PSI - Issue 47

Francesco Ascione et al. / Procedia Structural Integrity 47 (2023) 826–841 Author name / Structural Integrity Procedia 00 (2019) 000–000

828

3

mode II of two commercial epoxy resins, suitable for structural applications. Two ageing conditions were considered: immersion in sea water and tap water at the temperature of 30°C for at least one year. The whole experimental program involved the realisation and testing of one hundred and seventy-four specimens. The experimental procedure and the relative results are collected in the following three subsections. In the first one, the End Notch Failure test (ENF) and the Compliance-Based Beam Method (CBBM) in order to evaluate the fracture energy are presented; in the second subsection, the influence of the absorption phenomenon is studied; while, in the third subsection, the results relating to the influence of the desorption phenomenon are reported and commented. 2.1 Experimental procedure In order to evaluate the fracture energy in Mode II, the End Notch Failure Test (ENF), considered by current standards as the reference test, was adopted. The SLJ pieces, suitable for the ENF test, consist of two glass fiber reinforced pultruded profiles (GFRP) plates bonded with the two structural epoxy adhesives adopted: SikaDur 30 by Sika [32] and Araldite AV 5308 by Huntsman [Araldite]. The GFRP plates were 10 mm thick (t), 25 mm wide (b) and 340 mm long ( � ) as depicted in Figure 1a. A pre-crack 80 mm long ( � ) in the adhesive layer (Figure 1a) was assured in order to have a stable crack growth as reported in [34]. The Young Modulus of the plates along the fiber direction as well as in the transversal direction are equal to 24 GPa and 7 GPa, respectively, while the shear modulus was equal to 3 GPa. For a detailed description of the bonding step, the specimen curing, the hygrothermal conditioning, the absorption behaviour and of all the geometrical dimensions refer to paper [Ascione et al. 2021]. A schematic representation of the ENF test is shown in Figure 1b. The evaluation of the fracture data is based on the global force-displacement measurements. The sample was loaded with a force, P, normal to the bondline, applied at the mid-span position, in the displacement control at a rate of 0.25 mm /minute. The vertical displacement was measured by one transducer sensor placed under the specimen at the same cross section where the load was applied (Figure 1c). The joint was loaded continuously until the stable propagation of the fracture was observed.

a) c) Fig. 1. Test ENF: a) Specimen geometry (in mm); b) schematic representation of the deformed shape of the specimen; c) photo of experimental set-up. Representative experimental P- δ curves of the ENF specimens for both epoxy resins used are reported in Figure 2. Three different values of the load P are indicated in Figure 2: �� , ��� and � . The load �� corresponds to the loss of linearity (nl stands for non-linear) in the P- δ curve, ��� is the maximum value of the applied load P, while � is the ultimate value of the load P corresponding to the stop of the crack evolution (crack moves from an unstable state to stable one, the load restarts to increase). From Figure 2, it can be seen that the behaviour between SikaDur 30 and Araldite is quite different. In the first case, the adhesive shows a linear elastic behaviour up to failure followed by a vertical drop (brittle failure), while in the second case the presence of a post-elastic zone leads to a ductile behaviour. b)