The energies and configurations of interstitials and vacancies in the ordered compounds CuTi and CuTi2 were determined using atomistic simulation with realistic embedded-atom potentials. The formation energy of an antisite pair was found to be 0.385 and 0.460 eV in CuTi and CuTi2, respectively. In both compounds, the creation of a vacancy by the removal of either a Cu or Ti atom resulted in a vacant Cu site, with an adjacent antisite defect in the case of the Ti vacancy. The vacant Cu site in CuTi was found to be very mobile within two adjacent (001) Cu planes, with a migration energy of 0.19 eV, giving rise to two-dimensional migration. The vacancy migration energy across (001) Ti planes, however, was 1.32 eV, which could be lowered to 0.75 or 0.60 eV if one or two Cu antisite defects were initially present in these planes. In CuTi2, the vacancy migration energy of 0.92 eV along the (001) Cu plane was significantly higher than in CuTi. The effective vacancy formation energies were calculated to be 1.09 eV and 0.90 eV in CuTi and CuTi2, respectively. Interstitials created by inserting either a Cu or Ti atom had complicated configurations in which a Cu 〈111〉 split interstitial was surrounded by two or three Ti antisite defects. The interstitial formation energy was estimated to be 1.7 eV in CuTi and 1.9 eV in CuTi2.

An x-ray diffractogram was generated from a powder sample of solution treated and ice brine quenched Al–14.25 at.% Li alloy. The Bragg reflections obtained were characteristic of a very-nearly fully ordered Al-rich L12 phase based on δ'Al3Li together with two very weak reflections associated with δAlLi. All the lines were significantly broadened due to particle size effects. The average particle size associated with the 100 L12 superlattice line was found to be 4.2 (3) nm and with the fundamental lines, 26.8 nm. A simple structure factor calculation indicated the volume fraction of ordered phase to be around 0.77, assuming that the lack of maximum order was due to the presence of disordered fcc AlLi solid solution. These results suggest that the microstructure of this as-quenched alloy comprises ordered regions of about 4 nm in size in a sea of disordered matrix with a very small amount of δAlLi present. This conclusion is in excellent agreement with recent small angle x-ray and transmission electron microscope studies on similar alloys and suggests that AlLi alloys which are ostensibly disordered at high temperatures go through a disorder-order transformation and then decompose into regions of order and disorder which are associated with a composition spinodal.

The shear strength of materials after extensive plastic flow under a superimposed high hydrostatic stress component is most conveniently studied by means of Bridgman opposed anvils between which thin disk-shaped samples are sheared through relative rotation. A newly designed apparatus of this type permits, for the first time ever, the monitoring of sample thickness during shearing and thus the obtaining of averaged shear stress/shear strain curves. Such relations are needed for the better understanding of geological processes, behavior under shock or explosive impact, and of the surface layers during friction and wear, among others. A semiquantitative analysis shows that nonuniform pressure distribution in the samples cannot significantly falsify the results. It is concluded that distributed shearing is always accompanied by sample thinning and that, conversely, slippage between anvils and specimens is indicated whenever the sample stops thinning during rotation. Such slippage can be greatly reduced by raising the friction coefficient through etching of samples and/or anvils, but it apparently occurred undiscovered in previous studies, partly after initial distributed shearing. Further, in previous results slippage between directly contacting anvils was frequently mistaken for sample shearing. Additionally, the previously neglected sample material being extruded during shearing can falsify results, as can “turbulent flow” initiated at sample perforation. Correspondingly all prior data gained with Bridgman apparatuses are suspect. Present best results indicate (i) that in metals ordinary dislocation glide but apparently with strongly increased Peierls–Nabarro stresses continues to the highest pressures studied, (ii) that independent of thermal activation workhardening may cease at high strains, and (iii) that “turbulent flow” resulting through sample perforation or when the sample thickness decreases below a critical value, can give rise to mechanical alloying. The majority if not all of the data by Bridgman as well as Vereshchagin et al. probably involved such mechanical alloying. When plotted versus pressure in units of shear moduli, the apparent coefficients of friction of the five cubic metals examined so far follow nearly the same curve but they are lower for metals of lower symmetry.

The effect of excess Ca2CuO3 on the superconducting properties in the (Bi, Pb)2Sr2Ca2Cu3Oy, (2223) was investigated through the measurements of ac susceptibility, critical current density, and critical field. All the measurements have been performed on the samples that were identified by x-ray diffraction to consist of the mostly high Tc 2223 phase with varying amounts of excess Ca2CuO3. It was found that when the excess Ca2CuO3 is in the range 0.1 M to 0.3 M the loss peak of ac susceptibility appeared at higher temperature than that for an undoped sample, and the Jc showed a maximum at the excess Ca2CuO3 = 0.2 M, indicating that a proper amount of excess Ca2CuO3 enhanced the flux pinning. The upper critical field, Hc2, determined by using a pulse magnetic field showed a maximum at the excess Ca2CuO3 = 0.1 M, and increased linearly with decreasing temperature within the measurement regime.

Diffusion-induced phase transformations were studied in Ni/β-Ta multilayers between 523 K and 823 K, primarily using x-ray diffraction. The multilayers had a modulation length, A, of 20.3 nm and a composition of Ni48Ta52. They were polycrystalline without coherency between the Ni and Ta sublayers. Upon annealing at relatively low temperatures (up to 723 K) Ta dissolved in crystalline Ni, concurrently with the formation of an amorphous phase. The interdiffusion reactions did not take place only at the Ni/Ta interfaces, but also along the grain boundaries in the sublayers. The chemical diffusion coefficient in the amorphous phase was determined at 673 K, using a previously developed method. The results were compared with experiments on Ni/Ti multilayers, which show similar reactions. At 723 K and higher temperatures the fcc (Ni, Ta) solid solution transformed into the stable Ni3Ta compound.

The migration of a (100) θ = 43.6°(Σ29) twist grain boundary is observed during the course of a molecular-dynamics simulation. The atomic-level details of the migration are investigated by determining the time dependence of the planar structure factor, a function of the planar interparticle bond angles, and the location of the center of a mass of planes near the grain boundary. It is found that a migration step consists of local bond rearrangements which, when the simulation cell is made large enough, produce domain-like structures in the migrating plane. Although no overall sliding is observed during migration, a local sliding of the planes near the migrating grain boundary accompanies the migration process. It is suggested that a three-dimensional cloud of thermally produced Frenkel-like point defects near the boundary accompanies, and facilitates, its migration.

This paper examines the stresses and displacement gradients in atomistic models of cracks based on an EAM potential devised for aluminum. Methods for computing these quantities are described. Results are presented for two models differing in terms of the orientations of the crack relative to the crystal, a [100] (010) orientation that behaves in a brittle fashion and a [111] (110) orientation that emits partial dislocations prior to extending. Both models display lattice trapping. The stresses in the brittle crack model are compared with the linear elastic prediction and found to be in remarkably good agreement to within distances of about one lattice parameter of the crack tip and at the free surface where contributions from sources other than strain energy (e.g., surface tension) influence the results. Similar results are observed for the ductile model until dislocation emission occurs. The largest stresses that develop just prior to crack extension or dislocation emission are used to estimate the ratio of theoretical tensile strength to shear strength in this material. Eshelby's conservation integrals, F and M, are also computed. F is found to be essentially contour independent and in agreement with the linear elastic prediction in both models until dislocation emission occurs, at which point a large screening contribution arises from the emitted partials. The contour size dependence of M reveals some interesting features of the crack tip including a slight wobble of the crack tip inside its potential well with changing applied K and the existence of forces acting to move the crack faces apart as blunting occurs.

The temperature dependence of electrical resistivity of Nd2−xCexCuO4−y (x = 0.0, 0.1, 0.15, 0.18) was measured at oxygen partial pressures of 2.08 × 10−1 1.8 × 10−2, 8.1 × 10−4, and 3.3 × 10−6 atm. The oxygen-partial-pressure dependence of resistivity indicated that the charge carriers in Nd2−xCexCuO4−y were electrons. According to the resistivity data of these compounds at temperatures above 770 K, Ce doping affected the relation between the oxygen deficiency and oxygen partial pressure in Nd2CuO4−y: the doping of Ce worked to hinder the formation of oxygen vacancies in the lattice. Moreover, the carrier density after Ce doping was found to be much less than the value anticipated from the amount of the dopant. This suggested that not all the doped Ce ions worked as donors. That is, the relative amount of Ce3+ ions compared to that of Ce4+ ions increased as the total amount of the doped Ce ions increased.

Electrical and structural properties of coevaporated Fe–Si thin alloy films as a function of annealing temperature have been studied by in situ electrical resistivity measurements and structural analysis including 2 MeV 4He+ ion backscattering, x-ray diffraction, scanning microscopy, and roughness measurements. In the as-deposited state the coevaporated alloy film was amorphous. Upon annealing a sharp increase in resistivity occurred near 400 °C and the increase was determined to be amorphous-to-crystalline β–FeSi2 phase transformation. In cooling, the resistivity increased monotonically with decreasing temperature. Surface roughness is well pronounced for films heat treated at temperatures higher than 700 °C and is due to the difference in thermal expansion coefficients of substrate and β–FeSi2. The crystalline β–FeSi2 was determined to be a semiconductor with an energy gap of 0.80 eV. It is p-type, having a hole concentration and a Hall mobility of 1.4 × 1018 cm−3 and 4 cm2/V s, respectively, at room temperature. The kinetics of the transformation was determined by isothermal heat treatment over the temperature range of 362 °C to 390 °C, and an activation energy of 2.3 eV was measured. The nucleation and growth kinetics in the crystallization process show a change in the power of time dependence from 2.1 to 3.4.

The mechanism underlying the hydrogen-induced embrittlement effect in FeAl has been investigated using a local density functional total-energy approach. The bonding characteristics, the bond and cleavage strength between iron and aluminum layers, and the surface energy with and without interstitially absorbed H are calculated from first-principles band-structure and atomic-cluster methods. Our unique combination of techniques permits the simultaneous study of the metallic and localized bonding effects on an equal footing. Results from this study show that FeAl (in the absence of H) is intrinsically highly resistant to cleavage fracture in terms of the high theoretical cleavage strength. Hydrogen locally dilates the Fe–Al lattice, and this is accompanied by a sizable decrease in Fe–Al cleavage (or cohesive) strength. Our results suggest that the underlying mechanism of H-embrittlement in aluminides is a depletion of d-bonding charge on the Fe site resulting from the charge transfer from Fe to H. Results also indicate that the H-embrittlement effect is greater for H adsorbed in Fe-rich sites.

Interfaces formed by condensing Ni atoms onto YBa2Cu3O7−x (Y-123) and Bi2Sr2Ca0.8Y0.2Cu2Ox (Bi-2212) have been studied with x-ray photoelectron spectroscopy. For both Y-123 and Bi-2212, the Ni 2p3/2 and O 1s core level features indicate Ni–O reactions and changes in the Cu 2p3/2 emission that reflect reduction from nominal Cu2+ to Cu1+ oxidation states. Ni deposition onto Bi-2212 also reduces Bi–O bonding and releases Bi atoms. For Ni/Y-123, analysis of emission intensities as a function of coverage shows that O and Ba intermix in the growing metal overlayer but that Cu is trapped at the buried interface. For Ni/Bi-2212, similar analysis shows O and Bi intermixing but less effective Cu trapping.

We present results of a molecular dynamics simulation study of the effect of electron-ion interactions on the dynamics of the thermal spike in Cu. Interatomic forces are described with a modified embedded atom method potential. We show that the electron-ion interaction acts to reduce the lifetime of the thermal spike and therefore the amount of atomic rearrangement that takes place in energetic displacement cascades in Cu. The results point toward the important effect that inelastic energy losses might have on the dynamics of displacement cascades in the subcascade energy regime where the lifetime of the thermal spike is expected to exceed the electron-phonon coupling time.

Temporary alloying of Ti alloys with hydrogen is used to modify microstructure and hence enhance mechanical properties. In the present work the influence of the thermal history of the Ti–6Al–4V alloy on the final microstructure was studied. Two different kinds of microstructures were investigated: (a) as-hot isostatic pressed (HIP) having an α + β microstructure; (b) β-solution-treated and subsequently quenched (α'-microstructure). After each stage of the hydrogenation-dehydrogenation (HDH) process the material was analyzed by optical, scanning electron and transmission electron microscopy, and by x-ray diffractometry. It was found that fcc (γ) hydrides precipitate in both the α and the α' matrix. The orientation relationships were found to be the same in both cases. Hydrides precipitate from the β-phase when the hydrogen content of the alloy is sufficiently high. The minimum concentration that still led to precipitation was 1.4 wt. % H. The β/γ orientation relationship was determined. The HDH process resulted in either of two structures: (1) a fine, broken-up microstructure containing a fine mixture of α' + (α + β) when the starting microstructure was α', or (2) a modified α + β microstructure containing fine α and β precipitates and a high density of microdefects within the primary α in the as-HIP α + β microstructure specimens.

Gas-atomized spray deposition involves the creation of a spray of droplets by a gas atomizer and the consolidation and solidification of these droplets on a substrate. The present paper describes an investigation of the fundamental characteristics of heat transfer and solidification during spray deposition. Spray deposition was used to manufacture Sn-15 and 38 wt. % Pb preforms using atomizer-substrate distances of 180 and 360 mm, gas flow rates of 2.5 and 3.4 g/s, and melt flow rates of 61 and 35 g/s. Analytical and numerical models were developed to predict the thermal history of the spray deposit for a range of deposit-substrate heat transfer coefficients. A deposit-substrate heat transfer coefficient of ∼104 W m−2 K−1 was determined by comparing measured and calculated spray-deposit thermal histories both during and after spray deposition. Microstructural analysis of transverse sections of the spray deposits revealed maximum values of spray-deposit density and cell/grain size at specific distances from the deposit-substrate interface. The distance between the density and cell/grain-size maxima and the deposit-substrate interface increased from 0.9 to 10 mm for Sn–15 wt. % Pb and from 2.6 to 11.3 mm for Sn–38 wt. % Pb as the atomizer-substrate distance was increased from 180 to 360 mm and the melt to gas mass flow rate ratio was decreased from 24 to 10. The origin of these microstructural features is described in terms of heat transfer during spray deposition.

We have grown extremely flat (2D) electrodeposited copper leaves that exhibit ramified patterns. These patterns are stable and can be withdrawn from the solution. A fractal analysis based on correlation function and branch counting was carried out. Also, x-ray and SEM pictures give insight into the fractal structure of this new type of aggregates and help to find the lower crossover to nonfractal compact substructure.

Glancing angle x-ray reflectivity and EXAFS measurements have been made on a series of UHV prepared Al/Ni bilayers with varying amounts of oxygen impurities. These samples show an intrinsic reacted region prior to annealing, and for clean samples further reaction occurs at 250 °C. Oxygen is found to influence strongly the course of the reaction with an effect which depends on its location. A few percent O impurity within the Al film strongly suppresses the grain boundary diffusion path, which allows the growth of a smooth NiAl3 layer. Interfacial O exposures of 60 and 600 Langmuir both inhibit the initial reaction and raise the temperature at which further reaction occurs to as much as 300 °C with an effect which depends on exposure. The thickness of the intrinsic reaction zone is about 60 Å for clean samples, and is nearly eliminated for contaminated interfaces. The results indicate that surface/interface, grain boundary, and bulk diffusion all play important roles in the formation of these interfaces, and that each of these is influenced by O impurities.

Bilayer samples of U/Al, U/Ti, U/Si, and U/C have been ion beam mixed with 400 keV Ar and U/Al with Xe ions at doses from 2 × 1015 to 1 × 1017 ions/cm2. Mixing experiments were performed at various temperatures between 77 and 420 K. The amount of interfacial mixing, 4Dt, follows a linear dose dependence below a critical temperature, depending on the system studied. Below this temperature, the mixing efficiency, defined as ∂(4Dt)Φ where 4Dt is the mixing and Φ is the dose, is temperature independent. Its value, as well as the value of the transition temperature, agrees well with the thermodynamical model of chemically biased diffusion in a thermal spike for the four systems tested. The transition between the thermal spike regime and the temperature enhanced mixing regime was interpreted on the basis of an intracascade mechanism. The formation of an intermetallic compound in the U/Al system was detected and interpreted on a qualitative basis by crystallographic considerations.

Electron microscope studies have shown significant changes in the microstructure of the Nd14Fe80B6 melt-spun ribbons, depending on the quenching conditions. The average grain size of about 20 nm near the wheel side, increasing to 500–700 nm near the free side, can be observed for ribbons of the thickness of about 25–30 μm, indicating strong internal variations in quench rate across the ribbon thickness. For low quenching rates different morphologies of the crystalline phase Nd2Fe14B are observed. Equiaxed grains near the wheel side of the ribbon become columnar grains or even dendritic grains near the free side. Texturing effects in the equiaxed grains are observed only for grains larger than 30 nm and disappear for very large equiaxed grains (about 500 nm), as well as for dendritic structures. Most evident texture is observed in columnar grains, which crystallized with the c-axis perpendicular to the ribbon plane.

The new ternary intermetallic compound Zr3Ti2Si3, with a (Mn5Si3)16H crystal structure, was designed based on the information of the crystal structure of the related binary compounds Zr5Si3 and Ti5Si3 in order to be used in ultra-high temperature structural applications. By x-ray diffraction analysis, we demonstrate the possibility of substituting an entire layer of zirconium atoms with a layer of titanium atoms in the (Zr5Si3)16H. An analysis of atomic neighbor distances in each compound was done. It was found that the Zr–Si relative interatomic distance diminishes while the Ti–Si distance increases. This indicates that Zr–Si bond strength is maintained as in the binary Zr5Si3. The resulting ternary intermetallic compound has a 16H crystal structure and has a lower density than the original compound of zirconium silicide. This new compound, which is stronger than Ti5Si3 and lighter than Zr5Si3, is considered an excellent candidate of the next generation of intermetallic compounds for ultra-high temperature structural applications.

Ternary sublattice site occupancy in two L12-structured intermetallic compounds were evaluated by a transmission electron microscope technique called ALCHEMI, or atom site location by channeling enhanced microanalysis, and by x-ray diffractometry, through measuring the relative integrated intensity of fundamental and superlattice x-ray diffraction peaks. The x-ray diffractometry showed that in nickel-rich Ni3Al + Hf hafnium was found to occupy preferentially the aluminum sublattice, and in a multiphase alloy an L12-structured phase with the composition Al74.2Ti19Ni6.8 nickel atoms showed a strong preference for the titanium sublattice. The ALCHEMI data broadly agreed with the x-ray results for Ni3Al but gave completely the opposite result, i.e., a preference of nickel for the titanium sublattice, for Al3Ti. The methods of ALCHEMI and x-ray diffractometry are compared, and it is concluded that ALCHEMI data may be easily convoluted by peak overlap and delocalization effects.