The existing information on the behaviour of methyl ethers and acetates of pyranosides under electron impact is reviewed and certain fragmentation schemes have been reappraised. The mass spectra of several derivatives of fusicoccin, a naturally occurring glycoside, are reported. A part structure, which depends on mass spectral evidence,is deduced for the molecule. The principles laid down in the review are used to rationalise the fragmentations of the acetates of some partially methylated glucosides. Particular reference is made to the effect of the position of attachment of methoxy groups to the sugar ring. The methods of preparation, and the mass spectral decompositions of tetra-acetyl glycopyranosides are discussed. Eight fragmentation sequences are found in the mass spectra of these compounds, and the relative abundances of certain series of ions are related to the structure of the aglycone. The number of aglycone ions in the mass spectra of substituted phenolic glycosides is correlated with the Hammett substituent constants. Deuterium labelling and high resolution mass spectrometry have been used to elucidate the fragmentation mechanisms for 496-0-benzylidene derivatives of glucosides. Most ions are shown to be formed by rearrangement processes. The behaviour of allylic acetates under electron impact conditions is described, with emphasis on the mechanism of loss of acetic acid. For cinnamyl acetates this is accompanied by cyclisation to an indene ion. Pyrolysis of cinnamyl acetate gives an analogous reaction and indene is formed as the major product.

Glare is defined, and a brief discussion of ocular anatomy is given, together with some report and discussion of previous work in the field of disability glare. An investigation of the practical aspects of white and yellow glare in situations encountered while driving at night is described, and the two colours are shown to have very similar glare effects under certain conditions. A discussion of 'equality' of white and yellow stimuli is included to show that redefinition of these conditions can affect the conclusions. A binocular comparison instrument and ancillary equipment are described. Behaviour of sensitivity during recovery from short exposures to glare was investigated with this instrument for white and yellow glare sources and subsequently for purer spectral colours, with the aim of isolating chromatic effects. Measurements were made of the 'equivalent veils' from glare sources of different colours and an hypothesis is propounded concerning the intraocular location of scatter of different wavelengths. This agrees well with the earlier work of Le Grand and is here substantiated by separate methods. That different predominant wavelengths should be scattered by different elements within the eye can explain some inconsistencies in past reports. Definitions of stimuli which are adequate on the assumption of white light scatter in the eye must be expanded to include information about spectral composition as well as luminance. The concept of 'equivalent veiling luminance' as a method of specifying glare is found not to be valid for rapidly changing situations and this is used as an argument in support of neural inhibition as a contributary cause of glare, together with entoptic scattered light. This is quantitatively tested by an experiment using the Stiles-Crawford effect to reduce the receptoral response from a glare pencil admitted through the limb of a dilated pupil. This showed that neural inhibition may be small, but is not negligible over small angles.

Hollow core fibre (HCF) is a type of novel optical fibre which has lower refractive index in the hollow core than in the cladding. Total internal reflection (TIR) cannot explain the confinement of light to the core of a HCF. According to the confinement mechanism, the HCFs can be generally divided into hollow core photonic bandgap fibres and hollow core leaky mode fibres depending on their optical properties of the cladding structure. Hollow core negative curvature fibre (HC-NCF) is a kind of hollow core leaky mode fibre, which is defined by the negative curvature of the core boundary. This thesis presents my study of HC-NCFs over the last two years. My research has focused on developing low loss silica HC-NCFs and exploring the attenuation limit of HC-NCFs. Fifty different HC-NCFs were fabricated, which covered the spectral range from 800 nm to 4.5 μm. Minimum attenuations of 24.4 dB/km and 85 dB/km were measured at around 2400 nm wavelength and 4000 nm respectively, which are the best achieved in HCFs at these wavelengths to the best of my knowledge. The limits of HC-NCF attenuation were revealed by analysing the data from HC-NCFs scaled for minimum attenuation in different spectral regions. Other properties of HC-NCFs, including bending loss and dispersion, were also studied experimentally. By using white light interferometery, a low group velocity dispersion (GVD) was found in HC-NCFs, which agrees well with simulations. The bending loss of HC-NCFs was preliminarily studied by measuring the transmission spectra under different bending conditions. Significant bending loss was found when the bending radius was less than 15 cm. Numerical simulations were performed using COMSOL software to study the properties of HC-NCFs. In the simulations, it was found that the capillary thickness is the most important factor determining the attenuation of HC-NCFs. These results were used to explain the experimental results. This thesis is comprised of seven chapters. Chapter 1 and 2 supply background material that helps to understand the light guidance mechanism of HC-NCFs. My original work is presented in Chapter 3, 4, 5, and 6. Chapter 7 includes a summary and suggestions for future work.

The development of photonic crystal fibre from conventional optical fibre follows a trend in the development of materials, to create composites and structured materials on smaller and smaller scales. In fact the great success of photonic crystal fibre is largely due to the ability to structure it on scales comparable to the wavelength of light. It is this micron size structure that allows the creation of an (out of plane) optical bandgap in silica and allows hollow core fibre to guide light in an air core freeing the guided mode from the properties of bulk silica. This thesis focuses on the propagation and compression of high peak power optical solitons in hollow core fibre. As the Kerr nonlinear response of air is approximately a thousand times less than that of silica, the air core of hollow core fibre can support much higher peak powers than conventional optical fibre without the manifestation of nonlinear effects, making it ideal for the delivery of high peak power laser pulses. Coupled with this, hollow core fibre has a large region of anomalous dispersion in its transmission window allowing optical pulses to be transmitted as temporal solitons freeing them from the effects of dispersion. The author started his Ph.D. in 2006, three years after the first demonstration of soliton propagation in hollow core fibre and as the first demonstrations of soliton compression in hollow core fibre were being undertaken. Work by the author to build upon these early demonstrations is presented in this thesis in the following manner: Chapters 1, 2 and 3 are theory chapters. Chapter 1 explains the background waveguide theory and theory of nonlinear optics that is used throughout the thesis. Chapter 2 details the properties of photonic crystal fibres focusing on hollow core fibre. Chapter 3 details recent papers relevant to the propagation and compression of solitons in hollow core fibre. Chapters 4, 5 and 6 are experimental chapters reporting work undertaken by the author. Chapter 4 focuses on modifying the nonlinearity of hollow core fibre and measuring the dispersion of hollow core fibre accurately. Chapter 5 focuses on the compression of chirped and unchirped picosecond pulses in dispersion decreasing hollow core fibre tapers. Chapter 6 reports the compression in hollow core fibre of femtosecond pulses centred at 540nm wavelength through soliton effect compression.

Three methods of flame spectroscopic analysis have been investigated - atomic-fluorescence, atomic-absorption and flame-emission spectroscopy. The emphasis in the first part of this work has been on the development of the relatively new technique of atomic-fluorescence spectroscopy. This technique was found to be very sensitive and virtually free from all cation and anion interferences. The optimisation of the various operating parameters are described, and general methods for the measurement of atomic-fluorescence using commercially available sources are discussed. The preparation and operation of microwave excited electrodeless discharge tubes as atomic spectral line sources, for use in atomic-fluorescence and atomic-absorption spectroscopy, are fully described. These sources were found to be superior to commercially available atomic spectral line sources. Using electrodeless discharge tube sources, the fluorescence and absorption characteristics of 5 elements, in a variety of flames, have been determined. Sensitive methods of analysis for each of the 5 elements have also been developed. The second part of this work deals with new developments in flame-emission spectroscopy. The analysis of sulphur and phosphorus using the emission from unstable molecular species in a new type of 'cool flame' is described. The spectroscopic properties of the high temperature, low-background nitrous oxide-hydrogen flame are also investigated.

Part I An investigation of some of the crystal structures occurring in the Bi203-Nb205 binary system was undertaken by X-ray methods. The structure of triclinic BiNb04 was solved from visually estimated single-crystal diffraction data. The unit cell has a = 7.611, b = 5.536, c = 7.919 Å, ∝ = 89°54', β = 77°28', γ =87°9', the space group is Pī. The structure, which consists of sheets of niobium-oxygen octahedra linked through bismuth atoms, is of an entirely new type. The mode of joining of the octahedra in the sheets found in this compound is novel. Least-squares refinement with isotropic temperature factors converged to R = 0.122.

This thesis develops a theory of holographic recording valid for thin recording materials. Chapter I introduces holography and gives a historical survey of previous work. It serves to delineate the assumptions used in the remainder of the thesis and to define some terminology used in holography. Chapter II treats the linear theory of hologram recording. The Ta-E curve for the film is introduced and a method for measurement is described. The concept of modulation transfer function (MTF) for holograms is discussed and a method of measurement of MTF is given. Finally, the combined effect of MTF and film size is examined. It is shown that the holographic imaging technique is space-variant; however, it is also shown that under some conditions the technique is space-invariant. Chapters III and IV deal with nonlinear effects introduced by the film. The effect is examined using a Fourier transform technique and a power series expansion. The power series shows that the modified transmittance is the product of the linear transmittance and a polynomial in B, the normalized deviation from the mean of the light scattered by the object to the hologram. The results of the power series technique are used to examine the effect of the modified transmittance on the reconstructed image. A theory which predicts the magnitude of the effect is given along with experimental results which tend to confirm the theory. In Chapter IV, another.theory, which predicts both the magnitude and the distribution of the nonlinear effect, is given along with further experimental results. Chapter V examines the effect that the film-grains have on the holographic technique. A formula for the signal-to-film-grain-noise ratio is derived in terms of the intensity transmittance spectrum of the random transmittance fluctuations caused by the recording-film grains.

Absorption spectra of the radicals C2, C3 and CH have been studied quantitatively in acetylene-oxygen flames burning at atmospheric pressure. The effective light path through the flame was increased by using a new design of multiple reflection system. This focuses the light into the flame, gives better spatial resolution and is less affected by schlieren-type deflections of the light beam 0 than in the normal parallel-beam system. There is little evidence for thermal disequilibrium in the luminous mantle of fuel-rich flames. Concentrations of C2 and C3, about 2x1014 and 1x1015 molecules cm-3, are consistent with values calculated for equilibrium with solid carbon, but the concentrations of C2 and CH are much higher in the reaction zone. Measurements of C2 rotational and vibrational temperature support a chemiluminescent mechanism of formation. It is concluded that C3 is formed from evaporation of incipient carbon particles but the possibility of C2 radicals being important in the formation of carbon 'nuclei' cannot be ruled out.

The scientific applications of nuclear research emulsions are reviewed with special reference to their application for recording electron images generated in electronographic image tubes. The effect of electron irradiation of these emulsions is considered from a theoretical point of view. Apparatus is described which has been developed to investigate this effect quantitatively and results are presented for G5, L4 and XM emulsions exposed to electrons in the energy range up to 35kV. The agreement of these results with the theoretical considerations is discussed. The question of the granularity of nuclear research emulsions exposed to electrons is considered from a theoretical point of view. Experiments are described to evaluate the granularity of nuclear research emulsions and the agreement of the experimental results with theory is discussed. Prom the results obtained in this investigation, a figure is derived for the equivalent quantum efficiency of a single—stage electronographic device employing direct recording of 25kV electrons on G5 emulsion. Some aspects of the comparative performance of electronographic tubes employing direct and indirect recording of electron images are discussed. The benefits of operating eleotronographic image tube systems at higher accelerating potentials are indicated.

Nanoscale photonic functionalities based on light-induced structural transitions in nanoparticles have been investigated, and it hag been experimentally shown that nanoparticles can act as both low power nanoscale optical switches and as resonator-less optical memory elements. A system for in-situ growth and characterization of gallium nanoparticles, which combined technologies including atomic-beam deposition, ultra-high vacuum, cryogenics, and sophisticated fibre instrumentation including diode and ultra-fast laaers, has been developed. Optica,l switching has been observed in a gallium nanoparticle film on the end of a single mode optical fibre simultaneously in rejection and transmission, and under different regimes of excitation, for the first time. Measurements of the sub-microsecond dynamics of such light-by-light control allowed the first study of the fast kinetics of solid-solid and solid-hquid structural transformations in gallium nanoparticles to be performed. Single gallium nanoparticles have been grown from an atomic beam in the nanoaperture at the tip of a tapered optical fibre for the first time. Reversible light-induced reflectivity changes associated with a sequence of transformations between different structural forms (both solid-solid and solid-liquid) stimulated by optical excitation at nanowatt power levels, have been observed in such particles for the first time. The complex temperature hysteresis of the narioparticle's nonlinear response has been observed and it has been discovered that the extent of overcooling can be controlled by varying the optical pumping regime. The first demonstration of nanoscale all-optical resonator-less memory functionality baaed on phase transformations has been performed using a film of gallium nanoparticles. It has been shown that single 1 μs optical pulses of a few mW peak power can be used to 'write' information to the memory by converting the particles from a lower energy phase (logic state 0) to a higher energy phase (logic state 1). A high contrast method for 'reading' the state of the particle memory, based on measurements of the reflectivity change induced by a modulated pump beam, has been developed. Both volatile and non-volatile modes of memory operation have been demonstrated. For the 6rst time, an optical memory element based on a single particle has been demonstrated. It has been shown that an 80 nm gallium nanoparticle can act as a fourlevel nanoscale optical memory. Information is encoded on the particle by switching it between phases using single optical pulses with energies as low as 1.5 pJ, and by varying the pulse energy different states can be directly accessed from both ground and intermediate states, A closely packed array of such particles could provide a storage density of about 0.2Tb/in^. The experimental work has been underpinned by the development of appropriate qualitative physical models of the processes involved, so as to describe the relationships between excitation controlled phase coexistence in nanoparticles, their optical properties and the demonstrated functionalities.