Thesis (Ph. D.)--Rutgers University, 2007. / "Graduate Program in Ceramic and Materials Science and Engineering." Includes bibliographical references (p. 494-501).

Bio-composite hierarchical materials have attracted the interest of the academic community operating in the field of bio-inspired materials for their outstanding mechanical properties achieved via lightweight structural designs. Antler and mantis shrimp's cuticle are extreme examples of materials naturally optimised to resist impacts and bear dynamic loading. Firstly, a class of finite-element fibril models was developed to explain the origin of heterogeneous fibrillar deformation and hysteresis from the nanostructure of antler. Results were compared to synchrotron X-ray data and demonstrated that the key structural motif enabling a match to experimental data is an axially staggered arrangement of stiff mineralised collagen fibrils coupled with weak, damageable interfibrillar interfaces. Secondly, the cuticle of the crustacean Odontodactylus scyllarus, known as peacock mantis shrimp, was investigated. At the nanoscale it consists of mineralised chitin fibres and calcified protein matrix, which form plywood layers at the microscale. Lamination theory was used to calculate fibrillar deformation and reorientation and, in addition, an analytical formulation was used to decouple in-plane fibre reorientation from diffraction intensity changes induced by 3D lamellae tilting. This animal also attracted my attention for using its hammer-like appendages to attack and destroy the shells of prey with a sequence of two strikes. Inspired by this double impact strategy, I performed a set of parametric finite-element simulations of single, double and triple mechanical hits, to compute the damage energy of the target. My results reveal that the crustacean attack strategy has the most damaging effect among the double impact cases, and lead me to hypothesise, that optimal damaging dynamics exists, depending on the sequence of consecutive impacts and on their time separation values. These new insights may provide useful indications for the design of bio-inspired materials for high load-bearing applications.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / A Ãrea de CiÃncia dos Materiais utiliza sistemas de VisÃo Computacional para determinar tamanho e/ou quantidade de grÃos, controle de soldagem, modelamento de elementos de ligas, entre outras. O presente trabalho tem como principal objetivo desenvolver e validar o programa SVRNA (SegmentaÃÃo de Microestruturas por VisÃo Computacional Baseada em Rede Neural Artificial), que, combinado com Rede Neural Artificial, utiliza tÃcnicas de morfologia matemÃtica para realizar a segmentaÃÃo dos constituintes do ferro fundido branco de forma semi-automÃtica e a classificaÃÃo automÃtica da grafita nos ferros fundidos nodular, maleÃvel e cinzento. Os resultados da segmentaÃÃo e quantificaÃÃo destes materiais sÃo comparados entre o SVRNA e um programa comercial bastante utilizado neste domÃnio. A anÃlise comparativa entre estes mÃtodos mostra que o SVRNA apresenta melhores resultados. Conclui-se, portanto, que o sistema proposto pode ser utilizado em aplicaÃÃes na Ãrea da CiÃncia dos Materiais para a segmentaÃÃo e quantificaÃÃo de constituintes em materiais metÃlicos, reduzindo o tempo de anÃlise e obtendo resultados precisos. / CATERIALS Sciences field uses Computational Vision systems to determine size and/or amount of grains, welding control, modeling of alloy elements, among other. The present paper has as main objective to develop and validate the SVRNA system (Microstructure Segmentation for Computational Vision based on Artificial Neural Networks), which, combined with ArtiÂcial Neural Network, uses mathematical morphology technics to accomplish the constituent segmentations from white cast iron of semi-automatic form, and graphite automatic classiÂcation from nodular, malleable and gray cast iron. Segmentation and quantiÂcation results of this materials are compared between SVRNA and a commercial program more used in this domain. Comparative analysis between this methods showed that SVRNA present best results. It has concluded, therefore, which the proposed system can be used in applications in Material Sciences field for microstructure segmentation and quantification in metallic materials, reducing the analyze time, and obtained accurate results.