Table of contents

The energy spectrum of atomic ions emerging from a gas-filled chamber under the influence of a strong uniform axial electric field is investigated. Experiments were carried out with Ar⁺ ions in Ar at different gas pressures. Variation of ion energy distribution with pressure was measured. Observations covered the collision-free regime and extended into the collision-dominated regime.

By means of Townsend and Huxley's technique (1974) the ratio of lateral diffusion coefficient to mobility, D/ mu , in oxygen and dry air has been determined at ambient temperatures in the ranges of the reduced electric field 160<or=E/N<or=360 Td and 0.9<or=E/N<or=280 Td respectively. The present results have been compared with those obtained by other investigators.

Using a sensitive capacitance technique, the changes in length ( Delta l) of NaCl single crystals with temperature have been measured in the range 30-300 degrees C. These measurements have been taken on NaCl in the as-cleaved and later X-ray irradiated conditions; similar data are also given for quenched and X-ray irradiated NaCl. Delta l as a function of temperature in X-ray irradiated crystal exhibits two regions: (i) 30-170 degrees C, where the linear coefficient of thermal expansion alpha is 47*10^-6 K^-1); (ii) 170-300 degrees C where alpha is 40.5*10^-6K^-1 (the alpha of as-cleaved NaCl is 40.2*10^-6K^-1). Similar behaviour is shown by NaCl, which was quenched and later X-ray irradiated, but the first region seems to exist up to 150 degrees C, beyond which the second region takes over. The concentration of vacancies in X-ray irradiated NaCl at room temperature is calculated to be 4.5*10¹⁷ cm^-3. An attempt has been made to understand the results.

Measurements are presented of the transverse conductivity of TTF-TCNQ at room temperature under X-ray irradiation conditions. The conductivity decreases with irradiation dose without any saturation effect.

Thin neodymium-cobalt films prepared by vacuum evaporation were examined by transmission electron microscopy and defocused Lorentz microscopy. The examined films were polycrystalline with crystallite sizes about 40 AA and possessed magnetisation ripple structure. Interpretation of ripple in the framework of Hoffman (1964) and Harte (1968) linear ripple theory is presented.

A pure tin target was sputtered in a deliberately impure atmosphere to evaluate the discharge products using a differentially-pumped quadrupole mass analyser. Hydrogen increase was due to dissociation of water vapour, carbon monoxide and dioxide were generated by reaction between oxygen and hydrocarbons, and these discharge products were reduced by the gettering reaction of tin at a high discharge current density. It was estimated that tin oxide film prepared by argon-water-vapour sputtering had little contamination with carbon, compared with the film prepared by argon-oxygen sputtering.

A new method is proposed to calculate the expansion coefficients of separation of the variables for solving boundary-value problems of electromagnetic fields. It is a modification of Read's method (1969). The procedure for numerical calculation is explained by an example of a problem of electromagnetic fields. The results are compared with the ones by Read's method.

Stimulated Raman and Brillouin scatterings have been investigated in the presence of a magnetic field parallel to the direction of propagation of the pump electromagnetic wave. Fluid equations with phenomenological damping factors have been used for the evaluation of thresholds and growth rates. When both right and left circularly polarised pump modes are present the nonlinear interaction between the scattered circularly polarised waves introduces interesting modifications. Sharp increase in thresholds and similar behaviour in growth rates even for moderate magnetic fields is apparent. Effects due to the self action of the pump electromagnetic wave have been pointed out. Relevance to fusion studies has been discussed.

Error contour charts are presented which compare the approximate Fraunhofer diffraction theory with the exact solution of Maxwell's equations for infinitely long cylinders of circular cross-section. The comparison is made over angular ranges which extend well beyond the forward scattering lobe. For light of wavelength 632.8 nm the two theories agree well for absorbing cylinders as small as 4 mu m diameter. However, for non-absorbing cylinders having diameters less than 20 mu m the agreement is poor. The implications of these results to fibre sizing are briefly discussed.

During the normal impact of a hard particle on a metal surface, some of the kinetic energy of the particle is radiated into the target as elastic waves. The amount of energy involved is calculated for impacts where the metal is deformed plastically, and is found to be typically only a few per cent of the initial kinetic energy.

A simple model for the minimum obliquity required to induce ricochet of a high speed long rod projectile from a thick target plate is suggested.

Calculations have been made of the effect of absorbing electrodes on a continuous stream of electrons travelling in a gas in a uniform electric field. The effective electron drift and diffusion coefficients become a complex function of position, so that the electron density distribution is no longer given by the solution of the electron continuity equation using the usual equilibrium drift velocity and either the transverse or the longitudinal diffusion coefficients appropriate to the applied electric field and gas pressure. Calculations for a momentum transfer cross-section proportional to electron speed indicate an increase of the average electron energy of about a factor of two at the anode. Predictions using Monte Carlo methods and solutions of the Boltzmann transport equation are in good agreement. Monte Carlo calculations are also applied to predict the influence of absorbing boundaries on the Townsend-Huxley experiment (1974). Predicted current ratios compared with ratios obtained from a solution of the continuity equation lie between ratios obtained using both isotropic and conventional transverse and longitudinal diffusion coefficients.

The electron swarm development in SF₆ gas is studied for E/p₂₀ values from 15 approximately 200 V cm^-1 Torr^-1 (E/N=45.6-607 Td) by a Boltzmann equation method in which the effect of ionisation and electron attachment is considered properly. The momentum transfer, vibration, ionisation and attachment cross-sections are taken from experiments, but the electronic excitation cross-section is determined by fitting the calculated ionisation and attachment coefficients to previous measurements. The calculation is performed for the steady-state Townsend, pulsed Townsend and time-of-flight experiments. The results show that the deduced electron drift velocities and diffusion coefficients agree reasonably well with experiment and the value of a swarm parameter should depend on the type of experiment not only at high E/p₂₀ values, at which ionisation is appreciable, but also at low E/p₂₀ values, at which electron attachment is appreciable. The electron energy distribution, which is found to have depressions at low E/p₂₀ values, the electron mean energy and the excitation coefficients are also calculated and discussed.

The boundary layer integral method of analysis has been used to study the DC arc burning in the convergent-divergent nozzle of a gas blast circuit breaker. A particular case where the arc completely 'fills' the nozzle has been studied, since it is of special interest to designers of nozzles for circuit breakers. It is shown that for a given nozzle geometry the arc model is capable of predicting the current density at the nozzle throat where the mass flow approaches zero. This current density depends on the properties of the arc quenching medium, nozzle geometry and the stagnation pressure. Results are given for a particular nozzle shape, but the method of analysis is general.

A type-A vacuum arc on copper electrodes was subjected to an axial magnetic field up to a strength of 0.35 T. The arc current was about 500 A, with an arcing time of about 30-35 ms. A study of the dissipation of the extra energy input caused by the magnetic field was carried out, and the mass change of both electrodes at different field strengths was measured. The energy input to both electrodes increased with the magnetic field strength, and the input to the anode was a factor 1.7-1.9 greater than the energy collected by the cathode. At high field strengths the anode surface suffered gross melting due to anode spot formation. Probably, energetic ions are emitted from the anode surface, resulting in additional heating of the cathode.

Analytic expressions are derived to give approximate properties of free-burning arcs such as occur in switch-gear, arc welding and electric arc furnaces. Expressions are given for the arc voltage electric field, arc radius and plasma velocity in terms of the arc length, current, gas pressure and temperature. The energy balance equation relates arc temperature with current. The expressions involve the material functions of the arc plasma, so that predictions can be made for any arc medium whose material functions are known. Properties of free-burning arcs for low currents, for example, less than 30 A for air, are controlled by natural convection, it being assumed that the arc is vertical. At higher currents properties become controlled by convection induced by the self-magnetic field of the arc. Although gross simplifications are made, predictions for arcs in air for currents ranging from 1 to 20000 A are in fair agreement with experiment for a position 1 cm from the lower electrode.

Evidence is now brought forward to show that the reflection of evan a small amplitude pressure pulse at a free surface is accompanied by considerable distortion. An explanation of this fact, and of the low values of tensile strength of water inferred from underwater explosion and similar 'once only' pulses, is suggested in the light of existing ideas about the structure of a liquid surface. It is concluded that there is now no real disagreement between experiment and theory on tensile strength as there are now acceptable explanations of the low values found from ultrasonic and 'once only' pulse methods.

Both in-diffusion and out-diffusion experiments for Cr in GaAs have been performed, using radiotracer Cr so that profiles could be plotted. An analysis of the profiles shows that they can be explained on the basis of a model in which the Cr diffuses through the lattice interstitially. A simple interpretation of the solubility data implies that the Cr occurs on the group V site, in contradiction to previous work on the system. An attempt is made to resolve this paradox by considering the possible form of the Ga-As-Cr phase diagram at the temperatures used in these diffusions. Electrical measurements were also made on Cr-diffused samples. Interpretation of these measurements was difficult because conversion of the samples to p-type occurred during the experiment. It is demonstrated that this conversion was due to the heat treatment involved in the diffusion rather than to the introduction of Cr into the crystal. The electrical measurements are consistent with the idea that Cr introduces a deep acceptor level into GaAs.

Electrical ionic conductivity studies in superionic KAg₄I₅ thin films show good agreement with the best bulk studies. The activation energy for Ag⁺ ion conduction and the free-ion-like parameters such as mean free path and the lifetime in the excited state have been deduced from the conductivity plots. Studies on the compositional dependence of the electrical conductivity of the KI-AgI system reveal that the conductivity reaches a value of 0.25 Omega ^-1 cm^-1 for films with 80 mole.% of AgI. Electronic spectra of these films show a well defined exciton state 3750 AA at 300K. A forbidden internal transition in the Ag⁺ ion has been observed at 4250 AA. The existence of silver colloids in the films has been revealed through a broad band around 5000 AA.

Schottky barriers have been formed by vacuum evaporation of Cu on to CdS thin films. The behaviour of these samples has been investigated as a function of time and annealing by standard electrical methods: current-voltage analysis, capacitance-voltage analysis and analysis of the spectral dependence of the photoemission currents. The impurity profile deduced from the reverse differential capacitance shows evidence of copper diffusion occurring between 20 and 200 degrees C. An activation energy of 0.72 eV is found for the temperature dependence of the diffusion coefficient. The Cu-CdS interface has also been investigated by looking at the photovoltaic mechanisms in connection with the different heat treatments. Capacitance measurements performed under junction illumination have been used to obtain the true donor density after copper diffusion.

The authors investigate the influence of the time-dependent external conditions on the stability criterion of the critical state in a hard superconductor with respect to flux jumps. It has been shown that the existence of a non-linear region in the current-voltage characteristic can result in a significant delay of the onset of the instability. The methods developed by the authors make it possible to define the dependence of the stability criterion on the rate of change of the external parameters (magnetic field, temperature etc). The total theoretical results have been compared with the experimental data.

Measurements of magnetisation, Matteucci effect, and inverse Widemann effect have been performed on Fe₄₀Ni₄₀P₁₄B₆ amorphous ribbons under torsion. Influence of the torsional strain on the magnetic properties of the sample is analysed in two ways. Information about the distribution of the easy directions of magnetisation has been obtained from remanence measurements of magnetisation. Matteucci effect and inverse Widemann effect. The dynamic evolution of magnetisation has beeen deduced from coercive force determination in DC longitudinal and transverse of 'circular' fields, as well as from frequency response in AC fields. The magnetic behaviour of twisted amorphous ribbons seems to be caused by the inhomogeneous arrangements of magnetoelastic anisotropy induced by torsion when the latter is high enough to overcome the initial anisotropy distribution present in the sample. Possibilities of technological applications of the observed properties are also discussed.

Some aspects of the hard bubble suppression mechanism have been investigated in LPE grown magnetic bubble garnets, (Y_1.65Sm_0.18Lu_0.21Ca_0.96) (Fe_0.96Ge_4.04) O₁₂. Following implantation of neon ions at a standard suppression dose of 5*10¹⁴ cm^-2 collapse field measurements indicate effective hard bubble suppression. For this dose however, ⁵⁷Fe Mossbauer spectra indicate the pressure of a non-magnetic surface layer while X-ray diffraction traces and alpha particle backscattering measurements indicate that the layer is structurally disordered. These observations are inconsistent with the capping layer model of hard bubble suppression. Implantation of nitrogen ions resulted in similar hard bubble suppression but produced a totally different form of radiation damage. Axial channelling of 800 keV nitrogen ions along the (111) direction is observed.

An investigation of the proton second moment and nuclear spin-lattice relaxation time as a function of temperature in two Fischer-Tropsch waxes having average carbon numbers of 28 and 28.5 is reported. The temperature dependence of the motional behaviour of the chains in the waxes differs significantly from the behaviour in the normal compound n-C₂₈H₅₈. As revealed by the second moment measurements, the rotational-diffusional (ROD) motions of the chains in the waxes set in at temperatures much below the transition to the hexagonal phase. The results are consistent with a model in which the chains of unequal length in the crystalline zones of the waxes are packed in an irregular, staggered arrangement. The onset of the ROD motions in the waxes is preceded by structural phase transitions which occur at temperatures depending on the range of carbon numbers in each wax. Activation energies for the different molecular motions are reported and compared with corresponding values in n-C₂₈H₅₈.

The piezoresistivity coefficients in p- and n-channel MOS transistors manufactured on silicon wafers cut parallel to the (111) and (100) planes as a function of V_D, V_G and of the temperature, between -20 and 120 degrees C, have been measured with the particular aim of investigating the possible application of MOS transistors for strain measurements, by comparing the advantages and the disadvantages with bulk and diffused resistors. The characteristics of the MOS devices can be consistently better than the corresponding ones of bulk semiconductor strain gauges, particularly as regards the temperature stability of the resistivity and of the electrical characteristics.

The thermoreflectance dispersion curves of opaque films of Gd, Tb, and Dy, have been measured at 315K and 97K over the range 1.45 to 3.2 eV. The response at 315K has been shown to be largely determined by the optical functions alpha _p and beta _p which relate the thermoreflectance response to Delta epsilon ₁ and Delta epsilon ₂. When the temperature is lowered from 315K to 97K, pronounced structure develops in the spectra at 1.7 eV for Tb, 2.2 eV for Dy, and 1.6 and 2.8 eV for Gd; weaker structure is also evident at 2.0 eV for Gd. Measurements of the thermoreflectance as a function of temperature show the existence of anomalies at the ferromagnetic Curie point of Gd, and at the Neel points of Tb and Dy.

A correlation has been sought between the microwave performance of indium phosphide transferred electron devices and the metallurgical state of the silver metal-indium phosphide interface that forms the cathode contact in these devices. Both microwave performance and the current-voltage characteristic have been found to depend intimately upon the degree to which a grain structure has developed in the indium phosphide, close to the metal-semiconductor interface. The grain structure appears to act as sites for an alloying reaction involving the silver and the indium phosphide.