We present an Fe and Ni K-edge extended x-ray-absorption fine structure study of the L$1_2$-ordered ternary alloy ${\mathrm{Ni}}_{75}$(${\mathrm{Al}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Fe}}_{\mathit{x}}$${)}_{25}$. We find that Fe enters predominantly the Al sublattice of ${\mathrm{Ni}}_3$Al (γ’ phase). The distribution of Fe and Al atoms on this sublattice follows the expected distribution for an L$1_2$ structure. The Fe nearest-neighbor distances are consistent with the measured lattice parameters. The first coordination shell around Ni is composed of Ni, Fe, and Al atoms. The Ni-Ni and Ni-Fe bond lengths are consistent with diffraction data. On the other hand, an unexpected strong (≊0.2 Å) decrease of the Ni-Al distance with increasing Fe content is detected, the Ni-Al bond being also characterized by a large disorder. This result provides a local structural explanation to the large lattice strains recently measured in similar samples by x-ray diffraction.

We study the electronic properties of a deterministically aperiodic chain where the impurities are located at sites corresponding to prime numbers. We consider a one-dimensional tight-binding model with a discrete potential and calculate the wave functions, resistance, and density of states. It is found that both localized and extended states exist and that the wave function at a mobility edge is multifractal. We also discuss the energy spectrum and the physical origin of the extended states.

Non-Lorentzian depolarized Rayleigh-scattering spectra have been clearly observed by Fabry-Perot interferometry in bulk homopolymers, polyisoprene and 1,4-polybutadiene, far above the glass-transition temperature. We have developed two alternative phenomenological methods for analyzing this kind of spectra. First of all we have used a scattering model function based on the Havriliak-Negami relaxation function which, in fact, is compatible with a Kohlrausch-Williams-Watts relaxation function in the time domain. In the case of polyisoprene, the results obtained by this method are in good agreement with the obtained ones through other different relaxation techniques when a similar method of analysis is applied. Besides, we have extended to the frequency domain the contin algorithm which calculates an apparent distribution function of relaxation times from nonexponential time domain relaxation functions. The results obtained as well as the applicability of both methods are discussed.

We consider the tight-binding Hamiltonian on strongly modulated aperiodic chains (e.g., quasiperiodic, self-similar, random). The site energies are distributed according to a given binary sequence (${\mathit{V}}_{\mathit{n}}$=±V), the hopping integral t being a constant. We investigate the strong-modulation regime (V≫t) by means of degenerate perturbation theory. We thus derive systematic expansions in powers of t/V for the spectrum, the density of states, and other quantities of interest, like the total electronic binding energy. This investigation goes beyond the molecular approximation; it amounts to exploring recursively the pattern of resonances between molecular states along the structure, which can be either ‘‘local’’ or ‘‘itinerant.’’ The approach is worked out in detail on several classical examples of self-similar chains: Fibonacci, Thue-Morse, period doubling. The hierarchy of gaps thus obtained is put in perspective with the rigorous gap-labeling theory; the quantitative predictions for the density of states and the binding energy are extensively compared with numerical data.

In this paper we report the dielectric properties of ${\mathrm{Ba}}_{1\mathrm{-}\mathit{x}}$${\mathrm{K}}_{\mathit{x}}$${\mathrm{BiO}}_{3\mathrm{-}\mathrm{\delta }}$ (x=0, 0.2, 0.4, 0.625, and 0.75) glasses at different temperatures (80–450 K) and frequencies (1–10 kHz). Distinct anomalies (large increases) in the dielectric constant (ɛ’) and dielectric loss (ɛ’’) are observed for all the glasses between 297 and 320 K. A similar anomaly showing a change from semiconducting to metallic behavior is also observed in the electrical conductivities around the same temperature region. The anomalies observed around this temperature are considered to be associated with the orientational motion of the ${\mathrm{BiO}}_3$ or ${\mathrm{BiO}}_6$ structural units present in the glasses. Such motions might give rise to structural instability and different kinds of phase transitions including a superconducting transition in the crystalline phases.

We have found a simple criterion which allows for the straightforward determination of the order-disorder critical temperatures. The method reproduces exactly known results for the two-dimensional Ising, Potts, and Z(N<5) models. It also works for the Ising model on the triangular lattice. For systems that are not self-dual, our proposition remains an unproven conjecture. It predicts ${\mathrm{\beta }}_{\mathit{c}}$=0.2656. . . for the two coupled layers of Ising spins. The critical temperature of the three-dimensional Ising model is related to the free energy of the two-layer Ising system.

The photoconductivity and dark conductivity of carbon aerogels with various heat-treatment temperatures, ${\mathit{T}}_{\mathrm{HT}}$, and densities are studied. Unlike the trend for as-prepared samples, the magnitude of the photoconductivity for heat-treated (${\mathit{T}}_{\mathrm{HT}}$>1050 °C) carbon aerogel samples decreaes as the sample density and heat-treatment temperature increase. We propose a model which quantitatively explains the observed temperature dependence of the photoconductivity data. Based on this model, we attribute the observed photoconductivity to photoholes present in the system. The number of defect states and their corresponding energy decrease with increasing ${\mathit{T}}_{\mathrm{HT}}$ and increasing density. The decrease in the energy is attributed to a decrease in the screening effect of the defects on trapped charges. The effect of increasing the heat-treatment temperature causes the structure of the carbon aerogel to become more graphitic, especially for aerogel samples with higher mass density.

A fiber-bundle model in (1+1) dimensions for the breaking of fibrous composite matrix is introduced. The model consists of N parallel fibers fixed in two plates. When one of the plates is pulled in the direction parallel to the fibers, these can be broken with a probability that depends on their elastic energy. The mechanism of rupture is simulated by the breaking of neighboring fibers that can generate random cracks spreading up through the system. Due to the simplicity of the model we have virtually no computational limitation. The model is sensitive to external conditions such as temperature and traction velocity. The energy versus temperature behavior, the diagrams of stress versus strain, and the histograms of the frequency versus the size of cracks are obtained.

We report on anelastic-relaxation measurements of ${\mathrm{NbZr}}_{0.0045}$${\mathrm{H}}_{\mathit{x}}$ single crystals. The elastic energy loss below 50 K is ascribable to tunnel systems of H trapped by substitutional Zr, and consists of two components. A small peak near 2 K is observed under 〈100〉 extensional and 〈111〉 torsional vibrations but not under 〈100〉 torsional and 〈111〉 extensional vibrations. In terms of the symmetric stress components of the cubic system, relaxation is observed under 〈110〉 shear (E type) but not under 〈100〉 shear (T type) and hydrostatic (A type) stresses. The major component, extending up to 50 K, has the same dependence on the stress symmetry, although the reduction of its intensity under the 〈100〉 T and 〈111〉 E vibrations is smaller. The influence of stress on tunneling systems with various geometries is discussed, and it is shown that the observed dependence of the relaxation strength on the stress symmetry can be accounted for by tunneling over four equivalent sites.

We have studied the ultrasonic pulses that are generated when a picosecond light pulse is absorbed in a metallic film. The pulse shape is influenced by the distance that the hot electrons diffuse before losing their energy to the lattice. We show that this method can be used to test theories of the electron-phonon and electron-electron scattering rates. The results are in excellent agreement with a calculation in which the nonthermal character of the electron distribution is taken into account.

In a previous paper we studied the dynamics of self-interstitial atoms in bcc metals and calculated the local density of states of the 〈110〉 dumbbell in Mo using a Green’s-function method. In this paper, we calculate the local density of states of neighbors of the dumbbell which affect the properties of the irradiated metal significantly. The local density of states of the neighbors show the same well-known resonance and localized modes shown by the dumbbell spectrum. However, the amplitude of vibrations of neighbors in the resonant modes is, in general, very small so that only the spectrum of the closest neighbor in the (11¯0) plane shows clearly all the resonant modes found with the dumbbell spectrum. The result is discussed in the light of Mössbauer measurements of irradiated ${\mathit{Mo}}^{57}$Co. Contrary to earlier expectations, the vibrational modes of the Mössbauer atom located at a substitutional site nearest to a dumbbell atom do not explain the observed reduction in the Debye-Waller factor.

The experimental observations for the nonpropagation solitons with both spatial phase match and mismatch have been reported in a one-dimensional nonlinear lattice that is damped and vertically driven. Theoretically, using the method of multiple-scale solution, we develop a nonlinear Schrödinger equation for the experimental results including the kinks found by Denardo et al. Furthermore, under our order estimation the generating conditions for soliton and kink have been discussed. According to the generating conditions, the modes of the self-localized structures are determined closely by the intrinsic properties of the lattice rather than the driven parameters of the shake table.

We present an extensive investigation of the normal modes of vibration of a prototype hierarchical continuous system, consisting of a Cantor-like sequence of piezoelectric and resin elements. From a detailed analysis of the density of states, displacement profiles, and eigenvalue-spacing distributions, evidence is found for the existence of multiple fracton and phonon regimes. The role of resonant modes and the effect of disorder, which may be of primary importance in real systems, are also discussed in detail.

Magnetic properties of amorphous Co-Y alloys are investigated on the basis of a finite-temperature theory of amorphous magnetic alloys. The numerical results show that the enhanced Curie temperatures (${\mathit{T}}_{\mathit{C}}$) increase linearly from a critical concentration of ferromagnetic instability towards amorphous pure Co with increasing Co concentration. The enhancement of ${\mathit{T}}_{\mathit{C}}$ beyond 90 at. % Co is caused by the magnetic energy gain due to the structural disorder, while it is dominated around 60 at. % Co by the change of Co-Y neighboring pairs and related change of coordination number of Co atoms. The calculated magnetization vs concentration curves show good agreement with the experimental data. They are explained by the 3d-4d hybridization in the electronic states. It is predicted that the effective Bohr magneton numbers show a minimum around 55 at. % Co because of the occurrence of weak ferromagnetism.

Cu(11n) surfaces vicinal to Cu(001) are characterized by (001) terraces separated by monoatomic steps. Growth of thin epitaxial magnetic films on such terraced substrates, e.g., Co/Cu(1 1 13), leads to unexpected magnetic anisotropy associated with the surface steps. This anisotropy remains largely unexplained, and is of potential technological importance. We have studied the anisotropy of fcc Co/Cu (1 1 13) films using in-situ magneto-optical Kerr effect (MOKE), and find an in-plane uniaxial anisotropy favoring magnetization parallel to the steps. The strength of this anisotropy decreases with increasing film thickness. In this paper we focus on the interpretation of this anisotropy in terms of the Néel model. The Néel anisotropy energy for such a film has the form ${\mathit{E}}_{\mathrm{film}}$= ${\mathit{E}}_{\mathrm{bulk}}$-2${\mathit{E}}_{\mathrm{surface}}$/t-2${\mathit{E}}_{\mathrm{step}}$/(td), where t and d are film thickness and terrace width, respectively. The last term includes contributions from sites at both the upper edge and inner corner of the step and makes the dominant contribution to the step-induced anisotropy. This model properly accounts for the preferred direction of magnetization parallel to the steps in Co/Cu(1 1 13). bcc Fe/W vicinal to (001) shows a preference for in-plane magnetization perpendicular to the steps for t<2.5 monolayers. The Néel model also predicts this anisotropy for a bcc Fe film provided the magnetization lies in the film plane, not in the (001) plane.

High-pressure magnetization and neutron-diffraction experiments have been performed on CeSb under hydrostatic pressure up to 21 kbar in the temperature range 3–32 K. The magnetic ordering in CeSb is very sensitive to hydrostatic pressure. The Néel temperature ${\mathit{T}}_{\mathit{N}}$ changes from 16 K at P=1 bar to 31 K at P=21 kbar. The stability ranges in temperature of the so-called antiferroparamagnetic (AFP) phases decrease continuously with pressure and disappear completely above P=10 kbar. At P=2.5 kbar, the type-I phase stabilizes at ${\mathit{T}}_{\mathit{N}}$ and the transition at ${\mathit{T}}_{\mathit{N}}$ crosses over from first to second order giving rise to the critical end point at P=2.5 kbar, T=18 K. At P=10 kbar two phases with magnetic ordering sequences (++--+-) and (++--+--++-) of the ferromagnetic (010) planes are successively observed between type-I and type-IA phases. The results have been discussed in terms of the p-f mixing model.

Ferromagnetic resonance at 9.4 GHz has been used to characterize several samples of Fe single-crystal films and Fe/Cu multilayers prepared by electron-beam deposition on HF-etched, hydrogen-terminated Si(111). Resonance field data as a function of the field orientation in the film plane confirmed that Fe films and Fe/Cu multilayers grow epitaxially on Si(111) with excellent crystallinity in the [111] orientation. Moreover, we have found that the (111) plane is unique among the principal planes in cubic crystals, because it allows precise measurements of the small second-order and in-plane anisotropies as well as misorientations from the crystal plane.

We have investigated the effect of anion disorder on the spin-density wave induced by the magnetic field (FISDW) in (TMTSF${)}_2$${\mathrm{ClO}}_4$ (the TMTSF is the tetramethyltetraselenafulvalene molecule). Specific heat and magnetocaloric-effect measurements have been performed under a magnetic field for various cooling rates across the ordering transition of the ${\mathrm{ClO}}_4$ anions. The respective variations of the critical temperature and of the jump of the electronic specific heat at the metal-SDW transition are compared with the universal pair-breaking behavior of a Bardeen-Cooper-Schrieffer superconductor in the presence of magnetic impurities. The field-induced spin-density wave (FISDW), in the presence of nonmagnetic anion disorder, qualitatively behaves as the former but exhibits a stronger pair-breaking effect. The anion disorder is also shown to be responsible for the change in the criticality of the metal-SDW phase transition: a peculiar point in the phase diagram is found to evolve from tetracritical to bicritical as the anion disorder is increased. These behaviors are discussed in light of available models for the spin-density wave induced by the magnetic field.

Measurements of the magnetic moment and susceptibility of diamond, graphite along and perpendicular to the c axis, ${\mathrm{C}}_{60}$, and carbon nanotubes are reported as a function of magnetic field (0<H<55 kOe) and temperature (4.5 K<T<310 K). The data on ${\mathrm{C}}_{60}$ are diamagnetic, confirm previous results, and show the influence of the phase transition near 260 K. The data on the nanotubes were taken on a sample containing a distribution of tubes with diameters in the 2–10 nm range, and reveal a mostly diamagnetic behavior with a magnetic field and temperature dependence different from graphite.

The low-temperature specific heat of ${\mathrm{CeNi}}_{\mathit{x}}$${\mathrm{Pt}}_{1\mathrm{-}\mathit{x}}$ compounds has been investigated from x=0 to x=1. These compounds evolve from a Pauli paramagnet (CeNi) to a Kondo ferromagnet (CePt), with a crossover around x=0.90. The reduction of the magnetic entropy at ${\mathit{T}}_{\mathit{C}}$ has been used to illustrate the increasing 4f-sd hybridization with increasing Ni content, related to a decrease of the cell volume. This effect has been confirmed by the results obtained on Y and La diluted compounds which have opposite volume effects. The relations between the jump in the specific heat at ${\mathit{T}}_{\mathit{C}}$, the magnetic moment, and the ratio ${\mathit{T}}_{\mathit{K}}$/${\mathit{T}}_{\mathit{C}}$ have been analyzed. The values of the electronic heat capacity γ vary as ≊1/${\mathit{T}}_{\mathit{K}}$ in the nonmagnetically ordered compounds in accord with the standard theory. However in the magnetically ordered compounds γ increases with ${\mathit{T}}_{\mathit{K}}$ in accord with the resonance-level model developed by Bredl et al. [Z. Phys. B 29, 327 (1978)].