Dielectric elastomer (DE) actuators are low power, soft, electrically actuated 'artificial muscles' that offer the potential to emulate many of the actuators found in nature; including biological pumps such as cilia and peristaltic tubes. However, a low power pump based on tubular DE actuators is yet to have been produced. The literature describes several approaches towards achieving pumps based on soft actuators; however the majority of these are diaphragm pumps which are susceptible to bubbles and can be difficult to clean, leading to clogging and contamination. Meanwhile, the tubular pumps developed in the literature based on different types of actuator have suffered from low actuation speeds and limited output flowrates.

To perform long missions, small unmanned air vehicles (UAVs) need efficient, lightweight propulsion systems that can operate on energy dense fuels. Gas turbines offer better reliability, life, fuel flexibility, noise, and vibration than internal combustion (IC) engines, but they are uncompetitive due to fuel efficiencies around 6%. At this scale, conventional efficiency improvement approaches such as high pressure ratios and cooled metal turbines are impractical. Ceramic turbines could withstand high temperatures without cooling, but their life and reliability have been inadequate. This work explores the hypothesis that a low pressure ratio, highly recuperated ceramic engine design could overcome these problems. First, an accepted water vapor erosion model is extended to correctly account for the effects of recuperation, fuel type, and atmospheric humidity on the burned gas water vapor content. The results show that ceramic turbines without environmental barrier coatings can last 10,000 hours or more in highly recuperated engines, even at temperatures exceeding 1200[degrees]C. Next, a new design for a small recuperated ceramic engine is developed and analyzed, in which blade speeds are limited to 270 m/s – about half the typical value. A CARES slow crack growth analysis indicates this will lead to vastly improved life and reliability. The literature on foreign object damage and production costs suggests likely improvements in those areas, as well. Finally, an original ceramic recuperator is developed to fit the proposed engine design. Tradeoffs between fabrication constraints, weight, volume, effectiveness, pressure losses, and other considerations are explored through analysis, simulations, and experiments. For one design, these predict a thermal effectiveness in the 84-87% range at a specific weight of 44 grams per gram/second of airflow, surpassing the current state of the art by a factor of 1.25-1.5. A prototype designed for 1100[degrees]C operation was tested at 675[degrees]C exhaust inlet temperature. It did not crack or leak, and the performance roughly matched analytical predictions. With a heat exchanger of this type, a small, low pressure ratio turboshaft engine could achieve an efficiency of 23%, making it highly competitive with other state of the art propulsion systems by almost all performance metrics. In sum, this work contributes a novel ceramic recuperator that can enable low pressure ratio gas turbines to achieve high fuel efficiencies, and provides a significant extension of ceramic turbine life and reliability theory that shows such engines could achieve long service lives.

Articular Cartilage (AC) is a soft tissue covering the articulating surface of human and animal joints. The tissue has remarkable and highly complex mechanical and wear properties allowing the joint to undergo complex kinematics and function correctly for several decades. However, trauma and degenerative joint diseases such as osteoarthritis (OA) can cause damage and excessive wear of the tissue and due to its limited regenerative capabilities, can severely compromise joint movement and impair the quality of life. OA is the most common type of degenerative joint disease and the primary cause of joint replacement surgery leading to high associated healthcare costs. Although the exact cause of this pathology remains unknown, it is thought to be mechanically induced via excessive and abnormal stresses and strains in AC which cause altered biochemical properties and a gradual decrease in the mechanical quality of the tissue. There is currently no available cure for OA and the disease is currently being diagnosed only via imaging techniques which are based upon morphological changes of the tissue, when the pathology is already in its advanced stages and has caused irreversible changes to the AC. In this respect, one of the greatest challenges to now remains the early diagnosis of OA, potentially by assessing biochemical and mechanical changes, allowing early treatments and prevention of disability thus improving the patient’s life. Hence, there is a need to apply fundamental engineering principles to the medical world in order to shed light on the pathogenesis and progression of OA. Furthermore, the need for artificial substitutes of AC has called for a deep understanding of the mechanical behaviour of the tissue in order to design and mimic the response of the real tissue in the most accurate manner. In this research a combination of numerical (finite element) and experimental techniques involving mechanical and tribological tests were used to fully characterise the mechanical behaviour of the tissue. Selective degradation of the AC constituents was then induced to simulate OA (OA-like AC) and the effect of different stages of degradation on the mechanical and tribological response as well as the wear properties of the tissue was investigated. The mechanical properties of osteoarthritic AC were then evaluated and compared to the OA-like AC in order to correlate similarities in the variations to the structure and the mechanical response as a result of degradation. Quantifying the mechanical response of the tissue at different stages of OA and different levels of degradation was done to ensure both a thorough understanding of the effect of the pathology’s progression on AC as well as to provide a potential map of mechanical quality and degradation, contributing to the potential future diagnosis of OA via mechanical parameters rather than morphological alone. Having investigated structural and mechanical variation in early OA, a promising solution to treat localised early OA and AC defects was also investigated as part of this research. In particular, novel micro fibrous tissue engineered scaffolds have been mechanically and tribologically assessed and compared to AC demonstrating the strong potential of matrix-assisted autologous chondrocyte implantation (MACI). Finally, the numerical models developed to characterise the AC using numerical – experimental methods, namely advanced biphasic models incorporating fine material descriptions such as intrinsic viscoelasticity as well as transverse isotropy, were applied to a patient specific 3D menisectomised tibio-femoral joint contact model in order to demonstrate the implications that the implementation of different AC models have for the prediction of the joint response to repeated walking cycles. The results obtained from the models were then used to predict the most likely location for the origin of mechanical damage and OA.

Adhesive bonding has emerged as an appealing technique to join carbon fibre-reinforced plastics (CFRP) to other structural parts. The advantages that adhesive bonding offers include an even stress distribution, weight saving and superior fatigue resistance when compared to more traditional methods of joining. However, despite these advantages, the uncertainties regarding their durability have confined them largely to use in secondary structures. In the present work, a fracture mechanics methodology has been followed using both experimental and FE methods to predict the service-life of a CFRP-titanium alloy adhesive joint intended for use in a turbofan application. The methodology utilises the concept of the cohesive zone model to evaluate the performance of a simplified but representative structure, i.e. a Ti-to-CFRP tapered double-lap joint. The adhesive bondline was modelled by a layer of newly-developed cohesive elements, the kinematics and topology of which have been optimised to improve the mixed-mode behaviour and reduce the mesh-dependency. Their damage evolution has been enhanced to incorporate high-cycle fatigue degradation. Additionally, a simplified version of this formulation, specifically designed to predict only the fatigue threshold, has also been developed. To determine the various input parameters required for the models, a series of fracture mechanics specimens manufactured with a commercial film adhesive were tested quasi-statically in various modes and in mode I fatigue. Various data reduction schemes were evaluated and a version of corrected beam theory employing an effective crack length approach was found to be optimum for all tests. The fracture energies determined in the various modes were partitioned according to the theories proposed by Williams (Global) and by Davidson’s crack tip element singular field (CTE/SF) and non singular field (CTE-NSF) theories. The CTE-NSF partitioning strategy was found to be most suitable for the system under investigation. Fatigue tests were performed under wet and dry conditions, to investigate the effect of moisture on the joint performance. The fatigue results were fitted to a modified version of the Paris law and the required fatigue parameters were determined. The response of the various test specimens was simulated using the numerical scheme and good agreement with the experimental results was obtained. Significantly, the results obtained with a quadratic version of the cohesive element have been found to be independent of the element size, at least with respect to the global response. Finally, both the quasi-static and fatigue responses of the double lap joints were simulated using the cohesive element formulation and conservative predictions of the service life were obtained, in accordance with expectation, as only mode I fatigue data (lower bound Gc ) was inputted into the model.

Biofuels, are attracting great interest as an alternative to fossil fuels. n-Butanol has surfaced as a potential biofuel, mainly because it does not suffer from the drawbacks, that the current most widely used biofuel, ethanol, does. In this work a theoretical and experimental investigation of n-butanol combustion is performed, while a baseline investigation of fundamental combustion properties of methane is carried out. The computational work involves the investigation of the various reaction pathways for hydrogen abstraction from n-butanol by CH3 using quantum chemical calculations. The relative significance of hydrogen abstraction reactions from the specific carbon sites of n-butanol are compared with each other and with similar radical reactions initiated by OH and HO2 radicals. While the most stable structural conformer of n-butanol is expected to be the most abundant during the combustion process, the temperatures at which fuel burns, allow higher energy conformers to be accessible. A key feature of this work is to assess if any of these low lying, but not minimum energy conformers, have transition state barriers or product radicals lower in energy than those found for the most stable conformer. Based on the above ab initio calculations the rate constants and product branching ratios for hydrogen abstraction by CH3 from the different sites of n-butanol are computed using three available kinetic programs, namely CanTherm, MultiWell and Variflex, providing accurate data for future detail chemistry mechanism of n-butanol. An exhaustive comparison of the aforementioned kinetic programs is also carried out. The experimental work involves the development of a counterflow burner, specifically designed for the study of flames of pre-vaporised liquid fuels, which provides a very useful idealisation of the combustion process of a real combustor. While the present research focuses on n-butanol combustion, measurements of methane flames are used as a starting point of the experimental work, providing a basic understanding of the combustion aspects under investigation. The natural chemiluminescence of OH* and CH* radicals emitted in preheated premixed methane and n-butanol flames are measured using intensified high speed photography, in order to evaluate the dependence of the chemiluminescent ratio OH*/CH* on equivalence ratio and strain rate for various preheat gas temperatures and the performance of available chemical kinetics mechanisms. The effect of an electric field in counterflow flames is also investigated focusing on two principal themes: the effect of strain rate on the saturation current and the effect of high potential fields on the OH* chemiluminescence.

This thesis has explored the use of transient thermography for the detection of disbonds of minimum diameter 2mm located in a thermal barrier coating (TBC) system whose surface may be unpainted. The technique, the type/size of the defect and also the condition of the TBC system for the inspection has been specified by Alstom Power Switzerland, the sponsor of the EngD project. As for other Non Destructive Testing (NDT) techniques, reference test specimens are required for calibration, but unfortunately, real disbonds are very difficult to use because it is difficult to control their size, and larger ones tend to spall. Flat bottomed holes are commonly used, but these over-estimate the thermal contrast obtained for a defect of a given diameter. The thesis quantifies the differences in thermal response using finite element analysis validated by experiments, and proposes a form of artificial disbond that gives a better representation of the thermal responses seen with real defects. Real disbonds tend to have a non-uniform gap between the disbonded surfaces across the defect, and the effect of this on the thermal response is evaluated using finite element simulations. It is shown that the effect can be compensated for by adjusting the diameter of the calibration defect compared to the real defect. Surfaces of inspected specimens are usually covered by a black, energy absorbing paint before the transient thermography test is carried out. Unfortunately, this practice is not acceptable to some turbine blade manufacturers (including the project sponsor) since thermal barrier coatings are porous so the paint is difficult to remove. Unpainted TBC surfaces have very low emissivity, and after period of service their colour changes unevenly and with which also absorptivity and emissivity changes. The low emissivity gives low signal levels and also problems with reflections of the incident heat pulse, while the variation in emissivity over the surface gives strong variation in the contrast obtained even in the absence of defects. The thesis has investigated the effects of uneven discolouration of the surface and of Infra Red (IR) translucency on the thermal responses observed by using mid and long wavelength IR cameras. It has been shown that unpainted blades can be tested satisfactorily by using a more powerful flash heating system assembled with an IR glass filter and a long wavelength IR camera. The problem of uneven surface emissivity can be overcome by applying of the 2nd time derivative processing of the log-log surface cooling curves.

Arthroplasty of the knee has become one of the commonest orthopaedic procedures performed today. In the UK alone over 75,000 were performed in 2011. Patients requiring arthroplasty are getting younger and have higher demands on their replaced joints leading to continued evolution of prosthetic design. This biomechanical work has compared two different designs of Total Knee Arthroplasty (TKA) in relation to each other and the native un-resurfaced knee. The TKAs differed from each other in design of the femoral component. One had a single radius design and a trochlea that ran from the lateral side proximally, to the medial side distally, and the other prosthesis had a multi radius design with a symmetrical trochlea, essentially an unsided femoral prosthesis. The principal areas of study were the kinematics of the tibiofemoral articulation (TF), the patellofemoral joint (PFJ), the stability of the patella in the replaced knee joint and contact pressures of the tibiofemoral articulation. This was a cadaveric study using a knee navigation system to record the kinematic data for analysis. All the experiments involved cadaveric left legs of different genders and sizes. All the work was carried out at the same laboratory at Imperial College, London between July 2006 and October 2008. Both TKAs allowed significantly greater laxity than the intact knee with an anterior drawer force applied as the knees moved from 40 degrees of flexion to full extension. No significant difference was found between the two TKAs used in this study in the TF work. For the PFJ, the multiradius design was significantly more stable when the patella was displaced medially than the intact knee (p=0.016) at 30 degrees of flexion. It was also more stable than the single radius design from 0-30 degrees of flexion. There were no significant differences found between the single radius TKA and the intact knee during any of the PFJ work. Both TKAs appeared to behave differently when assessing patellar flexion with marked differences shown graphically but no statistically significant difference shown on post testing.In conclusion, both designs of TKA replicated the intact knee very well throughout all the experiments, apart from the differences noted above. This study was unable to show any significant advantage of using the newer single radius design when compared to the established multi-radius design. The single radius design did not appear to mimic the kinematics of the intact knee any closer than the established multiradius design.

Large Eddy Simulation (LES) has become an increasingly useful tool for the prediction of turbulent reactive flows with the increasing availability of cheaper and faster computing power. In the context of premixed combustion, LES encounters the challenge of resolving the flame thickness, which is normally smaller than the filter width used in typical engineering applications. This thesis considers the Flame Surface Density (FSD) approach to provide closure to the filtered LES reaction rate. The FSD can either be modelled algebraically (FSDA) or determined through a transport equation (FSDT) and both approaches are investigated in the LES of three different test cases. The first case explores the response of different FSDA models towards changes in turbulence levels, and compares the instantaneous flame structures and reaction rates predicted by FSDA and FSDT methods. The remaining cases examine the LES of two turbulent premixed burners. A relatively large range of FSDA models are tested under the same operating conditions for the first time, and the LES-FSDT equation is applied to premixed flames that involve a higher level of geometric complexity than earlier work. Generally, the results show that the performance of some FSDA models are inconsistent between the two premixed burners, suggesting that the models may operate optimally under different turbulent conditions. By contrast, the consistently good agreement of the FSDT results with experiments suggests that the method has much potential in the LES modelling of turbulent premixed flames. However, the improved FSDT predictions were dependent on the value of the model constant within the sub-grid curvature model, and the value yielded an additional dependency on filter width. For these reasons as well as for the higher computational expense, the effective use of FSDT requires further development, while the application of the FSDA models remains a viable alternative to the FSDT approach.

This research aims to examine the feasibility of a new concept in section rolling of thick strip, which either could not be rolled at present due to cracking at bent corners. Whereas, the second moment of area of sections could be increased through sharpened corners and increased gauge thickness. A heat assisted section rolling process is proposed. This process is based on application of high intensity heat on the inner surface of the strip, immediately prior to rolling. To investigate the new section rolling concept, the following work has been carried out. Firstly, the material property of the S450 steel has been determined using the Gleeble simulator, followed by thermal conductivity tests. Since a freon was used to increase temperature gradient, the heat transfer coefficient for the freon-hot surface interaction was determined. Finally, the four point hot bending tests were conducted to validate the simulation model. For this purpose, a hot bending test rig was designed and fabricated, utilizing an halogen heater as the heat source. The results between experiment and simulation were compared and a good correlation was found. Then, finite element analyses of a single pass hot rolling process has been adopted to investigate the neutral axis shift and section thickening effects. It is revealed that localised heating creates bulging on the compressed surface. The bulged surface affects the both neutral axis and thickening of the formed parts. This research has demonstrated that localised heating has a potential to be employed in section rolling operations. It shows that the neutral axis of the bent region shifted closer to the tensile surface would reduce the tendency for surface cracking. In addition, the increase in thickness that arises at a bend would enhance the stiffness of rolled sections. Ultimately a process window for heat assisted section rolling has been established.

Lubricants and lubrication have been of great interest to mankind since the introduction of machines with sliding/rolling surfaces into everyday life. With the recent trend of miniaturization, Micro-Electro-Mechanical Systems (MEMS) have taken centre stage, featuring components with scales in dimensions as small as nanometres. In this PhD study, two approaches to solving MEMS tribology problems have been pursued. First, a novel direct lubrication method using well-known lubricants such as perfluoropolyether (PFPE) and multiply alkylated cyclopentane (MAC) was developed and tested using reciprocating sliding and actual MEMS tribometry. The second approach utilized the concept of hydrodynamic lubrication and selective surface modification for MEMS. To combat spreading and starvation of lubricants in small contacts such as in MEMS, selective modification of the silicon surface with hydrophobic (non-wetting) and hydrophilic (wetting) portions was carried out and found to increase the force required to move a droplet of lubricant from a designated location on the surface. Octadecylamine and dodecylamine were also used as additives to successfully induce autophobicity in hexadecane, and the various spreading behaviours investigated. In conclusion, several new approaches to tackling tribological problems in MEMS have been researched. These methods are easily adapted to suitable MEMS devices and greatly reduce adhesion and friction, and increase wear and device life by several orders of magnitude.