The oxidation state and asymmetric coordination of iron control activity-stability tradeoff of Fe-N-C oxygen reduction catalysts

Fe-N-C catalysts are the most promising non-precious alternatives to platinum group metals (PGMs) for the acidic oxygen reduction reaction (ORR) in proton-exchange membrane fuel cells (PEMFCs). However, their practical application is significantly limited by the intrinsic trade-off between activity and stability. Herein, we propose an advanced integrated framework combining AI-based large language models (LLMs) with operational DFT calculations to elucidate the factors controlling the ORR activity-stability trade-off in Fe-N-C catalysts under acidic conditions. Results show that Fe single atoms coordinated with asymmetric N atoms at a moderate oxidation state lead to a moderate-strength Fe-N bond. This configuration significantly reduces Fe dissolution, improves the adsorption energetics of key ORR intermediates, and ultimately breaks the activity-stability trade-off. Leveraging this computational guidance, we experimentally synthesize a FeN4 site coordinated with a mixture of asymmetric pyrrolic and pyridinic N atoms in Fe-N-C catalysts. This catalyst intrinsically maintains a moderate Fe oxidation state that facilitates an optimal Fe-N bond strength, thereby delivering a high acidic ORR cycle stability (