Table of contents

The torsion giant impedance (TGI) ratio, (ΔZ/Z)_ξ, has been measured in as-cast (Co_0.94Fe_0.06)_72.5B₁₅Si_12.5 amorphous wire and after current annealing under torsion stress up to 33 π rad min^-1 at 450 mA for up to t_ann = 105 min. Strong changes have been induced in the TGI ratio by Joule heating under these conditions. The slightly asymmetric character of the (ΔZ/Z)_ξ dependence on applied torsion ξ with a broad maximum at around ξ = 2.5 π rad m^-1 has been observed in the as-cast state where the maximum value of (ΔZ/Z)_ξ reaches about 155%. Samples treated by Joule heating under applied torsion exhibit a (ΔZ/Z)_ξ dependence showing a tendency to achieve a finally sharp and rather asymmetric shape with a sharp maximum at a certain torsion, ξ_m.

Pre-annealing by Joule heating also introduces some changes in the observed dependences. A maximum (ΔZ/Z)_ξ ratio of 330% is obtained after optimal conditions of current annealing under torsion.

The asymmetric character of the (ΔZ/Z)_ξ in the as-cast state and after torsion annealing could be ascribed to the spontaneous or induced (after torsion annealing) helical magnetic anisotropy, which can be compensated by the application of certain torsion stress. The evolution of (ΔZ/Z)_ξ (ξ) dependences with annealing under torsion reflects the kinetics of the induced magnetic anisotropy in this studied amorphous wire.

Thin (300 nm) porous silicon layers have been formed on p-doped silicon substrate to be characterized by x-ray reflectivity. The objective was to determine the layer thickness, the depth profile of the porosity and the interface roughness as functions of the anodization conditions. A very intense synchrotron beam was used (≈10¹⁹ photons m^-2 s^-1) and under vacuum (≈0.13 Pa) was chosen for the measurements to limit as much as possible the well know phenomenon of oxide film growth on the pore walls during sample exposure to air. Results are reported for two silicon substrates anodized with different current densities and electrolyte compositions. They show that, despite the precaution of making measurements under vacuum, an unexpectedly fast sample evolution took place during irradiation, leading to a dramatic reduction in porosity. A plausible explanation of this phenomenon is reported, and the severe implication in comparing structural data obtained by different characterization techniques is underlined.

For single-crystalline samples of NdGaO₃ and LaAlO₃ the specific heat capacity c_p(T) (temperature ranges 2.24-100 K and 4.35-350 K, respectively) and the thermal conductivity κ(T) (15-300 K) are determined. From an analysis of the specific heat capacity of NdGaO₃ the crystalline electric field contribution and the high-temperature tail from the magnetic ordering are extracted. The thermodynamic standard values for enthalpy and entropy at 298.15 K are calculated: H° = 18.40(18) kJ mol^-1, S° = 121.0(1.2) J mol^-1 K^-1 (NdGaO₃); H° = 14.57(22) kJ mol^-1, S° = 86.6(1.3) J mol^-1 K^-1 (LaAlO₃). Significant contributions to the entropy from the magnetic ordering of the Nd³⁺ ions are included in the calculation. A recently published value of the standard entropy for the related NdAlO₃ is re-evaluated (S° = 105.1(3) J mol^-1 K^-1) in this context. The thermal conductivity of NdGaO₃ at low temperatures is limited by spin-wave scattering, while that of LaAlO₃ in the same temperature range is lessened by the short phonon mean free path, which is due to the presence of twin domains.

The deposition of AgCl in the human tooth dentine was studied as a function of the time and dc electrical field using a scanning electron microscope and by measurements of the frequency-dependent complex dielectric constant. Dielectric properties of the tooth dentine can be well described by the model which was recently developed for the dielectric response of hydrating porous cement paste. The fractal dimension of the tooth dentine was determined by the electron microscopy and dielectric spectroscopy techniques.

This paper reports the results of a study on the thermoluminescent (TL) kinetics parameters of the perovskite KMgF₃ activated by Ce and Er impurities. The kinetic parameters, i.e. the activation energy E the frequency factor s and the kinetic order b have been determined using the initial-rise technique as well as some methods related to the geometrical properties of the shape of a TL peak. The experimental E values given by the various peak shape methods have been checked using a simple acceptance test here introduced.

The applicability of the thermionic model for exoelectron emission was examined on a thin film (0.1 µm) of KCl by measuring the activation energies for exoelectron emission (0.73±0.02 eV) and thermoluminescence (0.64±0.02 eV) and the electron affinity (0.70±0.08 eV) of the solid. The electron affinity was measured through the study of the energy distribution of the secondary electrons. The exoelectron emission was found to be strongly correlated with thermoluminescence. In the temperature interval of -60 to 120 °C, in which the peaks in both phenomena were exhibited, no variation of the value of the electron affinity was observed, while its coefficient of temperature variation was found to be less than 7×10^-4 eV K^-1. The irradiation of the film with 600 eV electrons did not affect the electron affinity, even at the level of 2.5×10¹³ electrons cm^-2, a fluence (total number of particles per unit area) which was found to cause fatigue of the material as far as the two phenomena were concerned.

A dye-sensitized photoelectrochemical cell consisting of a film of SnO₂ crystallites coated with ultrafine particles of Al₂O₃ generates an exceptionally high open-circuit voltage as compared to a cell made only from SnO₂. Al₂O₃ coating on SnO₂ improves the efficiency and the fill factor while delivering reasonably high photocurrents. Photoexcited dye molecules on Al₂O₃ injects electrons into the conduction band of SnO₂ via tunnelling through the Al₂O₃ barrier. Suppression of recombinations of electrons with the dye cations and the acceptors at the electrolytic interface build up the quasi-Fermi level in SnO₂ with an impressive increase of the open-circuit voltage.

In this paper we report on the morphology of InSb quantum dots (QDs) grown by liquid phase epitaxy on InAs and GaAs substrates. Small and large QDs are observed on both these two substrates, the resulting structures being generally similar to those produced by MBE and MOVPE. The density of the small QDs is approximately 10¹⁰ cm^-2, with dimensions of 5-15 nm in height and 0.1-0.3 aspect ratio. The density, size and shape of the small QDs were found to change with growth temperature, supercooling and melt-substrate contact time. The large QDs appear to be at a much lower density, being approximately 2×10⁷ and 1×10⁹ cm^-2 for growth on GaAs and InAs substrates, respectively. Furthermore, it was found that the difference in QD diameters becomes smaller as the lattice mismatch decreases.

An analytical theory has been produced for describing the vector nature of the digital magnetic recording process in thin-film isotropic media. It is based upon assumptions on the way in which the magnetization follows the field during the recording process and shows, except for a scaling of their peak values, that the longitudinal and perpendicular components of magnetization behave, respectively, as if they were separately present in a longitudinal and perpendicular recording. The magnitudes of the longitudinal and perpendicular magnetization components are derived in particular cases.

Structure, magnetic properties, magnetostriction and anisotropy compensation of Tb_1-xDy_x(Fe_0.8Co_0.2)₂ (0⩽x⩽1) have been investigated by means of x-ray diffraction, ac initial susceptibility, SQUID magnetometery, and a standard strain technique. The x-ray diffraction results show that all samples crystallize in a C15 MgCu₂-type compound with a Laves phase cubic structure. The lattice parameter as well as the Curie temperature almost linearly decreases with increasing Dy content. The spin reorientation transition occurs at low temperatures when x<0.4, due to the substitution of Co for Fe. The spontaneous magnetostriction λ₁₁₁ decreases from 2460 ppm for Tb(Fe_0.8Co_0.2)₂ to 1900 ppm for Tb_0.4Dy_0.6(Fe_0.8Co_0.2)₂. When 0.6<x⩽1, the splitting of the (440) peaks almost disappears, which indicates that the composition for anisotropy compensation is shifted to the Tb-rich side compared with Terfenol-D. The polycrystalline magnetostriction exhibits similar behaviour to that of λ₁₁₁. The phase diagram for different spin configurations has been given for Tb_1-xDy_x(Fe_0.8Co_0.2)₂ according to the experimental results.

Several different mixture equations have previously been used to characterize the microwave properties of the ferrite particle-insulating medium mixtures. For very low particle concentrations, these equations give nearly the same results. But when the particle concentration is higher than a few per cent, these equations generally give different results. Based on microwave measurements and numerical calculations, the validity of five well-known mixture equations has been examined in this paper. The experimental results show that both the Bruggeman and QCA-CP (`quasi-crystalline approximation with coherent potential') equations can accurately describe the microwave properties of the ferrite-medium mixtures, and that the Lichtenecker, Logarithm and Maxwell-Garnett equations are not suitable for characterizing the ferrite-medium mixtures over a wide particle concentration range at microwave frequencies. The microwave intrinsic permeability and permittivity spectra of some BaZn_2-xCo_xFe₁₆O₂₇, Ba₄Zn_2-xCo_xFe₃₆O₆₀ and Ba₂Zn_2-xCo_xFe₁₂O₂₂ ferrite particles have been presented, which were calculated from the measurement data of the ferrite-wax mixtures using the Bruggeman equation. The microwave properties of these ferrite particles have also been discussed. In the 1-6 GHz range, both the real and imaginary parts of permeability of BaZn_2-xCo_xFe₁₆O₂₇ and Ba₄Zn_2-xCo_xFe₃₆O₆₀ ferrite particles evidently increase with the increasing Co concentration. Both the real and imaginary parts of permittivity of Ba₂Zn_2-xCo_xFe₁₂O₂₂ ferrite particles apparently decrease as the Co concentration increases.

The deposition of diamond layers from CH₄-H₂ microwave discharge operating in pulsed mode has been achieved. It has been shown that the variation of the time parameters of the process (frequency and duty cycle) leads to noticeable modifications of the deposited layers. From plasma diagnostic measurements, the change in plasma composition has been determined and correlated with the quality and growth rate of the diamond thin films. Particular attention has been paid to the concentration of H-atoms, CH and C₂ radicals and their evolution during the discharge regime and the afterglow. Indeed, these species are well known either as agents for graphite etching (H), or diamond precursors (CH_x imaged by CH) or graphite precursors (C₂H_x imaged by C₂). Optimum values of the power pulse repetition rate (500 Hz) and duty cycle (50%) have been found which are correlated with the variation of the relative concentrations of H, CH and C₂ with time, especially during the afterglow. It has been shown that these optimum conditions correspond to a minimization of C₂ in the afterglow while H and CH concentrations remain high enough to continue the diamond deposition process after the power is switched off.

A hypothesis by Akishev et al (1999 J. Phys. D: Appl. Phys.32 2399-409), that the photoionizing radiation emitted by the positive corona discharge in air is soft x-rays generated by electron bombardment of the anode surface, is discussed. Absorption curves of the photoionizing radiation generated by positive and negative coronae in atmospheric air are compared. The coincidence of the shapes of these curves confirms the earlier, widely accepted opinion that the photoionizing radiation is emitted by gas molecules and atoms, which are excited via electron collision. Photographic detection of this photoionizing radiation performed by the method proposed by Akishev et al is found to be improbable, because of the very high attenuation of radiation by the light protecting envelope around the photographic film.

The replacement of mercury in high-pressure discharge lamps by metallic zinc is studied. Pure zinc/argon discharges as well as lamps including zinc or mercury and metal halide additives are investigated. The studies focus on the physical properties as a function of the discharge geometries and lamp power. Experimental data are compared with model calculations of the energy balance involving the transport of heat and radiation. Since the excitation energies of relevant zinc transitions are lower than for mercury, axis temperatures of zinc lamps are about 300 K below the value of mercury arcs. In addition, the thermal conductivity of zinc, including the contribution of radiation diffusion, is larger than that of mercury. From lamp voltage measurements, it is found that the cross section for elastic electron scattering by zinc atoms is about the same as that for mercury. This fact needs to be verified by measurements of the momentum transfer cross section of zinc due to the lack of experimental data in the low-energy region below 1 eV. When adding metal halides, i.e. NaI, TlI, DyI₃, the axis temperatures decrease to about 5100 K due to strong radiation cooling. In these lamps the coldest spot and the mean wall temperatures must be increased when replacing mercury by zinc in order to obtain sufficiently large electrical fields. Temperatures of about 1350 K are adjusted by means of polycrystalline alumina oxide (Al₂O₃) as a wall material. In addition, the electrode distance of metal halide lamps containing zinc must be slightly increased by 25% in order to obtain lamp voltages of typically 80-90 V, of mercury metal halide lamps. The light technical data of the discharges are very close, since mercury and zinc do not contribute significantly to the radiation in the visible range. Efficacies of up to 93 lm W^-1 and 100 lm W^-1 are found in metal halide lamps with zinc and mercury, respectively. Consequently, zinc turns out to be an attractive replacement for mercury in this type of lamp, and not only from an environmental point of view.

Emission spectroscopy and the recording of electrical characteristics are used to study the influence of the granulometric composition of arc-quenching material. Six adjacent intervals of grain size are studied: each of them is 50 µm wide, and the upper limit of the global interval is lower than 1000 µm. The final dimensions of the fulgurite are found to be proportional to the granulometry. Concerning the electrical parameters, two groups are considered: the extinction time constant, the dissipated energy and the I²t value decrease with the granulometry, whereas the maximum value of the arc voltage increases with the granulometry. The rate of decrease of the temperature depends on the time interval observed: at the beginning of the arcing period, the decrease is fastest for the smallest granulometric parameters; and during the last part of the arcing period the trend is reversed. At the beginning of the arcing period, the electron number density increase varies between 10¹⁹ and 10²⁰ cm^-3 ms^-1. When the electric current decreases, the electron number density decreases exponentially with a time constant value ranging from 0.40 ms to 0.25 ms.

We propose a new discrete stochastic model for computer simulation of the lightning process and the breakdown process in long gaps in gaseous dielectrics. In this model we used cellular automata. Two different states of a conductive structure that correspond to streamers and channels of very high conductivity (leader phase) are introduced. The conductivity of the streamers is assumed to be negligible in comparison with highly conductive channels. The electric-field potential is obtained by solving the Laplace equation in a region outside the equipotential, highly conductive part of the structure. As the streamer growth criteria we use two multi-element models in which several conductive bonds can arise at each time step. The growth is stochastic in time, and the probability of streamer formation is proportional to a certain function of local electric field r(E) that depends on the properties of the dielectric. If the energy released in a segment of the streamer is larger than a certain critical value, the streamer transforms to a highly conductive phase. Patterns of conductive trees are obtained in computer simulations of breakdown under various conditions.

The influence of Ar addition to conventional Plasma Display Panel gases, Xe/Ne and Xe/He mixtures, on the radiation characteristics and the discharge onset voltage is studied using a Boltzmann equation method coupled with rate equations for electrons and excited species. When Ar is added to Xe/Ne mixtures, the efficiency of ultraviolet light radiation from Xe (1s_2,4) and Xe₂* increases 7-17% for E/p₀ between 3 and 100 V cm^-1 Torr^-1. In Xe/He mixtures, the efficiency increases over 20% for E/p₀<10 V cm^-1 Torr^-1. In addition, the visible light radiation of the 600 nm band from excited Ne, which makes the contrast worse, is effectively reduced. On the other hand, the discharge onset voltage increases with Ar addition because of a decrease in the effective secondary electron emission coefficient.

The optical model of vertical-cavity surface-emitting lasers (VCSELs) based on the effective frequency method is applied to formulate some guidelines useful for designers of nitride VCSEL resonators. Various materials for resonator distributed Bragg reflectors mirrors are investigated, namely SiO₂, TiO₂, HfO₂, MgO, Y₂O₃, ZrO₂, GaN, AlN, and AlGaN. For pulse-operating nitride VCSELs, both dielectric resonator mirrors (preferably SiO₂/TiO₂ stacks) are recommended, whereas in the case of continuous-wave operation, the bottom dielectric mirror should be replaced by a semiconducting one (e.g. GaN/Al_0.15Ga_0.85N stack) to enhance heat extraction. It is shown that the active region radius in the analysed lasers should not be less than 5 µm. The simulation also revealed that the barrier width has a strong influence on the threshold gain. Some thermoelectrical aspects of design have also been addressed.

Precise determinations below 200 K have been made of the complex susceptibility of Ti³⁺ and Ti⁴⁺-doped-sapphire crystal resonators using the whispering gallery mode method at microwave frequencies. The observed Ti³⁺ contribution to the frequency-temperature dependence of sapphire was interpreted as arising from the Van Vleck temperature-independent paramagnetic susceptibility. A small anisotropic dielectric contribution resulting from TiO₂ clusters also was observed in the Ti⁴⁺-doped crystal. In both crystal resonators, the compensation point in the frequency-temperature dependence was found to be dependent on the magnetic energy filling factor in the sapphire and the Ti³⁺ concentration. An exponential increase in the Q-factor was observed as the temperature was reduced below 10 K. A mechanism to explain this effect is proposed. Below 50 K, an Orbach double-resonant-phonon process was dominant, with characteristic energies equivalent to temperatures of 54±1 K and 27±1 K. Above 70 K it is suggested that Raman scattering was the major loss mechanism. The paramagnetic susceptibility of the doped crystal and the relaxation rate of the Orbach loss process were uniquely determined.

The analysis of complex peaks in the technique of thermally stimulated depolarization currents (TSDC) often makes use of a distribution of relaxation times. Here we suggest a different approach that emphasizes the manifestation, in the TSDC measurement, of the asymptotic power-law domains of the relaxation function in the frequency measurement. Calculations for the Cole-Cole response function show that the results of a typical TSDC experiment for systems showing time-temperature superposition contain information about the frequency response on both sides of the spectrum. The reason for this is that the equivalent frequency of TSDC measurements is only low in absolute terms, but not with respect to the system's relaxation time, which decreases rapidly during the temperature scan of a TSDC experiment. The analysis of the stretched exponential and Havrilak-Negami functions indicates that the TSDC technique is able to provide an accurate determination of the fractionary exponents of the loss peak.

Interfacial temperatures have been measured using fibre optic inserts in the stationary brake ring during ring-on-ring, subscale dynamometer aircraft brake testing. The brake materials are carbon fibre-reinforced, carbon-matrix (carbon-carbon) composites. The temperature distribution varies with dynamometer test conditions and demonstrates the non-uniformity of contact pressure within the interface. A two-dimensional, axis-symmetric finite element model (FEM) is presented that is used to estimate temperature profiles during braking using either a constant pressure or constant energy flux assumption. The model incorporates the measured temperature-dependence of the thermal diffusivity and specific heat capacity for the composite materials. The surface temperatures obtained from the FEM are compared with the measured surface temperatures. Substantial differences between the two results are observed and discussed.

The temperature profiles of a rectangular thin film heated with a scanning strip heat source are calculated using the Green function. Two boundary conditions are considered. One has both upper and lower surfaces insulated and the other has the upper surface insulated with the lower surface maintained at constant temperature. When the length of thin film along the scanning direction is infinity and temperature reaches steady state, the effect of scanning speed and specimen thickness on the temperature profile is investigated. The result is applied to an A4 paper sheet. Both the temperatures of the upper and lower surfaces are affected by the scanning speed, Gaussian radius and absorption depth. The criterion of paper feed to avoid paper curling is proposed.

Results are presented from investigations of multispark electric discharge in water excited along multielectrode metal-dielectric systems with gas supply into the interelectrode gaps. The intensity distribution of discharge radiation in the region covering the biologically active soft UV (190⩽λ⩽430 nm) has been determined and the absolute number of quanta in this wavelength interval has been measured. The potentiality of the slipping surface discharge in water for its disinfection is analysed. The energy expenditure for water cleansing is estimated to be as low as ~10^-4 kWh l^-1.

The optimal performance of the Feynman ratchet-and-pawl engine is analysed by taking the power and the efficiency of the engine as objective functions. The power-efficiency curves are also obtained. These curves show a loop shape similar to those characteristic of some real heat engines. Explicit analytical expressions are reported in the so-called linear regime.

The viewing of the erosion rate of the fuse element in high breaking capacity fuses is carried out using fast imagery. The rotating drum camera we have used provides up to 160 frames to observe the arc extinction throughout the phenomenon which lasts 4 ms. From these frames, we show that three stages follow each other with different values of burn-back rates: the maximum values are obtained at the beginning of the phenomenon and are equal to 6.65 m s^-1 and 5.81 m s^-1 for silver and copper fuse elements, respectively. The direct observation of the burn-back mechanism shows a reproducible disequilibrium depending on the nature of the electrode: the cathode erosion rate is 1.7 times that of the anode rate in the case of silver, and 1.2 times that of the anode in the case of copper.

Thermoelectric (TE) properties (resistivity and thermopower) of polycrystalline ceramics in the Bi_2-xPb_xSr_3-yY_yCo₂O_9-δ system prepared by a solid-state reaction in air have been investigated as a function of temperature up to about 300 K, changing the combination of x and y. These oxides are chemically very stable, and TE properties and the lattice structures do not change even after they are heated at high temperatures. Y doping has a profound effect in reducing resistivity. When y (the amount of Y) is small, ligand holes created by O 2p-Co 3d hybridization are mainly responsible for insulating conduction. With increasing y, resistivity decreases rapidly and insulating conduction transfers to metallic conduction. This is due to the shrinkage in the lattice constant by the Y doping, which leads to band crossing of the O 2p and Co 3d levels. Measurements of TE properties including thermal conductivity have been carried out on the material with x = y = 0.5 in the high-temperature range up to 800 K. The TE figure of merit increases from 2×10^-5 K^-1 to 7×10^-5 K^-1 as the temperature increases from 400 K to 800 K, which indicates that this compound has high potential as a good TE material at high temperatures.