A machine-learning approach to predicting and understanding the properties of amorphous metallic alloys

Abstract

There is a pressing need to shorten the development period for new materials possessing desired properties. For example, bulk metallic glasses (BMGs) are a unique class of alloy materials utilized in a wide variety of applications due to their attractive physical properties. However, the lack of predictive tools for uncovering the relationships between BMGs' alloy composition and desired properties limits the further application of these materials. In this study, a machine-learning (ML) approach was developed, based on a dataset of 6471 alloys, to enable the construction of a predictive ML model to describe the glass-forming ability and elastic moduli of BMGs. The model's predictions of unseen data were found to be in good agreement with most experimental values. Consequently, we determined that an alloy with a large critical-casting diameter would likely have a high mixing entropy, a high thermal conductivity, and a mixing enthalpy of approximately −28 kJ/mol, and that a BMG with a small average atomic volume would likely have a high elastic modulus. The efficacy of ML was demonstrated in furnishing a mechanistic understanding and enabling the prediction of metallic-glass properties.