Determination of stress intensity factors in cracked bonded composite joints

In this paper, a method is presented that facilitates the determination of stress intensity factors in composite bonded joints with interfacial or interlaminar cracks. The method is based on a global/local concept, where the displacements and stresses away from the crack tip are found by a conventional finite element analysis. In the neighborhood of the crack tip singularity, where the finite element results alone no longer converge, an elasticity solution is constructed using an eigenfunction expansion of the stresses and displacements. It was shown that the method allows accurate determination of stresses and stress intensity factors at the crack tip (within 5% of reference solutions) with a relatively small computational effort by using a coarse FE mesh. In addition, the method does not require special finite elements and can be used with any existing finite element code. An equivalent intensity factor K* = root K-I(2) + (1/K-IIc/K-IIc)K-2(II)2 that captures the combined effect of K-I and K-II on failure was defined. Numerical results indicated that, out of various interfacial and interlaminar cracks studied, the largest value of the equivalent intensity factor occurs at the interlaminar crack in the first ply interface of the adherend near the adhesive, but not at the adherend-adhesive interface. These results are consistent with the experimental observation that failure of composite bonded joints typically occurs by delamination of the adherend near. but not at, the adhesive.