Damage of PE HD due to both mechanisms is
Damage of PE-HD due to both mechanisms is usually considered to emerge from a local stress concentration in the material, leading to the initiation of a crack, which subsequently propagates further through the material, and finally leads to macroscopic failure. In this process, crack growth occurs owing to an overall stress well below the yield stress of the polymer [4,7]. The local stress concentration causing crack initiation is often due to inhomogeneities, surface scratches or defects in the material. Generally, for semi-crystalline polymers above their glass transition (such as polyethylene) crack propagation at low stress levels is usually discussed in terms of a crazing mechanism. After a crack is initiated emanating from a pre-existing defect, at first micro-voids or microcracks develop in the zone of plastic deformation at the forefront of the tip, which then grow to larger voids and, finally, fibrils are formed from the highly orientated inter-void material [8,9]. This stage is denoted as ‘craze’. Further growth of the crack occurs when these fibrils in the craze-zone start to break. Depending on stress level and rate of loading, this failure of the fibrils can be due to chain scission or disentanglement. For the slow processes SCG and ESC, disentanglement is regarded the KU55933 mechanism. The broken fibrils result in characteristic lawn-like texture on the fracture surface in the sub-micrometer range, as revealed by SEM [7,, , ]. To prevent failure caused by SCG/ESC, the occurrence of defects and inhomogeneities can be minimized, and a PE-HD material with enhanced resistance against both can be chosen. This resistance has origin in the molecular architecture, i.e. the molecular weight distribution, long- and short-chain branches, and the resulting complex semi-crystalline morphology and the interconnection of amorphous and crystalline regions, as well as inter-crystalline links (tie molecules) [4,7,11]. For checking and quantifying of this resistance, meaningful test methods are required. To realize such a test with durations in a practically acceptable time-scale, those methods include the application of a well-defined notch for crack initiation in the test specimen. During the test, a constant mechanical load impinged on notched specimens results in progressive crack growth due to craze-crack propagation and, finally, characteristic fracture of samples . The time to failure tf under certain specified conditions (geometry, mechanical stress, temperature and medium) is taken as a measure of the relative resistance to crack growth (SCG or ESC). One of these established testing methods is the full-notch creep test (FNCT) used in contrast study. The FNCT is widely applied for polyethylene materials, predominantly for pipe and container applications. The corresponding standard ISO 16770  specifies an aqueous solution of a neutral surfactant (Arkopal N100) as liquid environment, but other chemicals, air or distilled water may also be used. Due to its influence on the craze-crack mechanism, the Arkopal solution accelerates the test by a factor of about 5 compared to the test in water . The geometry of FNCT specimens (according to ISO 16770) ensures plane strain conditions as encountered in most practical applications . Tensile load and temperature conditions must be chosen to produce failure due to crack growth resulting in a predominantly brittle fracture surface. It must be noted that an increase of tensile stress may lead to a change of the damage mechanism and ductile shear deformation becomes dominant. In this case the test no longer represents the crack propagation behavior. Such an increase in tensile stress during an FNCT experiment is inevitable because the actual residual cross-sectional area is continuously decreasing due to progressing crack growth – so the very last stages of the test are always ductile in nature [11,14]. Consequently, the FNCT is only considered valid, i.e. representing the crack propagation behavior of the material under the chosen conditions, when areas corresponding to brittle behavior constitute the major part of the fracture surface, and crack growth is the major factor determining the observed time to failure. Failure caused by shear deformation (somewhat similar to the behavior in a conventional tensile test) finally results in a larger central ligament and the fracture surface is considered ductile. Following the simple classification of the ISO standard, such tests should be rejected.