Published January 2010 | Version v1
Journal article

Overview of statistical models of fracture for nonirradiated nuclear-graphite components

  • 1. NASA Glenn Research Center, Cleveland, OH (United States)
  • 2. Idaho National Laboratory, Idaho Falls, ID (United States)

Description

Nuclear-grade (low-impurity) graphite for the fuel element and moderator material for Next Generation (Gen IV) Reactors displays large scatter in strength and a nonlinear stress-strain response from damage accumulation. This response can be characterized as quasi-brittle. In this review, relevant statistical failure models for various brittle and quasi-brittle material systems are discussed with regard to strength distribution, size effect, multiaxial strength, and damage accumulation. This includes descriptions of the Weibull, Batdorf, and Burchell models as well as models that describe the strength response of composite materials, which involves distributed damage. Results from lattice simulations are included for a physics-based description of material breakdown. Consideration is given to the predicted transition between brittle and quasi-brittle damage behavior versus the density of damage (level of disorder) within the material system. The literature indicates that weakest-link-based failure modeling approaches appear to be reasonably robust in that they can be applied to materials that display distributed damage, provided that the level of disorder in the material is not too large. The Weibull distribution is argued to be the most appropriate statistical distribution to model the stochastic strength response of graphite.

Availability note (English)

Available from http://dx.doi.org/10.1016/j.nucengdes.2009.10.002

Additional details

Identifiers

DOI
10.1016/j.nucengdes.2009.10.002;
PII
S0029-5493(09)00496-8;

Publishing Information

Journal Title
Nuclear Engineering and Design
Journal Volume
240
Journal Issue
1
Journal Page Range
p. 1-29
ISSN
0029-5493
CODEN
NEDEAU

Optional Information

Copyright
Copyright (c) 2009 Elsevier Science B.V., Amsterdam, The Netherlands, All rights reserved.