EISCAT Svalbard Radar data, obtained during the IPY-ESR 2007 campaign at a period of deep solar minimum, has been examined to study ionospheric upflow events with fluxes exceeding 10¹³m⁻²s⁻¹. Ion upflow newly categorized, classifies the upflow events into low, medium and high flux upflows, and the incidence and seasonal distribution of these different classes are discussed. Over 300,000 field-aligned profiles have been considered and analysed. It is observed that, while high upflow fluxes are comparatively rare, low flux upflow events are a frequent phenomenon. Analysis of the ESR data shows that the occurrence frequency of the upward flux maximizes around local noon for all levels of upflow, with occurrence peaks of 31%, 16% and 2% being observed for low, medium and high upflow fluxes respectively during geomagnetically disturbed periods. Analysis of the seasonal distribution reveals that while high-flux upflow has its peak around local noon in the summer, with its occurrence being driven predominantly by high geomagnetic disturbance, the occurrence of low-flux upflow is broadly distributed across all seasons, geomagnetic activity conditions and times of day. The ambipolar electric force drives upflows about seven times more frequently as the Joule heating mechanism. This study finds that, though rare, Poynting flux leads to substantial upflow flux of 4.67×1013 m−2 s−1 even in the absence of precipitating flux. Upflow occurrence was found to correlate better with solar wind density than speed, and fits better with number density and dynamic pressure below about 20 cm−3 and 5 nPa respectively. Moreover, it was found that occurrence maximizes when the IMF clock angle is southward however, analysis showed a preference for duskward asymmetry of the cusp in comparison to dawnward’s. Analysis shows that reconnection is not compulsory for upflow flux, but only enhances its occurrence.

Natural wetlands, such as those in the Amazon, are important sources of methane (CH4), which is the second most important anthropogenic greenhouse gas in terms of radiative forcing. With a short atmospheric lifetime compared to carbon dioxide, reductions in CH4 emissions have the potential to mitigate global warming on much faster time scales. Currently, our understanding of these emissions is limited, with considerable disagreement between modelled wetland emissions estimates. Satellites can provide CH4 observations with high coverage and density that can provide the required observational constraints to improve CH4 emission estimates, especially for regions where in situ observations are sparse, such as the Amazon. In this thesis, I have carried out the first validation of CH4 from the GOSAT satellite using a series of dedicated aircraft in situ profile measurements demonstrating the high quality of GOSAT CH4 observations for this region. I have then used the satellite observations in conjunction with the aircraft profiles to investigate the characteristics of CH4 emissions from wetlands in the Amazon and their representation in state-of-the-art emissions inventories when combined with a chemical transport model. GOSAT observes large methane enhancements of up to 60 ppb between the wet and dry seasons in the Amazon coinciding with large Amazonian wetlands which are underestimated by approximately 15 ppb by models, pointing towards clear shortcomings in the inventories. To further assess the regionalCH4 emissions, a simulation system has been developed using a regional transport model based on a high resolution representation of atmospheric transport. This framework allows quick comparisons of different emissions inventories to GOSAT XCH4 on regional scales while giving a better representation of transport compared to global transport models. An assessment against global models and GOSAT data has shown that the model performs well and often agrees better with GOSAT than the global models do.

Modifications to an existing X-ray spectrometer are described which enable line profiles to be obtained undistorted by target ageing effects. Using the thin target technique, the Malpha spectrum of ytterbium fluoride was obtained and its dependence on exciting potential investigated. A similar investigation was made into the Malpha spectrum of hafnium dioxide. The profiles were analysed into Lorentzian components using a digital computer, and the satellites of ytterbium fluoride were found to attain a maximum integrated intensity of over 1100% of that of the parent line, whereas the Malpha satellites of hafnium dioxide attained a maximum integrated intensity of approximately 40% of that of the parent line. A vacuum oven is described, which was used to prepare, in situ, X-rey targets of ytterbium metal, reasonably free from oxide content. A method is described by which the Malpha spectrum of ytterbium metal was obtained. Comparisons are made between the Malpha spectra of hafnium dioxide, metallic ytterbium and triply ionised ytterbium, and the extraordinary magnitude of the Malpha satellites of triply ionised ytterbium is correlated with the presence of electron vacancies in the atomic core. An explanation of the large intensity of Malpha satellites in triply ionised ytterbium is given, in which they are regarded as pseudo-satellites arising from electron transitions in ionised ytterbium atoms, the parent line arising from transitions in atoms in which electrons have been excited into the core vacancies. Suggestions for further work have been made, which would establish the validity of these processes in the emission spectra of other rare earth elements. Also, it has been shewn that the Coster-Kronig theory of the origin of the M satellites of hafnium, proposed by Hirsh, cannot fully explain the observed facts, end a further suggestion has been made to explain their origin.

The problem of electron correlation in the field of atomic and molecular structure calculations is briefly reviewed. Various highly accurate atomic wavefunctions obtained from the literature have been analysed into the cluster form given by Sinano?lu and the orbital correction and pair correlation functions so obtained have been examined in detail. Computer programs have been developed to apply the method of Sinano?lu's Many Electron Theory to determine closed shell atomic and molecular wavefunctions in terms of Hartree-Fock orbitals and pair correlation functions. Various special techniques have been employed in the determination of the pair functions. These programs have been used to synthesise a series of wavefunctions for the Be-like ions and we have examined the effect on a number of one-particle expectation values of including only selected pair correlation interactions within the wavefunctions. In addition, we have investigated the effect of correlation on the 1S-1P generalised oscillator strength for the Be-like ions, the 1S-1P inelastic scattering cross-sections for electron and proton impact on beryllium and also on the photoionisation cross-section (or continuous oscillator strength) for both the He-like and Be-like ions. Approximate molecular pair correlation energies have been obtained for a series of diatomic hydrides and the variation of these quantities with bond distance has been discussed. Finally, we have determined spectroscopic constants for these molecules from the uncorrelated potential curves and also from potential curves obtained by representing the total correlation energy as the sum of our individual molecular pair energies.

The purpose of this thesis is to investigate the electrical and optical properties of a-GaAs thin films and also the related ternary compounds a-GaAlAs and a-GaAsP. The films were prepared by the technique of r.f sputtering in an argon atmosphere both with and without hydrogen. Experiments carried out include optical absorption, infrared absorption and electrical conductivity. The composition of the films was determined by EDAX. Information on structure and local bonding configurations for the a-GaAsP films was obtained from EXAFS measurements. Initially investigations were focussed on a-GaAs films prepared at room temperature, and in particular the effect of hydrogenation on film properties. Unhydrogenated a-GaAs has an optical gap of about 1.0 eV, although hydrogenation was found to increase this to about 1.3 eV. Infrared absorption data show that the main hydrogen mode in a-GaAs : H is the Ga-H-Ga bridging mode. It was also found that a-GaAs and a-GaAs : H films could be prepared at elevated substrate temperatures; similar measurements were also carried out on these specimens. Non-stoichiometric (As-rich) GaAs films were also prepared and characterized. Films of a-GaAlAs were successfully prepared by co-sputtering. For this system an interesting result is that the optical gap was found to decrease initially with increasing Al content before increasing at larger Al contents, in contrast to the crystalline case where the band gap increases monotonically with increasing Al content. Above 30 at.% A1, the films were found to be microcrystalline. A set of a-GaAsP films was also prepared. EXAFS measurements on these specimens reveal that Ga-P bonds in the amorphous network rapidly maximize at the expense of Ga-As bonds as the P-content increase. In other words the Ga bonding environment appears to be partially ordered. P-P and P-As "wrong" bonds are detected and, at high P-content, possibly also As-As bonds. The EXAFS results were found to correlate.

No description available.

The X-ray energy resolution achieved by conventional charge signal measurements using a gas proportional counter is determined by statistical fluctuations in the primary number of electrons and in the charge multiplication process. Conventional X-ray energy resolution thus increases as (E)-1/2 where E is the X-ray energy. The proportional counter, therefore, gives a very unsatisfactory X-ray energy resolution for X-ray energies below 1 keV. By eliminating the statistical fluctuations due to the charge multiplication, it is possible to improve the X-ray energy resolution by a factor of 2 over that achieved by conventional charge signal measurements. The implementation of this concept requires an 100% detection efficiency for the electrons present in the primary clusters. The present work is based on electron counting method which uses the charge signals due to single electrons avalanching at the anode wire. The main aim of this work was to determine the maximum possible electron counting efficiency. This required a detailed examination of the parameters relevant to the operation of an electron counting system. An experimental chamber consisting of a uniform field drift tube and a coaxial proportional counter was constructed. Experimental work was carried out to determine electron loss mechanisms such as electron loss by capture, electron loss below the discriminator threshold of the electron counting electronics and electron loss due to the finite resolving time of the electron counting electronics. This involved the measurements of electron mobility and electron lifetime at very low drift fields (Ed/p 0.02 V/cm Torr) for a number of different counter gas mixtures. Single electron response was also examined for these counter gas mixtures at a wide range of charge gains. It was found possible to achieve 89.0% electron counting efficiency at 1.49 keV using A-CH4(50%). The corresponding X-ray energy resolution was found to be 19.5%FWHM, compared to 28.0%FWHM achieved by the conventional charge signal measurements.

A detailed study of the sample of hot DA white dwarfs detected in the EUV and soft X-ray bands of the ROSAT all-sky survey is presented. Interpretation of the ROSAT data requires a priori knowledge of the temperature, gravity and visual brightness of all the white dwarfs detected. This information was obtained by a series of optical photometric and spectroscopic follow-up observations which are presented in detail. A by-product of this work is the mass distribution of the EUV selected sample of white dwarfs which shows an unexpectedly large proportion of high mass objects when compared with the optically selected population. It is suggested that this bias arises from the fact that hot white dwarfs with high gravities have lower heavy element abundances and are, therefore, more luminous (and easier to detect) than the lower gravity stars. In keeping with earlier results, it is found that the majority of stars 40,000K have more or less pure H atmospheres, while those above this temperature contain significant quantities of heavy elements. However, the increase size of the sample yields some important new results. A number of stars in the 40,000K to 50,000K range have nearly pure H atmospheres, a result at odds with the predictions of radiative levitation calculations. Furthermore, the dispersion in the observed EUV/X-ray opacity at a given temperature is much larger than expected from theory. An additional mechanism is then needed, other than radiative levitation and gravity, that can modify the atmospheric abundances. Finally, the expected strong dependence of opacity with surface gravity is not seen, except for very high mass stars.

Raman spectroscopy is a chemical and molecular structural identification technique which is sensitive to the vibrational modes of molecules. It is used in many fields, including: pharmaceuticals, security/defence and nuclear waste. In the past decade, significant advances in the miniaturisation and robustness of key technologies have also led to the consideration of Raman spectroscopy for application in the field of planetary exploration and in 2020 the first Raman instruments will be launched as part of two missions to the surface of Mars. The first is ESA’s ExoMars rover, which will include a compact macroscopic Raman instrument within its analytical laboratory. The second is NASA’s Mars2020 mission which will feature a stand-off Raman spectrometer: an instrument that provides a method of retrieving molecular information from remote distances, enabling rovers to investigate otherwise inaccessible regions. However, current stand-off and in situ instrument designs are limited by the challenging resource constraints associated with deploying systems on rover platforms, resulting in the need for detailed system software models and instrument prototypes that enable complex trade-offs and optimisations to be performed. This thesis presents work completed in preparation for the ExoMars, Mars2020 and other future planetary exploration missions that require information on molecular composition to fully address their science goals. It describes the development of sophisticated, end-to-end radiometric models and detailed software simulation tools for predicting and evaluating the performance of miniaturized Raman instrumentation. Details on the development of a prototype stand-off Raman instrument, designed to verify the software models are also provided, along with results from a number of instrument performance and model verification tests, including detector radiation damage campaigns. Several studies completed in order to assess and demonstrate the performance of various prototype and flight representative Raman instruments are also reported, including laboratory investigations of mission relevant sample standards and two field test campaigns.

Injecting a liquid jet into vacuum, and co-depositing this with atomic vapour, various nanomaterials were synthesised. Nanoparticles of zinc oxide, copper, silver and silica were produced. The silver vapour / ethanol liquid jet system was chosen to benchmark this new synthesis method. Particles of silver in ethanol had a log-normal size distribution, with a median size proportional to the logarithm of silver concentration, over three orders of magnitude. Particles showed a plasmon resonance characteristic of silver nanoparticles. It was shown that the plasmon absorption peak shape (height to width ratio) was approximately constant for silver/ethanol samples, regardless of concentration or ageing. The solvent used had a profound effect on both the particle size, which was measured varying from 2 nm in water to 10 nm in isopropanol, but also on the plasmon shape. This ratio changed dependent on the solvent chemistry, but was largely independent to other parameters. By dissolving commercially available spheres of silica in methanol, jetting, and codeposition with silver, silica particles were decorated with silver. These particles were catalytically active, and evolved away methanol at a rate of 1.65 x 10- 7 mol g-1s-1. A phase of ultra-small clusters was identified in both silica and silver samples by mircoscopy and spectroscopy. In the case of silver, these were tentatively ascribed to clusters of Ag2 < n < 5. Silver deposited with pure water was highly selective of these particles. Atomic force microscopy studies of arrays of these particles on surfaces showed evidence for the co-existence of three phases of a cluster fluid. Order of magnitude diffusion rates were estimated from experimental data. Liquid like cluster arrays had diffusion rate on the order of 10-2 nm2s-1 and gas phase clusters were estimated to have a diffusion rate on the order of 1 nm2s-1.