Diamond crystallites and continuous films were deposited on (100) silicon with various surface treatments by microwave plasma assisted CVD at times varying from 2 min to 1600 min. In each experiment, the average diameter of the crystallites increased linearly with time, while the density of crystallites was essentially constant. Thus, nucleation of the diamond occurred within a short time interval early in the deposition process. After the nucleation event, only growth occurred. Various surface treatments were used: untreated, polished with 1 μm diamond, scratched with 350 mesh SiC, scratched with 1 μm alumina, wiped with 350 mesh graphite powder, and spin coated with polymethyl methacrylate. Only the diamond polishing affected the crystallite density, and none of the surface treatments had any effect on crystallite morphology or growth rate. Growth rates were determined by least squares fits to average diameter versus time for crystals and average thickness versus time for films. The growth rate data extrapolate to zero size at zero deposition time. Applying the Volmer–Weber model, an activation energy for nucleation of diamond on silicon was calculated to be 52 kcal/mole.

NiZr powders produced by mechanical alloying become active NO decomposition catalysts after an activation period in reaction conditions. Although the initially glassy structure exhibits a high activity, the NiZr powder becomes temporarily inactive. After about 10 hours in reaction conditions at 673 K,the powder again becomes active with a lower,but still substantial turnover frequency for NO decomposition Oxygen produced by the reaction appears to be consumed by the catalyst during this second period of activity.

The nature of an amorphous solid is difficult to characterize, simply due to the lack of long range order, the presence of which allows crystalline surfaces to be efficiently characterized by the surface unit cell. Two recent model s of an amc1rphous surface [1, 2] have both portrayed a locally rough glassy surface. Garofalini’s pure metal element, one dimensional model yielded a cross sectional cut across an amorphous surface that would be traced by a surface diffusing atom [1]. Kowbel and Brower [2] utilized Gaskel’s trigonal prism morphology for the metal-metalloid chemical short range ordering in the bulk glass structure [3] to produce a two dimensional model and surface simulation of Pd80Si20 glass [2] and Fe80–B20 glass [4]. Both models predict an atomically rough glassy surface both normal to the surface and within the plane of the surface. The simulation of the surface of an Fe80B20 glass as generated by transsection lines across the hard sphere model is shown in Fig. 1. The fractal dimension was calculated to be 2.3, indicating a fine scale of roughness, and the average surface coordination number is 6, also indicating a locally rough surface as viewed by an adsorbant molecule. Such a scale of roughness on an Fe80B20 glass appears to be confirmed by the ion scattering experiments of Overbury et al [5].

Alloys of NiAu and RhCu were analyzed to determine the degree of extended solid solubility of the rapidly solidified structures. The rapidly solidified alloy of Ni58Au42 forms a one-phase extended solid solution, existing as featureless grains. During isothermal aging at 600 °C, the rapidly solidified NiAu alloy transformed into a two-phase structure of nickel-rich and gold-rich phases. The NiAu rapidly solified alloy was stable up to 350 °C under continuous heating at 10 °C min−1. The rapidly solidified alloy of Rh58Cu42 did not form a complete extended solid solution, but the rapid solidification did increase the solubility limits beyond the equilibrium phase boundaries. The micro-structure appears as refined dendrites with less micro-segregation. During isothermal aging at 900 °C, the rapidly solidified RhCu alloy undergoes increased segregation to rhodium-rich and copper-rich phases. The rapidly solidified RhCu alloy was stable up to 375 °C under continuous heating at 10 °C min−1. X-ray and electron diffraction and differential scanning calorimetry techniques were utilized to determine the microstructure, crystal structure and phase stability.

Conversion Electron Mossbuaer Spectroscopy, CEMS, along with conventional transmission Mossbauer spectroscopy is used to determine near surface crystallization behavior of Fe 80 B 20 (MG 2605) metallic glass. Single roller quenched samples were heat treated isothermally under inert atmosphere and in vacuum. The relative amounts of residual amorphous phase and crystallization products are determined. As predicted by Newtonian splat cooling conditions, no preferred crystallization is observed at the top or bottom of the sample as compared to the bulk during the course of the isothermal crystallization. The predominant crystalline product in the as quenched glass is Fe 3 B, indicating such short range ordering in the glassy state. However, upon isothermal aging, substantial amounts of α-Fe crystallize both near surface and in the bulk.

Pd/Si glasses, made by splat cooling, are active hydrogenation catalyst systems (catglas). During the hydrogenation with deuterium of cis-cyclododecene they produce more trans isomerization, more dideutero-saturate and less extensive exchange than crystalline Pd. Moreover, certain partially crystallized splat cooled glasses exhibit enhanced catalytic activity.

The reactions which occur during freezing in biological systems employing DMSO as a cryoprotective agent may well involve information given by a near equilibrium ternary H2O-DMSO-NaCl phase diagram. The initial freezing point depressions for solutions with three different DMSO-NaCl initial ratios (R) have been determined over the onefold surface of ice saturation. DMSO has been shown to be more effective in reducing NaCl concentration in the residual liquid than had been previously predicted. The temperature and the fraction solid which must be reached for the occurrence of second phase coprecipitation with ice have been shown to be a strong function of initial R value. Ternary invariant reactions have been identified at −35 °C, and tentatively identified at −115 and −105 °C for solutions having DMSO/NaCl ratios of R = 9, 5, and 1, respectively. Metastable nonequilibrium phase formation has been observed for slow cooling of a solution with R = 1. This metastable condition results in different phase relationships upon thawing than upon the initial freezing. By quenching the system after partial rewarming, it has been demonstrated that this metastable condition can be eliminated.