Optimizing BEN conditions

The influence of the process parameters on BEN was studied in Ref. [331] using an ASTeX reactor. The substrate used was Si(lOO) of 1-inch in diameter, which was placed on a Mo substrate holder. No carburization was done before the BEN process. The microwave power was 900 W, and the substrate temperature was 725 °C. Laser reflection interferometry (LRI) [330, 331] was used to monitor the diamond nucleation and the subsequent growth rate. The biasing time was determined from the inflection point of the intensity of the LRI. At this point, 

It seems, however, that the chamber volume was so large compared with the gas flow rate that the gas in the chamber was not completely refreshed by the cyclic change in the source gas. Hence, the effect mainly arose from the change in the average CH4 concentration rather than a complete refreshing of the source gas in the chamber. 

AC voltage was applied to the substrate. The bias voltage was actually appUed between the stainless steel chamber and the graphite susceptor, and the effects of AC and DC bias voltages on film morphology were compared. The process conditions are shown in Table 11.6. 

The nucleation process was monitored by LRI (see Section 11.10), and the application of bias voltage was terminated if diamond nucleation was detected by LRI. No appreciable diamond nucleation was observed in the AC-BEN treatment, if the voltage was less than 125 Vrms ( 175 V peak to peak). The optimum biasing time for the AC-BEN treatment was 45 min, much longer than the DC-BEN treatment that needed only 12 min. In terms of the number of oriented nuclei, it was more than 50% for the AC-BEN treatment, while it was only less than 10% for the DC-BEN treatment. It was hence inferred that concurrent processes of forming an epitaxial SiC layer and the diamond nucleation, induced by the AC-BEN treatment, were responsible for the increase in the number of oriented diamond nuclei. 

---