Benchmarking various levels of theory for first principles modeling of iron oxide surfaces and interfaces

Surfaces of iron and its oxides are scientifically relevant due to their performance as photo-anodes for water oxidation as well as their formation during corrosion processes. Finding an accurate ab-initio approach to appropriately describe surface phenomenon of iron and its oxides is critical to enable modeling of the charge accumulation, energetics and bond formation mechanisms governing the reactivity of iron oxide surfaces. Prior work has established that incorporation of the random-phase approximation (RPA) improves the accuracy of adsorption and surface energy calculations but these techniques have not been systematically applied to iron oxides. Here, we present a benchmarking study for various levels of theory such as onsite Coulomb interaction, hybrid functionals and quasiparticle GW methodology to appropriately modulate surface phenomenon in oxides. A detailed study of GW convergence from various starting points is presented for iron monoxide establishing hybrid-level orbitals as the optimum starting point for self-consistent GW calculations. The study also establishes the hybrid functional approach as an ideal trade-off between accuracy and computational cost towards efficient convergence of electronic wavefunction and formation energy for electrochemical applications.