The development of turbulent boundary layers immediately downstream of various strip-like transition devices is studied experimentally. Methods of predicting the lengths of boundary layer differing from undisturbed zero-pressure gradient state, and the effective increase in momentum thickness, due to these devices, are given. A possible model is suggested to describe the flow about the transition device, the breakdown into turbulence and the structure and development of the ensuing turbulent eddies. The turbulent energy balance is described and a method of using it to predict the development of the boundary layer discussed.

An investigation of the growth and properties of thin films of indium antimonide for use in active devices was made, together with field effect studies on InSb inert insulator solid-solid interfaces. The films were prepared by the protected re-crystallization method, namely vacuum evaporation and subsequent annealing of the resulting insulator-semiconductor-insulator sandwiches. The high annealing temperatures permissible without danger of re-evaporation of the more volatile component gave films with high mobility of approximately 103-104 cm2/V-s and relatively low carrier concentrations of approximately 1016-1017/cm3. The good bonding of the insulator on to an inherently clean semi-conductor surface resulted in a low fast surface state density of around 5 x 1011/cm2, which would permit the use of these films for field effect transistor applications. Thin films are very well suited for field effect investigations on low band gap semiconductors. Surface Hall effect and surface conductivity against surface potential gives the trapped and mobile charge density with the relevant Hall and field effect mobility directly. A model of fast surface states is proposed, and the behaviour of the surface conductivity mobility is compared with theoretical results for a degenerate, non-parabolic spherical band structure.

Tho research described in this thesis investigates the charge separation which occurs when liquids are broken up by the action of mechanical forces. It is shown that the problem of the charge separation is often the problem of the separation of charge in a single breaking jet. A series of experiments is described in which jets were produced under closely controlled conditions and by a variety of methods from volumes of liquids, usually aqueous solutions. The jets were produced by, the splashing of drops on a surface, the pulling of a capillary vertically from a horizontal liquid surface, and by the breaking of small isolated drops. The charges on the drops resulting from the breaking of these jets were measured as a function of the concentration of solutes in the liquids, and also as a function of certain geometrical factors. The results showed that great care must be taken to eliminate any charges which are separated by induction in the fields of contact potentials which are often present. It is shown that the charge separation can be explained in terms of the shearing of the electrical double layer which is found on the surface of the liquids, either during the deformation of the liquid to form the jet, or during the breaking of the jet into drops.A-theoretical expression for the charge separated by the later method is derived. This gives reasonable agreement with the experimental results provided that allowance is made for inaccuracies introduced into the theory by certain simplifying assumptions. It is also shown that charge separation in aqueous jets is limited by conduction for jets of radius greater than about one micron.

Three topics relating to the theory of aerodynamic sound are considered. The first is a mathematical study of the similarity between the behaviour of surface gravity waves on a shallow layer of water and sound waves in a compressible gas. It is concluded that various qualitative features of the two wave fields are the same, but that a complete quantitative similarity is not possible. In the second study, the current formulation of propeller noise theory is considered in the light of the more general aerodynamic sound theory. This reveals that some important sound sources have been overlooked in propeller theory, and their effect is discussed. Some extensions are also made to existing results concerning the frequency spectrum of a sound source in circular motion. The third chapter shows how the use of generalised functions leads to a greater understanding of the fundamental theory of aerodynamic sound. The classical results of this theory are more simply derived using the new methods. Also new results are obtained concerning the sound fields generated by rapidly moving sources, and these provide a better description of the fields than those results employed hitherto. In particular, the sound fields generated by rapidly convected turbulence and rigid surfaces moving at high speeds are discussed.

The general theory of magnon interactions in antiferromagnetic insulators is examined from a new view point, and the nature of the antiferromagnetic ground state is clarified. It is shown that the Anderson canonical transformation to free spin wave operators is non-unitary, with the magnon state vectors having large non-physical projections, and a projection technique which restores unitarity at low temperatures is developed. The non-physical projection of the state vectors leads to a large kinematic interaction, even at zero temperature. A group theoretic approach to the problem of calculating magnon self energies in complicated magnetic structures is developed, and this technique is combined with the Dyson-Maleey boson mapping in an effort to elucidate, the mechanism of the Morin phase transition in haematite. The free magnon dispersions are calculated for both magnetic phases of haematite, and it is shown that the optic modes can have unusually low energy. The equations for self-consistent renormalisation of the magnon energies are given for the low temperature phase, and order of magnitude estimates for the energy shifts are calculated at the centre of the Brillouin zone. The numerous experimental 'anomalies' are considered and it is proposed that the optic modes are unstable against magnon interaction.

Well known experimental techniques have been applied to molybdenum single crystals to study the orientation dependence of the work hardening. These techniques comprised slip line observations by white light microscopy, the use of carbon replicas of slip lines in a 100 key electron microscope, and the obtaining of thin foils by a two stage electro—polishing technique for the observation of dislocation structures. At and below 400°K {121} slip in the twinning sense is at least as prominent as {11O} slip in both <110> and <100> crystals. Observations of a predominently screw dislocation structure amid an array of debris edge dipoles in thin foils were used to explain slip line observations and the greater importance of temperature on the yielding and flow of <110> crystals as compared to <100> crystals. The orientation dependence of the yield and flow stress has been qualitatively explained in terms of the magnitude of the effective resolved shear stress required to constrict 1/8 a <110) partial dislocations on {110} planes.

In this thesis, some problems concerned with the large deflections of thin solid wings are investigated. First, an infinite strip of double wedge section subjected to spanwise bending, torsion and thermagradients is considered. Lateral deflections, stress resultants and stability boundaries are evaluated by a semi-exact process, in which numerical techniques are employed to avoid complicated integrations. The results agree well with those obtained using approximate polynomial expansions in the "small" large deflection regime, and also show close agreement with the results available for similar sections, like the biconvex wing. The second problem considered is that of the single wedge, infinite span wing under a pure spanwise bending moment, the analysis following directly from that of the double wedge. Difficulties arising from the unsymmetric section are overcome by a suitable definition of axes. To cater for sections not amenable to the analytic approach, a finite difference method is evolved in part three. The two previous problems are considered as examples, and the errors are found to be less than 1%. Finally, the practical wing problem of a cantilevered plate of finite aspect ratio, subjected to pressure and thermal loadings, is investigated. For simplicity, the example chosen is a square plate of constant thickness. The technique devised is shown to operate over the full region bounded by the two asymptotic solutions given by "small deflection" and "inextensional" theories. Based on the asymptotic formula, the accuracy of the method is estimated to be better than 1.5%. However, the governing differential equations of Von Karman are only valid under certain assumptions which do not hold for very large deflections, such as the approximate expression for curvature. It is shown that for the largest load considered, the error introduced in the curvature expression is 0(5%). Hence, above this value of loading parameter, the governing equations are seriously in error.

Detailed investigations of thermal behaviour of potassium acetate and some preliminary studies on rubidium and caesium acetates have been described. Properties have been studied by X-ray diffraction and other physical methods. Three crystallo-graphically distinct but closely related polymorphs of potassium acetate have been found: Form I : From 155°C upwards ... orthorhombic Form II : From 155°C to ~ 75°C ... monoclinic Form III : From ~ N 75°C to room temperature ... monoclinic with a multiple cell. The two solid-solid phase transitions have been studied. X-ray diffraction photographs of single crystal and co-operative pretransition-increase in dielectric and thermal properties have shown that the transition I → II is of "displacive" (martensitic) type whose structural basis has been established. The transition II → III which did not show up in dielectric and thermal analysis measurements has been structurally characterised by the appearance of additional layer lines, resembling superlattice reflexions in an ordered structure. Intensity of these reflexions has been used to study the hysteresis of this transition. Thermal expansion of the lattice parameters has been studied. Very high values of the thermal expansion along the a-axis and contraction along b and c axes have been measured. In the dielectric study, the relative permittivity and loss-factor rapidly increased at higher temperature, starting from about the Tc = 155°C of II → I transition. This has been attributed to thermal activation of the cations from lattice site to the void in the lattice. The basic structures of form I and II have been established as formed by ionic double layers parallel to {100} of potassium ions co-ordinated by six oxygen atoms from five different acetate ions, - four from four different acetate ions of the same half of the layers as the potassium ion and two from the same acetate ion of the other half of the layer. Only probable models of form III have been discussed. For comparison, some crystallographic and dielectric work have also been done with rubidium and caesium acetates.

The conductance and space-charge polarization capacity of single crystals of thallous chloride have been measured as a function of frequency between 350°K. and temperatures approaching the melting point. Correlation with available diffusion data indicates that Schottky disorder is present with the chlorine ion vacancy as the predominant charge carrier. Analysis of the capacitance results using a modified space-charge theory gives new values of the enthalpies of formation and motion of the vacancies present. Thus ∆Hs = 1.36 eV. ; ∆Hm(cv) = 0.40 eV. ; ∆Hm(av) = 0.09 eV. The defect concentration at the melting point is ~0.08 atomic %. Similar measurements on thallous bromide show that Schottky defects probably exist in this material with the bromine ion vacancy the more mobile species. The characteristic defect parameters for the bromide closely resemble those for the chloride.

The work reported here is largely an experimental description of some of the phenomena associated with the formation of an unstabilised linear pinched discharge. The work was stimulated by a theoretical account of the "inverse skin effect" in a rigid conductor, published by Dr. M.G. Haines(11). The aim of the work is to con— pare the derived current density distributions with the above theory and use measurements of the electrical conductivity of the plasma to make estimates of the plasma temperature. It is possible, with the technique employed here, to make only restricted measurements of the electrical conductivity owing to a lack of knowledge of the notional E.M.F. term in ohnis law. If the distribution of radial collapse velocity through the plasma column is known, it is possible to calculate the electric field in the moving frame of coordinates from the magnetic field data obtained. Measurements of the distribution of electric field in the laboratory frame of coordinates, taking into account the weakening of the field by the "skin effect", have been achieved, but not without difficulty. The reason for this is that in the early stages of the discharge, before any electric field appears on the axis of the discharge.tube, the total non—conservative field component is linearly related to the conservative field, but the straight line graph does not pass through the origin. This is interpreted in terms of either a cathode fall of potential which appears to be independent of current density, or to the presence of electric field on the axis even at the earliest times. A further aim of this work is to give interpretation to various gross discharge characteristics such as the current and voltage waveforms. Toward this end, as wide a range of discharge conditions as possible are examined. The chief conclusion of this section is that the inductance of the discharge column is not always maximum when most narrowly constricted. The reason for this is that, at the time of formation of the pinch, a significant fraction of the total discharge current flows in fresh plasma momentarily generated at the walls of the tube. At high initial filling pressures, the distribution of current density shows an anomalous behaviour. Initially the current density builds up at the wall of the tube in the usual manner — until the point is reached where the current peak just separates from the wall of the tube. At this point the current density in the shell decreases for a time, whilst a strong axial core of current flows. This behaviour can not be explained in terms of shock wave heating of the central regions. Explanation in terms of magnetisation of the plasma is attempted. A further phenomena requiring explanation is the flow of reversed current density in certain discharge zones after the pinch. This is not the straight-forward inverse skin effect described by Haines, but undoubtedly it does make a contribution in the correct direction. The reason for this is that the applied electric field is still positive in these regions of the discharge. Since motional E.M.F.'s after the pinch tend to enhance the growth of +ve,current density, this reversed current can only be explained in terms of pressure or temperature gradients. Then the product wt>>1, the charged particles are strongly coupled to the self-magnetic field. Under these conditions, gradients in the radial direction can cause axial current flow in the correct direction. The reason for invoking the more complex drift in a magnetic field, rather than a simple axial pressure or temperature gradient is that there is experimental evidence for the required radial gradients, whilst evidence for axial gradients, at these times, is non-existent.