Quantitative in situ study of short crack propagation in polygranular graphite by digital image correlation

M. Mostafavi, T. J. Marrow*

*Corresponding author for this work

Research output: Contribution to journalArticle (Academic Journal)peer-review

36 Citations (Scopus)

Abstract

This paper reports experimental observations that show non-irradiated Gilsocarbon polygranular nuclear graphite fits within the quasi-brittle class of materials. Such materials exhibit a degree of damage tolerance that depends on the stability of cracks that nucleate in the microstructure. Modelling efforts to predict the influence of microstructure on damage tolerance require direct observation of crack nucleation and growth to support them. Here, the technique of digital image correlation was applied to optical observations to measure the full field distribution of displacements on the surface of large (>100 mm dimension) specimens, loaded in uniaxial flexure. Repeated cyclic loading to strains approaching failure results in an inelastic (i.e. non-recoverable) strain, and a decrease in the static elastic modulus. Digital image correlation was also used for early detection and characterisation of fracture nuclei (short cracks). Such short cracks show a stable propagation stage before causing catastrophic failure. The displacement fields were used to calculate directly the energy release rate of the short cracks via a contour integral method. The value is consistent with the critical strain energy release rate for unstable fracture obtained from standard mode I fracture tests.

Original languageEnglish
Pages (from-to)695-707
Number of pages13
JournalFatigue and Fracture of Engineering Materials and Structures
Volume35
Issue number8
DOIs
Publication statusPublished - Aug 2012

Keywords

  • Crack Propagation
  • Digital Image Correlation
  • Gilsocarbon Nuclear Graphite
  • Inelastic Strain
  • Short Cracks

Fingerprint

Dive into the research topics of 'Quantitative in situ study of short crack propagation in polygranular graphite by digital image correlation'. Together they form a unique fingerprint.

Cite this