Predicting the propensity for thermally activated $\beta$ events in metallic glasses via interpretable machine learning

TL;DR

This paper develops an interpretable machine learning model to predict thermally activated beta events in metallic glasses from atomic structure, achieving high accuracy and generalizability, thus advancing understanding of their atomic-level excitations.

Contribution

The study introduces a novel ML approach that accurately predicts thermally activated processes in metallic glasses using atomic environment descriptors, with successful transferability to shear transformation predictions.

Findings

ML model achieves high accuracy in predicting thermal activation propensity.

Atomic environment descriptors effectively identify resistant or compliant atoms.

Model generalizes to predict shear transformation propensity.

Abstract

The elementary excitations in metallic glasses (MGs), i.e., $\beta$ processes that involve hopping between nearby sub-basins, underlie many unusual properties of the amorphous alloys. A high-efficacy prediction of the propensity for those activated processes from solely the atomic positions, however, has remained a daunting challenge. Recently, employing well-designed site environment descriptors and machine learning (ML), notable progress has been made in predicting the propensity for stress-activated $\beta$ processes (i.e., shear transformations) from the static structure. However, the complex tensorial stress field and direction-dependent activation would induce non-trivial noises in the data, limiting the accuracy of the structure-property mapping learned. Here, we focus on the thermally activated elementary excitations and generate high-quality data in several Cu-Zr MGs, allowing quantitative mapping of the potential energy landscape. After fingerprinting the atomic environment with short- and medium-range interstice distribution, ML can identify the atoms with strong resistance or high compliance to thermal activation, at an unprecedented accuracy over ML models for stress-driven activation events. Interestingly, a quantitative "between-task" transferring test reveals that our learnt model can also generalize to predict the propensity of shear transformation. Our dataset is potentially useful for benchmarking future ML models on structure-property relationships in MGs.