This thesis describes the growth and characterisation of four different III-V semiconductor materials. Growth was primarily performed by molecular beam epitaxy, while characterisation, which was largely structural, was carried out mainly using X-ray difraction and atomic force microscopy. Growth of low temperature(LT)GaAs was undertaken to investigate whether a phase transition accompanies the structural transition which occurs when GaAs is grown at temperatures below ~ 150C. It was found however, that LT GaAs remains zinc-blende, albeit with a signicant degree of disorder. Migration enhanced epitaxy was subsequently used to grow LT GaAs, resulting in single crystalline GaAs at growth temperatures down to 115C. The possibility of using AlN as a source for nitrogen, in the growth of GaAs based dilute nitrides was explored. No conclusive evidence has been presented to suggest that small amounts of nitrogen were incorporated into the GaAs lattice. The potential for ScN to be used as a buffer layer/interlayer to reduce the defect density in cubic GaN (c-GaN) was investigated. It was found that ScN grows on c-GaN(0 0 1)/GaAs(0 0 1) in a (1 1 1) orientation, leading to overgrowth of GaN occurring in the hexagonal phase. If the ScN interlayer was sufficiently thin (<3nm), then overgrowth of GaN was cubic, but no evidence of a reduction in stacking fault density was observed. Growth of ScN on GaAs(0 0 1) was also found to result mainly in a (1 1 1) orientation, but films were of poor quality. Growth of ScN on ScAs(0 0 1) was subsequently explored. ScN was found to grow in a (0 0 1) orientation, with both smoother surfaces and improved material quality than ScN(1 1 1). Growth of GaN atop ScN(0 0 1) was found to be c-GaN(0 0 1), but insuffcient studies have been carried out to determine the effect on material quality. During the growth of InGaN, it was found that unmounted substrates lead to large temperatures rises (>100C) for In rich compositions. Modelling heat absorption due to bandgap, phonon and plasmon absorption showed that this is due primarily to the large number of free carriers and not to the narrow bandgap (wrt substrate). The preliminary doping of In0.8Ga0.2N with Mn was investigated. The amount of Mn that can be incorporated without causing a signicant reduction in film quality was found to increase with decreasing growth temperature.

Phase images of the human brain acquired using gradient echo based Magnetic Resonance Imaging techniques show excellent contrast at 7T. This contrast is attributed to small variations in magnetic susceptibility that perturb the main magnetic field and thus yield a spatial variation of the NMR frequency. The work described in this thesis is primarily concerned with mapping the distribution of magnetic susceptibility within the human brain using these phase images. The main technical challenges of the project were first to extract accurate field maps based on phase data, and then to solve the ill-posed problem of inverting these field maps to reconstruct susceptibility (ϰ) maps. In initial work, simulations of field shifts based on known ϰ -distributions are compared to field maps acquired in vivo to highlight the non-local relationship between measured field offsets and the underlying susceptibility. These simulations were carried out using a recently derived Fourier method. The bulk of the thesis is then devoted to a detailed study of the process of inverting field maps generated from phase data using the Fourier relationship to yield quantitative 3D ϰ -maps. Unfortunately, the inversion problem is ill-posed and requires careful conditioning, either through rotation of the sample being imaged or through regularisation. A simple k-space threshold is introduced to condition the inversion and the preliminary results of applying this method to brain data from healthy subjects and patients with Parkinson's disease and multiple sclerosis are presented. The results suggest that susceptibility mapping is sensitive to iron deposition and could be a useful tool in investigating the progression of neurodegeneratived diseases. Iterative inversion algorithms, which deal with noise more robustly and allow more sophisticated filtering techniques to be employed, are then presented. These powerful regularisation methods are compared to previously described techniques, and are shown to yield high quality whole-brain ϰ -maps.

This thesis describes the characterisation of GaMnAs, related compounds and heterostructures. GaMnAs and other (III,Mn)V compounds have provided many interesting insights into fundamental physics, and are of considerable potential interest commercially in the field of spintronics. This study examines a set of samples grown by molecular beam epitaxy and characterised using several techniques: primarily this study makes use of the x-ray absorption techniques, x-ray magnetic circular dichroism(XMCD) and x-ray absorption spectroscopy (XAS). In addition, x-ray diffraction (XRD), transport measurements and super conducting quantum interference device (SQUID) magnetometry were used as complimentary techniques. GaMnAs layers with epitaxial Fe grown on top, are shown to have a sub-nanometre interfacial layer which remains polarised above room temperature. A detailed understanding of these systems is obtained by applying the element specific nature of XMCD in combination with two different probing depths to explore separately the nature of the coupling of the bulk and interfacial region. The coupling between the interfacial layer and the Fe is shown to be strongly antiferromagnetic (AF). A weaker coupling is also shown to exist between the Fe and the bulk of the \gamnas layer below the Curie temperature (Tc). This coupling is also AF at low fields, leading to an exchange bias for the entire layer. Doping of GaMnAs with P is shown to have several effects on the magnetic properties of the GaMnAs layer. Changes in the layer strain are observed using high resolution XRD. This strain also manifests itself in the Mn L_2,3 XMCD spectra and the relationship between the two is shown to be linear. A pronounced effect on the magnetic anisotropy is observed using SQUID measurements, with the easy axis switching from in-plane, in the compressively strained GaMnAs, to out-of-plane in the higher doped GaMnAsP layers. A decrease in total magnetic moment per Mn atom and Tc are observed with increased doping. This is inferred not to be due to a direct effect of the P on the local surrounding of the Mn ions, owing to the striking similarity of the XMCD spectra. This is instead attributed to reduced participation of Mn ions in the magnetic ordering. Finally, K edge XMCD is used to reveal the element specific nature of unoccupied states near the Fermi level in a set of GaMnAs and (In,Ga,Mn)As samples with differing Mn doping levels . The character of the holes in low-doped samples is shown to be markedly different than for those in the highly doped metallic samples. A transfer of orbital magnetic moment from the Mn to the As sites is observed on crossing the metal-insulator transition, with the large XMCD on Mn sites in low doped samples interpreted as a sign of hole localisation around the Mn ion.

This dissertation explores how decisions about the forecasting process can affect the evaluation of forecasting performance, in general and in the domain of residential electricity demand estimation. Decisions of interest include those around data sourcing, sampling, clustering, temporal magnification, algorithm selection, testing approach, evaluation metrics, and others. Models of the forecasting process and analysis methods are formulated in terms of a three-tier decision taxonomy, by which decision effects are exposed through systematic enumeration of the techniques resulting from those decisions. A computation platform based on the models is implemented to compute and visualize the effects. The methods and computation platform are first demonstrated by applying them to 3,003 benchmark datasets to investigate various decisions, including those that could impact the relationship between data entropy and forecastability. Then, they are used to study over 10,624 week-ahead and day-ahead residential electricity demand forecasting techniques, utilizing fine-resolution electricity usage data collected over 18 months on groups of 782 and 223 households by real smart electric grids in Ireland and Australia, respectively. The main finding from this research is that forecasting performance is highly sensitive to the interaction effects of many decisions. Sampling is found to be an especially effective data strategy, clustering not so, temporal magnification mixed. Other relationships between certain decisions and performance are surfaced, too. While these findings are empirical and specific to one practically scoped investigation, they are potentially generalizable, with implications for residential electricity demand estimation, smart electric grid design, and electricity policy.

This doctoral thesis presents density functionalcalculations applied in several domains of interest in solidstate physics and materials science. Non-collinear magnetismhas been studied both in an artificial multi-layer structure,which could have technological relevance as a magnetic sensordevice, and as excitations in 3d ferromagnets. The intricatebulk crystal structure of γ-alumina has been investigated.An improved embedded cluster method is developed and applied tostudy the geometric and electronic structures and opticalabsorption energies of neutral and positively charged oxygenvacancies in α-quartz. Ab initio total energycalculations, based on the EMTO theory, have been used todetermine the elastic properties of Al1-xLixrandom alloys in the face-centered cubiccrystallographic phase. The obtained overall good agreementwith experiment demonstrates the applicability of the quantummechanics formulated within the framework of the DensityFunctional Theory for mapping the structural and mechanicalproperties of random alloys against chemical composition.

Numerical simulation techniques play an important role due to their flexibility in dealing with a broad range of complex geometries and material responses. This flexibility requires substantial computational time and memory. Most numerical methods use structured grid for graphical discretization, although this approach is straightforward it is not ideal for smoothly curved boundaries. In this thesis the two-dimensional Transmission Line Modelling (TLM) method based on unstructured meshes is adopted. TLM is an established numerical simulation technique that has been employed in a variety of applications area. Using unstructured meshes to discretize the problem domain permits smooth boundary presentation which provides significant enhancement in the flexibility and accuracy of the TLM simulations. An algorithm is developed to implement Unstructured Transmission Line Modelling (UTLM) which is carefully designed for simplicity and scalability of model size. Several examples are employed to test the accuracy and efficiency of the UTLM simulations. Delaunay meshes, as a type of unstructured meshes, provide good quality triangles but have the disadvantage of providing close to zero transmission line length which has impact on the maximum permissible time step for stable operation. In this thesis, a simple perturbation method for relaxing the minimum link length and clustering triangles in pairs is presented, which permits substantial increase in time step and hence computational runtime to be made without compromising the simulation stability or accuracy. Also, a new model for relaxing the short link lines that fall on the boundaries is presented. UTLM method is based on temporal and spatial sampling of electromagnetic fields which results in dispersion. In this thesis, dispersion characteristics of the unstructured TLM mesh are investigated and compared against structured TLM results for different mesh sizes and shapes. Unlike the structured TLM mesh, the unstructured mesh gives rise to spatial mode coupling. Intermodal coupling behaviour is investigated in a statistical manner upon the change of the mesh local characteristics.

A dissertation submitted to the Faculty of Engineering, University of the Witwatersrand, in fulfilment of the requirements for the degree of Master of Sclenice in Engineering February 20th, 1998 / All available model used to determine the 50% breakdown voltage for rod-type ami conductor-type gaps subject to switching is impulse wave forms has been applied to a tower-conductor window gap configurution. The results for rod-plane, conductor-piane and tower-conductor window gaps have all been compared and correspond well with practical data. III order to app(v the model, a charge simulation technique has been adopted ill conjunction with the 'Coulomb 3D' charge simulation package. Additional tests have been performed where parameters of'the charge simulation method am! the electrode geometl:p hare been adjusted and consequent conclusions made. Recommendations for further application of tile model have been suggested. / MT2017

MSc., Faculty of Science, University of the Witwatersrand, 2011 / In this thesis, a novel type of electrostatic thruster is introduced. The Corona Ionization (CorIon) Space Propulsion system is an electrostatic propulsion system intended for use on satellites and for deep space probes. It makes use of the corona ionization mechanism to create the needed propellant ions. This same mechanism is also responsible for the thrust, thereby reducing its size and complexity. First, the effects of incomplete ionization of propellant molecules is discussed and conclusions drawn. Next, a mathematical model describing the electric field characteristics is derived. Considering the needle tip as a point charge and the exhaust plume to be cylindrically symmetric with a constant spread angle, the resultant electric field of both the needle tip and the produced ions obeys Poisson‘s equation. The charge density is obtained from the relationship between the drift velocity and the current. In order to solve the differential equation, we consider the electric field to only change in the radial direction so that Poisson‘s Equation is reduced to its radial part. This differential equation is solved to yield the electric field of the system. Some results are discussed. By integrating the electric field the relationship between the potential difference and the current of propellant ions is obtained. This relationship also yields insight into the ionization efficiency. Following this, an expression for the thrust is derived via two different methods: The first uses the energy conservation, and is termed ―Vector heating‖. The ions are viewed as a current heating the neutrals in the plume in the direction away from the needle. A temperature can be derived for the plume, and the resulting average gas velocity estimated from molecular theory. Finally, using the rocket thrust equation, an expression for the thrust is obtained. iv The second, more conventional method uses electrostatic repulsion to calculate the recoil on the needle: from the electric field computed for the system, an expression for the Coulomb forces on the ionized propellant can be derived. The recoil on the needle will experience the same force, resulting in thrust. Finally, the theoretical predictions for the various parameters are compared to experimental data. From this comparison, it is seen that there is a reasonable agreement between the experimental data and the model even though the electrostatic prediction underestimates the thrust of the system.

A dissertation submitted to the Faculty of Engineering, University of the Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree of Master of Science in Engineering. Johannesburg, 1997. / Results are reported on an initial investigation into corona power loss and its relationship to corona initiated audible noise. The work has the goal of being able to accurately predict the corona power loss, from a transmission line, from knowledge of the audible noise on the line. Corona power loss and its relationship to corona current, applied voltage and electric field is also exarr-ined. Experiments were performed using two single phase, different diameter, corona cages. Although the corona cage test results showed no correlation between corona initiated broad band audible noise and corona power loss, a linear empirical relationship, with excellent correlation, was found to exist between corona power loss and 100Hz Hum (dBA above 20jlPa). Results showed that, at the same magnitude of electric field, within the l h o different diameter corona cages, the corona current is the same magnitude and the corona power loss is proportional to the applied voltage. / MT2017

The thermally induced flux motion and the elemental pinning force, fp, for Nb thin films (1000-5000A) were measured for applied magnetic fields ranging from 0.3 to 7.5G, and temperatures from 4.22K to 5.72K. / The magnitude of f$\sb{\rm p}$(H,d,T) ranged from 10$\sp{-12}$ to 10$\sp{-11}$ N/m which is approximately 5 orders of magnitude smaller than Lorentz force depinning measurements made on Nb for the high field regime$\sp{16}$ (flux line lattice), as well as the low field regime$\sp{15}$ (isolated essentially non-interacting fluxoids). Some of these results are similar to the works of Huebener, et al, who first found a large discrepancy between the transport current method (J x B) and the thermal method, S$\nabla$T, when calculating the pinning force on a flux line lattice structure. A model was proposed to explain the discrepancy in terms of an electron scattering effect at or near the grain boundaries and extending into the grains which produces a channeling effect whereby the transport electrons flow away from the trapped fluxoids such that their local Lorentz force per unit length is less than the value calculated from J$\sb{\rm ave\Phi o}$. / The $\Delta\Phi$ vs $\Delta$T data for small $\Delta$T values, implied a spectrum of pinning force values where the fluxoids that were depinned and had substantially weaker pinning forces than the majority of the remaining trapped flux lines. Using statistical arguments, the qualitative features of the $\Delta\Phi$(H,d) data and f$\sb{\rm p}$(H,d) is explained. The data exhibited a magnetic field threshold, below which there is no flux motion for the temperature range studied. The value of the minimum required applied field necessary for flux motion increases with increasing thickness. / Source: Dissertation Abstracts International, Volume: 49-12, Section: B, page: 5358. / Major Professor: William G. Moulton. / Thesis (Ph.D.)--The Florida State University, 1988.