Sorption Databases for Opalinus Clay, Confining Geological Units and Bentonite: Methods, Concepts and Upscaling of Data

  pdf NTB 23-06 Sorption Databases for Opalinus Clay, Confining Geological Units and Bentonite: Methods, Concepts and Upscaling of Data(5.02 MB)

The safety analysis of deep geological repositories relies on values for the retention and transport properties of radionuclides under site-specific geochemical conditions. Retention on mineral surfaces of sedimentary hist rocks and their confining geological units (containment-providing rock zone) is one of the pillars on which the safety analysis is based. Consequently, solid-liquid distribution coefficients (R_d) for dose-relevant elements for the in-situ conditions (porewater composition, mineralogy) are required. 

This report describes the development of sorption databases (SDBs) that contain element-specific R_d values for the Opalinus Clay and confining geological units in the candidate siting regions and for MX-80 bentonite in their respective porewater types. Delivered in the form of look-up tables with R_d values for varying illite, smectite and illite-smectite mixed layer (ISML) contents in the confining rock units, R_d values were used to expand the available downhole data with high-resolution sorption profiles, and to produce the statistical distributions for the abstracted litho­stratigraphic units used in performance assessments. 

The R_d values were calculated within a "one-step" thermodynamic equilibrium system using GEMS codes, which rely on a comprehensive chemical framework based on the state-of-the-art PSI/Nagra TDB 2020, the updated ClaySor sorption model for illite and montmorillonite, and the in-situ porewaters for Opalinus Clay and bentonite. This approach has been validated through blind predictions of sorption measurements for key radionuclides on numerous rock samples from cores extracted from deep boreholes across the three siting regions.

Calculations at realistic solid-liquid ratios involved the relevant redox reactions and allowed for competition in cation exchange and surface complexation between divalent cations. For elements with existing thermodynamic sorption model parameters (i.e. selectivity coefficients and surface complexation constants), R_d values were calculated for the site-specific porewater conditions together with the upper and lower bounds of the 95% confidence interval, derived based on the model parameter uncertainties. The resulting R_d values were then arranged into look-up tables for given intervals of the total weight-% of the 2:1 clay minerals illite, smectite and ISML. 

The dominant factor influencing the sorption capacity of Opalinus Clay and the confining units is the weight content of illite/smectite/ISML in the rock samples. This factor outweighs other potential sources of variability, such as differences in porewater composition between litho­strati­graphic units, which, except for the calcareous units, are not expected to differ significantly. The variability of illite/smectite/ISML content is accounted for in statistical analysis of the sorption data for the abstracted lithostratigraphic units. Look-up tables, along with newly established correla­tions between the illite/smectite/ISML content and the R_d values, systemat­ically combined with the available high-resolution geological data, allow the calculation of R_d profiles along any strati­graphic profile of known clay mineralogical composition. In this way, the detailed downhole data plots for sorption R_d values can be prepared for reactive transport simulations. The profiles include reference, upper and lower bounding values for each abstracted unit used in performance assessment models. The present approach is superior to that applied in Stage 2 of the Sectoral Plan for Deep Geological Repositories where R_d values for only a few selected units, based on very few samples, represented the SDBs.