Reactive transport modeling of glass alteration in a fractured vitrified nuclear glass canister: From upscaling to experimental validation

Abstract Performance assessment of geological nuclear waste repositories entails modeling of the long-term evolution of the aqueous alterations of the fractured nuclear glass block, because the time scales under consideration are of several thousands of years and hence beyond the range of any direct experimental perspectives. In this study, our objective is to bridge the gap between the reservoir-scale flow and transport simulations and the micron-scale modeling of the glass-water interfacial processes by providing quantitative evaluation of the aqueous alteration of glass at the block scale. In particular, calculations of the equivalent diffusive, hydraulic, and alteration kinetics properties and reactive transport simulations are discussed. Prior to performing reactive transport modeling at the scale of the glass canister, the preferred upscaling techniques were first applied to a synthetic fracture network system with ends to compare the results of the borosilicate glass alteration with the discrete fracture modeling and the equivalent porous medium approach. The evolution of the altered glass obtained from reactive transport modeling applied to several realizations of the equivalent fracture network tessellation was compared to the experimental data of the aqueous alteration test of a nonradioactive full-scale SON68 glass canister. The proposed model agrees with the experimental data and offers, for the first time, an opportunity to better understand the impact of fracturing as the convection due to the radioactivity acting as a heating source on the corrosion of nuclear glass.