Rational Design of La 0.6 Sr 0.4 CoO 3− δ ‐Based Oxygen Electrodes for Solid Oxide Cells: Integrating Machine‐Learning Prediction With Experimental Characterization

Solid oxide cells (SOCs) are promising for efficient energy conversion, but their electrochemical performances are largely limited by the conductivity, catalytic activity, and durability of oxygen electrode materials. La 0.6 Sr 0.4 CoO 3− δ (LSC)‐based oxides as promising oxygen electrodes of SOCs, are widely studied. Here, we systematically studied Fe/Ni/Mn doped LSC‐based oxides by combining machine learning (ML) prediction with experimental characterizations. Among them, La 0.6 Sr 0.4 Co 0.8 Mn 0.2 O 3− δ shows the outstanding performance. Mn dopant not only suppresses the grain growth but also induces the highest concentration of oxygen vacancies than others, enabling the high proton uptake ability (0.05 mol unit −1 at 650°C) and catalytic activity. Moreover, the LSCM shows low thermal expansion coefficient of 17.5 × 10 −6 K −1 , lower than most of Co‐based oxides. The SOC with LSCM oxygen electrode achieves a high electrolysis current density of 1.45 A cm −2 at 1.3 V and 600°C. This work demonstrates the great potential of using the ML model prediction and experimental characterizations to optimize the electrode materials.