Transition metal doping of semiconductors by ion beams - diluted vs. granular magnetism

Diluted magnetic semiconductors (DMS) are based on common semiconducting material like GaAs, Si, Ge, GaN or ZnO doped with a few percent of a transition or rare earth metal. While early work has been performed mainly by polish groups in the 1970ies and 80ies, DMS have attracted worldwide scientific attention during the last 7 years due to their application potential in spintronics. This was triggered by the discovery of ferromagnetic GaMnAs and the theoretical prediction of room temperature ferromagnetism for ZnO:Mn and GaN:Mn by T. Dietl and H. Ohno [1]. One of the main obstacles while creating a DMS is secondary phase formation. Since solubility limits are rather low, non-equilibrium doping techniques like low temperature film growth are commonly used. On the other hand, ion implantation offers superb possibilities for low temperature doping but is always connected with lattice damage of the target material. Combining ion implantation with another non-equilibrium technique, i.e. rapid thermal annealing, leads to a diluted state while the crystallinity of the target material is restored. This has been shown for Si:Mn [2].
Recently, we investigated the secondary phase formation for Fe,Co and Ni as well as Gd, Tb implanted in ZnO single crystals. We found, that at elevated temperatures tiny superparamagnetic nanoparticles are formed in all transition metal (TM) doped samples. These phases can hardly be identified using lab X-ray diffraction (XRD), e.g. in the case of Fe. Only application of high resolution methods like synchrotron XRD, susceptometry, Mössbauer spectroscopy and transmission electron microscopy allows their identification [3]. On the other hand, low temperature implantation leads to ferromagnetic properties not originating from conventional superparamagnetism. The applicability of both the granular as well as the diluted magnetic materials will be discussed.

Reference
[1] T. Dietl, et al., Science 287, 1019 (2000).
[2] M. Bolduc, et al., Phys. Rev. B 71, 033302 (2005).
[3] K. Potzger, et al., Appl. Phys. Lett. 88, 052508 (2006).