• 2018-07
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • One possibility to differentiate between the two failure


    One possibility to differentiate between the two failure modes, and also to identify the corresponding characteristics of the resulting fracture surface, is a series of FNCT measurements with varying nominal (initial) stress. The dependence of the observed time to failure tf on the actual stress changes significantly, as demonstrated e.g. in [14]. Essentially, two linear regions are observed – one at higher stresses and shorter times tf representing ductile failure, and one linear region at lower stresses and longer tf related to brittle behavior and crack propagation leading to failure. As stated above, the FNCT can be performed in various media but results must be assessed carefully as the course of damage, the time to failure tf and also the ranking of different materials with respect to tf can change [15]. Depending on the test medium used, the failure occurring due to crack propagation may be regarded as SCG or ESC, respectively [13]. With the corresponding devices and procedures employed for FNCT both, SCG and ESC, can practically be addressed by selecting the appropriate medium. The FNCT device used in this POM 1 mg study is capable of continuously monitoring the specimen elongation, providing additional data characterizing the course of damaging during the test [14,15]. Several imaging techniques have been used in the past to describe and understand crack growth in PE and other polymers induced by SCG and ESC mechanisms, especially including SEM of fracture surfaces [1,3,10], sometimes after previous etching [7]. These studies mainly aimed to determine the interaction of penetrating or surface-active liquids with polymers or to evaluate crazing process itself. In combination with small-angle X-ray scattering (SAXS), the presence of cavitation and dilatation gaps in the material during deformation could be evidenced alongside highly deformed parts of material in the form of fibrils [16,17], supporting the craze-crack mechanism as the basis of SCG and ESC. Therefore, in this study, mechanically loaded FNCT specimens of PE-HD were examined in detail by different imaging techniques such as light microscopy (LM), laser scanning microscopy (LSM), scanning acoustic microscopy (SAM) and (industrial) X-ray micro computed tomography (CT). FNCT measurements were mainly performed to provide samples for imaging analysis, which were (partially) damaged under appropriate ESC conditions. Practically, the specimens were subjected to the usual FNCT procedure but the test was aborted at different stages before complete failure. Hence, multiple specimens were prepared for the series of measurements. This provides the possibility to study crack propagation at distinct points in time, chosen according to the overall time to failure tf* of the specific material.
    Materials and media
    Results and discussion In order to define the individual time-steps for FNCT exposure to obtain partially damaged samples, first, the duration of the complete test, i.e. the time to failure tf* has to be determined according to the standard procedure specified in ISO 16770 [12]. Subsequently, partially damaged samples were prepared using the appropriate time-steps (see Table 1) and analyzed with respect to fracture surface details and actual crack length using the different imaging techniques.
    Conclusions First, simple light microscopy was applied, which is also required by the standard test procedure specified in ISO 16770, to determine the actual initial residual cross-sectional area of the specimen. For this purpose, the contrast between notch and fracture surface is just sufficient. The contrast between crack surface and broken ligament area (manual cryo-fracture) is not sufficient for a reliable differentiation. This can be seen from the significant scatter of the respective crack length data shown in Fig. 12.
    Funding The authors are grateful for financial support of AiF (Arbeitsgemeinschaft industrieller Forschungsvereinigungen), Berlin, Germany [grant number IGF 18606 N].