Aborda-se, neste trabalho, a tecnologia do concreto auto-adensavel CAA aplicada a producao de pre-fabricados. Destaca-se a contribuicao feita ao Metodo IBRACON, largamente utilizado no pais para a dosagem do concreto comum, a partir de uma nova metodologia de dosagem proposta para o CAA, que considera duas inovacoes principais. A primeira foi o conceito de correcao da coesao do concreto fresco por substituicao de finos pozolanicos ou nao pozolanicos correspondentes, de maior area especifica, de modo que formulacoes mais pobres em cimento exigem maior teor de substituicao quando comparadas com composicoes mais ricas, para manter as mesmas caracteristicas de trabalhabilidade. Com isso, foi possivel acrescentar um 4° quadrante de correlacao dos parametros envolvidos na mistura ao Diagrama de Dosagem do metodo original. A segunda inovacao foi incorporar alguns ensaios de trabalhabilidade especificos. Para tanto, realizou-se uma analise critica dos principais testes disponiveis, apresentando suas vantagens e limitacoes, correlacionando com os parametros reologicos fundamentais, bem como com uma aplicacao pratica, validando-os para a qualificacao do CAA para a producao. Dessa forma, foram estudados tracos com dois niveis de auto-adensabilidade, sem adicoes e com metacaulim ou filer calcario, para resistencias compreendidas entre 50-70MPa, em um amplo programa experimental que totalizou 11 familias de concretos e mais de 55 tracos, no geral. Foram avaliados tambem os cuidados especificos para a producao do CAA, analisando aspectos de produtividade e viabilidade de implantacao, a partir de um estudo voltado a uma empresa especializada. Concluiu-se que o CAA e uma opcao vantajosa para o setor, sobretudo com emprego de adicoes minerais, porem e extremamente necessario haver algumas modificacoes de equipamentos e procedimentos para sua adequacao em fabrica. / This work focus the technology of the self-compacting concrete SCC, applied to the precast production. The main innovation was the contribution made to the IBRACON Method, which is widely utilized in Brazil for conventional concrete design, from the new proposed design methodology for SCC, which considers two main innovations. The first one was the concept of correction of the cohesion of the fresh concrete by substitution of pozzolanics or no pozzolanics corresponding mineral additives, with higher specific area, where the poorest formulations in cement requires higher content of substitution when compared with richer compositions, to maintain the desired characteristics of workability. It enables to add a 4th quadrant to the Dosage Diagram of the original method. The second innovation was to incorporate of some workability tests of the fresh concrete, that become important for the qualification of the self-compacting concrete for a practical application. Thus, it carried out a critical analysis of the main available tests, presenting its advantages and limitations, correlating with the parameters of rheology, and with a practical application, validating them for the qualification of SCC for production. Therefore, it was studied mixes of SCC with two levels of self-compactability, without additives and with metacaolin or limestone filler, for mechanical resistances between 50- 70MPa, in a wide experimental program with 11 families of concretes and more of 55 mixes. It was also evaluated the specific cares for the production of SCC, analyzing aspects of productivity and feasability of implementation, based in a case study. It was concluded that SCC is a worthwhile option for the sector, especially using mineral additives, however it is extremely necessary some equipment and procedure modifications in order to adapt the factories to this new technology.

Esta tese aborda as influências do agregado leve na dosagem, características físicas e mecânicas de concreto autoadensável (CAA) quando na fração de graúdo da mistura, substitui-se parte do volume absoluto da brita de basalto (máx 19 mm) pelo volume equivalente de argila expandida brasileira (máx 12,7 mm). O fato de conhecer as implicações na reologia do CAA, provocadas pelo uso conjunto de agregados com características físicas distintas e, apresentar este tipo de concreto como uma alternativa promissora para uso na indústria da pré-fabricação em concreto, justificam esta pesquisa. A substituição da brita de basalto pela argila expandida (AE-1506), em teores de volume absoluto, foi de 20%, 40%, 60%, 80% e 100%. Como resultados, produziram-se concretos autoadensáveis com consumo de aglomerantes (cimento Portland CP V-ARI e sílica ativa) da ordem de 510 kg/m³, que atenderam aos limites de autoadensabilidade propostos pela norma NBR 15823-1 (2010). Na condição endurecida, apresentaram massa específica seca de 2.358,3 a 1.720,7 kg/m³, resistência à compressão (fc28) de 60 a 43 MPa, módulo de elasticidade (Esc) de 23 a 34 GPa e eficiência estrutural (FEE) de 22 a 29 MPa.dm³.kg-1, sem sinais visíveis de frente de carbonatação. Obteve-se concreto leve autoadensável (CLAA) a partir de misturas com fração de graúdo foi composta por 60% de argila expandida e 40% de brita de basalto, que atingiram massa específica seca de 1.986 kg/m³, resistência a compressão (fc28) de 51,3 MPa e condutividade térmica () de 1,07 a 1,53 W/m.K. Constatou-se que a argila expandida interfere significativamente nas características dos concretos exigindo, na comparação com CAA confeccionado com 100% de brita de basalto, maior teor de argamassa e relação volume de água/volume de finos mais elevado. / This thesis discusses aspects related to the influence of lightweight aggregate in the mix design, physical and mechanical properties of the self-compacting concrete (SCC) when replacing part of the absolute volume of basalt crushed stone (máx19 mm) with a lightweight aggregate equivalent absolute volume Brazilian expanded clay (máx 12,7 mm). Understanding interference on the rheology of the SCC caused by the use of aggregates with different physical properties and recommend this type of concrete as a promising alternative for the pre-fabricated concrete industry, justify this research. The replacement of basalt crushed stone for lightweight aggregate (AE-1506), in equivalent absolute volume, was 20%, 40%, 60%, 80% and 100%. As a result, self-compacting concrete was produced with consumption of binders (cement Portland CP V-ARI and silica fume) of about 510 kg / m³, appropriate for self- compactibility limits established by the ABNT NBR 15823-1 (2010) standard. In the hardened condition, the dry density value ranged from 2.358,3 to 1.720,7 kg/m³, compressive strength (fc28) ranged from 60 to 43 MPa, elasticity modulus (Esc) ranged from 23 to 34 GPa, and efficiency structural (FES) ranged from 22 to 29 MPa.dm³.kg-1, with no visible signs of carbonation. The self-compacting lightweight expanded clay concrete (SCLC) was obtained from mixtures which its absolute volume fraction of aggregate coarse was composed by 60% of expanded clay and 40% of basalt crushed stone, with dry density of 1986 kg/m³, compressive strength (fc28) of 51.3 MPa and thermal conductivity () varied from 1,07 to 1,53 W/m.K. It was found that the expanded clay significantly interferes in the properties of concretes demanding in comparison with SCC made with 100% basalt crushed stone, mortar content and ratio higher volume of water/volume of higher fines.

The primary aim of this research was to develop a method for determining the true uniaxial tensile strength of concrete by conducting a series of cylinder splitting, modulus of rupture (MOR) and cylinder/cube compression tests. The main objectives were: • Critically reviewing previous published research in order to identify gaps in current knowledge and understanding, including theoretical and methodological contributions to the true uniaxial tensile strength of concrete. In order to maintain consistency and increase the reliability of the proposed methods, it is essential to review the literature to provide additional data points in order to add additional depth, breathe and rigor to Senussi's investigation (2004). • The design of self compacting concrete (SCC), normal strength concrete (NSC) and high strength concrete (HSC) mixes and undertaking lab-based experimental works for mixing, casting, curing and testing of specimens in order to establish new empirical evidence and data. • Analysing the data, presenting the results, and investigating the application of validity methods as stated by Lin and Raoof (1999) and Senussi (2004). • To draw conclusions including comparison with previous research and literature, including the proposal of new correction factors and recommendations for future research. 29 batches of NSC, 137 batches of HSC, 44 batches of fly ash SCC and 47 batches of GGBS SCC were cast and their hardened and fresh properties were measured. Hardened properties measured included: cylinder splitting strength, MOR, cylinder compressive strength and cube compressive strength. A variety of rheological tests were also applied to characterise the fresh properties of the SCC mixes, including: slump flow, T50, L-box, V-funnel, J-ring and sieve stability. Cylinders were also visually checked after splitting for segregation. The tensile strength of concrete has traditionally been expressed in terms of its compressive strength (e.g. ft = c x c f ). Based on this premise, extensive laboratory testing was conducted to evaluate the tensile strength of the concretes, including the direct tension test and the indirect cylinder splitting and MOR tests. These tests however, do not provide sufficiently accurate results for the true uniaxial tensile strength, due to the results being based upon different test methods. This shortcoming has been overcome by recently developed methods reported by Lin and Raoof (1999) and Senussi (2004) who proposed simple correction factors for the application to the cylinder splitting and MOR test results, with the final outcome providing practically reasonable estimates of the true uniaxial tensile strength of concrete, covering a wide range of concrete compressive strengths 12.57 ≤ fc ≤ 93.82 MPa, as well as a wide range of aggregate types. The current investigation has covered a wide range of ages at testing, from 3 to 91 days. Test data from other sources has also been applied for ages up to 365 days, with the test results reported relating to a variety of mix designs. NSC, SCC and HSC data from the current investigation has shown an encouraging correlation with the previously reported results, hence providing additional wider and deeper empirical evidence for the validity of the recommended correction factors. The results have also demonstrated that the type (size, texture and strength) of aggregate has a negligible effect on the recommended correction factors. The concrete age at testing was demonstrated to have a potentially significant effect on the recommended correction factors. Altering the cement type can also have a significant effect on the hardened properties measured and demonstrated practically noticeable variations on the recommended correction factors. The correction factors proved to be valid regarding the effects of incorporating various blended cements in the HSC and SCC. The NSC, HSC and SCC showed an encouraging correlation with previously reported results, providing additional support, depth, breadth and rigor for the validity of the correction factors recommended.

Disertační práce se zabývá studiem reologického chování materiálů na bázi cementu. Úvodní část obsahuje teoretické poznatky týkající se reologie a zkoumaných materiálů (cement, samozhutnitelný beton, superplastifikátory). Poté následuje detailní popis reologických vlastností čerstvých cementových materiálů. Kromě toho je dále uvedena hydratace cementu, kalorimetrie, Vicatův test, rentgenová difrakční analýza a elektronová mikroskopie. Experimentální část práce je rozdělena do dvou částí. První část se zabývá základní charakterizací, reologickým měřením a studiem statické meze toku směsí samozhutnitelných betonů a jejich matric. Druhá část se týká cementových past, které byly podrobeny testování pomocí reometru, Vicatova přístroje, isoperibolického a modulárního kalorimetru, rentgenového difraktometru a elektronového mikroskopu.

The present research is conducted to investigate the structural behaviour of continuously supported deep beams made with SCC. A series of tests on eight reinforced two-span continuous deep beams made with SCC was performed. The main parameters investigated were the shear span-to-depth ratio, the amount and configuration of web reinforcement and the main longitudinal reinforcement ratio. All beams failed due to a major diagonal crack formed between the applied mid-span load and the intermediate support separating the beam into two blocks: the first one rotated around the end support leaving the rest of the beam fixed on the other two supports. The amount and configuration of web reinforcement had a major effect in controlling the shear capacity of SCC continuous deep beams. The shear provisions of the ACI 318M-11 reasonably predicted the load capacity of SCC continuous deep beams. The strut-and-tie model recommended by different design codes showed conservative results for all SCC continuous deep beams. The ACI Building Code (ACI 318M-11) predictions were more accurate than those of the EC2 and Canadian Code (CSA23.3-04). The proposed effectiveness factor equations for the strut-and-tie model showed accurate predictions compared to the experimental results. The different equations of the effectiveness factor used in upper-bound analysis can reasonably be applied to the prediction of the load capacity of continuously supported SCC deep beams although they were proposed for normal concrete (NC). The proposed three dimensional FE model accurately predicted the failure modes, the load capacity and the load-deflection response of the beams tested. / Higher Education Institute in the Libyan Government

A criação de programas federais voltados para o desenvolvimento da infraestrutura brasileira no final da década passada resultou no aquecimento do setor de construção civil. Nesse contexto, o sistema de paredes de concreto moldadas no local se destacou devido à sua alta produtividade e repetitividade. Os requisitos e procedimentos para o dimensionamento de paredes de concreto moldadas no local são contemplados na NBR 16055:2012. Dentre as abordagens consideradas nessa norma, ressalta-se a recomendação do uso de concreto autoadensável. No entanto, alguns pontos de interesse não são contemplados na regulamentação vigente como, por exemplo, a incorporação de ar na composição do concreto, bem como os limites de teores de ar incorporado. Assim, a presente pesquisa busca avaliar a influência da incorporação de ar no comportamento de concretos autoadensáveis destinados à execução de paredes de concreto moldadas no local. Para isso, foram produzidos dois traços de concreto autoadensável de classes de resistência C25 e C40 e, a partir de cada um deles, produzidos outros dois traços modificados com aditivo incorporador de ar, variando o teor de ar incorporado, em um total de seis misturas cujas propriedades foram avaliadas tanto no estado fresco quanto no estado endurecido. Além disso, foram moldadas paredes de concreto a fim de analisar a estrutura interna do material com aparelho de ultrassom e o acabamento superficial do elemento por meio de análise de imagens. No caso da classe C25, os resultados demonstram que a influência do teor de ar para os traços modificados foi tão significativa que os concretos produzidos se tornaram inadequados para uso estrutural. Por outro lado, apesar da elevada perda de resistência à compressão aos 28 dias, os dois traços produzidos com incorporação de ar no concreto de classe C40 se adequam às recomendações da NBR 16055:2012. Alia-se a isso o fato de que, quando da execução das paredes em laboratório, os concretos utilizados demonstraram grande capacidade de preenchimento, resultando em elementos com bom acabamento superficial e estrutura interna homogênea, mesmo em regiões de difícil acesso. Sendo assim, o presente trabalho propõe traços muito promissores para utilização em paredes de concreto moldadas no local. / The creation of federal programs aimed at the development of Brazilian infrastructure at the end of the last decade resulted in the heating of civil construction sector. In this context, the system of concrete walls cast in place stood out due to its high productivity and repetitiveness. The requirements and procedures for the design of concrete walls cast in place are contemplated in the NBR 16055:2012. Among the several approaches considered in this code, the recommendation for the use of self-compacting concrete is emphasized. However, some points of interest are not considered in the current regulations, such as the incorporation of air in the concrete composition, as well as the limits for entrained air contents. Thus, the present research seeks to evaluate the influence of air entrainment in the behaviour of self-compacting concrete for the construction of concrete walls cast in place. For this, two self-compacting concrete mixes design of strength classes C25 and C40 were produced and, from each of them, two other mixes modified with air entraining admixture were produced, varying the entrained air content, in a total of six mixtures whose properties were evaluated in both fresh and hardened state. In addition, concrete walls were cast to analyse the internal structure with ultrasonic equipment and the element surface finish by image analysis. In the case of class C25, the results demonstrate that the influence of the air content for the modified mixes was so great that the concretes produced became inappropriate for structural use. On the other hand, despite the high loss of compressive strength at 28 days, the two mixes produced with entrained air in the concrete of class C40 respect the recommendations of the NBR 16055:2012. In addition, when the walls were cast in laboratory, the concretes used showed great filling capacity, resulting in structures with good surface finish and homogeneous internal structure, even in difficult access regions. Therefore, the present work proposes very promising mixes for use in concrete walls cast in place.

A busca por novos materiais estruturais visa a melhoria da qualidade e desempenho das estruturas, impulsionando o desenvolvimento científico e tecnológico. O concreto auto-adensável surgiu da necessidade de se dispensar o difícil e oneroso trabalho de vibração do concreto, sendo definido como um material capaz de fluir dentro de uma fôrma, passando pelas armaduras e preenchendo a mesma, sem o uso de equipamentos de vibração. Esta pesquisa caracteriza-se como um estudo teórico-experimental da aderência aço-concreto, utilizando concreto do tipo auto-adensável, mediante ensaios monotônicos de flexão em vigas e de arrancamento seguindo o modelo padrão do Rilem-Ceb-Fip (1973). O estudo considerou como parâmetros fundamentais o tipo de concreto (auto-adensável e convencional), a resistência à compressão do concreto e os diâmetros das barras. Ainda, realizou-se um estudo com relação à variabilidade do concreto auto-adensável nos estados fresco e endurecido, constatando que este possui pequena variação. De posse dos resultados, buscou-se verificar a previsão das formulações empregadas na literatura e pelos principais códigos internacionais. De acordo com os resultados, o comportamento dos modelos de viga e de arrancamento para ambos os concretos foi similar, mostrando que o concreto auto-adensável possui características semelhantes ao concreto convencional, com as vantagens da trabalhabilidade no estado fresco. Quanto à análise numérica, os modelos desenvolvidos representaram de forma satisfatória o comportamento dos ensaios, e forneceram uma idéia do comportamento da tensão de aderência na sua interface. Com relação às formulações teóricas, verificou-se que tanto para o concreto convencional quanto para o auto-adensável, ocorre uma superestimativa da resistência de aderência. / The search for new structural materials aims the improvement of the structures performance, pushing the scientific and technological development. Self-compacting concrete (SCC) origin was due to the need to avoid the difficult and expensive process of concrete vibration. It has been defined as a material capable to flow inside a formwork, passing through the reinforcement and filling it completely, without using of any special equipment. This research is characterized as a theoretical and experimental study of steel-concrete bond, using SCC, through pull-out tests and beam tests standardized by Rilem-Ceb-Fip model. The study considered as main parameters the concrete type (ordinary concrete and SCC), its compressive strength and the steel bars diameters. Yet, a study of the variability of the properties of SCC, both in fresh and hardened state was also performed, showing that SCC is a very reliable material. The obtained results were compared with the ones established by the related bibliography and by the main international codes. According to the results, the behavior of the beams and pull-out tests were very similar for both concretes (SCC and OC), with the well know advantages for the SCC in fresh state. About the numerical approach, the developed models represented satisfactory the test behavior (beam and pull-out) and gave an idea of the bond stress behavior on the steel-concrete interface. About the theoretical formulations, it was verified the they usually overestimate the bond strength, for both types of concrete.

Submitted by SBI Biblioteca Digital (sbi.bibliotecadigital@puc-campinas.edu.br) on 2017-03-23T12:34:35Z No. of bitstreams: 1 RENAN SERRALVO CAMPOS.pdf: 14804663 bytes, checksum: d21aa4db3ac6aef546c6f30dc795691b (MD5) / Made available in DSpace on 2017-03-23T12:34:35Z (GMT). No. of bitstreams: 1 RENAN SERRALVO CAMPOS.pdf: 14804663 bytes, checksum: d21aa4db3ac6aef546c6f30dc795691b (MD5) Previous issue date: 2017-02-15 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior - CAPES / The high consumption of natural resources and energy, coupled with the significant amount of solid waste generated by the construction activities, makes this sector one of the most responsible for environmental impacts. The recycling of mineral fraction of construction and demolition wastes (CDW), in form of aggregates, is seen as an alternative to reduce these impacts. In this way, the aim of this study was to investigate the influence of recycled aggregates on the rheological, mechanical properties, physical indexes and surface finish of self-compacting concrete (SCC). For this, eight compositions of SCC were analyzed, using the Repette-Melo method and Alfred model, which is based on the concept of particle packing, for the mix design of concretes. Thus, two reference mixtures, composed only of natural aggregates, were established. The other mixtures used recycled fine and/or coarse aggregates instead of natural aggregates, at 20% by mass. The properties of fresh concrete were evaluated with respect to the flowability, viscosity (measure of the speed of flow), passing ability and segregation resistance. In hardened state, compressive strength, splitting tensile strength, static modulus of elasticity, water absorption by immersion and by capillarity, density and voids index were determined. In this study, rheological characterization of cement pastes and mortars destined to the production of SCC was also carried out using rotational rheometry. Rheological characterization of pastes, mortars and concretes is a suitable way to evaluate the behavior of these composites in fresh state, providing useful information that will aid in mix design process and quality control. Another property evaluated was the surface finish of concretes and mortars and the possible correlation of this with the rheological parameters. Through the results obtained it was possible to verify the loss of fluidity in mortars and concretes that used recycled aggregates. Regarding the mechanical properties, different behaviors were observed in concretes produced with recycled aggregates: those that were dosed by Repette-Melo method did not show a significant decrease of these properties; while the concretes dosed by the concept of particle packing exhibited a great reduction in mechanical properties. Concretes produced with recycled aggregates still showed an increase in water absorption by immersion, voids index and capillary absorption, when comparing to the respective reference mixtures. It was possible to observe a good linear correlation between surface finish and shear stress of mortars. In the case of concretes, the slump-flow was the property that best correlated with the quality of surface finish. / O elevado consumo de recursos naturais e energia, associado ao expressivo montante de res?duos s?lidos gerados pelas atividades de constru??o civil faz deste setor um dos maiores respons?veis por impactos ambientais. A reciclagem da fra??o mineral dos res?duos de constru??o e demoli??o (RCD), na forma de agregados, ? vista como uma alternativa para a diminui??o destes impactos. Isto posto, o objetivo deste estudo foi investigar a influ?ncia dos agregados reciclados nas propriedades reol?gicas, mec?nicas, ?ndices f?sicos e no acabamento superficial de concretos autoadens?veis (CAA). Para isso, foram analisadas oito composi??es de CAA, sendo empregados o m?todo de Repette-Melo e o modelo de Alfred, que tem por base o conceito de empacotamento de part?culas, para a dosagem dos concretos. Assim, foram estabelecidos dois tra?os de refer?ncia, compostos apenas por agregados naturais. As demais misturas utilizaram agregados mi?dos e/ou gra?dos reciclados em substitui??o aos agregados naturais, no teor de 20%, em massa. As propriedades do concreto fresco foram avaliadas no que diz respeito ao espalhamento, tempo de escoamento, habilidade passante e resist?ncia ? segrega??o. No estado endurecido foram determinadas as resist?ncias ? compress?o e ? tra??o por compress?o diametral, o m?dulo de elasticidade est?tico, a absor??o de ?gua por imers?o e por capilaridade, a massa espec?fica e o ?ndice de vazios. Neste estudo tamb?m foi realizada a caracteriza??o reol?gica das pastas de cimento e das argamassas destinadas ? produ??o dos CAA?s, por meio de reometria rotacional. A caracteriza??o reol?gica de pastas, argamassas e concretos ? uma forma adequada para avaliar o comportamento destes comp?sitos no estado fresco, podendo prover informa??es ?teis que auxiliar?o no processo de dosagem e controle de qualidade. Outra propriedade avaliada foi o acabamento superficial dos concretos e argamassas e a poss?vel correla??o deste com os par?metros reol?gicos. Atrav?s dos resultados obtidos foi poss?vel constatar perda de fluidez nas argamassas e concretos que empregaram agregados reciclados. No que tange as propriedades mec?nicas, foram observados comportamentos distintos nos concretos produzidos com agregados reciclados: ?queles dosados pelo m?todo de Repette-Melo n?o apresentaram decr?scimo significativo destas propriedades; enquanto que, os concretos dosados com base no conceito de empacotamento de part?culas exibiram grande redu??o nas propriedades mec?nicas avaliadas. Os concretos produzidos com agregados reciclados ainda apresentaram aumento na absor??o de ?gua por imers?o, no ?ndice de vazios e na absor??o capilar, quando comparados aos respectivos tra?os de refer?ncia. Foi poss?vel observar boa correla??o linear entre o acabamento superficial e a tens?o de cisalhamento das argamassas. No caso dos concretos, o di?metro de espalhamento foi a propriedade que melhor se correlacionou com a qualidade do acabamento superficial.

La thèse comprend 8 chapitres, 3 annexes, 191 figures, 43 tableaux, 13 photos et 181 références bibliographiques. Le corps du texte totalise 135 pages et les annexes 120 pages. Le chapitre 1 constitue lintroduction dans laquelle le contexte et les objectifs de la recherche sont définis. Le chapitre 2 présente létat de la question concernant les colonnes faites de tubes en acier remplis de béton (CFSHS columns - concrete filled steel hollow section columns). Leur comportement à température ordinaire et sous conditions dincendie est analysé ainsi que les paramètres significatifs qui guident leur comportement. Les propriétés du béton auto-plaçant sont aussi examinées dans ce chapitre. Le chapitre 3 est consacré à la présentation du code de calcul SAFIR développé à lUniversité de Liège pour la simulation du comportement structural à température ordinaire et sous conditions dincendie. Le logiciel a été utilisé pour simuler des résultats dessais expérimentaux présentés dans la littérature, et quelques calibrations sont proposées. En utilisant le modèle décrit au chapitre 3, la charge ultime à température ordinaire des colonnes CFSHS est calculée en considérant différents paramètres. On considère essentiellement des tubes de petites dimensions avec un pourcentage élevé darmatures ou contenant un autre profilé métallique. Ce type délément nest pas couvert par les Eurocodes. Des simulations ont été réalisées en vue dexaminer si les méthodes classiques de lEurocode 4 (EN1994-1-1) sont applicables à ce type de section et de voir quelle courbe de flambement est la plus appropriée. Le chapitre 5 décrit la recherche expérimentale réalisée à lUniversité de Liège relative au comportement au feu des colonnes CFSHS. Les essais expérimentaux ont été simulés par SAFIR, tant en ce qui concerne la répartition de la température à lintérieur des éléments que le comportement structural. Un des objectifs de ces calculs était de vérifier si les propriétés thermiques et mécaniques du béton auto-plaçant sont proches de celles du béton ordinaire. En vue de fournir aux bureaux détudes des outils pratiques de dimensionnement, une formule permettant dévaluer la résistance au feu des colonnes CFSHS est proposée dans le chapitre 6. Le domaine dapplicabilité a été étendu par rapport aux études précédentes. Une formule pour colonnes courtes de section carrée a dabord été établie sur base de simulations effectuées à laide de SAFIR en considérant linfluence des paramètres principaux. On examine ensuite si cette formule est aussi applicable à dautres types de profilés. La formule a été étendue en cas des colonnes élancées et à celui des colonnes chargées excentriquement. Létude expérimentale du chapitre 5 est limitée à des sections dassez petites dimensions (< 300 mm), essentiellement pour des problèmes de capacité de mise en charge des laboratoires. En vue de recueillir des informations pratiques pour des profilés de plus grandes dimensions pour lesquels on peut atteindre des résistances au feu de 2 heures et plus, des simulations numériques additionnelles sont présentées dans le chapitre 7, avec des résultats utilisables en bureau détudes. Le chapitre 8 donne les conclusions de létude et présente des suggestions pour des recherches futures.

Self-compacting concrete is an innovative construction material / its priority to normal vibrated concrete is that there is not any vibration requirement. Bond strength of reinforcement is one of the key factors that ensures the usefulness of any reinforced concrete structure. In this study, 6 full-scale concrete beams spliced at the mid-span were tested under two-point symmetrical loading. Test variables were bottom cover, side cover, free spacing between longitudinal reinforcement, lap-splice length and presence of transverse reinforcements within the lap-splice region. Specimen SC_22_44_88_800 had cover dimensions close to the code limits and had 36db lap splice length. This specimen showed flexural failure. Specimen SC_44_44_44_710 had 32db lap splice and cover dimensions greater than code minimums. This specimen showed yielding primarily. With the increasing loading, however, bond failure occurred with side splitting. ACI 408 descriptive equation for normal vibrated concrete predicted bar stresses of the unconfined specimens produced with self-compacting concrete acceptably well. The predicted values were lower than the measured values to be on the safe side. The error varied between 3.4% and 6.5%. All predictions of the ACI408 descriptive equation was higher than the measured bar stresses of the confined specimens produced with SCC. All the calculated values were unsafe. The error varied between 10.6% and 34.5%. Specimen SC_44_22_22_530_T4 with 24db lap splice length had side cover and spacing between bars 63.3% and 56.7% less than the ACI 318 limits. The calculated bar stress was 21.6% higher than the measured value. The main reason of the deviation was inadequate cover dimensions. In specimen SC_44_22_22_530_T6, number transverse reinforcement was increased to 6 stirrups to overcome the small cover and spacing problem. However, increased number of stirrups inside a small side and face cover caused weak plane and measured bar stress decreased.