This thesis reports an investigation into the effects of fuel molecular structure on the emission of exhaust PAHs from a tube reactor and from a diesel engine. The study was underpinned by the results of experiments conducted in the pyrolysis tube reactor aimed at understanding the formation processes of PAHs. The PAHs found in the diesel engine exhaust, both in gaseous state and on the soot particles, were also measured and analysed. The thesis focuses on the US EPA 16 priority PAHs formed from fossil diesel, C1 to C7 model hydrocarbon fuels, and blends of C7 binary and tertiary fuels. Particular attention was paid to the B2 subgroup of PAHs which are possible human carcinogens. Particulate and gas phase PAHs were generated and sampled from the exit of the reactor and the exhaust of the diesel engine. The PAHs from the particulate and gas phase samples were then extracted using an accelerated solvent extraction (ASE) system. The PAHs were analysed qualitatively and quantitatively using gas chromatography coupled with mass spectrometry (GC-MS). The experimental results obtained in the laminar flow, oxygen-free conditions of the reactor showed that, depending on the temperature at which a fuel is pyrolysed, the degree of unsaturation, isomerisation, aliphaticity, aromaticity and carbon number of the hydrocarbon fuels played an important role on the identity and concentration of PAHs formed. The identity of PAHs produced, and their concentration influenced the overall carcinogenic potential of the gaseous and particulate effluent. In the diesel engine, the total PAH concentrations of the fuels decreased with increase in ignition delay and proportions of premixed burn fractions while the influence of fuel composition on the exhaust PAHs was greater than the influence of the combustion characteristics such as ignition delay, heat release rate and premixed burnt fraction.

This thesis investigates the blast response of simple structural components - fully clamped beams and plates - underwater and in air. Experimental work by others have shown that, with increasing loading intensity, these components deform in one of either three modes: mode I (large inelastic deformation), mode II (tensile tearing) or mode III (transverse shear failure). The aim of this thesis is to develop theoretical and numerical models that can accurately predict these damage modes, taking into account the effects of fluid-structure interactions, for both impulsive and non-impulsive blast loadings A fully-clamped ductile beam model is proposed that is capable of capturing large elasto-plastic deformation, progressive damage and failure through detachment from its supports. Predictions by the model were validated against experimental data in the literature and with finite element models developed in this thesis. Parametric studies were also performed to elucidate the effects of loading duration on the mode of deformation and the conditions governing their transition. Numerical evidence on elimination of pulse-shape effects using an effective rectangular pulse loading (Youngdahl's approach) has been provided. The effects of uid-structure interaction (FSI) are investigated for fully-clamped, elasto-plastic beams in deep underwater explosions and intense air blast loadings. The main objective is to understand how the introduction of fully-clamped clamped supports alter existing well known results grounded on rigid, free-standing counterpart; and, to quantify how different modes of deformation affects the impulse and energy transmitted to the structure by the blast wave. Sensitivity analyses were carried out to elucidate the dependence of the results on the beam's aspect ratio and inertial mass. The deformation and failure of fully clamped rectangular plates subjected to blast loading are modelled numerically using finite element method. The numerical results are validated against experimental data. Deformation maps delineating the different deformation regimes for different combinations of blast impulse and aspect ratio are constructed for plates of equal mass. The effects of imposing a finite period, as opposed to a zero-period, pressure pulse upon the deformation mode and maximum deflection are discussed.

Spent coffee grounds (SCG) are the main residues of the coffee industry, and a potentially valuable source of lipids for sustainable biodiesel production. However, feedstock properties, such as the high SCG moisture content and the relatively high free fatty acid (FFA) content of recovered oil, can impact on the efficiency of the extraction and the quality of extracted oil and derived biodiesel, thus reducing the possible environmental benefits of producing biodiesel from this waste stream. Therefore, a better understanding of feedstock properties and processing steps is required to improve the efficiency of SCG valorization as a biodiesel feedstock and contribute to its future industrialization. This work presents experimental studies including feedstock characterization of SCG, laboratory and pilot plant scale solvent extraction experiments and utilization of mechanical pressing for processing of coffee residues. The solvent extraction experiments investigated effects of solvent type, SCG moisture content and particle size, SCG-to-solvent ratio, and the duration, temperature and pressure of the extraction process on oil extraction efficiency and composition. Transesterification was performed with SCG oil containing high FFA content, and the combustion of derived biodiesel was investigated in a compression-ignition engine. Instant SCG were found to possess higher lipid and FFA content than retail SCG. Solvent extraction experiments showed that longer durations, higher temperatures, low moisture presence and mixed size SCG particles generally improved extraction efficiency, while the impact of pressure depended on temperature. A correlation was observed between longer extraction durations and lower FFA content, while extraction temperature and solvent selection affected the oil composition. Pilot plant extraction showed reduced sensitivity to moisture, while mechanical pressing was efficient in removing a fraction of residual moisture. A two-step transesterification process achieved a biodiesel conversion yield of 86.7 % relative to initial oil weight. SCG biodiesel showed similar combustion and emissions characteristics to commercial soybean and rapeseed biodiesel.

A two-dimensional boundary element method (BEM) based on potential flow theory is adopted to study the wave interactions with multiple structures at resonance. Here resonance refers to the wave resonance which appears in the gaps between structures. The wave-structure interactions problems are simulated using a mixed Euler-Lagrangian scheme, with fully nonlinear boundary conditions applied on the instantaneous free surface and wetted body surface. The numerical scheme is verified through the simulations of wave interactions with a single body. Results show that both the free surface elevation and the hydrodynamic forces can be calculated accurately enough. The first primary study proposes a numerical approach to calculate the dominant natural frequencies in the gap based on the understanding of free liquid sloshing in a tank. The effectiveness of this approach is verified through the ‘response amplitude operator’ (RAO) analysis in terms of the gap free surface elevation. The natural frequencies are found for twin barges, with various gap widths and draughts. The effects of resonance on wave forces and elevations are also analysed. The second primary study considers the resonance induced by forced heave, sway and roll of body motion at various amplitudes. Particularly, second-order resonance, which is due to the sum or difference frequency, is found especially significant when the gap width over draught ratio is large. Second-order resonance can sometimes be as pronounced as, or even stronger than, classical first-order resonance. The third primary study concerns the wave resonance induced by nonlinear regular incident waves. For hydrodynamic interactions when the two bodies are both fixed, the free surface elevations are captured, particularly the standing wave trains formed in front of the upwave structure and sheltering effect behind the leeside structure. The nonlinearity associated with incident wave steepness is taken into consideration. Then second-order resonance in the gap caused by incident waves is studied. Finally, the wave resonance behaviour in the gap when the two bodies are freely floating under incident waves is analysed.

Fuel Cell (FC) buses have been developed as a long term zero emission solution for city transportation and they have now reached levels of maturity to supplement the coming London 2020 Ultra Low Emission Zone implementation. A critical review of previous research in this field has highlighted promising potential for FC technologies applied to bus applications and also identified the associated challenges. This research analysed the current FC bus industry and addressed the most recent trend of applying FCs with hybrid technologies for city buses. This research developed a scaled laboratory Fuel Cell and Supercapacitor hybrid drivetrain model for investigating the design and performance of a low emission propulsion systems for city bus applications in its dynamic environment. The laboratory system has been used to validate a computer model to ensure it is suitably representative of practical and full sized FC bus power systems. A novel hybrid control strategy was developed for a FC hybrid system and evaluated with actual bus driving cycles. The power balancing strategy between multiple power sources in the FC hybrid system has been explored and investigated. The key finding of this research is that hybridising the FC with an energy storage medium showed superior performance over FC only system. Additionally, existing FC hybrid buses generally have an over-sized FC on-board which significantly increases the capital cost. A series of steps have been identified to determine the required FC / energy storage degree of hybridisation. An optimised degree of hybridisation for FC hybrid bus can potentially improve the system performance, reduce the size of the FC on-board and propulsion system costs.

Microbubbles generated by microfluidic techniques have gained substantial interest in various fields such as food engineering, biosensors and the biomedical field. Recently, T-Junction geometries have been utilised for this purpose due to the exquisite control they offer over the processing parameters. However, this only relies on pressure driven flows; therefore bubble size reduction is limited, especially for very viscous solutions. The idea of combining microfluidics with electrohydrodynamics has recently been investigated using DC fields, however corona discharge was recorded at very high voltages with detrimental effects on the bubble size and stability. In order to overcome the aforementioned limitation, a novel set-up to superimpose an AC oscillation on a DC field is presented in this work with the aim of introducing additional parameters such as frequency, AC voltage and waveform type to further control bubble size, capitalising on well documented bubble resonance phenomena and properties. Firstly, the effect of applied AC voltage magnitude and the applied frequency were investigated. This was followed by investigating the effect of the mixing region and electric field strength on the microbubble diameter. A capillary embedded T-junction microfluidic device fitted with a stainless steel capillary was utilised for microbubble formation. A numerical model of the T-Junction was developed using a computational fluid dynamics-based multiphysics technique, combining the solution of transport equations for mass and momentum (Navier-Stokes Equations), a Volume of Fluid algorithm for tracking the gas-liquid interfaces, and a Maxwell Equations solver, all in a coupled manner. Simulation results were attained for the formation of the microbubbles with particular focus on the flow fields along the detachment of the emerging bubble. Experimental results indicated that frequencies between 2-10 kHz have a pronounced effect on the bubble size, whereas elevated AC voltages of 3-4 〖kV〗_(P-P) promoted bubble elongation and growth. It was observed that reducing the mixing region gap to 100 μm facilitated the formation of smaller bubbles due to the reduction of surface area, which increases the shear stresses experienced at the junction. Reducing the tip-to-collector distance causes a further reduction in the bubble size due to an increase in the electric field strength. Computational simulations suggest that there is a uniform velocity field distribution along the bubble upon application of a superimposed field. Microbubble detachment is facilitated by the recirculation of the dispersed phase. A decrease in velocity was observed upstream as the gas column occupies the junction suggesting the build-up in pressure, which corresponds to the widely reported ‘squeezing regime’ before the emerging bubble breaks off from the main stream. The novel set-up described in this work provides a viable processing methodology for preparing microbubbles that offers superior control and precision. In conjunction with optimised processing parameters, microbubbles of specific sizes can be generated to suit specific industrial applications.

Nanoparticles are being widely investigated for food purposes and are beginning to see application in food production. There are numerous techniques to produce such nanoparticles, including emulsion-based techniques and spray drying, each with their advantages. Electrohydrodynamic spraying provides an alternative technique for preparing food component loaded particles in the nano-scale with a good control of important particle characteristics, such as size. In this work, Electrohydrodynamic spraying was used to investigate its potential for producing nanoparticles intended for nanoencapsulation of low solubility food components. Different processing parameters including flow rate, solute concentration, food component loading and their influence on nanoparticle characteristics and food component release were studied using food flavour as a model food component and ethylcellulose and stearic acid as a carrier materials. This work, EHD spraying was used to investigate its potential for producing nanoparticles intended for food delivery. Different processing parameters including flow rate, solution concentration, food component loading and their effect on nanoparticle characteristics and food flavour release rate were investigated. Polymeric and lipid nanoparticles were studied in detail with respect to nanoparticle characteristics and food component release kinetics and additional studies were performed for nanoparticles prepared with ethylcellulose as a model hydrophilic polymer to form the polymeric core; and stearic acid as a model lipid to form the lipid monolayer. EHD sprayed nanooparticles were prepared with diameters between 10-100 nm and a near-monodisperse size distribution was obtained in most cases. The flavour release rates were found that the release rate was a function of both the nanoparticle size and structure, and hence of the processing conditions. EHD sprayed nanoparticles generally had a slower flavour release rate compared with conventional techniques. The results indicated that EHD spraying is an attractive nanotechnology for generating food component loaded nanoparticles that can be tailored towards an intended food delivery application. Compared with the conventional techniques it provides better control of nanoparticle and demonstrated its suitability for producing nanoparticle formulations in which the food component and is released rate in a sustained manner to potentially improve bioavailability of low solubility food component such as flavour.

Prostate cancer remains a major healthcare issue. Limitations in current diagnosis and treatment monitoring techniques imply that there is still a need for improvements. The efficacy of prostate cancer diagnosis is still low. High intensity focused ultrasound ablation is an emerging treatment modality, which enables the noninvasive ablation of pathogenic tissue. Successful focal ablation treatment of prostate cancer is critically dependent on accurate diagnostic means and would be greatly benefited by a monitoring system. While magnetic resonance imaging remains the gold standard for prostate imaging, its wider implementation remains prohibitively expensive. Conventional ultrasound is currently limited to guiding biopsy. Elastography techniques are emerging as a promising imaging method, as cancer nodules are usually stiffer than adjacent healthy tissue, and even stiffer in the case of thermally ablated tissue. In this thesis, a novel transurethral elastography approach is proposed for the diagnosis of prostate cancer and its focal ablation monitoring, based on the transmission and detection of shear waves through the urethral wall. A viscoelastic wave propagation model is developed, using a finite difference time domain technique and based on a Kelvin-Voigt fractional derivative constitutive law. Validation of the model is achieved by high-speed camera tests carried out on translucent tissue-mimicking media. A Reverse Time Migration and a Genetic Algorithm techniques are proposed for reconstructing the parameters of the stiff lesion. A comparative study of the two techniques is presented. The Reverse Time Migration method finds the stiff lesion area in short computational time. The Genetic Algorithm provides full reconstruction of the location, size and stiffness of the lesion, however the computation time is much longer. A combination of both techniques achieves improved results by combining the speed of the Reverse Time Migration and the full reconstruction capacity of the Genetic Algorithm. Preliminary results support the feasibility of the method and encourage further investigation.

The biggest source of energy loss in shipping is found in the propulsion system. This study focuses on analysing, and working with, the concept of heat management for waste heat energy from the exhaust gas and scavenge air. Using waste heat recovery systems (WHRS) to make shipping more efficient represent a good area of opportunity. On board ships, a water-based Rankine cycle (RC) is typically installed; this has the task of providing steam and power. This work explores alternative waste heat technologies to assess the development and suitability, but also to find better solutions to the traditional RC. Different models coupled with advance optimisation processes were created to understand the marine WHRS. The results show that WHRS are sensitive to environmental and operational factors which must be considered at design stage. While water offers the possibility of producing both steam and power; organic Rankine cycles (ORC) produce larger power outputs at temperatures between 90˚C and 230˚C which translate to lower CO2 emissions. Organic WHRS will play an important role in the future as regulations push for tighter emission controls, and waste energy availability for power production reduces due to an increase in prime mover efficiency and waste heat utilisation for other processes (e.g. ballast treatment). The ORC technology can be applied to any kind of vessel type and size, keeping in mind that the ORC benefits depend on the waste heat temperature and availability, the ship's heat requirements and operational profile. It is also then important to bear in mind some of the drawbacks, such as larger mass flow rates and flammability of some of the organic fluids studied which will introduce additional safety equipment and costs.

This thesis is concerned with the formation of particulate matter, a topic of scientific and practical importance due to the toxicity of particulate emissions from automotive and other combustion sources. At present, fuels are predominantly derived from fossil sources, but as production technology improves, biofuels and synthetic fuels are expected to emerge as scalable long-term sources of liquid fuels. Efforts are being made to ensure that this next-generation of fuels is cleaner burning than the last. In order to inform the production and processing of cleaner burning fuels, more needs to be known about how molecular structure influences the formation of pollutant emissions. This thesis presents research that has been carried out in order to better understand the role of functional group chemistry on the conversion of carbon atoms in the fuel to the particulate matter (PM). In particular, the propensity of individual molecules or carbon atoms within molecules to form PM is reported quantitatively. To this end, a technique using carbon-13 (13C) labelled fuel molecules was used so to track the labelled carbon atoms in the fuel to PM. The technique required only very low levels of 13C enrichment, and isotope ratio mass spectrometry equipment (IRMS) was used as a means of 13C detection. Samples of particulate matter were formed using a tube reactor, and also in a compression ignition diesel engine. The tube reactor was designed and commissioned in order to study the pyrolysis of various fuel molecules under well-controlled, homogenous conditions. The contribution to PM of a number of molecules containing various functional groups was assessed, including: alcohols, esters, aromatics, double bonded carbon atoms, a ketone, and a carboxylic acid. Tests were conducted using single-component fuels, and blended in a binary mixture with n−heptane. The results show that the contribution of carbon atoms within molecules to PM, is not equal, but depends on the local molecular structure. For example, oxygenated molecules significantly reduced the contribution to PM of the carbon atoms directly attached to oxygen. The thesis presents one of only a handful of investigations that have been published on the conversion of specific carbon atoms of various molecules to soot and particulate. It advances the field of study by providing data for validation, at the sub-molecular level, for chemical kinetic models of soot formation, and advances fundamental understanding of how fuels convert to soot and particulates.