FE modeling of the interfacial behaviour of composite concrete members D. Dias-da-Costa a,b,⇑ , J. Alfaiate c , E.N.B.S. Júlio c a Civil Eng. Dept., University of Coimbra, Rua Luís Reis Santos, 3030–788 Coimbra, Portugal b INESC Coimbra, Rua Antero de Quental 199, 3000-033 Coimbra, Portugal c ICIST, Civil Eng. Dept., Instituto Superior Técnico, Av. Rovisco Pais, 1049–001 Lisboa, Portugal article info Article history: Received 4 November 2010 Received in revised form 1 May 2011 Accepted 13 June 2011 Available online 16 July 2011 Keywords: FEM model Interfacial behaviour Composite concrete bridges Push-off tests Shear joints Added reinforcement abstract The accurate prediction of the shear strength between concrete layers is important in: (i) new concrete bridges with precast and cast in-situ members as well as in (ii) existing concrete bridges strengthened with a concrete overlay. The monolithic behaviour of these composite members depends only on the interface strength. There are several experimental studies dealing with the influence of different param- eters on the bond strength of the concrete-to-concrete interfaces. However, an adequate numerical model concerning the behaviour of these structural interfaces is still missing. In this paper such a numerical model is presented, which is calibrated with experimental data. Results of push-off tests performed to evaluate the longitudinal shear strength between two concrete layers are considered. Different ratios of steel reinforcement crossing the interface were adopted. The comprehen- sive identification of the influence of the following parameters is achieved: elastic shear stiffness; inter- nal friction angle; dilatancy angle; cohesion; fracture energy; and bond–slip relation between steel connectors and concrete. The most relevant conclusions are presented. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction In the last three decades, Portugal has experienced an over- whelming increase on the existing highways network, starting from just 20 km to more than 1500 km presently. As a consequence, hundreds of bridges over these highways had to be built. The most commonly adopted bridge type consists of simply supported precast RC beams with cast-in-place concrete decks (see Fig. 1). In this type of composite structures, the behaviour of concrete-to-concrete interfaces plays an important role. Existing design expressions already deal with several relevant factors in this scope. For instance, Eurocode 2 [1] and CEB-FIB Mod- el Code 1990 [2] both consider the following parameters: concrete tensile strength; cohesion and friction coefficients for the inter- face; existence and orientation of steel reinforcement; and normal stress at the interface. However, the characterization of the rough- ness of the interface still remains qualitative [3]. Moreover, other influencing parameters are still missing. In the last years, the authors have been carrying out several experimental studies on this subject aiming to identify the role of the following parameters: (i) roughness of the substrate surface [4,5]; (ii) use of epoxy-based bonding agents [6]; (iii) compressive strength of added concrete [7]; and (iv) added reinforcement crossing the interface [8]. In this manuscript a comprehensive identification of the effect of each of the following parameters is aimed: elastic shear stiff- ness; internal friction angle; dilatancy angle; cohesion; fracture energy; and bond–slip relation between steel connectors and con- crete. With this goal, a numerical approach was followed having the following main steps: (i) to build a robust finite element model of a concrete-to-concrete interface; (ii) to calibrate this model using experimental results; and (iii) to conduct a parametric study to identify the effect of each of the most relevant parameters affecting the shear strength between two concrete layers. For cal- ibration purposes, results of push-off tests (see Fig. 2) conducted assuming different ratios of steel reinforcement crossing the inter- face were adopted. The paper is organised as follows: the experimental set-up is presented in Section 2. The constitutive models are briefly described in Section 3. The computation model is addressed in Section 4, being the importance of the accurate definition of the boundary conditions emphasised. In Section 5, the following material parameters are studied: elastic stiffness; internal friction angle; dilatancy angle; cohesion; fracture energy; and bond–slip relation between connectors and concrete. A strategy is established in order to clearly identify the role of each parameter both qualita- tively and quantitatively. The resulting model is presented in Section 6. Finally, conclusions are summarised in Section 7. 0950-0618/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2011.06.015 ⇑ Corresponding author. Tel./fax: +351 239 797 256/60. E-mail addresses: dcosta@dec.uc.pt (D. Dias-da-Costa), alfaiate@civil.ist.utl.pt (J. Alfaiate), ejulio@civil.ist.utl.pt (E.N.B.S. Júlio). Construction and Building Materials 26 (2012) 233–243 Contents lists available at ScienceDirect Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat