Search

Glass fibre reinforced polymers (GFRP) are obvious candidate materials for use in the structures of ocean energy devices due to their corrosion resistance, high strength and low cost. The polymers normally used in GFRP can absorb up to 5% water by weight when immersed for long periods and this can reduce the tensile strength of the material by 25% or more. The thesis describes a combination of experiments and modelling studies undertaken to (i) quantify the degradation in fatigue strength due to moisture saturation in a number of candidate materials, (ii) seek to understand the damage mechanisms which are important in the degradation of the material strength by moisture saturation and (iii) predict the degradation in tidal turbine blade life due to water saturation. The experimental work involved the fabrication of quasi-isotropic (QI) coupons of vinyl-ester or epoxy / E-glass and vinyl-ester / Advantex-glass using the vacuum assisted resin transfer moulding process followed by post curing at elevated temperature for 4 hours to ensure full cure of the laminate. Approximately one half of those coupons were acceleration aged in warm water for up to 2 years to simulate immersion in 12° C seawater for 15 to 20 years. The rest were stored at normal room temperature and humidity for a similar length of time. Constant amplitude fatigue testing of both dry and wet coupons established the stress-life curves for the materials and thereby quantified the degradation in the fatigue strength due to water saturation of the materials. The modulus of the coupons was also monitored during the fatigue testing to establish the damage levels due to fatigue cycling. A number of other experimental studies were also performed to investigate the effects of applied stress during ageing and different glass fibre material on the fatigue strength of the material. A preliminary fatigue design methodology for tidal turbine blades was developed using a tidal velocity model, a hydrodynamic forces model, a structural finite element model and a strain-life fatigue model. The methodology is applied here for the preliminary design of a three-bladed tidal turbine concept, including tower shadow effects, and comparative assessment of pitch- and stall-regulated control with respect to fatigue performance. This methodology was also used to predict the effect of moisture saturation on blade life for both pitch- and stall-regulated turbines. A multiaxial fatigue damage model for fibre reinforced polymer composite materials has also been developed. The model combines (i) fatigue-induced fibre strength and modulus degradation, (ii) irrecoverable cyclic strain effects and (iii) inter-fibre fatigue. The inter-fibre fatigue aspect is based on a fatigue-modified version of the Puck multiaxial failure criterion for static failure. The model is implemented in a user material finite element subroutine and calibrated against fatigue test data for unidirectional glass fibre epoxy. Validation is performed against the fatigue tests on epoxy / E-glass coupons in the experimental programme. The model is successfully validated across a range of stress levels. The model predicts both modulus degradation and fatigue life of GFRP laminates during constant amplitude fatigue cycling. It also predicts the fatigue strength knockdown factor due to moisture saturation of the materials.

This thesis presents the development of a computational and experimental methodology for predicting crack initiation and wear in fretting of engineering materials. A bridge-type fretting fatigue test rig is designed and developed to characterise the fretting fatigue behaviour of 316L stainless steel (SS). The plain and fretting fatigue testing of the material demonstrated a significant reduction in fatigue life due to fretting, as well as a significant effect of normal load. A cylinder-on-flat crystal plasticity (CP) finite element (FE) fretting model is developed to simulate the experimental test. A J2 cyclic plasticity model of the fretting test rig is also developed for comparative purposes. The CP, based on a UMAT user subroutine, and J2 material constants are calibrated against published cyclic plasticity data for 316L SS. A microstructure-sensitive crack initiation prediction methodology is implemented based on accumulated crystallographic plastic slip as a fatigue indicator parameter (FIP). This accumulated plastic slip parameter is calibrated against the experimental plain fatigue crack initiation data for subsequent application to fretting fatigue crack initiation. Significant non-uniform (inhomogeneous) micro-plasticity effects are predicted by the microstructure-sensitive model due to the representation of inhomogeneous material behaviour at the micro-scale. A mixed mode short crack growth propagation methodology is implemented for the CP model. The CP predicted total fretting fatigue lives are shown to be in agreement with the experimental data and, more specifically, give improved predictions over a critical-plane Smith Watson Topper (SWT) approach for the J2 model. An experimental fretting life reduction factor due to fretting is reasonably well captured by the CP model. Furthermore, a novel fretting wear prediction methodology was developed using the microstructure-sensitive crack initiation FIP and the CP model. This method was shown to give reasonable correlation to the measured wear scars and published wear coefficients. A CPFE model of the cylinder-on-flat fretting wear test rig at the University of Nottingham was developed. CP dual phase and single phase material models were developed and implemented for Ti-6Al-4V within a UMAT user subroutine. Published experimental data based on a laboratory fretting wear test rig is used to validate the microstructure-sensitive prediction methodology. The constitutive behaviour of individual phases of the Ti-6Al-4V material are calibrated against experimental cyclic stress-strain curves (CSSC) from the literature to obtain phase specific CP material constants. The accumulated crystallographic slip parameter is shown to capture the low cycle fatigue (LCF) response of the material vis-à-vis the published Coffin-Manson relationship. A crystallographic random orientation study demonstrated the ability of the methodology to generate scatter in fatigue life for plain fatigue and fretting situations, akin to measured experimental fatigue scatter. The microstructure-sensitive prediction methodology successfully captured crack initiation lives, crack locations and orientations, when compared to the published experimental data from the University of Nottingham. The microstructure-sensitive fretting damage methodology based on accumulated plastic slip is implemented to distinguish between fretting wear and cracking. This technique was used to predict wear scar profiles which gave reasonable correlation with experimental measurements, and also allowed for the estimation of a wear coefficient for Ti-6Al- 4V, that is consistent with the previously published values for the same test rig and Ti-6Al-4V material.

Cells and tissues continuously experience mechanical loading during daily activity. However, the mechanisms by which cells respond to mechanical stimuli are poorly understood. The focus of this thesis is to develop atomic force microscopy (AFM) techniques to investigate whole cell mechanics over physiologically relevant time scales, providing an in-depth understanding of the role of the actin cytoskeleton in osteoblast biomechanics. The work presented in this thesis can be divided into two categories: instrument development, and experimental cell biomechanics. In terms of instrument development, the first key contribution of this thesis is the adaptation of a standard AFM to apply high precision mechanical loading at the whole cell level. Correction factors for AFM force and indentation measurements are developed, for the first time, to account for constraints imposed on AFM cantilever bending due to the attachment of a sphere at the cantilever's free-end. It is demonstrated that uncorrected force-indentation data may result in a dramatic ~18 fold underestimation of a sample¿s elastic modulus. Using this modified AFM cantilever, high precision whole cell monotonic compression is applied to osteoblasts. It is found that the actin cytoskeleton contributes significantly (~40-60%) to the whole cell compression force. Additionally it is shown that the actively generated contractility of the actin cytoskeleton has a pronounced influence on cell and nucleus morphology. The second key contribution in terms of instrument development is the stability enhancement of a standard AFM system to achieve accurate displacement control over long time scales. The methodology developed in this thesis for the reduction of thermal drift provides a significant 17-fold enhancement in z-drift stability. Furthermore, a customised fluid cell setup is developed to eliminate liquid instabilities during long term cell mechanics experiments. This enhanced AFM system is used to implement cyclic single cell deformation, with a constant loading and unloading strain rate being applied to the cell. The range of applied deformation is altered during the experiment without altering the strain rate. It is demonstrated that steady state cell forces are largely unaffected by this change in deformation range. This phenomenon is not observed for non-contractile passive cells; measured forces for cells treated with cyto-D are found to be highly dependent on the applied deformation range.

Achieving sufficient initial implant stability is of paramount importance for orthopaedic device longevity. In order to improve the design of orthopaedic devices it is essential to gain a better understanding of the mechanical behaviour of the peri-prosthetic bone, both during and following implantation. The overall aim of this thesis is to provide a better understanding of the post yield behaviour of trabecular bone using experimental and computational analyses, and to investigate the importance of multiaxial trabecular bone plasticity in the implantation and failure of orthopaedic devices. An initial investigation of full and surface cementation techniques of the tibial component in total knee arthroplasty reveals a multiaxial stress state in the peri-prosthetic trabecular bone. High von Mises stress and pressure stress computed in the trabecular bone motivates the investigation of the multiaxial post yield behaviour of trabecular bone. Samples of natural and synthetic trabecular bone are tested under confined compression loading conditions. The ability of the Drucker-Prager, Mohr-Coulomb and crushable foam plasticity formulations to predict the experimental behaviour is investigated. In the case of the crushable foam plasticity formulation, increased pressure stress during confined compression results in material yielding followed by a stress plateau, as observed experimentally. In contrast the Drucker-Prager and Mohr-Coulomb plasticity formulations fail to capture the experimental results, as increased pressure stress inhibits material yielding. The role of the trabecular bone microarchitecture in the apparent level multiaxial yielding of trabecular bone is then considered using micro-computed tomography based finite element models. iii Following this, two macroscale orthopaedic device applications that involve multiaxial trabecular bone plasticity are considered: the subsidence of a vertebral interbody fusion device; and the press-fit stem implantation of a tibial component. Continuum based pressure dependent crushable foam plasticity formulations accurately predict the subsidence forces and plastic zone size, observed during device subsidence. It is also demonstrated that the crushable foam plasticity formulation provides accurate predictions of implantation forces for tibial component implantation. Additionally, the importance of friction at the implant-bone interface is highlighted. The work presented in this thesis provides insight into the multiaxial post yield behaviour of trabecular bone. The implementation of continuum based crushable foam plasticity formulations for the simulation of the inelastic behaviour of trabecular bone may aid in the future analysis, design and development of orthopaedic devices.