Abstract
Fatigue tests were carried out on compression mouldings supplied by a leading polymer manufacturer. They were made from three batches of ultra-high molecular weight polyethylene (UHMWPE) with weight-average relative molar masses, ¯¯¯¯MW, of about 0.6 × 106, 5 × 106 and 9 × 106. In 10 mm thick compact tension specimens, crack propagation was so erratic that it was impossible to follow standard procedure, where crack-tip stress intensity amplitude, ΔK, is raised incrementally, and the resulting crack propagation rate, da/dN, increases, following the Paris equation, where a is crack length and N is number of cycles. Instead, most of the tests were conducted at fixed high values of ΔK. Typically, da/dN then started at a high level, but decreased irregularly during the test. Micrographs of fracture surfaces showed that crack propagation was sporadic in these specimens. In one test, at ΔK = 2.3 MPa m0.5, there were crack-arrest marks at intervals Δa of about 2 μm, while the number of cycles between individual growth steps increased from 1 to more than 1000 and the fracture surface showed increasing evidence of plastic deformation. It is concluded that sporadic crack propagation was caused by energy-dissipating crazing, which was initiated close to the crack tip under plane strain conditions in mouldings that were not fully consolidated. By contrast, fatigue crack propagation in 4 mm thick specimens followed the Paris equation approximately. The results from all four reports on this project are reviewed, and the possibility of using fatigue testing as a quality assurance procedure for melt-processed UHMWPE is discussed.