Paul Friedel

Transition Metal Surfaces

Transition metals represent very interesting materials in which one can fairly readily analyze trends in their properties along the transition series. Although there have been recent calculations of their bulk and surface properties in the local-density approximation, our discussion here will be mainly based on the tight-binding method. This is because the latter technique, even at its simplest level of sophistication, has been very successful in describing chemical trends associated with the filling of the d-band. To begin with we thus recall the main features of the band structure of these materials and how it is possible to derive a simple model for their cohesive properties, elastic constants and even phonon dispersion curves.

General Methods for Calculating the Electronic Structure of Surfaces

There has been much progress recently in the theoretical understanding of the electronic properties of surfaces and interfaces. Such progress is at least partly due to the development of more or less sophisticated methods of calculation. Among these, two of them dominate, differing in their basic philosophy: the local density plus pseudopotential method and the empirical tight binding approximation. The first one is a first principles theory; the second one aims at simulating defect or surface properties with the help of simple rules.

Chemisorption on Semiconductor Surfaces

There have been many studies of chemisorption on metals and semiconductors and it is impossible to treat all aspects of this problem here. For this reason we have focussed our interest on two particular cases of chemisorption on semiconductors. The first one concerns the growth of a metallic layer for which there have been recent studies at low temperatures. This is directly connected to the Schottky barrier height which will be discussed in Chap. 7. The second system which is developed in Sect. 6.2 is the As layer on the (111) face of covalent semiconductors which represents an extremely well defined problem from the structural point of view and can be considered as a precise test for the theory.

Electronic States at Covalent Semiconductor Surfaces

The (111) and (100) surfaces of covalent semiconductors are among the most extensively studied systems, particularly in the case of silicon. We do not intend to give here a complete review of the numerous experimental as well as theoretical studies of these surfaces. Instead we try to select what we believe to be the most important results. We discuss them on the grounds of chemical-like models which allow us to obtain very simple pictures of the few essentially different situations.

Surfaces of Compound Semiconductors

Compound semiconductors are characterized by the fact that there is electron transfer between the electropositive and the electronegative atom. This means that self-consistency effects are very important, even for bulk properties such as the splitting between LO and TO phonon modes. Self-consistency will become even more important in the treatment of surfaces and interfaces. For this reason the first section is devoted to an analysis of such effects. This begins with general arguments separating long-range from short-range effects and showing the physical basis of the so-called “local neutrality condition”. Then a tight-binding formulation of self-consistency effects will be presented for further application to the prediction of core-level shifts and band offsets. Finally we give the values of the cation and anion dangling bond levels determined from the zero-charge condition which will help us in the understanding of surface and interface properties.