During the early stages of lattice mismatch epitaxy, island growth is observed in several systems. The evolution of this morphology is attributed to a large value of the substrate/layer interface energy. Origins of misfit and threading dislocations are also considered. Nucleation of misfit dislocations could occur either from steps present on surfaces of islands or from substrate/island edges. Arms of glide loops terminating at the surface in the first case form threading segments in the layer, whereas the coalescence of islands in which misfit dislocations are not aligned with respect to each other lead to threading dislocations in the second case. Existing approaches for lowering threading dislocation densities are also evaluated. A novel approach involving controlled, self-assembly of islands is presented to achieve the preceding objective.

Most types of GaN based Field Effect Transistors – MESFETs, MISFETs, Heterostructure FETs (HFETs), and Doped Channel HFETs (DC-HFETs) have demonstrated a good performance at both room temperature and elevated temperatures. The DC-HFETs demonstrated the best direct current and microwave characteristics among all wide band gap semiconductor devices because of excellent transport properties of two dimensional electron gas at the AlGaN/GaN heterointerface and a large sheet carrier concentration in the device channel. Other advantages of GaN-based materials for FET applications include a very high breakdown field, a very high saturation field, a high peak velocity, a large conduction and valence band discontinuities at the AlN-GaN heterointerface, and a reasonable thermal conductivity (comparable to that of Si). In this paper, we review our recent results obtained for different types of GaN-based transistors.

An overview of the development of a new dielectric material, cubic-GaS, from the synthesis of new organometallic compounds to the fabrication of a new class of gallium arsenide based transistor is presented as a representative example of the possibility that inorganic chemistry can directly effect the development of new semiconductor devices. The gallium sulfido compound [(tBu)GaS]4, readily prepared from tri-tert-butyl gallium, may be used as a precursor for the growth of GaS thin films by metal organic chemical vapor deposition (MOCVD). Photoluminescence and electronic measurements indicate that this material provides a passivation coating for GaAs. Furthermore, the insulating properties of cubic-GaS make it suitable as a the insulating gate layer in a new class of GaAs transistor: a field effect transistor with a sulfide heterojunction (FETISH).

In this study deposition temperature and post-deposition treatments, such as vacuum break, furnace anneal and RTA, were coupled together to improve Ti/TiN barrier properties. Raising substrate temperature during deposition resulted in lower sheet resistance and stresses, but worse Sheet resistance (Rs) uniformity, suggesting larger-grain growth and non-uniform silicidation. The RTA treatment appeared to be the most effective one to improve barrier integrity due to high temperature densification and oxygen rich along grain boundaries. To include contact geometry effect, the barrier processes were incorporated with hot Al on the contact level of 0.6µm SRAM. After alloying at 430°C for 45 min, the yields turned out to be similar, although the RTA treatment resulted in lower P+ contact resistance, implying more complete activation, silicidation and less dopant segregation. The result suggests that it is feasible to deposit Ti/TiN insitu with hot Al process to enhance throughput. The yield result after a second alloying set at 500°C for 30 min showed that RTA treated Ti/TiN was the only one to survive.

KBi6.33S10 and K2Bi8S13 were synthesized by the direct combination of K2S/Bi2S3 at high temperature (>700°C). The reaction of K2S/3.3Bi2S3 at 800°C revealed the presence of a new ternary sulfide KBi6.33S10 (I) (92% yield). The structure consists of blocks of Bi2Te3 and Cdl2-type units that are connected to form a three-dimensional network with K+ ions located in the channels that run along the b-axis. The same reaction but with a different ratio, at 750°C, gave the new ternary sulfide K2Bi8S13 (II) (94% yield). The structure of the shiny rod-like crystals is closely related to that of I. As in I, it also consists of Bi2Te3 and Cdl2-type fragments that connect to form K+-filled channels. The two potassium atoms and one bismuth atom are disordered over three sites. Some preliminary electrical and optical properties of these materials are discussed.

The microstructure of CdTe sources grown on bare 7059 and SnO2-coated 7059 glass substrates used in the close-space sublimation (CSS) growth of CdTe films, is affected by nucleation conditions during their fabrication. For CdTe sources grown on bare glass substrates, denser source microstructures with smaller grains are promoted by low-temperature nucleation conditions. Growth rates of CdTe films deposited using these source plates are inversely proportional to the packing density of the grains. CdTe sources grown on glass undergo significant texture changes during their use in subsequent CSS deposition. CdTe sources grown on SnO2-coated glass substrates exhibit dense structures both with and without low-temperature nucleation and increased grain size with low-temperature nucleation. These source plates show less texture change associated with thermal annealing than their bare-glass counterparts, however, they are much more susceptible to CdO formation during subsequent CSS deposition processes in which oxygen is present. Source oxidation results in a decrease of the (111)/(220) x-ray diffraction intensity. Analysis of the (111)/(220) intensity ratio with and without source oxidation suggests the preferential formation of CdO on the (111) planes. In general, source oxidation appears to be a inversely proportional to the grain size of the source material.

The structural and optoelectronic properties of polycrystalline CdS films, fabricated by three different methods, are compared to one another for the purpose of preparing CdTe/CdS solar cells. The three methods were: alternated spraying of cation and anion solution at room temperature, spray pyrolysis, and chemical bath deposition. We studied the surface morphology and crystal quality and texture by scanning electron and optical microscopy and x-ray diffraction. All films had a well-developed wurtzite structure. Films grown by the alternated-spray method and the chemical bath method consist of randomly-oriented crystallites with dimensions <0.5 microns. Annealing at 400°C increases the crystallite size slightly. Films which were grown by pyrolysis at substrate temperatures from 400°C to 500°C were oriented in the <002> direction. For growth by pyrolysis at 500°C, the surface is rough on a lateral scale of 0.1 to 0.3 microns.

Several n-type 4H-SiC samples grown by the physical vapor transport technique were studied using both thermal and optical admittance spectroscopy. The ionization energy ED for nitrogen donors in 4H-SiC was determined to be 0.053 eV and 0.10 eV below the conduction band. This agrees reasonably well with the values of 0.052 eV and 0.092 eV determined by IR absorption measurements. It is believed that EC-0.053 eV is the ionization energy of the nitrogen atom occupying the hexagonal site (h). The 0.10 eV activation energy has been attributed to the nitrogen atoms occupying the cubic sites(k).

The optical admittance studies reveal five conductance peaks. One of these corresponds to the band-to-band transitions from which the bandgap of 3.41 eV was determined at 40K. The other four conductance peaks correspond to transitions from defect levels to the conduction band. These defects are located at EC-1.73 eV, EC-1.18 eV, EC-0.87 eV, and EC-0.72 eV.

6H-SiC epitaxial layers with a background n-type dopant concentration of 1 × 1016/cm3 were hot implanted to doses ranging from 4.0 × 1013 to 1.8 × 1014 Al ions/cm2 at 65, 135, and 220 keV to achieve a box-type implant distribution to a depth of 300 nm. Electrical activation of dopants was carried out using a proximity annealing method at 1500°C in a buffer environment to retard surface degradation of the SiC samples. Measurements using atomic force microscopy illustrated the morphological stability of the SiC surface during the high-temperature annealing. Transmission line measurements showed some degree of dopant activation. Characterization of fabricated p-n junction diodes demonstrated p-type conduction in the aluminum-implanted SiC samples.

An explicit analytical expression relating the interface trap densities and transconductance is derived for enhancement mode field effect transistors without any simplifying assumptions regarding the energy distribution of traps. Using this relationship, the interface trap densities were calculated from transconductance data and compared to experimental data and that provided in the literature. Our expression provides a simple and convenient method to reliably estimate interface traps densities from the readily available transconductance data provided in the pertinent literature.

Fast and slow interface traps can considerably deteriorate the performance of field effect transistors. Slow interface traps, by slowly changing their charge occupancy, contribute to a drift in the quiescent operation point of the transistor, while fast traps deteriorate the device performance by contributing to both amplitude and phase current noise. They also result in a non-equilibrium surface depletion layer between gate and source which increases the gate-to-source parasitic resistance and deteriorates the device transconductance. We examine these different effects and present some preliminary data regarding interface traps in boron-doped 611-SiC.

Thermoelectric properties (Thermoelectromotive force and electrical conductivity) of porous silicon carbide ceramics are evaluated. Porous silicon carbide ceramics are made from rice hull by reacting silica and carbon, both of which are included in rice hull, using hotpressing method[1].When nitrogen gas is used as enviromental gas for hotpressing, this material becomes n-type semiconductor. It shows comparably high themoelectromotive force. Thermoelectric properties of this material change with the processing conditions.The relationship between thermoelectric properties and microstructure is also evaluated.

The roles which ligand size and stability play in influencing solid-state thermolyses products and phases are developed for a group of metal thiolate compounds. The exploration includes Pb(SR)2 [R = t-Bu, i-Bu, s-Bu], Cd(SR*)2 [ R* = i-Pr, I-Bu, Bz], ClHgSR** [ R** = i- Pr, neo-Pent, Bz] and In(SR***)3 [R*** = 2,4,6-(i-Pr)3C6H2, 2-CH3O-5-CH3C6H3, o- C6H4CH2N(CH3)2]. In several examples, the precursors have been characterized by single crystal X-ray diffraction. In most cases, solid-state decomposition products have been identified by XRPD, and volatile decomposition products have been identified by GC/MS. All precursors have been studied by TGA.

The synthesis, characterization and decomposition of several indium thiolates containing the bulky substituted aryl ligand 2,4,6-i-Pr 3C6H2 (Ar′) or the internally chelating ligands 2-CH3O,5-CH3C6H3 (Ar″) and o-C6H4CH2N(CH3)2 (Ar"‘) are described. Two synthetic methods have been utilized: metathesis reactions between lithium thiolates and the appropriate metal halides and the addition of elemental metal to diaryl disulfides. The thermal decomposition of each indium precursor results in the formation of 1n2S3, based on thermogravimetric data. The homoleptic compound In(SAr′)3 can be isolated as a yellow oil. This liquid precursor has been derivatized by the reversible formation of acetonitrile and tetrahydrofuran adducts. Although, the molecule exists as a monomer in both adducts, the coordination number of the metal and he orientation of the ligands are markedly different. The internally chelating In(SAr″)3 and In(SAr.″′)3 adopt contrasting dimeric and monomeric structures respectively.

Recent research on the growth and structure of carbide nanorods is reviewed. Carbide nanorods have been prepared by reacting carbon nanotubes with volatile transition metal and main group oxides and halides. Using this approach it has been possible to obtain solid carbide nanorods of TiC, SiC, NbC, Fe3C, and BCx having diameters between 2 and 30 nm and lengths up to 20 µm. Structural studies of single crystal TiC nanorods obtained through reactions of TiO with carbon nanotubes show that the nanorods grow along both [110] and [111] directions, and that the rods can exhibit either smooth or saw-tooth morphologies. Crystalline SiC nanorods have been produced from reactions of carbon nanotubes with SiO and Si-iodine reactants. The preferred growth direction of these nanorods is [111], although at low reaction temperatures rods with [100] growth axes are also observed. The growth mechanisms leading to these novel nanomaterials have also been addressed. Temperature dependent growth studies of TiC nanorods produced using a Ti-iodine reactant have provided definitive proof for a template or topotactic growth mechanism, and furthermore, have yielded new TiC nanotube materials. Investigations of the growth of SiC nanorods show that in some cases a catalytic mechanism may also be operable. Future research directions and applications of these new carbide nanorod materials are discussed.

Several novel transition metal nitrides and oxynitrides were synthesized via ammonolysis of solid state oxide precursors at temperatures ranging from 700°C-900°C and reaction times ranging from 12 hours to 4 days. The products were characterized by powder X-ray diffraction and their structures were determined by powder X-ray Rietveld refinement. The relationships between the structures of these nitrides and oxynitrides, and their similarity to the structures of the transition metal dichalcogenides, is discussed.

We demonstrate a simple electrochemical method suitable for preparing precursors for nine different metal-nitride ceramics. These precursor powders were calcined in flowing NH3 or Ar to yield metal-nitride ceramic powders. This electrochemical synthetic technique was applied to Al, Ga, Mo, Nb, Ni, Ti, V, W, and Zr and resulted in a nitride of each of these metals. The calcination conditions determined the resulting phase, composition, and morphology of the product. The ceramic materials were characterized primarily by powder XRD.

A new chemical synthesis approach has been developed in which ternary transition metal nitride powders are synthesized from hydrolyzed alkanoamine complexed precursors. In this technique, two different metal chlorides are dissolved in a common solvent such as acetonitrile, then reacted with ethanolamine to form a highly viscous solution which settles out of the solvent phase. Simultaneous evaporation of the solvent and forced hydrolysis of the remaining liquid results in the formation of a precipitate which can be filtered and dried. The precursor is heat treated in ammonia to form the corresponding ternary nitride. The approach has been successfully used to synthesize Ni3Mo3N, FeWN2, and Ti3AlN powders, each of which have been characterized using X-Ray Diffraction (XRD) and Scanning Electron Microscopy (SEM).

Irradiation of {[(CH3)3Si]2N}3Y, Zr[N(C2H5)2]4, and Nb[N(C2H5)2]4 by the 1.064 μm line of a pulsed Nd:YAG laser results in the decomposition of Zr[N(C2H5)2]4 and Nb[N(C2H5)2]4 while {[(CH3)3Si]2N}3Y remains unaffected. The decomposition is photolytic and is accompanied by a visible emission tracing the path of the incident beam. The decomposition products form corresponding carbide/nitrides on pyrolysis in vacuum and nitrides on pyrolysis in an ammonia atmosphere. The spectroscopic investigation of the visible emission, and the pyrolysis of the decomposition products to metal nitrides is described.

The synthesis of a new lithium tantalum oxynitride is described. The compound, LiTaO3-3xN2x was synthesized by reacting LiTaO3 with anhydrous ammonia at temperatures ranging from 773 K to 1173 K. Depending on the reaction temperature the sample color changes from white to tan (773 K) to bright orange (973 K) to red (1123 K). The color appears to be the result of a p-d transition consistent with a charge transfer from localized nitrogen states to a band of predominately metal character. Electrical and magnetic measurements indicate that the material is a diamagnetic insulator.