Grain-boundary melting in a lattice-gas model of a bicrystal is studied by Monte Carlo simulation using the grand canonical ensemble. Well below the bulk melting temperature ${\mathit{T}}_{\mathit{m}}$, a disordered liquidlike layer gradually emerges at the grain boundary. Complete interfacial wetting can be observed when the temperature approaches ${\mathit{T}}_{\mathit{m}}$ from below. Monte Carlo data over an extended temperature range indicate a logarithmic divergence w(T)∼-ln(${\mathit{T}}_{\mathit{m}}$-T) of the width of the disordered layer w, in agreement with mean-field theory.

The resistance versus pressure of NpSb single crystals has been measured up to 23 GPa and as a function of temperature from 1.3 to 300 K. Up to 2.3 GPa the Néel temperature, ${\mathit{T}}_{\mathit{N}}$, is found to decrease at a rate of about -11.5 K/GPa. We suggest that the direct effect of pressure on the spin may play a significant part in this variation. Between 2.3 and 2.7 GPa a dramatic change in the resistance versus temperature behavior below ${\mathit{T}}_{\mathit{N}}$ indicates some electronic or magnetic transformation in NpSb. At higher pressure the NaCl to tetragonal structural phase transition is observed by a decrease in the room temperature resistance and a maximum in the low-temperature resistance. The transition pressure is determined to be between 6 and 8 GPa and the transition shows considerable hysteresis on releasing the pressure. This is compared to previous data from high-pressure x-ray diffraction experiments which show a higher transition pressure. The high-pressure phase seems likely to be nonmagnetic and we show indirect evidence for delocalization of the 5f electrons at the phase transition.

The thermal conductivity λ of KOH-doped ice was measured using the transient hot-wire method at temperatures in the range 55–250 K and at pressures up to 0.15 GPa. With an isobaric increase of temperature at 0.08 GPa, λ decreased 17% at about 74 K. This decrease is associated with the known transition from ordered phase XI to phase Ih which exhibits orientational disorder of ${\mathrm{H}}_2$O. A model for λ indicates that the decrease of λ at the transition is due to an increase of the lattice anharmonicity which might be caused by the disorder in phase Ih. It was inferred from the temperature dependence of λ that phonon scattering in both phases Ih and XI is dominated by three phonon umklapp processes. At 58 K, λ of supercooled phase Ih decreased slightly with increasing pressure whereas λ of phase XI was independent of pressure.

We have investigated the temperature and concentration dependence of cross-relaxation processes in single-crystal fibers of YAG:${\mathrm{Pr}}^{3+}$ which were grown using the laser-heated pedestal growth technique. We have compared the time evolution of the ${}_{}{}^3{}_{}{}^{}$${\mathit{P}}_0$${\to }^3$${\mathit{H}}_6$ and ${}_{}{}^1{}_{}{}^{}$${\mathit{D}}_2$${\to }^3$${\mathit{H}}_4$ transitions under pulsed excitation to the solution of a rate-equation model. We have established values for cross-relaxation rates associated with the ${(}^3$${\mathit{P}}_0$${,}^3$${\mathit{H}}_4$)→${(}^3$${\mathit{H}}_6$${,}^1$${\mathit{D}}_2$) and ${(}^1$${\mathit{D}}_2$${,}^3$${\mathit{H}}_6$)→${(}^3$${\mathit{H}}_4$${,}^3$${\mathit{P}}_0$) processes and demonstrated experimentally that these rates increase with temperature. By comparing this dependence with a thermal activation we have been able to determine which of the individual sublevels of the above manifolds play a dominant role in the cross-relaxation process. In order to determine the mechanism responsible for the cross relaxation, we have compared our results to the Inokuti-Hirayama theory. The best fit was obtained for the case of dipole-dipole transfer.

Different cesium oxides have been prepared in situ and studied by near-edge x-ray-absoprtion fine structure (NEXAFS) measurements. The data have been analyzed by multiple-scattering calculations for two model structures, ${\mathrm{CsO}}_2$ and ${\mathrm{Cs}}_2$${\mathrm{O}}_2$ using self-consistent final-state potentials. The analysis shows that the sample grown by codeposition of oxygen and cesium has ${\mathrm{Cs}}^{+}$${\mathrm{O}}_2^{\mathrm{-}}$ structure with an O-O distance in the molecular anion of 1.40±0.05 Å, whereas the sample obtained on oxidizing a thin Cs film has ${\mathrm{Cs}}^{2+}$${\mathrm{O}}_2^{2\mathrm{-}}$ structure with an O-O distance of 1.55±0.05 Å. Both distances are larger than the values for the corresponding free ions. Models assuming charge transfer between the alkali and oxygen atoms other than quoted above can be excluded. The analysis also shows that the localized empty states of the ${\mathrm{O}}_2^{2\mathrm{-}}$ anion are sensitive to its surrounding matrix.

The crystal structure of the low-symmetry phase Sm(V) was studied by synchrotron x-ray diffraction under isothermal compression at room temperature up to 77 GPa (volume fraction=0.45). Measured lattice spacings and diffraction intensities for Sm(V) in the pressure range beyond the phase mixing show that this phase can be identified as a prototype structure with 3 atoms in the hexagonal unit cell. The occurrence of this low-symmetry structure gives strong evidence for a rapid increase of 4f-electron bonding in Sm in the pressure range above 37 GPa.

The probabilities for up-conversion involving two nearest-neighbor ${\mathrm{Ho}}^{3+}$${(}^5$${\mathit{I}}_7$) ions and ${\mathrm{Tm}}^{3+}$${(}^3$${\mathit{F}}_4$) ions in ${\mathrm{Y}}_3$${\mathrm{Al}}_5$${\mathrm{O}}_{12}$ have been determined. The final states are ${}_{}{}^5{}_{}{}^{}$${\mathit{I}}_5$ and ${}_{}{}^5{}_{}{}^{}$${\mathit{I}}_6$ for ${\mathrm{Ho}}^{3+}$ and ${}_{}{}^3{}_{}{}^{}$${\mathit{H}}_4$ and ${}_{}{}^3{}_{}{}^{}$${\mathit{H}}_5$ for ${\mathrm{Tm}}^{3+}$. A rate-equation model has been used to obtain an expression for the concentration dependence of the up-conversion-rate constant in the migration-limited regime. We have also measured the reverse cross-relaxation probabilities for two of the processes with ${\mathrm{Ho}}^{3+}$${(}^5$${\mathit{I}}_5$) and ${\mathrm{Tm}}^{3+}$${(}^3$${\mathit{H}}_4$) as donors. The results fit well to the energy-gap law for phonon-assisted energy transfer. They also show evidence that there may be a correlation between the deviations from the energy-gap law for opposite processes.

We present a theoretical study of the second-harmonic generation from pairs of two-level atoms in solids. The density-matrix formalism is used to obtain the nonlinear polarization governing the process. The second-harmonic polarization depends on the laser frequency, the interaction potential between the atoms, and the laser intensity. The signal line shape may exhibit power broadening and shift of resonances.

The sign of the formation energy Δ${\mathit{H}}_{\mathit{F}}$ of a compound indicates if the low-temperature long-range order (LRO) corresponds to compound formation (when Δ${\mathit{H}}_{\mathit{F}}$<0) or to phase-separation (when Δ${\mathit{H}}_{\mathit{F}}$>0). However, Δ${\mathit{H}}_{\mathit{F}}$ by itself does not tell us what type (ordering or clustering) of high-temperature short-range order (SRO) can be expected. The reason is that Δ${\mathit{H}}_{\mathit{F}}$ contains two types of contributions: an ‘‘elastic,’’ volume-deformation energy G(x) that reflects the energy invested in deforming the constituents A and B to the volume V(x) of the ${\mathit{A}}_{1\mathrm{-}\mathit{x}}$${\mathit{B}}_{\mathit{x}}$ alloy, and a ‘‘chemical’’ energy ɛ that reflects A-B interactions (charge transfer and atomic relaxations) at a fixed V(x). We show that the LRO and the SRO have the same behavior only when the signs of Δ${\mathit{H}}_{\mathit{F}}$ and ɛ are the same: ordering tendencies (‘‘type I’’) when both Δ${\mathit{H}}_{\mathit{F}}$<0 and ɛ<0, or phase-separating/clustering tendencies (‘‘type III’’) when both Δ${\mathit{H}}_{\mathit{F}}$>0 and ɛ>0. However, ‘‘type II’’ systems (Δ${\mathit{H}}_{\mathit{F}}$>0; ɛ<0) can exhibit a phase-separating LRO and an ordering-type SRO. Direct self-consistent local-density calculations of the total energy coupled with Monte-Carlo simulated annealing calculations of the ensuing Ising-like cluster expansion illustrate these type I, II, and III behaviors in Ni-V, Ni-Au, and Pd-Rh, respectively.

The complex dielectric constants of mixed ${\mathrm{Cs}}_{0.80}$(${\mathrm{NH}}_4$${)}_{0.20}$${\mathrm{H}}_2$${\mathrm{AsO}}_4$ single crystals have been measured in the temperature range from 13 to 200 K and the frequency range from 4.4 to 624 kHz along the a and c axis. Proton-glass behavior is first observed in the low-temperature region. The asymmetric concentration-temperature (x-T) phase diagram is obtained by measuring the complex dielectric constants of mixed ${\mathrm{Cs}}_{1\mathrm{-}\mathit{x}}$(${\mathrm{NH}}_4$${)}_{\mathit{x}}$${\mathrm{H}}_2$${\mathrm{AsO}}_4$ microcrystalline powder pellets for the full range of mixing concentration x from 0 to 1. Proton-glass behavior develops in the concentration range of 0.12<x<0.51. Coexistence of the proton glass phase with the respective long-range ordering is observed in the two phase boundary regions.

Total-energy local-density calculations on approximately 20 periodic crystal structures of a given AB compound are used to define a long-range Ising Hamiltonian which correctly represents atomic relaxations. This allows us to accurately calculate structural energies of relaxed substitutional ${\mathit{A}}_{1\mathrm{-}\mathit{x}}$${\mathit{B}}_{\mathit{x}}$ systems containing thousands of transition-metal atoms, simply by adding up spin products in the Ising Hamiltonian. The computational cost is thus size independent. We then apply Monte Carlo and simulated-annealing techniques to this Ising Hamiltonian, finding (i) the T=0 ground-state structures, (ii) the order-disorder transition temperatures ${\mathit{T}}_{\mathit{c}}$, and (iii) the T>${\mathit{T}}_{\mathit{c}}$ short-range-order parameters. The method is illustrated for a transition-metal alloy (${\mathrm{Cu}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Pd}}_{\mathit{x}}$) and a semiconductor alloy (${\mathrm{Ga}}_{1\mathrm{-}\mathit{x}}$${\mathrm{In}}_{\mathit{x}}$P). It extends the applicability of the local-density method to finite temperatures and to huge substitutional supercells. We find for ${\mathrm{Cu}}_{0.75}$${\mathrm{Pd}}_{0.25}$ a characteristic fourfold splitting of the diffuse scattering intensity due to short-range order as observed experimentally.

We present an x-ray-absorption-spectroscopy study of the electronic and crystallographic structure around the ${\mathrm{Nd}}^{3+}$ ions in 60% Nd-exchanged sodium β’’-alumina crystal. Using a multishell best-fit analysis procedure, we reconstructed the ${\mathrm{Nd}}^{3+}$ local environment and compared it with existing structural models. We found that neodymium ions are mainly located near the mO(9d) sites of the conduction region; they are strongly bonded to two O(5) oxygens which are located in the same plane and shifted from their ideal crystallographic positions in sodium β’’ alumina crystal. In particular, with respect to crystallographic parameters of sodium β’’ alumina, the Nd-O(5) distance is shortened by 0.36 Å, while the Nd-Al(6c) is slightly increased by 0.13 Å. Comparative analysis of Debye-Waller factors obtained for different coordination shells shows evidence of correlation in the vibrational motion of neodymium with the O(5) oxygens and with the Al(6c) aluminums.

A comprehensive theoretical treatment for the inelastic scattering of fast electrons in a crystalline environment is applied to account for variations in characteristic x-ray emission rates from spinel and chromia under zone-axis diffraction conditions for 300-keV electrons. X-ray counts as well as backscattered-electron contrast are recorded as the beam is tilted by about 70 mrad on a two-dimensional scan raster. Calculated K-shell ionization and backscattering cross sections, based on a model which takes into account interaction delocalization, thermal diffuse scattering, and the consequent thickness-dependent channelling effects in a realistic way, are in good agreement with experiment.

We find that recent low-temperature nonequilibrium dielectric experiments indicate that glasses have strongly interacting defects. While many of the features found in the experiments can be explained by the standard model of noninteracting two-level systems, we find that the frequency dependence cannot. Using a Monte Carlo simulation of a nearest-neighbor Ising spin glass, we show that interactions between defects can qualitatively explain the experiments because they lead to the formation of clusters and a hole in the distribution of local fields.

Photoemission spectroscopy measurements in the photon-energy range 35–150 eV have been used to determine the valence band of the stable icosahedral alloy ${\mathrm{Al}}_{70}$${\mathrm{Pd}}_{20}$${\mathrm{Mn}}_{10}$. Resonant photoemission near the Mn 3p→3d transition has been employed to show that the feature in the valence band at about 1.0 eV below the Fermi level is predominantly due to the Mn 3d-derived states. The effect of the Cooper minimum has been used to identify the feature at 3.6 eV below the Fermi level as being mainly due to states of Pd 4d character. The strong decrease of intensity towards the Fermi level has been interpreted as indicative of the presence of a pseudogap at the Fermi level. However, it is shown that the Fermi cutoff also contributes to the observed intensity decrease. No unusual features in the valence band of icosahedral ${\mathrm{Al}}_{70}$${\mathrm{Pd}}_{20}$${\mathrm{Mn}}_{10}$ which could be ascribed to its quasiperiodic nature have been observed within the experimental resolution. A review of published experimental results on the electronic structure of quasicrystals obtained with techniques directly probing the occupied and unoccupied electronic states is also presented.

Hydrogenated amorphous carbon structures, a-C:H, with densities of 1.8 and 2.0 g/${\mathrm{cm}}^3$, have been generated by semiempirical density-functions (DF) molecular-dynamics (MD) rapid cooling of a liquid phase of 128 carbon and 64 hydrogen atoms within periodically arranged cubic supercells. The electronic bonding properties of the model structures are analyzed within a local-orbital description. The structural properties are compared to relevant statistical and diffraction data obtained by neutron scattering and NMR in order to achieve a fundamental understanding of structure-related properties on the molecular level of chemical bonding.

We have studied the effect of quantum lattice fluctuations on the dimerized ground state of the half-filled Su-Schrieffer-Heeger (SSH) model. Our results show that, at least in the weak-coupling limit, quantum lattice fluctuations change the functional dependence of the gap Δ(0) on the coupling constant even if the ratio ${\mathrm{\omega }}_{\mathrm{\pi }}$/4t is small (but finite). In the spin-1/2 case our result is the same as that predicted by Fradkin and Hirsch. But in the spinless case we still predict a long-range dimerization order even if the coupling is weak and ${\mathrm{\omega }}_{\mathrm{\pi }}$/4t is finite. We have shown by numerical calculations that our variational approach gives a good description of the continuous variation of the gap Δ(0) in the fermionic spectrum as a function of the coupling constant v or the phonon frequency ${\mathrm{\omega }}_{\mathrm{\pi }}$. By using the same input parameters as those of SSH we get a 12% reduction of the gap Δ(0) compared with the adiabatic value. The calculated electron density of states does not have the inverse-square-root singularity but does have a peak around Δ(0). There is a significant tail below the peak. These are consistent with the observed optical absorption coefficient in undoped trans-polyacetylene.

A method for analyzing neutron-scattering data at intermediate and high-momentum transfer in liquids and solids is proposed. The aim is to separate the observed dynamic structure factor, S(Q,ω), into its impulse-approximation (IA) and final-state (FS) parts. When the separation is made, both the IA and FS effects can be determined from the data. The method is based on a cumulant expansion of the intermediate scattering function, S(Q,t), in powers of t. The expansion parameters are determined by fitting the corresponding S(Q,ω) to the data. Using the parameters, the IA and FS function are reconstructed. Variants of the method suitable for systems in which momentum distribution is (1) close to a Gaussian and (2) differs markedly from a Gaussian are proposed. The first variant is applied to recent data of Andersen et al. in normal ${}_{}{}^4{}_{}{}^{}\mathrm{He}$ at saturated vapor pressure and T=2.5 K and the kinetic energy, momentum distribution, and the FS broadening function are determined. In the second variant a model momentum distribution is introduced. This is applied to superfluid ${}_{}{}^4{}_{}{}^{}\mathrm{He}$ using a model n(k) having a condensate. The model parameters are determined and bounds on the condensate fraction (${\mathit{n}}_0$=0.10±0.03) are set. The method can be applied to any liquid, solid, or mixture. Full results for the momentum distribution and FS effects in normal and superfluid ${}_{}{}^4{}_{}{}^{}\mathrm{He}$ are presented in a forthcoming paper.

We present a study of the Raman and IR spectra of the ${\mathit{sp}}^2$-bonded carbon clusters ${\mathrm{C}}_{60}$ and ${\mathrm{C}}_{70}$, performed in the framework of a three-parameter bond-charge model based on a transferable interatomic potential. A consistent calculation of the relative amplitudes and of the static polarizability has been performed with only one additional parameter besides those entering the dynamics. The calculated spectra are compared with the existing experimental data for ${\mathrm{C}}_{60}$ and ${\mathrm{C}}_{70}$. Very good agreement is found for the Raman spectrum of both clusters, whereas the agreement for IR spectra is less satisfactory.

The motion of diffuse domain walls in thin garnet films (i.e., walls with extreme width), which was observed experimentally when an in-plane magnetic field was applied, is analyzed numerically using a spatiotemporal diagram technique and analysis of the time evolution of pattern entropy. It is found that the motion of such walls is chaotic and experimental observations of diffuse domain walls may be treated as a direct observation of chaotic motion in a magnetic system.