PAR Olga CHOwANIEC ii Principium cuius hinc nobis exordia sumet, nullam rem e nihilo gigni divinitus umquam. But only Nature's aspect and her law Which, teaching us, hath his exordium Nothing from nothing ever yet was born. Lucretius "De Rerum Natura" Acknowledgements My adventure which lead to this dissertation I totally due Prof. Karen Scrivener, who I would like to thank for giving me the opportunity to explore the world of cement science under her supervision, for all time spent discussing,

more »

... correcting, reading and believing that it is possible. The present work was possible thanks to financial support of Swiss Cement Industry -HOLCIM, for what I'm very thankful. Especially I would like to thank Dr. Silvia Vieira and Dr. Thomas Matschei from HOLCIM Group Support for their presence and collaboration. I would like to thanks to all my LMC colleagues, who number was growing during the four years spent at EPFL, for all the good time spent with you during, coffee brakes, sport classes, hiking, traveling and working. Je remercie tout particulièrement Emmanuel pour son aide pendant la première année de ma thèse, pour toutes les discussions que nous avons eues ainsi que ton soutien qui m'a beaucoup aidé. Merci de m'avoir permis d'apprendre de toi. Merci Lionel, pour ton précieux temps passéà préparer mes expériences età faire du béton, ainsi que ta bonne humeur et la patiente dont tu as fais preuve pour m'aiderà améliorer mon Français. Merci Kyle, Gewnn et Steve pour vos critiques constructives, votre soutien et votre collaboration. Merci Rouben pour sauver mes resulta de DRX. C'était un plaisir de travailler avec vous. Mercià tous mes collègues de bureau, plus particulièrementà Rodrigo qui a pris la responsabilité de me présenter tous les endroits importants du monde du laboratoire LMC, pour sa bonne humeur de tous les jours et sa diplomatie. Merci Patrick pour les nombreuses discussions que nous avons partagées en train, et devant le MEB. Merci Arnaud, John et Mathieu qui avezéclairé mes journées. Mercià mesétudiants, et tout particulièrement Gaëtan pour ton excellent travail, les discussions partagées et pour m'avoir motivéà apprendre le Français. Mercià tous les gens que j'ai pu rencontrer pendant tout ce temps, et pour toutes ces discussions sur mon travail partagées lors des meetings, séminaires ou autres conférences. Je remercie de tout mon coeur Tristan pour sa présenceà mes côtés, pour son soutien inconditionnel, en particulier la dernière année. Je te dédie ce travail. Abstract Addition of fine limestone provides an excellent means to reduce the amount of clinker in cement. It is now well accepted that limestone partially reacts in cementitious systems with C 3 A to produce hemi-and monocarboaluminate phases and as a consequence more sulfate is available to form ettringite and the total volume of hydrates increases. The mechanism by which limestone affects the hydration is crucial in understanding its influence on the properties of cementitious materials. Laboratory and Commercial Cements with two different types of clinker, low and high C 3 A with different gypsum and limestone addition, were investigated. Hydrated pastes and mortars were investigated in terms of kinetics, phases assemblage, microstructure development, porosity, mechanical properties and durability (sulfate attack and sorptivity). An improvement in sample preparation for XRD measurement was made, which allows preferential orientation to be avoid and improves Rietveld Analysis quantification. The elastic modulus was found to correlate well with compressive strength and could be used as a nondestructive method to measure compressive strength. Monocarboaluminate formation was found to increase with increasing C 3 A. For high C 3 A cement it is visible at 2 days of hydration and at 720 days 4.5% of monocarboaluminate is measured in the system. For Low C 3 A cement it is visible at 7 days of hydration and at 720 days 1.6% of monocarboaluminate is measured in the system. Mc is formed only after all gypsum, which is more reactive than limestone is consumed to produce ettringite. No monosulfate is observed in the limestone systems. The optimum gypsum, was found to have as much effect at early ages on the hydration as 10% of limestone addition. Consequently variations in the gypsum level were investigate but it was difficult to quantify differences between samples with increasing gypsum addition. Studies of behavior in sulfate solution indicated that C 3 A is the dominant factor. Limestone addition produce slight changes in the form of degradation but do not fundamentally change whether deterioration takes place or not.