Fiber reinforced polymers (FRPs) have gained popularity in upgrade projects for reinforced concrete structural elements within the last decade because of its ease of application and high strength-to-weight ratio. In the design of an effective retrofitting solution by means of an FRP system, the anchorage capacity has an important role. This study presents the results of an experimental program conducted to determine the strength of carbon fiber reinforced polymers (CFRPs) bonded to concrete prisms or hollow clay tiles that are finished with or without plaster. In the experimental program, different types of anchorage methods were tested in a double shear push-out test setup. A simple and effective strength model is proposed for strip type anchorages based on the existing analytical models and experimental observations from this study. This new model is suitable for determining the design capacity of CFRP-to-concrete and CFRP-to- hollow clay tile joints with or without plaster and accounts for the presence of embedment and concrete strength. Obtained results by using this model were found to closely match with the experimental observations.

El uso de barras de materiales compuestos (FRP) se propone como una alternativa efectiva para las tradicionales estructuras de hormigón armadas con acero que sufren corrosión en ambientes agresivos. La aceptación de estos materiales en el mundo de la construcción está condicionada a la compresión de su comportamiento estructural. Este trabajo estudia el comportamiento adherente entre barras de FRP y hormigón mediante dos programas experimentales. El primero incluye la caracterización de la adherencia entre barras de FRP y hormigón mediante ensayos de pull-out y el segundo estudia el proceso de fisuración de tirantes de hormigón reforzados con barras de GFRP mediante ensayo a tracción directa. El trabajo se concluye con el desarrollo de un modelo numérico para la simulación del comportamiento de elementos de hormigón reforzado bajo cargas de tracción. La flexibilidad del modelo lo convierte en una herramienta flexible para la realización de un estudio paramétrico sobre las variables que influyen en el proceso de fisuración. / The use of Fibre Reinforced Polymers (FRP) as reinforcement in concrete structures is considered to be a possible alternative to steel in those situations where corrosion is present. The full acceptance of FRP reinforcement in concrete construction is contingent on a complete study and comprehension of all aspects of their structural performance. This thesis investigates the bond behaviour between Fibre Reinforced Polymer (FRP) reinforcement and concrete. Two experimental programs were conducted. In the first program the role of the variables which affect the bond behaviour was studied through pull-out test. In the second program, GFRP RC members were tested in tension to study their cracking response. To conclude the thesis, a numerical model was developed to simulate the cracking behaviour of RC tensile members. Since the model was flexible enough to include any "user-defined" bond-slip law and variable materials' properties, a parametric study was conducted to analyze which are the variables that influence the cracking behaviour.

Research Doctorate - Doctor of Philosophy (PhD) / Inserting fibre reinforced polymer (FRP) strips into pre-cut grooves in the surface of masonry walls is an emerging technique for the retrofit of unreinforced masonry (URM) structures. This method, known as near surface mounting (NSM), provides significant advantages over externally bonded FRP strips in that it has less of an effect on the aesthetics of a structure and can sustain higher loading before debonding. As this technique is relatively new, few studies into the behaviour of masonry walls strengthened using this technique have been conducted. A combined experimental and numerical program was conducted as part of this research project to study the in-plane shear behaviour of masonry wall panels strengthened with NSM carbon FRP (CFRP) strips. In this project the FRP strips were designed to resist sliding along mortar bed joints and diagonal cracking (through mortar joints and brick units). Both of these failure modes are common to masonry shear walls. Different reinforcement orientations were used, including: vertical; horizontal; and a combination of both. The first stage of the project involved characterising the bond between the FRP and the masonry using experimental pull tests (18 in total). From these tests the bond strength, the critical bond length and the local bond-slip relationship of the debonding interface was determined. The second stage of the project involved conducting diagonal tension/shear tests on masonry panels. A total of four URM wall panels and seven strengthened wall panels were tested. These tests were used to determine: the effectiveness of the reinforcement; the failure modes; the reinforcement mechanisms; and the behaviour of the bond between the masonry and the FRP in the case of a panel. The third stage of the project involved developing a finite element model to help understand the experimental results. The masonry was modelled using the micro-modelling approach, and the FRP was attached to the masonry model using the bond-slip relationships determined from the pull tests. Reinforcement schemes in which vertical FRP strips were used improved the strength and ductility of the masonry wall panels. When only horizontal strips were used to reinforce a wall panel, failure occurred along an un-strengthened bed joint and the increase in strength and ductility was negligible. The vertical reinforcement prevented URM sliding failure by restraining the opening (dilation) of the sliding cracks that developed through the mortar bed joints. The finite element model reproduced the key behaviours observed in the experiments for both the unreinforced and FRP strengthened wall panels. This model would potentially be useful for the development of design equations.

Research Doctorate - Doctor of Philosophy (PhD) / Inserting fibre reinforced polymer (FRP) strips into pre-cut grooves in the surface of masonry walls is an emerging technique for the retrofit of unreinforced masonry (URM) structures. This method, known as near surface mounting (NSM), provides significant advantages over externally bonded FRP strips in that it has less of an effect on the aesthetics of a structure and can sustain higher loading before debonding. As this technique is relatively new, few studies into the behaviour of masonry walls strengthened using this technique have been conducted. A combined experimental and numerical program was conducted as part of this research project to study the in-plane shear behaviour of masonry wall panels strengthened with NSM carbon FRP (CFRP) strips. In this project the FRP strips were designed to resist sliding along mortar bed joints and diagonal cracking (through mortar joints and brick units). Both of these failure modes are common to masonry shear walls. Different reinforcement orientations were used, including: vertical; horizontal; and a combination of both. The first stage of the project involved characterising the bond between the FRP and the masonry using experimental pull tests (18 in total). From these tests the bond strength, the critical bond length and the local bond-slip relationship of the debonding interface was determined. The second stage of the project involved conducting diagonal tension/shear tests on masonry panels. A total of four URM wall panels and seven strengthened wall panels were tested. These tests were used to determine: the effectiveness of the reinforcement; the failure modes; the reinforcement mechanisms; and the behaviour of the bond between the masonry and the FRP in the case of a panel. The third stage of the project involved developing a finite element model to help understand the experimental results. The masonry was modelled using the micro-modelling approach, and the FRP was attached to the masonry model using the bond-slip relationships determined from the pull tests. Reinforcement schemes in which vertical FRP strips were used improved the strength and ductility of the masonry wall panels. When only horizontal strips were used to reinforce a wall panel, failure occurred along an un-strengthened bed joint and the increase in strength and ductility was negligible. The vertical reinforcement prevented URM sliding failure by restraining the opening (dilation) of the sliding cracks that developed through the mortar bed joints. The finite element model reproduced the key behaviours observed in the experiments for both the unreinforced and FRP strengthened wall panels. This model would potentially be useful for the development of design equations.

Fiber reinforced polymer (FRP) composites have been of interest to the structural engineering society since the earliest days of FRP composites industry. The use of such systems has been implemented in both new construction and for repair and rehabilitation of existing structures. Since the 1980s, researchers have developed a significant body of knowledge to use FRP composites in infrastructure applications; however, most of this established knowledge was concentrated on the use of traditional epoxy (EP) systems (epoxy matrix FRPs and epoxy adhesives). FRP composites with polyurethane (PU) matrices and adhesives have recently attracted the attention of a few researchers due to their potential advantages in constructibility and mechanical properties. The deployment of these systems is currently limited by a lack of knowledge on mechanical and durability performance. The objective of this research is to quantify the mechanical behavior of PU composites utilized in externally-bonded repair of common flexural and flexural-axial reinforced concrete systems. In addition, the mechanical performance, strength, and failure modes are compared directly with an epoxy-based composite by subjecting reinforced concrete specimens utilizing each of the matrix types (EP and PU) to the same protocols. The study presented therefore allows an objective comparison (advantages and disadvantages) between the two composite system used for repair and rehabilitation of concrete infrastructure. An experimental research program was designed with different length scales. Small-scale experiments were utilized to characterize the component level properties of the materials and bond to concrete, which include the flexural behavior as well as the pure shear behavior. The results of these small scale experiments were used to calibrate analytical models of the interface behavior between FRP laminate and concrete, and paved the way for the next level of the research which studied the behavior of each composite system at larger scales. The large scale experiments included flexural retrofitting of reinforced concrete girders and retrofitting of circular columns using FRP laminates. The large-scale experimental specimens were mechanically damaged prior to FRP repair and testing, making the testing more appropriate compared to common practice of repairing undamaged specimens.

The primary objective of this research is to generate reinforcing domains in situ during the processing of polymers by using phase separation techniques. Low molecular weight compounds were mixed with polymers where the process viscosity is reduced at process temperatures and mechanical properties are improved once the material system is cooled or reacted. Thermally induced phase separation and thermotropic phase transformation of low molar mass compounds were used in isotactic polypropylene (iPP) and poly(ether ether ketone) (PEEK) resins. Reaction induced phase separation was utilized in thermosets to generate anisotropic reinforcements. A new strategy to increase fracture toughness of materials was introduced. Simultaneously, enhancement in stiffness and reduction in process viscosity were also attained. Materials with improved rheological and mechanical properties were prepared by using thermotropic phase transformations of metal soaps in polymers (calcium stearate/iPP). Morphology and thermal properties were studied using WAXS, DSC and SEM. Mechanical and rheological investigation showed significant reduction in process viscosity and substantial improvement in fracture toughness were attained. Effects of molecular architecture of metal soaps were investigated in PEEK (calcium stearate/PEEK and sodium stearate/PEEK). The selected compounds reduced the process viscosity due to the high temperature co-continuous morphology of metal soaps. Unlike the iPP system that incorporates spherical particles, interaction between PEEK and metal soaps resulted in two discrete and co-continuous phases of PEEK and the metal stearates. DMA and melt rheology exhibited that sodium stearate/PEEK composites are stiffer. Effective moduli of secondary metal stearate phase were calculated using different composite theories, which suggested bicontinuous morphology to the metal soaps in PEEK. Use of low molecular weight crystallizable solvents was investigated in reactive systems. Formation of anisotropic reinforcements was evaluated using dimethyl sulfone (DMS) as the crystallizable diluent and diglycidyl ether of bisphenol-A (DGEBA)/m-phenylene diamine (mPDA) material system as the epoxy thermoset. Miscible blends of DMS and DGEBA/mPDA form homogenous mixtures that undergo polymerization induced phase separation, once the DGEBA oligomers react with mPDA. The effect of the competition between the crystallization and phase separation of DMS resulted in nano-wires to micro-scale fiber-like crystals that were generated by adjusting the reaction temperature and DMS concentration.

Pseudo-dynamic testing was employed to observe the seismic performance of three different retrofit methods on two story three bay reinforced concrete frame structures. The three test frames have hollow clay tile (HCT) infills in the central bay. All of the test frames represent the seismic deficiencies of the Turkish construction practice such as use of plain reinforcing bars, low strength concrete and insufficient confining steel. Two non-invasive and occupant friendly retrofit schemes suggested in the Turkish Earthquake Code, namely use of Fiber Reinforced Polymers and precast concrete panels integrated on the HCT infills and traditional approach of adding concrete infill wall were employed. Specimens were subjected to three different scale levels of North-South component of Duzce ground motion. Reference specimen experienced severe damage at 100% scale level and reached collapse stage due to the loss of integrity of the infill wall and significant damage on the boundary columns. The retrofitted test structures were able to survive the highest level 140% Duzce ground motion. Test results confirmed the success of the retrofit methods for simulated earthquake loads.

El uso de materiales compuestos de matriz polimérica (FRP) emerge como alternativa al hormigón convencionalmente armado con acero debido a la mayor resistencia a la corrosión de dichos materiales. El presente estudio investiga el comportamiento en servicio de vigas de hormigón armadas con barras de FRP mediante un análisis teórico y experimental. Se presentan los resultados experimentales de veintiséis vigas de hormigón armadas con barras de material compuesto de fibra de vidrio (GFRP) y una armada con acero, todas ellas ensayadas a flexión de cuatro puntos. Los resultados experimentales son analizados y comparados con algunos de los modelos de predicción más significativos de flechas y fisuración, observándose, en general, una predicción adecuada del comportamiento experimental hasta cargas de servicio. El análisis de sección fisurada (CSA) estima la carga última con precisión, aunque se registra un incremento de la flecha experimental para cargas superiores a las de servicio. Esta diferencia se atribuye a la influencia de las deformaciones por esfuerzo cortante y se calcula experimentalmente.Se presentan los aspectos principales que influyen en los estados límites de servicio: tensiones de los materiales, ancho máximo de fisura y flecha máxima permitida. Se presenta una metodología para el diseño de dichos elementos bajo las condiciones de servicio. El procedimiento presentado permite optimizar las dimensiones de la sección respecto a metodologías más generales. / Fibre reinforced polymer (FRP) bars have emerged as an alternative to steel for reinforced concrete (RC) elements in aggressive environments due to their non-corrosive properties. This study investigates the short-term serviceability behaviour of FRP RC beams through theoretical and experimental analysis. Twenty-six RC beams reinforced with glass-FRP (GFRP) and one steel RC beam are tested under four-point loading. The experimental results are discussed and compared to some of the most representative prediction models of deflections and cracking for steel and FRP RC finding that prediction models generally provide adequate values up to the service load. Additionally, cracked section analysis (CSA) is used to analyse the flexural behaviour of the specimens until failure. CSA estimates the ultimate load with accuracy, but it underestimates the experimental deflection beyond the service load level. This increment is mainly attributed in this work to shear induced deflection and it is experimentally calculated.A discussion on the main aspects of the SLS of FRP RC is introduced: the stresses in materials, maximum crack width and the allowable deflection. A methodology for the design of FRP RC at the serviceability requirements is presented, which allows optimizing the overall depth of the element with respect to more generalised methodologies.

Les techniques de renforcement de structures en béton armé (BA) par collage de polymères renforcés de fibres (PRF) trouvent un important champ d'applications dans le renforcement des poteaux en BA. Le chemisage par PRF confine le noyau du poteau et permet d'augmenter sa résistance et sa ductilité. Bien que de nombreux travaux expérimentaux aient été consacrés à l'étude de l'effet de confinement du PRF sur le comportement des poteaux en BA, la réalisation d'une simulation réaliste de la réponse structurelle de tels éléments présente de nombreuses difficultés liées aux modèles de comportement peu appropriés à reproduire précisément la réponse mécanique du béton confiné. Dans cette recherche, un modèle de comportement élasto-plastique endommageable est développé pour reproduire la réponse mécanique du béton sollicité suivant un chemin triaxial de contraintes. Ce modèle prend en compte différents mécanismes de comportement du béton tels que les déformations irréversibles, l'endommagement dû à la microfissuration, la sensibilité au confinement et les caractéristiques de dilatation. Un processus d'identification des paramètres du modèle est proposé sur la base d'essais classiques. La validation de ce modèle est ensuite démontrée en comparant des résultats de simulations à des données expérimentales de la littérature sur des bétons confinés activement puis des bétons confinés par des PRF présentant une large gamme de rigidité. Le modèle proposé est également comparé à différentes modélisations de la littérature. Les capacités du modèle sont illustrées et analysées sur des applications tridimensionnelles de poteaux en BA de taille réelle, non confinés et confinés par PRF. / For the past two decades, externally bonded Fiber Reinforced Polymers (FRP) has gained much popularity for seismic rehabilitation of reinforced concrete (RC) columns. In this technique, FRP wrap installed on the surface of a column acts as lateral confinement and enhance the strength and deformation capacity of the concrete element. Although many experimental works have been devoted to the study of confining effect of FRP on the behavior of RC columns, the numerical simulation of FRP-jacketed RC columns remains a challenging issue due to the lack of appropriate constitutive model for confined concrete. In this study, a damage plastic model is developed to predict the behavior of concrete under triaxial stress states. The proposed model takes into account different material behavior such as irreversible strain, damage due to microcracking, confinement sensitivity and dilation characteristic. A straightforward identification process of all model’s parameters is then presented. The identification process is applied to different normal strength concrete. The validity of the model is then demonstrated through confrontation of experimental data with simulations considering active confined concrete and FRP confined concrete with a wide range of confinement stiffness. The proposed constitutive model is also compared with other models from the literature and the distinguishing features of this new model are discussed. Furthermore, the capacity of the model in the three-dimensional finite element analysis of full-scale RC columns is demonstrate and discussed.

Orientador: Marco-Aurelio De Paoli / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Quimica / Made available in DSpace on 2018-08-06T15:37:34Z (GMT). No. of bitstreams: 1 Morita_AugustoTeruo_D.pdf: 5448584 bytes, checksum: b45ea2a6a097d79180fbd6e43996e982 (MD5) Previous issue date: 2006 / Resumo: O objetivo deste trabalho foi preparar placas e parafusos auto-reforçados a partir do copolímero poli(L/DL ácido láctico), para serem utilizados na fixação interna rígida de fraturas. Este copolímero pode ser hidrolisado quando inserido no corpo humano e seus produtos de degradação são expelidos através do ciclo metabólico. Foi construído um aparelho de autoreforçamento que deve ser acoplado a uma Máquina Universal de Ensaios, capaz de orientar uniaxialmente amostras poliméricas através da passagem por uma matriz aquecida de dimensões menores que a amostra. Foram realizados ensaios prévios de auto-reforçamento com amostras de poliestireno. Ensaios de auto-reforçamento com o copolímero poli(L/DL ácido láctico) mostraram que o auto-reforçamento induziu um aumento nas propriedades mecânicas, medidas sob flexão, de ~60% no módulo de elasticidade e de ~90% na tensão máxima. A amostra moldada apresentou linhas brancas quando dobrada a temperatura ambiente, indicando a fragilização. Por outro lado, as amostras auto-reforçadas podem ser dobradas a temperatura ambiente sem que ocorresse ruptura frágil ou fragilização. Através de ensaios de difração de raios-X determinou-se que a amostra é semi-cristalina após o auto-reforçamento, apesar dos ensaios de calorimetria diferencial de varredura não identificarem positivamente a formação desta estrutura semi-cristalina. A esterilização por raios-g causa uma grande queda na massa molar, porém não altera as propriedades mecânicas. Ensaios de degradação in-vitro, com solução tampão de fosfato numa temperatura de 37 °C, das amostras esterilizadas mostraram que a massa molar sofre uma queda durante todo o ensaio de degradação e as propriedades mecânicas sofrem uma pequena queda após 15 semanas. As amostras auto-reforçadas podem ser usinadas na forma de placas e parafusos. As placas usinadas do copolímero de poli(ácido láctico) ainda podem ser dobradas a temperatura ambiente sem que ocorra quebra ou formação de linhas brancas / Abstract: In this work self-reinforced plates and screws to be used for rigid internal fixation of fractures were produced using the poly(L/DL-lactic acid) copolymer. This copolymer can be hydrolyzed when implanted in the human body and its degradation products are expelled by the metabolic cycle. A self-reinforcement module was built. It can be attached to a Universal Mechanical Testing Machine, which is capable to uni-axially orient the polymeric sample though a heated die. Initial tests of processing and self-reinforcement were conducted with polystyrene. Self-reinforcement tests were conducted with the poly(L/DL-lactic acid) copolymer. The results show that the die-drawing process induced an increase in the flexural mechanical properties of ~60% in the elastic modulus and of ~90% in the maximum tension, both measured under flexion conditions. White lines (crazing) appears when the molded sample was bent at ambient temperature, whereas the self-reinforced sample could be bent in the same conditions without fracture or sample fragilization. X-ray diffraction results indicated that the self-reinforced sample exhibits a semi-crystalline structure; notwithstanding the differential scanning calorimetry results do not conclusively identify the sample morphology. g-rays sterilization caused a decrease in the molar mass, but the mechanical properties were not affected. In-vitro degradation test were conducted at 37 °C in a buffered aqueous solution. Sterilized samples show a larger decrease in the molar mass since the beginning of the test, but the mechanical properties exhibits a smaller decrease during the test. The self-reinforced samples can be molded to the form of plates and screws. The final sample could be bent at ambient temperature without fragilization / Doutorado / Quimica Inorganica / Doutor em Ciências