Date of Award
Master of Science in Electrical Engineering - (M.S.)
Roy H. Cornely
W. H. Warren Ball
An experimental Ionized Cluster Beam System similar in design to the one developed at Kyoto University in Japan was. designed and constructed in the Microelectronics Laboratory at NJIT. This involved making the working drawings, having the parts machined and then assembling the system in a vacuum system together with the necessary variable power supplies, meters, controls, gas inlets, cooling water connections, etc.
Eleven deposition runs were then attempted to grow gallium nitride utilizing the system constructed. In a typical deposition run, gallium was vaporized from a boron nitride crucible at 1000° C and allowed to expand through a 0.5 mm nozzle into a low pressure region containing nitrogen at 1-5 X 10-4 torr in order to form clusters of about 1000 atoms each. The gallium clusters were ionized by an electron source (300 mA) and deposited on heated quartz substrates (550° C) for reactive growth with nitrogen.
For the above deposition conditions, the film growth rate was only 1 Å/min and the films were highly insulating. Photoluminescence measurements and surface analysis of the grown films revealed boron contaimination which was then traced to water vapor contamination of the boron nitride crucible and heater. The water vapor if not driven out by pre-heating the crucible in an oven, caused the boron nitride to decompose at elevated temperatures introducing reactive contaminants at the substrate surface. Further experiments showed that for GaN deposition rates greater, than 1 Å /min (and up to 1 Å /sec) 0 the crucible temperature should be at least 1100° C but preferably around 1400° C.
Bakiri, Ghulum, "Gallium nitride film deposition by ionized cluster beam epitaxy" (1983). Theses. 1471.