Effect of high saline alkaline conditions onto radionuclide transport in a CEM V/A hardened cement paste

International audienceIn the French nuclear waste management context, some Intermediate-level long-lived (ILW-LL) and Low-level long-lived (LLW-LL) waste contain high levels of soluble salts, mainly composed of NaNO3 and Na2SO4. Under repository conditions and in contact with cementitious materials, the dissolution of these salts might induce changes in thermodynamic equilibria, mineralogical transformations and even-tually, might have an effect on the migration (uptake and transport) properties of radionuclides (RN) or toxic elements (Tox). The current work is part of the Andra’s CTEC group of laboratories project, dedi-cated to the characterization of the behaviour of RN or Tox in complex environments and particularly in high saline alkaline conditions. This study is a co-operative work between CEA and Subatech which main objective is the determination of transport parameters of RN/Tox, having different physico-chemi-cal properties such as HTO, Cs+, Ni(II), Cl- and Se(IV), for high saline alkaline conditions ranging from an artificial cement pore-water (ACW) solution to different salt mixtures with ACW solution using either NaNO3, KNO3, Ca(NO3)2, and Na2SO4 resulting to ionic strengths ranging from 0.2 to 4.3 mol/kgsolution. The current work is focused on HTO and Cs results.The experimental methodology is based on diffusion experiments using standard two-reservoir set-ups. Samples used are disks (2 to 5 mm thick) of a CEM V/A (Rombas, Ciments Calcia) hardened cement paste (HCP) prepared at CEA/L3MR (water/cement= 0.43) and cured at least 6 months in ACW (pH~13.5) solution before diffusion measurements.Based on Fick first law, apparent diffusion coefficients Dapp, are determined at Subatech by modelling the cumulative activity (and/or instantaneous diffusive flux) curves in the downstream reservoir with a routine based on an analytical solution of the diffusion equation with fixed boundary conditions. CEA interpretations are performed using an in-house numerical tool (I-Mode) especially developed for the interpretation of diffusion experiments data, modelling both the cumulative activity in the downstream reservoir and the remaining activity in the upstream reservoir for given fixed initial condition.For most of the experiments, HTO steady-state diffusive flux is achieved, after around 200 days. What-ever the salt mixture and considering measurements uncertainties, the mean value of Dapp is (8.6 2.2).10-13 m2.s-1. Cs experiments are still on-going due to a very low diffusivity compared to that obtained for HTO. All experiments, performed within this study, confirm that Cs migrates slower than HTO, which can then be regarded as conservative for cesium transfer. Nevertheless, Cs diffusion pro-files into HCP samples are necessary to confirm these first trends (see for illustration Figure 1 where 137Cs and HTO diffusion data are compared for experiments in ACW condition).(a)(b)Figure 1: 137Cs and HTO though-diffusion data (a) in upstream and (b) in downstream compartment