The crystal chemistry of Fe doped mayenite (Ca12Al14-xFexO33) samples prepared using solid state and sol-gel synthesis techniques were compared. Five samples were prepared using solid state process with varying Fe concentration (x) where x = 0, 0.05, 0.1, 0.25 and 0.3; two sets of samples were made via the sol-gel, the first set was prepared for studying the amount of Fe substituted by varying the Fe concentration where x = 0, 0.05, 0.1, 0.2, 0.3 and 0.4 and the second set was prepared for studying the firing temperature and did not contain Fe additions (x = 0). Samples produced via the sol-gel method were more likely to be single phase with incorporation of Fe2O3 while solid state samples contained multiple phases over the same range of Fe2O3 substitutions. The refined lattice parameters, of samples prepared using both methods were observed to increase with increasing Fe concentration, suggesting Fe is replacing Al since Fe has larger ionic radii than Al. Samples prepared via the sol-gel method were found to be single phase at lower temperatures compared to samples prepared via solid state synthesis. Samples synthesized using the sol-gel method were found to have multiple phases when fired at 800 ˚C but were single phase when fired at 900˚C. In comparison, samples synthesized using traditional solid state techniques showed single phase when fired at a temperature of 1350˚C.

The crystal chemistry of Fe doped mayenite (Ca12Al14-xFexO33) samples prepared using solid state and sol-gel synthesis techniques were compared. Five samples were prepared using solid state process with varying Fe concentration (x) where x = 0, 0.05, 0.1, 0.25 and 0.3; two sets of samples were made via the sol-gel, the first set was prepared for studying the amount of Fe substituted by varying the Fe concentration where x = 0, 0.05, 0.1, 0.2, 0.3 and 0.4 and the second set was prepared for studying the firing temperature and did not contain Fe additions (x = 0). Samples produced via the sol-gel method were more likely to be single phase with incorporation of Fe2O3 while solid state samples contained multiple phases over the same range of Fe2O3 substitutions. The refined lattice parameters, of samples prepared using both methods were observed to increase with increasing Fe concentration, suggesting Fe is replacing Al since Fe has larger ionic radii than Al. Samples prepared via the sol-gel method were found to be single phase at lower temperatures compared to samples prepared via solid state synthesis. Samples synthesized using the sol-gel method were found to have multiple phases when fired at 800 ˚C but were single phase when fired at 900˚C. In comparison, samples synthesized using traditional solid state techniques showed single phase when fired at a temperature of 1350˚C.

SESMR (Sorption Enhanced Steam Methane Reforming), SMR (Steam Methane Reforming) avec capture de CO2 in situ par un adsorbant solide, peut amener à une exploitation durable du gaz naturel pour la production de H2. La thèse, partie du projet de recherche ASCENT (Advanced Solid Cycles with Efficient Novel Technologies), concerne le développement de matériaux combinés adsorbants catalyseurs Ni-CaO-mayenite pour le SESMR, aux fins d’étudier les influences dues à la fraction de Ni, aux sels précurseurs du Ni (Ni acétate ou Ni nitrate), et à la fraction de CaO disponible. Les techniques ICP AES, XRD, BET/BJH, SEM/EDS, TEM/EDS, TPR et TGA ont été utilisés pour caractériser les matériaux synthétisés. La réactivité a été évaluée par des tests en lit fixe à l’échelle du microréacteur, qui ont aussi permis une sélection des matériaux les plus prometteurs pour une étude de l’applicabilité industrielle par tests multi cycliques SESMR/régénération de solides par un réacteur automatisé à lit fixe. / Sorption enhanced steam methane reforming (SESMR), steam methane reforming (SMR) with in situ CO2 sorption by a solid sorbent, can lead to a sustainable exploitation of natural gas to produce H2. (CSCM). This thesis, as a part of ASCENT (Advanced Solid Cycles with Efficient Novel Technologies) project, deals with Ni-CaO-mayenite combined sorbent-catalyst material for SESMR, to study the effect of Ni fraction, its precursor salt (Ni nitrate or Ni acetate), and free CaO fraction. ICP AES, XRD, BET and BJH methods, SEM EDS, TEM EDS, TPR and TGA were used to characterize synthesized materials. Their reactivity was evaluated by tests in a packed bed microreactor, which served also as a screening tool to choose the most promising materials. Their industrial applicability was assessed by multicycle SESMR/regeneration tests in an automated packed bed bench scale rig.

Améliorer la production en hydrogène dans les procédés de conversion des ressources renouvelables telles que la biomasse est un réel challenge dans le contexte de la production d’énergie propre et efficace. En effet, dans le procédé de conversion de la biomasse par vapogazéification, l’hydrogène est produit avec de quantités importantes de CO2, de CH4 et des molécules aromatiques lourdes, toxiques et complexes appelées goudrons. Ce travail s’intéresse à la production d’hydrogène par vaporeformage des goudrons avec capture in-situ de CO2 dans les procédés de vapogazéification de la biomasse. Nous proposons un système catalytique bifonctionnel de type Fe/CaO-Ca12Al14O33 dans lequel le fer est actif pour la production d’H2 tandis que la phase absorbante CaO-Ca12Al14O33 capture le CO2, ces deux processus se font simultanément dans un réacteur à lit fixe. Ce travail a permis de développer un système de type CaOCa12Al14O33/ olivine adapté pour la gazéification de la biomasse « Miscanthus » dans un réacteur à lit fluidisé. / Improve the hydrogen production from the conversion of renewable resources such as biomass is a real challenge in the context of the production of clean and efficient energy. In fact, during the biomass steam gasification, hydrogen is produced with significant amounts of CO2, CH4 and heavy, toxic and complex aromatic molecules called tars. This work focuses on the production of hydrogen by steam reforming of tar with in-situ CO2 capture. We propose a catalytic bi-functional material Fe/CaO-Ca12Al14O33 where iron favors the H2 production and simultaneously the CaO-Ca12Al14O33 capture CO2 in a fixed bed reactor. This work led also to the development of a CaOCa12Al14O33/ olivine system adapted for biomass gasification "Miscanthus" in a fluidized bed reactor.

Les chapes autonivelantes sont souvent composées de ciment alumineux et de sulfate de calcium pour assurer une prise rapide. Le ciment alumineux est généralement à base de monocalcium aluminate (CA). Récemment, un nouveau ciment alumineux riche en mayénite (C12A7) a été conçu pour optimiser l’application des chapes par l’augmentation de la quantité d’ettringite formée. Le développement de l’hydratation au jeune âge (pendant les premières 24 heures) de ce type de produit est pourtant très peu connu. L’objectif principal de cette étude est d’évaluer l’interaction entre l’hydratation au jeune âge et le séchage naturel à travers un modèle de couplage. Ce modèle est appliqué à l’étude d’une pâte pure ettringitique composée de ciment riche en C12A7 et de plâtre alpha. Tout d’abord, un modèle cinétique d’hydratation, initialement développé en système dilué, a été proposé pour étudier l’hydratation d’une pâte pure ettringitique en condition endogène. Il permet de décrire l’évolution en fonction du temps de la porosité, de la teneur en eau, des quantités de phases (réactifs et hydrates). Un modèle de distribution de rayon de pores a été aussi proposé. Ce modèle permet d’obtenir la courbe de pression capillaire au jeune âge nécessaire pour la modélisation des transferts hydriques. Ensuite, une modélisation complète des transferts hydriques a été présentée. Elle a montré que l’hypothèse d’une pression de gaz constante conduit à une surestimation de perte de masse pour des matériaux à faible perméabilité. Pourtant, pour le cas des ciments alumineux, cette hypothèse peut être retenue pour simplifier le couplage avec le modèle cinétique d’hydratation. Enfin, une modélisation des couplages entre l’hydratation et la dessiccation a été développée. L’originalité de ce modèle porte sur l’intégration de la courbe isotherme issue du modèle de distribution poreuse dans la modélisation des transferts hydriques via une fonction Matlab. Le modèle a été capable de reproduire la cinétique de perte de masse d’un échantillon de pâte pure ettringite réactive soumis au séchage précoce. La nécessité de la prise en compte de l’évolution de température a été mise en évidence. De plus, le modèle a prévu un arrêt précoce de l’hydratation à la surface séchante après 10 heures d’hydratation. L’effet de la dessiccation atteint une profondeur de 5 mm sur un échantillon de 3 cm d’épaisseur. / Self-leveling flooring compounds (SLC) are often composed of calcium aluminate cements (CAC) and calcium sulfate to ensure rapid setting. The mineral composition of calcium aluminate cements is usually designed around monocalcium aluminate (CA). Recently, a new cement with the main compound of mayenite (C12A7) has been designed to optimize the application of SLC by increasing the amount of ettringite in the hydration product. However, there is a lack of knowledge related to early-age hydration (during the first 24 hours) of this type of product. The main objective of this study is to evaluate the interaction between early-age hydration and natural drying through a coupling model. This model is applied to the study of an ettringite binder composed of cement rich in C12A7 and plaster First, a kinetic model of hydration, initially developed in a diluted system, has been proposed to study the hydration of an ettringite paste under endogenous conditions. It allows to describe the evolution as a function of time of the porosity, the water content, and the quantities of phases (reactants and hydrates). A model of pore size distribution has also been proposed. This model allows us to obtain the capillary pressure curve at early-age, which is necessary for the modeling of moisture transport. Then, a complete modeling of water transfers was presented. It has shown that the assumption of constant gas pressure leads to an overestimation of mass loss for weakly permeable materials. However, for the case of CAC, this hypothesis can be retained to simplify the coupling with the kinetic model of hydration. Finally, a modeling of the coupling between hydration and desiccation has been developed. The originality of this model concerns the integration of the isothermal curve resulting from the pore size distribution model in the modeling of water transfers via a Matlab function. The model was able to reproduce the kinetics of mass loss of a sample of ettringite binder subjected to early drying. The role of temperature was also elucidated. In addition, the model predicted an early stop of hydration on the drying surface after 10 hours of hydration. The effect of desiccation reaches a depth of 5 mm on a sample with a thickness of 3 cm.