Machine Learning Applied to High Entropy Alloys under Irradiation
Amin Esfandiarpour, Sri Tapaswi Nori, Silvia Bonfanti, Mikko J. Alava, Antoni Wadowski, Wenyi Huo, Ł. Kurpaska, Michał Pecelerowicz, Jan Wróbel
Abstract
High‐entropy alloys (HEAs) represent a frontier in materials science, offering many promising features suitable for high‐demand applications in nuclear and space sectors, such as exceptional mechanical properties. However, a major challenge in these fields is accurately predicting the behavior of HEAs under extreme conditions, such as radiation exposure or elevated operating temperatures, in order to maintain the integrity of the materials. Machine learning (ML) provides powerful tools to address this challenge. ML techniques, including ML interatomic potentials (MLIP), enable the modeling and prediction of complex behaviors in HEAs. This review focuses on ML to enhance the understanding of phase stability, mechanical properties, and radiation damage prediction in these complex alloys. The potential of ML to accelerate the discovery/optimization of new HEA compositions with good performance under extreme conditions is also discussed. Ultimately, the aim is to highlight the transformative role of ML in the field of HEAs under extreme conditions, in light of developing novel materials suitable for harsh environments.