A fibre-based frame element with explicit consideration of bond-slip effects

Sousa, Romain Correia, António Saraiva Esteves Pacheco De Almeida, João [UCL] Pinho, Rui

(eng) Reinforced concrete (RC) frames subjected to seismic loading often depict localized member-end deformations due to strain penetration effects between adjacent members, such as beam-column and column-footing joints. Past experimental programs indicate that the bond-slip deformations occurring at the interface between the reinforcement and the surrounding concrete can contribute up to 40% of the lateral deformation of the RC members. The employment of advanced bond-slip models within detailed finite element formulations, capable of simulating continuous domains with highly discretized meshes, has witnessed great advances over the recent years with encouraging results. Nonetheless, this modelling approach is computationally heavy and hence inapplicable for practical seismic (nonlinear) analysis of structures. Alternatively, the use of beam-column elements with lumped or distributed plasticity is a more computationally efficient and engineering-friendly modelling approach. Unfortunately, the elements of this type available in conventional numerical packages did not yet consider an explicit simulation of the interface between the reinforcing bars and the surrounding concrete along their embedment length. The present study aimed at overcoming the foregoing limitation by developing an explicit bond-slip model applicable to general fibre-based beam-column elements. Using a state-of-the-art bond-slip constitutive model, the current paper introduces a zero-length element that computes the localized member-end deformations accounting for the bond-slip response at each reinforcing bar of a given RC section. Along with the material properties and anchorage conditions, the proposed nonlinear model also accounts for cyclic degradation and rebar yielding effects. Validation studies conducted with the proposed numerical formulation reveal a good agreement with past experimental tests, evidencing an important stability and accuracy at the expense of an acceptable additional computational effort.