Influence of Epoxy Resin Treatment on the Mechanical and Tribological Properties of Hemp-Fiber-Reinforced Plant-Derived Polyamide 1010 Biomass Composites

2.2. Morphology of HF/PA1010 Biomass Composites

This section discusses the morphologies of various surface-treated HF/PA1010 biomass composites to clarify the internal microstructures of these biomass composites, such as the interfacial interaction between HF and PA1010 and fiber dispersion in these biomass composites. We carried out the scanning electron microscope (SEM) observation of the fractured surface cryogenically in liquid nitrogen. Figure 2 shows the SEM photographs of cryogenically fractured surfaces of various surface-treated HF/PA1010 biomass composites: HF_NaClO₂ (Figure 2a), HF_NaClO₂ + EP (Figure 2b), HF_NaClO₂ + A-1160 (Figure 2c), and HF_NaClO₂ + A-1160 + EP (Figure 2d). In terms of morphology, HF_NaClO₂ (Figure 2a) had a relatively clean and smooth fiber surface, suggesting poor chemical and physical interaction between HF and PA1010. On the other hand, the morphologies of HF_NaClO₂ +EP (Figure 2b), HF_NaClO₂ + A-1160 (Figure 2c), and HF_NaClO₂ + A-1160 + EP (Figure 2d) demonstrated good interaction between HF and PA1010, and there were no voids at the interfaces between the fibers and the matrix polymer. In particular, there was more residual polymer on the fiber surface of HF_NaClO₂ + A-1160 + EP (Figure 2d) than that of HF_NaClO₂ + EP (Figure 2b) and that of HF_NaClO₂ + A-1160 (Figure 2c). Thus, these results are supported by the fact that surface treatment using epoxy resin after combining NaClO₂ alkaline treatment and the use of an ureidosilane coupling agent (HF_NaClO₂ + A-1160 + EP) generates good interfacial interaction between HF and PA1010. This may be due to chemical reactions among the epoxy groups in the epoxy resin, the ureido groups in the ureidosilane coupling agent on the alkaline-treated HF, and the highly reactive functional groups such as amine and carboxyl end groups in the molecular chain of polyamide.

Next, Figure 3 shows the SEM photographs of the polished cross-sectional surface of various surface-treated HF/PA1010 biomass composites: HF_NaClO₂ (Figure 3a), HF_NaClO₂ + EP (Figure 3b), HF_NaClO₂ + A-1160 (Figure 3c), and HF_NaClO₂ + A-1160 + EP (Figure 3d). These SEM photographs show the differences in the fiber dispersion in the composites. The dispersion of HF in the composites improved with some types of surface treatments in the following order: HF_NaClO₂ (Figure 3a) < HF_NaClO₂ + EP (Figure 3b) < HF_NaClO₂ + A-1160 (Figure 3c) < HF_NaClO₂ + A-1160 + EP (Figure 3d). In particular, individual separation and good dispersion of HF were seen in the morphology of the polished cross-sectional surface of HF_NaClO₂ + A-1160 + EP (Figure 3d). There was also no agglomeration of fibers, which showed that surface treatment using epoxy resin after surface treatment by combining NaClO₂ alkaline treatment and the use of A-1160 improves the dispersion of fibers. This dispersion of HF is related to the change in the interfacial interaction between HF and PA1010 when melt-mixing using a twin-screw extruder according to the surface treatment of HF is performed. Especially, surface treatment using epoxy resin after surface treatment combining NaClO₂ alkaline treatment and the use of A-1160 (NaClO₂ + A-1160 + EP) is expected to cause stronger interfacial interaction between HF and PA1010 than other surface treatments.

2.3. Thermogravimetric Analysis

To evaluate the influence of surface treatment on the thermal stability of various surface-treated HF/PA1010 biomass composites, thermogravimetric analysis (TGA) of these biomass composites is discussed in this section. Figure 4 shows weight as a function of temperature T (TG curve) of various surface-treated HF/PA1010 biomass composites. As reported in our previous paper [24], the TG curves of various surface-treated HF/PA1010 biomass composites provide evidence of two types of weight loss. The first weight loss occurs between 80 °C and 200 °C due to the dehydration of HF as well as the thermal degradation of lignin and hemicellulose [14,24]. In particular, the first weight loss of HF_NaClO₂ was the greatest in the various surface-treated HF/PA1010 biomass composites used in this study. Although alkaline treatment using NaClO₂ leaves some HF sub-component residues such as lignin and hemicellulose, surface treatments using an ureidosilane coupling agent (A-1160) or epoxy resin covered the HF with a coating and inhibited the thermal degradation of lignin and hemicellulose. On the other hand, the second weight loss occurred at more than 300 °C due to the decomposition of cellulose in HF. The details of the temperature in the case of 10 and 15 wt % loss of various surface-treated HF/PA1010 biomass composites are listed in Table 1. The temperature at 10 and 15 wt % of various surface-treated HF/PA1010 biomass composites increased in the following order: HF_NaClO₂ < HF_NaClO₂ + EP < HF_NaClO₂ + A-1160 < HF_NaClO₂ + A-1160 + EP. These behaviors may be attributed to the fact that HFs are protected by A-1160 surface treatment.

Lu et al. reported similar tendencies, for example, the thermal stability of silane-treated hemp-fiber-reinforced polyethylene composites is higher than that of untreated and alkaline-treated ones [57]. In particular, they concluded that not only does silane treatment modify the surface of hemp but when silane-treated fibers come in contact with the polymer matrix, the organo-functional group of the silane molecule couples with the matrix and increases the strength of bonding. Moreover, the enhanced covalent bonding along with the physical compatibility with the polymer matrix increases the thermal stability of the resultant composites.

Therefore, in this study, surface treatment using an ureidosilane coupling agent (A-1160) increased the interfacial interaction between HF and PA1010 and was thermally more stable than that without interfacial interaction generated by A-1160. These results indicate why pre-treatment combining NaClO₂ alkaline treatment and the use of A-1160 is essential before surface treatment using epoxy resin.

2.4. Mechanical Properties

This section discusses the influence of epoxy resin treatment on mechanical properties, such as tensile, bending, and dynamic mechanical properties, of HF/PA1010 biomass composites. Figure 5 shows the tensile properties, namely tensile strength σ_t (Figure 5a), tensile modulus E_t (Figure 5b), and elongation at break ε_t (Figure 5c), of various surface-treated HF/PA1010 biomass composites. The σ_t and E_t of HF/PA1010 biomass composites improved with the surface treatment of the fiber using an ureidosilane coupling agent (A-1160) or epoxy resin (EP), although the improvement effects differed according to the type of fiber surface treatment performed. The σ_t and E_t of various surface-treated HF/PA1010 biomass composites increased in the following order: HF_NaClO₂ < HF_NaClO₂ + EP < HF_NaClO₂ + A-1160 < HF_NaClO₂ + A-1160 + EP. On the other hand, the ε_t of all HF/PA1010 biomass composites was more or less the same level, although the values of ε_t differed slightly according to the type of fiber surface treatment performed.

Figure 6 shows the bending properties, such as bending strength σ_b (Figure 6a) and bending modulus E_b (Figure 6b), of various surface-treated HF/PA1010 biomass composites. The σ_b and E_b of various surface-treated HF/PA1010 biomass composites had different tendencies from tensile properties and increased in the following order: HF_NaClO₂ < HF_NaClO₂ + A-1160 < HF_NaClO₂ + EP < HF_NaClO₂ + A-1160 + EP. In particular, the σ_b and E_b of HF_NaClO₂ + A-1160 + EP significantly improved compared to those of the other surface-treated HF/PA1010 biomass composites. The results for these static mechanical properties, such as tensile and bending properties, indicated that surface treatment using epoxy resin after pretreatment combining NaClO₂ alkaline treatment and the use of an ureidosilane coupling agent (NaClO₂ + A-1160) is the most effective surface treatment for enhancing mechanical properties such as the strength and rigidity, of HF/PA1010 biomass composites. This may be due to the change in the interfacial interactions between HF and PA1010 according to the type of surface treatment performed.

Specifically, these interfacial interactions are chemical reactions and physical interactions between HF and PA1010 surface-treated by HF_NaClO₂ + A-1160 + EP based on the following mechanisms:

1. NaClO₂ alkaline treatment removes sub-components such as lignin, wax, and hemicellulose from the HF surface, increases the surface roughness, and, moreover, generates hydroxyl groups.

2. The silane coupling agent generates alkoxysilane groups, which after hydrolysis are capable of reacting with OH-rich fiber surfaces [58]. Therefore, after HF is treated by NaClO₂, the condensation reaction between the silanol groups of the ureidosilane coupling agent (A-1160) and the hydroxyl groups on the surface of HF can occur more easily.

3. Ureido groups in A-1160 can cause hydrogen bonding with hydroxyl groups and epoxy groups in epoxy resin. This reinforces the interfacial interaction between A-1160 and epoxy resin, enhancing the compatibility between two phases.

4. Chemical reactions between epoxy groups in epoxy resin and highly reactive functional groups such as amine and carboxyl end groups in the molecular chain of polyamides can occur more easily.

Therefore, we found that epoxy resin treatment significantly improves mechanical properties such as strength and modulus by varying the interfacial interaction between HF and PA1010 based on the mechanisms mentioned above. On the other hand, the difference in the effects of surface treatment on tensile and bending test results is due to the different deformation modes in each test. Specifically, while the tensile stress against the force applied in the same direction as the strain is measured in the tensile test, the bending stress, which is a complex stress consisting of tensile and compressive stress against a force generated by the three-point bending method, is measured in the bending test. In addition, different test speeds and test geometries (thickness) were used in the two tests. The tensile and bending test results showed different tendencies, but it is difficult to determine the detailed reasons at this point.

The dynamic viscoelastic properties of various surface-treated HF/PA1010 biomass composites in the solid state were determined by dynamic mechanical analysis (DMA) experiments. The storage modulus E’ and the loss tangent tanδ (=loss modulus E”/storage modulus E’) were plotted as functions of temperature T of various surface-treated HF/PA1010 biomass composites, as shown in Figure 7a (E’ vs. T) and Figure 7b (tanδ vs. T), respectively. The E’ of HF_NaClO₂ + A-1160 + EP was the highest among the storage moduli of all surface-treated HF/PA1010 biomass composites in all temperature regions. Thus, a higher E’ indicates a higher mechanical rigidity of HF_NaClO₂ + A-1160 + EP than other surface-treated HF/PA1010 biomass composites. There was a significant decrease in the temperature in the regions between 40 °C and 70 °C, which is related to the glass transition temperature T_g of PA1010. These softening temperatures shifted slightly toward a high temperature according to the type of fiber surface treatment carried out and increased in the following order: HF_NaClO₂ < HF_NaClO₂ + EP < HF_NaClO₂ + A-1160 < HF_NaClO₂ + A-1160 + EP. These shifts may be attributed to the change in the molecular chain mobility in the amorphous segment of the matrix polymer according to the surface treatment using epoxy resin. Various studies on the DMA of surface-treated natural-fiber-reinforced biomass polymer composites have reported these shifts according to whether surface treatment was used or not used [18,24,59,60,61,62]. On the other hand, the tanδ curves in Figure 7b have two relaxation peaks. The first peak, between 40 °C and 70 °C, indicates α relaxation, which is related to the T_g of PA1010 and the composites mentioned earlier. On the contrary, the second peak, between −90 °C and −60 °C, indicates the relaxation of the hydrogen bonds between the PA1010 chains [24,63]. This second peak in the lower-temperature region shifted slightly toward higher temperatures with the surface treatment of fibers, due to restricted movement of the polymer chains.

From the results of the comparison of static and dynamic mechanical properties of various surface-treated HF/PA1010 biomass composites, it can be concluded that surface treatment using epoxy resin after pre-treatment by NaClO₂ alkaline treatment and the use of an ureidosilane coupling agent (HF_NaClO₂ + A-1160 + EP) is the most effective surface treatment method for enhancing the dynamic mechanical properties of these biomass composites. These findings are consistent with the observed morphologies of composites, such as SEM observations of cryogenically fractured surfaces (Figure 2) and cross-sectional polished surfaces (Figure 3), as mentioned earlier. The high strength and modulus of HF/PA1010 biomass composites with surface treatment using epoxy resin are considered to be due to the change in the interfacial interaction between HF and PA1010 and the morphological change in the fiber dispersion in the composites.

2.5. Tribological Properties

This section discusses the influence of surface treatment using epoxy resin on tribological properties using a ring-on-plate-type sliding wear tester under constant load and constant velocity and under dry conditions of HF/PA1010 biomass composites against the metal counterpart (carbon steel S45C). Figure 8 shows the relationship between the frictional coefficient µ and the specific wear rate V_s of various surface-treated HF/PA1010 biomass composites. The test conditions were a constant load P of 50 N, a constant velocity v of 0.2 m/s, and a sliding distance L of 600 m. µ was calculated as the average value from 400 m to 600 m, and V_s was calculated as the difference in the mass loss of the plate specimens made of polymer composites before and after testing. In this figure, friction and wear decreased toward the bottom left, indicating that high performance is required for application to tribomaterials. The µ of various surface-treated HF/PA1010 biomass composites decreased in the following order: HF_NaClO₂ > HF_NaClO₂ + EP > HF_NaClO₂ + A-1160 + EP > HF_NaClO₂ + A-1160. On the other hand, the V_s of these biomass composites decreased in the following order: HF_NaClO₂ > HF_NaClO₂ + A-1160 > HF_NaClO₂ + EP > HF_NaClO₂ + A-1160 + EP. Thus, the V_s of the composites has different tendencies from that of µ. These results may be attributed to the change in the mode of friction and wear mechanism according to the type of surface treatment of HF. It is rational to conclude that the tribological properties of these biomass composites are strongly influenced by the surface treatment of the fiber. In particular, the wear properties of these biomass composites notably improve with surface treatment using epoxy resin. The tribological behavior of polymer composites is strongly influenced by their ability to form transfer films on the metallic counterface and by the worn surfaces of materials and wear debris. Thus, it is essential to study these factors to understand the friction and wear mechanisms of the polymer composites [12,64,65].

Figure 9 shows the SEM photographs of the metallic counterface (carbon steel S45C) before and after sliding wear tests of various surface-treated HF/PA1010 biomass composites. Specifically, Figure 9a shows the surface before sliding, Figure 9b shows the surface after sliding for HF_NaClO₂, Figure 9c for HF_NaClO₂ + EP, Figure 9d for HF_NaClO₂ + A-1160, and Figure 9e for HF_NaClO₂ + A-1160 + EP. After the sliding wear tests, the metallic counterfaces showed differences in the polymeric transfer films formed on the counterfaces according to the type of HF surface treatment carried out. These polymeric transfer films are thin and have smooth surfaces. The apparent thickness of the transfer film can be determined from the appearance of the scratch marks left by the polishing paper on the metallic counterface to keep the surface roughness (R_a = 0.4 µm) constant before the sliding wear tests. Specifically, the thinner these transfer films are, the clearer are the scratch marks on the metallic counterface; the thicker these films are, the less obvious are the scratches. Their thickness decreased in the following order: HF_NaClO₂ (Figure 9b) > HF_NaClO₂ + EP (Figure 9c) > HF_NaClO₂ + A-1160 + EP (Figure 9e) > HF_NaClO₂ + A-1160 (Figure 9d). Small, thin scratch marks could also be seen on the whole surface of the counterfaces for all the biomass composites. These scratches were found to keep the surface roughness constant before the sliding wear tests (Figure 9a). Moreover, there was also considerable debris on the counterfaces of HF_NaClO₂ (Figure 9b), HF_NaClO₂ + EP (Figure 9c), and HF_NaClO₂ + A-1160 + EP (Figure 9e) because these biomass composites have relatively high frictional coefficients. On the contrary, there was no wear debris on the counterface of the HF_NaClO₂ + A-1160 biomass composite (Figure 9d), which has a low µ.

Figure 10 shows SEM photographs of wear debris after the sliding wear tests of various surface-treated HF/PA1010 biomass composites: HF_NaClO₂ (Figure 10a), HF_NaClO₂ + EP (Figure 10b), HF_NaClO₂ + A-1160 (Figure 10c), and HF_NaClO₂ + A-1160 + EP (Figure 10d). The wear debris of various surface-treated HF/PA1010 biomass composites was collected from the sliding surface after the sliding wear test. The shape and size of the wear debris differed according to the type of surface treatment of HF. The debris of HF_NaClO₂ (Figure 10a) was a mixture of many filaments and some big, flaky particles. The debris of HF_NaClO₂ + A-1160 (Figure 10c) was a mixture of small filaments and many big, flaky particles. On the contrary, the debris of HF_NaClO₂ + EP (Figure 10b) had more flaky particles but less filamentary ones. Moreover, the debris of HF_NaClO₂ + A-1160 + EP (Figure 10d) consisted of only big and flaky particles, without filamentary ones. Therefore, the shape of the wear debris of HF/PA1010 biomass composites changed from filamentary ones to flaky ones when surface treatment was performed using an ureidosilane coupling agent and epoxy resin. Furthermore, the size of the wear debris of these surface-treated biomass composites increased to larger flaky particles than those of HF_NaClO₂. These differences may support the assumption that the change in the mode of friction and wear mechanism according to the type of surface treatment of HF is caused by good interfacial interaction between HF and PA1010 and good dispersion of fiber in the composites.

Figure 11 shows the three-dimensional profiles of the worn surfaces of polymer composite specimens after the sliding wear tests of the carbon steel S45C of various surface-treated HF/PA1010 biomass composites: HF_NaClO₂ (Figure 11a), HF_NaClO₂ + EP (Figure 11b), HF_NaClO₂ + A-1160 (Figure 11c), and HF_NaClO₂ + A-1160 + EP (Figure 11d). The depth of the wear tracks and the roughness of the worn surface of these biomass composites decreased in the following order: HF_NaClO₂ (Figure 11a) > HF_NaClO₂ + A-1160 (Figure 11c) > HF_NaClO₂ + EP (Figure 11c) > HF_NaClO₂ + A-1160 + EP (Figure 11d). Figure 12 shows the SEM photographs of the worn surface of biomass polymer composites: HF_NaClO₂ (Figure 12a), HF_NaClO₂ + EP (Figure 12b), HF_NaClO₂ + A-1160 (Figure 12c), and HF_NaClO₂ + A-1160 + EP (Figure 12d). The morphologies of the worn surfaces of these biomass composites changed according to the type of surface treatment of HF. The worn surfaces of HF_NaClO₂ (Figure 11a and Figure 12a) had very rough and bumpy faces. In particular, there were two high-wear walls, where the wear debris accumulated, at the boundary between the worn surface and the non-worn surface (Figure 11a). Moreover, there were countless small wear-and-tear marks, and HF could not be clearly seen on the worn surface of HF_NaClO₂ (Figure 12a). On the other hand, the worn surfaces of HF_NaClO₂ + A-1160 (Figure 11c and Figure 12c) were smoother than that of HF_NaClO₂. HF could be clearly seen on the surface of HF_NaClO₂ + A-1160, although the HFs in HF_NaClO₂ + A-1160 were damaged and the surfaces showed many frictional marks (Figure 12c). The worn surface of HF_NaClO₂ + A-1160 showed slight fiber debonding with PA1010, which is a gap in the interface between HF and PA1010. On the contrary, the worn surfaces of HF_NaClO₂ + EP (Figure 11b and Figure 12b) and HF_NaClO₂ + A-1160 + EP (Figure 11d and Figure 12d) showed very smooth surfaces. In particular, HF could also be clearly seen on the surface of HF_NaClO₂ + A-1160 + EP, and it was not severely damaged. Especially, the fibers of HF_NaClO₂ + A-1160 + EP did bond with PA1010 and fiber breakage occurred. These differences may be due to the change in the mode of friction and wear mechanism according to the type of surface treatment of HF carried out or due to improvement in mechanical properties as a result of the strong interfacial interaction between HF and PA1010 and good fiber dispersion in the composites in surface treatment using epoxy resin after pre-treatment by NaClO₂ alkaline treatment and the use of an ureidosilane coupling agent. The results of the observation of the worn surfaces are supported by the significantly improved specific wear rate.

3.2. Processing

HF/PA1010 biomass composites were extruded using a twin-screw extruder and injection-molded. PA1010 and various surface-treated HFs were dried in a vacuum oven and then dry-blended in a bottle, and subsequently, the melt was mixed at 220 °C and 85 rpm using a twin-screw extruder (TEX30HSS, Japan Steel Works, Ltd., Tokyo, Japan). Test pieces for various experiments were injection-molded by an injection-molding machine (NEX30IV-2EG, Nissei Plastic Industrial, Co. Ltd., Nagano, Japan). The molding conditions were as follows: cylinder temperature of 220 °C, mold (cavity) temperature of 30 °C, and injection rate of 13 cm³/s. The test pieces were kept for at least 24 h at 23 °C in desiccators after injection molding to maintain them in a dry condition in all the measurements.

We would like to thank the Functional Microstructured Surfaces Research Center (FMS, MEXT, Japan) of Kogakuin University for funding some of the experimental apparatus.