Search

Many-body theories such as dynamical mean-field theory (DMFT) provide a description of the local electron correlation effects that are missing in current density functional theory (DFT) calculations. This thesis presents DFT with DMFT (DFT+DMFT) studies for various materials in which the predicted description of the local electron correlations yields improved agreement with the experiment for certain quantities. Here, the methodology of the newly developed ELK-TRIQS interface is presented, including how to calculate DFT+DMFT wave functions which can be used to calculate DFT+DMFT wavefunction-dependent quantities. This is first illustrated by calculating the electron localisation function (ELF) in monolayer SrVO3 and CaFe2As2, which provides a means of visualising their chemical bonds. Monolayer SrVO3 ELFs are sensitive to the charge redistribution between the DFT, one-shot DFT+DMFT and fully charge self-consistent DFT+DMFT calculations. In both tetragonal and collapsed tetragonal CaFe2As2 phases, the ELF changes weakly with correlation induced charge redistribution of the hybridised As p and Fe d states. The magnetic Compton profiles (MCPs) of Ni and the Compton profiles of V, using both DFT and DFT+DMFT, are presented. For Ni, the theoretical MCPs were calculated using the full potential linear augmented plane wave method with the numerically exact continuous-time quantum Monte Carlo DMFT solver, along with the Korringa-Kohn-Rostoker (KKR) method with the perturbative spin-polarised T-matrix fluctuation exchange approximation DMFT solver. The spin magnetic moments decrease with the intra-atomic Coulomb repulsion U, which is also reflected in the corresponding MCPs. The total magnetic moment obtained from the superconducting quantum interference device measurements can be reproduced by intermediate values of U. However, the MCP shape still disagrees with the experiment. The spectral function reveals that the minority X2 Fermi surface pocket shrinks and gets shallower with respect to the DFT calculations. For V, The addition of DMFT improves the Compton profile directional differences significantly, but these differences are fairly independent of the U values used. There are improvements in the spectral function with respect to the corresponding experimental quantities, except for the N-hole ellipsoid Fermi surface pocket sizes which worsen. For both Ni and V, there is a vital need to include the missing electron correlations beyond the DFT+DMFT picture. With the increasing interest in superlattice systems described by DMFT, here the metalinsulator transition (MIT) of strongly correlated 3d electrons in SrVO3 superlattices is shown to be due to quantum confinement. By producing excellent agreement between experiment and theoretical spectroscopic quantities, the underlying physics in these systems is captured by DMFT. New light is shed on the microscopic mechanism of the MIT and previously reported anomalous subband mass enhancement, both of which arise as a direct consequence of the quantization of V xz(yz) states in the SrVO3 layers. Therefore, quantum confinement can tune the strength of electron correlations.

In this thesis it is demonstrated how mass transfer can be induced within a network of reticulated channels through turbulent mixing, and how this mechanism can be applied as an integrated functional structure in a building envelope. The research is based on suggestions in literature that there are transient modes of mass transfer which are active in termite mounds of the species Macrotermes michaelseni and which do not rely on steady or cross flows: is the interaction of transient air flows, induced or natural, and the geometry found in the mounds - a complex reticulated network of tunnels with variable dimensions - conducive to controlled mass transfer? My original contribution is to show that such mass transfer is possible, to demonstrate the ways which geometry and input flow interact to create turbulent mixing, and to demonstrate how these parameters may be useful to control the flow within and across building elements. The research method is structured around a series of experiments where various geometries, based on the egress complex of termite mounds, are exposed to transient fluid flows, either a controlled, low amplitude oscillation or external turbulent flow. The resulting mass transfer is measured using tracer gas measurements and visualisations using planar laser and/or fluorescent dye. It is found that the oscillations in certain conditions, which involve a combination of geometry and oscillation properties, lead to the emergence of large scale turbulent convection which can increase the mass transfer rates with up to two orders of magnitude over an unperturbed system. It is concluded that such mechanisms are potentially useful to control the flows of heat and moisture within buildings and across building envelopes. The particular properties of the described flows can complement conventional steady flows and expands the repertoire available to the building designer, enabling new types of boundaries and functional integration in the building envelope.

This thesis draws conceptually and directly on music notation in its investigation of the temporality of landscape architecture. It differs from other studies that refer to music in two critical ways: firstly rather than their references to Baroque, Classical or Romantic music, this investigation focuses on music notation from a 27 year period of musical innovation in the twentieth century; secondly in referring to music notation as a source for landscape architectural notations, I address their omission of sound, which in importing aspects of music notation into design, they curiously left behind. The thesis studies three music notations in fine detail: Projection I1 by Morton Feldman, from 1950, Lontano by Gyorgi Ligeti from, 1969, and Green Meadows by Michael Finnissy, from 1977. Each study examines the notation of a musical score and its specific approach to time, investigating whether the temporality of this music is similar to that of aspects of landscape time. Each is juxtaposed with design notations representations by First nameDonald Appleyard et al from The View from the Road, 1965, Bernard Tschumi from Manhattan Transcripts, 1994, and William Kent from Rousham garden, 1748, before examining four landscape spaces through the development of new landscape architecture notations. The research makes available to a wider design audience the works of three influential composers of the latter half of the twentieth century, presenting a critical evaluation of their work within music, as well as a means in which it might be used in landscape architectural research. The thesis also offers valuable insights into the methods used by landscape architects for the benefit of musicians, and by bringing together musical composition and landscape designarchitecture through notation, it affords a focused and sensitive exploration of temporality and sound in both fields.

This thesis studies photogrammetric techniques applied to the ATSR instruments for the extraction of atmospheric parameters with the objective of generating new scientific datasets. The atmospheric parameters under observation are cloud top height, smoke plume injection height, and tropospheric wind components. All have important applications in various tasks, including the initialisation and validation of climate models. To generate accurate stereo measurements from the ATSR imagery the forward and nadir views need to be accurately co-registered. Currently this is not the case, with differences of up to 2 pixels in both axes recorded. In this thesis an automated image tie-pointing and image warping algorithm that improves ATSR co-registration to ≤1 pixel is presented. This thesis also identifies the census stereo matching algorithm for application to the ATSR instruments. When compared against a collocated DEM, census outperforms the previous stereo matching algorithm applied to the ATSR instrument, known as M4, significantly: RMSE ~700m vs. ~1200m; bias ~60m vs ~600m; R2 ~0.9 vs ~0.7. Furthermore, this thesis reviews the M6 algorithm developed for application within the ESA ALANIS Smoke Plume project. Using census a climatological cloud fraction by altitude dataset over Greenland is generated and demonstrated to agree well with current observational datasets from MISR, MODIS and AATSR. The 11μm channel stereo output provides insights into high cloud characteristics over Greenland and appears to be, in comparison with CALIOP, practically unbiased. The ALANIS Smoke plume project is introduced and the inter-comparison of the M6 algorithm against MISR and CALIOP is presented. M6 demonstrates some ability for determining smoke plumes injection heights above 1km in elevation. However, the smoke plume masking approach currently employed is demonstrated to be lacking in quality. Finally, this thesis presents the determination of cloud tracked tropospheric winds from the ATSR2-AATSR tandem operation using the Farneback optical flow algorithm. This algorithm offers accuracy on the order of 0.5 ms-1 at full image resolution, which is unprecedented in comparison to similarly derived datasets.

In this study, we have proposed several methods for sampling field-based representations of refraction attributes, and analysed these methods in terms of the quality and accuracy of the results produced. We have identified two types of anomalies that affect the correctness of some methods, and found that the best of these methods is capable of producing images of equal quality to surface-graphics techniques. Our approach produces good results with functionally defined models, and discrete volumetric data, in the absence of noise. We have also established that this approach allows the modelling and rendering of types of objects that cannot be represented in surface graphics (e.g. objects with a continuous, non-uniform refractive index). Where noise is present in volumetric data, it significantly affects the correctness of rendering refraction, usually resulting in images of poor quality. A number of methods for smoothing have been examined, including low-pass filtering and various types of nonlinear diffusion. We have shown that regularised anisotropic nonlinear diffusion is a powerful and effective method for processing noisy volumetric data sets in order to improve the quality of images featuring refraction. High quality images featuring refraction often take a long time to render. We have presented the design and implementation of a parallel, open source volume graphics rendered which supports refraction and multi-volume scenes. A large number of issues related to a efficient operation of a parallel renderer have been examined, and a high-performance, scaleable parallel renderer has been developed, which performs efficiently on both networked clusters, and shared memory multi-processor machines.

This Engineering Doctorate is concerned with change detection. The application of fuzzy clustering and fuzzy classification for online pipeline leak detection is investigated. Three different fuzzy methodologies are derived using one real data set and tested on multiple other real data sets. The first classification methodology, based on a fixed classifier, uses a novel merging algorithm developed to reduce the influence of cluster initialisation on the data space partition. Rules derived from this partition are subsequently used in the second methodology, to define a linguistic rule-based system, classifying without clustering. Instead, it evaluates membership functions to classify. The third methodology is a combination of the previous two, enhanced with cluster rejection and statistical change detection in the form of Shewhart charts.

In this dissertation, the large-scale geophysical behaviour of the Earth, Venus and Mars are compared, using data collected by the Magellan spaceprobe (for Venus) and the Viking and Mars Global Surveyor (MGS) probes (for Mars). Neither Venus nor Mars show evidence of plate tectonics operating at the present day. On Venus, the lack of water means the frictional resistance at faults and the viscous drag on the base of the moving lithospheric plates are too high to be overcome by the driving forces for plate tectonics. The high elastic thickness of Mars results in a large frictional resistance to fault motion, although the faults themselves are probably no stronger than those on the Earth, and means large compressive stresses are required to initiate subduction. The likely high viscosity of the martian mantle, a consequence of its probable dryness and low temperature, may also result in large drag forces on the base of the lithosphere. Plate tectonics may have operated in the past on both planets, providing a possible explanation for the rapid resurfacing of Venus required by the crater counts and the linear magnetic anomalies recently discovered on Mars.

Throughout the 1990's several investigators have used a mechanistic theory based upon third order kinetics to explain the rates of the reactions of several substrates in aqueous mixtures of protic (e.g. methanol) an aprotic (e.g. acetone) solvents. Such substrates investigated have included p-nitrobenzoyl chloride and p-nitrobenzenesulfonyl chloride. Reactions of these substrates in aqueous alcohols yield two products; an acid (RCO₂H and RSO₃H) and an ester (RCO₂R and RSO₃R). Measurements of the relative amounts of the products obtained from these reactions has enabled investigators to calculate selectivity values, S (S = [Ester]/[Acid] x [Water]/[Alcohol]). Both the rates and products (selectivities) of these reactions have been successfully explained by this third order theory. This thesis (Chapters 1-5) is primarily concerned with testing the third order theory further by investigating the reactions of selected phosphoryl chlorides (e.g. diphenyl-chlorophosphate, (PhO)₂POCl) in aqueous mixtures of protic and aprotic solvents. Some of the results obtained in this thesis strongly support the theory whilst some unexpected results have been explained by initial state solvation effects. Unexpected changes in selectivity in highly aqueous mixtures of methanol and ethanol (e.g. 10-30%) may be caused by complex solvent-solvent interactions in these solvents. Chapter six of this thesis is not concerned with the third order theory investigated in Chapters 1-5. Previous investigations of the solvolyses of p-methoxybenzoyl (anisoyl) chloride in aqueous methanol and ethanol have resulted in an S_N1 type mechanism involving solvent separated ion pairs being postulated. Selectivities for these solvolyses are constant over the whole range of methanol-water and ethanol-water mixtures investigated (10-90%, % v/v) with the selectivity in ethanol being less than in methanol. These constant selectivities have provided an opportunity to investigate aqueous mixtures of other alcohols (e.g. propanols and butanols) and to discover whether effects such selective solvation of anisoyl chloride by alcohol occurs. Any changes in the pattern of selectivity might be due to such effects and not a change in mechanism because anisoyl chloride is reluctant to react via associative mechanisms even in solvents of low ionising power (e.g. ethanol).

Thin coatings and other surface engineering techniques are widely used to improve the friction and wear properties of surfaces. However, current understanding of the behaviour of surfaces is relatively poor. Production methods are therefore to a large extent based upon empiricism. Many well-established materials characterisation tests are inadequate, and so novel techniques for the examination of surface layers have emerged, with the eventual aim of predicting the tribological performance of surfaces. One such technique is the micro-scale abrasive wear test, in which a ball is rotated against a specimen in the presence of a slurry of fine abrasive particles, producing a well-defined crater whose volume may be measured geometrically, allowing the wear coefficient of the specimen to be determined. It has been shown that the existing understanding of test is inadequate; further characterisation of the test has been performed. The effects of various parameters on the mechanism and severity of wear have been identified and explained by adaptation of existing models. Recommendations have been made for the optimisation of the accuracy and reproducibility of the test. The capability of the test to characterise thin coatings has been extended by shallow ball-crater testing which does not penetrate though the coating, thereby eliminating any influence of the substrate wear resistance. A number of different formulations of the wear equation, and a number of data analysis methods have been discussed, with the aim of minimising error. Since many surface coatings are exposed to high temperatures in service, the micro-scale abrasion test and the commonly used scratch test have been adapted for use at elevated temperatures in order to investigate changes in coating properties under these conditions. A new scratch test apparatus was designed and constructed for these tests. The behaviour of various PVD coatings has been investigated by these methods.

The purpose of this project was to show that aminyl radicals (nitrogen atoms with one unpaired electron) could be generated from the radical reactions between tri-n-butyltin hydride and sulfenamides. Sulfenamides are defined as a divalent sulfur atom bonded to a trivalent nitrogen atom. A wide range of sulfenamides were prepared in good yield using two methods. Benzenesulfenyl chloride and primary or secondary amines were reacted at room temperature with triethylamine as base to yield sulfenamides. Sulfenamides were also synthesised in good yield from the reaction between amines and N-(benzenesulfenyl)phthalimide. Initially, reactions were performed to show that aminyl radicals could be formed by reacting selected sulfenamides with tri-n-butyltin hydride in refluxing solvent. When these reactions proved to be successful, experiments were carried out to cyclise the intermediate aminyl radicals intramolecularly onto an olefin. The studies have led to a new method for the formation of nitrogen containing heterocycles.

The purpose of this thesis is to determine whether the Registered Land Act 1963 of Kenya has established an effective system of law and practice governing titles to land registered under the Act. Several key issues are addressed. First, how effective has been the process of land adjudication which brings onto the register land that was formerly subject to customary law; moreover, how successful has been the process of converting land that was subject to one of the preexisting systems of registration. Secondly, how effective is the conveyancing machinery provided by the Act. Thirdly, are the rights of registered proprietors, including those registered jointly or in common, as well as persons with third party interests in land adequately protected by the Act? Fourthly, to what extent have the provisions of the Magistrates' Jurisdiction (Amendment) Act 1981 undermined the provisions of the Registered Land Act 1963? In answering these questions the relevant provisions of the Registered Land Act 1963 are compared with those of the English Land Registration Act 1925. This thesis considers the extent to which judicial interpretation of the provisions of the Land Registration Act 1925 can assist in solving some of the problems created by the provisions of the Registered Land Act 1963. It is contended that the Registered Land Act has failed to provide a truly effective system of registered land.

The non-invasive nature or nuclear medicine techniques makes them particularly attractive for the investigation of cardiac patients. Since the early 1970's there has been a large research effort directed towards implementing nuclear cardiological techniques. Two broad types of LY functional tests have been proposed:- first pass of a bolus of radionuclide through the heart; and gated blood pool imaging of a radionuclide tracer that has equilibrated in the vasculature. The use of thallium to image the normal myocardium has also been proposed. This thesis is concerned with identifying sources of error in these radionuclide techniques. This is important both for the routine application, of the tests and for acing the tests as a research tool in cardiology. In particular, the use of rapid (about 2 minutes) LV functional testing would enable the tests to be used for serial measurement of LV function during stressing of the heart. The errors involved in measuring left ventricular ejection fraction (LVEF) and in assessing regional wall motion (RWM) are identified, and ways of correcting these errors are proposed. In order to evaluate RWM the concept of minimal detectable distance (mdd) is proposed. Following error analysis of LVEF and RWM measurement, it is proposed that rapid (about 2 minutes) acquisition is feasible for these techniques. The potential for using TAllium imaging for the detection of perfusion defects is considered, and the problems involved are identified. The use of seven pinhole tomographic imaging is proposed as a possible solution to some of the problems identified in nuclear cardiological techniques.