Review links computation, processing and AI to design high-entropy alloys

In 2026, an international team led by the Advanced Institute for Materials Research at Tohoku University and POSTECH published a review in Progress in Materials Science that integrates computational methods, processing science and artificial intelligence to guide the design of structural high-entropy alloys. Corresponding authors Renhao Wu and Hyoung Seop Kim argue that the complexity of these alloys, which mix several principal elements in comparable proportions, requires combining density functional theory, molecular dynamics, phase-diagram calculations, phase-field modeling and crystal plasticity simulations to connect atomic interactions, phase stability, microstructure evolution and mechanical behavior. The review further shows that machine learning and physics-based simulation can extend these tools into data-driven alloy design. It also links manufacturing simulations for additive manufacturing and severe plastic deformation to microstructural outcomes, including examples involving core-shell and reverse-gradient structures that improve mechanical performance and hydrogen-embrittlement resistance. The team envisions a closed-loop platform in which computation and artificial intelligence identify promising compositions, simulations explain the underlying mechanisms, and experiments validate results and feed new data back into the models. The framework aims to replace intuition-heavy trial and error with a more systematic, predictive path for developing high-entropy alloys.