High-resolution photoemission study ofV2−yO3

Abstract

High-resolution photoemission spectroscopy ͑HRPES͒ was performed on V 2Ϫy O 3 (yϭ0.00, 0.03, and 0.04͒, which has been considered as the prototype of the Mott-Hubbard model for the metal-insulator transition ͑MIT͒. The density of states at the Fermi level (E F ) observed by HRPES is one order of magnitude smaller than that of the band calculation based on the local-density approximation. We observed no distinguishable sharp structure at E F for hole-doped V 1.96 O 3 at 15 K as well as negligible temperature-dependence of the HRPES spectrum of V 2 O 3 between 200 and 300 K. The analysis of the V 3d spectrum with a local self-energy gives a too large quasiparticle effective mass compared to the specific-heat measurement, which implies the importance of spatial fluctuations near the MIT. All these facts strongly question the applicability of the dynamical mean-field theory of the Hubbard model to the V 2Ϫy O 3 system.

Local band gap modulations in non-stoichiometricV2O3films probed by scanning tunneling spectroscopy

“…Different hole concentrations in the oxide valence band are also responsible for changes in the band width and might contribute to the observed modulations of the gap size. 8 Due to compensation effects between acceptor and donor levels and missing information on their energy position in the gap, the local electron or hole concentration in the different oxide regions cannot be deduced from the dI / dV measurements. However, some general aspects shall be discussed in the following that may account for the shift of the Fermi level out of the midgap position.…”

“…The position of the O 2p valence band was determined to range from −4 to − 10 eV, whereas the V 3d derived conduction band starts at approximately −3 eV and exceeds the Fermi level. 7,8 However, the exact electronic structure of this material is still controversial.…”

Chemical bonding and many-body effects in site-specific x-ray photoelectron spectra of corundumV2O3

“…4,5 Photoelectron spectroscopy ͑PES͒ of the paramagnetic V 2 O 3 phase has been used intensively in an attempt to understand the underlying electronic structure in general, hybridization trends in particular, and their relation to the metal-insulator transition. [5][6][7][8][9][10][11][12][13] Unfortunately, interpretation of the PES data is not straightforward. Ultraviolet photoelectron spectroscopy ͑UPS͒ is highly surface sensitive, and its results may deviate from the bulk electronic structure.…”

Prominent Quasiparticle Peak in the Photoemission Spectrum of the Metallic Phase ofV2O3

“…However, in spite of continuing efforts for over twenty years, literature [5,6,7,8,9, 10] V 3d PES spectra for the PM phase of (V 1−x Cr x ) 2 O 3 have shown at most a near E F feature that is the smallest part of the spectrum. One could hypothesize that the distinctive central peak of the halffilled one-band model is obscured and washed out by the multi-band complexity of the actual electronic structure of (V 1−x Cr x ) 2 O 3 in which the two 3d-electrons of the V 3+ ion must be distributed among singly degenerate a 1g and doubly degenerate e π g orbitals derived from a small trigonal crystal field splitting of the cubic t 2g manifold of the V 3d states.…”