It has long been recognized that many epoxy resins offering a high glass transition temperature are inherently brittle materials absorbing insignificant amounts of energy during the fracture process. In an attempt to trigger energy-dissipating mechanisms in brittle epoxy resins, a number of workers have investigated the potential benefits of incorporating small inclusions such as rubber particles [1-5], thermoplastic spheres [6-8] and, more recently, core-shell particles consisting of a rubbery core surrounded by a thermoplastic shell [9]. Core-shell modified epoxies have recently attracted interest since an unusual craze-like damage has been observed in such systems [9]. At present, however, it is not clear if similar toughening mechanisms operate in highly crosslinked epoxy resins similar to those currently used in the aerospace industry. A number of workers have shown that the degree of toughness enhancement in a particle-modified epoxy depends upon the crosslink density of the network structure [10-12]. It is generally recognized that epoxy resins which have a high crosslink density, and therefore a high glass transition temperature Tg, are more difficult to toughen than their low crosslink density counterparts. This difficulty in achieving high toughness in certain epoxies appears to be related to the materials' inability to undergo appreciable plastic flow in high stress fields. The objective of this work was to investigate the influence of crosslink density of the toughness and the failure mechanisms in a series of core-shell modified epoxy resins. The materials examined and the curing procedures employed in this study are summarized in Table I. Two epoxide phenolic novolac resins, Araldite ® EPNl138 and Araldite®EPNl139, and two hardeners, bisphenol A (4,4'-isopropylidenediphenol) and DDS (4,4'-diamino diphenyl sulphone), were used. The bisphenol A cured systems resulted in materials A, B and C and the DDS-cured systems in materials D and E. The average crosslink density within each of the five epoxy systems was determined from rubber elasticity theory following the procedure used by Fischer [13], as well as from stoichiometry. The glass transition temperature of each polymer was measured by differential scanning calorimetry (DSC) at a heating rate of 20 Kmin -1. The core-shell modifier was supplied in powder form consisting of an acrylic/styrene shell and a butadiene-styrene core. The diameters of the particles lay in the range 0.1-0.2~tm. The modifier was added to the pre-polymer and stirred with a high speed mixer prior to the addition of the hardener. In general, a good distribution of the particles was achieved although large agglomerations were occasionally observed. For simplicity, the core shell modified systems are prefixed by the letter M. The mode I critical strain energy release rate Gic for each material was determined using the compact tension specimen geometry in accordance with the
Read full abstract