Developments in CVD-Diamond Synthesis During the Past Decade
- PDF / 1,387,196 Bytes
- 6 Pages / 576 x 777.6 pts Page_size
- 23 Downloads / 189 Views
it into less desirable graphitic carbons. The possibility of growing diamond with controlled properties and defect concentrations in unlimited sizes has led to recent enthusiasm in the scientific and technical communities. Reports of diamond growth from vapor phase began as early as the mid-1950s, with a continued low level of research in the 1960s and early 1970s because of extremely low growth rates and skepticism in the scientific community.7"9 Key papers demonstrating CVD of diamond appeared in the late 1970s, but it was not until the mid-1980s that the progress had proceeded to the point at which significant scientific and industrial interest was generated. By the mid-1980s, it was accepted that small crystals (ca. microns in size) and continuous polycrystalline thin films (ca. lOs-of-microns thick) could be grown by CVD. Typical CVD deposition of diamond is from a flowing gas mixture comprised of a small amount of methane (typically less than 1 at.%) in hydrogen at between 10 and 50 Torr, activated with a hot filament (ca. 2200°C) or plasma near a substrate heated to between 700 and 1000°C.10 A schematic of a generic diamond-CVD process appears in Figure 1." In the late 1980s, the first studies of the process' chemistry began to emerge using optical emission, in situ laser diagnostics, mass spectrometry, and other diagnostic techniques.12 These techniques demonstrated the presence of methyl radicals, acetylene, CH, C2, and atomic hydrogen near the growth surface, and led to computational studies of the gaseous and surface chemistries.13 Intense discussion raged over the predominant
growth species: methyl radical versus acetylene. Subsequent experiments have demonstrated that, under certain conditions, most of the deposited diamond results from methyl (or C i species) and that acetylene (or C 2 species) can also be important to growth under other conditions. No single surface mechanism or growth model has been sufficient to describe all the observations, and the community has come to understand that many reactions (models of growth) are often occurring simultaneously because of the complex chemical nature of the growth surface. Generic growth models that ignore the specific growth-site stereochemistry have been extremely successful in describing, both qualitatively and quantitatively, a wide variety of growth environments.1314 In these models, atomic hydrogen is the key chemical reactant driving both the gaseous chemistries and the surface chemistry, and influencing the deposition rate and material quality. Oxygen and halogen species have also proved to be important in certain deposition chemistries, with the growth in atmospheric combustion flames receiving great attention in 1988. As experience with many different activation schemes (plasmas, filaments, flames), reactor designs, and chemical recipes developed in the early 1990s, two important empirical patterns developed as follows: the H-C-O "phase diagram" '5 and the "alpha parameter"' 6 describing the morphology of individual crystals or the polycrystallin
Data Loading...
