Thermomechanical experiment and analysis on shape recovery properties of shape memory polymer influenced by fiber reinforcement

Abstract

Shape memory polymers (SMPs) are currently investigated as potential materials for large deployable space structures [1–3]. The thermomechanical properties of these polymers significantly change on reaching their glass transition temperature, which yields the excellent feature of shape fixity and shape recovery [4]. As another aspect, the modulus of these materials is not sufficient since they are polymeric materials. In actual applications, the fiber reinforcement is effective for ensuring the sustainability of the deployed structures. However, while the fiber reinforcement has advantages for increasing the stiffness, it has a negative influence on the shape recovery behavior of SMPs. Experimental studies for shape memory polymer composite were conducted by Gall et al. [5, 6] for SiC powder-reinforced nanocomposite, Ohki et al. [7] for short-glass-fiber reinforcement, and Lan et al. [8] for SMP reinforced with plain-weave fabrics, and the increase of residual strain after shape recovery process was confirmed. In order to maximize both the stiffness and shape-recovery behavior of fiber-reinforced SMPs, it is essential to know how the fibers block the shape recovery behavior of polymers. However, the mechanism was not modeled. Therefore, we focus on the mechanism underlying the degradation of shape recovery behavior due to fiber reinforcement. To investigate this mechanism, we first conducted thermomechanical cycle tests for pure SMP and SMP reinforced with short-carbon fibers. We used polyurethane series of thermoset SMP, Diary MP-5510 (curing temperature 100 C), provided by SMP Technologies Inc. The cured polymer has a glass transition temperature of 55 C, and the temperature range of glass transition between glassy state and rubbery state is 30 C. Short carbon fibers T700S (Toray Industries Inc.) were embedded in the polymer. The fibers with approximately 5 mm length were randomly embedded in the prepolymer before curing, and the weight fraction of fibers was set to 0, 2, and 4 wt%. It should be noted that the bundle of carbon fibers (12,000 fibers) was embedded and not dispersed in these model experiments, as also modeled later. After curing, strip specimens were cut out (approximately 40 mm gauge length, 20 mm width, and 2 mm thickness in average). The following thermomechanical cycle was applied to these specimens using a tensile test machine (INSTRON 5566), as shown in Fig. 1: