Table of contents

By using density functional theory with Becke-Lee-Yang-Parr (BLYP) gradient correction, we have studied the structural and electronic properties of beryllium clusters up to 21 atoms. The theoretical calculations successfully reproduce the experimental bond length and vibrational frequency of the Be₂ dimer and the overestimation of the binding energy is significantly reduced compared to that in previous theoretical work. The Be_n clusters with n = 4,10,17 show particularly high stability, consistently with the electron shell model. The size evolution of the electronic properties from van der Waals to covalent and bulk metallic behaviour in Be clusters is discussed.

Electrical resistivity measurements have been made under nearly hydrostatic pressure up to 18 kbar and at temperatures between 50 mK and room temperature on polycrystalline specimens of the ferromagnetic compound UGe₂. The Curie temperature decreases monotonically with pressure and can no longer be identified from the electrical resistivity measurements at a critical pressure P_c≃16 kbar. Plots of dρ/dT versus T reveal a feature in the ρ(T) curves at T₀~0.6T_C at P = 0 kbar that decreases monotonically with pressure and appears to vanish near P_c. The onset of superconductivity is observed in the range 8 kbar<P<14 kbar with a maximum onset temperature of 1.2 K at P~13 kbar. The polycrystalline specimens have a residual resistivity ρ₀ up to ~3 µΩ cm corresponding to an electron mean free path smaller than or of the order of the superconducting coherence length. These results suggest that high-purity specimens with long mean free paths are not necessary, at least in the case of UGe₂, in order to observe superconductivity near the critical pressure where the magnetic ordering temperature vanishes.

We analyse the possibilities of elastic matching and effective internal stress relief at the coexistence of polar phases in ferroelectric PbZr_1-xTi_xO₃ compositions close to the morphotropic phase boundary. The important role of the polydomain (twinned) monoclinic phase in this stress relief over wide concentration and temperature ranges is discussed. A new link between phase boundaries in the phase diagram, reported by Noheda et al (Noheda B, Cox D E, Shirane G, Guo R, Jones B and Cross L E 2001 Phys. Rev. B 63 014103), and optimal domain volume concentrations, calculated in the present work, is revealed.

We propose that the behaviour of asymmetric binary fluid mixtures with a large class of attractive or repulsive interparticle interactions can be understood by mapping onto effective non-additive hard-sphere models. The latter are best analysed in terms of their underlying depletion potentials which can be exactly scaled onto known additive ones. By tuning the non-additivity, a wide variety of attractive or repulsive generalized depletion potential shapes and associated phase behaviour can be `engineered', leading, for example, to two ways to stabilize colloidal suspensions by adding smaller particles.

We make the first report that a metallic pyrochlore oxide, Cd₂Re₂O₇, exhibits type II superconductivity at 1.1 K. The pyrochlore oxide is known to be a geometrically frustrated system, which includes a tetrahedral network of magnetic ions. A large number of compounds are classified in the family of pyrochlore oxides, and these compounds exhibit a wide variety of physical properties ranging from insulator through semiconductor and from bad metal to good metal. Until now, however, no superconductivity has been reported for frustrated pyrochlore oxides. The bulk superconductivity of this compound is confirmed by measurements of the resistivity and the alternating-current magnetic susceptibility. The upper critical field H_c2, which is extrapolated to 0 K, is estimated as about 0.8 T, using the resistivity measurements under an applied field. The plot of H_c2 versus T indicates that the Cooper pairs are composed of rather heavy quasiparticles. This fact suggests that frustrated heavy electrons become superconducting in this compound.

The temperature dependence of the relaxation time which governs the intense polarization mechanisms produced by the electrochemical interaction between the matrix and the humidity in the pore space of siliceous sedimentary rocks was obtained recently by employing high-resolution thermally stimulated depolarization current spectroscopy. The transformation of the results from the time to the frequency domain provides an informative picture of the low-frequency responses. Our results are compared with those obtained from complex-impedance measurements.

This review focuses on the effect of flow fields on the orientation of block copolymer nanostructures in the melt and in solution. In particular, the alignment induced by strong shear fields or large extensional flows applied to model diblock and triblock copolymers is considered. The distinct orientation states of lamellar, hexagonal-packed cylinder, cubic micellar and bicontinuous cubic structures under different deformation conditions are summarized and the mechanisms of orientation are discussed.

A wide variety of biomolecules, ranging over proteins, enzymes, antibodies and even whole cells, have been embedded within sol-gel glasses. They retain their bioactivity and remain accessible to external reagents by diffusion through the porous silica. Sol-gel glasses can be cast into desired shapes and are optically transparent, so it is possible to couple optics and bioactivity to make photonic devices and biosensors. The high specificity and sensitivity of enzymes and antibodies allows the detection of traces of chemicals. Entrapped living cells can be used for the production of metabolites, the realization of immunoassays and even for cell transplantation.

First, pressure formulae for electrons under the external potential produced by fixed nuclei are derived both in surface integral and volume integral forms for an arbitrary volume chosen in the system; the surface integral form is based on a pressure tensor which is the sum of the kinetic and exchange-correlation parts in the density-functional theory, and the volume integral form represents the virial theorem with subtraction of the nuclear virial term. Secondly, on the basis of these formulae, the thermodynamical pressure of liquid metals and plasmas is represented in the forms of a surface integral and a volume integral including the nuclear contribution. From these results, we obtain a virial pressure formula for liquid metals, which is more accurate and simpler than the standard representation. On the basis of our formulation, some comments are made on pressure formulae derived previously and on a definition of pressure used widely.

The potential of mean force for macroparticles at infinite dilution is computed for several models of solvent-solvent and solvent-macroparticle interactions. The reference hypernetted-chain (RHNC) closure of the integral equations for the distribution functions is used. The bridge functions taken from Rosenfeld's density functional theory are computed for the special case of macroparticles at infinite dilution. This method is found to significantly improve upon previous calculations as regards the agreement with simulations. In the few cases where the agreement is semi-quantitative, possible improvements such as going beyond the second-order expansion of the attractive part of the free-energy functional are suggested.

The structures of molten MgCl₂ and ZnCl₂ have been modelled using the reverse Monte Carlo method, based on neutron and x-ray diffraction data. We show that, although the structures are similar in terms of two-body correlations (e.g. partial radial distribution functions, average coordination numbers), there are important differences in the higher order correlations (e.g. bond angle and coordination number distributions). We have also analysed the models using bond-constraint counting. All of the results are consistent with the fact that ZnCl₂ has a high viscosity and is a glass former (intermediate between strong and fragile), whereas MgCl₂ is not a glass former, and with their different crystal structures.

The phase transformations of a metastable crystalline phase obtained under high pressure occurring upon heating are considered: amorphization and subsequent crystallization. A model for the kinetics of these processes taking into account the capability for competitive formation of crystalline and amorphous phases at the boundaries of the grains of the initial phase is constructed. Expressions for the volume fractions as well as for the nucleation rates and growth velocities of the phases formed are obtained. Differential scanning calorimetry curves are described. A numerical analysis of the equations of the kinetics and a comparison with the results from experiment are carried out with reference to Cd₄₃Sb₅₇ alloy.

Raman scattering experiment and molecular dynamics simulation have been performed on orientationally disordered crystal chloroadamantane. This investigation shows clear indication for a change of vibrational dynamics in the frequency range (2 cm^-1-230 cm^-1) at the temperature T_x corresponding to the onset of non-exponential relaxation and consistent with previous NMR and MD studies. This allows us to support the existence of a `landscape-influenced' regime above the critical temperature T_c of the mode coupling theory. Convincing experimental evidence of local orientational ordering which can be nicely probed from the so-called bending mode of the chloroadamantane molecule are also given.

The results of phase transformation studies carried out in four homologues of 4-n-alkyl-4'cyanobiphenyls, using positron annihilation spectroscopy, are presented. The homologues investigated are, hexyl-, heptyl-, octyl- and decyl-cyanobiphenyls (6CB, 7CB, 8CB and 10CB). In these compounds, the positron lifetime measurements were performed as functions of temperature. The positron annihilation parameters are found to exhibit strong dependence on temperature. It was found that the ortho-positronium pick-off lifetime shows changes which strongly support (i) a gradual disappearance, instead of an abrupt one, of some memory of the more ordered solid phases on passing to the liquid crystalline phases, (ii) the strong tendency for the molecules in the mesophases to undergo anti-parallel bimolecular association and (iii) the formation of cyabotactic groups of a smectic phase in a nematic medium. Changes were also observed in the ortho-positronium formation probability which apparently indicate a systematic transformation of the solid phase from a close-pack structure to an open-pack structure as one goes from a lower to a higher homologue of the compounds investigated. Solid crystalline polymorphism has been observed in 8CB. A change in molecular packing in the solid phase of 10CB has been observed.

We incorporate the role of free volume in the density function of amorphous structure and study its effects on the stability of such structures. Density functional theory is used to explore this `free-volume model' of supercooled structures. Free-energy minimization is carried out using the void concentration as a variational parameter. A critical value of this concentration exists corresponding to the free-energy minima of the amorphous structure. An increase in the stability is observed due to the inclusion of voids in the density structure. This study is conducted for both weakly and highly localized amorphous structures. The free-volume concentration shows a power-law decrease with density for the weakly localized states and a linear decrease for the highly localized amorphous structures.

The effect of the shape distribution of granular inclusions on the effective non-linear optical properties of granular metal/dielectric composites is considered. The study is based on a generalized Maxwell–Garnett type approximation for the spectral density function of two-component composites in which the metallic inclusions possess a `beta function' distribution of geometric shapes. Numerical results show that the spectral density function is increased (decreased) in the range 0.35 ≤ s' ≤ 0.5 (0.8 ≤ s' ≤ 1.0) with increasing shape distribution parameter α. By invoking the mean-field approximation, we calculate the optical nonlinearity and find the nonlinearity enhancement peak is separated from the absorption peak in the range 0.45 ≤ ω/ω_p ≤ 0.6, while a large enhancement of the optical nonlinearity is found in the range 0.8 ≤ ω/ω_p ≤ 1.0. Thus by introducing a shape distribution of metallic particles, we are able to make the figure of merit attractive in this frequency range. In the dilute limit, the shape distribution leads to an anomalous far-infrared optical absorption. Moreover, an exact formula for the effective optical nonlinearity is derived and a sharp nonlinearity enhancement peak is observed near the resonant frequency ω/ω_p ≈ 0.47.

The coherent response of electrons in a double-well superlattice with two minibands when the superlattice is photoexcited with an ultra-short laser pulse is studied theoretically. Since the gap between the two minibands oscillates with the variation of the quasimomentum, not only do quantum beats - which depend on the parameters of the superlattice - occur, but also a non-dissipative quenching of the photoinduced current oscillations appears over time. These modifications of the oscillations are examined within the framework of the Kane model, using the parabolic approximation for electron and heavy-hole states and neglecting the Coulomb interaction. Numerical calculations of the carrier coherent dynamics are performed for the case of a δ-pulse excitation with a phenomenologically introduced damping factor.

The effect of pressure on the structural behaviour of In alloys with Cd, Sn and Pb has been studied with diamond anvil cells using synchrotron radiation. The face-centred cubic phase (fcc) of an In alloy with 6 at% Cd transforms under pressure into a face-centred tetragonal In-type phase, fct, with c/a>1. The fcc phases of In alloys with 40 and 60 at% Pb transform under pressure to fct with c/a<1. For the ambient pressure fct phase (c/a<1) of an In alloy with 20 at% Sn the distortion increases with pressure. The stability of the tetragonal phases under pressure with respect to the cubic structure is discussed in terms of Brillouin-zone-Fermi-sphere interaction.

A Raman study of structural changes in the Zr-rich PbZr_1-xTi_xO₃ (PZT) system under hydrostatic pressures up to 5.0 GPa is presented. We observe that externally applied pressure induces several phase transitions in PZT ceramics among phases with orthorhombic (A_O), rhombohedral low-temperature (R_LT), and rhombohedral high-temperature (R_HT) symmetries (all found in PZT at ambient pressure and room temperature). Each of the compositions investigated (0.02⩽x⩽0.14) exhibits a high-pressure phase with orthorhombic (O_I') symmetry. We further report a detailed study of the pressure dependence of Raman frequencies to elucidate the phase transitions and to provide a set of pressure coefficients for the high-pressure phases.

Complex permittivity and Young's modulus provide relevant information on the role of point defects in the dielectric and mechano-elastic properties of ferroelectric materials. Low-frequency measurements as a function of the temperature performed on Bi₄Ti₃O₁₂ (BIT) have shown that point and dipole defects are frozen close to domain walls. Low-temperature dipole defect relaxation processes take place with characteristic times (τ₀) of the order of 10^-11 s and 10^-12 s and activation energies (E_a) of 0.70 eV and 0.65 eV for dielectric and mechano-elastic relaxations, respectively. At higher temperatures new dielectric relaxation peaks appear that can be attributed to jumps of de-iced oxygen vacancies (τ₀≅10^-11 s, E_a = 1.08 eV, T≅300 °C) and to vacancy migration (τ₀≅10^-15 s, E_a = 1.90 eV, T≅450 °C). Elastic relaxation peaks are also present close to 300 °C whose activation energy (1.50 eV) and characteristic time (10^-15 s) suggest a vacancy migration process. Close to 500 °C with E_a = 2.30 eV and τ₀≅10^-17 s another relaxation peak, which should correspond to domain wall viscous motion near the phase transition temperature, is observed. The Young's modulus has a smooth step at T≅300 °C that we attribute to a change in the mobility of oxygen vacancies with respect to the domain walls. Below 300 °C the vacancies are frozen in the domain walls and they are de-iced and distributed throughout the material at temperatures above 300 °C. The experimental results show that the material is softer when the vacancies are linked to domain walls than when they are distributed throughout the material. The diffusion of vacancies back to the domain wall traps at room temperature takes a long time (days).

Hydrogen in the C15 Laves-phase material TaV₂H_x has been studied by means of resonant ultrasound spectroscopy over the temperature range of 15-345 K for a series of hydrogen concentrations (x = 0.00-0.53). Ultrasonic loss peaks and frequency shifts (dispersion) associated with the hydrogen motion were observed, yielding parameters for the hydrogen motion. Hydrogen in these materials is known to occupy the tetrahedral g sites which form a series of interlinked hexagons. The ultrasonic results were associated with H hopping between g-site hexagons. The relaxation rates for x⩽0.18 were best described as a sum of two Arrhenius processes. For x = 0.34 and 0.53 only a single Arrhenius process was needed to fit the results, although the presence of a second Arrhenius mechanism could not be over-ruled. A single relaxation rate was sufficient to fit the data; a distribution of rates was not required. The magnitudes of the attenuation and dispersion depended linearly on the hydrogen concentration implying that it is the relaxation of isolated H atoms (the Snoek effect) that is responsible for the mechanical damping. The faster local motion of H reported from nuclear magnetic resonance measurements for motion within g-site hexagons was not observed in the present study. This suggests that the H hopping rate for the local motion remains above the ultrasonic frequencies over the temperature range of study, or perhaps that too few H atoms participate in the local motion.

Molecular dynamics simulations have been performed to investigate the early-time dynamic behaviour of gas-liquid phase separation in two dimensions by using the evolution of the potential energy. It is found that the average potential energy per particle u(t) shows a good linear relationship with the average number of nearest-neighbour particles n_nn(t) after a brief time. The results also show that u(t) follows a positive power law in time in the early stage of phase separation. The excess energy per particle u(t)-u_∞∝t^-1/2 is also confirmed in the late stage. These results suggest that u(t) may be considered as a parameter describing the entire phase separation process in gas-liquid systems.

Polarized electronic absorption spectra of Co²⁺ in trigonally compressed octahedra in brucite-type Co(OH)₂ have been measured at 290 and 90 K by microscope-spectrometric techniques and analysed in terms of the superposition model (SM) of crystal fields. The resulting SM and interelectronic repulsion parameters are ₄ = 5260, ₂ = 4920, Racah B = 825, Racah C = 3550 cm^-1 at 290 K and ₄ = 5320, ₂ = 3900, Racah B = 830, Racah C = 3500 cm^-1 at 90 K (R₀ = 2.1115 Å; fixed exponential and spin-orbit parameters t₄ = 5, t₂ = 3, ζ = 500 cm^-1). Together with a recent SM analysis of Li₂Co₃(SeO₃)₄, the _k refined for Co(OH)₂ further confine the magnitude of the hitherto unknown `correct' SM parameters of Co²⁺ for future application to structurally and/or chemically less well defined systems.

In order to reproduce the transient IR-FIR reflectivity and transmission due to an intense photo-plasma generated by a fast laser pulse at a frequency above the band gap, we have used a dynamical model of plasma evolution. In this way we have derived both the `quadratic' and `cubed' coefficients of the Auger recombination in indium antimonide, by analysing the transient reflectivity at 10.6 µ and 119 µ induced by a fast Nd pulse. In particular we have derived a cubic coefficient of about 7±3×10^-26 cm⁶ s^-1, a result larger than those derived in previous experimental works but quite in agreement with the theory.

We investigate tunnelling of electrons through a superconducting grain at nanometre scales. The conductance is computed from a network of many-body states, taking into account contributions from normal tunnelling and Andreev reflection. The contribution from Andreev reflection is maximized if the grain ground state is degenerate with a state having an extra pair of electrons. By increasing the Coulomb interaction within the grain, the system exhibits a crossover from the Andreev behaviour to the Kondo behaviour at a point where the charging energy is equal to the pairing energy.

We have studied the variation of superconducting critical temperature T_c as a function of charge density and lattice parameters in Mg_1-xAl_xB₂ superconducting samples. The AB₂ heterostucture of metallic boron layers (intercalated by A = magnesium, aluminum layers, playing the role of spacers) is made by direct chemical reaction. The spacing between boron layers and their charge density are controlled by chemical substitution of Mg by Al atoms. We show that high T_c superconductivity is realized by tuning the chemical potential at a `shape resonance' according with the patent for `high-temperature superconductors made by metal heterostructures at the atomic limit'. The energy width of the superconducting shape resonance is found to be about 400 meV.

Rietveld analysis was performed for the intermetallics Yb₃Co_4.3Sn_12.7 and Yb₃Co₄Ge₁₃ crystallizing with the closely related structure types, Yb₃Rh₄Sn₁₃ and Yb₃Co₄Ge₁₃. Below T_c = 3.4 K Yb₃Co_4.3Sn_12.7 crosses over into a type-II superconducting ground state with H_c2(0)~2.5 T. Yb₃Co₄Ge₁₃ stays in the normal state down to 300 mK. The γ value of 2.3(2) mJ gat^-1 K^-2 and the Debye temperature Θ_D = 207(5) K deduced from the specific heat as well as T_c correspond to that of elementary Sn, thus indicating conventional BCS superconductivity. Hydrostatic pressure applied to Yb₃Co_4.3Sn_12.7 reveals both an overall decrease of the absolute resistivity values, as well as a decrease of T_c, which vanishes for a critical pressure below 10 kbar. The magnetoresistance of both Yb-based compounds is positive at low temperature but does not exceed 8% in fields of 12 T. The Seebeck coefficient has a maximum value of about 18 µV K^-1 at T~250 K. L_III and magnetic susceptibility measurements reveal intermediate valence: 2.66(3) and 2.18(3) for Yb₃Co₄Ge₁₃ and Yb₃Co_4.3Sn_12.7, respectively.

We present two-photon excitation spectra for thulium III in the elpasolite Cs₂NaYCl₆:Tm. 37 out of the total of 40 crystal-field (CF) levels have been assigned, with the aid of a one-electron CF Hamiltonian, representing the most extensive data set so far reported for Tm III at a cubic site. Deviations from the predictions of the one-electron CF model are unusually large, the effective fourth-rank CF parameter being 60% larger for the singlet states than for the triplet states. This is discussed in terms of a spin-polarized covalency that is more pronounced in the singlet states of the metal ion.

The effect of the four-spin cyclic exchange interaction at each plaquette in the S = 1/2 two-leg spin ladder is investigated at T = 0, focusing especially on the field-induced gap. The strong-rung-coupling approximation suggests that it yields a plateau at half of the saturation moment (m = 1/2) in the magnetization curve, which corresponds to a field-induced spin gap with a spontaneous breaking of the translational symmetry. A precise phase diagram at m = 1/2 is also presented, based on the level spectroscopy analysis of the numerical data obtained by the Lanczos method. The boundary between the gapless and plateau phases is confirmed to be of the Kosterlitz-Thouless (KT) universality class.

A series of new layered 2D-network complexes [M(hfac)₂]₃(R_Δ)₂ of M = Mn(II) and Cu(II) with trisnitroxide radicals R_Δ has been prepared and the magnetic properties were studied. Each triradical R_Δ has a quartet ground state and contributes not only to the formation of extended structures but essentially to the overall magnetism. Several exchange interactions, between M and nitroxide and intraradical nitroxide–nitroxide interactions, are responsible for the development of the characteristic magnetic properties in these heterospin systems. Depending on the nature of the interlayer interactions, they show either ferro/ferrimagnetic or antiferromagnetic long range order. The hierarchy of the different exchange interactions is established and the Mn–nitroxide and Cu–nitroxide exchange integrals are evaluated from the analysis of the temperature dependence of the paramagnetic susceptibility. With increasing intraradical exchange interaction, the complexes exhibit more pronounced 2D behaviour.

Spin-lattice relaxation time T₁ was determined by the electron spin echo (ESE) method in the temperature range 4-60 K in a series of Tutton salt crystals M^I₂M^II(SO₄)₂·6X₂O (M^I = NH₄, K; M^II = Zn, Mg; X = H, D) weakly doped (⩽10¹⁸ ions cm^-3) with the ⁶³Cu²⁺ isotope. The ESE signal was undetectable at higher temperatures. The relaxation rate increases over the six decades in the studied temperature range with T₁ equal to 1 s at 4 K and 0.5 µs at 50 K. Various possible relaxation mechanisms are discussed with the conclusion that the relaxation is governed by two-phonon Raman processes without a noticeable contribution from the reorientations of Cu(H₂O)₆ octahedra between Jahn-Teller distorted configurations. Deuteration of the crystal has no effect in spin-lattice relaxation. For a few crystals, having the largest Cu²⁺ concentration among the studied crystals, a strong and linear in temperature contribution to the relaxation rate was found below 15 K. Possible explanations are discussed with the final conclusion that this effect is due to a non-uniform Cu²⁺ distribution in the host lattice producing effective relaxation via pairs and triads of the Cu²⁺ ions. From the T₁(T) dependence the Debye temperature Θ_D was determined for the all crystals studied. This varies from Θ_D = 166 K for K₂Zn(SO₄)₂·6H₂O to Θ_D = 238 K for (NH₄)₂Mg(SO₄)₂·6H₂O. The Θ_D values are discussed and used for calculation of the sound velocity which was found to be similar in all crystals and equal to ν = 4150(±150) m s^-1.

The assignment of the 1.193 eV zero-phonon line of the 3d² ion in GaN crystal to the internal luminescence of Cr⁴⁺ on the Ga³⁺ site is analysed. Based on this assignment, the g-shift Δg ( = g-g_s) of the groundstate of this luminescence is calculated by considering not only the conventional contribution due to the crystal-field mechanism, but also the contribution due to charge-transfer mechanism. The calculated result shows good agreement with the observed value, suggesting that the assignment is reasonable. The calculated Δg due to the charge-transfer mechanism is opposite in sign and 62% in magnitude, compared with that due to the crystal-field mechanism, and so it cannot be neglected. It appears that for the 3d² (or 3dⁿ) ion with a high valence state in crystals (particularly, in the case of the ligand having small optical electronegativity), the reasonable explanation of the g-shift should consider the contributions from both the crystal-field and charge-transfer mechanisms.

The effect of vibronic coupling (orbit–lattice interaction), which produces a mixing between the low-lying electronic states with the emission and absorption of phonons of varying energies, is considered to explain the observed temperature dependence of Mössbauer quadrupole splitting of Fe²⁺ ions in CsCoCl₃ over 22 to 300 K. The various lattice dynamical parameters are estimated and then quadrupole splitting is explicitly calculated as a function of temperature. One obtains a reasonably good agreement with the experimental data, which shows that the vibrating lattice model is quite realistic and important.