Decrease of radionuclide sorption in hydrated cement systems by organic ligands: Comparative evaluation using experimental data and thermodynamic calculations for ISA/EDTA-actinide-cement systems

Abstract

The sufficient retardation of radionuclide release by sorption is a central issue in assessing the safety of radioactive waste disposal systems and risks posed by contaminated sites. Various types of radioactive wastes and natural environments contain organic substances that can stabilize aqueous radionuclide complexes and therefore lead to a decrease of sorption. This effect can be quantified by sorption reduction factors (SRFs). The present study focuses on testing a methodology to quantify SRFs in the presence of organic ligands for cement systems. Three approaches for the estimation of SRFs were compared for a wide range of systems including representative organic ligands (iso-saccharinic acid: ISA and ethylenediaminetetraacetic acid: EDTA) and selected key radionuclides (trivalent-, tetravalent- and hexavalent-actinides): 1) analogy with solubility enhancement factors (SEFs) derived from published solubility data, 2) calculation based on radionuclide speciation as predicted by thermodynamic calculations, and 3) direct quantification from published experimental sorption data in ternary systems. Our approach allows to critically evaluate the dependence of SRFs in various systems on the chosen method of quantification, in accordance with the data availability for a given system. Our comparisons suggest that reliable SRFs can only be derived from sorption measurements in ternary systems, i.e, where the effect can be directly measured in the presence of solid surface, radionuclide and organic ligand. On the other hand, SRFs often need to be derived in the absence of such direct evidence, and approximations based on analogies and speciation information can be applied to estimate SRFs in such cases. However, such estimates may be subject to substantial uncertainties. This should not be seen as a fundamental hindrance to the use of approximations, but uncertainties need to be clearly acknowledged and an SRF should reflect this e.g. by providing (additional) conservative estimates.