EXPERIMENTAL INVESTIGATION OF THE EFFECTS OF STRESS RATE AND STRESS LEVEL ON FRACTURE IN POLYSTYRENE

The effects of stress rate and stress level on fatigue crack propagation in compression-moulded single-edge notched specimens (0.25 mm in thickness) of polystyrene are reported. Values of the stress rate sigma are obtained from the formula sigma = 2-nu (sigma-max - sigma-min), where nu is the frequency and sigma-max, sigma-min are the maximum and minimum stresses of the fatigue cycle. Different levels of sigma are achieved by changing the frequency while keeping sigma-max, sigma-min at fixed values. The effect of the stress level is investigated by keeping sigma and sigma-min constant and varying sigma-max and nu. The results show that when the kinetic data are plotted as DELTA-I/DELTA-t against the energy release rate G1, a relatively small effect of the stress rate is observed. If the same data are treated as DELTA-I/DELTA-N against G1, a decrease in DELTA-I/DELTA-N with test frequency is seen. The increase in the level of sigma-max results in a higher crack speed. The critical crack length is found to be practically the same for all stress-rate experiments. A decrease in the critical crack length is observed with the increase in stress level. Analysis of craze distribution around the crack path shows that the extent of crazing decreases with the increase in stress rate and increases with the increase in stress level. For all experimental conditions, the ratio of the second moment to the square root of the fourth moment of the histograms of craze density along directions normal to the crack path is found to be constant throughout the slow phase of crack propagation. This result supports a self-similarity hypothesis of damage evolution proposed in the crack layer model.