On the applicability of linear elastic fracture mechanics scaling relations in the analysis of intergranular fracture of brittle polycrystals

Journal Article (2019)
Author(s)

Z. Shabir (TU Delft - Applied Mechanics)

Erik van der Giessen (Rijksuniversiteit Groningen)

C. Armando Duarte (University of Illinois at Urbana Champaign)

A. Simone (Università degli Studi di Padova)

Research Group
Applied Mechanics
Copyright
© 2019 Z. Shabir, Erik Van der Giessen, C. Armando Duarte, A. Simone
DOI related publication
https://doi.org/10.1007/s10704-019-00381-x
More Info
expand_more
Publication Year
2019
Language
English
Copyright
© 2019 Z. Shabir, Erik Van der Giessen, C. Armando Duarte, A. Simone
Research Group
Applied Mechanics
Issue number
2
Volume number
220
Pages (from-to)
205-219
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Crack propagation in polycrystalline specimens is studied by means of a generalized finite element method with linear elastic isotropic grains and cohesive grain boundaries. The corresponding mode-I intergranular cracks are characterized using a grain boundary brittleness criterion that depends on cohesive law parameters and average grain boundary length. It is shown that load–displacement curves for specimens with the same microstructure and for various cohesive law parameters can be obtained from a master load–displacement curve by means of simple linear elastic fracture mechanics scaling relations. This property is a consequence of the independence of intergranular crack paths from cohesive law parameters. Perfect scaling is obtained for cases characterized by the same grain boundary brittleness number, irrespective of its value, whereas scaling is approximated for cases with different but relatively large values of the grain boundary brittleness number. The former case corresponds to grain boundary traction profiles that are identical apart from a scale factor; in the latter case, a large grain boundary brittleness number implies similar, apart from a scale factor, traction profiles. By exploiting this property, it is demonstrated that computationally expensive simulations can be avoided above a certain grain boundary brittleness threshold value.