FEM Analysis as a Tool to Study the Behavior of Methacrylate Adhesive in a Full-Scale Steel-Steel Shear Joint

The use of adhesive to joint structural elements, despite many advantages of this technology, is not a method commonly used in engineering practice, especially in construction. This is mainly due to the poor recognition of the behavior, both in terms of testing and analysis, of joints made on a scale similar to the actual elements of building structures. Therefore, this paper presents the results of model tests and then numerical analyses of adhesively bonded joints made of high-strength steel elements in a full-scale (double-lap joint). In order to properly model the adhesive connection, material tests of the methacrylate adhesive were performed in the field of tensile, shear (in two versions: single lap joint test and thick adherent shear test) and bond properties. Comparison of the results of the model and numerical tests showed very good agreement in terms of the measurable values, which makes it possible to consider the results obtained in the adhesive layer as reliable (not directly measurable in model tests). In particular, the distribution of stresses inside the adhesive layer, the range of plastic zones and areas of loss of adhesion are presented and discussed. The results indicate the possibility of a reliable representation of the behavior of adhesively bonded joints of high-strength steel, thus providing a tool for the analysis of semirigid adhesive in large-size joints.

Ratcheting and recovery of adhesively bonded joints under tensile cyclic loading

Polymers in general, and adhesives in particular, can exhibit nonlinear viscoelastic–viscoplastic response. Prior work has shown that this complex behavior can be described using analytical models, which provided good agreement with measured creep and recovery response. Under cyclic loading, however, some adhesives exhibit a temporal response different from what would be expected from their creep behavior. Ratcheting describes the accumulation of deformation from cyclic loading. The failure surfaces of adhesives subjected to creep and cyclic loads provide evidence of failure modes that depend on the loading history, suggesting a cause for the change in temporal response. The following considers two approaches to describe the ratcheting behavior of adhesives. Given the reduced time dependence, the first approach involved a nonlinear viscoelastic–plastic model. The second approach used a nonlinear viscoelastic–viscoplastic model, calibrated from the cyclic response, rather than the creep response. While both models showed good agreement with experiment for long exposure to cyclic loading, only the viscoelastic–viscoplastic model agreed with experiment for both short and long loading histories.