In the last few decades, there has been a shift to using more lightweight materials for the potential of fuel consumption reduction. In the Aerospace Industry, conventional metal structures are being replaced by advanced composite structures. The major advantage of an advanced composite structure is the huge reduction in the number of parts and joints required. Also composite materials provide better resistance to creep, corrosion, and fatigue. However, one cannot eliminate all the joints and attachments in an aircraft’s structure. Eliminating structural joints is impractical in present-day aircraft because of the requirements for inspection, manufacturing breaks, assembly and equipment access, and replacement of damaged structures. Currently, composite joints are overdesigned which leads to weight penalties. Understanding how to optimize the ultimate bearing strength of a composite joint by altering the cure cycle might be beneficial to the composite joint design process. This study investigates, through numerical and experimental analysis, the mechanical behavior of double shear joints. The first task is to test Aluminum double shear joint specimens inside the double shear joint fixture at a loading rate of 0.05 in./min. (quasi-static). The second task is to numerically model and validate the aluminum double shear joint specimen. The third task is to test the Unidirectional MTM 49 carbon fiber pre-preg double shear composite joint specimens with two different cure cycles and five different loading rates (0.05 in./min., 0.1 in./min., 1 in./min., 2 in./min. and 6 in./min.). The double shear composite joint specimens are made, using a heat press, with a quasi-isotropic laminate orientation of [0 0 +45 -45 +45 -45 90 90]s. The first cure cycle used is called the alternate cure cycle, which is Cytec’s MTM 49 Unidirectional Carbon Fiber pre-preg material cure cycle, and the second cure cycle used is called the datasheet cure cycle, which is Umeco's MTM 49 Unidirectional Carbon Fiber pre-preg material cure cycle. The recommended datasheet cure cycle and an alternate cure cycle are both compared to see how they affect the mechanical characteristics of the matrix along with the bearing stress. The fourth task is to adjust the Aluminum double shear joint numerical model for the double shear composite joint specimen. The numerical results for both the Aluminum and the composite specimens are in agreement with the experimental results. The theoretical in-plane material properties of the quasi-isotropic laminate were in agreement with the experimental results. One can see that at 0.05 in./min. and 0.1 in./min. (for both cure cycles) the composite double shear specimens carried more load compared to the higher loading rates of 1 in./min., 2 in./min. and 6 in./min. The tensile modulus of elasticity of an Aluminum sample is measured using a crosshead displacement, a strain gage and an extensometer. The crosshead displacement yielded very inaccurate results when compared to the strain gage and the extensometer.

Two failure mechanisms in a double lap joint are investigated. Analytical models of net-section and gross-section failure modes are proposed to describe these mechanisms. The effects of lamping force, interference fit, maximum axial load and WDCP on fatigue performance of the joint are included in the models. The effect of WDCP is assumed to give a reduction in friction coefficient. Three types of stress reduction factors are proposed in the net-section failure model to account for these parameters. The stress reduction factors modify stress range that is used in crack growth calculation. If there are no effects of these parameters, the stress reduction factors are equal to one. Two types of fretting stress are introduced in gross-section failure model to describe either sliding contact or incipient sliding contact on faying surface. The fretting stress is combined with body stress to modify stress range. The net-section failure model predicts that fatigue life is increasing as interference fit, clamping force and friction coefficient increase. The gross-section failure model predicts that fatigue life is decreasing as clamping force and friction coefficient increase. Both models predict that fatigue life is decreasing as maximum axial load increases. Transition of the failure mode occurs earlier as friction coefficient and interference fit increase, while it is delayed as maximum axial load increases. A transition parameter is proposed to establish a relationship between the four main parameters. The transition parameter is expressed in a polynomial equation. It gives an optimum combination of the four main parameters in order to achieve relatively higher fatigue life by having gross-section failure mode. Finite element analysis and fatigue testing are performed to validate the models. The finite element and the analytical models show that stress concentration factor at the edge of the hole is decreasing as clamping force increases. The rate of decrease of stress concentration factor is increasing as friction coefficient increases. While stress concentration factor on the faying surface is increasing as clamping force and friction coefficient increase. Fatigue testing reveals that the fatigue life of the joint is in good agreement with the predicted fatigue life of the proposed models.

Le collage par adhérence moléculaire est un collage basé sur la mise en contact de deux surfaces sans l'utilisation de colle ou matériaux additionnels. Ce procédé de collage est utilisé dans de nombreuses applications, notamment dans les domaines de l'optique terrestre et spatiale. Bien qu'un prototype ait déjà passé avec succès l'environnement spatial - où les contraintes d'utilisations sont différentes de celles rencontrées sur Terre - la spatialisation de cette technologie nécessite une caractérisation plus fine du procédé ainsi qu'une amélioration de la tenue mécanique des interfaces adhérées afin de valider les normes de l'Agence spatiale Européenne. Pour répondre à cette problématique de spatialisation de la technologie, des essais mécaniques ainsi que des analyses chimiques ont été réalisés dans le but d'étudier l'influence de certains paramètres du procédé ) sur la tenue mécanique et l'énergie de collage. Ces essais ont également été réalisés afin de comparer les deux matériaux étudiés : le verre de silice et le Zérodur vis-à-vis de l'adhésion. A l'issue de ces essais, les paramètres à appliquer permettant de doubler la tenue mécanique des interfaces adhérées ont été déterminés. Parallèlement, une loi phénoménologique reliant l'énergie de collage aux précédents paramètres du procédé a été développée ainsi qu'un modèle macroscopique visant à décrire l'intensité d'adhésion. Ces deux modèles une fois couplés permettent de modéliser le comportement normal de l'interface en fonction des paramètres du procédé. Enfin, ces deux lois sont implémentées dans un code éléments finis afin de simuler la propagation de la fissure lors de l'essai de clivage au coin. / Direct bonding consists in joining two surfaces without the use of any adhesive or additional material. This process is used in several applications, particularly in terrestrial and spatial optics. Although a prototype passed with success spatial environment - where constraints involved are very different from those encountered on Earth - this technology requires a more detailed characterization and an improvement of the mechanical strength of bonded interfaces in order to validate the European Space Agency standards. To address this issue, mechanical tests (double shear tests, cleavage tests and wedge tests) and chemical analysis (wetting tests and XPS spectroscopy) were performed in order to study the influence of some process parameters (roughness, relative air humidity during room temperature bonding, the annealing temperature and time) on the mechanical strength and the bonding energy. These tests compared the two materials used: fused silica glass and Zerodur glass. As a result of these tests, optimal parameters doubling the mechanical strength were also obtained. In the same time, a phenomenological law relating the bonding energy to the previous parameters is developed as well as a macroscopic model to describe the adhesion intensity. Both models when coupled describe the normal behavior of the bonded interface depending on the process parameters. Then, the both laws are implemented in a finite elements model in order to simulate the crack propagation during the wedge test.

This thesis concerns the mathematical modelling of phase transformations with a special emphasis on martensitic phase transformations and their application to the modelling of steels. In Chapter 1, we develop a framework that determines the optimal transformation strain between any two Bravais lattices and use it to give a rigorous proof of a conjecture by E.C. Bain in 1924 on the optimality of the so-called Bain strain. In Chapter 2, we review the Ball-James model and related concepts. We present some simplification of existing results. In Chapter 3, we pose a conjecture for the explicit form of the quasiconvex hull of the three tetragonal wells, known as the three-well problem. We present a new approach to finding inner and outer bounds. In Chapter 4, we focus on highly compatible, so called self-accommodating, martensitic structures and present new results on their fine properties such as estimates on their minimum complexity and bounds on the relative proportion of each martensitic variant in them. In Chapter 5, we investigate the contrary situation when self-accommodating microstructures do not exist. We determine, whether in this situation, it is still energetically favourable to nucleate martensite within austenite. By constructing different types of inclusions, we find that the optimal shape of an inclusion is flat and thin which is in agreement with experimental observation. In Chapter 6, we introduce a mechanism that identifies transformation strains with orientation relationships. This mechanism allows us to develop a simpler, strain-based approach to phase transformation models in steels. One novelty of this approach is the derivation of an explicit dependence of the orientation relationships on the ratio of tetragonality of the product phase. In Chapter 7, we establish a correspondence between common phenomenological models for steels and the Ball-James model. This correspondence is then used to develop a new theory for the (5 5 7) lath transformation in low-carbon steels. Compared to existing theories, this new approach requires a significantly smaller number of input parameters. Furthermore, it predicts a microstructure morphology which differs from what is conventionally believed.

Cette étude vise à bien comprendre puis à modéliser le comportement mécanique dans une large plage de vitesse de déformation des tôles d’acier austénitique AISI304 ayant subis un traitement d'attrition mécanique de la surface (SMAT). Ces tôles ainsi traités sont des matériaux multicouches avec un gradient de propriétés. Les principaux résultats obtenus sont résumés comme suit：(1) La sensibilité globale à la vitesse déformation des tôles d’acier austénitique AISI304 traités avec SMAT est caractérisée par des essais de double cisaillement sous chargements quasi-statiques et dynamiques, qui permet d’atteindre une grande déformation sans instabilité géométrique. Des essais de double cisaillement sous impact sont réalisés à l’aide des barres de Hopkinson de grande diamètre et un système d’attache qui a une même impédance acoustique que la barre. Une sensibilité significative a été révélée et on observe ce renforcement n’a pas induit une réduction importante de la ductilité.(2) Dans le but d’un meilleur dépouillement de ces essais de double cisaillement, leur conditions d’essai est analysé dans le détaillé. Le modèle numérique avec le système d’attache a été construit pour étudier l’influence du système d’attache au début de chargement. On trouve un effet limité pour les diverses conditions imparfaites des essais comme la souplesse de système d’attache, des champs mécaniques non-homogènes, l’état de non-équilibre, etc. Par contre, les études numérique et analytique ont démontré que l’hypothèse simple de petites perturbations habituellement utilisé pour le dépouillement de ces essais n’est pas suffisamment précise. La déformation Eulérien cumulée doit être utilisée pour obtenir un résultat numérique correct. A partir de ce résultat, la sensibilité à la vitesse déformation des tôles d’acier austénitique AISI304 traités avec SMAT obtenue expérimentalement a été retouchée.(3) Un modèle multicouche elasto-plastique en dommageable a été proposé pour décrire le comportement des tôles d’acier austénitique AISI304 traités avec SMAT. Les paramètres sont identifiées à partir des essai de traction. La partie elasto-plastique est calée par une loi d’écrouissage de type Ludwig. Par contre, les paramètres d’endommagement sont obtenus avec une méthode d’identification inverse sur la base de simulation numérique de ces essais de traction. Pour valider ce modèle multi-couche elasto-plastique dommageable, un essai d’indentation/perforation est réalisé sur des tôles d’acier austénitique AISI304 traités avec SMAT. Des simulations numériques correspondantes montres que ce modèle multi-couche elasto-plastique en dommageable une prédiction plutôt précise de ces essais de d’indentation/perforation.(4) Pour évaluer la performance anti-perforation des tôles d’acier austénitique AISI304 traités avec SMAT, des essais de perforation sous impact a été réalisés avec des barres de Hopkinson. Des simulations numériques de ces essais de perforation sous impact sont réalisées avec un modèle numérique comparable avec le cas quasi-statique. ́tant donne que la sensibilité globale à la vitesse déformation des tôles d’acier austénitique AISI304 traités avec SMAT est caractérisée par des essais de double cisaillement, la sensibilité à la vitesse a été introduite dans le modèle multi-couche elasto-plastique en dommageable. Le résultat numérique correspond bien à la mesure expérimentale, ce qui indique non seulement l’efficacité du modèle numérique mais aussi celle du modèle multicouche elasto-plastique en dommageable. / This Ph.D dissertation aimed at the comprehensive understanding and the constitutive modeling of the mechanical behaviours of the surface mechanical attrition treatment (SMAT) treated AISI304 stainless steel sheet under a large range of loading rates. SMAT treated AISI304 stainless steel sheets are multi-layered functionally graded materials (FGM). The main research results and conclusions are summarized as followed:(1) The overall rate sensitivity SMAT treated AISI304 stainless steel sheet is characterized by the double shearing test under quasi-static and dynamic loading where a large strain can be achieved without geometry instability. Impact double shear test are performed with a large diameter Hopkinson bar system and an adapted equal-impedance clamping device. Significant rate sensitivity is found. It is also observed that such a rate enhancement does not induce an important reduction of the ductility.(2) In order to extract accurate material information from the double shear tests, their testing conditions are thoroughly analyzed using numerical simulation. Numerical models including clamping devices have been built to investigate the influence of this clamping device at the early stage of loading. A limited effect was found for various imperfect testing conditions such as the clamping device stiffness, non-homogeneous stress and strain fields, non-equilibrium state, etc. On the contrary, numerical and analytical study shows that the simple small strain assumption usually used in double shear tests are not accurate enough. Eulerian cumulated strain definition should be used to get consistent numerical results. From this finding, the experimental rate sensitivity obtained for the SMAT treated AISI304 stainless steel sheet are recalculated.(3) A multi-layers elastic plastic damageable constitutive model is proposed to model SMAT treated AISI304 stainless steel sheet. The parameters are identified using tensile testing results. The elastic plastic behavior is curve fitted with a simple Ludwig hardening model. However, the damage parameters should be identified using an inverse method on the basis of numerical simulation of these tensile tests. In order to validate this multi-layer elastic plastic damageable constitutive model, indentation/piercing tests on SMAT treated AISI304 stainless steel sheet are performed. Numerical simulation of this indentation/piercing tests is also realized. It is found that the identified multi-layer elastic plastic damageable constitutive model allows for a quite accurate prediction of the experimental piercing tests.(4) In order to evaluate the impact anti-piercing capacity of the SMAT treated AISI304 stainless steel sheet, the impact perforation tests using Hopkinson bar are carried out. Numerical simulation of these impact perforation tests are realized with a similar FEM model as the quasi-static case. As the rate sensitivity of SMAT treated AISI304 stainless steel sheet is experimentally characterized with double shear test, a rate sensitive multi-layer elastic plastic damageable constitutive model is introduced. The numerical results agree well with the experimental ones, which indicates the effectiveness of the numerical model as well as the rate sensitive multi-layer elastic plastic damageable constitutive model.