PSI - Issue 37

Almudena Majano-Majano et al. / Procedia Structural Integrity 37 (2022) 492–499 Almudena Majano-Majano et al. / Structural Integrity Procedia 00 (2019) 000 – 000

497

6

The R -curves obtained from all the DCB tests are presented in Fig. 3 (right). The initial rising domain is characterized by the formation of the fracture process zone (FPZ). It is followed by a plateau tendency where a self similar crack growth takes place and so defines the value of critical strain energy release rate ( G Ic ), i.e., the material’s toughness to crack-growth.

Fig. 3. Load versus displacement curves (left) and R-curves (right) from DCB test on eucalyptus bonded joints.

As can be seen, most of the curves showed plateaus for considerable crack extent, which means that the FPZ was fully developed. The value of G Ic over the horizontal asymptote in each curve was calculated. A mean G Ic value of 0.73 N/mm was obtained from all the 1C-PUR bonded eucalyptus specimens tested. This value is slightly lower than that obtained for clear Eucalyptus globulus wood (0.77 N/mm) in previous research by the authors (Majano-Majano et al. (2019)), but both values remain in the same range. Comparing the mean strain energy release rate at maximum loads ( G I,Pmax ) between the two studies, the presence of the adhesive bondline again leads to values with approximately the same difference in relation to clear eucalyptus (0.77 N/mm for bonded joints vs. 0.80 N/mm for clear wood). As the difference between G Ic and G I,Pmax is minimal, the latter can be taken as a practical measure of the critical strain energy release rate in mode I. This G Ic value of 1C-PUR bonded eucalyptus specimens is considerably higher compared to DCB bonded joints using other species as adherents. Specifically, Veigel et al. (2012) reported an average fracture energy of 0.241 N/mm for spruce wood bonded using 1C-PUR (Purbond HB S309) with five different spreading quantities applying the direct compliance method (with crack propagation measurements) according to Gagliano and Frazier (2001). Also applying the direct compliance method, Harm (2006) reported an average specific fracture energy value of 0.44 N/mm for spruce specimens using fibre-reinforced PUR. Comparing investigations using the same data reduction method as the one applied here (CBBM), eucalyptus bonded joints also showed higher G Ic values than those resulted from DCB bonded joints using Pinus pinaster Ait. and epoxy adhesive (Araldite ® 2015) in Xavier et al. (2011) and Silva et al. (2013). Particularly, a mean G Ic value of 0.34 N/mm was obtained by the former, while a range between approximately 0.5 and 0.55 N/mm was reported by Silva et al. (2013), although the latter with few number of tests. A mean G Ic value of 0.31 N/mm was obtained from DCB specimens of Pinus pinaster Ait solid wood applying the same CBBM scheme in Xavier et al. (2014), which also confirms the small difference compared to glued joints. As mentioned above, t o the best of the author’s knowledge, there are no studies on wood bonded joints using 1C PUR adhesive and the CBBM as a data reduction scheme. 3.2. Cohesive law Fig. 4 (left) shows a representative experimental curve for the G I evolution correlated with the crack tip opening displacements measured by DIC during testing and the corresponding logistic fit according to Eq. (9). The cohesive laws of all specimen obtained after derivation (Eq. (1)) are plotted together in Fig. 4 (right). The average value of maximum traction (σ I,u ) was approximately 10.2 MPa.