Table of contents

Direct current electroluminescent (dcel) powder phosphors in the zinc sulphide: manganese, copper, system can now be readily prepared giving a luminance of 250 ft lamberts at 100 V and 5 mA cm⁻². Under continuous constant voltage operation at 100 V, the luminance falls to 100 ft lamberts in about 5 hours, and subsequently to 50 ft lamberts in about 25 hours. With constant power input a luminance of 100 ft lamberts has been maintained for a period of 400 hours. dcel has also been obtained in zinc selenide: manganese, copper powder phosphors. In addition to the yellow/orange emission due to manganese, green dcel emission has been observed in erbium activated zinc sulphide.

Electron probe microanalysis, associated with electron microscopy, of thin foils of gold-silver alloys has been made using thin foils of pure elements as standards.

Theoretical bases are discussed. Specimens and standards were prepared by vaporization in vacuum and their thickness accurately measured. The anomalous emission effect does not seem to introduce errors in these experiments.

A detailed study has been made of the properties of twisted filamentary superconducting composites. It is shown that such conductors can be used to construct high current density superconducting coils which will reach their full critical current without the necessity for additional stabilization. In addition such composites can be designed to have a sufficiently low ac loss for slow ac or pulsed applications.

The magnetic permeability of a number of specimens of iron and silicon iron, differing in grain size and heat treatment, has been measured over the range 500 Hz to 30 MHz; measurements were made on thermal and ac demagnetized states and in polarizing fields extending well beyond the knee of the magnetization curve. The measured dispersion of permeability, ascribed to eddy currents, has been analysed in terms of a simple but general domain model as in a previous paper by Collar et al. Consistent values of the rotational contribution to the permeability have been extracted from the measurements which are shown to be in quantitative agreement with theory provided due allowance is made for magnetic interactions between crystallites. There is found to be only a weak dependence of domain wall susceptibility upon grain size; the restoring force on a domain wall must therefore be caused primarily by metallurgical conditions within the crystal grains rather than by any direct influence of the grain boundaries.

Measurements of the magnetostriction constants h₁ and h₂ are presented for Co_xMn_1-xFe₂O₄ in the range 2 K less-than-or-eq, slant T less-than-or-eq, slant 400 K for 0·005 less-than-or-eq, slant x less-than-or-eq, slant 1·12. The contribution of Co²⁺ ions to the magnetostriction constants Δh₁ and Δh₂ have been extracted and compared with the theoretical predictions of the single-ion model. It is found that the magnetostriction varies linearly with cobalt concentration only for x<0·20. Close agreement between the theoretical and experimental temperature dependence of Δh₁ and Δh₂ over the whole temperature range implies a much larger spin-orbit coupling constant (αλ similar 270 cm⁻¹) than has hitherto been thought appropriate to this system, together with a relatively small exchange interaction (βH similar 125 cm⁻¹).

Measurements of the magnetocrystalline anisotropy constants K₁ and K₂ have been made on a number of 4% molybdenum permalloys in the range 69 to 89% nickel. The variation of K₁ with temperature over the range −200°C to 300°C has been studied. The results show that the influence of magnetocrystalline anisotropy on the variation of the initial permeability of the alloys is very small.

A summary is given of the analysis of pure absorptive and pure dispersive Lorentz modulation-broadened line shapes detected at either the first or second harmonic of the modulation frequency. Correction curves are derived and a procedure outlined to enable values of relative intensities, admixture parameters and linewidths of the absorptive component to be determined from modulation-broadened mode admixed Lorentz resonances.

A study has been made of the electrical resistance, the electrical noise and the change in resistance as a function of mechanical strain of single carbon fibres from seven different batches. The resistance of the fibres was found to be much larger than would be anticipated if they were entirely composed of aligned graphite crystallites and the electrical noise was found to have the characteristics of the contact noise associated with loose aggregates of carbon granules. Most fibres showed a linear increase in resistance above about 0·1% strain. Variable behaviour was observed in the initial strain region for all but a few fibres. In some cases fibres showed time dependent electrical properties while remaining mechanically stable. It is suggested that the electrical properties of the fibres are largely determined by the contact resistance between the fibrils which make up the fibres and the variation of this contact resistance with pressure. The possible electrical properties of a simple model of the fibres are discussed.

The conditions under which good quality single-crystal silicon films can be vacuum deposited on to sapphire substrates have been established. The critical deposition parameters, which are shown to be interdependent, are the silicon impingement rate and the sapphire substrate temperature. For any given silicon impingement rate there exists an etching-deposition boundary temperature T_E above which the silicon will etch away the sapphire and below which silicon is deposited. The etching-deposition boundary has been determined for the (011bar above 2) and (0001) sapphire orientations for silicon impingement rates in the range 0·1 to 1·5 μm min⁻¹. The silicon film quality, in terms of its electrical mobility, is presented for the two orientations as a function of the parameter T_D/T_E, where T_D is the substrate temperature and T_E is the etching-deposition boundary temperature corresponding to the impingement rate used. There is an optimum value of T_D/T_E; under optimum conditions (111) silicon films deposited on (0001) sapphire have mobilities up to 75% of bulk values but are not very reproducible, while (001) silicon films deposited on (011bar above 2) sapphire have mobilities about 45% of bulk values and are reproducible. Whilst the macroscopic surface finish of the sapphire substrates is important, the crystalline perfection is not and the limited mobilities are ascribed to the high fault density of the films and the high compressive stress resulting from the thermal expansion mismatch.

Although ultra-high vacuum techniques have been used throughout the work, evidence is presented which shows that the vacuum environment is not critical.

Antimony trisulphide was sintered on to glass blanks and their electrical and photoconductive properties such as spectral distribution of photoconductivity, current-voltage characteristics and lux-ampere relationship are presented.

The effect of impurities of praseodymium, neodymium, gadolinium, terbium, holmium and erbium on the plastic deformation of calcium fluoride single crystals has been studied. From the measurements of microhardness and the dimensions of the dislocation rosettes, it has been shown that hardness of a crystal depends on the atomic weight of the dopant.

Landau-Lifshitz-Looyenga's formula for correlation of bulk and powder dielectric parameters has been investigated at microwave and radio frequencies for a number of materials for wide enough experimental conditions of particle size, shape and packing fraction.

It has been found that the formula gives good accuracy (about 3%) for dielectric constant determination for particle size not less than 50 μm. For fine powders (particle size less than 30 μm) the deduced dielectric constant is too low by about 3-8%.

Dielectric loss determination from the formula is, however, less accurate, the deduced values being always too low by about 10-20%.

Vacuum breakdown and field emission measurements have been obtained for W and Mo wire cathodes as a function of annealing temperature using a variety of cylindrical anodes (Cu, W, Al and Mo). The cathode surface structure was investigated using a scanning electron microscope. These observations are shown to indicate that initial surface protrusions capable of giving reproducible Fowler-Nordheim plots consist mainly of impurities. Pure metallic protrusions are only formed after breakdown following high-temperature cathode annealing. Breakdown after the highest temperature annealing treatment appears not to be field emission dominated.

Electric fields, pulsed or dc, can initiate the formation of protrusions on tungsten hairpin cathodes used with plane anodes of varying materials. Topographical studies of both electrode surfaces, using scanning electron microscopy, have now shown that these protrusions appear to be the irregular lips of microcraters. This damage is almost entirely confined to the cathode surface, and its occurrence appears to depend upon the material initially used for the anode; a stainless steel anode having ten times the probability of giving a stable electrode pair than a copper anode. Neutron activation analysis failed to detect any anode material on damaged cathodes.

Two possible mechanisms are discussed that could account for the microcratering: (i) the instantaneous vaporization of field emitting micro-features on the cathode surface, and (ii) bombardment of the cathode by high velocity microparticles (clumps), originating at the cathode and subsequently acquiring an enhanced energy by a combination of elastic reflection and charge reversal at the anode.

The arcing behaviour of a number of refractory metals and graphite has been examined by negatively biasing specimen electrodes in the presence of a pulsed rf discharge. The main object of this work was to select anode materials for gas-filled valves which would reduce their tendency to arc-bank under severe commutation conditions.

The voltage at which about 5% of the pulses resulted in an arc, at an argon ion current density of 1·2 kA m⁻², was used as a figure of merit. Of the materials examined, tungsten was the best with a figure of merit of 15 kV and graphite the worst with 6 kV.

A good correlation was obtained between the results and the published values for the vacuum arc voltages.

Argon ring discharges (3·26 MHz) in the pressure range of 2-50 torr have been studied for applied axial magnetic field strengths between 0 and 4·5 kG. Spectroscopic measurements show very marked increases in low level neutral line intensities near the centre of those discharges which have an applied axial magnetic field. Electron temperatures and densities are derived from the spectroscopic measurements and electrical conductivities predicted by these temperatures and densities are compared with directly measured values of conductivity. The measured dependence of conductivity on magnetic field and pressure agrees well with that predicted by ion mobility.

A microwave cavity method was used to measure the electron density in the afterglow period of a pulsed dc glow discharge. The measurements were carried out in the mixture neon with 6 × 10⁻²% of helium for the pressure interval from about 5 to 25 torr.

The de-excitation cross section of metastable atoms was found to be 1·09 × 10⁻¹⁹ cm². It was also found that the late afterglow period is affected only by the ambipolar diffusion of ions and electrons at all measured gas pressures.

The value of an effective reduced mobility coefficient was calculated from the measured value of the ambipolar diffusion coefficient and it was found to be 8·37 ± 0·84 cm²V⁻¹ s⁻¹. This value is in a good agreement with the theoretical value calculated according to the Langevin formula for molecular (NeHe)⁺ ions in neon. It is supposed that these ions strongly affect loss processes of electrons in this mixture during the afterglow period.

The decay of neutral gas temperature in low pressure argon afterglows has been measured by monitoring the acoustic standing waves excited by the ionizing discharge, with a microwave interferometer. Calculations, based on heat loss by atom thermal conduction only, give very good agreement with experimental results over a wide range of plasma conditions.

The possibility of attaining a high non-equilibrium electrical conductivity in a high velocity flow of combustion products is of interest from the point of view of MHD power generation. Experimental evidence and theoretical predictions of the electron-ion recombination coefficient, which is of critical importance in this context, are discussed. An experiment is described, in which combustion products seeded with potassium were accelerated through a nozzle to a supersonic velocity. The electrical conductivity was measured and found to be a factor of six lower than that expected from the J. J. Thomson 3-body recombination theory.

In this communication the author has investigated the optimum seeding of helium with caesium in a linear MHD generator for three different electrode configurations; taking into account the dependence of electron collision cross sections on electron velocity. The result shows that the optimum ratio α of the number of caesium to helium atoms not only depends upon the electron temperature but also on the electrode geometry. The expressions for the electron temperature for all the three electrode geometries have also been derived. Numerical results have been presented from which the optimum ratio α and the electron temperature can be obtained for the relevant range of parameters.

A method of computing the trajectories of charged particles which would occur in a real quadrupole device is presented. A successive, over-relaxation procedure is used first to solve the Laplace equation for the specified electrode system. Subsequently a simple Runge-Kutta integration is used to compute the trajectory. A number of results are presented showing that the conditions necessary for stable trajectories in the particular system studied are more restrictive than those predicted by the hyperbolic solution.

The melting of materials during heating is modelled automatically on a resistance-capacitance network analogue using field effect transistor switches. Full details are given of the experimental techniques. Results are presented for the particular example of a flat slab heated by a concentrated heat source with melting occurring in the vicinity of the heat source. Non-dimensional curves are included from which the rise of temperature with time at a number of points can be determined. The latent heat of the material is also modelled on the network and for the problem of the concentrated heat source it is found to have no significant effect on the results. The results are shown to be independent of the conditions on the edge of the slab provided that the edge is a distance of at least 1·5 times the slab depth from the heat source.

A switched-voltage technique is described which makes it possible to set the pass-band of an electrically tunable electro-optic birefringent filter over a considerable range. A model incorporating the above technique and using dihydrogen phosphate-type crystals is investigated. The filter scans the optical spectrum from 250 nm to 750 nm with a bandwidth ranging in value from 8 nm at the violet end to 48 nm at the red end. The transmission losses for light in the principal plane of polarization for the five-unit filter considered are less than 25%. The switched-voltage program is obtained from an analogue computer controlled by a linear wavelength-voltage ramp. It is therefore possible to select a centre wavelength by manual control or alternatively to time-control the filter.

The liquid-vapour equilibria for the systems nitrous oxide-oxygen and carbon dioxide-oxygen are presented in the form of isotherms on a pressure-composition graph, and isobars on a temperature graph, up to 200 atmospheres (20·7×10⁶ N m⁻²). In the case of premixed cylinders of a 50/50 nitrous oxide-oxygen mixture used as an analgesic any liquid phase formed as a result of cooling the cylinder to not lower than −40°C can be dispersed by storing the cylinder horizontally for 24 hours at a temperature not less than 5°C.

A method is described for fitting complex data to a given function, with particular emphasis on the case where the fitting parameters are periodic. A procedure is given for the fitting of data when the relationship between the observed variables and the functional variables cannot be solved explicitly.

The method is illustrated by two examples taken from the field of dielectrics. These are the analysis of complex permittivity results and the calculation of the complex propagation constant of a liquid contained in a dielectric cell.

A three-dimensional isotropic continuum model is used to look at the gas-surface interaction problem. Results for energy transfer to the continuum when acted upon by a transient force pulse approximating that expected during a collision process indicate a negligible amount is propagated into the solid as stress waves. This appears to support claims of researchers using discrete lattice atom models which ignore energy transfer to any but a surface array of atoms.

A simple relation between the energy of formation E_f of Schottky defects in alkali halides and compressibility is developed. Values of E_f for alkali halides for which experimental values are not available are estimated and agree with the recent theoretical calculations.

A simple calculation is described by which it is possible to evaluate the well-known higher order pseudo-torques arising from the deviation of the magnetization M from the field direction H, when plotting torque curves in `non-infinite' fields. This then obviates the need to make graphical adjustments to the curves as plotted.

The result is applied with considerable quantitative success to the torque-reversal phenomenon. The non-equivalence of measured torque reversal field and anisotropy energy reversal field, unless angles are correctly chosen, is also pointed out.

Electrically induced conductivity switching has been observed in ZnS single-crystal platelets grown by the argon transport method. Changes in resistance in excess of 6 orders of magnitude have been observed in devices made by evaporating silver electrodes onto opposite faces of the crystals. The low resistance state obeys Ohm's Law over at least 4 orders of magnitude of applied voltage and has zero temperature coefficient of resistance. It is only stable below a critical temperature of about −40°C.

Electrical breakdown potentials of mercury vapour are measured between plane parallel nickel electrodes at various vapour pressures. The breakdown potential values are plotted as a function of the product of mercury vapour pressure and interelectrode spacing (p₀d) which is varied over the range 1·0×10⁻²-1·84 torr cm. The breakdown potential shows the normal behaviour associated with the Paschen curves. At low pressures, it decreases with increasing vapour pressure until a minimum sparking voltage is attained. At high pressures, the breakdown potential increases with increasing vapour pressure. A step rise in the breakdown potential is observed from a value of 521 V at p₀d of 0·29 torr cm, to 1652 at 0·2 torr cm. A minimum sparking potential of 330±10 V is observed at about p₀d=0·5 torr cm.

In 1963 Easterling found an order-disorder transition in Nichrome wire at 440°C. This temperature is shown to be in error by approximately 100°C because of the conductivity that arises, at high temperatures, in the cement.