Craze fibrillation

The stress on the craze fibril directly ahead of the crack tip was assumed to be... 

Transmission electron microscopy (TEM) and birefringence studies of strained and/ or fractured epoxies have revealed more direct experimental evidence that molecular flow can occur in these glasses. Films of DGEBA-DETA ( 11 wt.- % DETA) epoxies, 1 pm thick, were strained directly in the electron microscope and the deformation processes were observed in bright-field TEM 73 110). Coarse craze fibrils yielded in-homogeneously by a process that involved the movement of indeformable 6-9 tan diameter, highly crosslinked molecular domains past one another. The material between such domains yielded and became thinner as plastic flow occurred. 

For the studied materials D0 amounts typically to 28 to 150 nm. A plot of smax as a function of creep stress a at a given temperature reveals three distinct regimes each with a straight slope (Fig. 12) the slope is related to the energy r necessary to create the craze fibril surface [60,61] ... 

When the craze propagates over a certain length, the fibril located in the central part (midrib) of the craze breaks, yielding a crack in the middle of the craze. Such a craze fibril breakdown also occurs in the craze ahead of a crack tip and results in a crack propagation. The broken down fibril parts retract on each crack surface and can be observed on fracture surfaces. The fibril breakdown mechanisms will be described later on in this section. 

When considering fracture behaviour of polymers, an important feature, as mentioned, deals with craze fibril breakdown. Indeed, this latter mechanism leads to crack propagation and easier specimen fracture. 

As mentioned, the molecular weight between entanglements, Me, or equivalently the entanglement density, ve, is involved in crazing, in craze fibril stability and, thus, in crack formation and propagation. 

In the higher temperature range, where a single craze develops at the crack tip, it has been shown [46] that a breaking time r of the craze fibrils under a stress, ac> can be defined from the craze length S and the crack-craze velocity vc by ... 

Besides, analysing the effect of MW of PMMA on the increase in craze dimensions with time [44], leads to the conclusion that the molecules in the craze fibrils are not fully stretched, retaining folds and thus keeping entangled. So, it supports the consideration of r(crc) as a disentanglement time. 

The low temperature range is characterised by a toughness independent of MW. It extends to a temperature where all the /3 transition motions are not still active. Consequently, the lack of chain mobility within the craze fibrils avoids chain slippage and the craze fibrils fail by chain scission. 

Cavitation is often a precursor to craze formation [20], an example of which is shown in Fig. 5 for bulk HDPE deformed at room temperature. It may be inferred from the micrograph that interlamellar cavitation occurs ahead of the craze tip, followed by simultaneous breakdown of the interlamellar material and separation and stretching of fibrils emanating from the dominant lamellae visible in the undeformed regions. The result is an interconnected network of cavities and craze fibrils with diameters of the order of 10 nm. This is at odds with the notion that craze fibrils in semicrystalline polymers deformed above Tg are coarser than in glassy polymers [20, 28], as well as with models for craze formation in which lamellar fragmentation constitutes an intermediate step [20, 29] but, as will be seen, it is difficult to generalise and a variety of mechanisms and structures is possible. 

Young s moduli to represent the anisotropic craze fibrils structure... 

We present the major results established in the description of crazing and the recent developments in this field. Crazing has been investigated within continuum or discrete approaches (e.g., spring networks or molecular dynamics calculations to model the craze fibrils), which have provided phenomenological or physically based descriptions. Both are included in the presentation of the crazing process, since they will provide the basis for the recent cohesive surface model used to represent crazing in a finite element analysis [20-22],... 

These observations appear to be in contradiction with a creep mechanism for craze fibrillation, and the currently accepted description refers to the drawing-in mechanism due to Kramer [31,32], Kramer argued that fibrillation takes place within a thin layer (about 50 nm) at the craze/bulk interface, in which the polymer deforms into highly stretched fibrils similar to the mechanism of drawing of polymer fibers, as illustrated in Fig. 2. Craze thick-... 

When the craze thickness attains the critical value A , craze fibrils break down and a microcrack nucleates with a related vanishing normal stress. 

Fig. 12 Instantaneous plastic deformation for the set of craze parameters B at loading rate iCj° 3 x 10-2 MPaVrn/s. a Prior to craze fibril breakdown b, c during crack propagation, with K / (so /r ) 1.32 (from [22])...

---