Exploration and Modeling of Structural changes in Waste Glass Under Corrosion

The corrosion of nuclear waste glasses is a complex process involving adsorption, diffusion, ionexchange, hydrolysis and precipitation of mineral alteration phases. These processes occur in parallel and their relative rates change significantly over the long term. Corrosion models are required to accurately reflect the long-term durability of vitrified high level waste so that geologic repositories can be designed to meet the targets of stability and security, and at the same time, make better utilization of the storage volume by increasing the density of HLW in the glass. There are uncertainties in the existing corrosion models that have forced the current design of vitrified wasteforms to be extremely conservative. Of particular concern is the so-called residual dissolution rate which is observed in long-term laboratory leaching tests after 2-3 months. At this stage of the corrosion reaction, the combined effects of an altered surface layer and saturation of the contacting aqueous environment in silica reduce the initial dissolution rate by as much as 10,000 times. The residual rate is a critical design specification for both the wasteform and the repository, but validation of the atomic/molecular process or mechanism which controls this rate is still in question. Most workers agree that it is primarily dependent on the altered layer and its ability to limit mass transport, AND the effects of silica saturation in the contacting solution and its influence on the thermodynamic driving force for continued dissolution. This project aims to improve the understanding of these coupled phenomena by exploring new methods of analysis to probe atomic/molecular scale changes in both the altered layers and in the contacting solution, and thereby provide new information that can be used to update the existing long-term corrosion models. This study focuses on the use of nuclear magnetic resonance (NMR) to provide detailed information about atomistic connectivity (bonding), proximity (spatial localization) and order (crystal nucleation) within the altered surface layers of glass powders after static dissolution at a high surface area to volume ratio for times up to 6 months. The NMR studies explored several glass compositions (AFCI, simplified SON68 and ISG) using isotopic enrichment or depletion of selected species in the glass (Si, B and Li) to follow their exchange with solution species, and vice-versa. Other experiments were performed with fibers under dynamic conditions, simpler binary and ternary glasses, neutron depth profiling (NDP), saturated glass solutions (SGS) and thermodynamic modeling with Geochemists Workbench to both supplement the NMR results and to more specifically address the role of solution saturation and mineral precipitation. In addition to AFCI and SON68, simplified versions of these glasses were designed to reduce the complexity of the NMR spectra, including the international simplified glass (ISG) composition which was designed by a global coalition of researchers; very simple binary and ternary glasses were also employed to test specific effects. Some of these glasses were synthesized using raw materials enriched or depleted in Si, B or Li isotopes to follow their exchange with solution species. Saturated glass solutions (SGS) were made by total dissolution of finely ground powders to explore solution speciation effects at near equilibrium conditions. Very high surface area to volume ratios (~100,000) were used in the reaction experiments to provide sufficient surface sensitivity in the NMR analyses. Cross-polarization NMR was able to detect the so-called hydrated interphase (between pristine glass and the gel layer) based on changes in composition and Q-speciation. All of the glass powders, except the ISG, showed the retention of Al in the gel layer; in contrast to its tetrahedral coordination in the bulk glass, it