Oxygen matters

Abstract

Now, Chris-Kriton Skylaris and colleagues, in a collaboration between the University of Southampton and Johnson Matthey, study the impact of oxidation effects on Pt nanoparticles (NPs) to the ORR by computational means. To that end, they perform grand canonical minimization Monte Carlo simulations using the recently developed MACE-MP-0 machine learning interatomic potentials, which allows the treatment of the large Pt NPs (Pt₃₅₃ plus oxygen atoms, up to a total of 896 atoms) with nearly density functional theory accuracy at a fixed chemical potential.

The researchers obtain the Pt NP’s most stable oxygen content at each given voltage and derive a volcano plot that associates a rate-determining step and onset overpotential to each oxygen content. These results are thus a consequence of both thermodynamic considerations and site-blocking effects due to the presence of oxide species. The latter effects are studied in detail via a deletion energy analysis, which reveals the heterogeneity of surface oxide species. In addition, a shift in the d-band centre is found with the oxygen content highlighting two distinct regimes, which is related to the formation of an oxide phase at higher oxygen content. Finally, a kinetic model that includes the above considerations is derived.