First-principles studies of FeS2using many-body perturbation theory in theG0W0approximation

Abstract

We present a theoretical study on iron pyrite using density-functional theory (DFT) and the $GW$ approximation to many-body perturbation theory. The fundamental band gap of iron pyrite is determined by iron 3$d$ states at the valence band edge and a sulfur 3$p$-dominated conduction band at $\ensuremath{\Gamma}$. The gap is quite sensitive to structural changes as well as to the applied electronic structure method. We found that this $p$-dominated band does not play a significant role for the optical absorption, leading to a large difference between the optical and fundamental band gaps of iron pyrite. As a consequence the $GW$-corrected energies result in no considerable change of the optical band gap as compared to standard DFT, both being in reasonable agreement with experiment. However, we show that the fundamental band gap is reduced to about $0.3$ eV in $GW$, which may contribute to the low open-circuit voltage of about $0.2$ V observed in iron pyrite solar cells, representing a serious bottleneck for photovoltaic applications. To demonstrate that this unconventional reduction of the $p$-$d$ gap is not unique for iron pyrite, similarities for FeS${}_{2}$ in the marcasite structure are presented.