The Role of Cation Exchange in Controlling Groundwater Chemistry at A s p, Sweden

Transcription

1 UCRL-JC PR EPRl NT The Role of Cation Exchange in Controlling Groundwater Chemistry at A s p, Sweden Brian E. Viani Carol J. Bruton This paper was prepared for submittal to the Amsrican Nuclear Societl:'s International High Level RadioacfSve Waste M a n a g m m t Confermce Las V e p s, NV 1Uay 1-5,1995 January 1995 ~ This isa preprintof apaper intended forpublication in a journal orproceedings. Since changes may be made before publication, this preprint is made available with the understanding that it will not be cited or reproduced without the permission of the author. P

2 DISCLAIMER This document was of the United States the University of warranty, express responsibility for the information, apparatus use would not infring specific commercial pr manufacturer, or othe es not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or the University of California. The views and opinions of authors expressed herein do not necessarily statc or reflect those of the United States Government or the University of California, and shall not be used for advertising or product endorsement purposes. I

3 DISCLAIMER Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.

4 THE ROLE OF CATION EXCHANGE IN COmOLLLNG GROUNDWATER CHEMISTRY AT ASPO, SWEDEN Brian E. Viani Lawrence Livermore National Laboratory P.O.Box 808 Livermore, California (510) ABSTRACT Construction-inducedgroundwaterflow has resulted in the mixing of relatively dilute shallow groundwater with more concentrated groundwaterat depth in the underground Hard Rock Laboratory (HRL) at Sweden. The observed compositional variation of the mixed groundwater cannot be explained using a conservativemixing model. The geochemical modeling package EQ3/6, to which a cation-exchange model was added, was used to simulate mixing between the two fluids. Tile results of modeling simulationssuggest that cation exchange between groundwater and fracture-lining clays can explain the major element fluid chemistry observed in the =.The quantity of exchanger required to match simulated with observed fluid chemistry is reasonable and is consistent with the observed fracture mineralogy. r. INTRODUCTION Predicting the chemical changes that result from excavating a repository below the groundwater table in granitic terrain is a major focus of the Swedish nuclear waste management geochemistry program. The modeling study presented here demonstrates that cation exchange between groundwater and fracture-lining clays can explain the observed major element fluid chemistry that results when groundwaters of differing composition mix due to flow induced by excavation of the Hard Rock Laboratory (HRL) at As@, Sweden. The groundwater at Asp6 becomes more saline with depth, varying from a Ca-Na-HC03 type with less than 300 mg/l C1 for near surface waters, to a Ca-Na-C1 type with greater than 6500 mg/l C1 for groundwaters from depths greater than 500 m.' Groundwater flow has been induced due to the intersection of the HRL tunnel with transmissive Carol J. Bruton Lawrence Livermore National Laboratory P.O. Box 808 Livermore, California (510) fracture zones. The predominantly downward movement of groundwater into the tunnel results in mixing between relatively dilute shallow waters and more concentrated waters found at the tunnel depth. The groundwater composition in a fracture zone was monitored before and during groundwater mixing as part of h e Large Scale R ~ O XExperiment? If a conservative mixing model involving two endmember fluids (shallow and Mu, tunnel-depth) (Table 1) is used to predict the composition of mixed groundwater, discrepanciesbetween model predictions and observations are apparent for dissolved Ca, Mg, Na, and alkalinity (HC03-). For Ca and Mg, predicted concentrations are significantly higher than measured concentrations; for Na and H C O j the opposite is observed. Banwart et a1? suggested that cation exchange can account for this discrepancy. We used geochemical modeling to simulate fluid mixing and cation exchange and compared our results to the reported2 field data. Table 1. Characteristicsof shallow and tunneldepth endmemberfluids used to predict the composition of mixed groundwater. Groundwater Depth, m Ionic strength, M PH Ca, m a MgT mg/l Na,m a C1, m a HCO3, m g L Shallow Tunneldepth Viani, B.E

5 $) looo E c Figure Chloride, mg/l Measured concentrations of Na (solid circles), Ca (open circles), and Mg (open triangles) in redox experiment water samples are compared to predictions made with a conservative mixing model (dashed lines) and generated from EQ3/6 calculations (solid lines) for 0.1 equiv exchangerk of H20. II. METHODS The geochemical modeling package EQ3/6? to which a cationexchange model was added, was used with an associated thermodynamic data base to simulate the mixing between the two fluids. This version of EQ3/6 allows the simultaneous calculation of aqueous speciation in the fluid phase, mineravfluid reactions (e.g. dissolutiodprecipitation), and equilibrium between a fluid and a cation exchanger. All reactions simulated were assumed to attain equilibrium instantaneously,le., no kinetic constraints were considered. Simulations were carried out assuming a groundwater temperature of 10 OC.A cation exchange model employing the Vanselow convention4was used for the simulations. The exchanger phase was assumed to have Details of the implethe exchange energies of ~mectite.~ mentation of this option in EQ3/6 and exam les of its application may be found in Viani and Bruton6- f The Iimitations of EQ3/6 required the process of fluid mixing to be simplified. Mixing shallow and deep groundwaters was approximated by the following sequence: 1. x liters of shallow groundwater are added to y liters of deep groundwater to form a mixed groundwater (x+y =l). 2. z equivalents of a cation exchanger whose composition is in equilibrium with the deep groundwaterare added to the mixed groundwater. 3. Aqueous and exchange equilibria are allowed to occur; supersaturatedphases are allowed to precipitate. 4. Steps 1 to 3 are repeated for differing values of x and y so as to span the range in composition from shallow to deep groundwaters. III.RESULTS AND DISCUSSION A comparison of the results of modeling calculations using EQ3/6 with the experimental data set and with a conservative mixing model shows that a relatively small quantity of cation exchanger (0.1 equiv/l of H20)can account for the behavior of Na, Ca, and Mg during the mixing of the shallow and deep groundwaters (Figure 1). Precipitation of small quantities of hematite, pyrolusite, quartz and calcite were predicted to occur during fluid mixing. However, the impact on the cation exchange equilibrium is negligible because the major cation and anion concentnitions were not significantlychanged. In order to assess whether the quantity of exchanger used to simulate fluid mixing is physically reasonable, the thickness of a layer of cation exchanger on a fracture sur- Viani, B.E. 2

6 Status of the EQ3/6 Software Package for Geochemical Modeling, Chemical Modeling of Aqueous Systems II, D.C. Melchior and R.L. Bassett (eds), , American Chemical Society, Washington, D.C. (190). face necessary to contribute the required exchange capacity to one liter of groundwater was calculated as a function of fracture aperture. The calculation was made by assuming that fractures are completely filled with fluid and that the exchangerhas a density and exchange capacity equal to that of smectite. For fracture apertures between 200 and lo00 pm the calculated thickness varies between 4 and 20 pm. These thicknesses are consistent with the quantities of cation exchanger minerals that have been identified optically and by x-ray diffraction in fractures at As@. IV.CONCLUSIONS This preliminary study establishes cation exchange as a viable mechanism for controlling the chemical evolution of groundwaters in a fracture-dominated dynamic flow system.this modeling study also strengthens our confidence in the ability to model the potential effects of fracture-lining minerals on the transport of radionuclides in a high level nuclear waste repository. Greater confidencein our modeling capabilities would be gained by detailed field sampling and testing to measure the exchange capacity and exchanger compositions in HRL rocks. To advance our understanding of the chemistry of groundwater at the HRL, further modeling will be required to include additional exchanger phases (e.g illite, mica) and trace metals (e.g. Cs, Sr, Ba) in our znalysis. ACKNOWLEGMENTS G. Sposito, Thennodynamics of Soil Solutions, Oxford University Press, New York, NY (1981). P.Fletcher and G. Sposito, The Chemical Modelling of Clay-ElectrolyteInteractions for Montmorillonite, Cluy Minerals, (1989). 6. B.E. Viani and C.J. Bruton, Modeling Ion Exchange in ClinoptiloliteUsing the EQ3/6 Geochemical Modeling Code, Y. Kharaka and A.S. Maest, A.S (eds), Water-Rock Interaction, Proceedings of the 7th International Symposium, vol. 1,73-77, A.A. Balkema, i3rookfield. VT (1942). 7. B.E. Viani and C.J. Bruton, Modeling Fluid-Rock Interaction ut Yucca Mountain, Nevada: A Progress Report. Lawrence Livermore National Laboratory, Livermore, CA, UCRLID (1992). 8. E-L. Tullborg, B. Wallin, and 0. Landstron, Hydrogeochemical Studies of Fracture Mineralsfrom Water Conducting Fractures and Deep Groundwaters at Aspo. SKB Swedish Hard Rock Laboratory Progress Report , Stockholm, Sweden (1991). This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract W-7405-ENG-48. Financial support for this project was provided by the Geochemical Modeling activity of the Hard Rock Laboratory (HRL) Project Agreement between the U.S.Department of Energy and the Swedish Nuclear Fuel and Waste Management Company. REFBENCES 1. M. Laaksoharju, Measured and Predicted Groundwater Chemistry at Aspo. SKB Swedish Hard Rock Laboratory Progress Report , Stockholm, Sweden (1990). 2. S. Banwart,.M. Laaksoharju, A-C. NiIsson, E-L.Tul1borg, and B. Wallin, The Large Scale Redox Experiment: Initial Characterizationof the Fracture Zone. SKB Asp6 Hard Rock Laboratory Progress Report , Stockholm, Sweden (1992). 3. T.J. Wolery, K.J. Jackson, W.L. Bourcier, C.J. Bruton, B.E. Viani, K.G. Knauss, and J.M. Delany, Current Viani, B.E. 3