3.1. Extracted kenaf cellulose (KF)
The extracted cellulose fibers (KF) used in this study were verified via FTIR spectrum, as shown in Fig. 2(a). The spectrum show the characteristic peaks at 3300–3400 cm− 1, attributed to the stretching vibration of the OH groups of cellulose. The C-H stretching vibration at around 2800–2900 cm− 1, corresponding to the presence of C-H stretching. The peak at 1420–1430 cm− 1 and 1300–1400 cm− 1, are attributed to the symmetric bending of CH2 and the bending vibrations of the C-H and C-O groups of the aromatic rings in polysaccharide (Jonoobi et al. 2009; Kargarzadeh et al. 2012). The peak around the range of 1030–1150 cm− 1, are attributed to C-O stretching and C-H vibrations of the pyranose ring skeletal. The peak around 895 cm− 1, corresponds to COC, CCO, and CCH deformation and stretching vibrations. The FTIR spectrum show that the cellulose was successfully extracted from kenaf fiber.
Figure 2(b) shows the XRD intensities of our extracted kenaf cellulose fiber (Extracted KF) in comparison with commercially available cellulose fiber (Commercial Cellulose). Both cellulose samples presented typical semi-crystalline behavior, in which crystalline peaks at 2θ value of 22.5o were observed. The peaks obtained from the XRD intensities were the characteristic peaks for cellulose. The crystallinity index (CrI) of the commercial cellulose is about 77%, while the CrI of the extracted kenaf cellulose is relatively high at around 71%. Such high CrI of the extracted kenaf cellulose is due to the removal of amorphous hemicellulose and lignin during the bleaching and extraction process (Chan et al. 2013). The XRD results also proved that combined use of low NaOH concentration and relatively low temperature did not alter the cellulose structure compared to the commercially available cellulose.
Furthermore, from our previous study (Soatthiyanon et al. 2020), our extracted cellulose fibers (KF) were found to be decomposed at 337.3oC. The average widths and average length of the extracted kenaf cellulose fibers were 8.9 ± 2.1 µm and 101.7 ± 57.6 µm, respectively, with the aspect ratios of 11.5. The SEM image of the extracted kenaf cellulose fibers was shown in Sect. 3.5 Morphology, Fig. 10(a).
3.2. Thermal properties
The thermal properties of the neat PLA and all the biocomposites, including glass transition temperature (Tg), crystallization temperature (Tc), and melting temperature (Tm) of the PLA and the PLA/KF biocomposites, were characterized using the DSC measurement. The results from the second heating scan and first cooling cycles are shown in Fig. 3. The melting temperature (Tm) of the neat PLA and all the biocomposites remains almost constant at around 151oC, while the glass transition temperature (Tg) is slightly decreased for all the biocomposites with PEG incorporating. This may be attributed to the plasticizer effect of PEG, enhancing the PLA chains’ mobility. However, according to the DSC results, it could be concluded that the incorporation of KF, KFs, and PEG showed no significant effect on the melting and the glass transition temperatures of the obtained biocomposites.
For the PLA/KF and PLA/KFs biocomposites, the enhancement of Tc refers to the biocomposite filaments that are more difficult to undergo cold crystallization than the neat PLA filament and thus require higher temperature. As shown in Table 2, the crystallinity of the PLA biocomposites is gradually decreasing with a corresponding decrease in the enthalpy of crystallization as the KF and KFs are adding. However, the addition of PEG seems to diminish the effect of the KF and KFs on decreasing the crystallinity. The crystallinity of the PLA/PEG4/KFs biocomposite remains almost constant while the crystallinity increases for the PLA/PEG6/KFs biocomposite is approximately twofold increased compared to the neat PLA. This may be attributed to the combined effects of the PEG6000 and TEOS modification on improving the PLA chains' mobility, increasing the recrystallization ability (Murphy and Collins 2018).
Table 2
DSC results of PLA and PLA/kenaf cellulose biocomposites.
Sample | Tg [oC] | ΔHc [J/g] | Tc [oC] | ΔHm [J/g] | Tm [oC] | Xc [%] |
PLA | 59.7 | 16.83 | 122.2 | 17.35 | 150.8 | 0.56 |
PLA/KF | 59.5 | 7.28 | 126.2 | 7.51 | 151.3 | 0.27 |
PLA/KFs | 59.7 | 6.54 | 126.8 | 6.40 | 151.5 | - |
PLA/PEG4/KFs | 57.9 | 16.50 | 122.8 | 16.98 | 151.3 | 0.57 |
PLA/PEG6/KFs | 57.6 | 16.36 | 122.7 | 17.17 | 151.0 | 0.97 |
3.3. Rheological behaviors
Figure 4 shows the complex viscosity (ƞ*) as a function of angular frequency for the neat PLA and its biocomposites with the kenaf cellulose fibers at 190oC. The complex viscosity increases with the kenaf cellulose fiber addition. This may be attributed to the rigidity of the kenaf cellulose fiber restricted the capability to flow of the melted PLA chains. Remarkably, the surface modification of the kenaf cellulose fibers with TEOS leads to a decrease in the complex viscosity. This may result from the more hydrophobic surfaces of the KFs. The complex viscosities of the PLA/KFs biocomposite were almost similar to those of the neat PLA at all frequencies, which may be attributed to the better dispersion of the KFs in the PLA matrix comparing to the KF. Moreover, the addition of PEG led to a decrease in the complex viscosity since PEG acts as a plasticizer, enhancing the flowability of the composites. The complex viscosity of the PLA/PEG4/KFs composite was considerably lower than other materials. This might be because the smaller molecule of the PEG4 resulted in a lower viscosity as compare to the larger molecule of the PEG6.
Figure 5. (a) Storage modulus (G’) and (b) loss modulus (G”) of PLA and its biocomposite filaments at 190
oC as a function of angular frequency.
The modulus data is a factor to reveal the molecular rigidity. Figure 5 (a-b) shows the effect of the TEOS modified kenaf cellulose and the PEG addition on the evolution of the storage and loss moduli of the biocomposites. It is seen that the storage moduli increased with increasing the frequencies, which are consistent with the complex viscosity results. The addition of KF increased the storage moduli of the PLA matrix, indicating the effect of the rigidity of the kenaf cellulose fibers. The modified kenaf cellulose (KFs) showed a slight increase in the storage modulus compared to that of the neat PLA. The increase in the storage moduli was more noticeable for the PLA/KF as compared to the other biocomposites. Such an increase in storage moduli indicated that the addition of the KF and KFs resulted in prohibiting the deformation of the PLA chains. However, the presence of the plasticizers, PEG4 and PEG6, reduced the storage moduli at all frequencies. Furthermore, the addition of plasticizers improved the capability to flow of the biocomposites, at the same time, decreased stiffness.
The Cole-Cole plot is generally constructed from the variation in dynamic viscosities, η’ and η”. It is a useful tool to examine the mechanism of relaxation and homogeneity of polymer composites (Azizi and Ghasemi 2009; Manoharan et al. 2014; Wang et al. 2015; Walha et al. 2016; Aumnate et al. 2019). As shown in Fig. 6, the neat PLA and all the biocomposites present single relaxation behavior. The deviation of semicircular shape means the relaxation process of the PLA/kenaf cellulose biocomposites is affected by the type of the kenaf cellulose (KF and KFs) and the addition of the plasticizers (PEG4 and PEG6). It can be observed clearly that the PLA/KF biocomposite shows the longest relaxation time, in which the most deviation of the semicircular shape from the neat PLA is observed. This behavior indicates the heterogeneity and poor compatibility between the kenaf cellulose fibers (KF) and the PLA matrix. Thus, the motion of PLA chains is restrained significantly. The TEOS modified kenaf cellulose (KFs) enhanced the compatibility between such two phases, the modified kenaf cellulose fibers, and the PLA matrix, leading to improved PLA chain motion, as evidence by the smaller semicircular shape forming.
3.4. Optimization of extrusion-based 3D printing
In melt extrusion-based 3D printing, the crucial parameter to achieve high interfacial bonding between printed layers is the suitable extrusion temperature, in which the molten filament maintains this temperature in the most recently deposited layers. Knowing the extrusion temperature and flowability of the printed materials leads to more accurate and precise printing. The melt viscosity should be adequate to allow extrusion yet sufficient to provide structural support (Aumnate et al., 2018; Wang et al., 2017). Also, a high cooling rate leads to a higher and more uniform mechanical performance. Thus, tuning the cooling rates also matters.
The shear rate for the melt extrusion-based 3D printing processes, in particular the FDM system, is typically ranged between 100 to 200 s− 1 (Osswald et al. 2013; Aumnate et al. 2019; Spicker et al. 2019). Figure 7 shows the plot of steady shear viscosity versus shear rate for the neat PLA and its composites with the kenaf cellulose fibers. As expected, the viscosity of the biocomposite increases for the PLA/KFs composite. This may be due to the rigidity of the KFs interferes the mobility of the PLA chain. On the other hand, the viscosity of the PLA/KFs composites decreases with the PEG addition. The smaller molecule of PEG4, the lower the viscosity. The reduction in viscosity can be attributed to the plasticizing effect of the PEG, enhancing the flowability of the melt. According to the steady shear viscosity data, over the range of 100 to 200 s− 1, the PLA/PEG4KFs biocomposite filament is expected to flow most easily during 3D printing process.
Moreover, the cooling scan from the DSC measurement (see Fig. 1(b)) for the PLA and its biocomposite filaments gives the low bound for the extrusion temperature according to the biocomposite crystallization behavior. Changing in slope indicates the solidification of the materials. The results showed that PLA and all biocomposites were solidified at the temperature below 60°C. The temperature exceeding 60°C will prevent material solidification at the nozzle. Accordingly, a build plate temperature was set at 60°C for the FDM 3D printing test to diminish such shrinkage and warpage.
Figure 8 presents the 3D printed specimens of the PLA, PLA/KF, PLA/KFs, PLA/PEG4/KFs and PLA/PEG6/KFs biocomposites. All the samples were successfully 3D printed using the same printing conditions (see Table 1). The color of the sample deviated upon the material compositions in the biocomposites. The color of the printed biocomposites became lighter as the TEOS modification and plasticizers were presented.
Figure 9. SEM images of the printed samples from (a) PLA, (b) PLA/KF, (c) PLA/KFs, (d) PLA/PEG4/KFs and (e) PLA/PEG6/KFs biocomposites at 100x magnification.
Figure 9 exhibits the SEM micrographs of the fracture surfaces of the printed samples. It is seen that there are some voids appeared in all the samples. Such voids are common for the sample obtained from the melt extrusion based-3D printing due to the lack of molding pressure during the printing process. Moreover, many voids inside the deposited beads were noticed for the PLA/KF biocomposite, resulting from a weak fiber-matrix interfacial adhesion. The inter-bead voids decreased for the PLA/KFs, PLA/PEG4/KFs, and PLA/PEG6/KFs biocomposites. This suggests the improvement of interfacial compatibility between the KFs and the PLA matrix. This may be ascribed to the subsequent surface modifying kenaf cellulose with TEOS and incorporating the plasticizers, which can further enhance the fiber-matrix interfacial adhesion, thereby the mechanical performance improved.
3.5. Morphology
Figure 10 shows SEM micrographs of the KF and KFs at a magnification of 1000x. The surfaces of the KFs seemed to be smoother than those of the KF. This could be evidence of the presence of the TEOS layer that enveloped the fiber surfaces for the KFs.
The critical challenge to achieving the transfer of good mechanical properties of single-fiber nanocellulose to the macroscale properties of the polymer nanocomposites is also to obtain well-dispersed hydrophilic-reinforcing nanocellulose in the polymer matrices, but also optimization of the fiber-matrix interface and improved coupling of fiber and polymer for enhanced stress transfer.
Figure 11 shows SEM micrographs of a cross-sectional view of the tensile testing specimens for the neat PLA, PLA/KF, PLA/KFs, PLA/PEG4/KFs, and PLA/PEG6/KFs biocomposites, respectively. A dense structure can be observed for the neat PLA. There were two different failure modes observed from the test samples. The PLA and all the PLA biocomposite with the KFs depicted a ductile structure, as can be seen from the tiny fibrils that appeared in the images, while the PLA/KF biocomposite showed a brittle structure. Also, for the PLA/KF biocomposite, the fiber imprints and gaps between the fibers and the PLA matrix were noticed. This is due to the different polarities of the hydrophilic KF and the hydrophobic PLA matrix. These features indicate poor adhesion between KF and the PLA matrix. This negatively affected the tensile strengths of the biocomposites because the applied stress was challenging to transfer to the fibers. Meanwhile, compared to the PLA/KF biocomposite, the denser structure can be observed for the PLA/KFs, PLA/PEG4/KFs and PLA/PEG6/KFs biocomposites. It reveals that the KFs fibers are embedded and covered by the PLA matrix which suggested the compatibility between the KFs fibers and the PLA matrix. Also, the addition of PEG improves the dispersion of the KFs as can be seen from a smaller sizer of fiber bundle compared to the PLA/KFs biocomposite without PEG. The gaps between the fibers and the PLA matrix seemed to be diminished in the PLA/KFs biocomposite. This can be suggested that the modified kenaf cellulose surface with TEOS reduced the hydrophilicity of the kenaf cellulose fibers, thus increased the compatibility and interfacial adhesion between the kenaf cellulose fiber and the PLA matrix. The morphologies of the composites were found to be directly related to the tensile properties.
3.6. Tensile properties
Figure 12 shows the tensile strength of the PLA and its biocomposites. As expected, the addition of KF shows adverse effects on the mechanical properties. Due to the hydrophilic character of the cellulose, the poor interfacial adhesion and the poor dispersion of the KF in the PLA matrix can be promoted. It is seen that the addition of the KF significantly affected the tensile strength of the biocomposites, As expected, the Young’s modulus increased with the KF addition, which is due to the rigidity of the KF. However, the tensile strength and the elongation at break dropped significantly compared to the neat PLA. This can be attributed to the poor adhesion and poor compatibility between hydrophilic KF and the PLA matrix. Consequently, the weak spots would be generated during the tensile test. The TEOS surface modification seem to help on improving the tensile strength and the elongation of the biocomposites. As shown in Fig. 12(d), the elongation of the biocomposites with KFs were in the same level. However, Young’s modulus significantly decreased for the PLA/KFs and the PLA/PEG4/KFs. Remarkably, the biocomposite with the KFs and the PEG6 showed significantly improve in the tensile strength. The higher molecular weight PEG6 improved the tensile strength. From this study, the PLA/PEG6/KFs biocomposite shows the most promising property with an improvement in mechanical performance. This might be the combined effects of the TEOS modifying and the use of an appropriate plasticizer. As shown in Fig. 12, the tensile strength significantly dropped for the PLA/KF biocomposite.