Tesi sul tema "Chlorure. carbonation"

La carbonatation atmosphérique et la pénétration des ions chlorures sont les principales causes du déclenchement de la corrosion des armatures dans les bétons armés. Chacun de ses mécanismes a fait l'objet de nombreuses études dans la bibliographie. Pourtant les études prenant en compte en même temps les deux types d'agression sont rares. Le couplage des deux phénomènes existe cependant. La carbonatation intervient en effet dans tout ouvrage dès le décoffrage et il suffit que cet ouvrage soit aussi par exemple situé à proximité d'eau de mer pour que les embruns apportent des ions chlorures dans le matériau. L'objectif de cette thèse a été de quantifier les différentes phases de la matrice cimentaire et ses capacités de fixation des ions chlorures par le biais d'isothermes d'interactions pour des matériaux sains et carbonatés. Ces isothermes d'interactions obtenues notamment par la méthode des équilibres ont permis de voir que les matrices partiellement ou totalement carbonatées fixaient moins d'ions chlorures que les matrices saines. L'utilisation de moyens complémentaires d'analyses tels que l'analyse thermogravimétrique (ATG), la diffraction des rayons X (DRX) et la résonance magnétique nucléaire (RMN) a permis de mieux comprendre cette perte de fixation. Une diminution des sels de Friedel et des chlorures fixés par les C-S-H a été ainsi mise en évidence dans les matériaux carbonatés. Ces diminutions qui on été quantifiées sont attribuées à une modification sous carbonatation des phases capables de fixer les ions chlorures (phases aluminates, sulfo-aluminates et C-S-H). Des différences au niveau du réseau poreux ont été également mises en évidence par la porosimétrie par intrusion de mercure (diminution de la macroporosité et de la connectivité en général pour les matrices carbonatées par formation de CaCO3). L'objectif a été aussi de corréler toutes ces modifications observées avec les propriétés de transport des ions qui ont été obtenues par des tests de diffusion et migration des ions chlorures (et/ou sulfates) ainsi que des mesures de résistivité électrique. Les résultats expérimentaux ont montré que le coefficient de diffusion apparent est plus grand dans les matériaux carbonatés. Certaines expériences de diffusion faites en présence à la fois d'ions chlorures et sulfates ont aussi montré la concurrence de ces deux ions pour se fixer à la matrice cimentaire. Les matériaux cimentaires testés (bétons et pâtes) ont été formulés sans ou avec additions minérales (métakaolin, cendres volantes, laitier) et les propriétés de ces matériaux ont été observées à différents âges
Carbonation and chloride ingress are the main causes of reinforced concrete degradation. A lot of studies describe these two phenomena separately but only few studies state a possible coupling. However, the coupling of both phenomena exists. The corrosion of reinforcement due to chloride ions occurs more often in marine environments or in the presence of deicing salts while carbonation occurs systematically, in a more or less high degree depending on environmental conditions (humidity, temperature…). The aim of this thesis was to quantify the different phases of the cement matrix and its chloride binding by means of chloride binding isotherms for carbonated and non-carbonated cement based materials. These chloride binding isotherms, obtained by the equilibrium method, showed that the partially or totally carbonated cement matrix are bound fewer chlorides than the non carbonated ones. Complementary techniques such as the thermogravimetry (TGA), the X-rays diffraction (XRD) and the nuclear magnetic resonance (NMR) analyses were used and allowed to understand why this binding decrease occurs. A decrease of Friedel's salt and chlorides bound in the C-S-H was showed in the carbonated materials. These decreases were quantified and are due to a modification of the phases containing chloride (such as aluminates, sulfo-aluminates and C-S-H phases) during carbonation process. The porous network, was observed by mercury intrusion porosimetry (MIP), was also modified (decrease of the macroporosity and the connectivity for carbonated cement matrix due to formation of calcium carbonate). The other aim was to correlate all these modifications with the ions transport properties. Chloride diffusion and migration tests (with or without sulfates) were obtained as well as the measurements of electric resistivity. The experimental results showed that the apparent chloride diffusion coefficient is higher in the carbonated cement materials. Some diffusion tests were performed in presence of both chlorides and sulfates, which showed the competition of these two ions to bind on the cement matrix. Cement materials (concrete and cement paste) were designed without or with supplementary cementing materials (metakaolin, fly ash, slag) and the properties of these materials were followed at different curing time

La corrosion des armatures est reconnue comme la première cause de dégradation des structures en béton armé dans les domaines du génie civil et des monuments historiques. La corrosion peut survenir lorsque le béton d'enrobage est carbonaté ou lorsque les ions chlorure sont présents en quantité suffisante au niveau de l'armature.Les objectifs de cette thèse visent à appréhender les processus de corrosion des armatures lorsque le béton est doublement contaminé (carbonatation et ions chlorure) et d'évaluer une solution de réparation de type traitement électrochimique d’extraction des ions chlorure (ECE).Pour répondre à ces objectifs, trois phases ont été programmées : (1) des vieillissements artificiels, (2) un traitement ECE et (3) un suivi de la durabilité du traitement. Pour chacune de ces phases, les évolutions (1) de la corrosion des armatures dans le béton, (2) des produits de corrosion à l'interface armature/béton et (3) de la matrice cimentaire ont été investiguées. L'approche scientifique de l'étude, avec des analyses multiples d'observations à différentes échelles et l'utilisation de techniques non destructives, destructives ou in situ, a permis de rendre compte des processus de corrosion avant et après traitement et de discuter les résultats en fonction de critères d'efficacité proposés.L'étude a été réalisée en prenant en compte deux ciments (CEM I et CEM III/A) et en considérant deux modes de vieillissements artificiels. Trois cent vingt éprouvettes en micro-béton armé répondant à quatre formulations ont ainsi été étudiées. Les techniques principales utilisées sont des caractérisations électrochimiques non destructives, des caractérisations au MEB et EDS, des analyses destructives de déterminations quantitative pour la teneur en ions chlorure et qualitative pour l'évaluation du pH et enfin la microspectrométrie Raman qui permet de déterminer la nature des produits de corrosion in situ.L’exploitation de l’ensemble des résultats a mis en évidence l’évolution des caractéristiques étudiées au cours des phases de l’étude. Avant traitement, l'enrobage béton était entièrement carbonaté et la teneur en ions chlorure libres au voisinage de l'armature était comprise entre 0,9 et 2,1% par rapport à la masse de ciment en fonction des formulations. Les courants de corrosion étaient de l’ordre de 10 µA/cm² et une épaisseur de corrosion homogène de l’ordre de 5-10 µm était observée. Lors des analyses Raman in situ, de la rouille verte chlorurée a été identifiée, synonyme de corrosion active. Après traitement, plus de 90% des ions chlorure ont été extraits et le pH du béton au voisinage de l'armature a augmenté (pH>9) sur une épaisseur d’environ 1 cm. Les courants de corrosion sont devenus inférieurs à 0,1 µA/cm² (niveau de corrosion négligeable). La formation de magnétite a majoritairement été observée avec néanmoins la présence résiduelle très locale de rouille verte. Aucune évolution délétère significative de la matrice cimentaire n’a été identifiée. L’efficacité du traitement a donc été démontrée. L'évaluation de la durabilité du traitement sur une période de 4 à 12 mois a montré que les caractéristiques électrochimiques restaient stables. En revanche, la diminution de l'épaisseur de béton dont le pH avait augmenté pour les éprouvettes fabriquées avec le ciment CEM III/A peut laisser craindre une reprise de corrosion sur du plus long terme. Enfin, le suivi Raman in situ, montre des différences de la durabilité en fonction de la durée du traitement
Reinforcement corrosion is known as the first cause of reinforced concrete degradation in the fields of civil engineering structures and historical monuments. Corrosion occurs when concrete cover is carbonated or when chloride ions are present in sufficient quantity at the reinforcement vicinity.The main objectives of this thesis are to investigate the corrosion processes of the reinforcement when the concrete is contaminated by carbonation as well as chloride ions and to evaluate a cathodic electrochemical treatment as a repair solution.To meet these objectives, three phases were investigated: (1) artificial agings, (2) an electrochemical chloride extraction (ECE) treatment and (3) durability. For each of these phases, the evolutions (1) of the rebar corrosion in concrete, (2) of the corrosion products at the reinforcement/concrete interface and (3) of the cementitious matrix were studied.The scientific approach gathered multiple analyzes at different scales and the use of non-destructive, destructive or in situ technique. It allowed to study the corrosion processes before and after treatment and to discuss results according to efficiency criteria suggested.The study was carried out taking into account two cements (CEM I and CEM III/A) and considering two modes of artificial aging. More than three hundred specimens of reinforced micro-concrete, corresponding to four concrete designs, were therefore studied. The main techniques used were non-destructive electrochemical characterizations, SEM and EDS characterizations, destructive quantitative determinations (chloride ion content) and qualitative determinations (pH evaluation), and finally Raman microspectrometry to determine the nature of the corrosion products.The results highlighted some evolutions of different characteristics studied during the phases of the study. Before treatment, the concrete cover was entirely carbonated and the free chloride content at the reinforcement level was in the range of 0.9-2.1% by weight of cement depending on the concrete design. A 10 μA/cm² average corrosion currents was reported and a homogeneous corrosion layer thickness of about 5-10 μm was observed. During in situ Raman analysis, chloride green rust was identified indicating an active corrosion.After treatment, more than 90% of the chloride ions were extracted and the thickness of the concrete which had a pH value higher than 9 around the reinforcement, was about 1 cm. The corrosion currents became less than 0.1 μA/cm² (negligible corrosion level). The formation of magnetite was mostly observed with nevertheless a local presence of residual green rust. No significant deleterious evolution of the cementitious matrix was identified. The treatment efficiency was therefore demonstrated. The evaluation of the treatment durability over a period of 4 to 12 months showed that the electrochemical characteristics remained stable. On the other hand, a decrease in the concrete area which pH increased during the electrochemical treatment was observed for CEM III/A cement which could lead to a corrosion recovery over the longer term. Finally, the in situ Raman study revealed differences in durability as a function of the treatment duration

Le béton armé est connu l’un des matériaux les plus utilisés dans la construction. Néanmoins, après une période de mise en exploitation, il peut se corroder. Les chlorures et la carbonatation sont les principaux facteurs provoquant la corrosion des structures en béton armé. La thèse a pour objectif d’étudier le comportement à la corrosion du béton armé sous l’environnement de carbonatation ou des chlorures, correspondant à l'influence de l’effet de la qualité de l’interface acier/mortier (top-bar effet), du type de ciment ainsi que de la fissure transversale artificielle ou des fissures transversales induites par la charge mécanique.En général, la corrosion induite par la carbonatation est étudiée du point de vue de la corrosion uniforme. Le premier objectif de la thèse est de confirmer l’opinion de la corrosion par macro-cellule ou non uniforme dans la corrosion induite par la carbonatation. Selon les résultats expérimentaux, le taux de corrosion dans les armatures de béton carbonaté est susceptible d'être considéré du point de vue de la corrosion uniforme et macro-cellule, qui est fortement influencée par le rapport cathode/anode. De plus, l'application de CEM III utilisant des laitiers comme addition minérale a montré une réduction efficace de la corrosion non uniforme.La deuxième partie se concentre sur l'influence de l’effet de la qualité de l’interface acier/mortier (top-bar effet) sur le comportement de la corrosion dans le cas d'une corrosion induite par la carbonatation et de la présence d'une fissure induite par la charge mécanique. Le suivi du courant de corrosion par macro-cellule a été utilisée dans les essais expérimentaux. Puis la perte de masse ainsi que la cinétique de corrosion ont été calculée. La corrosion s'est développée principalement au niveau du pré-fissures. En raison de l’effet de la qualité de l’interface acier/mortier, les barres supérieures étaient plus corrodées que les barres inférieures. La valeur du courant de la corrosion mesurée sur les barres supérieures a été observée plus élevée que sur les barres inférieures. La perte de masse de l'acier calculée à partir de la mesure du suivi de courant ne correspond qu'à une petite partie de la perte de masse totale déterminée par la mesure gravimétrique. La corrosion uniforme semble être le phénomène principal.La dernière partie étudie l’effet de la qualité de l’interface acier/mortier (top-bar effet) dans des échantillons de béton fibré avec ou sans fissure artificielle sur la corrosion induite par les chlorures. Les barres supérieures sont plus corrodées que les barres inférieures et la partie supérieure de la barre supérieure présente une corrosion plus élevée que la partie inférieure. En présence de fissures artificielles, le ‘’top-bar effet’’ entraîne la propagation de la corrosion le long de la surface des barres d'acier. La corrosion se développe également le long de la barre supérieure lorsqu'il n'y a pas de fissure artificielle, dans un temps pas si différent du cas d'une fissure artificielle. Cela confirme que l'exposition de la surface supérieure et l’effet de la qualité de l’interface acier/mortier sont très préjudiciables à la corrosion. En comparant le comportement avec un béton sans fibres, il apparaît que l'ajout de fibres dans le béton armé entraîne une augmentation de la résistance à la fissuration induite par la corrosion.Dans chaque partie, les observations expérimentales sont couplées à des simulations numériques pour comparer et démontrer les résultats expérimentaux
Reinforced concrete is known one of the most popular materials applied in construction. Nevertheless after a period of time it can be corroded. Chloride and carbonation are the main factors causing to corrosion in reinforced concrete structure. The thesis aimed to study the corrosion behavior of reinforcement concrete under carbonation or chloride environment, corresponding to the influence of top-bar effect, the cement type as well as artificial transverse crack or load-induced cracks.In general carbonation induced corrosion is usually researched from the point of view that it generates uniform corrosion. The first objective of the thesis is to confirm the opinion of macro cell or non-uniform corrosion in carbonated induced corrosion. According to the results of experiments in this part, corrosion rate in carbonated concrete reinforcement is likely to consider from the point of view of both uniform and macro cell corrosion, which is highly influenced by the ratio cathode/anode. In addition the application of CEM III using slag as addictive exhibited reduction of non-uniform corrosion effectively.The second part focuses on the influence of top-bar effect on corrosion behavior in the case carbonation induced corrosion and presence of load induced crack. Macro cell corrosion current monitoring was utilized in the experiment following the loss of mass as well as corrosion kinetic was calculated. Corrosions mainly developed at the position of the pre-cracks. Due to the top-bar effect upper bars were more corroded than bottom bars. Current corrosion value of top bars was observed higher than bottom bars. Loss of steel mass calculated from macro cell current measurement corresponds to only to a small part of the total loss of mass determined by gravimetric measurement. Uniform corrosion appears to be the main phenomena.The last part investigates the effect of top-bar effect in fibers concrete samples with or without artificial crack on chloride induced corrosion. The top steel bars are more corroded than the bottom bars and the upper part of the top bar is recorded higher corrosion than the lower part. In presence of artificial crack the top casting effect results that corrosions spread along the surface of the steel bars. Corrosion also spread along the top bar when there is no artificial crack, in a time not so different from the case of artificial crack. It confirms that top surface exposure and top bar effect are highly prejudicial for corrosion. By comparing the behavior with concrete without fibers, it appears that the addition of fibers in reinforced concrete leads to an increase of resistance to corrosion induced cracking.For each part, experimental observations are coupled with numerical simulations to compare as well as demonstrate the experimental results

Les ciments sulfoalumineux (CSA) peuvent être une solution à l’amélioration de l’impact environnemental des bétons, grâce aux faibles consommations énergétiques et émissions de CO2 lors de leur fabrication. Leurs propriétés expansives, de rapidité de prise et de montée en résistance participent également à l’intérêt grandissant dans les domaines de la construction et de la préfabrication en béton pour ce type de liant. Toutefois, la durabilité des bétons sulfoalumineux reste encore méconnue ou discutée, notamment en milieux acides et face à la corrosion des armatures, en ambiance marine et par carbonatation. Les travaux présentés dans cette thèse se proposent ainsi d’étudier et de comparer la durabilité face à ces attaques, de bétons sulfoalumineux à celle de bétons de référence à base de CEM I et de CEM III/B. Ils s’appuient sur des caractérisations performantielles (indicateurs de durabilité, essais performantiels en conditions accélérées ou naturelles) et microstructurales (phases solides, porosité). Au préalable, l’application des caractérisations sus-citées aux bétons sulfoalumineux est vérifiée et discutée. La bonne résistance des bétons sulfoalumineux aux milieux acides est observée lors d’essais de lixiviation dynamique, à température et pH régulés. L’étude de la pénétration des chlorures (migration en régime permanent et transitoire, isothermes de fixation, exposition en zone de marnage) et de la carbonatation (naturelle et accélérée) de bétons à base de 8 ciments sulfoalumineux de laboratoire différents a permis d’établir des moyens d’améliorer leur durabilité potentielle. Ces moyens ont été appliqués à la fabrication d’un ciment industriel performant, dont la bonne durabilité reste toutefois à confirmer définitivement en conditions réelles
Calcium sulfoaluminate (CSA) cements can be a solution to improve the environmental impact of concrete, thanks to the lower energy consumption and CO2 emission during their production. Their expansive, fast setting and rapid hardening properties contribute to the growing interest of engineers for concrete construction and prefabrication. However, the durability of sulfoaluminate concretes is not clearly assessed yet, especially in acidic and marine environments, as well as carbonation and chloride induced rebars corrosion. Thus, the aim of this thesis is to study and compare the durability, related to these attacks, of concretes made with CSA cements to ordinary Portland and GGBS cements based reference concretes. This study is founded on microstructural (solid phases, porosity) and performancial (durability indicators, natural and accelerated tests) characterization. Beforehand, the application of these characterizations to CSA concretes is verified and discussed. These show a good resistance to acidic environments when submitted to dynamic leaching tests at pH 3 and 5. The study of chloride ingress (steady and non-steady state migration, binding isotherms and tidal zone exposition) and carbonation (natural and accelerated) in concretes of 8 different laboratory CSA cements enabled us to find achievable means to improve their potential durability. These means were applied to the production of a performant industrial CSA cement, whose good durability has still to be assessed on field conditions

Les laitiers de haut-fourneaux (LHF) sont utilisés dans les ciments notamment en raison de l'amélioration globale de la durabilité qu'ils confèrent aux matériaux de construction. Cependant, les liants à forte teneur en laitiers présentent une hydratation plus lente et donc un développement de résistances mécaniques moins rapide que les ciments portland (CP). Des solutions existent pour accélérer le durcissement des liants à base de LHF, notamment l'emploi d'activants chimiques à base de chlorures. Les effets de l'activation des systèmes CP/LHF sur les développements de résistances mécaniques sont abordés dans la littérature, mais leur durabilité est assez peu étudiée. Cette thèse a donc pour but d'évaluer les conséquences de l'utilisation d'activants chlorés sur la durabilité des liants contenant 50% de LHF. D'abord, une analyse des indicateurs de durabilité permet de voir comment l'activant modifie les propriétés de transfert du matériau durci ainsi que sur la structuration du réseau poreux. Cela permet de vérifier que l'activation à jeune âge n'hypothèque pas la durabilité des bétons avec laitier à long terme. Ensuite, une étude de l'état de corrosion des aciers dans ces mélanges est entreprise afin d'évaluer l'impact des activants chlorés, reconnus pour créer de la corrosion dans le béton armé. Des essais électrochimiques couplés à de l'observation directe ainsi qu'une étude microstructurale sur le devenir des chlorures au sein de la matrice cimentaire est effectuée. La faculté des mélanges à base de laitier à fixer les chlorures dans les hydrates est notamment étudiée. Enfin, le comportement des chlorures fixés est analysé, dans le cas où la carbonatation viendrait déstabiliser les hydrates. L'impact d'un relargage éventuel de chlorures est évalué à l'aide d'un modèle couplé carbonatation/diffusion des chlorures
Ground granulated blast-furnace slags (GGBS) are used in cements in particular because of the overall improvement in the durability of building materials. However, binders with a high content of GGBS have a slower hydration and thus a development of mechanical strength slower than Portland cements (PC). Solutions exist to accelerate the hardening of GGBS-based binders, including the use of chemical activators based on chlorides. The effects of activation on the mechanical strength developments of PC/GGBS blends are discussed in the literature, but the durability of these systems is not very developed. This thesis aims to evaluate the consequences of the use of chlorides activators on the durability of binders containing 50% of GGBS. Firstly, an analysis of the durability indicators shows how the activator modifies the transfer properties of the hardened material as well as the structuring of the porous network. This allows to verify that the activation at early age does not decrease the long-term durability of concretes with GGBS. Then, a study of the corrosion state of steels in these blends is realized to evaluate the impact of chlorides activators, known to initiate corrosion in reinforced concrete. Electrochemical tests coupled with direct observation as well as a microstructural study on how interact chlorides in the cement matrix are carried out. The ability of GGBS-based blends to bind chlorides in hydrates is studied in particular. Finally, the behavior of binded chlorides is analyzed, in the case where the carbonation would destabilize the hydrates. The impact of a possible release of chlorides is evaluated using a coupled model carbonation / chloride diffusion

La corrosion des armatures en acier est la plus grande cause de défaillance des ouvrages en béton armé. Ce phénomène électrochimique est déclenché par la présence d’ions chlorure en quantité suffisante au niveau de l’armature ou la carbonatation du béton d’enrobage (action du CO2). L’objectif de cette thèse est de développer des modèles utilisables par l’ingénieur dans une démarche d’approche performantielle pour la prédiction de la durée d’utilisation des ouvrages en béton armé soumis à l’attaque par les ions chlorure ou la carbonatation, suite à l’amorçage et au développement de la corrosion en leur sein. Il s’agit du développement de trois modèles : un modèle de transfert des ions chlorure, un modèle de carbonatation et un modèle de corrosion qui permettent d’estimer la durée d’initiation et la durée de propagation de corrosion. Ces modèles prennent en considération les facteurs liés au matériau (i.e. indicateurs de durabilité), à la mise en œuvre, à l’environnement et à la géométrie. La démarche adoptée pour le développement de ces modèles repose sur l’exploitation de plusieurs bases de données, sur des ouvrages vieillissants et des corps d’épreuve de bétons, issues de la littérature (BHP-2000, Perfdub, etc.). Ces exploitations ont permis d’améliorer la capacité prédictive de modèles existants (transfert des ions chlorure) et de développer de nouveaux modèles (carbonatation et corrosion)
Steel reinforcement corrosion is the major cause of failure in reinforced concrete structures. This electrochemical process is induced by presence in sufficient quantity of chloride ions at the reinforcement or by concrete carbonation (CO2 action). This thesis aims to develop engineering performance-based models for the service life prediction of reinforced concrete structures subjected to chloride-induced or carbonation-induced corrosion initiation and propagation. It consists in developing three models : a chloride ions ingress model, a carbonation model and a corrosion model that allows to estimate the corrosion initiation time and propagation time. These models consider factors related to the material (i.e. durability indicators), processing, environment and geometry. The approach used to develop these models is based on using several literature databases of ageing structures and concrete testing specimens (BHP-2000, Perfdub, etc.). The use of these data allowed to improve the predictive capacity of existing models (chloride ingress model) and to develop new models (carbonation and corrosion models)

L'experimentation est conduite sur des poutres de beton arme de grandes dimensions, conservees sous charge dans des ambiances agressives: brouillard salin ou atmosphere enrichie en gaz carbonique. L'etat des armatures est suivi periodiquement a l'aide de mesures du potentiel electrochimique (non destructives). Les evolutions de l'endommagement macroscopique et microscopique du beton, de la profondeur de carbonatation et de la penetration des chlorures sont egalement determinees. L'ensemble des observations permet de proposer un modele quadriphasique du processus de corrosion qui prend en compte l'influence de la fissuration de service. La premiere phase dite d'incubation correspond a la penetration des agents agressifs le long des fissures, jusqu'aux armatures. La phase d'initiation debute alors et se traduit par une corrosion localisee en fond de fissures. Lorsque les oxydes formes obstruent le fond des fissures, la corrosion est fortement ralentie: c'est la phase dormante. Durant cette phase, les agents agressifs continuent de penetrer a travers le beton d'enrobage et, lorsque le seuil critique de depassivation est atteint au niveau des armatures, commence la phase de propagation. La corrosion se developpe, conduisant a une fissuration secondaire ramifiee qui contribue a accelerer le phenomene. L'influence du beton d'enrobage est nettement mise en evidence alors que l'ouverture des fissures primaires, qui n'excede pas 0,5 mm dans les poutres experimentees, n'a pas un role determinant. En revanche, l'endommagement des interfaces pate-granulat accelere la penetration des agents agressifs dans les zones tendues des poutres. La comparaison entre les deux ambiances indique que le brouillard salin est plus agressif que l'atmosphere carbonique. De ce fait, le principal parametre a prendre en compte dans la reglementation ne devrait pas etre la seule epaisseur d'enrobage mais un enrobage equivalent integrant le niveau de sollicitation, l'agressivite du milieu et la position des armatures

Le béton armé est sans conteste le matériau le plus utilisé dans la construction, et permet de réaliser la plupart des infrastructures dans tous les pays du monde. Cependant, sa durabilité peut être compromise de façon prématurée par la corrosion des aciers, qui est la pathologie considérée comme la plus dangereuse vis-à-vis du maintien de l'intégrité des ouvrages de Génie Civil. Afin d'obtenir une meilleure résistance face au risque de corrosion, l'acier conventionnel (AC) peut être soumis à divers traitements, dont les plus connus sont le recouvrement superficiel par d'autres matériaux métalliques à base de zinc (AG), ou par des revêtements doubles métal-polymères (AD). Cependant, il existe aussi des barres d'acier thermiquement traitées (ATT) qui ne sont actuellement quasiment pas utilisées en tant qu'armatures. De ce fait, leur comportement face aux mécanismes de détérioration comme la corrosion en milieu cimentaire causée par les chlorures ou la carbonatation est encore très peu connu. Dans ce travail, le comportement des différentes barres d'acier mentionnées ci-dessus a été étudié. Des éprouvettes prismatiques de béton, incluant ces types d'armatures, ont été fabriquées avec deux rapports E/C : 0,45 et 0,65. Avant leur utilisation, les barres ont été caractérisées mécaniquement et métallographiquement. Ensuite, les éprouvettes ont été placées dans différentes conditions d'exposition : un environnement urbain/industriel ou côtier, et un environnement contrôlé en laboratoire. Des mesures du potentiel de corrosion, de résistance de polarisation linéaire, et de spectroscopie d'impédance électrochimique ont été mises en œuvre durant la période d'exposition. Pour chaque série, la teneur critique en chlorures a été déterminée, et la progression de la profondeur de carbonatation a été suivie. En outre, en induisant un couple galvanique par effet de la carbonatation, des mesures originales du comportement électrochimique de ces barres ont été effectuées puis une analyse, en s'appuyant sur une modélisation en éléments finis, en a été faite. Dans l'environnement contrôlé en laboratoire, l'ordre de dépassivation des différentes barres exposées a été observé de façon similaire pour les deux rapports E/C, à savoir : ATT, AC, AG et enfin AD. Une teneur critique en chlorures plus élevée a été obtenue pour les barres AG et AD. Cependant, lors de l'inspection visuelle, les dommages causés sur les armatures AG étaient plus élevés que sur les autres types de barres. Enfin, durant l'étape de propagation, la densité de courant de corrosion des barres ATT et des barres AG s'est révélée inférieure
Corrosion of reinforcing bars in concrete is considered as the most important problem that affects the integrity of the civil structures. In order to obtain a better resistance to corrosion, various superficial processes as coatings with zinc (AG) or such as the dual covering metallic-polymeric (AD) are applicate to ordinary steel bars (AC). On the other hand, steel bars with thermal treatments (ATT), principally developed as an alternative to improve the mechanical properties without the use of ferroalloys, are not used in concrete. The behavior of these kinds of bars in front of mechanisms of deterioration as the corrosion induced by chlorides or carbonation has not yet been studied. In this work, all these various steel bars (AG, AD, AC and ATT) were embedded in prismatic specimens of concrete made with two ratio water/cement: 0.45 and 0.65. Previously, steels bars were characterized by mechanical tests and metallographic identifications. Then, specimens were placed in several sites of exposition: urban/industrial environment, or coastal environment, or controlled atmosphere in laboratory. During these expositions, measurements of corrosion potential, linear polarization resistance, and electrochemical impedance spectroscopy were regularly carried out. For each type of steel bar, chloride threshold level and progress of the carbonation depth were determined. Furthermore, by means of the induction of a galvanic couple during design of new samples, the electrochemical behavior of the steels bars AC, ATT and AG was followed up experimentally and then analyzed with finite element model. It was founded that the different steels bars exposed in controlled atmosphere of laboratory followed a same sequence in depassivation for both ratios water/cement: ATT, AC, AG and AD. The chloride thresholds were higher for steels bars AG and AD. However, visual inspection showed that the morphology of damages caused on AG bars was most important compared with the other steel bars. In the propagation phase, the corrosion current density of the ATT bars was lower, even to that obtained by the AG bars

La corrosion des aciers est l’une des principales causes de dégradation des ouvrages en béton armé, notamment en façade maritime. Cette dégradation est due à la diffusion d’ions chlorures qui peut se produire dans les ouvrages immergés dans l’eau de mer, ceux subissant des cycles de marnage et également ceux soumis aux embruns marins. La corrosion peut être également due au processus de carbonatation du béton, responsable de la baisse du pH et par conséquent de la dépassivation des aciers. Cette thèse propose un modèle physico-chimique pour décrire la penetration des agents agressifs dans la première phase de la corrosion des aciers. Cette phase dite « incubation » correspond à la phase pendant laquelle le transfert des espèces agressives se produit dans le béton. Afin d’identifier clairement les paramètres de transport de ce modèle macroscopique une démarche multi-échelle est mise en oeuvre pour rendre compte de la diffusion dans les bétons. Cette démarche repose sur une description détaillée de la microstructure du béton (constituants et morphologie) par des méthodes de changement d’échelle. Cette approche se veut la plus exhaustive possible avec la prise en compte explicite de l’influence du rapport eau sur liant de la pâte de ciment, des propriétés des granulats et de la zone de transition entre la pâte et les granulats. Après identification de tous les paramètres d’entrée, le modèle physico-chimique macroscopique est utilisé pour évaluer la durée de vie d’un béton soumis au couplage carbonatation -chlorures en milieu insaturé. Suite à une validation du modèle par comparaison à des résultats expérimentaux, plusieurs applications de ce modèle sont présentées
The corrosion of steels is one of the principal causes of degradation of reinforced concrete structures, especially in front of the sea. This degradation is due to the diffusion of chloride ions that can occur in the structures immersed in the seawater, those undergoing tidal cycles and also those subjected to the sea spray. The corrosion can also be due to the carbonation process of the concrete, which is responsible for decreasing of the pH and consequently the depassivation of the steels. This thesis proposes a physicochemical model to describe the penetration of aggressive agents in the first period of steel corrosion. This so-called « incubation » period corresponds to the period during which the transfer of aggressive species occurs in the concrete. In order to clearly identify the transport parameters of thismacroscopic model, a multi-scale approach is implemented to consider the diffusion in concretes. This approach is based on taking into account the detailed description of the microstructure of the concrete (constituents and morphology) by upscaling methods. This multi-scale approach is intended to be the most comprehensive with the consideration of the influence of the water-to-binder ratio of the cement paste, the properties of the aggregates and the transition zone between the cement paste and the aggregates. After identifying all the input parameters, the macroscopic physico-chemical model is used to evaluate the service life of a concrete subjected to coupling carbonation -chlorides in unsaturated medium. Following a validation of the model compared to experimental results, several applications of this model are presented

L'objectif de ce travail est de contribuer à la valorisation du métakaolin en substitution du ciment dans les matrices cimentaires en levant certains verrous scientifiques. En effet l'utilisation d'additions minérales modifie les propriétés des bétons, que cela soit à l'état frais, à l'état durcissant ou à l'état durci. A l'état frais, la rhéologie a été étudiée avec l'adaptation et l'application d'une méthode de formulation pour béton auto-plaçant. Cette application a été développée jusqu'à l'échelle industrielle. A l'état durcissant, la phase d'hydratation a été étudiée en s'attardant particulièrement sur l'effet de la nature du ciment substitué sur la réactivité du métakaolin. Enfin à l'état durci deux caractéristiques de durabilité ont été explorées, à savoir l'effet du métakaolin sur la carbonatation dans une première partie, puis sur la diffusion des chlorures dans une deuxième partie
The objective of the present thesis was to promote the use of metakaolin as substitute for cement in cementitious matrix by unlocking several scientific challenges. The use of mineral additions modifies the properties of concrete in the fresh state, the hardening state and the durability state. In the fresh state, rheology was studied by adapting a formulation method for self-compacting concretes. This application was then further developed to be used at an industrial scale. In the hardening state, the hydration phases were analysed with a focus on the effect of the cement nature on the metakaolin reactivity. Finally in the durability state, two particular features were studied:1) the metakaolin effect on the carbonation kinetic and 2) the effect of the metakaolin on the chloride diffusion

The specifications for concrete patch repair mortars usually entail mechanical properties such as compressive strength and tensile strength. However, these properties may not directly relate to the desired performance in relation to durability and prevention of reinforcement corrosion. In addition, the literature does not show any direct relationship between compressive strength and durability properties of concrete and mortars in natural exposure conditions. Relevant performance requirements, such as carbonation resistance and chloride resistance, are usually not considered despite the fact that they have a direct influence on the durability performance of concrete repair mortars. The widespread premature failure of patch repairs which meet the existing compressive strength criteria implies that the use of compressive strength as a performance indicator may not provide a reliable measure of the durability performance. Therefore it can be argued that modern concrete repairs should be based on durability considerations, rather than compressive strength. In this study, an experimental investigation was conducted to determine the durability performance of patch repair mortars. Experimental results were analysed to investigate the correlations that exist between (i) electrical conductivity (Chloride Conductivity Index test) and rate of chloride-ion diffusion (bulk diffusion test), and (ii) gas permeability (Oxygen Permeability Index test) and rate of carbonation (accelerated carbonation test) in patch repair mortars. Eight mortar mixes were used in the investigation, including four commercially available repair mortars and four laboratory-made mortar mixes. To vary the pore structure of the laboratory mixes, different water/binder ratios (0.45 and 0.60) and binder types (100% Portland cement and 50/50 blend of Portland cement/blast furnace slag) were used to make the mortar specimens. Two curing conditions (dry and moist) were adopted with the aim of investigating the influence of curing on durability performance of patch repair mortars. Test results indicate good correlations between electrical conductivity and rate of chloride diffusion (correlation coefficient of 0.9112), and between oxygen permeability and rate of carbonation (correlation coefficient of 0.6751). This correlation was mainly attributed to the fact that these material properties largely depend on the pore structure (specifically the size, connectivity and tortuosity of pores). The good correlation further implies that electrical conductivity and oxygen permeability of repair mortars as evaluated by the CCI and OPI tests may provide a reasonable measure of chloride resistance and carbonation resistance respectively. However, the prediction of chloride ingress and carbonation depth from the electrical and gas permeability properties respectively, ought to be implemented within the range where reasonable correlation can be established. The results also showed that the durability performance of repair mortars in terms of chloride and carbonation resistance is sensitive to material factors, such as w/b ratio, curing type and binder type, which directly influence penetrability. Service life models for predicting chloride ingress and carbonation in the patch repair mortars used in this study were developed based on modified Fickian equations. The prediction profiles for chloride penetration were developed from a modified solution to Fick's second law of diffusion, while the carbonation depth prediction profiles were developed from the square-root-of-time law. Chloride penetration and carbonation depth could be predicted using the developed profiles. Though several assumptions that should be verified and/or modified in future work were made, the modelling results of this study serve as useful framework for evaluating the resistance to chloride ingress and carbonation in patch repair mortars.

The work described in this thesis has been performed in order to gain further understanding of the relationship between the microstructural characteristics of some common types of medium to high strength sustainable Self- Compacting Concretes (SCCs) (especially the internal pore structure, the interfacial transition zone (ITZ), and chemical composition) and the carbon dioxide and chloride diffusivities. This was done by evaluating the diffusion coefficients with the aid of some selected and modified accelerated tests at the macro scale. The internal composition and microstructure form were quantitatively analyzed for one normal vibrated concrete (NVC), one normal SCC mix (R-SCC) and three different types of sustainable SCCs with relatively high partial replacement of cement at micro and nano scales and linked with the macro scale tests. This was done by using a wide range of advanced techniques such as thermo gravimetric analysis (TGA), x-ray diffraction (XRD), mercury intrusion porosimetry (MIP) and scanning electron microscopy (SEM). In addition, the penetration parameters from the macro short term tests were used to simulate the results mathematically in order to assess the long term behaviour of the concrete mixes. Possible mechanisms, in multi-scale terms, are proposed to explain the overall response of both normal and sustainable medium to high strength grade SCCs to the degradation caused by carbonation and chloride penetration in harsh environments. The findings of the research will contribute to deeply understand the role of the internal microstructure of sustainable SCC in determining the carbonation and chloride penetration. The recommendations derived from this research are fundamental to achieving more durable sustainable SCC with longer service life for applications in aggressive environments.

La maîtrise et l'optimisation de la durabilité des structures en béton et des produits préfabriqués en particulier constituent un enjeu particulièrement important pour l'ensemble des acteurs (maîtres d'ouvrages, maîtres d'œuvres, industriels de la préfabrication). L'objectif des travaux développés dans le cadre de cette thèse de doctorat est de faire le point sur les connaissances actuelles et les besoins vis-à-vis de la tenue des armatures à la corrosion et de renforcer la compréhension phénoménologique de ces mécanismes de dégradation. In fine, ces travaux permettent d'aboutir à l'élaboration d'un modèle permettant d'étudier la pénétration des ions chlorure en conditions hydriques variables ou non, la carbonatation, le couplage des deux mécanismes, et le phénomène de corrosion des armatures en conditions hydriques variables. Une campagne expérimentale permet de caractériser les propriétés mécaniques et physico-chimiques de différents bétons et leur comportement au regard des principaux environnements agressifs pouvant conduire à la corrosion des armatures. L'objectif final de ces travaux réside dans l'élaboration d'un modèle simple et fiable permettant de considérer le couplage de tous les phénomènes importants liés à la durabilité des bétons, de la pénétration des espèces agressives à l'amorçage et à la propagation du phénomène de corrosion des armatures. En première application, le modèle peut être utilisé pour des calculs de durée de vie des ouvrages en béton armé dans un contexte probabiliste, cela en adoptant un critère de quantité de produit corrodé et non pas seulement de dépassivation, il permet alors d'optimiser le dimensionnement des enrobages en prenant en compte les propriétés intrinsèques des matériaux et les conditions environnementales. Cette étude permet de quantifier les gains de durabilité de produits exposés à un risque de corrosion des armatures obtenus en augmentant la compacité du béton, en optimisant les caractéristiques du liant et le process de fabrication
Mastering and optimizing the durability of concrete structures and especially prefabricated products is a major challenge for the entire profession. The goal of this work is to focus on the knowledge and the needs for assessing the performance of the reinforcement to the corrosion phenomenon and to heighten the mechanisms comprehension. Finally, this work allows coming up with a model able to simulate the penetration of chlorides in saturated environment or with wetting-drying cycles, the carbonation, the combination between these two mechanisms and the reinforcement corrosion in variable hydric conditions. A huge experimental campaign leads to a complete characterization of the mechanical, chemical and physical properties of several concretes and their performance towards to the main aggressive environment for the reinforcement corrosion. The final goal of this research project is to come up with a comprehensive and accurate model which should be able to consider all the most important coupled phenomena linked to the concrete durability, from the penetration of aggressive species to the real initiation of the corrosion phenomenon. First practical applications will consist in assessing the lifetime of the concrete structures in a probabilistic way and optimizing the design of concrete cover thickness by taking into account the intrinsic properties of the materials and the environmental conditions. This work allows quantifying the gain in durability performance of concrete product exposed to a risk of reinforcement corrosion obtained by increase the density level, optimize the cast process, or optimize the characteristic of the binder

Carbonation and chloride attack are the main causes of the corrosion of rebars in concrete. In the past, attempts have been made to develop various sensors to monitor these deterioration mechanisms. However, amongst the most popular sensors, the Electrical Resistance Sensor cannot differentiate each individual deterioration mechanism, whilst the long term stability and the need for recalibration are the two main limitations of the Fibre Optical Sensor. The optical fibre Raman system, combining the fingerprint characterisation capacity of Raman spectroscopy and the light transmission capacity of optical fibres, offers the potential for monitoring the deterioration mechanisms in concrete. This thesis reports a detailed study to investigate the qualitative and quantitative working capacity of the optical fibre Raman system for characterising carbonation and chloride attack in cementitious materials. Two optical fibre Raman systems were developed to characterise the deterioration mechanisms in hardened PC and PFA/GGBS blended pastes, namely, a ‘coaxial’ optical fibre Raman system for characterising carbonation and a ‘transmission’ optical fibre Raman system for characterising chloride attack, respectively. The results showed that the calcium carbonate polymorphs (e.g., calcite, vaterite and aragonite) formed during the carbonation were successfully identified and the variation in the content of calcium carbonates over depth was also clearly differentiated and verified by bench-mounted Raman and Thermogravimetry (TG) analyses, demonstrating its good potential for establishing carbonation profiles. On the other hand, the chlorides and the chloride content were also successfully established and verified by bench-mounted Raman spectroscopy and chloride meter. The results from this thesis demonstrate a good potential for the optical fibre Raman system to be developed into a novel sensor system for monitoring the durability of concrete structures. However, the insufficient working capacity and the interferences from the addition of PFA and GGBS still need to be addressed in the future.

The concrete lining of traffic tunnels is normally exposed to combined carbon dioxide (C02) and chloride environments. The ingress of CO2 and chloride ions into concrete in these environments could be different from that in I their independent effects, but relatively little is known about this, particularly for high performange concretes (HPCs) that are commonly used in tunnel linings. In this project, changes to hydrated cement pastes and concretes containing pulverised fuel ash (PF A) and microsilica (MS) during both independent and combined exposures to CO2 and chlorides were studied. Effects of these test regimes on the rate of carbonation and chloride ingress were quantified by analysing profiles of the consumed hydroxyl ions and chloride ions. The suitability of using Autoclam Permeability System, Permit Ion Migration Test and Covercrete electrode array for assessing the extent of carbonation and chloride ingress in normal and high performance concretes in the above exposure environments was explored. The results indicated that exposure to chlorides before subjecting to carbonation significantly slows down the progress of carbonation in concretes. However, the influence of carbonation on chloride ion ingress depends on concrete type and carbonation degree. It was found that this combination could severely compromise the tunnel lining durability, particularly for concretes containing PF A, due to the formation of microcracks in carbonated concretes. Autoclam and Permit enabled an assessment of the variations in permeation properties due to the test regimes, and the Covercrete electrode array could identify the carbonation depth and the extent of chloride ingress.

A busca pelo desenvolvimento sustentável impulsiona a pesquisa voltada ao uso de resíduos industriais na construção civil. Os resíduos da indústria de cerâmica vermelha podem ser utilizados como material pozolânico e já existem trabalhos que apontam para essa possibilidade. Porém, existe uma lacuna quanto à avaliação da durabilidade do concreto produzido com esse resíduo. Este trabalho analisa os efeitos da substituição de parte do cimento por resíduos de tijolos moídos (RTM) frente à corrosão da armadura provocada pela carbonatação e pela ação de cloretos. Para isso, foram preparados concretos convencionais (CCV) nos traços (aglomerante: materiais secos) 1:3,18; 1:4,11; e 1:5,58, cada um com 10%, 20% e 30% de RTM. Foram também analisados concretos autoadensáveis (CAA) com traço de 1:3 e 20%, 30% e 40% de substituição de resíduos, além dos traços sem resíduos usados como referências. O uso do RTM diminuiu o índice de vazios nos traços, porém houve perdas de resistência que alcançaram até 19% aos 90 dias. Os concretos autoadensáveis (CAA) mostraram melhores resultados de resistência apresentando melhoria em todos os traços para a mesma idade dos CCV, atingindo 11% de aumento. Em relação aos ensaios de carbonatação, os concretos com RTM tiveram maiores profundidades de carbonatação do que os concretos de referência, fato explicado pela redução da sua reserva alcalina. No entanto, esse aumento não inviabiliza o uso de concretos com RTM, principalmente para dosagens de menor porosidade e adições em patamares de substituição menos elevados Para a penetração de cloretos o RTM apresentou bom resultado, tanto para os CCV como para os CAA. Embora o RTM tenha diminuído o teor crítico de cloretos, ele aumentou o tempo necessário à despassivação das armaduras. Fato explicado pelo refinamento dos poros e pela maior capacidade de fixação de cloretos. Esse resultado é bastante animador para o uso deste resíduo em concretos convencionais expostos a ambientes marinhos ou como finos para concretos autoadensáveis.
The search for the sustainable development boosts the research focused on the use of industrial wastes civil construction. The industry wastes of red ceramics can be used as pozzolanic material and there are already works that point out for this possibility. However, there is a knowledge gap regarding the durability evaluation of the concrete produced with this waste. This research analyzes the effects substitution of part of the cement by crushed brick wastes (CBW) in relation to the reinforcement corrosion caused by carbonation and chloride action. For this reason, conventional cement concrete (CCV) was prepared in the traces (agglomerate: dry materials) 1:3,18; 1:4,11; and 1:5,58, each one with 10%, 20% and 30% of CBW. Self-compacting concrete (SCC) was also analyzed with trace of 1:3 and 20%, 30% and 40% of waste substitution, besides the traces with no waste used as references. The use of CBW decreased the rate of gaps in the traces, but there were losses of resistance that reached up to 19% in 90 days. The self-compacting concrete (SCC) showed better resistance results presenting improvement in all traces for the same age of the CCV, reaching 11% of increase. In regard to the carbonation tests, the CBW concretes had greater depths of carbonation than the reference concrete; fact explained by the reduction of its alkaline reserve. Nevertheless, this increase does not hamper the use of CBW concretes, especially for dosages of lower porosity and additions in lower substitution standards. For the chloride penetration, the CBW presented good result both for the CCV and the SCC. Although the CBW has diminished its critical chloride content, it increased the necessary period of time for reinforcement despassivation This fact was explained by the refining of pores and by the larger capacity of chlorides fixation. This result is very satisfactory for the use of this waste in conventional concretes exposed to marine environments or as fine aggregate for self-compacting concretes.

A durabilidade dos edifícios em concreto armado só pode ser alcançada se for atendido um conjunto de requisitos e critérios nas etapas de projeto, execução e manutenção, nas quais os materiais envolvidos precisam ser adequadamente especificados, produzidos e empregados, de acordo com as respectivas rotinas dessas etapas. Assim, o requisito genérico de proteção das armaduras de aço carbono, para a durabilidade de estruturas, é contemplado por vários critérios, desde a etapa de projeto. Esta dissertação trata dos critérios pertinentes à qualidade do concreto e visou contribuir para a evolução de procedimentos para o controle tecnológico de concretos estruturais de cimento Portland, em atmosferas marítimas e urbanas, em clima tropical, com vistas à proteção de armaduras de aço carbono. O programa experimental estudou três lotes diferentes de concretos estruturais, semelhantes quanto a especificações básicas de abatimento: 10 ± 2 cm, de fck 30 MPa e para ambiente classe III da NBR 6118 (2003). O objetivo foi caracterizar e analisar a influência de propriedades físicas do estado fresco sobre o estado endurecido de concretos, dentro da faixa de consistência citada, com ênfase nas propriedades relacionadas ao teor de ar das misturas e suas respectivas resistências à carbonatação e ao ingresso de íons cloreto. A amostragem, os ensaios no estado fresco e a moldagem dos corpos-de-prova foram realizados durante três concretagens conduzidas em duas obras de edifícios de múltiplos andares, na cidade de João Pessoa/PB. As composições e a produção dos lotes ficaram a cargo da central dosadora e fornecedora do concreto em cada obra. O lote inicial serviu para o treinamento da equipe, em campo e laboratório, com amostragem de seis caminhões. Os outros dois lotes, identificados como 1 e 2, foram caracterizados em condições bastante semelhantes, sempre pela mesma equipe, e representados pela amostragem de cinco e seis caminhões, respectivamente. As propriedades medidas no estado fresco foram: abatimento do tronco de cone pela ABNT NBR NM 67 (1998); teor de ar por método pressométrico da ABNT NBR 47 (2002); massa específica pela ABNT NBR 9833 (2008); relação água/materiais secos por analogia à ABNT NBR 9605 (1992); e compactabilidade dos concretos adensado e não adensado, por adaptação da BS EN 12350-4 (2008). Para o estudo das propriedades no estado endurecido, os corpos-de-prova eram cilíndricos, com 10 cm de diâmetro e 20 cm de altura, e foram maturados por dois métodos adaptados do Tipo A da ASTM C 684 (1999). Certas propriedades foram ainda caracterizadas para condições normais de cura da ABNT NBR 5738 (2003). As propriedades estudadas no estado endurecido foram: resistência à compressão pela ABNT NBR 5739 (2007); resistência à tração por compressão diametral pela ABNT NBR 7222 (1994); absorção de água por capilaridade pela ABNT NBR 9779 (1995); absorção de água total e índice de vazios pela ABNT NBR 9778 (2005); resistência à carbonatação em câmara com CO2 (5%; UR 65 + 10 %; 23 + 3 °C); e resistência à penetração de cloretos e CO2, por três ciclos de um dia de molhagem e 27 dias de secagem, entre sete e 91 dias. De modo complementar, foram moldados corpos-de-prova específicos de aço e concreto, para medidas de potencial de circuito aberto, com vistas à continuidade de pesquisas sobre envelhecimento acelerado. A maioria das propriedades foram medidas por duas repetições, constando os resultados individuais em apêndices. A análise inicial dos resultados dos três lotes foi descritiva e resumiu em tabelas o valor médio, desvio padrão, valor máximo, valor mínimo, a amplitude e o coeficiente de variação de cada lote. As propriedades de cada lote também foram comparadas por análise de variância e, ao final, foram correlacionadas de modo conjunto, independentemente do lote de origem. Nesse caso, foram destacadas as melhores correlações entre propriedades, independentemente dos materiais constituintes de cada concreto. Entre as propriedades do estado fresco, destacaram-se as seguintes: a) o abatimento do tronco de cone apresentou correlação forte e inversa (r² = -0,802, para lotes 1 e 2) com resistência à compressão após um dia de cura acelerada em temperatura moderada, em método similar ao Tipo A da ASTM C 684 (1999); b) o teor de ar pelo método pressométrico apresentou correlações de razoáveis a fortes com a resistência à compressão a 28 dias (r² = -0,698, para lotes 1 e 2), com a resistência à tração por compressão diametral por cura acelerada a sete dias (r² = -0,818, para lotes 1 e 2), com a profundidade de carbonatação em câmara de CO2 (r² = 0,699, para lotes 1 e 2) e com a profundidade de penetração de cloretos por três ciclos de imersão e secagem (r² = 0,625, para lotes 1 e 2); c) as medidas de compactabilidade do concreto adensado, ainda que realizadas em condições de campo, apresentaram várias correlações moderadas com outras propriedades do estado fresco e endurecido. No estado endurecido, destacaram-se as seguintes correlações: a) r² da ordem de -0,75 para a resistência à compressão a um dia dos concretos dos lotes 1 e 2, com cura acelerada de 0/24 horas em temperatura moderada, e a profundidade de carbonatação em câmara de CO2 (5%) e a de penetração de cloretos por três ciclos de imersão e secagem, ambas analisadas a 91 dias, tendo os corpos-de-prova recebido cura inicial acelerada de 24/48 h, em temperatura moderada, seguida de imersão normal por até sete dias; b) r² de -0,682 entre a resistência à tração por compressão diametral, com cura acelerada de 24/48h em temperatura moderada seguida de cura normal por até sete dias, e a profundidade de ingresso de íons cloreto, para os três lotes submetidos aos ciclos citados; c) valores de r² entre 0,521 e - 0,561 para as correlações entre a absorção de água por capilaridade, em corpos-de-prova submetidos à cura inicial acelerada em temperatura moderada por 24/48 h e em temperatura normal por até sete dias, e a resistência à carbonatação para os lotes submetidos aos ciclos citados. Assim, esta pesquisa conclui e propõe que, além do teor de ar no estado fresco, as resistências à compressão e à tração por compressão diametral entre um e sete dias, por cura acelerada do Tipo A da ASTM C 684 (1999) ou por duração adaptada da mesma, sejam propriedades que passem a ser avaliadas em concretos, com vistas a melhorar e controlar a sua resistência a agentes agressivos. Nesta pesquisa, a microestrutura nas primeiras idades do concreto mostrou ser mais determinante da rede de conexão de poros e do transporte de agentes agressivos do que a microestrutura em idades mais avançadas de hidratação; e trabalhos futuros devem confirmar esta interpretação. Espera-se que estes resultados possam estimular novas práticas de qualificação de concretos em estudos de dosagem ou no ato do recebimento de concretos pré-misturados, especialmente em ambientes mais agressivos às armaduras, com vistas à futura evolução de procedimentos da ABNT NBR 12655 (2006).
Durability of reinforced concrete buildings can only be achieved if a set of requirements and criteria is met in the design, execution and maintenance phases, in which the materials must be properly specified, manufactured and employed, according to the respective procedures of these phases. Therefore, the general requirement of protection of carbon steel reinforcements, for structural durability, is contemplated by several criteria as early as the design phase. This thesis addresses the relevant criteria concerning concrete quality and its objective was to contribute to the evolution of technological control procedures for Portland cement structural concretes, in urban and maritime atmospheres in tropical climate, aiming at the protection of carbon steel reinforcements. The experimental program studied three different batches of structural concrete that were similar in terms of slump test basic specifications: 10 ± 2 cm, fck = 30 MPa and class III of ABNT NBR 6118 (2003) environmental classification. The goal was to characterize and analyze the influence of physical properties of fresh concrete on hardened concrete, within the previously mentioned consistency range, with emphasis on those properties related to air content of the mixtures and their respective carbonation and chloride ion penetration resistances. Sampling, fresh concrete tests and specimen molding were conducted during three cast-in-place concretes in two construction sites of multi-storey buildings in the city of João Pessoa, state of Paraíba, in Brazil. The batch plant supplying the concrete for each construction site was responsible for batch composition and production. The initial batch was used to train the team, in the field and laboratory, with a sample of six trucks. The other two batches, identified as 1 and 2, were characterized under very similar conditions, always by the same team, and the samples consisted of five and six trucks, respectively. The properties measured in fresh concrete were: slump test according to Brazilian norm ABNT NBR NM 67 (1998); air content by the pressure method of ABNT NBR 47 (2002); bulk density according to ABNT NBR 9833 (2008); water/dry material ratio by analogy with ABNT NBR 9605 (1992); and compactability of compacted and non-compacted concretes, by adaptation of BS EN 12350-4 (2008). In order to study the properties of hardened concrete, the specimens were cylindrical, measuring 10 cm in diameter and 20 cm in height, and were matured by using two methods adapted from Type A of ASTM C 684 (1999). Some properties were also characterized for normal curing conditions according to ABNT NBR 5738 (2003). The properties studied in hardened concrete were: compressive strength according to ABNT NBR 5739 (2007); splitting tensile strength as per ABNT NBR 7222 (1994); capillary water absorption according to ABNT NBR 9779 (1995); water absorption by immersion and void ratio following ABNT NBR 9778 (2005); carbonation resistance in CO2 chamber (5%; 65 + 10 % RH; 23 + 3 °C); and chloride and CO2 penetration resistance, using three one-day wet and 27-day dry cycles, between seven and 91 days. As a complement, specific steel and concrete specimens were molded to measure open circuit potential, aiming at the continuity of researches on accelerated ageing. Most properties were measured twice and the individual results are presented in appendices. Preliminary analysis of results from the three batches was descriptive and summarized in tables the mean value, standard deviation, maximum value, minimum value, amplitude and coefficient of variation for each batch. The properties of each batch were also compared using analysis of variance and, at the end, were correlated as a whole, regardless of the batch of origin. In this case, the best correlations among properties were highlighted, regardless of the materials used in each concrete. Among the properties of fresh concrete, the following are noteworthy: a) the slump test presented strong and inverse correlation (r² = -0.802, for batches 1 and 2) with compressive strength after one day of accelerated curing at moderate temperature, using a method similar to Type A of ASTM C 684 (1999); b) air content in the pressure method presented reasonable to strong correlations with compressive strength at 28 days (r² = -0.698, for batches 1 and 2), with splitting tensile strength with accelerated curing at seven days (r² = - 0.818, for batches 1 and 2), with carbonation depth in CO2 chamber (r² = 0.699, for batches 1 and 2) and with chloride penetration depth after three immersion-drying cycles (r² = 0.625, for batches 1 and 2); c) compactability measurements of compacted concrete, despite being taken in field conditions, presented several moderate correlations with other properties of fresh and hardened concrete. In hardened concrete, the following correlations should be highlighted: a) r² was -0.75 for compressive strength at one day of the concretes from batches 1 and 2, with 0/24- hour accelerated curing at moderate temperature, and carbonation depth in CO2 chamber (5%) and chloride penetration depth after three immersion-drying cycles, both analyzed at 91 days, after the specimens underwent 24/48-hour initial accelerated curing at moderate temperature, followed by normal immersion for up to seven days; b) r² was -0.682 between splitting tensile strength, with 24/48-hour accelerated curing at moderate temperature followed by normal curing for up to seven days, and chloride ion penetration depth, for the three batches submitted to the aforementioned cycles; c) r² values between 0.521 and - 0.561 for the correlations between capillary water absorption, in specimens submitted to initial accelerated curing at moderate temperature for 24/48h and at normal temperature for up to seven days, and carbonation resistance for the batches submitted to the aforementioned cycles. Therefore, this research concludes and proposes that, besides air content in fresh concrete, compressive strength and splitting tensile strength between one and seven days, using accelerated curing of Type A of ASTM C 684 (1999) or for a duration adapted from that norm, are properties that should be evaluated in concrete, with the purpose of improving and controlling resistance to aggressive agents. In this research, the microstructure of the early ages of concrete proved to be more determinant of the pore structure connection and of the transport of aggressive agents than the microstructure at later ages of hydration; and future studies should confirm this interpretation. It is expected that these results will encourage new practices for the qualification of concrete in mixture proportion studies or when ready mixed concrete is received at construction sites, especially in environments that are more aggressive to reinforcements, with a view to developing future procedures of ABNT NBR 12655 (2006).

In this research, two phases of work were conducted. First, an investigation into the durability concerns for precast prestressed concrete piles exposed to marine environments was conducted. The investigation characterized the durability concerns of chemical, biological, and physical deterioration mechanisms. The results of this study were used to develop potential high performance marine concretes (HPMC) that would be capable of 100+ year service lives in marine environments. Extensive durability testing and service life modeling of the HPMC was performed. Chloride ingress resistance was investigated using accelerated and long-term test procedures and the results used to perform service life modeling to predict the time before corrosion initiation. Sulfate resistance characterization was performed using multiple techniques to characterize the physical and chemical behavior of binder compositions containing binary or ternary mixes containing cement and supplementary cementitious materials (SCM's) subjected to a sulfate-laden environment. Accelerated carbonation testing and material characterization led to the finding of relationships in the chemical composition of mix designs and the observed durability and the results used to perform corrosion initiation service life modeling. An investigation into the influence of self-healing of cracked concrete led to fundamental findings on the behavior of chloride ingress for cracked concrete structures in marine environments. The results of this research led to the development of concrete mix designs capable of providing service lives over 100 years in Georgia's marine environments, as well as the advancement of the current state of knowledge on the durability characteristics of ternary mix designs.

The corrosion study in actual concrete structures is a subject enough discussed as much in the academic ambience as in the professional one. The subject is relevant due the numberless pathology cases of corrosion in structural concrete that are reported, almost daily, in the technical sphere. Nevertheless, few are the studies on this phenomenon in structural concrete of the first half of the 20th century. Is common sense that these structures were executed in a different way from the actual, and its behavior in front of corrosion may be differentiated, however, little are the studies that analyze the behavior of these structures in front of this pathology. This dissertation, based on bibliographical study and author s experience, look for not only analyze the concrete structure characteristics of the first half of 20th century and its influence on the rebar corrosion, as also propose a methodology for its detection. The proposition of this theme is based in fact that many execution and calculus methods were enough different from the actual concerning the concrete structures. The sodium chloride and calcium chloride addition in concrete, with purpose to accelerate the set and curing of concrete, the use of wooden pins as spacers are some characteristics showed in this work. As conclusion, is proposed an evaluation methodology which permits analyze such structures under the optics of its peculiar characteristics, aiming at the rebar corrosion.
O estudo da corrosão em estruturas de concreto armado atuais é um assunto bastante discutido tanto no meio acadêmico, como no profissional. O tema reveste-se desta importância devido aos inúmeros casos de patologias de corrosão em estruturas de concreto armado que são reportados, quase diariamente, ao meio técnico. No entanto, pouco se tem estudado a respeito desse fenômeno em estruturas de concreto armado da primeira metade do século XX. É senso comum que essas estruturas foram executadas de forma diferente das atuais e que seu comportamento frente à corrosão pode ser diferenciado; no entanto, são poucos os estudos que analisam o comportamento desse tipo de estrutura frente a essa patologia. Este trabalho, tendo por base um estudo bibliográfico e a experiência do autor, busca não só analisar as características das estruturas de concreto armado da primeira metade do século XX e suas influências sobre a corrosão das armaduras, como também propor uma metodologia para sua detecção. A proposição desse tema baseia-se no fato de que muitos métodos de execução e cálculo eram bastante diferentes dos atuais no que concerne a estruturas de concreto armado. A adição de cloreto de sódio e cloreto de cálcio no concreto, com a finalidade de acelerar a pega e a cura do concreto, e o uso de tarugos de madeira como espaçadores são algumas das características mostradas neste trabalho. Como conclusão é proposta uma metodologia de avaliação que permita analisar tais estruturas sob a ótica de suas características particulares, com vistas à corrosão da armadura.

Actuellement, il est largement reconnu que la durabilité des structures en béton armé, due à la corrosion des armatures engendrée par la carbonatation ou la pénétration des chlorures, peut être affectée largement par les conditions de cure et de serrage du béton coulé en place. Toutefois, les effets de ces conditions sur la qualité du béton ne sont pas encore entièrement comprises, puisqu'elles sont habituellement négligées (ou traitées superficiellement) dans les méthodologies actuelles de performance utilisées pour la spécification et contrôle de sa durabilité. Dans ce travail sont étudiés les effets des conditions habituelles de mise en place (y compris le serrage) et cure sur les propriétés de durabilité du béton, à savoir, la résistance à la carbonatation accélérée, le coefficient de migration des chlorures (dans des conditions non stationnaires), l'absorption d'eau et la perméabilité aux gaz (méthode CEMBUREAU). À cette fin, plusieurs bétons de différent composition, sans et avec cendres volantes, ont été soumis à deux principaux programmes expérimentaux. Dans le premier programme, trois bétons ont été soumis à une cure humide dans le laboratoire à différentes températures, entre 5 °C et 60 °C, et testés à différents âges, entre 28 et 182 jours, pour quantifier l'effet isolé de la température de cure sur les propriétés de durabilité du béton. Dans le deuxième programme, plusieurs éléments (dalles, poutres et poteaux) ont été coulés sur chantier, pendant l'hiver et l'été, après avoir été soumis à deux conditions différentes de serrage, vibré et non vibré, et démoulés à différentes à 24 h et 72 h. Les propriétés de durabilité du béton près de la surface et de cœur des éléments (propriétés réelles) ont ensuite été mesurées à différents âges, entre 28 et 364 jours, et comparées avec les propriétés des échantillons vibrés et curés en conditions normalisées (propriétés potentielles)
It is widely recognized that the long-term durability of reinforced concrete structures related to carbonation- and chloride-induced corrosion can be detrimentally affected by on-site placing and curing conditions of concrete. However, the effects of these conditions on concrete durability are still not fully understood, being usually overlooked in current performance-based specifications and control of concrete durability. In this work, the effects of realistic placing (including compaction) and curing conditions on the concrete durability-related properties most used in performance-based specifications are studied, such as the accelerated carbonation resistance, chloride migration coefficient (non-steady state conditions), water absorption and gas permeability (CEMBUREAU method). For that purpose, several concretes of different composition, with and without fly ash addition, were subjected to two main experimental programs. In the first program, the concretes were cured in the laboratory under several temperature regimes, ranging from 5 ºC to 60 ºC, and then tested at different ages, from 28 to 182 days, in order to evaluate the isolated effect of curing temperature on their durability-related properties. In the second program, several concrete elements (slabs, beams and columns) were cast outdoors, during the winter and summer, and subjected to different compaction (vibrated and not vibrated) and curing (demoulded after 24 h and 72 h) conditions. The durability-related properties of the inner and outermost concrete of the elements (actual properties) were then measured at different ages, from 28 to 364 days, and compared with those of standard specimens made of the same concrete (potential properties)

La carbonatation atmosphérique et la pénétration des ions chlorures sont les principales causes du déclenchement de la corrosion des armatures dans les bétons armés. Chacun de ses mécanismes a fait l'objet de nombreuses études dans la bibliographie. Pourtant les études prenant en compte en même temps les deux types d'agression sont rares. Le couplage des deux phénomènes existe cependant. La carbonatation intervient en effet dans tout ouvrage dès le décoffrage et il suffit que cet ouvrage soit aussi par exemple situé à proximité d'eau de mer pour que les embruns apportent des ions chlorures dans le matériau. L'objectif de cette thèse a été de quantifier les différentes phases de la matrice cimentaire et ses capacités de fixation des ions chlorures par le biais d'isothermes d'interactions pour des matériaux sains et carbonatés. Ces isothermes d'interactions obtenues notamment par la méthode des équilibres ont permis de voir que les matrices partiellement ou totalement carbonatées fixaient moins d'ions chlorures que les matrices saines. L'utilisation de moyens complémentaires d'analyses tels que l'analyse thermogravimétrique (ATG), la diffraction des rayons X (DRX) et la résonance magnétique nucléaire (RMN) a permis de mieux comprendre cette perte de fixation. Une diminution des sels de Friedel et des chlorures fixés par les C-S-H a été ainsi mise en évidence dans les matériaux carbonatés. Ces diminutions qui on été quantifiées sont attribuées à une modification sous carbonatation des phases capables de fixer les ions chlorures (phases aluminates, sulfo-aluminates et C-S-H). Des différences au niveau du réseau poreux ont été également mises en évidence par la porosimétrie par intrusion de mercure (diminution de la macroporosité et de la connectivité en général pour les matrices carbonatées par formation de CaCO3). L'objectif a été aussi de corréler toutes ces modifications observées avec les propriétés de transport des ions qui ont été obtenues par des tests de diffusion et migration des ions chlorures (et/ou sulfates) ainsi que des mesures de résistivité électrique. Les résultats expérimentaux ont montré que le coefficient de diffusion apparent est plus grand dans les matériaux carbonatés. Certaines expériences de diffusion faites en présence à la fois d'ions chlorures et sulfates ont aussi montré la concurrence de ces deux ions pour se fixer à la matrice cimentaire. Les matériaux cimentaires testés (bétons et pâtes) ont été formulés sans ou avec additions minérales (métakaolin, cendres volantes, laitier) et les propriétés de ces matériaux ont été observées à différents âges

La corrosion des armatures est l'une des principales causes de dégradation du béton. Les deux mécanismes les plus courants contribuant au développement de la corrosion sont la pénétration des chlorures et la carbonatation du béton. Ceux-ci génèrent deux types de corrosion respectivement par piqûres et généralisée. Une des méthodes non destructives actuellement utilisée dans le Génie Civil, l'émission acoustique, tend à se développer pour évaluer la corrosion des câbles et des structures métalliques. L'objectif de cette thèse est, d'une part de définir un essai de corrosion accélérée réaliste et maîtrisé et d'autre part de vérifier la sensibilité de l'émission acoustique à la corrosion des armatures dans le béton. Des méthodes électrochimiques éprouvées sont utilisées pour valider les résultats obtenus par émission acoustique. Ces méthodes, comme la mesure du potentiel de corrosion et l'évaluation de la vitesse de corrosion faisant l'objet de recommandations RILEM, demeurent néanmoins des méthodes intrusives nécessitant l'accès à l'armature. A partir des essais rencontrés dans la littérature, différents essais ont été étudiés utilisant d'abord de forts courants de corrosion puis en les diminuant pour atteindre des courants proches de ceux existants dans les structures. Différentes conditions d'initiation de la corrosion comme l'attaque par les chlorures et la carbonatation du béton d'enrobage ont été également testées. Cela nous a permis de mettre au point un essai de corrosion accélérée avec des densités de courant de corrosion proches de la réalité. Cet essai consiste à créer une macro-pile en connectant deux armatures coulées dans un même béton, une des armatures étant dépassivée par carbonatation du béton, la seconde étant maintenue à l'état passif. .
Corrosion of reinforcement is a major cause of concrete deterioration. The two most common mechanisms contributing to the development of corrosion are penetration of chlorides and carbonation of cover concrete. These two mechanisms induce two types of corrosion, pitting and generalised corrosion. One of the non destructive techniques used in Civil Engineering, acoustic emission, extends to monitor the corrosion of prestressing cables and metal structures. The objective of this thesis is firstly to define an accelerated corrosion test able to provide conditions close to natural corrosion and secondly to check the sensitivity of acoustic emission for detecting the effect of early corrosion in reinforced concrete. Usual electrochemical methods are used to validate the results obtained by acoustic emission. These methods, potential of corrosion measure and corrosion rate measurement being RILEM recommendations, remain intrusive methods requiring access to the rebar. From a literature review, various tests were studied, first using high impressed corrosion currents and then by decreasing gradually the current to values existing in structures. Different conditions for initiating corrosion, as an attack by chlorides and carbonation of cover concrete, were also tested. This enables the development of an accelerated corrosion test with densities of corrosion current close to the natural ones. It's about creating a macro-cell by connecting both two rebars in the same concrete, one of the rebar being depassivated by carbonation of concrete, the second remaining in a passive state. Different galvanic currents are obtained with different anodic and cathodic surfaces and distances between electrodes. .

Les structures en béton armé exposées à des environnements agressifs subissent des dégradations qui affectent leur intégrité. La corrosion des armatures est l’un des mécanismes de dégradation les plus répandus et les coûteux en terme de maintenance et de réparation. Ce processus est dû à la pénétration des agents agressifs dans le béton, notamment les ions chlorures et le gaz carbonique. Les chlorures induisent une corrosion localisée ou par piqûre, alors que le gaz carbonique engendre une corrosion généralisée ou uniforme. Le déclenchement et la propagation de la corrosion dépendent de plusieurs facteurs liés aux matériaux, aux chargements, à la géométrie et à l’environnement. Ces facteurs présentent de grandes incertitudes qui doivent être prise en comptes à travers une approche probabiliste. Dans ce travail de recherche, nous nous intéressons au mécanisme de corrosion en général. Un intérêt particulier est porté à la prise en compte de l’impact des facteurs climatiques sur ce processus, notamment dans le contexte libanais. Ainsi, nous proposons une modélisation physique de la corrosion des aciers dans les poutres en béton armé qui se déroule en deux phases : - une phase d’initiation durant laquelle les agents agressifs (chlorures et gaz carbonique) pénètrent dans le béton et atteignent des concentrations critiques provoquant la dépassivation de l’acier ; - une phase de propagation durant laquelle il y a corrosion active des aciers et diminution de la résistance de la poutre jusqu’à la défaillance. Les facteurs présentant des incertitudes sont traités comme des variables aléatoires. Pour les modéliser, nous avons étudié, pour les différentes variables aléatoires, de nombreux modèles probabilistes proposés dans la littérature. Nous avons vérifié leur compatibilité vis-à-vis de notre problématique et la possibilité d’assurer les données nécessaires à leur bonne utilisation (notamment la cohérence entre les hypothèses). Ensuite, nous avons retenu les modèles probabilistes les plus adaptés à notre cas. Par ailleurs, l’application des principes fiabilistes nous permet d’évaluer la fiabilité des poutres sujettes à la corrosion vis-à-vis des deux états-limites (ELU et ELS). En effet, la perte de la section d’acier due à la corrosion induit d’une part, une diminution de la capacité portante de la poutre, et d’autre part une augmentation de la contrainte au niveau du béton tendu (provoquant un accroissement des ouvertures des fissures). Ainsi, pour l’état limite de service, la marge de sûreté s’annule lorsque l’ouverture des fissures dépasse la valeur limite préconisée par l’Eurocode 2. Quant à l’état limite ultime, la fonction d’état limite est la résistance en flexion : la défaillance a lieu lorsque le moment résistant équivaut au moment sollicitant. Le calcul fiabiliste est effectué au moyen de simulations de Monte-Carlo. Finalement, nous avons réalisé plusieurs applications aux modèles de corrosions proposées dans ce travail. La première application porte sur l’analyse des sensibilités des modèles de corrosion aux différents paramètres. L’effet des moyennes des paramètres aléatoires ainsi que leurs variabilités sur la réponse du modèle est examiné. Une attention particulière est accordée à l’impact des facteurs climatiques. Ainsi une application du modèle de corrosion induite par les chlorures avec des données réelles de température et d’humidité relatives à trois villes côtières ayant des caractéristiques climatiques différentes est présentée. Ensuite une étude comparative de l’effet du choix des diamètres des armatures et des épaisseurs des enrobages sur la fiabilité à l’état limite ultime et à l’état limite de service est effectuée. Les résultats obtenus ont permis de mettre en évidence l’aspect agressif des facteurs climatiques : un climat chaud et humide est très agressif vis-à-vis de la corrosion induite par les chlorures alors qu’un climat à humidité relative variable favorise la corrosion par carbonatation. (...)
When exposed to aggressive environment, reinforced concrete structures are subject to a degradation mechanism that affects their integrity. Among various environmental attacks, the corrosion of RC structures is considered the most dangerous. The process is launched by the penetration of aggressive agents, precisely the chlorides and carbon dioxide into the concrete. The chlorides induce a localized corrosion, also called pitting corrosion, while on the other hand the carbon dioxide leads to a general corrosion called uniform corrosion. This corrosion phenomenon depends on several factors such as the materials characteristics,loadings, geometry and the environment. All these components include different levels of uncertainties that are taken into account throughout a probabilistic approach. In this work, we propose two models for the corrosion mechanisms induced separately by the chlorides and the carbon dioxide. These models take into account the effect of the climatic condition that is mainly described by the temperature and the relative humidity. In addition to that, as a study case we have treated in details the Lebanese climatic context. We have proposed a physical model of steel corrosion in reinforced concrete beams that occurs in two phases : - An initiation phase where aggressive agents like the chlorides and carbon dioxide penetrate into the concrete and reach a critical concentration values causing the depassivation of the steel ; - A propagation phase in which the active corrosion of steel decreases the strength of the beam leading to its failure. All the factors that have uncertainties are treated as random variables. Several probabilistic models are listed and discussed in the literature while only the models that match with our context are selected. The reliability analysis allowed us to assess the reliability of beams subjected to corrosion in ULS and SLS. The loss of steel section due to the corrosion mechanism induces a decrease of the bearing beam capacity, and an increase in the tension stress in the concrete.This causes an increase of the width of cracks openings. Thus, taking into account the serviceability limit state, the safety margin goes to zero when the width of crack opening exceeds the acceptable width as recommended by the Eurocode 2. The limit state function in ULS is the bending strength. The failure occurs when the applied moment equals or surpasses the resisting moment. The reliability calculations are carried out using Monte-Carlo simulations. Finally, several applications to the corrosion model are proposed via this work. The first application concerns the sensitivity analysis of the corrosion models for the different parameters. The effects of the mean values and the variability of the random variables on the model response are also examined. The impact of climatic factors on the corrosion phenomenon took the biggest part of this work. We have applied the chloride’s corrosion model with the real temperatures and relative humidity of three coastal cities having different climatic characteristics. Then a comparative study showing the effect of the ba rdiameters and the cover thickness on the reliability of the RC beam subjected to aggressive environment is carried out. (...)

Les ouvrages en béton armé ont aujourd'hui en moyenne un âge qui approche de la période critique d'amorçage de la corrosion des aciers de renfort (de 30 à 50 ans), et leur remise à niveau représente un enjeu économique considérable. Les études dans le cadre de cette thèse ont pour but d'apporter des éléments de réponse à la question préoccupant les ingénieurs, à savoir l'influence de la corrosion des armatures sur la résistance mécanique des structures et les marges résiduelles par rapport à la ruine. Ces informations sont nécessaires pour obtenir l'évolution de la perte de capacité des ouvrages soumis à la corrosion. L'étude de cette évolution est calée sur les principales phases qui caractérisent la dégradation d'une structure. Ces informations sont néccessaires pour mettre en place une politique de gestion du patrimoine et décider en connaissance de cause d'une planification des travaux de réparation, du renforcement ou du remplacement des éléments d'ouvrage défaillants. Le mémoire comporte six chapitres. Le premier chapitre propose un modèle d'évolution dans le temps des performances d'un ouvrage selon trois étapes : étape d'initiation ti, étape de déterioration stable tsp et étape de déterioration imprévisible tup. Le chapitre 2 précise la première étape pour des éléments en béton armé et béton précontraint. Le chapitre 3 analyse les mécanismes de corrosion et leurs conséquences sur les structures en béton. Le chapitre 4 porte sur la prédiction de la corrélation entre l'adhérence et la corrosion. Le chapitre 5 développe une méthode de calcul de l'évalauation du comportement structurel des éléments. Cette méthode permet d'intégrer les conséquences de la corrosion des armatures dans le comportement de la structure. Le chapitre 6 propose une validation du modèle par comparaison avec des essais expérimentaux, notamment ceux du projet "Benchmark des poutres de la Rance"

Cette thèse s’intéresse à l’étude la corrosion des armatures du béton armé à la fois sur les conditions de son initiation et de sa propagation mais également sur ses conséquences à long terme sur le fonctionnement en service des ouvrages.Le démarrage d’une étude à long terme sur les conditions de l’initiation et de la propagation de la corrosion en ambiance saline est présenté et les premiers résultats confirment l’initiation précoce de la corrosion mais montrent que l’influence des fissures sur la propagation dépend des conditions de leur exposition et de l’enrobage.Une étude a été réalisée sur la carbonatation en présences de fissures et sur ses conséquences sur l’initiation et la propagation de la corrosion. Des échantillons annulaires de mortier renforcé par une armature HA ont été fissurés. Les résultats ont montrés que quelque soit l’ouverture des fissures mécaniques, il y a une carbonatation des lèvres de la fissure et de l’interface acier-béton. Après exposition à des cycles d’humidification séchage, la corrosion se développe tout le long de l’armature mais avec une épaisseur plus importante en sous-face de l’armature ou la qualité de l’interface est moins bonne. Les résultats montrent que les fissures de corrosion induites par le développement de la couche de rouille prennent naissance à partir des micro-fissures dues à l’endommagement mécanique. Ce résultat est cohérent avec le développement des fissures de corrosion observées dans la partie sur l’initiation et la propagation de la corrosion en ambiance saline.Une autre partie de la thèse est consacrée à l’étude des propriétés mécaniques résiduelles de poutres en béton armé corrodées par 26 à 28 années de conservation majoritairement sous charge de flexion en ambiance saline.On s’intéresse d’abord au comportement en traction simple des aciers corrodés. La corrosion induite par les chlorures conduit à des piqûres de géométries très diverses qui rendent difficile l’évaluation de la section résiduelle d’acier. Cependant, il apparait que la contrainte effective de plasticité est peu affectée, la contrainte ultime est légèrement augmentée significative d’un écrouissage plus important après corrosion et surtout l’élongation à rupture est drastiquement réduite par la présence des piqures de corrosion.L’effet de la corrosion des armatures sur le comportement en flexion est ensuite étudié. La corrosion se traduit par un changement du mode de rupture, de l’écrasement du béton comprimé après une phase de plastification des aciers à la rupture fragile et brutale des armatures tendues après corrosion. La diminution de la capacité portante est proportionnelle à la perte de section d’acier dans les sections les plus sollicitées. La perte de ductilité ou diminution de la flèche à rupture des poutres en béton armé est très importante et pourrait remettre ne cause la sécurité dans le fonctionnement des éléments corrodés.L’effet de la corrosion sur la résistance à l’effort tranchant est ensuite étudié. Pour ce faire des éléments courts (poutres épaisses) sont testés en flexion simple. Les résultats montrent que la corrosion peut conduire à des changements de fonctionnement mécanique mais que la résistance vis-à-vis de l’effort tranchant est globalement peu affectée par la corrosion des armatures longitudinales ou des cadres d’effort tranchant. La complexité des phénomènes mis en jeu sur des éléments courts et l’importance des bielles de béton comprimées dans les effets d’arc peuvent expliquer ces résultats. Par ailleurs, la capacité d’ancrage des armatures droites corrodées apparait très largement supérieure à celle attendue compte tenu de la présence des fissures de corrosion et des prévisions issues de la littérature. Le confinement dû à l’appui et le fait que la corrosion naturelle n’affecte pas tout le périmètre des barres d’armatures peuvent expliquer ces résultats surprenants
This thesis aims to study the reinforcement corrosion embedded in reinforced concrete structures on both initiation and propagation of corrosion and its long-term consequences on the service life of structures. Firstly, a long-term study on the conditions for the initiation and propagation of corrosion in saline environment is presented and the first results confirm the early initiation of corrosion, but shows that the influence of pre-cracks on the propagation of corrosion depends on their conditions of exposure and concrete cover thickness. A study was conducted in carbon dioxide condition with presence of cracks to examine its impact on the initiation and propagation of corrosion. The results showed that whatever the mechanical crack opening, the crack edge and the steel-concrete interface were carbonated. Carbonation of the steel-concrete interface is likely to be due to mechanical damage induced by loading and stress transfer from the steel to mortar when creating cracks. This damage is also caused by the presence of internal micro-cracks around the steel bar. Following exposure to wetting-drying cycles, corrosion develops throughout the reinforcement but with a greater thickness of rust layer on the underside of the reinforcement where the quality of the interface is weaker. The results showed that the corrosion cracking induced by the development of rust layer arises from internal micro-cracks due to mechanical damage. This result is consistent with the development of corrosion cracks observed previously in saline environment. Another part of the thesis is to study the residual mechanical properties of corroded reinforced concrete beams of 26-28 years exposed to a saline environment under bending load. The behavior of single tensile steel corroded extracted from these corroded beams is studied. It is very difficult to know the true stress of tensile steels corroded "naturally" in concrete contaminated by exposure to chlorides. Indeed, the corrosion induced by chlorides leads to create corrosion pits with various geometries that make it difficult to determine accurately the residual steel cross-section. However, it appears that the true yield stress is unaffected, the true ultimate stress is slightly increased by corrosion but the total elongation at failure is drastically reduced by the presence of pitting corrosion. The effect of reinforcement corrosion on flexural behavior is then studied. The corrosion resulted in a change in failure mode, from concrete crushing in compression after yielding of tensile steel to brittle failure of tension reinforcement. The decrease of the load bearing capacity is proportional to the loss of steel section in the mid-span section. The loss of ductility or decrease in deflection at failure of reinforced concrete beams is very important and could be the limiting factor for the service life of corroded RC structures. It appears that the change in ductility of corroded reinforced concrete beam is correlated with the change in ductility of the steel due to corrosion. The effect of corrosion on the shear behavior is then studied. To make this, the short-shear span beams (deep beams) were tested under three point bending until failure. The results showed that corrosion can lead to changes in mechanical behavior but load bearing capacity of deep beams is generally unaffected by corrosion of longitudinal reinforcement and stirrups. These results can be explained by a coupled behavior between arch action and beam action leading to change in load transfer mechanism and failure mode. In addition, the capacity of straight end anchorage of corroded reinforcement appears to be very much higher than expected despite the presence of corrosion cracks. Concrete confinement effect due to the end support reaction and the “natural” corrosion condition which do not lead to a homogeneous damage all around perimeter of re-bars may explain these surprising results

The diploma thesis is focused on a solution of the load reaction of the bridge construction from girders KA-73. Proposal methods of the nonlinear analysis by means of final elements on the stochastic and deterministic level are used for the solution of the load reaction. A simulation technique Latin Hypercube Sampling is used within the stochastic analysis. A material degradation in the form of the trussing corrosion is solved with the expected decrease of the construction lifetime. The conclusion of the thesis contains an evaluation of initial quantities of material parameters for the load reaction of the construction in the form of the sensitivity analysis.

碩士
國立成功大學
土木工程學系碩博士班
90
The variations of microstructure in mortar specimens under coupled deterioration of carbonation and chloride are investigated. Accelerated carbonation tests of four types mortar specimens containing chloride were conducted. The concentration of chloride in mixing water are 0﹪, 1﹪, 3.5﹪and sea sand, respectively. After the accelerated deterioration test, the specimens were quantitatively analyzed using thermogravimetric analysis (TGA) and mercury instrusion porosimetry (MIP), and the content of free chloride and total chloride was also measured. The test results show that the free and total chloride were significantly increased due to carbonation condensation effect of bound chloride. The pore structure varied during the process of carbonation. The porosity decreased for the pore diameters ranging from 0.01 to 0.1μm, while the porosity for the pore diameter less than 0.01μm had a trend of increasing. With the carbonation time increased, however, the carbonation depth did not varied significantly with the various contents of chloride concentration.

碩士
國立臺灣科技大學
營建工程系
101
This study explores the composite deteriorations between the alkali-silica reaction (ASR) and the carbonation, or the chloride attack. Strategies to mitigate the ASR expansion were proposed. Mortar specimens were prepared using Pyrex glass and limestone powders in partial placement of the Portland cement at two water-cement ratios of 0.47 and 0.6. Results showed that the specimens prepared in accordance to ASTM C1260 had ASR gel with morphologies changing at different stages of the acceleration. The calcium content of the gel was increased and the compressive strength of the specimen was decreased by the increase of the acceleration time. In addition, the limestone addition mitigated the ASR expansion, the most effectively at dosage of 10%. Carbonation prior to the ASR also mitigated the ASR expansion because of the reduced alkalis in the specimens. When the carbonation was applied after the initiation of ASR, the effective carbonation time was increased. Results of the cycling ASR and carbonation showed that the limestone addition had additional mitigating effects. Further results showed that the carbonation accelerated the calcite formation and modified the morphologies of the ASR gel. Even if the specimens were well cured, they showed similar ASR behaviors. Some additional experiments showed that the early ASR did not accelerate the salt intrusion.

碩士
國立臺灣海洋大學
河海工程學系
95
Generally speaking, structure needs long-term use, aside from elaborating about accuracy, quality, and suitability in the process of design, construction, and implementation management, repair and strengthening are very important afterwards. Owing to the Taiwan is an island-type climate, structures are severely subjected to chloride ingression. The service life of structure is reduced by the influence of carbonation of concrete. The main propose of this article was to study the service lives of existing reinforced concrete（RC）bridges separately due to chloride-ion and carbonation. When considering the service life of existing RC bridge induced by the chloride-ion penetration, the service life（t） is divided into three stages：initiation period （tc）, depassivation period （tp）and propagation period （tcorr）, i.e. t= tc + tp + tcorr. Some mathematical models are used to predict the values of tc, tp, and tcorr. Both the Tzyh-chyang and Dah-duh bridges were used as illustrative examples for verifying the mathematical models. The corresponding service lives prediction are 58.82 and 56.01 years, respectively. As to the influence of carbonation to existing RC bridge, the random model associated with reliability analysis is employed to predict carbonation life. Meanwhile, the field test data of the Chorng-ching viaduct, Wann-fwu bridge, and Shi-yuan viaduct are used to predict their carbonation lives. When the reliability index is zero, their carbonation life-time are 55, 55, and 60 years, respectively. The present study results can be extended and applied to predict the service lives, repair, and strengthening of existing RC bridges.

The present work had as objective the development of electrochemical sensors to monitor the ingress of chlorides. and the pH variation of the pore solution of cementitious materials due to the carbonation process and attack with sulfuric acid. For the detection of chlorides, two materials were tested: Ag/AgCl and Lucigenin. The Lucigenin film grown potentiodynamically on stainless steel did not present an acceptable potentiometric response. The synthesis of the Ag/AgCl electrodes was performed using cyclic voltammetry and potentiostatic and galvanostatic regimes. The electrochemically modified electrodes showed a potentiometric response for the detection of chlorides with a sub-Nernstian slope which mainly depends on the concentration of electrolyte used in film growth. Very alkaline environments lead to early degradation of Ag/AgCl electrodes. For pH determination, several metal oxides were tested — MnO2, IrOx and IrOx + RuO2 — and an electronically conducting polymer — polypyrrole. The electrochemically grown MnO2 films showed a linear response in the pH range between 13 and 8 with a slope close to Nernstian. However, these films dissolve in acidic environments (pH<6). The ellipsometry study identified the formation of a more resistive inner layer composed of MnOOH. The IrOx synthesis was carried out potentiodynamically and through a potentiostatic regime of single and double impulse. The IrOx electrodes grown by cyclic voltammetry showed the response with the greatest potential for application in environments as aggressive as cementitious materials, exhibiting a linear potentiometric response in the pH region between 13 and 5, with a super- Nernstian slope. The addition of 5 and 10 mM RuCl3 to the growth solution led to a decrease in the slope of the potentiometric response of the modified electrodes towards values closer to Nernstian. The potentiostatically modified electrodes showed a linear potentiometric response in the pH region between 10 and 2, with a slope closer to Nernstian than the films grown by voltammetry. Finally, the potentiostatically synthesized polypyrrole films exhibited a linear response in the pH region between 11 and 5. The ellipsometric study of steel passivation in 0.1 M NaOH showed the formation of an inner layer of Fe(OH)2 followed by the growth of two subsequent more porous layers composed of iron hydroxides. Ag/AgCl sensors have been successfully used to study chloride penetration in mortar samples with and without an additive with the potential to remove chlorides from solution (LDH). The presence of LDH in the mortar led to a decrease in the chloride content in the pore solution. However, the destructive analysis carried out showed the presence of a greater amount of total chlorides in the sample with LDH, which shows an increase in the amount of chlorides immobilized in the cement matrix. The removal of chlorides from the pore solution can be a consequence of ion exchange with the LDH or due to the formation of Friedel salts in the presence of aluminum ions from the dissolution of the LDH. The results with the iron electrodes corroborated the results obtained with the sensors about the effect of LDH on chloride capture. IrOx sensors have been successfully applied to study the pH variation of the pore solution of cementitious samples due to carbonation and acid etching processes. The carbonation study showed an immediate change in the pH of the pore solution near the surface after the initial release of carbon dioxide, but gradually recalinized after the release of the gas. However, the pure carbon dioxide regime promoted the obstruction of the superficial porous network, which led to the blocking of carbon dioxide entry to greater depths. Finally, IrOx sensors showed that the pH variation of the pore solution due to exposure to concentrated sulfuric acid is limited to a layer only 1 to 2 mm thick. The pH variation in depth happens as the attack front progresses.
O presente trabalho teve como objetivo o desenvolvimento de sensores electroquímicos para a monitorização do ingresso de cloretos e da variação de pH da solução de poro de materiais cimentícios devido ao processo de carbonatação e ao ataque com ácido sulfúrico. Para a deteção de cloretos foram testados dois materiais: Ag/AgCl e Lucigenina. O filme de Lucigenina crescido potenciodinamicamente sobre aço inox não apresentou uma resposta potenciométrica aceitável. A síntese dos eléctrodos de Ag/AgCl foi realizada através de voltametria cíclica e de regimes potenciostáticos e galvanostáticos. Os eléctrodos modificados electroquimicamente apresentaram uma resposta potenciométrica para a detecção de cloretos com um declive sub-Nernstiano que depende principalmente da concentração de electrólito utilizada no crescimento dos filmes. Ambientes muito alcalinos levam à degradação precoce dos eléctrodos de Ag/AgCl. Para a determinação de pH foram testados vários óxidos metálicos — MnO2, IrOx e IrOx + RuO2 — e um polímero electronicamente conductor — polipirrole. Os filmes de MnO2 crescidos electroquimicamente apresentaram uma resposta linear no intervalo de pH entre 13 e 8 com um declive próximo do Nernstiano. Contudo, estes filmes dissolvem em ambientes ácidos (pH<6). O estudo de elipsometria identificou a formação da uma camada interna mais resistiva composta por MnOOH. A síntese de IrOx foi efectuada potenciodinamicamente e através de um regime potenciostático de singular e de duplo impulso. Os eléctrodos de IrOx crescidos por voltametria cíclica apresentaram a resposta com maior potencial para a aplicação em materiais cimentícios, exibindo uma resposta potenciométrica linear na região de pH entre 13 e 5, com um declive super-Nernstiano. A adição de 5 e 10 mM de RuCl3 à solução de crescimento levou à diminuição do declive da resposta potenciométrica dos eléctrodos modificados para valores mais próximos do Nernstiano. Os eléctrodos modificados potenciostaticamente apresentaram uma resposta potenciométrica linear na região de pH entre 10 e 2, com um declive mais próximo do Nernstiano que os filmes crescidos por voltametria. Por último, os filmes de polipirrole sintetizados potenciostaticamente exibiram uma resposta linear na região de pH entre 11 e 5. O estudo elipsométrico da passivação do aço em 0.1 M NaOH mostrou a formação de uma camada interna de Fe(OH)2 seguida do crescimento de duas camadas mais porosas subsequentes compostas por hidróxidos de ferro. Os sensores de Ag/AgCl foram utilizados com sucesso no estudo da penetração de cloretos em amostras de argamassa com e sem um aditivo com potencial capacidade de remover cloretos de solução (LDH). A presença de LDH na argamassa levou à diminuição do teor de cloretos na solução de poro. Contudo, a análise destrutiva levada a cabo mostrou a presença de uma maior quantidade de cloretos totais na amostra com LDH, o que evidencia um aumento da quantidade de cloretos imobilizados na matriz cimentícia. A remoção de cloretos da solução de poro pode ser consequência da troca iónica com o LDH ou devido à formação de sais de Friedel na presença de iões alumínio provenientes da dissolução do LDH. Os resultados com os eléctrodos de ferro corroboraram os resultados obtidos com os sensores do efeito do LDH na captura de cloretos. Os sensores de IrOx foram aplicados com sucesso no estudo da variação de pH da solução de poro de amostras cimentícias devido aos processos de carbonatação e de ataque ácido. O estudo da carbonatação mostrou uma alteração imediata do pH da solução de poro perto da superfície depois da libertação inicial de dióxido de carbono, mas gradualmente realcalinizada após a libertação do gás. Contudo, o regime de CO2 puro promoveu a obstrução da rede porosa superficial, o que levou ao bloqueio da entrada de dióxido de carbono até maiores profundidades. Por fim, os sensores de IrOx mostraram que a variação de pH da solução de poro devido à exposição a ácido sulfúrico concentrado é limitada a uma camada de apenas 1 a 2 mm de espessura. A variação de pH em profundidade acontece à medida que a frente de ataque progride.
Programa Doutoral em Ciência e Engenharia de Materiais

Dissertação de mestrado integrado em Engenharia Civil
A cinza volante (CV) é o material pozolânico mais produzido em todo mundo. O aumento da quantidade de aluminatos, proporcionada pela substituição de cimento por CV, é um dos fatores que conduz a uma diminuição da quantidade de cloretos livres. Por outro lado, devido às reações pozolânicas, inerentes à utilização de CV, o Ca(OH)2 presente na matriz é consumido, facilitando o avanço da frente de carbonatação. Muito pouco é conhecido acerca da ação combinada da carbonatação e da penetração de cloretos em materiais cimentícios contendo CV. Assim, este trabalho investiga a influência da adição de CV no coeficiente de difusão de cloretos em argamassas carbonatadas. Para isso, foram moldados provetes em argamassa com 0%, 20%, 40% e 60% de CV, utilizando cimento CEM I 42.5R e uma relação água/ligante 0.54, 0.51, 0.5 e 0.69 respetivamente. Após um período de cura de 90 dias e 15 dias de pré-condicionamento, os provetes foram introduzidos na câmara de carbonatação (20ºC, 50% de HR e 5% de CO2) onde permaneceram por um período de 15, 60 e 90 dias. Terminado o período de carbonatação, os provetes foram submetidos de imediato, ao ensaio de difusão por migração de cloretos com base na especificação LNEC E 463 (2004). Os resultados alcançados sugerem que as argamassas carbonatadas apresentam um coeficiente de difusão de cloretos diferente do alcançado para as argamassas que não sofreram carbonatação, independentemente da percentagem de CV. Esta diferença deve ser tida em consideração, principalmente, em ambientes sujeitos à ação dos cloretos e, ao mesmo tempo, fortemente contaminados com CO2.
Fly ash is the most produced pozzolanic material in the world. The increase in the amount of aluminates provided by the substitution of cement by fly ashes is one of the factors that leads to a decrease in the amount of free chloride. On the other hand, due to the pozzolanic reactions inherent to the use of fly ashes, the Ca(OH)2 present in the cementitious matrix is consumed, facilitating the advance of the carbonation front. Very little is known about the combined action of carbonation and chloride penetration in cementitious materials containing fly ashes. This work investigates the influence of the addition of fly ashes on the diffusion coefficient of chlorides in carbonated mortars. For this, mortar specimens with 0%, 20%, 40% and 60% CV were molded using CEM I 42.5R cement with a water/binder ratio of 0.54, 0.51, 0.5 and 0.69 respectively. After a curing period of 90 days and 15 days of preconditioning, the specimens were introduced into the carbonation chamber (20ºC, 50% RH and 5% CO2) where they remained for a period of 15, 60 and 90 days. After the carbonation period, the test specimens were immediately submitted to the chloride migration diffusion test based on the specification LNEC E 463 (2004). The results suggest that carbonated mortars have a diffusion coefficient of chloride different from that achieved from mortars that didn´t undergo carbonation, regardless of the percentage of fly ashes. This difference must be taken into account mainly in environments subject to the action of chlorides and at the same time heavily contaminated with CO2.

Dissertação de mestrado Internacional em Sustentabilidade do Ambiente Construído
Garantir a durabilidade das estruturas de betão é, por si só, uma das formas mais importantes para assegurar a sua sustentabilidade. Neste sentido, optou-se por investigar a ação combinada dos principais agentes causadores da corrosão, a carbonatação e os iões cloreto. Individualmente, estes mecanismos de deterioração têm sido abordados de forma constante em diversas investigações. Ao contrário, estudos do efeito combinado no betão ainda são pouco expressivos em quantidade e não reúnem consenso. Isoladamente, a carbonatação modifica a microestrutura e altera o pH do betão. Quando combinada com os iões cloro, pode ter influência direta na difusão dos mesmos. A alteração na microestrutura do material poderá provocar uma diminuição na quantidade de cloretos penetrados. Por outro lado, a redução da alcalinidade, pode conduzir a uma libertação dos cloretos quimicamente fixados, aumentando a quantidade de cloretos transportados para o interior do betão. Desta forma, e avaliando também o papel das cinzas volantes (CV) neste contexto, estudou-se a influência da profundidade de carbonatação na penetração de cloretos, em diferentes composições de betão, com razão água/ligante de 0,55 e 0,5, para duas metodologias de ensaio distintas: determinação do coeficiente de difusão de cloretos por intermédio do ensaio de migração em regime não estacionário e determinação da resistência à penetração de cloretos por intermédio do ensaio de imersão. Realizou-se uma série de ensaios combinados, onde as amostras foram previamente carbonatadas e, em seguida, ensaiadas segundo as metodologias mencionadas. Avaliou-se também o efeito da ação combinada noutros indicadores de durabilidade das amostras testadas. Para isso foram utilizados os ensaios de absorção de água por capilaridade e por imersão. Os resultados obtidos permitiram concluir que a carbonatação influencia, de facto, o coeficiente de difusão de cloretos, sendo esta situação verificada com maior notoriedade na composição com incorporação de CV.
Guaranteeing the durability of the concrete structures is, by itself, one of the most important ways to ensure their sustainability. Therefore, the object of this research is the combined action of the main corrosive agents: carbonation and chloride ions. Individually, these deterioration mechanisms have already been object of research. However, studies of their combined effect on concrete are sparse and non-consensual. Carbonation alone modifies the microstructure of concrete, as well as its pH. When combined with chloride ions, it may have a direct influence on their diffusion. The change at the microstructural level can cause a decrease in the quantity of penetrated chlorides. On the other hand, alkalinity reduction might lead to a release of the chemically fixed chlorides, increasing the quantity of chlorides carried inside the concrete. In this way, while also assessing the role played by fly ashes (FA) in this context, the influence of the carbonation depth on the chlorides penetration has been studied using different concrete compositions, with water/binder ratios of 0,55 and 0,5 and two testing methods: evaluation of chloride diffusion coefficient by migration in non-stationary regime and evaluation of resistance to chloride penetration by immersion. Several combined tests have been conducted where the samples had been previously carbonated and then tested using the aforementioned methods. The effect of the combined action on the durability of the tested samples has been evaluated as well through water absorption tests, performed both by capillarity and immersion. The conclusion obtained from the obtained results is that carbonation has a real influence on the chloride diffusion coefficient and this can be more clearly verified in the composition that includes FA.

Research Doctorate - Doctor of Philosophy (PhD)
The Sorell Causeway Bridge, located in Tasmania, Australia, was completed in 1957 and was the first precast, post-tensioned bridge constructed in Australia. However after only 45 years of service, the bridge was replaced due to increasing concerns surrounding the level of corrosion of the prestressing strands in the beams. Prior to its decommission, an extensive and costly investigation program was carried out on the bridge in an attempt to determine the rate of deterioration and establish the remaining margin of safety. Despite the number of investigations and the resulting large quantities of information, the questions surrounding the safety of the bridge remained unanswered. The issue is thus raised: what do field investigations of reinforced or prestressed concrete structures with evidence of corrosion deterioration tell engineers about the actual condition of the structure? Three beams of varying condition (good, average, poor) were salvaged from the bridge demolition for further detailed examination to investigate the degree of correlation between pre-demolition field investigations and the physical condition of the steel post-demolition. The investigations included the use of conventional non-destructive techniques such as cover, half-cell potential and concrete resistivity surveys, and destructive techniques such as chloride profiling, carbonation depth measurement, and full-scale load testing, all of which were used to determine the likely risk of corrosion and likely corrosion rate for each beam. The results of these investigations were subsequently reviewed in relation to the physical condition of the steel. In general, all non-destructive tests were found to be inconclusive in relation to evidence of steel corrosion and the corrosion risk guidelines recommended in the literature. It was also apparent that these techniques were incapable of detecting steel pitting, a primary concern for the current investigation. Chloride profiles were variable and inconsistent in relation to steel corrosion and the chloride thresholds recommended in the literature. Carbonation was found to exist at prestressing levels in some locations and appeared to be influenced by the orientation and geometry of the beams. All beams did not achieve the estimated design capacity and corrosion had significantly impaired the ultimate capacity and ductility of beams in the worst condition. Aerobic and anaerobic corrosion products were identified via XRD analysis. These included Magnetite, Goethite, Akaganeite, Lepidocrocite, chloride based Green Rust (I), and Iron (III) Oxide Chloride. The phenomenon of “chloride weeping”, or droplets of highly acidic ferrous chloride, was observed forming on some steel/concrete interfaces on freshly cut concrete surfaces. Several other unexplainable observations were made during the course of the present investigations. These included bright, metallic pit surfaces; pits with concentric rings; black, wet rust covering bright, metallic surfaces; and unusual pitting profiles. A possible explanation for these observations may be the implication of microbiological activity in the corrosion process. Further research is required to confirm these observations.