Low-field magnetoresistance measurements were performed on a two-dimensional grid lateral surface superlattice (LSSL) fabricated on a GaAs/${\mathrm{Al}}_{\mathit{x}}$${\mathrm{Ga}}_{1\mathrm{-}\mathit{x}}$As heterojunction. From a one-parameter fit to the two-dimensional weak-localization theory we have obtained the effective inelastic scattering length at different temperatures. By subtracting the weak-localization term, we have also investigated the electron-electron interactions in this system. Both localization and interaction effects saturate at low temperatures when the inelastic scattering length and thermal diffusion length are longer than the period of the LSSL. At higher temperatures, when the thermal diffusion length becomes comparable to the period of the LSSL, the electron-electron interaction term causes the conductivity to decrease logarithmically with temperature. These results can be explained in terms of coherent backscattering of the electrons in a fraction of the electron puddles formed below the superlattice gate. Disorder ensures that the cells showing these results are not neighbors, so that this effect is an ensemble average of many unit cells several periods apart.

An analysis of the island size and separation distributions of Fe islands, observed in the initial stages of growth in the homoepitaxy of Fe on Fe(001) whiskers, shows scaling properties recently predicted in nucleation and growth theories. A critical size of one atom, where islands greater than the critical size undergo irreversible nucleation, is supported by the measured scaling functions for the Fe on Fe system in the temperature range of 20–250 °C.

Density-functional calculations have been performed to determine optimized geometries and energies of ${\mathrm{C}}_{240}$ using the divide-and-conquer method. Six initial geometries were considered, resulting in convergence to two optimized configurations. The formation energies of the optimized structures are separated by approximately 0.07 eV/carbon atom. The lower-energy structure is highly spherical in agreement with preliminary studies and experimental observations. The higher-energy structure is polyhedrally faceted. The results support the conclusion that the most stable form of large carbon clusters is that of dense spherical caged structures.

The process of x-ray absorption and emission in solids is considered as an x-ray resonant inelastic scattering process. This coherent inelastic scattering picture has recently been used to interpret the excitation energy dependence in the x-ray emission spectra excited with synchrotron radiation. It also suggests that with near threshold excitation, the x-ray emission spectra should be spatially anisotropic. We consider the validity and implication of this approach and factors affecting the spatial and temporal coherence in the scattering process. Taking into account the relaxation effects such as the electron-electron and the electron-phonon interactions, a significant fraction of the total emission intensity may be attributed to the coherent scattering. This picture of the x-ray absorption and emission process opens up the possibility for momentum-resolved x-ray absorption and emission measurements that can be used for band-structure determination. In addition, it has important implications on the fluorescence yield method of obtaining the absorption spectra.

Gaudin has determined the partition function at special temperatures for a lattice gas of particles confined to a ring and interacting with a potential that is logarithmic in the interparticle distance. This model is also relevant to the understanding of long-range order in a discrete version of the Sutherland model. Gaudin found that the pressure changes sign at β=2; it will be shown here that no phase transition is associated with this, and that the anomaly is equally present in the continuum version of the problem discussed by Dyson. The partition function and structure factor will be computed for a finite system of up to 32 sites for more general β, and the results interpreted in terms of the theory of a quantum liquid in the presence of an external periodic potential.

We have carried out molecular-dynamics simulations on BaX${\mathrm{F}}_4$ compounds, where X is Mg, Mn, or Zn. Ab initio potentials, with no adjustable parameters, were used to obtain short-range interactions between ion pairs. We found a polar ground-state structure which is in agreement with the A$2_1$am space group reported experimentally. We were able to reverse polarization in ${\mathrm{BaMgF}}_4$ at high temperatures, using large fields, but were unable to reverse polarization in the other compounds. The second-order phase transition in the Mn compound at 250 K was reproduced. We believe this to be the first extension of molecular dynamics to materials consisting of chains of F octahedra.

For the calculation of specular x-ray and neutron reflectivity from rough surfaces two expressions exist, which are valid under different circumstances. Another expression is derived using the second-order distorted-wave Born approximation, which smoothly connects the two existing expressions. The shape of the reflectivity curve depends not only on the average roughness, but also on the lateral correlation of the roughness profile.

Two-pulse photon-echo measurements were made on the ${}_{}{}^7{}_{}{}^{}$${\mathit{F}}_0$${\mathrm{\leftrightarrows }}^5$${\mathit{D}}_0$ optical transition in a number of crystalline samples of ${\mathrm{Y}}_2$${\mathrm{O}}_3$:${\mathrm{Eu}}^{3+}$ prepared by three different crystal-growth techniques. For crystals prepared by a laser-heated pedestal growth or arc-imaging furnace process, we report the observation of an additional sample-dependent contribution to the optical dephasing in comparison with a crystal grown by flame fusion. The dephasing rate 1/${\mathit{T}}_{\mathit{m}}$ at 1.4 K is up to two orders of magnitude faster than that previously reported for the flame-fusion-grown crystal, and is uncorrelated with concentration. Measurements up to 14 K show that 1/${\mathit{T}}_{\mathit{m}}$ increases linearly with temperature, in contrast to the expected ${\mathit{T}}^7$ behavior due to a two-phonon Raman process. The linear temperature dependence is similar to that seen in glasses and disordered crystalline systems, suggesting the presence of disorder in the samples with more rapid dephasing. The crystals were examined by x-ray diffraction, excitation spectroscopy, and near-field optical microscopy to provide information on the nature of the disorder. The effect of annealing on the crystal dynamics is also investigated.

We have used a first-principles ultra-soft-pseudopotential method in conjunction with an efficient preconditioned conjugate-gradient scheme to investigate the properties of a series of eight cubic perovskite compounds. The materials considered in this study are ${\mathrm{BaTiO}}_3$, ${\mathrm{SrTiO}}_3$, ${\mathrm{CaTiO}}_3$, ${\mathrm{KNbO}}_3$, ${\mathrm{NaNbO}}_3$ ${\mathrm{PbTiO}}_3$ , ${\mathrm{PbZrO}}_3$, and ${\mathrm{BaZrO}}_3$. We computed the total-energy surface for zone-center distortions correct to fourth order in the soft-mode displacement, including renormalizations due to strain coupling. Quantities calculated for each material include lattice constants, elastic constants, zone-center phonon frequencies, Grüneisen parameters, and band structures. Our calculations correctly predict the symmetry of the ground-state structures of all compounds whose observed low-temperature structure retains a primitive five-atom unit cell. The database of results we have generated shows a number of trends which can be understood using simple chemical ideas based on the sizes of ions, and the frustration inherent in the cubic perovskite structure.

In reduced ${\mathrm{LiTaO}}_3$, an axially symmetric EPR signal with ${\mathit{g}}_{\mathrm{\parallel }}$=1.948±0.003 and ${\mathit{g}}_{\mathrm{\perp }}$=1.827±0.003 has been identified. It is attributed to ${\mathrm{Ti}}^{3+}$ (3${\mathit{d}}^1$) by comparison with the already known EPR data of ${\mathrm{Ti}}^{3+}$ in ${\mathrm{LiNbO}}_3$. We are able to explain the g values of this center in ${\mathrm{LiNbO}}_3$, as well as in ${\mathrm{LiTaO}}_3$, by a model calculation involving a dynamic pseudo Jahn-Teller effect. Spin-orbit coupling, lattice vibration, pseudo Jahn-Teller interaction, and the Zeeman term are treated on equal footing. Intervalence transfer from ${\mathrm{Ti}}^{3+}$ to ${\mathrm{Nb}}^{5+}$, forming a ${\mathrm{Nb}}_{\mathrm{Li}}^{4+}$ polaron, has been identified. This transfer can be stimulated either by visible light or by thermal activation. By investigating the temperature dependence of the thermally activated intervalence transfer, we have determined the ionization energy of ${\mathrm{Ti}}^{3+}$ with respect to ${\mathrm{Nb}}_{\mathrm{Li}}^{4+}$ to be 0.11 eV ±0.02 eV.

We studied the pseudoelastic behavior of hypostoichiometric NiAl alloys using a very simple model, with interplanar interaction parameters extracted from recent neutron-scattering experiments. The behavior of the system under shear is studied numerically both in the static (quasiequilibrium) limit and with molecular dynamic simulations. Our results indicate that anomalous phonon softening can facilitate the nucleation of the martensitic phase near boundaries or defects as the system approaches the transformation temperature. For our simple model, we found that the criical stress for the nucleation of the martensitic phase is reduced by almost a factor of 2 as a result of anomalous phonon softening near the martensitic transformation temperature.

Combined first-principles pairwise simulations and quantum-mechanical ab initio perturbed ion (AIPI) calculations have been extensively performed to determine the static equations of state (EOS) of the cubic (fluorite-type) and orthorhombic (α-${\mathrm{PbCl}}_2$-type) polymorphs of ${\mathrm{CaF}}_2$. This theoretical investigation covers the range of pressures experimentally available (0–45 GPa). The elastic behavior and the equilibrium crystal parameters have been accurately determined by means of efficient numerical procedures involving a Richardson-iterated, finite-difference formula for the derivatives and the combination of downhill simplex and modified Powell methods for the multidimensional optimizations. For the bulk modulus and the effective elastic constants of the cubic phase, the simulations and AIPI calculations give increasing functions of pressure with small negative curvatures. Besides, the zero-pressure AIPI computations agree with the values of Catlow et al. [J. Phys. C 11, 3197 (1978)] for the bulk modulus and the elastic constants. The computed EOS also reproduces quantitatively the most recent experimental p-V data for the cubic phase. For the orthorhombic phase, we optimize nine crystal parameters for each value of pressure. This set provides a full structural characterization of this phase, as well as a global description of the p-V relationship that is consistent with the synchrotron-radiation x-ray-diffraction study of Gerward et al. [J. Appl. Cryst. 25, 578 (1992)]. Our simulation techniques are able to detect a first-order phase transition from the low-pressure fluorite-type to the high-pressure α-${\mathrm{PbCl}}_2$-type polymorph. The computed thermodynamic transition pressure lies below the experimental values, as it should for this kind of structural transformation exhibiting large pressure hysteresis.

The authors calculated values for the ratios of the double-electron to main-line strengths with their corresponding excitation energies for the ${\mathit{LN}}_{4,5}$ edges of elements found in the sixth row of the Periodic Table. In addition, the approximate analytical forms for the relative intensities in terms of monopole matrix elements are also given. A comprehensive view of double-electron processes which affect extended x-ray-absorption fine structure analysis of elements in the sixth row is obtained.

We present a calculation of bipolaron stability in doped conjugated polymers which emphasizes the role of the dopants as tunneling bridges between neighboring chains.

A numerical model for the dynamics of a classical wave equation in a two-dimensional Anderson disordered medium is integrated over times of the order of 27 000 inverse bandwidths. Excitations by narrow band sources lead to wave energy densities whose ensemble averages behave diffusively at early times. The behavior at more general times and distances is, however, not diffusive. The observed transport profiles are shown to be inconsistent with predictions from a simple hydrodynamical continuum model of Anderson localization, and to the hypothesis of exponentially slow diffusion. The evolution of the energy density distribution in systems with varying disorders and microstructures, and over a range of length scales, is found to collapse to a single function of rescaled space and time well approximated by e(x,t)≊exp{-x/ξ-(${\mathit{x}}^{2+\mathit{n}}$/4β${\xi }^{\mathit{n}}$t${)}^{\mathit{p}}$} with n≊0.46, p≊0.76, where x is the distance from the source, ξ is the localization length, and the β is a ‘‘residual diffusivity’’ not equal to the bare diffusivity. Dissipation is shown to have no delocalizing effect.

The problem of maximizing the density of quasiperiodic sphere packings is investigated. The investigation is limited primarily to the case where the centers of the spheres are determined by a cut through a higher-dimensional Bravais lattice of a single atomic surface. For certain cases, the sphere packing obtained by the largest connected atomic surface can be made more dense by replacing it with an atomic surface that is nonconnected. The nonconnected atomic surfaces differ from the connected ones only in specific regions near the atomic surface boundaries. These regions are determined by the quasicrystalline point-group symmetry and by the radius of the spheres. This method of improving sphere packings is applied to find nonconnected atomic surfaces for the dodecagonal, octagonal, and icosahedral cases. Using this method in conjunction with other methods, a packing fraction of approximately 0.5645 for a simple icosahedral b-c network is achieved.

Small-angle neutron scattering (SANS) has been used to study the adsorption and desorption behavior of a ${\mathrm{D}}_2$O/${\mathrm{H}}_2$O water mixture in porous materials. When the porous solid is partially saturated, a feature appears in the scattering having a fractal-like power-law form ${\mathit{Q}}^{\mathrm{-}\mathit{D}}$ (where Q is the scattering wave vector) in the very-low-Q region, extending to below the limit of the instruments used and which covers a maximum range of about one and one-half orders of magnitude in Q. The gradient measured during desorption D, being 1.75±0.1 for porous Vycor glass, 1.85±0.1 for Ludox, and 2.0±0.1 for a silica sol-gel, indicates that the percolation network forms a fractal-like geometry under these conditions. The gradient for adsorption is less than 1.0 for both Vycor and Ludox, and no power-law form of SANS is seen at all for the silica sol-gel. This implies that fractal-like percolation networks are not formed during the adsorption process.

We have investigated the extent of the spatial correlation of deuteron ordering in the solid solution ${\mathrm{Rb}}_{1\mathrm{-}\mathit{x}}$(${\mathrm{ND}}_4$${)}_{\mathit{x}}$${\mathrm{D}}_2$${\mathrm{PO}}_4$ (D-RADP-X, X=100x) by means of quadrupole-perturbed ${}_{}{}^{87}{}_{}{}^{}\mathrm{Rb}$ NMR. In contrast to the usual approach where glass transitions are discussed in terms of a departure from complete disorder, we are treating the ordering in D-RADP-X as a departure from the long-range-ordered ferroelectric (FE) and antiferroelectric (AFE) phase states. The composition parameter x is a parameter for the competing FE and AFE interactions which reduces the spatial correlation length of the ordered phases to the nanometric region (0.35<x<0.56). This approach allows us to make use of techniques employed in the classical distortive phase transitions and to expand them to short-range correlated systems. The results of this analysis clearly show that the glass state of D-RADP-50 consists of short-range-ordered AFE clusters of an average size 6 to 8 ${\mathrm{nm}}^3$: i.e., a correlated order of about 150–200 O-D...O bonds. This size is too small to exhibit features of the long-range-ordered AFE phase state, such as a phase transition with a well-defined transition temperature. Nevertheless, this local order can be distinguished unambiguously from a random freeze-out of the deuterons in the O-D...O bonds.

Li K soft-x-ray emission spectra excited with monochromatic synchrotron radiation have been measured for LiBr and LiCl. The white-light reflectivity of these samples has also been recorded as a measurement of the Li K excitation spectra. The phonon relaxations involved in Li K excitations to the conduction band were determined to be 1.0 and 1.2 eV in the LiBr and LiCl, respectively, based on the energies of the valence band maxima in the x-ray emission spectra and their comparison to the energies of the valence band maxima and the Li 1s core levels as measured by photoemission. On the other hand, energy offsets between the Li K exciton features in the emission and the excitation spectra provided a measure of the phonon relaxations for these localized excitations: the relaxations were found to be 1.0 eV in both materials, similar to the values determined for excitations to the delocalized conduction band states. The energies of the x-ray-emission-based valence band maxima following excitation of the Li K exciton are influenced not only by this phonon relaxation but also by the necessity to promote the initially-excited electron into another level when the valence electron recombines with the hole. The measured promotion energies are compared to the experimentally determined core-exciton binding energies.

When sufficient electrical power P is dissipated in a thin metal film at millikelvin temperatures, the electrons can be driven far out of thermal equilibrium with the phonons. For uniform power dissipation in a volume Ω we show that the electrons attain a steady-state temperature ${\mathit{T}}_{\mathit{e}}$=(P/ΣΩ+${\mathit{T}}_{\mathit{p}}^5$${)}^{1/5}$, where ${\mathit{T}}_{\mathit{p}}$ is the phonon temperature and Σ is a parameter involving the electron-phonon coupling. We have used a sensitive ammeter based on a dc superconducting quantum interference device (SQUID) to measure the Nyquist current noise in thin films of AuCu as a function of P, and thus inferred ${\mathit{T}}_{\mathit{e}}$. We fitted our data to the theory with the single parameter Σ, and found good agreement for Σ=(2.4±0.6)×${10}^9$ ${\mathrm{Wm}}^{\mathrm{-}3}$ ${\mathrm{K}}^{\mathrm{-}5}$. When we increased the volume of the resistor by attaching a thin-film cooling fin, there was a much smaller increase in ${\mathit{T}}_{\mathit{e}}$ for a given power dissipation in the resistor, in qualitative agreement with a simple model for nonuniform heating. We also measured the flux noise in dc SQUIDs at low temperatures, and found that the white noise was limited by heating of the electrons in the resistive shunts of the Josephson junctions. We were able to reduce these effects substantially by attaching cooling fins to the shunts.