Table of contents

A possible explanation for the low pulse-height tail observed in photo-multiplier pulse-height distributions is given. Some pulse-height distributions are presented for illustration.

Silicon oxide layers have shown irreversible reductions in electrical conductivity when first exposed to the atmosphere. With films evaporated slowly (rate of evaporation equal to 10 Å s⁻¹) the reduction is accompanied by polarization. It is found that significant conduction measurements following exposure to the atmosphere can only be made on films deposited more rapidly, i.e. at about 30 Å s⁻¹.

A study of the surface composition of low yield metal blacks was made using Auger electron spectroscopy. Auger peaks of the respective metals were observed whereas no argon was detected.

The six elastic compliances of CoF₂ have been obtained by measuring the resonant frequencies of a set of single crystal rods. The compliances have been inverted to give elastic constant values.

Reasons for the occurrence of etch pits with different shapes on NaCl cleavages, employing the etchant ethyl alcohol with HgCl₂ as poison were investigated. It is found that the inclination of the dislocation produces not only an asymmetric pit but also brings about a change in the pit morphology. This is probably due to the difference in the dissolution rates at different dislocation sites.

A relationship recently proposed by Date and Andrews for the analysis of the elastic constants of cubic polycrystalline metals has been examined in detail and shown to be one of a number of similar approximate relationships. Some examples suitable for use in connection with ultrasonic measurement techniques are given. The case of hexagonal materials and the estimation of errors are also discussed.

The depth of indentation of a thick layer by an axially symmetric rigid punch depends on its shape, the elastic constants of the layer and the force. A brief discussion of these relations and their physical significance is given in more general form than previously. The corresponding theory for very thin layers is given in detail. Results are calculated in particular for spherical indenters. Available experimental results are reviewed, and it is shown that for thin layers they need supplementing, in order to cover satisfactorily the conditions in which the theory is applicable.

The geometrical form of a specimen has been deduced which will give a constant temperature gradient along its length under direct electrical heating in a vacuum. For materials, available in the form of solid rods, this optimum form can be approximated to a solid cone with axial flow of current. The differential equation of heat flow then does not have to be integrated and the thermal conductivity at any temperature can be calculated irrespective of the temperature dependence of resistivity and the emissivity of the material. The method has been verified by determining the thermal conductivity of cobalt at high temperatures.

The preparation of lithium fluoride with thermoluminescence properties suitable for use in radiation dosimetry has been investigated. Variants based on magnesium and titanium additions have been prepared and examined from the viewpoint of activator concentration, role of secondary impurity and purity of the lithium fluoride starting material. The following properties were examined, impurity distribution and luminescence distribution within the crystal, emission spectrum, dose response characteristics and optical absorption. A sensitive thermoluminescent preparation may be obtained by the addition of magnesium alone to a readily available source of lithium fluoride.

It was previously found that thre was a simple empirical relationship between the average of certain measurements on spike-like magnetic domains observed on (110) surfaces of crystals in iron and silicon-iron and the coercive field of the samples. An approximate formula was calculated for the field strength required to cause such domains to expand irreversibly, and it was found to be similar to, but not in close agreement with the empirical formula for the coercive field. The similarity was sufficient to suggest that the shapes of the hysteresis loops of the specimens were mainly governed in the region of the coercive field by this irreversible expansion, rather than by the domain wall impedance effects which one would expect to predominate in magnetically harder materials.

A more detailed theoretical analysis now shows that the field strength at which irreversible expansion of the `spikes' occurs is quite close to the coercive field as obtained from the empirical formula.

The effect of a second phase on the sheath current carrying capacity of type II superconductors is investigated by measurements on suitable binary eutectic alloys. The results show that the presence of a normal second phase increases the sheath current. Furthermore, the orientation of the second phase with respect to the applied field is shown to affect the value of the sheath current. The results are explained by postulating surface sheath superconductivity occurring at the interfaces of the superconducting matrix with the second phase.

This paper reports the results of an experimental examination of the dc and ac properties of TiO₂ in single-crystalline and anodic thin-film form.

The single crystal and the thin film appear to obey the same dc conductivity law Iαexp V^1/2 over the range investigated. Their ac characteristics are similar as well. These results may be characteristic of the metal-insulator contact rather than of the oxide itself.

Resistance switching with memory was observed in almost all films, but it could not be induced in the single crystal.

No existing model is a complete success, and a more complete understanding of the structure of a metal-insulator-metal system is needed before a better model can be found.

It is generally believed that there is no possibility of deterioration occurring in silicon solar cells on the surface of the earth. As one of the tests for their environmental capabilities, the effect of electrical discharge on series resistance, curve power factor, conversion efficiency and spectral response is investigated. The cells are found to exhibit unexpectedly large and permanent deterioration in their characteristics. Some of the phenomena which might be responsible for the observed degradation are pointed out and discussed.

In recent years considerable research has been devoted to techniques for improving the performance of caesium plasma diodes. One such proposal is that increasing the ionization rate might enhance the performance; in fact, Badareu and co-workers have recently reported that they experimentally observed such an effect when seeding a caesium plasma diode with mercury or cadmium. This enhancement is believed to be due to the Penning effect. A theoretical investigation has been made of the effects of seeding a caesium plasma diode with mercury or cadmium. A model for both a pure caesium and a caesium seeded diode in the presence of the Penning effect has been developed; results from this model compare favourably with available experimental data. It was found that the electron temperature for a pure caesium diode operating in the ignited mode lies in the 2300-2800 K range. This model, covering both the ignited and unignited modes, shows clearly where significant enhancements due to seeding can be expected; unfortunately, as will be shown, the region where significant enhancements are obtained are not near the maximum power point.

Results are presented which show that a self-excited ion-sound instability in a plasma behaves in a similar manner to a classical Van der Pol oscillator, both (a) in a single mode régime, and (b) in a multi-mode régime. In the single mode régime various methods are described which enable the linear growth rate α₀ to be measured. These are checked by direct measurement and a consistent value is obtained. In the multi-mode régime, values are obtained for mode-mode coupling coefficients, and non-linear saturation coefficients for each mode present in the system, as well as for the parameters associated with the growth rates of these modes. Comparison with a theory developed to describe these two systems, shows good agreement with the experimental measurements.

This paper describes some aspects of the development of the Townsend discharge into a spark channel. Three different stages of the discharge growth are distinguished. The first of these is obtained when the discharge current is of the order of mA-A and this is followed by the second stage, i.e. a transition to a filamentary glow phase which may appear, depending on the energy supplied, either before or after a cathode spot is formed. The final stage, termed here the second glow phase, leads to the formation of the filamentary spark channel.

Arcs passing through azimuthally slotted grids in mercury arc rectifiers were found to move along these slots so that they rotated about the grid axis. The frequency of rotation varied with current for different grid designs; at 100 A the azimuthal velocity was 420 m s⁻¹ corresponding to rotation at 4·5 kHz.

A theory involving depletion of neutral particles in the grid region predicts this linear variation of velocity with current; the theory agrees numerically with the measurements within the accuracy to which the arc properties are known. The theory is also applied to a simple constriction under `current chopping' conditions and it predicts a critical charge, which is in good numerical agreement with the experimental data of Harrison et al.

We have studied the radiation produced in pre-breakdown discharges in Ar, Kr, and Xe, in a coaxial geometry, in the pressure range 50-600 torr. In these discharges, consisting of a single Townsend avalanche, radiation continues to be emitted for an appreciable time (up to some tens of microseconds) after the end of the current pulse. The intensity of the uv component of the radiation pulse is well expressed by a formula of the type I(t)similar, equals exp (−λ₁t) −exp (−λ₂t), where λ₁ and λ₂ are specific decay times. λ₁ is a function of the pressure; λ₂ is almost pressure independent.

The results are discussed on the assumption that excited dimers are formed in the avalanche by three-body conversion of metastable atoms. The three-body conversion frequency for the dimer formation is 6·6×10⁻³³ cm⁶s⁻¹ for Ar^m, 44×10⁻³³ cm⁶s⁻¹ for Kr^m and 25×10⁻³³ cm⁶s⁻¹ for Xe^m. The mean lifetime for the excited dimers is 3·7 μs for Ar₂^*, 1·7 μs for Kr₂^* and 0·5 μs for Xe₂^*. At the lower pressures, two-body collisions become a relatively important destruction mechanism for the metastable Xe levels.

Energy distributions of electrons in beams from hollow-anode gas discharge electron guns have been determined for electron flight paths of 0·1, 0·3 and 0·6 m. Guns have been operated at potentials between 5 kV and 20 kV in hydrogen, helium, oxygen, air, nitrogen and argon at pressures in the approximate range 10-100 N m⁻². A typical spectrum obtained has a sharp peak at an energy equivalent to that achieved by electrons accelerated to the full cathode potential. A second peak of amplitude 0·05-0·5 times the primary peak amplitude occurs at an energy lower by approximately one ionization potential. This secondary peak is broader, the amplitude falling to approximately zero within an energy range of about 100 eV.

In many instances, the electron microprobe analyser is required for quantitative analysis of composition in regions where concentration is a steep function of distance. The limitation imposed by a finite diameter probe has been analysed assuming Gaussian, conical, cylindrical, and square electron density distribution excited by the probe. These results are compared with experimental results for a one-dimensional step function composition profile, and the Gaussian beam is found to be the best approximation for the microprobe. A deconvolution technique, using Fourier analysis, is developed to correct for this finite probe diameter effect, thus increasing the effective resolution of the microprobe. However, attempts to reproduce the theoretical step function have been unsuccessful. Good results are obtained when this technique is used to reproduce smooth functions, e.g. error function profiles.

A generalized treatment of the influence of magnetic fringe fields on particle trajectories is proposed. The behaviour of the real field is compared with the ideal field, whose boundary is assumed to be curved. The position and shape of the virtual boundary are thus determined to simplify the resulting formulas. Thirdorder effects are disregarded.

The optical cavity of the oscillator section of a giant-pulse ruby laser has been split, by means of a pair of prisms, into two parts which can be Q-switched independently. The laser generates two equal multi-megawatt giant pulses during a single operating cycle, and the relative delay between two is continuously variable over several hundred microseconds.

The distortion of an intense light pulse by a saturable absorber is analysed on the basis of the dynamic transmission equation. The conclusions are in accord with an earlier numerical analysis for a particular pulse function, but are here proved for arbitrary pulse shape. Bounds are derived for maximum transmittance and energy loss, and the existence of stable periodic solutions is demonstrated. The region in which saturable absorption theory is valid is discussed and compared with that for self-induced transparency. The two phenomena are examples of completely incoherent and completely coherent absorption, respectively.

To understand heat transfer within and from a particle laden flame requires the knowledge of the radiation absorption and scattering properties of the particles. These properties can be defined in terms of the complex refractive index. A literature survey failed to reveal measurements of the complex refractive index of high temperature particles of any material. The most suitable experimental method appeared to be that of Avery (1952) and the variations of the method (discussed by Humphreys-Owen 1961) were investigated. The complex refractive indices were measured on a flat, macrosized sample of spectral graphite, on particles of SiO₂ and pulverized coal, all at room temperature and on pulverized coal in a flame at about 1700 K. The experimental method adopted used a photomultiplier and amplifying system to detect the modulated monochromatic light reflected by the sample, the polarized components of the reflected light being selected by a motor driven polarizing chopper. Two methods of calculation were compared and it was concluded that the least squares method of calculation using a computer was preferred to the more generally used graphical method. Some of the assumptions made and the limitations to the theory are discussed, including a modified definition of the complex refractive index for particles of various shapes.

The shape of the boundary enclosing analytical solutions to the generalized Fresnel reflectance equations for n and k in terms of reflectances and angle of incidence has been investigated and the optimum angle of incidence for experimental measurement determined to be 74°. The distribution of reflectance values for fixed n and k for which the method is accurate to within ±0·05 has been deduced to be n<3·0 and k<3·2.

A new method for measuring the velocity of charged particles or droplets moving in applied electric fields is described. It uses a Schlieren system which selects symmetrical plus and minus diffraction orders and allows them to overlap and interfere, so that the best frequency due to the differential Doppler-shift can be recorded at a point in the image plane. The method is illustrated by its application to a monodisperse spray of electrically atomized kerosene, using a He-Ne laser as the light source. Its advantages are discussed and its range is assessed theoretically.

A tungsten lamp of known colour temperature and intensity may conveniently be used as a standard of spectral power distribution. This paper describes a method of calculating absolute values of spectral radiant intensity using the two known parameters. The tables, which are based on De Vos' values for the emissivity of tungsten, list the spectral radiant intensity per candela Q(λ) for both tungsten and blackbody sources in μW sr⁻¹ cd⁻¹ Δλ⁻¹ for a 10 nm wavelength interval.

Specimen tables are reproduced for the three temperatures 2043·2, 2700 and 2855·54 K. Q(λ) is given at 10 nm intervals over the range 240 to 830 nm and at wider intervals to 2600 nm. The accuracy attainable by this method is about ±1% at the centre of the visible spectrum.

The method of scaling is applied to reduce the number of variables in the equations governing the elasto-hydrodynamic lubrication of a rubber cylinder. In this way the calculations by Roberts and Swales in 1969 are shown to be of general applicability.

The present study consists mainly in determining the speed of a conducting jet in a special disk-shaped tube. The results show that the mobility of the ionized jet depends, among other things, on the form of its movement in the tube.

For specific electromagnetic parameters, and also for given dimensions of the tube, it is possible to obtain a movement where the conducting jet behaves as a body moving in a fluid medium (mode I). In this case, the speed of the jet is very high. On the other hand, for different specific electromagnetic and geometric parameters of the experimental system, it is possible to obtain a low speed of the ionized jet, and the fluid in the tube appears to move as a single whole (mode II).

It must be noted that in each of the two types of the jet movement, both laminar and turbulent motion can occur, with the transition between them determined by the critical Reynolds number, which differs markedly for the two modes.

An expression for the opposing force (due to viscosity) was derived from the experiments in both modes.

Ferrofluids containing gadolinium and iron particles in the size range 80-750 Å in diameter have been prepared by evaporating gadolinium and iron onto a mercury surface in an argon atmosphere at a controlled pressure. The particle sizes are determined by electron microscopy and identified by microprobe analysis.

Apparatus used to measure the Seebeck coefficients of polymer films is described; values are presented for polyvinylchloride and Nylon 66.

Electron current in excess of the ideal square law occurred in p⁺-π-Al diodes and was interpreted in terms of double injection mechanism. The drain (Al) potential steps and the Richardson constants for electrons and for holes were determined using the excess electron current, and found to be ϕ_napproximate 0·504 eV, R_napproximate 461 A cm⁻² K⁻² and ϕ_papproximate 0·616 eV, R_papproximate 93 A cm⁻² K⁻² respectively.

Results are presented for ZnO acoustoelectric oscillators at room temperature of resistivities from 100 Ω cm to 10⁴ Ω cm. Pulsed operation as well as continuous wave operation was possible and frequencies up to 5 GHz and 4 GHz respectively were observed. The oscillations are divided into three different types depending on the thickness and the conductivity of the sample in question. The experimental results are compared with the existing theories for the acoustoelectric oscillator and with results previously obtained with CdS oscillators.

This technique of measuring the temperature of a discharge in local thermodynamic equilibrium uses the total continuum emitted by the discharge when seeded with hydrogen. The combined intensity of the continuum caused by the negative hydrogen ion and the positive ion from the discharge gas, when divided by the intensity of a spectral line from either of the constituents produces a function with a well defined maximum at a certain temperature, from which other temperatures corresponding to other ratios can be determined.

Experimental observations are given of the shape and structure of laser-induced sparks formed in molecular and atomic gases. Absolute measurements of the laser energy absorbed during the breakdown process are presented.