Tarfaoui Mostapha
ENSIETA/MSN, 2, rue Frangois Verny, 29806 Brest cedex 9
France
1. Introduction
The choice of composite materials as a substitute for metallic materials in high technological applications such as in the marine field is becoming more pronounced especially due to the great weight-savings these materials offer. In many of these practical situations, the structures are subjected to high impact loads like slamming, impact, underwater explosions or blast effects. Material and structural response vary significantly under impact loading conditions compared to static loading. The mechanical characteristics of these materials are well known for static loading; however, with the strain rate they are likely to evolve (Goldsmith et al., 1995; Shi et al., 1993; Tsai & Sun, 2004; Tsai & Sun, 2005; Gillespie et al., 2005; Brara & Klepaczko, 2007). The behaviour of structures subjected to impact has been of interest to many scientists for design purposes as well as for the purpose of developing constitutive models of the materials tested [7-8]. The study of the composite materials behaviour at high strain rates is still relatively new and reliable data on strain rate effects is very scarce. Even though the problem of obtaining reliable data is accentuated by difficulties encountered in design and conducting impact tests on composites [9], the qualitative relationship between the dynamic constitutive response and the dynamic damage evolution for composites at high strain rates is still far from being fully understood. To investigate the rate-dependent constitutive relations of materials at high strain rates, the Split Hopkinson Pressure Bar (SHPB) technique has been extensively accepted [10]. Experience of the use of SHPB for the investigation of metals has led to the adaptation of this technique for the characterization of laminated polymer composites at medium strain rates. Significant efforts have been made to examine the high strain rate behaviour of more brittle materials such as composites and ceramics using the split Hopkinson bar to measure dynamic response of materials under varying loading conditions (Kumar et al., 1986); (El-Habak, 1991), (Harding, 1993), (Sierakowski & Nevill, 1971). Ochola et al. (2004) studied the strain rate sensitivity of both carbon fibre reinforced polymer (CFRP) and glass fibre reinforced polymer (GFRP). The results show that the dynamic material strength for GFRP increases with increasing strain rates and the failure strain for both CFRP and GFRP is seen to decrease with increasing strain rates. Vinson & Woldensenbet’s (2001) results show that the ultimate stress
increases with increasing strain rate. Most recently, the study conducted by Hosur et al. (2004) presents the effect of in-plane off-axis testing of an 8-harness satin weave carbon fabric/SC15 composite specimen. The specimens were tested in the in-plane direction of 0°, 15°, 30°, 45°, 60°, 75°, and 90° in a range of strain rates from 1092 to 2425 s-1. From this study it was noted that the high strain rate-tested specimens showed a considerable increase in the stress to failure and stiffness of the composite compared with the quasistatic loaded specimens. Depending on the fibre orientation of the specimens, the ultimate strength and strain varied considerably and exhibited a nonlinear stress-strain response that increased with angles up to 45°. Gary & Zhao (2000) employed the use of low impedance materials such as nylon, for the incident and output bars of the split Hopkinson bar, to test the strain rate behaviour of glass epoxy composite panels. The failure strength of the glass epoxy panel tested by Gary and Zhao is reported to be strain rate sensitive. Fibre orientation effects on high strain rate properties were considered recently for a carbon epoxy system. Tsai & Sun (2002) have reported the difference between tensile and compressive behaviours in a unidirectional glass fibre-reinforced composite, and developed a nonlinear rate-dependent viscoplasticity model to characterize its compressive stress-strain relationship. Many different models (Gama et al., 2001), (Haque & Ali, 2005), (Vinson. & Woldesenbet, 2001) have been developed to predict failure stress and modes in composites subjected to quasi-static loading. However, few criteria have been developed and experimentally validated for high strain rate loading.
In this study, specimens of glass/epoxy composite were subjected to static and dynamic compression loading. Quasi-static tests were conducted on an Instron universal machine to evaluate the elastic properties and quasi-static response, while the split-Hopkinson pressure bar (SHPB) is used for dynamic tests. Samples were tested in-plane and out of plane direction. The fibre orientations of the samples were 0°, ±20°, ±30°, ±45°, ±60°, ±70° and 90°. Stress-strain curves at increasing strain rates were obtained for different cases. However, no experimental data for the intermediate range of strain rates between (80s-1 to 300s-1) was obtained, because the Instron universal testing machine and the SHPB employed in the experimental tests are designed respectively for low and high strain rates. Off-axis composites and angle-ply laminates exhibited significant nonlinear and strain-dependent behaviour. Finally, experimental observations enable us to draw up a history of dynamic damage in the specimens according to fibre orientation and load direction.
