Methyl versus acetylene as diamond growth species
- PDF / 642,024 Bytes
- 7 Pages / 593.28 x 841.68 pts Page_size
- 82 Downloads / 185 Views
L. Robbin Martin Aerophysics Laboratory, The Aerospace Corporation, P.O. Box 92957, Los Angeles, California 90009 (Received 22 February 1990; accepted 25 July 1990)
We have modeled plasma-assisted diamond growth on substrates placed in a high velocity 1-dimensional flow. The gas consisted of methane or acetylene injected into a flow of partially dissociated hydrogen gas at 800 °C. Diamond is formed only near the injector. More diamond is formed when methane is the additive, and Raman spectra show that the quality of the diamond films is also higher when methane is the additive. The model, which includes detailed chemistry, convection, concentration diffusion, and thermal diffusion, shows that with this experimental arrangement only methane and methyl radicals are present in significant quantities when methane is added, while only acetylene is present when acetylene is added. We conclude that (1) Diamond films can be grown directly from methyl radicals (or, possibly, from methane) and from acetylene. This suggests that a variety of hydrocarbons could act as growth species. (2) An environment containing methane and methyl is much more effective for growing diamond films than one containing acetylene. (3) The quality of the diamond film depends on the identity of the growth species, with acetylene producing lower quality films than methyl (or methane). (4) The fall-off in diamond formation with distance from the injector is due to destruction of species crucial to diamond growth on the silicon substrates.
I. INTRODUCTION
Ever since diamond formation at low pressure was discovered,1'2 researchers have speculated on the nature of the growth species—i.e., which gas phase species contributed carbon to the mass growth of diamond. Originally, such speculation was little more than a cataloging of species guessed to be present in the gas phase. However, because the hydrocarbon was exposed to a discharge or hot filament, which leads to a very complex mixture of species, and because there were no data on the concentrations of the species actually present at the substrate surface, little progress was made in correlating growth rates with the presence of particular species. Beginning in 19863'4 a more detailed search for growth species began with semiempirical quantum calculations describing a mechanism for adding CH 3 (methyl) radicals to an assumed positively charged diamond surface. Later, similar calculations5 were performed for the case of C2H2 (acetylene) adding to an electrically neutral surface. Harris proposed6 a detailed chemical kinetics mechanism for diamond growth from methyl radical addition to a neutral (100) surface and calculated a growth rate in reasonable agreement with experiment using an approach that contained no free parameters or phenomenological constants. Quantitative species measurements were first made by Kawato and Kondo,7 who detected C 2 H 2 and CH4 in J. Mater. Res., Vol. 5, No. 11, Nov 1990
the exhaust from their growth chamber and observed their dependence on starting conditions. They suggest
Data Loading...
