Surface elemental semiconductors

Both the Si(100) and Ge(100) surfaces have stable (2x1) reconstructions which involve the saturation of dangling bonds by the formation of dimers between the atoms in the top layer of the bulk termination of the solid. In the case of Si(100)(2xl), although the existence of surface dimers is no longer controversial, there have been contradictory reports of dimers oriented parallel to the surface (the symmetric dimer model) or tilted in the plane perpendicular to the surface (the asymmetric dimer model), see fig. 10. 

Recent studies which employ diffraction techniques, including medium energy ion scattering (MEIS) (Tromp et al., 1983), LEED (Holland et al., 1984) and grazing incidence X-ray scattering (Jedrecy et al., 1990), favor the asymmetric dimer model in which the top layer dimer is tilted by between 13.3 and 7.6 degrees. However, a kinematic LEED study (Zhao et al., 1991), photoemission studies (Johansson et al., 1990 Uhrberg and Hansson, 1991) 

The Ge(100)p(2xl) surface has been the subject of fewer studies that the analogous Si surface. In particular, the symmetric vs asymmetric dimer question remains unresolved for this surface. Photoemission (Kevan, 1985), He-atom scattering (Lambert et al., 1987), STM (Kubby et al., 1987) and an earlier XRD study (Eisenberger and Marra, 1981) indicate that the asymmetric is the majority species at the surface. However a recent, more detailed, X-ray diffraction study favors symmetric, or almost symmetric, dimer formation (Grey et al., 1988).The structural parameters for two relatively complete XRD determinations of the Ge(100)(2xl), reconstructed surfaces are given in table 15. 

Structural parameters for Si(100) and Ge 100) (2x1) and c(4x2) surfaces. zyi, in and z34 denote the distances between the atoms numbered as in fig. 10, measured perpendicular to the surface, dn is the Si-Si or Ge-Gc bond length of the top layer dimer. The tilt angle is measured relative to the surface plane and is derived from the coordinates. 

Reference Tromp Jedrecy Holland Zhao Zhao Eisenberger Grey 

---

We discuss next the surface structure of a compound semiconductor, GaAs, to illustrate the common features and the differences from elemental semiconductor surfaces. We examine the (110) surface of GaAs, which contains equal numbers of Ga and As atoms other surface planes in this crystal, such as the (001) or (111) planes, contain only one of the two species of atoms present in the crystal. The... 

This is a huge issue and one that will be covered at length throughout this and subsequent volumes. Very often, elemental semiconductors exhibit a much richer variety of reconstructions than metal surfaces do [115, 116], fortunately, much of the qualitative insight into semiconductor surfaces has been condensed into a series of general principles, first laid down by Duke [117, 118] and discussed at length in several other books [116, 119]. A discussion of the relevant principles along with a few examples is sufficient to provide a flavor of how some of the most common elemental semiconductor surfaces behave. 

Duke arrived at five principles. Only the four of relevance to elemental semiconductor surfaces are discussed here. The fifth states that surfaces tend to be auto-compensated , which is simply a principle prohibiting the accumulation of charge at the surface. Duke s principles were first... 

The degree of surface cleanliness or even ordering can be determined by REELS, especially from the intense VEELS signals. The relative intensity of the surface and bulk plasmon peaks is often more sensitive to surface contamination than AES, especially for elements like Al, which have intense plasmon peaks. Semiconductor surfaces often have surface states due to dangling bonds that are unique to each crystal orientation, which have been used in the case of Si and GaAs to follow in situ the formation of metal contacts and to resolve such issues as Fermi-level pinning and its role in Schottky barrier heights. 

A SSIMS spectrum, like any other mass spectrum, consists of a series of peaks of dif ferent intensity (i. e. ion current) occurring at certain mass numbers. The masses can be allocated on the basis of atomic or molecular mass-to-charge ratio. Many of the more prominent secondary ions from metal and semiconductor surfaces are singly charged atomic ions, which makes allocation of mass numbers slightly easier. Masses can be identified as arising either from the substrate material itself from deliberately introduced molecular or other species on the surface, or from contaminations and impurities on the surface. Complications in allocation often arise from isotopic effects. Although some elements have only one principal isotope, for many others the natural isotopic abundance can make identification difficult. 

We discuss the dissolution of surface atoms from elemental semiconductor electrodes, which are covalent, such as silicon and germanium in aqueous solution. Generally, in covalent semiconductors, the bonding orbitals constitute the valence band and the antibonbing orbitals constitute the conduction band. The accumulation of holes in the valence band or the accumulation of electrons in the conduction band at the electrode interface, hence, partially breaks the covalent bonding of the surface atom, S, (subscript s denotes the surface site). 

A shift of the flat band potential due to photoexcitation of the type shown in Fig. 10-18 results from the capture of holes in the surface state level, e , on the electrode as shown in Fig. 10-19. We now consider a dissolution reaction involving the anodic transfer of ions of a simple elemental semiconductor electrode according to Eqns. 10-24 and 10-25 ... 

Because of the potential importance for industrial-scale catalysis, we decided to check (i) whether an influence of a semiconductor support on a metal catalyst was present also if the metal is not spread as a thin layer on the semiconductor surface but rather exists in form of small particles mixed intimately with a powder of the semiconductor, and (ii) whether a doping effect was present even then. To this end the nitrates of nickel, zinc (zinc oxide is a well-characterized n-type semiconductor) and of the doping element gallium (for increased n-type doping) or lithium (for decreased n-type character) were dissolved in water, mixed, heated to dryness, and decomposed at 250°-300°C. The oxide mixtures were then pelleted and sintered 4 hr at 800° in order to establish the disorder equilibrium of the doped zinc oxide. The ratio Ni/ZnO was 1 8 and the eventual doping amounted to 0.2 at % (75). 

Reduction to the zero-valent state or formation of other metal solid phases like oxides, causes the element to deposit onto the semiconductor surface. The efficiency of the photocatalytic reaction depends on different factors. One of the most critical aspects is the high probability of electron-hole recombination, which competes with the separation of the photogenerated charges. On the other hand, as there is no physical separation between the anodic reaction site (oxidation by holes) and the cathodic one (reduction by electrons), back reactions can be of importance. The low efficiency is one of the most severe limitations of heterogeneous photo catalysis. 

High Resolution Electron Energy Loss Spectroscopy has been "discovered" at about the same time as the previous cited techniques - die first reported experiment is related to a study of small molecules adsorbed on a (100)W surface and is dated from 1967 (1). During the last 15 years, the characterization of adsorption states of molecules on metal and semiconductor surfaces was the principal attribute of HREELS information on the elemental composition, on the chemistry, and the kinetics of surface reactions (versus temperature and/or time) were studied. One significant "plus" of HREELS is its ability to identify adsorption sites on a metal, by using the "dipole-selection rule" it is therefore possible to gain information on the short-scale structure or morphology of a surface with HREELS. 

Given the very high level of technological infrastructure that already exists for these elemental semiconductors because of microelectronics applications, it is not surprising that both these materials were examined early on in the evolution of the field of photoelectrolysis of water. As mentioned in an introductory paragraph, cathodic reduction of the Ge surface is accompanied by H2 evolution.58,559 However, we are not aware of studies under irradiation of Ge electrodes from a HER or OER perspective. The extreme instability of this semiconductor in aqueous media coupled with its low band gap (Eg = 0.66 eV) make it rather unattractive for water photosplitting applications. 

The last decade has seen an explosion of activity in the field as electrochemists have wrestled with unfamiliar, and often intractable, problems generated by the very wide range of materials investigated, difficulties often compounded by the use of polycrystalline samples whose bulk and surface properties have proved resistant to control. In addition to the elemental semiconductors and the III/V materials, a huge range of n- and p-type oxides, sulphides, selenides, and tellurides have been described and surface and bulk modifications carried out in the hope of enhancing photoelectrochemical efficiency. New theoretical and experimental tools have developed apace and our fundamental understanding of the semiconductor-electrolyte interface has deepened substantially. 

In fact, this lull probably anticipates a new spurt of activity, perhaps directed at surface properties of inter metallic compound semiconductors, which present a far more difficult field of study than elemental semiconductors. 

Silicon and germanium are the most important elemental semiconductors. They have the diamond cubic structure with sp hybrid bonds. The structure of the low index crystallographic planes, the only ones to be considered here, is shown in Fig. 1. It is seen that in the ill surfaces the atoms are triply bonded to the layer below and thus have one unpaired electron (dangling bond). Each atom of the 110 surfaces also... 

An attempt was made in this paper to sketch the behavior of elemental semiconductors (with the diamond-type structure) and of the IH-V compounds (with the zinc blende strut ture) in aqueous solutions. These covalent materials, in contrast to metals, exhibit properties which sharply reflect their crystalline structure. Although they have already contributed heavily to the understanding of surfaces in general, semiconductors with their extremely high purity, crystalline perfection, and well-defined surfaces are the most promising of materials for surface studies in liquid and in gaseous ambients. 

Unraveling the relationship between the atomic surface structure and other physical and chemical properties is probably one of the most important achievements of surface science. Because of the mixed ionic and covalent bonding in metal oxide systems, the surface structure has an even stronger influence on local surface chemistry as compared to metals or elemental semiconductors [1]. A vast amount of work has been performed on Ti02 over the years, and this is certainly the best-understood surface of all the metal oxide systems. 

It is comforting that the elemental rules for predicting surface terminations outlined in section 3.1.1. work so well for predicting the structure of the (1x1) terraces and the step edges of all the orientations of both rutile and anatase. The extensive theoretical work has helped to refine the understanding of surface relaxations, and the level of detail on the atomic geometry of the TiO2(110) surface is certainly comparable to that of certain elemental semiconductors or metals. 

---