Optimization of Seeding for AlN Growth

Aluminum Nitride (A1N) substrates are attractive for III-nitride epitaxial growth due to their high thermal conductivity, close lattice and thermal expansion match to IlI-nitride compositions used for opto-electronic and electronic devices and their relative ease of growth when compared to Gallium Nitride (GaN). Vapor sublimation is an attractive method for growing A1N bulk crystals. Seeded growth is the best way to maximize the crystal growth yield. In standard crystal growth from the melt, a small seed crystal is dipped into the melt and the crystal diameter is slowly increased to reduce the probability of growth failures such as twins or polycrystal. The cone region uses a significant freaction of the melt and results in unusable crystal. A similar process can be used for sublimation growth but for large crystals this method is too inefficient and wasteful. Optimally, sublimation growth of A1N requires a low-defect A1N seed of the same diameter as the crystal to be grown. Unfortunately, A1N crystals of appropriate size are not yet available and other crystals which are not optimally matched to A1N need to be used as seed crystals. In Phase I we propose to determine at least one appropriate, large-area seed for A1N sublimation growth. The best seed will be optimized in Phase II. We recognize that when we have succeeded in growing large-area A1N crystals the hetero-seeding problem will have been solved. Nevertheless, seeding even with A1N seeds will continue to be a significant problem, as it still is with very well established crystal growth methods. After the successful growth of the first large-area A1N crystal, we will optimize A1N as the seed.