Shear transfer across cracks in steel fibre reinforced concrete

Existing models for the shear strength of steel fibre reinforced concrete (SFRC) in general, and for the transfer of shear stresses across cracks, typically assume that the shear strength of plain concrete is increased by a so-called "fibre contribution". This is usually assumed to be independent of the roughness of the cracks. While being relatively simple to implement in practice, such approaches fail to capture the physical-mechanical behaviour of the transfer of shear stresses across cracks. Hence, in order to exploit the full potential of SFRC in shear, rational and mechanically consistent models capable of describing this behaviour are essential. This paper presents such a model, which relates the shear and normal stresses transferred across cracks in SFRC to the crack opening and slip displacements. The novelty of the proposed model consists in the fact that it not only accounts for fibre stresses bridging the crack, as done in many previous models, but also for the aggregate interlocking behaviour along the faces of the crack, and the interaction of these effects. Regardless of the aggregate interlock relationship used, the proposed model predicts that except at large crack openings (in the order of half the size of the maximum aggregate particle), only a minor portion of the applied shear stresses are resisted directly by inclined fibres crossing the crack, even for high fibre contents. Rather, it is shown herein that shear stresses are primarily transmitted via the interlocking of aggregates along the crack faces. Nonetheless, the addition of fibres to concrete is highly beneficial to the transfer of shear across cracks since fibre stresses normal to the crack plane are equilibrated by compressive stresses on the crack faces. These compressive stresses greatly enhance aggregate interlock. A comparison of predictions by the proposed model with a wide range of reported experimental data shows a good correlation. However, the results also indicate that existing relationships for aggregate interlock in plain concrete are highly variable, and their applicability to SFRC needs to be carefully examined.