An embedded-defect (ED) interatomic potential for Fe is presented. The ED model is based on the same physical description of interactions as the embedded atom method (EAM) but including angular-dependent many-body local terms. In this work, the ED potential is fitted to several Fe properties and is used to simulate the vacancy and various self-interstitial configurations. The pair and many-body contributions to the formation energy of the defects are calculated, as well as the corresponding relaxation volumes. The results are compared to those obtained with an EAM potential fitted to the same properties. The vacancy displacement field is studied in detail for both potentials. Anharmonic effects are assessed and correlated with the unrelaxed force pattern.

A detailed structural study of the newly observed orthorhombic phase of silicon (space group Imma) has been made using angle-dispersive powder-diffraction techniques and an image-plate area detector. The Imma phase is found to be stable between 13.2(3) and 15.6(3) GPa, and both the β-tin-to-Imma and Imma-to-simple-hexagonal transitions are found to be first order with volume changes (ΔV/${\mathit{V}}_0$) of 0.2(1)% and 0.5(1)%, respectively. The volume discontinuities at the transitions are accompanied by pronounced discontinuities in Δ, the atomic coordinate of the Imma phase, which is found to vary from ∼0.3 to 0.4 over the stability range of the Imma phase.

We report results of atomistic simulation studies concerning the incorporation of extrinsic defects in yttrium iron garnet crystals. We calculate defect formation energies in order to determine the most favorable impurity incorporation mechanisms in yttrium iron garnet.

Thermodynamic, electronic, and magnetic properties of the P/FeΣ3[11¯0](111) grain boundary and the P/Fe(111) free surface are compared using the full potential linearized augmented plane-wave method. The calculated segregation-energy difference of 0.8 eV is consistent with the embrittling effect of P according to the Rice-Wang thermodynamic model. The clean FeΣ3[11¯0](111) boundary is found to undergo an ω phase transition which results in an antiferromagnetic coupling within the core of the grain boundary. Structural relaxations in the grain boundary alter the P segregation energy by ∼2.0 eV and are found to play an important role in defining the correct sign and value of the segregation energy difference controlling embrittlement. The P/Fe chemical interaction is found to be ‘‘embeddedlike’’ electrostatic rather than covalent in both environments. Long-range effects of P impurities are found for the Fe magnetization which reduce the embrittling potency of P.

We measured the refractive indices of undoped and Mg-doped lithium niobate by an interferometric technique in a wavelength range from 400 to 1200 nm. The composition of the undoped samples varied between 47 and 50 mol % ${\mathrm{Li}}_2$O in the crystal whereas the Mg-doped samples were grown from congruent melt with up to 9 mol % Mg. Our results can be excellently described by a generalized Sellmeier equation which takes into account the defect structure of Li-deficient and Mg-doped lithium niobate. Calculations of the phase-matching conditions of several nonlinear effects in Mg-doped lithium niobate reveal good correspondence with the respective experimental results in a temperature range from -50 to 250 °C.

The frequencies of transverse-optical Γ phonons in ${\mathrm{KNbO}}_3$ and ${\mathrm{KTaO}}_3$ are calculated in the frozen-phonon scheme making use of the full-potential linearized muffin-tin orbital method. The calculated frequencies in the cubic phase of ${\mathrm{KNbO}}_3$ and in the tetragonal ferroelectric phase are in good agreement with experimental data. For ${\mathrm{KTaO}}_3$, the effect of lattice volume was found to be substantial on the frequency of the soft mode, but rather small on the relative displacement patterns of atoms in all three modes of the ${\mathit{T}}_{1\mathit{u}}$ symmetry. The TO frequencies in ${\mathrm{KTaO}}_3$ are found to be of the order of, but somehow higher than, the corresponding frequencies in cubic ${\mathrm{KNbO}}_3$.

The evolution of individual He bubbles in thin Al foils during 200-keV Xe irradiation at room-temperature has been followed with in situ transmission electron microscopy. He bubbles were produced by room-temperature implantation of 3-keV He ions into prethinned aluminum samples. During subsequent xenon irradiation, several distinct processes were observed to cause individual He bubbles to increase or decrease in size. Bubble growth was observed to take place by radiation-induced coalescence of bubbles without bubble motion. This coalescence was a result of the net displacement of Al atoms out of the volume between bubbles initially in close proximity. The resulting nonequilibrium-shaped bubble evolved towards a more energetically favorable spherical shape whose final size was determined by equilibrium bubble pressure. Bubbles were observed to disappear as the specimen surface was removed by sputtering. Bubbles unaffected by sputtering were observed to decrease in size at an average rate of 0.1 to 0.2 nm/(${10}^{15}$ Xe/${\mathrm{cm}}^2$) or 0.024 to 0.048 nm/dpa (displacements per atom). This rate of bubble shrinkage can be understood on the basis of direct displacement of He out of the bubble while the bubble remains at equilibrium pressure. He resolution occurred at a rate of 0.005 to 0.01 (${\mathrm{He}}^{\mathrm{ejected}}$/He)/dpa. No examples were found that would indicate complete destruction of a bubble by a single Xe ion. Bubble centers remained fixed during bubble shrinkage indicating negligible bubble motion during room-temperature irradiation.

Potassium titanyl phosphate (${\mathrm{KTiOPO}}_4$ or KTP) is a quadriatomic material made of O, P, K, and Ti which crystallizes in an orthorhombic structure with space group Pna2. The virgin KTP at different axial orientation, 〈100〉, 〈001〉, and 〈010〉, has been investigated by the channeling technique. The obtained corresponding minimum yields are 3.8, 4, and 11 %, respectively. Damage in KTP was created by 200 and 400 keV He-, Xe-, and Hg-ion irradiations with the dose varying from 1×${10}^{13}$ to 1×${10}^{16}$ ions/${\mathrm{cm}}^2$. The effect of ion mass, energy, dose, annealing temperature, and orientation of KTP on the damage has been studied. The number of displaced atoms, the damage peak, and damage width are extracted from the experimental data. The numbers of the displaced atoms obtained are compared with the trim code (the transport of ions in the matters). The result shows that the experimental numbers of the displaced atoms are in agreement with theoretical ones predicted by the trim code within the order of the magnitude. Also for the case of He ion implanted in KTP to a dose of 1×${10}^{16}$ ions/${\mathrm{cm}}^2$, the recovery of the crystallinity of the KTP is observed after 800 °C annealing.

Based on an analysis of the temperature dependence of activation energies, performed for the rotational diffusion of water molecules and for their reorientation in the proton conductivity process, the relation γ=1+${\mathit{T}}_{\mathit{s}}$/${\mathit{T}}_{\mathit{b}}$ was deduced, where ${\mathit{T}}_{\mathit{s}}$ and ${\mathit{T}}_{\mathit{b}}$ are temperatures of singularity and boiling, respectively, and γ is the exponent in the power law η=${\mathrm{\eta }}_0$(T/${\mathit{T}}_{\mathit{s}}$-1${)}^{\mathrm{-}\gamma }$. The modified power law has successfully been applied to describe temperature changes in the viscosity η of water at normal and higher pressures.

This paper addresses the problem of how sound propagates in a granular medium composed of noncohesive particles. This propagation has many unusual aspects which are due to the fragile nature of the contacts between grains. Sound propagation in a granular material is very sensitive to the exact position of each of the grains. The thermal expansion of a single bead of about 3000 Å can produce a change as large as 25% in the total transmission of the sound even though this expansion is 4 to 5 orders of magnitude smaller than the wavelength of the sound or the size of a bead (5 mm). I interpret this as being due to sound propagating predominantly along force chains within the medium. By replacing some of the beads with local heaters, I have used this effect to investigate the spatial properties of the low-amplitude vibrations. The disturbance of a single heater can be characterized by a time scale ${\mathrm{\tau }}_{\mathit{h}}$ which reveals the elapsed time of the signal traveling along a path via the heater. Two heaters placed symmetrically with respect to the source and detector can produce very different disturbances. Finally, I show that the spatial pattern caused by heaters placed at different positions within the medium is very irregular in that two adjacent heaters can give very different responses. These experiments all indicate the presence of strong inhomogeneities and the existence of force chains within the medium along which the sound predominantly travels.

Electron-beam evaporation was used to deposit ${\mathrm{Ni}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Zr}}_{\mathit{x}}$ alloy thin films, 0<x<1.0, onto piezoelectric quartz-crystal substrates. For 0.1<x<0.87, the films are single-phase amorphous. Surface acoustic (Rayleigh) waves were generated and detected on the substrates using interdigital transducers predeposited on the substrate surface. The difference between the Rayleigh wave velocity before and after the deposition of the ${\mathrm{Ni}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Zr}}_{\mathit{x}}$ films is used to deduce the shear modulus, μ(x), of the films through a first-order perturbation theory. For all compositions, μ(x) is lower than the weighted average of the μ values of the pure Ni and Zr films. Within the range 0.1<x<0.87, μ(x) has five narrow maxima indicative of chemical short-range ordering at selected compositions.

The question of whether the deuteron glass transition is an equilibrium phase transition or a metastable kinetic phenomenon, observed because of the finite experimental observation time, is addressed. Two-dimensional (2D) exchange NMR of O-D . . .O deuterons in deuteron glasses provides a unique possibility of a direct determination of the O-D . . .O intra- and interbond exchange times as well as for a determination of the Edwards-Anderson order parameter deeply in the glassy phase where 1D NMR methods for order parameter determination fail. On the frequency observation scales of spin-lattice relaxation and line-shape studies ${10}^3$–${10}^8$ Hz, the deuteron glass phase appears static. 2D exchange NMR extends the frequency observation window into the mHz region. The O-D . . .O deuteron intrabond exchange time ${\mathrm{\tau }}_{\mathrm{exch}}$ in ${\mathrm{Rb}}_{0.68}$(${\mathrm{ND}}_4$${)}_{0.32}$${\mathrm{D}}_2$${\mathrm{AsO}}_4$ has been determined as a function of temperature below 45 K. A very slow fluctuation in the H-bond double minimum potential is detected which averages out the H-bond asymmetry and the glass order parameter to zero at long-enough times. The deuteron glass phase is nonergodic, showing a frozen-in disorder at times t≪${\mathrm{\tau }}_{\mathrm{exch}}$, whereas for times long compared to ${\mathrm{\tau }}_{\mathrm{exch}}$, ergodicity is restored.

The behavior of lead atoms implanted at 150 keV into (110) aluminum single crystals is studied by Rutherford backscattering (RBS) and a channeling technique. Channeling observed for the Pb signal in the as-implanted samples for implantation fluences in the region of 1×${10}^{15}$–3.1×${10}^{16}$ ${\mathrm{cm}}^{\mathrm{-}2}$ is ascribed to the formation of nanometer-sized Pb inclusions growing in topotaxy (epitaxy in three dimensions) within the Al matrix. Monte Carlo channeling simulations have been performed based on average sizes of the inclusions obtained from transmission electron microscopy and depth distributions of Pb atoms from RBS. It was shown that channeling in the inclusions strongly depends on their size. Large inclusions demonstrate by far the best channeling due to the relative reduction of the surface (interface) peak effect. In the simulations multiple intersections of the inclusion/matrix interfaces were found to be of importance for dechanneling of the analyzing ions. Comparison of the simulations with experimental data showed that in the as-implanted samples a considerable fraction of the Pb atoms is in supersaturated solution and in small (≤1 nm) inclusions which do not show significant channeling. Upon annealing the larger inclusions grow by absorption of atoms from the solution and coalescence with smaller precipitates. It is shown that the inclusions grow nearly perfectly aligned with the Al matrix in a cube-cube orientation relationship and with a possible misorientation of less than 0.1°. Features related to the coherency of the inclusion/matrix interface are discussed and a comparison with other channeling measurements for nanometer-sized crystallites is made.

A comparative study of the ${}_{}{}^{29}{}_{}{}^{}\mathrm{Si}$ spin-lattice relaxation behavior (induced by trace amounts of paramagnetic dopants in the glass) in phase-separated ${\mathrm{Li}}_2$${\mathrm{Si}}_4$${\mathrm{O}}_9$ and monophasic ${\mathrm{Li}}_2$${\mathrm{Si}}_2$${\mathrm{O}}_5$ and ${\mathrm{Na}}_2$${\mathrm{Si}}_2$${\mathrm{O}}_5$ glasses has been made in order to understand the nature of clustering and the resulting intermediate-range ordering. Optically clear tetrasilicate and disilicate glasses were prepared with 500 to 2000 ppm of ${\mathrm{Gd}}_2$${\mathrm{O}}_3$, a paramagnetic dopant. The constituent structural units (${\mathit{Q}}^3$ and ${\mathit{Q}}^4$ species) in all tetrasilicate glasses show strong differential relaxation following a power-law behavior. This is due to preferential partitioning of ${\mathrm{Gd}}^{3+}$ into the lower silica (${\mathit{Q}}^3$-rich) regions of these glasses, indicating the presence of Q species clusters too small to produce optical opalescence (a few nm to perhaps tens of nm). Preliminary results on ${}_{}{}^6{}_{}{}^{}\mathrm{Li}$ spin-lattice relaxation in these glasses support this hypothesis. Differential relaxation becomes more pronounced on annealing due to growth of such clusters. No such differential relaxation was observed in the monophase disilicate glasses. For spin-lattice relaxation induced by direct dipolar coupling to paramagnetic ions, the recovery of magnetization is proportional to time as M(t)∼${\mathit{t}}^{\mathrm{\alpha }}$ where α is a function of the dimensionality D of mass distribution of the constituent Q species around the ${\mathrm{Gd}}^{3+}$ paramagnetic centers in the glass. For tetrasilicate glasses D≊2.62±0.22 and the system behaves as a mass fractal up to a length scale of 2 to 3 nm. D is thus equal to, within error, the theoretical value of 2.6 for an infinite percolation cluster of one type of Q species in another. For disilicate glasses, D≊3.06±0.18 which indicates a three-dimensional (and thus nonfractal) mass distribution of the constituent Q species over the same length scale.

The structure of samples of amorphous hydrogenated carbon, prepared from acetylene and propane precursors, containing 35 and 32 at. % hydrogen, respectively, was investigated by time-of-flight neutron diffraction in the range 0.2–50 A${\mathrm{̊}}^{\mathrm{-}1}$ using the ISIS spallation source. The large dynamic range of the data ensures a real-space resolution sufficient to reveal directly the proportions of ${\mathit{sp}}^2$ and ${\mathit{sp}}^3$ hybridized carbon. The results show that, in these hard carbon materials, the carbon-atom sites are predominantly ${\mathit{sp}}^2$ bonded, and the carbon-carbon single bond:carbon-carbon double bond ratio is about 2.5:1. The detailed information on atomic correlations thus provided is used to discuss current structural models, and in particular the data are used to show that these models require significant modification.

Inelastic-neutron-scattering experiments have been performed on samples of amorphous hydrogenated carbon, prepared from acetylene and propane, containing about 35 and 32 at. % hydrogen, respectively. In these hard carbons, the hydrogen is predominantly bonded to ${\mathit{sp}}^3$ carbon, with approximately equal concentrations of CH and ${\mathrm{CH}}_2$ groups. There is little hydrogen in other bonding environments, and a small amount of ${\mathrm{H}}_2$ is trapped in cages in the material.

Our NMR studies give experimental evidence of bonding heterogeneity in samples of a-C:H on the nanometer scale. Two classes of protons were identified with distinctly different spin-lattice-relaxation behavior. The difference in the spin-lattice relaxation provides a means of spectral editing of cross-polarization magic-angle spinning, combined rotation and multiple-pulse spectroscopy, dipolar dephasing spectra, and multiple quantum NMR experiments. This combination of the various NMR techniques allows for a detailed structural investigation of a-C:H, e.g., the ${\mathit{sp}}^2$:${\mathit{sp}}^3$ ratio, the relative amount of hydrogenated and nonhydrogenated carbons, etc. A model incorporating the heterogeneity is established and discussed. The NMR results are compared with neutron spectroscopy and diffraction data.

Quasielastic and inelastic neutron-scattering experiments have been performed in order to determine the structural and dynamical properties of α-${\mathrm{LaNi}}_5$ hydride. A microscopic-jump model on the basis of the well-known ${\mathrm{CaCu}}_5$ structure of the compound (extended Chudley-Elliott model) has been used. It starts from the assumption of the hydogen atoms occupying two energetically different interstitial sites. This model has been applied to the long-range and localized diffusion of hydrogen in a single crystalline sample. To enhance the contrast for the incoherent scattering from the H nuclei, natural Ni was substituted with its ${}_{}{}^{60}{}_{}{}^{}\mathrm{Ni}$ isotope. An appreciable anisotropy of the long-range diffusion is not observable at 423 K. A fast and localized diffusive process was found which is confined to a hexagonal arrangement of six sites in the (001) lattice plane. The local hydrogen vibrational modes are resolved and assigned to specific vibrations taking into account the crystallographic directions in which they appear.

Previous theoretical studies and molecular dynamics simulations indicate that the anharmonic version of a zone-boundary-mode phonon in a periodic one-dimensional lattice with nearest-neighbor quartic and quadratic interactions is unstable, and that this instability can lead to the production of intrinsic localized modes. We show here that such an instability occurs when nearest-neighbor cubic anharmonicity is added, but that the zone boundary mode is stabilized when the cubic anharmonicity becomes sufficiently large. A direct connection is established between the existence of this instability and the existence of intrinsic localized modes. Furthermore, our analysis reveals the existence of a second type of zone-boundary-mode instability, which is not related to intrinsic localized modes. This ‘‘period-doubling’’ instability is also found to occur in one-dimensional lattices with realistic potentials, such as Lennard-Jones, Morse, and Born-Mayer, whereas the instability related to intrinsic localized modes does not occur for these potentials, owing to their inclusion of higher-order anharmonicity. Likewise, intrinsic localized modes are not found in direct numerical searches for these cases, and we conclude that they do not exist in monatomic lattices with these potentials.

We investigate solitons in the one-dimensional classical many-body system of particles interacting by an inverse-square pair potential. We use a combination of numerical results for finite systems, and exact results for the continuum limit, to deduce properties of solitons in the thermodynamic limit. Two remarkable conjectures are supported by these results, and the correspondence with the exact results for the quantum system is established.