Structure and morphology development in CVD diamond.

Abstract

The development of vapor deposited diamond for optical, electronic, and protective applications has been hindered by non-diamond carbon contamination, by our present inability to grow a thick, single crystal diamond film, and by our lack of understanding of how the various aspects of chemical vapor deposition interact. Our lack of understanding in this area leaves gaps in our ability to predict how microstructure on this, and many other film surfaces will develop. This dissertation reviews the factors that affect crystal growth rates and the development of microstructure. It also reviews the existing models of microstructure development with an eye toward the development of one that plays a general predictive role in the deposition of covalently bonded solids and in reactive deposition systems. The conclusions drawn point to a model which predicts where boundaries will lie on a morphology map using deposition parameters, such as temperature and pressure, as coordinates. Such a map will depend not only on the material being deposited, but also on the reaction path leading to deposition. Some deposition parameters will always dominate over others. Efforts are made to isolate some of the dominant processes, under a limited set of deposition conditions, for combustion deposited diamond. Previously published data on oriented polycrystalline growth is reconsidered with an eye on the same concerns, particularly around the observed transition between <110> and <100> orientations. Additional experiments in this vein are proposed. The existence of a metastable amorphous state is also proposed and considered. The dependence of amorphicity on bond angle variation makes possible superheated crystals that maintain their crystalline coordination amorphous. However, properties necessary to stabilize this state, high thermal transfer rates and a strongly periodic structure, are opposed to the soft bonding needed to make this state more probable. The state is, effectively, unobservable in silicon, and hence in diamond as well.