Table of contents

Today, the median time from receipt to publication for regular articles in Journal of Physics: Condensed Matter is about four months. Letters can be reviewed, possibly revised, and on the Web in little more than a week in favourable circumstances, and the median time is six weeks. When the Journal of Physics series was started, over thirty years ago, Letters took typically three months from receipt to print, and articles took substantially longer. Now that publication times for regular papers are of a similar order to those of Letters in the past, it makes sense to review the types of submission we accept as Letters and put a higher premium on urgency.

In the past, Letters have been of several different types. There have been Letters giving a first announcement of some important new result, and these have a justifiable urgency and need for priority. In addition, there have been what one might call short papers, self-contained pieces of work, but with no requirement for rapid publication. The Editorial Board of Journal of Physics: Condensed Matter has decided that in future all Letters published will have to satisfy criteria of significant importance and urgency.

To achieve this, all manuscripts submitted as Letters will be processed as follows. First, the Letter will be looked at by a Board Member, who will decide whether or not the proposed Letter has the right level of importance, urgency, and interest to appear as a Letter. The Board Member will not usually act as referee, unless the Letter is in a field in which they normally referee. If their decision is yes, then the manuscript will go to a regular referee. Special efforts will be made to ensure rapid treatment, both by the referee and in processing at Bristol.

The ideal Letter would address a significant topic in condensed matter physics. It would be recognized as important by a large number of condensed matter physicists, including those whose research area is a different one. So it is crucial that the Letter makes it clear (and credible) just why the work has this substantial and broad significance. A clearly written abstract, and an appropriate opening paragraph and conclusions will be essential. In addition, authors are strongly urged to enclose with their submission a brief statement giving reasons why the work should be regarded as urgent.

We expect to have this new process fully operational for submissions of Letters received in 2004, and we shall review the guidance and the process to ensure that it is effective. We hope that, within a short time, Letters to Journal of Physics: Condensed Matter will gain in visibility and in recognized importance.

During the sputter erosion of Pd(111) by 1 keV Ar⁺ ions in the temperature range we observe the development of clearly defined ridges that deepen as sputtering continues up to removed. Step edges observed by low energy electron microscopy appear as contour lines that define the detailed topography of the rippled surfaces. Different characteristic step profiles, and hence surface geometries, are observed after short and long sputtering times. The sequence resembles the interfacial structures reported for the Mullins–Sekerka instability of driven solidification interfaces. At long sputtering times the ridges adopt crystallographic orientations, as reported in recent studies of the ion erosion of crystal surfaces.

We propose a microscopic theory of relaxation of glass under pressure. We show how increased rigidity of the glass structure under pressure couples to local densification events to yield slow logarithmic relaxation and relate the microscopic parameters of the theory to the experimental densification data. We discuss two regimes of slow relaxation, one of which is the relaxation in the structure where rigidity percolates. Finally, we suggest that coupling of local densification events to structural rigidity can describe a wider range of relaxations in disordered media.

Multilayers of charged polymers can be formed by the alternating adsorption of polyanions and polycations. By this so-called layer-by-layer self-assembly method, planar surfaces or colloidal templates can be coated. This method is reviewed here with respect to the basic physical principles governing multilayer formation. Particular emphasis is put on the adsorption process of a single polyelectrolyte layer to a multilayer surface as the process controlling the charge complexation and the local molecular structure. Then, the implications for the properties of the multilayer assembly are discussed. In particular, molecular properties such as the internal stoichiometry, the local interactions, and the conformation on a molecular scale are reviewed.

We investigated the absolute acoustic band gaps for two-dimensional periodic arrays of silica cylinders in viscous liquid. Acoustic band structures, which have complex eigenfrequencies for these systems, are calculated with the plane wave expansion method. They show that when the viscous penetration depth is comparable to the structural length scale, the structures possess large absolute acoustic band gaps compared with those without viscosity. We also found that the magnitude of the viscosity plays an important role in these band gaps.

An efficient algorithm is presented to find the Blum–Høye mean spherical approximation (MSA) solution for mixtures of hard-core fluids interacting through multi-screened Coulomb plus power series (multi-SCPPS) potentials. The solution for hard-sphere mixtures is used as an initial estimate of the variables. From this initial estimate only a few Newton–Raphson iterations are required to reach the final solution. The algorithm is also able to reflect correctly their molecular characteristics into the appropriate macroscopic behaviour. All MSA thermodynamic properties are presented analytically. Our analysis will be applicable to a large variety of fluid mixtures because any smooth, realistic isotropic potential can be well reproduced by the multi-SCPPS closure with arbitrary accuracy. Demonstrations of the power of the algorithm are made for two binary-component systems of Lennard-Jones particles. One is composed of particles with nearly the same sizes, while the other mixture with widely different sizes. We have studied their thermodynamic stabilities based on the compressibility and the isotherms of virial and energy equations of state, and further from microscopic viewpoints in connection with radial distribution functions. The present method is able to detect vapour–liquid transitions and perhaps liquid–glass transitions. The multi-SCPPS closure includes the multi-Yukawa closure as a special case.

The relative distribution of rare-earth ions R³⁺ (Dy³⁺ or Ho³⁺) in the phosphate glass RAl_0.30P_3.05O_9.62 was measured by employing the method of isomorphic substitution in neutron diffraction and, by taking the role of Al into explicit account, a self-consistent model of the glass structure was developed. The glass network is found to be made from corner sharing PO₄ tetrahedra in which there are, on average, 2.32(9) terminal oxygen atoms, O_T, at 1.50(1) Å and 1.68(9) bridging oxygen atoms, O_B, at 1.60(1) Å. The network modifying R³⁺ ions bind to an average of 6.7(1) O_T and are distributed such that 7.9(7) R–R nearest neighbours reside at 5.62(6) Å. The Al³⁺ ion also has a network modifying role in which it helps to strengthen the glass through the formation of O_T–Al–O_T linkages. The connectivity of the R-centred coordination polyhedra in (M₂O₃)_x(P₂O₅)_1−x glasses, where M³⁺ denotes a network modifying cation (R³⁺ or Al³⁺), is quantified in terms of a parameter f_s. Methods for reducing the clustering of rare-earth ions in these materials are then discussed, based on a reduction of f_s via the replacement of R³⁺ by Al³⁺ at fixed total modifier content or via a change of x to increase the number of O_T available per network modifying M³⁺ cation.

Hydrostatic pressure in the 0.1–75 MPa range has been used as a non-invasive tool to study the crystallization process of the tetragonal crystal form of the protein thaumatin (M_r 22 200). Crystals were prepared within agarose gel and at temperatures in the range from 283 to 303 K. The solubility, i.e. the concentration of soluble macromolecules remaining in equilibrium with the crystals, decreases when the pressure increases and when the temperature decreases. High pressure was used to probe the nucleation behaviour of thaumatin. The pressure dependence of the nucleation rate leads to an activation volume of −46.5 cm³ mol⁻¹. It is shown that an increase in pressure decreases the enthalpy, the entropy and the free energy of crystallization of thaumatin. The data are discussed in the light of the results of crystallographic analyses and of the structure of the protein.

This paper studies the nanojet ejection process by spreading droplets on a solid surface using molecular dynamics simulation based on the Lennard-Jones potential. During the simulation, it was found that the spreading width increased as the aperture dimensions and time increased. An energy wave phenomenon was also found in the simulation. After the ejection process reaches a relatively steady state, the droplets' contact angle ranged from about 30° to 60°. When the atoms were deposited on a moving plate the layer phenomenon became apparent.

For the first time, an organic dye, morin, and another bio-macromolecule, human serum albumen (HSA), were embedded in porous alumina (P-Al) fabricated by anodization in oxalic acid. The photoluminescent (PL) spectra of morin and morin–HSA immersed in the P-Al pores were measured and compared with those of liquid solutions. It was found that their PL positions are similar to those of dye-Al³⁺ in ethanol solution, and the PL of the embedded morin was greatly enhanced by the introduction of HSA. We have ascribed the appearance of PL bands to the formation of morin–Al complexes in the nanopores, with its inner wall being involved. A likely luminescence model was proposed to explain the observed PL enhancement due to the coexistence of morin and HSA in the nanopores, which has been proved by UV and Fourier transform infrared measurements. Moreover, the PL intensity increased with increasing pore diameter, which may be used to develop a luminescent array to predict the pore size of a P-Al layer.

The depletion interaction between two and three big hard spheres in a small hard sphere fluid with volume fraction π/10 is calculated both from Monte Carlo simulations and the Asakura–Oosawa approximation. The Monte Carlo results are in good agreement with previous DFT results. The three-body interaction at this volume fraction of small hard spheres is found to be very small, which indicates that the pair additivity of the depletion force is a good approximation.

We present a detailed study of the validity of the slowly varying density approximation to calculate the surface tension and the surface energy of Lennard-Jones fluids. To do so, we consider three explicit analytical expressions for the radial distribution function of the liquid phase, including one proposed by our research group, together with very accurate expressions for the liquid and vapour densities, also proposed by ourselves. The calculation of the surface tension from the direct correlation function using the Percus–Yevick and the hypernetted-chain approximations is also considered. Finally, our results are compared with those obtained by other authors by computer simulation, and also with those estimated via the general expressions (i.e., not in the ambit of the approximation studied here). We demonstrate that although the slowly varying density approximation is in good agreement with more complex expressions near the critical point, it is not adequate to calculate the surface energy and the surface tension of Lennard-Jones fluids at every temperature.

Supercritical fluid decomposition of copper precursors in the presence of hexagonal mesoporous silica matrices are used to prepare ordered copper metal nanowire and nanotube arrays. Evidence is presented which shows that the pore-filling process is accompanied by large expansions of the pore-to-pore distance induced by the inclusion process. Preparation of these materials requires strict control as copper appears to catalyze the collapse of the mesoporous structure. The copper nanostructures exhibit the normal face-centred cubic crystal structure although there is a strong variation of the lattice parameter with size. UV–visible absorption spectra of copper nanowires within the mesoporous matrix as a function of copper loading exhibit effects related to the low dimensionality of the nanowires formed. Transmission electron microscopy images and changes in the surface area and pore size of the materials as a function of copper content suggest that the growth mode is radial rather than a random nucleation and growth mechanism.

The structural evolution at 300 K of the series Sr_xBa_1−xSnO₃ and Sr_xCa_1−xSnO₃ at x = 0.2 intervals has been determined by powder neutron diffraction. All Sr_xCa_1−xSnO₃ samples (x = 0–1) have the Pbnm superstructure. In the series Sr_xBa_1−xSnO₃, the x = 1.0 and 0.8 samples have the Pbnm superstructure. The x = 0.6 sample has a second orthorhombic structure with space group Imma. There is a tetragonal I4/mcm phase at x = 0.4 and the undistorted cubic structure for x = 0.2 and BaSnO₃. The octahedral tilt angles show a smooth variation with average A cation radius over the range of superstructures described by mean field theory close to the transition. The transitions are also analysed via the variation of spontaneous strains.

The DyCo₄Al and DyCo₄Ga compounds crystallize in the hexagonal CaCu₅-type structure (space group P6/mmm). Al atoms exclusively occupy the 3g crystallographic site of the structure, whereas most of the Ga atoms occupy the 3g site, the rest being located at the 2c site. Both compounds are ferrimagnetic, and remarkable changes in the magnetic properties are observed with respect to DyCo₅. Al or Ga substitution for Co causes a drastic reduction of the Curie temperature, from  K in DyCo₅ to 479(5) K (M = Al) and 475(5) K (M = Ga). This reduction is accompanied by an important increase of the saturation magnetization at 4 K: from 0.7 μ_B/fu for DyCo₅ up to 4.7 and 4.9 μ_B/fu for DyCo₄Ga and DyCo₄Al, respectively. The compensation temperatures T_comp are close to 300 K, while T_comp equals 124 K in DyCo₅. These phenomena are clearly linked to a decrease of the mean Co magnetic moment. For these two materials, thermomagnetic measurements have evidenced a spin reorientation transition at  K. Powder neutron diffraction investigations were performed, and the Rietveld refinements led to the characterization of the magnetic phase diagrams of both compounds in the whole ordered temperature range. The Co magnetic moments are typically about 1.2 μ_B.

By the use of a first principles density functional theory, two kinds of models, namely the Rice–Wang thermodynamics model and the Seah quasi-chemical model, are employed to evaluate the embrittling tendency of a grain boundary (GB) due to the 3d element segregation. The first principles method based on those two models is appropriate for calculating the chemical and structural relaxation contributions to the changes of GB cohesion with the 3d segregants. The effects of the 3d transition elements, such as Ti, V, Cr and Mn, on a stable fcc Fe GB are studied and the difference between these two models is interpreted. When the chemical and the structural relaxation effects are taken into account, the calculated results for these two models are coincident for most of the elements studied, except for chromium. After analysing their chemical bonding in detail, we find that this discrepancy may be attributable to a lower susceptibility of the Seah model to the bonding anisotropy caused by Cr in the GB. It is proposed that the Seah model should be prudently used for some elements, especially those lying in the middle of a transition period.

The complex lanthanide doping of electron manganites results in enhancement of various phase separation effects in physical properties of these compounds. Selecting Ca_0.8Sm_0.16Nd_0.04MnO₃ as a model case we show that the first order structural phase transition from the paramagnetic semi-metallic phase into the anti-ferromagnetic semi-metallic phase at  K is marked by an abrupt decrease in magnetization, a steplike anomaly ΔL/L = 10⁻⁴ in the thermal expansion and a large latent heat ΔQ = 610 J mol⁻¹. In a certain temperature range below T_S, the high field magnetization exhibits hysteretic metamagnetic behaviour due to a field induced first order transformation. The temperature dependences of ac susceptibility, magnetization and resistivity suggest rather a non-uniform state in Ca_0.8Sm_0.16Nd_0.04MnO₃ at low temperatures. The metal–insulator transition occurs at  K, accompanied by a steplike increase in magnetization. These features could be ascribed to 'sponging' of electrons from the neighbouring anti-ferromagnetic matrix by clusters undergoing the ferromagnetic ordering. The spontaneous magnetization of Ca_0.8Sm_0.16Nd_0.04MnO₃ at low temperatures is about 0.1% of the calculated saturation magnetization for this compound.

Magnetic and charge order in half-filled and doped finite two-dimensional (8 × 8) clusters described by the extended Hubbard Hamiltonian with medium to large on-site (U) and intersite (V) Coulomb repulsion were investigated using correlated wavefunctions. The treatment of spatially inhomogeneous correlations was possible thanks to a combination of the local ansatz and the local increments expansion methods. The surprising result is that in the half-filled case the correlations do not influence the transition from the antiferromagnetic (AF) to charge order (CO) phase found in the Hartree–Fock approximation. For the doped systems we found a variety of stripe or superlattice-like structures for small values of V, while for larger V we identified robust CO phases. For smaller doping (δ = 1/8 and 1/4) the charge-majority sublattice has typically inhomogeneous charge distribution and is decorated by various magnetic patterns, while for larger doping (δ = 1/2) the co-phase is homogeneous (ideal) and with superimposed AF order.

In the framework of density functional theory and the generalized gradient approximation, we investigate the effects of magnetism on four plutonium allotropes. We show that an antiferromagnetic configuration gives a structural hierarchy which is consistent with experimental features. Using a simple Debye–Grüneisen model and a corrective term, we are able to draw a P–T phase diagram of Pu. Harmonic contributions are not sufficient for obtaining a correct description of the thermal properties of Pu and we emphasize the important role played by electronic and anharmonic contributions.

Defect structures and nuclear electric hyperfine interactions have been investigated in a series of binary Ni–Al, Ni–Ga, Ni–In intermetallic compounds, and the shape memory compound NiTi using ²⁷Al, ^47,49Ti, ⁶¹Ni, ⁷¹Ga and ¹¹⁵In nuclear resonance. Components of the ⁶¹Ni NMR spectrum for a series of cubic Ni_1−xAl_x and Ni_1−xGa_x specimens on either side of x = 0.5 have been identified as due to Ni substitutions and Al (or Ga) vacancies. For stoichiometric NiAl, Ni₂Al₃ and NiAl₃ the ⁶¹Ni lines are narrow and distinguished by well-separated Knight shifts, although the ⁶¹Ni lineshape for the ordered vacancy compound Ni₃Al₄ indicates a substantial nuclear quadrupole interaction at the Ni site.

The substitution for Al of the group III elements Ga and In, which have isoelectronic outer shells, acts to increase the ⁶¹Ni Knight shift such that ⁶¹K(Al)<⁶¹K(Ga)<⁶¹K(In). This trend is observed not only for cubic NiAl and NiGa, but also for trigonal Ni₂Al₃, Ni₂Ga₃ and Ni₂In₃.

In NiTi the sharp first-order transition between the (high-temperature) cubic phase and the (low-temperature) monoclinic phase has been observed while monitoring the ⁶¹Ni and ^49,47Ti NMR lineshapes as a function of temperature.

In the non-cubic phases of these materials (including hexagonal ε-NiIn) the electric field gradient tensor components V_zz and η at the atomic sites have been determined from the nuclear quadrupole perturbed ²⁷Al, ⁴⁷Ti, ⁶¹Ni, ⁷¹Ga NMR lineshapes together with ⁶⁹Ga and ¹¹⁵In NQR transition frequencies, and compared with values derived from ab initio calculation.

We present zero-field muon spin relaxation (μ⁺SR) measurements on the rare earth antiferromagnet PrO₂. Oscillations in the time dependence of the muon polarization, characteristic of a quasistatic magnetic field at two distinct muon sites, are observed below T_N, along with relaxation dominated by the magnitude of the local field. Candidate muon sites in the material are suggested on the strength of dipole field calculations. Relaxation above T_N allows a prediction of the quasielastic neutron scattering linewidth, which is discussed in terms of a simple model of magnetoelastic relaxation involving a crystal field state at an energy Δ above the ground level. The obtained value of Δ may be taken as an approximate estimate of the splitting of the Γ₈ ground state of Pr⁴⁺ in the distorted fluorite structure.

Information on the energy that is needed to transfer an electron from the valence band of an inorganic compound to a trivalent lanthanide impurity is presented. The energy is a measure of the location of the ground state of the divalent lanthanide relative to the top of the valence band. A variation with type of lanthanide is found that is the same irrespective of the type of compound (fluorides, chlorides, bromides, iodides, oxides, sulfides). The variation is anti-correlated with the known variation in fd transition energies in divalent lanthanides. Because of the anti-correlation, the energy difference between the first 4fⁿ⁻¹5d state and the bottom of the conduction band is relatively invariant with type of lanthanide ion. The difference is largest for Eu²⁺, and decreases gradually towards the end of the lanthanide series by 0.5 eV for Y b²⁺. Based on the systematic variation in charge transfer energy and fd energy, a three-parameter model is presented to position the energy levels for each divalent lanthanide relative to valence and conduction band states. Using a similar model the levels of trivalent lanthanides are positioned.

We investigate the equilibrium and out-of-equilibrium Kondo effects in a single-level interacting quantum dot connected to two ferromagnetic leads. Within the noncrossing approximation, we calculate the total density of states (DOS), the linear conductance and the nonlinear differential conductance for both the parallel and the antiparallel alignments of the spin polarization orientation in the leads, followed by a brief discussion regarding the validity of this approach. Numerical calculations show that, for the antiparallel alignment, a single Kondo peak always appears in the equilibrium DOS, resulting in conventional temperature behaviour in the linear conductance and the zero-bias maximum in the differential conductance. The strength of the DOS peak is gradually suppressed with increasing polarization, due to the fact that formation of the Kondo-correlated state is more difficult in the presence of higher polarization. In contrast, for the parallel configuration the Kondo peak in the DOS descends precipitately and splits into two peaks to form a very steep valley between them. This splitting contributes to the appearance of a 'hump' in the temperature-dependent linear conductance and a nonzero-bias maximum in the differential conductance. Moreover, application of a bias voltage can split each Kondo peak into two in the nonequilibrium DOS for both configurations. Finally we point out that the tunnel magnetoresistance could be an effective tool to demonstrate the different Kondo effects in the different spin configurations found here.

Auger capture of majority carriers by a charged edge dislocation in n-type semiconductors is considered in the case of weak saturation of the dangling bonds existing on the dislocation core. The dependence of the capture radius on the depth of the one-dimensional dislocation band and on temperature is obtained. Values of temperature and free electron concentration are estimated when Auger capture is the dominant mechanism of nonradiative recombination in crystals with dislocations.

We propose a new mechanism for binding of two equally charged carriers in a double-layer system subjected to a magnetic field of a special form. A field configuration for which the magnetic fields in adjacent layers are equal in magnitude and opposite in direction is considered. In such a field an additional integral of motion—the momentum of the pair P—arises. For the case where in one layer the carrier is in the zero (n = 0) Landau level while in the other layer the carrier is in the first (n = 1) Landau level, the dependence of the energy of the pair on its momentum is found. This dependence turns out to be a nonmonotonic one: a local maximum and a local minimum appear, indicating the emergence of a metastable bound state of two carriers with the same sign of electrical charge.

The conformal mapping method (CMM) and the eigenvalue moment method (EMM) are employed to study the eigenvalue problem defined by a free particle in a quantum lens geometry. The characteristics of the spectrum and the corresponding spatial properties of wavefunctions are studied for varying symmetry quantization numbers and lens parameter values. It is shown that the states belong to two independent Hilbert subspaces corresponding to even and odd azimuthal, m, quantum numbers. The CMM analysis is used to reduce the Dirichlet problem for the Helmholtz's equation into a 2D problem defined by a region with semicircular geometry. This approach allows one to obtain explicit analytical solutions in terms of the lens geometry. In the case of small geometry deformations (relative to the semicircular case), the solutions can be found by a perturbative method. The exact and approximate solutions are compared for different values of the lens parameters. We compare these results with those derived through the EMM approach, which, although more computationally expensive, can yield converging lower and upper bounds. The particular formulation developed here differs significantly from an earlier application of EMM to the ground state case (within each symmetry class). Accordingly, we develop the new formulation and apply it to a limited number of states in order to confirm the results derived by the other, aforementioned, methods.

Through far-infrared studies of La_2−xSr_xCuO₄ single crystals for x = 0.063, 0.07, 0.09, and 0.11, we found that only of the total holes participated in the nearly dissipationless normal state charge transport and superconductivity. We have also observed characteristic collective modes at and 22 cm⁻¹ due to the bound carriers in an electronic lattice (EL) state, and the free carriers are massively screened by the EL. Our findings lead us to propose a composite picture of the charge system where the free carriers are coupled to and riding on the EL. This unique composite system of charge carriers may provide further insights into the understanding of the cuprate physics.

We present phonon frequencies and eigenvectors for the c-axis-polarized infrared- and Raman-active modes of A_2u and A_1g symmetry in the single-layer Bi₂Sr₂CuO₆ (Bi-2201) and the bilayer Bi₂Sr₂CaCu₂O₈ (Bi-2212) cuprate superconductors within our recently proposed microscopic formulation of the electronic density response and lattice dynamics in high-temperature superconductors (HTSCs). The model includes localized electronic charge-fluctuations as well as anisotropic dipole-fluctuations. The former have been shown to be the dominating polarization process in the CuO planes while the latter are particularly important for the ions in the ionic layers of the HTSC. The effect of both types of screening contribution on the c-axis lattice dynamics of the Bi-based cuprates is investigated. A characteristic and strong renormalization of certain A_2u and A_1g modes is shown to be related to the metallicity of the BiO layers. We obtain a good agreement between the calculated A_2u frequencies and new experimental results obtained by ellipsometric measurements on single crystals and propose an assignment of the modes. Moreover, the calculated A_1g Raman modes and eigenvectors are compared with measured frequencies and the contradictory assignments of the modes found in the literature.

We discuss the transition strength between the disordered ground state and the basic low-lying triplet excitation for interacting dimer materials by presenting theoretical calculations and series expansions as well as inelastic neutron scattering results for the material KCuCl₃. We describe in detail the features resulting from the presence of two differently oriented dimers per unit cell and show how energies and spectral weights of the resulting two modes are related to each other. We present results from the perturbation expansion in the interdimer interaction strength and thus demonstrate that the wavevector dependence of the simple dimer approximation is modified in higher orders. Explicit results are given in tenth order for dimers coupled in 1D, and in second order for dimers coupled in 3D with application to KCuCl₃ and TlCuCl₃.

We report analytical calculations of the Helmholtz free energy of non-integrable anisotropic quantum XXZ chains in the high-temperature regime for several values of the spin S. Single-ion anisotropy and interaction with an external magnetic field are taken into account. The seven lowest-order terms in the high-temperature expansion of Helmholtz free energy are obtained. Our results contribute to the existing literature on high-temperature expansions and numerical studies of those models by discussing the effects of anisotropy upon their high-temperature thermodynamic properties, such as the average energy per site, the specific heat and magnetic susceptibility.

Analytical and numerical methods are used to investigate the new localized states of a surface magnetostatic wave envelope in a ferromagnet–dielectric–metal structure in the framework of the extended non-linear Schrödinger equation. The problems of the modulation instability of homogeneous states and long-wavelength modulation of a monochromatic wave in the vicinity of a 'zero-dispersion' point are discussed. The excitation conditions for a multi-soliton state are analysed. It is shown that a certain threshold amplitude of the initial distribution should be reached for the excitation of these states, while no such threshold exists for the one-soliton state. The possibility of the experimental observation of the above-mentioned non-linear states is discussed.

A ferromagnetic spin glass (FSG) is one of the three isotropic and homogeneous phases of the long-range partially ordered magnets with spin and atomic disorder which are selected by symmetry (Andreev 1978 Sov. Phys.—JETP47 411) (the others are genuine and antiferromagnetic spin glasses). The linear dynamical response to a magnetic field of two sub-phases of a FSG with drastically different dynamics, a randomly canted ferromagnet, in which the component spins create an acute angle with the summary magnetic moment, and a less-ordered skewed spin glass is analysed in the spin-wave approximation in the framework of phenomenological theory. The spin-wave damping coefficients and frequency shifts due to a magnon–magnon interaction are evaluated as functions of temperature and wavevector as well as the spectral-weight functions of the linear response to a magnetic field and the neutron scattering cross section which provides the possibility for experimental verification of the results. Substantial differences in the spin-wave characteristics of the FSG compared to those of the Heisenberg spin glass and the Heisenberg ferromagnet are found to be due to non-linear anisotropy effects in a FSG.

To clarify the combined magnetic behaviour of BaCuO_2+x, which contains in the body-centred-cubic unit cell two Cu₁₈O₂₄ spherical clusters, eight Cu₆O₁₂ ring clusters, and six CuO₄ lone units, we made measurements of magnetization, ^63,65Cu nuclear magnetic resonance (NMR) near 6 T, and pure quadrupole resonance (PQR) in zero field on samples with well-defined oxygen contents x = 0 and 0.14. Quantitative analyses of Curie–Weiss-like susceptibility data for BaCuO_2.00 indicate that Cu spins within the Cu₆ clusters go into a ferromagnetically (FM) ordered ground state with a maximum spin S_r = 3 below  K, and those within the Cu₁₈ clusters into an FM-like ground state below  K with net spin that is less than the maximum possible spin S_s = 9. For BaCuO_2.14 the Cu spins within the Cu₆ clusters go into an FM ordered ground state at a lower temperature of  K, and these clusters exhibit no signature of long-range magnetic ordering down to 2 K. The NMR spectra for BaCuO_2.00 observed below  K have a trapezoidal structure that is known to be a characteristic of long-range magnetic ordering, and is consistent with the antiferromagnetic ordering of the Cu₆ clusters below T_N = 15 K predicted previously. On the other hand, the spectrum for BaCuO_2.14 observed in a temperature range between 1.2 and  K has an electric quadrupole-split powder pattern structure. This provides microscopic evidence that there is no long-range magnetic ordering in BaCuO_2.14. The addition of 0.14 moles of oxygen atoms results in a large suppression of some magnetic entities. We find, however, some anomalies in the NMR and PQR data for BaCuO_2.14 near  K, which could be associated with short-range ordering of the paramagnetic Cu₁₈ clusters.

We report on measurements of the Knight shift K and spin–lattice relaxation rate λ of positive muons (μ⁺) implanted in cubic CeAg, a compound which is known to possess ferromagnetic ( K) and ferroquadrupolar (T_Q = 15–16 K) order. The zero field (ZF) λ implies randomly and isotropically fluctuating Ce 4f moments above T_Q which show a trend to a, perhaps critical, slowing down on approaching T_Q. Below T_Q the fluctuations are restricted to the now tetragonally distorted axis of CeAg. In a longitudinal field the generally isotropic fluctuations show a tendency towards freezing near T_c, not T_Q. The Knight shift results reveal an anisotropic contact hyperfine coupling parameter A_c, which displays abrupt changes at T_c. ZF measurements below T_c reveal the presence of discrete spontaneous internal fields. The results on λ and A_c are interpreted as arising from quadrupolar effects, concerning in particular the induced spin polarization of the conduction electron system.

Magnetism of a misfit layered cobaltite [Ca₂Co_4/3Cu_2/3O₄]_x^RS[CoO₂] (, RS denotes a rocksalt-type block) was investigated by a positive muon spin rotation and relaxation (μ⁺SR) experiment. A transition to an incommensurate (IC) spin density wave (SDW) state was found below 180 K (= T_C^on); and a clear oscillation due to a static internal magnetic field was observed below 140 K (= T_C). Furthermore, an anisotropic behaviour of the zero-field μ⁺SR experiment indicated that the IC-SDW lies in the a–b plane, with oscillating moments directed along the c axis. These results were quite similar to those for the related compound [Ca₂CoO₃]_0.62^RS[CoO₂], i.e., Ca₃Co₄O₉. Since the IC-SDW field in [Ca₂Co_4/3Cu_2/3O₄]_0.62^RS[CoO₂] was approximately the same as those in pure and doped [Ca₂CoO₃]_0.62^RS[CoO₂], it was concluded that the IC-SDW exists in the [CoO₂ ] planes.

The dynamic hysteresis of the first-order phase transition around the transition point T_c in ferroelectrics is studied by investigating the dynamic response of the Landau–Devonshire (η²)³ model to a time-varying external field of frequency f and amplitude E₀. It is revealed that the single-loop hysteresis as obtained above the upper critical point T⁺ and below the absolute instability point T₀ shows dynamic behaviours very different from the double-loop hysteresis obtained between T_c and T⁺. An extensive calculation reveals a power-law scaling for hysteresis area as a function of E₀, while no reliable power-law scaling for the area as a function of f is available in the low-f regime. The scaling for the double-loop hysteresis seems not to fall into the same class as single-loop hysteresis. Furthermore, the power-law exponents, if any, are somewhat temperature-dependent, but this dependence is very weak in the regime of low E₀ and high f.

Raman scattering investigations of cyanoadamantane (CNa) plastic crystals were performed in the 0–10 GPa pressure range. A phase transition towards an ordered high-pressure (HP) phase is clearly identified. However, this transition can be easily bypassed upon pressurization, placing the initial plastic phase in a metastable situation. Between 0.3 and 0.7 GPa the escape from metastability is observed through the isobaric transformation towards the ordered HP phase. Above 0.7 GPa, the lifetime of the metastability appears to be infinite with respect to the experimental timescale, suggesting that the glassy crystal state was reached upon pressurization. Above 10 GPa, the metastable plastic phase, as well as the ordered HP phase, convert into a conventional amorphous state. The structural features of the ordered and the glassy crystal HP phases, determined from Raman investigations, were compared to those of the ordered and glassy crystal phases usually obtained upon cooling.

Strong photoluminescence bands range from 300 to 600 nm at room temperature have been observed in anodic alumina membranes (AAM). It was found that the photoluminescence (PL) intensity and peak position of AAM depend strongly on the excitation wavelength, and the PL intensity of AAM prepared in C₂H₂O₄ is much higher than for AAM prepared in H₂SO₄. There are two peaks in the PL bands; one is at constant wavelength of 460 nm, and the other increases almost linearly from 420 to 465 nm with excitation wavelength for AAM prepared in C₂H₂O₄ and from 360 to 465 nm for AAM prepared in H₂SO₄. Annealing treatment of the as-prepared AAM results in an apparent reduction of the intensity of the blue emission at shorter excitation wavelength, while at longer excitation wavelength (longer than 320 nm) the PL intensity firstly increases, and at 500 °C reaches a maximum value, then decreases. It is considered that there are two PL centres; one originates from the oxygen-related defects in the barrier layer, and the other is correlated with the aluminum incorporated into the anion-contaminated alumina layer in the AAM.