Influence of Dispersion and Orientation on Polyamide-6 Cellulose Nanocomposites Manufactured through Liquid-Assisted Extrusion

In this study, the possibility of adding nanocellulose and its dispersion to polyamide 6 (PA6), a polymer with a high melting temperature, is investigated using melt extrusion. The main challenges of the extrusion of these materials are achieving a homogeneous dispersion and avoiding the thermal degradation of nanocellulose. These challenges are overcome by using an aqueous suspension of never-dried nanocellulose, which is pumped into the molten polymer without any chemical modification or drying. Furthermore, polyethylene glycol is tested as a dispersant for nanocellulose. The dispersion, thermal degradation, and mechanical and viscoelastic properties of the nanocomposites are studied. The results show that the dispersant has a positive impact on the dispersion of nanocellulose and that the liquid-assisted melt compounding does not cause the degradation of nanocellulose. The addition of only 0.5 wt.% nanocellulose increases the stiffness of the neat polyamide 6 from 2 to 2.3 GPa and shifts the tan δ peak toward higher temperatures, indicating an interaction between PA6 and nanocellulose. The addition of the dispersant decreases the strength and modulus but has a significant effect on the elongation and toughness. To further enhance the mechanical properties of the nanocomposites, solid-state drawing is used to create an oriented structure in the polymer and nanocomposites. The orientation greatly improves its mechanical properties, and the oriented nanocomposite with polyethylene glycol as dispersant exhibits the best alignment and properties: with orientation, the strength increases from 52 to 221 MPa, modulus from 1.4 to 2.8 GPa, and toughness 30 to 33 MJ m−3 in a draw ratio of 2.5. This study shows that nanocellulose can be added to PA6 by liquid-assisted extrusion with good dispersion and without degradation and that the orientation of the structure is a highly-effective method for producing thermoplastic nanocomposites with excellent mechanical properties.


Introduction
Polyamide 6 (PA6), also called nylon 6, is a thermoplastic polymer used in a variety of applications owing to its excellent mechanical and thermal properties, as well as good adhesion and attractive surface appearance [1]. It has been reported that PA6 can be successfully reinforced with glass fibers [2], clay [3][4][5], and cellulose [6,7] to improve its properties. In the 1990s, studies on PA6-nanocomposites were carried out by Toyota [4,5] where PA6 was polymerized using montmorillonite (MMT) as a catalyst. The addition of MMT (4.7 wt.%) resulted in an increase in the tensile strength of these composites from 69 to 97 MPa and in their tensile modulus from 1.1 to 1.9 GPa. The use of cellulose nanocrystals

Preparation of NC Suspensions
The NC suspensions for the extrusion process were prepared by mixing commercial NC (2 wt.%) in distilled water and 0 or 10 wt.% PEG, with the aid of a magnetic stirrer. Afterward, the NC suspensions (2 wt.%) are further dispersed using an Ultra-Turrax (IKA T25, IKA Werke GmbH & Co. KG, Oxfordshire, UK) for 10 min, as shown in Figure S1a (Supplementary Information). The width of cellulose fibrils in the NC was measured using tapping-mode atomic force microscopy (Veeco MultiMode scanning probe, Santa Barbara, CA, USA). The sample was prepared by depositing a drop of diluted NC suspension (0.01 wt.%) on a fresh mica plate and letting it dry in a vacuum oven at 80 • C for 5 min. The measurements were recorded from images at height, using Nanoscope V software, and the average width values presented ( Figure S1b), varying between 5 and 40 nm, were based on 40 different measurements. The viscosities of the suspensions with different NC concentrations were measured using a Vibro viscometer (SV-10, A&D Company Limited, Tokyo, Japan). This method is used as a comparative metric of the viscosity with different NC contents (1.5, 2, and 3 wt.%), as shown in Figure S1c. The suspension viscosity increases with the increase in cellulose concentration, where 1.5 wt.% cellulose content results in a viscosity of approximately 400 mPa·s, and 3.0 wt.% cellulose in the suspension increases the viscosity almost 10-fold, to 3500 mPa·s. Subsequently, the NC suspensions with different concentrations were passed through the pump, and the flow rate was measured. More accurate measurements were obtained for 1.5 and 2 wt.% NC suspensions. Optical microscopy (OM) (Eclipse LV100POL, Nikon, Kanagawa, Japan) was used to observe the morphology of the fibers (Figure S1d) by placing a drop of the NC suspension (0.01 wt.%) on a glass slide.

Liquid-Assisted Extrusion and Film Preparation
Suspensions containing high-concentrations of NC are difficult to pump because NC suspensions have high viscosity (as explained in Section 2.2.1) even at low-concentrations, owing to the fiber length. Suspension viscosity is a limiting factor for the NC content in both the suspension and final composite. The water content should be as low as possible to guarantee slow and complete evaporation during extrusion. An overly fast water-evaporation can lead to agglomeration; moreover, the water remaining in the nanocomposite tends to interact with the polymer chains. Thus, an equilibrium must be found between the viscosity and liquid content. In this study, the maximum NC content was 2 wt.%, which was limited by the viscosity. To guarantee successful removal of water, a co-rotating twin-screw extruder W&P ZSK-18 MEGALab (Coperion, Stuttgart, Germany) is used because of its excellent degassing system, as shown in Figure 1a. A high-pressure syringe pump 500 D (Teledyne Isco, Lincoln, NE, USA) was used to feed the NC-containing suspension into the melt polymer ( Figure 1a). The feeding rate of PA6 was kept constant at 1.2 kg/h, and the screw speed was 200 rpm for all compounding trials. The used temperature setup (starting from 232 • C in the processing section to melt the PA6 and reduced to 225 • C) and used screw design in the extrusion process are shown in Figure 1a. The neat PA6 used as a reference material was prepared similarly to the PA6-nanocomposites rather than the NC suspension-only water was used as the pumping medium. The vaporized water was removed via atmospheric and vacuum venting. Table 1 lists the nanocomposites, their coding, compositions, as well as the feed and water evaporation rates. A rectangular die was used, and the PA6 and PA6-nanocomposites were collected and dried in an oven (40 °C for 12 h). Subsequently, approximately 4-5 g of the material was placed between Teflon-covered steel plates, pre-heated for 120 s at 225 °C and compression-molded at a pressure of 2.2 MPa for 60 s using a LabEcon 300 Fontijne Grotness press (Vlaardigen, The Netherlands). The films were then cooled to roomtemperature (25 °C) under pressure (approximately 5 min). The compression molding step was repeated to ensure homogeneous film composition as it was observed that the films were not completely homogeneous, and it was believed that the reason was the low molecular weight PEG. Therefore, the large films were cut into four smaller pieces which were melted again at 225 °C for 60 s and compression molded for another 60 s at the same pressure and temperature as the previous films. The second compression molding resulted in better film quality. SSD of the films was performed using a universal tensile testing machine (Shimadzu Autograph AGX, Kyoto, Japan) equipped with a temperature chamber and a load cell of 1 kN (Figure 1a). The drawing was performed at 140 °C, a rate of 100 mm/min, and the  A rectangular die was used, and the PA6 and PA6-nanocomposites were collected and dried in an oven (40 • C for 12 h). Subsequently, approximately 4-5 g of the material was placed between Teflon-covered steel plates, pre-heated for 120 s at 225 • C and compressionmolded at a pressure of 2.2 MPa for 60 s using a LabEcon 300 Fontijne Grotness press (Vlaardigen, The Netherlands). The films were then cooled to room-temperature (25 • C) under pressure (approximately 5 min). The compression molding step was repeated to ensure homogeneous film composition as it was observed that the films were not completely homogeneous, and it was believed that the reason was the low molecular weight PEG. Therefore, the large films were cut into four smaller pieces which were melted again at 225 • C for 60 s and compression molded for another 60 s at the same pressure and temperature as the previous films. The second compression molding resulted in better film quality.

SSD
SSD of the films was performed using a universal tensile testing machine (Shimadzu Autograph AGX, Kyoto, Japan) equipped with a temperature chamber and a load cell of  (Figure 1a). The drawing was performed at 140 • C, a rate of 100 mm/min, and the maximum draw ratio (DR) of 2.5 (the final gauge length divided by the original gauge length). The drawing temperature is chosen according to the DSC analysis of neat PA6 shown in Figure 1b, above the glass transition temperature (≈65 • C) [27] and below the cold crystallization temperature. The PA6-nanocomposite samples before and after drawing are shown in Figure 1c. The samples after SSD are marked with "O" before the sample name.

Characterizations
Optical and polarized optical microscopes (OM and POM, respectively) (Eclipse LV100POL, Nikon, Kanagawa, Japan) were used to investigate the dispersion of NC in the suspension, nanocomposite films, as well as the birefringence caused by the orientation of the PA6 and PA6-nanocomposite films. POM equipped with a Linkam hot-stage THM600 (Tadworth, UK) to study the crystallization behavior of PA6 and PA6-nanocomposites films. The films were heated between glass covers to 232 • C and then cooled to 200 • C which is a crystallization temperature of PA6. The nucleation and crystal growth were studied between a time interval of 0 to 120 s. High-resolution scanning electron microscopy (SEM) (FEI Magellan 4000 XHR-SEM, Thermo Fisher Scientific, Hillsboro, OR, USA) was used to analyze the effects of the SSD on the surface of the PA6 and PA6-nanocomposite films, at an acceleration voltage of 3 kV. The film was placed on the holder, and the surface was sputter-coated with platinum using a Leica EM ACE200 sputtering instrument (Wetzlar, Germany, 15 nm thick layer), to avoid charging.
The thermal stability and onset degradation temperature of the PA6 and PA6nanocomposites were measured using thermogravimetric analysis (TGA) (Q500, TA Instruments, New Castle, DE, USA) under a nitrogen flow, at a heating rate of 10 • C/min within a temperature range of 0-600 • C. Differential scanning calorimetry (DSC) (Mettler Toledo DSC821, Columbus, OH, USA) was used to study crystallinity and melt temperatures of the materials before and after SSD. The sample, weighing around 4-6 mg, was placed in a hermetic aluminum pan and tested under a nitrogen atmosphere from 25 to 260 • C, at a heating rate of 10 • C/min. The degree of crystallinity (X c ) was calculated using the enthalpy of 100% crystalline PA6 (230 J/g).
The mechanical properties of the materials (before and after SSD) were tested under controlled temperature and low-humidity (25 • C and 15% RH) using a Shimadzu Autograph AGX universal testing machine (Kyoto, Japan) with a 1 kN load cell. The samples were oven-dried prior to testing. The tests were performed with a gauge length of 20 mm and a crosshead speed of 5 mm/min. At least five specimens were tested for each sample. For the tested samples, a statistical analysis at a 5% significance level, based on the ANOVA and Tukey-HSD multiple comparison tests, was used to analyze the results.
The storage modulus and tan δ peak as a function of the temperature of the PA6 and PA6-nanocomposites (before and after SSD) were determined using DMA Q800 (TA Instruments New Castle, DE, USA) using a temperature ramp and single frequency of 1 Hz. The test was performed in tensile mode, within a temperature range between -50 and 150 • C, at a heating rate of 3 • C/min; the materials were oven-dried before the test.
FT-IR spectroscopy was used to evaluate the effect of SSD on the PA6-nanocomposites. The films were placed in a VERTEX 80 (Bruker, Billerica, MA, USA) FT-IR spectrophotometer with a range of 500-3500 cm −1 , and a 256-scan resolution was used. The curves were normalized by the highest intensity to minimize errors owing to differences in thickness.
The crystal type and crystal size of the PA6 and PA6-nanocomposites (before and after SSD) were studied using an X-ray diffractometer (XRD) (PANalytical Empyrean X-ray diffractometer, Malvern Instruments, Malvern, UK) with CuKα radiation (a wavelength of 1.540 Å), generated at an acceleration voltage of 45 kV and current of 40 mA. The 2θ scan angle range was 5 • -40 • . The apparent crystallite size was obtained from the half-width of the 2θ peaks using Scherrer's Equation (1) [28]: where D is the crystal size, k is a constant that is 0.89 [28], λ is the wavelength of the X-ray, β is the width of the peak at half the maximum intensity, and θ is the Bragg's angle. The degree of orientation of the PA6 and PA6-nanocomposites was studied using wide-angle Xray scattering (WAXS) analysis on an Anton Paar SAXSpoint 2.0 system (Anton Paar, Graz, Austria) equipped with a Microsource X-ray source (Cu K-alpha radiation, a wavelength of 0.15418 nm) and a Dectris 2D CMOS Eiger R 1M detector with a pixel size of 75 × 75 µm 2 . All measurements were performed with a beam size of approximately 500 µm in diameter and a beam path pressure of approximately 1-2 mbar. All samples were mounted on a Sampler for Solids 10 × 10 mm 2 (Anton Paar, Graz, Austria) holder. The transmittance was determined and used for scaling the intensities. The software used for instrument control was SAXS drive version 2.01.224 (Anton Paar, Graz, Austria), and post-acquisition data-processing was performed using SAXS analysis version 4.01.047 (Anton Paar, Graz, Austria). The orientation index (f c ) of the PA6 and PA6-nanocomposites was calculated based on the intensity distributions of the azimuthal angle using the following equation: where FWHM is the full width at half maximum of the azimuthal angle distribution.

Effect of the Addition of NC and NC/PEG to PA6
The dispersion of NC in PA6 was studied using OM ( Figure 2). The presence of dark spots (indicated by the circles in Figure 2a) is observed, which can be indicative of NC agglomeration, and the number of these spots decreases with the addition of PEG (5 wt.%), as illustrated in Figure 2b. PEG is expected to protect NC during the water-evaporation step, reducing agglomeration.
where D is the crystal size, k is a constant that is 0.89 [28], λ is the wavelength of the Xray, β is the width of the peak at half the maximum intensity, and θ is the Bragg's angle. The degree of orientation of the PA6 and PA6-nanocomposites was studied using wideangle X-ray scattering (WAXS) analysis on an Anton Paar SAXSpoint 2.0 system (Anton Paar, Graz, Austria) equipped with a Microsource X-ray source (Cu K-alpha radiation, a wavelength of 0.15418 nm) and a Dectris 2D CMOS Eiger R 1M detector with a pixel size of 75 × 75 µm 2 . All measurements were performed with a beam size of approximately 500 µm in diameter and a beam path pressure of approximately 1-2 mbar. All samples were mounted on a Sampler for Solids 10 × 10 mm 2 (Anton Paar, Graz, Austria) holder. The transmittance was determined and used for scaling the intensities. The software used for instrument control was SAXS drive version 2.01.224 (Anton Paar, Graz, Austria), and postacquisition data-processing was performed using SAXS analysis version 4.01.047 (Anton Paar, Graz, Austria). The orientation index (fc) of the PA6 and PA6-nanocomposites was calculated based on the intensity distributions of the azimuthal angle using the following equation: where FWHM is the full width at half maximum of the azimuthal angle distribution.

Effect of the Addition of NC and NC/PEG to PA6
The dispersion of NC in PA6 was studied using OM ( Figure 2). The presence of dark spots (indicated by the circles in Figure 2a) is observed, which can be indicative of NC agglomeration, and the number of these spots decreases with the addition of PEG (5 wt.%), as illustrated in Figure 2b. PEG is expected to protect NC during the waterevaporation step, reducing agglomeration. The thermal properties of PA6 and its nanocomposites are shown in Figure S2 in the Supplementary Information. There was no difference in the degradation temperature between PA6 and PA6/0.5NC, which suggests that NC (degradation onset around 273 °C) can be used with polymers with a high melting temperature, such as PA6. The liquid suspension decreases the processing temperature [9], resulting in lower processing temperatures in the last zones (shown in Figure 1a), which is beneficial for NC.
The degree of crystallinity of PA6/0.5NC (28%) is higher than that of PA6 (21%), as listed in Table S1 in the Supplementary Information. The addition of PEG/NC had a small effect on crystallinity (25%), explained by the similar polarities of PEG and PA6, which decreased the degree of crystallinity [29]. The PA6 shows two melting endotherms, Tm1 at ~205 °C and Tm2 at 218 °C, as illustrated in Figure S3. Khanna [27] found that the two The thermal properties of PA6 and its nanocomposites are shown in Figure S2 in the Supplementary Information. There was no difference in the degradation temperature between PA6 and PA6/0.5NC, which suggests that NC (degradation onset around 273 • C) can be used with polymers with a high melting temperature, such as PA6. The liquid suspension decreases the processing temperature [9], resulting in lower processing temperatures in the last zones (shown in Figure 1a), which is beneficial for NC.
The degree of crystallinity of PA6/0.5NC (28%) is higher than that of PA6 (21%), as listed in Table S1 in the Supplementary Information. The addition of PEG/NC had a small effect on crystallinity (25%), explained by the similar polarities of PEG and PA6, which decreased the degree of crystallinity [29]. The PA6 shows two melting endotherms, T m1 at~205 • C and T m2 at 218 • C, as illustrated in Figure S3. Khanna [27] found that the two melting-temperature peaks at 215 • C and 225 • C in nylon 6 are associated with two different crystal types, i.e., γ and α crystals, respectively. The double melting peaks observed in the DSC results are likely associated with these two crystal forms. The addition of only 0.5 wt.% NC increases the E modulus from 2.0 to 2.3 GPa, while the addition of PEG/NC reduces it to 1.4 GPa, as shown in Table S2 in the Supplementary Information. In terms of strength, no significant differences were observed for PA6 and PA6/0.5NC; however, the use of PEG reduced the strength of the nanocomposite. Similarly, in the work performed by Peng et al. [8], the addition of 0.5 wt.% CNCs in PA6 did not enhance the mechanical properties. The addition of PEG contributes to an increase in the elongation at break, from 19 to 63%, resulting in an increase in toughness, from 10 to 29 MJ m −3 , as shown in Table S2. Oksman et al. [20] reported that the addition of PEG (15 wt.%) as an additive in cellulose nanowhisker suspension improved the dispersion of NC in PLA with an increase in elongation at break but with a decrease in modulus and strength when compared with the corresponding nanocomposite.
The viscoelastic behaviors of PA6 and PA6-nanocomposites are shown in Figure 3. The α-relaxation peak assigned as the tan δ peak (≈65 • C) is related to the breaking of hydrogen bonds between the polymer chains, inducing chain motion in the amorphous region [30]. The storage modulus (E ) of PA6 and its nanocomposites remain at a similar level, while the tan δ peak moves to a higher temperature (from 61 to 69 • C) with the addition of NC, indicating an interaction between PA6 and NC. The addition of PEG/NC did not affect the tan δ peak position, in contrast to the addition of NC only. The use of PEG was expected to change the tan δ peak to a lower temperature due to the plasticizing effect of PEG [31], but the NC in this material counteracts the effect and therefore the peak position is not affected compared to PA6. melting-temperature peaks at 215 °C and 225 °C in nylon 6 are associated with two different crystal types, i.e., γ and α crystals, respectively. The double melting peaks observed in the DSC results are likely associated with these two crystal forms. The addition of only 0.5 wt.% NC increases the E modulus from 2.0 to 2.3 GPa, while the addition of PEG/NC reduces it to 1.4 GPa, as shown in Table S2 in the Supplementary Information. In terms of strength, no significant differences were observed for PA6 and PA6/0.5NC; however, the use of PEG reduced the strength of the nanocomposite. Similarly, in the work performed by Peng et al. [8], the addition of 0.5 wt.% CNCs in PA6 did not enhance the mechanical properties. The addition of PEG contributes to an increase in the elongation at break, from 19 to 63%, resulting in an increase in toughness, from 10 to 29 MJ m −3 , as shown in Table S2. Oksman et al. [20] reported that the addition of PEG (15 wt.%) as an additive in cellulose nanowhisker suspension improved the dispersion of NC in PLA with an increase in elongation at break but with a decrease in modulus and strength when compared with the corresponding nanocomposite.
The viscoelastic behaviors of PA6 and PA6-nanocomposites are shown in Figure 3. The α-relaxation peak assigned as the tan δ peak (≈65 °C) is related to the breaking of hydrogen bonds between the polymer chains, inducing chain motion in the amorphous region [30]. The storage modulus (E′) of PA6 and its nanocomposites remain at a similar level, while the tan δ peak moves to a higher temperature (from 61 to 69 °C) with the addition of NC, indicating an interaction between PA6 and NC. The addition of PEG/NC did not affect the tan δ peak position, in contrast to the addition of NC only. The use of PEG was expected to change the tan δ peak to a lower temperature due to the plasticizing effect of PEG [31], but the NC in this material counteracts the effect and therefore the peak position is not affected compared to PA6. The 1D-XRD scattering patterns, shown in Figure S4 in the Supplementary Information, reveal that the two main characteristic diffraction peaks for the PA6 and PA6-nanocomposite are located around 20.2° and 23.7°. According to the literature [32], the α-phase peaks for PA6 at 20.1° (200) and 23.8° (002/202) are related to the α1 and α2 peaks, respectively, and 21.3° (200/001) is related to the γ-phase. Other researchers [33] showed that α-phase and γ-phase can coexist in PA6 and that the γ-phase presents peaks at 22° (001) and 23° (200/201). The α-phase has a monoclinic structure, where polyamide molecules form an antiparallel chain arrangement hydrogen bonding. The γ-phase is also monoclinic, but the polyamide molecules adopt parallel chain arrangement hydrogen bonding. In Figure S4, no peaks from NC and PEG are visible owing to their low concentrations in the PA6-nanocomposites. However, the addition of NC increased the The 1D-XRD scattering patterns, shown in Figure S4 in the Supplementary Information, reveal that the two main characteristic diffraction peaks for the PA6 and PA6nanocomposite are located around 20.2 • and 23.7 • . According to the literature [32], the α-phase peaks for PA6 at 20.1 • (200) and 23.8 • (002/202) are related to the α 1 and α 2 peaks, respectively, and 21.3 • (200/001) is related to the γ-phase. Other researchers [33] showed that α-phase and γ-phase can coexist in PA6 and that the γ-phase presents peaks at 22 • (001) and 23 • (200/201). The α-phase has a monoclinic structure, where polyamide molecules form an antiparallel chain arrangement hydrogen bonding. The γ-phase is also monoclinic, but the polyamide molecules adopt parallel chain arrangement hydrogen bonding. In Figure S4, no peaks from NC and PEG are visible owing to their low concentrations in the PA6-nanocomposites. However, the addition of NC increased the apparent crystallite size in PA6, from 5.32 to 7.79 nm, and the addition of PEG in the nanocomposites further increased the size to 6.70 nm.
In addition, the crystallization of the PA6 and its nanocomposites was studied with POM and the results are shown in Figure S5, in the Supplementary Information. The images show fast crystallization behavior of the neat PA6, as well as the nanocomposites. The crystallization temperature for PA6 is at 200 • C and the visible nucleation (in the used magnification) of the crystallites is starting already after 40 s and appears to be completed at around 90 s, after that no difference is seen. The crystal size is very small, and no large crystals are visible in any of the materials. In addition, no difference of the nucleation and crystallization rate among the materials were visible, indicating that the NC is not acting as a nucleating agent. The exact crystal size cannot be defined from the POM images because of the small size. Figure 4 shows cross-polarized optical micrographs, with the orientation of the films at 0 • and 45 • to the cross-polarizers. The micrographs confirm the orientation of the films as their birefringence initially increases and reaches maximum power at 0 • (Figure 4a) and subsequently decreases to a minimum at 45 • (Figure 4b) as the film orientation is positioned parallel to one of the polarizers. All the materials show non-homogeneous birefringence, expressed by the heterogeneous orientation of the polymer chains, as indicated by the different colors.

Effect of Orientation on PA6 and PA6-Nanocomposites
Nanomaterials 2022, 12, x FOR PEER REVIEW 8 apparent crystallite size in PA6, from 5.32 to 7.79 nm, and the addition of PEG i nanocomposites further increased the size to 6.70 nm. In addition, the crystallization of the PA6 and its nanocomposites was studied POM and the results are shown in Figure S5, in the Supplementary Information images show fast crystallization behavior of the neat PA6, as well as the nanocompo The crystallization temperature for PA6 is at 200 °C and the visible nucleation (in the magnification) of the crystallites is starting already after 40 s and appears to be comp at around 90 s, after that no difference is seen. The crystal size is very small, and no crystals are visible in any of the materials. In addition, no difference of the nucleatio crystallization rate among the materials were visible, indicating that the NC is not a as a nucleating agent. The exact crystal size cannot be defined from the POM im because of the small size.  As shown in Figure 4a, the OPA6 and OPA6-nanocomposites have defects. O presents the highest birefringence with a small number of defects, followe OPA6/5PEG/0.5NC. Most defects are visible in PA6/0.5NC. Because OPA6 also pres small number of defects, their presence could be associated with the excessive stret during SSD, and the large defects seen with the addition of NC could be related t agglomeration. This further confirms the improvement in dispersion with the addit PEG. Barkoula et al. [34] discussed that drawing polar polymers, such as PA6, below melting temperature is difficult because of the strong hydrogen bonds between the c and their folding, creating a resistance of the crystallite to deformation and limitin drawing ratio. To clarify the defects shown in the POM, the OPA6 and O nanocomposite film surfaces are examined using SEM, as shown in Figure 5 micrographs confirm the presence of cavities on the surface of the OPA6 and O nanocomposite films. More accentuated cavities are observed in the OPA6/0.5NC As shown in Figure 4a, the OPA6 and OPA6-nanocomposites have defects. OPA6 presents the highest birefringence with a small number of defects, followed by OPA6/5PEG/ 0.5NC. Most defects are visible in PA6/0.5NC. Because OPA6 also presents a small number of defects, their presence could be associated with the excessive stretching during SSD, and the large defects seen with the addition of NC could be related to NC agglomeration. This further confirms the improvement in dispersion with the addition of PEG. Barkoula et al. [34] discussed that drawing polar polymers, such as PA6, below their melting temperature is difficult because of the strong hydrogen bonds between the chains and their folding, creating a resistance of the crystallite to deformation and limiting the drawing ratio. To clarify the defects shown in the POM, the OPA6 and OPA6-nanocomposite film surfaces are examined using SEM, as shown in Figure 5. The micrographs confirm the presence of cavities on the surface of the OPA6 and OPA6-nanocomposite films. More accentuated cavities are observed in the OPA6/0.5NC films than in the nanocomposite with PEG. Corroborating the POM images, the lower number of cavities in OPA6/5PEG/0.5NC is explained by the better distribution of NC in PA6. The cavities in OPA6 and its nanocomposites can also be attributed to the excessive stretching of the film. Nanomaterials 2022, 12, x FOR PEER REVIEW 9 of 14 than in the nanocomposite with PEG. Corroborating the POM images, the lower number of cavities in OPA6/5PEG/0.5NC is explained by the better distribution of NC in PA6. The cavities in OPA6 and its nanocomposites can also be attributed to the excessive stretching of the film. The DSC thermographs in Figure 6 show that the γ phase (Tm1) is more pronounced after the SSD for all samples. Furthermore, the degree of crystallinity increased, and both melting-peaks shifted to higher temperatures with the orientation; a relatively high increment in the intensity was observed for Tm1 (Table S1). The increase in crystallinity after the orientation process is explained by both the orientation of molecules and their much closer packing, allowing for the crystallites to be formed from the previously amorphous region [35]. In Figure 7, the mechanical properties of PA6 and PA6-nanocomposites before and after SSD are presented. The SSD has a significant effect on the strength, modulus, and toughness of OPA6 and OPA6-nanocomposites, when compared with non-oriented samples. For OPA6/0.5NC, the modulus improved by 30% (from 2.3 to 3.0 GPa), strength increased by 258% (from 60 to 215 MPa), and toughness improved by 140% (10 to 24 MJ/m 3 ), when compared with the non-oriented PA6/0.5NC. A comparison of OPA6/5PEG/0.5NC and PA6/5PEG/0.5NC shows that after SSD, strength improved by 325% (from 52 to 221 MPa) and the modulus doubled (from 1.4 to 2.8 GPa).

Effect of Orientation on PA6 and PA6-Nanocomposites
A comparison of the OPA6 and OPA6-nanocomposites showed that the addition of PEG (OPA6/5PEG/0.5NC) resulted in the highest tensile strength (221 MPa) and The DSC thermographs in Figure 6 show that the γ phase (T m1 ) is more pronounced after the SSD for all samples. Furthermore, the degree of crystallinity increased, and both melting-peaks shifted to higher temperatures with the orientation; a relatively high increment in the intensity was observed for T m1 (Table S1). The increase in crystallinity after the orientation process is explained by both the orientation of molecules and their much closer packing, allowing for the crystallites to be formed from the previously amorphous region [35]. Nanomaterials 2022, 12, x FOR PEER REVIEW 9 of 14 than in the nanocomposite with PEG. Corroborating the POM images, the lower number of cavities in OPA6/5PEG/0.5NC is explained by the better distribution of NC in PA6. The cavities in OPA6 and its nanocomposites can also be attributed to the excessive stretching of the film. The DSC thermographs in Figure 6 show that the γ phase (Tm1) is more pronounced after the SSD for all samples. Furthermore, the degree of crystallinity increased, and both melting-peaks shifted to higher temperatures with the orientation; a relatively high increment in the intensity was observed for Tm1 (Table S1). The increase in crystallinity after the orientation process is explained by both the orientation of molecules and their much closer packing, allowing for the crystallites to be formed from the previously amorphous region [35]. In Figure 7, the mechanical properties of PA6 and PA6-nanocomposites before and after SSD are presented. The SSD has a significant effect on the strength, modulus, and toughness of OPA6 and OPA6-nanocomposites, when compared with non-oriented samples. For OPA6/0.5NC, the modulus improved by 30% (from 2.3 to 3.0 GPa), strength increased by 258% (from 60 to 215 MPa), and toughness improved by 140% (10 to 24 MJ/m 3 ), when compared with the non-oriented PA6/0.5NC. A comparison of OPA6/5PEG/0.5NC and PA6/5PEG/0.5NC shows that after SSD, strength improved by 325% (from 52 to 221 MPa) and the modulus doubled (from 1.4 to 2.8 GPa).
A comparison of the OPA6 and OPA6-nanocomposites showed that the addition of PEG (OPA6/5PEG/0.5NC) resulted in the highest tensile strength (221 MPa) and In Figure 7, the mechanical properties of PA6 and PA6-nanocomposites before and after SSD are presented. The SSD has a significant effect on the strength, modulus, and toughness of OPA6 and OPA6-nanocomposites, when compared with non-oriented samples. For OPA6/0.5NC, the modulus improved by 30% (from 2.3 to 3.0 GPa), strength increased by 258% (from 60 to 215 MPa), and toughness improved by 140% (10 to 24 MJ/m 3 ), when compared with the non-oriented PA6/0.5NC. A comparison of OPA6/5PEG/0.5NC and PA6/5PEG/0.5NC shows that after SSD, strength improved by 325% (from 52 to 221 MPa) and the modulus doubled (from 1.4 to 2.8 GPa). elongation at break, the opposite behavior observed here can be attributed to the increase in the degree of orientation and fraction of the rigid amorphous phase [36], supported by the increase in crystallinity observed in the DSC results. Bugatti et al. [22] have also reported a decrease in elongation at break after drawing poly-(ε-caprolactone) nanocomposites. Given that, the high tensile strength attained through the orientation of the polymer chains and nanoparticles is very favorable to produce tough and stiff materials. The dynamic mechanical behavior is illustrated in Figure 8, which shows that the OPA6 and OPA6-nanocomposites present higher storage modulus values than the nonoriented PA6 and PA6-nanocomposites, as expected according to the tensile properties. In addition, the tan δ peak shifts to high temperatures after the SSD (from 60 to 90 °C). The shift of the tan δ peak to a high temperature can be attributed to the strain-induced crystallization and constrained-oriented amorphous phase, as noted by Mai et al. [21] after the SSD of PLA, as well as to the increment in the degree of crystallinity [32], supported by the DSC results. Similarly, the amplitude of the tan δ peak increases after SSD, indicating an enhancement in the material damping and that oriented PA6 and PA6nanocomposites can absorb and dissipate the energy better, resulting in an increased toughness, as observed from the mechanical properties. Penel-Pierron et al. [24] suggested that the difference in amplitude is caused by the different arrangements of the crystallites between different samples, modifying their damping behavior.  A comparison of the OPA6 and OPA6-nanocomposites showed that the addition of PEG (OPA6/5PEG/0.5NC) resulted in the highest tensile strength (221 MPa) and toughness, as well as comparable modulus and elongation at break (Table S3). The increase in the strength of OPA6/0.5NC and OPA6/5PEG/0.5NC, compared with OPA6, demonstrates that in addition to the presence of defects, the alignment of NC contributes to the enhancement in the mechanical properties. Even though SSD tends to increase the elongation at break, the opposite behavior observed here can be attributed to the increase in the degree of orientation and fraction of the rigid amorphous phase [36], supported by the increase in crystallinity observed in the DSC results. Bugatti et al. [22] have also reported a decrease in elongation at break after drawing poly-(ε-caprolactone) nanocomposites. Given that, the high tensile strength attained through the orientation of the polymer chains and nanoparticles is very favorable to produce tough and stiff materials.
The dynamic mechanical behavior is illustrated in Figure 8, which shows that the OPA6 and OPA6-nanocomposites present higher storage modulus values than the nonoriented PA6 and PA6-nanocomposites, as expected according to the tensile properties. In addition, the tan δ peak shifts to high temperatures after the SSD (from 60 to 90 • C). The shift of the tan δ peak to a high temperature can be attributed to the strain-induced crystallization and constrained-oriented amorphous phase, as noted by Mai et al. [21] after the SSD of PLA, as well as to the increment in the degree of crystallinity [32], supported by the DSC results. Similarly, the amplitude of the tan δ peak increases after SSD, indicating an enhancement in the material damping and that oriented PA6 and PA6-nanocomposites can absorb and dissipate the energy better, resulting in an increased toughness, as observed from the mechanical properties. Penel-Pierron et al. [24] suggested that the difference in amplitude is caused by the different arrangements of the crystallites between different samples, modifying their damping behavior. Figure S6a-c, in the Supplementary Information, shows the FT-IR spectra of PA6nanocomposites and OPA6-nanocomposites. It is possible to see that both types of crystals (α and γ) are presented before and after SSD, the peak around 1121 cm −1 is associated with γ crystals, and a higher-intensity peak is found after the SSD. A slight shift from 1238 to 1236 after SSD can be attributed to the vibration of amide III group in α-crystals (peak at 1238) to γ-crystals (peak in 1232), also a peak in 1530 cm −1 related to amide II group in γ-crystals is found for oriented samples [37]. In addition, the non-oriented PA6-nanocomposites show higher intensity peaks at~3300 cm −1 than oriented PA6-nanocomposites. This peak is related to the hydrogen-bonded N-H stretch in the crystalline phase [38]. During drawing, the hydrogen-bond network is broken and formed constantly in a dynamic system; therefore, it can be suggested that the hydrogen bonds are not fully formed during the drawing of PA6-nanocomposites, as reported previously by Xu et al. [32] during the stretching of the β-PA6 film.
The shift of the tan δ peak to a high temperature can be attributed to the strain-induced crystallization and constrained-oriented amorphous phase, as noted by Mai et al. [21] after the SSD of PLA, as well as to the increment in the degree of crystallinity [32], supported by the DSC results. Similarly, the amplitude of the tan δ peak increases after SSD, indicating an enhancement in the material damping and that oriented PA6 and PA6nanocomposites can absorb and dissipate the energy better, resulting in an increased toughness, as observed from the mechanical properties. Penel-Pierron et al. [24] suggested that the difference in amplitude is caused by the different arrangements of the crystallites between different samples, modifying their damping behavior.   Figure 9 illustrates the 1D-XRD results for the non-oriented and oriented PA6 and PA6-nanocomposites. The oriented PA6 and PA6-nanocomposites present broader peaks than the non-oriented ones. Moreover, for the OPA6-nanocomposites, the peak at 2θ = 20.3 • almost disappeared and the peak at 2θ = 23.7 • shifted toward low Bragg's angles (≈23.2 • , Table S4). This could be indicative of the crystal-to-crystal transition from the α-phase (2θ = 23.7 • ) to the γ-phase (2θ = 23 • ) [27], corroborating the results of DSC. In fact, the apparent crystallite size of all oriented nanocomposites decreased, as suggested by the peak broadening. Shi and co-workers [26] attributed the decrease in the crystallite size after the hot drawing of PA6 to the slipping and rupture of the lamellae in the spherulite.  Figure S6a-c, in the Supplementary Information, shows the FT-IR spectra of PA6nanocomposites and OPA6-nanocomposites. It is possible to see that both types of crystals (α and γ) are presented before and after SSD, the peak around 1121 cm −1 is associated with γ crystals, and a higher-intensity peak is found after the SSD. A slight shift from 1238 to 1236 after SSD can be attributed to the vibration of amide III group in α-crystals (peak at 1238) to γ-crystals (peak in 1232), also a peak in 1530 cm −1 related to amide II group in γcrystals is found for oriented samples [37]. In addition, the non-oriented PA6nanocomposites show higher intensity peaks at ~3300 cm −1 than oriented PA6nanocomposites. This peak is related to the hydrogen-bonded N-H stretch in the crystalline phase [38]. During drawing, the hydrogen-bond network is broken and formed constantly in a dynamic system; therefore, it can be suggested that the hydrogen bonds are not fully formed during the drawing of PA6-nanocomposites, as reported previously by Xu et al. [32] during the stretching of the β-PA6 film. Figure 9 illustrates the 1D-XRD results for the non-oriented and oriented PA6 and PA6-nanocomposites. The oriented PA6 and PA6-nanocomposites present broader peaks than the non-oriented ones. Moreover, for the OPA6-nanocomposites, the peak at 2θ = 20.3° almost disappeared and the peak at 2θ = 23.7° shifted toward low Bragg's angles (≈23.2°, Table S4). This could be indicative of the crystal-to-crystal transition from the αphase (2θ = 23.7°) to the γ-phase (2θ = 23°) [27], corroborating the results of DSC. In fact, the apparent crystallite size of all oriented nanocomposites decreased, as suggested by the peak broadening. Shi and co-workers [26] attributed the decrease in the crystallite size after the hot drawing of PA6 to the slipping and rupture of the lamellae in the spherulite. To confirm the effect of SSD found using the results of 1D-XRD, a 2D-WAXS analysis was performed ( Figure S7a in the Supplementary Information). The crystal peaks of PA6 were observed in all the samples ( Figure S7b). No peaks of NC and no distinct peaks of PEG were observed, which is consistent with the 1D-XRD results. A mechanically induced phase transition of a part of PA6 from the α-phase to the γ-phase can be suggested to occur in the oriented nanocomposites because the shift of the peak at 23.7° to a lower angle, corroborating the results of 1D-XRD. The PA6 orientation was estimated from the orientation index (fc) calculated at two peaks: 2θ = 20.1° and 23.7° (Table S5). The fc of PA6 is higher by 0.53-0.6 than those of PA6/0.5NC and PA6/5PEG/0.5NC. It is difficult to explain the high-orientation degree of PA6 after compression molding, but it may be suggested that crystal formation during cooling is stacking in a preferential orientation, while the presence of NC hinders this preferential orientation. However, the 2D-XRD Figure 9. The 1D-XRD scattering patterns of the PA6 and its nanocomposites before and after orientation.
To confirm the effect of SSD found using the results of 1D-XRD, a 2D-WAXS analysis was performed ( Figure S7a in the Supplementary Information). The crystal peaks of PA6 were observed in all the samples ( Figure S7b). No peaks of NC and no distinct peaks of PEG were observed, which is consistent with the 1D-XRD results. A mechanically induced phase transition of a part of PA6 from the α-phase to the γ-phase can be suggested to occur in the oriented nanocomposites because the shift of the peak at 23.7 • to a lower angle, corroborating the results of 1D-XRD. The PA6 orientation was estimated from the orientation index (f c ) calculated at two peaks: 2θ = 20.1 • and 23.7 • (Table S5). The f c of PA6 is higher by 0.53-0.6 than those of PA6/0.5NC and PA6/5PEG/0.5NC. It is difficult to explain the high-orientation degree of PA6 after compression molding, but it may be suggested that crystal formation during cooling is stacking in a preferential orientation, while the presence of NC hinders this preferential orientation. However, the 2D-XRD analysis shows that the orientation index for the OPA6-nanocomposites is higher than that for OPA6, in the range of 0.84-0.86. Although the results do not prove the orientation of the NC because the peaks for the NC cannot be observed and the diffraction peaks are only related to PA6, the enhanced mechanical properties of OPA6/0.5NC and OPA6/5PEG/0.5NC can be an indicator of a slight orientation of NC in the OPA6-nanocomposites. In addition, different intensities of the diffraction peaks observed in 1D-XRD and 2D-XRD can be attributed to different crystal orientations along the film.

Conclusions
In this work, we showed that liquid-assisted extrusion and the use of a PEG dispersant resulted in good dispersion and that the high-temperature process (232 • C) did not degrade the NC. The addition of a small amount of NC had a positive effect on the modulus, and the interaction between the NC and PA6 was excellent, as confirmed by DMA. SSD had a significant effect on the strength, stiffness, and toughness of PA6 and its nanocomposites. OPA6/5PEG/0.5NC was the nanocomposite with the best properties: a strength of 221 MPa, modulus of 2.8 GPa, and elongation at break of 23%, given the toughness of approximately 33 MJ m −3 with an orientation factor of 0.86. As the addition of 5 wt.% PEG improved the dispersion of the NC, increased the elongation and toughness but decreased the strength and modulus of non-oriented PA6-nanocomposites, a future study on the optimum concentration should be performed. In our findings, the use of never-dried nanocellulose in PA6 and the alignment of nanocomposites contributed to the improvement of mechanical properties.