Dresden 2003 – scientific programme
Parts | Days | Selection | Search | Downloads | Help
TT: Tiefe Temperaturen
TT 18: Metall-Isolator-Überg
änge in Quantensystemen I
TT 18.4: Talk
Thursday, March 27, 2003, 15:30–15:45, HSZ304
Metal insulator phase transition in VO2 — •M.W. Haverkort1, Z. Hu1, A. Tanaka2, W. Reichelt3, H.H. Hsieh4, H.-J. Lin4, C.T. Chen4, and L.H. Tjeng1 — 1II. Physikalisches Institut, Universität zu Köln — 2ADSM, Hiroshima University, Japan — 3Institut für Anorganische Chemie, Dresden — 4SRRC, Hsinchu, Taiwan
VO2 is a non-magnetic oxide that undergoes a metal-to-insulator transition at 340 Kelvin. Above this temperature, VO2 is metallic and has the rutile (TiO2) crystal structure (R-phase). At low temperatures, it is an insulator with a monoclinic structure (M1-phase), in which V-V pairs are formed.
The long-standing debate about this compound concerns the nature of the metal-to-insulator transition. The issue is whether the non-magnetic insulating state would be regarded as a Peierls-insulator with the character of a band insulator (one-electron picture), or whether it should be viewed as a Mott-insulator (many-body picture). In other words, the importance of electron-correlation effects in this lattice-coupled metal-to-insulator transition is at the heart of the discussion.
In order to probe the local electronic structure we used soft-xrays to study this metal-to-insulator transition at the L23 (V 2p to 3d) and K (O 1s to 2p) edge. We used linear polarized light with the polarization vector parallel and perpendicular to the crystal c-axes. We found that the orbital occupation changes as one goes through the phase transition. This strongly suggests that local electron-electron correlations are important for the metal-to-insulator transition in VO2. From that we infer that the metal-to-insulator transition is not a simple classical Peierls-transition.