We study electron Compton scattering with an electron microscope by means of a Castaing-Henry filter. In the electron-spectroscopic-diffraction mode the positions of the Compton maxima in the diffraction plane are determined. We find a nearly constant shift of this position with respect to the value given by E=${\mathit{q}}^2$/2. The intensity of Compton-scattered electrons does not peak at the scattering angle predicted by the binary collision mode. The energy dispersion of the Compton profile is well described by E=${\mathit{q}}^2$/2.

The electric-field gradient (EFG) at the Y site in ${\mathrm{YH}}_3$ has been investigated as a function of temperature by ${}_{}{}^{181}{}_{}{}^{}\mathrm{Ta}$ perturbed angular correlation (PAC) measurements. At T≤290 K the EFG shows a strong axial asymmetry, consistent with the ${\mathrm{HoD}}_3$ structure proposed for the heavy-rare-earth trihydrides. At higher temperatures the PAC spectra reveal the existence of a second phase, which is characterized by a much stronger EFG and a higher symmetry. The two phases were found to transform reversibly into each other between 300 and 400 K.

Single crystalline epitaxial Co/Mn superlattices have been grown on a (0001) Ru buffer layer onto mica substrates. The evaporation of a seed layer of 6 Å Mn is necessary to obtain a high-quality epitaxial growth. Reflection high-energy electron diffraction, x-ray diffraction, and ferromagnetic resonance experiments clearly show a modification of the Mn structure, when the thickness of the Mn interlayer increases. The Mn structure switches from a compact phase close to the fcc Mn-γ for few atomic planes, to a less compact one, probably a Laves phase (${\mathrm{MgCu}}_2$) which resembles Mn-α, for larger thicknesses of Mn. This behavior induces a variation of the structure in the Co layers where the stacking changes from fcc to hcp. Up to six Mn atomic planes, the Mn layers being highly strained, the stabilization of the fcc Mn and Co metastable structures occurs via elastic and chemical interactions. For larger Mn thicknesses, there is a trade-off between reduced strains and a higher density of epitaxial dislocations, leading to a lower coherence between the Mn and Co layers. This leads the Mn and Co to approach their bulk structure, Mn-α and hcp, respectively. However, the chemical interactions between Co and Mn favor the fcc Co stacking and consequently create a large density of stacking faults in the hcp Co, even when the Mn is no longer close packed.

Ab initio periodic Hartree-Fock (HF) theory was used to determine the elastic constants and selected phonon frequencies of bulk MgO; the accuracy of a posteriori correlation corrections to the periodic Hartree-Fock calculations is also discussed. Inclusion of diffuse atomic orbitals in the MgO basis was necessary to accurately describe elastic distortions and phonon vibrations of the solid. The computed HF lattice constant (4.195 Å) agrees with experiment (4.19 Å) and the elastic constants are within +15% of the observed values. Correlation corrections to these energetics shorten the lattice parameter to 4.09 Å and further stiffen the elastic constants. The HF phonon frequencies at the (Γ, X, and L) points in the Brillouin zone were within 15% of experiment and the correlation corrections softened the modes improving agreement with experiment. These data will be used to parametrize electrostatic shell models of MgO.

The spin-lattice interaction for ${\mathrm{Mn}}^{2+}$ ions as substitutional impurities in ${\mathrm{CaCO}}_3$ has been studied by electron-paramagnetic-resonance (EPR) experiments in crystals under externally applied uniaxial stresses. The shifts of the EPR lines of ${\mathrm{Mn}}^{2+}$ were measured as a function of the stress applied along the three different directions that accept uniaxial forces in this crystal. The values of six second-order spin-lattice coefficients for the ${\mathrm{Mn}}^{2+}$ site were obtained from the data. We also obtained two conditions relating linear combinations of the other four second-order coefficients. Contributions to the spin-lattice interaction of terms of fourth order in the effective spin have been detected. Our results are used to discuss the effect on the EPR spectrum of the different distortions of the carbonate ions surrounding the ${\mathrm{Mn}}^{2+}$.

The crystallographic structure of the ${\mathrm{Ni}}^{1+}$/${\mathrm{Ni}}^{2+}$ mixed valence ${\mathrm{Nd}}_{1.8}$${\mathrm{Sr}}_{0.2}$${\mathrm{NiO}}_{3.72}$ compound has been investigated over a wide temperature range (from 16 K to room temperature) by using neutron-powder diffraction. In contrast with stoichiometric ${\mathrm{Nd}}_2$${\mathrm{NiO}}_{4.00}$, the symmetry of this new oxygen-defective phase is monoclinic, and it can be described by the B112/n(C2/c) space group in the whole range of temperatures. Between 200 and 16 K, a smooth decrease of the monoclinic strain s=2(b sinγ-a)/(a+b sinγ) has been observed, which presents the same characteristics as the Bmab→P$4_2$/ncm first-order structural phase transition reported for ${\mathrm{Nd}}_2$${\mathrm{NiO}}_{4.00}$ at ${\mathit{T}}_{\mathit{c}1}$≊130 K. The oxygen vacancies are concentrated in the basal plane of the ${\mathrm{NiO}}_6$ octahedra along the [110] crystallographic direction and, in consequence, some of the Ni atoms are expected to be in square planar coordination. On the other hand, the strong additional broadening observed in the (hhl) reflections indicates the existence of microstrains, whose origin lies in the local fluctuations of the oxygen content along the ordering direction of the vacancies.

An experimental study of the Larmor frequency dependence of the proton spin-lattice relaxation time [${\mathit{T}}_1$(${\nu }_{\mathit{L}}$)] was carried out in the range of ${\nu }_{\mathit{L}}$=7 to 30 MHz. Two relaxation times describing the evolution of the longitudinal magnetization have been found to have different behaviors. One of them has been assigned to the relaxation of the protons belonging to the alkyl chain -${\mathit{T}}_1^{\mathrm{alkyl}}$(${\nu }_{\mathit{L}}$)-. The ${\mathit{T}}_1^{\mathrm{alkyl}}$(${\nu }_{\mathit{L}}$) dispersion behaves normally without signs of ${}_{}{}^1{}_{}{}^{}\mathrm{-}_{}^{35}{}_{}{}^{}$Cl cross relaxation. The second one, called ${\mathit{T}}_{\mathit{a}\mathrm{-}\mathit{c}}$, has a partial contribution from the protons bonded to the molecular core and those of the alkyl chain. Well-defined ${}_{}{}^1{}_{}{}^{}\mathrm{-}_{}^{35}{}_{}{}^{}$Cl quadrupole dips (QD) have been observed in the ${\mathit{T}}_{\mathit{a}\mathrm{-}\mathit{c}}$(ν) dispersion curve, due to a cross-relaxation process between the core protons and the chlorine nucleus of the molecule. The shape of the dips are not symmetric, and cannot be fitted by Gaussian or Lorentzian functions. Computer simulations with asymmetric functions previously used in line-shape analysis of incommensurate systems and organic glasses are discussed.

This paper presents a theoretical model for studying the dynamics of ordering in alloys which exhibit modulated phases. The model is different from the standard time-dependent Ginzburg-Landau description of the evolution of a nonconserved order parameter and resembles the Swift-Hohenberg model. The early-stage growth kinetics is analyzed and compared to the Cahn-Hilliard theory of continuous ordering. The effects of nonlinearities on the growth kinetics are discussed qualitatively and it is shown that the presence of an underlying elastic lattice introduces qualitatively interesting effects. A lattice Hamiltonian capable of describing these effects and suitable for carrying out simulations of the growth kinetics is also constructed.

The ${\mathrm{Al}}_{65}$${\mathrm{Cu}}_{20}$${\mathrm{Ru}}_{15}$ icosahedral alloy was studied by ${}_{}{}^{27}{}_{}{}^{}\mathrm{Al}$ nuclear magnetic resonance from 150 to 1110 K. The Knight shift of the unresolved resonance line was observed to significantly increase above 500 K. This uncommon temperature dependence of the Knight shift is interpreted in terms of the presence of a pseudogap at the Fermi level. The spin-lattice relaxation rate deviates from the linear temperature dependence of Korringa relaxation below 500 K, and above 500 K it is dominated by a thermally activated process with a small activation energy of 0.48 eV. This energy is distinctly different from the activation energy observed in simple metallic alloys.

The effect of spin fluctuations on the resistivity of ${\mathrm{Fe}}_{\mathit{x}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Zr}}_2$ metallic glasses was studied in the temperature range 4.2–80 K. After correcting for superconductivity and quantum interference effects on the resistivity at very low temperatures, the spin-fluctuation resistivity (${\mathrm{\rho }}_{\mathrm{sf}}$) as a function of temperature was obtained. It was found that ${\mathrm{\rho }}_{\mathrm{sf}}$ varies as ${\mathit{T}}^2$ at very low temperatures, and as T at higher temperatures in agreement with the predictions of both the two-band model of Kaiser and Doniach and the one-band model of Rivier and Zlatic. The one-band model of Rivier and Zlatic, however, gave an excellent description of ${\mathrm{\rho }}_{\mathrm{sf}}$(T) over the entire temperature range (4.2–80 K). The high-temperature saturation limit of ${\mathrm{\rho }}_{\mathrm{sf}}$ was found to vary quadratically with the Fe content of the glass. The spin-fluctuation temperatures varied between 15 and 92 K for x=1 and x=0.4, respectively, for the one-band model.

Ternary alloy energetics are examined for substitutional systems by a formal cluster expansion. In contrast to the binary-alloy problem, several distinct basis sets are possible for the ternary cluster expansion. Several of these sets of ternary basis functions are examined and compared, and relationships are derived between the expansion coefficients, or effective cluster interactions, expressed in various bases. The method of direct configurational averaging (DCA) (based on a tight-binding, linearized muffin-tin-orbital Hamiltonian) is extended to treat ternary alloy systems. Using the DCA, ternary, fcc-based effective pair and triplet interactions are computed for the Rh-V-Ti, Pd-Rh-V, and Ag-Pd-Rh systems, and convergence of the expansion is examined. By combining the cluster expansion with the results of the DCA computations, formation energies are obtained for the completely disordered state as a function of alloy composition. Both pair and triplet interactions are seen to be crucial towards obtaining quantitatively converged energetics.

The effect of the multiple elastic scattering of electrons in disordered media on the features of a type of weak localization is studied. This type of weak localization necessarily involves inelastic scattering and is the localization of electrons scattered inelastically with a fixed value of the lost energy. Earlier the possibility of such a localization was established in a model when electrons underwent multiple scattering through small angles and single scattering through a large angle. In this paper, an attempt is made to show that multiple elastic scattering of electrons through arbitrary angles does not destroy this type of weak localization. Also, the influence of the surface on the angular distribution of the scattering of electrons from a semi-infinite disordered system under conditions of quantum transport is treated. The results obtained can be applied to the theory of Raman scattering of light by randomly distributed scatterers.

This work deals with the simulation of excitonic absorption line shapes (ALS’s) of random aromatic chains based upon a description in terms of a Frenkel Hamiltonian. By averaging the ALS’s of single configurations over an ensemble of chain conformations, generated by means of a Pivot algorithm on a square lattice, a type of off-diagonal disorder is examined, which originates from the randomness of non-nearest-neighbor contacts in polymer coils. Dimensions of up to several thousand chromophores are studied by utilizing the recursive residue generation method [A. Nauts and R. E. Wyatt, Phys. Rev. Lett. 51, 2238 (1983)], which resorts to the Lanczos tridiagonalization. With this method ALS’s of pure donor chains and binary chains are investigated and the effects of diagonal and off-diagonal disorder, as well as convergence pattern of the recursive residue generation method in application to such systems is demonstrated.

We present heats of formation of ordered fcc transition-metal binary alloys as predicted by the non-self-consistent electron density functional corrected effective-medium (CEM) theory. Two forms of the CEM theory are used. The first includes explicit numerical evaluation of the kinetic-exchange-correlation energy functional, a computationally intensive task. The second approximates this term as a local function of electron density and leads to a formulation that is very similar to the embedded-atom method. Calculated values of the energies of formation from the two forms of the CEM method are compared to each other and to experimental data. We demonstrate that both forms of the CEM theory provide formation energies accurate to within about 0.05–0.1 eV/atom (5–10 kJ/mol) for 3d and 4d binary alloys.

Structural and dynamical properties of the anharmonic Fibonacci-chain quasicrystal have been studied using the unsymmetrized self-consistent-field method. Nonlinear integral equations forming the basis of this method are reduced to a set of transcendental or algebraic equations for the moments of one-particle functions. In the lowest orders of anharmonicity, these equations have been solved for an arbitrary one-dimensional model of various atoms interacting with the nearest neighbors. Herewith, the thermal expansion, the variances of the atomic positions, and high-order moments have been expressed in terms of the second, third, and fourth derivatives of the interatomic potentials. Quantum corrections have been calculated as well. The general results have been applied to the system with two kinds of atoms alternating with each other in a Fibonacci sequence. The interatomic distances, thermal expansions, and effective amplitudes of anharmonic atomic vibrations are calculated versus the temperature. The influence of anharmonicity on the thermodynamic functions of the Fibonacci chain is also discussed as well as possible applications to some real solids.

We investigate the localization of electrons on a ladder-shaped quasi-one-dimensional network of clean wires, with a quenched random magnetic flux across each of its square plaquettes. In the weak-disorder regime, the localization length ξ is much larger than the side of the plaquettes. Using perturbative analytic techniques, we derive scaling laws of the form ξ∼1/${\mathit{w}}^{\mathrm{\alpha }}$, with w being the width of magnetic disorder. The critical exponent α assumes different values in various energy ranges: α=4 when only one channel is open, α=2 when both channels are open, α=1 around external and internal band edges. The corresponding scaling functions and amplitudes are accurately determined by numerical simulations. Magnetic disorder and potential disorder thus pertain to different universality classes.

The experimental curves giving the temperature dependence of the thermal conductivity of glassy selenium are considered in detail. The observed behavior can be taken into account quantitatively if the densities of states for short-wavelength phonons as well as for low-energy librations arising from computer simulations are used for the calculations. In particular, it is shown that the lowest frequency excitations of a chain of selenium atoms can give due account of the plateau observed at temperatures about 2–10 K. The implications of the present findings for the current debate regarding the mechanisms for thermal transport in glasses are finally discussed.

The elastic intensity (EI) and the Debye-Waller factor (DWF) is calculated in the classical approximation for the anharmonic potentials of the soft-potential model. The two quantities are different for Hamiltonian motion. If frictional forces are included, the difference between EI and DWF may be found in a quasielastic line. The EI and the DWF are calculated in a small-momentum approximation, for realistic parameters of the soft-potential model. A configurational average is performed over distributions of the coefficients of the soft potentials that are typical for glassy systems. An observable difference between the EI and DWF is found for these systems.

We present a detailed theoretical investigation of the structure and the dynamical properties of icosahedral Al-Cu-Li alloys. Structural models of the quasicrystalline structure have been constructed on the basis of three-dimensional Penrose- and canonical-cell tilings. We show that the periodic approximants (up to τ∼5/3) are stable under molecular-dynamics (MD) annealing with realistic pair forces and that the partial-pair-correlation functions, powder- and single-crystal and neutron- and x-ray diffraction data are in good agreement with experiment, with a slight preference for the canonical-cell models. The vibrational spectrum (spectral functions, densities of state) and inelastic neutron-scattering intensities have been calculated using a real-space recursion technique and the same interatomic forces as used in the MD annealing. Well-defined propagating longitudinal and transverse acoustic modes are found in the vicinity of quasiperiodically distributed special points in k space, the Γ points of the reciprocal quasilattice. At higher energy, a hierarchy of stationary modes is predicted around other high-symmetry points corresponding to quasi-Brillouin-zone boundaries. For the propagating low-energy modes, the predictions are in very good agreement with inelastic neutron-scattering experiments on single crystals of i-Al-Cu-Li. However, certain discrepancies subsist between the predicted vibrational density of states (DOS) and the experiments with powdered samples. Quite generally, the calculated DOS shows more structure (with again only minor differences between the Penrose- and canonical-cell models) than the experimental result. This is probably to be attributed to the influence of phason disorder in the real quasicrystals.

The control of the transmission of electrons along a polyene chain by an impurity is studied. A semiconducting chain of atoms is modeled by a tight-binding Hamiltonian with alternating bonds into which the impurity is inserted. The energy-dependent transmission coefficient t and the change in the chain’s electronic density of states are extracted from the transfer matrix across the impurity. The switching ability of the impurity is considered by altering either its site energy or one of its bonds with the conducting chain. Invoking the Joyce-Dixon approximation enables us to write the Azbel’s energy-dependent effective-transmission coefficient ${\mathit{t}}_{\mathit{e}\mathit{f}\mathit{f}}$, in a for m where its Fermi level is temperature dependent, so that thermal effects can be investigated. The effects of site energy, impurity coupling, and bond ratio of both the impurity and chain are investigated by comparing their ability to alter the energy-averaged effective transmission coefficient as a function of either switching parameter to the ideal switch—a step function.