Theoretical justification of stable ferromagnetism in ferroelectric BiFeO3 by first-principles

Abstract

The multiferroic properties of the BiFeO3 (BFO) compound have received great attention for decades. To date, there are many detailed experimental and theoretical studies on the coexistence of ferroelectric (FE) and antiferromagnetic (AFM) orderings in the BFO which can show up a weak-ferromagnetism (w-FM) by canting of the ferromagnetic (FM) sublattice; however, there are only a few reports in the literature on BFO with FM ordering at the interfaces in heterostructures and, apparently, none on bulk FM BFO. Recently, a novel FM state coexisting with FE ordering at room temperature has been reported in (0 1 2)-oriented and strained BFO thin films [Ramirez-Camacho et al. Acta Materialia 128, 451–454 (2017)]. Based on such observed FM in BFO, the aim of our work is the theoretical study on the possibility of existence and stability of a macroscopic FM order coexisting with the FE arrangement in bulk BFO to explain the causes that lead to the observed FM-FE ordering. We outline the conditions that must be fulfilled by the crystalline structure, chemical bonding, and electronic structure employing the density functional theory (DFT) + Hubbard potential (U) formalism and using a combination of different representations as density of states, band structure, electron density, electron localization function, and molecular orbital. Our work demonstrates that the stable FM ordering is consequence of a small increase of the rhombohedral distortions and is strongly associated to higher degree of covalence of the spin-up eg orbitals in the FeO bonds, and it is established by indirect exchange interactions such as superexchange and/or double-exchange interactions which are mediated by the oxygen ions located between the Fe ions. In particular, the electronic structure features of the possible FM order in the BFO compound suggest potential applications as voltage-controlled magnetic semiconductor for spin inverters in Spintronics.