Table of contents

The authors have tested two explanations for the discontinuity-or 'kink'-in the DC conductivity versus 1/T characteristics of undoped a-Si:H. The statistical shift of the Fermi level leads one to predict the kink qualitatively but it is more difficult to see how it can be used in accounting for the data quantitatively. On the other hand, thermal equilibrium processes, although not quantitative, provide a general explanation for discontinuities in both doped and undoped a-Si:H. From the experimental tests carried out the authors infer that, although a contribution from the statistical shift cannot be excluded, the most likely cause of the kink is the metastability.

The crystal structure and absolute optical chirality of Tm₂Ge₂O₇ (M_r=595.04) have been determined by X-ray diffraction at room temperature. An optically laevorotatory crystal had space group P4₁2₁2(D₄⁴), with a=6.764(2), c=12.293(2) AA, V=562.4(4) AA³, Z=4, D_x=7.027 mg m^-3, Mo K alpha , lambda =0.71069 AA, mu =42.03 mm^-1, F(000)=1032. The final R-factors were R=0.0276 and wR=0.0369 for 1568 reflections. It is found that Tm³⁺ is unusual in being coordinated by seven oxygens. The refractive indices and optical rotatory dispersion have been measured. Using a point dipole polarisability theory it was possible to compute the optical rotatory power, and this was found to be in close agreement with the observed value. It is found that the main contribution to the optical rotation is from the O1, O2 and O3 oxygens, with O3 playing the major role.

The authors have calculated the constant volume lattice specific heat and the elastic constants of sodium and potassium using local, first-principles pseudopotentials. This kind of pseudopotential has been useful in the calculation of properties of aluminium and lithium and in the calculation of the phonon limited resistivity of sodium and potassium. It is obtained from the induced electron density around an ion in the corresponding electron gas. From each pseudopotential the authors obtained the interionic potential, the phonons (which are calculated by the harmonic approximation) and finally the lattice specific heat and the elastic constants for sodium and potassium. The results are in good agreement with experimental results.

A generalised Luttinger model is proposed and studied in the spinless and spin-dependent cases. The solution of the back-scattering and Umklapp scattering problems are obtained using a new strategy. For the standard Luttinger model the results already known are obtained again.

Fully relativistic first-principles electronic structure calculations of both the average structure and a supercell approximation of silver-free incommensurately modulated calaverite (AuTe₂) are presented. The differences between the results of both calculations are relatively small for the occupation numbers and the density of states, but quite dramatic for the shape of the Fermi surface. From the occupation numbers it is concluded that a previously proposed idea for explaining the modulation, based on mixed valencies for the gold atoms, is probably not applicable. The calculated Fermi surface of the average structure shows that the modulation cannot be understood in terms of Fermi-surface nesting either. The density of states in the supercell approximation compares very favourably with recently obtained X-ray photoelectron spectroscopy data. A rigid-potential calculation shows that the integral of the one-electron valence energies for the supercell is substantially more negative than the corresponding energy for the average structure, while the electrostatic energy difference has the opposite sign but is much smaller. This provides a qualitative indication of the electronic instability of the average structure with respect to the modulation of the supercell. Finally the authors conclude that Te s-like states a (Te p-Au d)-like complex dominate the energetics of the modulation.

The maximum anisotropic approximation for electron transport in an electric field is extended to the position-dependent case. Models utilising nonpolar optical, acoustic and piezoelectric phonon scattering processes are constructed. The piezoelectric model produces the Euler-Darboux equation in the position-dependent case and an analytical solution is given. Analytical solutions are given for each of the models in the position-independent case.

Very accurate measurements of the thermal conductivity of pure silver (RRR=1912) reveal abrupt changes in the temperature dependence of the inelastic electron-phonon scattering component of the thermal resistivity W_v at about 5, 7 and 12.8 K. Similar changes in the temperature dependence of the electrical resistivity of silver are found at about 4, 7 and 12 K. These results are consistent with previous results found in the electrical and thermal resistivities of Cu and Au and they remain unexplained.

The Landauer-Buttiker formalism for an electronic microstructure with many terminals is extended to account for temperature changes in the reservoirs and heat fluxes in the terminals. Terminal relations are developed in the presence of an arbitrary applied magnetic induction field which becomes uniform in the neighborhood of each terminal. They use temperature changes in the reservoirs and either chemical potential changes in the reservoirs or charge fluxes in the terminals as independent variables. In both cases formulae for the transport matrices are given in terms of the scattering matrix of the microstructure. Onsager symmetry relations and reciprocity theorems are given for electrical, thermal and thermoelectric configurations. The behaviour of quantum point contacts is outlined.

The authors show that the nonlinear response of certain direct gap semiconductor superlattices is greatly enhanced when the separation of the lowest conduction bands is comparable with the magnitude of the principal gap. The mechanism which results in this enhancement is virtual transitions to higher subbands. Examples of (InAs)_1-x(GaSb)_x/AlSb superlattices expected to exhibit the enhanced nonlinear response in the near-infrared region of the spectrum are presented. They also investigate nonlinear response in semiconductor superlattices at ultrashort times ( approximately 100 fs). In the authors scheme a full account is given fro the first time of band structure effects upon transients. They then use a three-level model to determine the conditions under which predictions, based on the steady-state response theory and the golden rule, break down in structures with closely spaced energy levels.

Magnetic susceptibility and Mossbauer spectroscopy studies were carried out on the alloy series FeAl_1-xCu_x (0<x<or=0.4). The variation of susceptibility with temperature showed transitions from a paramagnetic to a spin glass phase for copper contents 0.15<or=x<or=0.25. For x=0.325 transitions from a paramagnetic to a ferromagnetic and from a ferromagnetic to a magnetic glass phase were observed as the temperature was decreased. Mossbauer spectra gave information on the microscopic conditions at the iron sites. The electron density at the iron nuclei increased linearly with copper concentration x. Distributions of hyperfine fields were observed for all alloys at 4.2 K and related to iron cluster structure. Comparison between phase transition temperatures obtained by susceptibility and Mossbauer observations gave information on the relaxation of clusters. The size of these clusters as a function of x was studied via a simulation calculation and used to account for the occurrence of a ferromagnetic phase at a critical value of x.

The Stoner model of ferromagnetism in combination with self-consistent linear muffin-tin orbital paramagnetic calculations are used for the investigation of magnetic properties of MoPd₃. The authors calculations predict magnetic ordering in MoPd₃ and they calculate the magnetic moment per unit cell of this compound.

It is well known that spin-wave resonance (SWR) phenomena can be observed in single-layer ferromagnetic films provided that the sample exhibits certain (surface or volume) inhomogeneities in its respective magnetic properties. Here,, the authors show theoretically that the observation of SWR in bilayer (exchange-coupled) ferromagnetic films is possible even if both constituent sublayers are homogeneous provided that their bulk characteristics differ appropriately from each other.

The authors study the kinetics of the formation of antiferromagnetic order in a finite spin system. The system is initially forced to be an unstable ferromagnetic state by means of an external field. When the field is switched off the system decays toward equilibrium. The finite size of the system has relevant consequences on the time of persistence of the initial unstable state. The fluctuations driving the system to equilibrium become anomalously large in the transient. The authors obtain analytic results for the transient fluctuation statistics in the limit where the average squared magnetisation relaxes much more quickly than relative orientation of the sublattices. The results have been obtained in the framework of spherical model and of molecular field approximation. The relaxation dynamics are ruled by the nonlinear Langevin equations.

The complex permittivity of a composite dielectric containing conducting inclusion is derived. The shape of the inclusions is described by their aspect ratio. The resultant equations describe the frequency dependence of the permittivity exactly, without recourse to empirically derived parameters.

The authors measured the positron annihilation average lifetime in the sintered mullite-ZrO₂ ceramic material at different ratios A/Z and including Cr₂O₃ and at three temperature values. The experimental results show that the positron lifetime can sensitively reflect the influence of different ratios A/Z and different temperature values on the microscopic structure. According to the theory of microscopic defects and phase-change information supplied by positron lifetimes, some characteristics of the sintering reaction of the mullite-ZrO₂ ceramic are explained in this paper. The results were found to be in good agreement with those obtained using other methods.

Polycrystalline FeSi and FeSi₂ samples have been examined by ESCA experiments. The valence band spectra show features that can be attributed to bonding and non-bonding states, as is the case for other transition metal silicides, where bonding is expressed in terms of hybridisation. Furthermore the authors have investigated the photoelectron spectra of samples of Si and Ge implanted with Fe at an energy of 40 keV in order to draw some conclusions regarding the character of the Fe-Si bonding in the disordered layer. The concentration of Fe as a function of the sample depth was also determined and compared with the results of Monte Carlo calculations.

For viscoelastic binary liquid mixtures, generalised hydrodynamic equations are set up, in which the relaxation of the chemical potential due to structural relaxation is taken into account. A generalised diffusion equation is obtained, and the spectra of thermal fluctuations of concentration and mass density are derived. For very asymmetric mixtures of a mobile species in a very viscous solvent, such as a glass-forming liquid near its glass transition, coupled Langevin equations for concentration and mass density are derived, which may be applied to the calculation of the intensity of non-equilibrium fluctuations as observed in quench experiments.

Phase boundaries are calculated for the freezing of a binary mixture of colloidal hard spheres which are assumed to be immiscible in a single solid phase. Results are reported for mixtures of diameter ratio gamma =0.85 and 0.65. It is shown that the fluid phase is stable to a higher density in the binary mixture than the monodisperse system. The maximum fluid freezing density occurs at the binary eutectic in suspensions rich in small spheres. The proportion of small spheres in the eutectic fluid increases sharply with decreasing diameter ratio. At phase coexistence the hard sphere crystal is compressed relative to the monodisperse system. The increase in density is most pronounced for the crystal of large spheres. The hard sphere phase behaviour is described at constant volume which enables a direct comparison with experimental measurements on colloidal systems to be made.

Results are presented of molecular-dynamics experiments in which the Lennard-Jones liquid is cooled isobarically into the metastable temperature region. At a reduced temperature T approximately=0.45 a specific-heat anomaly is observed. Results are given for the density and current-density correlation functions. The variation of viscosity and coefficient of self-diffusion with temperature can be fitted to a power-law behaviour with an exponent mu =1.8, in close agreement with predictions of recently proposed mode-coupling models.

High-resolution ( Delta E=15 mu eV) quasielastic neutron spectra have been measured for liquid methanol (CD₃OH) and methane-thiol (CD₃SH) at two temperatures within their liquid ranges. The spectra have been analysed in terms of a previously developed dynamical model which allows for the separation of the translational incoherent parts from the total dynamic structure factors. The strong differences in linewidths of the incoherent functions between the two compounds are then analysed in terms of a phenomenological model which takes into account the existence of several association states in the alcohol. The sulphur compound is then used as a reference fluid in order to carry out a comparative study of the microscopic dynamics of the two materials.

The mean square displacement of a tagged particle in a liquid is known to exhibit a diffusive linear time dependence beyond a microscopic timescale. By making use of simple mode-coupling concepts the authors derive a set of analytic self-consistent equations for the relevant dynamical quantities in this regime, namely the diffusion coefficient and the intercept. The results of the theory are successfully compared with the data obtained by simulation experiments in different systems.

Quantum evaporation of ⁴He atoms by phonons and rotons in liquid ⁴He is measured for a wide range of conditions. The results confirm that an atom is evaporated in a single-excitation to single-atom process and that the boundary conditions of conservation of energy and parallel momentum are obeyed. Excitations in the liquid are generated by pulse heating a thin metal film heater, and the evaporated atoms are detected by the energy yielded on condensation at the surface of a superconducting transition edge bolometer. The heater and bolometer can be rotated about a common axis in the liquid surface, and the collimation of the excitations and atoms into beams allows the angles of incidence and evaporation to be defined. It is first shown that the input power and pulse length must be carefully chosen to produce beams of phonons and rotons that are fully ballistic. The time of flight from the heater to the bolometer is measured, and the angular distribution of the evaporated atoms is determined for different angles of incidence. In particular, the wavevector dependence of the roton to atom pulse shape can be seen. The authors see no sign of Pitaevskii roton decay over the long liquid path lengths involved ( approximately 6 mm), and there is no indication that ripplons are created in the evaporation process.

The authors propose an acoustic method to measure directly the density of the normal component in thin films of superfluid helium. The effect of the helium film on the velocity and the penetration depth of Love and Gulyaev-Bleustein shear surface waves is analysed. The advantage of the method based on the application of shear surface waves is that the change in their velocity, as is shown, is proportional to the normal component density and the helium film thickness.

The fate of topologically stable defects in the ³He-A₁ phase, which exists in a magnetic field, is analysed when this phase goes over to the A₂ phase at lower temperatures. It is shown that in magnetic fields that are not too large the A₁ monopoles become confined in the A₂ phase. It is argued that the Mermin-Ho relation breaks down in the presence of vortons.

The authors present a first-principles calculation of the atomic structure (pair correlation function, static structure factor), the electronic structure (total and partial densities of states), and of the photoemission intensities of liquid Li and Na. They find that in liquid Li the electronic density of states shows a strong structure-induced 'Brillouin-kink' which leads to a pronounced narrowing of the width of the occupied band compared to the prediction of the free electron model. The DOS of liquid Na on the other hand is close to the free electron parabola. The origin of the difference is in the strong p component in the electron-ion potential in Li, but not in Na. The observed deviation of the photoemission spectrum of Na from a free electron form is explained in terms of the partial photoionisation cross-sections. A detailed comparison with experiment shows that the accuracy of the theoretical predictions is limited only by the local density approximation for the many-electron interactions.

Processes in a radiation defect (called the E-cell) of a crystal lattice of hydrides A_XH_Y (the ordering number Z and the mass number N of the element A must be equal to one of the following pairs of numbers: (2,3), (3,6), (4,7) or (5,10)), formed by the capture of a thermal neutron in a crystal, are considered. As a result of this capture, one of the lattice cells has a surplus of electrons forming the electronic shell of atom A. The phenomena in E-cells are mainly characterised by the following features: (i) the average density of free electrons in the central region of the E-cell exceeds 10²⁵ cm^-3; (ii) H-nuclei rearrangement to a distance <0.1 AA leads to an energy gain in the electron-nuclear E-cell system; and (iii) the probability of H-nuclei tunnelling, determined as exp(-W), is characterised by W=20-30. The electrons are stably confined in the E-cell if the pressure in the crystal is in the Mbar range.