Surfaces of Compound Semiconductors

Semiconductor surfaces, in particular, the (001) planes of classical III-V and II-VI compounds, as well as the (0001) planes of group III nitrides, have been investigated for many years now. However, knowledge of surface formation and structure in the case of group III nitrides and other compound semiconductors is 

Sutface and Interface Science Properties of Composite Surfaces Alloys, Compounds, Semiconductors, 

Still insufficient compared to the classical III-V semiconductors. This applies even more to semipolar and nonpolar surfaces of these materials, which have started to gain more attention recently. One of the reasons is that nitride layers grown on nonpolar and semipolar surfaces are less influenced by the quantum-confined Stark effect, which hmits the performance of optoelectronic devices. 

In the following, we will discuss the principles of surface formation of compound semiconductors and treat some examples of differently oriented surfaces in detail. We will see that the following properties significantly determine the possible atomic structure formation  

The driving force behind all these mechanisms is the minimization of the surface total energy (TE). In order to deepen the knowledge, we will discuss certain examples of compound surfaces with regard to their atomic structure as well as related electronic properties. At the end, we will give a short outlook on the modification of the surface geometries by adsorbates and molecules and discuss the interplay between surface formation and epitaxial growth conditions. 

---

Ohta, M., et al. (1995), Atomically resolved image of cleaved surfaces of compound semiconductors observed with an ultrahigh vacuum atomic force microscope,/. Vac. Sci. Technol., 13(3), 1265-1267. 

We have demonstrated here that clean and atomically ordered surfaces of compound semiconductors can be prepared in electrolyte solutions if the surface is properly... 

The non-polar surfaces of compound semiconductors display a remarkable structural stability independent of how they are prepared and there is substantial experimental evidence that these surfaces are ideally terminated. Alternatively, the polar surfaces demonstrate a strong structural instability and experimental evidence suggests that they are in no case ideally terminated. These structural properties of both classes of surfaces can be traced to bonding ionicity. 

The bulk-terminated polar surfaces of compound semiconductor surfaces are typically unstable due to the partial electron occupancies of their dangling bonds... 

The (110) surfaces of compound semiconductors are very well understood systems. Owing to the relatively simple relaxation of the surface, in contrast to... 

Other preparation methods, such as cleaving or ion bombardment (sputtering) and annealing, do not allow one to prepare (001) surfaces of compound semiconductor surfaces (cleaving) or to vary the surface stoichiometry during deposition (sputtering and annealing). 

Most surfaces of compound semiconductors are polar, that is, the number of anions and cations per surface unit cell is not balanced. While for the zinc blende materials there is only one nonpolar exception, the (110) face, for the wurtzite structures, there are two nonpolar surfaces, the m-plane (1100) and a-plane (1120) [98]. In wurtzite materials, a (110) surface does not exist because of the different crystal structure. 

Very similar concepts of surface stability criteria, as those discussed, have been described for surfaces of compound semiconductors. In this respect, the concept of autocompensation, or electron counting, has been introduced. This refers to lowering the surface energy by pairing of dangling bonds leading to completely filled anion states, and completely empty cation states, respectively. The reader is referred to Refs. [22-25]. 

In the first chapter, on electrochemical atomic layer epitaxy, Stickney provides a review of experimental methodology and current accomplishments in the electrodeposition of compound semiconductors. The experimental procedures and detailed fundamental background associated with layer-by-layer assembly are summarized for various compounds. The surface chemistry associated with the electrochemical reactions that are used to form the layers is discussed, along with challenges and issues associated with device formation by this method. 

Fig. 3-10. Fonnation of standard gaseous ions A(sid, and fiom surface atoms of compound semiconductor AB.

Progress on understanding the surface chemistry relevant to the formation of compound semiconductors is being made. One major issue is the genesis of defects that appear in deposits formed with the flow deposi-fion sysfem. Probable defecf sources include fhe subsfrafe qualify, lattice mismatch problem, and problems associated with deposition of a compound... 

The problem with all three of the above scenarios is that they require an understanding of the surface chemistry of compound semiconductor in aqueous solutions. Much more is known about the surface chemistry and reactivity of Au in aqueous solutions. A prerequisite, then, to the use of a compound semiconductor as a substrate for compound electrodeposition is to gain a better understanding of the substrate s reactivity under electro-chemically relevant conditions. Our initial studies of compound reactivity in electrochemical environments involved CdTe single crystals [391]. The electrochemistry of CdTe is reasonably well understood from electrodeposition studies (Table 1), and single crystals are commercially available. 

In the following section, we provide a brief review of the structures of the major semiconductor surfaces for which the adsorption and reaction chemistry will be covered in this chapter. This includes the (100) and (111) crystal faces of silicon and germanium. Chapter 1 of this book also provides a brief overview of the structure of the silicon surface. The surface structures of compound semiconductors, including GaAs and InP, can be quite complex and are not covered here. A number of reviews describe the structure of these surfaces much more extensively [5,6,25-29], and the reader is referred to those references for more detail. 

The results in Section 3 and 4 demonstrate that steps on the surface of layered semiconductors are recombination sites, and hence predominantly responsible for the poor cell performance of structured electrodes. We therefore proceed in examining the role of steps in layered compounds. 

The behavior noted for ZnS and (Cd ZnjS has been reported by others for group II-VI (21) and III-V (23) compound semiconductors. Because the analysis of the surface reaction zone is based on conservation of mass without regard for the mechanism of transporting reactants to the substrate, the framework should be applicable for the engineering analysis of the deposition of a broad group of compound-semiconductor electronic materials by both... 

Surface and thin film analysis has been used more extensively in the study of epitaxial growth of compound semiconductors where there is the additional requirement to monitor stoichiometry. Epitaxial deposition of GaAs is most frequently achieved by VPE or LPE, however, the area of most active research is MBE of GaAs and Ga Al, As. In MBE, a molecular beam of... 

---