De manière générale, les composites oxyde/oxyde sont la plupart du temps élaborés par frittage, par voie sol-gel ou par infiltration en phase gazeuse, CVI (« Chemical vapor infiltration »). Ces techniques d’élaboration comportent de nombreuses étapes engendrant un temps long d’élaboration ce qui peut entraîner une détérioration des propriétés du composite. Cette thèse s’intéresse à un procédé original et rapide de densification de préformes fibreuses développé par le Commissariat à l’Energie Atomique et aux énergies alternatives (CEA) : la caléfaction. Ce procédé est connu pour élaborer des composites C/C ou C/SiC à partir d’un précurseur liquide. Cependant, la possibilité d’élaborer des composites oxyde/oxyde n’a jamais été testée. L’objectif de ce travail est d’étudier la réalisation de composites oxydes/oxydes par ce procédé. Plusieurs matrices ont été réalisées telles que la silice, l’alumine et le système ternaire aluminosilicate de baryum, BaSi2Al2O8. Plusieurs paramètres expérimentaux ont été étudiés tels que la température d’élaboration, le temps de manipulation et la composition du précurseur. Des caractérisations microstructurales et physico-chimiques ont permis de caractériser les matériaux élaborés. Plusieurs modifications ont été apportées au montage expérimental afin de permettre une meilleure reproductibilité des essais et un meilleur suivi thermique lors de l’élaboration de matrice oxyde. / Nowadays, oxide/oxide composites are most of the time developed by sintering, sol-gel process or CVI (Chemical Vapor Infiltration). These techniques include many steps of synthesis leading to a long time of synthesis and possible deteriorations of the properties of the composite. This thesis focuses on an original and rapid process developed by French Alternative Energies and Atomic Energy Commission (CEA): the film boiling chemical vapor infiltration. This technique is already used to synthesize C/C and C/SiC composites but works have never focused on oxide/oxide composites. The main goal of this thesis is to synthesize oxide/oxide composites by film boiling chemical vapor infiltration. Works were focused on alumina, silica and barium aluminosilicate matrices. Several experimental parameters were studied: temperature, time and liquid precursor. Microstructural and physicochemical characterizations were done on composites. Several modifications of the experimental setup have been made in order to allow a better reproducibility of the tests and a better thermal monitoring.

The challenges faced in today’s industry require materials capable of working in chemically aggressive environments at elevated temperature, which has fueled the development of oxidation resistant materials. All-Oxide Ceramic Matrix Composites (OCMC) are a promising material family due to their inherent chemical stability, moderate mechanical properties, and low weight. However, limited information exists regarding their behavior when in contact with other high-temperature materials such as superalloys. In this work three sets of tribological tests were performed: two at room temperature and one at elevated temperature (650 °C). The tests were performed in a pin-on-disk configuration testing Inconel 718 (IN-718) pins against disks made with an aluminosilicate geopolymeric matrix composite reinforced with alumina fibers (N610/GP). Two different loads were tested (85 and 425 kPa) to characterize the damage on both materials. Results showed that the pins experienced ~ 100 % wear increase when high temperature was involved, while their microstructure was not noticeably affected near the contact surface. After high temperature testing the OCMC exhibited mass losses two orders of magnitude higher than the pins and a sintering effect under its wear track, that led to brittle behavior. The debris generated consists of alumina and suggests a possible crystallization of the originally amorphous matrix which may destabilize the system. The data suggests that while the composite’s matrix is stable, wear will not develop uncontrollably. However, as soon as a critical load/temperature combination is attained the matrix is the first component to fail exposing the reinforcement to damage which drastically deteriorates the integrity of the component.