Search

The goal of this program is to develop vapor deposition processes for application to integrated circuit technology, particularly multilayer applications. Its purpose is to investigate vapor deposition techniques that offer potential for synthesis of materials having new, unique structures and/or higher quality than currently attainable. In this period, an in-situ argon ion- gun cleaning technique was to strip the native SiO2 film from the Si substrates prior to the actual deposition of SiO(x) and SiC films. Etch rate data for the native oxide on SiO2 and the relative film thickness for reactively sputtered SiO2 films as a function of substrate temperature for etched substrates was gathered. The relative deposition rate and film texture for single-source sputtered SiC films as a function of substrate type, orientation and film texture was obtained. Exploratory films of SiO(x), WO(x) and NbC(0.88) were fabricated using the sequential deposition of atomic layers technique. In the theoretical area, the CNDO/2 technique was used to determine the energies of formation for bridging oxygen vacancies, and Frenkel defects for a pair of silica structural units (dimer and tetramer ring). In addition, calculations were made for the adsorption of different molecular chemical species (in the parallel configuration) at the bridging oxygen site of a silica dimer.

The program has as its goal the development of vapor deposition processes for application to integrated circuit technology. Its purpose is to investigate vapor deposition techniques that offer potential for synthesis of materials having new, unique structures and/or of higher quality than currently attainable. Technological application of these materials will be a significant consideration in the selection of specific systems for study, particularly multilayer integrated circuit applications. In this period, experiments with the new cryopumped system were performed. Reactive sputtering research focused on a detailed study of SiOx and a more broad guage study of MOx films for a number of metals M (M = Si, W, Al, Zn, Zr, Mg) on 3-in. diameter wafers. High deposition rates, a wide range of stoichiometry and either amorphous or crystalline films were formed. The cleanliness of the new system is orders of magnitude better than the old system. Multiple source deposition studies focused on SiCx formation. X-ray analysis, refractive index and bandgap data all confirm the formation of SiC films. Good control of the sputter deposition of these films has been demonstrated. In the theoretical area, the constraints placed on closed n-ring assemblies (SinO3n) in SiO2 are being assessed to identify those constraints which differentiate the structure between the crystalline and vitreous forms.

This report describes studies of the growth of GaN by the reaction between gallium and ammonia, in the presence of a carrier gas, usually hydrogen. Experiments on the growth of GaN in a horizontal chamber have been modestly successful in that they have shown that crystals over 2.5 mm long can be grown in 10 days. These experiments have identified the temperature and ammonia partial pressure conditions for relatively stable growth and have yielded useful insights into the growth process. Attempts to grow GaN by a vertical gradient transport technique and by the use of gallium/tin alloys have been much less successful. A theoretical model which explains most of the data pertaining to the growth of GaN by this method is included. Finally, recommendations are presented for further study, including suggestions of methods for growing large crystals which should be tried as a continuation to this study. It should be emphasized that this was a short-term feasibility study so that long-term planning based on a detailed understanding of the system was not possible. The approach used was pragmatic and the numerical data was gathered as expeditiously as possible.

In addition to problems of multinucleation, low solubility, and slow convection experienced in earlier experiments, it is postulated that the solubility of GaN in gallium follows a reciprocal temperature dependence. In spite of this handicap, it has been found possible to achieve conditions under which the nucleation of GaN is reduced in comparison to previous experiments and so to produce larger GaN crystals. The major aim of experiments in the near future will be to obtain a more complete understanding of the conditions required to produce large crystals of GaN.

In the previous report, a breakthrough was reported in attempts to prepare large crystals of GaN, in that the maximum crystal size had been increased for the first time above 1 mm. The emphasis in the previous quarter has been to obtain an improved understanding of the factors leading to this improvement and so to determine the conditions necessary for the next advance. Experimental work has been concentrated on the horizontal crucible method which has shown the greatest promise to data. In order to study in more detail the reaction between ammonia and liquid gallium, a series of experiments has been performed in which ten samples were located in separate recesses in a compartmented quartz cruicible. The overall length of the crucible was 15 cm, as in the case of a single vitreous carbon boat, and the compartmented tube was located in the same temperature gradient as in previous experiments with the carbon boat (1010-825 C). The ratio of ammonia to hydrogen flows was 0.002, as in the experiments of Logan and Thrumond, and the hydrogen flow rate was about 225 cu cm/min. The reaction proceeded for 12 days in each case. Prior to crystal growth, the furnace was purged with hydrogen for 12 hours at room temperature, followed by 24 hours at about 1200 C.

The III-V semiconducting phosphides, arsenides and antimonides have been extensively investigated in recent years both with regard to single crystal growth and property evaluation. Relatively high quality material in the form of large single crystals and epitaxial layers can be routinely produced, and many of these materials have found use in device applications. The group III nitrides, on the other hand, have received little attention by comparison and techniques have not been sufficiently developed for the production of either adequate quality bulk single crystals or epitaxial layers. Of the group III nitrides, GaN is of particular interest because it has a wide, direct bandgap (3.4 eV at 300 K) and is potentially useful as an electroluminescent material and injection laser. The possibility of extending the laser frequency with this material further into the visible (into the blue-green region) is very attractive for a number of underwater applications. (Author)

Economical procurement of small quantities of high performance custom integrated circuits for military systems is severely impeded by inadequate process, device and circuit models that handicap accurate computer-aided design at low cost. The salient objective of this program is to formulate physical models of fabrication processes, devices and circuits to allow total computer-aided design of custom large-scale integrated circuit subsystems to reduce development cycle time and cost. The basic areas under investigation are: (1) ion implantation and diffusion of dopants; (2) thermal oxidation; (3) chemical vapor deposition of silicon epitaxy; and (4) device simulation and statistical circuit modeling. This report discusses the results of the second year of the program. (Author)

Economical procurement of small quantities of high performance custom integrated circuits for military systems is severely impeded by inadequate process, device and circuit models that handicap accurate computer-aided design at low cost. The salient objective of this program is to formulate physical models of fabrication processes, devices and circuits to allow total computer- aided design of custom large scale integrated circuit subsystems to reduce development cycle time and cost. The basic areas under investigation are: (1) ion implantation and diffusion of dopants, (2) thermal oxidation, (3) chemical vapor deposition of silicon, (4) device simulation and (5) analytical measurements. This report provides a detailed development of research results in developing computer-aids for modeling the IC fabrication process. The effort is vertically integrated and the outcome is the widely distributed SUPREM program. Moreover, device analysis capabilities have been developed based on SUPREM and applications have demonstrated. The impact of this effort on the design of custom VLSI is apparent and future research will explore, define and model the limits of technology for VLSI.

Contents: Epitaxial crystal growth; Device applications of GaAs; Relations between dislocations and mechanical properties and the production and characterization of defect structures in compound semiconductors; Scientific aspects of gallium arsenide crystal preparation; The synthesis, characterization and device application of gallium nitride: preparation of GaN light-emitting diodes., See also AD746882.

The migration of slab, cylindrical and spherical zones through a block of solid under the influence of a temperature gradient has been analyzed theoretically. The migration rate is found to be directly proportional to the temperature gradient in the system and to depend upon orientation dependent atomic kinetics of the melting freezing processes. The magnitude of the atomic kinetic effect depends upon the size of the molten zone. Stable zone migration occurs only in certain crystallographic orientations. The theory is discussed in terms of the limited experimental results available.

The migration of slab, cylindrical and spherical zones through a block of solid under the influence of an electric field gradient has been analyzed theoretically. It was determined that the zone could migrate either up or down the field gradient, E*, depending upon the magnitude and sign of (i) the effective ionic mobility of the solvent atoms, U, (ii) the Peltier coefficient at the solid-melt interface, and (iii) the imposed temperature gradient, G*.

The paper is concerned with the reorientation of the field of crystallization to a science-based, as opposed to an art-based, technology. A summary of the fields of science and the general knowledge which are necessary to engineer a crystallization application is followed by a discussion of the Crystallogenics approach and its successes. Finally, consideration is given to present exploitation opportunities in the areas of materials purification and concentration; single crystal growth, with particular attention given to the growth of silicon webs, large selenium single crystals, and large sapphire single crystals; and structure control of ingots and castings.