Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / It is certain that most of Brazilian highways are under unfavorable conditions of trafficability. Technological developments led to a large increase in load capacity of trucks, resulting in higher impacts on the highways. Another factor to be considered in this degradation process of pavements is climatic variations, very common in Brazil. Among many types of defects found on Brazilian highways, one deserves special attention, the fatigue cracking. Aiming to investigate this problem, this study aimed to study the fatigue phenomenon in laboratory for asphalt mixture used in the field (Av. Hélvio Basso, Santa Maria-RS) and evaluate in laboratory its possible improvement using polymer modified binder before to fatigue performance at three different temperatures: 10°C, 25°C and 35°C. Three asphalt mixtures were prepared in the la, two conventional binder used in the monitored section (CACAMPO asphalt content in the field, and CALAB 50/70 asphalt content obtained in laboratory dosage), and a third polymer modified binder with (CALAB 60/85-E). Mixtures reproduced in the laboratory were prepared according to the Marshall methodology and the materials were characterized according to traditional experiments. For all mixtures, the volumetric parameters were determined, stability, fluency, adhesion, Indirect Tensile Streght, Total Resilience Moduli and performance to fatigue by indirect tensile stress under controlled mode. These last three were held at the three test testing temperatures. In the laboratory, it was evident the improvement in thermal susceptibility properties and stability of the mixture by using the modified binder polymer, providing a blend less susceptible to variations in temperature and stable at high temperatures. The results obtained in the laboratory demonstrated the superiority of the mixture CALAB 60/85 E, for the three temperatures studied, with emphasis on the destructive tests at the higher temperature. The rigidity of the mixtures, measured by the resilience module, showed little influence by the use of the modified binder, however, showed very different behavior with variation in temperature. Regarding the fatigue performance, the modified blends proved to be far superior than the conventional mix, ensuring an increase in service life for the three temperatures studied, specially for the temperature of 35°C. The temperature variation of the test proved to be very crucial in the fatigue performance of the asphalt mixture, which is inversely proportional to the temperature increase. At the end of the study, it is concluded that the use of polymer modified binders for can be a great option to improve the performance of asphalt mixtures against the fatigue rupture mechanism and it can be used successfully in regions with large temperature variations. Thus, it is possible to ensure an increase in the service life of the asphalt layer and therefore to the pavement as a whole. / É notório que grande parte das rodovias brasileiras encontram-se em condições desfavoráveis de trafegabilidade. A evolução tecnológica permitiu um grande aumento da capacidade de carga dos caminhões, refletindo em maiores solicitações nas rodovias. Outro fator a ser considerado neste processo degradativo dos pavimentos são as variações climáticas, muito presentes no Brasil. Dentre tantos tipos de defeitos encontrados nas rodovias brasileiras, um deles merece destaque especial, a degradação por fadiga. Com intuito de investigar tal problema, o presente trabalho teve como propósito estudar o fenômeno de fadiga em laboratório para uma mistura asfáltica utilizada em campo (Av. Hélvio Basso, Santa Maria-RS) e avaliar, em laboratório, a sua possível melhora pelo uso de ligante modificado por polímero perante o seu desempenho à fadiga, em três temperaturas distintas: 10°C, 25°C e 35°C. Foram preparadas em laboratório três misturas asfálticas, duas com o ligante convencional utilizado no trecho monitorado (CACAMPO, teor de asfalto de campo, e CALAB 50/70, teor de asfalto obtido em dosagem de laboratório) e uma terceira, com ligante modificado por polímero (CALAB 60/85-E). As misturas foram preparadas conforme a Metodologia Marshall e os insumos foram caracterizados de acordo com ensaios tradicionais. Para todas as misturas, determinram-se os parâmetros volumétricos, estabilidade, fluência, adesividade, Resistência à Tração indireta, Módulo de Resiliência total e desempenho à fadiga por tração indireta à tensão controlada. Estes três últimos foram realizados nas três temperaturas de ensaio. Em laboratório, ficou evidente a melhora das propriedades de suscetibilidade térmica e estabilidade da mistura pelo uso do ligante modificado por polímero, proporcionando uma mistura menos suscetível às variações de temperatura e mais estável a temperaturas elevadas. Os resultados obtidos em laboratório demonstraram a superioridade da mistura CALAB 60/85-E, para as três temperaturas analisadas, com ênfase para os ensaios destrutivos à temperatura mais elevada. A rigidez das misturas, medida pelo Módulo de Resiliência, mostrou-se pouco influenciável pela utilização do ligante modificado, porém, apresentou comportamentos bem distintos com a variação da temperatura. Quanto ao desempenho à fadiga, as misturas modificadas apresentaram-se muito superiores às convencionais, garantindo um incremento na vida de serviço para as três temperaturas analisadas, sendo este, mais evidente para a temperatura de 35°C. A variação da temperatura de ensaio mostrou-se bastante determinante no desempenho à fadiga das misturas asfálticas, sendo este, inversamente proporcional ao aumento de temperatura. Ao término do estudo, conclui-se que o emprego de ligantes modificados por polímero pode ser uma ótima opção de melhora do desempenho de misturas asfálticas frente ao mecanismo de ruptura por fadiga, podendo ser utilizado com sucesso em regiões que apresentam grandes variações térmicas. Desta forma, garantindo um incremento na vida de serviço do revestimento asfáltico e consequentemente ao pavimento como um todo.

Ces travaux de thèse ont pour objectif de contribuer au développement de procédés innovants pour les assemblages contenant des métaux durs. Le procédé d’expansion à froid, dont l’efficacité a été prouvée à de nombreuses reprises sur les alliages d’aluminium et dont les qualités sont nombreuses, a attiré notre attention. L’objectif de la thèse est d’apporter des éléments de réponse quant à l’intérêt de l’utilisation du procédé d’expansion au sein des métaux durs, mesuré notamment par son impact sur la performance en fatigue. Plus généralement, il s’agit de comprendre son fonctionnement et son mode d’action au sein de ces alliages à hautes performances. Dans les travaux réalisés, les challenges scientifiques et techniques s’articulent autour de plusieurs thématiques d’étude. Des essais expérimentaux ont été effectués en vue de prouver la faisabilité du procédé mais aussi de mesurer de son impact sur la tenue fatigue d’alésages en alliages de titane (Ti-6Al-4V), en acier inoxydables à durcissement structural (15-5PH) mais aussi en superalliages à base Nickel (Inconel 718). L’efficacité du procédé a été prouvée et les gains apportés par la technologie sont importants. D’un autre côté, la problématique de la détermination des champs résiduels en bord de trou reste un réel défi pour la communauté intéressée par l’expansion à froid. La considération cette fois-ci des métaux durs apporte un degré supplémentaire de nouveauté et d’inconnu. La méthodologie générale employée réside dans la considération d’études numériques et expérimentales afin d’étudier la réponse des métaux durs à l’expansion à froid. Des modèles éléments finis axisymétriques ont été développés en vue de simuler le procédé complet d’expansion à froid réalisée au sein d’alésages et d’empilages de divers métaux durs. Les résultats numériques fournis ont été mis en parallèle avec ceux issus de mesures expérimentales, telles que la méthode du trou incrémental. Le but étant ainsi d’obtenir les cartographies les plus fiables possibles des contraintes résiduelles triaxiales générées en bord de trou mais aussi de rechercher de stratégies d’optimisations du procédé / This thesis aims to contribute to the development of innovative processes for mechanical components made of hard metals (Titanium, steels and superalloys). In the aeronautical field, the design of ever more efficient and reliable structures remains a technical challenge. In mechanical assemblies, hole edges are the seats of high stress concentrations and are a major risk site for crack initiation. To fight against this damage, manufacturing technologies such as cold expansion are widespread for aluminium applications but not for hard metal. The objective of this study is to provide answers concerning the impact of the cold expansion on the fatigue performance of holes and understand the mechanisms of the process in these high performance alloys. In the present work, the methodology proposed is to carry out both experimental and numerical studies of the response of hard alloy holes subjected to cold expansion. An extensive experimental campaign has been set up. It includes several materials (Ti-6Al-4V αβ and β annealed, 15-5PH & Inconel 718) and aims to test many parameters. The process has proved very effective on the fatigue performance in these high performance alloys. On the other hand, the main technical and scientific challenge lies in determining the stress fields generated within the material after cold expansion, especially in hard metals, the behaviour of which can be diverse and complex. The numerical modelling strategy chosen has led to the development of polyvalent axisymmetric models that are dedicated to simulation of the split sleeve process. Experimental and numerical results were compared using various methodologies, such as the incremental hole drilling technique. Measurements have validated the finite element simulations, with the purpose of mapping the residual fields in the expanded metallic section and the proposal of optimisation techniques.