Can pristine semiconducting oxides be ferromagnetic?

The recent finding of FM in HfO$_{2}$ thin films of Coey's group has urged us to re-judge the role of TM doping in introducing FM into semiconducting oxides. Our observation of FM in undoped TiO$_{2}$, HfO$_{2}$, In$_{2}$O$_{3}$, ZnO, and SnO$_{2}$ confirmed that magnetism is possible in pristine oxide thin films, and FM is likely due to oxygen vacancies. This assumption is confirmed by our XMCD measurement on TiO$_{2}$ films: The FM in TiO$_{2}$ films is indeed intrinsic, and stems from both O-2$p$ and Ti-3$d$ electrons. In semiconducting oxides, the origin of magnetism is not due to the doping, but oxygen vacancies/defects. A big issue is how to find a more appropriate model to explain better the mechanism. We propose a model based on an electronic structure calculation using the tight binding method in the confinement configuration. Vacancy sites in TiO$_{2}$, HfO$_{2}$, In$_{2}$O$_{3}$ films could create spin splitting and high spin state, so that the exchange interaction between the electrons surrounding the oxygen vacancy with the local field of symmetry could lead to a FM ground state. Calculations give the results of 3.18 $\mu _{B}$/vac for TiO$_{2}$, 3.05 $\mu _{B}$ /vac for HfO$_{2}$ and 0.16 $\mu _{B}$ /vac for In$_{2}$O$_{3}$. This model suggests that confinement effects play an important role in shaping up magnetic properties of low dimension systems.