The bias-enhanced nucleation(BEN) of diamond is characterized by its exceptionally high nucleation density. In an engineering point of view, a high nucleation density is important because of higher possibility of diamond film formation with a smooth surface. Recently, a heteroepitaxy between diamond and various substrate materials was reported to be possible through bias-enhanced nucleation. Several models were proposed to explain the enhancement of nucleation. However, the BEN mechanism of diamond is not elucidated, yet. In this study, a model of nucleation enhancement by BEN process was suggested. Based on the proposed model, various observed phenomena could be explained.
Firstly, the nucleation condition which yields the highest density during BEN process was determined. Nucleation density up to $2×10^9/㎠$ was obtained. The determined optimum condition was the filament temperature of 2300℃ the substrate temperature of 750℃, the bias voltage of 300V, the methane concentration of 20%, and the treatment time of 2 hours. The diamond nucleus was hemi-spherical and analysed to be a microcrystalline diamond. A carbon phase which showed a Raman spectrum of soot was co-deposited. Its shape was similar to a silkworm and its amount was increased with temperature. But it did not affect the nucleation density of diamond and was thought to be a amorphous phase.
The nucleation enhancement by arcing was observed. This new phenomenon occured by a positive bias applied on a W wire located between a filament and a Si substrate. Diamond particles were deposited or scattered prior to a arcing treatment and the nucleation was enhanced around the pre-existing diamonds. Sometimes, a very strange phenomenon was also observed. A pre-existing diamond particle was disappeared and small pieces of diamond were observed. And the center part of nucleation enhancement was seperated from the other part. This phenomenon was attributed to the fragmentation of diamond due to arcing. A diamond fragment is believed to act as a nucleation site of diamond and grow to a stable nucleus during the treatment period.
In terms of thermodynamics, it was dilemma that diamond particles were nucleated on Si substrate and diamond particles on a substrate holder were concurrently etched. This controversy can be solved by introducing an assumption that two parallel processes were engaged during the BEN process. One is the fragmentation of diamond and the other is the diamond growth from supersaturated carbon. The fragmentation is thought to be a result of electron emission of diamond but the details of the fragmentation process are not known, yet. This model is based on the experimental observation that diamond was etched away very quickly. The etching rate of 6㎛/h can not be explained by the conventional chemical etching. The DC glow discharge seems to play two roles during the BEN process: promotion of the electron emission of diamond and transport of the diamond fragment throughout the plasma.
