No abstract

The effects of such parameters as the filler volume fraction, particle size, aspect ratio, modulus and strength of filler, resin‐filler adhesion, and toughness of the matrix on the stiffness, strength, and toughness of particulate‐filled epoxy resins were evaluated. The mechanisms of crack initiation and subsequent crack propagation in these multiphase materials are discussed and illustrated by scanning electron microscopy of fracture surfaces.

No abstract

Existing methods for the measurement of crack position (and hence crack velocity) are briefly reviewed. Those methods are principally based on the optical, mechanical, electrical or acoustic behaviour of a cracking sample, or on an inspection of the fracture surfaces. None of the above techniques fulfills all conditions required to measure continually slow and fast crack growth in polymers up to 250 °C, the only method capable of doing this is a potentiometric gauge without support. The development and application of such a graphite gauge to crack propagation in polymers is described in this paper. The experimental arrangement in general and the results obtained from "instrumented samples" in Charpy impact tests and from crack propagation across an interface between dissimilar materials are discussed and the advantages (large sensitivity, range of crack speeds from 10 7 to 103 msec 1) and disadvantages (gauge is also sensitive to plastic deformations) are indicated.

Encouraged by the respective needs of the aerospace, nuclear, and automotive industries, much attention is at present being given to the problem of dynamic fracture in brittle materials (see for example [1][2][3][4][5][6]). At present, however, considerable disagreement and controversy still remain regarding the precise mechanisms responsible for accelerating a crack up to speeds of several hundreds of metres per second.Although most of this work is directed primarily towards fracture in metals, testing has generally been carried out on polymeric materials such as Homolite I00 and Araldite B since they offer superior photoelastic properties [2][3][4][5]. Currently, most workers supvort the view that a unique relationship exists between the dynamic stress intensity factor Kid and the crack velocity v [1][2][3][4], suggesting that the dynamic fracture process can be characterised in the form of a Kid-V diagram as shown for H0molite i00 in Fig. 1 [5]. Generally, such figures exhibit two quite distinct regions. At lower velocities, the slope of the curve is effectively infinite with a wide range of velocities achieved for one single value of Kid. Conversely, in the upper part of the curve significant changes in Kid frequently render little or no change in the velocity of the propagating crack. The uniqueness of such Kid-V relationships is further questioned when differing test configurations are employed. Kobayashi et al. [2] have shown that by varying the geometry of the specimen it is possible to shift the vertical part of the K_d-V quite considerably, suggesting such a relation-± ship is not an inherent basic material property. Other workers [8] have sought to explain the dynamic fracture process by considering the conditions local to the notch at the moment of instability.Here, it is suggested that large values of the critical strain energy release rate, Gic , precipitate rapid crack accelerations and higher crack velocities.In another publication [9], the present authors found that the fast fracture surface of a silica-filled epoxy resin consisted of a smooth, featureless region superceded by a very rough three dimensional zone. Upon correlating this fracture surface morphology with the corresponding crack velocity profile, it was found that the smooth region coincided with the region of crack acceleration and the rough zone corresponded with the region where the crack had achieved its limiting velocity and was not, therefore, capable of accelerating further.Consequently, it was suggested that over the smooth region all the excess elastic energy was dissipated in accelerating the crack to greater velocities. Once the crack had achieved its limiting velocity, it was impossible to accelerate it further and therefore another dissipating mechanism was sought, namely, crack branching.Subsequently, with the use of a photo-microscope and an image analyser, a unique relationship was found between the fracture surface roughness and the stored elastic energy at rupture for the wide range of test conditions examined (temp...