Epitaxial growth mechanism of diamond crystal in methane-hydrogen plasma.

Quantum chemical calculations were carried out in order to elucidate the mechanism of the astonishing observation that diamond crystal can be synthesized in a low temperature CH4-H2 plasma at a pressure of a few torr. A most probable mechanism was found to proceed in two steps, i.e., first, covering the (1 1 1 ) plane of the diamond surface by methyl groups, and second, epitaxial growth, where the three neighboring methyl groups standing on the ( 1 1 1 ) plane of the diamond surface are spontaneously bound to form the diamond structure via methyl cation intermediates. Detailed atomic displacements during the above reactions are shown to obtain a realistic understanding of the mechanism. Chemists have believed that synthetic diamond can only be produced under ultrahigh pressure (70 000 Kg/cm2) a t very high temperature (2300 K) from graphite by the help of a catalyst.' However, recent experiments in many laboratories show that small crystals or thin films of diamond are formed under a few torr of pressure in a low-temperature hydrogen plasma which contains a small amount of methane., This is an astonishing observation, and the mechanism remains obscure. Since no phase transition of graphite or amorphous carbon takes places under the conditions of low-temperature plasma reaction, the diamond formation in vacuum is considered to proceed by chemical reactions, different from the phase transition of graphite under ultrahigh pressure a t very high temperature. Then, how and by what kind of chemical species does the growth of diamond surface occur in this plasma atmosphere? This paper describes a most probable mechanism of the epitaxial growth of diamond crystal, which was elucidated by the method of quantum chemical calculations. Since the plasma used for diamond synthesis contains mainly hydrogen with just a few percent of methane, only hydrogen molecules may react with radicals and ions which a re produced by the dissociation of methane. The contribution of methane molecule in this stage will be negligibly small. So, the chemical species produced in the CH4-H2 plasma are CH4+, CH3+, CH2+, CH", C', H', H2+, CHS+ and H3+ as well as CH,, CH2, C H , C, and H. These active species can react with the surface of diamond crystal. However, provided that these active chemical species simply react a t random with the surface of diamond crystal, no diamond growth will occur, and the products will be highly crosslinked polyethylene or amorphous carbon. For example, if a singlet methylene radical (CH,) inserts itself into the C-H bond of diamond crystal surface, the product is X-CH,. And if this reaction repeats, highly branched polyethylene but no diamond will be produced. Therefore, a specific reaction which is strictly (1) For Liample, Atkins, P. W. Physical Chemistry, 2nd Ed. ; Freeman; San Fransisco. 1982, p 194. (2) (a) Gordeev, S. K.; Smirnov, E. P.; Kol'tsov, S. I.; Aleskovskii, V. B. Dokl. Akad. Nauk SSSR 1982, 262, 127. (b) Matsumoto, S.; Sato, Y.; Kamo, M.; Setaka, N. Jpn. J. Appl. Phys., Part 2 1982, 21, 183. (c) Matsumoto, s.; Sato, Y.; Tsutsumi, M.; Setaka, N. J . Mater. Sci. 1982, 17, 3106. (d) Matsumoto, S.; Sato, Y.; Kamo, M.; Tanaka, J.; Setaka, N. Proc. Znt. Conf. Vac. Metall., 7th 1982, I , 386. (e) Matsui, Y . ; Matsumoto, S.; Setaka, N. J. Mater. Sci. Lett. 1983, 2, 532. ( f ) Trefilov, V. 1.; Tesner, P. A,; Sawakin, G. I.; Borcdina, L. M. Dokl. Akad. Nauk SSSR 1983,273, 1431. (g) Kijima, K.; Matsumoto, S.; Setaka, N. Proc. Int. Zon Eng. Congr. 1983, 1417. (h) Doi, A,; Fujimori, N.; Yoshioka, T.; Doi, Y. Zbid. 1983, 1137. (i) Hao, 2.; Mitura, St.; Wender, B. Ibid. 1983, 1143. ti) Mori, T.; Namba, Y. J . Appl. Phys. 1984, 55, 3276. (k) Matsumoto, 0.; Toshima, H.; Kanzaki, Y . Denki Kagaku Oyobi Kogyo Bufsuri Kagaku 1984,52, 128. (1) Varnin, V. P.; Teremetskaya, 1. G.; Fedoseev, D. V.; Deryagin, B. V. Dolk. Akad. Nauk SSSR 1984, 276, 367. (m) Teremetskaya, 1. G.; Varmin, V. P.; Fedoseev, D. V. Zzc.. Akad. Nauk SSSR, Ser. Khim. 1984, 1888. (n) Fedoseev, D. V.; Kochergina. A. A.; Bukhovets, V. L. Z h . Fiz. Khim. 1984, 58, 2365. (0) Namba, Y.; Mori, T. J . Vac. Sci. Technol. 1985, A3, 319. _ _ _ . 0002-7863/86/ 1508-5780%0l.50/0 Table 1. The Reaction 3C-H 3C--CH? A. Methylene Insertion B. Hydrogen Abstraction + Methyl Radical Addition 3C-H + 'CH2 .+ >C-CH, 3C-H + *H 3C* + H-H 3C-H + 'CH, >C* + HCH, 3C-H + 'CH, + >C* + 'CH, 3C-H + 'CH 3C* + 'CH) Then, >C* + CH, 3C-CH1 oriented and spontaneously produces the crystal growth of diamond must be responsible for the observed growth. This paper discusses our search for such a reaction process or processes.