Steel-fiber reinforced concreteperformance-based design and mechanical characterization in static and dynamic regime

Resumen

Concrete, reinforced and prestressed, is the building material par excellence since the last century, whose evolution over the previous decades has paralleled the development of pro- cess technology and structural requirements. From the material, the appearance of high– performance concrete, such as self–compacting concrete, and the possibility of incorporating new types of reinforcement in its matrix, such as steel fibers, represents a set of advantages over primal concretes. Consequently, the characteristics of these materials require a reflec- tion on design procedures, which obviously cannot be the same as those used long ago. Nor is its mechanical behavior in a fluid state and a solid–state the same as that of the concrete of yesteryear. Hence, it is relevant to characterize them according to the load and strain rates they may experience. This doctoral thesis addresses the full study of concrete through the concept of performance design. It means that in the project of a concrete structure, everything is related, and everything must be taken into account. That is, the mechanical response in a hardened state is related to the behavior of its flow, which, in turn, derives from the design phase, the boundary conditions, and the constructive processes. The first topic treated in this thesis is the development of a design methodology for self–compacting steel–fiber reinforced concrete (SCSFRC) based on performance. Using a micromechanical model of a bi–phase suspension and from the content and geometric characteristics of the fiber, it is possible to estimate the value of the effective plastic viscosity of SCSFRC from the measurement of the plastic viscosity of the cement paste, which is determined by the target compressive strength of the concrete to be designed. The method has been analytically programmed and provides design charts and data files with possible concrete compositions. The design process contains no loops and eliminates trial and error testing by always providing concretes with self–compacting characteristics. Next, the variability in the measurement of plastic viscosity in a set of cement pastes of different composition has been analyzed. Measurements have been carried out through a rotational rheometer with double cone–plate sensor and two types of capillary viscometers, the Marsh funnel, commonly used in this type of suspensions, and the Cannon–Fenske vis- cometer, more common in other types of fluids. The plastic viscosity values calculated from several models applied to the Marsh funnel existing in the scientific literature have also been analyzed. The results indicate that the Cannon–Fenske viscometer shows a better approxi- mation to the rotational rheometer measurements so that its use can be recommended in the measurement of the plastic viscosity of cement pastes in the proposed design methodology for SCSFRC. Then the influence of the inclusion of steel fiber in the compressive behavior of concrete through a statistical–mathematical method such as that Response Surfaces methodology (RSM) has been studied. A bibliographical search has been carried out by selecting a set of uniaxial compressive experimental tests to which this methodology has been applied. The results show the energy absorption and ductility capacity of steel–fiber reinforced concrete (SFRC) during its compressive deformation. This aptitude of the material has been widely described by scientific literature; however, it is not taken into account in the codes of analysis and structural design, which is a waste of material capacity. With these premises, a model has been developed in a technological format that describes the uniaxial compressive behav- ior of SFRC. The model has a softening branch obtained so that the energy consumption after the peak load is equal to the value of the absorbed energy calculated in the softening branch according to the database. It defines a new parameter called non–dimensional com- pressive residual strength, which is equivalent to the dimensionless strength corresponding to a deformation equal to three times the strain under maximum stress. This model allows the non–linear calculation of any structural element in uniaxial compression. In this context, the classification proposed by the next Eurocode 2 draft for SFRC, which considers the residual flexural strength for different values of crack mouth opening displace- ment, has been revised. This classification does not take into account the contribution of the zone in compression of the element, which, as has been seen, is capable of absorbing energy during deformation and providing residual resistance. To obtain the complete flex- ural response of the material it is necessary to incorporate the capacity of the compressed zone. Following the same procedure, a set of results from experimental three–point bending tests has been compiled to which RSM has been applied. The results indicate that both compressive strength and fiber content are the factors that most influence the different flex- ural strengths. Also, as the value of crack opening increases, the geometric parameters of the fiber begin to gain importance, which is in accordance with the scientific literature. A new equation based on the law of effect size of Baˇzant has also been deduced that allows the conversion of the compressive strength of cubic specimens of any size to the compressive strength of the standard cylinder. Applying the concept of residual compressive strength and using the RSM, a relationship is established between residual strength values for different crack openings and the residual compressive strength. In this way, the flexural behavior of SFRC is completely characterized for each strength class defined by Eurocode 2. Finally, an experimental study of the dynamic mixed–mode of fracture of SCSFRC is presented. Being mixed–mode a fragile mode of fracture, the inclusion of steel fibers is particularly appropriate. Six orders of magnitude have been covered in the loading rate, from the quasi–static regime to impact conditions. Results suggest that the crack propagation may be stable or unstable depending on the energy stored in the specimen and that the maximum load value increases as the loading rate and the fiber content increases. Post–break analysis of the specimens shows that most fibers fail by pull–out and that the breakage of the aggregates is predominantly transgranular, appearing intergranular breaks at low loading rates, regardless of the fiber content of the concrete. The impact tests have been recorded with a high–speed camera and, from the visual identification of the cracking, its propagation velocity has been estimated, reaching values that are within the ranges collected by the scientific literature.