Table of contents

In situ energy dispersive x-ray measurements were performed on C₆₀ fullerene at pressures of 2.6, 5.7 and 12 GPa and temperatures between 300 and 1100 K. The polymerization process was followed in detail and the dynamics of the phase changes measured. This has enabled us to map the P-T phase diagram and look at the dynamics of the phase transformations.

Ab initio calculations based on the density-functional pseudopotential approach have been used to study the electronic structure and chemical bonding in layered machinable Ti₃SiC₂ ceramic. The calculations reveal that all three types of bonding - metallic, covalent and ionic - contribute to the bonding in Ti₃SiC₂. The high electric conductivity is attributed to the metallic bonding parallel to the basal plane and the high modulus and high melting point are attributed to the strong Ti-C-Ti-C-Ti covalent bond chains in the structure.

Orbital-resolved O Kα x-ray fluorescence spectra of in-CuO₂-plane [O(1)] and apical [O(2)] oxygen atoms were obtained for La_1.9Sr_0.1CuO₄ by making an appropriate choice of the energy of incident photons and the crystal orientation relative to the polarization vector of the incoming radiation. The site origin of 2p_x,y and 2p_z components in the Sr-doping-induced absorption prepeak near the O 1s edge, associated with states developed in the energy gap, was studied by O Kα monitoring at the excitation energy of the prepeak. The spectral analysis, based on the O Kα shape difference between inequivalent O sites, supports the assignment of the doping-induced 2p_x,y holes to the O(1) atoms but suggests that the 2p_z holes may have both O(2) and O(1) character in contrast to assignments in prior studies.

We compare the theoretical anisotropic upper critical field H_C(Θ,T) of a quasi-two-dimensional d-wave superconductor with recent H_c2-data for the layered organic superconductor κ-(BEDT-TTF)₂Cu(NCS)₂ (BEDT-TTF≡bis(ethylene-dithio)tetrathiafulvalene). We find agreement as regards both the angular and the temperature dependence of H_C. This supports the suggestion that the Fulde-Ferrell-Larkin-Ovchinnikov (FFLO) state exists in this material for exactly plane-parallel orientation of the magnetic field. Indications of precursor states, occurring for small deviations from the plane-parallel field direction, are also pointed out and further measurements for confirming the existence of the FFLO state are proposed.

The temperature dependence of peak widths in high-resolution angle-resolved photoelectron spectroscopy from quantum well states in ultra-thin Ag films on V(100) has been used to determine the electron-phonon coupling constant, λ, for films of thickness 1-8 layers. A strong oscillatory variation in coupling strength is observed as a function of film thickness, peaking at a two layer film for which λ≃1.0. A simple theory incorporating interaction of the photo-hole with the thermal vibrations of the potential step at the adlayer-vacuum interface is shown to reproduce the main features of these results.

We report our results about the fluctuation effects on the surface static structure factor and on the local Bose-Einstein condensate (BEC) in the low density surface region of ⁴He. This information has been obtained by a variational Monte Carlo calculation for a slab of ⁴He modelled by a novel shadow wave function with a glue term (glue-SWF). This glue-SWF describes the self-binding of ⁴He only via local interparticle correlations. BEC increases from the bulk-like value well inside the slab to a much larger value in the surface region (up to about 50%). Further out of the surface the condensate decreases and correspondingly there is an enhanced population of small momentum states. This is very different from the full BEC that was predicted using a simplified variational treatment of these ⁴He systems. This different behaviour is correlated with the presence in the glue-SWF of enhanced density fluctuations in the surface region as shown by the behaviour of the static structure factor at small wave vector. This is interpreted as an effect due to the zero point motion of ripplons.

Specular neutron reflection has been used to investigate the composition and structure of the surfactant-polymer mixture of sodium dodecyl sulphate (SDS) and hexaethylene glycol monododecyl ether (C₁₂E₆), and the cationic polymer poly-(dimethyldialyl ammonium chloride) (dmdaac), at the air-water interface. In particular, the effects of surfactant and polymer concentration are investigated. The addition of the cationic polymer results in an interface more rich in SDS. The variation of surfactant composition with polymer and surfactant concentration shows a complex behaviour, which is related indirectly to the bulk solution phase behaviour. The amount of polymer at the interface increases with increasing polymer concentration, but decreases with increasing surfactant concentration at a fixed solution polymer concentration. At a polymer concentration ~100 ppm there is a marked change in the structure of the adsorbed polymer layer: the thickness of the absorbed layer increases from ~20 Å to ~30 Å. In contrast the thickness of the absorbed layer, when only the surfactant is visible at the interface, shows no significant change with increasing polymer concentration.

In a charge-stabilized colloidal solution, the large colloidal particles are surrounded by microions that are up to four orders of magnitude smaller than the colloidal particles. Because of this size asymmetry, it is desirable to obtain an effective one-component description of the mixture where the colloidal particle plus its ionic atmosphere is treated as one, dressed particle. The effective pair potential between these dressed particles is a screened Coulomb potential. The screening depends, of course, on the density distribution of the small ions around and between the big colloidal particles. If the colloidal charge and the concentration of the ions is not too high, this distribution can be approximately determined from the linearized Poisson-Boltzmann equation, and the resulting effective pair potentials are Yukawa potentials. In concentrated suspensions, however, the full, non-linear Poisson-Boltzmann equation must be solved to determine the density distribution of the small ions. In this article, we suggest a way to obtain effective pair potentials for this case. We solve the non-linear Poisson-Boltzmann equation around a colloidal particle that is displaced a certain distance from the centre of its Wigner-Seitz cell. From the resulting density profile of the ions, we determine the total force acting on the shifted particle as a function of the displacement. From this function one can then estimate the non-linearly screened pair forces, and, thus, the effective pair potentials.

First-principles molecular dynamics simulations of liquid tellurium at 1123 K and at continuously varied densities are presented. The question of why recent studies within density functional theory overestimated the number of nearest neighbours is analysed in terms of structure factors and contributions of different neighbour shells to the pair distribution functions where the local density approximation and the generalized gradient approximation were used for the exchange-correlation functional. A recently proposed liquid-liquid phase transition under high pressure is also addressed.

The generalized collective mode approach is applied for the study of transverse dynamics in binary mixtures. The scheme is based on simultaneous treatment of the conserved total mass current and the mutual mass-concentration currents, as well as their time derivatives. The condition for existence of optic-like transverse modes in a binary system is derived. It is shown that high mutual diffusion and a tendency towards demixing prevent the emergence of transverse optic-like modes. Optic-like excitations are found in a Lennard-Jones Kr-Ar fluid and the liquid-metallic alloys Mg₇₀Zn₃₀ and Li₄Pb, while in the `fast-sound' dense gas mixture He₇₅Ar₂₅ they do not appear for small wavenumbers.

EXAFS, conductivity and Hall coefficient measurements have been carried out for liquid As-Te mixtures. EXAFS analysis reveals the presence of chemical disorder marked by the existence of homopolar As-As and Te-Te pairs. At high temperature around 500 °C the network structure composed of threefold coordinated As atoms and twofold coordinated Te atoms is transformed into the twofold chain structure. The network-chain transformation is accompanied by the gradual semiconductor to metal transition which is demonstrated by the results for conductivity and Hall coefficient. The microscopic origin of the semiconductor to metal transition is discussed in connection with the structural modification.

Ab initio molecular dynamics simulations are carried out for liquid Li_0.8Pb_0.2 and Li_0.5Pb_0.5 alloys to investigate the ionic structure and the electronic states. In our simulation, the existence of the `chemical complex' Li₄Pb is not found; rather, a salt-like ordering of Pb ions is seen in the liquid Li_0.8Pb_0.2 alloy. It is found from the calculated partial and total structure factors that this ordering leads to the characteristic behaviour of the total structure factor, which agrees well with the results of a neutron diffraction experiment. The composition dependence of the electronic states is explained on the basis of the ionic configuration. The tendency towards ionicity or charge transfer is seen in both liquid alloys, though the valence-electronic charge distribution is not so localized around the ions.

We present a model for the solidification process of two immiscible fluids interacting repulsively with mobile impurities on a two-dimensional square lattice. In the space of the fluids and impurity concentrations, the phase diagram exhibits a critical curve separating a percolating from a non-percolating phase. Estimated values for the fractal dimension and the exponent β of the order parameter reveal that the critical exponents do not vary along this curve, i.e., they are independent of the impurity concentration. The universality class is that of the ordinary percolation. On the basis of the ideas of the dynamic epidemic and invasion percolation models, we also propose a model that may be relevant to cleaning porous media by fluid injection. An analysis of the acceptance profile, the fractal dimension and the gap exponent strongly indicate that this model belongs to the universality class of the ordinary invasion percolation.

Optical second harmonic generation (SHG) as a particularly surface sensitive technique was employed for the first time to investigate the surface phase behaviour of a liquid alloy, Bi_xGa_1-x, in different heating and cooling cycles up to 280 °C. The aim of these experiments is to establish a relatively simple experimental access to the surface phase diagrams of liquid alloys. Measurements of the characteristic changes of the SH signal have been performed on pure Bi and different Ga-rich alloys (x_Bi⩽0.367). In pure bismuth the melting and freezing of the surface is indicated by a distinct polarization dependent variation of the SH intensities. Of particular interest is the characterization of the wetting transition found in Ga-Bi recently at the monotectic phase transition (x_m = 0.085, T_m = 222 °C). Surprisingly, on first heating of the Ga-rich alloys up to 280 °C the SHG signals give no indication of the dramatic compositional change at the surface induced by the wetting transition. From these observations we conclude that the main source for the nonlinear polarization is the outermost layer of the alloy which in the wet and the non-wet state consists of an adsorbed Bi-rich monolayer. It is shown that SHG is very sensitive to structural changes at the surface. Most interestingly, on cooling of the Ga-rich alloys a Bi-rich film crystallizes on top of the bulk liquid alloy.

The pretransitional behaviour of static dielectric permittivity in the isotropic phase of nematogens and in the homogeneous phase of a weakly off-critical binary solution is compared. The pretransitional anomalies are described by analogous relations, with critical exponents ϕ = 1-α≈0.50 and ϕ = 1-α≈0.89, respectively. Tests were conducted for pressure paths approaching the virtual critical point. The specific-heat-like dependences of derivatives of dielectric permittivities in both systems have been shown. This result agrees with the dependence suggested for critical mixtures by Mistura (1973 J. Chem. Phys.59 4563). Results obtained point to the validity of the recently formulated hypothesis on the fluid-like critical behaviour of the isotropic-nematic transition (Mukherjee P K et al 1995 Phys. Rev.  E 51 4570, Mukherjee P K 1998 J. Phys.: Condens. Matter10 9191).

We have studied the self-assembly of amphiphiles over wide ranges of temperature and concentration using Monte Carlo simulation of a lattice model introduced by Care. A new feature that we have observed is that for all the concentrations studied, the specific heat exhibits a peak as a function of temperature. The physical origin of this peak can be traced back to energy fluctuations occurring at the onset of micelle formations and can be used to characterize the critical micelle concentration (CMC). Our observations (i) that the temperature dependence of the specific heat fits rather well with the Schottky-type specific heat of a degenerate two-level system and (ii) that there is the simultaneous appearance of a knee in the cluster distribution function (CDF) explain the underlying physics of identifying the CMC with the appearance of this knee in the CDF, as proposed by Nagarajan and Ruckenstein. We show that the various moments of the cluster distribution functions can be expressed solely in terms of the second moment as predicted by a mean-field theory of Blankschtein, Thurston, and Benedek. We also calculate the fluctuation in the amphiphilic aggregate sizes and relate our simulation results to those from a simple but exactly soluble model proposed by Israelachvili.

The structural and electronic properties of liquid As₂Se₃ mixtures are studied by ab initio molecular dynamics simulations. We detail the temperature dependence of the atomic structure by calculating the correlation functions in a real space as well as in a reciprocal space. It is found that the network structure changes to a chain-like structure by bond breaking with increasing temperature, and that there are triangular atomic configurations in the chain-like structure at higher temperatures. We also investigate the relation between the semiconductor-metal transition and the structural change.

A three-body potential coupled with a molecular dynamics method have been used to simulate structural properties of CuI in the zincblende and tetragonal phases. It is found that the diffusion constant is well reproduced for the α-phase of CuI using this model rather than a two-body potential. This study predicts also the presence of cation disorder at elevated temperature within the tetragonal phase of CuI.

We studied, by neutron spectroscopy, local vibrations of hydrogen interstitials and hydrogen trapping by nitrogen interstitials in TaN_0.006H_0.003 and in a nitrogen-free TaH_0.086 reference sample between 1.4 and 295 K. The measurements on TaH_0.086 yielded vibrational energies E₁ = (113±1) meV and (doubly degenerate) E_2,3≃163 meV at 295 K where hydrogen is in solid solution, together with a higher energy E₁ = (120±1) meV and essentially unchanged energies E_2,3 at low temperatures (1.4 to 4.2 K) where hydrogen precipitates in a hydride phase. The energies agree with previous studies. The measurements on TaN_0.006H_0.003 yielded, at 295 K and at low temperatures, vibrational energies identical to those of the nitrogen-free sample at 295 K. The results show that the hydrogen interstitials are trapped by the nitrogen at low temperatures and do not precipitate. They show further that the trapped hydrogen interstitials occupy tetrahedral sites, i.e. the same type of site as occupied in the absence of traps.

A model N-electron system - a finite jellium - is considered first. The virial equation (VE) for it is obtained by adapting the result of Holas and March (Holas A and March N H 1999 Phys. Rev. A 60 2853) concerning a molecule in a homogeneous magnetic field B. Next, by applying a limiting procedure with N tending to infinity, the VE for an infinite-jellium system is established. This result extends the well-known zero-field VE by adding a term involving a derivative over B. Similarly, the VE for a periodic solid is obtained by applying a limiting procedure to the VE for a cluster (`finite crystal'). All VEs are valid for the systems in arbitrary (ground or excited) eigenstates.

Single crystals of insulating Bi₂Sr₂RCu₂O₈ (R = Dy, Er, Y and Ca_0.5Er_0.5) and superconducting Bi₂Sr₂Ca_1-xPr_xCu₂O₈ (x = 0, 0.10, 0.25, 0.36 and 0.42) were prepared, and the thermopower S(T) along the CuO₂ plane was measured from 4.2 to 420 K. All of the thermopowers including those for the insulators are roughly expressed as A-BT above a certain temperature T_scale, where A and B are positive constants. S(T)/S(T_scale) for all of the samples falls onto a single curve as a function of T/T_scale. Since T_scale is close to the pseudogap temperature for the superconductors, this scaling offers a rough estimate of the pseudogap temperature of high-T_c superconductors. The present study further suggests that the parent insulators have a pseudogap above room temperature.

We report on the magnetic effects induced by the substitution of Mn²⁺ ions in the quasi-one-dimensional Heisenberg-like antiferromagnet compound (CH₃)₂NH₃MnCl₃.2H₂O by non-magnetic Cd²⁺ and magnetic Cu²⁺ ions. In the undiluted compound three-dimensional ordering of the linear chains occurs at T_N = 6.36 K. In the diluted samples we observe the appearance of a ferromagnetic magnetization below T_N when the samples are cooled in small axial magnetic fields applied along the axis of antiferromagnetic alignment. At the lowest temperatures this magnetization is five times larger in the Cd diluted sample than in the Cu diluted one. Above T_N the susceptibility of both diluted samples shows an excess of magnetization with respect to that of the pure system. It is suggested that the uncompensated magnetic moments at the end of the `magnetically broken' chains are responsible for this behaviour. It is also suggested that some of these `free' paramagnetic ends can couple ferromagnetically at T_N, giving rise to the observed ferromagnetic component. The temperature dependence of the ferromagnetic magnetization of both samples is identical, indicating a common mechanism. The shape of this T-dependence suggests that the ferromagnetic correlations also have a low dimensional character.

The magnetic and transport properties of hard-soft giant-magnetoresistive-effect spin valves, with a Co/Cu/Co antiferromagnetically coupled sandwich as the hard layer, are investigated. The irreversibilities in the antiferromagnetically coupled Co/Cu/Co sandwiches prepared by DC-magnetron (DC = direct current) sputtering are shown to strongly depend on the magnetic history of the samples. The irreversibilities, evidenced from analysis of the giant-magnetoresistance and magnetization responses along minor loops, are attributed to domain-phase transformations. First, the effect of the buffer stack on the structural properties and on the quality of the antiferromagnetic coupling is investigated. Coupling strengths larger than 0.4 erg cm^-2 have been achieved for deposition on buffers containing a Cr/Fe bilayer. A method is given for estimating the amount of remanence in the Co/Cu/Co sandwich, based on the analysis of the giant-magnetoresistive response upon switching the magnetization direction of the soft magnetic layer located in the buffer. The perfect antiferromagnetic alignment at zero field is attributed to the formation of a magnetic state with relatively large domains. The small amount of remanence detected in some samples is ascribed to the persistence at zero field of a significant density of domain walls. Small fluctuations in the thickness of the layers causing a lateral distribution of the coupling strength are expected to influence the remanence magnetic state and induce differences between samples.

The structure of oxygen-related luminescence centres in oxygen-doped BaFCl was investigated by photoluminescence-detected electron paramagnetic resonance (PL-EPR). It turned out that one of the oxygen-related luminescence centres is due to an excited triplet state. The fine structure parameters of this S = 1 system are D = 188 mT and E = 28 mT with an angle of 17° in the ca plane between the z axis of the D tensor and the c axis of the crystal. The structure of the centre can be described as an oxygen-vacancy complex where the oxygen substitutes for fluorine with a next nearest chlorine vacancy.

Second-order nonlinear optical properties of Mg-doped lithium niobate single crystals are quantitatively studied from the chemical bond viewpoint. The results show that the second-order nonlinear optical response of Mg-doped LiNbO₃ single crystals at 1079 nm decreases linearly with increasing Mg concentration in the crystal. This linear correlation is quantitatively evaluated in the current work.

Polarized Raman spectra of betaine potassium bromide dihydrate were studied in the broad spectral range 5-3600 cm^-1 and in the wide temperature interval 30-292 K, with special emphasis on the structural phase transition at 160 K, recently discovered. It is shown that this phase transition is related to the freezing of the successive reorientations of the water molecules. Peculiar features and the temperature evolution of hydrogen bonding in this compound are discussed.

The optical absorption and photoluminescence (PL) of ZnO loaded in mesoporous silica have been investigated. These ZnO/SiO₂ composites are synthesized by impregnation of mesoporous silica with an aqueous solution of zinc acetate, followed by firing at 550 °C. A significant blue shift of the absorption edge compared to that of bulk ZnO in the absorption spectra is observed, and the band gap luminescence of ZnO is predominant in the emission spectrum, while the defect induced green luminescence, typical of ZnO, does not appear. However, the above sample, after additional annealing at 350 °C in air for three hours, exhibits an absorption edge (365 nm) near to that of the bulk ZnO; the colour of the sample also turns to yellowish from colourless, and green luminescence also appears. Combining the analysis of x-ray photoelectron spectroscopy (XPS), we propose that Zn-O-Si cross linking bonds formed in the interface between ZnO and the pore walls of silica have a great influence on the optical properties of ZnO/SiO₂ before the additional annealing at 350 °C and the corresponding optical properties reflect both those of SiO₂ and ZnO. After annealing, the Zn-O-Si cross linking bonds mentioned above are destroyed, and the optical properties of ZnO are exhibited alone.

Crystals, even of the same substance, grown naturally or artificially take a variety of crystal morphologies which depend on growth conditions. This work contains the analysis of the kinetic conditions corresponding to the growth of crystals of different morphologies. The crystal chosen for exemplary calculations of kinetic conditions is wulfenite, which most often takes two different habits: platelet-like and elongated along the c-axis depending on growth conditions. In particular, kinetic conditions of the growth of both types of wulfenite crystal are given. Additionally, the kinetic conditions which correspond to gradual transformation from one habit into another are presented.

The effects of internal stress and bias field on transformation strain in the Heusler alloy Ni₅₂Mn_24.4Ga_23.6 were investigated. It was found that both residual internal stress caused by the directional solidification during the growth and additional bias field with a direction consistent with the intrinsic preferential orientation would provide favourable conditions for exhibiting large transformation strain and magnetic-field-induced strain capability. These characteristics can be attributed to either nucleating favourable variants or increasing the volume fraction of favourable variants present through twin boundary motion.