The dependence of the growth velocity on crystal orientation has been studied during crystallization of thin polycrystalline silicon films from the melt. Two types of growth velocity anisotropy have been observed. In the first, competitive growth between (100) textured seeds and seeds with (110) and (111) textures indicates that the relative growth velocities are ν(100) > ν(110)ν(111). It is postulated that this textural growth velocity anisotropy is a result of the differences in the interal energy of grains with different textures. This assumption, combined with the data, yields estimates of the interfacial energy anisotropy for the Si–SiO2 interface: γ(111) − γ(100) = 0.069 J/m2 and γ(110) − γ(100) = 0.012 J/m2. Another type of growth velocity anisotropy is responsible for the development of in-plane orientation in competitive growth between (100)-textured seeds. Simple models, which describe development of these two types of crystallographic orientation via anisotropies in growth velocity, agree well with the experimental results.

A type of solidified “pure” C–S–H (calcium silicate hydrate) has been prepared by hydrating mixtures of synthesized Ca3SiO5(C3S) and silica fume at very low water/solid (w/s) ratios (< 0.15). The mechanical properties (splitting tensile strength), stoichiometry, and microstructure of hydration products have been investigated. Results are reported here in which 12.03%−41.56% (by weight) silica fume was added to C3S. It has been found that the strength of the materials with CaO/SiO2(C/S) molar ratio of 1.5 is comparatively higher than that of the others, after curing from 1 day to 28 days. The homogeneity and rheology are of direct importance to the formation of the calcium silicate bonding, which influences greatly the mechanical properties. The hydrates are primarily C–S–H with only small amounts of Ca(OH)2. The compositions of the solidified C–S–H are (1.33–1.37) CaO·SiO2. (1.02–1.21) H2O and (1.40–1.19)CaO·SiO2.(0.68–0.91)H2O for mixtures of C/S = 1.5 and 0.83, respectively. The characterization of the chemically bonded materials by Wn (nonevaporable water), density, x-ray diffraction, and scanning electron microscopy (SEM) studies has been carried out. The relationships between these characteristics and the strength are discussed.

The pH of the solution in contact with a hydrous zirconium oxide plays a dominant role in determining the crystal phase, tetragonal or monoclinic, in the calcined material. The substitution of low concentrations of hafnium for zirconium destabilizes the tetragonal phase so that only the monoclinic phase is formed; the amount of Hf required for destabilization depends upon the pH used for the preparation of the hydrous oxide. While this study has defined a phenomenon, the results do not permit a definition of the mechanism for it.

High-resolution nuclear magnetic resonance (NMR) spectra and synchrotron x-ray powder diffraction data have been obtained from a well-crystallized highly dealuminated sample of the zeolite ZSM-11. The Rietveld profile technique has been applied to the synchrotron data to give the first detailed refinement of the idealized structure derived ten years ago by distance least-squares modeling methods [G. T. Kokotailo, P. Chu, S. L. Lawton, and W. M. Meier, Nature 275, 119 (1978)], which involves 54 variable atomic positional parameters. The structure is tetragonal (a = 20.065 Å, c = 13.408 Å at 25 °C) and consistent with the previously reported tetragonal space group I \overline 4 m2, but the NMR spectra indicate local deviations from this symmetry that disappear at 100 °C.

Six MeV high-dose Ni implantation into silicon has been applied to synthesize deep-buried metallic layers. These layers have been analyzed by optical reflectivity and spreading resistance depth profiling as well as transmission electron microscopy and cross-section transmission electron microscopy. Already in the as-implanted state, at target temperatures of 450 K and doses above 1017 Ni/cm2, epitaxial precipitates of NiSi2 are formed. They grow in type-A and type-B orientations. In addition to these polyhedral crystallites, thin NiSi2 platelets on {111} lattice planes exist. At a dose of 1.3 × 1018 Ni/cm2, a continuous but highly defective layer of epitaxial NiSi2 is formed by coalescence of mainly type-A precipitates at the maximum of the Ni profile. Investigations indicate that damage gettering of nickel atoms as well as the atomic density increase during implantation influence the depth distribution of implanted metal atoms. Moreover, a suppression of silicon amorphization by nickel is evident.

Electrohydrodynamic atomization'has been adapted to produce Al2O3 powders ranging in size from 10 nm to 300 μm. Microstructural characterization using x-ray diffraction, scanning, and transmission electron microscopy reveals changes in phase selection as a function of particle size, hence supercooling. Amorphous powders are common below 100 nm in diameter. Cubic spinel γ is found in single phase form between 100 nm and 2μm, and partially transformed to δ between 2 and 20 μm. There is also evidence of σ and θ or a precursor of θ forming directly from the liquid above 5 μm. The stable corundum structure is consistently found above 20 μm but exhibits three different morphologies: faceted, dendritic, and cellular. Phase selection is examined on the basis of fundamental thermodynamic and kinetic considerations and results from computer models predicting the thermal history of the powders. It is concluded that metastable phases require the elimination of catalytic sites for the nucleation of a and are thus more likely to form in the smaller powders. Furthermore, submicron powders achieve sufficiently high cooling rates to preserve the metastable phases formed (γ), but those higher than ∼ 1 μm experience a thermal excursion long enough to transform γ to more stable forms of Al2O3.

Two quite different types of microstructures have been detected for MgTiO3-enriched LiTaO3 ceramics corresponding either to small size grains and open porosity or to large size grains with closed porosity and appearance of microcracks. The first structure appears in short sintering processes at relatively low temperature. The textural change is explained by transitory formation of a grain boundary phase that leads to quick coalescence of the small grains.

Zirconium tetrafluoride (ZrF4) is the major constituent of many heavy metal fluoride glasses of technological interest. Five samples of commercially available “zirconium fluoride” were analyzed to determine their exact nature. Two samples were found to be anhydrous β-ZrF4, one was found to be α-ZrF4, one was zirconium tetrafluoride monohydrate (ZrF4 · H2O), and the final sample was identified as a hydrated amorphous phase of ZrF4. The behavior of these samples on heating was investigated with particular interest in the removal of water and the loss of fluorine from the hydrated materials. Infrared spectroscopy, x-ray diffractometry, thermogravimetric analysis, and differential scanning calorimetry were used in this investigation.

Precipitated hydrous zirconium oxide can be calcined to produce either a monoclinic or tetragonal product. It has been observed that the time taken to attain the final pH of the solution in contact with the precipitate plays a dominant role in determining the crystal structure of the zirconium oxide after calcination at 500 °C. The dependence of crystal structure on the rate of precipitation is observed only in the pH range 7–11. Rapid precipitation in this pH range yields predominately monoclinic zirconia, whereas slow (8 h) precipitation produces the tetragonal phase. At pH of approximately 13.0, only the tetragonal phase is formed from both slowly and rapidly precipitated hydrous oxide. The present results, together with earlier results, show that both the pH of the supernatant liquid and the time taken to attain this pH play dominant roles in determining the crystal structure of zirconia that is formed after calcination of the hydrous oxide. The factors that determine the crystal phase are therefore imparted in a mechanism of precipitation that depends upon the pH, and it is inferred that it is the hydroxyl concentration that is the dominant factor.

The specific heat of liquid and solid phases and the heats of crystallization and fusion have been measured by differential scanning calorimetry (DSC) for a series of Te80Ge20 − x Pbx alloys (0≤x≤20). The enthalpy, entropy, and free energy of the undercooled liquid are quantitatively assessed with reference to the crystal phases. The available formulas for computing the free energy of the liquid are compared, and their relative merits are discussed. The glass transition temperature is shown to depend strongly on the ratio of the average excess specific heat of the liquid to the entropy of fusion. An anomaly in the liquid specific heat, which is particularly important for Te80Ge20 and Te80Ge15Pb5, leads to very good glass forming ability for these alloys; this is demonstrated by preparing amorphous samples by means of fluxing.

The formation of an insoluble gel under ion-beam bombardment is governed by ion-beam parameters and target parameters. Here reported is a study of the influence of the target molecular parameters on the sol–gel transition of ion-bombarded polystyrene with particular emphasis for the number-average molecular weight $\overline M$n. It is shown that the main parameter is the number of macromolecules of the film so that by adopting a “corrected” fluence F/n (ions per macromolecule), the different curves of the various polymers collapse in only one universal curve. The importance of the “corrected” fluence is shown also at molecular level and the MWD of the various polymers is similar at equal F/n values. An experimental model is outlined which explains the sol–gel transition on the basis of transition from an isolated-track regime to an overlap regime where the formation of insoluble giant macromolecules occurs.

In situ Raman spectra were taken, between 800 and 1200 cm−1, on films of polypyrrole during both the electrochemical growth and the reduction reactions. During the reduction reaction we observed a frequency shift of approximately 10 cm−1 of two in-plane vibrational modes. We attribute these shifts to the bond conjugational defects induced by doping and undoping of the polypyrrole films. A quantitative model for this phenomenon based on normal mode calculations is presented. The electrical conductivity and thermopower of polypyrrole doped with tetrafluoroborate and iron chloride ions is also reported for temperatures between 10 and 350K.

The substitution of fluorine for oxygen in TiO2 was investigated by the reaction of Ti2O3, TiO2, and TiF3 under conditions of 4–6.5 GPa and 700–1400°C. The single phase of TiO2−x Fx solid solution was obtained in the region of 0≤x≤0.7. According to the x-ray diffraction data, the a and c axes of the rutile-type structure linearly increased with increasing fluorine content. The electrical resistivities of TiO2−x Fx were in the range from 10 Ω cm for x = 0.3 to 850 Ω cm for x = 0.7 at 300 K and the relationship between In ρ and 1000/T was linear. The activation energies were estimated to be from 0.17 eV at x = 0.3 to 0.28 eV at x = 0.7. Also, the thermoelectric powers at room temperature changed from 250μV/K to + 50 μV/K. The mechanism of electric conduction was discussed on the basis of the extended band model of rutile.

The effects of gamma irradiation on groundwater and the reaction between groundwater and glass have been investigated at radiation exposure rates of 2 × 105 1 × 103 and 0 R/h. These experiments, which bound the conditions that may occur in a high-level nuclear waste repository located in tuff, have been performed using the actinide-containing glasses ATM-lc and ATM-8, and have been performed for time periods up to 278 days. The experimental results indicate that when only the repository groundwater is present, the pH of the system remains near-neutral, regardless of the radiation field, due to the buffering capacity of the solution. When glass is added to the system, the subsequent reaction is governed by the solution chemistry, which results from a complex interaction between radiolysis products, glass reaction products, and groundwater components. While no long-term reaction trends have been extracted from the current data, it is noted that there are no outstanding differences in the reaction of the glasses as measured by the release of the soluble components B, Mo, and Na, as a function of radiation exposure rate. However, there is a marked difference in the amount of U, Np, and Pu released from the glasses as a function of radiation exposure rate. This difference can be correlated with the pH values of the leachate, with more basic solutions resulting in lower actinide release.

The crystallizing phases from a BaZn YbTh fluoride glass were a BaYbTh fluoride, ZnF2, and YbF3, as identified with x-ray diffraction and composition-enhanced scanning electron microscopy. Blocky BaYbTh fluoride crystallizes first, at about 450 °C, and ZnF2 excluded from this phase crystallizes at its surfaces. At higher temperatures the BaYbTh fluoride phase decomposes partially to BaThF6 and YbF3 phases. The kinetics and temperatures of crystallization were also followed with differential scanning calorimetry.

New data on the production of electrical conductivity in poly (phenylene sulfide), PPS, by ion irradiation are presented. These and previously reported PPS data are investigated in the framework of a theoretical semiempirical model that relates observed conductivity to parameters associated with the deposition of energy in the polymer by the bombarding ions. It is shown that the onset of conductivity with increasing ion dose is dependent on overlap of individual ion damage regions. A straight line relationship is obtained between log of the ion stopping power and log of the effective overlap radius of the ion damage regions. Furthermore, the magnitude of the values of effective overlap radii are consistent with physical observations and theoretical predictions. At higher ion doses, the rapid increase in conductivity appears to be consistent with a multistage reaction mechanism for the production of conducting species in the polymer. A universal curve of conductivity as a function of dose above threshold dose is seen to fit all of the PPS data up to a dose of a factor of 50 above threshold. This curve and the predictable behavior of the threshold dose allow the selection of ion/dose combinations to produce a desired conductivity.

Polyimide-metal oxide (Co3O4 or CuO) composite films have been prepared via in situ thermal decomposition of cobalt (II) chloride or bis (trifluoroacetylacetonato) copper (II). A soluble polyimide (XU-218) and its corresponding prepolymer (polyamideacid) were individually employed as the reaction matrix. The resulting composites exhibited a greater metal oxide concentration at the air interface with polyamideacid as the reaction matrix. The water of imidization that is released during the concurrent polyamide acid cure and additive decomposition is believed to promote metal migration and oxide formation. In contrast, XU-218 doped with either HAuCl4 · 3H2O or AgNO3 yields surface gold or silver when themolyzed (300 °C).

A 25 kV pulsed electron beam was used to harden 0.5−3.0μm thick AZ-type, MacDermid, and polyamic acid (PMDA + ODA) resist patterns. The resist profiles are stable against high-temperature treatment that ranges between 200−350 °C. The short pulse ∼ 100 ns, electron beams employed in resist hardening are produced from a cold cathode in 30−50 m Torr air by discharging energy stored in a 7.5 nF capacitor producing a dose/pulse ∼ 1μC/cm2 at the processed surface. Comparisons with conventional hardening methods using ultraviolet emission from a high-pressure mercury lamp, a windowless, vacuum ultraviolet (VUV) lamp, and low-energy electron emission from a cw source are also made.