Page 385 - T. Anderson-Fracture Mechanics - Fundamentals and Applns.-CRC (2005)

1656_C008.fm Page 365 Monday, May 23, 2005 5:59 PM

Fracture Testing of Nonmetals 365

FIGURE 8.7 Stress-strain curves for the rigid PVC and polycarbonate. Taken from Cayard, M., “Fracture Tough-
ness Testing of Polymeric Materials.” Ph.D. Dissertation, Texas A&M University, College Station, TX, 1990.

dependence, which implies that the E 399 requirement is at least adequate for this material.
Further testing of thinner sections would be required to determine if the E 399 thickness requirement
is overly conservative for the PC.
The apparent lack of a significant thickness effect on the fracture toughness in PVC and PC is
broadly consistent with the observed behavior in metals. Recall from Section 2.10 and Section 7.2
that the apparent thickness effect on fracture toughness is due primarily to shear lip formation in
materials that fail by ductile crack growth. When a metal fails by cleavage, the fracture surface is
usually flat (no shear lips), and the fracture toughness is much less sensitive to the specimen
thickness. The PVC and PC specimens failed in a relatively brittle manner, and shear lips were not
evident on the fracture surface. Consequently, one would not expect a significant thickness effect.
One final observation regarding the ASTM K standard for plastics is that the procedure for
Ic
estimating P ignores time effects. Recall from the earlier discussion that nonlinearity in the load-
Q
displacement curve in K tests can come from two sources: yielding and crack growth. In the case
Ic
of polymers, viscoelasticity can also contribute to nonlinearity in the load-displacement curve.
Consequently, at least a portion of the 5% deviation from linearity at P could result from a decrease
Q
in the modulus during the test.
For most practical situations, however, viscoelastic effects are probably negligible during K tests.
Ic
In order to obtain a valid K result in most polymers, the test temperature must be well below T , where
g
Ic
rate effects are minimal at short times. The duration of a typical K test is on the order of several minutes,
Ic
and the elastic properties probably will not change significantly prior to fracture. The rate sensitivity
should be quantified, however, to evaluate the assumption that E does not change during the test.
8.1.4 J TESTING
A number of researchers have applied J integral test methods to polymers [3–6] over the last couple
of decades. More recently, ASTM has published a standard for J-R curve testing of plastics [13].
In addition, the European Structural Integrity Society (ESIS) has developed a protocol for J testing
of plastics [7]. Both J test methods are based on standards that were developed for metals such as
ASTM E 1820 [14].