3.1. Izod Impact
Table 1 shows all the results obtained for the Izod impact test of the different composites materials and neat epoxy resin.
The values found are similar to those already reported in the literature for other NLFs composites [
26,
27]. The specimens of neat epoxy resin,
Figure 3a, and composites with 10 vol% fiber,
Figure 3b, were all completely fractured and, thus, were validated by the standard.
Specimens of 20 and 30 vol% fiber,
Figure 3b,c, respectively, showed incomplete fracture. However, the fibers were broken and pulled out. Although the ASTM D256 [
21] standard invalidates the test in the absence of total fracture, there are indications that the energy supported by the composite was equal to or greater than that recorded by the pendulum. In other words, as there was no complete rupture, the absorbed energy was not enough to fracture the composite.
The results in
Table 1 can be better visualized in the graph in
Figure 4. With the increase in the fiber fraction, it is possible to notice an improvement in the impact energy, as already observed for other NFLs [
27,
28]. The high dispersion of the values given by the error bars associated with the fiber is also a known characteristic of NLFs, since they have a great heterogeneity [
18,
29].
When compared to the average values obtained for the neat epoxy resin, the effect of incorporating 30% by volume of caranan fibers as reinforcement in composites was evident, producing an increase of 637.30%.
A statistical analysis of ANOVA,
Table 2, was performed to compare the averages obtained and to verify if there was a significant difference in energy absorption between them.
Therefore, with a 95% confidence level (or 5% significance level), the hypothesis that the treatments have equal means is rejected, because the Fcalc was much higher than the Ftab (critical).
Table 3 shows the results of Tukey’s test. The calculated HSD was 48.36 J/m, and, thus, the differences above the HSD were considered significant. These values are marked in bold and showed that the impact strength of the 30 vol% caranan fiber composites was better than all other tested specimens.
To better understand the behavior of the composites analyzed in ballistic impact,
Figure 5 presents SEM images of broken epoxy-caranan (30 vol%) specimens.
In these images, different fracture mechanisms can be observed. One might note,
Figure 5, that cracks (resin failure) are formed in the matrix and their paths are blocked and interrupted by the fibers. Additionally, in this SEM, good fiber adhesion can be seen at the fiber-matrix interface. Similar behavior was reported by Costa et al. [
30] and Junio et al. [
16]. Other failure mechanisms that are influenced by fibers are also associated with the fragility of the polymer matrix failure mechanism. These fracture mechanisms are more complex and include the rupture of caranan fibers. Their pullout is evidenced by the circular holes shown in the fractography,
Figure 5b. These results can be related to the absorption of high impact energies, shown in
Table 1 and
Figure 4.
Figure 6 shows a comparison between good and poor fiber adhesion to the resin. To the left of the fiber, when the fiber-matrix interface is not adequate, the matrix rupture can be observed. This failure mechanism is not observed on the right side of the fiber. The difference in the nature of the polymer matrix and natural fiber explains this behavior. While caranan fibers, such as other NLFs, exhibit a hydrophilic nature, the epoxy resin has a hydrophobic character. This difference in nature impairs the interfacial adhesion of the reinforcement in the matrix, facilitating the delamination mechanism. Additionally, river marks are stopped when approaching the fiber. The sum of all these mechanisms directly contributes to the increase in energy absorption of the composite [
13,
31], showing the reinforcement of the caranan fiber in the epoxy matrix.
Due to the heterogeneity of natural materials, the same fiber might have different characteristics. This can occur for several reasons, among them we can highlight plant cultivation, plant age, fiber roughness, surface flaws, diameter variation, climate, extraction procedures and soil. Such factors may have influenced the results related to the fiber/matrix interface [
32].
Pullout testing was performed to characterize the caranan fiber/epoxy bonding and revealed a rather high interfacial shear strength of 17 MPa. As a result, one should anticipate good fiber/matrix adhesion and the possibility of caranan fiber reinforcing behavior [
19].
3.2. Ballistic Tests
In order to estimate the ballistic behavior of the 30 vol% composite of caranan fiber, a residual velocities test was performed.
Table 4 presents the absorption energy values for each one of the ballistic shots performed on the plate. During the tests, all the specimens were perforated so their residual velocities could be measured.
When considering the post-impact aspect of the plate, i.e., the physical integrity of the composite after the ten shots,
Figure 7, the sample did not fracture. Indeed, a very important criterion for application in ballistic protection [
33,
34] is the ability of the plate to receive projectile impact without disintegration.
Table 5 presents the average values of the composite mass
mc, projectile mass
mp, average impact velocity
Vi, average residual velocity
Vr) and absorption energy
Eabs of each composition (C30%—composite reinforced with 30 vol% of fiber).
Based on the results obtained, it can be inferred that the neat epoxy resin presenting the highest energy of absorption failed to resist an expressive number of shots, quickly suffering a rupture [
14]. For application in ballistic shielding, this is a negative aspect because target integrity is one of the evaluation criteria for effective protection. The higher
Eabs value observed for neat epoxy may be associated with its fragility, which tends to dissipate energy by generating fractured surfaces. This can be considered an indication that the fiber reinforcement was not carried out effectively or with adequate volume [
35].
For comparison, some limited velocity results, found in the literature for shooting tests with a 0.22-inch caliber, are presented in
Table 6. It is noteworthy that these results were found for shots with a compressed air apparatus and not with a firearm. The composites of the present work presented similar results of
VL when compared to the other related NFL.
After the ballistic impact, the brittle behavior of the epoxy matrix can be verified in
Figure 8. Such behavior is evidenced by the appearance of river marks and cracks in the epoxy matrix [
28,
36,
37,
38]. River marks on the surface usually means restricted plasticity on the crack tips and very quick crack propagation. However, no fiber failure is observed in this SEM, thus evidencing its reinforcement in the epoxy matrix by the caranan fiber.
Weibull statistical analysis,
Table 7, was also performed. It is possible to verify that the points are within the adjusted line, justifying the high value of R
2, above 0.9. In addition, it is worth noting that the characteristic value (
) is similar to the average found for
Eabs of the composite.
Figure 9 presents the graph of this statistical analysis for better visualization.
The results of this Weibull statistical analysis,
Figure 9, indicated the good reliability of the obtained results and revealed a homogeneous characteristic of the individual samples.