Theoretical studies of growth reactions on diamond surfaces

Diamond is of great interest because of its unique properties.1 It is the hardest substance found in nature due to its very high atom number density and strong covalent bonding. It also has the highest known thermal conductivity and the highest elastic modulus of any material. In addition, it is an insulator (E_g = 5.5 eV) and is transparent over a large range of wavelengths. Thermodynamically, diamond is slightly less stable than graphite and the two materials have very different structures and properties. These and other properties of diamond make it useful in a variety of applications and potential applications such as tool coatings, thermal management, optical windows and optoelectronics. Nanocrystalline diamond has potential for new applications such as seal coatings, electrochemical electrodes, microelectromechanical systems, biomolecular materials, and electron-emitting surfaces for flat-panel displays. As a result of these many uses, synthesis of diamond has been of considerable interest for many years. The initial synthesis work focused on methods based on high pressures and such methods are now widely used for production of commercial diamonds.2 There has also been an intensive effort aimed at low-pressure chemical vapor deposition (CVD) of crystalline diamond. 3 Efforts to synthesize crystalline diamond by CVD were initially limited by low growth rates, but in the last 20 years there has been much prog-ress in achieving higher growth rates. The advantages of CVD methods are low cost and flexibility with respect to the form of diamond that is deposited leading to many potential uses. Several types of CVD methods have been used for growing diamond films including thermal methods such as hot filament and thermal plasmas and nonthermal methods such as microwave plasmas. CVD methods have been the subject of several reviews.3,4 The most successful CVD growth methods have been from hydrocarbon-hydrogen gas mixtures. In most cases growth is initiated by the dissociation of a gaseous mixture of H₂ and a simple hydrocarbon precursor such as CH ₄. There are a variety of species that can be present in the gas dependent on the conditions and method used to activate the precursor gas. These include radicals, such as CH₃, CH₂, C₂H, and CH, acetylene, methane, and atomic and molecular hydrogen. In conventional methods atomic hydrogen present in the plasma is likely to saturate the diamond surfaces as well as etch non-diamond phases and abstract hydrogen from some surface sites. The latter process will provide surface radical sites that can react with the hydrocarbon species. The growth mechanisms are not well understood, but there has been considerable speculation about growth sites and vapour species involved. In nonthermal methods methyl radicals are believed to be the growth species, while in thermal methods acetylene is believed to be the major species involved. Recently, it has been discovered by Gruen et al.5,6 that diamond can be grown in low hydrogen argon plasmas with C2 as the growth species. The morphology of the diamond films obtained in the CVD processes depends on the C/H ratio in the plasmas. For example, in low hydrogen plasma growth it has been found that the crystallite sizes of diamond are very small (3-10 nm) due to a high renucleation rate.5 Understanding the growth mechanism is difficult because of the large number of experimental parameters and the difficulty of making accurate measurements under the CVD conditions. Theoretical studies have played an important role in gaining an understanding of the possible growth mechanisms of diamond in CVD. This work started in the late 1980's7 and has largely been based on quantum chemical studies of reactions of growth species on diamond surfaces with a focus on methyl radical or acetylene as the growth species. Generally reaction mechanisms are postulated to be initiated by the creation of a surface radical via the abstraction of hydrogen, followed by addition of the growth species to the surface radical site. In addition, some molecular dynamics and kinetics studies have been reported. An excellent review of the early theoretical work on mechanisms for CVD diamond growth has been presented by Spear and Frenklach. In this chapter we present a review of the theoretical studies of growth reactions on diamond surfaces. The focus of this review is largely on quantum chemical studies of reaction mechanisms of the major growth species methyl radical, methylene, and ethylene. In addition, theoretical studies of mechanisms involving C₂ are reviewed. This review is divided as follows. Section 2 contains a survey of the theoretical methodologies used for diamond surface reaction including two new promising approaches for dealing with surface reactions: density functional-based tight binding and SIMOMM. Section 3 reviews theoretical studies of the surface structure of diamond. Section 4 reviews theoretical studies of reactions involving the hydrogen abstraction and important growth species. In this chapter we also present some of our recent work on growth mechanisms involving C2 and methylene. Finally, Section 5 reviews other adsorbates on diamond surfaces that play a role in growth mechanisms.