Phase formability and nanomechanical properties in nanostructured multi-principal element alloys: Combinatorial and data-driven studies

The identification of multi-principal element alloys with optimal properties from a vast composition space is a daunting task, particularly for high-entropy alloys (HEAs) composed of five or more elements. Despite increasing theoretical and experimental investigations, the majority of HEAs are developed through a time-consuming trial-and-error approach. Here we demonstrate a combinatorial and data-driven methodology based on a composition gradient TiZrHfNbTa library. Using synchrotron X-ray diffraction and high-speed nanoindentation, we characterized the phase formability and mechanical properties, respectively. We correlated the compositions with the distributions of structures, elastic moduli, and hardness in the TiZrHfNbTa system and proposed a kinetic phase formation criterion based on critical cooling rates. Moreover, we evaluated high-throughput data extrapolation models including the volume fraction of triple junctions, the Hall-Petch relationship, and machine learning. Our methodology opens up new avenues to the accelerated discovery of multi-principal element alloys by revealing relationships among compositions, phase formability, and mechanical properties.