|Abstract or Summary
- The sudden failure of metal structures under load, which has
been a design problem for many years, is characterized by a crack
propagation rate which approaches the speed of sound in the material.
These failures, which are often known as catastrophic fractures,
may be divided into two types which depend on the particular material
in question. Fractures which cause a very small amount of plastic
deformation associated with low energy absorption are known as
brittle fractures, whereas high energy absorption during fracture,
due to large amounts of plastic deformation, is associated with
The Griffith theory for spontaneous fracture in brittle materials
states that the strain energy released during fracture must be greater
than the energy needed to form the new crack surface area. This
theory, which has been verified experimentally for brittle materials,
is not applicable to ductile materials since the energy of plastic deformation during crack propagation must be taken into account.
The revised and simplified theory for ductile materials states that
the strain energy must exceed the energy needed to cause plastic
deformation ahead of the crack. The steps in the fracture of ductile
materials are initiation, slow crack growth, and rapid or catastrophic
crack propagation, and the variable plastic deformation involved
in these steps causes the process to be complex in behavior and
analysis. However, an understanding of the causes of slow and rapid
crack growth can be obtained by studying the effects of plastic deformation
at the tip of a propagating crack.
The objective of this thesis was to investigate the factors which
affect the tendency for a ductile crack to propagate catastrophically.
This was accomplished through the use of two types of investigations.
In the first investigation an attempt was made to obtain catastrophic
failure in 18 x 32 inch uniaxially stressed sheets of commercial
household aluminum foil by introducing a crack in the center of the
sheet in such a manner that the crack was elongated outwards in
each direction. The catastrophic failure which was observed, however,
was due to the addition of extra energy to the system boundaries
which was not accounted for in the theory. Slow crack propagation
was also observed and led to the second investigation.
The plastic deformation at the crack tip was studied by visual
observation using a metallograph and by actual measurement
amount of deformation occurring during crack growth. This with a Tukon microhardness tester in which indentations were placed
on the foil specimens, which were later given a small crack at the center, and which were mounted in a drill press vise on the tester. The crack was then propagated by turning the screw on the vise and
measurements of the deformation between the indentations were
The photographs and measurements of the plastic deformation
at the crack tip showed that deformation existed throughout the entire
specimen and was not localized in a given area at the crack tip as has
been assured in some theories. The existance of elastic strain in
the material was shown by the occurrence of elastic recovery in the
strain relieved areas of the cracked material although no differentiation
could be made between the area of elastic and plastic deformation.
The existance of a stable configuration of iso-strain contours
surrounding the tip of the crack was also shown.
Calculations of the strain energy needed to satisfy the plastic
deformation energy requirements showed that a crack length of about
49 inches would be needed to cause catastrophic failure. This is
much larger than the specimen size used and would explain the lack
of rapid crack growth in the tests run under near theoretical conditions.