Table of contents

The time-dependent transmitted and reflected waveforms for an impulsive plane wave incident normally on an anisotropic lossy plasma half-space are computed by the use of an inverse Fourier transform. The amplitudes of the parallel and crosspolarized components of electric field are noticeably damped, while the period of oscillatory response is practically independent of collision frequency.

The spray electrification of xylene, both pure and contaminated with ionic additive, has been studied using fine stainless steel nozzles from 250 to 600 μm in diameter. A power dependence on flow velocity, nozzle diameter and conductivity has been found which is not explained by pipe flow theory.

Spontaneous endlight and sidelight measurements on an He/Cd segmented bore laser have revealed the contribution to the laser inversion made by moving striations.

The state-space method is used to obtain an exact solution for radiation from a spherical antenna coated by a lossy and inhomogeneous layer of plasma. It is assumed that the plasma is isotropic and the density decreases sinusoidally with radial distance. Between the antenna and the coating a plasma sheath is also introduced. The results show that for appropriate dimensions of the antenna and the plasma layer, a considerable increase can be obtained in the radiation of the antenna, by lowering the antenna frequency below the maximum frequency of the plasma.

The apparent voltamperes into a nonsinusoidal circuit can be considered as the resultant of three hypothetical components known as the active voltamperes S_R, the true reactive voltamperes S_X, and the apparent distortion voltamperes S_D. In systems with nonsinusoidal voltage, the active voltamperes S_R differs from the average power P, and the true reactive voltamperes S_X differs from the quantity Q=Σ₁ⁿE_nI_n sin I_n frequently quoted in the literature.

The reactive voltamperes of an inductive load can only be completely compensated by capacitance to give unity power when the load impedance is linear and the voltage is sinusoidal. If the load impedance is nonlinear or the load voltage is nonsinusoidal or both, some improvement of power factor may be relizable by capacitance compensation, but the highest power factor achievable is less than unity.

Formulae are developed to give the maximum power factor and the minimum reactive voltamperes achievable by capacitance compensation. The optimum value of capacitance to give maximum power factor operation is defined for the cases of nonsinusoidal voltage supplying a nonlinear load, nonsinusoidal voltage supplying a linear load, and sinusoidal voltage supplying a nonlinear load.

The amplitude distribution of prepulses - those pulses preceding ion afterpulses in photomultipliers - depends upon the point of origin of the afterpulse. This fact provides a method of confirming the discovery that, in RCA types 8850 and 8852, afterpulses were formed throughout the photomultiplier. A simple theory of afterpulse formation in the region between the cathode and the first dynode is used to derive the prepulse height distribution.

The degree of polarization of dye laser light (Rhodamine B) under excitation by a beam of linearly polarized light (second harmonic of a ruby or neodymium-doped laser) has been studied experimentally by measuring the Stokes parameters of dye laser light at a wavelength of 0·6328 μm. The sign of degree of polarization is negative in the case of the ruby second harmonic (0·3471 μm), whereas it is positive in the case of the neodymium second harmonic (0·53 μm). Even in the case of illumination with the ruby second harmonic, the magnitude of the degree of polarization increases with the intensity of unconverted ruby laser light. The experimental fact that the degree of polarization of dye laser light is nearly as high as unity has been explained fairly well on the basis of the theory proposed by the authors.

Cylindrical probe measurements have been made in an iodine vapour plasma containing a ratio of negative ions to electrons sufficiently large for electron effects to be neglected; that is, in a positive-ion/negative-ion plasma. The ion-ion plasma was produced by allowing plasma from the Faraday dark space of a DC iodine glow discharge to diffuse into a side-arm. Almost symmetrical current-voltage probe characteristics were obtained in the ion-ion plasma, and both the positive and negative current regions were found to follow a (current)²-voltage relationship, as would be expected if ion collection by the probe was orbital-limited. The characteristics were analysed using orbital-limited current theory to provide data on plasma potential, ion number density and Debye length. The results of the analysis are consistent with the dominant ion species in the ion-ion plasma being I₂⁺ and I₃⁻.

Investigations on the temporal development of single streamers performed with an electro-optical streak camera are reported. The starting electrons being released by the focused radiation of a giant-pulse ruby laser, any influence of the electrodes can be excluded. In rare gases the propagation of streamers was observed in a homogeneous field (gap distance 11 cm), beginning from low E/p and up to higher E/p corresponding to overvoltages of several hundred per cent. The velocities of different streamer stages are measured. Below a certain E/p value they are relatively low (some 10⁸ cm s⁻¹) and proportional to E/p. Above this value, a considerable increase in streamer velocity up to 5 × 10⁹ cm s⁻¹ is observed.

In metal discharge tubes composed of long isolated segments operated at high current densities, the axial electric field in the plasma is significantly greater than that in a solid dielectric-walled tube. A characteristic feature of such discharges is that the ion current j₊ to the tube wall appears to be a function of the potential V of the wall relative to the plasma. The discharge models investigated are those giving constant j₊ and j₊ is proportional to V^½, V^{fraction three-twos}, exp (cV) and {(cV)⁻¹ + [exp (−cV)−1]⁻¹} respectively.

The maximum length of a floating metal segment along which a discharge can be maintained is determined by the ratio of electron temperature to axial electric field. The constant of proportionality is shown to depend on the form of the ion current to the metal wall, and varies between 3 × 10⁻⁴ and 3 × 10⁻² V K⁻¹.

The conditions for the occurrence of partial discharges are analysed and shown to be governed by the trapping of charged particles on the insulation surface. According to the instant of trapping, each discharge is constituted either by a series of single avalanches or by a sort of degenerated spark which constitutes the partial discharge.

Charge deposits form a sharply defined site which is visible on powdering. If the deposited charge is positive, the site is star-shaped and its shape and size strongly depend on the electronegativity of the gas. If the discharge is negative, the site is circular in all the gases. By measuring the site diameter it is possible in several cases to calculate the capacity involved in the discharge, and consequently the energy of it. This technique is used to study the explosivity of vapours in spaces bounded by charged insulators.

Coherent anode oscillations occuring in an argon ion laser are examined. Time-resolved spectra of the light emission from the anode zone are compared with the voltage oscillations. A simple model for anode oscillations is compared with the experimental data.

Experimental results of the efficiency of current-limited spark chambers as a function of pulse delay are described for various types of chamber. These results are presented partially for their direct application to extensive air shower research, but primarily with a view to clarifying the spark mechanism operative in these chambers. A phenomenological theory describing the efficiency-pulse delay relationship is derived and compared with the experimental data. The results of this comparison are discussed and their relevance to a theory of the spark mechanism emphasized.

The electrostatic probe is an attractive diagnostic tool for DC welding arcs. The current-voltage characteristic of such a probe in a high-current (100 A), high-pressure (1 bar), DC argon (TIG) welding arc is found to give a very flat ion saturation region. Sheath thicknesses are estimated to be much smaller than 1 μm, ie less than the mean free path, so that conventional collisionless probe theory should be applicable - in contrast to the flame plasmas dealt with by Thomas (1969) and others. By using an Abel inversion technique the measured probe current can be converted to a radial distribution of current. The temperature variation across the arc can then be derived from this current variation. By measuring the radial temperature profile at several points along the length of the arc column, a complete isothermal map of the TIG arc has been obtained which is important for basic calculations of energy transfer.

The effect of an impressed axial flow on the characteristics of a low-current DC are burning between copper electrodes is examined. A model for the system is assumed in which the arc core is considered as a heated solid rod transferring heat to the surrounding gas flow by conduction. An analysis of the heat transfer process leads to an expression describing the arc characteristics under these conditions, and a good correlation of the experimental data. A transition from convection- to flow-dominated régimes is observed at air flows above 4-6 m s⁻¹.

Oscillations in a transient DC glow discharge at different pressures in air have been studied, and the effect of pressure, voltage and current on frequency, amplitude and cut-off current is shown. A mechanism for the generation of oscillation based on space-charge field modification is discussed.

The interaction of Trichel current pulses of neighbouring negative needle coronas is investigated. As the needles are brought together and the strength of the coulomb interaction increases, the experiments show that (i) the pulse frequencies decrease, (ii) the frequencies converge, and (iii) the relative phase of the pulses approaches π. A coupled circuit model of the interaction between the charge clouds is presented which shows that these observations are dictated by energy considerations. Localized Trichel corona spots also occur on negative corona wires, and the behaviour of these discrete discharge spots is explained by experiments and analysis.