This work proposes a methodology to translate the experimental pullout results of hooked-end steel fibers from cementitious matrix to a multiscale model with a discrete and explicit representation of steel fibers. The coupling scheme of non-matching meshes used to describe the fiber–matrix interaction is adapted to represent separately and explicitly the anchorage mechanism of hooked-end steel fibers. The difference of the experimental results of pullout tests of hooked-end steel fibers (DramixⓇ 80/60 BN) and straight fibers obtained by cutting the hooked-ends of the fibers is used as input parameters to describe the anchorage mechanism. Pullout tests with one and four fibers aligned to the applied loading and perpendicular to the crack plane are carried out to investigate the effect of instability on the experimental results. The methodology is also coupled with an analytical model for predicting the effect of fibers inclined to the crack surface. Initially, the effect to represent explicitly the anchorage of hooked-end steel fibers in multiscale models is assessed through the simulation of pullout tests. Then, the numerical model is applied to obtain the post-cracking parameters of steel fiber reinforced cementitious composites (SFRCC) through the simulation of flexural three-point-bending test (3-PBT) according to EN 14651, considering the same type of hooked-end steel fibers and a similar matrix strength. Thus, the bond–slip model derived from pullout tests is used together with the amount and orientation of steel fibers obtained from the inductive test carried out. The results demonstrate that the integrated experimental and multiscale methodology may be very useful to link the physical and numerical responses of SFRCC with hooked-end steel fibers, and the numerical tool obtained can contribute for better understanding the failure processes for this type of composite.
Read full abstract