A variety of gas-solids flows can be observed in the pipeline of a pneumatic conveying system. These flows may be classified as one of three modes: i. suspension flow; ii. non-suspension moving-bed type flow; iii. non-suspension plug type flow. The modes of flow that a bulk material can achieve are dependent upon its particle and bulk properties as well as the pipeline conditions. This work describes the development of mathematical models for these modes of flow as well as experimental investigations to determine the validity of the models proposed. The modelling technique was based upon the solution of the conservation equations for inter-dispersed continua. Mathematical models for phenomena, such as the aerodynamic drag force between the conveying gas and particles, were added to the general mathematical model so that the flow of the gas-solids mixture could be simulated. This resulted in successful development of models for the prediction of suspension flow and non-suspension moving-bed type flow. In addition to providing data for validation of the mathematical models, the experimental programme produced a number of other observations. For example, it was found that the solids velocity in non-suspension moving-bed type flow could be determined non-intrusively by pressure measurements due to the variation in height of the moving-bed with time at a fixed location. More importantly, observation of plug type flow has led to the proposal of a mechanism to describe the development of the flow along a pipeline.

In this work, the suitability of amorphous semiconductors as hard field optical sensors for application in optical process tomography (OPT) has been established. Two amorphous semiconductors were selected for the study, these being amorphous arsenic triselenide (a-As 2 Se3) and hydrogenated amorphous silicon (a-Si:H). The a-As2Se3 device was a single layered structure of 60|Um thickness fabricated upon a 2mm thick cylindrical aluminium substrate. The a-Si:H device was a multi-layered structure of 27.1fim overall thickness fabricated upon a 4mm thick cylindrical aluminium substrate. 20mm 2 samples were cut from the cylinders, their surface being left free for a xerographic investigation. For a tomographic investigation, semitransparent gold (Au) contacts were sputtered onto the surface of the devices to produce single contacts or contact arrays. The study comprised of the two fields of xerography and tomography. The xerographic study comprised of the measurement of such parameters as charge acceptance, dark decay, residual potential, and photoinduced discharge. The research project has concurred with other workers in that the dark discharge mechanism in a-As 2 Se 3 proceeds via a xerographic depletion discharge process, and a Poole-Frenkel type emission in a-Si:H. The tomographic investigation involved the study of such parameters as detectivity, responsivity, steady state photocurrent, and photoinduced fatigue. Detectivity has found to be dependant upon the magnitude of applied electric field and level of incident irradiance. Irradiance in the order of 3.45mW/cm 2 to 9.57mW/cm 2 for a- fj f\ As 2 Se3 and 5.31mW/cm to 28.32mW/cm for a-Si:H was required in order to produce a clean and repeatable photogenerated current pulse over the range of electric fields specified (0.66xl05 V/cm to 1.66xl05 V/cm). The production of steady state current has found to be dependant upon the magnitude of electric field, the level of irradiance, and the illumination period. Irradiance of 319mW/cm 2 to 1.46W/cm2 with an illumination period of 520ns was required to produce steady state photocurrent in a-As2Se3 , and 693mW/cm2 to 2.62W/cm 2 with an illumination period of 880ns for a-Si:H. A linear relationship between electric field and responsivity has been observed in both materials over a range of irradiance of 3.45mW/cm 2 to 9.57mW/cm2 . Responsivity in the order of 87.86|LiAAV to 145.19|iA/W for a-As2Se 3 and 14.19(iAAV to 103.81)J,AAV for a-Si:H has been demonstrated. An investigation as to the effects of photoinduced fatigue in both a-As 2 Se3 and a-Si:H has been carried out by the application of pulsed visible light of various flash repetition rate (FRR) under a constant high electric field over a 30 minute illumination period. It has been shown that the rate of fatigue is dependant upon the material, time, electric field, light intensity, and FRR. A maximum operating speed of 20Hz has been determined for a-As 2Se 3 and lOOHz for a-Si:H. The maximum operating speed of 20Hz for a-As 2 Se 3 was deemed unsuitable for OPT application and the a-As 2 Se 3 material was eliminated from further tomographic investigation. Tomographic prototypes were employed to establish the a-Si:H devices ability to produce qualitative and quantitative data. The results of this investigation demonstrated that 1mm changes in water level and 0.5% changes in fluid colour could be accurately determined by the a-Si:H device at speeds required for OPT. The use of an a-Si:H device containing a Au contact array facilitated the imaging of the curvature of a pipeline and a phantom object contained within the pipeline. The results of the overall investigation have confirmed that the a-Si:H device is suitable for application as a hard field optical sensor for OPT.

Theoretical investigations of two-phase flows have not so far produced a useful model since the interdependence of the many variables has been difficult to predict. Progress towards such a model is dependent on accurate experimental work on two-phase flows. Particle velocity is an especially important property, but most available techniques either disturb the flow or are slow or inaccurate. The laser-Doppler velocity meter, LDV, was developed for measurements in single-phase flows, but it has been demonstrated by a few authors to be practical for particle velocity measurements in air-solid flows. The aim of the investigation was to find the range for which the LDV was suitable, and also to make useful measurements in a pipe conveying a dilute suspension of solids pneumatically. Air and solid velocity distributions across the diameter of a vertically upward flowing air-solid suspension in a 50 mm diameter pipe were made using an LDV. The solids conveyed were spherical glass balls, mean diameter 455 um, and sand, mean diameters 176 um and 366 um. The maximum ratio of solids to air mass flow rate was 2.5 and the maximum mean air velocity was 50 ms. Significant slip between the phases was found. Some of the correlations postulated between the particle velocity and other flow properties, such as the pressure drop, were investigated. Velocity measurements were also attempted with an LDV on plastic pellets, with effective diameters of 2 to 3 mm and varying degrees of success were achieved. The optical properties of the particles appears to be important when applying the laser-Doppler particle measuring technique to flows conveying particles of this size.

Small scale laboratory experiments were conducted to study the dynamic mor- phology and rheological behaviour of fluid-particle mixtures, such as snout-body architecture, levee formation, deposition and particle segregation effects. Debris flows consist of an agitated mixture of rock and sediment saturated with water. They are mobilized under the influence of gravity from hill slopes and channels and can reach long run-out distance and have extremely destructive power. Better understanding of the mechanisms that govern these flows is required to assess and mitigate the hazard of debris flows and similar geophysical flows. Debris flow models are required to accurately deal with evolving behaviours in space and time, to be able to predict flow height, velocity profiles and run-out distances and shapes. The evolution of laboratory debris flows, both dry glass beads and mixtures with water or glycerol, released from behind a lock gate to flow down an inclined flume, was observed through the channel side wall and captured with high speed video and PIV analysis to provide velocity profiles through out the flow depth. Pore pressure and the normal and shear stress at the base of the flow were also measured. Distinct regions were characterized by the non-fluctuating region and the in- termittent granular cloud surrounding the flows. The extent of these regions was shown to be related to flow properties. The separation of these two regions allowed the systematic definition of bulk flow characteristics such as characteristic height and flow front position. Laboratory flows showed variations in morphology and rheological characteristics under the influence of particle size, roughness element diameter, interstitial fluid viscosity and solid volume fraction. Mono-dispersed and poly-dispersed components mixed with liquids without fine sediments, reveal a head and body structure and an appearance similar to the classic anatomy of real debris flows. Unsaturated fronts were observed in mono-dispersed flows, suggesting that particle segregation is not the only mechanism. A numerical simulation of laboratory debris flows using the computer model RAMMS (RApid Mass Movements Simulation) was tested with dry laboratory flows, showing close similarity to calculated mean velocities.

Lobe and cleft patterns are frequently observed at the leading edge of gravity currents, including non-Boussinesq particle-laden currents such as powder snow avalanches. Despite the importance of the instability in driving air entrainment, little is known about its origin or the mechanisms behind its development. In this work we seek to gain a better understanding of these mechanisms from a laboratory scale model of powder snow avalanches using lightweight granular material. The instability mechanisms in these flows appear to be a combination of those found in both homogeneous Boussinesq gravity currents and unsuspended granular flows, with the size of the granular particles playing a central role in determining the wavelength of the lobe and cleft pattern. When scaled by particle diameter a relationship between Froude number and the wavelength of the lobe and cleft pattern is found, where the wavelength increases monotonically with Froude number. This relationship, in addition to Particle Image Velocimetry analysis, provides evidence for the existence of pairs of counter-rotating vortices at the leading edge of these currents, which play a key role in the development of the lobe and cleft pattern. The internal pressure of these flows is found to scale with the dynamics of the large vortex-like structure that is observed within the head of the current.

The mechanism of unconventional superconductivity is still mysterious, despite extensive studies. Muon spectroscopy (μSR) and small angle neutron scattering (SANS) studies of some unconventional superconductors were carried out. The NCS superconductors: Ru7B3 and α-BiPd, Heusler compounds: ZrNi2Ga and NbNi2Ga and the marginal superconductor ZrB12 were investigated. This thesis reports new phenomena observed in those systems and advances the knowledge of the ground state properties of unconventional superconductors. Polycrystalline samples and single crystals of the non-centrosymmetric superconductor (NCS) Ru7B3 were studied. Evidence for time-reversal symmetry (TRS) breaking in the superconducting state of Ru7B3 were found using μSR. So far, it is the third known NCS superconductor which breaks TRS. Potentially multigap behaviour was observed. The presence of two distinct superconducting gaps was suggested in polycrystalline Ru7B3 using μSR. Additionally, SANS and μSR measurements indicate unusual behaviour for the observed distorted hexagonal flux line lattice (FLL) in Ru7B3, yet the nature of this behaviour remains inconclusive. μSR studies of another NCS superconductor α-BiPd give evidence that two superconducting gaps are present in α-BiPd, either two s-wave gaps or s-wave and d-wave gaps exist in α-BiPd. Additionally, unusual temperature evolution of the upper critical field of α-BiPd was observed. Two superconducting compounds belonging to the Heusler family, ZrNi2Ga and NbNi2Ga, were investigated using μSR. Temperature dependence of the penetration depth of ZrNi2Ga suggests an s-wave, conventional behaviour in this superconductor. Evidence of a structural change in the morphology of the FLL was observed: a square FLL potentially exists at low temperatures and a hexagonal FLL is present closer to Tc. Additionally, μSR experiments suggest that TRS is preserved in both measured Heusler compounds. SANS studies of a marginal, low-k superconductor ZrB12 were carried out. SANS studies were performed in a range of temperature and magnetic field so to access different parts of the superconducting phase diagram of ZrB12. A square vortex lattice was found in all measured temperatures and fields.

Chapter 1 is an introduction to the hierarchy in block copolymer self-assembly from a bottom up approach. To begin with, the control at each hierarchical level is introduced and the influences are reviewed. This chapter is concluded by reviewing some techniques to understand self-assembled block copolymers and the applications from bottom up self-assembly of block copolymers is presented. Chapter 2 reports the synthesis of a range of amine homopolymers and copolymers by RAFT polymerisation. These amino polymers are then explored further as lubricant additives and their solution self-assembly in non-polar media is explored. Chapter 3 investigates the performance of the amino polymers from Chapter 2 as ashless detergents in lubricant formulations using a range of industrial testing methods. Chapter 4 reports the synthesis and self-assembly of a range of pH responsive P(DMAEMA-co-DEAEMA)-b-PDMAEMA diblock copolymers, where the composition of the P(DMAEMA-co-DEAEMA) block was varied by copolymerisation, in aqueous solution studied by a range of scattering methods and microscopy. Chapter 5 reports the initial copolymer blending method protocol, here two P(DMAEMA-co-DEAEMA)-b-PDMAEMA diblock copolymers with varying P(DMAEMA-co-DEAEMA) block compositions are blended together. The structures formed on the nanoscale are analysed and further compared to the structures formed by a pure P(DMAEMA-co-DEAEMA)-b-PDMAEMA diblock copolymer with the same composition as the blend. Chapter 6 reports on the extension of the copolymer blending method protocol, here a range of neutral polymers and different self-assembly pathways are explored and the structures formed are analysed by detailed light scattering techniques and cryo-TEM.

This research work presents numerical simulations of supercontinuum (SC) generation in optical waveguides based on Ge11.5As24Se64.5 chalcogenide (ChG) material. Rigorous numerical simulations were performed using finite-element and split-step Fourier methods in order to optimize the waveguides for wideband SC generation. Through dispersion engineering and by varying dimensions of the 1.8-cm-long ChG nanowires, we have investigated dispersion curves for a number of nanowire geometries and identified a promising one which can be used for generating a SC with 1300 nm bandwidth pumped at 1550 nm with a low peak power of 25 W. It was observed through successive inclusion of higher-order dispersion coefficients during SC simulations that there is a possibility of obtaining spurious results if the adequate number of dispersion coefficients is not considered. We then investigate MIR SC in dispersion-tailored, air-clad, ChG channel waveguide employing either Ge11.5As24S64.5 or MgF2 glass and ChG rib waveguide employing MgF2 glass for their lower claddings. We study the effect of waveguide parameters on the bandwidth of the SC at the output of 1-cm-long waveguides. Our results show that output can vary over a wide range depending on their design and the pump wavelength employed. At the pump wavelength of 2 μm the SC never extended beyond 4.5 μm for any of our designs. However, SC could be extended to beyond 5 μmfor a pump wavelength of 3.1 μm. A broadband SC spanning from 2 to 6 μm and extending over 1.5 octave could be generated with a moderate peak power of 500 W at a pump wavelength of 3.1 μm using an air-clad, all-ChG, channel waveguide. We show that SC can be extended even further covering the wavelength ranges 1.8-7.7 μm and 1.8-8 μm (> 2 octaves) when MgF2 glass is used for the lower claddings of ChG channel waveguide and rib waveguide, respectively. By employing the same pump source, we show that SC spectra can cover a wavelength range of 1.8-11 μm (> 2.5 octaves) in a channel waveguide and 1.8-10 μm in a rib waveguide employing MgF2 glass for their lower claddings with a moderate peak power of 3 kW. Finally we present microstrucured fibre based design made with same glass to generate SC spectra in the MIR region. Numerical simulations show that such a 1-cm-long fibre can produce a spectrum extending from 1.3 μm to beyond 11 μm (> 3 octaves) with the same pump and peak power applied before. We consider three fibre structures with microstrucured air-holes in their cladding and find their optimum designs through dispersion engineering. Among these, equiangular-spiral microstrucured fibre is found to be the most promising candidate for generating ultrawide SC in the MIR region.

Experiments probing fundamental laser-plasma interaction physics have demonstrated some interesting and unexpected effects. Scattering from electron plasma waves with frequencies below the electron plasma frequency (called electron acoustic waves) has been observed in addition to conventional parametric scattering. Reflectivities observed in NIF early light experiments on long scale-length plasmas differed considerably from those predicted by existing fluid models. These effects are essentially kinetic in nature. The low frequency modes are supported by the trapping of electrons at low phase velocities and the saturation of instabilities at the intensities provided by the next generation of laser systems, such as NIF, is associated with the trapping of electrons. Numerical simulation of laser plasma interaction, therefore, benefits from an accurate treatment of the particle kinetics, in particular the evolution of the particle distribution functions. The direct solution of the Vlasov equation allows a high resolution, noise-free, representation particle distribution functions. Recent advancements in Vlasov codes, which draw a considerable expertise in the numerical solution of hydrodynamic codes, make such an approach to the simulation of laser plasma interaction viable. Here the development of a one dimensional electromagnetic Vlasov code is outlined. Thecode is applied to realistic laser and plasma parameters characteristic of single hot-spot experiments. Results are in qualitative agreement with experiments displaying both stimulated Raman and stimulated electron acoustic scattering [N. J. Sircombe, T. D. Arber and R. O. Dendy Kinetic effects in Laser-Plasma coupling: Vlasov theory and computations, Submitted to Plasma Physics and Controlled Fusion]. The amplitude of simulated electron acoustic waves is greater than that observed experimentally, and is accompanied by a higher phase velocity. These minor differences may be attributed to the limitations of a one-dimensional collisionless model. Furthermore, the interaction of a Langmuir wave with a density hole is investigated and shown to result in the acceleration of a small population of electrons [N. J. Sircombe, T. D. Arber and R. O. Dendy, Accelerated electron populations formed by Langmuir wave-caviton interactions, Phys. Plasmas, 12, 012303 (2005)]. This acceleration results from wave breaking and is dependent on the parameters of the background density profiles. In addition, pre-acceleration of electrons in supernova remnant shocks is considered as a kinetic problem [N. J. Sircombe, M. E. Dieckmann, P. K. Shukla and T. D. Arber, Stabilisation of BGK modes by relativistic effects, Astronomy and Astrophysics, In Press], [M. E. Dieckmann, N. J. Sircombe, M. Parviainen, P. K. Shukla and R. O. Dendy, Phase speed of electrostatic waves: The critical parameter for efficient electron surfing acceleration, Plasma Phys. Control.

The thesis is introduced by a discussion of the technological importance of developing superconducting materials having low losses at power frequencies. The following two chapters are devoted to a presentation of the reversible magnetic properties of superconductors using the Ginzburg-Landau equations, and a discussion of the irreversible behaviour of real materials in terms of flux pinning and the critical state concept. Following this background discussion, previous work on a.c. losses is reviewed. Particular emphasis is given to the losses below Hc1 and it is shown that, while losses above Hcl can be interpreted using the critical state model, below Hc1 they are only partially understood. Modifications to an electronic wattmeter which enable accurate measurements to be made at 10- 4 WHz- 1 m-2 are discussed. An improved vibrating sample magnetometer, is described, together with the theory and design of an a.c , permeability apparatus used to obtain critical current values. The preparation of a range of niobium samples is described. The techniques employed include annealing, spark-machining, mechanical, chemical and electro-polishing, neutron irradiation and ion-implantation. The magnetization curves, surface profiles and other data for the treated specimens are presented. Measurements of the low frequency losses in these samples at 4.2 K are described. It is shown that in rough samples hysteretic losses arise below Hcl·from the penetration of surface asperities and that the results are in good agreement with an expression, similar to Buchhold’s (1963), which is derived, namely : L =4 μOHm2 f(Hm/HCL) KD where Hm is the peak field, HCL the lower critical field, HCl or HC, D the centre line average height of the surface profile and K a ‘hysteresis factor’. The losses in cold-worked and annealed niobium samples and lead, tin and indium specimens with values for D between 0.33 and 20 μm are found to fit (within a factor of two ) the expression : L = ½ μ0Hm2 (Hm/HCL)2D Different results are observed in smooth samples and explanations for these are suggested. The losses in samples previously penetrated by a large a.c. field are found to be increased by a factor of up to thirty. This is related to trapped flux and ppossible loss mechanisms arc discussed. The losses above Hcl are shown to be reduced in damaged samples. Neutron irradiation induces heavy bulk damage but has little effect on the surface. Mechanical treatment produces large surface currents and is most effective reducing the loss, the dissipation at 50 Hz being below 0.1 W m-2 at fields up to 85 kA(RMS)m- 1 (1500 0e - Peak) in a mechanically polished, annealed polycrystalline sample. Implantation of niobium ions to a depth of 10 nm does not alter the losses. Measurements of the critical currents and flux profiles in two annealed samples are presented and it is shown that a critical state exists within them. The losses above Hcl in both irreversible and annealed samples fit the Ullmaier (1966) expression L = ⅔ μO(Hm - ∆H/2)3/JC, but it is found that equating ∆H/2 with HCl gives critical currents an order of magnitude smaller than those obtained in other measurements. It is shown that ∆H/2 is not constant and that the shielding currents are negligible except near HCl.