Table of contents

By the year 2000 electronics will probably be the basis of the largest industry in the world. Silicon microelectronics will continue to keep a dominant place covering 99% of the `semiconductor market'. The aim of this review article is to indicate for the next decade the domains in which research work in `physics' is needed for a technological advance towards increasing speed, complexity and density of silicon ultra large scale integration (ULSI) integrated circuits (ICs). By `physics' we mean here not only condensed matter physics but also the basic physical chemistry and thermodynamics. The review begins with a brief and general introduction in which we elucidate the current state of the art and the trends in silicon microelectronics. Afterwards we examine the involvement of physics in silicon microelectronics in the two main sections. The first section concerns the processes of fabrication of ICs: lithography, oxidation, diffusion, chemical and physical vapour deposition, rapid thermal processing, etching, interconnections, ultra-clean processing and microcontamination. The second section concerns the electrical operation of the ULSI devices. It defines the integration scales and points out the importance of the intermediate scale of integration which is the scale of the next generation of ICs. The emergence of cryomicroelectronics is also reviewed and an extended paragraph is dedicated to the problem of reliability and ageing of devices and ICs: hot carrier degradation, interdevice coupling and noise are considered. It is shown, during our analysis, that the next generation of silicon ICs needs mainly: (i) `scientific' fabrication and (ii) microscopic modelling and simulation of the electrical characteristics of the scaled down devices. To attain the above objectives a return to the `first principles' of physics as well as a recourse to nonlinear and non-equilibrium thermodynamics are mandatory. In the references we list numerous review papers and references of specialized colloquia proceedings so that a more detailed survey of the subject is possible for the reader.

The real AC electrical conductivity was studied as a function of frequency f, temperature T and composition x for a series of samples prepared using the double-sintering ceramic technique. The experimental results indicated that increases as the temperature and frequency increase and as the Ni ion substitution decreases. The Curie temperature and the activation energy for electrical conduction ( in the paramagnetic and in the ferrimagnetic region) increase as the Ni ion substitution increases. decreases while increases as the frequency increases and they become frequency-independent above the Curie temperature. The parameters n and B in the power law for were found to be composition- and temperature-dependent. n decreases while B increases as temperature increases. Both n and B decrease as the Ni ion substitution increases. Empirical formulae were suggested for the temperature-dependences of n and B.

General formulae for studying the distribution of the fringing field from a semi-infinite ring head with an arbitrary pole-angle, are derived by the method of the Schwartz - Christoffel transformation. The mapping functions can be calculated without any approximation. Based on these formulae, the distributive properties of the fringing field are studied in detail. It is found that the field intensity is monotonic with respect to the pole angle, while the optimum one does not exist. It is also found that the Karlquist's formula, which was obtained on the assumption that the magnetic scalar potential has a constant gradient across the top of the gap, can be applicable at the practical recording distance only for a rectangular head.

Experimental investigations on the strain-electric resistance relationships of an FeMnAlCr antiferromagnetic alloy show that resistance changes linearly with strain in both elastic and plastic areas. In the elastic area the strain sensitivity coefficient of resistance K = 4.0 - 4.8, and K decreases with increasing Cr content. In the plastic area . It is suggested that the high K of the FeMnAlCr alloy in elastic areas may be related to its antiferromagnetic spin structure. Finally the possibility of developing FeMnAlCr ambient temperature and cryogenic strain gauge materials are briefly discussed.

This paper deals with mixtures of bi-anisotropic media. The mixtures are assumed to consist of a bi-isotropic background medium in which bi-anisotropic, spherical, electrically small inclusions are embedded. A general formula is presented with which one is able to calculate the effective material parameters of the heterogeneous medium as functions of the corresponding parameters of the inclusions and their volume fraction. This formula is general in the sense that it gives the Maxwell Garnett, Bruggeman and coherent potential rules as special cases. Numerical results are given for chiral and bi-isotropic mixtures.

The efficiency of certain electrostatic technologies (precipitation of dust, removal of gaseous pollutants from flue gases and separation of mixed granular solids) can be improved by using pulsed instead of DC corona discharges. In this work, the behaviour of millimetre-sized conductive spheres and cylinders either in DC or in pulsed corona fields was studied with a three-electrode arrangement, which had the advantage that the space charge density and the electric field strength could be independently controlled. By energizing the upper corona electrode from a square-wave high-voltage generator (30 kV; 250 pulses per second), a space charge density of periodical variation in time was produced in the air gap of 10 to 25 mm between the earthed median metallic grid and the lower static electrode, connected to a DC high-voltage supply. It was found that particle movements in pulsed fields (small-amplitude vertical oscillations on the surface of the lower electrode) were less affected by the space charge than in the case of a DC corona. The results of numerical simulation of particle behaviour in ionized fields were in good agreement with the experimental findings. The observed phenomena can be explained in connection with the process of particle charging in corona fields.

The paper describes experiments on a metal ball rolled along parallel metal rails. An electrical force is developed on the ball in the direction of movement by passing current through it from one rail to the other. Contrary to the electromagnetic theory of the ball-bearing motor, the electrical force on a brass ball is of the same order as that on a steel ball. The results are discussed in terms of a torque exerted in the contact region.

We describe a method for calculating the magnetostatic fields of an iron free system of axisymmetric conductors. It is based on a series expansion that has an extensive and useful convergence region. The method is compared to the standard zonal harmonic method of Garrett. The convergence regions and the rates of convergence of the two methods are compared. For conductors of rectangular cross section, an algorithm is described in detail that calculates the conductors' magnetic fields. The implementation of the algorithm is computationally efficient using recurrence relations throughout and involves no numerical integration.

A theoretical treatment of modulation absorption spectroscopy, applicable when amplitude and frequency modulation are present simultaneously, is given. The effects of any relative phase difference between these two types of modulation are incorporated and the theory is applicable to an experiment involving detection at any harmonic order of the modulation frequency. The results obtained are given in terms of the derivatives of the absorption lineshape function and bring out the connection to Nth-order derivative spectroscopy naturally. General expressions for any gaseous molecular absorption line are given. It is shown that certain advantages are obtained when one performs higher harmonic detection, including an increase in the wavelength resolution. Experimental results are obtained and compared with the predictions of the theory developed.

We observed sharp resonances at certain frequencies in the acoustic and optical emission from metals during laser welding with a modulated laser beam. Strong emission is observed at high-order harmonics of the laser modulation frequency. Harmonics whose frequencies overlap with bands of vibrational frequencies corresponding to eigenmodes of the keyhole are greatly enhanced, relative to other harmonics whose frequencies lie outside these bands. The keyhole and its surrounding liquid layer act as a frequency selective amplifier for pressure fluctuations induced by changes in the interaction of laser radiation with the walls of the keyhole. When a CW beam is used, random fluctuations with frequencies within these allowed bands are amplified to produce a spectrum consisting of a large number of overlapping but discrete frequency components. When the laser beam is modulated, a forced response is elicited resulting in a much simplified emission spectrum. These results offer some new insights into the physical mechanisms that convert laser radiation to heat in laser welding, drilling and cutting applications.

Acoustic emission at frequencies between 20 Hz and 20 kHz during laser welding of steel contains important diagnostic information related to weld morphology, depth of penetration and the heat-affected zone. We have recorded acoustic spectra under a variety of welding conditions with mild and galvanized steel sheet and find that a well-defined spectrum is obtained under conditions that lead to an optimization of these weld characteristics. A statistical approach is developed which yields a predictive measure of weld quality, based on a comparison between real-time acoustic spectra, and that obtained under optimized conditions. This technique is applied to an analysis of bead on plate, lap and butt laser welds in mild and galvanized steel.

Employing absorption, excitation and luminescence spectroscopy in a wide spectral range, the upconversion processes are investigated in at low temperature. From the detailed analysis of the spectroscopic data and the power dependence of the upconverted intensity, excited state absorption and energy transfer between excited ions are found to be the important mechanisms. Structure in the Stark substate absorption suggests the ions occupy several different but well defined lattice sites for which the local symmetry is slightly perturbed, presumably by charge compensating defects. Site-selective measurements allow one to determine the ground state splittings for two of these sites. The observed energy transfer suggests the existence of interacting pairs of ions.

.

Two types (plastic and fused silica) of waveguides suitable for transmitting and Er-YAG laser radiation were prepared and characterized. The temperature of several points on the external surface of the waveguides was measured. Optical parameters (transmission, focusing, misalignment) of the two types of waveguides were measured and compared. The importance of heating (due to losses) on the long time delivery performance is also shown.

The relation between the structure of fibres extruded from mixed silver halides and their IR optical losses and luminescence properties is studied. The recently established ageing effect of the considerable growth of mean grain size of the fibre's polycrystalline structure (typically from 20 to 1000 nm in two years) is correlated with IR and luminescence measurements. IR spectra measured in the interval show growth of the concentrations of molecular impurities. The luminescence measurements are used for the characterization of the fibre's structure ordering. The luminescence decay kinetics is analysed theoretically in terms of the diffusion-controlled recombination of spatially well-correlated Frenkel defects. The considerable difference between the diffusion properties of monocrystals and fibres is established. This is caused by partial disorder (the grain structure effect) of the initially polycrystalline structure arising due to the fibre extrusion.

Metallic artificial index gratings are composed of metallic sub-wavelength structures. Appropriately designed metallic artificial index gratings show strong resonance effects if the wavelength is varied. This effect is demonstrated for several gratings.

Surface acoustic wave (SAW) properties of z-cut waveguides fabricated by proton exchange using octanoic acid are investigated. The phase velocity and electromechanical coupling coefficient of SAWs are significantly decreased by the increase of kd, where k is the wavenumber and d is the waveguide depth. The insertion loss IL of a SAW is increased by the increase of kd and becomes nearly constant at . The temperature coefficient of delay TCD of a SAW calculated from the frequency change shows an apparent increase with kd, reaching a maximum at , then slightly decreasing at higher kd. On the other hand, the effects of annealing on and IL are also investigated. The results show that post-annealing can result in a restoration of the decreased and an improvement of IL.

A Carnot-like irreversible refrigeration cycle is modelled with two isothermal and two non-adiabatic, irreversible processes. The generic source of internal irreversibility, measured by the Clausius inequality, is a general irreversibility term which could include any heat leaks into the Joule - Thompson expansion valve, the evaporator and compressor cold boxes. This cycle is optimized first for maximum refrigeration power and maximum refrigeration load, then for maximum coefficient of performance. Its performances are compared with those of the endoreversible refrigeration cycle, based on a propane stage of a classical cascade liquefaction cycle example. Both cycle models achieve optimum power and maximum refrigeration load at nearly the same refrigeration temperature, but only the coefficient of performance of the irreversible refrigeration cycle reaches a maximum. Moreover, its prediction of heat conductance allocation between evaporator and condenser appears to be not only more conservative, but also more realistic for actual design considerations of refrigeration cycles.

A novel wear test is decribed in which a hard tool steel wedge is pressed against the periphery of a bar of softer metal rotating in a lathe. The results obtained are used to investigate the roles of low cycle fatigue and ratchetting failure in producing wear.

This paper presents a heterodyne technique by which to solve an inverse problem on the determination of the media turbulence characteristics. Theoretical modelling of optical wave propagation through turbulent media is based on the use of the Kolmogorov - Oboukhov model. The coherence degree of the wave, measured by heterodyne detection, allows us to obtain the transverse correlation lengths and the index-of-refraction structure constant of the turbulent media.

By means of the Blevin - Hasan version of the Bradbury - Nielsen technique and Townsend - Huxley method, the drift velocity W, characteristic energy and the parameter have been determined at ambient temperature for reduced electric fields Td, Td and Td, respectively. The present results have been compared with experimental and theoretical ones of other investigators.

The temporal relaxation of the electron component in weakly ionized, collision-dominated plasmas is studied on the basis of the time-dependent Boltzmann equation. The relaxation process is treated using an expansion of the electron velocity distribution function in Legendre polynomials. In the past this expansion has been truncated after the first two terms. The equation system for the first two expansion coefficients which is obtained in this two-term approximation from the time-dependent kinetic equation has been solved by the so-called conventional approach and, very recently, by a strict time-dependent approach. In the conventional approach the distribution anisotropy is treated in a quasi-steady-state way. In generalization of the conventional and the strict time-dependent two-term approximation an efficient numerical method for solving the time-dependent electron kinetic equation in a multi-term approximation has been developed and its main aspects are presented in this paper. As an application, the temporal relaxation of an energetic electron group in a weakly ionized, collision-dominated neon plasma, acted upon by a DC electric field, has been investigated. Main findings of this relaxation study are that, depending on the reduced electric field strength, (i) the conventional two-term approximation can completely break down at early and intermediate stages of the relaxation process, (ii) the strict time-dependent two-term approximation can also be insufficient at intermediate relaxation stages and (iii) the converged solution of the Boltzmann equation is obtained throughout the relaxation in an approximation with six to eight terms from the expansion of the electron velocity distribution.

The physical processes determining the functions of an ionization system and especially the discharge stabilization by the distributed resistance of a semiconducting cathode in such a system are studied. The current - voltage (I - U) characteristics of the system with a semiconducting GaAs cathode are obtained experimentally as functions of the gap pressure P (16 - 760 Torr) and inter-electrode distance d ( to 5 mm), which are varied for the first time over very wide ranges. The experiments showed that the presence of the distributed resistance affects the passage of the discharge current so that the discharge glows uniformly throughout the cathode surface and is not divided into separate filaments. The loss of stability was primarily due to the formation of a space charge of positive ions in the discharge gap which changed the discharge from the Townsend to the glow type.

The refractivities of sulphur hexafluoride and thionyl fluoride have been measured at wavelengths of 632.99 and 1300 nm using a polarization interferometer. The value obtained for at 632.99 nm is in close agreement with results of previous workers, which gives credence to the new values for at 1300 nm and for at both wavelengths. is produced in electrical discharges through as a consequence of dissociation and these refractivity data may find an application in the interferometric monitoring of gas degradation in discharges.

An experimental and theoretical study of the drift oscillations in a non-equilibrium gas-discharge plasma with a density gradient across a magnetic field at low electron densities was carried out. The solution of the set of differential equations for the bounded cylindrical plasma is used to describe unstable oscillations in the plasma de-stabilized by the action of non-quasi-neutrality in perturbations and ion inertia simultaneously.

The high-frequency (HF) oscillations, which are driven by the electron beam on the high-potential side of an electric double layer, are investigated in a laboratory experiment. A new HF probe design has made it possible to achieve a combination of absolute amplitude calibration and spatial resolution which, for such high frequencies, has not been achieved before. The HF waves convert about 20% of the beam energy to oscillations within a region extending from 100 to 400 Debye lengths on the high-potential side of the double layer. This makes the HF waves the dominant mechanism of local tapping of the beam energy, and the modified energy and particle balance is investigated. The HF waves propagate with an approximately constant phase velocity which is slightly smaller than the beam velocity. Time-averaged measurements locate them to a HF region, extending approximately from 5 to 15 cm from the double layer, with a typical half width of 10 cm for the time-averaged electric field amplitude. Time-resolved measurements show, however, a much more narrow structure. The envelope of the electric field amplitude has a single maximum within the HF region and a typical half width of 1 - 2 cm (about one wavelength) along the beam. We call this envelope `the HF spike'. The position of the HF spike changes in an apparently irregular fashion within the HF region and it moves with velocities in the range . The spatial increase and decrease in amplitude of the wave is exponential over nearly two orders of magnitudes, with electron-folding distances of only about 5 mm, so the maximum of the envelope is very sharp. The amplitude increase agrees approximately with the growth rate from linear beam-plasma theory and the maximum amplitude observed agrees with saturation by beam trapping. The strong spatial decrease in the wave amplitude is not understood. It is proposed that the motion of the HF spike is caused by fluctuations, on the ion acoustic time scale, both in the growth length and in the starting point for wave growth at the double layer, which moves back and forth. This motion is shown to be strongly correlated to the motion of the HF spike in a particular case. Amplitude modulations, as observed by a stationary probe, are found on two time scales. On a slower time scale of typically , the motion of the HF spike, together with its limited spatial extent, gives rise to a temporal burst. Within these bursts the waves are also modulated on a much faster time scale of 10 - 30 ns (the wave period is 3 ns). The HF spike cannot be interpreted as a linear superposition of waves with the constant phase velocity measured, because this wave packet would have a spatial extention at least ten times larger than the width of the HF spike.

We have treated coagulation of microparticles of characteristic sizes cm in a low-temperature plasma at a gas pressure of the order of 1 Torr. The distribution function of the particles' charge, which is found before, allows one to calculate the rate of coagulation of two particles with given sizes. For the distribution function of the particles' sizes the non-stationary kinetic equation is solved numerically. The analytical formulae for the mean charge of particles, the fraction of the neutral particles that are charged and the time-dependence of the mean radius of the particles are deduced in this paper and comparison of the analytical and computed results is performed. It is shown also that the experimental data concerning the coagulation of silicon particles in an argon plasma coincide qualitatively with calculated values.

Two distinct modes of plasma sheath (PS) motion, the stable mode and the instability mode, in the radial compression and pinch phases in a small Mather-type plasma focus device (20 kV, 18 kJ) have been observed using laser differential interferometry. The two modes occur randomly from shot to shot in consecutive discharges and their probabilities are much influenced by the anode end structures. The instability mode is characterized by the rapid radial movement of the sliding front of the PS along the anode (centre electrode) end surface. The two modes show very different sheath and pinch configurations according to their different ways of motion. They also have notable influences on the neutron yield, in that the instability mode always produces a lower one and the stable mode a higher one. The mean neutron yield for consecutive discharges is made low and largely deviated by the co-existence of the two modes. The instability mode is found to be caused by the anode end geometry with a small transition arc, whose nature is of the m = 0 MHD type but only with a half wavelength, occurring at the plasma - electrode layer. The instability can be either avoided by a hemispherical-ended anode with a larger transition arc or stabilized by the stagnation of the radial movement of the sliding front of the PS in the initial radial compression by the hollow edge and the stable mode formed. The magnetic field introduced in front of the PS by the anode end discharge that occurs from the very beginning of the main discharge is also found to act as a stabilization mechanism for the instability and to cause some of the stable mode.

The distribution of atomic states for a helium plasma torch at high (atmospheric) pressure has been determined by calculating the populations of some excited states of the neutral helium system by means of different spectroscopic techniques. A plasma not close to local thermodynamic equilibrium was found. The overpopulation of some states showed that it was an ionizing plasma with the equilibrium deviation parameter of the ground state, , of about 100. The population of the excited states closely obeyed the law, where p is the effective principal quantum number.

This article presents a simple, unified theoretical description of the evolution of dust in plasmas consistent with recent laboratory experimental research and astronomical observation. The treatment is extended to explain how dust particles develop under the influence of magnetic fields, leading to the growth of dust needles, rather than spherical grains. Such elongated grains are shown to oscillate about the magnetic axis.

In this paper, we present and discuss a relationship between the leader current, the electrical charge, the velocity and the mode of propagation of the leader. By using energy considerations, we establish a mathematical expression that requires only physical parameters and allows one to evaluate the velocity of the leader or the fractional part of the energy consumed during propagation of the leader. The obtained value of gives leader velocities in good accordance in shape and amplitude with those measured experimentally for discharges in long air gaps.

An accurate numerical calculation of the electron emission using the Numerov integration method of the Schrödinger eigenvalue equation is presented. Such a calculation eliminating the WKB approximation and the use of elliptic functions can give the corrections that have to be applied to the well-known Fowler - Nordheim [1] or Murphy and Good results [2]. Because the very intense electron emission that takes place in the cathode spot is greatly affected by those corrections, we give the results of the calculation of the Nottingham effect, which plays an important role in the energy balance equation at the cathode surface. Finally, an exploratory study is presented on the influence of ions within a distance of a few tens of ångström units in front of the cathode surface on electron emission.

The influence of mono-energetic beam electrons on the behaviour of a plane plasma boundary is investigated assuming that the ion, electron and beam electron components in the pre-sheath act as three coupled fluids. The sheath edge is identified by the singularity in the electric field appearing under the quasi-neutral assumption, and the Bohm criterion is solved for a range of beam and plasma parameters. Collisions between the plasma particles and the neutral species are considered in a regime in which Coulomb interactions are unimportant. The plasma is generated by beam and thermal electron impact ionization of the neutral gas. Under collision-free conditions, sheath-edge potentials and ion energies are found to increase with the ratio of beam to thermal electron density, up to a critical ratio of about 0.3, above which no sheath edge is predicted to exist. Ion - neutral species collisions further raise the Bohm speed; however, the attenuation of the electron beam through scattering and ionization does not appreciably alter the sheath-edge conditions, but changes the width of the boundary layer. The results of this model, which are relevant to industrial plasma discharges containing beam electrons, are compared with those found from other treatments of the beam - plasma boundary.

The evolutions of the kinetics and of the energy distribution of the secondary electrons in argon gas during and following irradiation by a 350 ns 150 keV electron beam pulse have been studied quantitatively. The investigation was conducted by time-resolved infrared interferometry measurement of the secondary electron density for a wide range of experimental parameters and by comparison with a comprehensive time dependent kinetic code. From the comparison it was concluded that triatomic molecular ions play important roles in the kinetic reactions.

Noisy and spectrally poorly resolved molecular emission spectra of the 306.357 nm OH band were employed for evaluation of the rotational temperature in different plasma sources, in the temperature range 300 - 6000 K, when an apparatus function of a recording system is unknown. The described method is based on a comparison of experimental data with a theoretical spectrum. The following plasma sources are examined: a recombining jet of plasma torch, an atmospheric pressure glow discharge, a surface ionized wave device and a gliding arc. The temperatures of these weakly ionized plasmas, measured from the OH spectra, are in good agreement with temperatures obtained by using other techniques: thermocouples, tungsten wire resistance measurements and other spectroscopic methods. The results show that the OH spectra are very useful for temperature measurements of non-equilibrium plasmas.

The propagation of the long cathode-directed streamer in air is simulated in the approximation of a fixed radius of the streamer channel. The results of the simulation turn out to be sensitive to the channel radius and to a model of charged particle kinetics. A model including ionization expansion of the streamer channel and yielding results which depend only slightly on the initial channel radius is developed. For the streamer's propagation in a non-uniform field, our calculation agrees well with the available experimental values of the total anode current, the streamer length and the average electric field in the streamer channel. Simulation of the streamer's evolution in a uniform electric field shows a steady propagation if the field exceeds .

This paper deals with the modelling of the convection processes in high-pressure mercury arcs. Temperature and velocity fields have been calculated by using a 2D semi-implicit finite-element scheme for the solution of conservation equations relative to mass, momentum and energy. After validation, this model was applied to the study of the influence of arc external parameters such as mercury charge and tube diameter on the convective processes. It was found that stable laminar mono-cellular convection flow occurs at low values of tube diameter for a cylindrical burner and/or a low mercury charge. Finally, we considered in detail the region behind the electrodes, where an accumulation of mercury is observed. The evaluation of this amount of mercury `trapped' in these regions is of prime importance for a good description of the distribution of mercury in the burner and for a correct evaluation of the total discharge pressure.

The optical diagnostics of a DC pulsed plasma used for iron nitriding is carried out for various plasma conditions. The fluorescence of the first negative and the second positive systems of is observed in the time afterglow, in order to obtain some insights into the creation and the loss processes of and states. An iron emission line is also observed which is indicative of the sample sputtering. Particular attention is paid to the influence of the gas pressure on the respective durations of the discharge and the afterglow. Analysis of the plasma both during the discharge and during the post-discharge leads to the determination of favourable conditions for the surface treatment. It is also shown that the electron density in the discharge can be deduced from the line intensity decay in the afterglow.

A model of a weakly ionized, low-pressure positive column discharge is described. Of particular interest is the generation of atomic resonance radiation. The model computes species densities as a function of radial position using fluid continuity equations. Electron impact collision rates are found from the zero-dimensional Boltzmann equation. The appropriate electron impact cross sections, heavy particle collision rate coefficients, and radiative decay rates for xenon have been assembled from the literature, and are summarized. Model predictions compare favourably with measured axial electric field values in a xenon positive column discharge, while the model is found to overpredict the electron temperature at low xenon pressure.

An approximate method is presented to take account of interaction of positive streamers simultaneously propagating between a wire and a plate. The method is based on a quasi-2D positive streamer model modified by inclusion in the expression for electric field of an additional term describing the field of charges of all streamer channels in the gap. All streamers are assumed to be identical and to move parallel to each other to the plate from points equidistantly distributed along the wire surface. The model is used for calculation of streamer dynamics in atmospheric air for conditions under which simultaneously propagating streamers have been observed. It is shown that taking account of streamer interaction improves the agreement of the calculated dependence of streamer velocity on its length with the experimental one.

The electron emission mechanism which is active at the cathode of a pseudospark switch leads to thermal overloading of emission sites followed by the evaporation of cathode material. As a consequence, metal vapour should be present very shortly after the beginning of current flow. However, it has never been possible to observe metal vapour by laser-induced fluorescence (LIF) prior to the cessation of current. In the present work resonance absorption of laser light was used to detect molybdenum vapour from the cathode during the high-current phase of a pseudospark switch. To observe the narrow-band absorption the laser light which had passed the switch was introduced into a molybdenum hollow cathode lamp where it caused fluorescence from molybdenum vapour. The intensity of the fluorescence light from the hollow cathode lamp was proportional to the intensity of the laser light left after the absorption process in the pseudospark. For a current amplitude of 12 kA and a pulse duration of , it is shown that molybdenum vapour is definitely present shortly after the current maximum, well before the current stops flowing. The neutral vapour density in the high-current phase is estimated to amount to at least . It is concluded that LIF fails to detect metal vapour prior to the cessation of current, mainly because the fluorescence process is severely disturbed by electronic collisions which deplete the upper fluorescence level.

A high-voltage gaseous-switch triggering method based on a capacitively coupled (pulsed corona) discharge is described. The method enables the single closure discharges of the switch to be distributed over a large area of the electrodes' surface, yielding a relatively low erosion rate. Therefore, long-lifetime operation of the switch without maintenance is obtained. The switch was tested by discharging a 7.5 nF capacitor at repetition frequencies up to 50 Hz. The maximum voltage was 17 kV with a current peak of about 6 kA.

Synchrotron x-radiation in a double-crystal geometry has been used for topography and rocking-curve measurements to study imperfections in single-crystal HP/HT synthetic diamonds grown from (111) seeds by the reconstitution technique. The influence that clusters of small opaque particles (sometimes called `pin-point-like inclusions') have upon lattice parameter differences, as well as upon relative lattice tilts within each sample, was of particular interest in this study. Rocking curves measured with a large beam diameter (illuminating entire specimens) gave widths (FWHM) in the range 5 - 30 seconds of arc in a non-dispersive Bragg - Bragg diffraction geometry in which good diamond crystals give and for which the calculated theoretical minimum width is under . Rocking curves were also recorded for small regions of each crystal, using a translatable pin-hole, and these showed that some diamonds had relatively narrow rocking curves in selected areas (of diameter). Epochs of crystal growth were recorded in series of x-ray topographs, taken at intervals of 2.5 seconds of arc. Such series were taken for all four principal azimuths (, 90, 180 and ) of the incident beam by rotating the sample about the normal to the Bragg plane. This procedure yielded important information about lattice parameter distribution and lattice tilt across entire samples. For the samples heavily populated with pin-point-like inclusions it was concluded that there is a possible slight variation in lattice parameter values in some regions of the crystal but most of it falls below the adopted angular step resolution, although better accuracy can be achieved. Relative tilt measurements revealed that there are large misorientations of those parts of the crystal which are close to external outcrops of {111} boundaries. In this case of (111) oriented diamond samples, three separate regions with different tilts can be attributed to the three parts of the crystal close to optically visible {111} growth-sector boundaries.

This paper is devoted to an experimental investigation of the ferromagnetic aerosol behaviour in magnetic and electric fields. Under intense lighting some particles move in magnetic fields similarly to objects having a magnetic charge. It is shown that the magnitude of such charges may be equal to Dirac's one.

Hydration and hardening processes of Portland cement (type I) were studied by analysis of the one-dimensional projections (profiles) obtained periodically with the stray-field imaging technique over two days. The influence of additives, such as gypsum, in Portland cement (type IA) was also investigated. The decay of the signal intensity as a function of time was found to be bi-exponential for type I and mono-exponential for type IA.

Zinc oxide ceramic varistors with simplified compositions show nonlinear current - voltage characteristics with in the range 45 - 130. Addition of enhances the nonlinearity tremendously and the concentration range of needed for high is around 3 mol%. Formation and involvement of a liquid phase during sintering is indicated from the microstructure studies. Energy dispersive x-ray analyses (EDAX) show that Si is distributed more in the grain boundaries and within the secondary phase than in the grain interiors. X-ray diffraction and infrared spectroscopy studies indicate the formation of zinc silicate, . The capacitance - voltage measurements show that the donor density , barrier height , interface density and depletion width (W) are comparable to those of the ceramics without . Capacitance above the breakdown voltages shows negative values, indicative of oscillatory charge redistribution involving multivalent states of Co and the shallower interface states. Multiple trapping relaxations are evident from the complex-plane capacitance studies. The admittance spectroscopy data show that the type of trap state changes with the addition of , accompanied by a considerable increase in the density of traps.

We studied the newly synthesized quasi-one-dimensional compound . Resistivity measurements confirmed its pseudo-one-dimensional character and photoemission spectroscopy clarified the nature of its valence band states. This polytelluride is a narrow-gap semiconductor at room temperature, with a forbidden gap width of 0.34 eV.

The electrical characteristics of Schottky barrier diodes were studied considering the effect of shunt resistance. Both the DC and the AC behaviour of the device were investigated. It was found that the shunt resistance has a remarkable effect at low bias in contrast to the series resistance which influences the electrical characteristics at large bias. The standard evaluation techniques based on the DC characteristics are found to be inaccurate in the presence of the shunt resistance. Also, the AC conductance and capacitance of the device are significantly influenced, making our previously developed model limited in the presence of a shunt resistance. The well-known conductance and capacitance techniques for the characterization of these devices seem to be inapplicable at low bias because of the above-mentioned parasitic effect.

The space-time evolution of the flow-induced charge accumulation and fields in insulating tubes is analysed numerically. The proposed coupled finite-element and Laplace transform algorithm is an economical method for solving linear problems in which the electrical conductivity of the dielectric medium is negligible. The study focuses on the external charging mode under which the field arises from fully developed turbulent flow of a semi-insulating charged liquid through insulating tubes. The validity and accuracy of the advocated method is checked by applying it to certain models for which the experimental results have been documented by previous researchers or to some models for which analytical solutions can be found.

It has been found that dosing an electroformed device with chloroform leads to an increase in device current at voltages above a critical voltage of about 6 V. It is postulated that chloroform molecules help to release trapped charge which otherwise inhibits the regeneration of conducting filaments in the device. It is also found that the emission current increases more sharply than the device current upon exposure to chloroform and hence it is postulated that the emission of `hot' electrons is linked to filament regeneration (rather than filament rupture as previously believed). The influence of chloroform upon the voltage-dependences both of the device and of the emission characteristics has also been investigated. We discuss the possible exploitation of these effects in gas sensing.

We present the results of an experimental study of the dynamics of thermal self-oscillations of a normal zone in thin films of high-temperature superconductors (HTSC) connected to an external circuit with an inductance and cooled by boiling nitrogen. The response of the system to a harmonic driving force has been studied. Spectral and dynamical characteristics of various oscillative regimes have been determined. It is shown there arises a state of a deterministic chaos in some range of driving frequencies and amplitudes. An effective of synchronization of periodic self-oscillations of a normal zone and thermo-hydrodynamical oscillations caused by the boiling of liquid nitrogen near a film surface has been recognized.

Extensive structural studies were performed on different DC Joule-heated ribbons as well as on conventionally heat-treated samples. Optimum properties for the DC Joule-heated samples were obtained for I = 3.3 A and t = 60 whereas for the conventionally heated ribbons the optimum properties were obtained for a treatment at for 1 h. Grain size distributions of all samples were determined by x-ray powder diffraction analysis. Transmission electron microscopy (TEM) of the optimal DC Joule-heated sample was performed and compared with that of the conventionally heat-treated samples. Both x-ray analysis and TEM showed the known decomposition of the BCC structure into a mixture of -Fe plus plus phases. The lattice parameter of the BCC structure was measured and compared among the different heat-treated samples. TEM investigation revealed the existence of zirconium oxide precipitates in all ribbons.

In this paper we deal with a numerical model for the evaluation of the magnetic iron losses in one steel lamination. The magnetodynamic model for the magnetic field strength H is coupled with the Preisach model. The resulting boundary value problem for H is solved numerically by a modified finite element - finite difference method. The model is validated by the comparison of numerical experiments and measurements. The dynamic behaviour of two materials is investigated.

The partial discharge activity (the number of discharges per second) during electrical tree growth in the flexible epoxy resin CY1311 was measured using a phase-resolved synchronous counting system. The experimental conditions, voltage and pin - plane spacing were varied to produce a wide range of tree structures from branch to bush. Systematic and repeatable changes in the partial discharge activity occurred, depending on the experimental conditions and these correlated with the type of tree structure (branch density or fractal dimension) formed. The type of tree growth was principally determined by the applied electric field and the occurrence of regular changes (bursts) in the partial discharge activity which are associated with a sudden and temporary phase shift and broadening of the partial discharge phase distributions. The number of bursts an their dynamics determine the type of tree grown.

The dielectric response of hydrating cement paste has been measured in the frequency range 5 - Hz. The spectra of complex capacitance have been characterized in terms of the dispersive bulk and barrier responses at high- and low-frequency ranges, respectively, and by a generalized Maxwell - Wagner response at intermediate frequencies. We have developed a theoretical model for the bulk response of the cement paste by considering a lattice gas model for the diffusion of ions in pores exhibiting a fractal distribution in size. Using this model we have assigned physical meanings to parameters characterizing the bulk dielectric response and have determined the variation of these parameters during hydration.

This paper presents an approximate analysis for predicting the generation of elastic waves in multi-layered piezoelectric materials, when these waves are generated using interdigital transducers (IDTs). Each IDT finger is assumed to act as a source of plane (partial) waves propagating in all directions, and the condition for efficient wave generation is taken to be when the waves from adjacent pairs add in phase. The width of the IDT fingers is not explicitly taken into account and neither are the finite dimensions of the IDT. The conditions for surface or guided wave propagation between IDT transmit and receive pairs are then determined by satisfying the boundary conditions on the surfaces and at the interfaces of the substrate, using the partial waves already determined. The outer surfaces of the substrate may be either metallized or free. The model thus predicts discrete frequencies of operation of IDT devices, corresponding to the generation of surface or guided waves. No `width' is attached to these frequencies, because of the neglect of IDT finger width and of overall IDT dimensions. Wave profiles within the piezoelectric substrate are calculated by summation of the partial waves, to depict the depth-dependence of the mechanical displacement, electric potential and energy flow. This information is used to distinguish surface waves from guided waves and to show the depth of penetration of the surface waves. The paper is essentially divided into three parts. The model formulation is presented, computational methods and associated difficulties are discussed, and sample model predictions for Rayleigh and SH wave devices in ST cut quartz are compared with experiment, showing good agreement.

A novel use for magnetic birefringence experiments is in the measurement of the birefringence of commercially important colloidal vermiculite particles. When aligned in a magnetic field, results at low field strengths show a steady state birefringence proportional to the applied field. The effects on the relaxation time of the birefringence on removal of the field are found to decrease with increasing field strength. This has been attributed to polydisperity of particle size. A theory analogous to that of electric field relaxation and used for magnetic field relaxation data has been further developed and modified to incorporate the more general case of diamagnetism. The modified theory is then used to obtain a two-parameter log-normal size distribution function for the suspended particles. The experimental results obtained are in accord with data acquired from electron micrographs. It is therefore concluded that use of the modified theory is justified and forms the basis of a valid measurement technique applicable to rigid particulate systems of any uniform, regular spheroidal morphology.

Photoelectric properties were measured on a single crystal prepared by a special design based on the Bridgman technique. We report the results of an investigation of the dependence of DC and AC photoconductivity on light intensity, applied voltage and ambient temperature. The forbidden energy gap at room temperature and the thermal coefficient of the energy gap were calculated also and found to be 1.793 eV and respectively.

Dense, Yb-doped -sialon ceramics of the composition , with , together with some Yb-based - and -sialon ceramics containing an excess of glassy phase, have been prepared by hot-pressing at . These materials were subsequently heat-treated in the temperature range for different periods. It was found that Yb-doped -sialon forms within a wide composition region, with x ranging from 0.3 to more than 1.0, and that -sialon is stable over a large temperature interval and during heat treatment times of up to 30 days. The -sialon phase coexists with a liquid phase at and with crystalline phases such as the Yb garnet and/or J phase at . Further heat treatment at revealed that initially (during the first 24 h) the -sialon phase reacts with the residual liquid grain-boundary phase, yielding a material which contains an Yb-stabilized garnet phase, besides the -sialon phase, but no glassy phase. This phase assembly is stable upon further heat treatment (30 days). Samples containing Yb-doped -sialon in conjunction with an amorphous intergranular phase show very high hardness and fracture toughness in the range of at room temperature, and mechanical data in the same range were obtained for samples exposed to prolonged heat treatment at . Elongated -sialon grains were found in samples with high x values, but their presence did not affect the mechanical properties significantly.

A model of vertical dynamic penetration of projectiles in granular soils was developed based on known experiments and the theory of dimensions. The depth of penetration is derived as a function of initial velocity and material properties. Velocity and acceleration are obtained as functions of time and depth of penetration. Under certain conditions two acceleration peaks are observed, an initial one due to dynamic and a second one due to static characteristics of penetration. Static properties of soils are derived using dynamic measurements. Numerical examples are provided. Theoretical and experimental results coincide reasonably well.

A magnetically enhanced radio frequency (13.56 MHz) synchronized triode reactor with a multihole cathode used for etching of Si by a plasma is described. The ion energy is controlled independently of the plasma creation function by adjusting the phase difference between the two rf signals. The cathode shape allows a decreasing of the self-bias potential and an increasing of the ion current density when the plasma is created in the holes. We have studied the effects of ion energy and incident power on etch rate, contaminations and morphology of Si samples. The effects of the magnetic confinement and the electrode shape on the etch rate are also presented. Relative to a standard plate electrode, the etch rate is enhanced by a factor of about three at a pressure of 10 mTorr and a power of 500 W.

The active part of a linear translator is a shape memory alloy (SMA) made of nickel and titanium (NiTi) wire which is to be thermally cycled. We have achieved heating using electromagnetic radiation with a magnetic sheath and low-frequency waves at 8 kHz and without magnetic sheath and radio frequency waves at 28 MHz. The heating is equivalent for these two arrangements. In vitro experiments have been confirmed by computer simulations of the radiation distribution within the implant. We thus show that electromagnetic radiation could specifically heat a NiTi wire inside a stainless steel tube without heating the tube. An application could be a femoral prosthesis for the lengthening of the bone.

Thin carbon films have been prepared by ArF excimer laser ablation of hard fullerene-based carbon in vacuum. The target material was obtained from a pressure - temperature treatment of fullerene. Micro-Raman, electron energy loss and electron diffraction measurements of deposited films reveal distinct evidence of microcrystalline diamond. Appearance of the diamond fraction in the films can be related to a peculiar structure of the target substance.

Deep level transient spectroscopy (DLTS) has ben employed to study the cause of minority carrier lifetime degradation observed during the development phase of a single crystal silicon solar cell production process. Results on float-zone (FZ) samples showed that this was due to Fe unintentionally introduced at low levels (close to or below the detection limit of conventional analytical techniques) during cell processing and that DLTS of FZ wafers can be used as a highly sensitive monitor of process impurities. The DLTS spectra of processed FZ wafers showed a single trap, 0.53 eV away from the band edge and with a concentration of , attributable to an Fe - B complex. The feature was not present before processing. In Czochralski (CZ) wafers the situation was more complex. Two major Fe related majority carriers traps were observed in both seed-end and crucible-end wafers taken through the complete process cycle. One was consistent with the Fe - B complex whilst the other (at 0.33 eV from the band edge) was identified as an Fe - O complex. The concentrations of these traps lay in the range and showed a less marked increase after processing. The observations are again consistent with an increase in the Fe level and also the different oxygen contents of the starting wafers.