Alkali-aggregate reaction (AAR) is one of the most harmful distress mechanisms affecting the performance of aging reinforced concrete structures worldwide. Although several prediction models have been developed to assess the chemical reaction, a thorough and comprehensive approach with the capabilities to correlate important parameters that affect AAR and the mechanical properties of deteriorated materials, as well as the abilities to describe the current damaged state of AAR-affected structures (diagnosis) and predict the potential of further damage (prognosis) is still lacking. Such information is essential in selecting efficient remedial/rehabilitation actions for existing structures in the field. This project aims to develop a practical, yet accurate engineering-based finite element (FE) model for assessing AAR damage and predicting the future behaviour of affected infrastructure. The model is validated through three analyses. First, its capability to accurately simulate sound concrete under mechanical loading is verified by successfully simulating different beam failure mechanisms and cracking patterns, as well as predicting the members’ full force-deflection curves. Next, AAR anisotropic expansion under different stress-state (confinement) conditions is accurately simulated and verified by correlation with laboratory tests. Lastly, an AAR-affected slender reinforced concrete structure (Robert-Bourassa/Charest overpass) is successfully simulated by performing a condition assessment based on several tests performed prior to its demolition.

Alkali-aggregate reaction (AAR) is one of the most harmful distress mechanisms affecting the durability and serviceability of concrete infrastructure worldwide. Currently, there is a need to forecast future behaviour of AAR-affected concrete, and thus analytical and numerical models have been developed over the years. Larive developed a model in 1998 that is able to describe the behaviour of concrete samples in the laboratory. This model has been widely accepted and used by AAR community, even to predict the behaviour of concrete structures in the field. Larive’s model is based upon three main parameters and although they are normally set according to the concrete type under analysis, these parameters do not have a clear physicochemical meaning to describe AAR-induced expansion and development. Furthermore, it is widely known that AAR is influenced by several measurable variables such as the type and reactivity degree of the aggregates, temperature, moisture, and alkali content; those are currently not accounted for in Larive’s equation. This work aims to incorporate the previously mentioned parameters to Larive’s equation so that AAR kinetics and final expansion might be precisely described in the lab and/or field. Results indicate that the proposed modified Larive’s equation can predict AAR kinetics and final expansion in the laboratory although some modifications seem still necessary to assess field structures.

One of the biggest challenges nowadays when dealing with critical AAR-affected infrastructure is to determine the induced expansion and damage attained to date in the different locations of the structure (i.e. diagnosis), to forecast its potential for further distress over time (i.e. prognosis), as well as its potential structural implications. There are a number of prognosis techniques that have been developed in the past decades which are currently being used worldwide. Yet, most of these approaches use residual expansion procedures based on accelerated laboratory tests performed on cores extracted from damaged structures. However, the majority of the results gathered from these tests have been found to be inaccurate when compared to the swelling behavior of the respective structure in the field and several potential issues have been raised with respect to the test setup and alkali leaching. This work aims to evaluate the efficiency of the various commonly used laboratory setups to assess the residual expansion of AAR-affected concrete. Three different setups (i.e. 38°C and 100% RH; 38°C and soaked in 1M NaOH and; 38°C and wrapped in 0.7M NaOH) and two types of reactive aggregates (fine and coarse) were selected for this research. Expansion is monitored over time and four damage degrees (i.e. 0.05%. 0.12%, 0.20% and 0.30%) are selected for further chemical, microscopic (DRI) and non-destructive tests. Results demonstrated that the 1 M NaOH protocol is much more aggressive than the other two setups. Furthermore, it provides the samples with a unique damage pattern that is different from field affected concrete. Finally, the proposed setup shows to be reliable in providing tested samples with similar deterioration mechanisms than expected. Yet, more efficiency in the reaction kinetics and understanding of the alkalis exchange from the system is still required.

The alkali-aggregate reaction was first reported in 1940 as a cause of severe cracking in some concrete structures. It is only in recent years that papers have been published dealing with the effects of AAR on the performance of structures. This thesis outlines a methodology for numerical simulation of the response of concrete subjected to continuing AAR. First a constitutive model is presented based on the framework proposed by Pietruszczak (1996). The formulation incorporates an assumption that the rate of expansion depends on the confining pressure, the age of concrete and the temperature. The progress in the reaction is coupled with the degradation of mechanical properties, in particular the elastic modulus and the compressive and tensile strengths. Subsequently, the procedures for generating finite element models are described, including geometric modeling, mesh generation techniques, graphical representation of the results and interfacing between pre- and post-processor and the finite element solvers. The numerical analysis, undertaken in this thesis, pertains to the Beauharnois Powerhouse, situated in Quebec, Canada. The powerhouse has been experiencing problems related to a continuing expansion of concrete due to AAR since the early 1960’s. The progressive formation of macrocracks and the volumetric expansion in concrete has caused operational problems, such as the reduction in clearance between turbine runner blades and throat rings. In this study, typical structural units of the Beauharnois Powerhouse were selected for the numerical analysis. The AAR constitutive model was applied in a finite element framework. Mechanical properties of concrete were carefully evaluated based on available experimental data. Simulations were focused on the deformation and the time history of progressive macro/microcracking due to continuing reaction. Structural responses under isothermal as well as non-isothermal conditions were simulated. The results of the numerical analyses were then compared with in-situ measurements. / Thesis / Master of Engineering (ME)

Recycling composite glass with different colours in order to be manufactured into new glass products is at present not economically viable. Therefore, the search for new issues other than stockpile areas or dumping sites could be a serious opportunity. To a certain extent, one of the possible solutions is to use the recycled glass in manufacturing cements and in the preparation of concrete mixtures. However, it is essential to manage the two main behaviours that the glass can have when used in cement-based materials: (1) the use of glass as coarse aggregates reveals harmful behaviour related to alkali-silica reaction; (2) on the other hand, it can result in useful behaviour related to pozzolanic reaction if used as fine particles. Furthermore, the significant alkali content should not be overlooked as their mass corresponds to about 13% of the total mass of the glass and as they may activate the alkali-silica reaction. An experimental programme was conducted to provide answers to the various questions raised about the use of glass in cement-based materials. The first part of this work was primarily devoted to the evaluation of the reactive potential of glass in mortars (alkali and pozzolanic reactions). At this stage, nine classes of glass particles ranging from 3[mu]m to 2.5 mm were considered. Then, fine glass particles were used in order to counteract the negative effect of some classes of coarse aggregates having revealed alkali-reactive behaviour. The second part of this work was performed to study the mechanisms that could explain the behaviours of fine and coarse particles in aqueous and concentrated environments. Different answers have been proposed to explain the observed behaviour in terms of grain sizes of glass.

Orientador: Aparecido Augusto de Carvalho / Resumo: O desenvolvimento de sistemas capazes de monitorar e controlar informações estão sendo cada vez mais utilizados, permitindo maior refinamento e confiança nos resultados. Na área instrumental de laboratório, existem situações que exigem maior controle sobre o ensaio. Com o atual avanço na área de microprocessadores, microcontroladores, sensores inteligentes, entre outros, têm impulsionado a utilização dos mesmos na melhoria e confiabilidade de processos mecânicos, químicos, físicos, biológicos, de inteligência artificial, de melhoria de qualidade de vida de pessoas com deficiência. Os processos podem ser melhorados utilizando equipamentos que possuem sistemas embarcados capazes de processar informações e fazer o controle a partir de sistemas operacionais que antes só eram possíveis nos computadores. Na área das grandes estruturas de concreto não é diferente, deve-se monitorar algumas variáveis que resultam em uma manifestação patológica conhecida como reação álcali-agregado, que é um fenômeno de expansão que afeta a durabilidade e a resistência mecânica do concreto, este fenômeno está ligado à algumas variáveis, a umidade relativa interna é umas delas. Existem várias pesquisas no sentido de inibir as reações álcali-agregados, e é sabido que um importante passo antes do acontecimento da reação álcali-agregado em obras hidráulicas, como as usinas hidroelétricas é o monitoramento da umidade e da temperatura no interior do concreto e, acompanhamento das fissuras inerentes do proce... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The development of systems able to monitor and control information is increasingly being used, allowing greater refinement and confidence in results. In the instrumental area of the laboratory, there are situations that require greater control over the test. With the current progress in the area of microprocessors, microcontrollers, intelligent sensors, among others, have encouraged the use of these in the improvement and reliability of mechanical, chemical, physical, biological, artificial intelligence, improving the quality of life of people with deficiency. Processes can be improved by using equipment that has embedded systems that can process information and control from operating systems that were previously only possible on computers. In the area of large concrete structures is not different, it is necessary to monitor some variables that result in a pathological manifestation known as alkali-aggregate reaction, which is an expansion phenomenon that affects the durability and the mechanical resistance of the concrete, this phenomenon is connected to some variables, the relative humidity is one of them. There are several researches to inhibit alkali-aggregate reactions and is known that an important step before the occurrence of the alkali-aggregate reaction in hydraulic works such as hydroelectric plants is the monitoring of humidity and temperature inside the concrete and, the inherent fissures of the expansion process. The present work implemented a low cost, temperat... (Complete abstract click electronic access below) / Doutor

A reação álcali-agregado é uma patologia do concreto conhecida há mais de 60 anos no meio técnico, e envolve os álcalis, provenientes em maior parte do cimento, e certos constituintes mineralógicos dos agregados. O produto desta reação é um gel higroscópico, que, em contato com a água, aumenta de volume consideravelmente. Os danos decorrentes da reação envolvem principalmente o desenvolvimento de um quadro de fissuração intenso, o que acarreta outras diversas manifestações patológicas, e movimentações de peças estruturais. Os relatos mais freqüentes de ocorrência da reação sempre foram em estruturas hidráulicas, tais como as barragens, em função do permanente contato com a água. Entretanto, nos últimos anos, diversos outros tipos de estruturas têm sido afetados, inclusive fundações de edificios, deixando o problema em evidência. O único modo de combater a reação é prevení-la através do uso de adições minerais, e para tal se faz necessário avaliar o agregado em questão, dispondo de um método de ensaio confiável. Este trabalho buscou analisar um dos métodos de ensaio mais utilizados no mundo todo para avaliar a reação álcali-agregado, a ASTM C 1260. Foram modificados alguns parâmetros de moldagem para verificar a influência destes nos resultados de expansão fornecidos pelo ensaio. Avaliaram-se os fatores: relação água/cimento, consistência da argamassa e forma do grão de agregado, e no total foram conduzidos 63 diferentes ensaios de expansão para analisar a influência destes fatores. Ensaios auxiliares foram realizados para justificar os resultados, e utilizou-se tratamento estatístico para verificação da validade dos mesmos. Foi observado que a forma de britagem do agregado, para posterior realização do ensaio, interfere diretamente nas condições de moldagem. E, o mais importante: isto resulta em expansões diferenciadas na idade de avaliação do agregado. Para evitar classificações errôneas de agregados através do uso da ASTM C 1260, uma norma usada em vários países, sugere-se que sejam fixados o maior número possível de parâmetros de moldagem. / The alkali-aggregate reaction 1s a concrete pathology known for more than 60 years, involving the alkalis (mostly from the cement), and some mineralogical components of the aggregates. The product o f this reaction is a hygroscopic gel that, in contact with water, has its volume greatly increased. The damage resulting from this reaction is mainly a situation of intense craking, which causes other pathological manifestations and movements of the structure. The most frequent reports o f this reaction have always been in hydraulic structures, such as dams, caused by the permanent contact with water. However, in the last few years, many other kinds of structures have been affected, including building foundations; which brought the problem to international evidence. The only way to prevent the expansion caused by the reaction is the use o f mineral admixtures as a substitute o f part o f the cement content, but first it is necessary to evaluate the aggregate in use, with a satisfactory test method. This work' s purpose is to analyze one o f the test methods commonly used throughout the world to evaluate the alkali-aggregate reaction, the ASTM C 1260. Some molding parameters have been modified so as to check their influence on the results of expansion obtained through this test. These factors have been evaluated: water/cement ratio, mortar consistency and shape of the aggregate particle. A total of 63 different tests of expansion were conducted to evaluate their influence. Additional tests were done so as to justify the results, and statistical analysis was used to validate the results. It was observe that the way the aggregate is crushed, for subsequent execution of the test, directly interferes in the conditions of molding. And, the most important: the way the aggregate is crushed results in different expansions and it is increased with the age of the evaluation of the test. To prevent a erroneous classification o f the aggregates through the use of ASTM C 1260, a standard used in many countries, it is suggested that test conditions should be better defined.

A reação química entre os hidróxidos alcalinos presentes no cimento e fases minerais reativas presentes no agregado dá origem a um gel sílico-alcalino que ao adsorver água se expande, podendo causar expansão e fissuração do concreto, afetando a sua durabilidade. Este fenômeno é chamado de reação álcali-agregado (RAA) e embora seja conhecido há aproximadamente 70 anos no meio técnico, e principalmente estar relacionado às grandes barragens e estruturas de concretos diversas de Usina hidrelétricas, nos últimos anos outros tipos de estruturas têm sido afetadas, inclusive fundações de edifícios, deixando ainda mais o problema em evidência. As medidas preventivas para esta reação já são conhecidas há muito tempo e aplicadas rotineiramente em estruturas de usinas hidrelétricas quando se tem agregados reativos, sendo recentemente normalizadas no Brasil. Porém, ainda existem grandes carências em se combater o fenômeno quando já se encontra instalado no concreto. Diante do exposto, este trabalho, desenvolvido dentro do programa de P&D de Furnas/ANEEL, teve como principal objetivo verificar a potencialidade de compostos a base de Silanos através de testes laboratoriais com vistas à aplicação em concretos contendo agregados reativos na tentativa de minimizar as expansões residuais em estruturas que já se encontram afetadas. As atividades desenvolvidas contemplam ensaios em barras de argamassas que foram tratadas, em três etapas distintas, com compostos a base de silanos. A primeira etapa foi desenvolvida como projeto piloto, e seus resultados serviram de apoio para a realização das etapas seguintes. Nas segunda e terceira etapas foi verificada a atuação dos produtos na prevenção e na mitigação da RAA, respectivamente. Os resultados obtidos na primeira etapa não foram satisfatórios, pois as expansões foram maiores do que as da referência. Já para a segunda e terceira etapas, os procedimentos para aplicação dos tratamentos previram que as barras fossem tratadas sem que estivessem previamente expostas à solução de NaOH, conforme realizado na primeira etapa, o que possivelmente ocasionou o acréscimo das expansões, sem a eficiência do produto. Os resultados obtidos nessas etapas se mostraram eficientes tanto na prevenção quanto na mitigação das expansões residuais da RAA e, embora o lítio produza uma maior eficiência na redução das expansões, os tratamentos à base de silano, e principalmente do tipo octiltrietoxisilano, mostraram muito promissores. Após o lítio, o tratamento à base do silano OCTEO, no teor de 10%, com o surfactante tipo DMSO, em aplicação múltipla, se mostrou mais eficiente. Para complemento da parte experimental da pesquisa, foram realizados ensaios de absorção de água por imersão e análises por microscopia eletrônica de varredura e verificou-se o efeito dos produtos de tratamento nos poros das argamassas. / The chemical reaction between cement alkaline hydroxides and reactive silica present in some aggregates produces an alkaline silicate gel that expands when it adsorbs water and can cause concrete expansion and cracking, thereby affecting durability. This phenomenon is known as alkali-aggregate reaction (AAR) and was first detected over 70 years ago as the cause of deterioration especially in large dams and a number of other concrete structures in hydroelectric power plants in general. In recent years, AAR has also been noted in other types of structure, such as building foundations, bringing the problem more dramatically to the fore. Preventive measures for AAR have long been known and are routinely applied in hydroelectric power plant structures when there are reactive aggregates. These measures have been recently standardized in Brazil. Nevertheless there are still flaws in our knowledge regarding how to mitigate expansion when this phenomenon is already present in the concrete. Considering these aspects, the main purpose of this study, developed within Furnas/ANEEL Research and Development Program, is to verify, at laboratory level, the mitigating potential of silane compounds in residual expansion in affected mortars which contain reactive aggregates. Assays were performed on bars of mortar treated with silane compounds in three different stages. The first stage was developed as a pilot study with results serving as support for the following stages. The second and third stages verified the activity of the products in the prevention and mitigation of AAR, respectively. The results obtained in the first stage were unsatisfactory as the expansions were greater than the references. The second and third stage treatments required bars not previously exposed to NaOH, differently from the first stage and which may have been the cause of the increased expansion and product inefficiency. The results in these stages proved satisfactory both in preventing and mitigating AAR residual expansions, and although lithium is more efficient in reducing expansions, silane based treatments, mainly the octyltriethoxysilane types proved most promising. Following lithium, the most efficient treatment was 10% content OCTEO silane with a DMSO type surfactant applied in multiple layers. To complement the experimental part of the study, assays on water absorption through immersion were performed as well as scanning electron microscopy to verify the effect of the treatment products on the mortar pores.

Submitted by Erika Demachki (erikademachki@gmail.com) on 2014-09-24T21:42:24Z No. of bitstreams: 2 COUTO, T.A. - RAA Estudo do fenômeno em rochas silicosas 2008.pdf: 9649591 bytes, checksum: 0ecde13fd55a962ac1bc6a8f43f44415 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2014-09-25T10:37:27Z (GMT) No. of bitstreams: 2 COUTO, T.A. - RAA Estudo do fenômeno em rochas silicosas 2008.pdf: 9649591 bytes, checksum: 0ecde13fd55a962ac1bc6a8f43f44415 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2014-09-25T10:37:27Z (GMT). No. of bitstreams: 2 COUTO, T.A. - RAA Estudo do fenômeno em rochas silicosas 2008.pdf: 9649591 bytes, checksum: 0ecde13fd55a962ac1bc6a8f43f44415 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2008-08-29 / The symposium on Alkali-Aggregate Reaction (AAR) held in Goiânia in 1997 led to a greater dissemination of this deleterious reaction in Brazil, motivating engineers to find solutions for the problem. Recently, in Brazil, this occurrence has occurred in several building foundations in the Recife metropolitan area besides hydraulic powers plants, as can be seen in some publications from the last AAR Symposium in 2006. This study presents an investigation of several aggregates commercially used in the State of Goiás, Brazil, in order to determine their potential reactivity. In addition, their behavior were compared to the ones from aggregates from several Brazilian classic cases such as Furnas HPP (MG), Moxotó HPP (AL/BA), Pedro Beicht Dam (SP/MG) and Jaguara HPP (SP), beyond aggregates used in civil works of Pernambuco and also a highly reactive basalt from Paraná. The experimental program tested twenty-two rock samples with varied lithology and two local types of cements that had the same clinker (cement A and B). The research uses petrographic analyses, accelerated mortar bar test (NBR 15577), chemical test method (NBR 9774) and accelerated concrete prisms test (based on NBR 15577), to assess the reactivity of the rocks. The results indicate that aggregates show varied expansivity and reactivity among them and sometimes one specific aggregate present different behavior by comparing the results form the test methods performed. This study also discusses the possibility of preventing deleterious expansion from tested aggregates by using the correct type of cement. / Desde 1997, quando as reações expansivas do tipo álcali-agregado ganharam uma maior difusão no âmbito nacional com o primeiro Simpósio sobre Reação Álcali- Agregado (RAA), foi observado um envolvimento mais expressivo de diversos profissionais da área na busca de soluções para este problema. Atualmente, no Brasil, esta manifestação patológica que até então, de uma maneira geral, era exclusiva de obras hidráulicas, já vem sendo observada em estruturas residenciais e comerciais dentro de grandes centros, como os casos relatados da região metropolitana de Recife durante o segundo Simpósio sobre RAA em 2006. No estado de Goiás esta preocupação não foi diferente e, em virtude disto, este trabalho tem por objetivo apresentar a investigação da potencialidade reativa de agregados deste estado comparando com o comportamento de agregados procedentes de algumas obras que representam casos clássicos brasileiros com a UHE Furnas (MG), UHE Moxotó (AL/BA), Barragem de Pedro Beicht (SP) e UHE Jaguara (SP), além de agregados utilizados em obras civis de Pernambuco e um basalto altamente reativo do Paraná. Foram, portanto, selecionados vinte e dois agregados com litologias variadas para este trabalho. A utilização da técnica de petrografia e do método acelerado em barras de argamassa (NBR 15577) foram objeto de estudo da pesquisa, incluindo a análise dos agregados por meio do método químico (NBR 9774) e pelo método dos prismas de concreto (NBR 15577), adaptado na versão acelerada. Foram empregados dois tipos de cimento de mesmo clínquer (cimento A e B) procedentes do estado de Goiás. Os resultados indicaram potencialidades reativas variadas entre os agregados e comportamentos distintos entre alguns métodos de ensaio para o mesmo agregado, bem como a viabilidade do combate da RAA, dependendo do cimento empregado.

A reação álcali-agregado é uma patologia do concreto conhecida há mais de 60 anos no meio técnico, e envolve os álcalis, provenientes em maior parte do cimento, e certos constituintes mineralógicos dos agregados. O produto desta reação é um gel higroscópico, que, em contato com a água, aumenta de volume consideravelmente. Os danos decorrentes da reação envolvem principalmente o desenvolvimento de um quadro de fissuração intenso, o que acarreta outras diversas manifestações patológicas, e movimentações de peças estruturais. Os relatos mais freqüentes de ocorrência da reação sempre foram em estruturas hidráulicas, tais como as barragens, em função do permanente contato com a água. Entretanto, nos últimos anos, diversos outros tipos de estruturas têm sido afetados, inclusive fundações de edificios, deixando o problema em evidência. O único modo de combater a reação é prevení-la através do uso de adições minerais, e para tal se faz necessário avaliar o agregado em questão, dispondo de um método de ensaio confiável. Este trabalho buscou analisar um dos métodos de ensaio mais utilizados no mundo todo para avaliar a reação álcali-agregado, a ASTM C 1260. Foram modificados alguns parâmetros de moldagem para verificar a influência destes nos resultados de expansão fornecidos pelo ensaio. Avaliaram-se os fatores: relação água/cimento, consistência da argamassa e forma do grão de agregado, e no total foram conduzidos 63 diferentes ensaios de expansão para analisar a influência destes fatores. Ensaios auxiliares foram realizados para justificar os resultados, e utilizou-se tratamento estatístico para verificação da validade dos mesmos. Foi observado que a forma de britagem do agregado, para posterior realização do ensaio, interfere diretamente nas condições de moldagem. E, o mais importante: isto resulta em expansões diferenciadas na idade de avaliação do agregado. Para evitar classificações errôneas de agregados através do uso da ASTM C 1260, uma norma usada em vários países, sugere-se que sejam fixados o maior número possível de parâmetros de moldagem. / The alkali-aggregate reaction 1s a concrete pathology known for more than 60 years, involving the alkalis (mostly from the cement), and some mineralogical components of the aggregates. The product o f this reaction is a hygroscopic gel that, in contact with water, has its volume greatly increased. The damage resulting from this reaction is mainly a situation of intense craking, which causes other pathological manifestations and movements of the structure. The most frequent reports o f this reaction have always been in hydraulic structures, such as dams, caused by the permanent contact with water. However, in the last few years, many other kinds of structures have been affected, including building foundations; which brought the problem to international evidence. The only way to prevent the expansion caused by the reaction is the use o f mineral admixtures as a substitute o f part o f the cement content, but first it is necessary to evaluate the aggregate in use, with a satisfactory test method. This work' s purpose is to analyze one o f the test methods commonly used throughout the world to evaluate the alkali-aggregate reaction, the ASTM C 1260. Some molding parameters have been modified so as to check their influence on the results of expansion obtained through this test. These factors have been evaluated: water/cement ratio, mortar consistency and shape of the aggregate particle. A total of 63 different tests of expansion were conducted to evaluate their influence. Additional tests were done so as to justify the results, and statistical analysis was used to validate the results. It was observe that the way the aggregate is crushed, for subsequent execution of the test, directly interferes in the conditions of molding. And, the most important: the way the aggregate is crushed results in different expansions and it is increased with the age of the evaluation of the test. To prevent a erroneous classification o f the aggregates through the use of ASTM C 1260, a standard used in many countries, it is suggested that test conditions should be better defined.