Machine learning accelerated design of a family of AlxCrFeNi medium entropy alloys with superior high temperature mechanical and oxidation properties

This work implemented machine learning (ML) approach to map the relationship between temperature, alloying elements and yield strength in multi-component alloys. Then AlxCrFeNi medium-entropy alloys (MEAs) were developed and a two-phase structure, formed by the spinodal decomposition mechanism, was observed. With increasing Al content, the high temperature mechanical properties dramatically improved. Our developed AlxCrFeNi MEAs (x > 0.8) offer low density and excellent mechanical properties, superior to conventional alloys. The oxidation behavior of AlxCrFeNi MEAs (x > 0.8) at 1000 °C was explored and the oxidation mechanism was identified. This work has identified a promising family of MEAs for high temperature structural applications.