Hole transport in pure and doped hematite

Hematite (alpha-Fe2O3) is a promising candidate for use in photovoltaic (PV) and photoelectrochemical devices. Its poor conductivity is one major drawback. Doping hematite either p-type or n-type greatly enhances its measured conductivity and is required for potential p-n junctions in PVs. Here, we study hole transport in pure and doped hematite using an electrostatically embedded cluster model with ab initio quantum mechanics (unrestricted Hartree-Fock theory). Consistent with previous work, the model suggests that hole hopping is via oxygen anions for pure hematite. The activation energy for hole mobility is predicted to be at least 0.1 eV higher than the activation energy for electron mobility, consistent with the trend observed in experiments. We examine four dopants-magnesium(II), nickel(II), copper(II), and manganese(II/III) in direct cation substitution sites-to gain insight into the mechanism by which conductivity is improved. The activation energies are used to assess qualitative effects of different dopants. The hole carriers are predicted to be attracted to O anions near the dopants. The magnitude of the trapping effect is similar among the four dopants in their +2 oxidation states. The multivalent character of Mn doping facilitates local hole transport around Mn centers via a low-barrier O-Mn-O pathway, which suggests that higher hole mobility can be achieved with increasing Mn doping concentration, especially when a network of these low-barrier pathways is produced. Our results suggest that the experimentally observed conductivity increase in Mg-, Ni-, and Cu-doped p-type hematite is mostly due to an increase in hole carriers rather than improved mobility, and that Mg-, Ni-, and Cu-doping perform similarly, while the conductivity of Mn-doped hematite might be significantly improved in the high doping concentration limit. (C) 2012 American Institute of Physics. [http://dx.doi.org/10.1063/1.4730634]