During diamond synthesis under high pressure and high temperature (HPHT), the nucleation and growth of diamond occur consecutively and repeatedly. It is thus difficult to examine separately these two processes by experiments. In the present investigation, the nucleation mechanism of diamond has been studied and a method for control of diamond nucleation has been proposed.
In order first to understand the nucleation behavior of diamond, the effect of starting phase of carbon on diamond formation has been investigated. Two kinds of carbon, well-crystallized graphite and carbon black with poor crystallinity, were used for experiments. High pressure experiments were carried out in a belt-type high-pressure apparatus at 1390-1450℃, 4.3-4.9 GPa. When graphite was used as carbon source, diamond was easily formed at lower pressure and the morphology of synthesized diamond was irregular. On the contrary, when carbon black was used, higher pressure was needed to make diamond and the morphology of synthesized diamond was well-developed polyhedron. These differences in synthesized condition and crystal shape might be explained by difference in transformation path from the carbon source to diamond. Graphite was thought to transform directly to diamond, while carbon black seemed to transform first to graphite and then to diamond.
The role of graphite as nucleation site was tested in an experiment with seed diamond crystals. When graphite was used as carbon source, the seed diamonds grew with well-faceted morphology. On the other hand, when carbon black was used, the grown diamond had irregular shape and rough surfaces. This difference in morphology is believed to result from the difference in driving force for growth due to different structures of the carbon sources. Carbon black should deposit as diamond without graphitization when nucleation sites, such as seed diamond crystals, were available. The differance in diamond morphology suggests that the graphitization of carbon was necessary for diamond nucleation under the experimental condition.
The present experimental results also suggest that graphite can act as nucleation sites. The possibility of nucleation sites must depend on the dissolving rate of graphite into catalyst during experiments. At the initial stage, it is difficult to expect that graphite provides nucleation sites for diamond because graphite dissolves rapidly into catalyst. After rapid dissolution of graphite and its saturation in catalyst, remaining graphite may act as heterogeneous nucleation sites.
In chapter 6, the effect of SiC on diamond nucleation has been investigated. Three kinds of catalyst were prepared: (1) a fresh Ni-SiC powder mixture, (2) a heat-treated Ni-SiC mixture at 1300℃ for 2 h in vacuum, and (3) a heat-treated Ni-Si at 1100℃ for 3 h in vacuum. When carbon black with each catalyst was treated at 4.7 GPa, 1450℃ for 5 min, diamond was synthesized only in the specimen containing the fresh mixture of Ni-SiC catalyst. The added fresh SiC enhanced drastically diamond formation. SiC thus appears to act as direct nucleation sites of diamond under the experimental condition.
Direct nucleation sites of SiC was further confirmed in another experiment. When Ni catalyst was ground on SiC emery paper, diamond formation was much enhanced. This result also indicates that the pretreatment of catalyst with SiC can be an effective way to control the nucleation of diamond during HPHT synthesis.