Table of contents

The tuning of an atomic iodine photodissociation laser over 11 wavenumbers is reported. The Zeeman effect is used to tune the laser, which is operated in a longitudinal magnetic field of up to 75 kG.

Neel's theory of the Aleonard-Brissonneau experiments to determine domain wall specific surface energy gamma ₀ by observing the catastrophic torsional distortion of 180 degrees domain walls that span a monocrystalline platelet carrying a current density I is reviewed. It is shown that when I>0.924 gamma ₀/JL² (J is the saturation polarisation of the platelet of thickness 2L) Neel's domain wall profiles are only metastable against kink distortions of the Graham-Neurath type. Kinked profiles with lower energy have been calculated from the same equations and boundary conditions and these provide an alternative interpretation of the catastrophic transition observed. As it is not decisively clear which interpretation is correct, a degree of ambiguity would seem to exist concerning wall energies gamma ₀ deduced from the published observations.

The strain coefficient of resistivity of resistors deposited on a substrate is analysed, and an expression for its correlation to the measurable longitudinal and transverse gauge factors is given for isotropic (metal, amorphous semiconductor and cermet) resistors. The use of this expression for the evaluation of other physical quantities is exemplified by the calculation of the temperature coefficients of resistivity for resistors on substrates of different thermal expansivity.

Bombardment of Si by 50-500 keV Ar⁺ ions results in the emission of optical lines in the spectral range of the visible and vacuum ultraviolet. The authors present results of the measurements on the energy dependence of various lines emitted by sputtered atoms and ions. The results indicate that the photons are emitted by those atoms which are sputtered in a few collisions from the near-surface region.

Analyses conditions of oxide formation on silicon in a microwave plasma. Experimentally it is shown that oxide films can be created only in a plasma where the floating potential is close to zero or positive with respect to the grounded support of the Si sample.

A number of fundamental definitions of magnification in the field-ion microscope are derived in terms of parameters determined by the projection geometry. These expressions for magnification are then compared with analogous ones derived using an empirical approach. The precise dependence of magnification on emitter radius is discussed, and a particular example of an asymmetric emitter is worked through in detail.

The finite slab problem is reduced to a semi-infinite one by adding an infinitesimally thick layer such that both the added layer and the total layer are semi-infinite. The relation between the reflection and transmission functions for a finite slab and those for an infinite one are obtained in terms of an operator which satisfies a semigroup equation. The method is applied to anisotropic scattering with azimuthal dependence. Numerical calculations are made and the results compared with those of Kavenoky (1971).

The reconstruction of three-dimensional refractive index distributions from interferometric fringe data is required for interpreting interferograms. The layergram iterative method of reconstruction represents a new application of computer analysis techniques which is shown to have good convergence and is economical in computer store requirements.

Expressions have been derived for the irradiance distribution in the diffracted field of an aberration-free semitransparent and pi -phase aperture illuminated by a Gaussian laser beam. Calculations have been done for various values of the real part of the complex amplitude transmittance of the central region of the aperture. The results have been presented graphically.

Positive-ion drift velocities have been measured in ethane gas, using the pulsed Townsend-discharge technique, over the range 40<or=E/p<or=1400 V cm^-1 Torr^-1. For 40<or=E/p<or=825 V cm^-1 Torr^-1 positive-ion drift velocities were determined by the balanced-pulse technique, and in the range 490<or approximately=E/p<or=1400 V cm^-1 Torr^-1 by the positive-ion successor interval method. The experimental ion drift velocity was found to exhibit a linear dependence on E/p at low E/p changing to a square-root dependence at high E/p. Townsend primary ionisation coefficients have also been determined from balanced pulse measurements over the range 40<or=e/p<or=825 V cm^-1 Torr^-1. Experimental values of alpha /p are in moderate agreement with Townsend conductivity determinations, being lower at low E/p and higher at high E/p. In the range 100<or=E/p<or approximately=250 V cm^-1 Torr^-1 agreement is good. The effect of space-charge field distortion on the alpha coefficient, and hence the balanced pulse, is discussed. It is found that space-charge field distortion can produce different effects on the alpha coefficient in the pulsed and steady-state Townsend discharges.

A pulse technique has been developed for the measurement of the ratio of radial diffusion coefficient to mobility (D_r/ mu ) for electrons in gases. Measurements have been made in carbon dioxide for the experiment range 14<E/N<5650 Td, where E/N is the ratio of electric field to gas number density. The magnitude of D_r/ mu increases with increasing E/N; values vary from 0.85 V to 18.0 V and show good agreement with the results of Lakshminarasimha et al. (1974) obtained by using a steady-state technique for E/N up to 1553 Td. A Monte Carlo simulation of the present experimental range shows that the equilibrium conditions exist after the electrons have travelled a short distance within the gap.

Strong electric double layers are produced in a low-density plasma column confined by an axial magnetic field and maintained by single-ended inflow of plasma along the magnetic field. The double layer evolves from an anode sheath, and ionisation within the sheath is shown to be a significant process in this conversion. Once the double layer has been formed, its axial position can be controlled by the external electric circuit. The layer exhibits an axial motion back and forth with amplitudes somewhat larger than the layer thickness.

Coherent nonlinear coupling of ionisation waves, where the beat-wave is out of resonance, is treated experimentally by using space correlation functions. The observed nonresonant interaction of ionisation waves results in an onset of turbulent wave motion.

An experiment involving the breakdown in atmospheric air due to a CO₂ laser beam is described, which also shows, next to the familiar breakdown or detonation wave, a luminous feature which appears to re-enter the breakdown plasma just created. Using a two-dimensional Lagrangian numerical code, this phenomenon is analysed, and it is found that the laser beam re-enters the plasma at a rate governed by lateral expansion and cooling and hence change in absorption coefficient of the plasma. This passive mechanism is later supplemented by an active wave process in which the beam reheats the plasma and generates a motion closely related to the R-type ionisation front postulated by Axford (1961). Agreement between experiments and numerical analysis is good, though limited by the scarcity of available data on the absorptivity of air at high temperatures.

The problem of the calculation of the electron current to a retarding spherical probe at intermediate gas pressures is discussed. The current to the probe is expressed in terms of the probability with which an electron, from the undisturbed plasma, will reach the probe surface. This probability is obtained using the Monte Carlo method. It is shown that the computer simulation has to be performed only for the laser few free paths of the electron near the probe surface. An integral equation for the electron energy distribution function in the undisturbed plasma is received from the expression for the electron current.

The electron energy distributions of slow and secondary electrons in the negative glow of an argon sputtering discharge with a titanium cathode have been studied with single probes and grid probes. Measurements have shown that the distributions are generally of the Druyvesteyn type for the two electrons groups. In the range 40-120 mTorr, the slow group is characterised by a mean energy (E₁) approximately=0.5 eV and an electron density n_e1 approximately=1*10¹¹ cm^-3; the secondary group has a mean energy (E₂) approximately=7 eV and a density n_e2 approximately=1*10⁹ cm^-3. These values are dependent on the discharge parameters and are mainly influenced by the pressure for the slow group and by the discharge power for the secondary group.

It is shown how the results of Daalder (1977) can be adequately explained using the previous theory of Beilis (Beilis et al. 1972, Beilis 1974) for currents exceeding approximately 100 A.

The interfacial layer in a Schottky barrier solar cell plays an important role in reducing the dark current, which in turn improves the open-circuit voltage and the efficiency of the solar cell. From the transmission of electrons and holes across the interfacial layer the authors have analysed the current mechanism in a Schottky barrier solar cell under two approximations. It is shown that the efficiency of the cell increases at first with the interfacial layer thickness delta , and after acquiring a maximum value falls with a further increase of delta . The variation with delta of I_sc, the short-circuit current, V_oc the open-circuit voltage and FF the fill factor are also discussed.

Atomically clean (111) silicon surfaces have been produced in ultra-high vacuum both by cleavage and by heat cleaning. These surfaces were deliberately contaminated with oxygen in a controlled manner. During this contamination the work-function changes were studied using the Kelvin vibrating capacitor technique. Schottky diodes were then made by evaporation of copper or gold on to these surfaces without breaking the vacuum. The barrier heights of these devices have been measured by the I-V , C-V and photoelectric methods. In general, the barrier height obtained from C-V data was greater than that obtained from the other two techniques and this difference was found a decreasing function of oxygen exposure. With copper the devices produced after exposure to oxygen aged, over a period of months, in such a way that the barrier height differences increased. The experimental results have not revealed any simple correlation between the work-function changes caused by oxygen adsorption and the corresponding barrier height changes caused by the presence of oxygen at the interface.

Time lags to breakdown in n-hexane have been measured in high over-voltages, using approximately 100 ns width rectangular pulses. At a constant field, the time lags are divided into two types, N and B, at a 'critical distance' d. Above d_c, Laue plots have almost similar kinked-line characteristics; below d_c, a straight line is obtained. From the prebreakdown current estimated from the voltage waveforms, the ionisation coefficient, which increases with increasing applied fields, is estimated to be 6.4*10³ cm^-1 at 3.5 MV cm^-1, provided a single-electron avalanche is occurring.

The K-shell cross-section sigma _k( alpha ) for 80 kV electrons traversing thin single crystals of graphite and aluminium nitride is measured from electron energy-loss spectra generated within an energy-selecting electron microscope. These values, and the variation with collection semi-angle alpha , are compared with values derived from Bethe theory for an isolated atom.

Investigations on the adsorption, surface migration and thermal desorption of LaB₆ on the tungsten surface are made using as simple field emission microscope. The distribution of the absorbate over the surface, as the tip is heated progressively at higher temperatures, has been broadly classified in two stages corresponding to two different temperature ranges, 850-1100K and 1700-1950K. During the first stage, protrusions grow on (111) planes, with the estimated activation energy 1.0 eV and the pre-exponential factor 10⁴ s^-1. In the second stage, these protrusions are dispersed, resulting in a 'well-spread' pattern. Further heating in the same temperature range causes thermal desorption of the adsorbate, leaving the tungsten tip clean. The average activation energy for thermal desorption has been estimated to be 5.1 eV with a pre-exponential factor 10¹² s^-1. The results suggest that LaB₆ evaporates as atomic La and B from the tungsten surface. The behaviour of the adsorbate over the surface is discussed in the light of the results.

A quantitative method of testing adhesive-bonded joints is described based on the use of a very-high-resolution ultrasonic probe which is capable of resolving the echoes from the two interfaces of the bondline. The use of the method and its advantages are illustrated on some aluminium-to-aluminium joints bonded with Metalbond 328 of Narmco Corp.

In medical ionography an electrostatic latent image of the object is formed by collecting on an insulating sheet the ions formed in a layer of compressed gas or of insulating liquid. This latent image can then be rendered visible by powder cloud development. In order to reduce the radiation exposure to the patient it is desirable to improve the sensitivity of this development. In ionographic autoradiography the sensitivity of powder cloud development has been increased by using thicker dielectric layers instead of the 50 mu m foils normally used. The use of thicker sheets in medical ionography, however, resulted in a severe loss of detail. The contrast is reduced with increasing layer thickness for non-repetitive patterns typical of medical images. Contrast may be regained for a repetitive pattern. From these findings it may be concluded: (a) that an increased sensitivity in medical ionography is not feasible by powder cloud development on thick dielectric layers, and (b) that resolution measurements with a repetitive pattern need to be evaluated very carefully in this case.