PSI - Issue 68

P.M.D. Carvalho et al. / Procedia Structural Integrity 68 (2025) 398–404

401

4

P.M.D. Carvalho et al. / Structural Integrity Procedia 00 (2025) 000–000

placed at the bond edges to guarantee the designated value of t A . Spring clamps were used to ensure alignment of the different components during curing. The joints cured for a week at room temperature. A Universal Testing Machine (UTM) Shimadzu AGX-100 with a 100 kN load cell was used to perform tensile tests. The P m value of each test was recorded, then the average was calculated. The load-displacement ( P - d ) curves were also recorded. 2.4. FEM pre-processing The joint geometry corresponds to Fig. 1, with all configurations mentioned in section 2.2 being numerically simulated using in the ABAQUS ® environment (ABAQUS 2017, Dassault Systèmes, RI, USA) as 2D models assuming plane strain conditions (Adams and Peppiatt 1974). The joint symmetry on the YZ plane was taken in consideration to reduce the computation cost. Numerical modelling was accomplished by a CZM approach in the adhesive. Four-node plane strain continuum elements (CPE4R) were used to mesh the adherends in both approaches. Close to the adhesive layer the mesh was constructed using biased seeds to guarantee a finer discretization. Fig. 2 shows an example of the mesh used in these models. The adherends’ mesh elements varied from 2 mm to 0.2 mm.

Fig. 2. Mesh details for the model with t P2 =1 mm and cohesive elements.

The CZM models used a layer of four-node cohesive elements (COH2D4) with the through-thickness direction of the adhesive layer. A triangular CZM was considered. Damage initiation was predicted by a quadratic stress criterion, and damage growth by a linear energetic criterion (Campilho et al. 2013). The element size in the adhesive layer region was 0.2×0.2 mm 2 (Fig. 2). The left vertical edges had symmetry (Ux=0). To represent the grip of the testing jig, the right edge of the lower adherend was constrained in both directions (horizontally and vertically; Uy=Ux=0). 3. Results and analysis 3.1. CZM validation The CZM validation is conducted using the geometry and loading conditions of Section 2.2, and materials of Section 2.1, but under SAJ scenarios (Fig. 1). The geometry, loading, and boundary conditions are consistent with those employed in the numerical study of DAJ. Fig. 3 illustrates the experimental averages and deviations of P m along with the corresponding CZM predictions for the joints bonded with the 2015 and 7752 (SAJ configuration).

$

#

! ' ()*+

"

!

!

%

"

&

#

" ," (''+

"!%-./01/23'/4567 ::-"./01/23'/4567

"!%-.48'/23967 ::-".48'/23967

Fig. 3. Validation results: P m as a function of t P2 for the joints bonded with the 2015 and 7752.