Transition unstable fracture

The final transition, Transition II, occurs at a test speed about two decades above that at which Transition I is seen. The unstable fracture already occurs in the linear domain of the force-displacement curve (a in Figure 3). This transition is related to the total disappearance of toughening effects. The absence of a whitened zone on the specimen is noticeable. Under these conditions, the KImax = Klc and GImax = Glc values measured for the 2 L15 system are even lower at high test speeds than those measured for the neat PMMA. 

Transition I partially stable to fully unstable fracture Energy for initiation and propagation cannot be separated (Figure 5c) (drastic drop in force at Fmax)... 

However, when stable-unstable fracture occurs (between Transitions 0 and I), evidence of secondary crack nucleation has been obtained by SEM on postmortem fracture surfaces. These cracks lead to parabolic markings ahead of the main blunt crack propagating in the plastic zone. Figure 8a shows an example of such secondary cracks developed at the boundary between the stable growth region and the unstable fracture surface in a 2 L45 specimen tested at 0.1 m/s. 

The effects of moisture on epoxy fracture are not conclusive. Scott et al. reported that an amine cured epoxy, normally displaying stick-slip fracture at room temperature and low rates, exhibited stable behavior when immersed in distilled water. Also, they found that the rate necessary to promote the unstable to stable crack growth transition at room temperature was increased by two orders of magnitude in the presence of the water. Yamini and Young , on the other hand, found that testing in water tended to suppress stable behavior and promote stick-slip fracture in an amine cured epoxy over a wider range... 

By analogy with what have been done with the conventional LEFM approach, the first solution seems to be the most appropriate. This point is, however, ambiguous two values of rp and therefore of Ketr could not have been determined with certainty those in the transition zone at 0.4 and 0.7 m/s as direct consequence of the unstable-stable transition occurring in each series. We have indeed noticed that for an a/W, where both macroscopic behaviours have been observed, the values of Fmax were 5 to 10% higher under unstable crack propagation than for stable fracture growth. Table 6 shows the different values of rp and Ketr obtained ... 

Here Kq is the value of the stress intensity factor at the instant of the crack onset, Kc is the stress intensity factor critical value corresponding to the fracture transition into the stage of unstable crack propagation, while the difference Kq — Kq characterizes the residual strength of a specimen with a propagating crack. 

To understand ductile-to-brittle transitions, the Rice criterion in which the ratio between the surface energy (Griffith fracture energy) and the unstable stacking fault energy is often... 

In addition, Fig. 28 also indicates an effect of the toughness of the adhesive bonds on the directional stability of cracks. When the toughness of the bond increases, all the curves will shift down vertically, indicating that the transition between the directionally stable and unstable crack is less likely to occur. This result is again consistent with the experimental observations discussed in Section 2.3.8, which showed that as the rubber concentration in the adhesive increases, the fracture toughness of the bonds increases and consequently, the transition from directionally stable cracks to directionally unstable cracks is more unlikely to occur. 

---