Cobalt disilicide CoSi2 of a specific resistivity of 23 μω was formed by the solid phase reaction of cobalt and silicon in the phase sequence of Co2Si, CoSi, and CoSi2 by use of rapid thermal annealing. The through-metal arsenic implantation caused the mixing of cobalt with the silicon substrate and the formation of cobalt silicides. A significant lateral growth of cobalt silicides was observed in samples subjected to one-step rapid thermal annealing process at 900 °C without through-metal ion implantation. Ion beam mixing reduced this lateral silicide growth efficiently, but resulted in a higher density of cobalt atoms remaining in the silicon oxide film than after rapid thermal annealing, as revealed by vapor phase decomposition atomic absorption spectroscopy.

Elastic plane strain Green's function is calculated for an anisotropic bimaterial composite solid containing a free surface normal to the interface. An exact integral representation is obtained for the Green's function, which is evaluated numerically. The integral is also evaluated analytically, which gives a series representation for the Green's function. The singularities in the stress field associated with the presence of the free surface are identified and discussed. These singularities can be of the type r-δ, ln(r) as well as higher powers of ln(r), where δ is between 0 and 1 and r is the radial distance from the intersection of the free surface and the interface. The stress field may also contain an oscillatory factor of the type exp[ιg ln (r)] where g depends upon the material parameters of the two solids. For illustration, the formalism is applied to a cubic solid containing a Σ-5 grain boundary.

The Embedded Atom Method (EAM) is used to compute density, internal energy, and structure factor for bulk liquids of the fcc metals at several temperatures above and below the melting temperature. The calculated values are found to be in generally good agreement with experiment, although the volume expansion upon melting does differ by up to 50% from the expected result for some of the elements studied. The total energy of a liquid system with surfaces is calculated, and the results are compared with the bulk liquid results to determine the enthalpy and thickness of the liquid-vapor interface. Also, the surface tension is found for Cu near the melting temperature. The EAM values for surface enthalpy and surface tension are found to be smaller than experimental values, which is consistent with results for EAM calculations of the surface energy of crystalline solids.

A new diffraction imaging technique for the characterization of polycrystalline materials is proposed and applied to obtain direct information about individual grains and their size and shape distributions and, in turn, strains in these materials. Unlike traditional powder diffractometry, where divergent and focusing x-ray optics are essential to collect information from an ensemble of grains, the nearly parallel and monochromatic beam available from a synchrotron x-ray source is employed to observe and measure diffraction images from individual grains and component particles in consolidated materials prepared by various processes. Images can be recorded by traditional methods, such as film and pulse counting detectors, but modern image detectors, such as charge coupled device (CCD) detectors and image analyzers, make the proposed imaging technique more practical. Unlike traditional diffractometry, this new technique provides the ability to measure shape, size, and strain without model based analyses. The spatial distribution of strain within individual grains, displayed as a diffraction image (topograph), indicates the presence of defects, such as dislocations, subgrain boundaries, and precipitates, and sheds new light on the origins of residual strains (stresses) in industrial materials. The resolution of the imaging system used is limited to grains ∼10 μm or larger due to diffraction broadening (∼20” from the size effect) and the resolution of the recording medium.

The microstructure of the ordered intermetallic alloy with a nominal composition of Al66Fe9Ti24 is nearly single-phase L12 structure, with a few second phase agglomerates at some grain corners. Room temperature compression tests showed that this material exhibits a plastic strain of about 11% at fracture. Final fracture of the compression specimens occurred by a shear-off process along a surface oriented about 45 degrees to the compression axis. Fractographic analysis revealed that the fracture is transcrystalline and the fracture mode is mainly quasicleavage plus tearing. Transmission electron microscopy (TEM) was used to explore its deformation mechanisms. The dislocation density was low after homogenization, but is greatly increased during deformation. The deformation mode was found to be 〈110〉 {111} slip instead of twinning as in Al3Ti. The a〈110〉 superdislocations dissociated into two partials of a/3〈211〉-type, bounding a superlattice intrinsic stacking fault (SISF) on the {111} slip plane.

Implantation of Ta into single crystal Ni3Al was conducted to determine the degree of surface hardening in monolithic alloys in relation to its lattice location. Ta was implanted at 400 keV to doses of 0.07, 0.36, and 2.52 × 1016 cm−2 along the [100] axis of a [100] crystal of Ni3Al at room temperature. Composition versus depth profiles were determined by RBS, and lattice location of Ta was determined by channeling angular yield scans about the [100] axis. The hardness of the surface was measured by ultra-low load indentation. Results show that implantation softens the surface and that the Ta is randomly distributed between Ni and Al sites. Annealing at 1000 °C/1 h significantly reduces the damage and causes preferential occupation of Al sites by Ta, resulting in a slight increase in surface hardness. Further annealing at 1200 °C/0.25 h increases the surface hardness substantially and increases occupation of Al lattice sites to roughly 84%. Results are consistent with a model in which the as-implanted surface is softened by disordering, and subsequent diffusion of Ta to Al sites during thermal treatment causes hardening of the surface.

MoSi2-based composites reinforced with particles, whiskers, and continuous fibers were fabricated using hot pressing and hot isostatic pressing techniques. The microstructure, interface compatability, and interfacial properties between the reinforcements and matrix are discussed. The microstructural parameters which control the mechanical behavior of the MoSi2 composites were characterized. The need for developing a satisfactory reinforced MoSi2 composite for high-temperature structural applications is also addressed.

We have studied selective epitaxial growth of silicon in the SiCl2H2/HCl/H2 system. Our results on growth rate and selectivity are reported here. We found that the growth rate and selectivity are strong functions of the partial pressures of HCl, SiCl2H2, H2, exposed Si area, and wafer position. We propose that growth and selectivity are controlled by surface reactions between SiCl2H2 and HCl and the SiO2 and Si surfaces. Higher HCl flow rates decrease the adsorption coefficients and/or incorporation rates of the reactive SiCl2H2(g) species on both SiO2 and Si. When either is zero for SiO2 and both are greater than zero on Si, selective depositions occur. Finally, we present data on the effect various SiO2 surface treatments have on the density and distribution of Si nuclei on this surface.

This work presents the first experimental evidence of the formation of a fractal network of crystalline clusters in the first stage of the amorphous-crystal transformation. Although this transformation has been extensively studied, such an intermediate stage between amorphous and crystalline phases has never been experimentally revealed before. This fractal network was obtained by laser irradiation of an amorphous Te–Se–Br alloy. The irradiation conditions have been determined in order to be in a regime of limited diffusion, which is the basis of fractal formation. Moreover, the fractal dimension has been determined to be 1.55, which corresponds to the theoretical value obtained for fractal growth by a process of cluster-cluster aggregation with some structural readjustment.

Analytical results obtained for the transient kinetics of nucleation enable one to interpret N(gd,t), the accumulated number concentration of clusters at the instrumentally detectable size, gd. The new results enable one to extract kinetic and thermodynamic parameters of nucleation from experimentally measured cluster concentrations and to test nucleation theories experimentally. An approach to estimate the mean time to form the first nucleated cluster in a given sample is also presented.

An alloy of composition (Co78Fe22)3V, which orders to an L12 superlattice below a critical temperature (Tc) of 910 °C, was rolled to 25–50% reduction in the initially ordered condition and annealed at various temperatures above and below Tc and examined by hardness, tensile testing, optical and electron microscopy and dilatometry, in order to study the progress of recovery and recrystallization. Recrystallization was severely retarded on annealing below Tc; close to Tc, recrystallization was ≈ 300 × slower in the ordered than the disordered range. Although recrystallization started promptly, predominantly at grain boundaries, very rapid recovery-softening of the unrecrystallized regions progressively reduced the driving force for recrystallization and slowed it down drastically. However, at 770°and 500 °C, recovery-softening was replaced by some recovery-hardening (i.e., strain-age hardening). Above Tc, recrystallization was complete in a few seconds and a special annealing method was needed to measure such times accurately. Dilatometric measurements showed that most of the order destroyed by rolling was restored long before recrystallization began, but the restoration was never complete unless the alloy was heated up through Tc and then slow cooled. Electron microscopy showed no sign of any antiphase domains in recrystallized grains except for a few isolated domain boundaries on annealing at 770 °C. A model is proposed to rationalize the incidence of recovery-softening or strain-age hardening at different annealing temperatures.

Molybdenum nanocrystals (4 nm < particle size < 12 nm) were synthesized in the vapor by sputtering in a thermal gradient at argon pressures between 0.2 and 0.6 Torr. The gradient was achieved by cooling the substrate table with liquid nitrogen in order to enhance the deposition rate of the nanocrystals. The size of these nanocrystals depended on the sputtering argon gas pressure. Using appropriate deposition parameters, these nanocrystals were incorporated, either as a dispersed phase in a major phase produced by normal sputtering or as a layer in a layered structure to produce nanocomposite films. Results of Auger electron spectroscopy (AES) study of the Al/Mo layered nanocomposite films are presented. The Knoop microhardness of these films was increased by as much as a factor of four compared with the hardness of homogeneous Al films. A correlation of the microhardness to the microstructures as revealed by transmission electron microscopy (TEM) is discussed.

The reaction of SiH4/H2 mixtures with iron, cobalt, nickel, copper, chromium, molybdenum, and tungsten, at temperatures between 350 and 800 °C and 1 atm of total pressure, was studied. When the amount of water vapor in the gas mixture is carefully controlled, a metal silicide diffusion coating forms at the appropriate treatment temperature. Cu silicide coatings form at temperatures as low as 350 °C. Metal silicide coatings for Fe, Ni, Co, and Cr form at intermediate temperatures (500–700 °C), and higher temperatures (above 700 °C) are required for W and Mo. Composition and structure of the metal silicide coatings were determined by Auger depth profiling and x-ray diffraction. Kinetics of the surface reaction between SiH4 and the metal substrate as well as the behavior of these coatings in oxidizing environments at high temperatures were studied by a microgravimetric technique. The metal silicide coatings provide oxidation protection for Fe, Ni, and Cr in pure O2 up to 1000 °C, for W and Mo in air up to 900 °C, and for Cu exposed to air up to 700 °C.

Atomistic computer simulations are used to systematically investigate the role of interfacial disorder on the elastic behavior of composition-modulated superlattices of fcc metals, represented by simple Lennard–Jones potentials. The structures, energies, and average elastic properties of four types of superlattices with various degrees of interfacial disorder are computed as a function of the modulation wavelength along [001]. The four superlattice types studied include perfectly coherent, incoherent, and two types derived from these by introducing relative twists about [001] between alternating layers. A 20% lattice-parameter mismatch between the two modulating materials is assumed. Results are compared with our earlier work on unsupported thin films, grain-boundary superlattices, and incoherent superlattices with a 10% lattice-parameter mismatch. The degree of structural disorder at the interfaces is found to correlate well with the magnitude of the elastic anomalies, which cannot be accounted for by anisotropic lattice-parameter changes alone. The grain-boundary superlattices studied earlier are found to provide a good model limit for the elastic behavior of interfacially disordered dissimilar-material superlattices.

The microstructural changes and the tensile behavior of laser processed A356-Al alloy matrix composites reinforced with 10 and 20 vol.% SiC particulates are characterized. The autogenous bead-on-plate welds were made using a pulsed CO2 laser operating at a peak power level of 3.2 kW. The pulse on-time was constant at 20 ms and the off-time was varied from 20 to 2 ms (duty cycles of 50–91%). The microstructure of the laser melted region was investigated by optical, scanning, and transmission electron microscopy, and x-ray microchemical analysis techniques. The extent of microstructural changes varied directly with duty cycle, i.e., being a maximum for the longest (91%) duty cycles. Pulsed laser processing produced partial to complete dissolution of SiC particles and sometimes resulted in the formation of aluminum carbide. The associated rapid cooling also produced a fine distribution of nonequilibrium complex precipitates. In addition, the laser energy modified the SiC surface both physically and chemically. The results of tensile tests indicated that the modified SiC and the distribution of fine nonequilibrium precipitates enhance the mechanical properties of the laser processed composites. Optimum changes in microstructure and mechanical properties were obtained in the composites processed with intermediate (67 and 74%) duty cycles; therefore pulsed processing appears to be a strong candidate for successful joining of these MMCs.

The atomic structure and roughness on the surface of a carbon fiber have a great effect on the degree of bonding of that fiber in a carbon fiber composite. Although there have been many studies on the bulk structure of these fibers, this is the first study dealing with the atomic surface structure of several carbon fibers. With the advent of the scanning tunneling microscope (STM), it is now possible to study both the roughness and structure of these fibers on the atomic scale. Type II PAN based fibers were found to have a rougher surface than type II pitch-based fibers. Similar to what has been observed in the interior of pitch fibers, the percentage of graphitic structure on the surface increased with the degree of heat treatment and with the modulus of the fiber.

Analytical expressions for the time-dependent crystallized volume fraction are derived from new results for the transient rate of nucleation reported in Part I. Conventional formulations that have been used in interpreting crystallization experimental data and for assessing the stability of amorphous phases are shown to be large time limits of the newly derived expressions. An approach for assessing the stability of an amorphous phase is proposed.

We report x-ray diffraction experiments investigating the lattice expansion due to hydrogen in Nb/Ta superlattices and thin Nb films. For low hydrogen concentrations (cH < 0.09 H/metal), the lattice expands only normal to the plane of the film due to the epitaxial constraints imposed by the substrate. At higher concentrations, corresponding to strains on the order of the critical strains expected for the formation of misfit dislocations, we find a structural instability which leads to the formation of two metal-H phases, one of which has a lattice expansion in the plane of the film. These results differ significantly from the expected phase diagram and properties of the bulk metal-H systems.

Amorphous iron and titanium-based alloys containing various amounts of chromium, phosphorus, and boron exhibit high corrosion resistance. We report some physical properties of Fe and Ti-based metallic alloy films deposited on a glass substrate by a dc-magnetron sputtering technique. The films were characterized using differential scanning calorimetry (DSC), stress analysis, scanning electron microscopy (SEM), x-ray diffraction (XRD), secondary ion mass spectrometry (SIMS), electron microprobe, and potentiodynamic polarization technique.

Thin poly-Si layers deposited at 625 °C by LPCVD that are used in silicon technology for microelectronics exhibit a pronounced additional x-ray diffraction peak at about 0.334 nm. High-resolution electron microscopy (HREM) reveals that this peak stems from {010} reflections of a diamond hexagonal (dh) Si phase, which occurs as small inclusions with the orientation relationship (01) ‖ (0001), [011] ‖ [20] to the diamond cubic (dc) Si matrix. Due to the high density of planar faults on {111}, the dh-Si phase also exists in the form of the 2H silicon polytype with the orientation relationship (1) ‖ (0001), [011] ‖ [20]. In the first case the formation of the dh-Si phase may be understood by a multiple twinning transformation process, and in the second case by glide of Shockley partial dislocations on {111} planes. Various other hexagonal polytypes occur, which have all the {010} reflections in common and make a major contribution to the 0.334 nm peak. The medium temperature of 625 °C for layer deposition leads to a 〈011〉 preferential orientation and a high density of twins as well as to high compressive stress in the poly-Si layer itself. This seems to promote the formation of dh-Si. The strong twinning behavior produces a typical tilt grain boundary between adjacent dh-Si grains: [20], (016), Θ = 35°with a translation vector t = 1/2[031] parallel to it. The dh-Si phase vanishes in this poly-Si film after annealing at temperatures above 1000 °C due to grain growth by recrystallization.