Composite fracture behavior of concrete

In order to deduce systematically the main causes of size-dependence of experimentally observed fracture parameters in cement-based materials, concrete was simulated as a multi-phase composite material. Each individual phase in the composite was modeled to behave in accordance with the conventional single-phase fracture models such as linear elastic fracture model (LEFM) or cohesive crack model. As a result of applying the LEFM concept to each individual phase in the composite material, fracture parameters were found to be dependent on specimen size and stacking sequence. Although the size effect of fracture energy (G{dollar}sb{lcub}rm f{rcub}){dollar} was observed in the composite model, the theoretical G{dollar}sb{lcub}rm f{rcub}{dollar} values were found to underestimate the experimental results. Accordingly, the concept of a cohesive crack model was applied in the composite fracture analysis. As a result, the occurrence of significant multiple-cracks was observed during progressive fracture mode. By comparing with the use of LEFM, the theoretical fracture behavior based on the cohesive crack concept was in good agreement with the experimental results.; In terms of the fracture energy, it was found that the generation of spurious energy due to a snap-back instability is the main contributor to the increase of fracture energy value. The magnitude of the spurious energy value depends on the stacking sequences, fracture mode, beam size, notch-depth ratio, and material parameters of each individual phase composing the composite material.; The concept used in the multi-layer system was further extended for two different applications. One is to model the fracture behavior of a concrete beam reinforced with laminated composite. The theoretical result illustrated a significant increase of strength and stiffness in comparison with the behavior of pure concrete. For the other application, the effect of temperature on the crack formation in concrete pavements was studied. The result showed that the determination of sawcut depth is very sensitive to the existence of adjacent cracks in the pavement, temperature differential, pavement thickness, stiffness ratio between slab and subbase materials, and the timing of introducing a sawcut. It was further observed from the theoretical model that the timing of sawcut is, in most cases, more important than the introduction of the sawcut to control random crack formation.