Table of contents

The magnetocaloric effect (MCE) of half-metallic CrO₂ particles has been studied with respect to particle size on the nanometre scale. Results from superconducting quantum interference device (SQUID) measurements indicate that acicular CrO₂ particles with a length of 400 nm yield a large magnetic entropy change ΔS_M of 5.1 J kg⁻¹ K⁻¹ at an applied field of 15 kOe and an adiabatic temperature change ΔT_ad of 2.0 K near the Curie temperature (∼390 K). These results are among the highest for magnetic oxides, and are comparable to that for pure Gd. However, smaller CrO₂ particles with a length of 260 nm only exhibit ΔS_M = 2.25 J kg⁻¹ K⁻¹ and ΔT_ad = 0.95 K. The difference in MCE between these two sizes of CrO₂ particles results primarily from disordered spins on the surface of the particles. In addition, measurements and calculations of the specific heat capacity for the CrO₂ particles are presented. These results indicate that the total specific heat capacity is dominated by the magnetic specific heat contribution. Therefore, we believe that these CrO₂ nanoparticles may hold future promise in the development of new magnetic refrigerants.

Over the past two decades, unconventional superconductivity with gap symmetry other than s wave has been found in several classes of materials, including heavy fermion, high T_c, and organic superconductors. Unconventional superconductivity is characterized by anisotropic superconducting gap functions, which may have zeros (nodes) along certain directions in the Brillouin zone. The nodal structure is closely related to the pairing interaction, and it is widely believed that the presence of nodes is a signature of magnetic or some other exotic, rather than conventional phonon mediated, pairing mechanism. Therefore experimental determination of the gap function is of fundamental importance. However, the detailed gap structure, especially the direction of the nodes, is an unresolved issue for most unconventional superconductors. Recently it has been demonstrated that thermal conductivity and specific heat measurements under a magnetic field rotated relative to the crystal axes provide a powerful method for determining the shape of the gap and the nodal directions in the bulk. Here we review the theoretical underpinnings of the method and the results for the nodal structure of several unconventional superconductors, including borocarbide YNi₂B₂C, heavy fermions UPd₂Al₃, CeCoIn₅, and PrOs₄Sb₁₂, organic superconductor κ-(BEDT-TTF)₂Cu(NCS)₂, and ruthenate Sr₂RuO₄, determined through angular variation of the thermal conductivity and heat capacity.

Thin films of amorphous tungsten trioxide, a-WO₃, have been thermally evaporated onto glass substrate held at 350 K. Annealing at 723 K caused the formation of polycrystalline tungsten trioxide, c-WO₃, with a monoclinic structure. The dark DC electrical conductivity of both a-WO₃ and c-WO₃ was studied over a temperature range from 298 to 625 K in two environmental conditions (air and vacuum). A simple Arrhenius law, a polaron model and a variable range hopping model have been used to explain the conduction mechanism for a-WO₃ films. Using the variable range hopping model, the density of localized states at the Fermi level, N(E_F), was found to be 1.08 × 10¹⁹ eV⁻¹ cm⁻³. The mechanism of electrical conduction in c-WO₃ films is explained by means of the Seto model. The Seto model parameters were determined as the energy barrier (E_b = 0.15 eV), the energy of trapping states with respect to the Fermi level (E_t = 0.9 eV) and the impurity concentration (N_D = 4.05 × 10¹⁵ eV⁻¹ cm⁻³). The thickness dependence of resistivity of c-WO₃ films has been found to decrease markedly with increasing film thickness, which is explained on the basis of the effective mean free path model. Using this model, the mean free path of electrons in c-WO₃ films was evaluated. The temperature dependence of the thermoelectric power for a-WO₃ films reveals that our samples are n-type semiconductors.

Based on the Bean–Rodbell model, which assumes a linear variation of exchange coupling with atom spacing, the magnetovolume effects in LaFe_13−xSi_x (x = 1.2–2.0) have been systematically studied. A relation between phase volume and magnetization is first obtained by comparing the structural and magnetic data collected at various temperatures. The maximum spontaneous magnetostriction thus derived is dependent on the content of Si, linearly decreasing from ∼2.15% for x = 1.2 to ∼1.12% for x = 2. Based on these results and limited experimental data, the parameters involved in the Bean–Rodbell model are determined for the LaFe_13−xSi_x compounds. Further analysis indicates that the Bean–Rodbell model equipped with these parameters gives a satisfactory description of the magnetovolume effects produced by interstitial hydrogen for the LaFe_11.44Si_1.56 hydride. To explain the pressure effects, in contrast, changes of the parameters under pressure, which are a result of the enhancement of the first-order character of the phase transition, have to be taken into account. These results indicate that either the increase or the decrease of the Curie temperature is simply a consequence of the variation of the phase volume due to the introduction of interstitial atoms or the application of a high pressure, and can be described well by the Bean–Rodbell model.

Neutron diffraction and NMR relaxation measurements have been made of water/ice in SBA-15, a mesoporous silica constituting an ordered array of cylindrical mesopores of pore diameter ∼86 Å, over the temperature range 180–300 K in a cooling and heating cycle. The over-filled sample shows the initial formation of hexagonal ice on the outside of the silica grains, followed by the nucleation of cubic ice inside the pores at a lower temperature. Neutron scattering profiles for the cubic ice peaks are significantly broadened and indicate a defective structure, as observed in previous experiments on ice formation in sol–gel and MCM-type silicas. Below the pore freezing temperature the intensity of the cubic ice peaks exhibit a significant increase, down to the lowest experimental temperature, indicating a reversible conversion of defective ice to ordered ice crystals. The peak profile analysis for the two ice patterns indicates a systematic variation in the position as a function of temperature, giving values of the expansion coefficients that are slightly lower than other measurements for the bulk phase. NMR results on proton relaxation as a function of temperature indicate the presence of a mobile phase for temperatures below pore freezing that supports the view that there is interconversion between brittle and plastic phases of ice.

Starting from a quantum Langevin equation (QLE) of a charged particle coupled to a heat bath in the presence of an external magnetic field, we present a fully dynamical calculation of the susceptibility tensor. In a different 'equilibrium approach', we further evaluate the position autocorrelation function by using the Gibbs ensemble. This quantity is shown to be related to the imaginary part of the dynamical susceptibility, thereby validating the fluctuation–dissipation theorem in the context of dissipative diamagnetism. Finally, we present an overview of coherence-to-decoherence transition in the realm of dissipative diamagnetism at zero temperature. The analysis underscores the importance of the details of the relevant physical quantity, as far as coherence-to-decoherence transition is concerned.

Composite optical waveguides of Eu doped Y₂O₃ nanolayers with various theoretical thicknesses from 10 to 0.2 nm embedded in an amorphous alumina matrix were prepared by targeted alternating pulsed laser ablation. Their structural and fluorescence properties were compared to those of thick nanocrystallized Eu:Y₂O₃ films deposited under the same conditions. Whereas layers with thicknesses of 5 and 10 nm are continuous and crystalline, layers with thickness of less than 2 nm are constituted of isolated amorphous nanoparticles with a mean height of around 1 nm, according to transmission electron microscopy and atomic force microscopy. The Eu³⁺ fluorescence properties show that Eu³⁺ ions are localized in different sites inside the nanolayers or nanoparticles or at the interface with the matrix, depending on the size of the nanostructure. The refractive index of the composite waveguides was measured by m-line spectroscopy. Waveguiding propagation was observed.

In this paper we analyse how electron transport through a one-dimensional chain is modified by the presence of phonon dephasing mechanisms active in a limited strand of the chain. The treatment is based on the nonequilibrium Keldysh Green's function and the self-consistent first Born approximation, with a tight-binding description of the electronic states. A most remarkable feature of the calculated conductance curves is the occurrence of an exponential decrease for small lengths, followed by a slow asymptotic decrease inversely proportional to the strand length. The origin of such a different short-range and long-range behaviour of the conductance, and other observed scaling features, are interpreted with some intuitive understanding of the dephasing mechanisms.

The para-ferroelectric tricritical phase transition of a single crystal of triglycine selenate (TGSe) has been studied in the neighbourhood of the transition temperature, under weak electric fields, E, using a highly sensitive calorimetric technique. The specific heat, c_E, under fields in the range between 5 and 175 V cm⁻¹ and close to transition temperature (0.2 K), shows different behaviour on cooling and on heating at a temperature variation rate of ± 0.03 K h⁻¹; for T<T_c−0.5 K all sets of data match each other. The experimental data have been fitted to the 2–6 Landau potential obtained for E = 0; a term −ξQ, taking into account the coupling between E (ξ being dependent on E) and the order parameter Q, is included. This potential fits well with the experimental data in the ferroelectric phase. The different relation between ξ and E obtained on heating and on cooling runs is discussed and it is concluded that data on heating correspond to the thermal equilibrium.

We have undertaken a new study of the magnetic and crystalline structures of YMnO₃ and HoMnO₃ using neutron diffraction combined with neutron polarimetry. It is shown how the long-standing problem of distinguishing between magnetic structure models which give nearly identical diffraction intensities may be resolved. The experiments show that the magnetic structure of YMnO₃ has space group P6₃' (or P6₃) rather than P6₃'cm' (or P6₃cm) with Mn moments of 3.14(3) μ_B inclined at 11(1)° to [10.0] (or [12.0] for P6₃) in the (00.1) plane. For HoMnO₃ the experiments confirm the P6₃'c'm symmetry at 50 K and the spin-rotation transition leading to a low-temperature structure with magnetic space group P6₃'cm'. In this structure both the Ho1 and Ho2 moments can order with spins parallel to [00.1] in ferromagnetic layers, anti-ferromagnetically coupled. The Ho1 and Ho2 moments in neighbouring layers are found to be oppositely oriented. The ordered moments at 2 K are 3.32(8), 4.17(13) and 1.31(11) μ_B on the Mn, Ho1 and Ho2 atoms respectively. No evidence was found, in this zero-field study, for the magnetic transition at ≈5 K reported in the literature (Lottermoser et al 2004 Nature 30 541; Vajk et al 2005 Phys. Rev. Lett. 94 87601). Nuclear structure refinements show that in YMnO₃ at 10 K and in HoMnO₃ at all temperatures below 50 K any deviation of the manganese x parameter from 1/3 is less than 0.0008.

We have studied the structural and electronic properties of As-rich GaAs_1−xN_x and N-rich GaN_1−yAs_y alloys in a large composition range using first-principles methods. We have systematically investigated the effect of the impurity atom configuration near both GaAs and GaN sides of the concentration range on the total energies, lattice constants and bandgaps. The N (As) atoms, replacing substitutionally As (N) atoms in GaAs (GaN), cause the surrounding Ga atoms to relax inwards (outwards), making the Ga–N (Ga–As) bond length about 15% shorter (longer) than the corresponding Ga–As (Ga–N) bond length in GaAs (GaN). The total energies of the relaxed alloy supercells and the bandgaps experience large fluctuations within different configurations and these fluctuations grow stronger if the impurity concentration is increased. Substituting As atoms with N in GaAs induces modifications near the conduction band minimum, while substituting N atoms with As in GaN modifies the states near the valence band maximum. Both lead to bandgap reduction, which is at first rapid but later slows down. The relative size of the fluctuations is much larger in the case of GaAs_1−xN_x alloys. We have also looked into the question of which substitutional site (Ga or N) As occupies in GaN. We find that under Ga-rich conditions arsenic prefers the substitutional N site over the Ga site within a large range of Fermi level values.

Our corrected cohesive energy for the bulk Li is in fact 1.82 eV (in contrast to the value reported in the paper, 0.5 eV) against the experimental value of 1.63 eV. When the binding energies are now calculated after taking the bulk Li energy as the reference energy, the calculated binding energies turn out to be more or less in agreement with the values of Liu and Chan [1]. The authors apologise to Liu and Chen for the unfounded criticism of their work in the paper. None of the results of the above paper depends upon the bulk Li energy and therefore our results for the binding energies presented in the various tables which have been obtained after taking the isolated Li atom energy as the reference energy remain unaffected. Thus all the results for the binding energies, electronic structure, optical absorption and Raman active radial breathing mode frequencies obtained are technically sound and free from error.

The unit ^oC in figures 7(a), 7(b) and 7(c) should be K