Among various natural biopolymers, type I collagen gels have demonstrated the highest potential as biomimetic scaffolds for tissue engineering (TE). However, the successful application of collagen gels requires a greater understanding of the relationship between their microstructure and physical-mechanical properties. Therefore, a precise method to modulate collagen gel microstructure in order to attain optimal scaffold properties for diverse biomedical applications is necessary. This dissertation describes a new approach to produce collagen gels with defined microstructures, quantified by hydraulic permeability (k), in order to optimize scaffold properties for TE applications. It was hypothesized that the measurement of k can be used to study the role of microstructure in collagen gel properties, as well as cell function and cell-scaffold interactions. Applying increasing levels of plastic compression (PC) to the highly hydrated collagen gels resulted in an increase in collagen fibrillar density, reduced Happel model derived k values, increased gel stiffness, promoted MSC metabolic activity, osteogenic differentiation, and mineral deposition, while cell-induced gel contraction diminished. Thus, collagen gels with lower k and higher stiffness values exhibited greater potential for bone tissue engineering.Correlating between collagen gel microstructure, k, and fibroblast function within collagen gels indicated that increasing the level of PC yielded a reduction in pore size and an increase in fibril bundle diameter. Decrease in k values resulted in a decrease in gel contraction and an increase in cell metabolic activity. An increase in cell density accelerated contraction. Therefore, fibroblast function within collagen gels can be optimised by a balance between the microstructure, k, and cell seeding density.Developing a micromechanical model to measure experimental k of collagen gels during confined compression revealed the formation of a dense collagen lamella at the fluid expulsion boundary, thereby generating a two-layer model. By applying gel mass loss into Darcy's law, experimental k values of the lamella, along with the thickness of lamella (c) and hydrated gel layer (b) were measured. An increase in either compression level or compression time resulted in a decrease in k, decrease in b, and an increase in c. In conclusion, controlled compression of collagen gels can be used to produce multi-layered biomimetic scaffolds with defined microstructures and k in order to attain optimal properties for tissue engineering applications. / Parmi les biopolymères naturels couramment utilisés, les gels de collagène de type I se sont révélés être parmi les matrices biomimétiques les plus prometteuses pour l'ingénierie tissulaire. Cependant, le succès des applications thérapeutiques des matrices collagéniques nécessite une meilleure compréhension de la relation entre leur microstructure et leurs propriétés mécaniques. C'est pourquoi une méthode précise permettant de moduler la microstructure du gel de collagène est nécessaire pour pouvoir espérer atteindre les propriétés optimales de la matrice pour des applications médicales diverses. Cette thèse de doctorat décrit le développement et l'évaluation d'une nouvelle approche pour produire des gels de collagène avec une microstructure définie. Cette méthode permet de quantifier la perméabilité hydraulique (k) afin d'optimiser les propriétés de la matrice pour des applications en ingénierie tissulaire. Il a émis l'hypothèse que la mesure de k peut être utilisée pour étudier le rôle de la microstructure dans les propriétés du gel de collagène ainsi que la fonction cellulaire et les interactions matrice-cellules a été formulée.Appliquant des différents niveaux de compression plastique (PC) à des gels de collagène a entraîné une augmentation de la densité de fibrillaire, réduit les valeurs de k dérivées du modèle de Happel, augmentation de la rigidité du gel, stimulé l'activité métabolique des MSC, la différenciation ostéogénique et le dépôt de minéral, alors que la contraction du gel induite par les cellules a été réduite. Ainsi, les gels de collagène qui présentent une valeur de k plus faible et des valeurs de rigidité plus élevées ont présenté un potentiel plus élevé pour des applications en ingénierie tissulaire osseuse. Corréler la microstructure du gel de collagène, la perméabilité, et la fonction des fibroblastes cultivés dans des gels de collagène a indiqué que l'augmentation du niveau de PC résultait en la diminution de la taille des pores et une augmentation du diamètre des faisceaux de fibres. Diminution des valeurs de k résultait en une diminution de la contraction du gel et une augmentation de l'activité cellulaire métabolique. C'est pourquoi la fonction des fibroblastes, cultivés à l'intérieur de matrices de collagène, peut être optimisée en réalisant une balance entre les propriétés de microstructure, définie par k et par la densité cellulaire.Développement d'un modèle micromécanique pour mesurer la valeur expérimentale de k des gels de collagène pendant l'auto-compression radiaire confinée (SC) a révélé la formation d'une lamelle de collagène dense à la limite de l'expulsion de fluide, générant ainsi un model à deux couches. En appliquant la perte de masse de gel à la loi de Darcy, les valeurs expérimentales de k de la lamelle, ainsi que l'épaisseur de la lamelle (c) et hydratée couche de gel (b) ont été mesurés. Une augmentation soit au niveau de compression ou de temps de compression résultait en une diminution de k, diminution de b, et une augmentation de c.En conclusion, la compression contrôlée des gels hydratés de collagène peut être utilisée afin de produire des matrices multicouches biomimétiques présentant une microstructure définie et des valeurs de perméabilité permettant d'atteindre des propriétés optimales pour des applications en ingénierie tissulaire.

Cold spray is a thermo-mechanical process where powder deposition is achieved through particle acceleration to supersonic velocities using a preheated and pressurized gas and a deLeval nozzle. The operating temperatures are typically below the melting point of most metals, which limits the material oxidation and phase transformations. Thick, dense coatings can be produced at fast deposition rates (20 g/min), making cold spray a candidate for near net shape production and part repair technology for aerospace applications. However, the cold-spray process is not as well documented as thermal spray techniques and the lack of the reliable experimental data limits the use of the cold spray, especially in load bearing applications.The present study explored the mechanical and microstructural properties of the cold spray titanium splats and coatings with novel micro-mechanical testing techniques coupled with high resolution microscopy. Electron channelling contrast imaging and nanoindentation mapping were used to examine the evolution of the microstructure and hardness of individual cold spray splats. The grain refinement and recrystallization were observed at particle interfaces which contributed to the strain hardening of the material. Splats deposited above critical velocity (700 - 900 m/s) exhibited nanometer scale grains with average grain diameters in the range of 50 to 200 nm at the splat/substrate interface with regions of metallurgical bonding. A modified ball bond shear test was implemented to measure the adhesion strength of individual cold spray splats. Deposition conditions under which the adhesion strength of splats reached the theoretical shear strength of titanium were identified. Coatings, produced at the same deposition conditions as strongly bonded splats, showed strong particle cohesion strength and low coating porosity. The coating porosity and poor particle cohesion strength affected the microindentation hardness measurements and tensile properties of the titanium coating. A new multi-scale indentation method was developed, where the comparison between the nanoindentation and microindentation and the definition of a "hardness loss parameter" was used to quantify the effect of the coating defects on hardness measurements. Titanium coatings that demonstrated hardness loss parameter approaching that of the bulk titanium were also coatings that demonstrated tensile properties, measured with microtensile testing technique, of fully dense material. The splat adhesion and multi-scale indentation testing techniques developed in the present research were shown to provide new methods that can be used towards optimization of the cold spray process for deposition of titanium coatings with properties of commercially manufactured material. / La projection à froid est un processus thermomécanique où la déposition du revêtement est accomplie en accélérant les particules de la poudre, avec un gaz préchauffé et en pression, vers da vitesse supersonique. La température opérationnelle est, d'habitude, plus petite que le point de fusion de la plupart des métaux, ce qui limite l'oxydation et la transformation de phase. Des revêtements épais et denses peuvent être produits avec un taux de déposition rapide (20 g/min) permettant l'utilisation de la projection à froid pour la réparation des pièces et la formation des pièces proche de la forme finale pour les applications aérospatiales. Néanmoins, le processus de la projection à froid n'est pas aussi bien documenté que la projection thermique et manque de données expérimentales fiables, ce qui empêche l'application de la projection à froid aux pièces de portance. Le travail accomplis dans cette thèse est basé sur l'exploration des propriétés mécaniques et microstructurales des trempes et des revêtements par projection à froid de titane avec les nouvelles techniques micromécaniques et à l'aide de la microscopie à haute résolution. L'imagerie de contraste par canalisation et la cartographie par nanoindentation ont été utilisées pour examiner l'évolution de la microstructure et la dureté des trempes du titane. La diminution de taille des grains et la recristallisation des grains ont été observés à l'interface des particules ce qui a contribué à la hausse de dureté dans cette région. Les trempes, qui ont été déposées à une vitesse plus hautes que la vélocité critique (700 - 900 m/s), avaient des grains à l'échelle nanométrique (de 50 à 200 nm) à l'interface de trempe/substrat et présentaient des liaisons métallurgiques. Une nouvelle technique de balle au cisaillement modifié a été utilisée pour mesurer la force d'adhérence de ces trempes. Les conditions de la projection à froid, pour lesquelles les trempes ont démontré la force d'adhésion s'approchant de la force de cisaillement théorique du titane, ont était identifiées. Les revêtements, déposés aux mêmes conditions que les trempes démontrant une adhérence supérieure, ont présenté une forte cohésion des particules et une porosité du revêtement réduite. La porosité et la force de cohésion des particules avaient un effet sur les mesures de dureté et les propriétés de traction des revêtements. Une nouvelle méthode d'indentation multi-échelle a était développée dans laquelle la comparaison entre les mesures faites par nanoindentation et microindentation, définît comme le paramètre de perte de dureté, a été utilisé pour quantifier l'effet des défauts sur la mesure de dureté du revêtement. Les revêtements de titane qui avait un paramètre de perte de dureté s'approchant de celui du titane massif, ont démontré également les propriétés de traction du titane massif. Les techniques micromécaniques développées dans cette thèse, présente les nouvelles méthodes qui peuvent être utilisées envers l'optimisation du processus de projection à froid et le dépôt des revêtements de titane ayant des propriétés mécaniques de matériaux commerciaux.

The arc ion-plating technique is used in the industrial coating processes where TiN thin films are deposited onto various base materials. The overall objective of this research was to study the effects of low pressure nitrogen introduced into a continuous, titanium vacuum arc. An arc ion-plating system was designed and built to allow for as much flexibility as possible. Permanent magnets were placed behind the cathode surface to confine and rotate the arc. / Changes in cathode, arc and emission properties were noted with respect to vacuum, argon and nitrogen ambients. The introduction of nitrogen, above a critical pressure ($ sim$1 $ times 10 sp{-3}$ Torr), increased arc velocity and decreased crater diameter, erosion rate and ion emission. This occurred when arc rotation was combined with nitrogen introduction. Thermal properties of the cathode during arcing seemed to play an important role in determining the extent of nitrogen-cathode interaction. X-ray photoelectron spectroscopy (XPS) analysis has shown that nitriding occurred within the regions of arcing and was dependent on nitrogen pressure (with all other parameters remaining constant). The most encouraging result found was the complete elimination or macroparticles normally present in the coating/film.

High hardness, fracture toughness, modulus of rupture and fine grain size are material properties that are desired for components employed in aggressive wear environments. Since aluminum oxide offers such properties, this type of material is often selected for applications that demand wear resistance. Aluminum oxide materials commercially produced at Superior Technical Ceramics were investigated to find a correlation between their mechanical properties and wear resistant performance. A pin-on-disc apparatus and a tumbling test were employed to determine the relative difference in wear between different alumina compositions. The selected 88%, 96%, 98% purple, 98% alumina and zirconia toughened alumina were subjected to four different wear tests in order to determine their wear behavior in each system. The objective was to characterize the compositions according to their mechanical properties and to establish a wear material selection guide for Superior Technical Ceramics. (Abstract shortened by UMI.)

Equipment and techniques have been developed for the microstructural characterization of Zirconium Oxide films grown on Zr-2.5%Nb pressure tubes. A thin film texture apparatus was constructed and used to measure the texture and stress present in thin zirconium oxide films. The general techniques developed employ a grazing incidence geometry which allows the texture and stress present in thin films (<1mum) of any type to be examined. In addition, a technique for the quantitative phase analysis of textured ZrO2 films grown on zirconium alloys using pole figure data has also been developed. Moreover, equipment was constructed to determine the relative porosity of oxide films grown on a metal substrate using an electrochemical method that measures the effective non-porous oxide thickness. The described equipment and techniques were then used to characterize a test matrix of specimens whose relative hydrogen pick-up was measured by Differential Scanning Calorimetry. The application of beat treatments to the substrates prior to oxide growth was found to have a pronounced effect upon the sharpness of the oxide texture. A correlation between the degree of sharpness of the oxide texture and hydrogen pick-up and corrosion rate of the substrate was also determined. In addition, based upon the new techniques developed it was determined that the tetragonal phase of the oxide is stress stabilized in a region close to the metal/oxide interface.

Small cavities, called micropores, are common defects found in aluminum castings. Mathematical modeling of microporosity is difficult due to a lack of chemical parameters and a limited understanding of the mechanisms of pore formation and development. / The objective of the present work is to obtain a better understanding of the nucleation and growth of microporosity in Al-7%Si (A356) foundry alloy. This consisted of the measurement of the surface tension of A356 alloy, a study of the factors that affect the final amount of porosity in castings, and the measurement of the evolution of porosity during solidification. / The measurement of the surface tension of pure aluminum and A356 alloy was performed under vacuum and hydrogen atmospheres. The surface tension of A356 alloy under vacuum at 630°C was 0.889 N/m. The addition of hydrogen did not significantly alter this value. The addition of strontium decreased the surface tension by about 5%. It was concluded that the effect of strontium on the surface tension was not the cause of the increase in porosity observed in strontium modified castings. / The effects of hydrogen content, local solidification time and strontium modification on microporosity were studied. Two critical fraction solids were associated with the formation of microporosity. The first, the critical fraction solid for pore growth, is the point at which preexisting baseline porosity begins to grow due to hydrogen evolution. It varied with hydrogen content and strontium modification. The second, the critical fraction solid for nucleation, is the point at which shrinkage type porosity nucleates. It did not vary with hydrogen content or strontium modification, and remained constant at approximately 75--80% solid. / The measurement of microporosity was performed using image analysis. A new method for filtering and sorting the data, based on nearest neighbor cluster analysis, was developed. / A mathematical model based on the current experimental results was created. The model is capable of predicting the evolution of the percentage porosity during solidification, the final percentage porosity and the maximum pore size.

This study focuses on various aspects of solid-state diffusion bonding of two ceramic-metal combinations, namely: silicon carbide-molybdenum (SiC-Mo), and silicon nitride-molybdenum (Si$ rm sb3N sb4$-Mo). Single SiC-Mo and $ rm Si sb3N sb4$-Mo joints were produced using hot-uniaxial pressing. The microstructure of the resulting interfaces were characterized by image analysis, scanning electron microscopy (SEM), electron probe micro-analysis (EPMA), and X-ray diffraction (XRD). The mechanical properties of the joints were investigated using shear strength testing, depth sensing nanoindentation, and neutron diffraction for residual stress measurement. / SiC was solid-state bonded to Mo at temperatures ranging from 1000$ sp circ$C to 1700$ sp circ$C. Diffusion of Si and C into Mo resulted in a reaction layer containing two main phases: $ rm Mo sb5Si sb3$ and Mo$ sb2$C. At temperatures higher than 1400$ sp circ$C diffusion of C into $ rm Mo sb5Si sb3$ stabilized a ternary phase of composition $ rm Mo sb5Si sb3$C. At 1700$ sp circ$C, the formation of MoC$ rm sb{1-x}$ was observed as a consequence of bulk diffusion of C into Mo$ sb2$C. A maximum average shear strength of 50 MPa was obtained for samples hot-pressed at 1400$ sp circ$C for 1 hour. Higher temperatures and longer times contributed to a reduction in the shear strength of the joints, due to the excessive growth of the interfacial reaction layer. $ rm Si sb3N sb4$ was joined to Mo in vacuum and nitrogen, at temperatures between 1000$ sp circ$C and 1800$ sp circ$C, for times varying from 15 minutes to 4 hours. Dissociation of $ rm Si sb3N sb4$ and diffusion of Si into Mo resulted in the formation of a reaction layer consisting, initially, of $ rm Mo sb3$Si. At 1600$ sp circ$C (in vacuum) Mo$ sb3$Si was partially transformed into $ rm Mo sb5Si sb3$ by diffusion of Si into the original silicide, and at higher temperatures, this transformation progressed extensively within the reaction zone. Residual N$ sb2$ gas, which originated from the decomposition of $ rm Si sb3N sb4,$ dissolved in the Mo, however, most of the gas escaped during bonding or remained trapped at the original $ rm Si sb3N sb4$-Mo interface, resulting in the formation of a porous layer. Joining in N$ sb2$ increased the stability of $ rm Si sb3N sb4,$ affecting the kinetics of the diffusion bonding process. The bonding environment did not affect the composition and morphology of the interfaces for the partial pressures of N$ sb2$ used. A maximum average shear strength of 57 MPa was obtained for samples hot-pressed in vacuum at 1400$ sp circ$C for 1 hour.

Palladium, being impermeable to all gases except hydrogen, has been widely studied for hydrogen extraction in recent years. The specific surface area of the membrane is an important factor affecting the hydrogen permeation rate. How to obtain a palladium membrane with a high specific surface area is a great challenge for material scientists. In this study, a novel template-assisted technique was used to prepare nano-structured palladium membranes with a greatly increased hydrogen contacting surface. First, the anodic aluminum oxide (AAO) template was fabricated by anodizing electro-deposited aluminum film and commercially available aluminum foil. The template was then filled with palladium using the electrochemical and the sputter deposition techniques. Various factors affecting the preparation of the palladium membrane were analyzed and optimized. The preliminary hydrogen-permeation experiments clearly showed that the nano-structured palladium membrane is a promising candidate for the application of hydrogen separation. / Le palladium, étant imperméable à tous les gaz à l’exception de l'hydrogène, a été largement étudié pour l'extraction d'hydrogène dans les dernières années. La surface spécifique de la membrane est un facteur important qui affecte le taux de perméabilité d'hydrogène. Un grand défi pour la communauté scientifique est d’obtenir une membrane en palladium avec une surface spécifique élevée. Dans cette étude, une nouvelle technique utilisant une matrice ordonnée a été employée pour préparer des membranes de palladium nano-structurées avec une surface de contact avec l’hydrogène considérablement accrue.Tout d’abord, l'oxyde d'aluminium anodique (OAA) a été fabriqué en anodisant une couche d’aluminium électro-déposée et un papier d'aluminium commercial. Puis, la matrice d'OAA a été remplie de palladium utilisant la technique électrochimique et la technique par pulvérisation. De divers facteurs affectant la préparation de la membrane de palladium ont été analysés et optimisés. Les expériences préliminaires de la perméabilité d’hydrogène ont clairement prouvé que de telles membranes de palladium sont un candidat prometteur pour l'application de la séparation d'hydrogène.

An in-depth investigation is carried out to characterise the damage accumulation mechanisms and fatigue growth in single wood pulp fibres, which may be thought of as concentrically-layered, filamentary composite tubes that are approximately: 1-3 mm long, 20-40 $ mu$m in cross-section. The novel experimental methodology principally consists of the in situ apparatuses, on the one hand, which comprise the in-house designed and built single-fibre tensiometer, in conjunction with the confocal laser scanning microscope (CLSM) and, on the other, the computer hardware and periphery. The tensiometer's prime components, the loading jaws, are specifically designed to best emulate the force actions to which wood pulp fibres are subjected while between the discs of mechanical refiners (viz.: cyclic shear, radial compression and tension); and incorporate a mechanical fixation mechanism to ensure proper mounting of the single fibres. Moreover, the CLSM, which functions by scanning a diffraction-limited spot of light relative to the specimen in a raster-type scan, is a powerful tool for obtaining qualitative information on the morphology of fractured surfaces and structural behaviour of the fibres being fatigued, as well as providing accurate visual records of the history of crack propagation. The entire fully-automated set-up is controlled, in real time, via a computer algorithm specifically written for displacement-control fatigue-testing, while making efficient utilisation in terms of execution time, memory allocation, signal conversion and data acquisition. / The engendered conclusions may be summed up as follows. From a litany of tenuously-oriented microcracks, dominant macrocracks propagate along the axis of the fibre which may sharply deflect in the presence of natural bias (e.g. pits in the fibre wall). The material property degradation characteristics further include: volumetric expansion due to internal fibrillation in the cell wall, extensive external fibrillation, gradual delamination of the layers and partial peeling-off of the cell wall material. Cumulative damage due to cyclic shear is shown to be the most significant, further supported by the high structural collapsibility of the fibre wall layers. The mechanisms of fracture are either due to the development of transverse cracks at regions of high stress concentration (such as bordered pits), or owing to the gradual slippage of the fibre wall layers.

Thermoforming of heated polymeric sheets into three dimensional objects of various kinds, by means of pressure inflation against molding surfaces, presents a class of contact problems. This thesis develops the finite element formulations and computer programs for modelling the free and constrained inflation of thin polymer sheets in relation to thermoforming applications. In recognition of the generally time-dependent viscoelastic behaviour of polymers, and the large strains encountered in thermoforming applications, the material is modelled as non-linear viscoelastic. For this purpose the constitutive relation proposed by Christensen is adopted, assuming the relaxation function to be exponential. Most published work on nonlinear viscoelastic membranes deals with simple axisymmetric geometries, while the finite element formulations presented in this work are for both axisymmetric and nonaxisymmetric membrane inflations, including contact against constraining surfaces. Both frictionless and slipless idealizations of contact conditions are studied. The finite element solutions of free and constrained inflations of circular membranes serve as illustrative examples for the axisymmetric case, while those for elliptical membranes serve this purpose for the nonaxisymmetric case. Comparison of the finite element results with the analytical solutions obtained for some simple free and constrained inflation problems shows excellent agreement.