Machine-Learning-Based Composition Analysis of the Stability of V–Cr–Ti Alloys

Machine learning methods allow the prediction of material properties, potentially using only elemental composition a molecule or compound, without knowledge molecular crystalline structures. Herein, composition-based machine properties V–Cr–Ti alloys is demonstrated. Our machine-learning-based stability qualitatively consistent with composition-dependent experimental data ductile–brittle transition temperature and swelling. Furthermore, our computational results suggest existence region, Cr+Ti ~ 60 wt.%, at significantly low temperature. This outcome contrasts reportedly content less than 10 wt.% in conventional alloys. Machine-learning-based numerical useful for design analysis metal alloys, particularly multicomponent such as high-entropy to develop materials nuclear fusion reactors.