Abstract
Fe-N-C catalysts are emerging as potential alternatives to platinum in the oxygen reduction reaction (ORR) for fuel cell cathodes. The challenge in optimizing these catalysts lies in their structural complexity and the multiplicity of reaction pathways. Here, we employ a series of model catalysts with varying amounts of Fe-N x and Fe nanoparticles (NPs) and estimate their turnover frequency (TOF) for apparent H 2 O and H 2 O 2 production at different catalyst loadings. This approach highlights the importance of the surface site density (SD) of Fe-N x moieties in determining the overall ORR activity, selectivity, and even stability. We uncover that increasing the SD of Fe-N x moieties fosters the indirect 4e -ORR pathway and consequently promotes their TOF towards preferential H 2 O production. In contrast, Fe NPs, often formed at high Fe content, behave as anticatalysts (or spectators) in this context. Indeed, an online inductively coupled plasma-mass spectrometry (ICP-MS) study reveals that higher SD can lead to faster leaching of Fe-N x moieties during operation, resulting in accelerated activity decline. Taken together, the comprehensive understanding of the intricate dependence of catalytic activity and stability on the nature and amount of Fe species provides a basis for new design principles of next-generation Fe-N-C catalysts.