The Role of Iron in the Reduced Bioavailability of Arsenic in Soil V.L. Mitchell1; C.N. Alpers2; N.T. Basta3; T. Burlak4; S.W. Casteel5; R.L Fears1; A.L. Foster2; C.S. Kim6; P.A. Myers1; and E. Petersen7

(1) Department of Toxic Substances Control, Cal EPA, (2) U. S. Geological Survey, (3) Ohio State University, (4) California State University, Sacramento (5) University of Missouri, Columbia, (6) Chapman University, (7) University of Utah, Salt Lake City

ABSTRACT

Arsenic (As) is a naturally occurring element in soil and often the key chemical of concern at former mines in the California Mother Lode Region. Risk based screening levels are often below the naturally occurring background concentrations of As in soil. These screening levels assume 100% bioavailability, which can be greatly affected by the association of As with other elements. Iron(Fe)-bearing minerals are known to bind As and reduce its bioavailability; however the mechanism is not completely understood. The soils in the Mother Lode Region are relatively high in Fe, ranging from 2-10%. Total As in soils collected from the Empire Mine State Historic Park ranged from 16 to 9,700 mg/kg. Bulk soil samples were sieved to <250µm, the fraction that adheres to skin and is most likely to be ingested by humans. The mass of the <250µm fraction averaged 17.4% of the bulk sample (pre-sieved <6mm). In vitro bioaccessible As (IVBA As), a surrogate for in vivo bioavailable As, in the <250µm fraction ranged from 1-44%. The spatial association between As and Fe was visualized utilizing QEMSCAN that provides quantitative chemical and mineralogical data at the 2.5µm pixel scale. Polished sections of mine waste rock analyzed by QEMSCAN demonstrated variable As content in hydrous Fe-oxide rims on weathered sulfide minerals. Further analysis, using a variety of synchrotron-based x-ray techniques, mapped the various oxidation states of As and Fe in these samples. Similar analysis of the <250µm fraction soil samples should provide useful information on the effect of Fe content and speciation on bioavailability of As in mining environments.

MINING WASTE ROCK MINING SOILS

Funded by USEPA Brownfields Training, Research and Technical Assistance Grant TR-83415101

Dr. Foster with USGS analyzed the polished sections using micro x-ray fluorescence (µXRF) and micro x-ray absorption spectroscopy (µXAS ) at the Stanford Synchrotron Radiation Lightsource (SSRL). SSRL operates at a voltage of 3.0 GeV and 200-100 mAmp µ-XRF is used to map the distribution of multiple elements and/or multiple oxidation states of a single element; this information can be displayed in single element maps or as multiple element overlays, in which each element is represented by a color. µ-XAS is used to determine more specific chemical information about a single volume of the sample that is defined by the current beam size and penetration depth (this volume is represented by a point on the 2-d maps). Usually the identity of the mineral(s) or poorly-crystalline phase(s) in which arsenic is hosted can be determined quite specifically using this information.

SUMMARY

REFERENCES: 1.  Basta, N.T., J. N. Foster, E.A. Dayton, R. R. Rodriguez, and S.W. Casteel. 2007. The effect of

dosing vehicle on arsenic bioaccessibility in smelter-contaminated soils. Invited manuscript for the special JEHS publication "Bioaccessibility and human bioavailability of soil contaminants" J. Environ. Health Sci. Part A. 42:1275-1281.

2. Burlak, T., Alpers, C. N., Foster, A. L. , Brown,A., Hammersley, L.C., Petersen, E.. Tracking the fate of arsenic in weathered sulfides from the Empire Mine Gold-Quartz Vein Deposit by using Microbeam Analytical Techniques. AGU Fall 2010 Meeting, abstract #V51C-2220.

3.  George, G. N., and Pickering, I. J. (1993), “EXAFSPAK: A suite of programs for the analysis of x-

ray absorption spectra”,Stanford Synchrotron Radiation Laboratory (freeware). http://ssrl01.slac.stanford.edu/exafspak.html .

4.  Kelly, S. D., Hesterberg, D. and Ravel B. (2008), “Chapter 14: Analysis of Soils and Minerals using X-ray Absorption Spectroscopy”, in Methods of Soil Analysis. Part 5. Mineralogical Methods. SSSA Book Series, 78 p.

5.  Whitacre, S. D. 2009. M.S. Thesis. Soil controls on arsenic bioaccessibility: Arsenic fractions and soil properties. The Ohio State University, Columbus, OH.

Figure 1: Sample Collection: A) Field activities were performed by DTSC in conjunction with USGS at the Empire Mine State Historic Park in Nevada County, California. B) Samples were field sieved to reduce the volume of soil collected. Sample locations were chosen based on total and IVBA arsenic results from a previous sampling event. C) Twenty-five individual soil samples, totaling 2593 pounds, were collected.

In vitro Gastrointestinal Method:

Gastric Phase

• 0.1M NaCl, 1% Pepsin

• 1:150 soil:solution ratio

• Manual adjustment with HCl to pH 1.8 with constant stirring for 1 hr.

Intestinal Phase

• 0.563 g Bile, Pancreatin added • Manual adjustment with Na2CO3 to pH 6.1 with constant stirring for 2 hrs.

0

5

10

15

20

0

100

200

300

400

500

Iron:

As

% IV

BA

As

A B

C Figure 2: Sample Processing A) Soil samples were homogenized using a modified cement mixer in the laboratory of Dr. Basta at Ohio State University B) Following homogenization samples were sieved to <250 µm on a shaker table. C) Bulk and <250 µm fractions, respectively.

Figure 5: QEMSCAN Analysis of Mining Waste Rock: A) Mine waste rock collected in field showing quartz vein with scorodite B) Thick section of polished rock prepared for QEMSCAN Analysis. Shadow box indicates region shown in C) Arsenic map showing all areas containing measurable As. Square indicates region shown in D) Enlargement of arsenopyrite showing As-rich iron oxide rims (green) and lower As iron oxide zones (brown). Lower As zones contain approximately 7 % elemental weight of As. Color map shown in E)

E

Pyrite 44.14 Feldspar 27.68 Fe Oxides (G07) 13.21 Quartz 5.24 Chlorite 4.27 Background 2.10 Arsenopyrite 1.11 Fe Oxide- Low As 1.04 Fe Oxide As 1.00 Other Silicates 0.81 Ca-S/SO4/Gypsum 0.69 Jarosite 0.35 Others 0.21

A B

C D

A

A B

C

Arsenic

050

0010

000

1500

00

2000

4000

6000

8000

10000

<250 µm Fraction (mg/kg)

Bul

k Fr

actio

n (m

g/kg

)

R2= 0.9514

Iron

050

000

1000

00

1500

000

20000

40000

60000

80000

100000

<250 µm Fraction (mg/kg)

Bul

k Fr

actio

n (m

g/kg

)

R2= 0.2792

Iron: Arsenic

050

0010

000

1500

00

5000

10000

15000

<250 µm Fraction

Bul

k Fr

actio

n

R2= 0.978

C

Figure 3: Iron and Arsenic in Mining Soils: Total metal analysis was completed using US EPA Method 6010 on both bulk and <250µm soil fractions. Plots compare metals between the bulk and <250µm fraction. The trend is for an increase in Fe (A) and As (B) concentration in the fines fraction. C) There was a strong linear relationship between the ratio of Iron to Arsenic in the bulk and fines fraction(R2= 0.978)

A B C Arsenic in vitro Bioaccessibility

Metals in Bulk and <250µm Soil Fractions

Figure 4: Comparison of Iron:Arsenic Ratio and the % IVBA Arsenic in <250µm soil fractions

Figure 6: Synchrotron Analysis of mine waste rock: the area analyzed by µXRF is shown in an optical image of the sample (far left). The center map displays synchrotron µXRF data for: (1) reduced arsenic [As(-1), in pyrite and/or arsenopyrite ], (2) pentavalent As [As(V) the most oxidized form], and (3) sulfur. The pervasive green color indicates the predominance of As(V) in areas identified as Fe (hydr)oxides by QEMSCAN (far right, red box). The euhedral magenta-colored grain in the lower right of the tricolor map is pyrite; its color indicates the presence of both As(-1) and S.

Synchrotron Analysis

QEMSCAN Analysis

Primary Mineral Wt. % Arsenic Secondary Mineral Wt. % Arsenic

Arsenopyrite FeAsS

37-44% Scorodite FeAsO4�2H2O

32.5% (stoichiometric)

Arsenian Pyrite Fe(S,As)2

<0.05 to 3.5% Kaňkite FeAsO4�3.5H2O

29.1% (stoichiometric)

Goethite FeOOH

~10 to ~25%

Ferrihydrite 5Fe2O3�9 H2O

~2 to ~11%

Arsenic-Bearing Minerals in Empire Mine Waste Rock

What’s Next?

Relative Bioavailability in vivo Studies

Scheduled to begin in May 2011