One‐pot synthesized twin perovskite oxide composite of BaCe0.5Fe0.5O3−δ (BCF), comprising cubic and orthorhombic perovskite phases, shows triple‐conducting properties for promising solid oxide electrochemical cells. Phase composition evolution of BCF under various conditions was systematically investigated, revealing that the cubic perovskite phase could be fully/partially reduced into the orthorhombic phase under certain conditions. The reduction happened between the two phases at the interface, leading to the microstructure change. As a result, the corresponding apparent conducting properties also changed due to the difference between predominant conduction properties for each phase. Based on the revealed phase composition, microstructure, and electrochemical properties changes, a deep understanding of BCF's application in different conditions (oxidizing atmospheres, reducing/oxidizing gradients, cathodic conditions, and anodic conditions) was achieved. Triple‐conducting property (H+/O2−/e−), fast open‐circuit voltage response (∼16–∼470 mV) for gradients change, and improved single‐cell performance (∼31% lower polarization resistance at 600°C) were comprehensively demonstrated. Besides, the performance was analyzed under anodic conditions, which showed that the microstructure and phase change significantly affected the anodic behavior.

Fractionally doped perovskites oxides (FDPOs) have demonstrated ubiquitous applications such as energy conversion, storage and harvesting, catalysis, sensor, superconductor, ferroelectric, piezoelectric, magnetic, and luminescence. Hence, an accurate, cost-effective, and easy-to-use methodology to discover new compositions is much needed. Here, we developed a function-confined machine learning methodology to discover new FDPOs with high prediction accuracy from limited experimental data. By focusing on a specific application, namely solar thermochemical hydrogen production, we collected 632 training data and defined 21 desirable features. Our gradient boosting classifier model achieved a high prediction accuracy of 95.4% and a high F1 score of 0.921. Furthermore, when verified on additional 36 experimental data from existing literature, the model showed a prediction accuracy of 94.4%. With the help of this machine learning approach, we identified and synthesized 11 new FDPO compositions, 7 of which are relevant for solar thermochemical hydrogen production. We believe this confined machine learning methodology can be used to discover, from limited data, FDPOs with other specific application purposes.