Table of contents

We have studied the effect of pressure on the electrical resistivity of the antiferromagnet CeMg₂Cu₉ which crystallizes in the hexagonal structure. The structure is built up of alternating MgCu₂ Laves-type and CeCu₅-type layers along the [0001] direction. The Néel temperature T_N = 2.7 K at ambient pressure decreases with increasing pressure p and disappears at a critical pressure p_c≃2.5 GPa. Correspondingly, the residual resistivity ρ₀ and the coefficient A in a Fermi-liquid relation ρ = ρ₀ + AT² are found to have maximum values around p_c.

Nonlinear charge transport in semiconductor superlattices under strong electric fields parallel to the growth direction results in rich dynamical behaviour including the formation of electric field domains, pinning or propagation of domain walls, self-sustained oscillations of the current and chaos. Theories of these effects use reduced descriptions of transport in terms of average charge densities, electric fields, etc. This is simpler when the main transport mechanism is resonant tunnelling of electrons between adjacent wells followed by fast scattering between subbands. In this case, we will derive microscopically appropriate discrete models and boundary conditions. Their analyses reveal differences between low-field behaviour where domain walls may move oppositely or parallel to electrons, and high-field behaviour where they can only follow the electron flow. The dynamics is controlled by the amount of charge available in the superlattice and doping at the injecting contact. When the charge inside the wells becomes large, boundaries between electric field domains are pinned resulting in multistable stationary solutions. Lower charge inside the wells results in self-sustained oscillations of the current due to recycling and motion of domain walls, which are formed by charge monopoles (high contact doping) or dipoles (low contact doping). Besides explaining wave motion and subsequent current oscillations, we will show how the latter depend on such controlling parameters as voltage, doping, temperature, and photoexcitation.

One of the main goals of molecular biology is to understand the structure of biomolecules. With the emergence of single molecule manipulation techniques that structure can now be controlled by the application of stretching and torsional stresses. In this article we review some recent experiments on the stretching and twisting of single biopolymers, testing the elastic properties of DNA and proteins and studying their stress-induced structural transitions. Numerical simulations have emerged as a precious tool to interpret the experimental data and predict the associated structural changes. We shall explain how a combination of these experimental and computational tools open a new vista on the structure of biomolecules.

It is shown how the Langevin equation for the motion of the magnetization of a ferrofluid particle with uniaxial anisotropy in a strong uniform applied field reduces to those governing the Néel (i.e. the solid-state or internal) mechanism of reorientation of the magnetic moment in the non-axially symmetric potential created when a field is applied at an angle to the easy axis and a Larmor-like equation for the transverse motion. The field angle, unlike in the solid-state problem, is a function of the time due to the torques imposed by the fluid carrier. The Langevin equation for the Brownian rotational motion of the particle itself reduces to that describing Debye relaxation in the applied field but is coupled to the magnetic motion via the external field. The results indicate that the dissipation parameter of the internal solid-state mechanism is augmented by the external stochastic torques imposed by the carrier. However, the effect appears to be negligible because of the ratio of the Brownian (Debye) time to the free Néel diffusion time. Furthermore, just as in the pure solid-state process, pronounced precession-aided longitudinal relaxation and ferromagnetic resonance effects, having their origin in the breaking of the axial symmetry due to the strong field, will occur. The precession-aided relaxation disappears for weak fields since the potential becomes axially symmetric. Moreover, the equations of motion of the magnetic moment and the particle completely decouple and the overall decay function is simply the product of the decay functions of the internal (Néel) and Debye processes. It appears that the ferromagnetic resonance in this instance is accurately described by the known solid-state results, since the Brownian relaxation time greatly exceeds the effective relaxation times of the internal dipole and quadrupole modes associated with the ferromagnetic resonance. This conclusion is reinforced by the favourable agreement of the weak-field result with experimental observations of the complex susceptibility of four ferrofluid samples.

Phase transformations such as freezing typically start with heterogeneous nucleation. Heterogeneous nucleation near a wetting transition, of a crystalline phase, is studied. The wetting transition occurs at or near a vapour-liquid transition which occurs in a metastable fluid. The fluid is metastable with respect to crystallization, and it is the crystallization of this fluid phase that we are interested in. At a wetting transition a thick layer of a liquid phase forms at a surface in contact with the vapour phase. The crystalline nucleus is then immersed in this liquid layer, which reduces the free-energy barrier to nucleation and so dramatically increases the nucleation rate. The variation in the rate of heterogeneous nucleation close to wetting transitions is calculated for systems in which the longest-range forces are dispersion forces.

We investigate liquid crystalline phases with nematic order at (1/3)-filling of the valence Landau level (LL). We generalize Laughlin's fractional quantum Hall (QH) effect wavefunction at ν = (1/3) to include anisotropic nodal distribution by modifying the Jastrow factors. Lengthy Monte Carlo simulations are then used to determine with unprecedented accuracy the (anisotropic) pair distribution function g(r) and static structure factor S(q) for various degrees of anisotropy. The determination of the correlation energies at (1/3)-filling of an arbitrary LL is then performed by using standard mappings of g(r) and S(q) to higher LLs. Our results indicate that while Laughlin's state is stable in the lowest LL, there are regions of instability towards nematic order in higher LLs. Possible connections to the recently discovered QH liquid crystals are discussed.

We have carried out ab initio molecular-dynamics simulations for liquid phosphorus under high temperature and high pressure in order to investigate the microscopic mechanism of the recently observed liquid-liquid phase transition of liquid phosphorus. We have successfully shown by our simulation that the structural phase transition with increasing pressure corresponds to the structural change from the molecular liquid composed of stable tetrahedral P₄ molecules to the polymeric liquid with complex network structure and that the calculated structure factors are in good agreement with those obtained by the x-ray diffraction experiments. It is also found from our calculated electronic structure that this structural change gives rise to the nonmetal-metal transition, which is the transition from the nonmetallic molecular liquid to the metallic polymeric liquid.

The microwave and millimetre-wave dielectric response of Rb_0.5(ND₄)_0.5D₂PO₄ dipolar glass has been studied around the frustration temperature T_f≈120 K. It was found that above the frustration temperature T_f the soft relaxational deuteron mode is responsible for the whole dielectric dynamics below the phonon range. Its frequency decreases from that of submillimetre waves to that of microwaves following the classical law ν_R = A(T-T_f), with A = 0.75 GHz K^-1, but in the vicinity of T_f the soft mode gradually shifts towards the glass-like dispersion predominantly related to the diffusion of Takagi defects according to the Vogel-Fulcher law. From the experimental results, the distribution of the relaxation times and of the local polarization at various temperatures is calculated. It is shown that the local polarization distribution function obtained from the dielectric response ε*(ν,T) coincides well with that obtained from the NMR results.

As an extension of our previous study on the electron polarons and excitons in KNbO₃ and KTaO₃ [1,2], we present here results of semi-empirical intermediate-neglect-of-differential-overlap (INDO) calculations for free electron polarons, single-triplet excitons and the excitonic phase in BaTiO₃ perovskite crystal. Our INDO calculations confirm the existence of self-trapped electrons in BaTiO₃. The corresponding lattice relaxation energy is 0.24 eV and the optical absorption energy 0.69 eV. An electron in the ground state occupies the t_2g orbital of the Ti³⁺ ion. Its orbital degeneracy is lifted by a combination of the breathing and Jahn-Teller modes when four nearest equatorial O atoms are displaced by 1.53% a₀ outwards in the x-y plane and another two nearest oxygens shift 1.1% inwards, along the z-axis. Our INDO calculations show that creation of charge-transfer vibronic exciton (CTVE) in BaTiO₃ crystal is accompanied by a strong lattice distortion; the relevant energy gain due to CTVE formation is 2.2 eV. Moreover, our INDO calculations predict the existence of a new crystalline phase - that of CTVEs in BaTiO₃ where strongly correlated CTVEs are located in each unit cell of a crystal.

Diamond is a wide-band-gap material with large donor and acceptor ionization energies. In principle, at room temperature and below, the Fermi energy is pinned close to the donor or acceptor level, depending on which is present in the higher concentration. In semiconductors with shallow donors and acceptors the equilibrium charge states of defects are determined by the position of the Fermi level. However, in an insulating material like diamond we show that the calculated position of the Fermi level does not necessarily predict the correct charge state of a defect, and propose instead that the charge state is influenced by the proximity of the defect to a donor (or acceptor). Qualitatively this accounts for the dependence of the charge state on the concentration of isolated substitutional nitrogen and also explains why many optical centres can be present in two different charge states in the same diamond.

We consider a mean-field approach to the hole-mediated ferromagnetism in III-V Mn-based semiconductor compounds in order to discuss the dependence of the hole density on that of Mn sites in Ga_1-xMn_xAs. The hole concentration, p, as a function of the fraction of Mn sites, x, is parametrized in terms of the product m*J_pd² (where m* is the hole effective mass and J_pd is the Kondo-like hole/local-moment coupling), and the critical temperature T_c. By using experimental data for these quantities, we have established the dependence of the hole concentration on x, which can be associated with the occurrence of a re-entrant metal-insulator transition taking place in the hole gas. We also calculated the dependence of the Mn magnetization on x, for different temperatures (T), and found that as T increases, the width of the composition-dependent magnetization decreases dramatically, and that the magnetization maxima also decrease in magnitude, indicating the need for quality control of the Mn doping level in diluted magnetic semiconductor devices.

We report a detailed study of the magnetic-field-orientation dependence of the millimetre-wave magnetoconductivity of the superconductor Sr₂RuO₄. We find two harmonic series of cyclotron resonances. We assign the first, corresponding to a quasiparticle mass of 4.29±0.05 m_e, where m_e is the free-electron mass, to the β Fermi-surface section. We assign the second series, which contains only odd harmonics, to cyclotron resonance of the γ Fermi-surface section, yielding a quasiparticle mass of 12.35±0.20 m_e. A third, single cyclotron resonance, corresponding to a quasiparticle mass of 5.60±0.03 m_e, is attributed to the α Fermi-surface section. In addition, we find a very strong absorption mode in the presence of a magnetic field component parallel to the quasi-two-dimensional planes of the sample. Its dependence on the orientation of the magnetic field cannot be described in the context of conventional cyclotron resonance, and the origin of this mode is not yet clear. Finally, magnetic quantum oscillations due the α and β Fermi-surface sections are observed in the millimetre-wave magnetoconductivity of Sr₂RuO₄, and a detailed study of their magnetic-field-orientation dependence is presented.

Co₂(Nb_1-xV_x)Sn, Co₂NbSn_1-yGa_y and Co₂NbSn_1-zAl_z alloy systems exhibit itinerant ferromagnetic behaviour as observed from magnetization and susceptibility studies. Spin fluctuation effects are observed from magnetization, resistivity and thermopower data. The thermal expansion data gave evidence of structural transformation effects due to the formation of a pseudogap in the density of states near the Fermi level.

We describe a procedure for determining possible macroscopic symmetries of a multidomain ferroic crystal. The domain structure is represented by ferroic domain states and by their partial volumes which define its domain configuration. If such a crystal is exposed to external field(s), all possible domain states may no longer have equal free energy. Except for special cases, all states with same free energy are equivalent under the maximal subgroup H of the prototypic point group G that leaves the field(s) invariant; such states form an H-orbit. Provided that the ferroic crystal contains all states from a single H-orbit, and if these states occupy equal partial volumes, a coherent domain configuration arises. Its averaged symmetry is given by the stabilizer of the H-orbit, i.e. by the maximal subgroup of G transforming the H-orbit of states into itself. Within the model used, the possible macroscopic symmetry of the crystal is either the symmetry of a coherent domain configuration or an intersection of some of these symmetries.

The procedure is demonstrated on rhombohedral perovskite crystals for which all macroscopic symmetries and the external fields that produce coherent configurations are given.

Mineral zircon has been considered as a possible medium for luminescence dating. The development of a suitable material for luminescence dating requires detailed knowledge of the processes taking place during, for example, exposure to ionizing radiation, long-term storage, annealing at moderate temperatures, excitation with (visible-UV) light. In this paper we have described our efforts to obtain relevant dating information by investigating the electron paramagnetic resonance (EPR) spectra of a variety of paramagnetic defects in mineral zircon (ZrSiO₄) crystals as a function of the irradiation dose, annealing time and the temperature. The rare-earth ions Dy³⁺ and Tb³⁺, which play a crucial role as hole traps and recombination centres, have been investigated in detail and the behaviour of the intensity of the EPR spectra associated with these impurities can be understood in terms of a theoretical model describing the luminescence related processes in zircon. In addition, a number of defects, which can be characterized by SiO_m^n- have been identified and investigated. Also the behaviour of some of these centres has been analysed in the framework of the theoretical model.

The direct calculation of transition line strengths and relative intensities is presented for two intraconfigurational two-photon absorption (TPA) transitions of Eu³⁺ in the cubic Cs₂NaYF₆ host. Crystal field wavefunctions were utilized for the initial and final f^N-electron states and various approaches were used in constructing all the 4f^N-1 5d¹ intermediate-state wavefunctions. The calculated relative intensities of the (⁷F₀) Γ_1g→(⁵D₂)Γ_5g, Γ_3g TPA transitions are in reasonable agreement with experiment. The neglect of J-mixing in the initial state has only a small effect upon the calculation, whereas the neglect of spin-orbit couplings within the initial and terminal states drastically reduces the calculated transition linestrengths, but does not markedly change the intensity ratios. In the case of the (⁷F₀)Γ_1g→(⁵L₆)Γ_1g, aΓ_5g transitions, serious discrepancies between experiment and theory are found if the intermediate states are constructed from a 4f⁵ core comprising free ion states and the 5d¹ crystal field states. Satisfactory agreement is, however, found when the 4f⁵ crystal field states are utilized in constructing the intermediate states. The contributions to the transition moment have been evaluated for various Hamiltonian terms and the results are discussed.

Due to an error in the production process, our paper entitled: `Asymmetric binary mixtures with attractive forces: towards a quantitative description of the potential of mean force', was published with an incorrect set of figures. The correct set is given in the PDF file.