Use of mode - I cohesive - zone models to describe the fracture of an adhesively - bonded polymer - matrix composite

In this paper, the use of a cohesive-zone approach to model the mode-I fracture of adhesive joints made from a polymer-matrix composite is demonstrated. Cohesive-zone parameters were obtained by matching numerical results to experimental observations. It is shown that there is a distinction between the characteristic strength of the interface associated with the toughness, and the intrinsic cohesive strength of the interface. While the characteristic strength and toughness are often sufficient to describe fracture in the presence of a crack, the intrinsic cohesive strength is also required to analyze some geometries that have very small characteristic dimensions or crack lengths. It is shown that cohesive-zone models accurately predict the behavior of the joints studied. In particular, not only are the strengths and deformations accurately described, but the transition between failure of the composite and failure of the interface can also be predicted. This mode-I transition cannot be predicted by conventional fracture mechanics as it depends on both the energy-based and strength-based failure parameters associated with cohesive-zone models. The mechanical performance of structural assemblies containing adhesively-bonded components are generally analyzed using finite-element-based commercial codes. These codes do not have built-in procedures that are robust and well-tested for the purpose of modeling the failure of adhesively-bonded joints. Therefore, a strong need exists to develop techniques that can be used in conjunction with standard finite-element procedures to analyze structures containing adhesively-bonded joints. The field of linear-elastic fracture mechanics (L.E.F.M.) provides one theoretical framework by which the strength of adhesive bonds can be characterized. In principle, the toughness of a bonded interface can be determined as a function of the phase angle (which is a measure of the relative amounts of shear and opening at the crack tip [Hutchinson and Suo, 1992]). Provided the geometry of the bonded system, including characteristic dimensions such as the crack length, and the applied loads are known, it is possible to calculate the energy-release rate and phase angle for a crack. A comparison of the calculated energy-release rate with the experimentally-determined value of toughness at the appropriate degree of mode-mixedness then permits an assessment of whether a crack will propagate. In principle, it is possible to determine values of toughness for a variety of crack trajectories (in the adhesive, along the interface, or through the adherend), and to use the concepts of mixed-mode fracture mechanics to predict the appropriate crack path 3 Unfortunately, predicting the behavior of a …