A one-step nitric acid/ethanol pretreatment strategy for scalable and sustainable cellulose nanofibrils production from sugarcane bagasse†

Pengcheng Luan, Gang Wu, Jinming Wan, Yazeng Zhang*, Zhihao Liu, Yishan Kuang, Benping Lin, Chao Liu, Wenkun Luo and Qijie Chen*
School of Chemistry and Pharmaceutical Engineering, Hunan Provincial Key Laboratory of Cytochemistry, Changsha University of Science and Technology, Changsha 410114, China. E-mail: zhangyzeng@csust.edu.cn; chenqijie@csust.edu.cn

First published on 31st July 2025

---

---

1. Introduction

Cellulose nanofibrils (CNFs) derived from natural lignocellulosic biomass have gained great interest across a wide range of advanced applications—including sustainable packaging, biomedicine, energy storage and conversion materials, smart materials, and optoelectronics—owing to their exceptional mechanical strength, transparency, flexibility, biodegradability, and renewability.^1–4 As a result of these desirable properties, the global CNF market is expected to exceed USD 7 billion by 2031, growing at a compound annual growth rate (CAGR) of over 19%.⁵ Many factories around the world have been involved in commercial CNFs production, such as Innventia (100 kg CNF per day), Nippon paper (150 kg per day), and the US Forest Service (500 kg per day).⁶

Typically, CNFs are produced from commercial pulp or cotton fibers by applying intense shear forces to disrupt the hydrogen bonds between cellulose microfibrils. A variety of mechanical methods, including high-pressure homogenization, microfluidization, disk refining, grinding, and ultrasonication, have been developed for this purpose.⁷ Among these, high-pressure homogenization remains the most widely used method for industrial production. However, the energy consumption associated with homogenization alone can reach as high as 20000–30000 kWh MT⁻¹, which hampers the large-scale production and practical deployment of CNFs.⁷

In recent years, several pretreatment strategies have been developed to reduce the energy required for fibrillation of cellulose fibers into CNFs, including adding charged species to the cellulose surface, reducing the polymerization degree or length of fibers, or using suitable solvents to swell and loosen the cell wall matrix.⁸ Typical pretreatment methods, such as TEMPO oxidation for increasing anionic groups of cellulose, the use of organic solvents for swelling, and refining, cryo-crushing, and enzymatic pretreatments for decreasing the fiber length, can significantly decrease the energy requirements for CNF production to less than 1000 kWh MT⁻¹.^7–9

Despite these advancements, several critical limitations hinder the scalability and sustainability of the existing pretreatment techniques. Notably, most pretreatment strategies rely on the use of purified pulp fibers as the starting materials, yet the extraction of these fibers from lignocellulosic biomass itself demands significant energy and chemical inputs. Furthermore, many pretreatment processes involve the use of environmentally harmful reagents, posing significant challenges in terms of recovery and waste management. Thus, an integrated strategy combining lignocellulosic feedstocks (e.g., agricultural residues) with recyclable, cost-efficient pretreatment systems is essential to enable economically viable and scalable CNFs production.

In this work, we present a novel integrated strategy utilizing sugarcane bagasse—an agro-industrial residue—directly as feedstock. A recyclable nitric acid/ethanol system simultaneously achieves pulping, delignification, fiber size reduction, and surface carboxylation, bypassing conventional pulp isolation steps. This method reduces chemical and energy inputs while producing CNFs with tailored dimensions (1–10 nm diameter) and enhanced suspension stability. By addressing key challenges related to feedstock processing and energy efficiency, this method presents a promising pathway for the economically viable and scalable manufacturing of CNFs.

2. Experimental section

2.1. Materials and methods

Sugarcane bagasse used in this study was kindly provided by Guangxi Boguan Environmental Products Co., Ltd, and was obtained from local plantations in Guangxi, China. Prior to use, sugarcane bagasse was washed several times with water and oven-dried at 60 °C.

2.2. Nitric acid/ethanol treatment of sugarcane bagasse

Sugarcane bagasse was treated with nitric acid/ethanol to remove lignin, liberate individual fibers, introduce surface carboxylation, and reduce fiber dimensions, accomplishing single-step integration of pulping and pretreatment processes. The system's effectiveness for CNF production was evaluated through fiber yield, chemical composition, fiber dimensions, and the characterization of the produced CNFs. The experimental process was as follows: sugarcane bagasse was immersed in a specific ratio of nitric acid/ethanol/water mixed solution (2:7.6:0.4 v/v) with a solid–liquid ratio of 1:30 and then heated at 60–100 °C with stirring at 500 rpm for 1–8 h. Post-treatment, fibers were washed repeatedly with deionized water and then sieved using a flat screen pulp machine (sieve size: 0.25 mm) to remove non-fibrous impurities. Fiber yield was calculated, and the resulting fibers were stored in a refrigerator at 4 °C to prevent mold growth and spoilage (Table 1).

---

2.3. Preparation of CNFs

Fibers (T100t5) from nitric acid/ethanol treatment were dispersed in water at a concentration of 1 wt% and mixed for 5 min at 1500 rpm using a disintegrator with a smooth blade. The suspension was then processed through a high-pressure micro-jet homogenizer (YL-JZJ-B, Shanghai Yinglu Auto Parts Co. Ltd., China) equipped with a diamond process nozzle (D5 type, 130 micron) at 20000 psi. Homogenization was performed for 5 and 10 cycles to produce CNFs. The obtained CNFs were stored at 4 °C until they were used for further characterization.

2.4. Characterization

3. Results and discussion

3.1. Behavior analysis of the nitric acid/ethanol process

Previous studies have demonstrated that lignin significantly enhances the physicochemical stability of lignocellulosic biomass by forming a densely crosslinked network with cellulose fibers and microfibrils. Targeted weakening of lignin–cellulose interactions facilitates microfibril delamination, thereby enabling substantial energy reduction during subsequent mechanical homogenization processes.¹² Consequently, efficient lignin removal or partial delignification has been identified as a critical step for effective nanofibrillation. In this work, sugarcane bagasse—an agro-industrial waste biomass—was treated with a nitric acid/ethanol system to simultaneously achieve defibrillation and delignification. The effects of reaction parameters (temperature and time) on the chemical composition and screened pulp yield (0.25 mm sieve) of sugarcane bagasse are shown in Fig. 1.

The results revealed that both the elevated temperature (60 → 100 °C) and the prolonged reaction time (1 → 5 h) significantly enhanced delignification efficiency (20.8% → 94.3%). This directly facilitated fiber liberation, as evidenced by the increasing screened pulp yield (1.6% → 22.2%) shown in Fig. 1c. Contour map analysis of composition and screened pulp yield changes (Fig. 1b and d) demonstrated limited efficiency when the reaction time was extended below 80 °C for both delignification and defibrillation, whereas temperatures exceeding 100 °C dramatically accelerated both the processes. Notably, the optimized reaction parameters within a narrow range (100 °C, 2–5 h) achieved effective removal of lignin and hemicellulose while minimizing cellulose degradation (screened pulp yield: 28.0% → 22.2%). This balance indicates that controlled thermochemical treatment can selectively disrupt lignocellulosic matrices without causing excessive cellulose degradation, thereby facilitating subsequent nanofibrillation processes.

The mechanical homogenization process primarily facilitates CNFs production by applying intense shear forces to disrupt hydrogen bonding between cellulose microfibrils, thereby inducing their delamination into nanoscale structures.¹³ Pretreatment strategies aimed at reducing fiber dimensions have been demonstrated as effective approaches to minimize energy consumption during subsequent homogenization. The effects of nitric acid/ethanol treatment at varying temperatures and times on fiber dimensions are illustrated in Fig. 2a–c and e. The prolonged reaction time coupled with elevated temperatures synergistically promoted fiber fibrillation, yielding a 68.0% reduction in fiber length (1.164 mm → 0.372 mm) and a 813.5% increase in fines content (5.2% → 47.5%) in T100t5 compared to T60t1. Compared to the commercial bleached sugarcane bagasse pulp (a common nanocellulose feedstock), sample T100t5 exhibited 59.0% shorter fiber lengths (0.372 mm vs. 0.907 mm) and 442.9% higher fines content (47.5% vs. 8.75%), highlighting its potential for energy-efficient CNFs production.

In addition, despite the similar chemical composition and screened pulp yield between samples T100t5 and T100t2 (Fig. 1a and c), the prolonged reaction time significantly reduced average fiber dimensions (from 0.596 mm to 0.372 mm, a 37.6% reduction) (Fig. 2e–f). These results demonstrate that nitric acid/ethanol treatment effectively achieves a controlled fiber size reduction without inducing excessive cellulose degradation. Furthermore, T100t5 maintained a 22.5% yield, indicating that nitric acid/ethanol treatment effectively balances between high yield and substantial dimensional reduction. This dual advantage mitigates energy demands during homogenization while obviating additional pretreatments, positioning nitric acid/ethanol treatment as a sustainable and efficient strategy for CNFs production.

To investigate the effects of nitric acid/ethanol treatment on the crystalline structure of cellulose fibers, XRD analysis was conducted on samples at varying temperatures and times. As illustrated in Fig. 3a, the crystallinity (Cr) of the treated cellulose fibers increased remarkably from 48.7% in the raw material to ∼75%, demonstrating effective removal of amorphous components. A strong positive correlation between Cr values and treatment intensity was observed, attributable to selective cleavage of lignin–carbohydrate complexes (LCC) via nitric acid-catalyzed depolymerization. Beyond specific severity thresholds (>100 °C for 1 h or >80 °C for 5 h), Cr values stabilized at ∼75% without further enhancement.

Complementary to the XRD analysis, FTIR spectroscopy was performed to investigate the chemical structural changes of cellulose fibers during nitric acid/ethanol treatment. As shown in Fig. 3b, characteristic peaks at 1730 cm⁻¹ (ester linkages in hemicellulose) and 1620–1649, 1510, and 1595 cm⁻¹ (aromatic rings in lignin) showed substantial intensity reductions.¹⁴ In contrast, the peak at 890 cm⁻¹ (β-glycosidic bonds in cellulose) exhibited limited attenuation, suggesting restricted oxidative hydrolysis in cellulose amorphous regions.¹⁵ Critically, FTIR analysis revealed no detectable nitro group signals at 1385 cm⁻¹ (symmetric stretching) or 1550 cm⁻¹ (asymmetric stretching),¹⁵ while elemental analysis results confirmed no nitrogen content increase (Table S1, ESI†), collectively excluding cellulose nitration. Moreover, T100t5 exhibited new peaks at 1637 cm⁻¹ (carboxylate, –COO⁻) and 1735 cm⁻¹ (carboxylic acid, –COOH), demonstrating selective oxidation of hydroxyl groups to carboxyl groups (Fig. S1, ESI†).¹⁵ These results confirm the selective degradation mechanism of the nitric acid/ethanol system, which preferentially cleaves lignin β-O-4 bonds and hemicellulose ester linkages while preserving the cellulose crystalline structure. Collectively, XRD and FTIR results demonstrate that the nitric acid/ethanol system achieves efficient delignification and hemicellulose removal while maintaining cellulose crystallinity, making it particularly suitable for biomass pretreatment requiring structural preservation of cellulose matrices.

3.2. Mechanism analysis of the nitric acid/ethanol reaction process

To elucidate chemical structural transformations and mechanistic insights during nitric acid/ethanol treatment, FTIR analysis and NMR spectroscopy were performed on nitric acid/ethanol treated dissolved lignin (NDL) and milled wood lignin (MWL) of sugarcane bagasse. As shown in Fig. 4a, distinct nitro group (–NO₂) absorption bands at 1385 cm⁻¹ (symmetric stretching) and 1550 cm⁻¹ (asymmetric stretching) appeared exclusively in NDL, confirming lignin-specific nitration. Moreover, aromatic C–C skeletal vibrations at 1600 cm⁻¹ and 1510 cm⁻¹ exhibited significant attenuation in NDL, indicating aromatic ring modification. The C–H/C–O bending region (1110–1160 cm⁻¹), attributed to overlapping signals from aromatic rings, ether linkages, and aliphatic groups, exhibited a pronounced signal reduction in NDL, suggesting cleavage of β-O-4 ether bonds and aliphatic side chains. Furthermore, a new carbonyl (CO) signal emerged at 1725 cm⁻¹ only in NDL, directly evidencing oxidative conversion of hydroxyl groups to carboxyl groups.¹⁶

¹H NMR analysis revealed molecular transformations in lignin following nitric acid/ethanol treatment (Fig. 4b). Methoxy group signals (3.4–4.0 ppm) persisted in both NDL and MWL. While MWL exhibited a broad band in the aromatic hydrogen region (6.1–7.9 ppm), indicative of hyperbranched lignin structures, the NDL displayed sharp, well-resolved peaks in this region, suggesting that lignin had been depolymerized into smaller oligomeric fragments with simplified aromatic configurations.¹⁵ These spectral changes, supported by the diminished spin–spin coupling effects in NDL, demonstrate lignin structural reorganization through selective bond cleavage during nitric acid/ethanol treatment.

To obtain detailed information about the structural evolution and depolymerization mechanisms of lignin, comparative 2D-HSQC NMR analyses of MWL and NDL were performed. The α-, β-, and γ-carbon correlations of lignin substructures were identified (Fig. 5a–e). The β-aryl ether (structure A), representing the most abundant β-O-4 linkage in lignin, underwent significant transformation. A marked increase in Iγ signals and a decrease in A signals were observed in the aliphatic region, directly evidencing β-O-4 cleavage and depolymerization. The signals of Aα and Aβ shifts, coupled with the emergence of A′′ signals, were mechanistically attributed to the oxidation of α-position hydroxyl groups to carbonyl groups (CO), demonstrating selective oxidative modification of lignin side chains during nitric acid/ethanol treatment.

MWL exhibited distinct guaiacyl (G) units in the aromatic region (Fig. 5b), whereas NDL showed a marked reduction in G-unit-derived signals (Fig. 5d). The disappearance of aromatic carbon signals indicates profound lignin structural reorganization, though persistent methoxy group signals confirm retained aromatic frameworks. This spectral loss likely stems from altered bonding environments of aromatic carbons. Introduction of nitro groups (–NO₂) via electrophilic substitution eliminates protonated carbon signals in HSQC spectra due to C–H bond conversion to C–N bonds. Elemental analysis corroborates this mechanism, showing elevated nitrogen content in NDL (4.91 wt%) compared to MWL (0.80 wt%), quantitatively validating nitro group incorporation during nitric acid/ethanol treatment (Table S1, ESI†).

To corroborate these findings and obtain complementary information on carbon atoms, which cannot be detected by HSQC analysis, supplementary ¹³C NMR spectroscopy was performed (Fig. S2, ESI†). The aromatic region of MWL exhibited characteristic signals in 110–120 ppm corresponding to C2, C5, and C6 positions in guaiacyl (G) units. In contrast, NDL displayed a complete signal loss, due to electrophilic nitration at the aromatic ring. MWL showed a broad peak in 144–151 ppm from overlapping signals from etherified and non-etherified C3/C4 in G units, while NDL exhibited a sharpened C3 signal at 148 ppm, consistent with lignin depolymerization-induced structural simplification.¹⁷ Moreover, a distinct peak at 161 ppm emerged in NDL, assigned to conjugated carboxylic acid (–COOH) groups from aromatic ring opening oxidation, which was previously proposed in lignin nitration studies.¹⁸ Collectively, these transformations evidence the dual nitro-oxidative transformation of lignin under nitric acid/ethanol treatment.

Furthermore, the signals originating from C–N groups present challenges due to the spectral overlap with other functional groups, complicating unambiguous identification. Nitro groups (–NO₂) attached to aliphatic chains are expected in the 55–75 ppm range, overlapping with signals from α-, β-, and γ-position carbons in lignin side chains. Comparative analysis of the MWL and NDL spectra revealed marked differences in the relative intensities of multiple signals within this region, evidencing lignin depolymerization and nitration.

Comprehensive characterization supports a reaction mechanism for lignin nitration, depolymerization, and oxidative transformation during nitric acid/ethanol treatment (Fig. 6). Nitro group (–NO₂) functionalization proceeds via three primary pathways designated as N₁ (aliphatic chain attachment), N₂ (aromatic ring substitution), and N₃ (side-chain modification) and combinatorial routes (e.g., N₁ + N₂ or N₂ + N₃). These nitration processes occur alongside lignin depolymerization, increasing aromatic and aliphatic hydroxyl group content. During depolymerization (D₁), newly formed phenolic hydroxyl groups undergo either nitric acid-mediated oxidation (O₁-1) or aromatic ring-opening reactions (O₁-2). Simultaneously, terminal hydroxyl groups in lignin side chains (D₂) are oxidized to carboxyl groups (O₂) or α-carbonyl (CO) groups (O₃).

3.3. Morphological analysis of the prepared CNFs

CNFs were successfully prepared from nitric acid/ethanol-treated cellulose fibers (T100t5, 1% concentration) using a high-pressure micro-jet homogenizer with 5 and 10 cycles. Notably, no nozzle clogging occurred—a common challenge in cellulose nanofibrillation—likely due to the reduced fiber dimensions and enhanced hydrophilicity from carboxyl group formation. AFM images (Fig. 7a, b, d, and e) revealed elongated nanofibrils with uniform diameters forming interconnected CNFs networks. Diameter distributions were calculated from AFM height profiles (Fig. 7c and f), while length distributions were statistically analyzed using NanoBrook Omni (Fig. S3, ESI†) due to the challenges in measuring elongated fibrils via AFM. The CNFs exhibited diameters of 1–10 nm and lengths ranging from 100 nm to 2 μm, consistent with conventional CNFs. Quantitative analysis demonstrated that 5-cycle homogenized produced CNFs with average diameter and length values of 4.6 nm and 1447.4 nm, respectively, while 10-cycle processing refined these values to 4.4 nm and 1063.5 nm, confirming cycle-dependent size reduction.

The partial oxidation of hydroxyl groups to carboxyl groups imparted a negative surface charge to CNFs, evidenced by a zeta potential of −21.8 ± 1.2 mV—significantly higher than purely mechanically processed CNFs (−8.3 ± 0.9 mV)—thereby enhancing suspension stability. Remarkably, CNF suspensions maintained exceptional stability over 6 months (Fig. S4, ESI†). Furthermore, the nitric acid/ethanol solvent was efficiently recycled via distillation with nitric acid replenishment (Fig. S5, ESI†), highlighting process sustainability.

4. Conclusions

In this study, a novel nitric acid/ethanol pretreatment method was successfully developed for the sustainable production of CNFs directly from sugarcane bagasse. Under optimal conditions (100 °C, 5 h), this process effectively removes lignin (94.3% delignification), preserves crystallinity (Cr ∼75%), reduces the fiber length while increasing the fines content (a 59.0% reduction in fiber length and a 442.9% increase in fines compared to commercial bleached sugarcane bagasse pulp fibers), and minimizes cellulose degradation (22.5% yield). Mechanistic studies revealed selective lignin–carbohydrate complex cleavage, aromatic nitration (4.91 wt% N), and hydroxyl-to-carboxyl oxidation. The process generates cellulose fibers with carboxyl groups, enhancing their electronegativity, which leads to the production of uniform CNFs (1–10 nm in diameter and 100 nm to 2 μm in length) with excellent suspension stability (>6 months, ζ = −21.8 mV). Importantly, the process integrates solvent recovery via simple distillation, minimizing waste and enabling solvent reuse, thus enhancing sustainability. By integrating pulping, delignification, functionalization, and fiber size reduction into a single-step, low-energy process, this work could promote the commercialization of CNFs and advance the utilization of agro-industrial waste.

Conflicts of interest

The authors declare no competing interests.

Data availability

The data that support the findings of this study are available from the corresponding author, Qijie Chen, upon reasonable request.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 22308030), the Hunan Provincial Natural Science Foundation of China (No. 2023JJ30009), the Hunan Provincial Education Department Foundation of China (No. 22A0208), and the Open Fund of Hunan Provincial Key Laboratory of Cytochemistry (2022xbhx02).

References

1. C. Liu, P. Luan, Q. Li, Z. Cheng, P. Xiang, D. Liu, Y. Hou, Y. Yang and H. Zhu, Adv. Mater., 2021, 33, 2001654 CrossRef CAS PubMed .
2. A. Vikulina, D. Voronin, R. Fakhrullin, V. Vinokurov and D. Volodkin, New J. Chem., 2020, 44, 5638–5655 RSC .
3. Y. Kong, H. Zhang, X. Xu, G. Tong and P. Li, New J. Chem., 2024, 48, 19974–19982 RSC .
4. L. Zhang, D. Du, W. Dong, T. Kaneko, H. Zhang, M. Chen and D. Shi, New J. Chem., 2024, 48, 6696–6703 RSC .
5. S. Antony Jose, N. Cowan, M. Davidson, G. Godina, I. Smith, J. Xin and P. L. Menezes, Nanomaterials, 2025, 15, 356 CrossRef CAS PubMed .
6. J. Li, F. Zhang, Y. Zhong, Y. Zhao, P. Gao, F. Tian, X. Zhang, R. Zhou and P. J. Cullen, Polymers, 2022, 14, 4025 CrossRef CAS PubMed .
7. P. Luan, X. Zhao, K. Copenhaver, S. Ozcan and H. Zhu, Adv. Fiber. Mater., 2022, 4, 736–757 CrossRef CAS .
8. I. Siró, D. Plackett, M. Hedenqvist, M. Ankerfors and T. Lindström, J. Appl. Polym. Sci., 2011, 119, 2652–2660 CrossRef .
9. J. Jose and T. Vinod, New J. Chem., 2025, 49, 9475–9483 RSC .
10. J. Gong, J. Li, J. Xu, Z. Xiang and L. Mo, RSC. Adv., 2017, 7, 33486–33493 Search PubMed .
11. R. Sun, J. Fang and J. Tomkinson, J. Wood Chem. Technol., 1999, 19, 335–356 CrossRef CAS .
12. H. Wang, C. Chen, L. Fang, S. Li, N. Chen, J. Pang and D. Li, Cellulose, 2018, 25, 7003–7015 CrossRef CAS .
13. S.-Y. Lee, S.-J. Chun, I.-A. Kang and J.-Y. Park, J. Ind. Eng. Chem., 2009, 15, 50–55 CrossRef CAS .
14. P. Luan, J. Liao, L. Chen, Y. Kuang, X. Zhang, Y. Zhang, Y. Zhu, Y. Dai, L. Mo and J. Li, Biomass. Convers. Bior., 2024, 14, 4055–4070 CrossRef CAS .
15. A. Karnitski, L. Natarajan, Y. J. Lee and S.-S. Kim, Int. J. Biol. Macromol., 2024, 281, 136408 CrossRef CAS PubMed .
16. U. Kamran and S.-J. Park, J. Colloid. Interface Sci., 2021, 594, 745–758 CrossRef CAS PubMed .
17. J.-L. Wen, S.-L. Sun, B.-L. Xue and R.-C. Sun, Materials, 2013, 6, 359–391 CrossRef PubMed .
18. S. Bylin, T. Wells, Q. Sun, A. Ragauskas and H. Theliander, BioResources, 2014, 9, 6002–6018 CrossRef .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5nj02114f

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2025