PSI - Issue 51

Zahra Silvayeh et al. / Procedia Structural Integrity 51 (2023) 141–144 Z. Silvayeh et al. / Structural Integrity Procedia 00 (2023) 000–000

143

3

3. Results and discussion Fig. 2 (a) exemplarily shows the typical fracture behavior of the samples under quasi-static loading. The joint was the weakest part of the sample, since neither fracture of the aluminum alloy sheet nor fracture of the beech veneer plate occurred. Fig. 2 (b) shows the tensile force-displacement curves as monitored during testing of the samples. The tensile force increased almost linearly to the maximum of about 5 kN (T/L/T) or 6 kN (L/T/L), res-pectively. If the fibers of the cover veneers were oriented perpendicular (transverse, T/L/T) to the loading direction, the bending deformation of the veneer plate at the bonding zone was stronger. The deformation promoted debonding of the aluminum sheet from the beech veneer plate. Hence, the maximum tensile force in T/L/T stacking order was about 1/5 lower than in L/T/L stacking order. At the displacement of about 1.5 mm the force abruptly decreased by almost 2/3 for both stacking orders, once brittle failure of the adhesive layer between the aluminum alloy sheet and the veneer plate occurred. Hence, the static strength (i.e., the maximum tensile force) of the hybrid joints was mainly determined by the adhesive layer between the aluminum alloy sheet and the cross-laminated veneer plate, as proposed by Imakawa et al. (2022) and Domitner et al. (2023b), but not by the screws. However, the maximum tensile force achieved in this study was only about half of the maximum force achieved by Domitner et al. (2023b), although the dimensions of the bonding areas were identical (90 mm × 20 mm). Hence, the adhesive used in the present study was unsuitable for dissimilar joining of aluminum with wood. After failure of the adhesive layer the screws were pulled out of the veneer plate. The tensile force increased again from about 2 kN at the displacement of 1.5 mm to about 2.6-2.8 kN at the displacement of 3-3.5 mm for the samples with T/L/T stacking order (blue lines), but the tensile force steadily decreased for the samples with L/T/L stacking order (green lines). This strongly indicates that the pull-out resistance of the screws was different. As the thickness of the veneer plates was identical for both stacking orders, the different fiber orientations of the cover veneers were identified to influence the pull-out resistance. The orientation-dependent response of the fibers to the load transferred by a single screw is schematically illustrated Fig. 2 (b). If the fibers of the cover veneers were parallel (longitudinal, L/T/L) to the loading direction, the pull-out resistance of the screws was lower, as adjacent fibers were easily separated when the screws were pulled out from the plate. However, if the fibers of the cover veneers were perpendicular (transverse, T/L/T) to the loading direction, the pull out resistance of the screws was higher, as the fibers were compacted (increase of tensile force) before they were torn (decrease of tensile force) when the screws were pulled out from the plate. The reinforcing effect of the T/L/T stacking order vanishes, if the screws are located very close to the edge of the veneer plate, but the effect strengthens, if the number of veneers with fibers perpendicular to the loading direction increases.

(a) (b) Fig. 2. (a) Setup used for quasi-static shear-tensile testing of the screw-bonded samples. (b) Comparison of force-displacement curves monitored in shear-tensile testing of samples with L/T/L and T/L/T stacking orders of the cross-laminated beech veneers.