作者
Olha Marinich,George Dan Miron,Dmitrii A. Kulik,Maria Marques Fernandes,Bart Baeyens
摘要
The ClaySor 2023 model package within the GEM-Selektor software includes an updated version of the two-site protolysis non-electrostatic surface complexation and cation exchange model for illite and montmorillonite, as well as the first implementation of the generalised caesium sorption model for illite. These models have been harmonised with the most recent PSI Chemical Thermodynamic Database 2020, resulting in the updated sorption thermodynamic database for illite and montmorillonite (STDB 2023). STDB 2023 includes surface complexation constants and cation exchange selectivity coefficients for a wide range of radionuclides and transition metals, including Cs(I), Cd(II), Co(II), Fe(II), Mn(II), Ni(II), Pb(II), Ra(II), Zn(II), Am(III), Cm(III) (for illite only), Eu(III), Pu(III, IV), Np(IV, V), Sn(IV), Th(IV), Nb(V), Pa(V), and U(IV, VI). These parameters were derived from extensive experimental sorption data collected over decades at the Paul Scherrer Institute and supplemented with data from the open literature (160 datasets in total). Surface complexation constants and cation exchange selectivity coefficients are now provided with uncertainty estimates, determined using the 95% confidence intervals obtained via the Monte Carlo sampling method implemented in the GEMSFITS parameter optimization tool. To address gaps caused by missing experimental data, the surface complexation constants for Cd(II) and Fe(II) on illite, as well as those for Pu(III, IV), Np(IV), and U(IV) on both illite and montmorillonite, were estimated based on linear correlations between the stability constants of aqueous and surface complexes. For geochemical calculations, STDB 2023 is supplemented with selectivity coefficients for naturally occurring major cations, sourced from recent literature. The applicability and performance of the model are discussed, and versions of the model and database adapted for the PHREEQC and the Geochemist’s Workbench codes are also provided. • The sorption thermodynamic databases (STDB 2023) for illite and montmorillonite were expanded to cover over 20 dose-relevant radionuclides, and were updated and harmonised with the PSI Chemical Thermodynamic Database 2020. • The 2SPNE SC/CE model, implemented in GEM-Selektor and used together with STDB 2023, has proven reliable in predicting radionuclide sorption on real containment-providing rocks and has been applied to develop sorption and diffusion databases for Opalinus Clay and MX-80 bentonite in their respective porewater chemistries for the safety assessment of the Swiss deep geological repository. • Two-level uncertainty evaluation – of individual parameters and overall model performance for pH-edge and sorption isotherm data – supports robust model reparameterization. • Published and newly generated experimental datasets used in model reparameterization are presented in the Supplementary Materials, with propagated confidence intervals for radionuclide distribution ratios.