Design of corrosion-resistant eutectic high-entropy alloys via hybrid data-driven and expert-guided strategies

Eutectic high-entropy alloys (EHEAs) leverage multiphase eutectic microstructures to achieve exceptional cast-ability and mechanical properties. However, their inherent heterogeneity often leads to micro-galvanic corrosion, limiting practical utility in aggressive environments. Conventional alloy design approaches struggle to navigate the vast compositional space of EHEAs, necessitating the development of innovative strategies to balance performance trade-offs effectively. The present work proposes a hybrid approach that combines machine learning (ML) techniques, including Multilayer Perceptron Classification, Gradient Boosted Regression, Support Vector Regression models, and uncertainty quantification, with corrosion expertise to rapidly identify corrosion-resistant EHEAs. This yields the novel AlCoCrFeNi1.7 EHEA, which shows a 33.3 % improvement in toughness and 58.97 % increase in pitting corrosion resistance compared to the benchmark AlCoCrFeNi2.1 EHEA. By strategically reducing Ni content in Al-Co-Cr-Fe-Ni EHEAs, we optimize the size and Cr concentration of Cr-rich nanoprecipitates (Cr-NPs) within the B2-phase matrix. In the AlCoCrFeNi1.7 EHEA, the dislocation loops around the larger Cr-NPs enhance strength, while the increased Cr content in Cr-NPs promotes Cr2O3 enrichment in the passive film, thereby improving its corrosion resistance. This phenomenon demonstrates the dual mechanisms of performance enhancement. The integration of ML techniques with materials science expertise provides a generalizable framework for designing multifunctional alloys.