ALBERT

Description: The use of thermal oxides of Al(x)Ga(1-x)As (x = 0-1) as masking materials for selective-area epitaxy by a organometallic chemical-vapor deposition has been investigated. It was found that the thermal oxide of GaAs is only applicable for low growth temperatures (less than or equal to 600 C), and the addition of aluminum significantly improves the thermal stability of the oxide. The oxide of Al(0.4)Ga(0.6)As is suitable for high-temperature deposition, but there are criteria for the thickness and oxidation temperature. Thin layers of AlAs oxidized at 475 C are excellent masks and allow precise thickness control. Promising results of selective-area deposition using these aluminum oxide masks have been obtained. High-quality single crystal grew in mask openings uniformly surrounded by dense and fine-grain polycrystalline deposits, producing a planar duplication of the original pattern.

Description: This final report describes a three-year research effort, aimed at developing new types of THz low noise receivers, based on bulk effect ('hot electron') nonlinearities in the Two-Dimensional Electron Gas (2DEG) Medium, and the inclusion of such receivers in focal plane arrays. 2DEG hot electron mixers have been demonstrated at 35 and 94 GHz with three orders of magnitude wider bandwidth than previous hot electron mixers, which use bulk InSb. The 2DEG mixers employ a new mode of operation, which was invented during this program. Only moderate cooling is required for this mode, to temperatures in the range 20-77 K. Based on the results of this research, it is now possible to design a hot electron mixer focal plane array for the THz range, which is anticipated to have a DSB receiver noise temperature of 500-1000K. In our work on this grant, we have found similar results the the Cronin group (resident at the University of Bath, UK). Neither group has so far demonstrated heterodyne detection in this mode, however. We discovered and explored some new effects in the magnetic field mode, and these are described in the report. In particular, detection of 94 GHz and 238 GHz, respectively, by a new effect, 'Shubnikov de Haas detection', was found to be considerably stronger in our materials than the cyclotron resonance detection. All experiments utilized devices with an active 2DEG region of size of the order of 10-40 micrometers long, and 20-200 micrometers wide, formed at the heterojunction between AlGaAs and GaAs. All device fabrication was performed in-house. The materials for the devices were also grown in-house, utilizing OMCVD (Organo Metallic Chemical Vapor Deposition). In the course of this grant, we developed new techniques for growing AlGaAs/GaAs with mobilities equalling the highest values published by any laboratory. We believe that the field of hot electron mixers and detectors will grow substantially in importance in the next few years, partly as a result of the opportunity given us through this grant, which represents the major effort in the US so far. We note, however, that parallel research on hot electron mixers in thin film superconductors in Russia, and recently in Sweden, have demonstrated mixing up to 1 THz, with the potential for low-noise receivers for frequencies up to many THz. The three groups recently assessed the relative adtantages of 2DEG and superconducting film mixers in a joint paper (Kollberg et al., 1992; see Appendix II).