Table of contents

A direct-current transformer has been constructed in which the primary consists of a thermocouple and the secondary consists of a thermomagnetic element. It is shown that this arrangement can be operated as a variable transformer, the voltage multiplication being controlled by the applied magnetic field.

It is shown that the threshold of parametrically excited fractional harmonics (subharmonics) can be used to determine the attenuation of ultrasonic waves in liquids. From this attenuation the viscosity can be calculated. The existence of cavitation in the liquid is not essential to such measurement.

Expressions are derived for the resolution functions appropriate to chromatic and spherical aberration in electron microscopy. The resolution functions are given for the two components of the scattered electron beam (i.e. the elastic and inelastic components). The analysis given here for the elastic component is valid only in the case of incoherent radiation (i.e. using a large condenser aperture). The treatment of aberrations caused by inelastic scattering in the specimen assumes that the inelastically scattered radiation is incoherent, an assumption which is valid for amorphous specimens.

Numerical results are presented for the angular distributions for plural elastic and inelastic electron scattering in `amorphous' carbon films. It is shown that the contribution of the inelastic scattering to the intensity scattered within the objective aperture (semi-angle in the range 10⁻³-10⁻² rad) exceeds the elastic scattering by an order of magnitude for film thicknesses in the range 50-2000 Å and for incident electrons of energy 100 keV. The variation of the electron energy loss spectrum of carbon with the angle of scattering is given. The angular and energy distributions are used to calculate the spherical, chromatic and combined aberration functions appropriate to carbon specimens of thickness 50-1000 Å and for 100 keV electrons. In the incoherent approximation, it is concluded that for carbonaceous materials, inelastic electron scattering dominates all resolution considerations; in particular the chromatic aberration effect due to the inelastic scattering is far more significant than the contribution from spherical aberration.

The present treatment of aberrations would seem relevant to low voltage electron microscopy and to thick (about 1000 Å) films of carbonaceous materials at conventional voltages; in both these situations inelastic electron scattering determines the resolution attainable.

Dynamic refocusing in microspot cathode ray tubes is seen to be necessary to compensate for an increase in the focus coil image distance with deflection, for deflection astigmatism and for axial yoke field components. The dynamic focus current required for each of these three focusing effects is shown to vary as the square of the deflection, to a first order, and the dynamic focus current required to correct for deflection astigmatism is shown to be half that required to correct for the increase in the focus coil image distance.

Deflection astigmatism, which is the principal spot aberration in large (9-inch) microspot cathode ray tubes, is seen to vary as the square of the deflection, to a first order. Deflection astigmatism correction by analytically determined diquadrupole currents should, therefore, be possible. The application of these dynamic astigmatism correction currents prevents the spot from growing by more than 5 micro-metres at a ± 20° deflection angle on 9-inch microspot cathode ray tubes.

This discussion describes the development of a small thermionic emitter, which for simplicity will be called the LM-cathode. The most suitable fabrication and manufacturing techniques are outlined. This cathode has several attractive features when compared with the tungsten filament, which is quite frequently employed in thin electron beam-forming apparatus. Besides its durability in use, emission tests reveal promising features. In pulse operation at 1400 K, densities of the order of 40 A cm⁻² can be seen. For continuous operation at the same temperature, the emission is 5 A cm⁻². On the basis of its continuous operation, it was estimated that the source brightness would be enhanced by a factor of 4 or 5 when compared with the usual operation of the tungsten filament. The LM-cathode is shown to be rugged in comparison to the tungsten filament, as it appears to have a superior life expectancy, and it can also be operated in conventional vacuum devices. Because of these promising features, its role as a substitute for the tungsten filament in thin electron beam devices should be investigated.

The primary object of this investigation was to undertake a practical survey of the operational features of a scaled-down electron gun with an LM-cathode. In order to accomplish this, a study has been made on the effect of reducing the gap between the anode and the cathode-grid assembly on the current density profile across the section of the beam beyond the anode. In particular the value of the peak current density j_m, and the agreement this value has with Langmuir's limit, has been studied. The other parameters which are also included in the discussion are the transmission efficiency of the current through the anode and the perveance of the gun. At each of these electrode gaps, the practical values of the above mentioned parameters have been observed for different anode-to-cathode voltage in the range 50 to 400 V, in steps of 50 V. For easy reference, these values have been plotted on graphs. The profiles of the current density distribution across the section of the beam in the post-anode region, have been found to be very similar to Gaussian, and the parameters of the Gaussian in each case have been computed. The interesting feature of this study has been the fact that with this convergent-flow electron gun it is possible to produce beams with peak current densities of the order of 10 A cm⁻² with a low dc voltage of the order of 400 V between the anode and the cathode. The effect of applying a bias on the grid with respect to the cathode is to cause a general deterioration of the performance. A slight improvement is, however, noticed with a small bias of the order of 0·5 V.

Details are given of an electron beam spectrometer using a magnetic deflection prism as a momentum analyser. The spectrometer is designed for the measurement, as a function of scattering angle, of the discrete energy losses of 10 keV electrons transmitted through thin solid films. The energy resolution is 0·7 eV and the angular resolution before unfolding is 0·8 mrad. The spectrometer has been used to verify the intensity dependence on scattering angle of the volume and surface plasma loss lines of aluminium to a high accuracy and also to obtain an accurate value for the volume loss, namely 15·19±0·03 eV. A folding technique used to analyse the intensity against angular scattering results is described together with a least-squares method for fitting the experimental data to the folded intensity-angle distribution.

Two-layer anti-reflection coatings on crown glass have been made for helium-neon lasers, using arsenic trisulphide and magnesium fluoride as the high and low refractive index layers. The thicknesses of both films were varied and the reflectivity spectra measured. The most efficient coating had a reflectivity of 0·005±0·002% at λ632·8 nm. The coatings are durable and stable. Good agreement was obtained between the measured and calculated reflectivities; the efficiency of a coating is limited by non-uniformities in the film thicknesses. The effect on the reflectivity of departures in film thicknesses from their optimum values was calculated. The optimum thicknesses of the films were calculated for substrates of different refractive indices and for wavelengths from 0·58 to 1·0 μm.

The angular dependence of the intensity of light scattered by solutions of ellipsoidal macromolecules is computed using the assumptions of the Rayleigh-Gans theory. By comparing light scattering measurements with these results, it should be possible to characterize such molecules by three size parameters rather than the usual two.

Thermal diffusion factor α_T of the system ²⁰Ne-CH₄ has been measured as functions of composition and temperature by an effective eight-tube swing separator. The α_T against composition curve shows a dip which cannot be explained by the available elastic or inelastic theories of thermal diffusion in gases.

The thermal conductivity of the semiconducting liquid Tl₂Te was found to be thickness dependent. This phenomenon was explained in terms of a normal thermal conductivity modified by a photon heat transport.

Measurements of the anisotropy in the nuclear magnetic resonance fourth moment for proton resonance of a number of drawn polyethylene films are reported. A general expression is derived for the fourth moment of a partially drawn polymer in terms of a set of coefficients which describe the molecular chain orientation, on the basis of a model in which the drawn film is regarded as an aggregate of units aligned by the drawing process. The orientation coefficients have the form IMG1 where Δ is the angle which a typical unit makes with the draw axis. The IMG2 and IMG3 coefficients determined from fourth moment considerations are in good agreement with values derived in an earlier paper from second moment data, and these coefficients therefore provide a firm basis for the description of the distributions of molecular chains in the drawn polymer. Higher order orientation coefficients are estimated on the basis of a smoothing procedure which suggests itself from x-ray diffraction measurements on the drawn polymer.

The distributions predicted by the combined nmr and smoothed orientation coefficients are consistent with distributions obtained from x-ray diffraction data. The results confirm that the anisotropy of the nmr signal can be attributed to preferred orientation of the crystalline regions of the polymer. The predicted distributions are very non-affine in character.

Finally there is reasonable agreement between the measured mechanical anisotropy and that predicted on the basis of the aggregate model using both nmr and x-ray diffraction data.

Uranium adsorption on the (110), (112), (100), (111), (113) and (116) planes of tungsten field emitter substrates has been studied. Measurements of work function against coverage at 295 K, 950 K and 1120 K were found to be in good agreement with previous contact potential difference and photoelectric measurements. A Topping point dipole depolarization model has been used to determine dipole moments and effective adsorbate polarizabilities. This model is found to be only partially adequate in describing the adsorption process, deviation from the theory being most evident for the (100) plane.

The adsorption of zirconium on the (110), (112), (100), (111) and (310) planes of tungsten field emitter substrates has been studied. Work function against coverage measurements have been made and the dipole moment and effective adsorbate polarizability determined for (100) adsorption. Comparison is made with contact potential difference measurements. As for uranium adsorption, the modified Topping point dipole depolarization model is found to be inadequate in describing adsorption on the (100) plane.

A method is presented, of sharpening metal points which are used in field emission and field-ion microscopy. By sputtering electropolished tungsten tips using neon ions, the tip radius can be controllably reduced from that which results from thermal cleaning (typically 1000 Å) to below 200 Å. These tips, which can exploit the full resolution of the field ion microscope, can be formed on thermally cleaned shanks. The rate of reduction in radius is presented as a function of neon pressure and sputtering current. Changes in the tip shape which result from sputtering, are examined by scanning electron microscopy, and are used as a basis for a simple model of tip sputtering, which leads to the expectation that the method of sharpening is applicable to a wide variety of metals.

Interferometric studies of domain structures in KNbO₃ single crystals, grown by using excess of K₂CO₃ as a flux are reported. The surface deformations produced on the pseudocubic (001) planes by the stable patterns of 90°, 60° and 180° domains have been studied and discussed. If the observed plane is not the plane of the shear, the 90° domains are inclined at 30prime. The angle between the 60° domains is 57prime for the domain line, parallel to a pseudocubic edge. The head to tail coupling of dipoles across the 60° domain walls gives a step ladder structure on the surface. The nature of dipole coupling in a step ladder due to 90° domains cannot, however, be distinguished. The wedges are all 60° wedges, with the wedge angle 57prime. The theoretical formulae have been derived, and the values found to agree closely with the experimental values.

The angles between the planes and the other surface angles are all temperature dependent as they are functions of the lattice parameters. In the tetragonal phase, the domain behaviour is found to be quite similar to that of BaTiO₃. The domain patterns are seen at times to have a memory indicating the presence of traps in the crystals for these domain walls.

The initial permeability of annealed, strain-free 77Ni-14Fe-5Cu-4Mo wt% alloy sheet is reduced by an order of magnitude by prolonged heat treatment at a temperature of 850°C in a vacuum of about 2 μtorr. The reduction in permeability is associated with the loss of copper through evaporation, leading to the formation of a concentration gradient of copper through the thickness of the sheet. The observed decrease in permeability is probably due to a combination of changes in alloy composition across the concentration gradient and strain introduced by variations in lattice parameter across the gradient and by differential thermal contraction on cooling from 850°C to room temperature.

The Faraday rotation of a ferrimagnetic calcium-bismuth-vanadium iron garnet sample has been measured in the wavelength range 1-5·5 μm. The results suggest that the vanadium and bismuth ions do not each occupy only one lattice site, but that these ions can occur in more than one valence state and can each partially occupy two lattice sites.

Details of the Fe L_II, L_III x-ray emission spectra of metallic iron and haematite have been investigated by an electron microprobe using operating voltages between 1·5 and 30 kV. It is shown that the measured ratio of the Fe L_II: L_III intensities is strongly dependent on operating voltage and that simple measurement of the L_II and L_III peak heights gives erroneous values for the ratio. More accurate values appear to be obtainable by integrating techniques which have been developed and are described. Differential self-absorption effects on the L_II and L_III bands of iron metal are extreme, requiring a correction factor of about 900% at 30 kV; none of the absorption correction formulae applied appear entirely satisfactory for making such corrections, but a true value for the ratio close to 0·5 for metallic iron is indicated at all operating voltages. For haematite (Fe₂O₃) the absorption corrected intensity ratio does not remain constant but increases progressively with increasing operating voltage from close to 0·5 near the critical excitation voltage to about 1·3 at 30 kV. This increase is thought to be associated with delocalization of electrons into molecular orbitals in the Fe-O bonds, since electrons of the same wave function may be involved in the production of the L_III band of iron and the K spectrum of oxygen.

The possibility of determining the Fe²⁺: Fe³⁺ ratio in iron compounds and minerals by measurements of the Fe L_II: L_III intensity ratio must be re-examined in the light of these results.

A method for the rapid measurement of initial impact velocity and contact time for collisions between two bodies is described. Results are given for normal impact between spheres and compared with the predictions of Hertz law of contact.

The phenomenon of generation of a third harmonic due to the interaction of a microwave with a plasma is investigated in the non-steady state. The results show that in the limit when time of observation is greater than the period of the wave but much less than the energy relaxation time of electrons, the third harmonic current density may be as much as 20 times the steady-state value for easily obtainable fields.

The axial voltage distribution of spark discharges in hydrogen, nitrogen and carbon dioxide is studied during the transient glow phase over a wide range of pressures. The results indicate that a glow mode develops in discharges which is initiated by the Townsend mechanism and is independent of external impedance. It is characterized by fixed values of cathode fall voltage and the ratio electric field to pressure in the positive column.

Using an iterative technique to obtain the exact solutions of the cubic Christoffel equation, the 21 elastic constants of copper sulphate pentahydrate have been determined at 25°C by the ultrasonic pulse echo method. The elastic constants, referred to the IRE recommended system of axes, are c₁₁=5·65, c₁₂=2·65, c₁₃=3·21, c₁₄=−0·33, c₁₅=−0·08, c₁₆=−0·39, c₂₂=4·33, c₂₃=3·47, c₂₄=−0·07, c₂₅=−0·21, c₂₆=0·02, c₃₃=5·69, c₃₄=−0·44, c₃₅=−0·21, c₃₆=−0·16, c₄₄=1·73, c₄₅=0·09, c₄₆=0·03, c₅₅=1·22, c₅₆=−0·26 and c₆₆=1·00 in units of 10¹⁰ N m⁻².

Some results are presented showing the existence of cylindrical `bubble-type' domains in cobalt crystals about 10³ Å thick. The diameter of the bubbles is small, of the order of 10³ Å, and they are observed by the technique of Lorentz electron microscopy. The magnetization configurations are interpreted in terms of the magnetic contrast. To investigate the nucleation method, structures have been examined in various conditions of applied magnetic field. The dimensions and stability of the cylinders are discussed in relation to recent theories which state the necessary conditions for bubble formation. Tentative theories are put forward regarding the nucleation of the cylinders.

A method of obtaining thin films of mercury sulphide is reported. Dispersion and absorption results are given using the method of analysis developed by Hall and Ferguson.