Crack growth predictions in polyethylene using measured traction–separation curves

Abstract

Listen

Translate article

The accuracy of predicting the crack growth in any cohesive zone model calculation depends critically on the choice of cohesive law. A novel experimental method was used to measure directly such a cohesive law or ‘traction–separation curve’ in polyethylene. Deep notched tensile specimens were tested under constant displacement rate conditions, which facilitated a localisation of the damage mechanisms thought to precede crack growth and allowed a quantification of these processes independent of bulk deformation. Results showed that both the fracture energy and cohesive strength measured in this manner are a function of the applied rate and specimen geometry. Here we present a cohesive zone model within the finite-volume method to predict crack initiation and propagation history in three grades of polyethylene of different toughness, using the experimental measurements described above. The choice of cohesive law is crucial as it has a fundamental bearing on the predicted crack growth rates, particularly in tough polymers, where changes in the prevailing rate, constraint and temperature may affect the magnitude of the holding tractions within the damage zone. Initially a single experimentally measured, fixed rate traction–separation curve was used in the model as the fracture criterion but was unable to provide satisfactory crack growth predictions. By contrast, use of a more physically realistic family of curves measured at different rates provided better agreement of the prediction with experiment for the tough polyethylenes and very good agreement for the more brittle polyethylene. It was concluded that along with a rate dependent cohesive law, an accurate prediction of the crack growth history of tough polyethylenes would also require an incorporation of the effects of variations in constraint and perhaps also temperature. The ultimate goal may therefore be the development of a physical material model, sufficiently calibrated by experimental data, which would be able to accurately describe the local fracture process via a rate, constraint and temperature dependent traction–separation law.