Table of contents

The paramagnetic Meissner effect (PME) has been observed in superconductors at 40 G. The origin of the effect has been considered on the basis of the local oxygen variation which implies non-superconducting or lower- regions embedded inside the higher- layers of superconducting (SC) grains or portions. These shell-like varying- SC layers in turn apparently create `cumulatively compressed flux trapping' and hence the PME. The exhibition of PME seemingly depends on the type of defect as well as their density.

The thermodynamics of two-phase coherent equilibrium is analysed in the case of a phase transition without change of composition. The elastic contribution to the Gibbs free energy associated with a distortion of the matrix of the parent phase due to the transformation misfit strain is considered. The phase transition from the parent to the product phase is studied using Landau theory, with the transition volume change being coupled with the phenomenological order parameter. The minimization of the free energy with respect to the volume change and order parameter gives the dependence of the Gibbs energy on the volume fraction of the product phase. The transformation proceeds in a finite-temperature region with the equilibrium volume fraction depending on temperature rather than at a fixed temperature as would be expected from the Gibbs phase rule for the first-order transition.

A model of the diffusion of a tagged particle moving by hopping on a lattice, where a finite concentration c of background particles gives rise to blocking due to forbidden multiple occupancy of sites, is extended to the situation where disorder exists on the lattice. In this paper the specific case of variable bond hopping rates is considered in the strongly disordered limit where a finite concentration p of the bonds are completely blocked. The resulting self-diffusion coefficient takes the form where is the percolation limit and f is the dynamical correlation factor. It is expected that f is affected by the disorder and this is estimated by random-walk theory through a calculation of where is the angle between successive jumps of a particle and a vacancy. Also a quite comprehensive simulation study of tracer diffusion in a two-dimensional square lattice for 0 < c < 1 and is performed. The results are in good agreement with the analytical results which are known for small concentrations. Our approximate theory gives a good description over the entire range provided that the corrected form of is used.

An approximate theory for tracer diffusion on a lattice containing randomly distributed traps and in the presence of a finite concentration of diffusing particles which preclude double occupancy of any site is developed by extending earlier theories of diffusion in a many-particle system on perfect lattices using random-walk concepts. Both blocking and dynamic correlation effects are considered. The theoretical results are compared with computer simulation for a two-dimensional square lattice with two types of trap over the entire concentration range for particles and traps. The agreement between the simulation results and the theory is satisfactory and gives confidence that the approximation will be valid more widely in models of diffusion in disordered lattices.

We study a rotation-invariant Majorana fermion model in one dimension using diagrammatic perturbation theory and numerical diagonalization of small systems. The model is inspired by a Majorana representation of the antiferromagnetic spin- chain, and it is similar in form to the t - J model of electrons, except that the Majorana fermions carry spin 1 and charge. We discuss the implications of our results for the low-energy excitations of the spin- chain. We also discuss a generalization of our model from three species of Majorana fermions to N species; the SO(4) symmetric model is particularly interesting.

Based on the analysis of the group-chain scheme, the crystal-field-level fitting has been carried out for :, in which the ions occupy site symmetry positions. The RMS of energy-level fitting is . The wavefunctions obtained were used to the study of magnetic, thermal and spectroscopic properties of the crystal. The calculated g-factors confirm Karayianis' partial g-sum rule. The temperature dependence of Schottky specific heat and magnetic susceptibility agrees well with the experimental data. Using the Judd - Ofelt parameter analysis, the parameters are fitted to the absorption spectra with , and (in ). Considering the radiative, nonradiative transition and cross-relaxation in , one set of rate equations is established to study the fluorescence dynamics. The relation between the ion's population density in the excited state and the doped ion concentration is discussed, and some useful inferences are drawn.

We present the results of an investigation, both theoretical and experimental, of partially occupied and unoccupied electronic distributions in orthorhombic . The occupied Al states near are p - d hybridized in interaction with Ru 4d states, whereas they are almost purely s-like in character beyond 6 eV from . The Ru 4d states are found close to ; their interaction with the Al states produces an important depletion of the Al 3s - d and Al 3p distributions. As a result, the Al edges are somewhat pushed back from . Thus a wide range of energy with almost no states is present in the occupied band side. lies close to the bottom of the unoccupied band where both unoccupied Al states and unoccupied Ru states are present. These results are compared to previous data for icosahedral quasicrystalline .

New crystal structure maps have been proposed on the basis of the molecular orbital calculations of electronic structures. Two electronic parameters have been introduced and employed as new parameters for the classification of crystal structures. One is the bond order and the other is the d-orbital energy level of elements. Both of them change following the position of elements in the periodic table. With these parameters crystal structure maps have been constructed for aluminides, silicides, and some transition-metal-based compounds. There is a clear separation of the crystal structures on the maps. These maps are found to be applicable to the prediction of crystal structures not only for binary compounds but also for ternary compounds. The possibilities of structural modification of and by alloying are also discussed with the aid of these maps.

We study the electronic band structure and the local density of states of different GaAs - AlAs Fibonacci superlattices grown along the [001] direction. We use an empirical tight-binding Hamiltonian including spin - orbit coupling together with the surface Green function matching method. We have analysed second-to-fifth-generation superlattices with different generating layer thicknesses. A selective localization of the local density of states in the thickest GaAs slabs is found for both the highest valence and lowest conduction band states in all of the cases considered in our study.

The dielectric function and the density - density correlation function are calculated for a compositional superlattice of type II, which consists of alternate electron and hole layers (a two-component plasma) in an inhomogeneous dielectric background. The dielectric background of the electron gas is considered to be different from that of holes and the finite width of an electron (hole) layer is considered to allow both intrasubband and intersubband transitions. Our model superlattice consists of electron plasma, hole plasma, lattice vibrations of the background of the electron gas and lattice vibrations of the background of the hole gas. Electron - electron, electron - hole, hole - hole, electron - phonon, hole - phonon and phonon - phonon interactions take place in our model superlattice. Our calculation is applied to the superlattice. Variation of plasmon - phonon coupled modes and their lineshapes with (x, y) and unit cell width has been investigated in order to study the effects of semiconductor to semimetal or vice versa phase transitions. It is found that phase transition prominently affects the plasmon modes, while phonon modes remain almost unaffected. The inhomogeneity in the background of the electron - hole gas also produces a significant change in plasma frequencies. Lineshapes of coupled plasmon - phonon modes for both semimetal and semiconductor phases are calculated and are computed with those of the homogeneous background. Significant changes in peak height and half width are observed due to inhomogeneity in the dielectric background and the semiconductor to semimetal phase transition.

An approach has been suggested for integrating the superior properties of the BeTe semiconductor with the mature technology of group IV semiconductors. In a semiempirical tight-binding scheme, detailed calculations of the electronic structures of the , and superlattices are performed over a wide range of . It is found that the fundamental energy gap increases (up to 2.11 eV for BeTe/Si and 1.93 eV for BeTe/Ge at the point for n = m = 2) with decreasing superlattice period, and that the silicon or/and germanium layer plays an important role in determining the fundamental energy gap of the superlattice system due to the spatial quantum confinement effect. Two interface bands are identified in the upper region of the thermal gap in these systems, which extend over a quite different region of k-space. Furthermore, the calculated electronic structure of is found to be quite different from those of II - VI compounds grown on pure group IV semiconductors, but fairly close to their average.

We present a simulation of the motion of electrons in a mesoscopic Hall bar, scattered by a local inhomogeneous magnetic field. In the low-field regime, the Hall resistance is found to be determined precisely by the average magnetic field in the cross junction, which implies a valuable device application of non-invasive access for measuring magnetic flux, like SQUIDs do, but on a rather small (submicron) scale. The bending resistance is found to depend sensitively on the local magnetic field profile, which may also imply certain device applications, such as detecting the local magnetic properties of small objects. We also discuss briefly the asymmetric effect due to non-identical leads and asymmetric location of the field profile in the cross junction.

The kinetics of cation ordering in complex crystal structures, such as minerals and ceramics, is discussed with regard to the applicability of time-dependent Ginzburg - Landau theory (TDGL). Deviations from the predictions made by TDGL may arise due to partial conservation of continuous and non-uniform order parameters at low temperatures. Under these conditions the movement of areas with large order parameter gradients (`walls') provides a local alternative to the continuous ordering kinetics described by TDGL. A simple phenomenological description of order parameter kinetics taking these constraints into consideration can be obtained. The underlying assumption is that the geometry of spatial order parameter variation close to equilibrium resembles that of a phase-separated system. The resulting kinetics differs substantially from the predicted TDGL behaviour.

Current-driven vortex dynamics of type-II superconductors in the weak-pinning limit is investigated by quantitatively studying the current-dependent vortex dissipation of an untwinned single crystal. For applied current densities (J) substantially larger than the critical current density , non-linear resistive peaks appear below the thermodynamic first-order vortex-lattice melting transition temperature , in contrast to the resistive hysteresis in the low-current limit . These resistive peaks are quantitatively analysed in terms of the current-driven coherent and plastic motion of vortex bundles in the vortex-solid phase, and the non-linear current - voltage characteristics are found to be consistent with the collective flux-creep model. The effects of high-density random point defects on the vortex dynamics are also investigated via proton irradiation of the same single crystal. Neither resistive hysteresis at low currents nor peak effects at high currents are found after the irradiation. Furthermore, the current - voltage characteristics within the instrumental resolution become completely ohmic over a wide range of currents and temperatures, despite theoretical predictions of much larger -values for the given experimental variables. This finding suggests that the vortex-glass phase, a theoretically proposed low-temperature vortex state which is stabilized by point disorder and has a vanishing resistivity, may become unstable under applied currents significantly smaller than the theoretically predicted . More investigation appears necessary in order to resolve this puzzling issue.

The Hamiltonian for two-sublattice Heisenberg ferromagnets and ferrimagnets with different sublattice anisotropies, which is applicable for rare-earth - transition-metal (R - T) intermetallics, is established. In order to study spin-waves in easy plane or easy cone configuration, a transformation of spin-vector coordinates is performed by rotating the quantization axis frame by Eulerian angles and accordingly the Hamiltonian. Spin-wave spectra at low temperatures of the present system are determined by performing the standard Holstein - Primakoff transformation and a four-step diagonalizing procedure consisting of two coupled Cullen transformations, an extended Bogoliubov transformation, two independent Bogoliubov transformations and two independent Holstein - Primakoff transformations. The results for the ground states of the easy axis, the easy plane and the easy cone configurations are compared with those obtained by the mean-field theory. The border lines between the different spin structures are derived in either the pure classical limit or the large-exchange limit. Continuous transitions, accompanied with the continuous change of the angle between the averaged sublattice magnetizations, are found in both cases. It is found that splittings of the spin-wave spectra of the two-sublattice Heisenberg ferromagnets or ferrimagnets exist. A gap can appear in the spin-wave spectra, depending on the competition among the exchange and the anisotropies. Other physical properties, such as sublattice magnetization and specific heat, are discussed also.

The temperature dependence of infrared reflection spectra of betaine phosphate single crystal is reported for polarizations parallel to the a- and b-axes. The spectra were fitted with the factorized form of the dielectric function. A mode assignment is proposed on the basis of a comparison with the spectra of betaine phosphite and deuterated betaine phosphate. Some very asymmetric phonon lines are observed, and the importance of mode couplings is emphasized. A picture of a tendency towards transverse optical mode softening of the low-frequency spectrum upon approaching either side of the phase transition temperature emerges for the polarization parallel to the b-axis, consistent with the peak observed in the temperature dependence of the dielectric constant. On the other hand, the temperature dependence of the ionic effective charges does not display any appreciable discontinuity at the phase transition, contrary to what is observed at the temperature of the transition to the ferroelectric phase of the parent compound betaine phosphite. This is consistent with the very low level of the spontaneous polarization below in betaine phosphate. The similarities of the behaviours with those of the compounds of the KDP family are discussed.

We study the size dependence of the antiferroelectric-to-paraelectric phase transition in ultrafine particles of (with x-ray domain size down to ) prepared by the sol - gel method. The phase transition was investigated by means of dielectric and thermal measurements. The nature of the particle size dependence of various properties of antiferroelectric was found to be generally similar to that of typical `displacive' ferroelectrics such as and . In both cases, a size-induced structural distortion towards the high-symmetry paraelectric phase appears to dictate the behaviour of the nanoparticulate system. Nanoparticles of `order - disorder' systems such as behave in a markedly different manner. With decreasing particle size, we observe an increasing deviation of the dielectric response from the Curie - Weiss behaviour. Certain physical situations that may lead to the observed smearing out of the transition in nanoparticles of and are discussed.

One- and two-dimensional momentum distributions of positron annihilation radiation in Ge are calculated using the independent-particle model (IPM), the local density approximation (LDA) and the generalized gradient approximation (GGA). It is found that these three approximations are all capable of reproducing the main structures of the momentum distributions observed in the one- and two-dimensional angular correlation of positron annihilation radiation measurements. The positron annihilation rate is, however, found to be overestimated in the LDA in the low-momentum regime. While the GGA alone gives the correct total annihilation rate, it is found that the GGA is essentially just scaled with the LDA, and introduces no new momentum dependence to eliminate the overestimation of the LDA in the low-momentum regime.

The results of x-ray fluorescence measurements, performed with monochromatic-photon excitation at Pr absorption edges of , are presented and discussed in the framework of an Anderson impurity model. The dependence of the fluorescence structures on varying excitation energies is found to be in agreement with the localized, many-body description of the fluorescence process. Experimental spectra are reproduced by model calculations using values for the model parameters in the ground state of the system which are somewhat different from those obtained earlier in an analysis of core-level photoemission and absorption. Referring to this difference, the importance of resonant x-ray fluorescence data for the spectroscopic analysis of strongly correlated systems and for deriving a unique set of model parameters is stressed.

We describe an approach to the positron trapping at vacancies based on the optical theorem. We calculate the imaginary part of the optical potential in an ab initio model at the time when electron - hole excitation takes place. The obtained analytical formulae for the optical potential are tested as a function of the positron incident energy, the positron binding energy, and the electron density parameter . The calculation of the optical potential range agrees well with the vacancy radius. The approach is used for the description of observed temperature dependences of trapping rate at vacancies in pure metals as reported in the literature. The results are satisfactory and the obtained values of the optical potential are quite reasonable.

Errors were unfortunately introduced into this paper. On page 6737 equation (1) should read:

On page 6741 the line before equation (IA) and equation (IA) should read:

... limiting cases (i.e. ignoring the contribution from exchange) by Buckingham in the form