The earliest analysis along the above lines was conducted by
Griffith [1921] in 1920. Griffith used
the crack geometry and loading configuration shown in Figure
2.2.11 and assumed that the stress would be constant during any incremental
growth of the crack. Griffith also
neglected the plastic work term in Equation 2.2.9 since he was trying to test
his fracture hypothesis with a brittle material, glass. Griffith’s analysis showed that the input
work rate (G) was equal to the
negative of the derivative of potential energy of deformation (Us ) as shown by Equation 2.2.12, and
the resistance (R) was equal to the
rate of increase in potential energy due to surface energy (UT ) during crack extension:
|
(2.2.13)
|
Figure
2.2.11. Griffith Crack and Loading
Configuration, Uniformly Loaded, Infinite Plate with a Center Crack of Length 2a
The potential energy of deformation (Us ) was found to be
|
(2.2.14)
|
while the potential energy due to surface tension (UT
) was given by
|
(2.2.15)
|
with surface tension T,
and for plate thickness B.
The crack area A is
given by
|
(2.2.16)
|
So the energy balance equation becomes
|
(2.2.17)
|
where E¢ is dependent on the stress state in the following way
E¢ =
|
E
/ (1-n2), for
plane strain
|
(2.2.18)
|
E, for plane stress
|
Solving Equation 2.2.17 for the critical stress (scr)
associated with the point at which the crack (a) would grow, one finds
|
(2.2.19)
|
Later, Irwin [1948] and Orowan [1949] incorporated the effects
of crack tip plasticity into the analysis by taking the plastic dissipation
term in Equation 2.2.9 as a constant, i.e. they assumed that
|
(2.2.20)
|
so that the resistance in Equation 2.2.17 was defined as the
combination of surface energy absorbed and plastic work dissipated. Thus, the Griffith-Irwin-Orowan energy
balance equation became
|
(2.2.21)
|
and the critical stress was
|
(2.2.22)
|
Both Irwin and Orowan noted that the plastic dissipation rate
for metals was at least a factor of 1000 greater than the surface energy
absorption rate so that Equation 2.2.22 could be approximated by
|
(2.2.23)
|
Irwin also noted that the driving force or input energy rate G was directly related to the square of
the magnitude of the crack tip stress field for the Griffith center crack
geometry (Figure 2.2.11), i.e., that
|
(2.2.24)
|
Later, Irwin [1960] reported this result to be general for any
cracked elastic body based upon a virtual work analysis of the stresses and
displacements associated with crack tip behavior during an infinitesimal crack
extension.