This dissertation focused on development of a novel apparatus and techniques for crystal growth of bulk gallium nitride (GaN) using the sodium flux method. Though several methods exist to produce bulk GaN, none have been commercialized on an industrial scale. The sodium flux method offers potentially lower cost production due to relatively mild process conditions while maintaining high crystal quality. But the current equipment and methods for sodium flux growth of bulk GaN are generally not amenable to large-scale crystal growth or in situ investigation of growth processes, which has hampered progress.

A key task was to prevent sodium loss or migration from the sodium-gallium growth melt while permitting N2 gas to access the growing crystal, which was accomplished by implementing a reflux condensing stem along with a reusable sealed capsule. The reflux condensing stem also enabled direct monitoring and control of the melt temperature, which has not been previously reported for the sodium flux method.

Molybdenum-based materials were identified from a corrosion study as candidates for direct containment of the corrosive sodium-gallium melt. Successful introduction of these materials allowed implementation of a crucible-free containment system, which improved process control and can potentially reduce crystal impurity levels.

Using the new growth system, the (0001) Ga face (+c plane) growth rate was >50 mum/hr, which is the highest bulk GaN growth rate reported for the sodium flux method. Omega X-ray rocking curve (?-XRC) measurements indicated the presence of multiple grains, though full width at half maximum (FWHM) values for individual peaks were <100 arcseconds. Oxygen impurity concentrations as measured by secondary ion mass spectroscopy (SIMS) were >1020 atoms/cm3, possibly due to reactor cleaning and handling procedures.

This dissertation also introduced an in situ technique to correlate changes in N2 pressure with dissolution of nitrogen and precipitation of GaN from the sodium-gallium melt. Different stages of N2 pressure decay were identified and linked to distinct modes of GaN crystal growth. Based on this analysis, the highest growth rate may have up to 90 mum/hr, and did not appear to be mass-transport limited, in contrast to previous reports.

Several refinements and improvements of the novel equipment and methods introduced were also explored, all of which have potential to further advance state of the art sodium flux growth of bulk GaN.