High-strength medium-entropy alloy designed by precipitation-strengthening mechanism via machine learning

Precipitation-hardened high/medium entropy alloys (PH-HEAs/MEAs) have attracted extensive interest due to their excellent mechanical properties at both room and high temperatures. Strengthening mechanism models can be used to guide the discovery and design of new alloys, however, the parameters of the models are often lacking, limiting their use for compositional screening from a large space. We proposed a strategy combining machine learning with precipitation strengthening mechanism for high-performance PH-HEAs/MEAs. Using this strategy, we successfully synthesized an alloy Al5Ti8Co40Fe7Ni40 with a yield strength approaching 1.15 GPa, 10% higher than the best value of the existing AlTiCoCrFeNi HEAs/MEAs strengthened by L12-type nanoprecipitates. The contribution of precipitation strengthening from the L12 phase is approximately 717 MPa. Our strategy that unites the strengthening mechanism and machine learning may inspire the rapid design and discovery of precipitation-hardened alloys.