Table of contents

I first became aware of Professor Rosen the physicist in 1966 when he started to measure (and understand) the magnetic and magnetoelastic properties of the rare earth elements. His pioneering work in this area provided both a wonderful example of careful measurement and intelligent interpretation but also a significant concern since I was at the same time carrying out similar measurements for my PhD. This, however, was only a portent of things to come. Whenever I had a bright idea or a new topic for research it soon became clear that Moshe Rosen had been there already and in many cases skimmed the cream off the top! His work on ultrasonic techniques and their application in materials science generally has been a significant and important contribution to the field. Indeed, his term as Head of Department at Johns Hopkins University in Baltimore, USA has helped to keep the Department of Materials Science at the forefront of world-wide activity in the field. However, his most important contribution to Journal of Physics D: Applied Physics has been his eight-year term as a member of the Editorial Board. He has brought to our activities a deep insight in the materials field, a keen eye for the new emerging topics and a wealth of contacts with potential authors. He played a key role in driving the new subsections of the journal in Applied Magnetism and Magnetic Materials and in Applied Interfaces and Surfaces. All these, though, are relatively minor - it was his enthusiasm for applied physics and his collegiate approach to the success of the journal that influenced all of us who have been involved with Journal of Physics D: Applied Physics during the last ten years. Moshe will be sorely missed but we are privileged to have been able to call him a friend and colleague.

Professor Stuart B Palmer

Polarization switching and dielectric spectroscopy of KTiOAsO₄ (KTA) crystals have been studied. This has allowed, for the first time, the fabrication by low temperature poling of periodically-poled domain structures in KTA crystals (0.5 mm thick, 10 mm long) and also the observation of optical parametric oscillation in the mid-infrared region. The integrity of the periodic domain pattern and the structural perfection of the periodically poled crystal have been demonstrated using scanning electron microscopy (SEM) and high-resolution x-ray diffraction imaging.

This review presents a wide-ranging broad-brush picture of dielectric relaxation in solids, making use of the existence of a `universality' of dielectric response regardless of a wide diversity of materials and structures, with dipolar as well as charge-carrier polarization. The review of the experimental evidence includes extreme examples of highly conducting materials showing strongly dispersive behaviour, low-loss materials with a `flat', frequency-independent susceptibility, dipolar loss peaks etc. The surprising conclusion is that despite the evident complexity of the relaxation processes certain very simple relations prevail and this leads to a better insight into the nature of these processes.

The hydrodynamics of steady laminar film flow of a viscoelastic fluid down a vertical wall is considered. The rheological model used is known as Walters' liquid B´´ and exhibits normal-stress differences even in simple shear flows. The equations governing the fluid motion are subject to boundary-layer type approximations. The momentum integral approach is then adopted to obtain the variation of film thickness with distance in the flow direction. The resulting analytical expression for the film thickness reveals that viscoelastic films develop more rapidly than Newtonian liquid films towards the downstream asymptotic state.

Cathodoluminescence (CL) spectra from a silicon dioxide (SiO₂) film on silicon (Si) substrate were studied. The temperature-dependent results of the 2.7 eV peak suggest that the quantum efficiency increases but the build-up of electron-beam-irradiation-induced luminescence centres decreases upon specimen cooling. The voltage-dependent behaviour of the 2.7 eV peak does not show any luminescence enhancement from the SiO₂-Si interface.

Substantially improved uniformity and enhanced electron field emission properties of hydrogen-free diamond-like carbon (DLC) films were obtained using a titanium intermediate layer after the annealing process. Large emission current densities of 2.08 mA cm^-2 at 14.3 V µm^-1 and 7.20 mA cm^-2 at 25.7 V µm^-1 were achieved for DLC/Ti/Si film annealed at 430 °C for 0.5 h. Its field emission was much more uniform than that of as-prepared DLC/Ti/Si and DLC/Si films. Secondary ion mass spectroscopy (SIMS) showed that C has been amply diffused into the Ti layer. An x-ray photoelectron spectroscopy (XPS) spectrum of the annealed DLC/Ti/Si film after 10 min of argon ion sputtering showed the formation of TiC at the interface between the DLC and Ti/Si substrate. This interaction and interdiffusion of C and Ti could significantly lower the Schottky barrier height between the DLC and Ti/Si substrate. The result was that electrons induced from the Ti/Si substrate can be easily penetrated into DLC films, which enhances the field emission properties.

The influence of interstitial H, C and N atoms on structure and intrinsic magnetic properties of Nd(Fe,M)₁₂Z_y (M = Ti, V, Mo; Z = H, C, N) was studied. Interstitial hydrogen uptake is reversible, expands the lattice and has a small, but positive effect on the intrinsic magnetic properties. Upon C and N absorption, the ThMn₁₂-type crystal structure is preserved in the case of M = Mo, but a decomposition process and strong growth of the content of -Fe is observed at temperatures above 450 °C for M = Ti or V. Along with this decomposition process, a deterioration of the magnetic properties (measured at room temperature) occurs. For M = Ti or V, the highest values of the Curie temperature, saturation polarization and anisotropy field were found just before the decomposition process starts. For M = Mo, the Curie temperature and anisotropy field have the highest values for Z = N, with maxima for a nitrogen concentration of 1 per formula unit. At low temperatures, the high field magnetization curves of all carburized compounds investigated show a metamagnetic transition when the field is applied perpendicularly to the aligned c-axes of the epoxy bonded powders.

We report on the measurement, over the frequency range 50 MHz to 18 GHz, of the complex susceptibility, () = ´()-i´´(), of a magnetic fluid as a function of the polarizing field. Measurements were obtained by means of the short-circuit transmission line technique, with the frequency range being realized by the use of an HP Network Analyser which operates up to 40 GHz. From these measurements more complete ´() and ´´() profiles have been obtained, from which the values of resonant frequency, f_res, and frequency of maximum absorption, f_max, can be identified. The variation in f_max/f_res as a function of the polarizing field, H, over the range 0-100 kA m^-1, was investigated and found to be in accordance with that predicted by Raikher and Shliomis. The extended susceptibility profiles have enabled the normalized after-effect function of the sample, b(t)/b(0), to be investigated within the 10^-11 s time region.

Three Fe-rich SmCo₅Fe_x (x = 0, 1 and 2) ribbons were prepared directly by melt spinning using a wheel speed of 25 m s^-1. The x = 2 ribbons possessed the best hard-magnetic properties, intrinsic coericivity 8.2 kOe, saturation magnetization 117.72 emu g^-1, and maximum energy product 16.06 MGOe. The microstructures were investigated by transmission electron microscopy. The x = 0 ribbons showed striated grains with varied compositions, such as SmCo₅, SmCo₃, Sm₅₄Co₄₆, etc. In contrast, the x = 1 and 2 ribbons showed cellular microstructures. However, the boundary phases were found to be different, Sm(Co,Fe)₃ for the x = 1 ribbons and both Sm(Co,Fe)₃ and Sm₂(Co,Fe)₇ for the x = 2 ribbons. The correlation between magnetic properties and microstructures is discussed in detail. The coercivity mechanism is explained according to the temperature dependence of coercivity by the nucleation model. In turn, the dependence of the small microstructure parameters _k and N_eff on intergrain exchange coupling were analysed in nanoscale grains of SmCo₅Fe₂ ribbons.

The decomposition of pyrolytic boron nitride (p-BN) during milling is studied as a function of the milling time. It has been found that the p-BN compound can be partially decomposed by milling until an amorphous p-BN phase is formed so that the content of nitrogen in the p-BN system will not continue to be changed by the milling process. Furthermore, the structure and magnetic properties of Nd₂Fe₁₄BN_x-based alloys prepared by mechanical alloying using either p-BN or milled p-BN as starting material have been investigated. The Nd₂Fe₁₄BN_x phase with x up to 0.25 coexists with some amounts of NdN, the Nd-rich phase and -Fe. A pre-milling process of p-BN favours the formation of the Nd₂Fe₁₄BN_x phase. The magnetic properties of Nd₁₆Fe₇₆B₈N_x alloys prepared by using milled p-BN are better than those made of non-milled p-BN. The Curie temperature of the Nd₂Fe₁₄BN_0.25 phase is 335 °C, which is slightly higher than that of the Nd₂Fe₁₄B compound. A coercivity higher than 20 kOe is achieved for Nd₂Fe₁₄BN_x-based alloys by adding excess Nd, which is close to the value of Nd₁₆Fe₇₆B₈ prepared by using pure B.

The magnetization processes of small (350 Å) and nearly spherical particles of -Fe₂O₃ were studied at room temperature using the initial, hysteresis and remanence magnetization curves. The effective interparticle interactions were estimated for a number of virgin and alternating current demagnetized samples of -Fe₂O₃-SiO₂ mixtures with different magnetic phase concentrations. A non-monotonous dependence of the interactions on the magnetic phase concentration was observed. It was related to the changes in the domain state of some of the particles from the single-domain to the multi-domain.

Some theoretical and experimental studies are presented on magnetostrictive anisotropy (stress induced anisotropy) in a long circular steel bar. An explicit analytical solution to the problem of uniform axial alternating current (ac) distribution in a cylindrical bar which retains magnetic and electrical anisotropy was established. By relating the stress level to the amount of magnetic anisotropy in the material, an electromagnetic stress model was then developed for cylindrical bars under uniaxial stress using the analytical solution. For the application of this model, experimental stress measurements on mild steel bars with different radii were conducted based on the alternating current potential difference (ACPD) technique. During the test, a form of alternating current field measurement (ACFM) instrumentation, known as the ACFM crack microgauge, was used to inject a uniform ac current into the steel bars and to measure the electrical field on the bar surface. The ACPD stress measurement results showed that the theoretical stress model could predict the stress values with the calibration of a magnetic anisotropic parameter.

A plume consisting of vapour and ionized particles from the workpiece is commonly formed during various types of laser materials processing. The process parameters such as the laser power, spot diameter, scanning speed, material properties and shielding gas affect the properties of the vapour-plasma plume. A mathematical model is presented in this paper to predict the plasma properties, such as its temperature and absorption coefficient, and the partitioning of laser energy between the plasma and workpiece for different process parameters. The effect of plasma on the surface temperature of the liquid metal and the vaporization rate are modelled using the Stefan condition at the liquid-vapour interface. A new experimental technique named as the pinhole experiment is presented in this paper to measure the partitioning of laser energy between the plasma and the workpiece.

In this work we consider the interaction of a laser-generated cavitation bubble with a solid boundary for the case of a cavity created very close to the wall. Using a combination of a thin-film transducer placed on the surface of the boundary and schlieren photography techniques we observe the induced pressure stresses on the solid boundary and in the surrounding fluid. By studying bubble shapes and identifying the formation of a liquid jet and shock wave emission, we speculate on the dominant pressures stresses induced by the bubble around the time of its first minimum volume for this chosen creation site of the cavity.

Although the method based on the Mueller matrix for the experimental determination of optical-device polarization behaviour is a powerful tool, it has rarely been applied to optical fibre. This paper introduces an experimental methodology for measuring the Mueller matrices of monomode fibre under uniform strains. Using a theoretical model derived from coupled-mode equations, we were able to first estimate the physical parameters of the fibre, then to use them to both test the model validity and assess their influence on the polarimetric behaviour of the fibre.

Conductive-radiative heat transfer through an inhomogeneous, anisotropic scattering, turbid plane-parallel medium is considered. The medium is taken to be of diffuse and specular reflecting boundaries. The specular reflecting coefficients of the boundaries are considered to be angular dependent. The variational technique is used to solve the radiative problem while an iterative method is taken to include the nonlinearity effect of the temperature distribution of the medium from the conductive energy equation. The dimensionless temperature distribution and conductive, radiative and total net fluxes through the medium are calculated. The calculations for matched boundaries and a homogeneous, isotropic scattering slab are calculated and compared with other calculations to show good agreement. The calculations are carried out for mismatched boundaries, with homogeneous and inhomogeneous media. The results are given for isotropic and forward linear anisotropic scattering slabs.

Based on a previously developed electron cascade model, a study is performed to investigate the electron kinetics in the breakdown of argon under two sets of experimental conditions in which argon over a pressure range 10-3 × 10³ Torr is irradiated with focused beams of laser radiation of wavelengths 0.53 µm and 0.248 µm and pulse duration 15 ns and 18 ns, respectively. The model takes into account all the possible electron, atom and photon interactions. The calculated breakdown threshold intensities are found to be in accordance with the measured ones over the whole pressure range for = 0.248 µm. However, the agreement for = 0.53 µm was poor below 200 Torr. Moreover, the study of the electron energy distribution function and its parameters revealed the competing role of multiphoton and cascade collisional ionization mechanisms against loss processes over the pressure range examined in this analysis.

Argon gas breakdown in RF and dc combined electric fields (dc-biased RF field) is investigated experimentally and theoretically for two parallel electrode systems of bare metals and glass-covered metals. To simulate breakdown, electron and ion density equations based on a charged particle balance model are numerically solved taking account of the secondary electron emission according to different electrode surface conditions. The calculated breakdown voltages are in agreement with the data observed with the bare electrode. In the insulated electrode system, RF breakdown is achieved with a time lag after application of RF and dc voltages to the electrode and the RF breakdown voltages at fixed pressures decrease with the time lag while increasing with the dc voltage. The generation of electric field by the charge accumulation on the glass surfaces is considered to affect the RF breakdown. In the case of short-time-lag breakdown for which the charge accumulation is ignored, the RF breakdown voltage increases with the dc voltage and then saturates at higher dc biases in contradiction to the prediction that the RF voltage should increase continuously.

We used a pulsed Townsend technique to measure the effective ionization coefficients, electron drift velocities and positive ion drift velocities in methane-argon mixtures over the combined density-normalized electric field intensity, E/N, range from 0.05 to 700 × 10^-17 V cm². The mixture studied contained 0.5, 3, 25, 50 and 75% CH₄, including pure methane and pure argon. We found a well defined dependence of the effective ionization coefficient on the amount of CH₄ in the mixture at low E/N, while at the higher E/N end, all the curves merged into a single one. The electron drift velocities in the mixture show a pronounced negative differential conductivity region, the maxima and minima of which depend on the mixture ratio.

The effect of near-sheath dusts on the rf power loss in a surface-wave-sustained gas discharge is studied. The planar plasma is bounded by a dielectric and consists of an inhomogeneous near-wall transition layer (sheath), a dusty plasma layer and an outer dust-free plasma. The discharge is maintained by high-frequency axially symmetrical surface waves. The surface-wave power loss from the most relevant dissipative mechanisms in typical discharge plasmas is analysed.

Low-current self-generated oscillations in a rectangular hollow cathode discharge in helium gas were investigated experimentally and by means of a two-dimensional self-consistent hybrid model. The model combines Monte Carlo simulation of the motion of fast electrons and a fluid description of slow electrons and positive ions. The low-frequency (20 kHz) oscillations were found to arise as an effect of the interaction of the gas discharge and the external electric circuit - consisting of a stable voltage source, a series resistor and a capacitor formed by the discharge electrodes. Good agreement was found between the experimentally observed and calculated oscillation frequency and current wave forms. Beside these characteristics the modelling also made it possible to calculate the time dependence of numerous other discharge characteristics (e.g. electron multiplication, ion density, potential distribution) and provided detailed insight into the mechanism of oscillations. The advantage of the present model is that considerable deformation of the electric field (due to space charge accumulation) can be taken into account.

La mesure du rayonnement du fond continu d'un plasma permet de déterminer la densité électronique sans faire d'hypothèse sur l'état d'équilibre thermodynamique du milieu, à condition de connaître les facteurs correctifs de Biberman correspondant aux différents continuum de recombinaison. En particulier, dans un plasma d'argon à haute température (T_e>18 000 K), la contribution à l'émission continue de la recombinaison des atomes d'argon deux fois ionisé (Ar III) doit être prise en compte. Malheureusement, il n'existe dans la littérature que très peu de données relatives au facteur ^{Ar III} pour les ions Ar III. Mettant à profit le diagnostic précis de la zone cathodique d'un arc électrique fonctionnant dans l'argon, nous avons mesuré pour la première fois ce facteur pour = 468,8 nm à une température T_e~25 000 K.

A simulation technique for the analysis of the transverse evolution of electron swarms in gases was developed based on moment equations derived from the Boltzmann equation. A numerical calculation of the moment equations for an electron swarm was performed using a propagator method and it was demonstrated that the propagator method can be used to calculate the higher-order transverse diffusion coefficients stably. Applying a Hermite expansion technique, the electron distribution in real space and other electron swarm parameters were derived as functions of the transverse position. The calculation result was verified by comparisons with those by a Monte Carlo simulation and other methods. Features of the transverse electron swarm evolution were presented.

SF₆/N₂ mixtures with a majority of nitrogen are currently highly recommended, at the international level, in gas insulated transmission lines as an alternative to pure SF₆; indeed, these mixtures are much more friendly to the atmosphere and particularly cheap. Among the areas of investigation of such gas mixtures, their electrical decomposition as a function of impurity content and type of discharge must be studied. The present study concerns the decomposition rate of SF₆ and SF₆/N₂ (10:90) mixtures at 400 kPa under negative and 50 Hz alternating current corona discharges carried out without and in the presence of 0.3% H₂O added, up to 10 C of transported charge. The corona discharges were generated at 23 °C in a 340 cm³ experimental cell between a stainless steel point (radius of curvature 10 µm) connected to high voltage and a plane layer of aluminium (gap space 2.3 mm for pure SF₆ and 3.4 mm for SF₆/N₂).

The gaseous decomposition products SOF₄, SO₂F₂, SF₄+SOF₂, SO₂, S₂OF₁₀, S₂F₁₀, S₂O₂F₁₀, S₂O₃F₆, SF₅NF₂, NF₃ and (SF₅)₂NF were assayed by gas chromatography at the end of each run.

The comparison of the formation rates of the byproducts detected in SF₆ and SF₆/N₂ mixtures led to the following conclusions.

(i) The use of wet SF₆/N₂ presents an additional advantage with respect to pure wet SF₆ besides those mentioned above; the production rates of all the usual SF₆ decomposition products are much lower with the wet SF₆/N₂ (10:90) mixture than with wet SF₆. (ii) More exactly the addition of 0.3% H₂O to SF₆/N₂ greatly affects the production rates of all the compounds whose formation needs SF₅ radicals as SOF₄; so the lower quantity of products formed in SF₆/N₂ is in fact more due to the particularly inhibiting effect of water than to the low percentage of SF₆ in the mixture. (iii) The production of NF₃, SF₅NF₂ and (SF₅)₂NF remains very low compared to the other gaseous byproducts.

A new experimental set-up has been designed to observe spectral studies in fuse arcs. The fuses studied here are high breaking capacity fuses.They consist of a fuse element (in silver or copper) surrounded by silica sand. In this study, the arc dissipates 1200 J, the peak of electric current is about 2000 A and the duration of the arc is 3 ms. The maximum cross section of the squeezed arc is about 0.5 mm × 5 mm. The whole of the visible spectrum has been registered, from 360 nm to 800 nm, in the exploding part of the arc. The spectrum consists of continuous light together with spectral lines. The continuous light intensity is closely correlated with the current intensity. The spectral lines appear mostly when the current decreases. The behaviour of the spectra shows that there is a big gradient of species concentrations in the arc, together with a high gradient of temperature. The temperature is measured from Si II lines in the arc core. The calculation gives about 30 000 K at 1.3 ms. The measured values of electronic densities vary between 10¹⁷ and 10¹⁸ cm^-3.

An analysis of six different definitions of the rate of reaction is made and it is shown that some of the definitions are not consistent with the mass action law. Kinetic equations are derived whose solution is the time dependence of the composition of an arbitrary closed gaseous system in an ideal state for a given time dependence of temperature on the assumption that either the time dependence of pressure in or the time dependence of the volume of is given. Results are given of the computation of composition of a system of SF₆ dissociation and ionization products at a temperature drop from 12 000 K to room temperature and a comparison is made of the time-dependent composition with equilibrium composition for pressures of 0.1, 0.5, 1 and 2 MPa.

So far in the literature there mainly exist two quite different forms of the mass action law or the Saha equation modified to a two-temperature plasma. It has been shown in the present paper that the modified Saha equation deduced from the minimum Gibbs or Helmholtz free-energy principle is incorrect. The minimum Gibbs or Helmholtz free-energy principle, which is valid for a single-temperature system as a substitute of the maximum entropy principle in an isolated system, cannot be extended to the two-temperature plasma. A more rigorous thermodynamic derivation of the Saha equation modified to a two-temperature plasma system is presented based on the basic thermodynamic principle that the entropy of an isolated system will assume its maximum at the equilibrium state. It is pointed out that one has to pay close attention to the chemical potential expressions for the case of the two-temperature plasma, since any negligence may cause mistakes in the derivation of the two-temperature Saha equation.

A series of sterically-stabilized polystyrene latex particles in the size range 0.1-5.0 µm have been coated with ultrathin (<50 nm) overlayers of either polypyrrole, polyaniline or poly(3,4-ethylenedioxythiophene). In each case the conducting polymer overlayer allows the latex particles to acquire surface charge and hence be accelerated up to hypervelocities (>1 km s^-1) using a Van de Graaff accelerator. These coated latexes have two main advantages compared to the sterically-stabilized polypyrrole particles of 0.1-0.3 µm diameter reported previously. First, a wider particle size range can be accessed. Second, the particle size distributions of the coated latexes are much narrower than those of the pure polypyrrole particles reported earlier.

Preliminary studies confirm that, after charging and acceleration, these conducting polymer-coated latex particles have very similar mass-velocity profiles to those reported for colloidal iron particles in the hypervelocity literature. The hypervelocity impact generated ionization has been measured for latex spheres impacting copper targets. This is compared to previous work for impact ionization by iron particles, thus demonstrating the ability to study the dependence of impact ionization on widely different projectile materials.

In order to investigate single electron transport the lateral dimensions of a resonant tunnelling diode have been confined to the submicrometre range by oxygen ion implantation. Depending on the temperature sequence used in the annealing step subsequent to the implantation, these small area resonant tunnelling diodes in the single electron regime either show clearly developed staircase-like features as a consequence of single electron tunnelling through discrete zero-dimensional states or a sequence of current peaks due to single electron tunnelling through coupled zero-dimensional states. Magnetotransport measurements suggest impurity states related to implantation defects as an origin of the zero-dimensional states.

An expression for the electron-wave phase shifts caused by the potential of an electric dipole is extended to a general equation representing the phase shifts caused by the electrostatic potential across a parallel-plate capacitor. The capacitor is then used to construct a simple model for the phase shifts caused by the potential barriers present across internal interfaces in solids. The phase simulations yields phase shifts for electroceramic interfaces in good agreement with the phase images reconstructed from off-axis electron holograms of acceptor-doped grain boundaries in SrTiO₃.

The paper deals with the development of a mathematical model for the calculation of melting rates obtained in gas-shielded arc welding with a multiple-wire electrode or in submerged arc welding with a multiple-wire electrode. The first part provides a very general and short description of welding with a multiple-wire electrode, the main advantages and characteristics, as well as variants of welding with a multiple-wire electrode applicable in practice to various cases. The major part of the paper treats the development of the mathematical model for the calculation of the melting rate on the basis of the physical principles of the welding arc and of the wire extension heating due to current conduction and mutual influence of the welding arcs. Finally, a comparison is made between the melting rate results obtained by practical measurements and those obtained theoretically by the mathematical model. The mathematical model for the calculation of melting rates in welding with a multiple-wire electrode is accurate enough to be used in practice and for further studies.

The photobleaching of the lasing dye Rhodamine 6G embedded in the solid matrix poly(methyl methacrylate) was investigated using a photoacoustic technique. Chopped laser radiation from an argon ion laser at four different wavelengths was used for the study. Experimental results indicate that the photobleaching rate is directly proportional to the incident laser power while it decreases with increase in concentration of the dye molecules. In the present case we have not observed any dependence of photobleaching on the chopping frequency. One-photon absorption is found to be responsible for the photobleaching of the dye within the selected range of laser power.

It is shown that a mechanical stress relaxation spectrum may be obtained directly from measurements of the dispersion of shear waves in viscoelastic media. The effectiveness of this spectrum, in terms of its sensitivity and accuracy in parametrizing distributions of stress relaxation in the frequency domain, is demonstrated with reference to several published spectra. Of the three different forms of stress relaxation spectrum which may be obtained directly from oscillatory shear measurements, only that based on the group and phase velocities of shear waves is found adequate in terms of sensitivity to the presence of adjacent relaxation mechanisms in the systems considered herein.