Deciphering Short-Range Order Driven Mechanical Enhancement in CrTaTiMo Refractory Alloys via Machine-learning Potential and Experimental Validation

Refractory high and medium entropy alloys (RH/MEAs) are promising for extreme environments owing to their superior long-term oxidation resistance and mechanical strength. In these alloys, short-range order (SRO) is a critical structural feature, yet its influence on microscale chemical heterogeneity and deformation mechanisms remains unclear. To this end, we developed a machine-learning potential (MLP) for a CrTaTiMo refractory medium-entropy alloy (RMEA), achieving near first-principles accuracy while enabling large-scale molecular dynamics simulations. The study demonstrates that the compositional segregation of two chemically distinct regions is attributed to chemical affinities of Ta–Mo and Cr–Ti, correlated with SRO. These regions impact mechanical behaviors: a Ta–Mo enriched region enhances elastic stiffness as a structural backbone, and a Cr–Ti enriched region promotes deformation twinning and dislocation activities, contributing to plasticity. These findings uncover possible mechanistic links between atomic-scale chemical ordering and both microscale compositional segregation and deformation behavior in CrTaTiMo RMEA.