A New Approach to Growth of Bulk Zno Crystals for Wide Bandgap Applications

  • PDF / 674,148 Bytes
  • 6 Pages / 414.72 x 648 pts Page_size
  • 9 Downloads / 193 Views

DOWNLOAD

REPORT


temperatures. In a recent article [12], researchers at Bell Labs and Lucent Technologies reported growing GaN films using ZnO substrates. They reported that ZnO had a good lattice match and ideal structure for GaN epitaxy. However, the major drawback in using ZnO as a substrate material was the lack of availability of high quality crystals, cost and instability of the crystal at high temperature in a reducing atmosphere. It is known that zinc oxide has a band gap of 3.5 eV, which is very similar to that of GaN. Recently it has been demonstrated that ZnO can be used as a source for short wavelength lasing [13]. The intensity of optical excitation in ZnO epitaxial films grown on sapphire substrates was comparable to GaN. This is perhaps the most tantalizing use for ZnO. By combining established ZnO MOCVD technology with readily available, low cost, high quality ZnO substrates for homoepitaxy of ZnO films, the major problem of lattice mismatch currently facing the GaN technology is avoided. Single crystal ZnO is unavailable because the material can not be grown using standard

melt growth processes. In standard melt growth atmospheres, the ZnO decomposes upon heating into a very defective structure. Currently, the only techniques for growing bulk single crystal ZnO are sublimation and hydrothermal growth [14, 15]. The hydrothermal technique is relatively less expensive, but the crystal quality is poor, and the ultimate size is limited. The sublimation technique is a slow process, which increases the cost of crystals produced. Standard crystal growth techniques rely on growth from a melt or liquid. The melting point of stoichiometric ZnO has been difficult to obtain, since zinc oxide decomposes above 1800'C. At atmospheric pressure and temperatures near the melting point, ZnO rapidly loses lattice oxygen and reduces to a highly defective, non-stoichiometric ZnO(1 X). Estimates for the melting point are approximately 1880'C. At this temperature, only noble metal crucibles are candidates for the containment of molten ZnO. However, the melt atmosphere required to protect the crucible from oxidation reduces the ZnO lattice. Conversely, the melt atmosphere required to maintain ZnO stoichiometry quickly oxides even iridium at 1880'C. The oxidation from the crucible would readily react with the ZnO in such hostile growth environments, introducing impurities into the lattice. The technique used by Cermet eliminates the problem of crucible reactivity, crystal contamination and decomposition by using a high pressure variation of the established skull melting technique (for which patents will issue in Fall of 1998). The system is capable of melting materials at temperatures in excess of 3600*C and pressures in excess of 100 atmospheres. In general, rf melting is accomplished by placing a suitably conductive material in an alternating magnetic field. Induced fields in the material produce eddy currents which produce joule heating in the charge. The rf melting technique utilizes a segmented, water-cooled, copper crucible, which co