Subcritical crack growth law and its consequences for lifetime statistics and size effect of quasibrittle structures
For brittle failures, the probability distribution of structural strength and lifetime are known to be Weibullian, in which case the knowledge of the mean and standard deviation suffices to determine the loading or time corresponding to a tolerable failure probability such as 10 −6 . Unfortunately,...
Published in: | Journal of Physics D: Applied Physics |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
IOP Publishing
2009
|
Subjects: | |
Online Access: | http://dx.doi.org/10.1088/0022-3727/42/21/214008 https://iopscience.iop.org/article/10.1088/0022-3727/42/21/214008 https://iopscience.iop.org/article/10.1088/0022-3727/42/21/214008/pdf http://stacks.iop.org/0022-3727/42/i=21/a=214008/pdf |
Summary: | For brittle failures, the probability distribution of structural strength and lifetime are known to be Weibullian, in which case the knowledge of the mean and standard deviation suffices to determine the loading or time corresponding to a tolerable failure probability such as 10 −6 . Unfortunately, this is not so for quasibrittle structures, characterized by material inhomogeneities that are not negligible compared with the structure size (as is typical, e.g. for concrete, fibre composites, tough ceramics, rocks and sea ice). For such structures, the distribution of structural strength was shown to vary from almost Gaussian to Weibullian as a function of structure size (and also shape). Here we predict the size dependence of the distribution type for the lifetime of quasibrittle structures. To derive the lifetime statistics from the strength statistics, the subcritical crack growth law is requisite. This empirical law is shown to be justified by fracture mechanics of random crack jumps in the atomic lattice and the condition of equality of the energy dissipation rates calculated on the nano-scale and the macro-scale. The size effect on the lifetime is found to be much stronger than that on the structural strength. The theory is shown to match the experimentally observed systematic deviations of lifetime histograms from the Weibull distribution. |
---|