Table of contents

The magnetization of platinum-cobalt multilayers has been investigated using a combination of vibrating sample magnetometry and polarized neutron reflectometry. The enhanced magnetization due to spin polarization of the platinum atoms effected by the neighbouring cobalt atoms is evaluated for films with cobalt layer thicknesses in the range 6-12 AA, the range over which there is a transition from perpendicular to in-plane anisotropy. We have observed increased moments above the bulk cobalt value for the set of films.

Permeability due to domain wall motion and spin rotation is studied on CoZr and CoZrRe thin films presenting different distributions of anisotropy. The influence of the amplitude and frequency of the exciting field on the two magnetization processes, domain wall motion and spin rotation, is studied, showing the existence of a critical field for domain wall depinning and a domain wall relaxation. We show that it is possible to separate and evaluate the contribution of the two magnetization processes to the permeability and analyse the influence of the anisotropy dispersion on the frequency spectra of the permeability.

The field-dependence of magnetization, Young's modulus and magnetostriction have been used to investigate the domain structure of FeSiB amorphous wires. The experimental observations of the as-cast wire are explained in terms of a modified core-shell model in which the shell is magnetized in a series of domains oriented approximately perpendicular to the wire axis and the magnetization in the core is oriented at an average angle of about 44 degrees to the wire axis. The presence of reverse spike domains in the core at the ends of the wire is also suggested. The critical length below which magnetic bistability is lost may be associated with these reverse domains. The behaviour of wires annealed at 425 degrees C can be interpreted in terms of a similar model, except that the core volume and the length of the reverse spike domains are increased and the average magnetic moment angle in the core is decreased.

Fine tip structures are required for many applications in different technologies, for example in scanning probe microscopy or vacuum microelectronic devices. Tips can be produced with nanometre precision by the electron-beam-induced deposition technique. The deposition process, however, is still not really understood. This paper describes the results of a Monte Carlo simulation of electron scattering in tip structures, which reveal some fundamental differences from flat and tilted bulk substrates. An analytical expression is found for the total energy loss of electrons in tips consisting of various materials deposited at 8-30 keV beam energy.

A pulsed self-collimated intense electron beam of diameter less than 1 mm, a few centimetres in length and up to 280 A peak current, 20 ns pulse duration and 50 Hz repetition rate is produced along the axis of a cylindrical discharge tube using the synergy of two discharges. It needs neither bore holes nor internal electrodes and can be obtained in various gases and for a large range of tube diameters.

For weak scatterers, simple relations have been discovered between physical properties and the Fourier transforms of their scattering fields, but they are invalid under strong-scattering conditions. In this paper, a microwave networking technique is used to investigate the one-dimensional inverse scattering problem, from which a novel inverse scattering solution for the permittivity profile is derived by considering the equivalent microwave network of the permittivity profile discontinuity at the interface of free space with the medium. This novel solution is essentially different from previous relations and is valid for both weak-scattering and strong-scattering conditions. Closed-form and numerical reconstruction examples show that the novel solution is accurate in comparison with exact theory and is applicable to both continuous and discontinuous permittivity profiles.

The fast Fourier transform spectrum of acoustic emission during CO₂ laser welding of Al 1100 shows frequency components in the 3-9 kHz range that can be identified with the presence of a keyhole and correlate with penetration. In addition, a study of the effect of anodization and surface pretreatment of Al 1100 with excimer laser radiation has shown that acoustic emission at 9-10 kHz arises from burning off of surface oxide. A comparison of these results with those predicted from simple thermal and fluid dynamical models yields good agreement with theory. These results indicates that acoustic emission over specific frequency ranges may be highly diagnostic of laser processing conditions.

Planar waveguides were produced by K⁺-Cs⁺ exchange of Z-Cut KTiOPO₄ single domain crystals. The analysis of the refractive index profile has shown an exponential Cs⁺ distribution in the exchanged layers. By exciting different waveguide modes, Raman spectra were obtained from guide depths with varying Cs doping. Significant changes have been observed between Raman spectra of Cs:KTiOPO₄ and those of pure KTiOPO₄ only in the region of external vibrations while the internal modes, due to vibrations of the TiO₆ and PO₄ groups, are practically the same. The bands corresponding to the external vibrations were found to be strongly influenced by the substitution indicating strong disorder in the lattice, and possibly a structural phase transition. Furthermore, Raman spectroscopy proves to be an alternative technique for determining variation in Cs composition.

The linear stability of two dielectric inviscid fluids separated by a cylindrical interface is investigated. The interface allows mass and heat transfer. The system is stressed by a radial periodic electric field that allows the presence of surface charges at the interface. The standard normal-modes approach is utilized. A general dispersion equation valid for all (axisymmetric and non-axisymmetric) modes of disturbances is derived. Some previous studies are compared using appropriate data. Parametric excitation of electrohydrodynamic surface waves is obtained in the case of Rayleigh-Taylor instability. The transition curves are obtained by means of Whittaker's technique and the results are confirmed numerically. Contrary to the case of uniform electric field, the periodic one frequently shows the effect of mass and heat transfer on the conditions of stability. The presence of surface charges made the radial electric field play a dual role in the stability criterion, which shows some analogy to the nonlinear theory of stability.

An effective method for calculating the electron energy distribution function in a plasma pumped by an electron beam is suggested. The electron impact excitation rates of different N₂ and Hg states are obtained. The evolution of populations of the N₂ excited states and of the Hg radiative states in a N₂-Hg mixture is considered. The characteristic time tau _e of equilibrium establishment is shown to be strongly dependent on the pressure of the mixture. The variation of the N₂(A³ Sigma ) population is found to be determined by the dynamics of the Hg emissions if the measurement time is longer than tau _e.

The evolution of the argon metastable states density has been studied by absorption spectroscopy in power-modulated plasmas of argon and a mixture of 4% silane in argon. A small concentration of silane suppresses the argon metastable states density by molecular quenching. This molecular quenching adds to the electronic collisional dissociation to increase the silane dissociation rate as compared with pure silane plasmas. Using time-resolved emission spectroscopy, the role of metastable states in excitation to the argon 2P₂ state has been determined in comparison with production from the ground state. In silane plasmas, emission from SiH* is due essentially to electron impact dissociation of silane, whereas in 4% silane-in-argon plasmas, emission from SiH* seems to be due to electron impact excitation of the SiH ground state. These studies demonstrate that argon is not simply a buffer gas but has an influence on the dissociation rate in plasma-assisted deposition of amorphous silicon using argon-diluted silane plasmas.

We investigate a source of H atoms generated by a low-pressure surface wave discharge (2.45 GHz). We study the influence of microwave power both on the discharge characteristics on the H atom density, which has been measured by actinometry. Dissociation levels of H₂ are much higher (75%) at low microwave power than at high power (10%). Unlike what has been found in oxygen surface wave plasmas, discharge characteristics depend strongly on microwave power, due to an important coupling between discharge equilibrium and kinetics of the atomic hydrogen. These results are explained taking into account the effect of discharge tube wall temperature on atomic recombination. The wall recombination probability gamma is estimated as a function of the microwave power: it ranges from 6*10^-3 to 6*10^-2, which is very high in comparison with values determined previously under post-discharge conditions.

A study of spatial and temporal dependences of plasma-relevant parameters has been done in the frame of a general quasi-one-dimensional model for ablative capillary discharges and for a specific time-dependence of the input energy. Particular attention was given to the effect of the so-called 'second boost of input energy' that leads to a substantial increase of some of the main plasma parameters such as exit velocity (26%), mass flux (177%) and energy flux (exit power) (285%). Calculations and experiments are in good agreement. The reliabilities of the model and numerical approach are supported by energy conservation condition fulfilment and the critical value of the exit Mach number.

The main forces acting on the molten tip of an arc electrode are surface tension and electromagnetic force. Both of these forces depend on the shape of the droplet. Conversely, the shape of the droplet is determined by these forces. The equation to describe the shape of the droplet pending at the arc electrode is proposed and solved. This allows calculating the size of the detaching droplet significantly more accurately and over a wider current range compared to both existing models (stable force balance theory and pinch instability theory). Furthermore, the equation describes the main changes of the metal transfer as the arc current increases: slow decrease of the droplet size at low currents, sharp decrease at higher currents, and absence of the stable droplet above a certain current value. For higher currents a simple non-stationary model of the droplet development is proposed.

This work was performed to screen potential cathode materials for use in arc heaters with cylindrical electrodes. It focused on hot isostatic pressing consolidated copper composites; 10% composites of copper with tungsten and niobium as well as 1.5% niobium were used. The composite cathodes were tested in pure Ar, He and N₂ as well as in mixtures of Ar with CO, N₂ and H₂, at atmospheric pressure and currents up to 150 A. The arc was driven around the circumference of the cathode with an external magnetic field. The results are compared with those using pure copper cathodes. The Cu-W cathode showed a better arc stability and lower erosion rates than copper. The Cu-Nb cathode showed much lower erosion rates and much better arc stability than copper for almost all the different operating conditions and may be considered a promising new electrode material. The experimental results corroborated previous interpretations of the relationships between arc velocity, erosion rate, surface drag and electrode surface composition.

Radial direct current electric fields were applied to a cylindrical alumina (94% Al₂O₃) insulator, which was operated under ultra-high vacuum (pressure <10^-8 mbar), with current magnitudes limited to pre-breakdown values of less than 50 mu A. A high-resolution monochrome charge-coupled device camera was used to capture images of photon emission from the alumina; the spectrum of the emitted light was quantified using a high-speed, high-resolution scanning monochromator with a photo-multiplier tube detector. Optical images of the observed luminescence are presented, together with wavelength spectra recorded at fixed values of direct current bias voltage. An observed time-dependence of the photon emission intensity is correlated with an apparent time-variation in the direct current-voltage characteristic of the bridged alumina insulator. The features of the recorded spectra are discussed in terms of solid-state photon-emission processes, arising from radiative electron recombination at impurity and structural defect centres within the surface layers of the alumina ceramic.

The longitudinal structure of a gamma -type RF discharge is computed and shown to have much in common with the structure of a direct current glow discharge. In a sufficiently long gap or at rather high pressures, there are regions of cathode fall, negative glow and Faraday dark space near each of the electrodes. A uniform positive column exists between the electrodes. This pattern is obtained when taking into account non-local effects of the electron spectrum and abandoning Townsend's coefficient to describe the ionization, electron diffusion and electron losses as functions of a local field. The reverse field regions obtained numerically agree with experimental values. It is shown that very low magnitude of the field amplitude and mean electron energy in the plasma are not intrinsic properties of the gamma discharge. The low magnitudes are typical only for sufficiently short gaps and low pressures when the plasma is concentrated in the negative glow region and the Faraday dark space and there is no room for the positive column to form. In long gaps and for high pressure the positive column is formed where the field and electron temperature are as high as in an alpha discharge or in the positive column of a glow discharge.

The electron energy distribution function (EEDF) is measured with a Langmuir probe in a symmetrically capacitively-coupled RF (13.56 MHz) helium discharge over a range of gas pressures from 0.1 to 1.0 Torr while keeping the discharge current constant at 100 mA. The bulk electrons are Maxwellian over this range, while the average electron energy decreases with increasing pressure, together with an increase in electron density. The power dissipation calculated from the EEDF and the discharge current-voltage relation agrees with that measured.

The influence of excited (metastable) states of helium on the ionization and excitation processes in surface-wave sustained plasmas is studied experimentally as well as theoretically. The experimental studies are performed employing emission and absorption spectroscopy methods. The theoretical investigations are based on a collisional-radiative model coupled to self-consistent electron-energy distribution functions. The influence of the stepwise ionization on the axial distribution of the electric-field strength is discussed. Also the impact of the radial distribution of excited atoms is taken into account. Reasonable agreement between measured and calculated population densities of the 2³S level is found. The axial changes of the spectral line intensity ratios of a helium singlet line (4¹S-2¹P) to a triplet line (4³S-2³P) are found experimentally. By comparison with the theoretical model it is shown that these modifications can (to a great extent) be attributed to the changing contribution of stepwise excitation of the triplet level.

Electron drift velocities W_SC and W_FD obtained respectively from Schlumbohm's and Frommhold's experiments are calculated by a Monte Carlo simulation for five cases of model gases. The obtained drift velocities are compared with the theoretical drift velocities defined as the average velocity V_d of the steady state Townsend condition and the average velocity W_v, the centre of mass drift velocity W_r and the mean arrival time drift velocity W_m of an isolated electron swarm, which are calculated by a separate Monte Carlo simulation in a free space for the respective model gases. It is found that W_SC is almost equal to W_m, and that W_FD is almost equal to drift velocity W_w represented as ¹/₂ (W_r+W_m), for the E/N values studied here. It is also found that the difference between electron drift velocities depends not only on the E/N value but also on the value of the effective ionization frequency.

At low pressures, diffusion plays an important role in determining the number density n(r) of metastable particles created in an electrical discharge. The appropriate surface boundary condition for solving the metastable diffusion equation is known to be of the form n(r) varies as s. Del n, where s. is the unit vector normal to the surface. In this paper a new expression for proportionality constant is derived using the arguments of gas kinetics. This expression is used to estimate the reflectivities of a variety of types of metastable particle.

Measurements of the temperature of a 200 A free-burning arc in nitrogen, performed using a laser-scattering technique that does not require frequency resolution of the scattered signal, are presented. The technique is applicable to the full temperature range observed in free-burning arcs. However, signal-to-noise problems limited the present measurements to temperatures below about 16000 K. The laser-scattering results show excellent agreement with temperatures measured by emission spectroscopy. The spectroscopic measurements are shown not to be affected by deviations from local thermodynamic equilibrium at temperatures as low as 8000 K, below which the emission was too weak to allow temperatures to be determined. This is contrasted with the case of free-burning arcs in argon, for which departures from local thermodynamic equilibrium, thought to be caused mainly by resonance radiation trapping, render spectroscopic measurements unreliable below about 11000 K. An explanation for the differences between arcs in the two gases, in terms of different rates of excitation of high-energy states by resonance radiation trapping, is suggested.

Films of a new form of carbon allotrope, C₆₀, also known as fullerenes are deposited on Si(111) substrates by the ionized cluster beam deposition technique under an accelerating field less than 100 V. Raman spectra and X-ray photoelectron spectroscopy measurements are carried out to analyse the electronic properties of the films, to indicate the existence of C₆₀ soccer-balls in the films. The resistance of this C₆₀ film deposited here to oxygen contamination is better than that deposited by molecular beam epitaxy. Binding energies of C 1s peaks for C₆₀ and highly oriented pyrolytic graphite are 284.7 and 284.3 eV, respectively. X-ray theta -2 theta diffraction investigations show that C₆₀ films deposited under V_a=0 V have highly textured close-packed structure with X-ray diffraction assignment (110), while those deposited under V_a=65 V turn out to be more polycrystalline. C₆₀ soccer-balls are found to be broken into fragments as accelerating field exceeds about 400 V, indicated by the results of X-ray photoelectron spectroscopy, Raman spectra, X-ray diffraction, and ultraviolet visible absorption spectra.

The problem of capacitance between a rough sphere and a rough plane very close to each other is studied. The problem is simplified by gathering the roughness of both members onto the plane. We begin by considering the case where the sphere and the plane are both ideally smooth. Starting from a known exact formula of the capacitance involving an infinite sum, we derive a simple and easy to use approximate expression. Then we consider the case where the sphere is smooth and the plane rough. The new formula for the capacitance involves an integral in which the weighting function is the height distribution. To validate this formula, the results of two experiments, during which the derivative of the capacitance with respect to the separation has been recorded, are considered. Comparison of the measurements with theoretical calculations indicates the approach is valid.

Direct current resistivity and the Seebeck coefficient ( alpha ) have been measured in the temperature range 300-800 K for compositions x<or=0.40 in the Sr_1-xLa_xTi_1-xCo_xO₃ system. Alternating current conductivity has been measured as a function of frequency and temperature. Positive values of alpha over the entire range of measurement show that holes are the majority charge carriers. Alternating current conductivity shows that conduction occurs due to hopping of charge carriers among localized sites.

We propose an electron wavesplitter, which employs resonant tunnelling through a quantum well between electron waveguides to transfer part of the electrons in one waveguide to another. The device consists of one input and two output waveguides coupled through a symmetric barrier-well-barrier system. Maximum resonant tunnelling of electrons through the well is achieved by adjusting the direct current voltage across the well so as to align the incident energy (< the barrier height) with the energy level in the well. The mode-matching method is adopted in the two-dimensional numerical calculations. The results show that transmission from one arm to another reaches sharp peaks (half-width<4*10^-4 eV) at certain energies.

ZnO and ZnO:Li films (1-3 mu m thickness) were deposited on a soda lime glass substrate by a spray pyrolysis technique using ZnCl₂ and LiCl as starting materials. Physical properties of these films were studied as a function of substrate temperature T_s and lithium concentration. The best ZnO film was obtained for T_s=400 degrees C where it exhibits a conductivity of 6.3*10^-3 Omega ^-1 cm^-1 and a mobility 104 cm² V^-1 s^-1. X-ray and EDAX studies confirm that it is possible to obtain good transparent conducting ZnO films preferentially oriented along the c axis. When doped with lithium, amorphous films are obtained.

Studies on polarization and depolarization charges have been performed for triglycine sulphate single crystals. The measurements have been taken at temperatures in the range between liquid nitrogen temperature and 85 degrees C and for voltages in the range 1-1000 V. The polarization time has been varied in the range 10^-2-10⁵ s. A strong relation between the above parameters and values of the measured polarization and depolarization charge has been found. The activation character of the dependence between the (de)polarization charge and the temperature has been confirmed. For weak electric fields (7 and 70 V cm^-1) and temperatures in the range -193 degrees C to 10 degrees C, (de)polarization charges increase with increasing temperature (activation energy 0.01 eV), whereas for high field (3.3 kV cm^-1) depolarization charges decrease (activation energy 0.013 eV). In the temperature range from 10 degrees C to the Curie point (49 degrees C) (de)polarization charges increase with increasing temperature and activation energy equals 0.68 and 0.31 eV for polarization fields 7 V cm^-1 and 3.3 kV cm^-1 respectively. Depolarization charges have a local minimum appearing for some polarizing field, which value depends on the temperature. No influence of ageing time on depolarization charge has been observed, while variation of ageing time caused changes in polarization charge values.

A series of GaAs/Al_xGa_1-xAs MBE structures, in which x ranged from 0.10 to 0.40, was studied using microscope spectrophotometry (MSP). The MSP reflectance measurements were found to be extremely sensitive to compositional variation, with a clear inverse relationship between Al content and reflected intensity, and a sympathetic relationship between composition and dispersion. Theoretical optical models were simulated and fitted to the measured data to give layer dimensions which fitted remarkably well with SIMS data. It is concluded that MSP has great potential for the non-destructive metrology and quality control of simple and complex multilayer structures.

Some empirically founded expressions are established that aim to describe the backscattering coefficient and the energy distributions of electrons backscattered from a bulk solid target (atomic number Z>or=4) within the incident energy region 0.5<or=E₀<or=100 keV, and for any incidence angle 0<or= Theta <or=60 degrees . Comparisons are made with measurements in order to check the applicability of the proposed formalisms.

ZnSe_xTe_1-x films were prepared by co-evaporating ZnSe and ZnTe powders from a two-zone hot wall evaporation jig onto glass substrates. The optical band gaps for different x were determined and this showed a bowing behaviour. The refractive indices and extinction coefficients have been determined as a function of wavelength. Variations of surface roughness with composition and microstructural details were also reported. Grain boundary scattering effects were found to be a dominant factor controlling electron transport processes in these films.

The preparation of epitaxial thin films of Pb(Zr_xTi_1-x)O₃(0.75, 0.25), Pb(Zr_xTi_1-x)O₃(0.52, 0.48) Pb_1-xLa_x(Zr_yTi_z)_1-(x4)/O₃(0.09, 0.65, 0.35), PbMg₁3/Nb₂3/O₃ by means of pulsed laser ablation on YBa₂Cu₃O_7-x coated (100)SrTiO₃ and (110)SrTiO₃ substrates is reported. The films have been characterized by X-ray diffraction, Rutherford backscattering, scanning electron microscopy and electrical measurements. High-quality ferroelectric films were obtained at substrate temperature of 550, 520 and 550 degrees C for Pb(Zr_xTi₁-_x)O₃, Pb_1-xLa_x(Zr_yTi_z)_1-(x4)/O₃ and PbMg₁3/Nb₂3/O₃, respectively. X-ray diffraction studies of the Pb(Zr_xTi_1-x)O₃/YBa₂Cu₃O_7-x/(100)SrTiO₃ show that the layers grow in the c axis orientation with a rocking angle of 0.4 degrees -0.6 degrees for the YBa₂Cu₃O_7-x and 0.8 degrees -1 degrees for the Pb(Zr_xTi_1-x)O₃ layers. The remanent polarization and coercive field of the Pb(Zr_xTi_1-x)O₃(0.52, 0.48) film were 42 mu C cm^-2 and 55 kV cm^-1.

Thin molecular films of copper iodide have been grown by direct chemical reaction within a molecular beam using laser-assisted molecular beam deposition. Molecular iodine vapour entrained in a helium carrier gas stream was supersonically expanded into the laser-ablated plasma plume of vaporized copper, producing copper iodide. Films of this material were grown on substrates situated in the path of this molecular beam. The oxidation state of copper and the chemical composition of these films were studied by electron spectroscopy for chemical analysis and energy-dispersive X-ray spectroscopy, and, the results were compared with those of a high-purity Cul powder. The binding energy of the Cu 2p_3/2 photoelectron line and the kinetic energy of the X-ray-induced Cu L₃M_4.5M_4.5 Auger line are determined to be the same for both samples. The surface morphology and extent of surface inhomogeneities in copper iodide films were found to be largely dependent on the fluence of the incident laser and the expansion conditions. Secondary electron images of these films indicate clustering of copper iodide on the surface of the films during film growth. The copper in these films is found to be in the Cu⁺ ionization state.

Current leakage is a serious problem in electro-rheological fluids. In this study, the electrical current leaking through water-activated electro-rheological fluids has been investigated as a function of particle volume fraction, water content, electric field strength and temperature by both static and dynamic experiments. The effect of electric field strength on current leakage is presented in terms of a power-law relation between current density and applied field strength. The nonlinear characteristic of the I-V behaviour of water-activated electro-rheological fluids is a consequence of the nonlinearity of the resistance of the fluid. The exponent in the relation between current density and applied electric field strength was found to depend strongly on the water content but only weakly on the particle volume fraction. The electrolysis of water molecules under the applied electric field is proposed to explain this nonlinearity in electrical resistance of the electro-rheological materials. We note that the temperature-dependency of current density through our electro-rheological fluids is similar to that observed for semiconducting materials, that is the conductivity increases with increasing temperature. The Arrhenius-type activation energy of water-activated electro-rheological fluids was also estimated to be about 3 J mol^-1 a similar value to that of pure water. An increase in thermal motion of water in the particles (silica in this case) may be a reason for this behaviour. Furthermore, an equilibrium structure consisting of linear chains of particles with branches has been determined from the power-law relation between electrical current and volume fraction of dispersed particles. The electrical current was also measured as the suspension was being subjected to shear. The current decreased as the shear rate increased, implying that the particle chains, formed by the applied electric field, were destroyed by the shear field.

The spatial variations of the magnitude and phase of the electric field produced by electrodes used in electrorotation studies have been computed, using the charge density method at 10 degrees intervals for one complete AC cycle of the applied voltages. These results, together with the development of a more general theoretical treatment than has hitherto been applied to this problem, enable the time-averaged rotational torque and dielectrophoretic force acting on a particle to be determined as a function of its location within the inter-electrode space. Correction factors that can be applied to electrorotation spectra to take account of the relative position, or change in position, of the test particle are provided for electrodes of polynomial and circular geometry.

We show for the first time that a particle in a non-uniform AC electric field experiences a dielectrophoretic force arising from spatial non-uniformities of the magnitude and phase of the field interacting, respectively, with the in-phase and out-of-phase components of the induced dipole moment.

Nanometre-sized silicon crystallites were achieved using the crystallization of amorphous silicon thin films on silicon substrates by means of a thermal annealing process. Transmission electron microscopy and X-ray diffraction measurements indicate grain diameters from 3 nm to 10 nm, depending on the annealing conditions. The crystallization consists of a substrate induced process forming needle-like silicon crystallites and a strain induced process forming nanocrystalline regions. The crystallized thin films show intense violet and blue photoluminescence at room temperature which exhibit separated peaks at energies ranging from 2.5 eV to 3.5 eV and shows no intensity degradation. The luminescence from such small crystallites is suggested to be due to an enhancement effect on the oscillator strength of the confined levels in zero-dimensional systems. The observations of the violet and blue light emissions from silicon crystallites should be very important for novel optoelectronic device applications of silicon-based materials.