C. P. Bosnyak

The effect of a process zone on the fracture path in a complex stress field

The importance of the process zone (also called the damage zone, plastic zone or active zone) has been recognized since the works of Irwin [1], Orowan [2] and the celebrated models of Dugdale [3] and Barenblatt [4]. Recently, a thermodynamic analysis of the crack and process zone growth as an evolution of one system called the Crack Layer has been advanced [5-9]. The Crack Layer analysis is quite tedious, particularly due to the complexity of the crack tip field resulting from the crack-process zone interaction problem [10]. In the present work the effect of the process zone on the fracture path is analyzed experimentally using a new technique based on controlled exposure of a poly(ethylene-co-carbon monoxide), ECO, to ultraviolet radiation, (UV). UV exposure to ECO results in hardening and embrittlement of the sample [11]. A comparison of the findings in this work with previous studies is also presented [ 12].

An energy analysis of crack-initiation and arrest in epoxy

The objective of this work is to study fracture processes such as crack initiation and arrest in epoxy. A compact tension specimen with displacement-controlled loading is employed to observe multiple crack initiations and arrests. The energy release rate at crack initiation is significantly higher than that at crack arrest, as has been observed elsewhere. In this study the difference between these energy release rates is found to depend on specimen size (scale effect), and is quantitatively related to the fracture surface morphology. The scale effect, similar to that in strength theory, is conventionally attributed to the statistics of defects which control the fracture process. Triangular shaped ripples, deltoids, are formed on the fracture surface of the epoxy during the slow sub-critical crack growth, prior to the smooth mirror-like surface characteristic of fast cracks. The deltoids are complimentary on the two crack faces which excludes any inelastic deformation from consideration. The deltoids are analogous to the ripples created on a river surface downstream from a small obstacle. However, in spite of our expectation based on this analogy and the observed scale effect, there are no ‘defects’ at the apex of the deltoids detectable down to the 0.1 micron level. This suggests that the formation of deltoids during the slow process of sub-critical crack growth is an intrinsic feature of the fracture process itself, triggered by inhomogeneity of material on a sub-micron scale. This inhomogeneity may be related to a fluctuation in the cross-link density of the epoxy.

Design of impact modifiers for thermoplastic polymers based on micromechanics

Efficient impact modifiers for lowering the ductile-brittle transition temperature of thermoplastic blends have been designed by modeling the stress distribution near the notch of an Izod impact test sample and the nature of stresses in spherical particle-filled polycarbonate. The model considers the inhomogeneity of a soft phase inside a relatively rigid phase, particle interaction, and the effects of thermal residual stresses imposed as a consequence of processing and differences in matrix and particle thermal coefficients of expansion. Polycarbonate blends are used as an example for the modeling. The predictions of the ductile-brittle transition temperature of blends provide guidelines for selection of impact modifier type. The model predicts that there is no further advantage in toughening by increasing the ratio of the moduli of matrix and rubber particle more than 1000. The model also predicts that the glass-rubber transition temperature, Tg, and the nature of transition (i.e., sharp or smooth transition) dominate the ductile-brittle transition temperature of blends. An energy criterion for yielding is proposed to be an improved necessary condition for the yielding of polymer instead of the Von Mises stress-yielding criterion. The energy creterion can be used to predict an optimal volume fraction of rubber particles for ductility.

Observations of the micro-mechanisms of fatigue-crack initiation in polycarbonate

The mechanisms of crack initiation in tensile fatigue of single-edge notched specimens of polycarbonate of varying thickness have been elucidated. At low stresses and long times microcracking and localized yielding occurred to form regular diamond-shaped cells on a scale of 2–4 Μm. On increasing the stress level with thin specimens (<1 mm), the microshear bands coalesced to form macroscopic damage zones of yielded material around the notch, followed by crack tearing from the notch surface. With increasing specimen thickness, restriction of shear banding ensued and a stable, semi-elliptical cavitation, or pop-in, formed about 10–100 Μm ahead of the notch, dependent on specimen geometry. As a result, the ligament formed between the notch and pop-in consists of yielded material. Brittle behaviour resulted with further increases in specimen thickness on loading, i.e. when the ligament could not be stabilized.

Real time study of failure events in polymers

Experimental methods have been developed so that in situ transmission electron microscope (TEM) tensile studies can be performed on bulk polymer sections, and failure processes observed; real time can be correlated with failure in bulk parts. Using specially designed support grids, polymer section geometry and in situ tensile procedures, the submicrometre failure response of polycarbonate-poly(ethylene terephthalate) phase morphology to crack propagation has been studied. This paper focuses on the design of the tensile grids, sections and procedures, which had to be devised for these studies. The techniques developed allow quantification of strain rates and crack velocities. TEM experiments performed showed that artefacts, such as vacuum or radiation damage, were not significant factors influencing the morphological response to crack propagation. A companion paper presents the failure processes found in situ and correlations with failure processes found in bulk tested parts.