Table of contents

The experimental results given in a recent paper by Holmes and Freeston are found to be in good agreement with a theory due to Lam in 1964. Their poor agreement with the thin-sheath formula of Clements and Smy is also explained.

The proper and improper free electron density integrals are represented as sequence limits by two different methods. It is then shown how these sequences can be developed into power series. Coefficients and convergence radii of these power series are given. The results may be used both for analytical investigations and numerical calculations.

Computer programs based on least-squares numerical methods are described and their application to the analysis of complex internal friction spectra discussed. Various ways in which the programs can be used are demonstrated with reference to typical internal friction spectra due to Snoek and grain boundary relaxations.

The solutions for contact between specific rough and curved bodies are formulated based on the assumptions that the deformations of the surface roughness and waviness are fully plastic and elastic respectively, and that the roughness height distribution is Gaussian. Utilization of this model establishes a relationship between the applied load, mechanical and surface properties of the materials and the size of the effective contact region. The size of the effective contact region is shown to depend uniquely on the hardness number and the load number both of which are related to the modulus of the elasticity, Poisson's ratio and surface parameters of the specimens.

The reflectivity and curvature of a shock front moving into air at 1 atmosphere at velocities of up to Mach 10 are determined experimentally and the results compared with theory. It is found that whilst the reflectivity of the shock front is in good agreement with the expected values, the curvature is greater than that predicted at the highest Mach numbers.

This paper describes an experimental and analytical approach to the calculation of variations in temperature with distance and time downwind of a fire burning along the timber lining of a ventilated tunnel. Experimental observations of temperature variations downwind of a moving source of heat in a ventilated tunnel 9·1 m (30 ft) long and 63·5 mm (2½ in) square enabled simplifications to be made in the analysis of the problem. In the analysis it is assumed that the fire behaves as a source of heat of constant temperature and moves at a constant speed along the tunnel. The formulae resulting from the analysis give a satisfactory description of the experimentally observed temperature distributions.

An experimental investigation was conducted to determine the mechanisms by which electrically charged dust particles are collected electrostatically in the wake of a grounded perforated plate (or grid) placed at right angles to an airstream moving at about 180 cm s⁻¹. The electric field was applied by means of a second perforated plate placed just downstream of the first and at high negative potential. Dust passing through the first of these two grids was collected on the downstream face of the first grid as a result of the electrical forces acting on the particles in the direction opposing the gas flow. A theoretical model for the collection mechanism describes how this is greatly assisted by the formation of eddy pockets in the wake of the first grid, and by the entrainment of dust particles into these eddies. Using industrial fly-ash samples, experimental measurements of collection efficiency as functions of applied grid voltage and dimensionless grid aspect ratio indicate that the theoretical model is basically correct.

Total cross sections for the production of slow negative ions in collisions of O⁻ and O₂⁻ ions with O₂ molecules have been measured for incident ion energies in the range 4 to 100 eV, using a single-collision beam technique. Over the whole energy range for O₂⁻ and for O⁻ energies greater than approximately 35 eV the dominant contribution to the cross section is attributed to simple charge transfer, while, for low-energy O⁻ beams, elastic scattering has an important effect. The results are compared with those of Rutherford and Turner and Bailey and Mahadevan and reasons are suggested for the large discrepancies which appear for O⁻ in O₂ cross sections.

This paper describes the characteristics and potential uses for plasma chemistry of the sac which occurs at a constriction in a dc glow discharge (SESER). It is found that the rate at which molecules are dissociated is increased in the presence of a plasma sac; in the sac itself the dissociation rate per unit volume can be three orders of magnitude greater than in the positive column. Plasma sacs have been produced in an ammonia discharge with currents in the range 50 to 600 mA, pressures between 0·02 and 2·0 Torr and flow rates of 0·1 to 14 mg s⁻¹, the direction of flow being through the plasma sac from anode to cathode. The dissociation of ammonia to hydrogen and nitrogen has been measured with a mass spectrometer. Optical spectra show that the intensity of all lines is higher in the plasma sac than in the positive column; hydrogen lines, Hα and Hβ are seen in the sac but not in any other region of the discharge. Strong NH lines are seen but none from NH₂ and it is concluded that excited NH radicals may play an important role in the dissociation process. The high dissociation rates and intense spectra appear to be caused by the high-energy electrons in the plasma sac. A constriction in a gas discharge (SESER) appears to have interesting potentialities for plasma chemistry.

The paper describes the spectral properties of the optical emission from plasmas generated using a transversely excited high-pressure CO₂ laser and solid targets. The plasma emission spectra obtained using CO₂ and Nd³⁺ glass lasers are compared. As a consequence of the longer wavelength of the CO₂ laser radiation, heating of the plasma by optical absorption occurs at lower plasma densities. Strong line emission is therefore observed at greater distances from the target surface, which improves discrimination between the line and continuum background emission, and pressure broadening of the spectral lines is much reduced. The application of this technique to atomic spectroscopy and spectrochemical analysis is discussed.

Measurements have been made of the mobility of ions in a helium + 1% hydrogen admixture. A time of flight technique was used and the zero-field mobility of a single ion species determined as 22 cm² V⁻¹ S⁻¹. A secondary experiment, using a quadrupole mass spectrometer to analyse ions in such an admixture, suggests the measurements relate to H⁺.

A study has been made of the voltage required to initiate an electrically augmented propane-air flame between a copper anode and a pointed graphite or tungsten cathode. The voltage was applied as a ramp input of less than 30 V S⁻¹. The initiation voltage, which varied between 1200 and 2000 V, was measured at propane concentrations of 3·2 to 6·3% and electrode gaps of 0·4 to 1·8 cm, with total gas throughputs of between 525 and 1400 cm³ S⁻¹.

The minimum initiation voltage occurred at the stoichiometric propane-air ratio, that is at the highest flame temperature, and decreased with increasing gas velocity. The use of a propane-oxygen-argon mixture of equivalent composition to a propane-air mixture resulted in a 40% reduction of the breakdown potential. The voltage reduction with argon rather than nitrogen as the diluent gas is attributed to the lower heat capacity and breakdown potential of argon, relative to nitrogen. High-speed photographs of the initiation and the discharge path are presented which showed that the discharge initiated from the cathode and developed in approximately 2 ms to a discharge of 6-10 A.

The pulsed gas discharge in crossed electric and magnetic fields was studied spectroscopically in the wavelength range between 3100 and 4660 Å. A 5 μF, 800 V condenser bank was discharged through a cylindrical magnetron with an incandescent tungsten filament operated at argon pressures between 5 × 10⁻⁵ and 5 × 10⁻⁴ Torr. The axial magnetic field was 600 G. The spectrum contains lines of ionized argon; no atomic argon lines were observed. Ionized oxygen and atomic lines of tungsten were the main impurities. All wavelengths emitted by the pulsed discharge lie between 3900 and 4660 Å. The majority of ion lines have upper 4p states (in the case of ArII) and 3p states (in the case of OII).

Measurements of pre-breakdown currents in hydrogen have been made in uniform fields, in sphere-plane systems and in concentric cylindrical gaps, using the same gas sample. Measurements of Townsend's first ionization coefficient α in the uniform fields were made for E/p (electric field/pressure) between 100 and 1100 V cm⁻¹ Torr⁻¹ and in the pressure range 0·2 to 10 Torr. The multiplication factor exp integral operator α dx was then calculated for the sphere-plane and concentric cylindrical gaps and compared with the measured values. The full current-growth equation for a non-attaching gas was used in each case. The agreement between calculated and measured values in non-uniform fields establishes the validity of the current-growth equation even when the electric field changes markedly over an electron mean free path.

A method has been developed for measuring the transit times of charges injected into a dielectric while maintaining a stationary average current. This has enabled estimates of the mobility of photo-injected charges in n-hexane to be made over a range of nearly three decades of power input. Absence of any dependence of apparent mobility upon power suggests that liquid motion is not a significant factor in the transport of charges at the level of injection normally associated with this technique. The value of mobility obtained is (7·5 ± 1·0) × 10⁻⁴ cm² V⁻¹ S⁻¹, close to that obtained by Swan who used a similar injection technique.

The crack growth in the non-uniform stress field due to the contact loading of a spherical ball on an elastic half-space is described quantitatively for the case of an isotropic brittle solid. This theoretical understanding provides a basis for the Hertzian test which may be used to measure three important surface properties of strong solids and consequently their strength. These are fracture toughness, surface crack size densities and residual stress. All examples of applications of the test are described in detail for glass but the application to a wider variety of strong materials is implied and gives the test a wider significance.

The conditions of growth of hexagonal and a tetragonal form of lithium iodate are described and some properties of the latter recorded.

Plastic deformation of single crystals of MgO has been investigated by etching its indented (110) cleavage faces. The rosettes formed around the indentation marks have been critically studied by successively etching and polishing the crystal face. The configuration of dislocations forming the rosette pattern has been worked out and it has been shown that the configuration of dislocation loops produced in the body of the crystal by indenting (110) faces differs from that produced by indenting (100) faces, as reported by Keh.

A theoretical and experimental study has been made of silicon Schottky diodes in which the metal and semiconductor are separated by a thin interfacial film. A generalized approach is taken towards the interface states which considers their communication with both the metal and the semiconductor. Diodes were fabricated with interfacial films ranging from 8 to 26 Å in thickness, and their characteristics are related to this model. The effects of reduced transmission coefficients together with fixed charge in the film are investigated. The interpretation of the current-voltage characteristics and the validity of the C⁻²-V method in the determination of diffusion potentials are discussed.

The oxide thickness δ of a tunnel MOS diode is varied over the range 10 to 45 Å. This is done in an effort to establish the restrictions upon δ for which thermal equilibrium in the semiconductor is a valid approximation under the application of bias. Particular attention is paid to the reverse-bias case, and most of the experimental results are for δ>25 Å. A transition is observed from the behaviour of the ideal Schottky barrier to that of the thick-film MOS device. The ac conductance and capacitance together with dc current characteristics are studied as continuous functions of bias. From these results, information is obtained which relates the quasi-Fermi levels for (i) majority carriers, (ii) minority carriers and (iii) electrons in the interface states to δ. Thermal equilibrium statistics are found to be applicable to the semiconductor in the presence of a bias voltage when δgreater, similar30 Å, which compares with a theoretical prediction of δgreater, similar13 Å.

The change of the absolute Seebeck coefficients ΔS of thermocouple wires (iron, copper, Chromel, Alumel, and Constantan) due to magnetic fields of up to 27 kOe were measured near room temperature. ΔS of the paramagnetic materials was smaller than 10 nV/°C. Only ferromagnetic wires showed ΔS values of more than 10 nV/°C at 27 kOe. The magnetoresistance Δρ of these wires followed a similar trend to the magneto-Seebeck coefficient. Δρ/ρ increased with increasing ΔS values.

By using a pulse measurement procedure, we have studied the electronic emission coefficient and the energy distribution of the secondary electrons when a NaCl single crystal is bombarded by He⁺, Ne⁺, Ar⁺, and Kr⁺ ions, the energy of which varies from 1 to 30 keV. The dependence of the coefficient of electronic emission on temperature is interpreted by the interaction of the excited electrons with phonons.

The demagnetizing energy or magnetostatic self energy of a spherical particle is calculated for two groups of magnetic `free pole' configuration. The groups are chosen so that the calculated energies are directly applicable to problems concerning (a) non-magnetic inclusions in uniaxially magnetized domains and supporting `spikes' of reverse magnetization and (b) spherical magnetic particles of uniaxial materials containing few domains. The results are presented in graphical form but the various equations should be easily applied to different problems.

Changes in electrical conductivity of CdS films, deposited under ultra high vacuum and subsequently exposed to hydrogen adsorption, have been measured. From a plot of the rate of conductivity increase against reciprocal absolute temperature, an activation energy of 4·7±0·1 eV has been deduced for the adsorption of atomic hydrogen on CdS.

Domain structures are considerably influenced by the effect of conductivity, imperfections, etc. in a real crystal. The domain walls are found in positions which are not normally expected in a perfect insulating crystal. This leads to head-to-head or tail-to-tail couplings, while the normal way of the coupling of dipoles is head-to-tail, for no charge condition on the wall. The step ladders on pseudocubic (001) planes have been studied from this point of view using the technique of multiple beam interferometry. It has been found that an association of 180° domains with 60° domains can also be indicated in a simple way.

The theory of light modulation using the linear electro-optic effect in class 4bar above 2m crystals is briefly reviewed. Errors resulting from deviations in the direction of crystal cut and from the finite beam angular aperture are calculated.

The cross section for stimulated emission for the 1·06 μm transition of Nd³⁺ in glass has been determined by both laser and fluorescence measurements. Results from both methods were found to agree for six glasses studied. The change of cross section determined by the fluorescence method was established for a systematic change of glass composition over a wider range of glasses within the silicate system. The cross section was found to vary by a factor of 3·0, the highest value being obtained for a high Li₂O/low SiO₂ glass.

The role of SBS-generated hypersonic phonons in the formation of laser-induced oriented optical damage in polymethylmethacrylate was investigated by means of the Bragg diffraction of a probing He-Ne laser beam. No evidence whatsoever of SBS-generated phonons was found during damage formation.

The heat of formation of tin-nickel electrodeposit was measured by solution calorimetry. A single phase obtained between 47·5 and 60·3 at.% Ni had a heat of formation (at 298 K) best represented as a linear function of composition. The constant partial molar enthalpies were, for nickel, −11·6 kJ/g atom, and for tin, −47·4 kJ/g atom; the estimated error is 5% or more. The alloy, not a phase appearing in the accepted phase diagram, decomposes slowly above 575 K with a small positive enthalpy change for the reaction considered at 298 K. It can be argued that the alloy is in fact a phase stable at its usual temperature of formation 340 K. The volume change on formation of the alloy is constant within the single-phase composition range and is numerically quite large (−1·3±0·1 cm³/g atom at 298 K). It is believed that this is the first study made of an alloy intermetallic compound which can only be prepared at relatively low temperatures.