Data-Driven Design of High-Entropy Ceramics for Extreme Environments

High-entropy ceramics present a promising material class but represent a design challenge due to the massive compositional design space. Through the use of machine learning approaches, design rules linking compositions, for example of CeO₂, Y₂O₃, and Eu₂O₃ components, with mechanical properties for accelerated selection of compositional refinement. These machine learning techniques identify optimal compositions and reveal underlying relationships by capturing the intricate multidimensional relationships between composition, processing conditions, microstructure, and mechanical performance. Our findings present a set of promising compositions that exhibit enhanced mechanical properties, ideal for extreme environment applications, such as hypersonic applications.