|
M. Mulazzi (1,2), A. Barbier (3), R. Belkhou (4), M. Hochstrasser (5), I. Vobornik (2), J. Fujii (2), G. Panaccione (2), G. Rossi (1,2)
(1) Physics Department of the University of Modena and Reggio Emilia, Modena (Italy) (2) APE beamline, TASC laboratory (INFM), Trieste (Italy) (3) CEA/Saclay, Paris (France) (4) LURE, Orsay, Paris (France) & Synchrotron Soleil, Gif-sur-Yvette (France) (5) Lab. für Festkörperphysik, ETH Zuerich, Zuerich (Switzerland) We have investigated the NiO(111) surface by means of angle-resolved photoemission spectroscopy (ARPES) with polarised synchrotron radiation. The sample was a single crystal with very high crystalline quality (mosaic spread <0.01°). The (111) surface plane orientation was previously checked by X-Ray Diffraction and then by LEED in the photoemission experimental chamber. NiO is a strong insulator, with a gap of 4.3eV at 0 K. It is therefore very difficult to measure photoemission due to charging. For this reason, all the measurements were collected at high temperature, in the conducting and paramagnetic phase, since in the used preparation conditions thermal decomposition could be avoided. Data were collected at the APE beamline of the TASC-INFM laboratory on the ELETTRA storage ring. Normal emission energy distribution curves were measured for different photon energies (from 22eV to 58eV) obtaining the (111) dispersion law. Off-normal spectra measured with different excitation energies allowed us to trace the electron state dispersion along the [110] and [100] directions. The use of linear horizontal and vertical polarisations allowed probing states of different symmetries because of the selection rules determining the photoemission transition probabilities. A first result is that in the conductive phase there are no states at the Fermi level in agreement with previous measurements, but in contrast to some theoretical predictions [1]. The top of the valence band approaches the Fermi level when heating from room temperature, leaving a residual gap of 0.5eV in the conducting phase. Our photoemission spectra look very different from data present in literature [1,2]. One of the possibilities for these discrepancies is the sample preparation. Our method has been thoroughly checked for surface stoichiometry conservation. Previous investigations were performed on cleaved single crystals surfaces that are not stoichiometric and presenting many defects. Previously published spectra may thus be understood if metallic Ni is accumulated at surfaces defects. We stress on the fact that we have been able to measure ARPES spectra from a single crystal sample with a very low density of defects and insulating in the antiferromagnetic phase. To our knowledge this is the first time an experiment has been carried out on a sample fulfilling all the three above mentioned conditions. [1] Z.-X. Shen et al., Phys. Rev. B. 44, 3604 (1991) [2] Z.-X. Shen et al., Phys. Rev. Lett. 64, 2442 (1990 |