This thesis presents an experimental study on the influence of the molecular structure of potential biofuels on combustion in a compression ignition engine. The molecular structure of a fuel is amongst the most fundamental parameter controlling its physical and chemical characteristics, and is thus critical to the combustion process within an engine. The approach employed in this work was to study the combustion of several individual molecules in a series of experiments whilst varying a single feature of the molecular structure at a time. This yielded information about how a particular structural feature of a molecule affects the combustion process. During the course of this project, a special fuel injection system was developed, which allowed the injection of small fuel samples into the engine at high pressure. This allowed tests to be carried out on purposely synthesised fuel samples that were only obtainable in small quantities. Detailed studies on the combustion of fatty alkyl esters (commonly termed biodiesel), acetals, ethers and alcohols were conducted. The combustion chamber pressure of the engine, the energy release of combustion, the engine efficiency, the exhaust gas composition and the emission of particulate matter were measured and analysed. It was observed that in the diffusion combustion of biofuels, the emission of nitrogen oxides from the engine depend primarily on the ignition delay of the fuel, which governs the combustion stoichiometry and peak cylinder pressures and temperatures within the combustion chamber, and secondly on the adiabatic flame temperature of the biofuels. It was found that the number of double bonds present in biofuel molecules correlated with the amount of particulate mass emitted from the engine. It was further observed that oxygenated biofuels such as fatty acid alkyl esters, acetals, ethers and alcohols produced much lower levels of particulate mass from their combustion than petroleum-derived diesel fuel. The emission of particulates depended on the fuel oxygen content, as well as on the boiling point of the fuel. Combustion experiments conducted in homogeneous charge compression ignition combustion demonstrated that ethers of low molecular mass could be amongst the most-suited liquid fuel molecules for this type of combustion method.

Electrospun fibrous structures are providing solutions to many problems such as biomedical, environmental and energy issues. Despite the exponential growth of interest, limitations still exist in optimising the process. In this dissertation, new advances and applications of electrohydrohynamic spinning and processing were explored to meet current demands and challenges encountered in the processing of electrospun biomaterials. Good electrospinnability refers to consistent fibre production with minimal ‘bead-onstring’ formation during electrospinning. One challenge is the solvent selection process, which significantly affects electrospinnability and the as-spun fibre morphology. This is mostly done by trial-and-error and the effect of solubility in solution-electrospinning have been neglected or taken for granted. The general concept is that a suitable solvent should have solubility parameters close to that of the polymer. However, literature has abundantly suggested a solvent may dissolve a polymer well, but whether the resultant solution is electrospinnable cannot be guaranteed. A novel method to systematically select and mix solvents for solution-electrospinning was developed in the first section of this work. The results were combined on the Teas graph to create a new “spinnability–solubility map”, allowing topographic analysis on the potential electrospinnability of a solvent or mixed solvents. Marginal solvents were shown to allow lower critical minimum solution concentrations – a parameter required for electrospinning to occur. Since fibre diameter depends on concentration, a lower critical concentration translates to finer fibres. Key properties of a solvent promoting elongation during electrospinning were also identified and the significant effects of the spinning environment on the polymer solution and the final fibre morphology were discussed. Unique fibrous relics were electrospun from poor solvents at elevated temperatures. Collaborating with the food industry, the application of electrohydrodynamic spinning and processing in food sciences was investigated, using edible biomaterials such as ethyl cellulose and chocolate. The potential to control fibre aspect ratio and directly electrospin short microfibres was studied.

B cells perform several immunological functions, including the production of antibodies, release of cytokines and presentation peptide or lipid-antigen respectively to T or invariant natural killer T (iNKT) cells. Patients with systemic lupus erythematous (SLE) display dysregulated B cell responses and reduced peripheral iNKT cell frequencies. The significance of these defects and how they relate to SLE pathogenesis remains elusive. I examined the role of B cells in the maintenance of iNKT cells in healthy individuals and SLE patients. I report that B cells are essential for iNKT cell expansion and activation in healthy individuals but fail to exert a similar effect in SLE patients. Defective B cell-mediated iNKT cell stimulation in SLE was associated with rapid internalization and reduced CD1d surface expression on SLE B cells compared to healthy B cells, a defect that could be recapitulated in healthy B cells after simultaneous stimulation with interferon-alpha and anti-immunoglobulin. Strikingly, iNKT cell number and function were restored in SLE patients responding to anti-CD20 treatment upon normalization of CD1d levels exclusively in repopulated immature B cells. Furthermore, I demonstrated that B cells are essential for the maintenance of iNKT cells during experimental arthritis. Whereas alphaGalCer treatment ameliorated arthritis in wild-type mice, alphaGalCer treatment had no effect in arthritic B cell deficient mice. Amelioration of arthritis was paralleled with the proliferation of iNKT cells and a skew towards a Th2-like cytokine milieu. Further, I show that iNKT cells are necessary for the differentiation of regulatory B cells (Bregs). Whereas WT Bregs were capable of suppressing arthritis, Bregs isolated from iNKT cell deficient mice failed to exert the same effect despite producing IL-10. Impaired Breg suppression was associated with a potential trafficking defect. These results suggest that B cell-iNKT cell cross-talk is pivotal for the maintenance of tolerance in both human and mice.

The osteoconductive and osseointegrative nature of hydroxyapatite (HA) has made it a popular coating material for orthopaedic implants. HA coated metallic prostheses, which combine the osteoconductivity of HA and high strength of metallic alloys, have been increasingly favoured by surgeons for the ageing population. Topography is known to provide a powerful set of signals for cells to improve initial bonding of implant to the body. This project focuses on controlling the chemistry and topography of the implant surface in order to improve the initial attachment of bone cells to the implant surface by the creation of ordered topographical surfaces. Silicon-substitute hydroxyapatite (SiHA) was chosen as the material of choice for patterning due to its enhanced biological responses. Template-assisted electrohydrodynamic atomisation (TAEA) is an electrically driven jet-based deposition technique capable of producing patterns of regular and uniform nature on a substrate, which has been developed using the principles of electrohydrodynamic atomisation. The deposition of nanoSiHA patterns on titanium surface has been optimised by the set up configuration of the electrode and processing parameters, namely, flow rate, applied voltage, distance from substrate. It was found that decreasing the flow rate was able to increase the coverage of the template and reached to 80% of template space. The optimum distance between the needle and substrate was 50 mm, which allowed solvent evaporation to occur and resulted in clear, well-defined structures. The thickness of patterns was increased with spraying time, and the deposition rate was higher on the track patterns in comparison with that of pillar patterns. The template surface finishing also affected the resultant pattern with ridged edging reducing the surface coverage. To further understand how cells interact with surface topography and quantify the response, in vitro studies were carried out on human osteoblasts (HOB), murine osteoblasts (MC3T3-E1) and dental pulp stem (DPS) cells. In vitro cellular studies found that both pillar and track nanoSiHA patterns were able to encourage the attachment and growth of osteoblast cells, providing anchor point for cell attachment. Actin and cell body preferentially aligned parallel to the track pattern, linking the actin orientation to the entire cell placement. HOB cells were found to respond to pattern topography by stretching nearly three times of their original size when the distance between the tracks increased. AlamarBlue™ tests showed the proliferation rates of HOB cells increased when cultured on patterned substrates than those on nanoSiHA coating. An increase of alkaline phosphatase production, an osteoblast differentiation marker, was found on the HOB cells cultured on patterned surfaces, as well as on the pillar and track patterns when the thickness increased from 1 to 5 µm. The expression of ALPL, a gene involved in mineralisation, was higher for the HOB cells cultured on pillar patterns. Runx-2 expression and collagen production of HOB cells was also increased on track patterns when compared to pillar patterned and nanoSiHA coated Ti. In order to understand the influence of pattern topography on the mechanics of the cell, shear force modulation force microscopy was used to determine the elasticity of osteoblasts (MC3T3-E1) cell and dental pulp stem cell (DPS). This study found that increasing the thickness of the pattern, in both track and pillar, lead to an increased modulus of the MC3T3-E1 cells. MC3T3-E1 cells were found to have significantly higher moduli than DPS cells, regardless of location on the substrate or thickness of the pattern. Therefore, the study has provided an insight for future design of implant surfaces to control and guide cellular responses, while TAEA patterning provides a controllable technique to provide topography to medical implants.

The thesis describes an experimental investigation of the flow in a shaken bioreactor of cylindrical geometry with a flat bottom. Several reactor designs can be distinguished that attain mixing in different ways: oscillatory flow mixers (OFM), static mixers, stirred vessels and shaken bioreactors. Shaken bioreactors are often small-scale mixers (microwells) employed in the early stage of bioprocess development (i.e. microbial fermentation, bioconversion and product recovery techniques), before the developed process is implemented in a large-scale industrial stirred tank. However, despite their wide use, little is known about the fluid mechanics of these systems. In the present study Particle Image Velocimetry (PIV) measurements are carried out to determine the variation of the flow dynamics in a cylindrical shaken geometry for different operating conditions such as medium height, shaking frequency, orbital shaking diameter, cylinder inner diameter and fluid viscosity. In the first part phase-resolved measurements are carried out with PIV to provide a thorough characterisation of the flow and mixing dynamics occurring in a cylindrical shaken bioreactor for a fluid of low viscosity (i.e. water). From this analysis a flow scaling law based on the Froude number, Fr, is identified, which correlates the shape and inclination of the free surface to the occurrence of a flow transition. More specifically it was found that at low Fr the mean flow ischaracterised by a toroidal vortex with its axis aligned along the azimuthal direction, while after flow transition the free surface exhibits a phase lag and a vortical structure with a vertical axis that precesses around the cylinder axis is present. In the second part of the thesis flow characteristics, such as the interfacial area, circulation time, vortex size and location, kinetic energy and viscous dissipation rate of kinetic energy for a fluid of low viscosity are analysed in depth. The free surface interfacial area was directly measured by image analysis to assess oxygen transfer potential and was compared to an analytical solution valid for low Fr. The non-dimensional time and length scales of the vortical structures occurring in the cylindrical bioreactor were determined to provide an insight into the mixing dynamics, while a Reynolds decomposition analysis of the kinetic energy was carried out to assess the onset of a laminar-turbulent flow transition with increasing Fr. Direct measurements of the viscous dissipation rate of the kinetic energy, ǫ, were obtained across the tank to help assess micro-mixing and identify regions in the bioreactor experiencing higher levels of viscous stresses that can potentially affect cell growth. In the third part of the thesis the flow obtained with Newtonian fluids of higher viscosity is investigated and the flow scaling law determined for water is extended to a broader range of viscosity. A flow transition map based on Fr and Re is identified and four main regions characterised by different mean flow dynamics are shown. The turbulent kinetic energy levels and shear rate magnitudes are assessed for different combinations of Fr and Re. The results offer valuable new information for the design of mixing processes and crucial data to validate computational fluid dynamics simulations of cylindrical shaken bioreactors.

With the advent of fast digital computers, the mathematical methods used for modelling, assessing and controlling large complex engineering systems are being strained to the utmost. For example, space and aircraft research, nuclear engineering and power systems now use advanced computer techniques for their design, construction and operation. Areas of analysis which were confined mainly to the realm of pure mathemeticians only two decades ago are now being utilised by engineers. Probability spaces, eigen-:" vectors and state-spaces are some of the many concepts employed in such analyses. Multivariable engineering problems are now often expressed, mathematically, in terms of abstract spaces and subspaces.

Environmental legislation has pressured fuel and automotive industries to alter their technologies for compliance. Changes made to a fuel to achieve compliance can push specific fuel qualities, such as lubricity, to a level away from that required by the automotive engine, creating a fuel/engine requirement gap. Use of fuel additives offer an economical route to bridge the gap, however, their effects on stages of the diesel combustion process is not yet fully understood owing to measurement difficulty. Greater understanding of additive effects requires precision control of operational test conditions with the ability to apply high fidelity measurement techniques. To enable this a high pressure, high temperature optical facility was developed which allows for the acquisition of large data-sets of spray parameters for a more accurate study of the effect fuel additives have on the diesel combustion process. For facility commissioning, tests on diesel fuel sprays into elevated pressure and temperature environments were carried out, with data obtained from high speed backlight illuminated imaging. Macroscopic spray measurements such as penetration length and spreading areas of the sprays were performed, so that effects of ambient pressure and temperature on these parameters could be identified and discussed. From the tests, injector opening times were a main cause of variability in the observed characteristics which were taken into account in the study. Following commissioning, a systematic study was carried out using combustion improving and detergent additives for the first time, where low and high concentrations were tested. Similar behaviours in penetration lengths for each of the additives tested were seen, however no statistical confidence could be applied to the observation as the penetration lengths of the additised fuels for the tested back pressures, since these values did not shift from the base fuels' measured data by a magnitude greater than the experimental error. Penetrating spray area data variance was large, and changes due to additives were unidentifiable. To further clarify, laser droplet sizing was employed at atmospheric conditions to identify whether additives cause changes in microscopic measurements of the spray. The tests showed no change in the droplets' Sauter Mean Diameters (SMD) were observed due to additives. The study carried out clearly indicates that the additives added in the tested concentrations did not change the statistically determined transient parameters of diesel sprays.

Alongside deploying weapons and sensors what makes a warship distinct is survivability, being the measure that enables a warship to survive in a militarily hostile environment. The rising cost of warship procurement, coupled with declining defence budgets, has led to a reduction in the number of ships in most western navies. Moreover, cost cutting is often aimed at aspects which may be difficult to quantify, such as survivability, and can lead to potentially inadequate designs. Innovation in both the design process and the design of individual ships is, therefore, necessary, especially at the crucial early design stages. Computer technology can be utilised to exploit architecturally orientated preliminary design approaches which can address innovation early in the ship design process and in issues such as survivability. A number of survivability assessment tools currently exist; however, most fail to integrate all survivability constituents (i.e. susceptibility, vulnerability and recoverability), in that they are unable to balance between the component aspects of survivability. Some are qualitative, therefore less than ideal for requirement specification, others are aimed towards the detailed design stages where implementing changes is heavily constrained or even impractical. Since a ship’s survivability is dependent on layout, the approach adopted in this research takes advantage of an architecturally orientated ship design approach applicable to early stage design. Such a method is proposed and demonstrated on five combatant (including a trimaran configuration) and two auxiliary ship design studies. The proposed method combines various tools used by UCL and the UK Ministry of Defence, as well as a new approach for recoverability assessment and, therefore, tackles difficulties currently associated with the latter (e.g.: lack of data, human performance and time dependence) by using weighted performance measures. An overall approach for survivability assessment has been applied across the range of designs produced and conclusions drawn on their relative merits for overall survivability. The approach and implications of the integration of survivability assessment in the preliminary ship design stages, as well as the identification of major survivability design drivers, are discussed. Through the identification of problematic topics, areas for further research are suggested. It is envisaged that this research will assist in developing the design process of what are, according to Captain C. Graham, USN, “the most complex, diverse and highly integrated of any engineering systems” produced today on a regular basis.

Over the last two decades there has been increasing interest among naval ship designers to adopt design style elements, standards and practices from commercial shipbuilding. Notable among this is a transition from the highly complex structural styles prevalent during the Cold War to simpler, more readily produced structure. It is generally presumed that this will reduce ship procurement costs, but may also have an effect on operational characteristics, including vulnerability to hostile action. Many naval weapon systems employ shock from underwater explosions as their damaging mechanism. In severe cases shock can cause catastrophic loss of watertight integrity, but in even moderate cases of shock the resulting acceleration environment inside the ship can damage or destroy vital equipment. The research presented in this thesis attempted to quantify the effect of adopting simpler structural styles upon this damaging acceleration environment. A number of different frigate structural models were specifically designed, using different structural styles but to meet the same design strength criteria. These models were subjected to simulated underwater explosions using Fluid Structure Interaction Finite Element Analysis techniques and the resulting motions at likely equipment mounting points computed. Results are presented in the form of comparative shock response spectra and also compared against existing shock response prediction techniques. This thesis concludes that the adoption of certain simplified structural styles in warships can lead to significantly elevated shock response motions, compared to those expected from a ship with a more typical naval structural style. In particular, the adoption of reducing the number of stiffeners, or adopting lower cost stiffener profiles, may result in motions increased by a degree significant enough that they should be taken into account when specifying the shock tolerance or mounting arrangements for on-board equipment.

Conventional methods of preparing biomedical-based porous structures from biomacromolecules offer great potential in a number of application areas within the healthcare sector. The key requirements for these fabrication methods are enhanced control over structural features, scale-up potential, robustness, convenience, non-toxicity, ease of use and economic viability. Electrospun fibrous structures have drawn considerable attention particularly for tissue engineering applications. In this thesis, a technique based on the same principles; co-axial electrohydrodynamic flow processing, commonly used to generate encapsulated macromolecular structures for applications in drug delivery and medicine, was adapted for the generation of microbubbles as templates for porous structures. The applicability of the electrohydrodynamic bubbling processing and the generated bubble structures were investigated for their use in the field of tissue engineering. Microbubbles coated with proteins bovine serum albumin (model protein) and/or silk fibroin (SF) were generated and the effect of processing parameters (applied voltage and flow rate) on the bubble size, uniformity and stability were investigated. Porous structures were fabricated by the layer by layer deposition of microbubbles and investigated by their pore morphologies and internal structures with varying concentrations, blends and with the use of a cross-linking agent. The formed structures were analysed and compared in terms of their chemical interactions, thermal, mechanical and degradation properties. In addition, cell studies using mouse fibroblast cell lines were used to investigate the effects of processing parameters, ratio of SF to bovine serum albumin and the use of cross-linking on cell viability and proliferation. Furthermore, silk and hydroxyapatite composite scaffolds were formed to demonstrate their possible use in a specific application: bone tissue engineering. The applicability of microbubbled scaffolds was investigated using osteoblast bone cells. The results demonstrated the fabrication of porous structures with controllable pore size with a novel facet of electrohydrodynamic bubbling process.