Characterization of the Strain Rate-Dependent Deformation Response and Fracture Behaviour of a Three-Part Snap-Cure Epoxy Resin Under Tension and Compression Loading

Abstract

The recent development of snap-curing epoxy resins and rapid manufacturing processes has led to increased usage of fiber-reinforced plastic composite materials in high-volume production vehicles. For energy absorbing composite structures, characterizing the epoxy strain rate-dependent response and local failure behaviour is critical to assess their crashworthiness performance. The current study investigated the tensile and compressive deformation response of a three-part snap-cure epoxy resin over a range of strain rates (i.e., ${10}^{-4} {\text{to}} {10}^{3} {{\text{s}}}^{-1}$) using a suite of testing apparatuses. The average tensile elastic modulus, yield strength, and ultimate strength increased by 32%, 55%, and 49%, respectively, with increasing strain rate over the range considered. Tensile specimens fractured in a brittle manner despite deforming plastically, while fracture surface morphology was slightly influenced by the strain rate. The compressive elastic modulus was insensitive to increasing strain rate, while the compressive yield strength increased by 81% over the strain rates considered and was higher than the tensile yield strength at similar strain rates. This study addresses a critical gap in the literature by providing a comprehensive data set for a snap-cure epoxy material, which will support future development of a virtual multiscale-modeling framework aimed at predicting impact performance of composite automotive structures.

Appendices

Appendix A

The cylindrical and dog-bone specimens shown in Table 1 for the tensile tests were analyzed using the commercial Finite Element (FE) software, Abaqus (Dassault Systèmes, Vélizy-Villacoublay, France). FE models were constructed for both specimen geometries (Fig. 17), employing the Johnson–Cook plasticity material model calibrated with quasi-static tensile test data presented in this paper. Mesh sensitivity was investigated, and a 0.2 mm mesh size was chosen for the models. Boundary conditions were set to emulate quasi-static tensile tests, applying displacement to the reference point, and corresponding reaction forces were calculated through simulations. The extension of the specimens was measured in the gauge section, mirroring the experimental digital image correlation (DIC) analysis. Numerical simulation results indicated that the stress and strain curves for the two specimen types differed by less than 1% (Fig. 18). Consequently, the impact of the two specimen geometries on the tensile response of the polymer is deemed to be negligible.

Fig. 17

FE models and simulation results for the cylindrical and rectangular specimens subjected to quasi-static tensile loading

Fig. 18

Calculated engineering stress-train curves for the indicated specimens under quasi-static tensile loading

Appendix B

To firmly mount the specimens onto the polymer bars (minimize the free movement of the specimen), a tapered thread clamping fixture was designed and manufactured (refer to Fig. 19). The clamping fixture consists of four parts: one RH connector (linked to the RH male thread side of the specimen), one LH connector (linked to the LH male thread side of the specimen), and two RH female thread rings. The downside of employing the fixture is an elevation in geometric impedance and mass. In comparison to a solid bar, the fixture exhibits an 8% increase in mass.

Fig. 19

Custom clamping fixture for the tension SHPB specimens