A systematic study on the photocatalytic activity of well-defined, macroscopic bulk single-crystal TiO(2) anatase and rutile samples has been carried out, which allows us to link photoreactions at surfaces of well-defined oxide semiconductors to an important bulk property with regard to photochemistry, the life time of e-h pairs generated in the bulk of the oxides by photon absorption. The anatase (101) surface shows a substantially higher activity, by an order of magnitude, for CO photo-oxidation to CO(2) than the rutile (110) surface. This surprisingly large difference in activity tracks the bulk e-h pair lifetime difference for the two TiO(2) modifications as determined by contactless transient photoconductance measurements on the corresponding bulk materials.

The adsorption of NO on single crystalline rutile TiO(2)(110) surfaces at 100 K and the subsequent formation of N(2)O via NO dimer intermediates was studied by reflection-absorption infrared spectroscopy using a UHV-FTIR system. Analysis of the IR data reveals that the occurrence of s-polarized adsorbate vibrational bands always increases the reflectivity, giving negative bands, whereas p-polarized bands are either positive or negative, depending on the reduction state of the substrate. This sign reversal of optical bands is caused by vacancy doping, which also affects the optical constants governing the complex interplay between reflection and transmission of p-polarized light on a dielectric substrate.

234 3214182 y These authors contributed equally to this work.Metal oxides and metal nanoparticles dispersed on oxide substrates have gained increasing interest in surface science because of their widespread applications, especially in heterogeneous catalysis. In this review we summarize our recent vibrational spectroscopic studies on pure and metal-covered oxide surfaces (ZnO, TiO 2 , Au/ZnO and Cu/ZnO) using a number of small molecules (H 2 , CO, CO 2 , NO and HCOOH) as probes and/or reactants. High-resolution electron energy loss spectroscopy (HREELS) turned out to be a powerful tool to investigate well-defined oxide and metal/oxide model systems. The application of a novel ultrahigh vacuum IR spectroscopy (UHV-FTIRS) apparatus allowed us to record high-quality IR data on oxide surfaces of both single crystals and polycrystalline powder particles. We will particularly focus on following important issues: (i) the interaction of hydrogen with ZnO; (ii) structure and reactivity of polar and nonpolar ZnO surfaces; (iii) the role of defects in surface chemistry of oxides; (iv) the origin of significant difference in photocatalytic activity between anatase and rutile TiO 2 and (v) the interaction between metal nanoparticles and oxide supports. We will demonstrate that the data from HREELS and UHV-FTIRS provide detailed insight into structural, electronic and chemical properties of the studied systems.

Water bilayers deposited on Ru͑0001͒ have become a benchmark system for water interacting with solid surfaces. Here, we report diffraction data recorded using He-atom scattering demonstrating the presence of an only weakly disturbed bilayer of D 2 O on Ru. Results obtained by this method cannot be affected by e-beam damage which so far made the experimental determination of the structure of this important model difficult. We find a high-order commensurate, stable water bilayer exhibiting a structure which is not compatible with any of the previously published structural models for this system. The cross section for e-induced damage has also been determined.Understanding the wetting of surfaces by water is a topic of pronounced importance in connection with many interface phenomena, including tribology, corrosion, and catalysis. On oxide surfaces, as a rule, water dissociates, 1 but there are exceptions. 2,3 On metals, the situation is reversed; here, the fragmentation of water is rather the exception. 1 This general assumption has been questioned recently when older data from low-energy electron diffraction ͑LEED͒ were reinterpreted so as to imply that water molecules adsorbed on Ru surfaces are partially dissociated. 4-7 Soon the dissociation of water on Ru͑0001͒ became the subject of a rather controversial debate. [7][8][9][10][11][12][13][14] In the course of the discussion, it emerged that for the interpretation of the original LEED data, 15 e-beam damage has to be considered. In later experiments, the corresponding cross section for electrons impinging on intact water adlayers on this surface has been found to be on the order of 10 −16 cm 2 . 16 This fairly large cross section implies that the ͑ ͱ 3 ϫ ͱ 3͒R30°pattern observed in the early experimental work 15 has to be related to a partially dissociated adlayer containing O and OD species in addition to intact D 2 O molecules.In the course of the debate, it also became clear that there are fundamental problems in connection with using densityfunctional theory ͑DFT͒ to study H 2 O adsorption on Ru͑0001͒. The interaction of water with a metal surface is governed by a rather complex interplay between different types of forces, namely, chemical bonding, van der Waals forces ͓which can presently not be well described by DFT ͑Refs. 17-20͔͒, and hydrogen bonding. 6 The present difficulties of DFT as regard to this system are illustrated best by the fact that calculations using presently available DFT functionals predict that water does not wet the Ru͑0001͒ surface ͑see Ref. 6 and discussion in Ref. 21͒-in pronounced contrast to the experimental findings, see below.Recent careful studies using thermal desorption spectroscopy ͑TDS͒ and IR spectroscopy-where e-induced effects can be rigorously excluded-have allowed to demonstrate the presence of intact water molecules wetting the surface 22 and to precisely determine their binding energy. 16,22 The geometrical structure of the water adlayer on this Ru surface is, however, still unknown. This is a disturbing situation,...

No abstract