water chemistry geology /mineralogy 02 gmw.pdf · p-gmw-01 international meeting, december 9-12,...

Poster [GMW]

Geology /Mineralogy / Water Chemistry

P-GMW-01

International Meeting, December 9-12, 2002, Reims, FranceClays In Natural And Engineered Barriers For Radioactive Waste Confinement 31

APPROACH OF SEDIMENTOLOGICAL FACTORSCONTROLLING THE VARIOUS SCALES OF

SPATIAL VARIABILITY OBSERVED IN CALLOVO -OXFORDIAN ARGILLITES OF THE SITE OF

BURE / SAUDRON (MEUSE / HAUTE - MARNE)S. Bregoin 1, C. de Fouquet 1, H. Accarie 1, A. Trouiller 2, P. Elion 2

1. Ecole des Mines de Paris, CGES-Laboratoire de sédimentologie, 35 rue St Honoré77305 Fontainebleau cedex

2. Andra, DS/GG, Parc de la Croix Blanche, 1-7 rue Jean Monnet 92298 Châtenay-Malabry cedex

IntroductionIn order to evaluate the possibilities of building a high activity and long life wastes disposal ingeological formation, an underground research laboratory is under construction by ANDRAin the East of the Paris basin (site of Meuse/Haute-Marne). The host formation chosen for thisresearch corresponds to an argillite formation attributed to part of Callovo-Oxfordian. Thisformation is approximately 130m thick. The principal level of investigation rests at 490mdepth.The objectives of this study is to evaluate the characteristics and the geological history of thecallovo-oxfordian argillites. The approach is focused on the identification of spatial variabilitywithin the callovo-oxfordian formation and sedimentological factors controlling the differentscales of variability. The study is based on well-logging and core data (geochemistry-mineralogy, geotechnical parameters) from three drillings (HTM102, EST103 and MSE101),located at 3km and 300m at the SE and 12km at NW respectively of the access well of theunderground laboratory. Due to the general dip of the layers, the layer investigated by theURL principal level is 420m deep at HTM102 and 570m deep at MSE101.The study is conducted using exploratory analysis and geostatistical methods. Results will beillustrated on geochemical and mineralogy parameters from the HTM102 drilling.

ResultsThe univariate study of the vertical evolutions and histograms of clay minerals (chlorite, illite,kaolinite, interlayered illite/smectite) present three modes. These allow to divide theformation in three reference intervals, with a first limit between 410-412m and a second at460m depth (HTM102 bore hole). The multivariate study on the scatter diagrams supports thepattern in three intervals and brings precision to the position of the limits. Thus, the first limitis statistically positioned at 412m depth, the second one at 460m.Similar analyses conducted on carbon stable isotopes indicate limits of the reference intervalsare at 440m, 407m, 393m depth. Six negative events are also observed and considered asreference levels situated at 441m, 438m, 435m, 388m, 385m and 373-374m. For oxygenstable isotopic data, the limits of reference intervals are at 440m, 420m, 393m depth and thereference levels (positive values) are positioned at 441m, 388m, 385m, 373-374m, 368m,364m. Although δ13C and δ18O reference levels coincide with carbonate beds, some of themare identified only on the isotopic signal δ13C or δ18O.The multivariate study by the Principal Components Analysis allows the estimate of therepresentation of variability, and the development of links between parameters. PCAconducted on clay minerals, carbonates (calcite and dolomite), and stable isotopes shows a

P-GMW-01

International Meeting, December 9-12, 2002, Reims, FrancePage 232 Clays In Natural And Engineered Barriers For Radioactive Waste Confinement

first axis that represents 34% (for comparison 62% for well-logging data). It shows a goodcorrelation between kaolinite, illite and chlorite and a cluster of δ13C and interlayeredillite/smectite.The variographic analysis leads to the approach of higher scale variabilities: the δ13Cvariogram presents a cyclic signal which gives evidence of a variability within the referenceintervals.

DiscussionBased on sedimentary study, the sequential analysis of the callovo-oxfordian argilliteformation shows transgressive surfaces at 440m (S1) and 362m (S2) depth (HTM102borehole), maximal floading surfaces at 417m(S1B) and 342m (S2B) depth. These twosedimentary sequences are considered as 3rd order cycles.Our statistically based reference intervals present thickness similar to that of the 3rd ordersedimentary cycles, but some limits are different. The transgressive surface S1 is wellidentified by the stable isotopes analysis but not by the clay minerals one. We proposed thatisotopes vertical evolutions partly depend on eustatism variations, whereas clay mineralsdistribution is subjected to detrital contributions and sensitive to changes in sources.Reference levels of metric thickness could correspond to 4th or 5th order sedimentary cycles.The fluctuation of oxygen isotopic signal can be attributed to climatic variations or fluidpercolations. Those of carbon isotopic signal associated to carbonate beds and pyrite could berelated to early diagenesis.The statistical analysis is a powerful tool to identify primary and secondary signalsconfirming sequential analysis and bringing new elements for the geological history of theformation.

P-GMW-02

International Meeting, December 9-12, 2002, Reims, FranceClays In Natural And Engineered Barriers For Radioactive Waste Confinement Page 233

BARRIER CHARACTERISTICS OF CLAY ANDCLAYSTONE : THE BETTON DATABASE

Johannes Gerardi

Federal Institute for Geoscience and Natural Resources, Stilleweg 2, 30655 Hannover,Germany, [email protected]

ObjectivesThe barrier characteristics of the clays and claystones in Germany are determined by theirgeochemical, mineralogical, physical, mechanical and hydraulic properties. The correspon-ding data sets and data sources are stored in BETTON together with the lithological andstratigraphic descriptions and the geographical position including the burial depth. The data-base is to contain only data describing stratigraphic horizons and not single samples withoutreference to stratigraphy or location. The BETTON database was developed to provide thedata available about a given layer at a specific place or area of interest as rapidly andcomprehensively as possible to the members of the BGR claystone research group. In thisway it shall be a tool for the site selection of future nuclear waste disposals in clay orclaystone.

Data sourcesNumerous reports are stored in the archives of BGR and the state geological surveys of theFederal Republic of Germany. These reports deal with surveys and tests on nuclear wastedeposits or disposal site selection projects, exploration of raw materials deposits, such as clayand hydrocarbons, as well as geological mapping surveys. Other sources included in theBETTON database are open reports of scientific results provided by other scientific institutesand universities, from international joint ventures of the BGR in the field of the clay andclaystone research.

Description of the BETTON databaseThe BETTON is a MS-ACCESS 2000 database within Windows NT. The user has freeaccess to the all of the data. Data input, however, requires administrator rights. A formsubdivided into registers (sub-forms) is used for data input. The geographical andstratigraphical position of the sampling site, the rock name and a layer description togetherwith a site map and a photo are stored for each sample. In addition, the database containsvitrinite reflexion data and sedimentological data, e.g., grain size (numerical values andgraphs), as well as descriptions of diagenesis, lithological texture, REM images, andphotographs of thin sections. The geochemical, mineralogical and hydraulic data make up themain part of the database. Moreover, the database contains geomechanical, geophysical, andmineral properties, CEC and sorption values, as well as numerous annotations and references.Plausibility checks are automatically conducted during input of the data. The sources of thedata in the Betton database are contained a literature database together with secondaryliterature. The results of a search in this database can be viewed either on the screen orprinted.

Searches in the BETTON databaseUser-friendly formats are provided by Betton for the output of the results of a search for data.These formats can be used for data on a selected stratigraphic horizon at a selected site, forexample. Other formats are for the results of a search for the horizons in a given area with the

P-GMW-02


properties of most interest with respect to barrier-rock properties. An format is also availablefor the output of all figures and photographs and the raw data sets used, for example, forgraphical representation of mineralogical evaluations or to calculate variance or mean values.The area of interest can be defined by the number of the topographic map sheet (scale1:25,000) or by the coordinates (Gauss-Krueger type). The results can be shown on thescreen, printed, or exported into a file. Moreover, the filter functions of ACCESS2000 areavailable. A glossary provides explanations and definitions of terms related to geologicalbarriers and other geological terms.

Availability of the databaseAt present, the database is available on CD-ROM for the internal use of the BGR clayresearch project. To facilitate revision of the database or updating of the data, the Betton wassubdivided into the interface and the database tables. After conversion of BETTON toMS-SQL Server2000, input and output of data via the BGR intranet will be possible. Thepossibility to make the database available via the internet is also being considered. AnEnglish version would then be necessary, too.

ReferenceGERARDI, J. (2002): Barriereeigenschaften von Ton- und Tonstein- Zwischenberichtzum AP 9Y32140100.– BGR, unpubl. report, 82 p., CD-ROM, Hannover.

P-GMW-03


ANDRA UNDERGROUND RESEARCHLABORATORY: INTERPRETATION OF THEMINERALOGICAL & GEOCHIMICAL DATA

ACQUIRED IN THE CALLOVIAN - OXFORDIANFORMATION BY INVESTIGATION DRILLINGS

E. Gaucher 1, C. Robelin 1, JM Matray 2, 5, G. Négrel 2, A. Vinsot 3, H. Rebours 3,A. Cassagnabère 4, A. Bouchet 4

1. BRGM, Orléans, 45, France2. ANTEA, Orléans, 45, France3. ANDRA, Bure, 55, France4. ERM, Poitiers, 86, France5. present address IRSN, Fontenay-aux-Roses, 92, France

In the scientific program of the ANDRA Meuse/Haute Marne underground researchlaboratory, two boreholes (EST204 and EST205) of 510 m in depth were drilled in a purposeof scientific characterisation. The two boreholes were set up in the place of the two highdiameter shafts in construction at the present time for the access of the future undergroundlaboratory. The scientific program dedicated to the boreholes concerns the geology, thehydrogeology, the hydrochemistry, the geomechanic, and the mineralogical and geochemicalcharacterisation of the different crossed formations. The boreholes were drilled between Juneand October 2000. The present paper is focussed on the mineralogical and geochemicalcharacterisation of the Callovian-Oxfordian formation replaced in the geological context.The works realised led to the establishment of a mineralogical and geochemical log morecomplete and precise than in the previous works. Indeed, 29 core samples (borehole EST205)were taken in the formation, every 3 m between 422 and 504 m. Physical measurements(water content, porosity, density, specific surface), geochemical analyses (major and traceelements, Cations Exchange Capacity and surface cations occupancy, leaching anions, redoxstate, organic matter concentration), and semi-quantitative mineralogy were conducted on thesamples.

As the formation is in a reduced state, a procedure of storage under nitrogen of the coresamples was applied immediately after the drilling to limit the oxidation. The core samplesare sealed in a double plasticized aluminium bag, purged of atmospheric gas and refilled withN2. The data obtained during this work show that the samples are not or low perturbed by theoxidation and can be considered as representative of the in situ conditions.

The results of major and trace analyses and of physical measurements are consistent with thevariation of mineralogical composition. The formation is more carbonateous at the top of theformation with an alternance of clayey layers and carbonateous strata. The formation is morehomogeneous in its central part with a concentration in clay minerals of 45-50%, whichcorrespond to a maximum of flooding within the area. In the upper part, the group of micasand the interstratified Illite/Smectite R0 dominate, whereas in the lower part there is a shorttransition to the interstratified Illite/Smectite R1 associated with kaolinite.An exhaustive list of minerals was established for each sample by the crossing of theobservations realised by optical and electronic microscopy and X-ray diffraction. Theelementary analyses realised with the EDX probe of the Scanning Electronic Microscope

P-GMW-03


permit a better characterisation of the structural formula of the minerals. For example, it wasshown that there is no sensu stricto dolomite in the formation but calcite-Mg or dolomite richin Ca. The characterisations by SEM establish that there are no iron oxides in the formation.A statistical analysis crossing the data of majors and traces elements with the semi-quantitative mineralogy allowed the identification of some mineralogical trap for the traceelements. This type of information is interesting to predict the migration of the radionuclidesfrom a potential nuclear waste disposal.The values of cation occupancy at the surface of the clay minerals provided good image of thechemistry of the porewater. Indeed the porewater is in equilibrium with the clay surfaces. Theleaching experiments permitted to know of the salinity of the porewater and profiles of Cl andBr concentrations were established. The Cl/Br values in the centre of the formation are closedto the seawater ratio. This can be an indication for the origin of the porewater. Somemeasurements of total reduced capacity were made for the first time and quantify the reducedstate of the rock.The treatment of the data of water content, helium and petroleum pycnometry allowed tocalculate the total porosity of the rock and to give a range of incertitude on this value. Finallythe high BET(N2) specific surface are consistent with the clayey nature of the rock.

P-GMW-04


CLAY MINERALOGY CHANGES AT THECALLOVIAN - OXFORDIAN BOUNDARY OF THE

PARIS BASIN : A SIGNAL FOR PALEO -ENVIRONMENTAL MODIFICATIONS ?

R. Mosser-Ruck, V. Huault, M. Elie

Université Henri Poincaré UMR 7566, BP 239, 54506 Vandoeuvre-lès-Nancy, France

Our pluridisciplinary study of the Callovian-Oxfordian boundary in the Paris basin wasinitially justified by the fact that palynological analysis pointed out different evolution inthree cores, located in the center part (A and B) and in the eastern part (Andra HTM 102borehole) of the Paris basin. More homogeneous results, obtained on clay and organic mattershow that this difference is controlled by the paleogeographical positions of the 3 cores andthat the origin of changes is due to eustatic rather than paleoclimatic reasons.In sedimentary formations, the variations of the clay mineralogy can be explained by severalprocesses such as i) the existence of submerged oolitic barriers or coral reefs, able topreferentially filter some clays from the continent, ii) changes of clay origins correlated withthe eustatic variations, related to emersions and increase of erosion, iii) the effect of climaticchanges, or/and iv) tectonic events. Discrimination between sources and processes may beadressed through an integrated study, involving mineralogy, palynology and organicgeochemistry.From paleogeographic maps, the location of the 3 boreholes in the sedimentary basin is thefollowing : A and B cores are far from the shoreline, and consequently from detrital sources.Late Callovian and Early Oxfordian deposits produced condensed series (around 7 meters andless than 1 meter respectively). HTM102 is closer to the shore, characterized by moreexpended deposits (80 meters at least). These assertions are in good agreement with dataobtained according to the sporomorph ecogroups (SEG) model (Abbink et al., 2001). In A andB cores, coastal SEG are in greater part, whereas in HTM 102 core, SEG are dominated bylowland and upland types. As no influence of thermal diagenesis was demonstrated byanalyses performed on organic matter (Elie & Landais, 1997) and fluid inclusions(Cathelineau et al., 1997), the mineralogical evolutions are rather explained by modificationof detrital sources and major palaeogeographical changes instead of burial diagenesis(Pellenard et al., 1999). The study of clay mineral assemblages by X-ray diffraction show thatthe Callovian-Oxfordian boundary is characterized by an assemblage of four main clays: illite,chlorite, I/S (enriched in illite), kaolinite involving a terrigenous contribution. Illite andchlorite abundances are relatively constant whereas kaolinite and I/S (and smectite in mixedlayer illite/smectite minerals) abundances are anticorrelated. Late Callovian is marked by thepresence of kaolinite and smectite poor I/S in the clay mineral assemblage (15 to 20 % each).Early Oxfordian shows the gradual disappearance of kaolinite and enrichment of smectite inI/S minerals. Concerning the SEG, the Callovian-Oxfordian boundary shows a highercontribution of the upland SEG in the 3 cores. High Pristanes/Phytanes ratios (Pr/Ph>2)correlated with the predominance of C29-steranes and βα C29-hopanes are observed at theCallovian-Oxfordian boundary. These data explain an important contribution of terrestrialorganic matter input (Gymnosperm) and oxic deposit conditions. Early Oxfordian and thenMiddle Oxfordian are characterised by low Pr/Ph ratios (0.8-1.7), the predominance of

P-GMW-04


αβ hopanes, and C27-steranes, which involve an higher marine (algal) contribution and anoxicpreservation conditions (Elie et al., 2001).These integrated data support a change of erosion conditions after the Callovian-Oxfordianboundary, related to i) a climatic evolution (wet tropical weather is in favour of kaoliniteformation) and/or ii) topographic modifications related to eustatism (kaolinite is preferentiallyformed if the drainage is strong). From the palynological point of view, a climaticmodification is unlikely. Thus, the mineralogical, organic and palynofacies changes observedin the three boreholes at the Early Oxfordian are interpreted by a modification of the distancesfrom detrital sources, related to a transgression.

Abbink O., Targarona J., Brinkhuis H., Visscher H. (2001) Global and Planetary Change 30,231-265.Cathelineau M., Ayt Ougougdal M., Elie M., Ruck R. (1997) Journée scientifique de l’Andra,Communications et Posters, Bar-le-Duc, 13.Elie M., Landais P. (1997) Journée scientifique de l’Andra, Communications et Posters, Bar-le-Duc, 13.Elie M., Huault V., Mosser-Ruck R. (2001). 20th International Meeting on OrganicGeochemistry. Nancy, France. September 10-14, 2001.Huault V. (1998) C. R. Acad. Sci. 326, 521-526.Pellenard P., Deconinck J-F., Marchand D., Thierry J., Fortwengler D., Vigneron G. (1999)C.R. Acad. Sci. 328, 807-813.

P-GMW-05


PHYSICAL, TEXTURAL, MINERALOGICAL ANDHYDROCHEMICAL CHARACTERISATION OF THE

OPALINUS CLAY ARGILLACEOUS FORMATIONA. Mª Fernández 1, M.J. Turrero 1, J. Peña 1, A.M. Melón 1, M.D. Sánchez 1, P. Rivas 1,

D. Arcos 2, P. Hernán 3

1. CIEMAT, Departamento de Impacto Ambiental de la Energía, Spain2. Enviros QuantiSci, Barcelona, Spain3. ENRESA, Empresa Nacional de Residuos Radiactivos, Spain

Clay formations are considered as favourable materials to be used as potential host rock forthe disposal of high-level radionuclide wastes. In order to determine their suitability for wastedisposal, evaluations of the hydrogeochemistry and transport mechanisms from that system tothe environment must be undertaken. A general understanding of the basic physical andchemical processes that govern solute transport through these formations is required besidessite-specific data from a site-characterisation programme. For this goal, a number activitieshave been developed in the Opalinus Clay formation at the Mont Terri URL (Switzerland).In the context of some experiments (WS-C, WS-D, WS-E, GM Task, DI, etc.), ENRESAworking groups have been performed studies of characterization of the Opalinus Clayformation to develop a conceptual, thermodynamic and transport model. The framework ofthese works is the necessity to know the mechanism of the possible migration of radionuclidesand the chemistry of the pore water, as essential data for the performance assessment of suchargillaceous systems. The methodology followed and results are presented in this paper.As part of this characterization program a complete physical, textural, physicochemical andmineralogical analysis of different drillcores has been done. The pore water chemistry hasbeen analysed by squeezing. Also the soluble salts at different solid to liquid ratios,exchangeable cations and selectivity coefficients of exchange reactions have been determined.The tests were performed with core samples of the Opalinus Clay, shaly facies, selected fromboreholes BWS-A4, BWS-A5, BWS-A6, BDI-A1 and BDI-B1.The Opalinus Clay is a consolidated claystone with a very low permeability (≈ 2·10-13 m/s).Total water content ranges from 6.5 to 7.1%, the dry density from 2.28 to 2.31 g/cm3 and thegrain density from 2.737 to 2.748 g/cm3. The porosity is 8.9% and the mean pore diameter is80 Å. The total and BET specific surface is ≈ 142 m2/g and ≈ 28 m2/g, respectively. TheOpalinus Clay has a complex mineralogy containing 58-65% of clay minerals (illite,illite/smectite mixed-layers, kaolinite, chlorite), 18% of quartz, 17-25% of carbonates, 0.7%of pyrite, 1.5% of organic carbon and numerous accessory and trace minerals. Scanningelectron microphotographs (SEM) show that some pyrite crystals present framboidal shapesand sometimes are cemented with calcite forming nodular aggregates. Calcite sometimesoccurs as foraminifera debris. Other minerals detected are K-feldspar, siderite, dolomite-ankerite, aragonite, celestite, barite, gypsum in some veins, apatite, zircon, monazite, biotite,muscovite, sphalerite and chalcopyrite. The Opalinus Clay has a cation exchange capacity of≈ 10-12 meq/100g. The pore waters are Na-Cl with high ionic strength (0.38 M) and highelectrical conductivities (11 mS/cm).As current investigations and on the basis of the experience gained from the WS experimentsand all GM Task contributions, new methodologies have been proposed. in the frame of theDI-B water sampling experiment at Mont Terri to: a) design an appropriate methodology forperforming in situ pore water sampling in the Opalinus Clay formation minimising anyalteration of the chemical composition, b) acquire experience and knowledge working in such

P-GMW-05


indurated clay formations, c) develop and evaluate new techniques to provide in situparameters controlling the pore water chemistry, such as pH, Eh, conductivity, alkalinity andpCO2, because these parameters are fundamental to broach any modelling (hydrochemistry,water-rock interaction processes, reactive transport).On March of 2002, the BDI-B1 borehole was drilled in the New Gallery. The borehole is 10m depth, upwardly inclined and crosses the bedding at high angles. This borehole has beeninstrumented to collect water that will be seeped from the formation into a sintered porousfilter. The borehole instrumentation is composed of a single inflatable packer system, 5 m-long stainless steel filter section, a central stainless steel tubing and a water pressure sensorcontroller in the sampling interval. The sampling interval is also equipped with a 1/8”stainless steel water-sampling line and two stainless steel tubes for flushing with a mixture ofAr-1% CO2 gas. The CO2 content is for avoiding outgassing problems. The water can be bledoff directly to an anoxic glove box with Ar+1%CO2 and <1 ppm of O2, preventing thepossibility of any oxidation phenomena. Inside the glove box, pore water will be distributed todifferent bottles of 2-10 mL. Some chemical determinations (pH, Eh, electrical conductivityand alkalinity) will be done in situ and other ones at CIEMAT laboratories. In case enoughamount of water would be available, the determination of pCO2, Fe(II)/Fe(III) and S2- will betried, depending on both the concentrations in solution and the analytical detection limits. ThepH, Eh and electrical conductivity will be measured by needle electrodes. The alkalinity willbe determined by an specially designed method for small volumes (2 mL) in saline waters.Due to the low permeability of the medium and the low pore water pressures detected in thiszone of the Mont Terri URL, over 6-12 months will be necessary to collect some water.

AcknowledgementsThis work has been supported by ENRESA, NAGRA and the Mont Terri Consortium.

P-GMW-06


LATERAL VARIATIONS IN THE INTERNALLAYERING OF THE BOOM CLAY FORMATION

(BELGIUM).DEPOSITION OF THE FORMATIONJeroen Mertens 1, Laurent Wouters 1, Noel Vandenberghe 2

1. ONDRAF / NIRAS. Kunstlaan 14, BE-1210 Brussels, Belgium2. K.U.Leuven. Afdeling Historische Geologie. Redingenstraat 16, BE-3000, Leuven,

Belgium

The Boom Clay Formation is a marine Oligocene deposit of several tens of meters thickness.It is the unit stratotype of the Lower Oligocene Rupelian stage and a well published exampleof cyclostratigraphy (Vandenberghe 1978, Van Echelpoel & Weedon 1990, a.o.). The Claywas deposited in the southern part of the North Sea basin. The water depth in the depositionalenvironment is estimated to be between 50 and 100m. The depositional characteristics,however, slightly changed as documented by the typical and lateral very continuoussubhorizontal layering of alternating clayey and silty layers. Each bed is several tens ofcentimeters thick. The difference between adjacent layers is due to the relative proportion ofthe different lithological components of the clay: the silt-sized, mainly quartz grains, and thefine-grained clay minerals and carbonate particles.

The internal layering, especially in the lower half of the approximately 100 m thick formation,can be easily seen in the clay pits and can be traced in borehole measurements throughout thewhole north of Belgium.

The most useful tools to enlighten the internal layering are resistivity measurements. Due tothe difference in resistivity response of a silt and a clay layer, the alternation of the layers caneasily be characterised. In some of the boreholes, an FMI or FMS was performed (FullboreFormation MicroImager). This micro-resistivity device is equipped with a large quantity ofelectrodes evenly distributed over 4 pads and 4 flafs providing high borehole coverage(typically approximately 80% in an 8-inch hole) and a high image resolution in the orders of atenth of an inch. On these FMI's, the banded structure of the clay is fully visible.

Using these measurements, it is possible to compare thicknesses of the internal layeringthroughout the area. In the figure, the internal layering of the Boom Clay Formation betweentwo boreholes located, nearly two kilometers away from each other, is compared (Dessel-1and Mol-1). Results were obtained from the two FMI plots from the boreholes. Although thelateral continuity of the layers is great, differences in thickness are already significant at thisshort distance.

The figure represents the cumulative thickness difference starting from the base of the BoomClay to its top. It clearly shows that the sedimentation rate changed gradually in the course oftime.

A detailed study of the bed thickness variations is preformed throughout the Campine Basin.

P-GMW-06


ReferencesVandenberghe N. 1978. Sedimentology of the Boom Clay (Rupelian) in Belgium. Verh. Kon.Acad. België, Klasse der Aardetenschappen. XL, pp 147.Van Echelpoel, E., Weedon, G.P. 1990. Milankovitch cyclicity and the Boom ClayFormation: an Oligocene siliciclastic shelf sequence in Belgium: Geological Magazine. 127(6). P 599-604.

Figure : Cumulative thickness differencebetween Dessel-1 and Mol-1 in meters

P-GMW-08


RADIATION - INDUCED DEFECTS IN CLAYMINERALS, TRACERS FOR PAST OCCURRENCE

OF RADIOELEMENTS AT THE EARTH'SSURFACE. GEOCHEMICAL APPLICATIONS AND

PROSPECTIVESTh. Allard, S. Sorieul, G. Calas

Laboratoire de Minéralogie Cristallographie de Paris, UMR 7590, Universités Paris 6et 7, IPGP, Case 115, 4 place Jussieu, 75252 Paris cedex 05

Understanding the processes controlling migrations of radioelements at the Earth's surface isan important issue for the long-term safety assessment of high level nuclear waste repositories(HLNWR). Geochemical and mineralogical analyses of far-field natural analogues ofHLNWR may provide useful data on ancient mobilization events. However, this may belimited when past migrations are not accounted for by the present day distribution ofradioelements. Complementary evidence of past radionuclide occurrence and transfers ingeosystems can be brought using radiation-induced defects in minerals. Clay minerals areparticularly relevant because of their widespread occurrence at the Earth's surface and theirfinely divided nature which provides high contact area with radioactive fluids. Owing to theirsensitivity to radiations, these minerals can be used as natural, in situ dosimeters. During thepast decade, main research effort has concerned kaolinite and dickite, most recent studiesconcerning native defects in montmorillonite. We will present a synthetic overview of thework done by our group on radiation defects in clay minerals.Kaolinite is known to contain radiation-induced defects (RID) which have been detected byElectron Paramagnetic Resonance (EPR), and interpreted as hole-centers located on oxygenatoms of the structure (Muller and Calas, 1993 ; Clozel et al., 1994). They are differentiatedby their nature, their production kinetics and their thermal stability. One of these defects isstable at the scale of geological periods and provides a record of past radionuclide occurrence(Ildefonse et al., 1990 ; Muller et al., 1990, 1992). Artificial irradiations have been performedusing gamma 60Co source and Van de Graaf accelerators providing beta rays, He+ and heavyion beams to simulate natural ionizing radiations and recoil nuclei effects (e.g. Allard et al.,1994). Dosimetry parameters describing the production of defects are related to the degree ofcrystalline order of kaolinites. A methodology has been subsequently proposed to determineradiation dose cumulated by kaolinite (paleodose) since its formation. The paleodose can beused to derive equivalent radioelement concentrations, provided that the age of kaoliniteformation can be constrained. This allows quantitative reconstruction of past transfers ofradioelements in natural systems. Examples of application to the study of natural analogues ofHLNWR include sediment and hydrothermal alteration systems.A first example is given for kaolinites from the Nopal I U-deposit (Chihuahua, Mexico),hosted in hydrothermally altered vocanic tufs. This study reveals past accumulation ofuranium in the mineralized breccia pipe and past leaching in the fissure network of the presentbarren rock (Allard & Muller, 1998). Dickites are also shown to have potentials for dosimetryapplications. Native RID’s are identified in fissure dickites from the El Berrocal graniticU-deposit (Spain). These defects are similar to those in kaolinites in terms of nature andstability. First results using kaolinite dosimetry parameters as models for dickite suggest thatseveral dickite generations and U-migration episodes have to be considered in the El Berrocaldeposit. Two promising prospectives appear from the data obtained on radiation defects in

P-GMW-08


clay minerals. The recent discovery of native defects in some montmorillonites widens theapplication field of dosimetry to parageneses which are also relevant for natural analogues ofHLNWR. Moreover, the dosimetry approach can be applied to the datation of clay mineralgenerations in environments where the radioelement mobility was not significant or can beconstrained using geochemical models.

RéférencesAllard T., Muller JP., Dran J-C, Ménager M-T (1994) : Radiation-induced paramagneticdefects in natural kaolinites : Alpha dosimetry with ion beam irradiation. Physics andChemistry of Minerals. 21, 85-96.Allard T. and Muller JP. (1998) Kaolinite as an in situ dosimeter for past radionuclidemigration at the Earth's surface. Appl. Geochem., 13 (6), 751-765.Clozel B., Allard T. and Muller J-P. (1994) Nature and stability of radiation-induced defectsin natural kaolinites: new results and reappraisal of published works. Clays Clay Miner.,42(6), 657-666.Ildefonse P., Muller J.P., Clozel B. and Calas G. (1990) Study of two alteration systems asnatural analogues for radionuclide release and migration. Eng. Geol., 29, 413-439.Muller J.P. and Calas G. (1993) Genetic significance of paramagnetic centers in kaolinites. InH.H. Murray, W. Bundy, and C. Harvey, Eds. Kaolin genesis and utilization, p. 261-289. TheClay Minerals Society, Boulder.Muller J.P., Clozel B., Ildefonse P. and Calas G. (1992) Radiation-induced defects in kaolinite: indirect assessment of radionuclide migration in the geosphere. Appl. Geochem., 1, 205-216.Muller J.P., Ildefonse P. and Calas G. (1990) Paramagnetic defect centers in hydrothermalkaolinite from an altered tuff in the Nopal Uranium deposit, Chihuahua, Mexico. Clays ClayMiner., 38(6), 600-608.

P-GMW-09


RESEARCH OF BENTONITE ANDMONTMORILLONITE - RICH CLAY FROM THE

CZECH DEPOSITSRichard Prikryl 1, Frantisek Woller 2, Irena Kolarikova 1, Radek Hanus 1, Jan Vejsada 1

1. Institute of Geochemistry, Mineralogy and Mineral Resources, Faculty of Science,Charles University in Prague, Albertov 6, 128 43 Prague 2, Czech Republic

2. Radioactive Waste Repository Authority, Dlazdena 6, 110 00 Prague 1, CzechRepublic

All national programmes of deep geological repository development, which consider thefractured media as the host-rock consider simultaneously the montmorillonite-bearingargillaceous rocks as the important part of engineered barriers. The Czech Republic belongsamong these countries.Two different genetic types of montmorillonite-rich raw materials are distinguished in theCzech Republic :- bentonite - in-situ weathered residual rocks related to the Tertiary alkaline volcanites in

the north and north-western parts of state territory, andmontmorillonite-rich clay for which the transportation and sedimentation in basins is

significant. The deposits of this raw-material are scattered in the sedimentary basins ofsouthern, south-eastern and western parts of the Czech Republic.

Systematic exploration of bentonite has been started in 1960’s. At present, about 20 bentonitedeposits and some 7 deposits of montmorillonite-rich clay are registered in the CzechRepublic. For reserves balance the above mentioned genetic types are traditionally divided tobentonite for foundry (high grade) with ion exchange capacity min. 45 meqv. per 100 g andcommon (low-grade) bentonite with ion exchange capacity 30 - 44 meqv. per 100 g. Reservesof the mentioned deposits are estimated to 270 mil. t. The highly predominant part of reserveshas been estimated during 1970’s and 1980’s. Bentonite was tested mainly for foundry, iron-ore peletizing, agriculture and drilling mud production. Clay was tested for agriculture andpartly for the pet litter absorbent. At present, only high-grade bentonite is mined andpredominant part of production is treated with soda ash to produce sodium-activatedbentonites.Research of bentonite from most promising deposits has been started in the middle of 1990’swith the aim to test mechanical properties of these materials and evaluate possibility to usethese materials as the buffer in the deep geological repository.At the end of last year the research programme has been started with the aim to studychemical, mineralogical and mechanical properties of bentonite and montmotillonite-richclay. According to results of single tests, some low-grade bentonites and montmorillonite-richclay exhibit most promising properties (sorption capacity, favourable swelling) in comparisonwith high-grade bentonite. Beside the detailed evaluation of betonite and its possibilities, theresearch will follow the possibility to substitute bentonite (for example in sealing) bymontmorillonite-rich clay.Results of the first year of the mentioned research will be presented in the full paper. Theseresults will be completed by the brief description of individual promising deposits.

P-GMW-10


IDENTIFICATION OF SUITABLE ARGILLACEOUSFORMATIONS FOR LILW DISPOSAL IN SLOVENIA

Janko Urbanc 1, Jasna Šinigoj 1, Peter Tomše 2

1. Geological Survey of Slovenia, Dimičeva 14, Ljubljana, Slovenia2. Agency for Radioactive Waste Management, Parmova 53, 1000 Ljubljana

Due to the growing need for a final disposal of LILW, the final solution for the short-livedLILW is the key issue of radioactive waste management in Slovenia at the moment. ARAO –the Slovenian Agency for Radwaste Management – and Geological Survey of Slovenia areintensely involved in the re-initiated site selection process for a LILW repository. By acombination of technical and volunteer approach to the site selection we wish to guaranteehigh public involvement and sufficient flexibility of the process. In the technical phase, ourtendency is to retain a larger number of potential areas/sites. As an outcome of the area surveystage, a number of potentially suitable areas were identified in 2001.

The final confirmation of site suitability will be gained by detailed field investigations duringsite characterisation and site confirmation stages. The progress of this phase of the siteselection process will strongly depend on the response of local communities where potentiallysuitable areas are identified.

With regard to the geological setting of Slovenia, argillaceous rocks are mainly considered forLILW disposal. In the geological suitability assessment the required natural predisposition ofSlovene territory was assessed in order to locate geologically suitable formations. It wasperformed by the multi-criteria decision-making evaluation with ARC/INFO technology. Theresults are compiled in a map, showing potential areas for underground and surface disposalof LILW in Slovenia. The main natural predisposition for LILW disposal was identified forsurface type of LILW disposal and for the underground type of disposal.

The argillaceous formations identified as potentially suitable for LILW disposal can beclassified as :

a) Oligocene, Miocene and Pliocene mainly unconsolidated sedimentsThese sediments were deposited in different parts of the Pannonian basin and their lithologicalcomposition can vary. Clays, marls and clayey schists alternating with sand and gravel can befound. They are extended in the NE and E parts of Slovenia and to a minor extent also in thecentral part. Special emphasis has to be given to the oligocene clay called "sivica", whichbuilds thick low permeable strata.

b) Palaeocene flysch rocksFlysch rocks are built of marls and sandstones, which alternate in layers of different thickness.Main flysch areas are in the SW part of Slovenia.

c) Carboniferous/permian claystones and shalesThese rocks are most common in the central part of Slovenia. Their thicknes can be very great- over 1000 m in some parts.

P-GMW-10


For the further assessment of the suitability of these formations for LILW disposal somelimited filed investigations have also been performed. These were mainly focused on theirhydrogeological properties.

The lowest permeabilities were found in oligocene clays (about 4 . 10-10 m/s), while inpalaeocene flysch and carboniferous/permian rocks permeabilities are normally higher mostlydue to their heterogeneus lithological composition.

P-GMW-11


BENTONITES FROM ISHIRINI (JABAL ALHASAWNAH, LIBYA)

Irena Kolarikova, Radek Hanus

Charles University, Faculty of Natural science, Prague

The Ishirini deposit (Jabal al Hasawnah, Libya) has been examined as a potential naturalanalogue of the bentonite clay barrier. The objective of this project is to qualify the observedeffects of thermal loading in several bentonites of the Ishirini area.

Although these processes could be simulated in the laboratory, the results could not beextrapolated to the time scale which is expected for the repository. For this reason it isnecessary to study natural analogues, where these processes have been acting for a long time.

The Ishirini area is a volcanic region which is represented by the number of phonolite bodiesand basaltic rocks (lava flows, volcanic cones and dikes). Two periods of volcanic activityhave been distinguished. The first (middle miocene) is composed by phonolites, in the secondperiod (plio-pleistocene) large volumes of basaltic rocks were extruded (Abdelkader, 1997).

Bentonites (Hishar formation) at Ishirini deposit are crosscut by sub-volcanic domes andbasaltic dikes which injected the bentonites after their formation. These intrusions produce alarge metamorphic aureole (silicified rings) due to the high temperature of intrusion, althoughthese intrusions get cool very quickly - tens of years (Abdelkader, 1997).

The intrusion temperature of the basalt dike has been extrapolated by using clinopyroxenegeothermometers. The temperature values are in the range of 850 to 900oC.

Bentonite samples (collected in and near the intrusive contact) have been studied in detail.Analyses were made of the whole rock and of the < 2 µm fraction. The quartz contentincreases closer to the contact, whereas the smectite content in the < 5 µm fraction increaseswith increasing distance from the contact.

Montmorillonites from the southern part of Ishirini area exhibit higher magnesium contentand occasionally high quartz, pyrite, chalcopyrite, calcite, dolomite and siderite contents. Thehigh Mg-montmorillonite can be a consequence of the fluid interaction betweenmontmorillonite and dolomite layers. Apart from the fluid composition, bentonite chemistrywas also controlled by the original composition of the altered rocks.

The effect of the thermal gradient is also evident from the oxygen isotopic composition of themontmorillonite and kaolinite fraction.

Aditionally, it is also useful to use fluid inclusion studies for the determination ofmetamorphic fluids. It could be very helpful to evaluate the temperatures and salinity of thefluid in the contact with bentonites and to establish a thermal gradient from the intrusivecontact.

P-GMW-11


ReferenceAbdelkader V. A. (1997): Radiometric age determination of the basalts and phonolites at theIshirini area. - Sebha University Archiv.

P-GMW-12


ILLITE - SMECTITE MIXED LAYERING :STRUCTURE AND STABILITY STUDIED BY HRTEM,MICROCALORIMETRY, AND ENERGY MODELLING

R. Perbost 1, J. Olives 1, M. Amouric 1, J. Rogez 2, H. Gailhanou 1, *, A. Inoue 3

1. CRMC2-CNRS, Campus de Luminy, case 913, 13288 Marseille cedex 9, France2. TECSEN, Faculté des Sciences de Saint Jérôme, case 251, 13397 Marseille cedex 20,

France3. Department of Earth Sciences, Chiba University, Chiba 263, Japan* Doctorante Andra

Interstratified structures very frequently occur in layered clay minerals. They generallyconsist of ordered or disordered mixing of two types of layers, and their study is of highinterest in various scientific fields, concerning both applied and fundamental research. In thiswork, illite-smectite mixed layering (structure and stability) is studied (1) experimentally, byusing HRTEM observations, AEM analyses, and microcalorimetric measurements, and (2)theoretically, by using energy modelling, based on interatomic potentials.

HRTEM observations, on various samples from a hydrothermal smectite-to-illite conversionseries, reveal different ordering types in the illite-smectite layer sequences: from ramdomlydisordered (R0), in the less transformed samples, to R1 and R2 ordered, with, locally, regular…isis… and …iisiis… sequences (i = illite, s = smectite) (Fig. 1). AEM microanalysesindicate a general chemical trend from montmorillonite to illite.

By microcalorimetry, performed on the same samples, the enthalpy of dissolution of eachmixed-layer sample, in a reference solution, was then measured. The enthalpy of mixing ofthe (illite and smectite) layers may then be estimated, for each sample (Fig. 3). The values ofthese enthalpies of mixing, although small, seem to be slightly negative (a few kJ per mole ofO10(OH)2, for the 50%i-50%s sample). These first experimental results seem to indicate somestability of mixed-layer minerals, with respect to the two-poles assemblage.

From an original theoretical point of view, using energy modelling, based on appropriateinteratomic potentials, it is shown that the actual (illite and smectite) layers which areinterstratified, are O1/2TITO1/2 layers (O, T, I = octahedral sheet, tetrahedral sheet, interlayer,respectively; 1/2 = half octahedral sheet) (Fig. 2). This leads to TOT polar units (TiOTs), foradjacent is layers. In addition, first simulated illite, smectite, and illite-smectite mixed-layerstructures (atomic coordinates) were obtained, by relaxation of the atoms (in various cases:order/disorder of interlayer cations, dehydrated/hydrated interlayers).

The energies of such simulated structures were also calculated from this theoretical model.The obtained energies of mixing of the layers are found to be close to 0, the lowest value, –2kJ/mol, being obtained for identical (illite and smectite) octahedral composition, orderedinterlayer cations, and one-layer hydrated smectite. It is noteworthy that such theoreticalvalues are in good agreement with the above experimental ones.

In conclusion, both experimental and theoretical results seem to indicate some stability ofillite-smectite mixed-layer minerals, although very close to the stability of the two-phasesassemblage, illite and smectite.

P-GMW-12


i

s

m

mFeuillet

TOTpolaire

TO

IT

T

T

T

T

O

O

I

i

s

m

i

s

m

mFeuillet

TOTpolaire

FeuilletTOT

polaire

TO

IT

T

T

T

T

O

O

I

TO

IT

T

T

T

T

O

O

I

i

s

m

Figure 1. : Locally ordered …isis…sequence, observed by HRTEM

Figure 2. : Structure of …isis… interstratification, withO1/2TITO1/2 actual layers, as deduced from energymodelling

-40

-30

-20

-10

0

10

20

0 0.2 0.4 0.6 0.8 1

Xi

y = m*(1-Xi)*Xi

ErreurValeur10.303-21.845m

∆ m

(kJ.mol )-1

H Figure 3. : Enthalpies of mixing ofthe (illite and smectite) layers, forthe various samples, obtained bymicrocalorimetry

P-GMW-13


CHARACTERISING PORE WATER IN CLAY ROCK :LESSONS LEARNED FROM THE MONT TERRI

PROJECTH.N. Waber 1, F. J. Pearson 2

1. Rock-Water Interaction, Bern University2. Ground-Water Geochemistry, New Bern NC

One of the goals of the geochemical program within the International Mont Terri Project wasto test and validate practical techniques for characterising pore water in clay rock. Theprocedures tested were in situ sampling from sealed boreholes and techniques for collectingpore water from core in the laboratory. A geochemical modelling task was added to providea synthesis of the geochemical data being collected and evaluate their internal consistency.The properties of the Opalinus Clay at Mont Terri are such that all experiments succeeded.Thus, it was possible not only to develop a detailed, confident understanding of the pore watergeochemistry, but to evaluate how each technique contributed to that understanding and howwell it would have succeeded on its own. This is the transfer value of the Mont Terri projectbecause other mudrocks of interest as potential repository host rock may have properties thatallow only one or two of the techniques to be successful.Four boreholes were constructed for water sampling. All were drilled upward using air about20 metres with the deepest 6 metres isolated by packers. Three boreholes drilledperpendicular to the bedding yielded sample totals of 6 to 23 l between 1996 and 2000. Thefourth borehole, drilled parallel to the bedding, yielded no water. Water samples werecollected for analyses of dissolved and suspended constituents, and in-line, fieldmeasurements of pH, Pt-electrode potential and alkalinity were made. These samples – whencorrected by geochemical modelling for pH and redox state changes during collection –represent the in situ pore water chemistry and are the standard against which data from allother techniques were judged.Direct samples of water were also extracted by high-pressure squeezing. Core samples withwater contents above 6 % yielded water at pressures of 75 Mpa and below, but pressures to512 Mpa were also used. Ion exclusion effects appeared at pressures above 200 MPa. Atlower pressures, non-reactive ion (Cl) concentrations were those of the pore water butartefacts, including oxidation of pyrite in the core, pressure dissolution and CO2 loss, effectedother solutes.Geochemical modelling evaluated the quality of all types of samples and could correct forsome sampling artefacts. The identity and quantity of reactive minerals was known frommineralogical work on the core. In situ cation exchange properties are also required and aredetermined on core samples. To fully define a model water chemistry, the concentration ofnon-reactive solutes such as Cl must also be known, as well as the pH or some proxyparameter such as the pCO2. Squeezed samples can provide the non-reactive ion contents butthe pCO2 will still be undefined. If the formation does not yield water to squeezing, aqueousleaching of the core provides the only measure of non-reactive ions.Leaching gives concentrations of soluble salts per mass of core leached. All Cl leached waspresent in the pore water but to find the pore water Cl content requires knowledge of theamount of Cl-bearing water in the sample. This is lower than the water content measured bydrying because Cl is not present in water bound to mineral surfaces. The Cl porosity of theOpalinus Clay, found by comparing rock Cl contents with those of borehole and squeezedwater samples, is 50 to 60% of the porosity based on drying.

P-GMW-13

International Meeting, December 9-12, 2002, Reims, France4 Clays In Natural And Engineered Barriers For Radioactive Waste Confinement

Absent squeezed or borehole samples, the Cl porosity must be measured on core. The relativeamounts of free and bound water can be found from surface area measurements. The freewater content of the Opalinus Clay seems consistent with its Cl-bearing water content. The Clgeochemical porosity should also be the same as the Cl diffusion porosity. The few diffusionstudies yet made are inadequate to evaluate the effects of core collection, storage andpreparation procedures on the diffusion properties measured.Measurements of pCO2, the last unknown, were made in the laboratory and in the field. In thelaboratory, core outgases into a small volume and the pCO2 after stabilisation is taken as thatof the formation. In the field, a borehole is filled with an artificial water like that estimated tobe in the formation. With time, this water takes on the properties of the pore water. Theresults of early experiments are reasonable but could still over-estimate the in situ pCO2because of bacterial activity and/or acid production by pyrite oxidation.We conclude that if borehole samples can be collected, in situ pore water chemistry can befound with some confidence. It can also be found from only squeezed or leached samplesusing geochemical modelling, but with decreasing confidence and provided pCO2, cationexchange, and the Cl-porosity are available. To minimise perturbations of in situ conditionsgreat care is required during borehole construction and instrumentation, and core samplehandling and preservation.

P-GMW-14


CATION EXCHANGE PROPERTIES OFCLAYSTONES : METHODOLOGY COMPARISONAND IMPLICATIONS ON OPALINUS CLAY PORE

WATER COMPOSITIONH.N. Waber 1, E.C. Gaucher 2, A.Ma Fernández 3

1. Rock-Water Interaction, Bern University2. BRGM, Orléans3. CIEMAT, Madrid

One approach to characterise the pore water composition in a claystone of low-permeability isbased on geochemical modelling using data from cation exchange and aqueous leachateexperiments combined with mineralogical investigations. Aqueous leachate data providegeneral information about the chemical type and - when combined with adequate (i.e.geochemical) porosity data - the salinity of the pore water and are required for the derivationof exchange selectivity coefficients. Cation exchange reactions together with carbonate andsulphide/sulphate mineral reactions largely determine the dissolved cation concentrations insuch pore water. The fast kinetics of these reactions relative to the residence time of the porewater in the claystones at Mont Terri allow a geochemical equilibrium modelling approach forthe characterisation of the in-situ pore water composition.Cation exchange and aqueous leachate data were produced by different methods on drillcorematerial from numerous boreholes across the Opalinus Clay at the Mont Terri RockLaboratory. The mineralogical composition of the samples is dominated by carbonates(calcite, dolomite/ankerite, and siderite), quartz and clay minerals. The clay fraction rangesfrom 43 wt.% to 71 wt.%. It is composed of illite, kaolinite, illite-smectite mixed-layers andminor chlorite and chlorite-smectite mixed-layers. Pyrite is present in all samples at between0.9 wt.% to 3.2 wt.%.Great precautions were taken to minimise exposure of the material to the atmosphere anddifferent storage times and techniques for sample preparation were tested. It is demonstratedthat oxygen induced by air-pressured drilling leads to pyrite oxidation and the formation ofsecondary gypsum and jarosite in the still coherent core samples. The formation of secondarygypsum consumed pore water and reduced the water loss at 20ºC by 5-10% after one monthand by 10-22% after five or more months of storage time when compared to the water loss at105ºC. Pyrite oxidation also changed the original exchange population and the formation ofsecondary jarosite falsified the K-concentration in the aqueous extraction tests by up to 200%.Sample processing in the laboratory was performed under ambient and controlled conditionsin glovebox with an oxygen content of <0.1%. Under ambient conditions, additional sulphideoxidation is induced leading to cation exchange properties that are not representative for insitu conditions. In the glovebox experiments such effects are minimised.Cation exchange properties were determined using BaCl2, NH4-Na-Acetate, NH4Cl, Ni-ethylenediamine (Ni-en), Co-ihexamine (Co-ihex) as highly selective reagents, and bydetermining exchange isotherms with Co-ihex and Ni-en. Cation exchange capacitiesdetermined by different methods agree fairly well. Significant differences exist, however, inthe fractional cation occupancies. The data produced with the BaCl2, NH4-Na-Acetatemethod under ambient conditions cannot be regarded as representative for in-situ conditions.Closer agreement is found for the Co-(ihex), Ni-(en) and NH4Cl methods, althoughuncertainties remain for the concentrations of exchanged K and Mg.

P-GMW-14


The exchange selectivity coefficients obtained from whole rock samples by the combinationof aqueous leaching and leaching with a highly selective cation agree to within about ± 0.1log unit. However, selectivity coefficients derived in this way are subjected to uncertaintieswith respect to primary mineral dissolution during the experiments. A significant error isinduced on the Na-K selectivity coefficient by effects related to pyrite oxidation duringdrilling and sample storage. The erroneous concentrations of K in the aqueous leachate leadto selectivity coefficients for the Na-K exchange that are too low by 0.2 to 0.4 log units.The modelling of exchange isotherms allows the deduction of the exchange capacity andselectivity coefficients of different active surface sites on the clay minerals. It also shows thatproton exchange cannot be completely neglected in the Opalinus Clay. In the pH rangeexpected in the pore water, Na-H and Na-K exchange take mainly place on illite-type andsmectite-type sites of the clay mixture. Only one global type of active site is observed for Na-Ca and Na-Mg exchange. The selectivity coefficient obtained for the illite-type site from theNa-K exchange isotherm compares well with that obtained from leaching and corrected forthe perturbation of the K concentrations in the aqueous extracts.Pore water compositions were modelled using the differently derived cation exchangeproperties and compared to the compositions of pore water sampled from boreholes, whichacted as reference. The best agreement is obtained with data obtained from samples processedin a glovebox within days after sampling and selectivity coefficients determined on the basisof exchange isotherms.

P-GMW-15


ASSESSMENT OF THE IN SITU pH AND Eh OF BOOMCLAY

H. Moors 1, L. Wang 1, F. Vandervoort 1, P. De Cannière 1, A. Dierckx 2

1. SCK·CEN, Waste and Disposal – R&D Geological Disposal, Boeretang 200,2400 Mol, Belgium

2. NIRAS/ONDRAF, Kunstlaan 14, 1210 Brussel, Belgium

As a candidate host rock for the disposal of nuclear waste, the Boom Clay Formation(Belgium) will act as the main barrier. In view of this role: “nuclear waste/biosphereisolator”, the Boom Clay Formation is extensively studied. Many geological claycharacteristics need to be thoroughly evaluated. Amongst them are the two most importantgeochemical parameters: pH and Eh. Both are key parameters for understanding waste-hostrock interactions. The assessment of the pH and Eh prevailing in situ is an ongoing challengefor many geochemists.The necessity of in situ measurements is that in laboratory systems it is very difficult toachieve and maintain important geochemical conditions such as: the correct content oforganic and inorganic carbon, representative anaerobic conditions and, the continuous contactof the measurement solution with the solid phase of the Boom Clay.The electrochemical principle of using the classical indicator/reference electrode set-up, is forchemists recognised as the most accurate and reliable measurement technique available forboth parameters. However, the implementation of electrode technology for in situmeasurements is not evident. Many difficulties have to be solved to profit from theadvantages of potentiometric measurements. The general encountered problems linked toelectrochemical in situ measurements are: high water pressure, recognition of the differentappearing disruptions and how to guarantee long term measurement-system stability.A design, based on the classical electrode flow-through cell concept, adapted for Boom Clay,has been used to assess pH and Eh under the most representative and geological equilibratedin situ conditions. A piezometric filter-screen is placed in a closed water circuit, together witha pump and the required number of flow-through cells. The flow-through cells are equippedwith rugged solid polymer filled electrodes (Xerolyt®, Mettler-Toledo) which have proven towork well for extended time periods (more then 460 days of continuous operation) under highwater pressures (up to 1.8 MPa). With this first preliminary set-up, steady-state values forBoom Clay water have been obtained: pH ≈ 8.0 and Eh ≈ -310 mV.An improved experimental design called: “ORPHEUS” (acronym for: Oxidation ReductionPotential and pH Experimental Underground Station) is already operational, but its multi-piezometer has not yet attained geochemical equilibrium with the surrounding Boom Clay.This piezometer is one of the major improvements of ORPHEUS, as it is the first piezometerthat contains no metal parts. This non-metallic design is meant to avoid contact of BoomClay water with metal species. A similar experimental set-up has been installed in theunderground Mont-Terri laboratory (Opalinus Clay Formation, Suisse) in the frame of thePore water Chemistry Experiment (project coordinated by NAGRA).During the in situ measurement of pH and Eh, we experienced that gas-disruptions, eitherendogenous or exogenous from origin, disturb the measurements to big extents. The timeneeded for the Boom Clay system to regain its initial geochemical equilibrium can takeseveral months.To support the in situ observations, laboratory experiments are done in special designedgeochemical-reactors. The pH is dominated by the reversible inorganic carbonate chemistry.

P-GMW-15


For Eh, the in situ – laboratory parallelism is far from evident. Although both environmentsshow the predicted “reducing” tendency of the Boom Clay, the magnitude and evolution arenot the same.In parallel with the experiments, geochemical modelling has been performed to suggestmechanisms that might rule the observed variation of pH and Eh. Preliminary modellingresults indicate that the pH of Boom Clay is likely controlled by the inorganic carbonateequilibrium while the predicted Eh is governed by oxidative pyrite dissolution. However,model simulation at this stage assumes the partial pressure of CO2 as free variable. To betterdefine the in situ Boom Clay chemistry, an accurate determination of the partial pressure ofCO2 under in situ total pressure is still a challenge.Solving the geochemical puzzle of Boom Clay will be an important asset in the evaluation ofBoom Clay as host rock for geological disposal of nuclear waste.

AcknowledgementThis work is undertaken in close co-operation with, and with the financially support of,NIRAS/ONDRAF, coordinating the National Belgian Radioactive Waste ManagementProgram.

P-GMW-16


HIGH SULPHATE CONCENTRATIONS IN SQUEEZEDBOOM CLAY PORE WATER : EFFECT OF

OXIDATION OF CLAY CORES ?M. De Craen, M. Van Geet, L. Wang, M. Put

Belgian Nuclear Research Centre (SCK•CEN), Waste and Disposal Department,Boeretang 200, 2400 Mol, Belgium

Pore water has been extracted from Boom Clay by mechanical squeezing. Clay cores wereobtained from the Mol-1 borehole, drilled in April 1997 at the SCK•CEN domain (Mol,Belgium) in the frame of a regional hydrogeological study. Immediately after drilling, thesamples were vacuum-packed in aluminium-coated plastic sheets to protect them from dryingand oxidation. After three years of storage, the pore water was extracted from the clay bymechanical squeezing (squeezing conditions: hydraulic press EMAC HP100, 40 MPa,1 week). Sample preparation was performed in anaerobic conditions.

The most striking result of the chemical analyses is the high and variable SO42- concentration

in the pore water, ranging from 30 to 20 000 mg/l. The samples with the highest sulphateconcentrations also contain more Na, K, Mg, Ca, Fe and some trace elements such as Rb, Sr,Ba and U. Note that in situ Boom Clay pore water collected from different piezometersgenerally contains less than 1 mg/l SO4

2- (Dierckx, 1997). The high and variable sulphateconcentrations measured in Boom Clay pore water may suggest that the clay cores, inparticular pyrite, has been partly oxidised. Oxidation can occur in situ (as a result of coring),can be the result of ineffective sample preservation, or can be induced during samplepreparation and squeezing.

Additional squeezing tests were performed to evaluate sample preservation and duration ofstorage, sample preparation and squeezing conditions, with respect to oxidation of the claycore. Boom Clay samples were obtained from freshly drilled clay cores (SCK•CEN boreholes2000/10 and 2001/02). Pore water extraction was performed immediately after coring (nostorage), and for a few samples after a few months of storage.

From the results of these tests, it is clear that the lowest amounts of sulphate were measured inpore waters squeezed from clay cores immediately squeezed after drilling (5 to 58 mg/l).Higher amounts of sulphate were measured in the pore water when the clay cores werepreserved for a few months (generally between 100 and 350 mg/l). In the clay cores of theMol-1 borehole (squeezed after 3 years of storage), still higher amounts of sulphate weremeasured (up to 20 000 mg/l). From these observations, it seems that there exists a correlationbetween the sulphate concentration (degree of oxidation of the clay cores) and the time ofstorage. This indicates that the preservation of the clay cores in Al-coated plastic sheets wasnot effective to protect the clay from oxidation.

Since the latter observations, the storage conditions of the clay cores are improved to betterprevent them against oxidation. Clay cores are vacuum packed in Al-coated plastic sheets asbefore, but preserved in anaerobic conditions at 4°C. Squeezed pore water of these clay coresgenerally contain less than 20 mg/l sulphate.

P-GMW-16


A preliminary geochemical modeling has been attempted to explain the measured highconcentration of sulphate. The model is based on dissolution of Boom Clay consisting ofprimary minerals identified by independent mineralogy analyses. The modeling resultsindicate that the high concentration of sulphate may result from the oxidative dissolution ofpyrite. The extent of dissolution of other minerals, which controls the concentrations of majorions, depends on the variation in pH and partial pressure of CO2. Model simulation may haveto consider CO2 degassing in order to explain the observed variation in carbonateconcentration. Iron concentration remains high comparing to the solubility of ferric hydroxidesuggesting that iron concentration may be controlled by the dissolution of primary Fe bearingminerals. Calcium concentration is governed by the dissolution of calcite at the beginningstage of Boom Clay oxidation but limited by the precipitation of gypsum as sulphateconcentration increases.

Dierckx (1997) Boom Clay in situ porewater chemistry. SCK•CEN Report. BLG-734.

P-GMW-17


HYDROCHEMICAL CHARACTERISTICS ANDGROUNDWATER EVOLUTION MODELING INSEDIMENTARY ROCKS OF THE TONO MINE,

JAPANHiroshi Sasamoto 1, Mikazu Yui 2, Randolph C. Arthur 3

1. Japan Nuclear Cycle Development Institute, Paris Office, 4-8 rue Sainte-Anne,75001 Paris, France

2. Japan Nuclear Cycle Development Institute, Tokai Works, 4-33, Muramatsu,Tokai-Mura, Naka-Gun, Ibaraki 319-1194, Japan

3. Monitor Scientific L.L.C., 3900 S. Wadsworth Blvd. #555, Denver, Colorado 80235,USA

IntroductionThe Japan Nuclear Cycle Development Institute (JNC) has carried out in-situ tests at the Tonouranium mine as part of a study to obtain data on the characteristics of various geologicalenvironments and to understand processes that occur in them. Results will be used generallyto help develop a scientific basis for the safe disposal of high level radioactive waste in deepgeological formations in Japan. Toward this end, the present study evaluates whetherequilibrium-based concepts of water-rock interaction are consistent with observed variationsin the mineralogy and hydrochemistry of the Tono mine area.

Geological settingThe Tono mine is located in central Japan. Tertiary sedimentary rocks in this area overlieCretaceous granitic rocks. The sedimentary rocks comprise a sequence of fluvial/lacustrineand marine formations within the Seto Group and Mizunami Group (the latter including indescending order the Oidawara Formation, Akeyo Formation and the Toki lignite-bearingFormation), with a combined total thickness of less than 200 m. The Toki lignite-bearingFormation unconformably overlies the basement Toki granite and consists of quartz (0.1%),plagioclase (21.5%), biotite (0.3%), amphibole (0.6%), calcite (0.1%), zeolite (0.4%), clayminerals (76.8%) and opaque minerals such as pyrite etc (0.8%). The dominant clay mineral,a dioctahedral smectite (probably montmorillonite), has a relatively high ion exchangecapacity [(cation exchange capacity of 0.56 meq/mg and exchangeable Na content of0.12 meq/mg (Iwatsuki et al., 1995)].

Groundwater chemistrySeo and Yoshida (1992) and Iwatsuki et al. (1995) describe the hydrochemistry of the Tonoregion. Hydrogen and oxygen isotope analyses indicate that the groundwaters are of meteoricorigin. Tritium concentrations are less than 1 T.U., indicating that the groundwaters have beenisolated from the surface for more than 40 years. Carbon-14 dating of groundwaters from theToki lignite-bearing Formation suggest that these solutions range in age from 13,000 to15,000 years BP (Before Present). The groundwaters vary from neutral to mildly alkaline,and from Ca-Na-HCO3 type solutions to Na-HCO3 type solutions, with increasing depth.Groundwater in deeper sections of the sedimentary rocks is strongly reducing (Eh ~ -300 mV;pH ~ 9.1).

P-GMW-17


Groundwater evolution modelingA reference groundwater composition is defined for the purpose of testing groundwaterevolution models. The compositional range of several tens of samples from boreholes is usedfor the reference values for concentrations of major cations, anions, dissolved silica and pH.The results of long-term monitoring are used for the reference Eh values for redox potential.The groundwater evolution model assumes that rain water is the primary source of Tonogroundwaters. Carbon dioxide, produced by the decomposition of organic matter, is added tothese solutions as they infiltrate into the soil zone, and the partial pressure of CO2(g) thereforeincreases. The pH, Eh and solute concentrations of the groundwater are controlled by a set ofmineral-water reactions in the rock zone. Equilibrium minerals and the respective ions whoseconcentrations are assumed to be fixed by solubility equilibrium are as follows: chalcedony(Si), calcite (Ca), kaolinite (Al) and pyrite (Eh and sulfate). Ion-exchange reactions involvingclay minerals are assumed to control the concentrations of Na, K, Mg and Ca. Ion-exchangeconstants are estimated based on batch experiments using samples from the Lower TokiFormation. The partial pressure of CO2(g) is assumed to vary from 10-4.78 to 10-2.55 bar(averaging 10-3.55 bar), based on inorganic carbon concentrations in groundwaters from thisformation. The geochemical modeling code, PHREEQE (Parkhurst et al., 1980), and itssupporting thermodynamic database, modified by JNC, were used to evaluate this conceptualmodel. Results suggest that the effect of different log PCO2 values on modeled groundwatercompositions is significant, and that model results are, overall, closely consistent with thecompositional range of the reference groundwater.

ConclusionsSimplified geochemical models of water-rock interaction are used in this study to interpret thechemistry and chemical evolution of groundwaters at the Tono in-situ tests site. Resultssuggest that the models can simulate measured groundwater compositions with reasonableaccuracy. This result helps build confidence in the use of geochemical modeling techniques todevelop an understanding of dominant geochemical reactions controlling groundwaterchemistry in rocks similar to those that could be used for the geological disposal ofradioactive wastes.

ReferencesSeo T and Yoshida H (1992), in Proceedings of 5th CEC Natural Analogue W.G., pp.141-146.Iwatsuki T, Sato K, Hama K and Sato T (1995), in MRS Proceedings, vol.353, pp.1251-1257.Parkhurst D L, Thorstenson D C and Plummer L N (1980), USGS report 80-96, 210pp.

P-GMW-18


PORE WATER CHEMICAL COMPOSITION OFOPALINUS CLAY CORE SAMPLES OBTAINED BY

SQUEEZING AT DIFFERENT PRESSURESAna Mª Fernández

CIEMAT, Departamento de Impacto Ambiental de la Energía, Spain

The argillaceous formations are considered as favorable materials to be used both asengineering and geological barriers for the disposal of high level radioactive wastes (HLRW).In this context, the knowledge of the pore water chemistry is essential for performanceassessment purposes, since the pore water composition controls the processes involved in therelease and transport of the radionuclides. However, to obtain in situ reliable data on porewater chemistry in such media of very low permeability is difficult. For this reason, differentlaboratory techniques have been developed ex situ in order to collect or determine the porewater composition of a clayey media. One of them is the squeezing technique at highpressures, in which an unaltered core sample is consolidated gradually inside a squeezer bymeans of a hydraulic compression machine. As a consequence, the intersticial fluid from thevoids is expulsed from the saturated argillaceous material, thus dissipating the excess of porewater pressure due to the stress applied.A number of factors influence how reliable compositions measured on squeezed watersamples recovered from drillcores represent the chemistry of pore water in situ: a)contamination of core samples during drilling, storage or handling (oxidation of samples,evaporative loss of moisture), b) variation of the chemical and isotopic composition ofsqueezed water due to fractionation processes as water is forced through the compacted clay,causing a decrease in ion concentration and a change in the relative proportions of ions, c)modification of squeezed water chemistry and water-rock equilibria due to thepressure/temperature régime, outgassing from squeezed water, or oxidation during squeezing.In order to investigate potential variations on chemical pore water composition obtained bymeans of the squeezing technique, several drillcores from Opalinus Clay formation have beensequentially squeezed from 25 to 200 MPa. The results of this study are presented in thispaper. Furthermore, to determine if the pore water extracted by squeezing can be consideredrepresentative of the in situ formation water, the squeezed pore water compositions werecompared with pore water compositions obtained in situ by seeping into packed-off intervals.For this study a pore water squeezer device has been used, which allows a one-dimensionalcompression of the sample by means of an automatic hydraulic ram operating downwards.The compaction chamber is made of type AISI 329 stainless steel with an internal diameter of70 mm and 500 mm high. The filtration system (a 0.5 µm stainless steel porous disk), allowsextraction of interstitial water by drainage at the top and at the bottom of the cell, and waterexpelled is collected in a polyethylene syringe. The whole system is under ambientconditions, although N2 or Ar can be flushed through the sample, at the beginning of the test,displacing all the air inside the cell.The Opalinus Clay in the Mont Terri Underground Research Laboratory (Switzerland) hasmoisture contents generally between 6.5 and 8.0%. Squeezing tests have been performed withOpalinus Clay core samples collected during drilling the BWS-A4, BWS-A5, BWS-A6boreholes, located at different positions along the Reconnaissance Gallery; and from the BDI-A1 borehole, located at the New Gallery.Pore water could be extracted by squeezing from rock samples from this site by applyingpressures over 70 MPa, between 70 and 200 MPa. About 12% to 14% of the total pore water

P-GMW-18


originally present in the rock samples was extracted. Some differences among the chemicalcomposition of pore waters obtained in sequential squeezing tests under increasing pressuresup to 200 MPa were detected. However, they are not significant. Most of the variations can beattributed to external processes to the squeezing, i.e., to handling and to the procedure ofsqueezing test. The main problem detected in some of the core samples (BWS-A4 and BWS-5samples) is the oxidation of the sample. However, the time elapsed in the squeezing test is akey parameter to avoid the oxidation. This process mainly increase the sulfate concentrationin the pore water, influencing also the concentration behaviour of the major cations. Insqueezing tests performed over shorter periods, this alteration does not occur (BWS-A6 andBDI-A1 samples). The comparison between squeezed pore waters and in situ seep watersseem to indicate that pore water samples obtained at 70-100 MPa in short period squeezingtests, are representative of the Opalinus Clay formation. pore waters. A threshold squeezingpressure of 175-200 MPa have been established for the Opalinus Clay material at Mont Terriat which the composition of the extracted pore water does not change.

ReferencesFernández, A.M., Turrero M.J. & Rivas P. 2001. Analysis of squeezed pore waters as afunction of the applied pressure in Opalinus Clay material (Switzerland). Water-RockInteraction-10. Cidu (ed.), pp. 1323-1326.Fernández, A.M, Cuevas, J. & Rivas P. 2001. Pore water chemistry of the FEBEX bentonite.Mat. Res. Soc. Symp. Pro. 663, pp. 573-588.Pearson, F.J.; Arcos, D.; Bath, A.; Boisson, J.Y.; Fernández, A.M.; Gäbler, H.E., Gaucher, E.;Gautschi, A.; Griffault, L.; Hernán, P.. & Waber, H.N. (in prep.): Geochemistry of water inthe Opalinus Clay formation at the Mont Terri laboratory. SNHGS Report Nº 30.

AcknowledgementsThis work has been supported by ENRESA, NAGRA and the Mont Terri Consortium.

P-GMW-19


EVOLUTION OF PORE WATER IN THE OPALINUSCLAY AT MONT TERRI, SWITZERLAND

Adrian Bath 1, Andreas Gautschi 2

1. Intellisci, Loughborough UK-LE12 6SZ2. NAGRA, CH-5430 Wettingen

Thick clay rock formations in some locations may be effective as natural barriers in whichradioactive waste repositories can be safely placed. The Opalinus Clay formation at MontTerri is the location of an underground laboratory in which investigation methods, propertiesand processes are being studied to assist safety assessments at repository sites in clay rocks.Geochemical data show that pore waters in the clay rock exchange very slowly with waters inadjoining formations, taking millions of years for perceptible changes to occur. This is ofinterest for three reasons – firstly it helps to explain how the pore water evolved, secondly it isevidence of how solutes move through the rock (i.e. diffusion, advection), and thirdly it is abasis for general confidence in this type of formation as a natural barrier.

The clay was deposited in marine conditions in Middle Jurassic time, subsequently buried tomore than 1000 metres, and is presently about 160 m thick at Mont Terri, 200 to 300 m belowground. It was uplifted at around early Tertiary time, buried again in mid Tertiary anduplifted again from late Tertiary time to present day. During early Tertiary erosion, a freshwater karstic system was developed in Upper Jurassic limestones. The effects of theseprocesses are evident in the pore waters that have been sampled.

Spatial variation of pore water compositions through the Opalinus Clay is the key tounderstanding geochemical evolution. Therefore hydrochemical and isotopic ‘profiles’ alongthe gallery section, obliquely across the clay rock formations, are examined. Hydrochemicalcontinuity indicates that the Opalinus Clay and Liassic claystones act as a single lowpermeability unit. Two particular aspects of how the system has evolved are considered: theorigins of water and solutes, and the transport and mixing of solutes. Mobile (i.e. non-reactive) solutes, stable oxygen and hydrogen isotope ratios and dissolved noble gas contentsprovide the evidence.

Chloride (Cl) concentrations in pore waters reach a maximum of 12,000-14,000 mg/L, abouttwo thirds that in sea water, and vary regularly towards dilute compositions in groundwaterseepages on both sides of the clay rock formation (Figure 1). The absence of a hydrochemicalanomaly in the principal zone of shear and micro-fracturing (‘main fault’) is consistent withthe observation that hydraulic conductivity of this zone is comparable with that of undisturbedrock.

P-GMW-19


Figure 1. : Chloride concentrations in Opalinus Clay pore waters

Br/Cl and SO4/Cl ratios are very close to the values for sea water and are considered to have amarine source. The oxygen isotope ratio of water, 18O/16O, varies similarly to Cl reaching amaximum value near the base of the Opalinus Clay and having minimum values in the freshwater seepages. However the correlation line for O and H isotope data does not pass throughthe isotopic composition of sea water. The saline end member has been depleted in heavy Oand H isotopes relative to sea water. This is most likely due to diffusional fractionation ofwater from the chloride content (anion exclusion effect).

Helium is produced radiogenically from natural uranium in the rock and dissipates from zonesof high helium contents to zones with lower contents, i.e. from pore waters that are lessexchanged to those that are more exchanged. This allows the minimum in situ age of porewater to be estimated at about 9 million years, being the time for steady state to be reachedbetween in situ production of He and diffusive loss into mobile groundwaters in adjacentformations (Rübel et al., 2002). Dilution of Cl and other mobile solutes and the mixing ofmolecular water as indicated by O and H isotopes are occurring by a similar diffusion process.Cl and the stable isotope ratios are effectively declining 'sources' whereas 4He is a constantsource at steady state with respect to in situ production.

Rübel, A.P., Sonntag, C, Lippmann, J., Pearson, F.J., and Gautschi, A. (2002) Solutetransport in formations of very low permeability: Profiles of stable isotope and dissolvednoble gas contents of pore water in the Opalinus Clay, Mont Terri, Switzerland. Geochim.Cosmochim. Acta (in press).

0

2000

4000

6000

8000

10000

12000

14000

16000

-50050100150200250

NW Distance from Opalinus Clay contact with Dogger limestone (m) SE

Chl

orid

e co

ncen

tratio

n (m

g/L)

Borehole waters & seepages

Squeezed waters

Leachates (reconstructed)

DoggerlimestoneOpalinus Clay

Jure

nsis

Mar

lPo

sido

nia

Shal

e

Liassiclimestones

Obt

usus

Cla

yG

ryph

aea

Lim

esto

neR

haet

ian

clay

Keuper marl &

Gipskeuper

sand

y

silty

-sha

ly

sand

y

silty

-sha

ly

mai

n fa

ult

silty

-sha

ly

carb

-rich

san

dy

Agence nationale pour la gestion des déchets radioactifs

Parc de la Croix Blanche -1/7, rue Jean Monnet - 92298 Châtenay-Malabry Cedex Tél. : (33)1 46 11 80 00 - http://www.andra.fr

©An

dra

- 205

- N

ovem

ber 2

002

- cré

dits

pho

tos

: Offi

ce d

u To

uris

me

de R

eim

s - c

once

ptio

n gr

aphi

que

: ada

msk

ides

igns

.com

water chemistry geology /mineralogy 02 gmw.pdf · p-gmw-01 international meeting, december 9-12,...

Documents