Chemical vapor deposition of diamond in flames and fluidized beds

Citation

Shin, Ho Seon (1996) Chemical vapor deposition of diamond in flames and fluidized beds. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/25pq-ng74. https://resolver.caltech.edu/CaltechETD:etd-06152005-130942

Abstract

NOTE: Text or symbols not renderable in plain ASCII are indicated by [...]. Abstract is included in .pdf document.

An experimental and computational study of chemical vapor deposition (CVD) of diamond in low pressure flames, and an experimental study of microwave plasma-enhanced CVD of diamond on particles in fluidized beds are presented. Diamond film growth experiments were performed in low pressure (30-52 Torr) acetylene/oxygen flames and the effects of varying substrate temperature, equivalence ratio, and pressure on diamond growth were examined. Uniform diamond films fully covering 5 cm diameter substrates at growth rates of up to 2.3 [...]/hr were grown at 30 Torr using a 4 cm diameter flat flame burner. A carbon-conversion efficiency of up to 3.5 x [...] was obtained which was comparable to that observed in the atmospheric pressure torch method. The Raman spectrum of the deposited film showed good diamond film quality.

To extend the combustion synthesis technique for diamond to fuels other than acetylene, and to reduce the cost of diamond produced by combustion synthesis, growth experiments using several alternative hydrocarbon fuels (MAPP, propylene, ethylene, and propane) were performed at 50-180 Torr. Well-faceted diamond films at growth rates of up to 1.0 [...]/hr were grown in these alternative fuel flames. The Raman spectrum analysis showed that good quality diamond films were grown in MAPP/oxygen and propylene/oxygen flames. An economic comparison study showed that switching from acetylene to propylene may be able to lower the fuel cost per unit mass of diamond by roughly a factor of three.

A numerical modeling study was performed to analyze the growth environment. The model predicts peak flame temperatures above the adiabatic flame temperature, and a chemical environment near the substrate far from its equilibrium state. Surface concentrations of H and CH3, and the ratio of H to CH3 mole fractions in low pressure acetylene/oxygen flames are similar to those observed in hot filament reactors. The simulations of low pressure (25-30 Torr) acetylene/oxygen flames near diamond growth conditions suggest that increasing the mass flow rate while reducing the pressure is favorable for increasing the growth rate, and high quality diamond can be grown in leaner flames.

Although the values are slightly lower than for acetylene/oxygen flames, alternative fuel flames are predicted to have high enough H and CH3 concentrations at the substrate to grow diamond at a reasonable growth rate. The results indicate that nonequilibrium flame chemistry is important in the low pressure combustion environment. This suggests that still other fuels may be worth considering for diamond growth at low pressures. Further studies employing alternative fuels other than acetylene could potentially further reduce the cost of diamond produced by combustion synthesis, and could potentially reduce the barriers to commercializing the combustion synthesis of diamond for many applications.

To grow continuous, conformal diamond coatings on small, irregular objects, experiments were performed using microwave plasma-enhanced fluidized beds. Studies were carried out to map the parameter space leading to diamond growth and to determine the relationship between gas composition and diamond growth rate. The effects of varying gas composition and pressure on deposited carbon morphology, growth rate, and nucleation density were examined in these experiments.

Oxygen addition had a strong influence on growth rate and morphology over the range of gas compositions studied. No diamond deposition was obtained without O2. Well-faceted diamond at a growth rate of up to 6 [...]/hr was observed to grow on 0.25-0.7 mm diameter silicon and SiO2 seed particles using up to 15.0% CH4 in H2 with addition of O2. Unlike diamond deposition on bulk substrates, no surface pretreatment was necessary for diamond nucleation. Well-faceted continuous diamond coatings were deposited on seed particles after 8 hours at a pressure of 9 Torr and 120 Watts of microwave power with flow rates of 160 sccm of 2.0% CH4 in H2 and 3 sccm of O2. The micro-Raman spectrum of the deposited diamond crystal confirmed good diamond quality. These results show that plasma-enhanced fluidized beds can be effectively used to deposit diamond coatings on small objects of complex shape.

Thesis Files