In 2000, Dielt theoretically predicted that the ferromagnetism 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.
Reports of Curie temperatures well above room temperature for wide-bandgap oxides doped with a few percent of transition-metal cations have triggered intense interest in these materials as potential magnetic semiconductors.

The origin of the magnetism is debated; in some systems, the ferromagnetism can be attributed to nanoparticles of a ferromagnetic secondary phase, but in others, properties are found which are incompatible with any secondary phase, and an intrinsic origin related to structural defects is implicated: Our experimental results on TiO2, HfO2, In2O3, ZnO, and SnO2 confirmed that magnetism is certainly possible in pristine oxide thin films, and the observed ferromagnetism is most probably due to oxygen vacancies. The assumption for FM due to oxygen vacancies/defects in TiO2 thin films is strongly confirmed by our X-ray magnetic circular dichroism measurements (XMCD): There is a presence of XMCD signals at both O K and Ti L2,3 edges. It shows that the FM in TiO2 films stems from both O-2p and Ti-3d electrons. Our theoretical model also suggests that confinement effects must play some key role in shaping up magnetic properties of low dimension systems.

A new picture of defect-based magnetism is emerging. At present, progress is impeded by the lack of reproducibility of magnetic properties between laboratories. There is a desperate need for a dilute magnetic ‘fruitfly’ — a system that is easy to prepare and reproduce. Device applications can follow by design or by serendipity. Once the mechanism is better understood, the challenge will be to generate stable and controllable defect structures where we can reap the benefit of this unusual high-temperature ferromagnetism.