Anomalous crack propagation under linear elastic conditions

This paper analyzes fatigue crack propagation (FCP) data obtained from experiments on single-edge notched specimens, prepared from rigid thermoplastic polyurethane. The tests were conducted in tension-tension, at constant R-ratio of 0.1 and a frequency of 0.5 Hz. Two maximum remote stresses were employed: 26.2 MNm(-2) (=0.34 sigma(y)) and 22.1 MNm(-2) (=0.29 sigma(y)). The average crack speed at the higher stress was expectedly higher, however, the data plotted in terms of the stress intensity factor range (Delta K) according to Paris equation yield two intersecting straight line with notably different kinetic parameters, i.e., slope (m) and intercept (A). This anomaly questions the usefulness of the stress intensity factor as a similitude parameter under linear elastic conditions which has been substantiated in other materials. Plastic zone analysis in both cases established the dominance of plain strain and small scale yielding (SSY) conditions and that requirements for K-dominance were satisfied. It was therefore suggested that K-dominance appears to be necessary, but insufficient criterion for transferability of crack propagation data. Application of the Crack Layer (CL) theory to the same data produced a single value of the specific enthalpy of fracture. Differences in FCP kinetics could be related to the change found in the dissipation coefficient.