PSI - Issue 47

J.B.S. Nóbrega et al. / Procedia Structural Integrity 47 (2023) 408–416 Nóbrega et al./ Structural Integrity Procedia 00 (2023) 000–000

415

8

The numerical prediction shows an increase of P/b as a increases, which can be attributed to the incremental nature of the imposed displacement; the peel strength goes from 2.39 N/mm to 2.81 N/mm along the standardized interval (Figure 6). Nevertheless, the slope of the curve is minimal. The average peel strength was found to be 2.66 N/mm. In this case, the value is higher than the observed experimentally, as shown in Figure 7, but lower than the value reported by the manufacturer (Huntsman, 2015).

0 0,5 1 1,5 2 2,5 3

P/b [N/mm]

0

50

100

150

a [mm]

2015 numerical

2015-1 2015-4

2015-2

2015-3

Figure 7. Comparison between numerical and experimental peel strengths.

The proposed model was displacement driven, in which the amount of displacement imposed increases on each iteration until the value of  is reached. Such loading mode has no effect on static models, as it has been observed in the literature; however, in this work, the models were quasi-static and were solved using a different solver; therefore, the actual implication of such loading approach is still unknown. On the other hand, force-driven models were also explored during the development of this work, which did not yield the expected outcomes and took even longer. Regarding the differences between the average peel strength, the mechanical properties employed in the numerical model were taken from different experimental tests but the FRPT, these data are already available in the literature from previous work. In consequence, the testing speeds used to obtain those properties were significantly lower than those employed for the FRPT; furthermore, it is known that the testing speed influences the overall mechanical behaviour of the materials, so further work is necessary for both gathering mechanical properties experimentally and in refining the numerical model, although the model provides a conservative estimation. It is worth noting that, to the authors’ best knowledge, this is the first numerical methodology that fully represents the FRPT while considering the elastic effect the substrates have on the final peel strength. 4. Conclusions The present work aimed to develop a methodology to determine the peel strength of adhesive joints using numerical modelling and mechanical properties obtained from more common experimental tests. The proposed methodology considers for the first time the floating behaviour of the test, the elastic effects of the adherends and their influence on estimating the peel strength while requiring moderate computational power. The methodology was tested with mechanical properties gathered from the literature and then the average peel strength was calculated. The obtained curves are qualitatively similar to those obtained experimentally. On the other hand, the predicted peel strength was found within the range of the experimental data (obtained in this work) and the value reported by the manufacturer.