Magnetic oxide semiconductors

Abstract

In 2000, it was theoretically predicted that ferromagnetism (FM) at high temperature could be obtained in many semiconductors such as ZnO, GaAs, GaN, etc., if we dope Mn plus a certain concentration of holes into these systems. The magnetic ordering in those compounds was suggested to originate from the Ruderman-Kittel-Kasuya-Yoshida interaction of localized moments of dopants via 2p holes or 4s electrons. Inspired by this idea, many experimentalists have tried to dope transition metals (TM) into many oxides such as ZnO, TiO2, SnO2, In2O3, etc., with the hope to obtain room temperature FM in semiconductors, in order to be able to exploit both charge and spin in the same devices. Actually, room temperature FM was observed; however, the phenomenon is not exactly as what theorists have proposed. The finding of FM in undoped HfO2 thin films in 2004 has first given some alert to the magnetism community to rejudge the real role that a dopant indeed plays. More recently, experimental observations of FM for various oxides such as TiO2, HfO2, In2O3, ZnO, CeO2, Al2O3, and MgO in low-dimensional structures have confirmed that FM is certainly possible for undoped oxide semiconductors. It was suggested that FM might stem from oxygen vacancies and/or defects that were formed at the surface and interfaces. Research on very thin films and nanoparticles of Diluted Magnetic Oxide semiconductors (DMSO) has pointed out that downscaling magnetic oxide semiconductors to nanometer scale should be an important step, in order to make them ferromagnetic. It opens a door to exploit the bright side of nanoworld in this field of spintronics. The domain of DMSO research still requires a lot of efforts of the world wide research groups toward higher levels with the hope to bring it closer to a realization of spintronic devices based on DMSO materials.