En raison de leur porosité élevée, les roches carbonatées forment d'importants réservoirs d'eau et d'hydrocarbures, et conviennent également à d'autres applications telles que le stockage de CO2 ou des déchets nucléaires. Cependant, le compactage localisé dans les roches carbonatées affecte le champ des contraintes et les propriétés hydromécaniques de ces roches, entraînant des déformations inélastiques et des dommages avec des impacts économiques, environnementaux et sociaux potentiels. Des études antérieures sur le terrain et des études expérimentales en laboratoire que dans les roches carbonatées poreuses, contrairement aux grès, une variété de micromécanismes tels que l'effondrement des pores, le broyage des grains, la petre de cimentation, la plasticité des cristaux et la dissolution sous contrainte peuvent entraîner une densification inélastique. En raison de la coexistence de ces multiples processus inélastiques et de leur interactions, le micromécanisme prédominant à l'origine du processus de localisation et de rupture reste mal compris.Cette thèse de doctorat présente une étude expérimentale sur les mécanismes de déformation régissant le comportement mécanique et le mode de rupture des roches carbonatées à haute porosité. A cet effet, le Tuffeau de Maastricht, un calcaire sédimentaire bioclastique présentant jusqu'à 52% de porosité, a été testé à l'état sec. Cette étude s'est concentrée sur la façon dont le chemin des contraintes, la pression de confinement et l'orientation de la stratification influent sur l'apparition et la propagation des bandes de compaction. Trois campagnes expérimentales principales sont menées sur des éprouvettes cylindriques pour étudier la transition fragile-ductile :compression isotrope,compression uniaxiale et compression triaxiale. Une analyse systématique du comportement anisotrope du Tuffeau de Maastricht est effectuée sur des échantillons qui ont été forés perpendiculairement, obliquement 45° et parallèlement au plan de la stratification. La micro-tomographie à rayons X à haute résolution est utilisée pour obtenir des images 3D de l'ensemble de l'échantillon en cours de chargement. Les images acquises sont traitées et des mesures de plein champ ont été utilisées pour élucider les mécanismes d'initiation et de propagation des zones de compactage localisées. Les variations de porosité pendant le chargement sont mesurées macroscopiquement et localement. L'utilisation systématique des rayons X combinée à l'utilisation de l'analyse d'image avancée et de la corrélation d'images numérique fournit une information 3D quantitative du champ de déformation et de son évolution pendant un test.Deux modes de rupture sont identifiés, selon les mesures de porosité et la CIVD: les bandes de cisaillement contractantes développées à basse pression de confinement et les bandes de compaction formées perpendiculairement à la contrainte principale majeure de compression à un confinement plus élevé. Ces bandes se développent à contrante déviatoire presque constante et se propagent dans tout l'échantillon, phase marquée par des chutes épisodiques de la contrainte deviatoire. Les essais de compression triaxiale conduit à des niveux de déformation axiale plus élevée présentent trois phases distinctes: une phase initiale quasi-linéaire, suivie par un plateau de contrainte déviatoire, enfin un durcissement post-plateau. L'observation essentielle de ces expériences est l'existence d'un mécanisme de dégradation del la cimentation entre grains qui transforme l'échantilon cohésif et frottant en un miliu granulaire purement frottant. D'autres expériences réalisées sur du sable obtenu par dégradation artificielle de la roche originelle mettent l'accent sur cette phase de déstructuration qui se produit lors du plateau de la contrainte déviatoire. Les résultats expérimentaux mettent également en évidence la forte influence due à l'anisotropie sur le comportement mécanique du matériau étudié. / Given their high porosity, carbonates form important water and hydrocarbons reservoirs, and they are also suitable for other applications such as CO2 storage and nuclear waste disposal. However, localized compaction in carbonates affects the stress field and the hydromechanical properties of these rocks leading to inelastic deformation and failure with potential economic, environmental and social impacts. Previous field and experimental studies have shown that in porous carbonates, unlike sandstones, a variety of micromechanisms such as pore collapse, grain crushing, debonding, crystal plasticity and pressure solution can potentially lead to inelastic compaction. Due to the coexistence of such multiple inelastic processes and the interplay among them, the dominant micromechanism responsible for failure remains poorly understood.This doctoral thesis presents an experimental investigation into the deformation mechanisms governing the mechanical behavior and failure mode of high porosity carbonate rocks. To this end, Tuffeau de Maastricht, a bioclastic sedimentary limestone exhibiting up to 52% porosity, has been tested under dry conditions. This study focuses on how stress path, confining pressure and bedding orientation affect the onset and propagation of localized compaction. Three main experimental campaigns are conducted on cylindrical specimens of 11.5 mm diameter and 22 mm height to study the brittle-ductile transition: (i) isotropic compression, (ii) uniaxial compression, and (iii) triaxial compression tests at confining pressures ranging from 1 to 5 MPa. A systematic analysis of the anisotropic behavior of Tuffeau de Maastricht is conducted on samples cored perpendicular, oblique 45° and parallel to the bedding plane. High resolution x-ray computed tomography (CT) is used to obtain 3D images of the entire specimen under loading. The acquired images are processed and full-field measurements have been used to elucidate the mechanics of initiation and propagation of localized compaction. Porosity variations during loading are measured macroscopically as well as locally. The porosity measurements are performed over a REV, which is defined with the use of statistical tools. The systematic use of x-ray micro tomography combined with the use of advanced image analysis and Digital Image Correlation (DIC) provides a quantitative 3D information the strain field inside a sample and its evolution during a test.Two failure modes are identified, based on porosity measurements and DIC: compactive shear bands at low confining pressure, and compaction bands (perpendicular to the maximum compressive stress) at higher confinement. These bands develop at essentially constant deviator stress and propagate through the whole sample punctuated by episodic stress drops. Triaxial compression tests at much higher axial strain present three distinct phases: (1) an initial quasi-linear increase of deviator stress, followed by (2) a plateau and (3) a post-plateau hardening. The essential observation from these experiments if the occurrence of a debonding phase which converts the specimen from rock-like to sand-like. A second localization, typical of dense sand, eventually occurs for very axial strain. Additional experiments that are performed on artificially debonded specimens emphasize this destructuration phase during the plateau of deviator stress. The experimental results also highlight the strong anisotropy of the mechanical behavior of the studied material.

Made available in DSpace on 2016-04-28T20:38:11Z (GMT). No. of bitstreams: 1 Elizabeth Medeiros Pacheco.pdf: 61847230 bytes, checksum: d35038669cedb4e1e85e9a0778f42e95 (MD5) Previous issue date: 2012-05-28 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The research seeks to accommodate, without hierarchy, the resonances between the aesthetic experimentation, the investigations of the physics and the corporal physiology of the affects from a possible dialogue of their speeches. This seems to be a political gamble in the mixing between this fields that arises since the dawn of the XX century with the crisis of fundamentals. The XIX century had already removed God from the monopoly of the Other. Now subatomic physics faces the uncertainty of the substance, and man is more and more dissonant with all the knowledge about himself, since the positive in the field of cognition becomes intolerable in the field of his experience. As the breaking of identity patterns extending from the borders of the countries till the memory of the cells, the differences are no more treated as essences. The difference begins to be addressed as what is produced between the bodies; what become interesting are the processes of differentiation. How to ensure that "ephemeral I will be stronger than everything that the eternal had always been in me? (Lucio Cardoso). This is the most alive question ever, not by the body, but by the thought / A pesquisa busca acolher, sem hierarquia, as ressonâncias entre a experimentação estética, as investigações da física e a fisiologia corporal dos afetos a partir de uma possível interlocução de seus discursos. Essa parece ser uma aposta política da mestiçagem entre campos que se anuncia desde a aurora do séc. XX com a crise dos fundamentos. O séc. XIX já havia destituído Deus do monopólio do Outro, agora a física subatômica injeta incerteza na substância, e o homem contraindo só dissonância com os saberes sobre si, já que o positivo no regime do entendimento se torna intolerável no regime da experiência. Com a quebra dos padrões identitários se estendendo das fronteiras dos países à memória das células, as diferenças deixam de ser essências para serem abordadas como aquilo que se produz entre os corpos; o que passa a interessar são os processos de diferenciação. Como garantir que efêmero serei mais forte do que tudo o que eterno foi sempre em mim (Lucio Cardoso)? Esta é a pergunta do poeta. Mais vivo do que nunca, não pelo corpo, mas pelo pensamento

O desenvolvimento de biomateriais com estrutura porosa visa permitir uma boa ancoragem biológica através do crescimento do tecido ósseo no interior dos poros, além de no caso de algumas ligas de titânio, proporcionar valores de módulo elástico semelhantes ao do osso, prevenindo assim o insucesso por reabsorção óssea na interface com o implante. O objetivo deste estudo foi avaliar o processamento de implantes porosos à base de titânio produzidos por metalurgia do pó, utilizando polímeros naturais como aditivos (albumina, amidos de arroz, batata, milho e a gelatina), e analisar a osteointegração desses implantes por ensaio in vivo. Foram obtidas, amostras de titânio comercialmente puro (Ticp) e da Liga Ti-13Nb-13Zr pelos processos de space-holder e suspensão; os polímeros naturais foram removidos termicamente (350°C/1h) previamente a sinterização (1300°C/1h e 1300°C/3h, respectivamente). As metodologias de processamento utilizando aditivos de baixo custo propiciaram a obtenção de implantes metálicos porosos de maneira simplificada, com porosidades diferenciadas, boa homogeneidade estrutural, grau de porosidade adequado (entre 40% e 60%), distribuição e tamanho dos poros conforme o polímero natural formador. Na caracterização biológica, o teste in vitro de citotoxicidade validou a utilização dos implantes para a realização do teste in vivo. A avaliação da osteointegração foi realizada em coelhos da raça New Zealand em períodos de 49 dias. Na análise histológica por microscopia eletrônica de varredura e fluorescência, foram obtidos os resultados quantitativos e qualitativos da osteointegração, com crescimento ósseo em todos os implantes, apresentando osteocondutividade. Os melhores resultados ocorreram em implantes com porosidade homogênea, e com tamanho médio de poros entre 100 e 200 μm. Assim, os materiais porosos de titânio e da liga Ti-13Nb-13Zr produzidos com os polímeros naturais (amido de arroz, amido de batata, amido de milho, gelatina e albumina), permitiram a nutrição e manutenção necessárias para a sobrevivência das células, comprovando a osteointegração e osteocondução dos implantes desenvolvidos. / The development of biomaterials with porous structure, seeks to allow a good biological anchorage through bone tissue ingrowth within the pores, besides to propitiate elastic modulus values similar to bone, thus preventing the failure by bone resorption, on implants interface. The goals of this study was to evaluate the processing of titanium based porous implants by powder metallurgy (PM), using natural polymers as additive (rice, potato and corn starches, albumin and gelatin), and evaluate these implants osseointegration by in vivo assay. Samples were obtained of commercially pure titanium (cpTi) and Ti-13Nb-13Zr alloy by space-holder and suspension techniques; the natural polymers were removed by thermal treatment (350C/1h) prior to sintering (1300C/1h and 1300C/3h respectively). The processing methodologies using low cost additives propitiate the production of porous metallic implants in a simplified manner, with differentiated porosities, good structural homogeneity, proper porosity degree (between 40% and 60%), distribution and pore size as the related natural polymer. The biological characterization was first performed in vitro with a cytotoxicity test, which allowed the use of implants in vivo. The osseointegration evaluation was performed in New Zealand White rabbits for 49 days period. Histological analyses with scanning electron microscopy and fluorescence microscopy were obtained quantitative and qualitative results of osseointegration, with bone growth in all implants, presenting osteoconductivity. The best results were in implants with homogenous porosity and with mean pore size between 100 and 200 μm. Thus, titanium and Ti-13Nb-13Zr alloy porous materials produced with natural polymers (rice, potato and corn starches, albumin and gelatin), allowed the nutrition and nourishing needed to bone cells survival, proving the osseointegration and osteoconduction features of developed implants.

O conhecimento de propriedades do solo ligadas diretamente à produtividade das culturas é uma busca incessante. As propriedades fortemente correlacionadas com o espaço poroso do solo tornam-se muito importantes, principalmente porque têm ação direta no desenvolvimento vegetal: é pelo espaço poroso que ocorre o deslocamento de água e ar para a rizosfera das plantas. Cada tipo de solo e mesmo cada horizonte pedológico possui uma geometria de poros que o caracteriza e que permite uma maior ou menor facilidade de transportar água e, consequentemente, o ar. Como as propriedades do solo relacionadas ao transporte de água e de ar são altamente variáveis no espaço, o objetivo deste trabalho foi estudar a variabilidade espacial: a) dos parâmetros da equação que correlaciona a condutividade hidráulica com o conteúdo de água no solo e b) da permeabilidade do solo ao ar. Para tanto, foi instalado um experimento no campo, num Latossolo Vermelho Amarelo textura média, constituído por 60 tubos de acesso a uma sonda de nêutrons, numa malha regular de 5 x 5 m, com a finalidade de medir o conteúdo de água em função do tempo, nas profundidades de 0,20; 0,40; 0,60; 0,80 m e, então, determinar-se a condutividade hidráulica pelo método do perfil instantâneo. A permeabilidade do solo ao ar (ka) foi determinada pelo método da pressão decrescente, realizada em laboratório, utilizando amostras de solo com estrutura indeformada, equilibradas nas tensões de 6 e 10 kPa. De posse desses atributos do solo, realizou-se o estudo da variabilidade espacial. Os valores mais elevados de condutividade hidráulica do solo saturado e de permeabilidade intrínseca do solo ao ar foram encontrados nas porções da área experimental com menores cotas altimétricas, o que indica uma influência da morfologia do terreno na estrutura do solo, principalmente porque nessas porções há uma maior quantidade de argila. Os mapas de predição mostram que a condutividade hidráulica do solo saturado está correlacionada diretamente com o conteúdo de água do solo saturado e inversamente com o coeficiente angular da reta do logaritmo da condutividade hidráulica em função do conteúdo de água no solo. Os mapas de predição das permeabilidades do solo ao ar foram semelhantes quanto à localização dos maiores e menores valores, mas houve um aumento das porções de área com valores mais elevados de ka(m=-10 kPa) em comparação ao de ka(m=-6 kPa), devido ao esvaziamento de poros com raios menores, que aumenta a conectividade e consequentemente o fluxo de ar. / The knowledge of soil properties directly related to crops yield is na incessant search. The soil properties strongly correlated with the soil porosity are very important for their direct action on plant development: it is through the soil porous space that occurs water and air displacement to plant rhizosfere. Each kind of soil and even each soil horizont have a pore geometry that the chracterizes and permits a higher or lower facility in transporting water and air. Since these soil properties are highly variable in space, the objective of this work was to study the spatial variability: a) of the parameters of the equation correlating soil hydraulic conductivity and soil water content and, b) of the soil air permeability. For this, 60 acess tubes for a neutron probe were installed in a medium texture Yellow Red Latosol, in a regular grid of 5 x 5 m, in order to measure the soil water content a long the time, in depths of 0.2; 0.4; 0.6; 0.8 m and to determine the hydraulic conductivity by instantaneous profile method. The soil air permeability was determined by the decreasing pressure method in laboratory, using undisturbed soil samples under stabilized matric soil water potentials of -6 and -10 kPa. The higher values of saturated hydraulic conductivity and soil air permeability were found in the lower altitudes of the experimental area, showing that topography can influence soil structure, due mainly the higher amount of clay in lower altitudes. The predict map of saturated hydraulic conductivity shows that this soil propertie is directly correlated with the saturated soil water content and inversely correlated with the slope of the curve of logarithm of hydraulic conductivity as a function of soil water content. Predict maps of soil air permeability were similar, since the higher and lower values were found in the same place. However, there were an increase of the area with higher air permeability, measured under -10 kPa, due to higher amount of air filled pores, increasing the porous connectivity and the air flow.

Neste trabalho investigamos a porosidade nanométrica de materiais lignocelulósicos derivados do bagaço de cana-de-açúcar que foram submetidos à compressão úmida. A produção dos materiais estudados a partir do bagaço de cana-de-açúcar utilizou processos de tratamento hidrotérmico seguido de processos de deslignificação organossolve (etanol-água) e soda (hidróxido de sódio). Os tratamentos hidrotérmicos utilizaram a fração fibra do bagaço de cana-de-açúcar, no estado bruto e moído, seguindo planejamento experimental fatorial de 2² com ponto central mais configuração estrela. O tratamento hidrotérmico ocorreu em temperaturas de 150-190°C com tempos de 20-60 minutos. Os processos de deslignificação utilizaram temperaturas de 160°C e 190°C para o processo soda e organossolve respectivamente, com tempos de 20, 40, 60, 80 e 100 minutos de tratamento. Os ensaios de compressão úmida foram realizados com cargas de 5, 10, 15, 20 e 25 toneladas e mostraram que materiais mais homogêneos e com menor granulometria, como o material moído e as polpas, são mais fáceis de comprimir. No entanto, os ensaios mostraram que os materiais comprimidos não são homogêneos, apresentando uma variação no teor de humidade do material comprimido (o centro da pastilha é mais seco que a periferia). O adensamento dos materiais também não é homogêneo, sendo o centro mais denso que a periferia das pastilhas. A perda de água durante a compressão foi de 74-85% para o material tratado hidrotermicamente, 66-85% para as polpas obtidas no processo soda e 81-94% para as polpas obtidas no processo organossolve. As análises de termoporometria mostraram que fração apreciável da porosidade nanométrica dos materiais deslignificados é colapsada com as menores pressões aplicadas (21 MPa). Incrementos de pressão (até carga de 107 MPa) promovem reduções comparativamente muito menores na porosidade nanométrica. / In this work we investigate the nanometric porosity of lignocellulosic materials derived from sugarcane bagasse and tested for wet pressing. The production of the studied materials from sugarcane bagasse employed hydrothermal treatments followed by organosolv (ethanol-water) and soda (sodium hydroxide) delignifications. For the hydrothermal treatments, we used bagasse fiber fractions in crude and milled states, following the factorial experimental design of 2² with central point plus star configuration. Hydrothermal treatment used temperatures of 150-190°C and times of 20-60 minutes. The delignification processes used temperatures of 160°C and 190°C for the soda and organosolv, respectively, with treatment times of 20, 40, 60, 80 and 100 minutes. Wet pressing was carried out with loads of 5, 10, 15, 20 and 25 tons and showed that materials with small and homogeneous particles such as the ground materials and the pulps are easier to compress and form a mass of material. However, the tests showed that the compressed materials are not homogeneous, presenting variations in the moisture content of the compressed materials (the center of the tablets were dryer than the periphery). Densification of the materials is also not uniform, the center being denser than the periphery. Water loss during compression was 74-85% for hydrothermally treated material, 66-85% for soda pulps and 81-94% for organosolv pulps. Thermoporometry analysis showed that appreciable fraction of the nanoscale porosity of the delignified materials collapse with the lowest applied pressures (21 MPa). Pressure increments (up to 107 MPa) promotes comparatively much lower reduction on nanoscale porosity.

Frente a um período no qual a escala de ação para muitos agentes sociais sobretudo as empresas hegemônicas no mercado globalizado corresponde ao mundo, os territórios nacionais são mobilizados e transformados de maneira a viabilizar a organização e eficiência dos circuitos espaciais produtivos internacionalizados. A criação dos Centros Logísticos e Industriais Aduaneiros (CLIAs) é, além de muitas outras, uma dessas adaptações do território. São recintos que exercem função aduaneira de despacho/admissão de mercadorias importadas ou a serem exportadas, mas que se situam fora da zona primária de portos e aeroportos, constituindo-se como alternativa às alfândegas tradicionais para a importação ou exportação de mercadorias. São, então, novos objetos geográficos que aumentam a porosidade territorial brasileira criando caminhos mais fluidos para as trocas internacionais, principalmente para alguns circuitos espaciais produtivos. A presença deles no estado de São Paulo liga-se diretamente ao processo de industrialização desse território uma vez que são utilizados em grande parte pela indústria química e pela indústria automotiva. Torna-se importante, assim, a individualização das relações estabelecidas entre os CLIAs e as indústrias às quais são funcionais. Por fim, além da participação dos CLIAs nesses circuitos espaciais produtivos, também podem implicar diferentes relações com o lugar onde são instalados, criando modalidades diversas de integração ao território verticais e horizontais. É justamente o entendimento dessa dinâmica territorial complexa que se constitui no objetivo central do presente trabalho. / Facing a time in which the action scale for social agents especially the hegemonic companies in the globalized market comprises the world, national territories are mobilized and changed in order to allow the organization and efficiency of internationalized spatial productive circuits. The establishment of Logistic and Industrial Customs Centers (CLIAs) is one among several of those territory adaptations. The CLIAs are entities responsible for the customs services of regulating the flow of imported and exported goods, yet they are based out of the primary zone of airports and harbors, hence consisting of an alternative to the traditional customs for the importation and exportation of goods. Therefore, CLIAs are new geographic agents that increase the Brazilian territorial porosity, creating more fluid pathways for international trade, mainly for some spatial productive circuits. The presence of these entities in the State of São Paulo is directly related to the industrialization process of this territory, since they are largely used by the chemical and automotive industries. We believe that it is important to disentangle the relationships established between the CLIAs and the industries for which they function. Besides the involvement of CLIAs in these spatial productive circuits, they may also develop different relationships with the place where they are based, generating diverse means of integration with the territory vertical and horizontal. It is precisely the understanding of this complex territorial dynamics that constitutes the central objective of this work.

Le procédé de fusion sélective « lit de poudre » (SLM) permet d'élaborer des pièces métalliques bonne matière (denses) directement à partir de la fusion de couches de poudres successives. De nombreux problèmes techniques doivent encore être surmontés pour faire du SLM un processus de fabrication entièrement viable. C'est le cas de l’état de surface et de l'apparition systématique de porosités, qui nécessitent des étapes de post-traitements. Jusqu'à présent, l'origine de la porosité reste incertaine mais est supposée être liée à la stabilité du procédé. Cette thèse propose une étude originale de l'interaction laser-poudre-bain liquide sur 316L et sur deux alliages d’aluminium (5086 et 4047) avant d’étudier les conditions de densification de la matière. Le travail de cette thèse s’articule en deux parties. Dans la première partie, une étude expérimentale de l’interaction laser-matière a été effectuée sur un banc instrumenté à partir d’imagerie par caméra rapide (>10 000 images /s). Les conditions de formation des éjections métalliques, de dénudation en poudre, l’hydrodynamique des zones fondues (dont le humping) ont été caractérisées et quantifiées. Tous ces phénomènes sont liés aux fortes densités de puissance utilisées en SLM, qui favorisent le régime de keyhole et la vaporisation. La deuxième partie de ce travail de thèse a consisté à évaluer l’origine et le taux de porosités sur une machine SLM. Une formulation analytique de la densification, dépendante d’un paramètre énergétique VED, a été validée par une étude expérimentale de l’évolution du taux de porosité, quelle que soit l’épaisseur de poudre. Un premier lien a été réalisé entre les dimensions des cordons de fusion et les conditions de densification. Enfin, une forte interaction (diffusion de Rayleigh ou absorption) a été observée entre le faisceau laser incident et les nanoparticules contenues dans la colonne de vapeur métallique, à l’origine de la dispersion importante des profondeurs de fusion. / The selective laser melting (SLM) process allows to produce dense metal parts directly from the melting of successive powder layers. However, many technical issues are still to overcome for making SLM a fully viable manufacturing process. This is the case of surface finishing and the systematic occurrence of porosities, which require post machining steps. Up till now, the origin of porosities remains unclear but is expected to be related to the stability of the process. This thesis proposes an original study of the laser-powder-melt pool interaction on 316 L and on two aluminum alloys (5086 and 4047) before studying the material’s densification conditions. The work is structured in two parts. In the first part, an experimental study of the laser-matter interaction has been carried out on an instrumented SLM setup equipped with a fast camera (>10 000 images /s). The conditions of formation of metal ejections, denudation and hydrodynamics of melt pool (including humping) have been characterized and quantified. All these phenomena are related to the high power densities used in SLM, which favor keyhole regime and vaporization. The aim of the second part of this work was to characterize the origin and the porosity fraction on an SLM machine. A first correlation has been made between the dimensions of the fusion beads and the densification conditions. A strong interaction (Rayleigh scattering or absorption) has been observed between the incident laser beam and the nanoparticles contained in the metal vapor column: this interaction is responsible for the significant dispersion of melting depths.

L'un des objectifs de l'industrie éolienne est de produire de grandes pièces de structure à moindre coût. Dans ce contexte, la fabrication de pièces composites à partir de renforts quasi unidirectionnels(quasi-UD NCF) avec le procédé d’infusion est compétitive tant sur le plan mécanique que financier. Le procédé d'infusion engendre un phénomène de décompaction dû à la flexibilité de la bâche à vide. De plus les NCF présentent un écoulement à double échelle pendant leur imprégnation. La modélisation des deux phénomènes est souvent réalisée en supposant que la préforme fibreuse est un milieu continu à perméabilité variable. Néanmoins, la perméabilité est influencée par la répartition et la taille des mésopores, qui dépendent de l'état de compaction. Le but de cette thèse est de caractériser expérimentalement l'évolution d'un quasi-UD lors de l’infusion et d’évaluer l'impact de la réorganisation microstructurale sur des quantités macroscopiques d’intérêt, tels que la perméabilité et le temps de remplissage des pièces.Des infusions ont été réalisées à l'intérieur d'un tomographe pour capter l’évolution d’une même microstructure avant et après infusion. Un modèle simplifié a été proposé pour prédire la perméabilité dans le plan et ainsi évaluer l'influence de la réorganisation microstructurelle sur celle-ci. De plus,un outil numérique a été développé pour prendre en compte un écoulement double échelle dans un milieu fibreux déformable bidisperse. L'impact de la décompaction sur le temps de remplissage des pièces a été établi. Une étude mécanique expérimentale du comportement de la mèche tout au long de l’infusion a également été réalisée afin de mieux comprendre le comportement du quasi-UD. Un modèle hyperélastique a finalement été proposé pour prédire le comportement mécanique 3D des mèches pendant la phase de chargement à sec, avant l'infusion. / A current aim of wind turbine industries is to produce large structural parts at reduced costs. In this context, manufacture composite blades made of quasi-unidirectional non-crimp fabrics (quasi-UD NCF) using the infusion process is competitive on both mechanical and cost aspects. The infusion process involves an unloading phenomenon due to the vacuum bag flexibility. Additionally, during the impregnation, NCFs exhibit a dual-scale flow. Usual modeling of both phenomena assumes that the fibrous preform is a continuous medium with a varying permeability. Nonetheless, the permeability is affected by the meso-pores size and spatial distribution, which depend on the compaction state. The goal of this thesis is thus to characterize experimentally the flow-induced microstructural evolution of a quasi-UD NCF during the infusion process, and to quantify the impact of thismicrostructural reorganization on relevant macroscopic parameters, such as modelled in-plane permeability as well as computed filling time of parts. In situ infusion process has been conducted inside X-ray Computed Tomography device to capture a dual-scale fibrous microstructure prior and after the infusion process. Additionally, a simplified model has been proposed to predict the in-plane permeability and thus to evaluate the influence of the microstructural reorganization on it. Then, a numerical tool has been developed to account for dual-scale flow in a bidisperse deformable fibrous media. The impact of the dual-scale unloading on themacroscopic filling time of parts has been established. A mechanical investigation of the towbehavior during the infusion process has been additionally carried out experimentally to better understand the quasi-UD NCF behavior. From these results, a hyperelastic model has been proposed to predict the 3D mechanical behavior of tows during the dry loading phase, prior to the infusion process.

Moisture-induced damage is one the major causes of deterioration of asphalt pavements and extensive research has been conducted on this topic. Theoretical and experimental results have led the researchers to believe that moisture-induced damages are caused mainly by the generation of pore water pressure in asphalt mixtures when traffic passes over a pavement. The Moisture Induced Sensitivity Tester (MIST) has been recently developed to simulate the phenomenon of repeated pore pressure generation and deterioration in the laboratory. The objective of this study was to evaluate moisture-induced damage in typical Maine Department of Transportation (DOT) asphalt mixes, with the use of MIST, pre and post testing, and analysis of data. The MIST was used to condition Hot Mix Asphalt (HMA) samples that were compacted from eight typical Maine DOT mixes, with different types of aggregates and asphalt binder. A modified Dynamic modulus test in Indirect Tensile Mode was used for the determination of damage. A layered elastic model, along with a fatigue-cracking criterion, was utilized to assess the total impact on the pavement lives. Monte Carlo analysis was conducted to determine the distribution of number of repetitions to failure of pavements that are subjected to moisture damage. The major conclusions are that most of the mixes are likely to experience a reduction in their life due to the effect of moisture and that the Micro-Deval and the fine aggregate absorption test results can be related to such damage. A composite factor, consisting of both of these test results, is recommended for regular use by the DOT to screen mixes with high moisture damage potential.

Several Arctic council reports conclude that oil spills are the most significant threat to the Arctic ecosystem. Some studies have shown that in-situ burning (ISB) of oil spills over water can remove more than 90% of the oil, and is the most promising technology for an efficient response to oil spills in the Arctic region. The definition of "In situ" is intentional, controlled burning of oil in place (i.e., without extracting or removing the oil first). Earlier studies [Bellino (WPI 2012), Farahani, (WPI 2014)] have investigated burning behavior of crude oil on ice, similar to what one would expect in sea-ice or bare lake ice conditions. The focus of the current study is to investigate the burning behavior of crude oil in snow, similar to oil spills in snow-covered land, or in snow covered sea ice in the Arctic. Understandably, due to the difference in packing density between ice/water and snow, the parameters that influence burning behavior of oil in snow are different compared to burning oil in the sea or ice conditions. The current experimental study shows that the snow behaves as a porous medium, and depending on the porosity and volume of the oil spill, two extreme behaviors are exhibited. In the case of an oil spill on snow with low porosity, the oil sinks easily to the bottom, and the burning involves, significant thermo capillary effects enabling the oil to rise up and burn. On the other hand, if the snow is less porous, most of the oil layer remains on the surface, approaching the case of an ice bed. However, the melting of snow due to flame heat flux causes a circulating flow pattern of the oil, whereby the hot layer at the surface moves down and comes back up due to capillary action. These processes, which have not been observed in the earlier studies, are physically explained in this study. The implications to overall efficiency of the burning process, which represents the amount of crude oil left in the snow after the burning process is discussed. The results will ultimately improve the strategies and the net environmental benefit of, and by it the success of, oil clean-up after an accidental spill on snow.