2.1. Eco-Friendly Preparation of NC and HBL from CS
CS represent a significant agricultural waste in both China and the United States.[32–33] However, they contain abundant carbohydrates such as cellulose, hemicellulose, and lignin, which can be converted into functional materials.[33] In this study, CS was chosen as the source material for producing NC and HBL. Notably, instead of employing traditional alkali treatments[32] or TEMPO[34] treatments, a potentially environmentally friendly DES, glycine-lactic acid (DESLG), was utilized to process the CS powder. The resulting insoluble solid was collected and further processed to obtain NC, whereas the remaining DESLG solution was used to prepare HBL and recover the DESLG.
After treating the insoluble solid with a bleaching solution, the resulting powder was analyzed using Fourier-transform infrared spectroscopy (FT-IR), X-ray photoelectron spectroscopy (XPS), and solid-state 13C nuclear magnetic resonance (SS 13C-NMR). The FT-IR spectra (Fig. 2a) indicated the disappearance of the characteristic peak at 1,520 cm− 1, which corresponds to the C = C stretching vibration of lignin.[35] Instead, characteristic peaks at 1,050 cm− 1 and 837 cm− 1 emerged, indicating C-O stretching and glycosidic linkage vibrations between the d-glucose units of cellulose.[36] These findings confirmed that the powder consisted of cellulose isolated from CS (CSC). Additionally, a new peak at 1,730 cm− 1 in the FT-IR spectra (Fig. 2a), attributed to C = O stretching vibration,[37–38] suggested that the CSC underwent modification involving lactate acid or glycine during DESLG processing. XPS survey of CSC (Fig. 2b and c) revealed a peak at 288.3 eV assigned to C = O, whereas no N was detected in the XPS spectra,[39] indicating that CSC was modified solely by lactic acid. To further elucidate the chemical structure of the prepared CSC and confirm the presence of ester or ether bonds between cellulose and lactic acid, SS 13C-NMR analysis was employed. The obtained spectra displayed a chemical shift at δC 173.8 ppm, corresponding to C = O,[38] and another peak at δC 20.5 ppm, representing a methyl group,[38, 40] both characteristic of lactate acid. Signals attributed to glycine (C-N) were absent, reinforcing the conclusion that CSC was modified by lactate. Moreover, curve-fitting of the peak at δC 173.8 ppm (assigned to C = O) revealed two divided peaks at δC 174.6 ppm and 171.6 ppm, corresponding to ester bonds and carboxyl groups,[40] respectively (Fig. 2d and e). The SS 13C-NMR results confirmed that CSC was indiscriminately modified by lactic acid through both ester and ether bonds. Next, transmission electron microscopy (TEM) was employed to observe the CSC suspension. TEM images revealed the presence of nanofiber-like cellulose in the suspension, with an average aspect ratio of 112.0 ± 52.0 (Fig. 2f). These aspect ratios were larger than those of cellulose nanocrystals (CNCs) but smaller than those of cellulose nanofibrils (CNFs). Therefore, the name "NC" was assigned to them.[41] Additionally, the aspect ratio of NC obtained through DESLG pretreatment was significantly higher than that obtained through sulfuric acid hydrolysis for 30 min (61 ± 24) (Figure S1a). This can be attributed to the mild reaction conditions in DESLG pretreatment and the inhibitory effect of non-cellulosic components in CS on cellulose hydrolysis during the DESLG processing. In contrast, sulfuric acid is too strong and can hydrolyze both the amorphous and crystalline regions of cellulose.[41] Furthermore, the NC obtained through DESLG processing showed excellent dispersion in distilled water with a polydispersity index of 0.7, and the NC suspension remained stable without aggregation for up to 3 months. This is likely due to the presence of lactate moieties on NC, which introduce negative charges that repel each other, as indicated by the zeta potential of -24 mV for the NC suspension.
Meanwhile, water was used to precipitate the dissolved substance in the DESLG solution.[42–43] The resulting brownish powder (Fig. 2g) was analyzed using 2D-HSQC NMR and FT-IR. The 2D-HSQC NMR spectroscopic analysis revealed that the chemical constituents of the obtained powder consisted of Syringa (S) (S2,6, δC/δH 103.5/6.70), Guaiac (G) (G2, δC/δH 110.4/6.99; G5, δC/δH 116.2/6.78; G6, δC/δH 119.5/6.79), p-coumaric acid (PCA, δC/δH 130.1/7.46), p-benzoate (PB, δC/δH 131.6/7.77), and p-hydroxycinnamyl alcohol terminal groups (I, δC/δH 66.2/4.20), which are characteristic units of lignin. Furthermore, these units were primarily interconnected by β-O-4' aryl ether linkages (Aα, δC/δH 72.4/4.92; Aβ, δC/δH 83.8/4.39 and Aγ, δC/δH 60.2/3.50) and Cβ-Cβ' bonds (Bα, δC/δH 83.8/4.97 and Bγ, δC/δH 71.2/3.81), forming a three-dimensional structure[44–48] (Fig. 2h-i and S1c). Similarly, the FT-IR spectrum of the powder showed characteristic peaks associated with lignin (Figure S1b). Collectively, these analyses confirmed that the brownish powder constituted natural lignin and demonstrated the successful preparation of HBL.
Notably, the preparation process of NC and HBL from CS was established on a pilot scale, with yields of 22.1 ± 0.5% and 13.7 ± 0.6%, respectively (Table S1). The total biomass utilization of CS reached 35.8%, significantly higher than that of CS treated with alkali (26.4%). Furthermore, the DESLG could be recycled by evaporating water under a vacuum atmosphere, with a recovery rate of 73.1% (Table S1). The recovered DESLG could be reused four more times for processing CS, with consistent performance achieved by replenishing the lactic acid to its initial weight to compensate for the loss caused by cellulose modifications (Table S1). This demonstrated the favorable recyclability of DESLG. Additionally, the processing of CS using DESLG (four times) imposed a low environmental burden, as indicated by an E-factor of 27.3. These characteristics highlight the eco-friendly advantages of this technique and suggest its potential for industrialization. While, further endeavors should be devoted to the recovery of the residuals in the used DESLG, probably hemicellulose, and develop their utilizations, hoping to achieve full recycling of CS.
2.2. PNC Nanocomposite Film
The obtained NC was employed as an enhancer to improve the mechanical properties of pullulan films. As a result, the addition of NC increased the tensile strength (TS) of the PNC film from 43.7 ± 2.6 MPa to 76.6 ± 1.9 MPa as the NC content increased from 0 mg/g pullulan to 25 mg/g pullulan (Fig. 3a). However, a notable decrease in TS was observed when the NC content reached 30 mg/g pullulan (Fig. 3a). Furthermore, the elongation at break (EB) decreased significantly from 12.2 ± 0.3% to 4.7 ± 0.2% with an increasing NC content of 0 mg/g pullulan to 30 mg/g pullulan (Fig. 3a). PNC20 and PNC25 films exhibited considerably higher TS (approximately 75.6 MPa) than other natural polymeric materials such as DNA-based bio-plastic and poly(epichlorohydrin-co-ethyleneoxide)-based ionomers (1 MPa),[49] chitosan-cellulose nanofiber-hemicellulose composite films (38.6 MPa),[50] and soybean protein nanostructured films (15.6 MPa).[51] Moreover, the TS of PNC20 and PNC25 films outperformed that of conventional fossil-based packaging plastics such as polystyrene (PS) (30.0–35.0 MPa), polypropylene (PP) (25.0–40.0 MPa), polyethylene terephthalate (PET) (75.0 MPa), and low-density polyethylene (LDPE) (15.0 MPa)[52] as well as recyclable synthetic materials, including poly(γ-butyrolactone) (PγBL) and poly(trans-hexahydrophthalide) (PT6HP) (7.4–28.8 MPa),[9] and bio-based plastics like poly-hydroxyalkanoates (PHAs) (40.0 MPa) and PLA-poly(butyleneadipate-co-terephthalate) (PBAT) composites (65.0 MPa)[53] (Fig. 3b). Additionally, the EB values (approximately 7.3%) of PNC20 and PNC25 films were significantly lower than those of the aforementioned materials (Fig. 3b and Table S2). While the reduced EB may limit the use of PNC films as tensile packaging plastics like cling films, it can prevent PNC-based packaging from excessive stretching when used for PLA-based packaging plastics and lifting of heavy objects. Moreover, the TS and EB of PNC20 and PNC25 films showed no significant differences across varying thicknesses (0.03 mm to 0.13 mm) (Table S3). The maximum force at break (MF) for the thinnest PNC20 film (thickness: 0.03 mm) was 35.4 N, whereas that for the 0.03 mm-thick PLA-PBAT film was 2.5 N (Table S3). As a result, a 0.03 mm-thick PNC20 film belt (length: 8 cm; width: 0.5 cm) could easily lift 1 kg of weight without being fractured or noticeably stretched (Figure S2a). Overall, these results indicate that the mechanical performance of pullulan films can be significantly enhanced by incorporating pullulan-NC composites. PNC nanocomposite thin films demonstrate mechanical properties competitive with those of other plastic/bioplastic materials currently in practical use or under investigation. This development meets the fundamental requirement for their potential application as alternative materials for food packaging.
Importantly, no significant differences were observed in the TS and EB between the PNC20 and PNC25 films, highlighting the need to confirm the ideal film formulation. Additionally, the substantial enhancements in the mechanical properties of the PNC films prompted us to identify a general principle for enhancing the mechanical strength of other polysaccharide-based materials. To achieve this objective, the distribution of NC within the PNC films was initially investigated using X-ray diffraction (XRD) patterns, atomic force microscopy (AFM), and polarized light microscopy (POM). The XRD patterns revealed a single broad diffraction peak at 19° in the spectra of the PNC5-PNC20 films, indicating the presence of amorphous pullulan.[54] However, this peak split into two peaks at 19° and 22.6° for the PNC25 and PNC30 films, respectively (Fig. 3c). The appearance of the 22.6° peak suggested the occurrence of crystalline-state NC[55] and the phase separation of NC from the pullulan matrix. Furthermore, AFM and POM observations of the PNC films demonstrated that the surfaces of the PNC20 films were highly smooth without any noticeable aggregates. In contrast, the surface of the PNC25 film exhibited slight roughness with the formation of small aggregates, whereas the PNC30 films displayed obvious roughness with visible aggregates (Fig. 3d-f and S2b). This can be attributed to the gradual aggregation of NC within the pullulan matrix as the concentration increased. Consequently, NC was uniformly dispersed in the PNC5-PNC20 films, whereas aggregation began to occur in the PNC25 film. However, this aggregation became significant in the PNC30 film, which likely contributed to its substantial decrease in mechanical strength. Overall, based on the findings, the optimal film formulation was determined to be PNC20. The improved dispersion of NC within the pullulan matrix in the PNC20 film likely resulted in stronger internal interactions,[56] as subsequently confirmed by the following assays.
As observed in the preceding analysis, modification of NC with lactic acid, which has received limited attention in previous studies, was conducted in this study. The disparate dispersions of NC in the PNC20 and PNC30 films (Fig. 3d-f and S2b) may give rise to diverse intermolecular interactions. Subsequently, the internal interactions within the PNC films were examined by investigating the rheological behavior of the pullulan-NC composite solution (PNS) and conducting FT-IR analyses of the PNC films. The rheological behavior of the PNS revealed that the intermolecular interactions were strengthened by the incorporation of NC into PNS, reaching a maximum effect at an NC content of 20 mg/g pullulan (Figure S3). Moreover, the FT-IR spectra of the PNC films displayed peak shifts in the hydrogen group region, ranging from 3,000 cm− 1 to 3,700 cm− 1,[57] when compared with those of the single pullulan film. This finding indicated the formation of hydrogen bonds in the tested PNC films (Fig. 3g-i and Table S4). Notably, the fraction of hydrogen bonds (FH−OH) in the PNC films progressively increased with the increasing NC content, reaching a peak value of 0.742 ± 0.008 in the PNC20 film (Fig. 3i). These results suggested that the formation of additional intermolecular hydrogen bonds played a vital role in enhancing the mechanical properties of pullulan through the inclusion of a nanofiller such as NC.
Furthermore, the higher FH−OH observed in the PNC20 film compared with those in the PNC30 films corresponded to its more homogeneous dispersion within the pullulan-NC composites (Fig. 3d-f and S2b). This observation indicated that the well-dispersed NC facilitated the formation of a greater number of intermolecular hydrogen bonds. In contrast, the acetylation of hydroxyl groups in NC (yielding acetyl-NC) led to a reduction in the TS of the pullulan-acetyl-NC (PANC) nanocomposite films (Figure S4). This finding demonstrated the significance of intermolecular hydrogen bond quantity in enhancing the mechanical strength of the pullulan-NC nanocomposite film. Therefore, the fundamental approach to enhance the mechanical properties of a film-forming polysaccharide is to employ a nanofiller capable of forming robust intermolecular hydrogen interactions with the polysaccharide, and simultaneously to achieve a homogeneous dispersion of the nanofiller within the polysaccharide matrix.
2.3. Food Packaging Material Consisting of Inner PNC Film and Outer Kraft Paper
Although the PNC20 film showed sufficient mechanical strength, which would benefit the protection of the packaged contents against external forces during storage and transportation, it exhibited an unsatisfactory vapor barrier capability (water vapor permeability [WVP] = 0.032 ± 0.002 g·m·[m2·d·kpa]−1) compared with that of recycled chemical synthetic polymer film (Fig. 4a).[9] Consequently, this film demonstrated intolerance to relative humidity (RH) exceeding 53% after prolonged exposure, leading to a significant decrease in mechanical strength as indicated by its TS (Fig. 4b). However, the PNC20 film exhibited advantageous oxygen barrier capability (oxygen permeability; peroxide value [PV] = 49.4 ± 0.4 meq/kg) (Fig. 4a), folding endurance (Fig. 4c), heat sealability (heat sealing strength = 1.60 ± 0.08 kN/m) (Figure S5a), high transparency (Figure S5b), and non-toxicity (Figure S5c and d). Collectively, these properties indicate that the PNC20 film is suitable for use as a food inner-packaging material, commonly used in food products with individual small packages such as Oreo cookies and Nescafé sticks. However, to prevent moisture ingress, it is necessary to use an outer packaging material to enclose the PNC20 film. In this study, kraft paper, the most conventional and readily available outer packaging material, was selected. When the PNC20 film was sealed within a kraft paper wrap, its TS showed no significant deterioration under various humidity conditions for 7 days (Fig. 4b). This suggests that the combination of the PNC20 film and kraft paper wrap may be suitable for use in food products with individual packages.
Next, to evaluate the performance of this food-packaging system, three different food items, namely soluble coffee powder, biscuits, and olive oil, were individually packed in heat-sealed PNC20 bags and then wrapped with kraft paper (Fig. 4d). The preservation effects were assessed after storing the packed items at 53% RH for 12 months. Figure S6a-d illustrates that the quality of the food items wrapped in the PNC20/kraft paper packaging was comparable to that packed in PET bags (Figure S5e) in terms of moisture content for soluble coffee powder and biscuits and PV for olive oil and crisp biscuits. These results demonstrate that the PNC20/kraft paper packaging can effectively preserve these foods, extending their shelf life. Moreover, upon peeling off the kraft paper wrap, the individual PNC20 bags were capable of maintaining the quality of the foods for up to 3 days without significant moisture adsorption or softening (Figure S6e-f). Additionally, the PNC20 bag demonstrated rapid decomposition in hot water, fully dissolving the coffee within 50 s of stirring (Figure S6g). Therefore, it can also serve as edible packaging for food products that require dispersion in hot water. This packaging combination, employing PNC20 as the inner-packaging material and kraft paper as the outer layer, presents a promising alternative for preserving instant drinks, convenient snack foods and oil flavors (Movie 1).
Importantly, after being buried in soil for only 10 days, the PNC20 bags underwent almost complete degradation, with no visible remnants (Fig. 4e), with a degradation rate of 83.4 ± 1.1% (Table S5). In contrast, under the same conditions, PLA and PBAT films retained their original shapes and showed an average degradation rate of only 1.1 ± 0.3%.[14, 58] This finding is of great significance, as it indicates that the PNC20 inner packaging can undergo rapid degradation in the natural environment, thereby significantly reducing the additional environmental impact and labor required for waste sorting and recycling. Furthermore, kraft paper, which is naturally biodegradable, is an environment-friendly material. Overall, the PNC20 film introduced in this study holds promise as an alternative to conventional plastic food-packaging material.
2.4. Sandwich-Like Lignin-Pullulan-NC (LPLNC) Nanocomposite Film as Food Outer-Packaging Material
To address the inadequate water resistance of the PNC20 film and enable its use as a food outer packaging material, we utilized HBL (prepared as described in section 2.1) to create a sandwich film (LPLNC). The LPLNC film consisted of HBL-based films as the outermost layers and the PNC20 film as the middle layer (Fig. 5a), which were formed through hot pressing. However, notably, HBL alone cannot form films. Therefore, we mixed HBL with NC (initial lignin/NC weight ratio, L/C = 1:2) to fabricate an HBL-NC composite (LNC) film. The resulting LNC film had a thickness of 0.03 mm, which is the thinnest dimension that allows film formation. However, when two layers of this thickness of LNC and one layer of the PNC20 film were hot-pressed together, the resulting LPLNC film became highly brittle and prone to cracking upon folding. To address this issue, we initially adjusted the L/C ratio of the LNC composite film to 1:6 or 1:1, but it had no effect on the brittleness of the LPLNC film. Interestingly, we discovered that the brittleness of the LPLNC films (L/C = 1:2) decreased with an increase in glycerol content in the PNC20 film, while their TS showed a slight decrease. When the glycerol content reached 300 mg/g pullulan in the PNC20 film (referred to as PNC20-G300 film), the resulting LPLNC-2 film exhibited the most balanced performance, with a TS of 36.7 ± 1.3 MPa and folding endurance 1,000 times in magnitude (Fig. 5b-d). Although the L/C ratio of the LNC film did not affect the brittleness of the LPLNC film, as mentioned earlier, it did influence the water resistance properties of the LPLNC-2 film. Specifically, an increase in the L/C ratio enhanced the water resistance of the LNC film. When the L/C ratio was 1:3 in the LNC-2 film, the resulting LPLNC-3 film demonstrated favorable water resistance properties (water absorption: 136.8 ± 11.7%; contact angle: 63.6 ± 2.3° in 350 s) and a TS of 35.1 ± 1.2 MPa (Fig. 5e-g). Thus, the LPLNC-3 film was identified as the optimal film and was fabricated using the PNC20-G300 film (glycerol content: 300 mg/g pullulan) and the LNC-2 film with an L/C ratio of 1:3. The TS of the LPLNC-3 film was higher than that of PLA-PBAT (8.5 MPa) (Figure S7a) and comparable to that of commercial polyolefin films such as PE and PP (20–40 MPa).[59] However, its EB was significantly lower than that of biodegradable PLA-PBAT and commercial polyolefin films (Figure S7a). These mechanical properties indicate that the LPLNC-3 film can be utilized to develop high-strength food-packaging materials. However, further efforts are required to develop LPLNC-based thin films with extensibility, as the current version primarily serves as an elastic food-preserving thin-film wrap.
Different food-packaging materials require varying thicknesses for practical use. For instance, plastic bags used for food preservation and shopping generally have a thickness ranging from 0.05 to 0.07 mm, whereas plastic straws are approximately 0.4 mm thick, and disposable plastic cups are approximately 0.5 mm thick. The LPLNC-3 film mentioned earlier, with a thickness of 0.065 mm, fulfilled the requirements for making food-preserving shopping bags. This same LPLNC film was utilized for making straws and disposable cups. Initially, attempts were made to increase the thickness of the LNC-2 and PNC20-G300 films to 0.07 mm by layering them. However, during hot pressing, the PNC20-G300 film extruded from the interlayer of the two LNC-2 films. This indicated that the thickened film contained excessive material and could not be fully integrated with the outer layers. A new film, LPLNC-M, was therefore created by hot-pressing multiple layers of the LNC-2 and PNC20-G300 films. The layers were alternately overlaid on each other, with the LNC-2 film serving as the outermost layer on both sides (Fig. 5a). By adjusting the number of layers to 21 and 25, the thickness of the LPLNC-M films could be increased to 0.393 ± 0.003 mm and 0.508 ± 0.004 mm,
respectively. Importantly, the TS values of these films were not significantly different from that of the LPLNC-3 film (three layers) (Table S6).
Based on the progress made, LPLNC-M films with various thicknesses were developed for applications in making food-preserving shopping bags (three layers), disposable cups (21 layers), and straws (25 layers) (Fig. 6a-d). The performance of these films was investigated, yielding the following findings. First, a preserving bag made of LPLNC-M successfully stored fresh beef at -20°C for 2 months without becoming brittle or sustaining damage from freezing (Fig. 6a and Figure S7b), and this LPLNC-M bag could be stored at 25℃ and 53% RH for 12 months without significant change in mechanical strength (Figure S7c). Second, an LPLNC-M shopping bag demonstrated its high strength by easily supporting a weight of 700 g without deformation or breakage (Fig. 6b). Third, an LPLNC-M cup exhibited the ability to hold water overnight without leaking (Fig. 6c), even when initially filled with boiling water. However, despite multiple layers being hot-pressed, molding LPLNC-M films into bottles proved challenging. This hurdle may be attributed to the insufficient rigidity of the LPLNC-M films, which warrants further investigation in future studies. Lastly, an LPLNC-M straw (Fig. 6d) successfully siphoned water, maintaining its shape without obvious deformation after being submerged for 1 h (Figure S7d). This demonstrates the potential of LPLNC-M films as a substitute for conventional plastic straws and degradable bioplastic straws, such as the Starbucks straw made of coffee grounds and PLA composites (Fig. 6d). All the performance can also be overviewed in Movie 2.
The LPLNC-M film-based food-packaging materials and expendables demonstrated performance comparable to that of commercial products in recent studies.[51, 60–61] These films exhibited easy biodegradability in natural environments, completely disintegrating after being buried in soil for 5 weeks (Fig. 6e). This biodegradability can be attributed to the components of the films, including pullulan, lignin, and cellulose. In contrast, commercial PLA-PBAT packages maintain their original shape when buried in soil for the same duration (Fig. 6e). However, they degrade after an additional 6 months of composting[62] and take hundreds of years to degrade in the oceans.[58, 63] The only minor drawback of the LPLNC-M film is its dark brown color, which hinders clear observation of the food and beverage inside the packages. Addressing this issue requires the development of advanced techniques to improve the transparency of the film using non-lignin biomass-based materials, which can also enhance water resistance.
Importantly, all the raw materials used for the food-packaging films in this study, including pullulan, NC, and lignin, were obtained from biofabrication processes. These processes involved yeast fermentation and recyclable biorefinery of agricultural stalk waste using DESs. Furthermore, the fabrication of these films involved physical methods such as high-pressure homogenization, drying, and heat pressing. These approaches are much simpler to implement than the chemical processing of wood chips and maize cobs with aldehydes, necessary to convert these inedible biological materials into biodegradable polyesters like dimethylglyoxylate xylose.[16] Consequently, the techniques employed in this study for fully biomass-based food packaging present novel strategies for the production of bioplastic materials with favorable eco-friendly characteristics. However, the sustainability of these materials is yet to be assessed, as bio-based plastics do not inherently offer sustainability superior to fossil-based plastics.[2] Additionally, concerns regarding their cost and energy efficiency must be resolved. Nevertheless, the bioplastic packages in this study exhibit rapid natural degradation, enabling them to easily enter Earth's natural carbon cycle at the end of their lifespan. As a result, they can be converted into carbon sources for fermentation and contribute to the production of stalks for NC and lignin, effectively closing the loop of matter cycling in bioplastic packaging, as demonstrated in this work. This eliminates the need for specialized composting, labor-intensive sorting, and recycling processes typically required for PLA-based plastics. Notably, in several underdeveloped countries, there may be insufficient resources to sort and recycle PLA-based plastics, making the rapid natural degradation of these new bioplastics an advantageous feature. Overall, our findings suggest that bioplastic packaging would have minimal end-of-life impact on the environment and the economy, indicating its potential as an alternative to fossil-based food packaging.