Machine learning-assisted design of strong and ductile BCC high-entropy alloys

Machine learning-based feature importance analysis was utilized to systematically identify the critical material parameters and elemental effects governing the yield strength of BCC high-entropy alloys (HEAs). We found that the shear modulus mismatch and Mo element exhibit highest contribution to yield strength among these parameters and elements, respectively. Through adjustment of Ti/Mo content ratio, an as-cast single-phase BCC refractory HEA was successfully developed with a yield strength of 1169.3 MPa and an elongation of 18.8%, which originates from the activation of multiple slip systems. Our findings provide a new strategy for accelerating design of HEAs with outstanding mechanical properties.

• NN
  National Natural Science Foundation of China

  Awards: Grant Numbers 52441409, 52474389, 52441409
• NK
  National Key Research and Development Program of China

  Award: 2024YFB3714400
• BA
  Basic and Applied Basic Research Foundation of Guangdong Province

  Award: 2025A1515010505