Table of contents

The short-circuit coaxial transmission line technique has been used to investigate ferromagnetic resonance in magnetic fluids over the frequency range 0.1-6 GHz. Measurements of the complex susceptibility, chi ( omega )= chi '( omega )-i chi "( omega ), of magnetic fluid samples of magnetite and cobalt particles, of approximate median diameters 10 and 5.8 nm are presented. The presence of loss peaks in the chi "( omega ) components at approximate frequencies of 1.2 and 2.5 GHz, respectively, coupled with the transition of the chi '( omega ) components from positive to negative values at slightly higher frequencies, is indicative of ferromagnetic resonance. Appropriate equations for the calculation of the complex susceptibility are presented.

In this paper, the second-order and fourth-order relativistic chromatic variational functions for all kinds of relativistic chromatic aberrations of a combined electrostatic-magnetostatic focusing-deflection system have been derived in detail. Here, the relativistic chromatic aberrations produced by the variations of accelerating voltage, currents of magnetostatic round lenses, voltages of electrostatic round lenses and excitations of deflector multipoles are considered.

Passive Q-switching of Er:glass and Ho:YLF lasers utilizing the saturable absorption properties of multiple quantum well (MQW) structures has been demonstrated. Pulses as short as 15 ns with energies of a few mJ were obtained. The limits imposed by optical damage and device quality were investigated. The multi-shot damage intensity was measured to be approximately 25 MW cm^-2 for 50 ns pulses, which is more than three orders of magnitude greater than the saturation intensity. The MQW samples were observed to cause beam distortion though the cause of this has not been isolated. This limited the optimization of the experiments.

The effect of a regenerative heat exchanger in a gas turbine is analysed using a regenerative Brayton cycle model, where all fluid friction losses in the compressor and turbine are quantified by an isentropic efficiency term and all global irreversibilities in the heat exchanger are taken into account by means of an effective efficiency. This analysis, which generalizes that reported by Gordon and Huleihil for a simple, nonregenerative Brayton cycle, provides a theoretical tool for the selection of optimal operating conditions in a regenerative gas turbine. For a fixed ratio of the compressor inlet temperature to the turbine inlet temperature and in terms of the isentropic and regenerator efficiencies we present explicit results for the behaviour of the maximum efficiency, efficiency at maximum power, maximum power and power at maximum efficiency as well as for the behaviour of the pressure ratios required for maximum efficiency and maximum power.

Using the theory of finite-time thermodynamics, an optimal performance analysis of an internally and externally irreversible Carnot-like single heat engine was studied to determine the maximum power and the efficiency at the maximum power output. Based on this study, the maximum power and the efficiency at the maximum power output of an irreversible combined heat engine (two single irreversible heat engines in a cascade) were obtained for steady state operation. The optimal performance of the combined cycle was expressed in terms of two design parameters. The effects of these parameters on the performance of the combined cycle were examined. The upper and lower bounds to the maximum power and the efficiency at the maximum power were determined. These results were compared with those obtained for a single cycle. It was shown that it is possible to design a combined engine which generates more power and is more efficient than a single cycle.

Similarity parameters and scaling laws are derived appropriate to an arc burning in a strongly accelerating gas flow. The flow is assumed to be laminar and transport of energy by radiation is neglected. The freedom available in selecting the normalization scheme is used to reflect observations from experiment and computer simulations of are behaviour. The similarity theory developed relates arcs which share the same values of two similarity parameters in affinely related nozzles. The theory is verified using numerical calculations of an arc burning in the supersonic nozzle of a gas-blast circuit-breaker.

We have studied the fluid permeability and the electrical conduction in artificially made impregnated sandstones in order to determine the surface conductivity approximately. The electrical conductivity of the impregnated porous medium is not proportional to the conductivity of the impregnation liquid, but a linear dependence is found which permits us to calculate an electrical formation factor, F, and an intercept with the ordinate axis. We propose that 'Archie's law' for the formation factor's porosity-dependence is composed of two factors: a factor inversely proportional to the porosity and a tortuosity factor. The tortuosity factor is dependent on the grain size distribution of the sandstone. The formation factor has been shown to relate to the permeability, k, in two ways: either as k alpha F^-1 or as k alpha F^-2, where the proportionality factors are different geometrical properties of the porous medium. We also show that it is possible to calculate the surface conductivity from a theoretical relation between the conductivity intercept and the permeability.

Values of the ratio D_T/K, the transverse diffusion coefficient to mobility, for Rb⁺ ions drifting in nitrogen and carbon dioxide gases are experimentally measured for the first time at E/N ranging from 5 to 600 Td and from 5 to 800 Td respectively. Results were obtained with estimated accuracies better than 3% at about 303 K. The drift tube system which is capable of mass identification of the ions and of variation of drift distance enhances the reliability of the experimental results. In addition, calculations of D_T/K are made using the generalized Einstein relation technique based on the three-temperature theory.

The paper makes use of the solution of the system of Saha equations derived for any mixture of ionized monatomic gases. This solution is specifically applied to the case of two-component mixtures of the gases used as shields in the process of laser welding, namely Ar plus He, Ar plus O₂, Ar plus N₂ and Ar plus H₂. The number densities of electrons and ions, the degrees of ionization of the plasma as well as the refractive indices and the inverse Bremsstrahlung absorption coefficients are presented as functions of temperature for a variety of volume ratios of the gases in the mixtures. The properties of the mixture of helium and argon change dramatically as the volume ratio of the two gases is altered. This mixture is also most interesting insofar as laser welding is concerned because it causes only very slight defocusing of the laser light above the keyhole.

The electrical conductivity of SF₆-Cu and Ar-Cu plasmas has been calculated by using new values of the cross section of the electron-copper momentum transfer. The comparison between the new values of the electrical conductivity and the previous ones, in the temperature range 2500-15000 K, shows the actual influence of copper vapour on this transport coefficient and allows quantification of the errors made with the various approximations.

Time-resolved laser-induced fluorescence (LIF) of He(2³S) and emission spectroscopy of the N₂⁺(B) first negative system are used in a parallel-plates RF post-discharge of He-N₂/O₂ to study the Penning ionization mechanism in this device. The LIF technique and a theoretical model for its interpretation are discussed in detail. Langmuir probe measurements are also presented to corroborate and enrich the experimental data. The main results are: (i) most of the N₂⁺(B) excitation is due to He(2³S) Penning ionization; (ii) in pure helium and mixtures of low N₂ content the He(2³S)/He density ratio is of the order of 10^-5; (iii) it is very likely that, at low N₂ content, the Penning ionization contributes significantly to the discharge ionization balance; and (iv) a measurement of the rate coefficients for quenching of He(2³S) by N₂ and O₂ Penning ionization is given, that agrees very well with literature data.

Radial profiles of gas density have been measured in positive and negative SF₆ coronas at 101.3 kPa using optical interferometry. The results show that, in DC SF₆ coronas, the electrical power input is confined to a small volume around the point electrode and energy transfer by the corona wind efficiently dissipates the heat generated there throughout the discharge chamber. Consequently, temperature rises and neutral species density changes throughout the chamber are relatively small. The gas density profiles reveal a thin channel of reduced density extending well out into the discharge gap towards the plane electrode and regions of increased density in the outer regions of the discharge chamber. This is consistent with the idea of a corona-driven circulation of gas around the discharge chamber.

We present a simple analytical derivation of the position of the field reversal in the negative glow of a dc glow discharge. In the absence of charged particle losses, this position is shown to depend only on the relaxation length of the energetic electrons entering the glow, the sheath length and the gap length. The model shows that the position of the field reversal, if it exists, can be deep in the negative glow plasma; it is not necessarily at the sheath-negative glow boundary. A large part of the total number of ions created in the negative glow can therefore enter the sheath and return to the cathode. These model results are independent of the electron temperature in the glow.

Measured temperature profiles for various oxide-tungsten cathodes and for pure tungsten cathodes are presented for high-current arcs burning in argon at atmospheric pressure. Temperature profiles are also presented for thoriated tungsten cathodes with different cathode cone angles, are currents and composition of the gas provided to the arc. Evidence is also presented that the temperature and the behaviour of the cathode are sensitive to the oxygen concentration in the argon.

The decaying behaviour of the axial temperature distribution in the post-arc channel after current zero is measured around the nozzle throat in a flat-type SF₆ gas-blast quenching chamber. In the quenching chamber, iron is intentionally used as the electrode material. Two iron atom spectral lines at wavelengths 426 and 443 nm are detectable even up to 100 mu s after current zero for an arc with peak value of 5 kA. This is because the spectral lines for iron have higher radiation intensities than those for other species such as sulphur ions or copper atoms. From the radiation intensities measured at each of the five axial positions around the nozzle throat, the temperatures are estimated by means of the two-line method. The axial temperature distribution shows a local maximum at the nozzle throat up to a time of 20 mu s after current zero. However, the temperature at the nozzle throat decays more rapidly than at the other axial positions. At 100 mu s after current zero, the temperature at the nozzle throat reaches 3500 K, the lowest magnitude among the axial positions.

The growth behaviour of a Y123 crystal was investigated in a melt-textured Y_0.75La_0.25Ba₂Cu₃O_7-y specimen in which La had been substituted for Y. A chi -shaped pattern, which is likely to be related to the crystal orientation, was observed. The concentration of Y and La atoms varied gradually from the centre to the surface of the grains. The growth pattern of a La-doped Y123 crystal, which is very similar to the annual rings of a tree, was explained by a solubility difference between Y and La atoms in a melt that in turn led to the different formation temperature of a La-doped Y123 crystal.

The phase diagram of the water-AOT-undecane system (AOT is sodium bis(2-ethylhexyl) sulphosuccinate), which is a ternary microemulsion, was determined in the P- phi plane (P is pressure and phi is the volume fraction of dispersed matter). The microemulsions were observed at T=20 degrees C with the molar ratio n=(water)/(AOT)=30, over the pressure interval 1-40 MPa. The electrical conductivity sigma was used to trace the percolation threshold line in the single-phase zone of the P- phi plane. The extension to pressures of a method which we had already used in the T- phi plane at atmospheric pressure made it possible to evaluate quantitatively the variations of interactions with pressure and the range of interactions. With this model it is possible to adjust very satisfactorily both the percolation threshold line and the cloud-points curve in the P- phi , plane, despite the approximations made to determine the interactions.

A new amphiphilic compound NMOB (2-nitro-5-(N-methyl-N-decosyl) aminobenzoic acid) with both good film forming property and high optical nonlinearity was synthesized for applications of Langmuir-Blodgett (LB) films in waveguide-type second harmonic generation (SHG) devices. The near-quadratic dependence of the SHG intensity on the bilayer number, the small angle X-ray diffraction (SAXD) patterns and the temporal stability of alternating CdA (cadmium arachidate)/NMOB multilayers confirmed their good film structures. We obtained the hyperpolarizability beta of NMOB of 6.3*10^-50 C m³ V^-2 (1.7*10^-29 esu) at a fundamental wavelength of 0.064 mu m from transmitted SHG measurements which agreed with those deduced from our linear electro-optic (EO) modulated attenuated total reflection (ATR) spectra at 832 nm, 633 nm and 488 nm in order of magnitude. A resonant enhancement of the EO response near the absorption band in NMOB monolayers was observed for the first time.

The capacitance-voltage (C-V) characteristics of multiple-quantum-well (MQW) semiconductor heterostructures are studied using a proposed numerical model. The MQW structures with a Schottky emitter contact exhibit stepwise C-V characteristics, with constant jumping of the inverse capacitance and with the distance between steps increasing with applied bias. The structures with a tunnelling emitter barrier show a strong variation of capacitance, starting from large values, corresponding to the emitter barrier width at low bias, and saturating at low values, corresponding to the length of the NQW structure at high bias. These features of the C-V characteristics are associated with the re-charging of the quantum wells with applied voltage. In turn, the re-charging effects are closely related to the electron injection and transport properties of the MQW structures.

The rectifying barrier of poly(dithienothiophene-dithienopyrrole) (pDTDP) conducting copolymer deposited on n-Si has been investigated. Current-voltage (I-V) and capacitance-voltage (C-V) measurements were used to explore the junction properties. The results presented can be understood in terms of the thermionic emission and space charge limited current model. The pDTDP-Si contact barrier has a height of 0.21 eV from I-V measurements and 0.23 eV from C-V measurements.

We have studied the interference pattern of multiple reflections from a uniaxial, optically active and non-absorbing plane parallel plate under normal incidence. Due to birefringence, a beam of polarized light splits into an ordinary and an extraordinary component. Because of optical activity, the number of beams is doubled after each reflection. During crossing of the plate the phase difference between beams is increased by fixed amounts. The calculated results show that the interference pattern of the transmitted light is more complex than that of an isotropic parallel plate. It consists of strong fringes at large phase intervals and weak doublets in between, at nearly the same phase intervals as in the isotropic plate.

An electroluminescent diode utilizing molecularly doped poly(3-alkylthiophene) has been fabricated and its novel characteristics have been clarified. A remarkable enhancement of the electroluminescence intensity but suppression of the current have both been observed for a diode using poly(3-alkylthiophene) doped with oxadiazole derivative (PBD). The possible origins of these anomalous phenomena are discussed.

An efficient method is proposed to analyse numerically thermoluminescence (TL) glow curves. The new method reduces the number of variable parameters in the rate equations by applying integral calculus to glow curves available for the analysis, a boundary condition, and a maximum condition. This allows for the determination of the best-fit values for all trap parameters, based on a three-level model, when a single pure TL glow curve is given.

Although there are some obvious variances in both electrical and optical characteristics describing the frequency dependence in ACTFEL devices, a method for the quantitative analysis of frequency characteristics is still lacking. This paper presents a method for including the frequency dependence. We assume that there is a phase shift between the applied voltage and the equivalent current emitted from the phosphor-insulator interface. Combining the continuity equations and the tunnel current formula, we have obtained the voltage in the phosphor layer, the phase shift between the applied voltage and the tunnel current, the average current and the total amount of space charge in the phosphor layer at varying applied voltages and at different frequencies. Some physical parameters of the phosphor layer can be obtained by fitting the theoretical model proposed to the experimental results in the range of 500 Hz to 2000 Hz. The results are satisfactory.

The effects of composition and thickness of amorphous thin films of (As₂Se₃)_1-xIn_x (0<or=x<or=0.5) on their optical properties have been investigated. Optical absorption measurements showed that the fundamental absorption edge is a function of glass composition, the optical absorption is due to indirect transition and the energy gap increases with x=0.01 and decreases with x=0.05. On the other hand, the width of the band tail E_e exhibits the opposite behaviour. This behaviour is believed to be associated with the compensation of hole traps as well as the generation of excess electronic localized states. The optical band gap E_opt was found to be almost thickness-independent (100<or=t<or=250 nm). Also, it was observed that the prepared films are almost stable against gamma -doses up to 22 Mrad.

We present data on the NIR/VIS/UV optical constants of thin composite layers, built up from a copper phthalocyanine matrix with embedded metal clusters. The metal clusters (copper, gold and silver) have diameters on a nanometre scale and act as strong absorption centres. In particular, local plasmon resonances and metal interband transitions could be observed from a fit of the experimentally determined dielectric functions by means of a Lorentzian multi-oscillator model. Average dipole transition matrix elements with respect to a single metal cluster have been estimated from the oscillator parameters.

In the energy range 100 eV-2 keV, we applied the Monte Carlo method to analyse the elastic reflection coefficient and the angular distribution of electrons elastically backscattered from the solid surface of an isotropic and homogeneous medium. Results indicated that elastically backscattered electrons arose substantially from only a few scatterings with a single scattering event contributing to approximately half of these electrons. Thus, neither the multiple scattering model nor the single scattering model is sufficient to describe the angular distribution. To improve these models, we evaluated the contribution from one, two and three scatterings exactly and higher scatterings by the P₁-approximation, an approximate method to solve the Boltzmann transport equation assuming multiple elastic scattering of electrons in the solid. This approach allowed us to derive analytical formulations for the elastic reflection coefficient and the angular distribution of elastically backscattered electrons. Results calculated based on these formulations were in good agreement with those using Monte Carlo simulations and experimental data.

The surface properties of lead zirconate titanate (PZT) and its interaction with a metallic overlayer have been investigated. X-ray photoelectron spectroscopy (XPS) was used to study the composition and chemical states of the substrate and adsorbate ions. Results in PZT were compared to those in iso-structural SrTiO₃. It was found that, on polished PZT and SiTiO₃ surfaces at room temperature, a fraction of the first monolayer of evaporated Ni is oxidized to form Ni-O type bonds. On the other hand, if the surface is ion bombarded prior to deposition, the extent of this chemical interaction is strongly reduced in both samples even though there is no evidence of oxygen depletion at the surface. The relationship between this behaviour and sputtering-related defects has been studied by quantifying bombardment related changes to obtain defect profiles. In PZT, there is depletion of Pb²⁺ accompanied by chemical reduction to metallic Pb over a depth of 30 or more monolayers. This species is found to be insensitive to surface adsorbates. In both PZT and SrTiO₃, Ti⁴⁺ ions in the outer two to three monolayers are reduced to Ti³⁺. This defect reacts easily with adatoms (metals, atmospheric contaminants, etc) to change back to its original Ti⁴⁺ state. Since this defective species acts as a source of electrons, and is common to both solids which show similar adsorption behaviour, it must be responsible for reduced oxidation of Ni on the sputtered surface. The scientific significance of these observations and their implication to practical properties have been discussed.

It is shown how the rate of deposition of ionized clusters with a given energy range (from a cluster source) is dependent on the impact area, the energy spread in the beam and the source emittance. A novel technique, SLED (soft landing by electrostatic dispersion), is described for reducing the impact energies whilst still depositing a significant fraction of the total beam.

Single, double and multiple Q-switch Nd:YAG laser pulses are used to vaporize material from solid steel samples and to induce a plasma. The material ablation of different pulse bursts, emission intensities of iron lines and electron temperatures and densities are determined. Material ablation is found to increase with multiple pulses compared to single pulses of fixed total energy as well as electron temperatures and densities. Line intensities can be increased by a factor of about two using double pulses. Quantitative microchemical analysis of low-alloy steel is performed with single and double pulses. The analytical performance is improved by the double-pulse technique.

On the basis of the Brinkman model (BM), the main objective of this paper is to predict theoretically the effects of viscous shear stresses on the dynamical behaviour of pure squeeze films in short porous journal bearings under cyclic loads. The results obtained from the BM, as the value of permeability parameter Phi approaches zero, show good agreement with that of a solid bearing and, hence, provide support to the present prediction. Comparing with that derived by the Darcy model (DM), the influence of viscous shear stresses of the BM on the squeeze-film performance is slight for the thin-walled bearings. But for the thick-walled bearings, the effects of viscous shear stresses provide a significant increase in the load-carrying capacity. Under dynamic loading, the viscous shear effects reduce the eccentricity ratio and, thus, provide a longer time to prevent the journal-bearing contact. Hence, the viscous shear effects of the BM improve the squeeze-film characteristics for a system with thick-walled bearings.

A two-dimensional, time-harmonic eddy-current problem is examined in which a uniform field is perturbed by a long, surface-breaking crack in a non-magnetic, conducting half-space. The crack is assumed to be ideal in the sense that it has infinitesimal opening and yet forms a perfect barrier to the passage of electric current. A solution is sought which is accurate both at high frequencies, at which the skin depth is small compared with the crack depth, and at intermediate frequencies, where the skin depth and crack depth are of similar magnitude. The Wiener-Hopf technique is used to derive a Fredholm integral equation of the second kind for the scattered magnetic field. An approximate closed form solution of this equation is found as a series of exponentially decreasing terms. At lowest order, local solutions at the buried crack edge and at the corners where the crack meets the surface of the conductor are decoupled. Higher order terms in the series account for the coupling which occurs between the fields perturbed by the crack edge and corners.