Advances and challenges in high-performance cathodes for protonic solid oxide fuel cells and machine learning-guided perspectives

Protonic solid oxide fuel cells (P-SOFCs) have garnered significant attention due to their high power density and efficiency in operating at 400–700 oC. The development of high-performance cathode materials, characterized by excellent proton, oxide-ion, and electron conductivity, catalytic activity for oxygen reduction reaction, and long-term stability, is essential and urgently needed for realizing high-efficiency P-SOFCs. Recently, machine learning (ML) has emerged as a powerful tool in materials science, playing a central role in transitioning away from traditional approaches for developing new materials. In this review, recent advances of high-performance cathodes are summarized, and the challenges associated with their developments are highlighted. Furthermore, the potential ML-guided perspectives in terms of predicting proton, oxide-ion, and electron conductivity, catalytic activity, hydration ability, and stability for addressing these challenges are detailedly discussed, providing insights into the design and optimization of high-performance cathodes. Finally, the difficulties faced are presented for better utilization of ML in developing high-performance cathodes. In a word, this review not only presents the latest advances and challenges in high-performance cathodes for P-SOFCs but also highlights the promising role of ML in guiding their development.