This PhD research project mainly examined the important yet incompletely understood field of the mechano-sorptive creep of softwood. It included the acquisition of accurate creep data during constant and cyclic moisture conditions. The study only concerns the loading of softwood in the longitudinal, parallel to the grain direction in bending mode. Mechano-sorptive creep deformation takes place when the wood is subjected to load in varying humidity conditions, where the creep deformation is substantially higher than the creep deformation under constant humidity conditions. Visco-elastic creep is defined as the creep strain that is primarily dependent on time. Further analysis on the pseudo-creep, which is reversible with moisture changes, was also investigated. The Young's Modulus (E) and density (p) of the samples used in this research were measured and the ratio of (E/p) was used to characterise the quality of the wood specimen. A program was written in Quick-Basic for the computer to take readings from the creep machines at all times during all creep tests for acquiring maximum creep data possible for obtaining better results. A 'steaming' technique was implemented to accelerate creep recovery, in order to allow the re-use of the same sample for repeated creep testing, so reducing the effects of variability. Measurements of visco-elastic creep, mechano-sorptive creep limit and pseudo-creep rate as functions of percentage relative humidity and E/p were investigated and analysed. The mechano-sorptive creep limit was successfully reached by the load-reduction method. The pseudo-creep rate was similar for samples with the same values of E/p at different relative humidity. The higher the relative humidity, the larger the magnitude of visco-elastic and the 'reference' creep at the mechano-sorptive creep limit. A higher value of E/p produced lower pseudocreep rate, visco-elastic creep and mechano-sorptive creep limit. Equations were derived from the one week visco-elastic creep experimental data in order to extrapolate visco-elastic creep to longer time durations, (medium, long and permanent periods, and compared to the new European code of practice for timber design, namely the Draft Eurocode 5), and at two critical relative humidity conditions, (63% and 90%rh), which correspond to the Service Classes 1 and 2 in the Draft Eurocode 5. Also, an equation to predict the mechano-sorptive creep limit at various values of E/p for the two critical relative humidity conditions, (63% and 90%rh), was acquired. The prediction of creep at various relative humidities along the pseudo creep line was also quantified. The extrapolated visco-elastic creep along with the total upper-bound creep values were compared to the creep allowances of the Draft Eurocode 5 for both Service Classes 1 and 2. Moreover, a design scheme for estimating the upper-bound creep, consisting of the addition of visco-elastic, mechano-sorptive and pseudo creep for Draft Eurocode 5 Service Classes 1 and 2 was achieved. A study of the degree of interaction between visco-elastic and mechanosorptive creep was further investigated. Two different experiments were performed to reach some conclusion. The results of the first experiment showed that mechanosorptive creep acceleration decreases the visco-elastic creep rate and there was some kind of interaction between the two creep mechanisms. Hence the two creep mechanisms should not be considered as a separate phenomenon. On the other hand, the results of the second experiment showed that the visco-elastic and mechanosorptive creep are not a single process but are related processes. Moreover, it was deduced that the mechano-sorptive creep does not depend on the accumulation of the visco-elastic creep rate.

Proefschrift Universiteit van Amsterdam. / Met lit. opg. - Met samenvatting in het Nederlands.

L'objectif de notre étude est de mettre en évidence les synergies existantes entre les différents facteurs de vieillissement lors de la dégradation d’un revêtement polymère modèle, représentatif d’une peinture anticorrosion.Notre travail a consisté à suivre le vieillissement hygrothermique du système DGEBA/TETA sous forme de film libre et de revêtement déposé sur substrat acier dans différentes solutions (eau pure et NaCl 3%) pour différentes températures et contraintes mécaniques visco-élastique appliquées. Nos résultats ont montré que le programme de cuisson affectait les propriétés physico-chimiques et mécaniques du système initial et que l’absorption d’eau, non homogène, suivait une loi non-Fickienne, suite à un phénomène de gonflement.La prise en eau et la diffusion ont été caractérisées par gravimétrie et spectroscopie d’impédance électrochimique. Nos résultats montrent que les propriétés barrière du revêtement diminuent avec la quantité d’eau absorbée et avec la température de vieillissement. Cependant, la contrainte plane appliquée sur le système tend à contrarier cet effet. La formalisation de nos observations expérimentales dans le cadre d’une approche thermodynamique a montré l’effet prépondérant de la partie entropique par rapport à la partie enthalpique de l’énergie d’activation des processus de transport de l’eau, en raison de l’orientation des chaînes sous l’action de la contrainte. De plus, une diminution de la permittivité relative du revêtement sec a été observée en fonction de la température et de la contrainte appliquée, indépendamment de son signe (tension ou compression). Dans ce cas, la contribution enthalpique semble être prépondérante. / The objective of this study is to highlight synergies existing between different ageing factors during the degradation of a polymer model coating system, representing an anti-corrosion coating.We first monitored the hygrothermal ageing of DGEBA/TETA system as free films and as attached films onto a steel substrate in different solutions (water and NaCl 3wt% solution) for different temperatures and applied mechanical visco-elastic stresses. Our results showed that the curing programs affected the physic-chemical and mechanical properties of the cured systems and that the non-homogeneous water uptake process followed a non-Fickian law, due to a swelling phenomenon.The water uptake and diffusion were characterized by gravimetry and electrochemical impedance spectroscopy. The results indicate that the barrier properties of the coating decrease with the quantity of absorbed water in system and with the increasing of the ageing temperature. However, the planar stress applied onto the system tends to counteract this effect. The formalization of our experimental observations within a thermodynamic approach showed the predominant effect of the entropic part in comparison with enthalpic part of the activation energy of the water uptake process, due to the reorientation of polymer chains under the applied stress. In addition, a decrease in the relative permittivity of the initial dry coating was observed as a function of the temperature and the applied stress, regardless of its sign (tension and compression). In this case, the enthalpic contribution seems to be predominant.

Durant la phase de construction des puits pétrolier, le ciment est coulé dans l'espace annulaire entre le cuvelage et la roche environnante. La gaine de ciment a pour but de tenir le cuvelage, garantir l'étanchéité des puits pétroliers, ou des réservoirs de stockage de CO2, et de protéger le cuvelage de la corrosion. Au cours de la vie du puits, cette gaine de ciment est soumise tout le long du puits à des sollicitations thermiques et mécaniques qui varient au cours du temps et qui peuvent modifier ses propriétés et altérer son étanchéité. L'objet de cette thèse est d'étudier l'effet de la température et des contraintes sur les propriétés mécaniques de la pâte de ciment en cours de prise et de la pâte de ciment durcie. L'approche est basée sur une étude expérimentale qui combine des essais calorimétriques, des mesures de vitesse des ondes et des essais oedométriques avec le système des cellules STCA (Slurry To Cement Analyzer) sur le ciment en cours de prise ainsi que des essais de compression uniaxiale et triaxiale sur la pâte de ciment durcie. Les résultats expérimentaux ont montré que la température et la pression accélèrent la cinétique d'hydratation et que la température affecte significativement les propriétés élastiques du matériau tandis que la pression ne les influence pas pour une gamme de pression limitée à 20MPa. Une modélisation de la cinétique d'hydratation associée à une technique d'homogénéisation est proposée afin d'interpréter les essais. On a mis aussi en évidence que lorsque la prise se fait sous contraintes mécaniques, des déformations irréversibles peuvent se développer dans la pâte de ciment, ce qui peut conduire à la formation d'un micro-annulaire entre la gaine ciment, le cuvelage et la formation géologique. Le comportement différé de la pâte de ciment durcie a été étudié à partir d'essais de fluage uniaxiaux et d'essais de compression isotrope. Les essais ont mis en évidence que le fluage sous chargement uniaxial est plus important pour un ciment hydraté à une température plus élevée, ce qui est attribué à une microstructure plus hétérogène. Un modèle visco-endommageable permettant de reproduire les phases de fluage primaire et tertiaire a été développé. Les essais de compression isotrope drainés et non-drainé isothermes sous forte contrainte ont montré un comportement différé avec hystérésis lors de cycles décharge-recharge. Ces essais ont été analysés à partir d'un modèle de comportement poro-visco-plastique. Le comportement élastoplastique à court terme a été abordé à l'aide des essais triaxiaux sous chargement déviatorique drainé. Ces essais ont mis en évidence que la température d'essai affecte fortement la surface de charge initiale et les déformations tandis qu'elle ne modifie pas significativement la surface de rupture. Un modèle de plasticité avec une surface de charge fermée et un écrouissage dépendant des déformations plastiques accumulées a été développé pour décrire ces essais. Enfin, une étude préliminaire sur les effets des cycles mécaniques et thermiques a été menée. Des cycles thermiques ne dépassant pas la température d'hydratation ne semblent pas affecter les propriétés mécaniques du matériau. Cependant, une dégradation très rapide avec le nombre de cycles mécaniques a été mise en évidence lorsque la contrainte dépasse 60% de la résistance en compression simple / During the construction phase of oil wells, a cement slurry is pumped into the annular space between the casing and the rock. The cement sheath aims to support the casing, provide zonal isolation in the well and reservoirs of CO2 storage and protect the casing against corrosion. During the life of the well, the cement is submitted to various thermal and mechanical solicitations along the well that can modify its mechanical properties and damage its sealing performance. The aim of this thesis is to study the effect of temperature and stresses on the mechanical properties of cement paste during hydration and in the hardened state. The used approach is based on an experimental study that combine the calorimetric tests, waves velocity measurement and oedometric tests in STCA system (Slurry To Cement Analyzer) on cement paste during hydration as well as the uniaxial and triaxial compression tests on hardened cement paste. The experimental results showed that temperature and pressure accelerate the kinetics of hydration. Temperature affects significantly the elastic properties of the material whereas the pressure does not modify them for a range of pressure limited to 20MPa. A hydration kinetics modelling associated to a homogenization method is used to interpret the tests. It is shown that for hydration under stress, the irreversible strains can evolve in the cement paste and conduct to the formation of a micro-annular between the cement sheath, the casing and geological formation. The time-dependent behaviour of hardened cement paste is studied using creep tests under uniaxial loading and also from the results of isotropic compression tests. The results show that the uniaxial creep is more important for cements hydrated at higher temperatures, which is attributed to a more heterogeneous microstructure. A visco-damaged model allowing to reproduce the primary creep and tertiary creep is developed and calibrated. Isothermal isotropic drained and undrained compression tests show a time-dependent behaviour with hysteresis during unloading-reloading cycles. These tests are analyzed on the basis of a poro-visco-plastic model. The elastoplastic behaviour in short terms is studied from triaxial tests under drained deviatoric loading. These tests bring to light that the test temperature affects highly the initial yield surface and the strains but it does not significantly modify the failure surface. A model of plasticity with a closed yield surface and hardening, depending on the accumulated plastic strains is developed to describe these tests. Finally, a preliminary study on the effect of mechanical and thermal loading cycles is performed. The thermal loading cycles with temperatures lower than the hydration temperature seem to do not affect the mechanical properties of the material. The mechanical loading cycles show a rapid degradation with the number of loading cycles when the axial stress exceeds 60% of the uniaxial strength

A obtenção de dados de viscosidade, é fundamental para o projeto de equipamentos, tubulações e otimização de processos, já que esta propriedade afeta tanto a perda de energia por atrito em fluidos, como os mecanismos de transferência de calor e de massa. O presente trabalho de Mestrado teve como objetivo medir experimentalmente dados de viscosidade para misturas formadas por ácidos graxos comerciais e solventes alcoólicos de diferentes tamanhos de cadeia carbônica (etanol, 2-propanol, 1-butanol, 1-pentanol e 1-hexanol), com ou sem a adição de água, variando-se a temperatura desde o ponto de fusão do composto graxo até o ponto de ebulição de cada solvente. Estes dados foram agrupados aos dados experimentais de viscosidades de misturas de triacilgliceróis (TAG) e ácidos graxos (AG) já disponíveis na literatura, para a obtenção de parâmetros de interação entre os grupos funcionais formadores dos componentes presentes nos sistemas estudados, utilizando-se o modelo de contribuição de grupos UNIFAC-VISCO. Em uma primeira abordagem, apenas os parâmetros inexistentes na literatura foram ajustados, obtendo-se um erro relativo médio igual a 10,21 %. A modelagem também foi realizada ajustando-se todos os parâmetros de interação, incluindo os já publicados, obtendo-se nesta segunda abordagem, um erro relativo médio igual a 7,54 %. A capacidade preditiva dos parâmetros obtidos foi testada em misturas semelhantes, formadas por óleos vegetais ou gordura animal, ácidos graxos e solventes alcoólicos, cujas viscosidades também foram determinadas neste trabalho, porém não fizeram parte do banco de dados utilizado na modelagem. Os resultados obtidos foram comparados aos de outros modelos para o cálculo de viscosidades de misturas conhecidos da literatura, dentre eles, a Regra de Kay, o Modelo de Kendall e Monroe e o Modelo CG-UNIMOD. Para fins comparativos, este conjunto de dados também foi ajustado a equação de Grunberg-Nissan, da qual foram obtidos parâmetros de interação entre os componentes das misturas. Os testes preditivos mostraram que o modelo UNIFAC-VISCO proporcionou os menores erros relativos para a maioria das misturas estudadas, porém ainda bastante elevados em alguns casos. Por outro lado, o Modelo de Grunberg-Nissan proporcionou erros relativos bem menores, o que era esperado, uma vez que as viscosidades de cada sistema foram correlacionadas pelo modelo. / The acquisition of viscosity data is critical for the design of equipments and pipes, as well as for the process optimization, since this property affects both the energy loss by friction in fluids, such as the mechanisms of mass and heat transfer. The present work aimed the measurement of experimental viscosity data for mixtures containing commercial fatty acids and alcoholic solvents of different sizes of carbon chain (ethanol, 2-propanol, 1-butanol, 1-pentanol and 1-hexanol), with or without the addition of water, varying the temperature from the melting point of the fatty compound to the boiling point of each solvent. These data were grouped together to experimental viscosity data of triacylglycerols (TAG) and fatty acids (FA) mixtures that were already available in the literature, to obtain interaction parameters between the forming functional groups of the components present in the systems studied, using the group contribution model UNIFAC-VISCO. In a first approach, only the parameters not available in the literature were adjusted obtaining an average error equal to 10.21%. The modeling was also performed by adjusting all interaction parameters, including those already published, obtaining in this second approach, an average error equal to 7.54%. The predictive ability of parameters obtained was tested in similar mixtures containing vegetable oils or animal fats, fatty acids and alcoholic solvents, whose viscosities were also determined in this work, but were not part of the database used in the modeling. The obtained results were compared to other known models for the calculation of viscosities of mixtures, such as the Kay\'s Rule, the Kendall and Monroe model and the CG-UNIMOD. For comparative purposes, this data set was also adjusted to the Grunberg-Nissan equation, in which interaction parameters between the components of mixtures were obtained. The predictive tests showed that the UNIFAC-VISCO model provided the lowest relative errors for most of the mixtures studied, but still quite high in some cases. On the other hand, the Grunberg-Nissan model provided relative errors much smaller, which was expected, since the viscosities of each system were correlated by the model.

A obtenção de dados de viscosidade, é fundamental para o projeto de equipamentos, tubulações e otimização de processos, já que esta propriedade afeta tanto a perda de energia por atrito em fluidos, como os mecanismos de transferência de calor e de massa. O presente trabalho de Mestrado teve como objetivo medir experimentalmente dados de viscosidade para misturas formadas por ácidos graxos comerciais e solventes alcoólicos de diferentes tamanhos de cadeia carbônica (etanol, 2-propanol, 1-butanol, 1-pentanol e 1-hexanol), com ou sem a adição de água, variando-se a temperatura desde o ponto de fusão do composto graxo até o ponto de ebulição de cada solvente. Estes dados foram agrupados aos dados experimentais de viscosidades de misturas de triacilgliceróis (TAG) e ácidos graxos (AG) já disponíveis na literatura, para a obtenção de parâmetros de interação entre os grupos funcionais formadores dos componentes presentes nos sistemas estudados, utilizando-se o modelo de contribuição de grupos UNIFAC-VISCO. Em uma primeira abordagem, apenas os parâmetros inexistentes na literatura foram ajustados, obtendo-se um erro relativo médio igual a 10,21 %. A modelagem também foi realizada ajustando-se todos os parâmetros de interação, incluindo os já publicados, obtendo-se nesta segunda abordagem, um erro relativo médio igual a 7,54 %. A capacidade preditiva dos parâmetros obtidos foi testada em misturas semelhantes, formadas por óleos vegetais ou gordura animal, ácidos graxos e solventes alcoólicos, cujas viscosidades também foram determinadas neste trabalho, porém não fizeram parte do banco de dados utilizado na modelagem. Os resultados obtidos foram comparados aos de outros modelos para o cálculo de viscosidades de misturas conhecidos da literatura, dentre eles, a Regra de Kay, o Modelo de Kendall e Monroe e o Modelo CG-UNIMOD. Para fins comparativos, este conjunto de dados também foi ajustado a equação de Grunberg-Nissan, da qual foram obtidos parâmetros de interação entre os componentes das misturas. Os testes preditivos mostraram que o modelo UNIFAC-VISCO proporcionou os menores erros relativos para a maioria das misturas estudadas, porém ainda bastante elevados em alguns casos. Por outro lado, o Modelo de Grunberg-Nissan proporcionou erros relativos bem menores, o que era esperado, uma vez que as viscosidades de cada sistema foram correlacionadas pelo modelo. / The acquisition of viscosity data is critical for the design of equipments and pipes, as well as for the process optimization, since this property affects both the energy loss by friction in fluids, such as the mechanisms of mass and heat transfer. The present work aimed the measurement of experimental viscosity data for mixtures containing commercial fatty acids and alcoholic solvents of different sizes of carbon chain (ethanol, 2-propanol, 1-butanol, 1-pentanol and 1-hexanol), with or without the addition of water, varying the temperature from the melting point of the fatty compound to the boiling point of each solvent. These data were grouped together to experimental viscosity data of triacylglycerols (TAG) and fatty acids (FA) mixtures that were already available in the literature, to obtain interaction parameters between the forming functional groups of the components present in the systems studied, using the group contribution model UNIFAC-VISCO. In a first approach, only the parameters not available in the literature were adjusted obtaining an average error equal to 10.21%. The modeling was also performed by adjusting all interaction parameters, including those already published, obtaining in this second approach, an average error equal to 7.54%. The predictive ability of parameters obtained was tested in similar mixtures containing vegetable oils or animal fats, fatty acids and alcoholic solvents, whose viscosities were also determined in this work, but were not part of the database used in the modeling. The obtained results were compared to other known models for the calculation of viscosities of mixtures, such as the Kay\'s Rule, the Kendall and Monroe model and the CG-UNIMOD. For comparative purposes, this data set was also adjusted to the Grunberg-Nissan equation, in which interaction parameters between the components of mixtures were obtained. The predictive tests showed that the UNIFAC-VISCO model provided the lowest relative errors for most of the mixtures studied, but still quite high in some cases. On the other hand, the Grunberg-Nissan model provided relative errors much smaller, which was expected, since the viscosities of each system were correlated by the model.

Motivado por diversos processos de manufatura, tais como conformação de metais a frio ou mesmo manufatura aditiva, este trabalho consiste no desenvolvimento de um código computacional para a simulação numérica de problemas bidimensionais que abordam três tipos de não-linearidade: a geométrica, presente em situações de grandes deslocamentos; a física, presente no modelo constitutivo do material; e a de contato. Na primeira etapa, desenvolve-se um programa para análise dinâmica bidimensional de sólidos elásticos, utilizando a abordagem posicional do método dos elementos finitos, que engloba naturalmente a não-linearidade geométrica em sua formulação. Em seguida, implementam-se modelos constitutivos não-elásticos para problemas com grandes deformações. No modelo elastoplástico, adota-se o critério de von Mises com encruamento cinemático baseado na lei de Armstrong-Frederick. Essa formulação é então generalizada para o caso visco-plástico, onde é considerado o modelo de Perzyna em conjunto com a lei de Norton. No caso visco-elástico, utiliza-se uma formulação que parte do modelo reológico de Zener. Por fim, apresenta-se um modelo visco-elasto-plástico que consiste no acoplamento dos modelos visco-elástico e visco-plástico descritos anteriormente. Em todos os casos, utiliza-se a decomposição multiplicativa do gradiente da função mudança de configuração. Com respeito à aplicação 2D, consideram-se as hipóteses de estado plano de deformações e estado plano de tensões, onde a última é resolvida numericamente por um procedimento local de Newton-Raphson. Para o problema de contato, aplica-se a estratégia Nó-a-Segmento, sendo as condições de não-penetração impostas com a introdução de multiplicadores de Lagrange. A formulação é testada em cada uma das etapas por meio de exemplos numéricos de verificação. Além disso, para mostrar as potencialidades do código desenvolvido, são propostos diversos exemplos numéricos, sendo alguns inspirados por processos de manufatura existentes. Nesses exemplos, são estudados os efeitos de diferentes parâmetros dos materiais e diferentes taxas de deformação na resposta numérica, permitindo uma análise do comportamento dissipativo decorrente da plastificação e da viscosidade, incluindo a influência desses sobre o amortecimento dinâmico. / Motivated by several manufacturing processes, such as cold metal forming or even additive manufacturing, in this work we develop a computational code for numerical simulation of two-dimensional problems addressing three types of nonlinearities: geometric nonlinearity, present in large displacements situations; physical non-linearity, present in the material constitutive model; and contact non-linearity. In the first step, we develop a computational program for dynamic analysis of two-dimensional elastic solids using the positional finite element method, which naturally takes into account geometric non-linearity in its formulation. Following, we implement inelastic constitutive models for large strain problems. In the elasto-plastic model, we adopt von Mises yeld criteria and kinematic hardening based on the Armstrong-Frederick law. The formulation is then generalized to the visco-plastic case, where we consider Perzyna model associated with Norton\'s law. In the visco-elastic case, Zener\'s rheological model is employed. Finally, we present a visco-elasto-plastic model by coupling the visco-elastic and visco-plastic models described previously. In every case, the multiplicative decomposition of the deformation gradient is employed. Regarding the 2D application, we consider both plane strain and plane stress hypothesis, where the latter is solved numerically by a local Newton-Raphson procedure. For the contact problem, we employ the Node-to-Segment strategy, imposing non-penetration conditions with the introduction of Lagrange multipliers. The resulting computational code is tested in each step by means of numerical verification examples. In addition, to show the potentialities of the developed code, several numerical examples are proposed, some of which inspired by existing manufacturing processes. On these examples, we study the effects of different material parameters and strain rates on the numerical response, allowing an analysis of the dissipative behavior due to plasticity and viscosity, including the influence of these on the dynamic damping.

Hot mix asphalt (HMA) is a granular composite material stabilized by the presence of asphalt binder. The behavior of HMA is highly influenced by the microstructure distribution in terms of the different particle sizes present in the mix, the directional distribution of particles, the distribution of voids, and the nucleation and propagation of cracks. Conventional continuum modeling of HMA lacks the ability to explicitly account for the effect of microstructure distribution features. This study presents the development of elastic and visco-plastic models that account for important aspects of the microstructure distribution in modeling the macroscopic behavior of HMA. In the first part of this study, an approach is developed to introduce a length scale to the elasticity constitutive relationship in order to capture the influence of particle sizes on HMA response. The model is implemented in finite element (FE) analysis and used to analyze the microstructure response and predict the macroscopic properties of HMA. Each point in the microstructure is assigned effective local properties which are calculated using an analytical micromechanical model that captures the influence of percent of particles on the microscopic response of HMA. The moving window technique and autocorrelation function are used to determine the microstructure characteristic length scales that are usedin strain gradient elasticity. A number of asphalt mixes with different aggregate types and size distributions are analyzed in this paper. In the second part of this study, an elasto-visco-plastic continuum model is developed to predict HMA response and performance. The model incorporates a Drucker- Prager yield surface that is modified to capture the influence of stress path direction on the material response. Parameters that reflect the directional distribution of aggregates and damage density in the microstructure are included in the model. The elasto-visco-plastic model is converted into a numerical formulation and is implemented in FE analysis using a user-defined material subroutine (UMAT). A fully implicit algorithm in time-step control is used to enhance the efficiency of the FE analysis. The FE model used in this study simulates experimental data and pavement section.

It is well known that surface water waves interact with fluid mud on the sea bed. Wave mud interaction results in high wave energy dissipation and mud mass transport. This kind of wave energy dissipation, which generally is much more significant than wave dissipation due to bottom friction, should be considered in the simulation of wave evolution and transformation in muddy coastal environments. In this research, a two-dimensional horizontal wave propagation and morphodynamic model for muddy coasts was developed. The model can be applied on a general three dimensional bathymetry of a soft muddy coast to calculate wave damping, fluid mud transport and resulting bathymetry change under wave action. In addition to the effect of wave-mud interaction on wave propagation, the dissipation due to wave-mud interaction was also implemented in SWAN (a third generation numerical model for Simulating WAves Nearshore) using a multilayered wave mud interaction model. These two models combined can be used for generation and propagation of waves in muddy coastal areas. The nonlinear constitutive equations of the visco-elastic-plastic model are adopted for the rheological behavior of fluid mud in this research. The results of the numerical model are compared against a series of wave-basin experiments, wave flume experiments and field observations. Comparisons between the simulated results with the both field and laboratory data reveal the capability of the proposed model to predict the wave transformation and mud mass transport. / Thesis (Ph.D, Civil Engineering) -- Queen's University, 2009-07-24 11:18:18.622

L’intégration 3D est la suite de l’évolution de la microélectronique vers la miniaturisation à haute performance. Les pièces à pas fins permettent la conception de produits dits 3D (ou 2.5D) qui maximisent l’espace pour les données et accélèrent les transferts, tout en minimisant l’espace physique et l’énergie requise par le dispositif. Le procédé de fabrication le plus largement utilisé est celui de puces à protubérances inversées (ou flip-chip) et il implique l’alignement de la puce sur son substrat à l’aide d’une tête automatisée. Le procédé de puces inversées est particulier du fait que les puces comportent une matrice d’interconnexions couvrant la majorité de leur surface, en comparaison au procédé de fils soudés (wire bond) où les connexions sont en périphérie des puces. Elles doivent donc être renversées pour être placées pour la soudure. Le rôle primaire du flux est de s’assurer que les boules de soudure sont bien désoxydées avant l’étape de brasure afin de garantir la qualité et la conductivité de cette dernière. L’objectif de la recherche était de trouver un flux trempable qui satisfasse aux exigences des produits à pas fins soudés par thermocompression. Pour ce faire, les flux candidats ont été caractérisés tant sur le volet physico-chimique que sur le volet des phénomènes visco-inertiels en trempage. Cette caractérisation a permis de prédire le comportement des flux en trempage pour leur mouillabilité et la quantité de flux retirée, ainsi que la température où survenait un changement structurel important dans le flux, indiquant le moment où la réaction chimique de désoxydation culminait. La partie sous-jacente de l’objectif du projet devait aussi s’assurer de la compatibilité du nouveau flux avec les étapes d’encapsulation suivantes. Dans un tel procédé, cela signifiait la propreté après lavage, la compatibilité avec l’agent de remplissage époxy qui protège les pièces, l’ajout du capot métallique et les tests de fiabilité. Des trois candidats de flux testés, un seul a été choisi pour valider la méthode de trempage développée au travers des tests de fiabilités. Ces tests n’ont démontré aucune défaillance électrique et une fiche parfaite pour le test accéléré en humidité et température sous tension (HAST). Le test en cyclage thermique a fait apparaître des défauts mineurs sur 12% des pièces du flux choisi et d’un flux témoin.