A review on viscosity and structure of high-temperature borosilicate melts

Borosilicate melts with optimal viscosity are critical for the production of high-performance glass, serving as functional materials in metallurgy and metal hot-working processes. The viscosity of these melts is intrinsically linked to their structure, which is governed by chemical composition and temperature. Consequently, the interplay between viscosity and structure in borosilicate melts has attracted significant research interest. This study reviews commonly employed methods to analyze the structure and viscosity of borosilicate melts, focusing on representative studies from 1964 to 2025 that investigate the effects of various oxide components. The addition of Na₂O, K₂O, Li₂O, CaO, BaO, CaF₂, and CeO₂ disrupts the melt network and decreases viscosity, whereas the impacts of MgO, Al₂O₃, TiO₂, La₂O₃, and Y₂O₃ remain less well understood. The observed discrepancies exist between viscosity measurements and theoretical predictions from the Arrhenius and Vogel-Fulcher-Tammann (VFT) models. This is particularly true for multicomponent oxide melts. However, these established models are still widely used in practical applications, due to the lack of alternative models with better performance across various material systems. Developing precise viscosity models may require optimizing critical parameters, integrating structural data, and employing machine learning techniques. Moreover, molecular dynamics (MD) simulations often yield viscosity predictions with substantial errors when compared to experimental results, underscoring the necessity for improved potential functions. This highlights the importance of uncovering the microscopic mechanisms linking MD structural parameters to melt viscosity.