Fracture Prediction in Brittle Materials
Abstract
The use of ceramics in scientific and industrial applications is limited by their relatively poor mechanical properties, brittleness and variability in strength. Design with brittle materials requires an entirely new concept from that with ductile materials. One must think in terms of probability rather than virtual certainty. Statistical flaw theory based on the weakest-link hypothesis has been applied to problems of fracture and failure of brittle materials under test loads. One such theory, the Weibull distribution, which is probably the most widely applied, addresses the statistical variation of strength and the size effect due to flaws, because in a variety of conditions both its mathematical formulation and the estimation of the Weibull parameters from experiments are simple. Many techniques exist to determine the Weibull parameters from the series of data obtained from experimental tests. The Weibull distribution expressions have been developed for the pure bending and torsion stress states for the fracture of hollow alumina tubes. Such tests are relatively inexpensive and provide useful data reliably but no closed-form solution exists for the probability formulations. An iteration method based on the least-square minimization of residual errors in the test results is used to determine the Weibull parameters.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 29, 1981
- Accession Number
- ADA112452
Entities
People
- George J. Filatovs
- William J. Craft
Organizations
- North Carolina Agricultural and Technical State University