Effects of alteration product precipitation on glass dissolution

Abstract

Understanding the mechanisms that control the durability of nuclear waste glass is paramount if reliable models are to be constructed so that the glass dissolution rate in a given geological repository can be calculated. Presently, it is agreed that (boro)silicate glasses dissolve in water at a rate dependent on the solution concentration of orthosilicic acid (H4SiO4) with higher [H4SiO4] leading to lower dissolution rates. Once the reaction has slowed as a result of the buildup of H4SiO4, another increase in the rate has been observed that corresponds to the precipitation of certain silica-bearing alteration products. However, it has also been observed that the concentration of silica-bearing solution species does not significantly decrease, indicating saturation, while other glass tracer elements concentrations continue to increase, indicating that the glass is still dissolving. In this study, we have used the Geochemist’s Workbench code to investigate the relationship between glass dissolution rates and the precipitation rate of a representative zeolitic silica-bearing alteration product, analcime [Na(AlSi2O6)⋅H2O]. To simplify the calculations, we suppressed all alteration products except analcime, gibbsite (Al(OH)3), and amorphous silica. The pseudo-equilibrium-constant matrix for amorphous silica was substituted for the glass pseudo-equilibrium-constant matrix because it has been shown that silicate glasses act as a silica-only solid with respect to kinetic considerations. In this article, we present the results of our calculations of the glass dissolution rate at different values for the analcime precipitation rate constant and the effects of varying the glass dissolution rate constant at a constant analcime precipitation rate constant. From the simulations we conclude, firstly, that the rate of glass dissolution is dependent on the kinetics of formation of the zeolitic phase. Therefore, the kinetics of secondary phase formation is an important parameter that should be taken into account in future glass dissolution modeling efforts. Secondly, the results indicate that, in the absence of a gel layer, the glass dissolution rate controls the rate of analcime precipitation in the long term. The meaning of these results pertinent to long-term glass durability is discussed.