Data-Driven Exploration of Critical Factors for Single-Phase High-Entropy Oxide Anode Materials

High-entropy oxides (HEOs) are attracting significant attention owing to their compositional tunability and structural robustness. However, the identification of specific compositional combinations that yield a single-phase structure in HEOs remains unclear owing to the immense combinatorial complexity inherent in multielement systems. This study adopts a materials informatics approach that integrates experimental synthesis data with machine learning to identify key compositional factors enabling single-phase HEO formation via solid-state synthesis. This approach extracts compositional rules and constraints favoring the formation of homogeneous rock-salt or spinel phases. Applying these insights allowed the compositional space to be efficiently explored, leading to the successful synthesis of a single-phase cobalt-free HEO exhibiting high reversible capacity and outstanding cycling stability as a lithium-ion battery anode. These findings demonstrate the effectiveness of data-driven methodologies in rational material design and highlight the potential of HEOs as sustainable materials for next-generation energy storage technologies.