Structural defects in ${\mathrm{SiO}}_2$ glasses implanted with ${\mathrm{Li}}^{+}$, ${\mathrm{N}}^{+}$, ${\mathrm{O}}^{+}$, ${\mathrm{F}}^{+}$, ${\mathrm{Si}}^{+}$, and ${\mathrm{P}}^{+}$ ions were examined by vacuum-ultraviolet-absorption and electron-paramagnetic-resonance spectroscopies as well as thermal-gas-release analysis. The chemical interaction of implanted ions with substrate structure was considered on the basis of the obtained results. It is found that the type of predominant defects is controlled by the electronegative nature of implants. Silicon-silicon homobonds, which are oxygen-vacancy-type defects, are produced by electropositive implants (i.e., Li, P, and Si) at concentrations comparable to those of the implants. On the other hand, in the case of electronegative implants (F and O) ${\mathrm{O}}_2$ molecules and peroxy radicals (POR), both of which may be regarded as oxygen-interstitial-type defects, are the major defects and the total concentrations of these two defects are comparable to implant concentrations. These results indicate that chemical interaction of implanted ions with ${\mathrm{SiO}}_2$ is primarily controlled by the electronegative nature of implants. Electropositive implants (M) react chemically with oxygen atoms in the substrate structure to form M-O bonds, leaving Si-Si bonds. Electronegative implants (A) react chemically with silicon atoms to form Si-A bonds and oxygen atoms recoiled with implants combine with each other to form ${\mathrm{O}}_2$ molecules or react with the silica-network structure to form POR’s. Concentrations of these predominant defects relative to implants can be used quantitatively to describe the strength of chemical interactions. When the chemical interaction is strong, both concentrations are comparable. On the other hand, when the chemical interaction is weak, concentrations of these defects are much smaller than those of implants because the major fraction of implants occur in a neutral state without forming chemical bonds with constituents of the substrate. Nitrogen is an example of this category and the major fraction of implanted nitrogen atoms are present as ${\mathrm{N}}_2$ molecules.

Frozen aqueous solutions of sodium hydroxide doped with a small amount of paramagnetic ${}_{}{}^{65}{}_{}{}^{}\mathrm{Cu}_{}^{2+}{}_{}{}^{}$ ions have been investigated in the amorphous and polycrystalline phases by means of continuous-wave and pulsed-electron-paramagnetic-resonance spectroscopy. The continuous-wave spectra of the amorphous phase recorded at 77 K showed a significant g and A strain if compared to that of the polycrystalline phase, indicating random distortions of the local environment of the paramagnetic ions. On the other hand, spin-lattice relaxation rates obtained from magnetization-recovery data, put into evidence dynamical anomalies typical of the glassy state of matter. The low-temperature (from 2 to 5 K) behavior of the relaxation rates in the amorphous phase is consistent with the existence of two-level states coupled to the paramagnetic centers, while the behavior of the Raman relaxation process (from 5 to 100 K) indicates an anomaly in the density of vibrational states, probably due to mode-localization phenomena.

${}_{}{}^7{}_{}{}^{}\mathrm{Li}$ nuclear spin-lattice relaxation (NSR) and Li-ion electrical-conductivity relaxation (ECR) data of (${\mathrm{Li}}_2$S${)}_{0.56}$(${\mathrm{SiS}}_2$${)}_{0.44}$, a glassy fast-ionic conductor, have been reanalyzed. Both NSR and ECR data are fitted by using a stretched exponential, exp[-(t/${\mathrm{\tau }}^{\mathrm{*}}$${)}^{\mathrm{\beta }}$], for the correlation function in each relaxation. It is found that ${\mathrm{\tau }}_{\mathit{s}}^{\mathrm{*}}$ for NSR can be several orders of magnitude longer than ${\mathrm{\tau }}_{\mathrm{\sigma }}^{\mathrm{*}}$ for ECR at the same temperature, ${\mathrm{\beta }}_{\mathit{s}}$ and ${\mathrm{\beta }}_{\mathrm{\sigma }}$ are not the same with ${\mathrm{\beta }}_{\mathit{s}}$<${\mathrm{\beta }}_{\mathrm{\sigma }}$, and the activation energy ${\mathit{E}}_{\mathit{s}}^{\mathrm{*}}$ of ${\mathrm{\tau }}_{\mathit{s}}^{\mathrm{*}}$ is significantly larger than ${\mathit{E}}_{\mathrm{\sigma }}^{\mathrm{*}}$ of ${\mathrm{\tau }}_{\mathrm{\sigma }}^{\mathrm{*}}$. In this contribution we show how these pronounced differences of the parameters in the correlation functions of NSR and ECR are explained in the framework of the coupling model. An additional predicted relation: ${\mathrm{\beta }}_{\mathit{s}}$${\mathit{E}}_{\mathit{s}}^{\mathrm{*}}$=${\mathrm{\beta }}_{\mathrm{\sigma }}$${\mathit{E}}_{\mathrm{\sigma }}^{\mathrm{*}}$ is found to be consistent with the experimental data and the common value of the two products is identified naturally with the true single ion activation energy. The constraints that these experimental data impose on any viable theory of the dynamics of carriers in glassy ionic conductors are discussed.

We have measured the low-temperature specific heat C and thermal conductivity κ (1.5 K≤T≤30 K) of amorphous ${\mathrm{As}}_{\mathit{x}}$${\mathrm{Se}}_{1\mathrm{-}\mathit{x}}$ in the concentration range of 0≤x≤0.72. The maximum in C/${\mathit{T}}^3$ and the plateau in κ display consistent but nonmonotonic concentration dependences, similar to the earlier specific-heat results for the same samples below 1 K. Our results give unambiguous evidence that the behavior below 1 K (attributed to tunneling states) and the intermediate-temperature anomalies are intimately related, hence, they might have a common structural origin. Possible microscopic mechanisms are discussed, suggesting two different types of low-energy excitations. The specific-heat maximum in C/${\mathit{T}}^3$ can be scaled by two parameters, and this feature can be generalized to other amorphous systems. Finally, our results are discussed in terms of current theoretical models.

Numerical simulations of large-amplitude cellular dynamics during thin-film, directional solidification of a succinonitrile-acetone alloy demonstrate strongly nonlinear local dynamics that involve oscillations of the deep grooves between the cells. The oscillations begin as breathing modes of the grooves and become more complex, showing period doubling, when the simultaneous dynamics of more than one groove is considered. These oscillations are not predicted by amplitude equations derived for cells near the planar interface. Simulations of the onset of cellular solidification demonstrate the importance of this mechanism to explain the scattered finite-amplitude cells that are observed in experiments near the onset of cellular solidification.

The time-of-flight spectrum of high-frequency phonons with both anharmonic decay and mass-defect (isotope) scattering is studied numerically for silicon. For the quantitative description of the spectrum in the whole time domain (from t≃${\mathit{t}}_{\mathit{b}}$ to t≫${\mathit{t}}_{\mathit{b}}$, where ${\mathit{t}}_{\mathit{b}}$ is the ballistic time of flight), we make the following substantial improvements beyond previous simulations: (1) For the anharmonic decay we consider, the multibranch decay where the energies and propagation directions of the decayed phonons are determined from the energy-momentum conservation of the process. (2) For the isotope scattering polarization-dependent anisotropic scattering is used when we discuss the phonon intensity at a time t close to ${\mathit{t}}_{\mathit{b}}$, otherwise an approximate isotropic scattering is employed. (3) The elastic anisotropy is explicitly incorporated through the group velocity of phonons to locate the positions where the scattering events occur. The simulated phonon intensity versus time of flight compares favorably with recent experiments. Specifically, the exponential decay for late arrival times observed experimentally is well reproduced. We also find the phonon focusing due to elastic anisotropy is crucial to explain the shape of the phonon intensity at t≃${\mathit{t}}_{\mathit{b}}$. The size of the source for the ballistic phonons is also studied.

Two-body potentials of the Toda, Born-Mayer, Lennard-Jones, and Morse type are used in a one-dimensional diatomic lattice to demonstrate both analytically and numerically that stable anharmonic gap modes are a general feature of these perfect lattices. Associated with each vibrational mode is a localized dc lattice expansion, which forms an integral part of this dynamical configuration.

The heat capacity of Spectrosil-WF (vitreous silica containing <20 ppm of OH) in the temperature range from 2 to 16 K as well as the Raman measurements are presented. The heat capacity is similar to that of Spectrosil-B (vitreous silica containing 1200 ppm of OH) but is about 20% higher than the Heralux (vitreous silica containing 130–180 ppm of OH). The density of low-frequency vibrational states, g(ν), has been determined. It has been found that the form of g(ν) is nonquadratic, which can be explained on the basis of a soft-potential model. About 110 atoms participate in a soft mode. The Raman and infrared coupling constants have similar dependence on frequency. These are explained on the basis of a single-coupling-constant bond model. It is found that the form of the Raman coupling constant cannot be explained on the basis of Martin-Brenig theory.

The theory for the large acoustic polaron (soliton) in a one-dimensional lattice is extended to a general case with any interatomic (intermolecular) potential including both hardening and softening anharmonicities. The soliton solution is given by general expressions that allow us to calculate the binding energy and the effective mass of the polaron in any particular case. Our results show that the introduction of softening nonlinearity increases both the binding energy of the polaron and its effective mass.

We study the temperature dependence of the low-frequency dielectric properties of the ferroelectric ${\mathrm{LiTaO}}_3$ by generating phonon polaritons with frequencies in the terahertz regime. The phonon polaritons are impulsively excited and phase-sensitively detected with 60-fs laser pulses. The propagation and damping of the polaritons are investigated as a function of frequency and temperature. The experimental results are compared with a quantum-mechanical model for the low-frequency dielectric response of ${\mathrm{LiTaO}}_3$. In this model the lowest-energy ${\mathit{A}}_1$ mode is described with a one-dimensional anharmonic potential in a single unit cell. We find that at 300 K the polariton dispersion and damping are determined by a strong resonance at 6 THz and a weak resonance at 1 THz. The latter resonance is due to a tunneling transition and leads to the observation of phonon-polariton beats. The model predicts that with increasing temperature the strength of the resonance at 6 THz decreases and that a new broad resonance at 3 THz and a central mode arise. The measured polariton dispersion and damping are in quantitative agreement with these theoretical predictions. Thereby we obtain evidence that the ferroelectric phase transition in ${\mathrm{LiTaO}}_3$ takes place without mode softening.

We have studied the magnetization of the antiferromagnet dicalcium ferrite ${\mathrm{Ca}}_2$${\mathrm{Fe}}_2$${\mathrm{O}}_5$ in magnetic fields up to 5.5 T and in the temperature range 5–330 K. Temperature dependences of the weak ferromagnetic moment ${\mathit{m}}_0$ and susceptibilities along the main crystallographic axes have been measured. Effective intrinsic parameters, such as an exchange field and the Dzyaloshinskiy-Moriya field have been obtained as a function of temperature. Possible magnetic structures have been analyzed by means of comprehensive group-theoretical consideration and a phenomenological free energy has been derived. Problems involving hedgehog structure, parallel susceptibility, field-induced spin reorientation, magnitude of the Dzyaloshinskiy-Moriya interaction, and the origin of the weak ferromagnetism are also discussed. It is shown that ${\mathrm{Ca}}_2$${\mathrm{Fe}}_2$${\mathrm{O}}_5$ can be described as two canted antiferromagnets putting one into another with an antiferromagnetic interaction between the weak ferromagnetic components.

Self-consistent electronic-structure calculations have been performed for the ferromagnetic Laves-phase compound ${\mathrm{GdCo}}_2$ and ${\mathrm{GdCo}}_2$${\mathrm{H}}_4$. Experimentally, the magnetic moment for ${\mathrm{GdCo}}_2$ is only a little reduced when the hydride is formed, but the Curie temperature drops from ${\mathit{T}}_{\mathit{c}}$=395 K to ${\mathit{T}}_{\mathit{c}}$=90 K in the hydride. Using parameters that are calculated ab initio (the susceptibility and the exchange coupling) the Curie temperature is calculated from a simple model and found to agree with experiment for ${\mathrm{GdCo}}_2$. For the hydride, the reduction in ${\mathit{T}}_{\mathit{c}}$ is accounted for, but we calculate the Curie temperature to be slightly too large (157 K). By simply scaling the calculated ${\mathit{T}}_{\mathit{c}}$ for ${\mathrm{GdCo}}_2$ and ${\mathrm{GdCo}}_2$${\mathrm{H}}_4$ with the de Gennes factor, the Curie temperature for the other heavy rare-earth cobalt compounds and their hydrides are estimated.

We have measured the temperature dependence of low-field SQUID magnetization for random-mixture graphite intercalation compounds (GICs), stage-2 ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ GICs, stage-2 ${\mathrm{Ni}}_{\mathit{c}}$${\mathrm{Mn}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ GICs, and stage-2 ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Mn}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ GICs. For stage-2 ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ GICs (0≤c≤1), stage-2 ${\mathrm{Ni}}_{\mathit{c}}$${\mathrm{Mn}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ GICs (0.8≤c≤1), and stage-2 ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Mn}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ GICs (0.9≤c≤1) where ferromagnetic intraplanar exchange interactions are dominant within each island in the magnetic intercalate layers, the zero-field-cooled (ZFC) magnetization of these compounds deviates downward from the field-cooled (FC) magnetization below a critical temperature ${\mathit{T}}_{\mathit{c}}$, and shows a broad peak at a temperature ${\mathit{T}}_{\max }$(<${\mathit{T}}_{\mathit{c}}$). This irreversible effect of magnetization indicates that a cluster glass phase appears below ${\mathit{T}}_{\mathit{c}}$. The ferromagnetic spin order is established within each island, forming a ferromagnetic cluster. The spin directions of these ferromagnetic clusters are frozen because of frustrated interisland interactions consisting of both the dipole-dipole interaction between ferromagnetic clusters and antiferromagnetic interplanar exchange interaction between different intercalate layers. The broad peak of ZFC arises as a result of a competition between thermal energy and these frustrated interisland interactions.

Magnetic dichroism is investigated in second-order processes, especially for x-ray-resonance scattering, in a semirelativistic treatment. The matrix element of the cross section is expressed by the Green function of the system for the unoccupied states and by the Fourier-transformed density for the occupied states. The lattice harmonic expansion of the Green function is used, the coefficients of which can be calculated by a band-structure investigation. For cubic systems and resonances at the K and L edge, respectively, the matrix elements are evaluated for dipole and quadrupole contributions. It is analyzed how to measure different contributions independently by observing Bragg peaks in and outside the scattering plane and by changing the polarization of the incoming and of the scattered photon.

Analytic steady-state solutions of the density-matrix equation are derived by means of a formalism suitable for multilevel spin systems in dilute paramagnetic solids at high temperature and subjected to several strong fields. It is shown that solid-state masers without inversion may be obtained in the presence of one, two, or three quasimonochromatic strong fields at resonance. In the usual cases of two fields, the maser action with or without inversion cannot achieve high-efficiency microwave generation or detection, since for strong emitted fields, the saturating and nonlinear effects are strong too. On the contrary, whenever three frequency-correlated fields fulfill the ‘‘spectroscopic bridge’’ conditions, the saturating effects do not occur, and linear effects prevail, even when all these fields are strong. In this case, high-efficiency microwave generations or phase-sensitive detections may be achieved.

The dynamic scaling relations for the frequency-dependent magnetic susceptibility are derived, using the spin autocorrelation function q(t)∝${\mathit{t}}^{\mathrm{-}\mathit{a}}$ exp[-(t/τ${)}^{\mathit{b}}$], which successfully describes results from Monte Carlo simulations on a three-dimensional Ising spin-glass model. This functional form accounts very well for the observed frequency and temperature dependences of the imaginary component of the ac magnetic susceptibility in ${\mathrm{Cd}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Mn}}_{\mathit{x}}$Te, above the spin-glass transition temperature ${\mathit{T}}_{\mathit{c}}$. The parameters a and b obtained from the scaling analysis are very comparable to those found for ${\mathrm{Fe}}_{0.5}$${\mathrm{Mn}}_{0.5}$${\mathrm{TiO}}_3$, close to ${\mathit{T}}_{\mathit{c}}$, emphasizing the similar critical behavior of short-range insulating spin glasses, of Heisenberg and Ising type.

EPR studies on ${\mathrm{Mn}}^{2+}$-doped ${\mathrm{LiCsSO}}_4$ single crystals were performed in the temperature range 3.8–301 K, as well as on ${\mathrm{Cu}}^{2+}$-doped ${\mathrm{LiCsSO}}_4$ single crystals at room temperature. At higher temperatures the crystals were found to consist of twin domains. There were detected two phase transitions, one of which occurs at 205 K, and the other one is spread over the temperature range 155–175 K. In addition, the occurrence of a third phase transition somewhere in the range 3.8–31 K has been confirmed by the present data. The values of the isotropic g factor and the hyperfine-interaction constant for ${\mathrm{Mn}}^{2+}$ have been estimated at room temperature. Further, the values of all the ${\mathrm{Mn}}^{2+}$ spin-Hamiltonian parameters g, ${\mathit{b}}_2^0$, ${\mathit{b}}_2^2$, ${\mathit{b}}_4^0$, ${\mathit{b}}_4^2$, ${\mathit{b}}_4^4$, Q’, Q’’, ${\mathit{A}}_{\mathrm{\parallel }}$, ${\mathit{A}}_{\mathrm{\perp }}$ have been estimated at 90 and 152 K in the untwinned monoclinic phase. The intensity of the central broad ${\mathrm{Mn}}^{2+}$ line (hyperfine structure unresolved) is described well as a functions of temperature, near ${\mathit{T}}_{\mathit{c}}$, by the expression c‖T-${\mathit{T}}_{\mathit{c}}$${\mathrm{\Vert }}^{\gamma }$+b (${\mathit{T}}_{\mathit{c}}$=205 K), with the critical exponent γ=1, in accordance with that determined previously for the elastic constant ${\mathit{c}}_{66}$.

The magnetic-anisotropy energy (MAE) versus the d-band filling is systematically studied for a number of high-symmetry free-standing transition-metal monolayers and bilayers. We show that a simple tight-binding model combined with the recursion method is very useful in the elucidation of the complicated MAE problem. A canonical MAE curve is proposed as the simplest one curve consistent with the oscillation theorem and is interpreted in electronic-band-structure terms. Departures from this simple picture found in calculations are discussed in detail based on a symmetry analysis and the generalized theory of Wigner and von Neumann describing the energy-band crossing geometry. Some of our conclusions seem to agree with recent theoretical results and may shed light on the existing controversies regarding the Fe(001) monolayer. Other topics such as the crystal-field splitting, magnetic orbital momentum, the applicability of perturbation theory for MAE evaluation, and the substrate role are touched upon.

A classical theory of the deflection of a single-domain ferromagnetic cluster in a Stern-Gerlach experiment is presented. Two cases are discussed in detail: (1) superparamagnetic relaxation, in which the thermal fluctuation of the cluster spin occurs on a time scale much smaller than the transit time of the cluster through the Stern-Gerlach apparatus; and (2) anisotropy-induced relaxation, in which the cluster is sufficiently isolated from the thermal bath during its transit through the apparatus so that its dynamics is governed by the interaction of the cluster spin with the anisotropic magnetic forces caused by the lattice and the shape of the cluster. Average spin orientation and average spin rocking during transit, as well as Stern-Gerlach intensity profiles, are presented.

The propagation of nonlinear dipole spin waves in a film consisting of a two-sublattice, uniaxial, antiferromagnetic material has been investigated. The external magnetic field is assumed to be parallel to the anisotropy axis of the antiferromagnetic film and is directed parallel, or perpendicular, to the film surface. The existence of envelope solitons of spin waves is predicted and conditions for their creation are discussed.