Table of contents

The question of atomic and electronic structure of grain boundaries in Si is addressed within the tight-binding approximation with a self-consistent charge neutrality condition. The authors use simultaneous recursion in real space to calculate inter-atomic forces and relax reconstructions of Sigma =3(112) and Sigma =5(130) grain boundaries. Consequences of over- and under-coordination at defects are illustrated, and relaxed atomic structures are compared with recent lattice imaging and electron diffraction experiments.

The phonon spectrum of YBa₂Cu₃O_y depends on the oxygen concentration. With decreasing oxygen content and LO mode at 655 cm^-1 increases rapidly while the mode content between 220 and 400 cm^-1 decreases. These changes are accompanied by an equally rapid increase of a band of TO modes between 230 and 365 cm^-1. The feature at 655 cm^-1 may be associated with an A_2u mode involving asymmetric Cu-O motion parallel to the c axis. Removal of the chain oxygen atoms reduces the damping of this motion and allows the A_2u mode to grow while leading to a redistribution of phonon energy from LO to TO modes below 300 cm^-1.

The expression of the von Weizsacker term consisting of the kinetic energy density of an inhomogeneous fermion gap is derived from a variational standpoint when the density varies rapidly. It may provide insight or severe as a starting point for a Kohn-Sham-type self-consistent calculation in density functional theory.

It is pointed out that the restricted Hartree-Fock singularity at the Fermi wave-number for a metallic chain is not limited to the partially filled uppermost band(s), but occurs in all bands. Thus the density of states vanishes in all bands at the energy corresponding to the Fermi wavenumber. Possible implications for the characterisation of the Peierls' instability are discussed.

The authors report photo-induced infrared absorption and photo-induced electron spin resonance measurements on crystalline powders of polydiacetylenes (PDAS). The results imply long-lived, localised, charged, spinless photo-excitations. They have observed metastable photo-induced absorption from the vibrational modes of the PDA backbone and an associated overlapping broad electronic transition. The authors conclude that bipolarons are created through inter-chain charge transfer and coexist with the intra-chain singlet and triplet (neutral) excitons. Weak inter-chain hopping inhibits recombination and leads to metastable bipolarons (with lifetimes of the order of 30 minutes) at low temperatures.

The authors have calculated the band structure for positrons in the high-temperature superconductors YBa₂Cu₃O₇ and YBa₂Cu₃O₆ using the LMTO ASA method. On the basis of the results they argue that it is very unlikely that oxygen vacancies will trap positrons in these compounds.

Time-resolved photoconductivity measurements are reported for pulsed excitation of highly oriented poly(p-phenylenevinylene) (PPV) films. These measurements reveal a large, short-lived photoconductive response separate from the long-lived component that has been investigated by several other groups. The authors report the dependence of this short-lived photocurrent upon the sample temperature, upon the polarisation, energy and intensity of the excitation light and upon the magnitude and orientation of the applied DC bias field.

The current-voltage characteristics for a Bi-Sr-Ca-Cu-O bulk specimen incorporating a manually carved constriction show both zero-voltage and finite-voltage pair currents, flux flow and quasi-particle tunnelling regimes. The finite voltage pair current regime may be understood to arise due to non-equilibrium enhancement of Josephson supercurrents. The pair current amplitude modulates with applied DC magnetic fields. Microwave-induced Shapiro steps are clearly seen in the I-V curves. The Josephson effects arise due to the presence of three-dimensional arrays of inter-grain junctions.

The authors have measured the susceptibility, magnetisation, resistivity, thermopower and Hall coefficient for a number of samples of the recently discovered superconductor of nominal composition Bi_2+xCa_1-xSr₂Cu₂O_{8+ delta} . They find that sintered ceramic pellets show a small volume fraction superconducting at 110 K and a further transition with onset at 85 K to a zero resistance state at 70 K. The Meissner effect shows 13% by volume flux exclusion at 20 K. In the normal state, they find a temperature-independent magnetic susceptibility of about 2.8*10^-4 emu/(mol Cu); this value is relatively insensitive to variations between batches. In contrast, both the thermopower and Hall coefficient show a wide spread in behaviour, both p- and n-type, and this they attribute to the presence of other metallic phases in addition to the superconducting phases.

The cumulant-averaging method for dilute systems with off-diagonal and environmental disorder is discussed. The averaging procedure used by several authors is shown to produce inequivalent results when applied to a number of equivalent equations for the conditional Green function. The difference seems to be due to an incomplete projecting-out of unphysical contributions to the average Green function from the vacancy sites.

Raman scattering has been measured in amorphous and crystalline diglycidyl ether of bisphenol A (DGEBA) and in epoxy resins made from DGEBA hardened with ethylene diamine. The spectra of the amorphous materials show a constant frequency dependence beyond a certain crossover frequency which depends on the characteristic molecular length. These results are compared with similar data obtained from inelastic neutron scattering and good agreement with the phonon-fraction hypothesis of Alexander and Orbach (1982) is obtained. The results yield a fracton dimensionality of 4/3 if the superlocalisation exponent d_phi is taken to be equal to the resistivity exponent zeta . In the crystalline DGEBA, characteristic peaks in the Raman spectrum compare well with those obtained by inelastic neutron scattering.

A new phase transition is observed in the KLiSO₄ crystals below 77 K using MoO₄^3- as a paramagnetic probe. The EPR spectra of this molecular ion, which substitutes for the host SO₄^2- ion, show triclinic symmetry at 77 K due to the joint action of a trigonal crystal field and a trigonal Jahn-Teller effect of E(X)e type. Above 77 K, motional averaging in the a-b plane occurs; thereby an axial spectrum is observed. Below 64 K, the two sulphate sites are no longer equivalent resulting in the appearance of two distinct EPR spectra of the molybdate ion. The possible space groups of the crystal in the vicinity of the phase transition are discussed. The authors' results indicate that one needs to distinguish carefully between the local dynamics and the structural changes in this crystal.

The analytic theory of continued fractions is used to discuss the dynamic properties of a system described by a tridiagonal eigenvalue equation, equivalent to a one-dimensional tight-binding Hamiltonian in which only the nearest-neighbour hopping energy coefficients are finite. Expressions for the density of states and the dynamic susceptibility are reported. The expressions for these response functions reduce to simple closed forms when the coefficients in the eigenvalue equation are periodic functions. A model of linear transverse spin fluctuations in a longitudinally modulated magnet is chosen to motivate the analytic work, and an example of the spectrum of spontaneous spin fluctuations is provided together with the corresponding Lyapunov exponents.

A parameterised tight-binding model is used in a CPA description of the electronic states in substitutional nonisoelectronic alloys Cd_1-xPb_xS in the rock-salt phase. The calculated densities of states and spectral densities show a major reconstruction of the band-gap region. In addition, new hybridisation-induced features in the upper valence band are predicted.

Using the augmented spherical wave methods the authors calculate the electronic structure of the transition-metal monoborides of the form TM-B (TM=Sc, Ti, V, Cr, Mn, Fe, Co, Ni or Cu) for five different structure types (FeB, CrB, NaCl, CsCl and CuAu). From the curves of total energy against volume, they determine the preferred crystallographic modification which confirms the experimental trend. For FeB and MnB, their spontaneous magnetic ordering is taken into account. The equilibrium volumes and magnetic moments are in agreement with measurements. The binding mechanism in the FeB and CrB structure types exhibits not only transition-metal-boron interactions but also strong boron-boron bonds along the chains which are responsible for their stability.

Using the augmented spherical wave method, the authors calculate the electronic structure of the transition-metal semi-borides of the form TM₂B (TM identical to Ti, Cr, Mn, Fe, Co, Ni or Mo) in the Al₂Cu structure. Besides the transition-metal-transition-metal interaction, the bonding mechanism exhibits also TM d-B p hybridisation which also influences the formation of a magnetic moment in Fe₂B and Co₂B. The calculated magnetic moments and hyperfine field are in agreement with experiment.

Measurements of the Knight shift performed at helium temperatures are reported for heavily doped, n-type, indium arsenide and indium antimonide in magnetic fields up to 8.0 T. Such fields, in samples of the doping densities used here, are sufficient to fully polarise the electronic spins in some of the samples, so that all ambiguity about the state of the electronic magnetisation is removed. In such circumstances the shift is a direct measure of the magnitude and character of the conduction electron wavefunction at the group III and the group V nuclear sites. The results show strong indium character at the bottom of the band, particularly so in the indium arsenide samples. Quantitative comparisons are made with nonlocal pseudopotential calculations of band structures and charge densities in the two semiconductors.

The thermopower and resistivity of a-Tl_100-xTe_x have been measured in the temperature range 5-60 K for x<30. There is evidence in both measurements of an impending metal-insulator transition as the concentration of Te is increased. The data gives some support to recent predictions regarding the thermopower behaviour near a metal-insulator transition. Annealing curves are also presented.

The electrical conductivity and differential thermal analysis (DTA) characteristics of RECoO₃ (RE=La, Nd, Gd, Ho, Y) have been reexamined. In general, the data are quite different from those previously reported. In particular, first-order semiconductor-to-metal transitions of YCoO₃, LaCoO₃ and HoCoO₃ are not observed. Instead, broad, high-order semiconductor-to-metal transitions are observed over the approximate temperature ranges La: 385-570 K; Nd: 400-590 K; Gd: 490-770 K; Ho: 540-795 K; Y: 575-850 K. The form of the transitions is similar to that observed for Ti₂O₃ where a thermally activated band overlap model has successfully been employed. A similar scheme for the rare earth cobaltates can be constructed in which Co 3d electron character is considered collective. This is contrary to the localised-to-collective phase transition model previously employed. In addition, a first-order semiconductor-to-semiconductor transition of NdCoO₃ at 342 K is observed in electrical conductivity data. The tenfold increase in conductivity accompanying this structural transition can also be understood within a collective 3d-electron model.

An analytical Green function approach to the calculation of the diffusion coefficient of hot electrons in two-valley semiconductors at high electric fields is presented. Correlation functions for velocity fluctuation and population fluctuation are derived, respectively, to determine the thermal fluctuation term and carrier transfer term of the diffusion coefficient. New terms related to those cross-correlation functions for the velocity and population are introduced. Reasonable agreement between calculated and measured diffusion coefficients is obtained when the method is applied to n-type GaAs at room temperature.

The local structural excitations (LSEs) in the new superconducting oxides, based on the migration of the oxygen atoms in the Cu-O planes, are analysed with a simple microscopic model which involves two different local configurations and the transitions between them. A new electronic pairing mechanism via exchange of LSEs is suggested. By taking into account both the electron-phonon and the electron-LSE interactions, a formula for T_c in the strong-coupling case is derived. The results give a new insight into the relation between the oxygen vacancy structure and the superconducting transition temperature in these materials. It is shown that the orthorhombic structure with an ordered distribution of oxygen vacancies is more favourable for high T_c than is the tetragonal structure with disordered oxygen vacancies; for superconductors with a dominant LSE mechanism the isotopic effect is not easily observed under ordinary conditions. These arguments are in qualitative agreement with the current experimental results. Methods for obtaining further experimental confirmation of this are suggested.

A spin-1 Ising model with both linear and biquadratic interactions is studied in the effective-field approximation of Honmura and Kaneyoshi (1978). It is found that the Curie temperature is double valued under the condition that the ratio of biquadratic to bilinear exchange constants is negative, suggesting the occurrence of the reentrant phenomenon. The results for square and cubic lattices are presented.

Single-crystal neutron diffraction measurements have been used to study the long-range magnetic ordering in cupric oxide, CuO. An incommensurate antiferromagnetic structure forms below the Neel temperature on 230(1) K, with a propagation vector (0.506(1)a*-0.483(1)c*) which remains constant down to a magnetic phase transition at 213(1) K. Below the latter temperature, the structure remains antiferromagnetic with a commensurate propagation vector (1/2 0-1/2), and this structure persists to the lowest temperature reached in the investigation (20 K). The arrangement of the copper moments in both phases is such that the n-glide perpendicular to the b axis of the monoclinic cell, space group C2/c, does not reverse the direction of the spin. The two magnetic sublattices related to the C-face-centring scatter in phase quadrature and the relative directions of the spin on them could not be determined. Good agreement is obtained for the commensurate phase with a multipole model for the copper magnetisation density and spins of 0.65(3) mu _B directed parallel to b. The lower sublattice magnetisation in the incommensurate phase precluded a meaningful multipole fit, but a reasonable agreement is obtained with a model in which the spins rotate in the a-c plane following an elliptical envelope with major axis directed 33(2) degrees to c in beta obtuse and a maximum moment of 0.38(2) mu _B at 215 K. The paramagnetic scattering at ambient temperature and 550 K was measured to try to find the origin of the peak in the susceptibility. No significant paramagnetic scattering could be obtained from a powdered sample although the sensitivity of detection was some five times that required to observe the scattering from a Cu²⁺ ion in an ideal paramagnet.

The alkali metal thioferrates are quasi-one-dimensional magnetic compounds. The authors have studied the intra-chain exchange and anisotropy constants (J and D_x respectively) by means of Mossbauer spectroscopy. The spectra at temperatures greater than the transition temperature have been fitted and the rate W of electron spin flips has been determined. Then, taking into account the relation between W and the magnetic parameters, they have obtained that square root (JD_x)=(107+or-14) K. Finally, a comparative discussion has been given.

The authors have studied the saturation behaviour of the electron paramagnetic resonance (EPR) spectrum of defects in crystals when they undergo a reorientational motion with particular reference to the H_i⁰(Li) defect in MgO:Li. The EPR spectra of this centre have been measured as a function of temperature and microwave power. Two features appear as a consequence of the simultaneous presence of saturation and motion. The saturation degree of a particular line depends on its allowed or forbidden character and also on the extent to which motion affects that line. Different lines of a given spectrum can show different dependencies on microwave power. In order to understand this behaviour, they have performed a lineshape simulation analysis. The stochastic Liouville formalism, which has been extensively used in the nonsaturated case, has been extended to include saturation phenomena. Several approximations have been made to reduce the dimension of the problem. The calculations reproduce the experimental facts qualitatively, although a quantitative fitting has not been attempted owing to the presence of unresolved interactions that deform the lineshape.

For the description of photoluminescence spectra of highly doped III-V compounds, the authors present a lineshape model without fitting parameters. Their calculations have been achieved in the framework of Klauder's fifth approximation. Using a screened and factorised Coulomb potential, they have applied this theory for the first time to the analysis of experimental spectra measured on n-InP and n- and p-GaAs. For slightly compensated material, their computations are in good agreement with experiment. On the basis of a quasi-particle concept that influence of doping on the energy dispersion near the band bottom is discussed. It appears that in highly doped direct-gap material any deviations in the corresponding band from the parabolic k dependence usually assumed to be related to high k-values are not the only factor to be considered.

The luminescence spectra of RbMnCl₃, Rb₂MnCl₄ and CsMnBr₃ crystals doped with Cr²⁺ exhibit a feature at approximately 650 nm which must be assigned to an impurity emission. Time-resolved studies as well as the excitation spectra indicate that the impurity emissions result from energy transfer from the host lattice (exciton capture) rather than from direct absorption. The impurity emission is assigned to the ³E to ⁵E transition of the Cr²⁺ ion. In addition, there is evidence for a weak emission in the infrared, between 780 and 1050 nm, which appears to arise from the ³E to ³T₁ transition of the Cr²⁺ ion. At low temperatures the ³E state lifetimes range between 150 mu s and 1.1 ms depending on the host lattice. Emission from the ³E state is an unexpected observation since the ³E state is not the lowest-energy excited state of the Cr²⁺ ion.