A Reference Database of Radionuclide Sorption to Support the Safety Assessment of Deep Geological Repositories

The long-term safety assessment of a deep geological repository requires comprehensive and consistent sorption thermodynamic data for modelling geochemical scenarios in the components of the geological and engineered containment system.

Currently, no internationally recognized database exists that provides quality-assured sorption thermodynamic data for governing radionuclide transport through the barrier system of a deep geological repository. The SOREDA (Sorption Reference Database) joint project aims to develop a unified database that fulfils the need for sorption thermodynamic data on the most important dose-relevant radionuclides and the key mineral phases forming the buffer and natural barriers.

The results of sorption experiments on natural minerals often show high variability due to different mineral origins, reactive impurities, mineral crystallization, chemical pre-treatment and laboratory specific experimental conditions. Consequently, the development and parameterization of a robust sorption model requires a critical evaluation of existing sorption data and their acquisition conditions complemented by a fundamental knowledge of mineral surface chemistry and uptake mechanisms.

The sorption database developed in this study includes the following elements: Cs, Sr, Ra, Ni, Am, Cm, Np, Pu, Sn, Th, and U, as well as the low-sorbing elements I, Se, and Tc in different oxidation states. Their sorption data, along with titration data for minerals being identified as most relevant (quartz, kaolinite, illite, montmorillonite, ferrihydrite, goethite, hematite, magnetite, pyrite, calcite, and dolomite) were rigorously analysed. The data were obtained from in-house experiments, open literature, the surface complexation/ion exchange database of the Lawrence Livermore National Laboratory (Zavarin et al., 2025 [1]), and the online sorption database supported by the Japanese Atomic Energy Agency [2].

The Diffuse Double Layer (DDL) model was used for sorption modelling on quartz, carbonates, and iron minerals. Due to significant challenges in applying this model to sorption on clay minerals, the quasi-mechanistic two-site protolysis non-electrostatic surface complexation and cation exchange (2SPNE SC/CE) model (Baeyens & Bradbury, 1997 [1] and Bradbury & Baeyens, 1997 [2]) was used for these minerals instead.

Given the noticeable data gaps in the available sorption datasets for the investigated element-mineral systems, alternative approaches were considered. It was shown that the linear free energy relationship (LFER) can be utilized successfully to address these data gaps. In cases where sufficient data for LFER were unavailable, chemical and mineral analogies were reliably used.

This comprehensive sorption database supports a reliable and robust safety assessment of the barrier system in deep geological repositories. The database has strong potential for expansion to include a broader range of dose-relevant radionuclides, including their organic complexes, to be sorbed on rock-forming and other minerals.

This research was funded by the Federal Company for Radioactive Waste Disposal (BGE, Germany) as part of the SOREDA project.

 

References:

[1] Zavarin, M. et al. (2025) Proc. 19th Conf. Migration '25, New Orleans, LA, USA, in press.

[2] JAEA. “Sorption and Diffusion Database System”. https://migrationdb.jaea.go.jp/nmdb/db/sdb/search_1.jsp

[3] Baeyens, B. and Bradbury, M.H. (1997) J. Contam. Hydrol. 27(3-4), 199–222.

[4] Bradbury, M.H. and Baeyens, B. (1997) J. Contam. Hydrol. 27(3-4), 223–248.