PSI - Issue 41

P.M.D. Carvalho et al. / Procedia Structural Integrity 41 (2022) 24–35 Carvalho et al. / Structural Integrity Procedia 00 (2019) 000 – 000

31

8

t P2 . The percentile P m increase, over t P2 =1 mm, for t P2 =2, 3 and 4 mm, was in this order of 81.2 %, 197.8% and 403.7% (2015) and 110.9%, 227.1% and 358.6% (7752). According to previous stress analyses to identical joints (Carneiro and Campilho 2017), this variation is mainly due to the peak  y stress reduction at the stress initiation zone. Between the two tested adhesives, the 7752 performs best, although it is less strong than the 2015 (Table 1). The P m improvement of the joints bonded with the 7752, compared with the 2015, varies between a minimum of 78.5% ( t P2 =4 mm) and a maximum of 128.1% ( t P2 =2 mm). These results show that, under peel-predominant loading conditions, in which stresses are concentrated in small zones, the flexibility and ductility of the adhesive is more important than its strength. Actually, flexibility enhances the load-transmitting portion of the adhesive, while ductility enables plasticization at the loci of peak stresses, thus resulting in higher P m . The CZM prediction were accurate, especially for the 2015, but also acceptable for the 7752. The highest relative errors over the average experimental P m were +4.7% for the 2015 ( t P2 =3 mm) and -7.2% for the 7752 ( t P2 =4 mm). The more significant deviation for the 7752 is considered to be due to the marked ductility of this adhesive. Under these conditions, a triangular law may present minor variations to the expected behavior (Campilho et al. 2013). Nonetheless, this law is adequate and provides good qualitative results between different geometric conditions. Thus, the CZM law is validates, leading to the numerical dual-adhesive analysis that follows. 3.2. Numerical evaluation of dual-adhesives A CZM comparison study was performed to assess the advantages of using the dual-adhesives technique. The numerical failure modes are initially described. The following stress analysis is divided into the evaluation of peel and shear stresses in the adhesive layer. The dual-adhesive technique is assessed by comparison of P m and U between the proposed technique and SAJ. 3.2.1. Failure modes The numerical analyses revealed four different types of failure initiation, which then propagates to the entire adhesive layer leading to failure. Table 2 shows the overall distribution of failure modes for the different joint configurations, according to the nomenclature defined in the following discussion. In SAJ, failure starts cohesively in the adhesive layer either at the edge near the curvature (A), for smaller t P2 , or at the adhesive layer’s free edge (B) for bigger t P2 . For the dual-adhesive configurations, most failures originate at the transition region between adhesives. Identically to the SAJ, when the value of t P2 is smaller, failure onset typically takes place at the transition region closest to the curvature (C), while for bigger t P2 it initiates at the free edge (D). It is important to notice that, for the case of the DA-1/3 configuration, failure mode A takes place for t P2 =1 mm. On all other dual-adhesive configurations, failures initiate at the transition regions (either failures C or D).

Table 2. Distribution of failure modes for the tested joint configurations.

t P2 (mm)

Setup

1

2

3

4

Araldite ® 2015 Sikaforce ® 7752

A A C A

A A C C

B A D C

B B D D

DA 1/8 DA 1/3

3.2.2. Stress analysis Peel and shear stresses in the adhesive layer were also analyzed for both SAJ and DAJ using the ABAQUS ® software during the elastic loading and for a prescribed displacement of 0.015 mm. 3.2.2.1. Peel stresses Fig. 6 shows  y stresses for the different joint configurations with t P2 =1 (a) and t P2 =4 mm (b), as representative of the limit geometrical conditions tested. In this figure, the SAJ 7752 and DAJ 1/3 curves are overlapped.