PSI - Issue 72

Thomas Steffen Methfessel et al. / Procedia Structural Integrity 72 (2025) 105–112

111

Fig. 8. Incremental energy release rates of the analytical and numerical model for steel adherends connected with adhesive layer of epoxy under a tensile load of 8kN. 6. Results for effective strength For the same kind of single-lap joint, the effective strength (failure load) was then determined depending on the adhesive thickness. As can also be seen from Fig. 8 the mode II portion of the energy release rate and the total energy release rate show a clearly non-monotonic behavior. As a consequence, the optimization procedure for the effective strength is a bit more demanding, however, with a sufficiently general optimization algorithm (also implemented in MATLAB) the effective strength and the corresponding crack length can be identified. Figure 9 shows the resultant critical failure load in dependence of the adhesive thickness (upper curves) and also the generated critical crack lengths (lower curves) in comparison to respective finite element analyses and, even more important, for the critical failure load also in comparison to experimental findings of da Silva et al. (2006).

Fig. 9. Mechanical failure loads and crack lengths for different adhesive thicknesses t according to analytical model (AM) in comparison with finite element analyses (FEA) and experimental findings (EXP) of da Silva et al. (2006).

Obviously, there is again a very satisfying agreement and the presented analysis approach leads to realistic predictions. The impact of the adhesive thickness is thus that an increase of the thickness results in a decrease of the critical failure load. This so- called “thickness effect” originates from the energy subcriterion as thicker adhesive layers mean more stored strain energy and in the case of crack propagation an increased incremental energy release rate. The