Long-term alteration of basaltic glass: Mechanisms and rates

The long-term behavior study of archaeological artifacts and natural minerals and glasses revealed discrepancies between laboratory and field data. For a better understanding of the cause of these discrepancies and to reinforce the use of basaltic glass as an analog for nuclear waste glasses, this study focuses on the determination of alteration rates and processes of synthetic basaltic glass in residual rate regime. Laboratory batch experiments were performed at high surface-to-volume ratios at 90 and 30 C for more than 1000 days. In all the experiments, the residual rate regime was reached after about 6 months. The residual alteration rates at 30 and 90 C were 4.0 ± 1.0 * 10^-6 and 9.5 ± 3.2 * 10^-6 gm^-2d^-1, respectively. At 90 C, this residual alteration rate is five orders of magnitude lower than the forward alteration rate (0.8 gm^-2d^-1). Altered powders and monoliths were characterized by Transmission Electron Microscopy and Time-of-Flight Secondary Ion Mass Spectrometry. From glass core to solution, the altered materials are structured as follows: pristine glass, gel (corresponding to the palagonitic layer of natural glasses) and intergranular clays. To assess the passivating properties of this alteration film, we used solid characterization, an isotopically-tagged post-leaching experiment and the measurement of mobile species diffusion coefficients through the alteration film at different stages of reaction using various techniques (solution analysis and X-ray Reflectometry). These characterizations showed that the alteration film formed during residual rate alteration is passivating even without clogged porosity within the gel. Diffusion coefficients of water and alkali metals - respectively diffusing to and from the pristine glass - through the alteration film dropped from 10^-20 to 10^-19 m²s^-1 during the first alteration stages to 10^-25 m²s^-1 in residual rate regime. (authors)