Electronic Structure of Oxide Surfaces

As stated in the introduction, oxides may cover a wide range of electronic properties, in particular, from insulating ionic to superconducting materials. Consequently, the electronic structure of oxides covers wide band gap insulators, semiconductors, and metals. Examples are MgO or AI2O3, which show properties of insulators with band gaps of 7.5 and 8.5 eV [123], respectively Ti02 or TiOj with semiconducting 

Through the formation of surfaces, bonds are broken and the above-addressed relaxations, rumpling, or reconstruction phenomena affect the interatomic potentials and bonding characteristics. The surface electronic structure is modified with respect to the bulk. In the following, we will address a few topics connected with surface electronic structure of some selected oxides. 

---

Articles dealing with the structure and chemistry of solid and crystal surfaces include Tabor (1981) and Forty (1983), who discusses metals and catalysts in particular. The surface of diamond is discussed by Pate (1986), metal oxides by Henrich (1985), transition-metal compounds by Langell and Bernasek (1979), and transition-metal oxides by Henrich (1983). Some of these articles deal with the electronic structures of the surfaces as well as the surface atom geometry the volume edited by Rhodin and Ertl (1979) on the nature of the surface chemical bond and the review paper by Tsukada et al. (1983) on the electronic structure of oxide surfaces concentrate on this aspect. One of the few reviews directed specifically towards minerals is that of Berry (1985). 

Tsukada, M., J. Adachi, and C. Satoko (1983). Theory of electronic structure of oxide surfaces. Prog. Surf. Sci. 14, 113-74. 

This chapter is concerned with experimental studies of the electronic structure of oxide surfaces. Obviously the complexities of surface electronic structure cannot be understood without first understanding the factors which determine bulk electronic structure the next section outlines some of the key ideas that are needed to describe the bulk electronic structure of oxides. However, it is beyond our scope to dwell at length on experimental determination of bulk electronic structure instead the main focus of the chapter is on aspects of electronic structure associated with surfaces that cannot be understood in terms of the bulk electronic structure alone. The publication of... 

Peculiarities in the electronic structure of oxide surfaces are related to the strength of the electrostatic potential which acts on the surface atoms and to the number of broken anion-cation bonds. In a qualitative way, the more open the surface, the stronger the Madelung potential reduction and the less efficient the electron delocalization. 

Supported metal oxide catalysts are a new class of catalytic materials that are excellent oxidation catalysts when redox surface sites are present. They are ideal catalysts for investigating catalytic molecular/electronic structure-activity selectivity relationships for oxidation reactions because (i) the number of catalytic active sites can be systematically controlled, which allows the determination of the number of participating catalytic active sites in the reaction, (ii) the TOP values for oxidation studies can be quantitatively determined since the number of exposed catalytic active sites can be easily determined, (iii) the oxide support can be varied to examine the effect of different types of ligand on the reaction kinetics, (iii) the molecular and electronic structures of the surface MOj, species can be spectroscopically determined under all environmental conditions for structure-activity determination and (iv) the redox surface sites can be combined with surface acid sites to examine the effect of surface Bronsted or Lewis acid sites. Such fundamental structure-activity information can provide insights and also guide the molecular engineering of advanced hydrocarbon oxidation metal oxide catalysts such as supported metal oxides, polyoxo metallates, metal oxide supported zeolites and molecular sieves, bulk mixed metal oxides and metal oxide supported clays. 

Simply from symmetry considerations, the electronic structure of any surface, where the atoms or ions are necessarily coordinatively unsaturated, should be different from that of the bulk. The magnitude and type of the differences between surface and bulk electronic structure depend on the particular oxide, of course this is addressed in Chaps. 2, 4 and 14 in this volume. However, a few general observations can be made. 

The characteristics of the charge distribution in the bulk and in the outer layers are thus key factors to understand the physics of polar surfaces. A zeroth-order description of the electronic structure of oxides is the fully... 

The chemical reactions which occur at oxide surfaces are important in a wide variety of industrial, geological and environmental contexts. Nevertheless, the study of the atomic and electronic structure of these surfaces is poorly... 

In a third part, the known adsorption properties of CO, and adsorbates after a CO oxidation TPR, both on Pt(ll 1) are investigated as a test reaction. The EES results show the capability of MIES to probe the electronic structure of a surface reaction, i.e. their adsorbates respectively. 

One of synthetic approaches for the iron nanoparticles is based on the widely used decomposition of iron pentacarbonyl [19, 361, 362], The novelty of the approach is the surfactant system used. Studies with a number of strongly bound surfactants have resulted in decreased magnetic response, due to surface oxidation, disturbing the electronic structure of the surface atoms, or some other mechanism. With this in mind, ones chose to work with a weak surfactant, a p-diketone. P-diketones do have a history as adhesion promoters in bonds between metals and polymers [363], The limited reactivity of p-diketones is as an advantage the P-diketone is much weaker oxidizer than carboxylic acids or alcohols and will not oxidize iron, it is not as nucleophilic as phosphines, yet it is known to be capable of chelating iron. 

The main numerical approaches allowing a calculation of the bulk electronic structure of oxides have been recalled in the first chapter. Treating surface effects requires some care, especially concerning the geometry of the systems... 

The electronic structure of polar surfaces does not follow the trends discussed at the beginning of the chapter for non-polar surfaces, primarily for two reasons. The first one comes from the number of broken bonds at the surface which may be large in some cases. For example, on the rocksalt lll faces, three in every six anion-cation bonds are broken. But the major reason is linked to the behaviour of the electrostatic potential discussed above, which induces large charge redistributions. In some cases, these charge redistributions may provide the compensation of the macroscopic electric field required to stabilize the surface. The electronic structure of three polar oxide surfaces has been calculated MgO lll, ZnO(OOOl) and (0001) and the weakly polar SrTiOsjOOl face. 

Many questions related to the electronic and atomic structures of oxide surfaces remain unanswered. For example, the respective roles of bond-breaking and structural distortions in the interpretation of surface densities of states and gaps are not well elucidated. Similarly, for complex crystal structures, it is difficult to relate the observations to models, since one does not know precisely the surface terminations. More systematic studies of stoichiometry in surface layers, and of the electronic and atomic features of polar surfaces are needed to better disentangle the origin of the deep surface states which may result either from dangling bonds or from structural and stoichiometric defects. Finally, the modification of correlation effects on oxide surfaces represents a completely virgin field for future investigations. 

In the first chapters of this book, the electronic structure of oxides in the bulk and at the surface in the absence of adsorbates has been analyzed, and some theoretical models were given as guidelines for their understanding. It is possible to use them now to point out how the arguments of electronegativity and partial charge developed above are related. 

As an example to study the valence electronic structure of oxides, we consider one of the least disturbed and relaxed surfaces, namely, the MgO(lOO) surface. Figure 15.25a shows angle-resolved photoemission (Volume 1, Chapter 3.2.2) data taken with Hell radiation at different polar angles 9 along the T X azimuth defined in the figure [124]. Via the simple formula... 

---