Interlaminar fracture toughness and the associated fracture behavior for glass fiber reinforced polymers (GFRP)

Abstract

Laminated fiber reinforced polymers (FRP) have excellent in-plane stiffness and strength, but suffer from transverse impact loading due to the lack of fiber reinforcement in that direction. Under transverse impact, extensive delamination can occur in the interlaminar resin-rich regions, which leads to a reduction of stiffness and strength. Therefore, many researchers have attempted to establish a standard testing method that characterizes the interlaminar fracture toughness for delamination [1]. In addition to the measurement of the delamination toughness, analysis of the fracture surface using optical microscopy or scanning electron microscopy (SEM) can also provide useful information for the delamination resistance characterization. Recently, we have evaluated mechanical properties of two glass-fiber-reinforced polymers (GFRP) [2], using transverse impact, double cantilever beam (DCB) and end-notched flexure (ENF) tests. Results from the transverse impact and DCB tests showed a clear difference in delamination resistance between the two GFRP, but such a difference was not found from the ENF test. Based on the ENF test results, the two GFRP should have similar delamination resistance in the shear mode (Mode II). Although this discrepancy in delamination resistance may be caused by the toughness variation in different modes of loading, we believe that the discrepancy is mainly due to a problem with the ENF test for the characterization of the delamination resistance. This issue has been raised in several studies [3, 4], and is further investigated here through the examination of fracture surfaces to provide additional evidence. Two GFRP used in this study have polymer matrices that are the same as those used previously, that is, isophthalic polyester (TMR300 isopolyester, provided by Viking Plastics, Edmonton) and polyurethane-based resin (PUL-G polyurethane, provided by Resin System Inc., Edmonton). The PUL-G polyurethane resin contains 15% CaCO3 particulates for stiffness enhancement. For convenience, the two GFRP will be named PI-GFRP and PU-GFRP in this paper, for isopolyesterbased GFRP and polyurethane-based GFRP, respectively. The glass fiber used is 9-oz/yd2 warp, unidirectional woven fabric (provided by ZCL Composites, Edmonton) which consists of parallel fiber bundles stitched together with a gap of around 1 mm. The gap between the fiber bundles generated resin-rich zones in the GFRP. Therefore, the GFRP have resin-rich zones in the intra-laminar, inter-fiber-bundle regions and the interlaminar regions. GFRP plates of nominally 6.0 mm thick were fabricated using a resin transfer molding (RTM) technique. The fiber lay-up of the transverse impact specimens is [(0/90)5]s, and that of the DCB and ENF specimens is [(0/90)402]s of which the four central 0-degree layers were to provide a uni-directional fiber environment in the direction of crack growth, in accordance with most of the test coupons used in the past. The specimens for DCB and ENF tests contain an aluminum insert film, 15 μm thick, that is placed between the 2nd and the 3rd of the four central layers, acting as a starting defect for delamination crack growth. Dimensions of the specimens are 93 × 93 mm2 for the transverse impact test, and 120 × 20 mm2 for the DCB and ENF tests. Overall fiber volume fraction of the specimens is around 40%, estimated using the following equation [5]: