Polystyrene craze studies

The structure of crazes in bulk specimens was studied by Kambour [15], who used the critical angle for total reflection at the craze/polymer interface to determine the reliactive index of the craze, and showed that the craze was roughly 50 per cent polymer and 50 per cent void. Another investigation involved transmission electron microscopy of polystyrene crazes impregnated with an iodine-sulphur eutectic to maintain the craze in its extended state [33, 34]. The structure of the craze was clearly revealed as fibrils separated by the voids that are responsible for the overall low density. 

A quantitative analysis of craze shape and mass thickness contrast within the craze allowed Lauterwasser and Kramer [382] to derive the stress profile existing along a polystyrene craze. Kramer and his coworkers have extended this study to many other polymers, relating the mean density of craze material to entanglement density in the polymer glass and to toughness [395] without a basic change of preparation technique. 

Kambour et al. performed extensive studies on the mechanisms of plasticization [18-25]. The correlation observed between the critical strain to craze and the extent of the glass-transition temperature (Tg) depression speaks strongly in favor of a mechanism of easier chain motion and hence easier void formation. In various studies on polycarbonate [19,24], polyphenylene oxide [20], polysulfone [21], polystyrene [22], and polyetherimide [25], Kambour and coauthors showed that the absorption of solvent and accompanying reduction in the polymer s glass-transition temperature could be correlated with a propensity for stress cracking. The experiments, performed over a wide range of polymer-solvent systems, allowed Kambour to observe that the critical strain to craze or crack was least in those systems where the polymer and the solvent had similar solubility values. The Hildebrand solubility parameter S [26] is defined as... 

Bucknall and Smith (17) concluded that crazing is the dominant mechanism to toughen high impact polystyrene and related polymers. One important function of the rubber particles is to serve as craze initiators and stabilizers in the glassy matrix. However Newman and Strella (18) concluded from optical microscope studies that cold drawing is responsible for toughness in ABS. 

The ESCR performance of a resin is not easily modeled. A laboratory technique for the preparation of thin films of HIPS materials for the study of deformation processes by microscopy allows the deformation process to be better understood. The transmission electron microscope (TEM) allows direct visualization of the crazes themselves in thin films. For good contrast between the crazes and the bulk polystyrene, thin, cast films from 0.5 to 2 p,m are required, and also staining of the rubber phase with a heavy atomic species to provide contrast between the rubber and the polystyrene. Another intricacy of this method requires a solution of the HIPS material in a 65 35 methyl ethyl ketone-toluene solution to prevent significant swelling of the rubber particles during the preparation process. 

Early in the study of crazing it was recognized that there was usually a time delay between the application of stress and the visual appearance of crazes. This delay time is evidence of a barrier to craze nucleation. The experimental situation may be suimnarized by Figure 1, which shows the density of crazes as a function of time under various states of stress in polystyrene (PS) from the work of Argon and... 

In contrast to observations in polystyrene, we do not observe permanent bands our specimens exhibit no residual birefringence upon release from stress. Neither do we observe crazing before failure. However, the specimens do whiten just before failure when viewed edge-on, and this whitening disappears within a few seconds after fracture occurs. We think the oscillations in intensity we observe are likely to be due to incipient shear deformation which disappears after specimen failure. Unpublished results of other workers are reported (see References 11 and 14 in the present Reference 12) to be consistent with the idea that such bands should be difficult to observe in PMMA and in polycarbonate because of their lower draw ratios compared to polystyrene. Studies of an unfilled epoxy polymer (14) in cyclic tensile deformation indicate that shear bands do not remain after removal of stress until a threshold amplitude of deformation is exceeded. 

Thus crazing has been directly identified in typical rubber-modified plastics in a plane perpendicular to the direction of the applied stress and initiated at the rubber-matrix interface. These findings are quite consistent with macroscopic model studies performed by Matsuo et al. (1972), who studied the behavior of polystyrene-containing rubber balls (as an analog of ABS morphology). It was found that equatorial crazes developed in tension, as expected, and also that stress field interaction occurred when the balls were close together, resulting in a heavier craze density between the balls. This shows that a principal role of rubber particles is to induce many... 

Kuboky and co-workers [40] used transmission electron microscopy (TEM) to study block and white crazes in high impact polystyrene (PS). They examined the mechanism of block craze formation and found that the rubber molecules were not necessarily diffused into the entire crazes. The length of the block crazes varied before they turned to white and in some cases only white crazes were generated from the rubber particles. TEM should thus be used with caution in examining stained rubber-toughened polymers to ensure that all crazes, including the white crazes, were considered for evaluation of the extent of the deformation behaviour [41]. 

Recently, Brown and Kramer have reported a study of the rise in stress after changing the environment of crazed polystyrene specimens under load from methanol, water or their mixtures to air [12]. They derived an equation relating the change in the surface component of the stress to the surface tension and the craze fibril geometry. 

The microstructure observed for thick films shows fibrils, about 4-10 nm in diameter for polystyrene, in agreement with SAXS measurements on the crazes in the bulk polymer. Very thin films of polystyrene (100 nm) show modification in the craze structure as there is no plastic restraint normal to the film [397]. Deformation zones have also been studied in polycarbonate, polystyrene-acrylonitrile and other polymers [398]. Crazes in thermosets can be studied in thin films spun onto NaCl substrates which can be washed away when the film has been cured. Mass thickness measurements are difficult to make in radiation sensitive materials that is why most TEM work has been done on polystyrene and least on PMMA. After developing the techniques described above for TEM Donald and Kramer [398] applied similar methods in optical microscopy to study radiation sensitive materials and the kinetics and growth of deformation zones. Thin films were strained on grids in situ in a reflecting OM. Change of interference color, which depends on the film thickness, was a very sensitive method for observing film deformation. 

A study of a PS-PB block copolymer showed variation in craze behavior as a result of rubber particles added to modify the otherwise brittle, glassy polymers. Such copolymers were studied under the high strain of physical laboratory testing where the polybutadiene in the copol)mier was stained with osmium tetroxide prior to microtomy [333]. The brittle behavior of the glassy polymers was shown by TEM and STEM to be modified by the rubber particles which provide toughening by control of the craze behavior. In a study of the craze behavior in isotactic polystyrene [146], films of polystyrene were drawn from dichlorobenzene solution and cast onto glass microscope slides, followed by... 

---