A multi parameter approach for predicting initial dissolution rate of silicate glasses

The prediction of a glass’ initial dissolution rate (r0) is crucial for various applications in industry, nuclear waste management, and healthcare. The primary limitations of existing approaches are the lack of dissolution behavior classification and the choice of appropriate and robust criteria to relate the glass composition to its dissolution rate. This work introduces congruence-focused, multi-variable, and MD-agnostic methodology to estimate the dissolution rate of silicate glasses. The employed criteria reliably discriminate between congruent and incongruent dissolution behaviors and the corresponding descriptors demonstrate reasonable (R² > 0.9) and proportional correlation with ln r0 for all studied glasses. Therefore, holistic structural models that consolidate the fundamental chemistry and topology of the glass network offer a systematic means to capture complex property relationships and evaluating durability through r0.