Table of contents

We have studied the ground-state properties of the half-filled Hubbard model on the triangular lattice by means of the slave-boson technique. The combined effects of high correlation and frustration in this non-bipartite lattice produce a rich phase diagram, including paramagnetic, commensurate spiral, and linear spin-density-wave phases, as well as a metal-insulator transition. Unlike previous Hartree-Fock studies of this model, the slave-boson technique predicts no metallic incommensurate spiral state to be the stable ground state in any region of parameter space.

The two-dimensional doping by charge transfer and/or diffusion of metal atoms into C₆₀ has been studied by in situ electronic transport measurements during the deposition of metal-C₆₀ ultrathin bilayers. The results show that the transport properties of these interfacial systems are significantly altered by such doping processes. Some useful information about the charge transfer from metal to C₆₀ and the electronic transport properties of the metal-doped-monolayer C₆₀ can be obtained after careful analysis.

It is believed that the mean distance between water molecules increases with a decrease in water temperature below room temperature due to the increase in 'hydrogen bonds', which implies an increase in the local fourfold coordination. In order to investigate how far this effect extends over a wider range, we have made constant-density measurements of the principal peak of the neutron structure factor of heavy water as a function of temperature. Data were taken at densities from 10% greater to 8% less than the maximum density at 1 atm, and for temperatures extending from -20 to 220 degrees C. The effects seen at ambient conditions were found to extend over this whole range.

On the basis of the analytical solution of the Ornstein-Zernike equation in our previous paper, a model is presented for the calculation of the fluid density profiles near curved walls. The result of the model is compared with the computer-simulation data, and the agreement is reasonably good. An effect of wall curvature and the cavity free energy are discussed.

We report an inelastic neutron scattering (INS) experiment on liquid ⁷Li at 470 K, 526 K and 574 K using thermal neutrons. Fully corrected structure factors S( kappa , omega ) were obtained. We have assessed the quality of the data by means of the first frequency moment of S( kappa , omega ) and by the detailed balance condition. The coherent dynamic structure factor S_c( kappa , omega ) and the incoherent dynamic structure factor S_s( kappa , omega ) were separated by fitting models. We compare the data for S_s( kappa , omega ) at small kappa with several models using no adjustable parameters. The Lovesey model for S_s( kappa , omega ) gives very good agreement with the experimental data at all three temperatures. S_s( kappa , omega ) is not well described by the predictions from simple diffusion, mode-coupling theory or the model of Nelkin-Ghatak. The Lovesey model also gives the best agreement with the data when we fit one parameter.

Structural information on the pair distribution function g₂(r) for single-component disordered systems is usually obtained from the experimental structure factor S(k) measured by diffraction techniques. Complementary short-range information can be provided by the analysis of the extended X-ray absorption fine structure chi (k) associated with a certain X-ray absorption edge. The intrinsic differences in the nature of the chi (k) and S(k) signals are discussed and particular effort is devoted to connecting the chi (k) signal with usual quantities familiar to the distribution function theory in disordered matter. An example of the short-range chi (k) sensitivity is presented showing signals associated with g₂(r) functions of liquid Cu at 1150 degrees C and 1300 degrees C. The necessity to fit realistic g₂(r) models to EXAFS spectra satisfying both long-distance behaviour and the compressibility sum rule is emphasized. A method to combine these constraints and previous information on g₂(r) with available chi (k) data is proposed and applied to recent EXAFS data on liquid palladium.

X-ray absorption spectra at the Au L edges of Au(PPh₃)Cl, Au(PPh₃)₂Cl and KAuCl₄.2H₂O have been collected as a function of temperature. Data analysis of the extended X-ray absorption fine structure was carried out with the GNXAS package performing an appropriate treatment of the multiple-scattering effects involving several ligand atoms. The accuracy of the current analysis procedures for the Au L edges is assessed and the sensitivity to coordination numbers and distances is established. Evidence for the presence of a (2p4f) double-electron excitation in the atomic Au background is obtained for the first time. The results presented in this paper are of general importance for improving data analysis at the Au L edges and in particular for improving the EXAFS results in Au cluster compounds.

Quasielastic neutron scattering experiments have been carried out to study the thermally activated rotational process of the NH₄⁺ ion in the perovskite family NH₄MF₃ (M=Mg, Zn, Mn and Cd). The activation energies of this process in the high- and low-temperature structural phases have been obtained, and range from 44.8 meV for NH₄MgF₃ to 128 meV for NH₄CdF₃ in the high-temperature phase, and from 75 meV for NH₄ZnF₃ to 246 meV for NH₄CdF₃ in the low-temperature phase. Geometric models for the orientations of the NH₄⁺ ion occupied during the rotational process are proposed in both phases. In the high-temperature phase, the dominant rotations are around twofold axes. A preferential axis, determined by the orientations of the ion in the crystal, shows the greatest jump rate, in the low-temperature phase the rotational disorder persists through two energetically non-equivalent orientations.

Elastic and inelastic neutron scattering was used to probe the structural and magnetic properties of polycrystalline CuGeO₃. The diffraction data show that the chemical lattice remains stable in the measured temperature range 1.5 K<or=T<or=296 K. At low temperature the inelastic experiments revealed a well defined magnetic excitation at 15 meV, which persists up to T=50 K. The analysis of the spin-wave spectrum is based on a one-dimensional Heisenberg model within the spin-wave theory. On the other hand, no evidence of energy-gap opening, which would result from a dimerization of the Cu chains, could be found at low temperature.

Using a pulsed-laser-beam and electron beam coevaporation method, we have embedded CdTe microcrystallites into a CaF₂ thin film in a high-vacuum chamber. Nanometre-scale CdTe particles embedded in the CaF₂ thin film are observed in transmission electron micrographs. The blue shifts of the absorption bands and in particular the corresponding photoluminescence bands are easily obtained in the CdTe-microcrystallite-doped CaF₂ thin film. A broadened optical phonon Raman structure and low-wavenumber acoustic phonon structure are also obtained. These results are considered to be a size effect of the CdTe microcrystallites.

We present self-consistent LMTO-ASA cluster calculations of the solution energy Q and the nearest- and next-nearest-neighbour configurational interactions, V₁ and V₂, for the dilute BCC Ti_1-xV_x and FCC Ni_1-xAl_x alloys. The lattice relaxation effects (LRE) are described with a proper choice of atomic volumes for the impurity and its neighbours, which are found employing the conventional phenomenological estimates for atomic displacements near an impurity. For the Ti-V system LRE are found to be important, while the charge transfer effects (CTE) are small. The results of our calculations for this system seem to indicate that the phenomenological estimates overestimate LRE contributions to Q and V_i by a factor of about two. We also present phenomenological estimates of LRE contributions to Q for impurities in 24 cubic host metals. For the Ni-Al system LRE are found to be small while CTE are significant, and taking them into account in the cluster calculations of Q and V_i seems to noticeably improve their accuracy as compared with single-site approaches.

A method has been developed to calculate the texture contribution to the anisotropy of the physical properties of polycrystalline materials. It starts from an approach proposed earlier for modelling the rolling textures in FCC and BCC metals by a superposition of partial fibre components. The averaged values of the elastic constants and Young's modulus have been calculated for copper sheets with different textures. The method computes the component-by-component physical property anisotropy and thus predicts the required component ratios for any required anisotropy. In our approach the texture is described by the minimum number of components, with the number of lasts being rigorously equal to the number of structure components.

The magnetic phase boundary of the commensurate-incommensurate spin density wave (SDW) transition in CrFe alloys was determined more precisely using neutron diffraction of single crystals. The re-entrant incommensurate SDW phase recently reported in CrFeMn ternary alloys was not observed in CrFe binary alloys.

The specific heat of the blue bronze Tl_0.3MoO₃ is measured using an adiabatic continuous heating method from 100 to 220 K. A specific heat anomaly is observed at 172.3 K. The specific heat jump, and enthalpy and entropy changes associated with the CDW phase transition are estimated. The results suggest that the phase transition is of second order and that the lattice plays an important role in thermodynamics in this system, by analogy with the behaviour of its isostructural compound K_0.3MoO₃. The mean-field theory seems not to be adequate for describing the CDW phase transition in this compound.

Thin W films were deposited by RF sputtering onto a glass substrate and then covered with an Se layer. The WSe₂ layers were obtained either by annealing the W layers in a vacuum sealed Pyrex tube or by the tarnishing reaction of Se vapour on W films. The properties of the thin WSe₂ films made by synthesis and by tarnishing are described. The layers were examined by X-ray diffraction, X-ray photoelectron spectroscopy, scanning and transmission electron microscopy, electron microprobe analysis, optical transmission and resistivity measurements. The films crystallize in the hexagonal structure. It has been found that stoichiometric layers are obtained after synthesis while layers obtained by tarnishing are partly oxidized. The optical gap of the synthesized layers is in good agreement with the expected value. The electrical resistance is governed by grain boundary scattering mechanisms.

First-principles studies are carried out on the electronic structure and Compton profile for the low-symmetry system of calcium dihydride using the discrete variational local-density-functional method with embedded cluster model. The results demonstrate that, with the ionic bonding picture, experimental data from photoelectron spectroscopy with both x-rays and synchrotron radiation and from Compton profile measurements can be reasonably understood. Based on the bonding character revealed in the electronic structure, we suggest that CaH₂ doped with a monovalent element could be a superconductor.

We develop a theory of the core-valence-valence Auger spectra from 3d transition metals including core-hole screening effects. First, we analyse the limiting case when the d electrons fully screen the core hole in the initial state. We employ the variational method including singly excited states to construct the ground state with a completely screened core hole and solve the resulting many-body problem for correlated valence electrons using the self-consistent version of the perturbation theory in powers of U/w (where U is Hubbard on-site interaction and w, is the bandwidth) within the T-matrix approximation. The resulting lineshape has a contribution proportional to the one-particle density of valence states and, in addition, also the two-hole-one-electron contributions, but not the two-hole term that had been suggested in the literature. The latter arises only for filled bands. The comparison with experimental data is qualitatively good for early 3d metals, giving a natural explanation of the so-called 'negative U' behaviour, but deteriorates with increasing atomic number. Next, we present a simple approximation to describe the effects of incomplete relaxation in the initial state. Contributions from the unrelaxed initial state are proportional to a two-hole density of states convolved with a suitable core lineshape. This removes the difficulties with the late transition metals. Our results indicate that the core-hole screening becomes progressively less complete towards the end of the 3d series. The theory then reduces to the standard one, based on two-hole propagators in the low-density approximation.

The non-linear transport properties of sliding charge-density waves (CDWS) in pure Tl_0.3MoO₃ and impurity-doped (Tl_0.2K_0.8)_0.3MoO₃ blue bronzes are measured in the temperature range 80-150 K. It is found that the analytical form of non-linear current in a moderate field range fits the power-law relation, i.e. I_CDW varies as (V/V_r-1)^alpha , where the exponent alpha is temperature dependent and not a constant 1.5. Such a relation seems to be universal for blue bronzes over the temperature range below the Peierls transition temperature and does not depend on the quality of samples. The results are in agreement with those observed by Mihaly et al for pure K_0.3MoO₃ blue bronze in the low-temperature range and suggest that the dominant damping of the collective mode arises from the dissipative normal current induced by dynamical deformation of the CDWS, not only in the low-temperature range, but also in the high-temperature range.

Carbon substitution in bulk silicon has been investigated using the super-cell approach, in conjunction with the periodic ab initio Hartree-Fock method. The convergence of the defect formation energy and of the relaxed defect geometry as a function of the super-cell size is discussed with reference to super-cells containing 8, 16, 32 and 64 atoms. It turns out that the convergence of the unrelaxed defect formation energy is rapid, in spite of the large local charge redistribution around the defect (the net charge on carbon is 1.2 mod e mod ); the relaxation effects are very large (about 2.0 eV) and involve mainly the first and second neighbours; however, the relaxation of the fifth neighbours of the defect (which is possible only with the biggest super-cell considered) lowers the energy by a further 0.06 eV. The defect formation energy and the atomic displacements obtained with the 32 and 64 atoms super-cells are similar, whereas the energy difference between the 16 and 32 atoms cells is as large as 0.4 eV.

We report the electrical resistivity rho , thermopower S and thermal conductivity lambda of Ce_1-xLa_xAu₂Si₂, Ce_1-yY_yAu₂Si₂, Ce_1-xLa_xAg₂Si₂ and Ce_1-yY_yAg₂Si₂ (0<or=x, y<or=1) systems. The rho curves show a sudden change in slope at the antiferromagnetic ordering temperature T_N^rho . In the CeAu₂Si₂-based alloys, the T_N depression, due to the dilution of Ce atoms by La or Y, is accounted for solely by the weakening of the inter-site interaction temperature scale T_RKKY. On the other hand, in the CeAg₂Si₂-based alloys, apart from T_RKKY, the single-ion Kondo temperature T_K also plays an important role in determining the nature of the T_N variation. The rho and lambda behaviours, at both low and high temperatures, show clear evidence for the Kondo effect in the presence of crystal fields.

Composites of silica glass and metal nanoparticles of copper or nickel have been synthesized by the hot compaction of sol-gel-derived glass-metal powders. The metal particle sizes range from 5.1 to 7.2 nm. The electrical resistivity of the composites is characterized by semiconductor-like behaviour in the temperature range 235-340 K. This is believed to arise owing to electron hopping between localized states within the band gap of amorphous silica formed by the dispersed metal atoms. A low activation energy of the order of a few millielectronvolts in the specimens containing copper in the temperature range 140-235 K may occur because of a quantum size effect, the metal particles forming continuous chains with diameters of the order of 3 nm.

A measure of the Mn 3d partial density of states (DOS) of NiAs-type MnTe has been obtained from resonant photoemission experiments in the Mn 3p-3d core excitation region. On the basis of an analysis of the Mn 3d partial DOS spectrum in terms of a configuration interaction on a Mn²⁺ (Te^2-)₆ model cluster, we find that the spectral intensities in the top 6 eV are predominantly due to transitions to the d⁵L, final-state configuration, where L represents a ligand hole. On the other hand, those at 6-14 eV originate mainly from transitions to the d⁴ configuration.

Optical absorption spectra of Ge²⁺ ions in KCl and KBr crystals have been investigated in the temperature range 15-300 K. In the triplet-structured spin-orbit-allowed C band, the C₁-C₂ band splitting was observed to be larger than that of the C₂-C₃ band at low temperatures, while the two become equal to each other with increasing temperature. The absorption lineshape has been calculated, by means of the Monte Carlo integration method, for Ge²⁺ ions with a linear electron-lattice interaction and tetragonal crystal field. Comparing the calculated lineshape with the observed spectra, we conclude that the Ge²⁺ ion is at the off-centre position with the tetragonal field at low temperatures, but is moved into the on-centre position with increasing temperature.

The optical absorption spectrum of NiO is measured at low temperatures and energies between 7000 and 30000 cm^-1. An assignment to cubic crystal field components of the 3d⁸ configuration of Ni²⁺ is proposed. The fine structure of the A band, showing two sharp lines at the lowest energies, 7811 and 7852 cm^-1, is explained by additional spin-orbit interaction, rhombohedral exchange striction and an orthorhombic transverse molecular field due to antiferromagnetic long-range order.

Magic-angle-spinning nuclear magnetic resonance (NMR) spectroscopy of BaFBr and BaFCl doped with up to 6 mol.% oxygen (previously enriched to 40% in ¹⁷O) has detected sites on which oxygen impurities were incorporated. In BaFBr:O, monatomic O^2- is found preferentially on fluoride sites, and subsidiarily on bromide sites, but in BaFCl:O, substitutional O^2- falls below the threshold of detection. In both systems the diamagnetic molecular anion O₂^2- is detected, on one site in BaFBr:O, on two in BaFCl:O; one of the latter is assigned to the chloride site, while the other is attributed to interstitial O₂^2-. Although it was intended to introduce oxygen only through BaO, ¹H NMR indicates OH^- on several sites. The NMR spectra, including those of ¹⁹F, ⁸¹Br and ³⁵Cl, are compared with results obtained previously by electron paramagnetic resonance and magneto-optical techniques on ¹⁷O-containing, X-irradiated samples.

The optically active process of higher-order bands (HOBs) in fast-neutron-irradiated float-zone silicon has been investigated at low temperature using a Fourier-transform infrared spectrometer with optical excitation. It is found that the photoexcitation process follows an exponential time dependence and that the decay is logarithmic with a decay time constant of 10⁵ s at 7 K. The saturation value of the absorption coefficients depends on the logarithm of the illumination-light intensity. These characteristics are associated with the slow relaxation of photoexcited carriers originating mainly from defect clusters. The relaxation behaviour of photoexcited carriers has been analysed using the macroscopic-barrier model, which is in good agreement with the observations.

The complex dielectric constant ( epsilon ), the absorption coefficient (A), the reflectivity (R), and the optical joint density of states (JDS) of the misfit-layer compounds (MS)_1+xTS₂ (M=Sn,Pb,Bi,Ce,Sm; T=Ti,Nb) have been derived from reflection electron-energy-loss spectra via Kramers-Kronig analysis. Loss structures arising from interband transitions and a partial plasma resonance have been studied systematically. It is considered that most of the main structures result predominantly from a TS₂ layer. A main contribution from an MS layer causes a loosely varied background as a whole, leading to a decrease in depth of a valley at the optical gap of TS₂. It also gives a high optical JDS at 3-4 eV for the Sn and Pb compounds and at 6-7 eV for the rare-earth compounds. A reasonable interpretation of most experimental results obtained has been made within the rigid-band model in terms of the electronic structure of each layer, although a small deviation from the model must be considered.

Defects in 3 MeV electron irradiated vitreous SiO₂ (v-SiO₂) were probed by the positron annihilation technique. For unirradiated v-SiO₂ specimens, almost all positrons were found to annihilate from positronium (Ps) states. The high formation probability of Ps was attributed to the trapping of positrons by open-space defects. The formation probability of Ps was decreased by the electron irradiation. The observed inhibition of the Ps formation was attributed to the trapping of positrons by point defects introduced and/or activated by the irradiation. From measurements of the lifetime distribution of Ps, it was found that, by the electron irradiation, the mean size of open-space defects was decreased and the size distribution of such defects was broadened.