Table of contents

The full text of this article is available to view as a PDF file.

The T₁ N point and ω point phonon anomalies for body-centred cubic (bcc) Zr are studied using an approach which goes beyond the traditional quasi-harmonic approximation and perturbation theory. We are able to reproduce, for the first time, the anomalous phonon behaviour in bcc Zr.

The total structure factor of molten TbCl₃ at 617ºC was measured by using neutron diffraction. The data are in agreement with results from previous experimental work but the use of a diffractometer having an extended reciprocal-space measurement window leads to improved resolution in real space. Significant discrepancies with the results obtained from recent molecular dynamics simulations carried out using a polarizable ion model, in which the interaction potentials were optimized to enhance agreement with previous diffraction data, are thereby highlighted. It is hence shown that there is considerable scope for the development of this model for TbCl₃ and for other trivalent metal halide systems spanning a wide range of ion size ratios.

By means of a novel exact-diagonalization technique, we show that bound pairs propagate between repulsive Hubbard clusters in a superconducting fashion. The size of the matrices that must be handled depends on the number of fermion configurations per spin, which is of the order of the square root of the overall size of the Hilbert space. We use CuO₄ units connected by weak O–O links to model the interplanar coupling and c-axis superconductivity in cuprates. The numerical evidence on Cu₂O₈ and Cu₃O₁₂ prompts the proposal of a new analytic scheme describing the propagation of bound pairs and also the superconducting flux quantization in a 3D geometry.

The properties of materials (mechanical, electronic, magnetic, etc) derive ultimately from the identity and spatial arrangement of their constituents. Nowadays, with the dimensions of technological devices and nanostructures reaching a few atomic constants, descriptions in terms of macroscopic concepts appear to be frequently inadequate and must give way to atomistic formulations based on elementary processes. Focusing on metallic materials, and more specifically on low-dimensional systems such as ultrathin films, superlattices or nanostructures, this paper reviews the atomic scale phenomena responsible for the most common types of defects (interfacial alloying, etching and roughness, formation of dislocations and pinholes, film discontinuities and twinning). It is shown that many of these features are related to the different mechanisms of strain relaxation in heteroepitaxial systems as well as to specific characteristics of atomic diffusion, such as the presence of Ehrlich–Schwoebel barriers hindering step crossings. Some special growth techniques (use of surfactants and codeposition) are also presented together with experimental examples demonstrating their usefulness to overcome the elements' natural limitations and produce accurately controlled, custom-designed epitaxial samples. Finally, a brief overview is given of different phenomena that can be exploited to produce self-assembled or self-organized structures.

The Rydberg-like series of image-potential states is a prototype system for loosely bound electrons at a metal surface. The electronic structure and the femtosecond dynamics of these states is studied by high-resolution energy-and time-resolved two-photon photoemission spectroscopy. The electron trapped in the image potential moves virtually freely laterally to the surface where it is subject to inelastic and quasielastic scattering processes which cause decay of population and phase relaxation. The influence of surface corrugation on these processes has been investigated for adsorbates on Cu(001) and stepped Cu(117) and Cu(119) surfaces which are vicinal to Cu(001). The dynamics depend on both the distance of the electron in front of the surface and the parallel momentum. For CO molecules on Cu(001) inelastic scattering into bulk states and adsorbate-induced resonances determine the decay rate. For small numbers of Cu adatoms on Cu(001) and the vicinal surfaces the decay rate of image-potential states is significantly modified by interband and subsequent intraband scattering. On the vicinal surfaces the origin of the interband-scattering process is clarified as quasielastic scattering caused by disorder of the lateral superlattice. It occurs for electrons with group velocity perpendicular to the step edge and, moreover, exhibits a sizeable asymmetry. Electrons are mainly scattered into states with momentum in the upstairs direction. This asymmetry in quasielastic scattering explains the direction dependence of the lifetime of the first image-potential states on stepped Cu(119).

The capabilities of small-angle x-ray scattering (SAXS) and wide-angle x-ray diffraction (XRD) to recognize structural changes in periodic multilayers were compared on Fe/Au multilayers with different degrees of structural degradation. Experimental results have shown that both methods are equally sensitive to the multilayer degradation, i.e., to the occurrence of non-continuous interfaces, to short-circuits in the multilayer structure and to the multilayer precipitation. XRD yielded additional information on the multilayer crystallinity, whilst SAXS could better recognize fragments of a long-range periodicity (remnants of the original multilayer structure). Changes in the multilayer structure were initiated by successive annealing at 200 and 300ºC. Experimental data were complemented by numerical simulations performed using a combination of optical theory and the distorted wave Born approximation for SAXS or the kinematical Born approximation for XRD.

Rare-earth/iron multilayered thin films were magnetron sputtered to investigate the thickness-dependent transition from amorphous iron to polycrystalline body-centred-cubic iron. To characterize this transition it is necessary to get information about the average short-range order (SRO) in the multilayers. A unique technique for measuring this SRO is calculating pair distribution functions (PDFs) from reduced intensity functions by using high-energy electron diffraction in a 300 kV transmission electron microscope. With a maximum resolution in the PDFs of 0.024 nm, this method offers a high sensitivity for the investigation of the SRO. Due to the planar probing characteristics of the experiment, one gets structure information on interfaces rather than from the bulk structure. A further advantage of this method is that no specimen preparation is necessary. Therefore preparation artefacts can be excluded.

The electronic structures of the γ/γ' interface for two-phase Ir-based alloys (Ir/Ir₃Ta and Ir/Ir₃Ti) have been investigated by performing first-principles quantum mechanics DMol3 (a type of density functional theory for molecules) calculations. The Mayer bond order (MBO) is used to represent the shear and cohesion strengths of the interface by a local sum of the horizontal and vertical MBOs. By comparison with those for single-crystal Ir, the results show that both the cohesive and shear strengths of the γ/γ' interface for the Ir/Ir₃Ta alloy increase. The cohesive strength of the interface for the Ir/Ir₃Ti alloy increases, whereas the shear strength of the interface for Ir/Ir₃Ti decreases. The electron charge density, the Hirshfeld charge, and orbital charge transfers are also calculated and analysed. An electronic mechanism for the γ/γ' interface strength of Ir-based alloys is then suggested.

An Au thin film system, deposited on silicone oil surfaces by the thermal deposition method, has been fabricated and its structure as well as electrical properties has been studied. A web-shaped characteristic surface morphology of the films is observed. The dc sheet resistance R of the metal films on the liquid surfaces is measured during and after deposition in situ by the four-probe method. The time dependence of the sheet resistance can be explained in terms of the film growth mechanism on the oil surface. The anomalous I–V characteristics of the film system can be interpreted as a competition among the local Joule heating, hopping and tunnelling effects. It is found that the dc third-harmonic coefficient B₀ and the zero-power resistance R₀ satisfy the power-law relation B₀ ∝ R₀^2+w and the exponent w is close to zero. This result indicates that the hopping and tunnelling effects in the samples are much stronger than those of the other film systems. We also find I_m ∝ R₀^−β with β = 0.79 ± 0.08 and I_m ∝ B₀^−γ with γ = 0.49 ± 0.07, where I_m is the critical current. The physical origins of the phenomena are discussed.

The sign of the exchange bias in field cooled nanocrystalline antiferromagnetic–ferromagnetic bilayers (Co–O and Ni–O/permalloy) is reversed at temperatures approaching the antiferromagnetic (AFM) blocking temperature. A similar phenomenon is observed after magnetic training processes at similar temperatures. These effects can be explained assuming that the boundaries of nanocrystalline grains in AFM layers exhibit lower transition temperatures than grain cores.

The initial stage of Er/Si(100) interface formation has been investigated by using synchrotron radiation photoelectron spectroscopy combined with low-energy electron diffraction. Both the valence band and the core level peaks of the Si photoemission spectra shift rigidly with increasing Er coverage in the submonolayer region. Upon depositing 0.6 monolayers of Er on the Si(100) surface at room temperature, the surface Fermi level is ultimately pinned at 0.29 eV above its initial value, which is equivalent to a Schottky barrier height of 0.67 eV. No evidence is found for the formation of Er silicides at the as-deposited surfaces. Annealing of the Er-covered Si(100) surfaces at 600ºC results in the appearance of a new peak located 1.2 eV below the Si 2p peak, indicating the presence of some sort of Er silicide. Meanwhile, the Er 4f spectrum measured for samples upon annealing exhibits a well-resolved fine structure, implying that only monospecies of Er silicide may exist on the surface.

Size-controlled nanocrystalline silicon (nc-Si) has been prepared from a-SiN_x/a-Si:H/a-SiN_x ('a' standing for amorphous) structures by thermal annealing. Transmission electron microscope analyses show that the lateral size of the nc-Si is controlled by the annealing conditions and the a-Si sublayer thickness. The deviation of the nc-Si grain size distribution decreases with the a-Si sublayer thickness, so thinner a-Si sublayers are favourable for obtaining uniform nc-Si grains. In the a-Si:H (10 nm) sample annealed at 1000ºC for 30 min, an obvious bi-modal size distribution of nc-Si grains appears, but no obvious bi-modal size distribution is found in other samples with thinner a-Si:H sublayers. On the basis of the experimental results, we discuss the process of transition from the sphere-like shape to the disc-like shape in the growth model of the nc-Si crystallization. The critical thickness of the a-Si sublayer for the constrained crystallization can be determined by the present model. Moreover, the increase of the crystallization temperature in the ultrathin a-Si sublayer is also discussed.

We apply a comprehensive Monte Carlo simulation for the global oscillations in the CO catalytic oxidation system on the Pt(100) surface based on a well defined lattice gas model. The basic reactive steps of the present model follow the Langmuir–Hinshelwood mechanism incorporated with adsorbate-induced surface restructuring which follows the statistical theory of first-order phase transitions. It is shown that oscillation only exists in a certain range of CO fraction in gas and the production rate of CO₂ forms a double-periodic oscillation, while the coverages of CO and O₂ and fraction of (1 × 1) phase are in single-periodic oscillation. We give an explanation for these phenomena. Moreover, the simulation results clearly show that the diffusion of adsorbed CO works to synchronize local oscillators and the oscillation in the system will disappear with the decrease of the diffusion.

We study the frequency response of large-amplitude oscillations of a trapped Bose condensed gas. On the basis of the Thomas–Fermi approximation, we deduce the hydrodynamical equation including the excitation source of the velocity drive, and obtain the analytical expression for the frequency as a function of velocity fluctuation amplitude, trap geometry, and symmetry of modes. By solving the expansion equation after switching off the trap, we find a simple relationship between the velocity amplitude and the oscillation amplitude. The theoretical calculations on the response frequencies agree well with the existing experimental observations.

The behaviour of a three-dimensional Ising-like system at temperatures above the critical value of T_c is studied in the approximation of the sextic distribution for modes of spin density oscillations (ρ⁶ model). The original collective variables method is developed in this higher non-Gaussian approximation for calculating the thermodynamic characteristics of the system near T_c taking into account the first confluent correction. The calculations are illustrated by an example of a simple cubic lattice and an exponentially decreasing interaction potential. The main distinctive feature of the method is the separate inclusion of the contributions to the thermodynamic functions from the short-wave and long-wave spin density oscillation modes. The dependences of the phase transition temperature, leading critical amplitudes, and correction-to-scaling amplitudes for the specific heat and susceptibility on the microscopic parameters of the system are investigated.

High-resolution x-ray rocking diffraction has been used to measure the spontaneous strain associated with the cubic–rhombohedral phase transition in LaAlO₃ in the range 10K ≤ T ≤ 750 K. The data are consistent with a second-order Landau-like model at high temperatures, with T_C = 834(2) K. At lower temperatures, the strain data display order parameter saturation, related to the quantum saturation of the phonon modes. Comparison of the saturation temperature for the spontaneous strain (θ_S = 95 K) with the saturation temperatures for independent measurements of the rotation (θ_S = 260 K) and distortion (θ_S = 150 K) of the AlO₆ octahedra indicates that the phase transition consists of two coupled processes, and that the coupling does not have the same effect in the classical and quantum saturation limits.

High resolution neutron spectra of Ga(CH₃)₃ show tunnelling transitions between 4.5 and 19 μ eV. The spectrum can be explained within the single-particle model on the basis of the monoclinic C2/c (Z = 16) low temperature crystal structure of Ga(CH₃)₃ with six inequivalent methyl groups in the unit cell. The overlapping tunnelling lines prevent the extraction of temperature dependent linewidths which would allow us to assign the librational energies measured in the phonon density of states. Classical rotational motion is studied by quasielastic neutron scattering. Three activation energies could be extracted. Methyl librations, tunnelling energies and barrier heights are combined with consistent intensities into rotational potentials. Only the concerted application of all spectroscopic techniques yields a conclusive description.

High resolution neutron powder diffraction and single crystal measurements on the ferromagnetic shape memory compound Ni₂MnGa have been carried out. They enabled the sequence of transformations which take place when the unstressed, stoichiometric compound is cooled from 400 to 20 K to be established. For the first time the crystallographic structure of each of the phases which occur has been determined. At 400 K the compound has the cubic L2₁ structure, and orders ferromagnetically at T_C ≈ 365 K. On cooling below ∼ 260 K a super-structure, characterized by tripling of the repeat in one of the ⟨110⟩_cubic directions, forms. This phase, known as the pre-martensitic phase, persists down to the structural phase transition at T_M ≈ 200 K and can be described by an orthorhombic unit cell with lattice parameters a_ortho = 1/√2a_cubic, b_ortho = 3/√2a_cubic, c_ortho = a_cubic and space group Pnnm. Below T_M the compound has a related orthorhombic super-cell with b_ortho ≈ 7/√2a_cubic, which can be described within the same space group. The new modulation appears abruptly at T_M and remains stable down to at least 20 K.

We extend to bianisotropic structures a formalism already developed, based on the Bloch method for defining the effective dielectric tensor of anisotropic crystals in the long-wavelength approximation. More precisely, we provide a homogenization scheme which yields a wavevector-dependent effective medium for any 3D, 2D, or 1D bianisotropic crystal. We illustrate our procedure by applying this to a 1D magneto-electric smectic C^*-type structure. The resulting equations confirm that the presence of dielectric and magnetic susceptibilities in the periodic structures generates magneto-electric pseudo-tensors for the effective medium. Their contribution to the optical activity of structurally chiral media can be of the same order of magnitude as the one present in dielectric helix-shaped crystals. Simple analytical expressions are found for the most important optical properties of smectic C^*-type structures which are simultaneously dielectric and magnetic.

Cd_1−xZn_xTe crystals are annealed in Cd/Zn vapours to eliminate Te-rich phases. During the saturated Zn partial pressure annealing and the high-temperature annealing under low P_Zn, large Te-rich phases gather at the slice surface through thermo-migration. This process increased the stress in the bulk crystals and caused the formation of defects such as twins, stacking faults, and dislocations. The IR transmission is greatly improved by annealing the slices under saturated P_Zn and then removing a layer from the surface of the slices with a thickness of about 50μm. For Cd_1−xZn_xTe slices annealed at low temperature and low Zn partial pressure, the reduction of the Te-rich phases greatly depends on the diffusion of the atomic defects.

We verify the functional form of the asymptotics of the spin–spin equal-time correlation function for the XX chain, predicted by the hypothesis of conformal invariance at large distances and by the bosonization procedure. We point out that the bosonization procedure also predicts the functional form of the correlators for the chains of finite length. We found the exact expression for the spin–spin equal-time correlator on a finite lattice. We find excellent agreement of the exact correlator with the prediction given by the leading asymptotics result up to very small distances. We also establish the correspondence between the value of the constant before the asymptotics for the XX chain with the expression for this constant proposed by Lukyanov and Zamolodchikov. We also evaluate the constant corresponding to the subleading term in the asymptotics in a way which is different from previous studies.

We use a jellium model and the classical approach to study plasma oscillations of a carbon peapod. The dipole–dipole interaction between the C₆₀ molecules in the array of C₆₀ molecules encapsulated in a carbon nanotube is found to affect the longitudinal and transverse modes differently. The energy of the longitudinal mode is lowered, while the energy of the transverse mode is raised. The coupling between the nanotube and C₆₀ molecules is found to be weak for longitudinal modes and strong for transverse modes. A carbon peapod is predicted to have Luttinger liquid behaviour similar to that of a carbon nanotube alone.

The structural and magnetic features of the TbCo₄M compounds (M = Al and Ga) have been studied by means of neutron powder diffraction and magnetic measurements. X-ray diffraction experiments show that the samples are single phased, crystallizing in the CaCu₅-type structure. Neutron powder diffraction measurements demonstrate a preferential substitution of Al and Ga at the 3g crystallographic site of the CaCu₅-type structure. Together with neutron powder diffraction investigations, magnetization measurements and x-ray powder diffraction studies performed on field-oriented samples are combined in order to clarify the effects of the Al and Ga substitutions on the magnetic features of these compounds. The modifications of the magnetic properties upon substitution are significant: a drastic decrease of the mean Co magnetic moment and of the Curie temperature is observed. Unlike what was found earlier for TbCo₅, no sign of a spin reorientation transition has been observed in the present TbCo₄M compounds. Moreover, both compounds present a planar magnetocrystalline anisotropy over the whole temperature range. Finally, the thermal evolutions of the magnetization of terbium and cobalt sublattices in TbCo₄Al and TbCo₄Ga are presented.

Temperature-dependent magnetization and magnetoresistance of the Na doped La_1−xCa_{x −y}Na_yMnO₃ type samples with x = 0.3 and 0 ≤ y ≤ 0.15, showing a rhombohedrally distorted perovskite structure with space group Rc, have been studied under a pulsed magnetic field. With the increase in temperature from the low temperature ferromagnetic phase, the samples exhibit a sharp metal to insulator transition around T_p accompanied by a ferromagnetic (metallic) to paramagnetic (semiconducting) phase transition with a well defined Curie temperature T_C (almost equal to T_p). Interestingly, small Na doping largely increases the conductivity and T_p, which is associated with a decrease of the (e–ph) interaction constant (γ). Doping of monovalent Na in the Ca site of La_1−xCa_x−yNa_yMnO₃ drives the system from a high resistivity regime with lower T_p to a lower resistivity regime with higher T_p values. Low temperature resistivity data fit well the relation ρ = ρ₀ + ρ₂T², signifying the importance of the electron–magnon scattering process (the ρ₂T² term). On the other hand, the high temperature (T > T_p up to 320 K) conductivity data satisfy the variable range hopping (VRH) model. For T > 320 the small polaron hopping model is more appropriate than the VRH model. Even with a very small change of y, the density of states at the Fermi level N(E_F) changes considerably. The resistivity of these materials measured under pulsed and continuous DC magnetic fields behaves in an identical fashion. The relaxation time (decay time of the magnetic pulse within the sample) varies with field strength, which indicates that, with a change of magnetic field, the ordering of spin in the ferromagnetic regime changes.

First-principles calculations have been used to study the effect of vacancies on the structural and electronic properties in substoichiometric TiC_x and TiN_x. The effect of vacancies on equilibrium volumes, bulk moduli, electronic band structures and density of states of the substoichiometric phases was studied using a full-potential linear augmented plane-wave method. A model structure of eight-atom supercells with ordered vacancies within the carbon and nitrogen sublattices is used.

We find that the lattice parameters of the studied stoichiometries in both TiC_x and TiN_x are smaller than that of ideal stoichiometric TiC and TiN. Our results for the variation of the lattice parameters and the bulk moduli for TiC_x are found to be in good agreement with experiment. The variation of the energy gaps with the atomic concentration ratio shows that these compounds present the same trends. Results for TiC_x are compared to a recent full-potential calculation with relaxed 16-atom supercells.

First principles calculations are performed on possible structures of silicon monoxide solids. The chemical character of all of the bonding interactions is systematically quantified in real space. It is found that the most stable SiO structure possesses the highest number of inequivalent bond paths. This process reveals a novel metallic Si–Si interaction and provides an explanation for the origin of the unexpectedly high conductivity in thin silicon oxide layers. In this paper a new measure for quantifying metallic character (in direct space) present in a bond has been introduced. Furthermore it has been possible to determine the directional properties of this metallic character in real space using the charge density. This finding is very important for future complementary metal oxide semiconductor technology.

We have carried out DC magnetization measurements on spherical glassy carbon exposed to high temperatures and high pressures. The observed magnetic signals clearly depend on the sintering temperature. On the basis of the graphitization temperature (around 1400ºC), the behaviour can be classified into three regions: (1) paramagnetism in the no-graphitization region; (2) ferromagnetism in the near-graphitization region; (3) diamagnetic behaviour after graphitization. The magnetic transitions associated with the process of graphitization are discussed.

For isolated rare-earth impurities substituting for Mg atoms in the superconductor MgB₂ the crystal field parameters are calculated by the ab initio density functional electron theory with constraints for the 4f charge and spin density. The crystal field parameter A₆⁶ is extremely small due to the structure and bonding properties of MgB₂, and therefore the crystal field levels are nearly exclusively determined by one magnetic quantum number M. Implications for the pair-breaking mechanism of the superconductivity in MgB₂ are discussed.

A theory of neutron scattering by magnetic ions is developed and applied to diffraction by crystals in which ions in the unit cell are not equivalent on account of a lack of translational symmetry in their environments. The development demonstrates a close connection between interpretations of magnetic neutron and x-ray Bragg diffraction, in terms of an atomic model. Cross-sections for neutron diffraction by powders and single crystals, including polarization-induced interference between nuclear and magnetic amplitudes, are considered. By way of illustrating the theoretical development, cross-sections are predicted for Mott–Hubbard insulator, V₂O₃, on the basis of findings from extensive studies using resonant x-ray diffraction.

Static magnetization measurements have been performed on the mixed fluoride series Fe_xCr_(1−x)F₃ (0 ≤ x ≤ 1). The evolution of the shape of the hysteresis loops reveals drastic changes as a function of the degree of concentration x and the temperature, whereas the coercive field and the magnetization exhibit singularities. These data are interpreted from a phenomenological model based on the existence of two uncompensated weak ferromagnetic sublattices randomly located on a cubic network. By including a spin relaxation process, such a mechanism explains well the presence of magnetic clustering phenomena, in agreement with⁵⁷Fe Mössbauer spectrometry and magnetic susceptibility measurements.

We report measurements of the thermoelectric power (TEP) on the La_0.65Ca_0.35Mn_1−x Fe_xO₃ system for 0.00 ≤ x ≤ 0.07. The ferromagnetic and metallic transition temperatures are lowered and the TEP shows an increasingly positive trend with the addition of Fe. We also observe a clear magnetic contribution that manifests itself as a peak in the TEP close to the critical temperature. The activation energies determined from the TEP are seen to be insensitive to the Fe content. The data are interpreted firstly as showing a decrease in the density of active holes, i.e. holes that can participate in the hopping process, with increasing Fe content. Secondly the data suggest the role of magnetic scattering due to the clusters formed by the antiferromagnetically coupled Fe. Abrupt changes in the variation of the TEP are observed at the concentration region x ∼ 0.04 consistent with the hole density variation and with previously reported transport and magnetic measurements.

The uncommon Sc ions are employed to partially replace Mn in La_0.7Sr_0.3MnO₃ perovskite. X-ray diffraction patterns indicate a pure single phase with rhombohedral lattice symmetry for La_0.7Sr_0.3Mn_1−xSc_xO₃ (0 ≤ x ≤ 0.1). The Curie temperature (T_C) is tuned to near room temperature by Sc substitution with x = 0.03–0.05, and consequently a significant enhancement in magnetoresistance (MR) at room temperature is obtained in the Sc-doped compounds. Magnetic measurement shows no spin coupling between Sc and Mn ions. We suggest that the reduction in T_C and the enhancement in MR are ascribable to the suppression of ferromagnetic order due to the structural distortion caused by the large Sc³⁺ ions.

We report transport, magnetization and transmission electron microscopy studies of the effects of A-and B-site substitution, and the addition of metal ions such as Pt, Ag and Sr, on doped ABO₃ perovskites, where A = La, Pr etc and B = Mn. Disorder induced by such substitution changes the behaviour of the charge-ordered (CO) state significantly. A-and B-site substitution suppresses the CO phase due to size mismatch and disorder produced by inhomogeneity. On the other hand, addition of metal ions such as Pt and Ag improves several colossal-magnetoresistance properties significantly due to microstructural effects and enhanced current percolation through grain boundaries.

Electron paramagnetic resonance (EPR) of Fe³⁺ ions in Al₂O₃ is studied in powder samples prepared by different routes and/or modified by thermal or mechanical treatments, with different doping levels and grain sizes. The measurements are performed in various frequency bands (S, X, K, Q and W) and with bimodal detection in X-band. Simulations of the spectra are achieved with a code designed for computing EPR powder spectra described by any spin Hamiltonian including second-, fourth-and sixth-order ZFS terms (S ≤ 7/2). The linewidths, intensities and lineshapes are accounted for. The lineshape is Gaussian at low Fe³⁺ concentration whereas it is Lorentzian for higher concentration. The linewidths are interpreted as the superimposition of three main contributions: intrinsic linewidth, dipolar broadening and broadening due to lattice imperfections. The latter is tentatively interpreted in terms of quadrupolar spin Hamiltonian parameter distributions treated using first-order perturbation theory. Whatever the sample, only the b₂² spin Hamiltonian parameter is found to be distributed around a mean zero value which corresponds to rhombic distortions. Angle and bond length distributions are tentatively extracted from the b₂² distributions which gives some insight into the local order around the spin probe in relation to the preparation and treatment of the samples.

We propose modified Bloch equations (MBEs) with specific power-dependent relaxation and dispersion parameters characteristic for two-pulse excitation and when the magnetic dipole–dipole interactions in the electron spin system control the dephasing. We discriminate between the 'active' (excited by both pulses) and 'passive' (excited by the second pulse only) spins: it is shown that the 'active' spins participate in a new effect, an active spin frequency modulation effect giving rise to the power-dependent dispersion and multiple electron spin echoes (ESEs); the 'passive' spins contribute to the power-dependent relaxation. The MBEs are solved and a general expression for the two-pulse ESEs is obtained. Detailed numerical analysis of this expression gives results in good quantitative agreement with the recent experiments on the two-pulse ESEs at conventional low applied fields. The developed theory is applied also to high field ESEs, which are promising for future investigations. On the basis of published results it is deduced that the instantaneous diffusion mechanism is ineffective.

Our high-pressure Raman scattering experiments on pentaerythritol (C(CH₂OH)₄) show that this compound undergoes at least three phase transformations up to 25 GPa. Splitting of various modes at ∼6.3, ∼8.2 and 10 GPa suggests that these phase transformations result in lowering of crystalline symmetry. A very small discontinuous change in slope of most of the Raman-active modes is observed at 0.3 GPa. However, no other signature of a phase transition was observed at this pressure. The observed correlation of the pressures for the onset of the two phase transformations with the limiting values of the distances between various non-bonded atoms in the parent phase suggests that the molecular rearrangements across the phase transformations are not very drastic. In addition, our earlier Fourier transform infrared and present Raman investigations indicate that high-pressure compression leads to increase in strength of the hydrogen bond present in this compound.

We have studied the charge dynamics of Sr-doped La₂CuO₄ at hole concentration p = 0.07. We observed a clear indication of the competition between two new characteristic collective modes of the Wigner lattice orders suggested earlier: one corresponds to a T_c = 15 K superconductivity and the other to T_c = 30 K. Our results naturally explain the recent observation of metallic transport in extremely low (1%) Sr-doped La₂CuO₄ and establish a novel insulator to metal transition mechanism and superconductivity in the two-dimensional Wigner lattice ground state. We suggest that the fluctuations of the relative phase between the superconducting condensate and the Wigner lattice order can provide an answer to the unidentified mode in the c-axis conductivity that appears only below T_c.

We develop an approach for derivation of quantum–classical relaxation equations for a two-channel problem. The treatment is based on the adiabatic channel wavefunctions and the system–bath coupling is modelled as a bilinear interaction in momentum representation. In the quantum–classical limit we obtain Liouville equations with the relaxation operator containing diffusion terms diagonal in Liouvillian space and the off-diagonal part which is responsible for thermal interlevel transitions. The high-frequency interlevel quantum beats are fully taken into account in this relaxation term. In the framework of the present formulation and as a consequence of the momentum-dependent interaction the Smoluchovsky diffusion limit can be reached without invoking Fokker–Planck equations as an intermediate step. The inherent property of equations so obtained is that the partial rates of interlevel transitions obey the principle of detailed balance. This result could not be gained in earlier treatments of the two-level diffusion problem.