There is an urgent need for characterisation of group V donors in silicon necessary for the development of quantum information processing (QIP) devices, and my PhD work has been contributed towards this objective. In this thesis, three different group V donors were individually studied with scanning tunnelling microscopy (STM) and spectroscopy (STS), combined with DFT calculation and simulated STM images where necessary. Si(001) dosed with phosphine (PH3) at room temperature prior and imaged at 77 K a few minutes later was investigated. Novel phosphine-related features were observed and compared to the room-temperature results [4]. Some features were the same as the room-temperature dissociative products but a new dissociative mechanism was proposed and novel bonding configurations were assigned to the features. A transformation between adsorbates was seen to occur on the surface at 77 K and was attributed to a tip-induced effect. The appearance of the Si-P heterodimer was found to be different at 77 K and room temperature [5] and the reasons for this difference was discussed. Subsurface As and Bi donors below the Si(001):H surface were separately investigated at 77 K. Special sample annealing (flashing) procedures were created for both studies. There were two classes of As features commonly observed while there were three types for Bi. The appearance of the As features related to the As wavefunction informed that the subsurface As donors were electrically neutral but could be reversibly switched to being ionised (positively-charged) by changing the sample bias. Some subsurface As donors can also be negatively-charged, depending on their distances from the surface. DFT calculations were performed and simulated STM images were generated to compare with the experimental data, allowing us to assign the features to As donors at their exact lattice positions. The appearance of the Bi features indicated that the Bi donors were negatively-charged but could be changed to being positively-charged. With the unequal rate of tunnelling in and off the donor energy level, the Bi donors could be switched back to being negatively-charged again. The Bi wavefunction projection was speculated, based on the Bi features.

Warming of human tissue causes vasodilation and therefore, increase in blood volume. Such thermal responses allow the assessment of hemodynamics in the tissue, providing physiological and clinically important information of the diagnosed subject. Local warming is often accomplished on the skin because of its accessibility and simplicity. To allow the investigation into deeper tissue such as the muscle, an innovative hybrid microwave-optical system has been developed. This comprises of a microwave system, an optical monitoring and cooling system. The tissue warming is induced by a novel microwave applicator, which was based on microstrip patch design operating at 2.45 GHz with a superstrate interface layer to improve the coupling of electromagnetic (EM) waves into the skin. The active cooling was introduced to reduce skin heating. While the optical sensors based on Near-Infrared Spectroscopy (NIRS), was used to measure the changes in tissue oxygenation including the muscle. This thesis demonstrates the development procedure, covering the design and operation of the entire system. Moreover, the majority of the work is based on the four developed applicators, where each design was evaluated using EM and thermal simulation based on numerical phantoms. The study evaluates the distribution of absorbed EM energy in the tissue known as the specific absorption rate (SAR). The applicators are developed in the following order: (i) Applicator I was fabricated for preliminary study for general tissue heating with the integrated optical probes. This early study provided an insight to the importance of superstrate thickness and material. (ii) Applicator II, which introduces a new approach to skin cooling based on Thermoelectric Coolers (TEC) and high thermal conductive superstrate. This design could cool the skin and monitor tissue oxygenation, skin perfusion and temperature. (iii) Applicator III was an updated model of the predecessor, resolving cooling configuration and the discrepancy in operating frequency, and was capable of minimising skin heating effectively (iv) Circularly polarized (CP) Applicator aimed at reduction of the SAR in the superficial layer, and hence skin heating. The simulated thermal study of all developed applicators was validated with exvivo (mimicked phantom) and in-vivo experimental trials. The measurements and the simulation model were in agreement, apart from the CP applicator due to the complexity of measurement. The results from the phantom and human calf indicated superficial heating was reduced by about 5.0-6.0 ° C when skin cooling was applied, while the temperature change in muscle was not significantly affected. The measurement with mimicked tissue showed the applicator was capable of elevating muscle temperature by approximately 3.0-4.0 ° C. This is a sufficient increase to cause tissue dilation, and therefore, change in the thermal response. The hybrid microwave-optical system has been developed and examined on three human calves during in-vivo physiological study. The results using Applicator II illustrated that the device can successfully stimulate and measure thermal responses in terms of oxy/deoxy/total haemoglobin concentrations changes ( HbO2/ HHb/ HbT). The slope (rate of change) of HbT curve during microwave exposure is defined as the thermal response. This parameter is essential in studying physiological responses between different subject, particularly in vascular diseases, transplanted free flaps and other conditions, including chronic spinal cord injury. Subjects with such conditions will have a distinguishable response to tissue heating than a healthy subject. The monitored haemodynamic signals of Applicator II are primarily based on superficial responses. However, measurements with Applicator III showed the potential of the applicator. The measured thermal response was 0.83 10 3×10⁻³ μM/s without skin cooling, which was dedicated by skin heating. The introduced cooling system has reduced the skin temperature and maintained the local skin micro-circulation, which was monitored with the secondary optical system based on Laser Doppler Flowmetry (LDF). This probe measures blood flow at superficial depth, and consequently, was used as a validation tool to demonstrate the cooling efficiency. The measured thermal response with skin over-cooling was -0.08 10 3×10⁻³ μM/s. The negative response indicates arterial constriction, and therefore, the skin heat was eliminated while the simulations study to indicate the muscle temperature was elevated by 3 ° C. However, the response was dominant by the superficial response. Obtaining a response from muscle only was challenging and currently being solved in numerous applicator and cooling technique, which have been presented in the thesis.

The aim of the research in this thesis was to design and model multimode lightguides for optimising visible light for liquid crystal display systems and to design, model and experimentally test infrared light propagation within polymer multimode waveguides as board-to-board interconnects for high data rate communication. Ray tracing models the behaviour of a novel LCD colour separating backlight to optimize its efficiency by establishing the optimum dimensions and position for a unique micro-mirror array within the light guide. The output efficiency increased by 38.2% compared to the case without the embedded mirror array. A novel simulation technique combined a model of liquid crystal director orientation and a non-sequential ray tracing program was used first time to compute the reflected intensity from a LCOS device for a rear projection TV system. The performance of the LCOS display was characterised by computing the contrast ratio over a ±15° viewing cone. Photolithographically manufactured embedded multimode waveguides made from acrylate Truemode® polymer are characterized by measuring the optical transmission loss of key waveguide components including. straight, bend and crossing. Design rules derived from the experimental measurement were used to optimize optical PCB (OPCB) layout. A most compact and complex optical interconnects layout up-to-date for data centres, including parallel straight waveguide sections, cascaded 90° bends and waveguide crossing other than 90° angles, was designed, tested and used in an optic-electrical demonstration platform to convey a 10.3 Gb/s data. A further new method for reducing the end facet roughness and so the coupling loss, by curing a thin layer of core material at the end of the waveguide facet to cover the roughness fluctuations, was proposed and successfully demonstrated giving the best results reported to date resulting in an improvement of 2.8 dB which was better than the results obtained by using index matching fluid.

Whilst diamond nanoparticles have been synthesised by the detonation method since the middle of the 20th century, it was not until the development of their fully dispersed form in 2008 that the hugely varied array of potential applications opened up. These prospective uses are described in chapter 2, spanning the breadth of the sciences; from the chemical (ion sensors and electrochemical electrodes), the biological (drug delivery and intracellular monitoring), to the physical (composite materials and growth nucleation seeds). Highlighting this versatility, chapter 5 of this thesis covers the use of ND as a neuronal biomaterial, investigating for the first time the role of the size, production method and surface functionalisation on neurite extension, where it was found that the only significant variable was the size of the ND used, suggesting the curvature of particle is of high importance. A patterned ND surface was also fabricated and this resulted in the successful growth of neuronal networks along the ND patterning, down to ND track widths of 10 μm. Chapter 6 presents a study into the electrical characteristics of nanodiamond. Here for the first time, ND layers with various surface terminations were produced and verified using Fourier transform infrared spectroscopy. These layers were then probed using impedance spectroscopy to obtain modelled values of the layers’ resistance and capacitance as a function of measurement temperature. This data was manipulated into Arrhenius plots to extract activation energies of the observed conduction paths. The hydrogen terminated ND layers were shown for the first time to be stable for short heating durations up to 475°C, although longer heating duration at this temperature did cause permanent damage to the ND layer. Oxygen terminated ND was found to be less stable when heated in atmospheric conditions, with permanent degradation occurring at 200°C. However when measured in vacuum, these layers showed resilience similar to that seen on the hydrogen terminated ND, suggesting the hydrogen termination is providing protection against oxidative degradation in atmospheric conditions. The next chapter documents the attempts to fabricate thin, conformal layers of ND within the channels of a microchannel plate (MCP) – the electron amplification stage of a night vision device. This work resulted in the first successful seeding of ND throughout the first 60 μm of the MCP channel, as requested by the industrial sponsor of this work, Photonis. This chapter also presents the iterative designs of a custom cooler and sample mount for the microwave plasma enhanced chemical vapour deposition chamber used for low temperature diamond growth both in this chapter and the final experimental chapter. The last experimental chapter details a novel investigation into the optimal conditions for ND to be grown into a thin film for use as a secondary electron emitter. The results presented include Raman spectroscopy (for film quality assessment), scanning electron micrographs (to provide topographical information), atomic force microscope line scans (to estimate film thickness) and finally secondary electron yield experiments upon each of the films. It was found that the films grown with the highest applied microwave power -­‐ resulting from the most aggressive cooling -­‐ gave the highest secondary electron emission yield, slightly under 10 (meeting the objective set by Photonis) along with a sharper sp3 peak at 1333cm-­‐1.

Networks of interacting users arise in many important modelling applications. Commuters interact with each other and form traffic jams during peak-time. Network protocols are users in a communication network that control sending rate and server choice. When protocols send with too high rates, network links get overloaded resulting in lost data and high delays. Although these two example users seem very different, they are similar on a conceptual modelling level. Accurate user models are essential to study complex interactions in networks. The behaviour of a user with access to different paths in a network can be modelled as an optimisation problem. Users who choose paths with the highest utility are common in many different application areas, for example road traffic, Internet protocol modelling, and general societal networks, i.e. networks of humans in everyday life. Optimisation-based user models are also attractive from the perspective of a modeller since they often allow the derivation of insights about the behaviour of the entire system by only describing a user model. The aim of this thesis is to show, in four practical studies from telecommunications and transit networks, where optimisation-based models have limitations when modelling users with path choice. We study users who have access to a limited number of paths in large scale data centers and investigate how many paths per user are realistically needed in order to get high throughput in the network. In multimedia streaming, we study a protocol that streams data on multiple paths and path properties matter. We also investigate complex energy models for data interfaces on mobile phones and evaluate how to switch interfaces to save energy. Finally, we analyse a long-term data set from 20,000 transit commuters and give insights on how they change their travel behaviour in response to incentives and targeted offers. We use tools from optimisation, simulation, and statistics to evaluate the four studies and point out problems we faced when modelling and implementing the system. The findings of this thesis indicate where user models need to be extended in order to be of practical use. The results can serve as a guide towards better user models for future modelling applications.

Active target location systems capable of measuring both range and bearing have niche applications, including maritime navigation where a seafaring vessel is manoeuvring in the vicinity of a harbour or an oil rig. Such systems can also be used to determine the location of other vessels for vessel-to-vessel personnel or material transfer. The usual approach is to combine FMCW radar with a mechanically or electrically steered beam, establishing both range and bearing to a target, respectively. The radar system described in this thesis is an innovative alternative approach, one that combines FMCW radar with a crossed-dipole antenna, which conveniently functions as a circular array, thereby simultaneously determining the range and bearing of an active target. By using phase mode excitation, neither mechanical nor electrical beam steering is required to locate the active target, as the receive antenna is able to monitor 360 in azimuth continually. However, due to the use of +1st and -1st order phase modes, the radar can only operate in an 180 sector unambiguously. The usual inherent problems with circular arrays being aected by multipath are also easily mitigated by the range resolution of the radar. This thesis describes in detail the development of a 2.44 GHz crossed-dipole antenna structure and its associated feed network. It also describes the first prototypes that led to its current form and goes on to discuss in detail the design and construction of the radar system and frequency shifted active target. Frequency shifting was implemented within the target to overcome the increased clutter power due to the omnidirectional receive antenna. However, firstly this thesis lays the foundation of radar theory, active targets, phase modes and basic antenna theory. Some of the literature associated with radars currently used in this type of scenario is also discussed. Appropriate analysis, modelling and experimental validation is conducted to assess system performance in relation to the predicted behaviour. The radar system was then tested in an open field, with the active target detected to a range of 125 m.

The Engineering Doctorate (EngD) was conducted in conjunction with BT Research on state-of-the-art Wireless Sensor Network (WSN) projects. The first area of work is a literature review of WSN project applications, some of which the author worked on as a BT Researcher based at the world renowned Adastral Park Research Labs in Suffolk (2004-09). WSN applications are examined within the context of Machine-to-Machine (M2M); Information Networking (IN); Internet/Web of Things (IoT/WoT); smart home and smart devices; BT’s 21st Century Network (21CN); Cloud Computing; and future trends. In addition, this thesis provides an insight into the capabilities of similar external WSN project applications. Under BT’s Sensor Virtualization project, the second area of work focuses on building a Generic Architecture for WSNs with reusable infrastructure and ‘infostructure’ by identifying and trialling suitable components, in order to realise actual business benefits for BT. The third area of work focuses on the Open Geospatial Consortium (OGC) standards and their Sensor Web Enablement (SWE) initiative. The SWE framework was investigated to ascertain its potential as a component of the Generic Architecture. BT’s SAPHE project served as a use case. BT Research’s experiences of taking this traditional (vertical) stove-piped application and creating SWE compliant services are described. The author’s findings were originally presented in a series of publications and have been incorporated into this thesis along with supplementary WSN material from BT Research projects. SWE 2.0 specifications are outlined to highlight key improvements, since work began at BT with SWE 1.0. The fourth area of work focuses on Complex Event Processing (CEP) which was evaluated to ascertain its potential for aggregating and correlating the shared project sensor data (‘infostructure’) harvested and for enabling data fusion for WSNs in diverse domains. Finally, the conclusions and suggestions for further work are provided.

Two-photon microscopy has become an invaluable tool for imaging neural activity at the cellular level. The non-linear excitation process of two-photon imaging has enabled neuroscientists to observe multiple brain cell dynamics at a higher depth, while causing lower photo-damage in comparison to confocal microscopy. However, conventional galvanometer-mirror based two-photon microscopes are not agile enough to follow the action-potential path between firing neurons. A fast scanning and refocusing 3D two-photon microscope has been developed by the Department of Neuroscience, Physiology and Pharmacology of University College London capable of rapid accessing hundreds of points of interest. The instrument achieves this by employing a new type of 3D scanner that uses four Acousto-Optic Deflectors (AODs) to form a high performance Acousto-Optic Lens. The aim of this research is to develop a highly sophisticated system capable of generating the necessary signals to drive the four AODs. Such a system would enable arbitrary 3D scanning and compensate for inherent aberration of the op-tical systems as well as tissue-induced aberrations, and revolutionise neural ac-tivity imaging. Different paths to developing the control system and synthesiser were investi-gated in depth and evaluated, in order to yield a confident selection of tools, technologies, platforms and techniques. Since the early research stages, it was realised that the system should be developed around a Field Programmable Gate Array (FPGA) device, a new technology but ideal for fast parallel systems. The resulted system is capable of generating amplitude modulated, non-linear frequency ramps at a very high precision. Using several experimental setups, it was shown that with this driver the AOL is not only capable of rapid access 3D focusing, but also rapid scanning and correcting for aberrations, making it a via-ble adaptive optics device. A fast scanning, rapidly focusing lens, which can compensate for specimen and/or optical path aberrations, can revolutionise functional imaging and open new horizons in the exploration and understanding of the human brain.

The original work presented in this thesis concentrates on the origin of the visible and near-IR luminescence from silicon nanocrystals (Si-NCs) embedded in two different matrices, namely SiO2 and SiOxNy, and prepared by three different chemical vapour deposition (CVD) techniques and ion implantation. The optical properties of these materials were studied by time-integrated photoluminescence, time correlated single photon counting (TCSPC) and time-resolved PL spectroscopy (TRPL) while the structural and compositional characterisation was carried out using ellipsometry, Fourier Transform Infrared spectroscopy (FTIR), Rutherford backscattering spectroscopy (RBS) and X-ray photoelectron spectroscopy (XPS). As opposed to the near-IR emission, well investigated and commonly associated with phonon-assisted excitonic recombination, the visible emission from Si-NCs in SiO2 is still subject to debate, with the majority of studies pointing to a defect-related origin. Recent works suggested that it is possible to achieve no-phonon direct-like transitions in Si nanostructures and thus extract a more efficient emission in the visible range. These works have sparked new interest in the high-energy emission from Si-NCs and motivated that part of the present research that aims at exploring the possibility of radiative recombinations in a process that does not involve phonons. In this work, the analysis of the PL decay curves evidenced the coexistence of multiple lifetimes, with components mainly grouped around two regions in the microsecond and picosecond domain. The decrease in PL intensity with annealing is correlated with the shortening of PL lifetime, characteristic of a class of non-radiative defects. Although my time resolved PL study shows the presence of ultra-fast lifetime components that could play a role in the populations of energy levels at the  point, there is no evidence in my samples for no-phonon recombinations via direct channels. Ion implanted samples of Si-NCs embedded in SiO2 were investigated by means of PL and TRPL at different photon fluxes. No significant dependence of the PL dynamics on emission energy was observed, indicating that decay lifetimes are not linked to different discrete sizes in the NCs size. I proposed a model of diffusion of excitation between neighbouring nanocrystals, with the saturation of acceptor nanocrystals providing a switching-off mechanism of the excitation transfer. The evolution of the PL emission with increasing silicon excess and annealing temperatures was found to be in agreement with a diffusion limited, Ostwald ripening process. For lower temperature treatments, a factor of 5 PL enhancement was observed and attributed to a thermal-activated carrier “recovery mechanism”, i.e. de-trapping of carriers from localized states within the band-gap to the states of Si-NCs. I propose that the PL contribution at around 2 eV observed in the same low temperature regime (10-100 K) arises from Si-NC-sensitized luminescent defects. At higher temperatures a monotonic quenching of the PL peak emission was observed and attributed to the enhancement of non-radiative recombination from defect states. Finally the Si-NC PL intensity in SiOxNy films was studied and found to be strongly dependent on the annealing conditions. RBS and XPS measurements showed that the composition of the thin films is significantly affected by oxygen contamination. Surprisingly, only the PL spectra of the as-deposited samples are well correlated with the evolving Si-NC size according to the quantum confinement (QC) model in which thin films containing larger clusters emit at lower energy. The formation of cracks after annealing the samples at temperatures from 400 °C to 1150 °C for 1 hour in forming gas, results in the suppression of the emission in the near-IR and in the arising of a defect-related emission peaking at higher energy.

The topside of a naval combatant must accommodate a plethora of highly sophisticated and computerised electromagnetic sensors in order for the ship to effectively fight, both to attack and defend. The electromagnetic sensors serve as the eyes and ears of the ship, and without them the ship would be very vulnerable. Since the topside of a naval ship has limited space, antennas must be sited close to each other (co-site). Many of the topside antennas are required to transmit and receive at similar bands of operating frequencies and, at times, they may be required to operate simultaneously. This gives rise to electromagnetic interference (EMI) which causes performance degradation of the equipment, blockage of communication channels, impairment of the on board sensors and even burning out of the inadequately protected equipment. One of the key challenges faced by ship designers at the concept phase of ship design is the need to effectively distribute topside electromagnetic (EM) sensors to avoid EMI. This is difficult to predict especially as in concept design the ship configuration and the EM systems characteristics will change as different arrangements are explored. The likely interference between shipboard antenna systems can be assessed by computational electromagnetic (CEM) tools which model the ship and its topside antenna systems. To tackle this problem, a general purpose and commercially available electromagnetic simulation package, Computer Simulation Technology (CST), has been employed as a rapid and cost effective method for handling the EMI problems in Early Stage Ship Design (ESSD). Using CST, the project initially modelled the recently in service Royal Navy Type 22 Batch II Frigate and its topside sensors. The EM sensor models on the Type 22 Batch II Frigate model were then simulated against each other in order to determine the EMI coupling between them. To benchmark these EMI coupling simulations, validation of a certain number of CST simulations has been carried out using two physical scale models of the Type 22 Batch II Frigate. After obtaining the required confidence in the reliability of the simulation package, a CST based approach for prediction of topside EMI/EMC has been developed. The approach has then been applied on an early design study for a Future Patrol Ship to predict the likely EM interactions between its topside antennas. Finally, in order to assess the likely free space antenna interference, MATLAB based codes have been developed. These will allow quick evaluations of the magnitude of antenna interference in free space. Thus, they would enable the project sponsor to determine antenna interactions between certain shipboard EM sensors without requiring to use a CEM tool.