The disturbed fracture process zone theory for the assessment of the asymptotic fracture energy of concrete

Abstract

The fracture energy constitutes an important input parameter for non-linear finite element analysis of concrete structures. Characterization of such parameter based on standard RILEM three-point bending test or wedge splitting test is known to depend on the size of the specimen. For the assessment example of large hydraulic structures using non-linear finite elements, it is important to develop an experimental protocol for characterizing a size-independent fracture energy, or asymptotic fracture energy representative of the size of these structures.Characterization of the asymptotic fracture energy requires specimens with very large sizes and thus represents an important experimental challenge. This work aims to develop a new experimental protocol based on wedge splitting test on specimen with moderate size and using digital image correlation technique.The new protocol is based on the disturbed fracture process zone (DFPZ) theory developed in this work, as an interpretation theory for the boundary effect theory of Duan et al. [8]. It uses the same hypothesis of bilinear distribution of the local fracture energy as in the simplified boundary effect method (SBEM) (Abdalla and Karihaloo [1], Karihaloo et al. [21]) but requires only one set of specimens with a specific notch over depth (a/D) ratio rather than two sets of specimens with two different a/D ratios.In order to validate the developed approach and theory, an experimental programme of 14 concrete specimens is conducted. Two concrete mixtures are considered with two maximum aggregate sizes: 38 mm and 76 mm, representative of hydraulic structures.Assessment of the fracture energy using the new protocol and the protocol based on SBEM resulted in similar values for both mixtures and thus validates the new developed theory.