Bayesian exploration of the composition space of CuZrAl metallic glasses for mechanical properties

TL;DR

This paper introduces a Bayesian exploration method to efficiently identify optimal compositions of CuZrAl metallic glasses with desired mechanical properties, using simulations and an automated data loop.

Contribution

It presents a novel Bayesian optimization framework for in silico design of metallic glasses, focusing on composition and process parameters with limited data.

Findings

Optimal Al concentration around 15% for yield stress

Zirconium concentration near 30% enhances properties

Cooling rate significantly affects mechanical performance

Abstract

Designing metallic glasses in silico is a major challenge in materials science given their disordered atomic structure and the vast compositional space to explore. Here, we tackle this challenge by finding optimal compositions for target mechanical properties. We apply Bayesian exploration for the CuZrAl composition, a paradigmatic metallic glass known for its good glass forming ability. We exploit an automated loop with an online database, a Bayesian optimization algorithm, and molecular dynamics simulations. From the ubiquitous 50/50 CuZr starting point, we map the composition landscape changing the ratio of elements and adding aluminium to characterize the yield stress and the shear modulus. This approach demonstrates with relatively modest effort that the system has an optimal composition window for the yield stress around aluminium concentration $c_{\rm Al} = 15$ \% and zirconium concentration $c_{\rm Zr} = 30$ %. We also explore several cooling rates ("process parameters") and find that the best mechanical properties for a composition result from being most affected by the cooling procedure. Our Bayesian approach paves the novel way for the design of metallic glasses with "small data", with an eye toward both future in silico design and experimental applications exploiting this toolbox.