Understanding Active Sites in Pyrolyzed Fe–N–C Catalysts for Fuel Cell Cathodes by Bridging Density Functional Theory Calculations and ⁵⁷Fe Mössbauer Spectroscopy

Abstract

Pyrolyzed Fe–N–C materials are promising platinum-group-metal-free catalysts for proton-exchange membrane fuel cell cathodes. However, the detailed structure, oxidation, and spin states of their active sites are still undetermined. As such, <sup>57</sup>Fe Mössbauer spectroscopy has identified FeN<sub><i>x</i></sub> moieties as the most active sites, with their fingerprint being a doublet in low-temperature Mössbauer spectra. However, the interpretation of the doublets for such materials has lacked theoretical basis. Here, we applied density functional theory to calculate the quadrupole splitting energy of doublets (Δ<i>E</i><sub>QS</sub>) for a range of FeN<sub><i>x</i></sub> structures in different oxidation and spin states. The calculated and experimental values are then compared for a reference Fe–N–C catalyst, whereas further information on the Fe oxidation and spin states was obtained from electron paramagnetic resonance, superconducting quantum interference device, and <sup>57</sup>Fe Mössbauer spectroscopy under external magnetic field. The combined theoretical and experimental results identify the main presence of FeN<sub><i>x</i></sub> moieties in Fe(II) low-spin and Fe(III) high-spin states, whereas a minor fraction of sites could exist in the Fe(II) <i>S</i> = 1 state. From the analysis of the <sup>57</sup>Fe Mössbauer spectrum under the external magnetic field and the comparison of calculated and measured Δ<i>E</i><sub>QS</sub> values, we assign the experimental doublet D1 with a mean Δ<i>E</i><sub>QS</sub> value of around 0.9 mm·s<sup>–1</sup> to Fe(III)N<sub>4</sub>C<sub>12</sub> moieties in high-spin state and the experimental doublet D2 with a mean Δ<i>E</i><sub>QS</sub> value of around 2.3 mm·s<sup>–1</sup> to Fe(II)N<sub>4</sub>C<sub>10</sub> moieties in low and medium spin. These conclusions indicate that D1 corresponds to surface-exposed sites, whereas D2 may correspond either to bulk sites that are inaccessible to O<sub>2</sub> or to surface sites that bind O<sub>2</sub> weaker than D1.