In this study, cellulose nanocrystals (CNC) were extracted from rice straw and incorporated into polyvinyl alcohol (PVOH) as reinforcement nanofillers. Multiple nanocomposites with different CNC contents were prepared. Extracted CNC appear as long, well-defined rodlike crystals with a high aspect ratio (41). Nanocomposites with 3 wt% of CNC significantly exhibit improved tensile strength (60.4%) and maximum degradation temperature (287°C). Moreover, they demonstrate a decrease in water vapor permeability rate and in the swelling and solubility indices of PVOH/CNC. Significant improvements were observed when nanocomposites were crosslinked specifically in terms of tensile strength (104.8%) and maximum degradation temperature (364°C). They also demonstrate greatly reduced water vapor permeability rate, swelling, and solubility indices. The optimum CNC amount for both nanocomposites is 3 wt%.

1. Introduction

Plastic is a synthetic polymer that is versatile and durable and has a low production cost; it is also the material of choice for numerous applications worldwide. However, plastics are nondegradable. The significant increase in the use of and dependence on plastics has also increased the waste stream, and this is an issue for the ecosystem and the environment [14]. The common nondegradable polymers, such as polyethylene, polypropylene, polyvinyl chloride, and polystyrene, are consumer market dominators that constitute the vast majority of the products that end up in landfills; moreover, they are extremely resistant to degradation [510]. In response to this, researchers must design environmentally friendly substitutes (e.g., polymers that are biodegradable when exposed to the environment). One of the most perfect examples of biodegradable synthetic polymers is polyvinyl alcohol (PVOH). PVOH is a conventional water-soluble polymer with a substantial value to the manufacturing industry as it is nontoxic and fully biodegradable and has exceptional biocompatibility [1113]. PVOH can be adaptively manufactured to suit multiple applications by varying its degree of hydrolysis and its average molecular weight. In addition, PVOH is used in packaging industries owing to its excellent film-forming capability (but only in niche applications due to its water sensitivity).

Researchers are currently focusing on the enhancement and improvement of the properties of neat polymers through the addition of small amounts of inexpensive and environmentally friendly fillers. The conversion of “waste to wealth” is a common interest among the scientific and global communities; thus, agricultural wastes, such as rice straw, sugarcane bagasse, and soy hulls, are the main materials used in the current research. Nanotechnology has warranted exciting opportunities within the realm of material science, especially in the use of nanocellulose derived from renewable sources to improve the properties of polymers as reinforcement filler materials. Nanocellulose exists in multiple forms, namely, cellulose nanocrystals (CNC) and cellulose nanofibers (CNF), depending on the extraction method [14]. CNF are mainly obtained through mechanical fibrillation (a process that requires a huge amount of energy), whereas the production of CNC depends on strong chemical treatment, such as acid hydrolysis [1518]. Compared with CNF, CNC retain only the crystalline regions of cellulose and are thus preferred [1923]. However, the inherent hydrophilic nature of CNC with –OH side groups limits their widespread use as reinforcement fillers. The mechanical strength of hydrophobic polymer-reinforced CNC is unfavorable due to the incompatibility between the two counterparts, leading to weak dispersion and serious agglomeration of the CNC [2426]. To solve these compatibility issues, surface functionalization of nanocellulose, such as TEMPO oxidation, acetylation, acylation, silylation, and polymer grafting, is performed to promote dispersibility of the CNC in hydrophobic polymers by changing the hydrophilic nature of CNC. However, these methods require tedious processing routes and may disrupt the structural integrity of the CNC [27]. Hence, CNC are best utilized in a natural state.

Despite its limitation in widespread applications, neat PVOH can be modified to suit the needs for its end-use application. PVOH crosslinking is one of the best solutions as this method is capable of inducing a tight three-dimensional structured network of molecules that demonstrate hydrophobicity and enhanced properties [2832]. The hydroxyl groups present in PVOH are the target of reactions, most commonly with chemical compounds, such as dialdehydes, anhydrides, and carboxylic acids [3338]. Research on crosslinkers such as glutaraldehyde [3941], epichlorohydrin [42, 43], formaldehyde [44, 45], hexamethoxymethylmelamine [46], and sodium trimetaphosphate [47, 48] indicates that they are not preferable owing to their toxicity. Therefore, the focus of this study is the crosslinking reaction between a natural dicarboxylic acid and ethanedioic acid (EA) on PVOH with different CNC ratios as it is the simplest crosslinking reaction and is environmentally friendly. The tensile properties, degree of crystallinity, and water and thermal stability of non-crosslinked and crosslinked nanocomposites are investigated.

2. Methods

2.1. Materials

Dried rice straw was obtained from a paddy field located in Arau, Perlis, Malaysia (6°2558.0N+100°1356.8E). PVOH (98% hydrolyzed) with an Mw of 89,000–98,000 was purchased from Sigma-Aldrich (Malaysia). Sodium hydroxide (NaOH), sodium chlorite (NaClO2), and acetic acid glacial (CH3COOH) was purchased from Merck (Malaysia), Acros Organics (Malaysia), and HmbG (Malaysia), respectively. The sulfuric acid (95%–97%) used for the CNC extraction was purchased from Merck (Malaysia). EA was purchased from Sigma-Aldrich (Malaysia). Deionized water was used consistently throughout the experiment.

2.2. Microcrystalline Cellulose and CNC Extraction

Dried rice straw was washed with water to remove dirt and other impurities before drying in an enclosed environment for 48 h. The rice straw was then converted to a powder form by using a mill grinder prior to separation through a 150 μm mesh screen. Microcrystalline cellulose (MCC) and CNC from rice straw were extracted through alkaline treatment, bleaching, and acid hydrolysis with slight modification to the process described by Chan and coworkers [49]. Next, the separated rice straw powder was treated with a 4% aqueous solution of NaOH () for a total of 2 h at 80°C under mechanical stirring. Then, the rice straw fiber was washed several times with deionized water until the pH reached 7. Subsequently, the rice straw fiber was bleached with a solution comprising equal parts () acetate buffer (40 g NaOH, 75 mL glacial acetic acid, diluted to 1 L with distilled water) and sodium hypochlorite (1.7 wt% NaClO2). This treatment was performed at 80°C for 2 h under mechanical stirring. The fiber was washed with deionized water until the pH reached 7 again. A total of four cycles of alkaline and bleaching treatment were completed to obtain a milky white MCC pulp. Acid hydrolysis was performed with 64 wt% H2SO4 at the optimal temperature of 45°C for 45 min. This process is important to eliminate the amorphous regions of cellulose to achieve the desired highly crystalline CNC. Immediately after 45 min, tenfold of cold deionized water was added to halt the hydrolysis reaction. The suspension was then washed with a copious amount of deionized water and centrifuged at 10,000 rpm. This process was repeated until a CNC suspension with a pH of 6–7 was achieved. Gravimetric method which consist of convection oven drying of CNC at 105°C was adopted to determine the final concentration of CNC in the final suspension after sonication [50].

2.3. Preparation of PVOH/CNC Nanocomposites

PVOH/CNC-film nanocomposites were produced using the solution casting method. PVOH powder was dissolved in water at 90°C with mechanical stirring for 2 h in order to obtain a 2 wt% solution. CNC of known weight percentages (1, 3, 5, and 7 wt%) were added concurrently. For nanocomposite films with EA, 5 wt% of EA was added during the final 30 min to prevent acidolysis. The final solution was poured into a glass mold and oven-dried at 40°C for 24 h. Thermally crosslinked films were obtained via thermal curing of the films with EA in a convection oven at 130°C for 30 min. All the films were stored in a desiccator for 72 h before the measurements were performed.

2.4. Fourier Transform Infrared Spectroscopy

Fourier transform infrared spectroscopy (FTIR) spectra were obtained through a small cut of PVOH/CNC and PVOH/CNC/EA nanocomposite films in the range of 4,000 to 650 cm−1 wavenumbers with 32 scans using a PerkinElmer Spectrum 10 spectrophotometer. Beforehand, the samples were washed thoroughly with water and acetone to remove the unreacted excess EA on the surface of the samples. The samples were then oven-dried to remove moisture before analysis was conducted.

2.5. X-Ray Diffraction

A Bruker D2 Phaser X-ray diffractometer with a monochromatic Cu–Kα radiation source was used to obtain the X-ray diffraction (XRD) patterns of the CNC and the PVOH nanocomposites. The measurement parameters were 0.02° step size in the range of 10° to 40° of a angle. The crystallinity index (CrI) was determined by calculating the peak area from the diffraction peaks using the Segal method. The formula of the calculation is as follows: where CrI denotes the degree of crystallinity (%), is the maximum intensity of lattice diffraction, and is the diffraction intensity at .

2.6. Tensile Properties

The tensile properties of the nanocomposite films were examined on a universal testing machine (Instron 5567) in accordance with the ASTM D-882 standard procedure. A total of five rectangular samples were cut precisely from the nanocomposite films, and the test was conducted at a crosshead speed of 20 mm/min. The tensile strength, elongation at break, and Young’s modulus were obtained.

2.7. Field Emission Scanning Electron Microscopy

The surface morphology of the tensile-fractured surfaces of the nanocomposite films was observed using a field emission scanning electron microscope (FESEM, Carl Zeiss Leo Supra 50 VP Field Emission). A small cut of the fractured surface from each of the nanocomposites was adhered onto the specimen stage with carbon tape. All the samples were sputter-coated with platinum before they were examined at an accelerated voltage of 5 kV under a high-vacuum condition.

2.8. Transmission Electron Microscopy

To determine the morphology of the CNC, 10 μL of 0.005 wt% CNC suspension was placed on a transmission electron microscopy (TEM) grid for 2 min. Excess liquid was removed with filter paper, followed by negative staining with uranyl acetate solution for another 2 min. Excess uranyl acetate solution was blotted with filter paper before letting the CNC dry at an ambient temperature. The CNC were observed under a transmission electron microscope (LIBRA 120) at 100 keV. The average length and diameter of the CNC were determined from 50 representative CNC using the ImageJ and Origin software. Gaussian distribution curves were obtained using the Origin software.

2.9. Thermogravimetric Analysis

Thermogravimetric analysis (TGA) and derivative thermogravimetry (DTG) curves were obtained through the thermal behavior analysis of the nanocomposites using a thermogravimetric analyzer (TGA Q50 V20.13 Build 39). A nanocomposite film (estimated weight: 5 mg) was cut and placed on the heating pan before heating from room temperature to 650°C at a rate of 10°C/min. Nitrogen gas purging was performed before the start of the measurement. The temperature values obtained when the weight loss of the sample reached 10% and 30% of the original weight are denoted as and . is the temperature at maximum weight loss of the sample.

2.10. Moisture Vapor Permeation Rate

The moisture vapor permeation (MVP) rate was determined in accordance with the wet-cup method (ASTM E96). Glass bottles with a mouth opening of 1 cm2 were filled with deionized water and sealed with adhesive to prevent the escape of water vapor. The film samples were placed on top of these bottles and in a humidity chamber that was set to maintain an ambient temperature of 25°C with a relative humidity of 40%. The weight of the bottle was determined every 24 h for 7 days. The MVP was calculated using

is the initial weight of the bottle filled with water and nanocomposite film. is the mouth area of the bottle. is the time (hours).

2.11. Swelling and Solubility Indices

Nanocomposite films () were prepared and immersed in 100 mL deionized water for 24 h at room temperature. After 24 h, the nanocomposite films were removed and dried using filter paper before the weight gain and swelling were measured. The swelling index (SwI) was calculated using Equation (3). Due to the difficulty of the dissolution of fully hydrolyzed PVOH in water at room temperature, the solubility index (SoI) was determined through film immersion in heated deionized water (90°C) for 15 min. SoI was calculated using Equation (4).

is the initial weight of the sample. is the weight gain of the sample. is the remaining weight of the sample.

3. Results and Discussion

3.1. FTIR

Figure 1 presents the FTIR spectra of pure PVOH, crosslinked PVOH, and crosslinked and non-crosslinked nanocomposites with 3 wt% CNC. There are minor differences in the spectrum curves between the parallel compositions of nanocomposites. The broad peak at 3,230 cm−1 of all the nanocomposites indicates the availability of the –OH groups in the intermolecular and intramolecular hydrogen bonding between PVOH and CNC [51]. The peak at 2,921 cm−1 for all the nanocomposites indicates the vibration of the aliphatic C–H alkyl groups. The bands appearing at wavenumbers of 912 and 842 cm−1 show the presence of the stretching vibrations of the C–H bond in PVOH [52]. For the crosslinked nanocomposites, the appearance of a small peak at 1,750 cm−1 proves the success of the crosslinking reaction due to the formation of ester C=O bonds [53]. The peak at the wavenumber of 1,680 cm−1 for all nanocomposites is due to the presence of adsorbed water, whereas the peak at 1,380 cm−1 indicates the presence of the C–H vibration band. The peak intensity increased for the nanocomposites with CNC at 1,040 cm−1 (but it did not for pure PVOH) due to the C–O–C vibration from the pyranose ring present in the cellulose [32].

3.2. Interactions between Polyvinyl Alcohol, Cellulose Nanocrystals, and Ethanedioic Acid

As presented in Figure 2(a), the crosslinking mechanism is attributed to the esterification reaction between PVOH and EA. Intermolecular diester bonds are formed as a result of the linkage between the carboxyl groups of EA with the hydroxyl groups of PVOH in high-temperature environments (as was proven in Section 3.1). Either one or both carboxylic acid end groups of EA will react with the hydroxyl end group of PVOH, resulting in the formation of strong crosslink bridges that improve the SwI, SoI, and tensile properties of the nanocomposites. Conversely, Figure 2(b) presents the interaction between the hydroxyl groups of PVOH and CNC through the formation of hydrogen bonds.

3.3. X-Ray Diffraction

Figure 3 presents the XRD diffraction patterns of MCC, CNC, pure PVOH, crosslinked PVOH, and crosslinked and non-crosslinked nanocomposites (3 and 7 wt%). The well-defined characteristic peaks of the cellulose type I lattice structures, occurring around 16.1° and 22.4°, are observed for CNC [54]. The unaltered diffraction peaks of the CNC in comparison with MCC confirm that the acid hydrolysis treatment employed to extract CNC through the removal of unwanted amorphous regions did not affect the crystalline nature of the cellulose. The narrower and sharper peak demonstrated by CNC at 22.4° indicates that it exhibits a better crystal lattice structure and a higher cellulose index than MCC [55, 56]. This translates to a greater reinforcing ability on the tensile properties of PVOH (discussed in Section 3.6). As presented in Table 1, the CrI of CNC increased to 74% due to the release of individual crystallites from the hydrolytic cleavage of the glycosidic bonds in the amorphous regions. For both the non-crosslinked and crosslinked nanocomposites, the CrI improved with the addition of CNC due to the increase in the amount of crystalline material in the system [57]. At a higher filler content of CNC (7 wt%), the CrI decreased. This is mostly due to the effect that the steric hindrance of CNC, which is the retardation of chemical reactions that disrupts the formation of intermolecular bonds between molecules, has on the organized structure of the carbon–carbon polymer chains in PVOH. A similar occurrence was reported by Hoseini et al. [58] when the CrI dropped during high carbon nanotube loading in a polyamide-6 polymer. The addition of EA as the crosslinker decreased the CrI of the nanocomposites. This occurred because after crosslinking, the interactions between the PVOH chains limited the mobility of the molecular chain motion, hampering the crystallization process and negatively affecting the crystallinity of the crosslinked films. A similar decrease in crystallinity findings was observed by Nataraj et al. [59] when chitosan films were crosslinked with citric acid. Wu et al. [60] also reported a decrease in peak intensity (indicating a drop in the percentage of crystallinity) after crosslinking starch films with citric acid.

3.4. Tensile Fracture Surface

Figures 4(a)4(f) present the FESEM micrographs of the tensile-fractured surfaces of non-crosslinked and crosslinked nanocomposites exhibiting different morphologies. From Figures 4(a) and 4(b), it can be observed that the crosslinking of pure PVOH resulted in a much smoother fracture surface when compared with pure PVOH. Similar findings were reported by Jain et al. [61] on crosslinked PVOH with hydrochloric acid. Figure 4(c) presents a rougher, bristly cross-sectional fracture due to the existence of strong interactions between the CNC and PVOH matrix. These findings indicate that the non-crosslinked 3% nanocomposite suffered from a brittle fracture. This is also supported by the results presented in Figure 5, in which the incorporation of CNC increased Young’s modulus, thereby increasing the brittleness of the nanocomposite. As can be seen from Figure 4(d), the crosslinked 3% nanocomposite exhibited a smoother, wavelike fracture surface structure, suggesting a relative homogeneous dispersion of CNC within the nanocomposite. Similar findings were also obtained in PVOH/functionalized CNC composite films by Chen et al. [62]. There are also no observable internal cracks or phase separations attributed to the greater tensile strength of crosslinked nanocomposites when compared with non-crosslinked nanocomposites [62]. In Figures 4(e) and 4(f), the voids, aggregates, and coarser surfaces shown throughout the fracture surfaces led to a weaker transfer of strength. This directly translated to a decrease in tensile strength and Young’s modulus.

3.5. Morphology

As can be seen from Figure 6(a), the CNC extracted from rice straw appeared as long, well-defined rodlike crystals. The Gaussian distribution for the CNC diameter presented in Figure 6(b) confirmed that it was in the range of 3–14 nm. The standard deviation, reduced chi-squared, and -squared values obtained were 2.1 nm, 1.3, and 0.88, respectively. As can be seen from Figure 6(c), the length of the CNC obtained was in the range of 100–800 nm. The standard deviation, reduced chi-squared, and -squared values obtained were 139.0 nm, 9.74, and 0.68, respectively. The average length was 331.97 nm, and the average diameter was 8.11 nm. The aspect ratio of the CNC, calculated from the ratio of the length over the diameter, was 41. This value was higher than the aspect ratio of the rice straw CNC extracted by Lu and Hsieh [63]. This result demonstrated that the preliminary cyclic treatment performed on the CNC, prior to extraction, was effective. Moreover, the aspect ratio reported in this study was higher than that of the CNC extracted from other reported cellulosic sources, such as commercial MCC (16.8), corncob (37), switchgrass (38), cotton (13), and onion skin (10.1) [6467].

3.6. Tensile Properties

Figure 7 presents the comparison of the tensile strengths of the crosslinked and non-crosslinked nanocomposites with varying CNC contents. The tensile strength of non-crosslinked nanocomposites improved (increasing in value to 60.4%) as the CNC content increased up until 3 wt%. The crosslinked nanocomposites also demonstrated an upward trend, but the increase was much more pronounced (up to 104.8%) due to additional factors, such as the improved transfer of strength from the crosslinking network to the nanocomposite. The excellent compatibility of PVOH with CNC facilitated the homogeneous distribution of the CNC throughout the PVOH, as is shown in Figure 4(c). This, coupled with the success of hydrogen intermolecular and intramolecular bond formation, led to the positive enhancement in tensile strength. The nature of CNC, which is widely known to be strong and possess large interfacial surface area, also contributed to such an enhancement [68]. In CNC above 3 wt% for both the crosslinked and non-crosslinked nanocomposites, factors such as reaching the filler-saturation point and agglomeration start to exert detrimental effects on the tensile strength of the nanocomposites [69]. In the case of the crosslinked nanocomposites, the improvement in tensile strength is attributed to the formation of the crosslink bridges through the intermolecular ester bonding between the PVOH chains (as is illustrated in Figure 2(b)). This interaction within the matrix contributed to the development of a more compact structure than the non-crosslinked nanocomposites. The drop in strength beyond 3 wt% may be due to the reduced availability of the hydroxyl groups as they were already used to form the crosslinks. A reduced number of hydroxyl groups prevents excess CNC from forming hydrogen bonds with PVOH, leading to unreacted CNC. This unreacted CNC agglomerates, forming random localized weak points on the nanocomposite.

Figure 8 presents the comparison of elongation at break values between crosslinked and non-crosslinked nanocomposites with varying CNC contents. In all cases, the increase in the amount of CNC slightly reduced the elongation of break in nanocomposites due to the addition of CNC with known stiff and rigid structures as reinforcement filler materials [68]. Consequently, the ductility of the nanocomposites was reduced (due to the diminishing content of PVOH) when the CNC content increased. A similar trend was observed in crosslinked nanocomposites in which the elongation of break was found to be much lower than that of non-crosslinked nanocomposites. Such a reduction in elasticity after crosslinking indicates that the nanocomposites lost their ductile nature as their polymer chain mobility was reduced by the formation of crosslinked networks. Jose reported similar findings when citric acid was used as the crosslinker for PVA/starch/graphene nanocomposites [70]. The elongation at break values was also found to decrease when glutaraldehyde was used a crosslinking agent in polyvinyl alcohol/chitosan composites [71]. There is even a phenomenon demonstrating that crosslinkers exert a plasticizing effect, but this is at the expense of tensile strength. Wu et al. [60] demonstrated that when a higher citric acid content was used for crosslinking, the elongation at break was favored, but it significantly reduced the tensile strength. The unreacted excess citric acid content acted as the plasticizer, reducing the interactions between the PVOH carbon chains. Therefore, the incorporation of crosslinking agents in optimum amounts can improve the properties of the nanocomposites. However, any excess crosslinking agent will act as a plasticizer and cause a negative impact.

The significant increase in Young’s modulus for both crosslinked and non-crosslinked nanocomposites up until 3 wt% is presented in Figure 5. With 1 wt% CNC content in non-crosslinked and crosslinked nanocomposites, the increase in Young’s modulus was 12% and 17%, respectively. With 3 wt% CNC in non-crosslinked and crosslinked nanocomposites, the highest increase in Young’s modulus was approximately 49% and 59%, respectively. The addition of stiff CNC to the polymer matrix led to the formation of rigid CNC networks that translated to the stiffness and improved strength of the nanocomposites. The 3 wt% CNC non-crosslinked nanocomposites exhibited a greater increase in Young’s modulus than PLLA-grafted CNC with a similar CNC weight percentage [72]. Young’s modulus also significantly increased after crosslinking. Ben Shalom et al. [73] reported that crosslinking of PVOH/CNC with 1,2,3,4-butanetetracarboxylic acid further improved the stiffness and strength of the composites. These findings are in agreement with the results presented in Figure 5.

3.7. Thermogravimetric Analysis

TGA was conducted to analyze the thermal behavior of crosslinked and non-crosslinked nanocomposites of varying CNC contents. The three important regions of weight loss are disclosed in the thermogravimetric curve presented in Figure 9. The initial weight loss occurred at approximately 40°C–100°C and is observed for all nanocomposites. This event is attributed to the loss of water through evaporation, mostly of free bound water molecules that are loosely attached to the surface of the nanocomposites [44]. The next weight loss event occurred at around 210°C–400°C and is attributed to the degradation of the nanocomposites. The nanocomposites lost approximately 70% of their total weight, and this region is the most important for indicating their thermal stability. Weight loss is mainly attributed to the degradation of volatile products such as intrinsic water content and unsaturated compounds.

The last weight loss event occurred at greater than 400°C (at approximately 440°C) and is attributed to the degradation of polyene, hydrocarbons, and chain scission of the nanocomposites [74]. As presented in Table 2, crosslinked nanocomposites were found to exhibit greater , , and values. After the occurrence of crosslinking, the , , and values of pure PVOH increased from 239°C, 273°C, and 275°C to 243°C, 291°C, and 342°C, respectively. A similar increase was observed for the , , and values of crosslinked PVOH with the addition of CNC, from 241°C, 285°C, and 287°C to 280°C, 348°C, and 364°C, respectively. The increase in the thermal decomposition temperature indicated the occurrence of crosslinking. These results are in agreement with the findings of Stelescu et al. [75] in which the occurrence of crosslinking of ethylene propylene diene monomers with butyl rubber composites was proven by the resulting improvement in the , , and values. This improvement is highly beneficial as the crosslinked nanocomposites are more thermally stable and more resistant to thermal degradation.

Figure 10 presents the derivative weight loss curve of the nanocomposites. Crosslinked PVOH had four degradation steps, whereas pure PVOH and the other nanocomposites had only three degradation steps. Sonker et al. [76] reported similar findings on crosslinked PVOH with suberic acid and terephthalic acid. For the crosslinked nanocomposites with 3 and 7 wt% of CNC, the addition of CNC caused the second and third steps to coalesce, forming only one strong derivative weight peak. Song et al. [77] presented similar findings on the thermal degradation steps for the Fenton free radical crosslinking of PVOH with the addition of CNC. It is also evident in Figure 10 that the addition of CNC increased the maximum degradation temperature of the nanocomposite with 3 wt% CNC but reduced it when 7 wt% CNC was added. At higher filler content, factors such as CNC agglomeration dominated, and the thermal stability of the nanocomposites was reduced. A comparable occurrence on the thermal behavior of PVOH/CNC extracted from commercial cellulose at higher filler loading was also demonstrated in the report by Song et al. [77]. The degradation peak from the DTG curve indicates an increased thermal stability of the nanocomposites with 3 and 7 wt% CNC when compared with pure PVOH. As presented in Table 2, the value significantly increased after crosslinking, most prominently for the nanocomposites with 3 wt% (from 287°C to 364°C). The obvious shift of the peak of the crosslinked nanocomposites to the right indicated that there is a great improvement in thermal stability.

3.8. Water Vapor Permeation Rate

The MVPs of crosslinked and non-crosslinked nanocomposites with varying CNC contents are presented in Figure 11. Moisture vapor transmission through films (especially hydrophilic films) independently varies with its degree of hydrophilicity. Moisture tends to interact with the polymeric matrix of the film, increasing its moisture permeability. Non-crosslinked nanocomposites are more inclined to have a higher MVP value compared with crosslinked nanocomposites owing to their higher hydrophilic behavior. During the initial exposure of nanocomposite films to moisture, swelling slowly occurs, causing distortions in the polymer chain conformation within the films [78, 79]. At 48 h, the MVP value starts to taper and remains at a steady state. Water, acting as a plasticizer, loosens the polymer chains, which leads to an increase in the MVP [80]. Pure PVOH most clearly demonstrates this effect. The addition of CNC substantially reduced the MVP value for all the nanocomposites. At higher CNC concentrations, most of the water molecules are blocked from passing through the polymer matrix, as the increase in CNC disrupts the water molecule pathways. CNC significantly reduces the diffusibility of moisture vapor owing to its size and swelling constraints, as well as its strong network formation in nanocomposite films. Nair et al. [81] obtained similar findings on the reduction in the MVP of nanocellulose films. The reduction in the MVP is more pronounced with crosslinked nanocomposites with increased CNC content. This is due to the synergistic effect from the crosslinked bridges with CNC, thereby favoring their application as thin packaging films for moisture-sensitive materials, such as food and beverage products [82].

3.9. Solubility and Swelling Index

The SoI and SwI are presented in Figure 12. The results indicate that the SoI of the crosslinked nanocomposites significantly decreased when compared with that of the non-crosslinked nanocomposites, from an estimated index of 1 to 0.25. This result proves that the proposed crosslinking reaction with EA (as was discussed in Section 3.2) was successful as it increased the hydrophobicity of the nanocomposites to water. However, the addition of CNC only minimally affected the SoI index. This is likely due to CNC’s similar hydrophilic nature to the polymer matrix. The SwI was at its lowest in the 3 wt% of PVOH/CNC nanocomposite when compared with pure PVOH and the nanocomposite with 7 wt% of CNC. A similar trend was observed for the crosslinked nanocomposites. PVOH is a hydrophilic material that tends to easily attract water molecules owing to its free hydroxyl groups. The difference in the swelling behavior is attributed to its crystallinity, where the non-crosslinked nanocomposite films with higher CrI values augmented the difficulty of the water to penetrate the PVOH network [83]. The formation of dense, three-dimensional network structures as a result of crosslinking additionally reduced the intermolecular spaces. This favorably affected the SwI value (even at a lower CrI value) for the non-crosslinked nanocomposites [84, 85]. According to Alessandra Bersanetti et al. [84], the reduction in the SwI after crosslinking was due to the diminished availability of excess hydroxyl groups to form hydrogen bonds with water molecules. In the case of PVOH with 7 wt% CNC, for both crosslinked and non-crosslinked nanocomposites, agglomerated and unreacted CNC imparted voids and imperfections throughout the film, which are presented in Figures 4(e) and 4(f), respectively. These findings consequently resulted in an increase in the SwI.

4. Conclusion

This study successfully extracted cellulose nanocrystal (CNC) from rice straw and used it as a reinforcement filler on polyvinyl alcohol. The utilization of EA as a crosslinker proved successful and produced very favorable results. A distinctive improvement in the tensile strength and Young’s modulus of non-crosslinked nanocomposites and an even greater improvement in crosslinked nanocomposites were observed. The XRD data indicated that there was an increase in the degree of crystallinity of the nanocomposites when CNC was added. After crosslinking, the nanocomposites exhibited a decrease in the degree of crystallinity. The extracted CNC had a high aspect ratio of 41 and appeared as long, well-defined rodlike crystals. The maximum degradation temperature of the nanocomposites increased at the optimum content of 3 wt% CNC. An additional increase was observed after crosslinking. Crosslinking reduced the values of MVP, solubility index, and swelling index, favoring its application in thin-film packaging for moisture-sensitive materials. Although the addition of CNC is promising and this study revealed its benefits by improving on certain niche properties of PVOH, it does not favor widespread use. Crosslinking of the nanocomposite to drastically reduce its hydrophilicity and significantly improve its tensile and thermal properties must be performed before it can be widely used. In conclusion, the optimum CNC content was 3 wt% for both crosslinked and non-crosslinked nanocomposites.

Data Availability

The experimental data used to support the findings discussed in this study are included within the article.

Conflicts of Interest

The authors declare no conflict of interest.


The authors are thankful to Universiti Malaysia Perlis for providing the facilities necessary to perform this research. This research was performed as a part of the employment of Universiti Malaysia Perlis.