Comparative Evaluation of Surface Complexation Models for Radionuclide Sorption by Diverse Geologic Materials

This chapter summaries a major international modelling exercise, co-ordinated by the OECD Nuclear Energy Agency, in which independent scientific teams applied thermodynamic sorption models (TSMs) to a number of experimental adsorption data sets. A wide variety of models was employed to simulate and predict the data. In all the test cases, reasonable, broadly similar TSM approaches were adopted, but based on wide diversity of assumptions and methods of parameter estimation. The models were able to realistically, and with some predictive power, simulate the experimental data for a range of substrates, radionuclides and chemical conditions.However, sorption modelling has not reached a stage approaching standardisation. Basic features such as the stoichiometry and structure of surface complexes and mathematical formulations for such model components as the EDL are subject to debate. In addition, key model input parameters such as site populations are not well defined (particularly for natural substrates).As a result, the numerical values of optimised model parameters are highly model- dependent, which means that, in the present study, it has not been meaningful to compare individual model parameters (such as log K values for surface complexes). If a consensus is reached on model components, and uniform modelling approaches are adopted, it will be appropriate to do such a comparison.Until that time, modellers need to recognise that model parameters can typically not be used directly in other models, but that they need to be scaled or re-fitted.Although the modelling strategies differed among the teams, all were guided by a single principle, representation of sorption in terms of mass action and mass balance laws. The generally satisfactory results of this intercomparison suggest that these types of models have inbuilt chemical plausibility and predictive capabililty.