Page 373 - T. Anderson-Fracture Mechanics - Fundamentals and Applns.-CRC (2005)

1656_C008.fm Page 353 Monday, May 23, 2005 5:59 PM

8 Fracture Testing of Nonmetals

The procedures for fracture toughness testing of metals, which are described in Chapter 7, are fairly
well established. Fracture testing of plastics, composites, and ceramics is relatively new, however,
and there are a number of unresolved issues.
Although many aspects of fracture toughness testing are similar for metals and nonmetals, there
are several important differences. In some cases, metals fracture testing technology is inadequate on
theoretical grounds. For example, the mechanical behavior of plastics can be highly rate dependent,
and composites often violate continuum assumptions (see Chapter 6). There are also more pragmatic
differences between fracture testing of metals and nonmetals. Ceramics, for instance, are typically
very hard and brittle, which makes specimen fabrication and testing more difficult.
This chapter briefly summarizes the current procedures for measuring the fracture toughness
in plastics, fiber-reinforced composites, and ceramics. The reader should be familiar with the
material in Chapter 7, since much of the same methodology (e.g., specimen design, instrumentation,
and fracture parameters) is currently being applied to nonmetals.

8.1 FRACTURE TOUGHNESS MEASUREMENTS
IN ENGINEERING PLASTICS

Engineers and researchers who have attempted to measure the fracture toughness of plastics have
relied almost exclusively on metals testing technology. Existing experimental approaches implicitly
recognize the potential for time-dependent deformation, but do not specifically address viscoelastic
behavior in most instances. Schapery’s viscoelastic J integral [1,2], which was introduced in
Chapter 4, has not seen widespread application to laboratory testing.
The Mode I stress-intensity factor K and the conventional J integral were originally developed
I
for time-independent materials, but may also be suitable for viscoelastic materials in certain cases.
The restrictions on these parameters are explored below, followed by a summary of procedures for
K and J testing on plastics. Section 8.1.5 briefly outlines possible approaches for taking account
of viscoelastic behavior and time-dependent yielding in fracture toughness measurements.

8.1.1 THE SUITABILITY OF K AND J FOR POLYMERS

A number of investigators [3–7] have reported K , J , and J-R curve data for plastics. They applied
Ic
Ic
testing and data analysis procedures that are virtually identical to metals approaches (See Chapter 7).
The validity of K and J is not guaranteed, however, when a material exhibits rate-dependent mechanical
properties. For example, neither J nor K are suitable for characterizing creep crack growth in metals
1
(Section 4.2); an alternate parameter C* is required to account for the time-dependent material
behavior. Schapery [1,2] has proposed an analogous parameter J to characterize viscoelastic
v
materials (Section 4.3).

1 The stress-intensity factor is suitable for high-temperature behavior in limited situations. At short times, when the creep
zone is confined to a small region surrounding the crack tip, K uniquely characterizes crack-tip conditions, while C * is
appropriate for large-scale creep.

353