RESULTS

Overall tensile mechanical properties

The mechanical properties of all specimen groups determined by uniaxial tensile testing are summarized in Table 2. One-way ANOVA revealed significant between-group differences in all five mechanical parameters: Emod (F(7,32) = 9.253, p < 0.001), yield strength (σy) (F(7,32) = 7.198, p < 0.001), yield strain (εy) (F(7,32) = 5.200, p < 0.001), ultimate stress (σu) (F(7,32) = 18.635, p < 0.001), and ultimate strain (εu) (F(7,32) = 36.411, p < 0.001). Normality of residuals was violated for four of the five parameters (Shapiro–Wilk test, p < 0.05), and homogeneity of variance was violated for three of the five parameters (Levene's test, p < 0.05); however, one-way ANOVA was retained for all parameters because it is considered robust to moderate violations of these assumptions when group sizes are approximately equal (Supplementary Table S1). The results of post hoc pairwise comparisons using Tukey's HSD test are presented in Figure 3. The coefficient of variation for Emod was consistently lower in the epoxy groups (1.4%–4.5%) than in the polyester groups (5.9%–14.5%), and this pattern persisted across both unreinforced and all fiber-reinforced groups (Table 2).

Effect of fiber reinforcement on E and R composites

In the absence of fiber reinforcement, R exhibited significantly higher yield strength (11.51 ± 3.97 MPa; d = 2.011, p < 0.001) and yield strain (1.12 ± 0.23%; d = 2.171, p < 0.001) than E (6.02 ± 0.88 MPa and 0.75 ± 0.10%, respectively). However, Emod, ultimate stress (σu), and ultimate strain (εu) did not differ significantly between the two unreinforced resins (p > 0.05 for all). The maximum fiber weight fraction at which hand-pour casting into the silicone mold remained feasible was 5 wt%.

No. | Group | n | Emod (MPa) | σy (MPa) | εy (%) | σu (MPa) | εu (%) | Coefficient of variation for Emod (%)
1 | E | 6† | 1,099.52 ± 48.62 | 6.02 ± 0.88 | 0.75 ± 0.10 | 33.17 ± 5.24 | 5.96 ± 0.79 | 4.4
2 | E+1%RF | 5 | 1,234.84 ± 24.43 | 9.28 ± 1.07 | 0.95 ± 0.08 | 18.52 ± 2.22 | 2.69 ± 0.44 | 2.0
3 | E+3%RF | 5 | 1,333.06 ± 18.42 | 10.38 ± 2.02 | 0.98 ± 0.14 | 15.78 ± 2.87 | 1.93 ± 0.55 | 1.4
4 | E+5%RF | 4‡ | 1,513.12 ± 67.85 | 12.37 ± 0.59 | 1.02 ± 0.06 | 17.71 ± 2.27 | 1.66 ± 0.35 | 4.5
5 | R | 5 | 1,234.28 ± 179.32 | 11.51 ± 3.97 | 1.12 ± 0.23 | 31.28 ± 8.92 | 4.82 ± 1.08 | 14.5
6 | R+1%RF | 5 | 1,387.52 ± 103.47 | 10.71 ± 1.00 | 0.99 ± 0.13 | 14.75 ± 2.36 | 1.67 ± 0.48 | 7.5
7 | R+3%RF | 5 | 1,387.24 ± 81.92 | 11.93 ± 1.50 | 1.06 ± 0.12 | 14.19 ± 2.00 | 1.39 ± 0.27 | 5.9
8 | R+5%RF | 5 | 1,397.44 ± 125.41 | 8.32 ± 1.02 | 0.80 ± 0.04 | 12.89 ± 2.00 | 1.73 ± 0.81 | 9.0
9 | PU | 5 | 57.72 ± 7.69 | 1.14 ± 0.25 | 2.17 ± 0.24 | 1.76 ± 0.38 | 4.04 ± 1.26 | 13.3

Figure 3: Tensile mechanical properties of epoxy and polyester resin composites across fiber reinforcement concentrations. (A) Elastic modulus. (B) Yield strength. (C) Ultimate stress. (D) Yield strain. (E) Ultimate strain for each specimen group. ns = p ≥ 0.05; * = p < 0.05; ** = p < 0.01; *** = p < 0.001.

When reinforced, the Emod of E increased with increasing fiber weight fraction: E+1%RF (1,234.84 ± 24.43 MPa), E+3%RF (1,333.06 ± 18.42 MPa), and E+5%RF (1,513.12 ± 67.85 MPa). Significant differences from unreinforced E (1,099.52 ± 48.62 MPa) were detected at 3 wt% (d = −6.103, p < 0.01) and 5 wt% (d = −7.307, p < 0.001), representing a 37.6% increase at the highest concentration. The difference at 1 wt% did not reach statistical significance (p > 0.05). No significant change in Emod was observed across any fiber weight fraction in R (p > 0.05 for all pairwise comparisons; Figure 3A; Supplementary Table S2).

Yield strength increased significantly in epoxy composites at 3 wt% (10.38 ± 2.02 MPa; d = −2.914, p < 0.01) and 5 wt% (12.37 ± 0.59 MPa; d = −8.076, p < 0.001) relative to unreinforced E (6.02 ± 0.88 MPa); however, the difference at 1 wt% (9.28 ± 1.07 MPa) was not statistically significant (p > 0.05). In polyester composites, no significant pairwise difference in yield strength was detected among any of the four fiber weight fractions (p > 0.05 for all). E+5%RF exhibited significantly higher yield strength than R+5%RF (12.37 ± 0.59 MPa vs. 8.32 ± 1.02 MPa; d = 4.710, p < 0.05; Figure 3B).

Ultimate stress decreased significantly with fiber addition in both resin systems relative to their respective unreinforced controls (Figure 3C). For epoxy composites, significant reductions from unreinforced E (33.17 ± 5.24 MPa) were observed at 1 wt% (18.52 ± 2.22 MPa; d = 3.505, p < 0.001), 3 wt% (15.78 ± 2.87 MPa; d = 3.996, p < 0.001), and 5 wt% (17.71 ± 2.27 MPa; d = 3.534, p < 0.001). For polyester composites, significant reductions from unreinforced R (31.28 ± 8.92 MPa) were observed at 1 wt% (14.75 ± 2.36 MPa; d = 2.535, p < 0.001), 3 wt% (14.19 ± 2.00 MPa; d = 2.645, p < 0.001), and 5 wt% (12.89 ± 2.00 MPa; d = 2.847, p < 0.001).

Among the five mechanical parameters, yield strain (εy) showed the least sensitivity to fiber reinforcement. In epoxy composites, only E+5%RF (1.02 ± 0.06%) differed significantly from unreinforced E (0.75 ± 0.10%; d = −3.129, p < 0.05); differences at 1 wt% and 3 wt% were not significant (p > 0.05). In polyester composites, only R+5%RF (0.80 ± 0.04%) differed significantly from unreinforced R (1.12 ± 0.23%; d = 1.903, p < 0.01); differences at 1 wt% and 3 wt% were not significant (p > 0.05; Figure 3D).

Ultimate strain decreased significantly with fiber addition at all three concentrations in both resin systems (Figure 3E). For epoxy composites, reductions from unreinforced E (5.96 ± 0.79%) were significant at 1 wt% (2.69 ± 0.44%; d = 4.960, p < 0.001), 3 wt% (1.93 ± 0.55%; d = 5.790, p < 0.001), and 5 wt% (1.66 ± 0.35%; d = 6.488, p < 0.001). For polyester composites, reductions from unreinforced R (4.82 ± 1.08%) were significant at 1 wt% (1.67 ± 0.48%; d = 3.771, p < 0.001), 3 wt% (1.39 ± 0.27%; d = 4.363, p < 0.001), and 5 wt% (1.73 ± 0.81%; d = 3.230, p < 0.001).

Stress–strain behavior of the resin composites

The stress-strain curves for all specimen groups are shown in Figure 4. Curves for all groups exhibited an initial toe region followed by a linear elastic region and abrupt brittle failure. Unreinforced E specimens displayed a two-slope response, characterized by a steeper initial elastic region transitioning to a reduced-slope region before failure, which is consistent with the 0.2% offset method used for yield determination. Unreinforced R specimens showed a more continuously curvilinear ascent with greater inter-specimen variation (coefficient of variation for Emod = 14.5%; Table 2). With increasing fiber weight fraction in both resin systems, the post-toe response became progressively steeper and more linear, with correspondingly more abrupt failure. All fiber-reinforced specimens failed within the gauge section by abrupt brittle fracture (representative failure shown in Figure 2B), whereas unreinforced epoxy and polyester specimens exhibited visible localized deformation at the fracture plane before failure.

Mechanical comparison between E+3%RF and PU foam

The mold-constrained apparent density of the PU foam specimens was 280 ± 4 kg/m³ (mean ± SD; n = 5). Welch's independent-samples t-test detected significant differences between PU foam and E+3%RF across all five mechanical parameters (Figure 5). Elastic modulus (Emod) (57.72 ± 7.69 MPa; t(5.35) = 142.891, d = 90.372, p < 0.001), yield strength (σy) (1.14 ± 0.25 MPa; t(4.12) = 10.174, d = 6.435, p < 0.001), and ultimate stress (σu) (1.76 ± 0.38 MPa; t(4.14) = 10.817, d = 6.841, p < 0.001) were all substantially lower in PU foam than in E+3%RF (1,333.06 ± 18.42 MPa, 10.38 ± 2.02 MPa, and 15.78 ± 2.87 MPa, respectively). Yield strain (εy) (2.17 ± 0.24%; t(6.55) = −9.436, d = −5.968, p < 0.001) and ultimate strain (εu) (4.04 ± 1.26%; t(5.46) = −3.427, d = −2.167, p < 0.05) were significantly higher in PU foam than in E+3%RF (0.98 ± 0.14% and 1.93 ± 0.55%, respectively).