Adhesively bonded joints are widely used in large-scale composite structures. However, their damage and failure behavior, particularly when flexible adhesives are employed, has not yet been fully clarified. In this study, the static and fatigue damage behavior of carbon fiber reinforced plastic scarf joints bonded with flexible urethane adhesives was investigated. Static and fatigue tests were conducted on scarf joints, and the damage evolution was observed in situ using optical microscopy. For the fatigue tests, a trigger system synchronized with the fatigue testing machine was employed to capture microscopic images at the maximum load of each cycle. In the static tests, transverse cracking in the composite adherends and subsequent interlaminar crack propagation were observed prior to final failure. Shear deformation of the urethane adhesive layer was evident compared with the case of epoxy adhesives. When the joint strength was evaluated in terms of the shear stress acting on the adhesive layer, the strength of joints bonded with urethane adhesives became nearly independent of the scarf angle, in contrast to epoxy-bonded joints reported in a previous study. In the fatigue tests, transverse cracking occurred at an early stage of cyclic loading, followed by interlaminar crack growth, whereas shear deformation of the adhesive layer accumulated continuously until final failure. Fracture surface observations revealed that the dominant failure mode depended on the loading condition. Interfacial failure was predominant under static loading, whereas cohesive failure dominated under fatigue loading. These results demonstrate that the interaction between mesoscopic damage in composite adherends and shear deformation of flexible adhesive layers governs the static and fatigue failure behavior of composite scarf joints. The findings provide important insights for strength evaluation, durability assessment, and the reliable design of large-scale composite structures bonded with flexible adhesives.
Oshima et al. (Thu,) studied this question.