Aluminoborosilicate waste glass dissolution under alkaline conditions at 40°C: implications for a chemical affinity-based rate equation

Abstract

Environmental context. The production of nuclear materials has generated a very large amount of highly radioactive wastes that need to be disposed of in a manner that will keep them from posing a danger for millions of years until the radioactivity decays. The process being considered for this daunting task is to contain the wastes in glass. Although studies with ancient and natural glass suggest the weathering of glass is slow, experiments are being conducted to determine the impact of this material on the natural environment and attempt to predict its long-term behaviour. The present paper briefly discusses three models that are being considered for implementing this process and the one that appears to hold the most promise. Abstract. Single-pass flow-through experiments were conducted with aluminoborosilicate waste glasses to evaluate how changes in solution composition affect the dissolution rate (r) at 40°C and pH (23°C) = 9.0. The three prototypic low-activity waste (LAW) glasses, LAWE-1A, -95A and -290A, used in these experiments span a wide range covering the expected processing composition of candidate immobilised low-activity waste (ILAW) glasses. Results suggest incongruent release of Al, B, Na, and Si at low flow-rate (q) to sample surface area (S), in units of (m s–1), (log10(q/S) < –8.9) whereas congruent release is observed at high q/S (log10(q/S) > –7.9). Dissolution rates increase from log10(q/S) ≈ –9.3 to –8.0 and then become constant at log10(q/S) > –7.9. Forward (maximum) dissolution rates, based on B release, are the same irrespective of glass composition, evident by the dissolution rates being within the experimental error of one another (r1A = 0.0301 ± 0.0153 g m–2 day–1, r95A = 0.0248 ± 0.0125 g m–2 day–1, and r290A = 0.0389 ± 0.0197 g m–2 day–1). The results also illustrate that as the activity of SiO2(aq) increases, the rate of glass dissolution decreases to a residual rate. The pseudo-equilibrium constant, Kg, (log10(Kg) = –3.7) predicted with these results is slightly lower than the K for chalcedony (log10(K) = –3.48) at 40°C. Finally, these results support the use of a chemical affinity-based rate law to describe glass dissolution as a function of solution composition.