Importance of the gas-phase error correction for O2 when using DFT to model the oxygen reduction and evolution reactions

DFT modelling of the oxygen reduction and evolution reactions (ORR and OER) habitually makes use of semiempirical corrections to oxygen in the gas phase. Although such corrections are tacit in the model, they should not be overlooked. In this article, we calculate the errors in the total energy of oxygen for commonly used exchange-correlation functionals, PW91, RPBE, PBE, and BEEF-vdW, to show that, for all functionals tested, the error is at least 0.3 eV. We discuss the impact this sizeable error in oxygen has on the modelling of the ORR and the OER. The error due to oxygen affects not only the overall equilibrium potential of the reaction, but also the energies of individual mechanistic steps. This illustrates that understanding the reasoning behind the semiempirical corrections for oxygen is important for researching new catalysts which may have different potential limiting steps.

Automated summary

The study shows that errors in the total energy of oxygen commonly used in DFT modelling of ORR and OER can have a significant impact on the overall equilibrium potential of the reactions and the energies of individual mechanistic steps. Understanding the reasoning behind the semiempirical corrections for oxygen is important for researching new catalysts with potential limiting steps.