Table of contents

The basic principle of a simple method is explained for the determination of the damping element characteristic. The theory upon which the method relies does not involve approximations.

There have been long-standing inconsistencies among the values of different experimenters for the Townsend first ionization coefficient α/p in nitrogen at low values of E/p. It is shown that the inconsistencies have arisen because the analytical procedures employed have been insensitive to the effect of secondary emission by N₂(A ³Σ_u⁺) metastable molecule bombardment of the cathode. The neglect of this effect has led to erroneously high values of α/p in the cases where very high gas purity was achieved.

A brief discussion is given of a priority matter associated with the problem of calculating the potential distribution in a finite layer of intrinsic semiconducting material.

Evidence is put forward which suggests that the polarity ratio in the dielectric breakdown of NaCl with point-plane electrodes may be time-dependent. The direct breakdown voltage is highest when the point electrode is the cathode, but for impulse voltages this is reversed.

Average spectral irradiance values for global and solar radiation for air mass 2 are given. The results were calculated from data obtained during measurements made at Pretoria during 1964-65. The values are considerably lower than those of Moon (1940).

Detailed numerical studies are made of recent measurements by Sumita and Takahashi of average and most probable slowing down times to thermal energies in graphite. Good agreement is obtained between computed and measured values at 5·2, 4·91, 1·46, 0·28, ·20 and ·11 eV. Similar studies are successfully made of the lead slowing down time spectrometer for a variety of detectors covering the energy range 1 to 3 keV. Estimates are made of the effects of data uncertainties on the theoretical values.

Appreciable scatter has been reported in the measured Young's modulus of whisker crystals determined by microtensile testing. In the most extensive investigation by Mehan et al appreciable viscoelastic flow was observed in the adhesive used to secure the whisker. Suitable corrections for extraneous elastic deformation have been presented. The present paper shows that it is possible to account quantitatively for the viscoelastic flow in the adhesive. A viscoelastic three-element elastic model is used to represent the whisker machine system, and relationships between elastic and viscoelastic parameters and curve characteristics are derived. A good fit to continuously recorded experimental data is obtained. An example shows that the elastic modulus is underestimated by 30%, even when the observed hysteresis loop is quite narrow, if no correction is made for viscoelastic flow of the adhesive. The extension to a five-element viscoelastic model is indicated and makes the method applicable to a wide range of viscoelastic problems involving cyclic loading.

Impact between two bodies has long been recognized as a cause of ignitions of gases and vapours. It is not immediately obvious, however, that friction during impact plays an important part in the production of surfaces hot enough (usually > 1100 °C) to cause such ignitions. The classical theory of impact has been applied to the measurements made of the temperatures and permanent deformations produced by the impact of cylinders on rigid inclined planes. A satisfactory correlation has been obtained between the energy calculated as dissipated by friction and the heating produced by impact, and also between the energy dissipated by the normal impulse and the work done in producing permanent deformation.

Calculations show how, according to the theory, the coefficients of friction and restitution and the length-to-diameter ratio of cylinders affect the dissipation of energy by friction during impact. The conclusion is that, given the physical properties of two bodies undergoing impact, it is possible to forecast the heat energy dissipated at the surfaces in contact and the resulting surface temperature.

A simple optical transmission method has been developed to estimate the uniformity of film thickness of thin metal films deposited on nontransparent ceramic substrates. The method is capable of detecting a few ångström units of variation in the film thickness at optical magnifications up to ×200.

The RF heating of a plasma produced by a coaxial plasma gun is studied. The resonance absorptions at ω=ω_ce and ω=ω_ce/2 are observed, where ω_ce is the electron cyclotron frequency. In the frequency range where ω<ω_ce/2, nonresonant absorption occurs. This may be attributed to the Whistler wave mode.

A new method of measuring the local velocity of shock and detonation waves, based on measuring the Doppler shift in frequency of laser light using Schlieren interferometry, is reported and its limitations are assessed. Expressions are given for the Doppler shift in terms of all the relevant parameters, including the refraction produced at the windows of the detonation vessel, for the spatial resolution, and for the accuracy of the method. Velocity measurements made on stoichiometric oxygen-acetylene mixtures at an initial pressure of 75 Torr show that, prior to the onset of detonation, local shock wave velocity measurements can be made over a distance of 4·4×10⁻³ cm with an accuracy of ±11%, and 4·4×10⁻² cm with an accuracy of ±2%. For the detonation wave the error in the velocity measurement was about ±20% when averaged over distances of 1 to 2×10⁻¹ cm.

A method is described which enables measurements taken on a coaxial line to be automatically recorded onto computer paper tape. The description is divided into two parts: the standing wave technique for the low loss liquids and the travelling wave for those of higher loss. Unlike in existing methods, the travelling wave results are obtained by observing the output of a microwave bridge as it goes off balance. Another feature is that both methods have involved a departure from the usual superheterodyne detection system.

The techniques are illustrated with typical results which have been obtained for water between 300 MHz and 4 GHz and which show that as well as reducing the experimental time the new methods are capable of yielding greater precision than the non-automated method.

Brief details of the computer analysis of the data are also given.

An experiment has been done in which the electrodes of a vacuum gap were operated at widely different temperatures over a range 4 to 300 K.

The temperatures of both the anode and cathode were found to affect the prebreakdown current and breakdown voltage of a 0·8 mm uniform field gap between niobium electrodes of industrial surface finish.

An approximate expression describing the surface concentration at a given time is derived for a diffusivity D dependent on the concentration c when diffusion occurs in columns with the surface flux a prescribed function of the time. Estimates of the surface concentration during sorption into semi-infinite columns initially at zero concentration, for the cases of D constant, D is proportional to c and D is proportional to c² when the flux is constant and when it varies linearly with time, are compared with exact values obtained from the concentration profiles calculated after applying a transformation used by Barenblatt. Estimates have also been compared with exact values for these diffusivities and for D is proportional to 10^c and D is proportional to 1000^c for the case of constant surface concentration when the flux is inversely proportional to the square root of time. These illustrative examples show that the error in using the approximate expression is often small, especially in sorption cases in which the diffusivity increases with concentration.

Low-angle electron diffraction patterns observed in an electron microscope from thinned carbon fibres show features due to both the shape of the fibre and its internal structure. Simple optical analogue experiments using a laser are described, in which a reconstituted image is formed from an idealized diffraction pattern. In this way it is possible to distinguish those features in the diffraction pattern which arise from the fibre shape from those which carry information about the structure of the fibre. It is then possible to measure, for example, the spacing of striations along the fibre axis. Equatorial intensity maxima in the diffraction patterns are shown to arise from deflection of the main beam by electrostatic charging effects; this does not prevent meaningful measurements being taken from the patterns.

The longitudinal magnetostriction of polycrystalline grain-oriented cubic materials, having either a Goss, a cubic or a fibre texture, is studied as a function of the degree of orientation of the grains.

The theory has been developed by introducing, in addition to the material magnetostrictive constants λ₁₀₀ and λ₁₁₁, two further parameters, τ and W, which are related respectively to the grain dispersion and the volume of cross domains when the material is in the demagnetized or remanent state.

The expressions obtained, which yield, for τ=0 and τ -> infinity respectively, the known results for the magnetostriction of a single crystal and a perfect polycrystal, can be used to calculate the cross-domain volume W from magnetostriction measurements, when τ has been independently measured by other methods.

Some experimental results on grain-oriented Fe-3% Si strips are given.

Apparatus is described for the measurement of the axial (α₁) and transverse (α₂) thermal expansion coefficients of uniaxially drawn Perspex sheets over the range 0 °C to 30 °C, the orientation being represented in terms of birefringence Δn. It is found that α₁ decreases while α₂ increases with increasing orientation, the dependence on Δn being linear in both cases. At the higher temperatures the mean expansivity αbar above decreased slightly with increasing orientation, while at the lower temperatures it increased, the change-over occurring at about 11 °C.

Tangential edge stress distributions for orthotropic plates containing holes were computed for various orientations of the principal material axes with respect to load. The results were compared with material strength distributions to predict failure in various GRP laminates. Elastic stress distributions can be used to predict the position and remote stress level for the onset of debonding between matrix and fibres with reasonable accuracy. The stress concentrator is almost fully effective at this stage but becomes less effective at more advanced states of damage. An examination of large specimens revealed a substantial adverse size effect. It is suggested that the onset of debonding forms a suitable basis for design and that a further examination of size effects is required.

Glass chopped strand mat/polyester resin specimens were subjected to tensile or fatigue loading. From examination of polished specimen edges fibre debonding and resin crack damage were quantified. The damage depended on the topography of the site observed, but by expressing damage as a fraction of that present at failure all the results could be reduced to single, nonlinear curves for debonding and cracking. Resin cracking damage and the loss of residual tensile strength of damaged specimens gave curves of similar quadratic form. It was concluded that the development of resin cracking was primarily responsible for the progressive loss of strength in a fatigue specimen. Damage depended on cycle ratio but not on fatigue stress. It was suggested that a suitable cumulative fatigue damage law would be nonlinear and stress-independent.

By considering several analytical forms for the instrumental profile and following the linewidth relations of Ruland, a comparative study on the basis of true profile integral breadths has been made in cold-worked silver- and copper-base alloys to ascertain the nature of the instrumental profile. It is suggested that the instrumental profile can be approximated by two other functional forms fΔ and fy, apart from the gaussian and parabolic forms observed earlier.

The distribution in depth of boron ions implanted into silicon crystals has been measured using the French `CAMECA' ion microprobe analyser. Two problems associated with the technique, namely the normalization of the curves and the appearance of a tail on the deeper side of the gaussian, have been studied. The technique provides a rapid and simple procedure for profiling, and determines values of projected range which show good agreement with other techniques.

Iodine is known to diffuse through polyethylene, and, because of its high electron affinity, forms donor-acceptor complexes with several organic materials. On this basis, equations are derived predicting the form of the current transients obtained in electrically stressed polyethylene films after the addition of iodine to the electrode solutions. Experiment and theory are found to be in good qualitative agreement. From a curve-fitting process, reasonable values for the diffusion constant of iodine in polyethylene (10⁻⁸-10⁻⁷ cm² s⁻¹) and for the number density of possible charge transfer sites (10¹²-10¹⁴ cm⁻³) are derived.

Electrical and optical properties of n and p type Cd_xPb_1-xSe alloys with x less, similar 0·10 and Mg_xPb_1-xSe alloys with x less, similar 0·05 have been measured over a range of temperatures. In both systems the direct energy gap at 300 K was observed to increase linearly according to the relationship E_G=0·265+2·4 x eV, and was found to depend less strongly on temperature with increasing alloy fraction x. The density of states effective mass for the principal valence and conduction bands was found to be approximately proportional to the direct energy gap, and the observed reduction in mobility at 300 K with increasing alloy fraction could be explained by this increase in mass. Conduction in a second valence band, such as occurs in the Cd_xPb_1-xTe alloy system, was not observed.

The growth of large single crystals of LiNbO₃ by a special technique is described. The crystals show the presence of c and a ferroelectric domains, the polar axis being the c axis with respect to the hexagonal axes. Polishing, etching and microscope techniques were used to study the domains. The polar axes of adjacent a domains, which are the result of twinning on rhombohedral faces, were found to be inclined at nearly 35° to each other, the calculated angle being 35° 39minute.

Electroforming and conduction have been studied in thin amorphous films of ZnS and CdS sandwiched between Al, Ag and Au electrodes. The forming process differs from that observed in most amorphous oxides in a number of ways. Forming appears to be a two-stage process and the subsequent conduction properties are dependent upon the background pressure during device fabrication. The voltage at which the peak circulating current occurs is found to be time dependent in both ZnS and CdS.

Single crystals of CdS were prepared in an excess pressure of either sulphur or cadmium vapour. Attempts were made to incorporate copper during the growth. Trapping parameters, photoconductivity response, infrared quenching, and photochemical effects in these crystals were studied. It is suggested that the cation vacancies and the incorporated copper play important parts in determining these properties. Significant changes in the above properties were observed when the copper concentration in the crystal was 10 to 20 ppm.

Photovoltaic cells have been prepared by plating a variety of undoped and doped single-crystal slices of CdS with Cu₂S. The dopants investigated included copper, indium and chlorine. It is shown that the optimum starting material for maximum photovoltaic efficiency is copper doped with CdS with a dark resistivity of the order 10² Ω cm. Devices prepared on copper-doped crystals have such desirable features as fast speed of response, good temperature stability and saturation of the open-circuit voltage at low light intensities. Good efficiency is only obtained in cells formed on undoped CdS after a period of baking in air at 200 °C, which reduces their speed of response and introduces some other undesirable features. These effects are thought to originate with optical absorption processes in a high-resistivity, photoconductive region of CdS which is produced by copper diffusion at the interface.

Measurements of the energy and angular distribution of electrons photoelectrically generated from clean gold surfaces by the radiation produced by DC glow discharges in helium and nitrogen have been made. A significant number of electrons from gold due to helium radiation have energies of up to 16 eV, whereas at helium pressures of the order of 10⁻² Torr up to 10% of the radiation has been shown to be produced by the HeII resonance line at 40·8 eV (304 Å). The significance of these results for low-pressure breakdown and for the interpretation of probe data is briefly discussed.

The angular variation and field dependence of the initial permeability and eddy current loss in (110) [001] textured silicon-iron sheet are interpreted in terms of the characteristic slab domain structure. The material is represented as an assembly of weakly interacting grains containing slab domains with a gaussian distribution of orientations about the texture axis. This model not only forms a natural basis for the approximate independence of magnetization processes in the easy and hard directions (discussed in a previous paper by the authors) but also provides an explanation for the relatively high eddy current losses measured in the hard direction. Initial permeability and loss data for 0·004 inch 3% Si-Fe laminations, analysed on the basis of this model, provide three consistent independent estimates of the variance of slab domain orientations which are in fair accord with a Kerr magneto-optic survey. The effect of magnetostatic interaction between grains is investigated and shown to be small for the thin material studied.

Based on a simple atomic model giving the potential between electrons and atoms as V(r)=Ze²a^{s −1}/sr^s, the range of electrons penetrating solid targets is derived. Starting from the generalized power law involving the energy loss, the Lenard-type absorption law, and the assumption that the distribution of the secondary electrons due to both incident and backscattered electrons within the target is isotropic, a theoretical universal reduced yield curve of secondary electrons and the resulting maximum yield, which are found to be in good accordance with results obtained experimentally, are deduced as a function of three parameters such as atomic number, resonance potential and backscattering coefficient.

The change in cooled-KDP temperature due to 337 μm irradiation of the order of 0·1 mW is measured to be +1·3 K during 120 s by means of the rapid change in reflectivity of KDP near 109 K.