A Ni3Al-based alloy containing Cr is examined after preparation by Osprey deposition. The microstructure consists of solidified spherical regions showing cellular segregation interspersed with regions of finer, equiaxed segregation morphology. The segregation structure is characterized by cell interiors rich in aluminum and poor in chromium, while the cell walls are poor in aluminum and rich in chromium. This segregation pattern is the inverse of that expected, based on earlier melt spinning experience, and is explained in terms of the undercooling of the melt prior to solidification. Very high temperature annealing is required to homogenize the material, despite the use of this rapid solidification process for material fabrication.

The structure and phase relations of Te–Ge–Sn thin films are examined with application to erasable optical storage media. Free energy data from the literature predict that the region of the Te–Ge–Sn phase diagram between Ge, Sn, and the TeGe–TeSn pseudobinary consists of one two-phase field [α–Ge and Te50 (GexSn1−x)50] and one three-phase field (α–Ge, β–Sn, and TeSn). Electron diffraction from five different Te–Ge–Sn films annealed at 623 K experimentally confirms this prediction. One composition from the two-phase field is deposited as a tri-layer film with the structure 150 nm SiO2/75 nm Te36.3Ge47.4Sn16.3/150 nm SiO2 on a grooved disk substrate, and the microstructure resulting from low-power (12 mW) CW and higher-power (∽50 mW) pulsed laser exposure is studied by transmission electron microscopy and electron diffraction. Of particular significance is that laser-induced crystallization produces a single-phase structure consisting of the Te–Ge–Sn compound phase which is supersaturated with respect to the excess Ge. This supersaturation leads to a disordering of the equilibrium NaCl-type structure of this phase. Crystallization of a micron-sized amorphous spot on a ∼200 ns time scale occurs by a diffusionless process. The fast erase times required by a phase-change optical recording application can thus be achieved in off-stoichiometric compound compositions by way of a nonequilibrium crystallization process.

The conductivity and photoconductivity are measured on a high-surface-area disordered carbon material, i.e., activated carbon fibers, to investigate their electronic properties. This material is a highly disordered carbon derived from a phenolic precursor, having a huge specific surface area of 1000–2000 m2/g. Our preliminary thermopower measurements show that the dominant carriers are holes at room temperature. The x-ray diffraction pattern reveals that the microstructure is amorphous-like with Lc ≃ 10 Å. The intrinsic electrical conductivity, on the order of 20 S/cm at room temperature, increases by a factor of several with increasing temperature in the range 30–290 K. In contrast, the photoconductivity in vacuum decreases with increasing temperature. The magnitude of the photoconductive signal was reduced by a factor of ten when the sample was exposed to air. The recombination kinetics changes from a monomolecular process at room temperature to a bimolecular process at low temperatures, indicative of an increase in the photocarrier density at low temperatures. The high density of localized states, which limits the motion of carriers and results in a slow recombination process, is responsible for the observed photoconductivity.

Oxide thin films on Si substrates were prepared from carboxylate precursors by the reaction of the metal nitrates and ammonium trimethylacetate. Precursor salts were characterized with respect to purity, structure, thermal pyrolysis, and phase development during calcination. A solvent system, based on carboxylic acid/amine mixture, was developed to dissolve the synthesized precursors, resulting in increased solubility, viscosity, and stability. Smooth, fine-grained ZrO2, Y2O3, and YSZ films were obtained on Si wafers by spin-coating and subsequent heat treatment above 500 °C. Films heat treated below 700 °C were generally adherent, amorphous, or microcrystalline, while YSZ and ZrO2 showed (111) preferred orientation above 700 °C. These oxide films show promise as protective or buffer layers on Si wafers.

A two-step process has been developed to form highly oriented thin films in material systems with dissimilar crystal structures and interatomic spacings. This processing method utilizes current polycrystalline thin film deposition techniques. In this method, a polycrystalline thin film is first deposited and heat treated to promote its breakup into isolated grains. The breakup process favors those grains that have a low substrate interfacial energy and so produces a film of highly oriented but isolated grains. In the second process step, another polycrystalline thin film is deposited. The remnant grains act as seeds for the growth of a highly oriented thin film. The process is demonstrated through the growth of highly (100) oriented thin films of cubic ZrO2 (25 mol % Y2O3) on (0001) Al2O3 single crystal substrates, a material system in which film and substrate have dissimilar structures and interatomic spacings. Implications for the growth of epitaxial films using this method are discussed.

Novel apparatus and methodology have been developed for measuring the fracture toughness of ceramics using small disk-shaped samples 3 mm in diameter. The method involves the Vickers indentation of specimens ranging in thickness from 300 to 700 μm, and testing them in a ring-on-ring bending mode. Experiments on ZnS grown by chemical vapor deposition (CVD) have been used to evaluate the technique. The apparent fracture toughness of this material increases with crack length (R-curve behavior). This behavior is analyzed using established equations in conjunction with a new graphical method that obviates the need for measuring the apparent toughness of samples containing very large cracks. The fracture toughness at “infinite” crack length, 0.86 ± 0.08 MPa · m1/2, is in very good agreement with the values obtained by conventional methods. The effect of sample size on the accuracy of the results is thoroughly discussed. The thickness, t, of the sample should exceed by a factor of ten the initial crack length produced by the indentation. Also, since samples are tested at various indentation loads, P, the ratio of their thicknesses should be chosen to satisfy the condition t ∝ P2/3. The importance of several parameters in the design of the apparatus is also discussed.

Solid-state amorphization reaction (SSAR) between GaAs and Co thin films was investigated by transmission electron microscopy and Auger electron spectroscopy. Upon annealing of GaAs/Co thin-film couples at 260–300 °C, an amorphous phase was observed to form. Annealing at higher temperatures or for longer times led to the crystallization of the amorphous phase into a supersaturated CoAs solid solution phase with the B31 structure. Amorphization is attributed to the rapid diffusion of Co in the rather open GaAs structure. In order to consider the thermodynamic driving force for amorphization and subsequent crystallization, the phase diagram of CoGa–CoAs was investigated using DTA and metallography. The pseudobinary system was modeled thermodynamically to yield relative stability data for the various phases between GaAs and Co. These data were used to rationalize the amorphization process.

In situ gravimetric measurements and microscopic examinations were used to determine the mechanisms of oxygen removal from synthetic ilmenite disks between 823 and 1353 K. Under a hydrogen atmosphere, iron was observed to form a layer of low porosity on the surface of samples early in the reduction. This created diffusion limitations for hydrogen to the reaction front and for the escape of water vapor. A shrinking core reduction model, modified to include the growth of this iron film, was capable of predicting the conversion-time relationships of ilmenite samples. An activation energy of 43.2 ± 2.6 kcal/gmole was determined to be representative of reaction control over the temperature range 823–1023 K.

Polycrystalline CdSe and CdTe layers were fabricated by putting liquid Cd in contact with Se or Te vapors under constant Ar flow. The crystalline structure, surface properties, and semiconducting properties of these materials have been determined. Both materials were found to be n-type semiconductors. The results show that, under the proper experimental conditions, the liquid metal-vapor reaction enables the synthesis of polycrystalline CdSe photoelectrodes with a 6.9% energy conversion efficiency when used in an electrochemical photovoltaic cell under 80 mW/cm2 of white light illumination. This efficiency rates amongst the highest ones measured under similar conditions using polycrystalline CdSe. These CdSe layers have a majority charge carrier density of ND = 2.6 × 1017 cm−3 and possess a highly textured surface which is assumed to be mainly responsible for the high photovoltaic efficiency. The highly textured CdTe samples obtained by this process, however, show a photovoltaic conversion efficiency of only 0.2%, and this is seen to be mainly due to their high majority charge carrier density of ND = 7.8 × 1019 cm−3.

A quasicrystalline transition state was found in an Al65Cu20Fe15 alloy after heat treatment at 700 °C and subsequent slow cooling of samples originally containing the icosahedral quasicrystalline phase. The transition state is characterized by the simultaneous presence of two types of quasicrystalline domains. Their structure can be obtained by the introduction of one-dimensional periodicity along one of the fivefold axes of the original icosahedral quasicrystal. The two types of domains differ in their period lengths which are obtained by multiplying a fundamental length of the icosahedral structure by the Fibonacci numbers 5 and 8.

Chemical vapor deposition (CVD) of boron nitride (BN) is most readily performed using BCl3 and NH3, which are brought into the deposition zone through two separate tubes. This causes some problems: inadequate mixing leading to a nonuniform deposit, formation of solid intermediates, etc. To avoid these problems, the process was performed by mixing BCl3 and NH3 at elevated temperatures (120–220 °C) prior to entering the deposition zone. The reaction between them took place by the forming of volatile stoichiometric B–N compounds (trichloroborazine and iminochloroborane), which were then transported through a single tube into a deposition zone. The resulting deposit was found to be hexagonal boron nitride.

Point-defect-mediated atomic diffusion in GaAs and AlxGa1−xAs–GaAs superlattices is examined thermodynamically by focusing on activation enthalpy of diffusion. Through a review of available experimental results of impurity diffusion of Si, Zn, and Be, their diffusion phenomena are discussed by taking the characteristics of column-III-site-related point defect, such as Ga vacancy and arsenic-antisite, into consideration. It is suggested that Zn and Be diffusion should be mediated by As-antisite defects. On the other hand, Si diffusion is mediated by either Ga vacancy or As-antisite, depending on the growth method of materials and diffusion conditions. It is argued that As-antisite should be a mediator of diffusion in As-antisite-rich materials and with using As-rich diffusion source under p-type conditions. Cation self-diffusion or interdiffusion is also discussed in the same manner. Impurity-enhanced layer-disordering phenomena are examined by considering the reduction of defect energy of Ga vacancy and As antisite under n-type and p-type conditions, respectively. Beryllium enhanced and suppressed interdiffusion (cation self-diffusion) in MBE-grown AlxGa1−xAs–GaAs superlattices are interpreted in view of point-defect-mediated cation diffusion on the basis of the Fermi-energy dependence of point defect. In order to explain the phenomena, crystal-growth methods and surface-localized point defects which is responsible for Fermi-level stabilization are taken into consideration.

As part of a research program devoted to the microstructural characterization of Al2TiO5-based compounds, high resolution electron microscopy (HREM) has been undertaken in order to study the crystallographic arrangement, especially ordering possibilities, of Al and Ti cations in the metallic sublattice of aluminum titanate. It is seen that adequate experimental conditions, mainly defocus setting, for a resolution of at least 2.5 Å point-to-point, enable the disordered model to be directly and unambiguously proved on 100-oriented micrographs.

Bulk samples of single crystalline and polycrystalline alpha alumina have been neutron-irradiated in the Experimental Breeder Reactor-II (EBR-II) to doses of 1026 n/m2 at temperatures of 925 K and 1100 K. The samples were found to swell macroscopically between 3% and 6%, depending on the temperature of irradiation and the form of the material. The damaged microstructures were investigated via transmission electron microscopy in order to understand the origin of the macroscopic swelling. In both single crystals and polycrystals the damage consists of a high density of dislocations containing predominately b = 1/3<101> dislocation loops on the (0001) planes coexistent with a high density of voids, which are aligned along the c-axis in this rhombohedral material. The established theory of void formation in metals is utilized to explain the formation of voids in alumina. The polycrystalline samples were extensively microcracked, and this is thought to be due to anisotropic swelling of the grains which in turn leads to stresses and fracturing at the grain boundaries.

Solid-state reactions between bulk samples of copper oxide and alumina have been studied using scanning electron microscopy and electron microprobe analysis. Both CuAl2O4 and CuAlO2 were found to form during reactions in air at 1100 °C between CuO powder and single-crystal alumina substrates. The relative position of the CuAl2O4 and CuAlO2 layers was observed to depend on the crystallographic orientation of the surface of the alumina substrate: CuAl2O4 formed in contact with (0001) alumina substrates while CuAlO2 formed when the alumina substrate surface was (110). Faceted Cu–aluminate/alumina phase boundaries were observed to develop when single-crystal alumina rods were reacted with CuO, although the interfaces invariably tended to be wavy.

Melt-grown crystals of HgS and HgS : Co2+ were used to measure their optical absorption spectra. The optical energy band gaps of these crystals were 2.030 eV and 1.870 eV at room temperature, respectively. The impurity optical absorption peaks of Co2+ were observed at 4030, 5988, 12285, 12672, and 12905 cm−1. These peaks can be attributed to the electronic transitions between the split energy levels of Co2+ ion located at the Td symmetry site, where the crystal field, Racah, and spin-orbit coupling parameters were given by Dq = 403, B = 427, and Λ = −155 cm−1, respectively.

Reaction sintering in the Na2O–MgO–Al2O3 system was studied with α–alumina green compacts containing Na2O and MgO in the pores of a controlled structure to produce β, β″–alumina polycrystalline at low temperatures. HCOONa and (CH3COO)2Mg were infiltrated into the alumina powder compacts (average particle size 0.2 μm) with relative densities of 51 to 66% at the compositions of Na2O/MgO/Al2O3 = 1/0.10/4.0–7.6. The alumina powder compacts were formed by filtration of the aqueous colloidal suspensions through gypsum molds. Phase change from α to β, β″–alumina proceeded fast at low temperatures in the samples with high Na2O/Al2O3 ratios. However, no significant densification occurred below 1500 °C owing to the volume increase of powder compacts associated with the phase change of α to β, β″–alumina. At 1600 °C, the green compacts were rapidly densified to relative densities above 99%. This densification was related to the liquid phase sintering based on the partial decomposition of β″–alumina. The dense β, β″–alumina polycrystalline consisted of needle-like grains of 2–5 μm in length and 1–2 μm in width.

Solid state reactions between SiC ceramics and Co, Ni, and Pt metals have been studied at temperatures between 800 and 1200 °C for various times under He or vacuum conditions. Reactions between the metals and SiC were extensive above 900 °C. Various metal silicides and carbon precipitates were formed in layered reaction zones. Interfacial melting was also observed at certain temperatures; teardrop-shaped reaction zones, porosity, and dendritic microstructure resulting from melting/solidification were evident. The metal/ceramic interfaces exhibited either planar or nonplanar morphologies, depending upon the nature of the metal/ceramic reactions. Concave interfacial contours were observed when interfacial melting occurred. By contrast, planar interfaces were observed in the absence of interfacial melting. In all cases, the decomposition of SiC was sluggish and may serve as a rate limiting step for metal/ceramic reactions. Free unreacted carbon precipitates were formed in all the reaction zones and the precipitation behavior was dependent upon the metal system as well as the location with respect to the SiC reaction interface. Modulated carbon bands, randomly scattered carbon precipitates, and/or carbon-denuded bands were formed in many of the reaction zones, and the carbon existed in a mixed state containing both amorphous and graphitic forms.

Cross-sectional TEM studies of ion implantation induced amorphization in a large number of semiconductors have been performed. Samples of Si, AlAs, GaAs, GaP, GaSb, InP, InAs, and ZnSe were simultaneously implanted at 77 K with 20 keV Si+ at doses between 1 × 1014/cm2 and 1 × 1016/cm2. A dose of 1 × 1015/cm2 minimized the ion beam induced epitaxial crystallization and sputtering effects. The depth of the amorphous layer at this dose was compared with Monte Carlo damage density distribution calculations (TRIM'90). The threshold damage density (TDD) necessary for amorphization was determined for each compound. The values of the threshold damage density vary from as low as 2.4 × 1019 keV/cm3 for InAs up to 7.3 × 1020 keV/cm3 for AlAs. ZnSe never became amorphous and GaSb exhibited an unusual disordering after the highest dose. The values of the threshold damage density for the various compositions were compared with known thermochemical data and several bond energy estimates. No single calculation explained all of the trends observed.

Mg–Ti–spinel formation along the interface of epitaxial TiN(100) films to MgO(100) substrates has recently been investigated by transmission electron microscopy (TEM) in the diffraction-contrast mode in samples grown at substrate temperatures higher than 800 °C and in such post-annealed at 850 °C. This phenomenon has now been investigated by high resolution electron microscopy of cross-sectional samples, at an acceleration voltage of 300 kV. Emphasis is given to the TiN/spinel and the spinel/MgO interfaces with respect to their structure and morphology. The results obtained confirm the previously drawn conclusions on the atomic mechanism of the solid state reaction during the spinel-forming process: The spinel, which most likely is of the composition Mg2TiO4, forms by counterdiffusion of the cations Ti4+ and Mg2+ in the rigid oxygen frame provided by the fcc oxygen sublattice of MgO. The latter is completely taken over by the spinel lattice. This “host” character of the MgO substrate lattice for the topotaxial growth of the spinel lattice and the coherency of the solid state reaction with respect to the lattices of all the phases involved are demonstrated. Misfit dislocations at the TiN/MgO, TiN/spinel, and the spinel/MgO interfaces, as well as antiphase boundaries of the cation sublattice of the spinel phase, have also been observed.