Interpretation of aggregate volume fraction effects on fracture behavior of concrete Burcu Akcay a,⇑ , Ayda Safak Agar-Ozbek b , Fikret Bayramov c , Hakan Nuri Atahan c , Cengiz Sengul c , Mehmet Ali Tasdemir c a Civil Engineering Department, University of Kocaeli, 41040 Kocaeli, Turkey b Department of Civil Engineering and Geosciences, Delft University of Technology, Delft, The Netherlands c Civil Engineering Faculty, Istanbul Technical University, 34469 Istanbul, Turkey article info Article history: Received 14 May 2011 Received in revised form 25 August 2011 Accepted 29 August 2011 Available online 4 November 2011 Keywords: Aggregate volume fraction Notch depth Fracture energy Meso-mechanical modeling abstract Fracture process zone has a strong effect on fracture energy particularly for concretes with coarse aggre- gates. In order to demonstrate this effect, four groups of concrete mixtures with various aggregate vol- ume fractions were designed. Three point bending tests were applied on beam specimens with a single size but different notch size to obtain the boundary effect model based fracture energy. The results were evaluated to determine the effect of both aggregate volume fraction and notch size on fracture energy of concretes. In addition, meso-mechanical relations based on toughening mechanisms were employed in determining the fracture energy of concrete. The fracture energies obtained from the meso-mechanical approach and the boundary effect model were compared and found to be in good agreement. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction Concrete as a composite material consists of aggregate, cement paste and the interfacial transition zone (ITZ) between them. Since aggregates occupy the considerable volume of concrete, they have significant effects on various properties of concrete. The type, shape, size distribution and volume fraction of aggregate play an important role on fresh, hardening and hardened properties of concrete, as well as its fracture behavior. There are some studies specifically focused on the aggregate volume fraction. Feng et al., for instance, found that with increasing coarse aggregate content (from paste to concrete with 75 volume% of aggregate) both the critical stress intensity factor (SIF) and the fracture energy increase [1]. Chen and Liu reported that in low strength concrete with increasing aggregate volume from 40% to 80%, both the critical SIF and the fracture energy increase, while the maximum values achieved at 60 volume% of aggregate in high strength concrete [2]. On the other hand, Amparano et al. stated that since coarse aggregates are used almost always with fine aggregates, the total aggregate content should be a testing parameter [3]. They showed that with increasing total aggregate content (from 45% to 75%) fracture energy varied within 25%, and there was not any critical volume fraction which gives the maximum fracture energy. They found that fracture energy decreased with an increasing aggregate content and then started to increase after its minimum value at 65 volume% of aggregate. They also noted that with increasing total aggregate content the fracture process zone (FPZ) decreases and they explained this by the change in coarseness of grain structures [3]. Roziere et al. [4] found that increasing the volume of paste resulted in slight decrease of FPZ width and fracture toughness in self-compacting concrete. They also concluded from the fracture and acoustic emission analyses that self-compacting concretes behave more brittle with increasing volume of paste and, therefore, decreasing aggregate content [4]. Fracture mechanics based approaches have been proposed in order to determine the fracture parameters of concrete [5–12]. In very large structures, the effect of size can be neglected because of the fact that a large specimen with a crack tip away from bound- aries obeys Linear Elastic Fracture Mechanics (LEFM) and shows no size effect or crack length or ligament dependence in its fracture behavior. The size dependence of fracture energy can be explained by the existence of non-uniform fracture energy dissipation along the crack growth path. This non-uniform energy dissipation dem- onstrates the influence of specimen boundary and the FPZ, defined as the inelastic zone around the crack tip where the toughening mechanisms are operative [13–15]. The existence of the FPZ and the influence of the specimen boundary are among the underlying reasons of the size dependence of fracture energy. The boundary effect model, which is based on this influence, can be utilized in determining the size-independent specific fracture energy by using the experimental results obtained from single size specimens that differ in their notch characteristics [12]. The toughening mechanisms, which are operative at the meso- level themselves, contribute to the macroscopic non-linear stress 0950-0618/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2011.08.080 ⇑ Corresponding author. Tel.: +90 262 3033265; fax: +90 262 3033003. E-mail address: burcu.akcay@kocaeli.edu.tr (B. Akcay). Construction and Building Materials 28 (2012) 437–443 Contents lists available at SciVerse ScienceDirect Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat