Mapping Shear Transformation Zones in Silica Network Glass

ABSTRACT Understanding nanoscale behavior in glasses is challenging due to their inherent long‐range disorder and the absence of clearly defined defect regions. However, shear transformation zone theory suggests that certain “soft spots,” which are prone to plastic rearrangements, can be identified even in stress‐free configurations. Various predictive methods, such as energy approximations and local structural indicators, have been proposed with varying success. The local yield stress (LYS) method, however, has demonstrated strong potential by identifying these soft spots based on local mechanical properties. In this study, we use the LYS method to analyze the distribution of soft spots in 2D silica network glass, where the local critical stresses are probed using mechanical simulations. Samples with different heterogeneities are generated through a Monte Carlo bond‐switching algorithm and subjected to shear via an athermal quasistatic (AQS) deformation protocol. Our results reveal a high correlation between regions with low critical stresses and the locations of rearrangement events, especially for events that happen early in the shear simulations, though this correlation decreases as the system approaches failure. These findings highlight the method's effectiveness in predicting plastic activity in structural glasses, and confirming that soft spots can be identified from the initial configuration. Additionally, we optimized the probing length scale for different levels of heterogeneity and analyzed the resulting statistical distributions of the local critical stresses, which are essential for accurate macroscopic modeling of silica glass.