PSI - Issue 47

L.A.R. Gomes et al. / Procedia Structural Integrity 47 (2023) 94–101 Gomes et al. / Structural Integrity Procedia 00 (2019) 000 – 000

99

6

that of the AV138, reducing the peak stresses on 38.5% for L O =12.5 mm and 44.1% for L O =50 mm. Thus, it is noticeable that the efficiency of the load transfer of 7752 adhesive is higher than AV138, benefiting small L O , since the corresponding stress distributions are more uniform.

10

10

8

8

6

6

4

4

 y /  avg

 y /  avg

2

2

0

0

-2

-2

0

0.2

0.4

0.6

0.8

1

0

0.2

0.4

0.6

0.8

1

x / L O 12.5 25 50

x / L O 12.5 25 50

a)

b)

Fig. 3.  y /  avg stress distributions in the adhesive mid-thickness for the DLJ with AV138 (a) and 7752 (b).

Fig. 4 shows the τ xy / τ avg stress plots as a function of L O for both adhesives: AV138 (a) and 7752 (b). For both adhesives, τ xy peak stresses are located at the overlap edges, and gradually reduce until reaching the joint centre, as found in the σ y / τ avg stress curves. Although this behaviour is expected on account of the widely known data produced for static loads (de Sousa et al. 2017), it relates to a different phenomenon, namely the shear-lag effect (Fernandes et al. 2015). For the curves shape, L O has a relevant effect, since lower L O lead to small stress gradients, which increase for higher L O . For the AV138, peak τ xy / τ avg stresses increased from 4.20 to 6.84 between the two limiting L O . For the 7752, the corresponding peak quantities are 1.07 and 3.01, thus showing a smaller aggravation of peak τ xy / τ avg stresses at the overlap edges, which is related to the adhesive’s compliance. In this case, the shear-lag effects across the adhesive layer are directly related with this variation. Generally, when dealing with brittle adhesives, these high peak stresses cause a reduction of the P m / L O ratio. Comparing τ xy / τ avg peak stresses between both adhesives, a major reduction for this parameter is found for the 7752, over the AV138: 22.2% ( L O =12.5 mm) and 57.1% ( L O =50 mm). The 7752 behaviour leans to improved efficiency in the load transfer and provide higher strength. Despite this fact, comparing with the AV138, P m can be limited by the smaller bulk strength.

8

8

6

6

4

4

 xy /  avg

 xy /  avg

2

2

0

0

0

0.2

0.4

0.6

0.8

1

0

0.2

0.4

0.6

0.8

1

x / L O 12.5 25 50

x / L O 12.5 25 50

a)

b)

Fig. 4. τ xy / τ avg stress distributions in the adhesive mid-thickness for the DLJ with AV138 (a) and 7752 (b).

3.3. P-  curves and joint strength

Initially, the numerical P -  curves are presented for the SLJ bonded with both adhesives and all L O (Fig. 5). The oscillatory behaviour identified in the validation study (Fig. 2 a) is confirmed. A noteworthy increase of P m with L O is also found with both adhesives, along with higher energies to failure (area beneath the P -  curves). Fig. 6 presents