A critical evaluation of the dissolution mechanisms of high-level waste glasses in conditions of relevance for geological disposal (GLAMOR)

Different theories have been proposed worldwide in the past for interpreting the results from experimental programmes on the dissolution of nuclear waste glasses. The common and important trend is that the dissolution rate of nuclear waste glass decreases with increasing time of exposure to the contacting solution under static conditions. The interpretations are diverging in that the decrease may be attributed to saturation effects in the solution, to protecting effects of the reaction layer formed on the glass surface, or to ion exchange processes between solution and glass. Currently these different interpretations relate to either a thermodynamically (affinity) or with kinetically controlled dissolution behaviour. The first objective of the GLAMOR1 project was to achieve a common understanding amongst the participants on the interpretation of the decrease in dissolution rate of nuclear waste glasses as observed in most experimental programmes and referring to geological disposal conditions. The project started from a selection of existing experimental data and existing analytical models as the basis of its work. A group of acknowledged international experts identified the experimental data to be used in the project and applied the models to the data. The experimental data were produced by various laboratories and under various experimental conditions (i.e. different glass compositions, pure solutions, solutions loaded with solids referring to disposal concepts). The second objective of the project was to better define and quantify the uncertainties associated with the calculations performed in modelling, for instance the uncertainty of the parameters used in the models. This modelling was done so that laboratory data would be properly interpreted and correctly extrapolated beyond the laboratory time scale of several years. As a result of this project we achieved a common view of the different glass dissolution processes, and the different individual views on certain processes were integrated into an overall picture characterised by the following observations: - Glass dissolution is strongly linked to the concentration of dissolved silica and as much as 10 000 times lower long-term dissolution rates relative to the initial dissolution rate can be achieved when Si concentrations are high. - The principal source of high Si concentration is the glass itself. As long as the silica from the glass is not consumed by reaction with near-field materials, high silica concentrations in solution are expected to be achieved in a few months, and correspondingly low glass dissolution rates. Even though there is a general agreement on these observations and on their mathematical treatment in conceptual models, there is still large uncertainty on how to interpret in detail the mechanism by which high silica concentrations slow down the dissolution rates. Two key interpretations were evaluated in detail in the project: (1) affinity-limited dissolution and (2) rate decrease by transport limitation through the reaction layer. We conclude that both affect the rate decrease. By applying the analytical models we realised a better understanding of the role of the different parameters considered in the models and of the impact of the uncertainties of these parameters. We were however unable to quantify the uncertainties associated with the calculations performed in modelling because of the large uncertainties on some of the modelling parameters still existing. The following report provides a summary involving detailed discussions of the two main models and their potential use to explain the long-term behaviour of nuclear waste glasses under conditions relevant for geological disposal conditions.