Polymer Testing (Jan 2021)

Influence of sample thickness, curvature and notches on the Charpy impact strength - An approach to standardise the impact strength of curved test specimens and biological structures

  • Nina Graupner,
  • Niels Kühn,
  • Jörg Müssig

Journal volume & issue
Vol. 93
p. 106864

Abstract

Read online

The specimen geometry has a significant influence on the Charpy impact strength. This is often a problem when it comes to the analysis of materials that can only be prepared with a curved shape, or which, like some biological structures, are naturally given in curved form like nutshells. The question repeatedly arises to what extent the curvature influences the impact strength. Hence, the present study deals with the dependence of the specimen geometry like specimen thickness, curvature and notches on the impact properties of concrete, gypsum, polyurethane (PUR) and epoxy-based (EP) samples. Increasing sample thickness from 2 to 10 mm resulted in increased toughness for brittle materials, whereas more ductile materials did not show any significant change in toughness. Increasing span length showed an increase of the unnotched impact strength for brittle gypsum, while toughness increased for a more ductile PUR sample. For the curved specimens both, the unnotched and the notched Charpy impact strength could be shown to increase with a reduction of the specimen radius (higher curvature) while the notch-sensitivity was not significantly affected. Material-specific linear relationships between the impact strength and the curvature of the samples were found. These relationships were used to create a model to calculate the influence of the curvature on the impact strength. For all materials, the characteristic values of a normalised flat standard sample could be well predicted from measured values of curved specimens. This calculation approach was used to predict the impact strength of an isotropic biological sample (sweet potato) concerning a flat sample and different curved specimens. The results show a good trend with the measured values.

Keywords