)
c)
Ch Bt----(m2 L-1)----
ChromiteCongruent Dissolution
Figure 312
98
Pathway 1 Chromite Dissolution
1a) Cr dissolution from chromite FeCr2O3(s) + 4H+ rarr 2Cr(OH)2+
aq + Fe2+(Oxyhydroxide) + H2O
1b) Adsorption of CrOH2+ on the surface of birnessite Cr(OH)2+
aq + MnO2(s) rarr Cr(III)-MnO2(s) + OH- 1c) Oxidation of CrOH2+ on the surface of birnessite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
1d) Desorption of Cr(VI) from birnessite Cr(VI)-Mn(II) + 4H2O rarr Mn2+
(aq) + HCrO4-(aq) + 7H+
Pathway 2 Birnessite Dissolution 2a) Mn4+ dissolution from Birnessite MnO2(s) + 4H+ rarr Mn4+
(aq) + 2H2O 2b) Mn4+ adsorption on the surface of chromite Mn4+
(aq) + FeCr2O4(s) rarr Mn(IV)-FeCr2O4 2c) Oxidation of Cr(III) on the surface of chromite
1st electron transfer Cr(III)-Mn(IV) rarr Cr(IV)-Mn(III) 2nd electron transfer Cr(IV)-Mn(IV) rarr Cr(V)-Mn(III) Cr(IV)-Mn(III) rarr Cr(V)-Mn(II) 3rd electron transfer Cr(V)-Mn(III) rarr Cr(VI)-Mn(II)
2d) Desorption of CrO24- from chromite
Cr(VI)-Mn(II) + 4H2O rarr Mn2+(aq) + HCrO4
-(aq) + 7H+
Pathway 3 Solid Solution Reaction 3) Solid solution interaction between chromite and birnessite
Figure 313
99
CHAPTER 4
CHROMIUM GEOCHEMISTRY OF SERPENTINE SOILS
ABSTRACT
Serpentine soils derived from the weathering of ultramafic rocks mainly
ophiolitic serpentinites are characterized by Cr concentrations in excess of 200 mg kg-1
higher concentrations Fe Ni Co and Cd low concentrations of plant nutrients and
higher concentrations of ammonium acetate extractable Mg compared to Ca The
chemistry of Cr in serpentine soils including its protolith is reviewed focusing on
serpentine soils collected from New Caledonia Oregon and California Overall
serpentine soils are slightly acidic (average pH of ~6) contain a variety Fe(III) oxides
(magnetite and hematite) phyllosilicates (serpentine and chlorite) and clays (smectites
and vermiculites) and contain concentrations of Cr (gt200 mg kg-1) Ni (gt1000 mg kg-1)
and Mn (gt200 mg kg-1) exceeding values of non-serpentine soils Although Cr
concentrations in serpentine soils have been reported as high as 6 Wt in New
Caledonia Cr in New Caledonia Oregon and California serpentine soils evaluated in this
study range from 827 to 9528 mg kg-1 Chromium(III) is the only valence state observed
in the serpentine soil solids however Cr(VI) has been identified in New Caledonia and
California serpentine soil solutions at concentrations below 30 microM The enrichment and
range of Cr concentrations in serpentine soils are directly related to the presence of Cr-
spinels specifically chromite and Cr-magnetite These spinels are resistant to weathering
and are preserved in the soil environment however oxidation of Cr(III) from Cr-spinels
by high valent Mn oxides or other strong oxidants is a potential source of Cr(VI)
identified in serpentine soil solutions Due to the chemically and physically resistant
nature of the Cr-spinels Cr-bearing silicates including clay minerals Cr-chlorite Cr-
garnet Cr-mica and Cr-epidote are more viable sources of Cr identified in vegetation
soil extractions soil solutions and related waters
100
INTRODUCTION
Serpentine soils a generic term used to describe any soil derived from an
ultramafic rock (peridotites and pyroxenites) or serpentinite (metamorphosed ultramafic
rock) are often characterized by 1) low concentrations of plant nutrients such as N P
and K 2) high concentrations of biologically toxic elements including Ni Co and Cd 3)
low ammonium acetate extractable CaMg ratios and 4) unique flora and physical
properties (Brooks 1987) Measured Cr concentrations in serpentine soils range from
~200 mg kg-1 to 6 Wt values enriched compared to non-serpentine soils lt200 mg kg-
1 (Kabata-Pendias and Pendias 1984 Schwertmann and Latham 1986 Brooks 1987)
Chromite and Cr-magnetite are common primary sources of Cr in these soils (Rabenhorst
et al 1982 Schwertmann and Latham 1986 Alexander et al 1989 Gough et al 1989
Gasser and Dahlgren 1994 Kaupenjohann and Wilcke 1995 Becquer et al 2003)
however the plethora of factors relating the chemistry of Cr in serpentine soils and the
weathering of chromite has not been fully addressed using an integrative approach
Overall serpentinites and serpentine soils cover ~1 of the earthrsquos exposed
surface and these rocks and soils are commonly associated with ophiolite complexes
(Coleman 1977) Serpentinites and serpentine soils related to ophiolite complexes are
the primary interest of this paper A map illustrating the worldwide distribution of major
ophiolitic serpentinite bodies and their related soils is shown in Figure 41 A majority of
the sites are near convergent boundaries (Figure 41) Although these rocks and soils
cover a minimal area of the earthrsquos surface they are present in populated areas within the
Circum-Pacific Margin and the Mediterranean Due to the wide-distribution and
occurrence of serpentinites and serpentine soils weathering of these rocks and soils will
differ from location to location due to varying climatic conditions as well as the nature of
the parent material and other factors including topography biota and time (Proctor and
Nagy 1992) As a result the chemistry of these soils as well as the nature of Cr in the
serpentinites and serpentine soils will vary from site to site potentially allowing Cr to be a
source of non-anthropogenic contamination
The objective of this paper is to review the chemical characteristics of serpentine
soils globally evaluate mineralogical and compositional variations between several soils
derived from ultramafic sources and assess the geochemistry of Cr in diverse soil-
101
forming environments Additionally chemical and spectroscopic analyses were
performed on soils derived from ultramafic rocks collected from 1) New Caledonia
South Pacific 2) Eight Dollar Mountain Selma OR 3) Nickel Mountain Riddle OR 4)
Harvard Mine Jamestown CA 5) Jasper Ridge Biological Preserve Stanford CA and 6)
Pillikin Mine El Dorado CA shown geographically in Figure 41 A basalt-derived soil
from Hawaii is included in this study in order to compare a soil formed from a mafic
protolith to those from ultramafic protoliths The various field sites are shown in Figure
42 These soils allow a comparative examination of how soil-forming factors such as
climate topography biota parent material and time affect the chemistry of Cr
REVIEW
Chromium Geochemistry
Chromium is a group VIB transition metal with two common oxidation states
Cr(III) and Cr(VI) differentiated by noticeably different chemical characteristics and
toxicities It should be noted that Cr(II) has been observed in lunar basalt olivines and in
high temperature terrestrial olivines (Sutton et al 1993 Li et al 1995) however this
oxidation state is not observed in soil environments A majority of Cr commonly persists
in ultramafic rocks serpentine soils and related waters as Cr(III) a non-hazardous
species and a micronutrient in human nutrition (Daugherty 1992 Cohen et al 1993
Fendorf 1995 James et al 1997) Chromium(III) has the highest crystal field
stabilization energy for all transition metals strongly favors octahedral coordination and
has 3 d-electrons with a high spin state Chromium(III) is strongly hydrolyzed in
aqueous solutions and the predominant species based on pH redox potential and O2(aq)
are shown in Figure 43 Chromium(III) species are more prevalent under reducing
conditions and its mobility is increased only under strongly acidic conditions At pH
values greater than 9 the hydrolyzed Cr(III) species is negatively charged demonstrating
the amphoteric properties of Cr(III) (Figure 43) Due to the factors described above
Cr(III) forms low solubility compounds and adsorbs strongly on mineral surfaces and
organic matter at pHgt4 (Rai and Zachara 1988 Rai et al 1989 Fendorf 1995) For
example the adsorption of Cr(III) on HFO (hydrous ferric oxide) as a function of pH is
shown in Figure 43 illustrating an adsorption edge at a pH of ~4 (Leckie et al 1984
102
Stumm 1992) Chromium(III) oxidation to Cr(VI) has been observed in reactions
involving hydrogen peroxide (H2O2) and high valent Mn oxides (Eary and Rai 1987
Johnson and Xyla 1991 Fendorf et al 1992 Fendorf et al 1993 Rock et al 2001)
Chromium(VI) is a strong oxidizing agent a highly corrosive substance a toxin to
living cells and a Class A human carcinogen by inhalation (Daugherty 1992 Cohen et
al 1993 Fendorf 1995) The hydrolyzed Cr(VI) species are also shown in Figure 43
with respect to pH redox potential and O2(aq) Chromium(VI) is more favorable under
oxidizing conditions and its mobility is enhanced under basic conditions In solution
Cr(VI) is a weakly adsorbing (oxy)anion (HxCrO42-x) on common weathering products
meaning that it forms highly labile complexes that can be readily displaced by ions
including phosphate and sulfate (Rai and Zachara 1988 Fendorf 1995 Ball and
Nordstrom 1998) Additionally the adsorption of CrO42- on HFO shown in Figure 43
demonstrates the anionic character in which the mol bound of CrO42- increases with
decreasing pH and an adsorption edge is at pH ~72 (Leckie et al 1980 Stumm 1992)
The reduction of Cr(VI) to Cr(III) is known to occur by reactions with Fe(II)aq Fe(II)-
containing minerals such as magnetite H2S and organic matter including microbial
processes (Ilton and Veblen 1994 Fendorf 1995 Peterson 1996)
Mineralogy of Serpentine Soils
Serpentine soils have been identified on every continent excluding Antarctica
(Figure 41) The occurrence and distribution of ultramafic rocks specifically related to
ophiolites are closely associated with convergent plate boundaries (Figure 41 Coleman
1977 Coleman and Jove 1992) Tectonic blocks of ultramafic rock are hypothesized to
be detached from the down-going slab of oceanic crust and incorporated into the
subduction meacutelange Subsequent serpentinization of peridotite forms low density
serpentine group minerals (245-250 g cm-3) resulting in the upward diapiric migration of
the serpentinite masses along zones of structural weakness (fault zones) and through the
denser enclosing subduction meacutelange (~265 g cm-3) For example North America has
numerous exposed ophiolites containing serpentinite located in western California and
Oregon where subduction zones were once active in the Early Cretaceous (Harden
1998)
103
Serpentinization of ultramafic rocks is a significant factor controlling mineral
assemblages and textures in serpentinite Hydration and oxidation of olivine and
pyroxene during serpentinization forms serpentine (lizardite chrysotile and antigorite)
talc chlorite tremolite brucite and magnetite (Coleman 1977 Brooks 1987 Coleman
and Jove 1992 Malpas 1992) Additionally magnetite formation creates favorable
conditions for native metal compounds (such as Fe Co and Ni alloys) to form (Coleman
and Jove 1992) As a serpentinite mass is emplaced the extensive physical and chemical
alteration of the serpentinite results in numerous crosscutting alteration veins and highly
sheared textures Ultimately the mineral assemblages and textures in serpentinites are
complex prior to pedogenesis due to the multiple phases of metamorphosis tectonic
deformation and weathering of the rock
Serpentine soils consist of a mixture of residual minerals from the soil protolith
eolian and illuvial minerals and minerals formed within the soil Additionally highly
weathered serpentine soils enriched in Fe-oxides are also termed laterites soils common
in tropical and subtropical regions Soil protolith minerals may include primary minerals
of ultramafic rocks including olivines pyroxenes and oxides (chromite and magnetite)
as well as a large variety of metamorphic minerals the most common being serpentine
(lizardite chrysotile and antigorite) chlorite talc and actinolite Silica phases occur as
eolian grains concretions andor net type structures or silica webs (Hotz 1964
Coleman 1977 Schwertmann and Latham 1986) Eolian and illuvial aluminosilicate
minerals are important sources of Al in serpentine soils an element of low abundance in
ultramafic rocks
Layer silicates observed in serpentine soils are varied in type and abundance
They include smectites (often as montmorillonite) vermiculite chlorite brucite
serpentine and a variety of intrastratified clays (Veniale and Van der Marel 1963 Hotz
1964 Schellmann 1964 Wildman et al 1968 Rabenhorst et al 1982 Schreier et al
1987 Graham et al 1990 Bulmer et al 1992 Gasser and Dahlgren 1994 Lee et al
2003) Compositional variations crystal structures and parageneses of layer silicates
have been characterized for a number of serpentine soils (Wildman et al 1968
Rabenhorst et al 1982 Schreier et al 1987 Graham et al 1990 Bulmer et al 1992
Lee et al 2003) In addition to the layer silicates goethite hematite maghemite and
104
amorphous Fe (oxy)hydroxides are soil-forming phases in serpentine soils (Coleman and
Jove 1992) Gibbsite is not a common phase in most serpentine soils however limited
isomorphic substitution of Al(III) in goethite and hematite was noted in New Caledonia
(Schwertmann and Latham 1986) Manganese oxides such as birnessite (δ-MnO2) are
also present in serpentine soils and often coat mineral surfaces
Sources of Chromium for Serpentine Soils
Chromium primarily occurs in oxide and silicate phases in ultramafic rocks and
serpentinites Chromium-spinel often referred to as chromite is a common primary
(igneous) spinel in mafic and ultramafic rocks and often resists low-grade metamorphic
processes related to serpentinization (Hoffman and Walker 1978 Malpas 1992)
Isomorphic substitution of Al3+ Fe3+ and Ti4+ into the octahedral sites and Mg2+ Ni2+
Zn2+ and Mn2+ into the tetrahedral sites is common however end-member chromite
(FeCr2O4) is atypical (Cooper 1980 Raase et al 1983 Von Knorring et al 1986
Treloar 1987b Pan and Fleet 1991 Sack and Ghiorso 1991 Burkhard 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996 Mathiesen 1999) Serpentinite chromite compositions reported in the literature are
plotted in Figure 44 to demonstrate the wide-range of Cr concentrations possible for Cr-
spinels Isomorphic substitution of octahedral Cr and octahedral Al for these chromite
compositions are also shown in Figure 44 Serpentinite pyroxenes containing Cr include
augite (le1 Cr Wt ) and enstatite (le10 Cr Wt ) (Grapes 1981 Von Knorring et al
1986 Treloar 1987a Challis et al 1995 Sanchez-Vizcaino et al 1995 Deer et al
1996)
Magnetite and serpentine minerals may incorporate varying amounts of Cr into
their structures during serpentinization Chromium(III) from primary Cr-containing
phases may isomorphically substitute into the Fe(III) site of magnetite due to size
(octahedral radii Cr3+ = 0615 Aring and Fe3+=0643 Aring) and charge similarities Chromium
concentrations in magnetite have been observed as high as 13 Wt (Oze et al 2003)
Chromate minerals such as crocoite (PbCrO4) are extremely rare in serpentinites and the
occurrence of these minerals are mainly limited to specific localities such as in Australia
Tasmania and the Urals (Williams 1974 Crane et al 2001) Serpentine minerals have
105
been reported to contain as much as 1500 mg kg-1 Cr based on spectrographic analyses
by Page and Coleman (1967) however serpentine minerals typically contain only trace
amounts of Cr
Hydrothermal and CO2 metasomatism of ultramafic rocks during the evolution of
convergent plate margins commonly produce Cr-silicate minerals including
fuchsitemariposite (Cr-muscovite lt19 Cr Wt ) uvarovite (Cr-garnet lt20 Cr Wt )
tawmawite (Cr-epidote lt13 Cr Wt ) and kaumlmmererite (Cr-chlorite lt10 Cr Wt ) as
shown in Figure 44 (Whitmore et al 1946 Leo et al 1965 Chen and Lee 1974 Neiva
1978 Cooper 1980 Phillips et al 1980 Grapes 1981 Schreyer et al 1981 Max et al
1983 Nutman et al 1983 Raase et al 1983 Von Knorring et al 1986 Kerrich et al
1987 Treloar 1987b Treloar 1987a Morand 1990 Pan and Fleet 1991 Schandl and
Wicks 1993 Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995
Mathiesen 1999 Proenza et al 1999) Figure 44 illustrates the relationship of Cr and
Al substitution for Cr-silicate compositions based on Cr and Al Wt Additionally Cr-
silicate compositions are plotted based on octahedral Cr and octahedral Al in Figure 44
Compositions for Cr-muscovite Cr-epidote and Cr-garnet plot near the tie lines
connecting Cr and Al end-member minerals Chromium isomorphically substitutes for
octahedral Al in these silicates due to similarities in charge octahedral preference and
octahedral radii (Al3+ = 061 Aring) The numerous isomorphic substitutions possible for Cr
and Al in the chlorite structure results in analyses that do not plot near the tie line
connecting clinochlore and kaumlmmererite Overall any of these minerals are potential
sources for Cr in soils produced by the weathering of serpentinites and their hydrothermal
and CO2 metasomatic derivatives
Chromium in Serpentine Soils
Chromium concentrations vary widely in serpentine soils worldwide (Table 41)
New Caledonia soils appear to be the most Cr-enriched serpentine soil with a Cr
concentration of ~60000 mg kg-1 (Schwertmann and Latham 1986) however Jaffre
(1980) report values ranging from 6300 to 80000 mg kg-1 for New Caledonia soils
Some serpentine soils have measured Cr concentrations as low as 29 and 66 mg kg-1
atypical of most values reported in Table 41 (Proctor and Woodell 1971 Brooks et al
106
1992 Jeffrey 1992) Chromium concentrations as a function of depth for serpentine soil
profiles in Maryland (Rabenhorst et al 1982) Tehama County CA (Gough et al 1989)
the Klamath Mountains CA (Hotz 1964) Newfoundland Canada (Roberts 1992)
Harvard Mine CA (this study) and Jasper Ridge CA (Oze et al 2003) are reported in
Figure 45 The pedogenic process(es) responsible for the varying Cr concentrations as a
function of depth for each soil have not been characterized however Oze et al (2003)
determined that a majority of Cr in Jasper Ridge serpentine soils is the result of
weathering-resistant residual Cr-bearing minerals from the protolith mainly chromite
and Cr-magnetite Additionally Oze et al (2003) demonstrated that soil samples from
Jasper Ridge contain higher concentrations of Cr with respect to their corresponding
bedrock due to the Cr-spinels accumulating over time The observation that Cr
concentrations decrease with depth towards the protolith is characteristic for most of the
profiles in Figure 45
In addition to examining total Cr concentrations chemical extractions performed
on serpentine soils provide additional information with regards to Cr availability and the
minerals and complexes in which Cr may reside in the soils Table 42 summarizes a
number of extraction experiments performed on serpentine soils For a majority of the
soils listed in Table 42 total Cr concentrations for the serpentine soils were not
measured however these extractions do provide a quantitative assessment of where Cr
may be bound in soil (such as silicates organic matter and secondary Fe-bearing phases)
Extractions performed by Schwertmann and Latham (1986) were able to remove ~20000
Cr mg kg-1 from the serpentine soil using a dithionite-citrate bicarbonate (DCB) solution
which preferentially dissolves crystalline Fe-oxides Solutions such as concentrated
nitric and hydrochloric acid were only able to remove a maximum of ~1400 Cr mg kg-1
from Unst serpentine soils (Shewry and Peterson 1976) which preferentially dissolve
Cr-bearing silicates and to a lesser extent the Fe-oxides Where bulk soil Cr values were
measured extraction experiments are only able to remove a small portion (lt1) of the
total Cr present in the serpentine soils (Table 42) (Gasser and Dahlgren 1994 Oze et al
2003) Based on the extractions presented in Table 42 it appears that Cr(III) resides in
minerals and phases resistant to common extraction techniques
107
In serpentine soils Fe(III) oxides and Fe(III) hydroxides provide a host of
structural and surface sites available to sequester Cr(III) released to soil solutions by
dissolution of Cr-bearing minerals during soil-forming processes Substitution of Cr(III)
for Fe(III) in Fe-oxides Fe-(oxy)hydroxide minerals and amorphous compounds is
expected based on crystallographic considerations synthesis experiments and analysis of
extracts from selective dissolution of Fe(III) compounds in serpentine soils
(Schwertmann and Latham 1986 Rai and Zachara 1988 Amonette and Rai 1990
Gerth 1990 Gasser and Dahlgren 1994) Chromium(III) also readily sorbs on surfaces
of Fe(III) oxides and (oxy)hydroxides via inner-sphere surface complexes (Figure 43)
and forms low-solubility precipitates (Charlet and Manceau 1992) The low
concentration of Cr(III) in soil solutions has been attributed to the low solubility of
Fe(III)-Cr(III) amorphous hydroxides (Sass and Rai 1987)
Layer silicates provide structural and surface sites capable of uptaking Cr(III)
Octahedral Cr(III) in clays such as smectites and vermiculites is expected and Cr
concentrations in the clay-sized fraction can achieve concentrations of ~1000 mg kg-1 in
serpentine soils (Oze et al 2003) Sorption of Cr(III) and Cr(VI) on edge and interlayer
sites of phyllosilicates have been evaluated by Ilton and Veblen (1994) Ilton et al (2000)
and Brigatti et al (2000) indicating that layer silicates (montmorillonite chlorite biotite
phlogopite and muscovite) provide sites for both Cr(III) sorption and Cr(VI) reduction
Kaupenjohann and Wilcke (1995) concluded that Cr in serpentine soils from Bavaria is
residually bound mainly in silicate structures including clay minerals based on extraction
experiments
Soil Solutions and Related Water
The addition of water gradually changes the spatial and temporal characterization
of pore fluids in soils including the ionic strength pH and redox potential Leaching of
a soil should progressively enrich the soil with less chemically mobile elements such as
Cr and Fe Elements identified in soil solutions include Mg Ca Fe Ni and Cr and the
ionic strengths of serpentine soil solutions are typically low (lt001) (Anderson et al
1973 Johnston and Proctor 1981 Gasser and Dahlgren 1994) Soil solutions obtained
from serpentine soils using tension-free lysimetry demonstrate Cr concentrations ranging
108
from 01 to 32 micromol L-1 (Gasser and Dahlgren 1994) This study also revealed a
majority of Cr Fe Mg and Ni in extracted soil waters were in the form of colloids
suggesting that nanoparticles are an important factor of Cr translocation in soils
The pH of a soil is a dominant factor controlling the nature of many chemical and
microbial reactions including the geochemistry of Cr (Brady and Weil 1999) Ranges in
pH for serpentine soils worldwide are plotted in Figure 46 The highest soil pH was
measured in a South Africa serpentine soil with a value of 9 (Cole 1992b) whereas the
lowest pH was reported in Ireland with a value of 38 (Jeffrey 1992) This wide range of
pH is a function of the adsorbed H+ and Al3+ the base-forming cations the organic
matter and the mineralogy present within each soil The pH range of serpentine soils
shown in Figure 46 is plotted on the stability diagram in Figure 43 demonstrating that
both Cr(III) and Cr(VI) aqueous species are favorable The Eh range for the serpentine
soils in Figure 3 is based on the soils being in contact with the atmosphere as well as
being a transitional environment Although the activity of H+ varies by four orders of
magnitude the pHs of most serpentine soils are slightly acidic (pH ~6) which is near that
of average rainwater (pH~57 Figures 43 and 46) Additionally serpentine soil pH
values have been observed to increase with depth towards the protolith for a number of
soils (Rabenhorst et al 1982 Gough et al 1989 Gasser and Dahlgren 1994) The
increase in pH with respect to depth may be the result of rainwater being progressively
influenced by the mineral chemistry and hydrolysis reactions as it descends through the
soil towards the protolith
Groundwater from serpentinized ultramafic rocks have pH values ranging from 8
to 12 (Barnes et al 1972) and these waters have been divided into two categories as
shown in Figure 46 Mg bicarbonate (Mg2+-HCO3-) type and Ca hydroxide (Ca2+-OH-)
type (Barnes and ONeil 1969 Barnes et al 1972) The Mg bicarbonate waters (pH~8)
are characteristic of most surface waters and shallow groundwaters from serpentinites
and ultramafic rocks The Ca hydroxide waters (pHasymp10-11) are associated with
incompletely serpentinized peridotites and these fluids are comparatively low in Mg2+
Fe2+ Mn2+ Crtotal and SiO2total The Ca2+ in solution originates from the serpentinization
of Ca-rich orthopyroxene (Barnes et al 1972) The pH of serpentinite groundwaters are
109
plotted on the Cr stability in Figure 43 demonstrating that both Cr(III) and Cr(VI) are
potential valence states in these waters
Chromium predominates as Cr(III) in ultramafic rocks and serpentine soils
however Cr(VI) has been identified in serpentine soil solutions and serpentinite
groundwater in California and New Caledonia at concentrations of lt20 microg L-1 (Gough et
al 1989 Becquer et al 2003) Calcium hydroxide waters (pH~11) collected from the
same spring evaluated by Barnes et al (1972) in the Cedars ultramafic body (Cazadero
CA) contains 12-22 Cr(VI) microg L-1 and no Cr(III) (this study) Additionally water from
chromite-containing ultramafic rocks in the Siniktanneyak Mountain ultramafic complex
of Alaska with pH values ranging from 67 to 93 were measured with low total Cr
concentrations lt2 Cr microg L-1 (Taylor et al 1998) A mechanism for Cr(III) oxidation in
serpentine soils and related water suggested by (Gough et al 1989 Becquer et al 2003)
may be the result of high valence Mn oxides common in serpentine soils High valence
Mn-oxides such as birnessite (δ-MnO2) are naturally occurring oxidants capable of
oxidizing Cr(III) to Cr(VI) at pHlt9 (Eary and Rai 1987 Fendorf et al 1992 Fendorf et
al 1993) It is also important to note that Cr(VI) reduction is hypothesized to occur in
soils due to electron transfer involving organic matter and Fe(II)-bearing minerals such as
magnetite (Peterson 1996 Fendorf et al 2000) Whether toxic concentrations of Cr(VI)
can be present in serpentine soils and related water over geologic time is important to
consider especially when high valence Mn oxides Fe(II)-bearing minerals and organic
matter are capable of significantly influencing the oxidation state of Cr
Serpentine Soil Vegetation
Vegetation associated with serpentine soils has received considerable attention
due to the apparent infertility of these soils The inhabitability of serpentine soils for
most plants has been attributed to the imbalance of available Mg relative to Ca or low
CaMg ratios (Walker 1954) the deficiency of plant nutrients such as K N and P
(Turitzin 1991) elevated heavy metal concentrations such as Ni and Cr (Robinson et al
1935) and poor drainage (Brooks 1987) Linking plant productivity to toxic levels of
Cr in the soil has not been fully resolved (Brooks 1987 Kruckeberg 1992) however a
majority of Cr in the soils is not available for plant uptake due to being in high stable and
110
weathering resistant phases discussed in the previous sections (Table 42) In general Cr
concentrations in plants typically do not exceed 100 mg kg-1 (Brooks 1987)
Table 43 provides a global list of serpentine soil plant species and their measured
Cr concentrations with an emphasis on Cr hyper-accumulators ie plants able
incorporate higher than average concentrations of Cr A majority of these measurements
are based on the dry weight of Cr Aerobryopsis longissima from New Caledonia has the
highest reported Cr concentration with values ranging from 200 to 700 mg kg-1 (Lee et
al 1977) Other noteworthy plants include the Lophostachys villosa from Brazil (Brooks
et al 1992) Arenaria humifusa from Newfoundland (Roberts 1992) the Calluna
vulgaris from Unst (Shewry and Peterson 1976) with Cr concentrations of 203 530 and
417 mg kgndash1 respectively Even mosses such as the Funaria flavicans (Maryland)
incorporate Cr concentrations as high as 90 mg kg-1 (Shaw 1991) By examining
specific parts of the plant Gough et al (1989) determined that Cr concentrations are
highest in the basal burl (600 mg kg-1) of the manzanita bush compared to the roots (30
mg kg-1) stems (20 mg kg-1) and leaves (20 mg kg-1) The study by Gough et al (1989)
demonstrates that Cr is not homogeneously distributed throughout the plant and that
evaluating a leaf versus a stem could result in significantly different Cr concentrations
Soil Organic Matter
Soil organic matter (SOM) including the biomass (living organisms such as
bacteria) detritus (nonliving plant residues) and humus (amorphous and colloidal
mixtures of organic substances) affects the geochemistry and buffering capacity of
serpentine soils Furthermore biologic reduction of Cr(VI) to Cr(III) via cometabolic
and metabolic pathways coupled with carbon oxidation has been demonstrated for a
number of bacterial strains including aerobes (Gopalan and Veeramani 1994 Park et al
2000) and anaerobes (Turick et al 1996 Tebo and Obraztsova 1998) Additionally
aqueous Fe(II) produced by microbial Fe(III) reduction is a known biogenic pathway for
the reduction of Cr(VI) (Fendorf et al 2000) Fe(III)-Cr(III) (oxy)hydroxides are the
resulting secondary phases (Rai et al 1989 Patterson and Fendorf 1997 Hansel et al
2003) Bacteria oxidizing Cr(III) to Cr(VI) has not been observed
111
Most serpentine soils in temperate regions appear to be either Mollisols or
Alfisols which often contain high concentrations of SOM Alexander et al (1989) and
Bulmer et al (1992) reported that serpentine soils can contain organic carbon
concentrations in excess of 50 mg g-1 Humus has a high cation exchange capacity and
provides much of the pH buffering capacity in these soils (Brady and Weil 1999)
Organic acids such as fulvic and humic acids are known to complex cations such as
Cr3+ Fe3+ and Mn2+ forming stable organometallic complexes (Kaupenjohann and
Wilcke 1995 Brady and Weil 1999) Additionally metal leaching experiments
performed by Schreier et al (1987) established that organic acids produced by soil
organic matter such as oxalic acid and citric acid were extremely effective in releasing
Cr Mg Fe Ni and Co from serpentinite rocks and sediments
CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
Serpentine soils were obtained from California Oregon and New Caledonia and
analyzed to further evaluate the factors influencing the Cr geochemistry in serpentine
soils Analyses for these soils include pH electrical conductivity total organic matter
total N ammonium acetate extractable Ca K Mg and Na soil texture soil color
chemical composition of the soil and the chemical composition of Cr-bearing minerals
The general soil characteristics are described in the following section and the analytical
methods are explained in Appendix 41 Although a majority of these soils are located in
the western United States factors influencing the pedogenesis of these soils vary
considerably allowing a comparative analysis of how the fate of Cr might be influenced
by climate topography biota parent material and time
Sample Locations and General Soil Characteristics
Six serpentine soils were collected from locations shown in Figures 41 and 42 to
evaluate the chemistry of Cr in serpentine soils These soils include 1) a laterite (Oxisol)
collected in New Caledonia South Pacific 2) a laterite (Oxisol) exposed at Eight Dollar
Mountain Selma OR 3) a laterite (Oxisol) near Nickel Mountain Riddle OR 4) a
serpentine soil (Inceptisol) with a CO2-metasomatized serpentinite protolith from
Harvard Mine Jamestown CA 5) a serpentine soil (Mollisol) obtained from a preserved
112
grassland at Jasper Ridge Biological Preserve Stanford CA and 6) a serpentine soil
(Inceptisol) collected in the proximity of Pillikin Mine El Dorado CA Lastly an Oxisol
formed from basalt obtained in Oahu Hawaii is also included for comparison
These seven sites include a range of soil environments from different protoliths
under varying conditions of rainfall temperature and topography Each soil was collect
near the highest and topographically level point for each serpentine soil site Serpentine
grassland soils at Jasper Ridge formed under Mediterranean-like climates (xeric moisture
regime) whereas higher temperatures and less rainfall (ustic moisture regime)
characterize soil formation at the Harvard and Pillikin mines in the Sierra Nevada
Foothills metamorphic belt The vegetation of these three sites is dominantly grasses with
some manzanita These modern soil environments contrast with the Miocene Ni and Cr-
rich laterite soils of Nickel Mountain and Eight Dollar Mountain that were probably
formed under an aquic moisture regime near sea level in a subtropical climate Grasses
manzanita and pine trees are the dominant forms of vegetation at Nickel and Eight
Dollar Mountain An ultramafically-derived laterite formed under tropical conditions
from New Caledonia was collected The vegetation and topography related to the New
Caledonia laterite was not recorded during sampling An Oxisol from Hawaii is also
included in this study to evaluate the differences between a highly weathered soil
originating from an ultramafic rock and one from a mafic rock (basalt) under tropical
conditions The Hawaiian soil was collected from a topographically level area however
the vegetation related to this site was not noted
BedrockProtolith Mineralogy
A broad characterization of the protoliths for the serpentine soils is briefly
summarized due to the bedrock not being collected from all the field sites New
Caledonia is composed of variably serpentinized peridotites predominantly harzburgites
and some dunites (Leblanc 1995) Nickel Mountain and Eight Dollar Mountain are
predominantly serpentinized harzburgite and dunite The bedrock for soils collected at
Harvard Mine is mainly composed of quartz ankerite and mariposite (also referred to as
QAM) caused by the CO2 metasomatism of serpentinite associated with gold
mineralization (Bohlke 1989) The average serpentinite at Jasper Ridge is composed of
113
serpentine (mainly lizardite and antigorite) chlorite talc magnetite and chromite (Oze et
al 2003) however brucite has been reported as a major constituent in related Franciscan
Complex serpentinites (Hostetle et al 1966) Pillikin Mine serpentinites dominantly
consist of talc and serpentine (lizardite antigorite and chrysotile) and these serpentinites
contain podiform chromite deposits (Cater et al 1948) Mafic bedrock for the Hawaiian
Oxisol is tholeittic basalt
Soil Chemistry and Physical Properties
As shown in Figure 47 the colors of the collected soils range from the red New
Caledonia laterite to the greenish gley of the Pillikin Mine Inceptisol The Munsel color
values are also shown in Figure 47 The dark grey color of the Jasper Ridge soil and the
greenish gley of the Pillikin Mine soil are atypical of soils derived from serpentinites due
to the lack of any apparent redness The range of these colors is often attributed to the
presence of Fe-oxides however organic matter and Mn oxides are also dominant factors
controlling color (Brady and Weil 1999)
Chemical properties for the six serpentine soils and the Hawaiian soil are listed in
Table 44 The slightly acidic pH values range from 53 to 69 similar to those shown in
Figure 43 and 46 The New Caledonia laterite is the most acidic soil with a pH of 53
compared to the Harvard Mine carbonate-buffered soil at 69 The pH of the Hawaiian
soil is also slightly acidic with a pH of 604 The electrical conductivities (EC) of the
soils range from 01 to 07 dS m-1 a range indicative of relatively few dissolved inorganic
solutes The highest organic matter content is in the very dark grey Jasper Ridge soil
with a value of 32 and the lowest is in the red New Caledonia laterite (05) The
Hawaiian soil has an organic matter content of 1 Nitrogen concentrations for all the
soils are lt012 with Jasper Ridge having highest N concentration which is consistent
with the soil organic matter content
Ammonium acetate extractable Ca K Mg and Na are also listed in Table 44
Magnesium concentrations are gt6923 mg kg-1 in all of the serpentine soils The lowest
concentration of Mg was measured in the basalt-derived Hawaiian Oxisol with a value of
168 mg kg-1 Calcium concentrations in the serpentine soils range from 167 to 998 mg
kg-1 The 998 mg kg-1 of Ca in the Harvard Mine soil is directly related to the abundance
114
of ankerite (gt20 of the bedrock) Potassium concentrations are low for most of the soils
(lt55 mg kg-1) however Jasper Ridge and Harvard Mine soils have the highest
extractable K with values of 237 and 301 mg kg-1 respectively The extractable K value
for the Harvard Mine soil is due to the K-metasomatism related to gold mineralization
involving the serpentinite soil protolith The extractable concentration of Na is lt44 mg
kg-1 for all the serpentine soils whereas the Hawaiian soil has the highest Na (53 mg kg-
1) due to plagioclase and basaltic glass
Textures for the soils range from clay loam clay loam to sandy clay loam (Table
44) The laterite soil in New Caledonia and Oxisol in Hawaii are extremely different in
which the Hawaiian soil is mostly clay (66 clay) compared to the loam at New
Caledonia (25 clay) Additionally both ancient laterites (Eight Dollar Mountain and
Nickel Mountain) have clay textures Harvard Mine Jasper Ridge and Pillikin Mine
soils are all loamy in character
Major and trace element compositions for the soils are listed in Table 45
Chromium concentrations in the serpentine soils range between 827 and 9528 mg kg-1
The Eight Dollar Mountain laterite has the highest Cr concentration and the New
Caledonia laterite contains the lowest Cr concentration The Cr concentration for the
New Caledonia soil is lower than the values reported in Table 41 The Hawaiian soil has
a Cr concentration well below the analyzed serpentine soils with 388 Cr mg kg-1 but still
greater than the 200 mg kg-1 characteristic of serpentine soils Nickel concentrations in
the serpentine soils range from 1895 to 14100 mg kg-1 with the highest concentration
being identified in the Nickel Mountain laterite Manganese concentrations range
between 257 (New Caledonia) to 3017 (Eight Dollar Mountain) mg kg-1 in the serpentine
soils Element concentrations in the Hawaiian Oxisol contrast from the serpentine soils
with higher Al Cu P Pb and Zn concentrations
X-ray absorption spectroscopy (XAS) was used to deduce the oxidation state of
Cr in the soil solids for Jasper Ridge New Caledonia Pillikin Mine Eight Dollar
Mountain and Harvard Mine Although the Cr concentrations in these soils exceed 200
mg kg-1 (Table 45) X-ray absorption near-edge spectra (XANES) shown in Figure 48
illustrate Cr(VI) levels are below the detection limits of this method (~10 Cr mg kg-1) due
to the lack of a pre-edge feature The box in Figure 48 illustrates where a Cr(VI) pre-
115
edge peak should be present Chromium(VI) concentrations in these soils are possibly
present in the parts per billion range
Soil Mineralogy
Backscattered electron images and microprobe analyses were used to identify the
Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49
and corresponding Cr-bearing mineral compositions are listed in Table 46 The New
Caledonia soil exhibits a wide range of particle sizes and a Fe-(oxy)hydroxide was
identified containing 165 Cr2O3 Wt Chromium-spinel has been identified in the New
Caledonia laterite by Becquer et al (2003) Eight Dollar Mountain soils exhibit a coarse
texture and both Cr-magnetite (2387 Cr2O3 Wt ) and chromite (3343 Cr2O3 Wt )
are present Nickel Mountain soils are also coarse and chromite (1315 Cr Wt ) was
identified Soils at Harvard Mine contain numerous chromite grains and the mineral
trevorite (a Ni-bearing spinel) was identified with 3991 Cr2O3 Wt and 2054 NiO Wt
Chromite and magnetite are abundant in soils at Jasper Ridge as well as a fine-
grained mixture of chromite magnetite and an unidentifiable silicate with up to 1878
Cr2O3 Wt (Figure 49) Veins of Cr-chlorite crosscut the spinel-silicate mixture in
Figure 49 and this Cr-chlorite is the only Cr-silicate identified using this method
Chromite (5355 Cr2O3 Wt ) is the dominant Cr-bearing mineral detected in soils at the
Pillikin Mine and these grains contain the highest Cr concentrations compared to the
other minerals in the seven soils The Hawaiian Oxisol possesses ulvoumlspinels (007
Cr2O3 Wt and 2465 TiO2 Wt ) which are evenly distributed throughout the soil
Overall microprobe observations reveal that the Cr-bearing spinels are present in every
soil Additionally Cr-bearing spinels at Jasper Ridge have been shown to undergo
incongruent dissolution progressively enriching the spinel towards a FeCr2O4 end-
member (Oze et al 2003)
DISCUSSION
The physical and chemical characteristics of serpentine soils are directly related to
the alteration and tectonic history of the serpentinite as well as the soil-forming factors
including the parent material climate biota topography and time Identifying Cr-
116
bearing minerals their chemical compositions and abundance in the soils allows a more
definitive explanation for Cr enrichment in these soils as compared to total Cr
concentrations listed in Tables 41 and 45 and sequential extraction experiments reported
in Table 42 Detrital chromite grains from the serpentinite protolith are resistant to
weathering and accumulate in the soil over time Although serpentine soils contain Cr
concentrations typically gt200 mg kg-1 the presence of Cr-spinels and their resistance to
weathering demonstrate that a majority of the Cr is not chemically available to interact
within the soil The role of organic matter and vegetation are important factors related to
the chemistry of serpentine soils however their role for influencing a majority of the Cr
geochemistry is limited due to the nature of the Cr-spinels
Chromium and Serpentine Soils
Total Cr concentrations exceed 200 mg kg-1 for most of the serpentine soils
reported in Table 41 and in the six serpentine soils of this study (Table 45) These Cr
concentrations can be accounted for by assessing the Cr-bearing minerals associated with
these soils and their relative abundance Small grains of Cr-spinels or chromite identified
in the soils (Figure 49 and Table 46) have the potential to appreciably impact Cr
concentrations Table 47 demonstrates the mineral masses and volumes of Cr-bearing
minerals required to cause predetermined Cr concentrations in soil samples A minimal
amount of chromite (lt02 of a soil sample mass or 005 of the volume of serpentine)
can result in Cr concentrations of 1000 mg kg-1 A few grains of chromite could easily
be responsible for the Cr concentrations in Tables 41 and 45 In fact residual grains of
Cr-enriched spinels are observed in every soil analyzed (Figure 49 and Table 46)
Ultimately grains of spinel in the soil do appreciably impact Cr concentrations however
they do not provide a readily available source of Cr based on extraction experiments
performed by Oze et al (2003) Serpentine minerals could also provide minor Cr
concentrations to the soils however soil extractions have consistently shown that a
majority of the Cr is in highly resistant phases capable of resisting dissolution even in
concentrated hydrochloric acid nitric acid and hydroxylamine with acetic acid (Table
42) Chromium end-member minerals including eskolaite Cr-muscovite (fuchsite) Cr-
garnet (uvarovite) Cr-epidote (tawmawite) and Cr-chlinochlore (kaumlmmererite) are also
117
included in Table 47 to demonstrate that relatively low abundances of these minerals are
all that is necessary to increase Cr concentrations in rocks and soils Chromium-
clinochlore is often observed developing adjacent Cr-spinels in serpentinites and is
perhaps a more accessible mineral source of Cr in the serpentinites compared to Cr-
spinels (Phillips et al 1980 Schreyer et al 1981 Kerrich et al 1987 Treloar 1987a
Pan and Fleet 1991 Schandl and Wicks 1993 Christofides et al 1994 Challis et al
1995 Sanchez-Vizcaino et al 1995 Mathiesen 1999 Oze et al 2003)
As shown in Table 44 and Figures 43 and 46 the chemical properties (pH EC
N etc) of serpentine soils are not correlated to Cr concentrations measured in the soil
solids (Tables 41 Figure 45) due to Cr-spinels accounting for a majority of the Cr
(Figure 49 and 410 Tables 46 and 47) However extraction experiments and soil
solution analyses demonstrate that a limited fraction of the total Cr (lt1) is extractable
If only 1 of the Cr in a serpentine soil is extractable a serpentine soil containing 10000
Cr mg kg-1 could possibly contribute up to100 Cr mg kg-1 still a significant Cr
concentration The chemical properties of the soil such as pH will influence this limited
fraction of the Cr in the soils Acidic pHs may aid in mobilizing Cr(III) however this
will only occur at low pHs (lt4) which exceed the range of most serpentine soils (Figures
43 and 46) pH values greater than 8 like those in some serpentine soils and
serpentinite groundwater may help sustain Cr(VI) concentrations in solution (Figure
43) In addition to pH organic matter in the soils may be a significant factor related to
the mobility of Cr due to the production of organic acids promoting the dissolution of the
Cr-bearing phases but particulate organic matter is known to bind Cr(III) in stable stable
organometallic complexes (Kaupenjohann and Wilcke 1995 Brady and Weil 1999)
Evaluating the valence state of Cr is crucial for understanding the chemistry as
well as the potential toxicity related to serpentine soils The soil solids do not contain
detectable concentrations of Cr(VI) (Figure 48) however Cr(VI) concentrations are
present in soil solutions and groundwater at values lt30 microg L-1 (Gough et al 1989
Becquer et al 2003) A majority of serpentinite Cr-bearing minerals only contain
Cr(III) therefore there must be a reaction or process by which Cr(III) is oxidized in a
water saturated soil High valence Mn oxides such as birnessite are present within
serpentine soils and provide a feasible explanation for Cr(III) oxidation Additionally
118
Mn oxides or other strong oxidants in the soil might accelerate the weathering of Cr-
bearing minerals including chromite and generate the Cr(VI) concentrations observed in
soil solutions and groundwater After Cr(III) is oxidized Cr(VI) is a chemically mobile
species and would be transported away from the serpentine soils before it can be
observed in the solid phase Additionally Cr(VI) may be subsequently reduced by
reacting with organic matter or Fe(II)-bearing minerals such as magnetite which are
present in a majority of serpentine soils Perhaps the best way to evaluate the potential of
serpentine soils being a non-anthropogenic source of Cr contamination is to monitor
Cr(III) and Cr(VI) in the soil solutions over time as demonstrated by Gasser and
Dahlgren (1994)
It is curious that plants do uptake Cr in serpentine soils especially when Cr is
tightly bound in the soil matrix (Table 42) In order for plants to uptake Cr it must be in
a form to be absorbed by plant roots Soil pH in the rhizosphere can be altered by plant
roots by as much as 2 units within two days (Fischer et al 1989) This acidification is
known to increase the solubility of the minerals in the soils and could serve as a
mechanism by which plants can incorporate this relatively insoluable element Another
possibility is that Cr(III) in the soil is oxidized by either organic or inorganic means and
then incorporated into the plant in its more chemically mobile and bioavailable form
Cr(VI) How and why plants incorporate as much Cr as they do needs to be examined
more thoroughly
Chromium Spinel in Soils and Rocks
Compositions of chromite identified in serpentinites worldwide (Figure 44) and
Cr-bearing spinels identified in the six serpentine soils are compared in Figure 10 in order
to determine whether the soil spinels are geochemically similar to those in serpentinites
This comparison is ascertained by examining how Al Fe and Mg Wt values varied
with respect to Cr Wt Trends for magnetitehematite chromite and the
microcrystalline spinel-silicate mixture observed in the soils are noted in each diagram in
Figure 410 Additionally the highest Cr concentrations for soil chromite
magnetitehematite and the microcrystalline spinel-silicate mixture are ~40 Cr Wt
~25 Cr Wt and ~15 Cr Wt respectively Based on Al Wt soil
119
magnetitehematite deviates significantly from the chromite trend with Al Wt values
lt5 Magnetitehematites are enriched with Fe relative to serpentinite and soil chromite
and Mg concentrations in the soil chromite are slightly higher than serpentinite chromite
Soil magnetitehematite and chromite are similar by demonstrating a linear trend when
Al Fe and Mg are summed The microcrystalline spinel-silicate mixture (CSM) does
not follow this trend because of Si present in this mineral mixture Overall soil
chromites are similar to serpentinite chromite and soil magnetiteshematites can be
differentiated from the chromites when Cr Wt is plotted against Al Fe and Mg Wt
CONCLUDING STATEMENT
Serpentine soils are abundant among convergent margins and the chemistry of
these soils is ultimately a function of the ultramaficserpentinite rock mineralogy and
texture as well as all the common soil-forming factors Although ultramafic rocks and
serpentinites produce groundwater that is highly basic most serpentine soil pHs are
slightly acidic Although Proctor and Nagy (1992) argued that it is futile to categorize
serpentine soils based on chemistry and biology (otherwise referred to as the lsquoserpentine
factorrsquo) as proposed by Brooks (1987) Cr concentrations greater than 200 mg kg-1 in
serpentine soils are consistently present Chromium in the soils originates from minerals
in the bedrock where the valence is primarily Cr(III) the exception being Cr(VI) in the
rarely present mineral crocoite (PbCrO4) Increased Cr-concentrations in serpentine soils
are largely the result of chromite or other resistant Cr-bearing minerals causing a
localized enrichment in the soils Although Cr in serpentine soils is not readily released
from the soil matrix serpentine soil vegetation can incorporate Cr concentrations as high
as 700 mg kg-1 (Table 34) Finally the limited extractability of Cr and site dependent
factors such as the presence of organic matter clays Fe(II)-bearing minerals pH and the
poor drainage indicates that the possible formation and sustained presence of Cr(VI) is
not a portentous matter related to serpentine soils
120
ACKNOWLEDGEMENTS
We appreciate technical and field support from Robert Jones Anders Meibom
Rob Dunbar and Guangchao Li Funding from the JRBP Mellon Grant Stanford Shell
Foundation Stanford McGee Fund GSA Graduate Student Research Grants and the
Office of Teaching and Learning at Stanford University is gratefully acknowledged
121
Table 41 Measured Cr concentrations in serpentine soils worldwide
Location Cr concentration (mg kg-1) Reference
Canyon Creek area OR 920 (White 1971) 66-13200 (Brooks et al 1992)
Ireland 29-5000 (Proctor and Woodell 1971 Jeffrey 1992) Italy 1300-3900 (Gambi 1992) Japan 5200 (Shewry and Peterson 1976) Maryland 1500-6000 (Rabenhorst et al 1982) Mt Tamalpais CA 1700 (White 1971) New Caledonia 6300-80000 (Jaffre 1980) New Caledonia 2-6 (Schwertmann and Latham 1986) New Idria CA 1640 (White 1971) New Zealand 850-3250 (Lee 1992) Newfoundland Canada 309-1447 (Roberts 1992) Nickel Mountain OR 1730 (White 1971) Santa Barbara County CA 2080 (Woodell et al 1975) Southern Africa 045-14 (Morrey et al 1992) Tehama County 023-12 (Gough et al 1989) Transvaal Lowveld South Africa 120-800 (Cole 1992) Unst Shetland 3000-4000 (Proctor 1969) Wenatchee Mnts WA 895 (White 1971) Zimbabwe 110-46000 (Soane and Saunder 1959 Proctor et al 1980)
Goias State Brazil
122
123
Table 42 Summary of Cr Fe and Ni extractions from serpentine soils
Range of Extractable Cr Fe and Ni Location Cr Fe Ni
Method Reference
Italy 2-5 microg g-1 305-640 2300-2700microg g-1 25 CH3COOH (Gambi 1992)
NE Portugal 02-09 microg g-1 120-540 microg g-1 40-290microg g-1 EDTA (Menezes de Sequeira and Pinto da Silva 1992) New Caledonia 22-23 384-581 02-216 DCB (dithionite-citrate bicarbonate) (Schwertmann and Latham 1986) New Caledonia 02-11 microg g-1 02-66microg g-1 Method Unknown (Jaffre 1980)
Norway 357microg g-1 316microg g-1 HClO4HNO3 (concentrated) (Zeien and Bruemmer 1989) Norway 254microg g-1 134microg g-1 Ascorbic and Oxalate Acid pH=325 (Zeien and Bruemmer 1989)
Norway 62 microg g-1 53microg g-1 NH4-EDTA (Zeien and Bruemmer 1989) Norway 22-60 microg g-1 8-25 mg kg-1 1-7 microg g-1 pH-stat reactor EDTA (Kaupenjohann and Wilcke 1995)
Swiss Alps 08 micromol L-1 28micromol L-1 228microg g-1 Lysimeter (Nonfiltered) (Gasser and Dahlgren 1994) Trinity Ophiolite CA 004-045 gkg 27-210 g kg-1 005-245microg g-1 Citrate-Dithionite Extracts (Alexander et al 1989)
Unst lt01 microg g-1 19-77microg g-1 Ammonium Acetate (Shewry and Peterson 1976)
Unst 500-1390microg g-1 1540-3330microg g-1 HNO3-Soluble (Shewry and Peterson 1976) JRBP CA 0 microg g-1 01microg g-1 103microg g-1 BaCl2 Method (Oze et al 2003)
JRBP CA 78 microg g-1 513 microg g-1 150 microg g-1 Hydroxlyamine+Acetic Acid (Oze et al 2003) JRBP CA 31 microg g-1 2039microg g-1 35microg g-1 10M Hydrofluoric Acid (Oze et al 2003)
123
124
Table 43 Serpentine soil plant Cr concentrations
Plant Cr Concentration (microg g-1) Location Reference
Lophostachys villosa 203 Goias State Brazil (Brooks et al 1992) Ruellia geminiflora
56 Goias State Brazil (Brooks et al 1992) Minuartia ophiolitica 83 Italy (Gambi 1992) Sanguisorba minor 194 Italy (Gambi 1992) Plantago lanceolata 135 Italy (Gambi 1992) Cerastium latifolium 110 Italy (Gambi 1992) Funaria flavicans 34-899 Maryland (Shaw 1991) Avena sativa 168-583 New Zealand (Lee 1992) Aerobryopsis longissima 200-700 New Caledonia (Lee et al 1977) Arenaria humifusa 530 Newfoundland Canada (Roberts 1992) Senecio pauperculus 406 Newfoundland Canada (Roberts 1992) Rhalomitrium lanuginosum 246 Newfoundland Canada (Roberts 1992) Santolia semidentata 95 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Asplenium cuneifolium 228 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992) Armeria eriophylla 38 Northeast Portugal (Menezes de Sequeira and Pinto da Silva 1992)
Arctostaphylos viscida 029-705 Tehama County CA (Gough et al 1989) Calluna vulgaris 417 Unst (Shewry and Peterson 1976) Armeria maritima 149 Unst (Shewry and Peterson 1976) Agrostis stolonifera 120 Unst (Proctor and Woodell 1971) Dicoma niccolifera 77 Zimbabwe (Proctor 1992) Xerophyta equisetoides 68 Zimbabwe (Proctor 1992)
124
125
Table 44 Chemical and textural properties of the collected soils
Ammonium Acetate Extractable ------------------------Hydrometer--------------------- Ca K Mg Na
Sand Silt ClaySample pH EC dSm
Organic Matter ------
LECO Nitrogen ------ -------------------mg kg-1-------------------
------ ------ ------
Texture
New Caledonia 534 01 05 lt001 279 29 692 44 43 32 25 LoamHawaii 604
03 10 002 839 46 168 53 8 26 66 ClayEight Dollar Mnt 651 07 16 006 599 62 3464 25 28 28 44 Clay
Nickel Mnt 665 05 07 001 167 42 3289 17 24 32 44 ClayHarvard Mine 691 03 11 003 998 301 2897 43 41 30 29 LoamJasper Ridge 671 04 32 012 505 237 3291 36 36 31 33 Clay LoamPillikin Mine 642 04 05 002 185 55 3421 29 45 23 32 Sandy Clay Loam
125
126
Table 45 Total major and trace element concentrationsdagger of the collected soils
Element Concentration New Caledonia Hawaii Eight Dollar
Mnt Nickel Mnt Harvard Mine Jasper Ridge Pillikin
Mine Ag mg kg-1 lt05 lt05 lt05 lt05 02 lt05 lt05Al 931 1325 314 095 179 314 179Ba mg kg-1 lt10 180 40 20 30 230 lt10Be mg kg-1 05 16 05 lt05 lt05 08 lt05Ca 003 012 085 018 007 079 002Cd mg kg-1 lt05 lt05 18 lt05 05 lt05 lt05Co mg kg-1 56 183 338 403 85 155 122Cr mg kg-1 827 388 9528 5761 2900 5976 1977Cu mg kg-1 78 248 43 25 51 33 4Fe 638 1296 1635 17 538 727 468K 002 046 004 003 004 039 001
Mg 072 026 753 829 792 1025 gt150Mn mg kg-1 257 2706 3017 2554 920 1664 903Na 001 008 005 001 003 057 lt001Ni mg kg-1 2780 345 5510 14100 1895 2180 2760P mg kg-1 50 260 160 150 130 140 20Pb mg kg-1 6 31 4 3 2 8 6Ti 016 041 008 002 lt001 02 002V mg kg-1 63 132 158 54 65 92 27W mg kg-1 10 10 10 10 lt10 lt10 lt10Zn mg kg-1 43 218 160 132 66 111 47
126
dagger ICP-AES analyses
127
Table 46 Compositionsdagger for soil Cr-containing minerals identified in Figure 49
Wt New Caledonia A
Hawaii B
Eight Dollar Mnt C
Eight Dollar Mnt D
Nickel Mnt E
Harvard Mine F
Jasper Ridge G
Jasper Ridge H
Pillikin Mine I
SiO2 491 008 175 026 002 000 1940 001 000Al2O3
756 185 272 2235 2041 088 559 1575 839NiO 023 005 001 006 011 2054 014 002 007FeO 7766a 6212 6633b 3420b 1784 2445 2629 2314 3105MnO 004 037 181 123 133 031 155 146 216Cr2O3 165 007 2387 3343 4682 3991 1878 5024 5355MgO 048 121 202 653 1315 1062 1668 926 497CaO 001 000 013 008 000 000 015 000 000Na2O 000 000 002 001 000 000 006 000 000TiO2 004 2465 012 006 012 000 000 003 018
TOTAL 9257 9041 9879 9821 9980 9671 8864 9920 10037Mineral Cr-Hydroxide Ulvoumlspinel Cr-Magnetite Chromite Mg-Chromite Trevorite CSM1 Chromite Chromite
127
dagger Electron microprobe analyses This is the form of iron unless otherwise stated a FeO is in the form Fe2O3 b FeO is in the form of Fe3O4 1CSM Microcrystalline chromitesilicate mixture
128
Table 47 Chromium-bearing mineraldagger masses and volumes required to influence Cr concentrations in soils
Cr Concentration Mineral mass (grams) required for 1000 g bulk soil analysis Mineral Mass Percent () (mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite
500 11 07 22 24 48 57 011 007 022 024 048 057 1000 22 15 43 48 95 115 022 015 043 048 095 1152000 43 29 86 96 191 230 043 029 086 096 191 2305000 108 73 216 241 477 574 108 073 216 241 477 574
Cr Concentration Mineral Volume Percent () in Serpentine Mineral Volume Percent () in Quartz
(mg kg-1) Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite Chromite Eskolaite Fuchsite Uvarovite Tawmawite Kaumlmmererite500 005 004 02 02 03 15 006 004 02 02 04 06
1000 01 007 04 03 07 30 01 007 04 03 07 112000 02 014 07 06 14 59 02 015 07 06 14 225000 06 036 18 16 34 147 06 037 18 16 35 55
128
daggerMineral formulas eskolaite Cr2O3 chromite FeCr2O4 fuchsite KCr2(AlSi3O10)(OH)2 uvarovite Ca3Cr2Si3O12 tawmawite Ca2CrAl2Si3O12(OH)2 kaumlmmererite Mg5CrAlSi3O11(OH)8
Note Molar volumes used in the calculations above are from their Al-enriched counterparts in Robie and Hemingway (1995) The molar volume for kaumlmmererite was estimated using unit cell volumes from XRD unit cell measurements from Robie and Hemingway (1995) Christofides et al (1994) and Phillips et al (1980)
129
Figure Captions
Figure 41 Selected serpentine soil localities as well as soils collected for analyses are
shown on a world map Active plate boundaries are shown Serpentine soil localities not
adjacent the plate boundaries on this map are related to ancient plate boundaries that are
no longer active
Figure 42 Field pictures of soils collected from a) Eight Dollar Mountain b) Nickel
Mountain c) Harvard Mine d) Pillikin Mine and f) Jasper Ridge Biological Preserve
Figure 43 Stability field diagram of Cr under a range of Eh and pH conditions (log a
Cr+3 = -16 and log a H2O = 0) modeled after Ball and Nordstrom (1998) Adsorption
isotherms for Cr3+ and CrO42- on hydrous ferric oxide (HFO) are also shown (Leckie et
al 1980 Leckie et al 1984)
Figure 44 Microprobe analyses demonstrating Cr versus Al Wt for Cr-spinel
(chromite) and Cr-silicates including Cr-mica (fuchsitemariposite) Cr-garnet
(uvarovite) Cr-clinochlore (kaumlmmererite) and Cr-epidote (tawmawite) (Whitmore et al
1946 Leo et al 1965 Chen and Lee 1974 Neiva 1978 Cooper 1980 Phillips et al
1980 Grapes 1981 Schreyer et al 1981 Max et al 1983 Nutman et al 1983 Raase
et al 1983 Chao et al 1986 Von Knorring et al 1986 Kerrich et al 1987 Treloar
1987b a Morand 1990 Pan and Fleet 1991 Tracy 1991 Schandl and Wicks 1993
Christofides et al 1994 Challis et al 1995 Sanchez-Vizcaino et al 1995 Mathiesen
1999 Proenza et al 1999) Chromium-spinel and Cr-silicate compositions based on
octahedral Cr and octahedral Al are also shown with tie-lines (dotted) denoting ideal
isomorphic substitution between Cr and Al end-members
Figure 45 Chromium concentrations (mg kg-1) as a function of depth (meters) for
selected serpentine soils indicated in the key above Each dotted line connecting the
points represents a separate soil profile at each site
129
Figure 46 Ranges of pH for serpentine soils and related groundwater worldwide Soil
pH values were obtained from A(Kruckeberg 1992) B (Graham et al 1990) C
(Alexander et al 1989) D (Gough et al 1989) E (Rabenhorst et al 1982) F
(Roberts 1992) G (Bulmer et al 1992) H (Proctor 1992a) I (Jeffrey 1992) J
(Gambi 1992) K (Verger 1992) L (Proctor and Cole 1992) M (Cole 1992b) N
(Morrey et al 1992) O (Proctor 1992b) P (Mizuno and Nosaka 1992) Q (Cole
1992a) and R (Lee 1992) Groundwater pH values from serpentinite are from Barnes
and ONeil (1969) and Barnes et al (1972) Note the range of pH environments possible
for ultramafic rocks and their related soils
Figure 47 Soil pictures with Munsel color values and names from soils collected from
a) New Caledonia b) Hawaii c) Eight Dollar Mountain d) Nickel Mountain e) Harvard
Mine f) Jasper Ridge Biological Preserve and g) Pillikin Mine
Figure 48 Cr XANES spectra for soils analyzed from Jasper Ridge New Caledonia
Pillikin Mine Eight Dollar Mountain and Harvard Mine The pre-edge area indicative of
Cr(VI) peaks is noted in the diagram No definitive Cr(VI) peaks are present in any of
these spectra
Figure 49 Backscattering images of mineral phases within each soil are presented
Electron microprobe analyses are noted by a circle and letter for the Cr-bearing minerals
and the results are listed in Table 46
Figure 410 Chromite magnetite and hematite compositions from the collected soils
(gray circles) as well as analyses for serpentinite chromite (white squares) from Figure
44 are shown in these diagrams with Cr Wt versus Al Fe Mg and Al+Fe+Mg Wt
130
Figure 41
131
Figure 42
132
Figure 43
133
Cr Wt
0 10 20 30 40 50
Al W
t
0
5
10
15
20
25Cr-SpinelCr-MuscoviteCr-EpidoteCr-GarnetCr-Chlorite
Cr-SpinelBased on 4 Oxygens
Cr0 1 2
AlVI
0
1
2
ChromiteFeCr2O4
HercyniteFeAl2O4
Cr-MuscoviteBased on 11 Oxygens
Cr0 1 2
AlVI
0
1
2
MuscoviteKAl2(AlSi3O10)(OH)2
FuchsiteKCr2(AlSi3O10)(OH)2
Cr-GarnetBased on 12 Oxygens
Cr0 1 2
AlVI
0
1
2
GrossularCa3Al2Si3O12
UvaroviteCa3Cr2Si3O12
Cr-EpidoteBased on 125 Oxygens
Cr0 1
AlVI
2
3
ClinozoisiteCa2Al3Si3O12(OH)2
TawmawiteCa2CrAl2Si3O12(OH)2
Cr-ChloriteBased on 14 Oxygens
Cr0 1
AlVI
0
1
2
Kammererite(Mg5Cr)(AlSi3)O10(OH)8
Clinochlore(Mg5Al)(AlSi3)O10(OH)8
Figure 44
134
Maryland (USA)(Rabenhorst et al 1982)
Chromium (mg kg-1)
0 1000 2000 3000 4000 5000 6000 7000
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Tehama County CA (USA)(Gough et al 1989)
Chromium (mg kg-1)
0 2000 4000 6000 8000 10000 12000 14000
Dep
th (m
eter
s)
00
02
04
06
08
Kalamath Mountains CA (USA)(Hotz 1964)
Chromium (mg kg-1)
2000 4000 6000 8000 10000 12000 14000 16000
Dep
th (m
eter
s)
00
05
10
15
20
25
30
35
40
45
50
Newfoundland (Canada)(Roberts 1992)
Chromium (mg kg-1)
800 1000 1200 1400 1600 1800
Dep
th (m
eter
s)
00
02
04
06
08
Harvard Mine CA (USA)(Unpublished Data)
Chromium (mg kg-1)
0 500 1000 1500 2000 2500
Dep
th (m
eter
s)
00
02
04
06
08
10
12
14
16
Jasper Ridge CA (USA)(Oze et al 2003)
Chromium (mg kg-1)
1500 2000 2500 3000 3500 4000 4500 5000
Dep
th (m
eter
s)
00
01
Figure 45
135
LocationsZ A B C D E F G H I J K L M N O P Q R S
pH
0
1
2
3
4
5
6
7
8
9
10
11
12
Nor
th A
mer
ica
Can
ada
Brit
ain
Italy
Far E
ast
Japa
n
Zim
babw
e
Sout
h A
fric
a
Wes
tern
Aus
tral
ia
New
Zea
land
Mg-Bicarbonate Waters pH=8
Ca-Hydroxide Waters pH=11-12
Average RainpH=57
Sout
h A
fric
a
Italy
Irela
nd
Cal
iforn
ia
Cal
iforn
ia
Brit
ish
Col
umbi
a
Mar
ylan
dCal
iforn
ia
Figure 46
136
Figure 47
137
-2
-1
0
1
2
3
4
5
5980 5990 6000 6010
Energy (eV)
Nor
mal
ized
Abs
orba
nce
New Caledonia
Pillikin Mine
Eight Dollar Mnt
Jasper Ridge
Harvard Mine
Cr(
VI)
Figure 48
138
Figure 49
139
Cr Wt 0 10 20 30 40 50
Al W
t
0
10
20
30 Soil Serpentinite
Cr Wt 0 10 20 30 40 50
Fe W
t
0
20
40
60
80
Cr Wt 0 10 20 30 40 50
Mg
Wt
0
2
4
6
8
10
Cr Wt 0 10 20 30 40 50
Al +
Fe
+ M
g W
t
0
20
40
60
80
MagnetiteHematite
Chromite
MagnetiteHematite
Chromite
Chromite
MagnetiteHematite
SpinelSilicate
SpinelSilicate
SpinelSilicate
SpinelSilicate
Figure 410
140
- Chapter 1 Introductionhelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip1
-
- Chapter 3 Oxidative Promoted Dissolution of Chromite by
- Manganese Dioxide and Concurrent Production of Ch
-
- Chapter 4 Chromium Geochemistry of Serpentine S
-
- List of References
-
- Chapter 2
- Chapter 3
-
- Chapter 4
-
- Appendix A
- Chapter 2
-
- Chapter 3
-
- Chapter 4
-
- Appendix A
-
- CHAP1pdf
-
- Toxicity and Regulations of Chromium
- Mineralogy of Chromium in Serpentinites
-
- Chapter Summaries
-
- CHAP2Tablepdf
-
- JR305CLAY Depth 0-5cm
- JR3515CLAY Depth 5-15cm
- JR33045CLAY Depth 30-45cm
-
- HCAP2Figspdf
-
- Figure 27
-
- CHAP3Tablepdf
-
- Figure 36
- Figure 37
-
- CHAP4pdf
-
- ABSTRACT
- REVIEW
-
- Chromium Geochemistry
-
- Soil Organic Matter
-
- CALIFORNIA OREGON AND NEW CALEDONIA SERPENTINE SOILS
-
- BedrockProtolith Mineralogy
- Soil Chemistry and Physical Properties
- Soil Mineralogy
- Backscattered electron images and microprobe analyses were used to identify the Cr-bearing minerals in the seven soils backscattered images are shown for in Figure 49 and corresponding Cr-bearing mineral compositions are listed in Table 46 The New C
-
- DISCUSSION
-
- Chromium Spinel in Soils and Rocks
-
- CONCLUDING STATEMENT
- ACKNOWLEDGEMENTS
-
- CHAP4Tablepdf
-
- Ulvoumlspinel
-
top related