Table of contents

Ions in an optically pumped mercury vapour plasma, principally by the reaction Hg(6³p₀)+Hg(6³p₁) to Hg₂⁺+e, have been extracted using a novel device. Initial experimental results show that the ratio of the flux of atomic ions to that of molecular ions formed in the plasma and arriving at the device varies in a fashion reasonably well described by a theoretical model but differs substantially in absolute value.

Recent criticism of the theory of the Taylor cone is refuted.

Effects of free convection currents and mass transfer on the unsteady flow of an electrically conducting and viscous incompressible fluid around an accelerated infinite vertical porous plate subjected to variable suction proportional to (time)^-12/, in presence of a transverse magnetic field, have been studied. Analytical solutions to the unsteady flow and the skin friction have been derived for small magnetic field parameter M. During the course of discussion, the effects of the Grashof number G_r' the modified Grashof number G_e (depending on the species concentration difference), and the Schmidt number S_e have been discussed.

Cathode-directed electrical discharges in air at 100 Torr, advancing under the influence of single modulated high voltage pulses, have been found to have the appearance of a necklace of luminous beads. Photomultiplier studies reveal the monotonic advance of the wavefront, the oscillatory nature of the bright regions and a suppressed emission from the dark spaces.

The equation a_T=a₀ exp(CE_T) where a_T is the lattice constant of a cubic metal at TK, C is a constant and E_T= integral ₀^T C_V(T)dT where C_V(T) is the atomic heat of the metal at TK and constant volume was observed by Mitra and Mitra (1963) to yield the value of a₀, the lattice constant at absolute zero, for copper, and has since been confirmed by Kroeger (1977). In the present work, the same formula has been used to extrapolate the value of a₀ for some cubic elements. These values of a₀ have been compared with experimental measurements of a₀ for these metals at temperatures very close to 0K and the agreement between these two sets of observations is found to be excellent. This gives us a method for predicting a₀ from measurements at relatively high temperature.

The spectral behaviour of the kinetics and polarity of the transverse acoustoelectric voltage in the separated-medium surface acoustic wave configuration is studied experimentally. It is shown that the polarity of the effect can be different for one and the same type of conductivity in agreement with a recent theoretical study. Two types of transient characteristics are revealed and explained on the basis of a simplified equivalent circuit. Indications are given on possible spurious effects of the external measurement circuitry.

The optical absorption spectra of facet core free and facet core parts in Bi₁₂SiO₂₀ have been measured precisely at various temperatures between 30K and 300K, and analysed in the region below the fundamental absorption edges. In both cases, it can be seen that the absorption coefficients plotted semilogarithmically vary linearly with photon energy h(cross) omega (Urbach's rule). The values and the temperature dependence of Urbach parameter in facet core free are larger than those in facet core in Bi₁₂SiO₂₀.

A Malvern particle sizer was used to measure the size of droplets leaving the throttle plate of a carburettor fuelling a four-cylinder test-bed engine. The effect of engine vibration on the measured values was investigated and techniques were developed to keep the viewing windows clear of impinging droplets. Droplet size distributions are reported for various engine loads.

The authors examine the Doppler profile of a collimated effusive atomic beam for two situations of experimental interest. First, they show that there is an advantage in making the source and collimating slit of comparable size. Secondly they discuss the distribution along a line of sight which is not normal to be beam axis. They compare the predictions with some preliminary experiments.

X-ray distributions in aluminium, gold and carbon have been measured using the tracer technique. The results have been compared with calculations using the simplified Monte Carlo method of Love et al. (1977). It is concluded that the Monte Carlo model predicts too high a peak to the X-ray distribution and that the discrepancy is worse at lower overvoltage ratios. Calculated surface ionisation functions Phi (0) are, however, closer to experimental values and the general shape of the Phi ( rho z) curve is considered satisfactory. The more rigorous model of Myklebust et al. (1976) also overestimates the height of the peak in the Phi ( rho z) curve and, whilst it performs rather better than the simplified model for gold, it gives even more unrealistic data on carbon.

For pt.I see ibid., vol.18, p.1233 (1985). A new absorption correction is developed which is shown to work well for quantitative analysis of all systems including light elements. It is based upon representing the X-ray depth distributions as a quadrilateral shape defined by the mean depth of X-ray generation and the position and relative height of the peak in the X-ray depth profile. Equations for describing these parameters in terms of target material, incident electron energy, etc. are derived by reference to tracer and Monte Carlo determinations of X-ray depth distributions.

For pt.II see ibid., vol.18, p.1245 (1985). Four recently proposed correction procedures for quantitative electron-probe microanalysis have been evaluated using a fairly comprehensive collection of microanalysis results. The Love-Scott II model, which incorporates the new 'quadrilateral' absorption correction, gives the best performance for all specimens included in the database. The methods proposed by Bastin (1984) and by Tanuma (1983) are reasonable for heavier element analysis but are less satisfactory when applied to the lighter elements (Z<10). The Love-Scott I model, with its simpler 'square' Phi ( rho z) profile, is not recommended for use when absorption effects are large, say 50%, although it can deal satisfactorily with heavy element analyses.

Aggregation phenomena in water-based magnetic liquids have been observed using a polarising microscope fitted with a phase compensator. Macrochains formed in an external magnetic field are shown to have a greater birefringence than the surrounding 'unchained' liquid. A simple model of the colloidal suspension allows the optical data to be interpreted and hence a numerical estimate of the birefringence of the macrochains to be made. This indicates that for fluid diluted five-fold, the chain birefringence is some four times greater than the initial bulk birefringence. It is also found that the bulk fluid birefringence remains essentially constant for times up to 2 h, independent of the formation of macrochains.

Radiation at 121 nm has been generated by sum-frequency mixing of the 277 nm radiation, from a Raman-shifted KrF laser beam, with 431 nm dye laser radiation tuned to a two-photon resonance in Mg vapour. The generated vacuum ultraviolet power was limited by an anomalously low breakdown intensity of the Mg vapour (24 MW cm^-2) for 277 nm light. No phasematching effects were observed when either the negatively dispersive Kr or the positively dispersive Ar was added. Both of these effects are ascribed to the close coincidence of the 277 nm radiation with the Mg((3s²)¹S₀-(3s3p)¹P₁) transition. Conversion efficiencies were measured to be of the order of 10^-6.

The VUV line emission of a high-current DC hollow cathode was investigated in the wavelength region 10 nm to 100 nm. Spectra of quadruply ionised atoms could be observed. The radiance in the Al IV lines at 13 nm and 16 nm and in the He II Lyman-series was determined by a comparison with the spectral concentration of radiant intensity of the synchrotron radiation emitted by the electron storage ring BESSY. The authors found the radiance of the lines to be reproducible within +or-25%.

The temperature variation of the thermal conductivity of pitch-derived carbon fibres measured from 3 to 300K is reported and discussed. The results are compared with those previously obtained on benzene-derived fibres (BDF) and with the room-temperature thermal conductivity of PAN-based carbon fibres, the GY70. The data confirm that the structure of pitch fibres is less perfect than that of BDF. At room temperature a thermal conductivity comparable to that of copper is found.

The 3-dimensional form of a draining oil drop is complex and varies with time. Dynamic similarity between the time development of drops of different size may be predicted on the assumption of dependence only on body force, viscosity and surface tension. The edge conditions, which introduce an additional edge height parameter, modify this to an approximate, experimentally verified, similarity with slightly different power laws. The edge velocity versus edge slope variation is found to control the shape and progress of the drop. The form of the head may be predicted by a differential equation based on 2-dimensional flow if allowance is made for transverse curvature. In the tail of the drop the influence of surface tension is mainly an indirect one. The tapering plan form which results from the edge velocity condition modifies the parabolic profile which is obtained in 2-dimensional flow.

New photoabsorption measurements in the 110-200 nm range using an electron storage ring and a low-resolution monochromator have been made for a number of important gases of relevance to both semiconductor etching and gas discharge studies. Important effects of low-level corona disruption of SF₆ and the resultant UV absorption changes are also reported.

A filament discharge plasma reactor, with electrostatic confinement, has been constructed. An Ar plasma has been produced in the reactor at a pressure of 10^-1 Pa and has been investigated with a movable Langmuir probe. The plasma parameters have been determined and the influence of the confining potential on them has been studied. Three electron populations with different temperatures have been found. In addition to the electrons emitted by the filament and those generated in the plasma ionisation processes, there exist the electrons originated by secondary emission at the walls. These are energetic enough to contribute to the ionisation in the plasma. It has been demonstrated that by electrostatic confinement a significant increase of the plasma density can be achieved.

Experiments are described in which a high-current electron accelerator is used to generate, by Al wire explosion, super-high-density (N_e=5*10²² cm^-3), high-temperature (T_e approximately=400 eV) plasmas. The X-ray emission characteristics of the plasma are analysed.

Current and radiation noise spectra of normal and abnormal obstructed glow discharges were analysed. In the normal glow discharge ionisation waves are usually present. The high number of harmonics of such waves is the source of intense radiation and current noises below 1 MHz. Thermal, generation-recombination and flicker current noise, shot and flicker radiation noise spectra were measured and are discussed.

The Townsend first ionisation coefficient alpha _T and the secondary coefficients gamma _T have been measured in binary and ternary mixtures of CO₂, N₂ and He gases by the steady-state Townsend method for E/p₀ values from 54 to 215 V cm^-1 Torr^-1. The results show that the alpha _T/p₀ value in mixtures of CO₂ and N₂ increases monotonically with increasing N₂ partial pressure whereas in binary mixtures of He and N₂, or of He and CO₂ there is a maximum value of alpha _T/p₀ as the partial pressure of He is increased for E/p₀>or=108 V cm^-1 Torr^-1. The alpha _T/p₀ values of ternary mixtures of these gases are shown on a trigonal prism plot. The value of gamma _T is found to decrease rapidly with increase of the component having a small value of gamma _T.

Polycrystalline samples of copper sulphide having a predominantly tetragonal phase have been investigated. The thermal activation energy of the electric conductivity Delta E_a=0.15 eV associated with the activation energy of an acceptor level has been determined. Low-temperature photoconductivity has been observed for the first time. The photoconductivity spectra measured at 80-120K reveal an appreciable surface recombination rate. From the analysis of these spectra, the minority carrier diffusion length (3.3*10^-5-2.5*10^-4 cm) and the possible lower limit of the band gap ( Delta E_g approximately=1.2 eV) have been evaluated.

The turning points on the curve of resistance versus atomic ratio ( gamma =H/Me) were correlated with the phase boundaries of the H/Zr and H/Sc systems at 295K. The changes of work function with r showed that, at low value of r, some H atoms adsorbed on the surface and increased the work function slightly for Zr but decreased it for Sc. They also showed that for H/Zr the Fermi level mu of the FCC hydride was 0.2 eV higher than mu of the FCT hydride and at least 0.07 eV higher than mu of the clean metal. However, mu for Sc was about 0.2 eV higher than mu of its (FCC) hydride.

Very low frequency dielectric spectra (10³-10^-4 Hz) show space charge build-up in alpha PVDF under the influence of a low field as the temperature is raised to 80 degrees C or above. Preheating changes very little the medium frequency alternating conductivity when gold electrodes are used, whereas with aluminium electrodes the conductivity increases substantially as excess fluorine collects under the electrodes. Experimental results are interpreted through a two-carrier (H⁺ and F^-) conduction model based upon three assumptions: (i) an increase in H⁺ ion concentration during the first heating: (ii) partial deep trapping of H⁺ ions at high temperature; (iii) field-induced extraction of F^- ns from the bulk leading to H⁺ enhancement of pre-exponential mobility factor. Orders of magnitude are proposed for H⁺ and F^- ion activation energies.

A simple method to compute approximate cross sections for the elastic scattering of electrons by atoms in the intermediate energy range ( approximately=500 eV-50 keV) is proposed. The method starts from the static field approximation, the atomic potential being derived from self-consistent procedures. Calculation results are in good agreement with experimental ones for free atoms. For atoms in solids, the solid structure is taken into account by using Wigner-Seitz boundary conditions in the self-consistent calculation. This method is suitable to be used in Monte Carlo simulation of electron transport.

In this paper a method is presented to measure the dielectric constant of materials. It is based on the analysis of the discontinuity problem of a circular cylindrical dielectric post centred in a rectangular waveguide. By expanding the field the reflection coefficient can be found. Curves have been made showing the relationship between the reflection coefficient and the dielectric constant of the post. For anisotropic materials the reflection coefficient will also depend on the direction of the cylinder. A rotation of the cylinder gives information on the dielectric constant in all directions. An experimental application is described to illustrate the procedure.

A method of measuring tilt angles in uniformly aligned thin films of a liquid crystalline material is described. Within the range 15 degrees to 75 degrees , the method is accurate to better than +or-0.5 degrees . It depends upon measuring the change in optical path length of one arm of a Michelson interferometer balanced to give the zero-order white light fringe, when a liquid crystal cell is inserted in one of two orientations relative to the incident electric vector of the incident polarised light. Accurate data on the principal refractive indices and their dispersion are required. The interpretation of the data is remarkably straightforward.

This paper considers, largely in the context of metal transfer in arc welding, the stability of a viscous, current-carrying fluid cylinder with surface charge. Simple expressions are obtained between the characteristics of instability growth and the dependence of these on 'process parameters'. For liquid metals it is found that viscous effects are negligible while electrostatic effects exert a stabilising influence and may have a number of interesting effects on stability (particularly at low wire currents).

For pt.I see ibid., vol.18, p.1431 (1985). A model of metal transfer in open arc welding is developed which provides a framework for predicting the influence of welding parameters in steady and pulsed current welding. It is found possible to present results in terms of simple expressions which are in broad agreement with available observations. Considered here are aspects such as droplet evolution, detachment frequency and size, selection of pulse parameters, influence of electrostatic stress on transfer, drop dynamics and the relationship between balance of force approaches to transfer and the present instability model.