**3. Experimental Results**

In the SHPB tests, the striker bar was launched by varying the gas pressure to achieve different average nominal strain rates. Figure 4 shows the stress–strain curves of the FFRCs under quasi-static and SHPB compression experiments. The experimental results exhibited good repeatability (see Figure 4a–e). In addition, Figure 4f shows that FFRC is strongly affected by strain rates. The yield stress, as well as the flow stress, of the FFRCs markedly increased with an increase in strain rate. Therefore, the FFRCs showed evident strain rate sensitivity. This finding was highly similar to that reported by Omar et al. [27].

**Figure 4.** Stress–strain curves at different strain rates: (**a**) 0.006 s<sup>−</sup>1, (**b**) 0.06 s<sup>−</sup>1, (**c**) 0.6 s<sup>−</sup>1, (**d**) 1300 s<sup>−</sup>1, (**e**) 2200 s<sup>−</sup>1. (**f**) Comparison of typical curves.

The representative stress–strain curve in Figure 5 shows the trend of this stress–strain curve can be divided into three distinct stages—elastic region (oa), yield stage (bc), and plastic stage (cd)—similar to that of metal [25]; however, the specimen is made of a fiber-reinforced polymer material. The yield

strength, *σ*1, of the representative curves were extracted from the figure and listed in Table 1, consistent with the previous study (100–200 MPa) [33]. Evidently, *σ*1 increased with an increase in strain rate. For example, when the nominal strain rate reached 1300 s<sup>−</sup>1, *σ*1 markedly increased to 152 MPa, which was about 1.5 times higher than that of 0.006 s<sup>−</sup>1. The increase in yield strength from 1300 s<sup>−</sup><sup>1</sup> to 2200 s<sup>−</sup><sup>1</sup> was 11.7 MPa.

**Figure 5.** Different stages of a typical stress–strain curve: elastic region (oa), yield stage (bc), and plastic stage (cd).

**Table 1.** Yield strength of representative curves.


In addition, Young's modulus exhibited an appreciably increasing trend with an increase in strain rate, which was consistent with reference [27]. They attributed the increase in stiffening to the increase in strain rate, thereby decreasing the molecular mobility of polymer chains. However, dynamic Young's modulus could not be accurately measured by SHPB [34]. In the current study, the value of Young's modulus was not given to avoid inconsistency.

Figure 6a–c shows the final deformation morphologies of the crushed specimens. At a lower strain rate, the specimen only exhibits a reduction in thickness, where no obvious damage in appearance is observed. In the case of 1300 s<sup>−</sup>1, the margin of the specimen was damaged to a certain extent and compression deformation was evident. In the case of 2200 s<sup>−</sup>1, the specimen broke into two fragments, along with small cracks, and the fracture angle was approximately 45◦ (Figure 6c). The magnitude of the shear fracture angle mainly depended on the interfacial bond strength: a strong interface resulted in a larger shear fracture angle, whereas a weak interface generated a small fracture angle [27]. The failure could be inferred to have been initiated by matrix plasticity, followed by cracks passing through the layers of the laminate and forming a shear fracture with an angle of 45◦.

The microscopic failure mechanism was analyzed by scanning electron microscopy (SEM, Zeiss Auriga, manufactured in Oberkochen, Germany) to highlight the dominant failure modes at selected locations on the specimens. Prior to SEM observation, the specimens were coated with an ion sputter coater to obtain enhanced conductance. Figure 6d–f presents micrographs of the fractured surface of the crushed specimen at the strain rate of 2200 s<sup>−</sup>1. Fiber pull-out from the matrix is clearly shown in Figure 6d. Almost no matrix residue could be found on the surface of the fibers (Figure 6d,e). This observation could be attributed to the poor adhesion between the hydrophilic flax fiber and the hydrophobic epoxy matrix [18,35]. As seen in Figure 6e, a crack occurs along the fiber's longitudinal

direction, which could be attributed to the shear failure of the fiber when the matrix fractured. The crack also indicated a reduction in the shear strength of the flax fiber. In Figure 6f, flax fragments were stuck to the matrix after fiber pull-out, and superficial flax shavings exhibit partly separated from the fibers—that is, not completely from the reinforcements. This occurrence was highly consistent with the observation of Liang et al. [35] that this could be considered as an additional type of damage mechanism for NFRCs.

**Figure 6.** Specimens after experiments (**a**) 0.6 s<sup>−</sup>1, (**b**) 1300 s<sup>−</sup>1, (**c**) 2200 s<sup>−</sup>1; Micrographs of fractured edges: (**d**) fiber breakage and fiber pull-out, (**e**) fiber–matrix debonding and crack along the fiber, (**f**) superficial fiber shavings.
