Combined Experimental and Theoretical Investigations of n-Type BiFeO3 for Use as a Photoanode in a Photoelectrochemical Cell

Combined experimental and theoretical investigations were performed to evaluate the potential of n-type BiFeO3 as a photoanode. While previous experimental and theoretical studies on BiFeO3 mainly focused on its ferroelectric properties, several studies have reported the advantages of BiFeO3 as a photoelectrode for solar water splitting (e.g., bandgap energy and band-edge positions relative to water reduction and oxidation potentials). However, the photoelectrochemical properties of n-type BiFeO3 have not yet been thoroughly investigated. In our experimental investigation, we developed an electrodeposition-based synthesis to prepare uniform n-type BiFeO3 thin-film electrodes. Furthermore, using a heat treatment under a N2 environment, we intentionally introduced additional oxygen vacancies into the pristine n-type BiFeO3 electrodes to increase the majority carrier density. The bandgaps, flatband potentials, photocurrent onset potentials, photocurrent generation, and photoelectrochemical stabilities of the pristine and N2-treated BiFeO3 photoanodes were investigated comparatively to improve our understanding of BiFeO3 photoanodes and to examine the effect of oxygen vacancies on the photoelectrochemical properties of BiFeO3. In our theoretical investigation, we performed first-principles calculations and demonstrated the formation of a small polaron when an extra electron was introduced into the BiFeO3 lattice. Changes in electronic states caused by the small polaron formation were carefully investigated. We also examined the effects of oxygen vacancies on electron-polaron formation and carrier concentration in BiFeO3. Using charge formation energy calculations and referencing charge transition levels to the free electron-polaron level instead of to the conduction band minimum, we showed that the oxygen vacancy is capable of serving as a donor to enhance the carrier concentration of BiFeO3. Our theoretical results agree well with our experimental findings. Together, the new experimental and theoretical results and discussion provided in this study have considerably improved our understanding of n-type BiFeO3 as a photoanode.