Play the heavy: An effective mass study for α-Fe2O3 and corundum oxides.

Abstract

Iron(iii) oxide (α-Fe2O3) is a known water splitting catalyst commonly used in photoelectrochemical cells. These cells are severely impaired by poor conductivity in α-Fe2O3, and resolving the conductivity issue is therefore crucial. One of the most intrinsic properties of matter, which governs conductivity, is the carrier effective masses. In this work, we investigate the carrier effective masses in α-Fe2O3 and other corundum oxides, including Al2O3, Cr2O3, Ga2O3, and In2O3 with different theoretical constructs: density functional theory (DFT), DFT+U, hybrid DFT, and G0W0. We find DFT sufficiently describes the carrier masses and a quasi-particle theory is only required for accuracies better than 30% for the conduction band effective mass. Additionally, we compare the density of states (DOS) and band effective mass approximations and conclude the DOS effective mass provides poor results whenever the band structure is anisotropic. We find that the charge carriers in Fe2O3 "play the heavy" since they have large effective masses that reduce conductivity and device efficiency. Finally, we conclude that the less heavy electron effective masses of other corundum oxides studied relative to Fe2O3 could contribute to efficiency improvements in Fe2O3 upon Al2O3, Ga2O3, and In2O3 coverage.