Impacts of glass composition, pH, and temperature on glass forward dissolution rate

John D. Vienna,Joseph V. Ryan,Sebastien N. Kerisit,James J. Neeway

doi:10.1038/s41529-018-0042-5

John D. Vienna, Joseph V. Ryan + Show 2 more

Open Access

https://doi.org/10.1038/s41529-018-0042-5

Copy DOI

Abstract

Nuclear waste glasses dissolve at the forward dissolution rate (rf) in very dilute aqueous solutions, which can isolate the impact of the glass composition from solution feedback and alteration product effects. While it has long been known that pH and temperature (T) strongly impact rf, the impacts of glass composition have remained uncertain. In this work, rf data from 19 nuclear waste glasses were used with the aim of identifying the effect of glass composition on rf. The rf values were modeled as: rf = k010–ηpHexp(−Ea/RT), with k0, η, Ea, and R, respectively, being the intrinsic rate constant, pH coefficient, apparent activation energy, and gas constant. However, no predictive correlation could be established between the individual model parameters (log[k0], η, and Ea) and glass composition for the glasses considered in this study, an outcome that was attributed to the strong positive correlation between the log[k0] and Ea parameters. Therefore, a model was fitted directly to the combined rf from all 19 glasses. This approach showed that 90% of the variation in rf data could be accounted for solely by T and pH effects. Therefore, any composition effects must be relatively small. After normalizing for differences in pH and T, the only notable differences in rf between the glasses were found to correlate with variations in the fraction of glass forming tetrahedra contributed by tetrahedral boron, f([4]B), with an abrupt threshold at a high value of f([4]B) (~0.22), where higher rf are predicted with no discernable composition effects below the threshold.

Highlights

A common method to treat radioactive wastes is through vitrification, where the waste is converted to a glass matrix, commonly in the borosilicate family.1,2 In scenarios of long-term waste glass disposal, the most likely mechanism of release of radioactive and chemical hazards from the glass is through reactions with water
Of the characteristic rates listed above, only rf is independent of solution feedback from dissolved glass and near-field materials. This rate is the focus of this work, which is aimed at isolating the intrinsic impact of waste glass composition on dissolution rates
The present study focuses on the range of rf data available on borosilicate nuclear waste glasses

Summary

INTRODUCTION

A common method to treat radioactive wastes is through vitrification, where the waste is converted to a glass matrix, commonly in the borosilicate family. In scenarios of long-term waste glass disposal, the most likely mechanism of release of radioactive and chemical hazards from the glass is through reactions with water. Of the characteristic rates listed above, only rf is independent of solution feedback from dissolved glass and near-field materials This rate is the focus of this work, which is aimed at isolating the intrinsic impact of waste glass composition on dissolution rates. Statistical analysis of a large data set covering a wide compositional range and including data at multiple pH and T values is needed to maximize the likelihood of reliably identifying correlations between rf and glass composition For this reason, the present study focuses on the range of rf data available on borosilicate nuclear waste glasses. The present study focuses on the range of rf data available on borosilicate nuclear waste glasses Within this range of glass compositions, we attempt to statistically determine the effect of the various glass components and glass structure on the dissolution rate of the glass. The root mean squared error (RMSE), an estimate of the differences between

AND DISCUSSION

Findings

METHODS