Surfaces of In 2 O 3 and tin-doped In 2 O 3 (ITO) were investigated using photoelectron spectroscopy. Parts of the measurements were carried out directly after thin film preparation by magnetron sputtering without breaking vacuum. In addition samples were measured during exposure to oxidizing and reducing gases at pressures of up to 100 Pa using synchrotron radiation from the BESSY II storage ring. Reproducible changes of binding energies with temperature and atmosphere are observed, which are attributed to changes of the surface Fermi level position. We present evidence that the Fermi edge emission observed at ITO surfaces is due to metallic surface states rather than to filled conduction band states. The observed variation of the Fermi level position at the ITO surface with experimental conditions is accompanied by a large apparent variation of the core level to valence band maximum binding energy difference as a result of core-hole screening by the free carriers in the surface states. In addition segregation of Sn to the surface is driven by the surface potential gradient. At elevated temperatures the surface Sn concentration reproducibly changes with exposure to different environments and shows a correlation with the Fermi level position.

The interface formation of Nb-doped SrTiO 3 single crystals and ͑Ba, Sr͒TiO 3 thin films with Pt has been studied by using photoelectron spectroscopy with in situ sample preparation. For the single crystal sample, a Schottky barrier height for electrons of 0.5-0.6 eV is determined after deposition of Pt in vacuum environment. After annealing in 0.05 Pa oxygen pressure, a strong increase in the barrier height to Ն1.2 eV is observed. X-ray induced photovoltages of up to 0.7 eV are observed in this case and have to be taken into account for a proper determination of the barrier height. A subsequent annealing in vacuum reduces the barrier again. Hence, the barrier height can be reversibly switched between an oxidized state with a large barrier height and a reduced state with a low barrier height. Quantitative analysis of the barrier heights indicates that the changes are related to the changes of interfacial defect concentration. Due to the occurrence of a Ti 3+ related signal, the defects are identified as oxygen vacancies. The same effects are observed at interfaces between Pt and ͑Ba, Sr͒TiO 3 thin films with a smaller absolute value of the barrier height in the oxidized state of ϳ1 eV. Deposition of ͑Ba, Sr͒TiO 3 onto a metallic Pt substrate also results in a barrier height of 1.0 eV.

No abstract

Carefully prepared bulk ceramic specimens of In2O3 and Sn-doped In2O3 (ITO) were analysed with x-ray and UV photoelectron spectroscopy before and after heat treatment in vacuum and oxygen atmosphere. The results on ex situ prepared ceramic specimens were shown to be comparable to those of in situ deposited-measured thin films in terms of core levels, Fermi levels and ionization potentials. This suggests a viable path for rapid synthesis and screening of surface electronic-defect properties for other transparent conducting oxides (TCO) materials. A strong correlation exists between the surface electronic-defect structure of In2O3-based TCOs and their underlying electronic-defect structure, owing to the unique crystal-defect properties of the bixbyite structure. This leads to formation of a chemical depletion at the surface and the formation of a peroxide surface species for higher preparation temperatures. The results are discussed with respect to the use of ITO as hole injection electrode in organic light emitting devices.

No abstract

A determination of the Schottky barrier height at the interface between ferroelectric Pb(Zr,Ti)O3 thin films and Pt by photoelectron spectroscopy is presented. Stepwise Pt deposition was performed in situ onto a contamination-free Pb(Zr,Ti)O3 thin film surface. The substrate surface is reduced in the course of Pt deposition as evident from the observation of metallic Pb. The Fermi level is found at EF − EVB = 1.6 ± 0.1 eV above the valence band maximum of the as-prepared interface. Annealing of the sample in an oxygen pressure of 0.1 and 1 Pa strongly reduces the amount of metallic Pb and leads to a reduction in the Fermi level position at the interface to EF − EVB = 1.1 ± 0.1 eV. Storage in vacuum at room temperature strongly reduces the interface leading to a significantly higher Fermi level position (EF − EVB = 2.2 ± 0.1 eV). The reduction is attributed to the presence of hydrogen in the residual gas. The change in barrier height might be a severe issue for stable device operation with Pt contacts even at ambient temperatures.

The interface formation between Pb(Zr,Ti)O 3 (PZT) and RuO 2 and between PZT and In 2 O 3 :Sn (ITO), respectively was characterised using in-situ X-ray photoelectron spectroscopy (XPS). No interface reaction was observed for the interfaces studied. The Fermi level position at the interface (Schottky barrier height) is strongly different for the two electrode materials. A Fermi level position of 1.0± 0.1 eV above the valence band maximum (VBM) is observed for the contact between PZT and the high work function oxide RuO 2 . For the contact between PZT and the low work function oxide ITO a Fermi level position of 2.1± 0.2 eV above the VBM is found.

The interface formation between PbTiO 3 and SrTiO 3 has been studied by in situ photoelectron spectroscopy. A valence band offset of 1.1 ± 0.1 eV, corresponding to a conduction band offset of 1.3 ± 0.1 eV, is determined. These values are in good agreement with the band offsets estimated from measured ionization potentials of SrTiO 3 and PbTiO 3 surfaces. The observed band offsets are also in line with a ∼1.1 eV difference in barrier heights of PbTiO 3 in contact with different electrode materials as compared to barrier heights of SrTiO 3 with the same electrode materials. The results indicate that the band alignment is not strongly affected by Fermi level pinning and that the barrier heights are transitive. The limits of Fermi level variation observed from a number of thin films prepared on different substrates with different conditions are the same for both materials when these are aligned following the experimentally determined band offsets. By further comparing electrical conductivities reported for SrTiO 3 and PbTiO 3 , it is suggested that the range of Fermi level position in the bulk of these materials, which corresponds to the range of observed conductivities, is comparable to the range of Fermi level position at interfaces with different contact materials. In particular the possibly low barrier height for electron injection into SrTiO 3 is consistent with the metallic conduction of donor doped or reduced SrTiO 3 , while barrier heights 1 eV for PbTiO 3 are consistent with the high resistivity even at high doping concentrations. The variation of barrier heights at interfaces therefore provides access to the range of possible Fermi level positions in the interior of any, including insulating, materials, which is relevant for understanding defect properties.