Table of contents

The geometrical magnetoresistance scattering factor has been measured for vapour-phase epitaxial GaAs in the temperature range 77 to 400 K. These measurements eliminate experimental difficulties associated with other determinations. The factor is approximately unity throughout the temperature range and is 0·96±0·08 at 300 K, which is in fair agreement with theoretical expectations.

In a recent paper in this journal, Hwang presented data concerning two high-temperature thermoluminescent (TL) peaks and related optical absorption bands in LiF. Unfortunately, Hwang's analysis is based on certain misunderstanding of information in the literature.

Knowledge of the influence of magnetic fields on the electron energy distribution function is applied to the problem of establishing values for the equivalent pressure pprime that would represent the change in the transport properties of electrons in hydrogen gas at pressure p subjected to a uniform electric field E and a transverse magnetic field B.

The importance of changes in the form of the distribution function due to the magnetic field is established and a re-evaluation of the significance of the effective collision frequency is given.

Electrostatic properties of the point-plane electrode configuration are reviewed in the hyperbolic approximation. Simple expressions for the field in the neighbourhood of the tip are derived under the assumption that space-charge distortion is absent. Current induced by a charge carrier moving along a field line is investigated and the theoretical expressions are compared with experimental results. Finally, the problem of the space charge is solved along the tip axis.

The operating characteristics of perforated-mesh hollow-cathode gas discharge electron guns (hc guns) have been investigated paying particular attention to the energy spectrum of the electrons forming the beam. While it was found impractical to attempt a complete systematic study, sufficient quantitative data was obtained to apparently confirm certain findings made by Boring and Stauffer in 1963 relating to the potential of the plasma within the cathode. Further observations made visually help towards a more complete understanding of the operation of the gun.

There are similarities between the spectra produced by this gun and those produced by the hollow-anode (ha) gun. The most obvious difference is that electrons from the hc gun lie within one of two distinct energy ranges, both of which are often present, while in the case of the two-electrode ha gun only one range is present. Of these two energy groups one is associated with the full cathode to anode potential difference while the other seems to be the result of acceleration from the internal cathode plasma potential to anode potential.

It has been found that similar, but not identical, spectra can be obtained from an ha gun modified to include a third electrode at a potential near that of the cathode. This also indicates that the electrons from the hc gun are produced at two different sites and that the mechanism which is necessary for the production of electrons in the ha gun also operates in the hc gun.

An exact solution is obtained for the trajectory of an electron moving in a magnetic field which is inversely proportional to the distance of the electron from a fixed plane.

A new hollow-cathode effect has been observed in the residual plasma of a mercury arc discharge. A pair of negatively biased plane parallel electrodes were found to give rise to an orderly sequence of high frequency electrical bursts. The bursts made negligible contribution to ionization in the plasma. Similar bursts were subsequently observed in a self-sustained hollow-cathode discharge. The timing of the bursts and their variation with pressure in the range 2 to 15 mTorr are in good agreement with a simple theory based on multiple-mode resonant interaction of secondary electrons with a decaying plasma. In the present tube, the frequencies of the oscillation within each burst are predicted to lie between 100 MHz and 20 GHz.

The development of high-current transient glow and arc discharges over a wide overvoltage range in a 3 cm uniform-field gap has been investigated in nitrogen, oxygen and mixtures of these gases at sub-atmospheric pressures, using an `Imacon' image converter and an EMI 4-stage image intensifier. Streak photographs have been obtained which show clearly the propagation of luminous fronts in both anode and cathode directions during the development of ionization from the prebreakdown stage of the discharge to the final arc.

In a dielectric liquid with high applied voltages, the current-voltage relationship is known only in two extreme cases: emission-limited injection and space-charge-limited injection. In an actual case, such as in the electrofluiddynamic generator, neither situation is true. In this paper, a more generalized theory of which the extreme cases lead to the above limiting cases is developed. Comparison with the experimental results of Tobazéon showed good agreement.

We make an analogy with the scattering of electromagnetic waves by a randomly rough surface and obtain a scattering kernel for gas atoms reflected by a surface. The functional form of this kernel is identical with that obtained recently by other authors who have based their work either on the Fokker-Planck equation or simply intuition. The present work demonstrates that the constraint of detailed balance restricts very severely the class of functions that can be used for semi-empirical representation of experimental gas-surface interaction data.

The conditions of growth for optical quality single crystals of cadmium and silver thiogallate are described. Some information about the growth of two other thiogallates is presented and the growth faults observed in all the crystals grown are discussed.

A general analysis is given for the determination of the effective elastic constants of a macroscopically isotropic and homogeneous two-phase composite material. Approximate formulae are given and their validity discussed in terms of the general features of the sample geometry.

In 1967 Stoneham showed how the width of a zero phonon line may be determined from the strain field of point defects in the crystal. The direct application of the theory is difficult because of the need to evaluate numerically a two-dimensional integral. It is shown here how the integral may be trivially evaluated to within a few per cent for any zero phonon transition between non-degenerate electronic states. The method exploits the change in linewidth which results from the defects being forbidden to approach closer than some minimum distance to the optical centres. This change is found to be approximately the same for all transitions between non-degenerate levels.

Electrical conduction is studied in polyethylene terephthalate and to a limited extent in polyethylene over a range of fields up to 1·2×10⁶ V cm⁻¹. In common with previous investigations, it is found that the current decays with time according to an inverse power law, but this can be obviated by a conditioning procedure which then allows steady values of current to be measured over a range of temperature and field. The results suggest that the current has an activation energy which is field dependent, but attempts to fit the results to a Schottky-type law are not very successful.

By adopting a more generalized form of potential barrier than the coulombic one usually chosen in the Schottky law, it is possible to obtain consistent agreement with experiment. The form of potential barrier chosen appears to refer to the cathode-dielectric interface and is probably determined by space charge in the dielectric. Values of the metal-dielectric contact potential for the samples of polyethylene terephthalate and polyethylene investigated were 2·58 and 2·14 eV respectively for aluminium electrodes. It is suggested that the time dependence observed before conditioning is completed is due to the slow build-up of the full interfacial barrier.

An approximate method of estimating hysteresis losses in hard super-conductors of complex geometry is described. It is based on approximating contours of current penetration by polygons and adjusting the polygons so that the vector potential is the same at all the corners; this corresponds to zero flux crossing the contour between corners. Two examples are worked: for a square-sectioned conductor and for a pair of closely spaced thick strips carrying antiparallel currents.

Several models have been proposed for the intermediate state which occurs when superconductivity is destroyed in type I wires by a current. Experimental testing of these models has, so far, only been possible through measurements of the resistance and the intermediate state core radius during the transition. We describe an experiment in which a wire specimen is placed along the axis of solenoid whose inductance is monitored as a steadily increasing direct current destroys superconductivity in the wire. The change in inductance of the solenoid is a function of the intermediate state structure, and the experimental results are compared with theoretical values obtained on the basis of the various models. Our results appear to favour the model presented by Baird and Mukherjee in 1971.

A simple method is described for the measurement of surface-state density and surface potential in a MIS capacitor. It is similar in principle to the low frequency capacitance method in so far as a dc relationship is established between the surface potential and the voltage applied to the gate electrode. This is achieved by measuring the saturation surface photovoltage. The method differs from the low frequency capacitance method in that it measures the surface potential directly, requiring no additive constant. Experimental results are presented which agree well with those obtained by the low frequency capacitance method.

The present method is complementary to a second method to be described by Lam and Rhoderick (companion paper) in that the surface-state density measurable by this first method has a lower limit while that by the second method has an upper limit, the two overlapping at the value of the space-charge capacitance.

This paper describes a surface photovoltage method which is complementary to the one reported in part I. The present method relies on the fact that a change in illumination level causes a change in electron occupancy of some of the surface states. It involves only moderate light intensity and yields results for the surface-state density from the slope of the surface-photovoltage against log-of-light-intensity curve. The limits of surface potential within which the method is valid as well as the range of surface-state densities measurable will be examined. Results are presented and are compared with those obtained by other methods. The possible advantage in measurements at low temperatures is also discussed.

Projections formed on tungsten emitters, as a result of an applied electrostatic field and the adsorption of contaminant gases, are believed to have relevance in the study of vacuum breakdown. By examining the energy spectrum of the electrons, emitted from single projections, temperatures at the projection ends have been estimated to exceed 2800 K on the basis of arguments which assume the applicability of the free-electron model to the emission process. The energy spectrum was obtained with the aid of a retarding potential energy analyser. Cathodic projections are found to have resistivities (between 0·55 and 2·2 × 10⁻³ Ω m) and positive temperature coefficients of resistance indicative of a metallic nature, unlike anodic ones. Cathodic projection growth is retarded by low temperatures or by previously subjecting the emitter to very high electrostatic fields in the absence of residual gases. The shape of the cathodic projection (needle or dendritic) is influenced by the nature of the contaminant gas; electron micrographs and field ion microscope patterns of the two types were obtained. The growth mechanisms of the anodic and cathodic projections are briefly discussed.

Experiments concerning dynamic aspects of threshold switching in devices employing chalcogenide glasses are compared with `computer experiments' based upon a simple one-dimensional thermal theory. Despite the inadequacies of the one-dimensional description, a good semiquantitative agreement with experiments in high and intermediate temperature regions suggests that the thermal mechanism of switching is operative under these conditions. Departure from the predictions of the thermal model at low temperatures (less, similar - 70°C) is pointed out and discussed.

Using the theory of Walker the eigenfrequencies are calculated for the magnetostatic modes of a ferromagnet containing a spherical cavity. The solutions obtained for the non-uniform magnetization are localized around the cavity. The resonance frequencies are, with one exception, all higher than that of the uniform mode ω₀, but they approach ω₀ at high values of the applied static magnetic field. Equations are given for ellipsoidal cavities.

A new method for determining optical constants n and k, for thin absorbing films supported by a dielectric substrate, has been devised. Measurement of the ratio of two reflection coefficients R_parallel/R_{perpendicular} is made at two angles of incidence, and the film thickness d is determined independently. Optimum angles of incidence for experimental measurement have been determined for 1·0less-than-or-eq, slantnless-than-or-eq, slant4·0 and 1·0less-than-or-eq, slantkless-than-or-eq, slant4·0 for many optical film thicknesses d/λ in the range 0·01less-than-or-eq, slantd/λless-than-or-eq, slant0·25. The method, when used at these optimum angles, yields n and k accurate to ±0·05 for values of d/λgt-or-equal, slanted0·05, providing angles are measured to better than ±0·1°, reflection ratios are measured to better than ±0·01 and values of d/λ are measured to better than ±0·005.

The electron spin resonance characteristics of fibres derived from a polyacrylonitrile based precursor heat treated to between 1000 and 2800°C have been examined. The resonance is in general similar to that observed in graphitizing carbons and has been analysed by methods developed for the latter materials. One striking feature of the results is that a g-value anisotropy is observed when the processing temperature exceeds approximately 1750°C. The appearance of this anisotropy is associated with discontinuities in several other parameters. The resonance intensity decreases rapidly up to 1750°C then much more slowly above this temperature, and it would appear that the localized component in the resonance is constant above this transition point. Various mechanisms for the appearance of the anisotropy are discussed including increased perfection of the layers in the crystallites and the existence of a separate three-dimensionally ordered graphite phase. It is suggested that the elimination of nitrogen from the carbon structure may play an important part in the changes which occur in this region. The g-anisotropy seems to be a reliable and sensitive measure of the degree of graphitization in these fibres and its magnitude is consistent with the limited degree of graphitic crystallite development indicated by x ray methods.

Homogeneous nickel - cobalt binary alloys were analysed in the electron probe microanalyser to determine the magnitude of the atomic number and K_β-K_α secondary fluorescence corrections. In these alloys, the atomic number effect is due almost entirely to the ionization-penetration factor. A relative error of 4-6% between the nickel intensity ratios and the true composition was found in the high cobalt alloys. This error was an additional 2-3% greater than that predicted by the Philibert and Tixier atomic number correction. When used with the Philibert and Tixier correction none of the well-known expressions for the mean ionization potential were able to account for the difference between the experimental and predicted values. Secondary fluorescence of Co K_α radiation by Ni K_β radiation was significant in the high nickel alloys, causing relative errors of up to 4% and overriding the atomic number effect. The Reed fluorescence correction did not completely correct for this error.