Measuring fracture properties of adhesives in Mode II is often problematic. Indeed, no method can today be regarded as established by the community. In this paper a number of methods are presented. Experiments show that the evaluated properties of the same adhesive sometime vary considerably with the choice of specimen. Even just modest variations in loading conditions using the same specimen can yield considerable variation in the evaluated properties. Sources for these deficiencies are identified.
It has long been understood that Mode II testing using the end-notched flexure specimen (ENF) is conditionally stable. That is, the length of the crack has to be large enough to achieve a stable experiment. This is also the case for other Mode II specimens. A condition for stability is derived leading to an easily evaluated equation. Moreover, careful studies of the crack tip area during Mode II experiments often reveal an expansion of the adhesive during the final phase of loading. That is, negative Mode I loading. Due to the stiffness of the adherends, this leads to a compressive transversal loading of the process zone. Experiments and simulations show that the evaluated fracture energy depends on this constraint. A more detailed analysis of Mode II loading considering large-scale process zones gives some insight into the problem. It is also clear that Mode II has to be more carefully defined than is necessary for Mode I. Due to the transversal expansion of the process zone associated with shear, we may choose to define Mode II as a state of pure shear deformation or a state of pure shear stress. In experiments, none of these states are easily achieved. Moreover, transversally loaded short specimens can result in a process zone extending under the loading point. The result is compression of the process zone and exaggerated evaluated fracture energy. This problem is especially important to consider when evaluating soft and tough adhesives. If better understood and modelled, these effects might also be used in design so that an adhesive joint is loaded in a more favourable way.