Table of contents

It is shown that, because of dipolar forces, above the Curie temperature the linewidth Gamma (q) for small wavevector q cannot be fitted with the diffusion law corresponding to spin-conserved dynamics. The theoretical results obtained are qualitatively conformable to the experimental results obtained by F. Mezei (1982) with new high-resolution neutron scattering techniques.

The authors have carried out pseudopotential calculations concerning the electronic structure of (100) GaAs-Ga_0.8Al_0.2As superlattice with period approximately (70+70) AA. They recover the states reported earlier in optical experiments and in envelope function calculations. They also find new well localised states lying close to the top of the confining barriers. These states have large amplitudes in the confining alloy layers.

The electronic structure of Au⁺, Au⁰ and Au^-substitutional impurities in silicon are calculated within the framework of the self-consistent-field multiple-scattering molecular cluster model. The results show that centres behave basically as a 'perturbed' silicon vacancy with a filled gold-derived atomic d level localised below the bottom of the crystal valence band.

Addition gaps in the energy spectrum of a two-dimensional electron gas in a strong magnetic field due to electron-electron interaction seem to be responsible for the fractional quantum Hall effect (FQHE). The authors have analysed the structures attributed to the FQHE in the components rho _xx, rho _xy and sigma _xx of the resistivity and conductivity tensor as a function of the magnetic field at Landau filling factors up to nu =3. For the first time the values for the energy gaps at nu =⁴/₃ and nu =⁵/₃ are determined and resistivity values for the corresponding Hall plateaux are reported.

The author presents initial results for the one-electron theory of electron states at a semiconductor heterojunction under the influence of a periodic potential supplied by an interdigitated finger gate structure. The effects of magnetic field and the likely many-body corrections or instabilities are outlined.

A phenomenological expression for the high-temperature susceptibility is obtained by making a small modification of the binomial expansion of a power law. The equation contains the 'universal' exponent which pertains to the to the approach to T_c and an additional exponent to describe the power law in the high-temperature limit. The expression provides an alternative to the method of Pade approximants for analysing high-temperature series expansions in powers of inverse temperature. The conclusions of M. Fahnle and J. Souletie (1984) that there is interesting physics in the high-temperature power laws is supported.

The P_In antisite centre has been observed in electron-irradiated InP by monitoring the magnetic circular dichroism at 0.95 mu m (1.3 eV). A direct comparison of the EPR and the MDC-ODMR results shows that the optical method is approximately 100 times more sensitive. A new broad resonance with g=1.93+or-0.01, Delta B=75 mT which may be due to vacancies has been observed via an absorption at 1.08 mu m (1.15 eV). The P_In resonances have not been observed in undoped InP.

The photothermal displacement response of GaTe as a function of frequency is presented and anomalies in this behaviour are interpreted in terms of free carrier lifetime and diffusion effects.

Photoinduced absorption shows a peak in neutral polythiophene film at 1.35 eV. Chemical doping gives rise to more complex absorption in the region 0.5 to 2 eV. These absorptions are discussed in a polaron framework and comparisons are drawn with observations on other polymers.

A theory of the quadrupolar relaxation rate, R_Q, in liquids is developed. In the process, the time dependence of the electric field gradient correlation function is investigated, and a formally exact result is obtained for the leading term in an expansion about t=0. An earlier theory, which has serious defects in this limit, is corrected, and the authors derive an expression for the correlation function which displays the correct physical behaviour. By means of this result, it is shown that when the field gradient is short-ranged, the quadrupolar relaxation rate is determined by rapid adjustments in the positions of the nearest neighbours of a relaxing nucleus. Calculations of R_Q in liquid neon are in good agreement with experiment. When applied to liquid rubidium near the melting point, the relaxation rate obtained lies within the (wide) range of experimental uncertainty.

The thermodynamics is studied for two exactly soluble models by means of the Bethe ansatz method: the lattice quantum sine-Gordon model and the lattice massive Thirring model proposed by Izergin and Korepin (1981-82), and Weiss and Schotte (1983), respectively. The specific heat is analysed for each of these models as regards the effect of the cut-off parameter. The classical limit of the free energy is also discussed for the lattice quantum Sine-Gordon model.

The second-order elastic constants, their hydrostatic pressure derivatives and the third-order elastic constants of cubic K₂SnCl₆ have been measured between room temperature and the temperature of the phase transition of the soft rotary optic mode, T_c1, at -10.8 degrees C. It is found that the acoustic-mode vibrational anharmonicity of K₂SnCl₆, especially that of the (110) polarised shear mode propagating in the (110) direction, is strongly enhanced as the phase transition is approached: coupling between the softening optic mode and the acoustic modes markedly influences the acoustic-mode anharmonicity. Study of the Gruneisen parameters associates the lambda -point anomaly in the specific heat with the mode softening.

Linear response formulae are derived for the phenomenological coefficients for matter transport in a system characterised by a set of jump probabilities. The expressions for the isothermal transport coefficients are the same as those originally derived by Allnatt (1981-82) but the expressions for the anisothermal transport coefficients are different. The details of the derivation differ somewhat from those of the earlier work and are clearer and technically more satisfactory. The matrix method developed by Okamura and Allnatt (1983) for calculations of the isothermal transport coefficients in dilute binary alloys is extended to the anisothermal coefficients. As an example, new exact formulae for the terms in these coefficients which are of first order in the concentrations of paired and unpaired defects are calculated for the five-frequency model of the vacancy mechanism in a dilute FCC alloy. The results reduce to earlier less exact expressions in the appropriate limit.

An extension of effective-medium theory for bond percolation problems is given to include effects of hopping to more than adjacent neighbours. In this approximation the phase boundary separating the conduction and insulating phases is a Z-1-dimensional hyperplane in the space of occupation probabilities if Zth neighbours are included. Long-time and critical properties remain the same as for the nearest-neighbour model. A quasi-transition is found if the hopping rates for the nearest-neighbour model. A quasi-transition is found if the hopping rates for neighbours of different order differ grossly. The dynamical features are discussed numerically in two-dimensions.

The results from three different theoretical approaches to the non-perturbative optical susceptibility of semiconductors near the biexciton resonance are analysed and the origin of their differences is elucidated.

In the exciton-neutral-donor complex the electron-hole exchange interaction mixes states of opposite symmetries with respect to the electron interchange. The present variational perturbation procedure has the advantage of scaling the results by a perturbation parameter proportional to the interband exchange integral. The calculated oscillator strength for the bound exciton recombination is in reasonably good agreement with the measured radiative lifetime of exciton bound to a neutral donor in cadmium sulphide.

The non-linear Poisson-Boltzmann equation for the self-consistent electrostatic potential of the 'Coulomb gas' of point defects in an ionic crystal is solved in the plane-symmetry case. The obtained distributions of defects form a planar periodic 'substructure' in the crystal. The periodic potential corresponding to this 'substructure' is included in the Schrodinger equation, and the energy band spectrum following from this equation is obtained. The influence of the doping of the crystal on the width of the energy bands and the change of the electron's 'effective mass' due to the periodic potential are investigated.

Using a tight-binding parametrisation of the band structure of SrTiO₃ together with the Green function method, the authors calculate the energy levels of the Fe⁵⁺, Fe⁴⁺, Fe³⁺ and Fe²⁺ ions, with and without spin polarisation. The shape of the optical transition spectra is deduced and the results are compared with experimental results. A comparison is also given with a previous X alpha calculation.

Reports resistivity measurements along the high-conductivity axis of Per₂Pt(mnt)₂, Per₂Pd(mnt)₂ and Per₂-Au(mnt)₂. At low temperatures a clear metal-insulator phase transition is observed at 6.5K and 28K, respectively for the first two salts, while it appears very gradually for the Au salt. The X-ray study shows above these critical temperatures one-dimensional structural fluctuations for Per₂Pt(mnt)₂ and Per₂Pd(mnt)₂, which are attributed to the M(mnt)₂ stacks, and no such scattering for Per₂Au(mnt)₂, down to 12K. The diffuse scattering 'condenses' in superstructure reflections below 28K in Per₂Pd(mnt)₂, which thus might undergo a spin-Peierls phase transition. The nature of these various phase transitions is discussed.

Infrared absorption studies are reported for CdTe containing simultaneously a group IIIA impurity (Al, Ga or In) added during crystal growth and lithium (⁶Li and ⁷Li isotopes) introduced by diffusion. Absorption bands observed in the spectral region 250 to 330 cm^-1 are explained in terms of localised vibrational modes of two kinds of defects: (i) isolated Li occurring as an interstitial species, and (ii) M_Cd-V_Cd-Li_i defect complexes, where M_Cd denotes Al, Ga or In on a Cd site, V_Cd denotes a cadmium vacancy and Li_i denotes lithium in an interstitial position.

The fluorescence line narrowing (FLN) of the N₁ and N₂ lines of pairs of Cr³⁺ ions in concentrated ruby has been studied. The structure of the narrowed lines has been explained and measurements of the homogeneous width have been improved. By use of the FLN and time-resolved spectroscopy, spectral transfer within the N₁ line has been investigated. A two-phonon-assisted process has been found to be responsible for the transfer.

The electronic contribution to the energy of surface phonons is decomposed into a wavevector-dependent term which depends only on the response function in the surface region, plus a wavevector-independent term giving the energy change due to a uniform surface displacement. The wavevector-dependent term is evaluated for the unreconstructed Mo (001) surface, using the matching Green function method to calculate the electronic response function, and the rigid muffin-tin approximation to determine the electron-phonon matrix elements. It is found that the large density of the surface states at the Fermi energy, and the states bonding the surface atoms to the bulk, favour lateral displacements, through the wavevector of the corresponding surface phonon anomaly disagrees with the experimentally observed reconstruction. Periodic vertical displacements of the surface atoms are energetically unfavourable.

Resistance measurements and magnetoresistance measurements on thin Ru and RuO₂ films at low temperatures are reported. The results of these measurements are confronted with recent theories of electron localisation and electron-electron interaction in two-dimensional systems. The measurements on the Ru films can be explained with these theories. It is concluded that the temperature dependence of the resistance of the Ru films is largely due to electron-electron interaction effects. The magnetoresistance of the Ru films consists of a temperature-independent part caused by localisation effects and a temperature-dependent part caused by interaction effects. In the case of RuO₂ films a transition from strong to weak localisation is observed as a function of film thickness. All these films, however, fall in a region where theory is not so well developed. If the authors' measurements on the RuO₂ films are confronted with existing theories, their measurements can be explained in qualitative terms. The conclusion is that Zeeman effects are extremely important in these films. Finally a saturation of the resistance was observed at extremely low temperature, in Ru as well as in RuO₂ films, and an explanation for this is suggested.