Table of contents

We examine a two-dimensional Fermi liquid with a Fermi surface which touches the Umklapp surface first at the four points as the electron density is increased. Umklapp processes at the four patches near lead the renormalization group equations to scale to strong coupling, resembling the behaviour of a two-leg ladder at half-filling. The incompressible character of the fixed point causes a breakdown of Landau theory at these patches. A further increase in density spreads the incompressible regions so that the open Fermi surface shrinks to four disconnected segments. This non-Landau state, in which parts of the Fermi surface are truncated to form an insulating spin liquid, has many features in common with phenomenological models recently proposed for the cuprate superconductors.

We explore the effect of the self-energy, , having a single pole, , with spectral weight and quasi-particle lifetime , on the density of states. We obtain the set of parameters , , and by means of the moment approach (exact sum rules) of Nolting. Due to our choice of self-energy, the system is not a Fermi liquid for any value of the interaction, a result which also holds in the moment approach of Nolting without lifetime effects. Our self-energy satisfies the Kramers - Kronig relationships since it is analytic in one of the complex half-planes. By increasing the value of the local interaction, , at half-filling , there is a transition from a paramagnetic metal to a paramagnetic insulator (a Mott metal - insulator transition) for values of of the order of (W is the bandwidth) which is in agreement with numerical results for finite lattices and for an infinite number of dimensions . These results expose the main weakness of the spherical approximation of Nolting: a finite gap for any finite value of the interaction, i.e., an insulator for any finite value of . Lifetime effects are absolutely indispensable to making our scheme work better than that based on improving the narrowing band factor, , beyond that obtained from the spherical approximation of Nolting.

Gas permeability of aerogels was studied as a function of the aerogel concentration in non-densified and densified aerogels. To represent numerically the aerogel structure we have used the off-lattice diffusion-limited cluster-cluster-aggregation model which is a good model for base-catalysed aerogels. The sintering process is modelled by using a `dressing' and `contracting' process which has been previously introduced in the literature. A random ballistic motion of a gas particle inside the pores of the aerogels represents the gas motion. We also obtain a theoretical expression of the gas permeability as a function of the aerogel density.

We discuss the viscosity damping effect on capillary waves in a binary-liquid system, using the linearized Navier-Stokes equation. The damping correction for the dispersion relation depends on the wave vector k as well as the interfacial tension. The calculated k-dependence of damping is characterized by a critical capillary-wave-number value , which separates the regions of weak and strong damping. The surface and interfacial roughnesses of a binary liquid system with large liquid depths are calculated and compared to experiments. Although the analysis has been restricted to the classical, macroscopic level, we obtain a noticeable modification from earlier hydrodynamic results for capillary wave damping and liquid-liquid interfacial roughness.

The density dependence of the velocity of sound in liquid mercury along the liquid-vapour coexistence curve is investigated by means of a large-scale molecular-dynamics simulation using the effective pair potential derived from the experimental structure factor by the inverse method. The resulting velocity of sound is in very good agreement with experiment and its density dependence changes at , at which point the metal-nonmetal transition occurs. It is shown that the repulsive part of the effective pair potential plays a crucially important role in the density dependence of the velocity of sound in liquid mercury.

Structural and thermodynamic properties of cubic boron nitride (c-BN) under pressure and for varying temperature are studied by molecular-dynamics (MD) simulation with the use of a well-tested Tersoff potential. Various physical quantities including the thermal expansion coefficient and heat capacity are predicted. Our simulation is extended to study liquid boron nitride at various densities.

The influence of local coordination ions on the optical properties of and in fluorochloride glass has been investigated by using steady-state and time-resolved laser excited site selective spectroscopy. From the thermal behaviour of lifetimes of the state, a dependence for the nonradiative Nd-Nd relaxation process has been found in the 10-90 K temperature range for concentration higher than 0.5 mol% which is in agreement with a two-site nonresonant process. In spite of the high content of fluorine in this glass, the presence of chlorine anions causes a significant effect on the spectral behaviour and relative quantum efficiency of emission as compared with those of pure fluoride glasses. The excitation wavelength dependence of the lifetimes of the state points to the existence of some site segregation for in this fluorochloride glass. This result is also confirmed by fluorescence line-narrowing experiments performed with in the same glass. The behaviour of both the line-narrowed fluorescence of the hypersensitive transition and the linewidth of the time-resolved resonant line-narrowed emission band as a function of the selective excitation wavelength indicates that homogeneous linewidths are dominated by relaxation time broadening caused by rapid phonon relaxation processes. These results together with the crystal field strength calculated as a function of excitation energy along the absorption band might be interpreted as indicating the existence of a crossover between the behaviour of fluorinelike and chlorinelike coordination anions. A discussion on the origin of this behaviour is given.

Extended x-ray absorption fine structure (EXAFS) spectra at the Cu K-edge and Ag K-edge were measured for copper(I) halide-based and silver halide-based glasses, respectively. Molecular dynamics (MD) simulation was also performed for the glasses with the same compositions. The bond distances, A-X, in the copper(I) halide-based and the AgI-based glasses, which were obtained from the EXAFS analyses, were almost the same as those in the CuX (X = Cl, Br, I) and AgI crystals. The coordination numbers of halide ions around the monovalent metal ions were nearly four as well as those in the crystals. These results were consistent with those of the MD simulation. On the other hand, the EXAFS analyses showed that the bond distances in the AgCl-based and the AgBr-based glasses were 0.1 Å shorter than those in the AgCl and AgBr crystals. These values are in the range of those of the halogenoargentate crystals having three- or fourfold coordination. However, the coordination numbers obtained from the EXAFS analyses were larger than four. The MD simulation suggested that the coordination shells in the AgCl-based and the AgBr-based glasses are highly disordered.

Two Fe-Nb-Cu-Si-B alloys, (B9) and (B6), prepared with the isotope, have been analysed using data obtained by means of in situ neutron diffraction. This technique allows one to scrutinize crystallographic phases during thermal treatments, avoiding problems due to sample handling. The B9 sample develops Fe(Si) nanometric crystals (10 nm) with 19 at.% Si in the phase when it is annealed at for one hour. An increase to favours the growth of Fe(Si) grains and the crystallization of other phases, mostly Fe borides. A Rietveld analysis of these phases results in a good reproduction of the nominal composition of the alloy. It also elucidates the crystallographic structure of the Fe(Si) phase. This is similar to the structure, but with some of the Fe atoms occupying some (45%) of the Si 4a sites. The compositions and amounts of the phases derived are in agreement with Mössbauer spectroscopy results for the same sample. Knowledge of the Fe(Si) composition enables one to compare the different magnetic behaviours observed for bulk and nanocrystalline alloys. By contrast, B6 alloy does not show the presence of a Fe(Si) structure, presumably due to the lower amount of Si in the Fe(Si) phase. The thermal expansion of the phases that appear is fairly linear and the corresponding thermal expansion coefficients for the different phases have been extracted. The magnetic structure of the Fe(Si) phase is ferromagnetic collinear, without any trace of antiferromagnetic ordering. The thermal variation of the (1, 1, 1) magnetic peak intensity of the Fe(Si) phase matches well with reported DC magnetization results.

The Peierls phase transition in the quasi-one-dimensional conductor is investigated by means of elastic and inelastic neutron scattering. The effective critical exponent , extracted from the temperature dependence of the integrated intensity from the CDW satellite reflections, is anomalously low, suggesting that the phase transition may be of first order. The intensity distribution among symmetry-related satellite reflections indicates a domain structure with slowly fluctuating domain populations. Correlation lengths associated with the diverging `central peak' are determined and are found to be nearly isotropic, at variance with results obtained on other quasi-one-dimensional compounds, such as platinum chains (KCP) or blue bronze, . Doping with 1.2% Nb has a severe effect on the modulated state. The low-temperature satellites are replaced by a diffuse scattering distribution elongated along . The absence of a phonon soft mode and the presence of a diverging central peak at the phase transition is interpreted within the framework of strong electron-phonon coupling. Finally, we propose a Ginzburg-Landau phenomenological model, where the interplay between the electronically coupled optical-like order parameter (Ta-atom tetramerization along the chain axis) and the elastic deformations lies at the origin of the phase transition in .

We present the results of a theoretical study of the structural phase transformations of ZnS under high pressure, using first-principles pseudopotential and full-potential linear muffin-tin orbital methods, in which the semicore Zn d electrons are treated as valence states. The zinc-blende, NaCl and cinnabar forms of ZnS have been considered. The structural properties and the band structures of these systems have also been studied. In the case of the FP-LMTO approach, an optimal choice of the empty spheres, atomic radii and filling percentage is introduced, which gives results in excellent agreement with those of the present pseudopotential method. It has been found that cinnabar phase is not a stable phase in ZnS under high pressure. The cinnabar phase is predicted to be a semiconductor with a direct band gap of about 3.6 eV.

The ab initio mixed-basis pseudopotential method based on the density-functional theory is applied to study the cohesion, ferromagnetism and electronic structure of iron and iron monohydride with cubic crystal structures. Spin-unpolarized and spin-polarized calculations are used to assess the transferability of norm-conserving ionic pseudopotentials for iron, and the level of accuracy obtainable for structural equations of state with reasonable effort. The influence of generalized gradient corrections on the cohesive properties is investigated. The results are compared directly to corresponding all-electron results obtained by using both FLAPW and LMTO-ASA methods.

The energetical ordering and magnetic states of hexagonal and double-hexagonal close-packed (hcp, dhcp) as well as face-centred cubic (fcc) Fe and FeH crystals are studied via spin-polarized ab initio total-energy calculations in the local spin-density approximation and with generalized gradient corrections by means of the mixed-basis pseudopotential and the all-electron LMTO-ASA methods. In all three structures, the magnetic spin moments go to zero under volume compression. For pure Fe in the compressed non-magnetic state, the hcp structure is found to have the lowest energy, fcc the highest, and the dhcp structure lies in between. The two hexagonal structures have significantly smaller than ideal ratios. For compressed non-magnetic FeH the energetical ordering of the structures is reversed, compared to that for pure Fe, with fcc ground-state structure and almost ideal ratios for both hexagonal structures. In the ferromagnetic states at expanded volumes, the energetical orderings are again opposite to those of the non-magnetic states both for Fe and FeH. In ferromagnetic FeH these energy differences are particularly small, yielding almost an energetical degeneracy of all three close-packed structures.

Adiabatic potentials, energy levels and wavefunctions for collective vibrational states of hydrogen isotopes in monohydrides of transition metals with face-centred cubic lattices, and NiH, and with body-centred cubic lattices, and CrH, are investigated by means of ab initio total-energy calculations in the local density and local spin-density approximations (LDA, LSDA). The study for the different transition-metal monohydrides, including PdH and NbH studied earlier, yields a general insight into the microscopic vibrational potential wells: the topology of their spatial shapes is specific to the metal lattice, but their depths and curvatures change quantitatively in a systematic manner through the transition-metal series. The calculated excitation energies agree very well with results of inelastic neutron scattering (INS) experiments both for non-magnetic NiH, treated within the LDA, and ferromagnetic , treated within the LSDA. The theoretical data for the two considered hydrides, with bcc structure, for which corresponding experimental data do not exist, provide an ab initio database for the construction of metal-hydrogen interaction models, e.g., for studies of self-trapped vibrational states of isolated H atoms in transition metals.

On the basis of 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 value of the energy-level fitting is and the group character of each Stark level compares well to the experimental assignment on the whole. The wavefunctions obtained were used in the study of the Zeeman interaction and the fluorescence spectra. The calculated g-factors obey Karayianis's partial g-sum rule, and the g-factor of the ground state of is predicted. According to the Judd-Ofelt theory, the odd crystal-field parameters have been quantitatively determined by the fit to the experimental transition rates from . The line-to-line transition rates and thus the fluorescence spectra from at room temperature have been calculated, which are in agreement with the experiment by and large.

Dielectric constant and ac and dc electrical conductivity measurements have been performed in single crystals of the fluorite compound , in the temperature range 100 to 770 K. The dielectric response shows both dipolar and charge carrier contributions. The electrical conductivity varied more than 11 orders of magnitude and showed the existence of two thermally activated processes, with activation energies of 1.71(1) eV and 1.17(1) eV, for the higher and lower temperature regions, respectively. The conduction mechanisms have been discussed in terms of the different conduction pathways for the fluorine vacancy motion in the crystal.

The efficient operation of polymer light-emitting diodes (LEDs) requires balanced injection and transport of electrons and holes. This has stimulated much research into suitable electron-injecting and transporting materials. We report the use of polypyridine as an efficient electron-transporting polymer. We have achieved much-improved LED performance by incorporating polypyridine as an electron-transporting layer in a poly(p-phenylene vinylene) (PPV) LED and optimizing layer thicknesses to balance transport of electrons and holes. The external quantum efficiency of these LEDs is 0.25%, 60 times greater than similar devices without the electron-transporting layer.

Fluctuations can break down mean-field critical behaviour. For the random-field Ising model, fluctuations caused by the randomness are important. The Ginzburg criteria based on two kinds of mean-field theory are discussed. One is the mean-field theory using the replica method and the other is the site-dependent mean-field theory. An argument that justifies the use of the site-dependent mean-field theory to obtain critical properties for the random-field Ising model is given.

Various types of mixed-spin two-dimensional Heisenberg network are investigated by means of Monte Carlo simulations. This study aims at interpreting quantitatively the thermodynamical properties of two-dimensional molecule-based magnets that have been recently synthesized. The proposed model requires that: (i) one of the two magnetic centres has a spin large enough to be treated as a classical spin; (ii) the zero-field Hamiltonian is isotropic; (iii) the quantum spins have only classical spins as neighbours. The quantum Hamiltonian is then replaced by a classical one with effective ferromagnetic interactions. The temperature dependence of both the specific heat and the magnetic susceptibility are calculated. The effects of the lattice geometry are analysed. We obtain for the specific heat a typical curve which is independent of these effects.

We try to interpret angle-resolved photoemission spectroscopy (ARPES) cross sections directly in band-structure terms by approximating the one-particle propagator needed to evaluate the lowest-order Keldysh diagram using the bulk Green function. As applications to and demonstrate, this is a sensible approach, since it reproduces important features of experimental curves. Matrix element effects strongly enhancing the contributions of individual bands at certain energies and depressing them at others are well described. Self-energy corrections included in an average way improve the agreement and allow for a rough estimate of the magnitude of the band broadening. In the case of our investigations demonstrate the significance of correlations, since adding self-interaction corrections to the local density approximation leads to considerable improvement of the calculated ARPES cross sections.

We consider thermodynamic properties, e.g. specific heat and magnetic susceptibility, of alternating Heisenberg spin chains. Due to a hidden Ising symmetry, these chains can be decomposed into a set of finite chain fragments. The problem of finding the thermodynamic quantities is effectively separated into two parts. First we deal with finite objects; secondly we can incorporate the fragments into a statistical ensemble. As functions of the coupling constants, the models exhibit special features in the thermodynamic quantities, e.g. the specific heat displays double peaks at low enough temperatures. These features stem from first-order quantum phase transitions at zero temperature, which have been investigated in the first part of this work.

The current-voltage characteristic (CVC) of quantum wires at low temperatures is considered theoretically. Three mechanisms of electron relaxation are taken into account: (i) elastic momentum relaxation by impurities or defects, (ii) inelastic energy relaxation by acoustic phonons and (iii) optical phonon emission. The latter causes a very fast energy loss for electrons reaching the energy of optical phonons. If the two first mechanisms are characterized by energy-independent relaxation times, the CVC has a sublinear shape with saturation. If the elastic relaxation time increases with electron energy, which is typical for impurity scattering, the CVC is characterized by a non-monotonic field dependence of differential resistance with a minimum at some critical field. Most analytical results are obtained for a wire with one occupied subband but the main features of a multi-subband case are also discussed.