This thesis describes the study of the molecular packing and organisation of collagen molecules within a fibril. The first two chapters describe the background to the study. In Chapter 1, a review of the extracellular matrix concentrates on the structure and organisation of type I collagen. Chapter 2 summarises the theory of X-ray diffraction by fibres, and Chapter 3 describes X-ray sources and equipment used in data collection. Data treatments and data extraction methods (such as simulated annealing) are also discussed. Chapters4 and 5 present the results of the study. Chapter 4 describes the determination of the one-dimensional structure of type I collagen to 0.54 nm resolution using X-ray diffraction and isomorphous derivative phase determination. The significance of the electron density map is interpreted in light of the known amino acid sequence, showing possible variations in the nature of the helix pitch. More importantly, the conformations of the intermolecular crosslink forming non-helical telopeptides were determined. Chapter 5 provides a detailed background to the current understanding of the three dimensional packing structure of collagen, and presents the first model-independent phase determined structure of a natural fibre - the lateral packing structure of type I collagen in rat tail tendon. The data extraction methods described in Chapter 3 are employed to calculate an electron density map of anisotropic resolution, from which the 4 crosslink forming telopeptide segments within the quasi-hexagonal packing structure are identified. Conclusions are drawn concerning the nature of order/disorder within collagen fibrils and the validity of the compressed microfibril model of collagen molecular packing and organisation is discussed. Chapter 6 summaries the results and evaluates the success of the study. The potential for development of the techniques and results found for further studies are also discussed.

This thesis describes an integrated approach to fatigue cracking, reliability and inspection of offshore structures. The basis of the approach is statistical in nature and draws on recent experimental data and field measurements. It is intended as a working tool for those engaged in design, structural appraisal and sub-sea inspection of steel jacket structures. A review of current practice has been made and the requirements of an integrated approach are established. An approach is proposed comprising a series of compatible models dealing with fatigue cracking, the reliability of cracked joints and the inspection of welds for fatigue cracks. The primary linking parameter is the distribution of fatigue crack size which is considered as a time dependent variable. An integral part of the approach is a new statistically-based fatigue crack growth model. This is developed and the parameters involved in the model estimated from an analysis of experiment and oceanographic data. For any fatigue calculation the model allows the corresponding fatigue crack growth distribution to be estimated for any time during, or beyond, the nominal fatigue life. A number of example calculations are included; and using one of these a Bayesian procedure for revising fatigue lives in the light of inspection results is demonstrated. The effect of fatigue cracking upon the various modes of tubular joint failure is considered using linear statistical models. Example calculations are performed for a typical joint. An inspection strategy is proposed based on the concept of minimising life costs, including risk costs arising from the consequences of possible structural failure. This allows alternative inspection plans to be evaluated and compared, and a typical example calculation is included. The approach is discussed in the context of possible alternative approaches and areas for further related research are identified.

No description available.

A new, two-dimensional model is presented for predicting the behaviour of tall, thin piezoelectric transducer structures, typical of those encountered in many ultrasonic phased array assemblies. Based on linear systems theory, the model may be represented in block diagram format, utilising feedback and feedforward loops to model mechanical and piezoelectric cross coupling, in addition to secondary and tertiary piezoelectric generation. As a result, the model permits a ready understanding of the electrical, mechanical and piezoelectric interactions which take place when two principal vibrational modes are present within a transducer structure. Comprehensive experimental and theoretical results are shown to produce close agreement, and compare favourably with alternative modelling strategies. It is considered that this novel approach will be of considerable benefit to the analysis and understanding of the transduction process within such structures.

Limit and shakedown analyses are powerful methods in designing pressure vessel components and other engineering structures. With the development of computer technology the use of finite element analysis as an alternative tool for engineering structure design has become ever more increasing. In this thesis the finite element method utilises the novel elastic compensation method to carry out the limit and shakedown analyses on variety engineering structures: nozzles in pressure vessels, plates with a central hole and frame structures. There are two main objectives of the present study. One of them is to conduct limit and shakedown analyses on a series of thick cylinders, nozzle/sphere intersections under internal pressure, plates with a hole and frame structures under multiple loading conditions using the initially developed elastic compensation method based on 2-D solid element models. A comparision of the lower, upper bound limit loads and shakedown solutions is made with the result s available in literature or with new elasto-plastic analyses. The results obtained using the elastic compensation method were found to be of useful accuracy. Another is to further develop the elastic compensation method using generalised yield criteria. Then the procedure is implemented to beam and shell finite elements to calculated limit loads for beam and shell structures. A number of 2-D and 3-D frames were examined using a general yield surface. The obtained results were compared with that of theoretical plastic analysis and with the results available in literature and were found to be in good agreement. Parametric studies of nozzle/sphere intersections and nozzle/cylinder intersections under internal pressure were carried out using Ilyushin's and Ivanov's generalised yield criteria. The results calculated were compared with the solutions obtained using the initially elastic compensation method and with the solutions available in literature and were found to also be in good agreement. The newly developed elastic compensation procedure using generalised yield criteria was found to be more economic and useful in engineering design. From this study, some new design methods based on limit and shakedown loads are proposed for nozzle/sphere intersections and for other engineering structures. The newly developed elastic compensation procedure using generalised yield criteria is highly recommended in structural design for a quick limit load estimation.

Chez les plantes à fleurs, où une immense variété de stratégies de reproduction est rencontrée, la dispersion des gènes s’opère classiquement via des flux de pollen pour la voie mâle, et via des flux de graines pour la voie femelle. La gynodioécie correspond à un système de reproduction original, caractérisé par la coexistence de plantes femelles et de plantes hermaphrodites au sein de populations naturelles. Ce système de reproduction suscite depuis longtemps un intérêt particulier, du fait d’un paradoxe évolutif apparent : les individus femelles, ayant perdu une voie de transmission de leur information génétique, devraient être désavantagés par rapport aux hermaphrodites, dotés des deux voies de transmission des gènes. L’objet de ce travail thèse était d’expliquer le maintien des femelles et les importantes variations spatiales de sex ratio fréquemment observées dans les populations naturelles de betterave maritime (Beta vulgaris ssp. maritima), chez laquelle le déterminisme du sexe implique des interactions entre des gènes cytoplasmiques induisant la stérilité mâle (CMS) et des allèles nucléaires de restauration de la fonction mâle. En mêlant des approches de biologie et de génétique des populations, trois thématiques ont ainsi été abordées. (i) La valeur sélective des différents types sexuels a été mesurée, en faisant appel à des mesures phénotypiques en conditions contrôlées et à des analyses de paternité en populations naturelles. Ceci a permis de révéler un avantage femelle extrêmement restreint ainsi que des variations importantes du succès reproducteur mâle chez les hermaphrodites, liées notamment au génotype des individus et à l’existence d’un coût de la restauration. (ii) La caractérisation moléculaire de la diversité génétique au sein de plusieurs populations naturelles et entre plusieurs cohortes successives nous a permis de mettre en lumière l’importance relative des effets de la migration, des évènements de fondation et de la dérive génétique sur la structure spatiale des phénotypes sexuels. Les variations observées de sex ratio à très fine échelle semblent ainsi expliquées par des effets de fondation multiples couplés à des flux de pollen et de graines restreints dans l’espace. (iii) Finalement, nous mettons en évidence l’effet important que cette forte structure spatiale des sexes peut avoir, à la fois sur les individus hermaphrodites et sur les individus femelles. L’ensemble de nos résultats montre qu’une structure spatiale à fine échelle, générée par des évènements de fondation et des flux de gènes limités, peut affecter de façon importante la dynamique de la gynodioécie dans la nature. / In flowering plants, which exhibit a spectacular diversity of reproductive strategies, gene dispersal generally occurs through two distinct pathways: pollen for the male function, and seed for the female function. Among sexually polymorphic flowering plants, gynodioecy refers to a particular breeding system in which females and hermaphrodites co-occur in natural populations. Since females reproduce only through seeds, they apparently transmit their genes only half as frequently as hermaphrodites, which gain fitness through both seed and pollen production. This apparent evolutionary paradox has attracted the attention of evolutionary biologists as far as the mid-nineteenth century. The aim of this PhD thesis was to understand the successful maintenance of female individuals and the important spatial variations in sex ratio that are often observed in natural populations of gynodioecious Beta vulgaris ssp. maritima, in which sex is determined by interactions between cytoplasmic male sterility (CMS) genes and nuclear restorers of male fertility. Using population biology and population genetics approaches, three distinct themes were considered. (i) First, male and female fitness of the different sexual types were compared, using both measures in controlled conditions and paternity analyses in the wild. While our results only suggest a very restricted female advantage, we detected strong male fitness differences among hermaphrodites that were partially explained by the genotype of individuals and by the occurrence of a cost of restoration. (ii) The study of the distribution of genetic diversity in several natural populations and in several consecutive cohorts allowed us to quantify the relative impact of migration, founder events and genetic drift on the spatial distribution of sexes. Overall, random founder effects with spatially restricted pollen and seed flow appeared to be the primary determinants of sex ratio variations. (iii) Finally, we explore how such sex ratio variation can affect the reproductive output of hermaphrodites and females. Altogether, we show that fine-scale spatial structure, resulting from the joint action of founder events and limited gene flow, can notably modify the dynamics of gynodioecy in natural populations.

Network structure plays a significant role in determining the performance of network inference tasks. An interactive tool to study the dependence of network topology on estimation performance was developed. The tool allows end-users to easily create and modify network structures and observe the performance of pole estimation measured by Cramer-Rao bounds. The tool also automatically suggests the best measurement locations to maximize estimation performance, and thus finds its broad applications on the optimal design of data collection experiments. Finally, a series of theoretical results that explicitly connect subsets of network structures with inference performance are obtained.

Because of the need for data on the geometry of nitrogen in arylamines, the determination of the crystal and molecular structures of tri-(p-fluorophenyl)-amine (TFPA) and tri-(p-iodophenyl)-amine (TIPA) was undertaken as the subject of this dissertation.

No description available.

No description available.