Periodic table-based compositional descriptors for accelerating electrochemical material discovery: Li-ion conductors and oxygen evolution electrocatalysts

The discovery of high-performance electrochemical materials is essential for sustainable energy technologies, yet conventional methods rely on trial-and-error experiments, time-consuming computations, and detailed structural data. To address these challenges, we introduce a periodic table-based compositional descriptor that requires only chemical formulas, enabling efficient material discovery with reversible design. We applied this approach to two key applications: fast Li-ion conductors for solid-state electrolytes and platinum-group metal (PGM)-free oxygen evolution reaction (OER) electrocatalysts. Our model identified both known and new Li-ion conductors, including anti-fluorite structures with high ionic conductivity at 600–700 K — significantly lower than traditional compounds like Li2S and Li2Se. For OER electrocatalysts, we predicted Fe0.1Co0.1Cu0.1Ag0.1W0.6 oxide, which exhibited experimentally validated performance comparable to RuO2 but at a lower overpotential. The periodic descriptor offers a scalable and efficient framework for accelerating the discovery of green energy materials, including next-generation batteries and hydrogen production, contributing to carbon neutrality.