As stated above, the effectiveness of FRP externally bonded on RC members is strongly related to the interface behaviour that depends on the mechanical and geometrical properties of materials. Thus an FRP-to-concrete bond strength model is the key to the accurate prediction of debonding failures in FRP-strengthened RC beams, including plate end debonding failures and shear crack-induced debonding failures, as well as intermediate flexural or flexural-shear crack-induced debonding failures (Teng et al., 2002).
Plenty of pull-out tests have been carried out by researchers during last years in order to experimentally investigate the influence of FRP and concrete mechanical properties on the bonding behaviour.
An overview of most common bond test techniques is reported in the first subsection. Then, the most significant results achieved by such bond tests are showed in the second one.
1.2 Overview of testing techniques
Yao et al. (2005) classified the existing test setups into the following types: (a) double-shear pull tests; (b) double-shear push tests; (c) single-shear pull tests; (d) single-shear push tests; and (e) beam (or bending) tests. These arrangements are based on the definition of the loading condition of the element and on the symmetry of the system (a double or single test refers to the contemporaneous loading of two or one FRP reinforcement applied on the specimen sides). Collectively, the first four test methods, may also be referred to as pull tests, as the plate is always directly pulled by a tensile force. Pull-pull (a) and push-pull (d) test setup are the most popular test methods. In end debonding failures in FRP flexural – strengthened RC beams with longitudinal laminates as well as debonding failures in FRP shear-strengthened RC beams with transverse laminates, the bond strength model developed from pull tests is directly applicable. Furthermore, in intermediate crack-induced debonding failures in FRP flexural – strengthened RC beams with longitudinal laminates, the stress state in the critical region of the beam is also closely similar to that of the concrete prism in a single-shear pull test. In order to extend the results of bond tests to various types of strengthening (flexural, shear, torsion), the pull-pull test setup (a and c) are probably the configuration giving the loading condition more similar to the actual one in RC elements, but it is also the most difficult to realize with a reliable setup. Furthermore the test setup (a), as well as (b) could lead to
underestimate the bond strength due to the influence of detailing (Blontrock et al., 2002; Ceroni et al., 2008). On the contrary, the push-pull test (d), where the concrete is partially in compression, is easier to be realized but it gives comparable predictions of the bond strength, if the pushing force is applied on the concrete block sufficiently far from the external reinforcement (Yao et al., 2005). Indeed, if the compressed area of concrete is too extended, the volume of material involved in the debonding failure can be lower and, thus, the related fracture energy decreases.
Asymmetrical schemes (c and d) are, in general, preferable to the symmetrical ones (a and b) mainly because the latter are more influenced by the alignment detailing of the two strengthened sides. Moreover the specimen symmetry is however lost when the debonding starts on one side and prevents from following correctly the post-peak behaviour.
As concerns the asymmetrical push-pull test setups (d), they are commonly realized by positioning a single concrete block in a stiff steel frame with an upper plate compressing the specimen, while the end of the FRP reinforcement, glued on one side, is clamped in the grips of a tensile machine (Nigro et al., 2008; Savoia et al. 2009; Ceroni and Pecce, 2010). An alternative configuration can be also realized by fixing the end of the concrete block, placed horizontally, and applying tension to the FRP reinforcement with an hydraulic jack (Yao et al., 2005; Mazzotti et al., 2009). As concerns the asymmetrical pull-pull test setups (c), they can be realizes by installing metallic threaded rods inside the concrete specimen which can be clamped in the testing machine grips (Bilotta et al. 2011).
Asymmetrical single-shear test setup (c or d) appear to be promising candidate as a standard setup for determining the FRP-to-concrete bond strength; hence they were adopted to perform the tests showed in the following section.