Fe doped ZnO – a diluted magnetic semiconductor?

The fabrication of diluted magnetic semiconductors (DMS) by transition-metal (TM) doping of ZnO has attracted tremendous interest within the last 3 years. However, there are still vivid discussions if the ferromagnetic state stems from Zener-interaction between diluted TM ions or from magnetic secondary phases. In order to prove or exclude the possible formation of TM-secondary phases in ZnO we have investigated iron doping, since Fe-ZnO (n-type) DMS are theoretically predicted to exhibit ferromagnetism [1].
For these investigations Fe-ions have been implanted with an ion energy of 180 keV (projected range Rp=80 nm) at 250 K and 620 K into ZnO single crystals. Two fluences of either 0.4 or 4x1016 ions per cm2 were chosen which correspond to 0.5 and 5 at%, respectively. The samples were characterized by CEMS (conversion electron Mössbauer spectroscopy), XRD (X-ray diffraction) using synchrotron radiation, RBS (Rutherford back scattering), TEM (transmission electron microscopy) and SQUID (superconducting quantum interference device) magnetometry.
For the as implanted sample a high solubility of Fe was found. The ionic states are 2+ and 3+ but none of the ionic fractions could be clearly determined to occupy substitutional lattice sites. The sample implanted at high fluence and 620 K exhibits ferromagnetic behavior at room temperature as was observed by CEMS and SQUID. However, CEMS and XRD measurements confirm that the origin of the ferromagnetic behavior is due to Fe-nanocluster formation. These nanoclusters show the magnetic moment and hyperfine field of metallic bcc-Fe. For samples implanted at 250 K and a low fluence also a ferromagnetic behavior could be detected at room temperature. No secondary phase formation occurs suggesting the formation of a diluted magnetic semiconductor (DMS)

[1] K. Sato, H. Katayama-Yoshida, Jpn. J. Appl. Phys. 40, L334 (2001)