γ-Valerolactone Pulping as a Sustainable Route to Micro- and Nanofibrillated Cellulose from Sugarcane Bagasse

Abstract

The study explores γ-valerolactone (GVL) pulps as a sustainable approach to producing microfibrillated (MFC) and nanofibrillated (NFC) cellulose from sugarcane bagasse, a widely available agro-industrial by-product. Pulp was obtained by acid-catalyzed organosolv delignification with a GVL–water system. MFC was generated through a simple disc refiner, while NFC was produced by TEMPO-mediated oxidation followed by mechanical treatment in a colloidal mill. NFC and MFC produced using the same methodology from a commercial sugarcane totally chlorine-free (TCF) soda–anthraquinone (soda–AQ) pulp served as a reference. Structural and physicochemical characterization involved optical transmittance, turbidity, conductimetry, X-ray diffraction, viscosity, FTIR, carboxyl content, cationic demand, degree of polymerization, and morphology by scanning electron microscopy (SEM). Results demonstrated that xylan and residual lignin contents influenced MFC formation, and the NFC showed properties comparable to those of the commercial pulp with fewer fibrillation passes. The study highlights GVL pulping as a greener, efficient alternative to conventional processes, opening new pathways for producing viscosity-controlled nanocellulose suspensions suitable for advanced applications.

1. Introduction

The replacement of products derived from non-renewable fossil resources has increased significantly in recent years, and lignocellulosic biomass is a promising option for a transition to a circular economy. The by-products of agricultural and forest industries, such as wood sawdust and crop wastes, are becoming an excellent source due to their high availability, low cost, and renewability. Moreover, these second-generation resources do not compete with food or feed compared with sugarcane and corn [1]. These resources could be used in integrated biorefineries to produce bioproducts, biofuels, and bioenergy.

A green chemistry-based biorefinery roadmap involving catalysts to reduce pollution in secondary streams has led to the development of more sustainable and environmentally friendly industries [2]. One possible alternative is the use of γ-valerolactone/water systems (GVL systems) for achieving optimal biomass fractionation, resulting in the use of low amounts of acids and high rates of solvent recovery [3,4,5,6].

GVL pulping is a fractionation process based on green chemistry using a binary mixture containing γ-valerolactone (a green biodegradable solvent derived from sugars) [7]. The green GVL process offers several advantages; the temperatures, pressures, and acid concentrations applied during the treatment allow efficient removal of lignin and xylans, retaining the cellulose fraction [5]. Compared with other processes (e.g., kraft or soda/anthraquinone), the GVL process has lower environmental impact, reduced energy consumption, and milder chemical requirements. The organosolv pulping with GVL systems produces highly delignified pulps with similar brightness levels to those obtained after a bleaching sequence [5]. Furthermore, GVL can recover from the spent liquor and recycle [3]. Two-step lignin precipitation, distillation at reduced pressure, and liquid CO₂ extraction have proven to be an effective process for solvent recovery after eucalyptus chip treatment, achieving a recovery of more than 85% [4].

GVL systems have the potential to be incorporated in a biorefinery scheme to produce added-value products. For example, an organosolv/GVL system was used to elaborate textile fibers, polyhydroxyurethane, and xylose from birch wood in a biorefinery concept with solvent recovery [6]. However, scalability is still influenced by some challenges, like those during GVL production from biomass (stability and cost of catalysts for producing, bottleneck control in the levulinic acid hydrogenation stage, design of equipment to control exothermic reactions, corrosion, and separation of co-products) [8]. Despite this, pretreatment with GVL appears economically promising.

As described in our previous study [3], lignin dissolved during acid treatments can be precipitated by adding a low-boiling antisolvent and then centrifuging. Subsequently, GVL can be efficiently recovered from the liquor by distillation, with recovery yields exceeding 90 wt.%. The moderate operating conditions make the process suitable for implementation in existing facilities. The remaining sugar-rich residue could serve as a feedstock for furfural production via heterogeneous catalysis or for xylitol synthesis [3].

Lignocellulosic biomass valorization with GVL system fractionation has been studied previously to produce sulfur-free lignin or regenerated lignin [4,9], high-purity cellulose or cellulose derivatives [3,10], nanocellulose [5], and bioethanol [11,12]. NFC and MFC production from GVL pulps aims to optimize the process by introducing “recipe changes” that reduce the number of stages and employ more economical and technically viable chemical pretreatments. In one study, nanocellulose was obtained from aspen chips, and its properties were comparable to bleached kraft pulp from the same raw material [5]. These results demonstrate the possibility of simplifying the bleaching sequence. Another study involved the production of nanocellulose from GVL pulp produced with bamboo [13]. The authors demonstrated the capacity of the nanocellulose to absorb dyes; this could be interesting in waste treatment applications [13]. NFC obtained from GVL Eucalyptus globulus pulps with variable lignin content were evaluated by Lê et al. [14], demonstrating the strong influence of initial lignin content on the rheology and morphology of the final suspensions [14]. However, the effect of lignin content on fibrillation performance may differ between feedstocks. For example, sugarcane bagasse and wood have distinct structural and chemical characteristics.

Although numerous studies have evaluated biorefinery products from GVL system organosolv pulps, there is a lack of research on obtaining NFC/MFC, particularly from sugarcane bagasse pulps. Moreover, comparative studies of mechanical MFC versus TEMPO-mediated NFC from the same GVL pulp are lacking.

In a previous study exploring different variable combinations, various types of pulps were obtained. Based on a Drapen–Lin central composite design, each pulp showed potential for several applications, e.g., MFC or NFC suspensions, dissolving pulps, and substrates for hydrolysis in sugar production platforms, among others. A synergistic effect between GVL and acid was observed, even at low acid concentrations, resulting in high delignification. The acid content influenced the breaking of chemical bonds in lignin by hydrolysis and subsequent solubilization. Also, time and temperature significantly influenced the evaluated properties in both the solid and liquid fractions and utilizing the severity factor allowed the comparison of process conditions for specific applications [3]. Following these results, the present study evaluates the effect of acid-catalyzed GVL organosolv sugarcane bagasse pulps on the final properties of NFC and MFC. The MFC was obtained using a simple disc refiner, while the NFC was produced by TEMPO oxidation followed by mechanical fibrillation with a colloidal mill. The properties of the pulps, NFC, and MFC were compared with those of NFC and MFC produced using the same methodology from a commercial sugarcane totally chlorine-free (TCF) soda–anthraquinone (soda–AQ) pulp, which was used as the control. The results are expected to inform potential applications of NFC and MFC in papermaking, composites, and other high-value materials.

2. Materials and Methods

2.1. Materials

Sugarcane bagasse and ethanol were supplied by a local mill (San Javier Sugar Mill, San Javier, Argentina). Sugarcane bagasse was wet-depithed in a two-step process [15,16]. Commercial bleached sugarcane bagasse soda–AQ pulp obtained by TCF bleaching (control pulp) was kindly provided by an industrial paper mill (Ledesma, Libertador General San Martín, Argentina). The chemical composition for the control pulp was 1.89% lignin odp (on dry pulp base), 23.0% xylans odp, and 67.1% glucans odp. The intrinsic viscosity and ISO brightness were 823 mL/g and 85.4% ISO, respectively. All chemical reagents were reagent grade and purchased from Sigma-Aldrich (St. Louis, MO, USA).

2.2. GVL Process Conditions, Characterization, and Scaling-Up

GVL pulping was performed under conditions previously optimized by González et al. [3]. First, 80 g of air-dried sugarcane bagasse (wet-depithed using a Bauer simple disk refiner and a sieve) was delignified using a 10/1 w/v ratio of acid-catalyzed GVL/water as solvent (43 wt.% GVL in water, 0.005 M sulfuric acid) in a 1 L reactor at 155 °C for 90 min. After separating the black liquor from the pulp by vacuum distillation, the pulp was washed in three cycles with 50% v/v ethanol–water and two cycles with hot water (~80 °C), with vacuum distillation applied between each step to remove the solvents. Finally, the pulp was sieved by a 0.15 mm slot Somerville screen to remove shives.

Lignin and structural carbohydrate content in pulps were determined according to NREL-LAP 42618 report [17] using an AMINEX-HPX87H (BIO-RAD) column with refractive index detectors and a diode array [17]. The intrinsic viscosity of pulps, NFC, and MFC suspensions was measured according to ISO5351/1 standard [18]. ISO brightness and color were measured by a Color Touch Model ISO (Technidyne Corporation, New Albany, IN, USA) according to ISO 2470 standard [19]. All measurements were performed in duplicate.

2.3. NFC/MFC Production

Figure 1 shows the experimental setup for the production and characterization of MFC and NFC.

NFCs were obtained using the GVL and control pulps. TEMPO oxidations were performed according to reference [20]. Wet pulps were treated at 1% consistency with 1.6 wt.% odp of TEMPO (2,2,6,6-tetramethylpiperidine-1-oxyl), 10 wt.% odp of NaBr, and 20 mmol odp of NaClO, maintaining a pH of 10 with drops of 0.1 N NaOH. The time required from NaClO addition (t₀) to achieve a constant pH without NaOH addition was recorded.

The resulting suspension was washed with 4 L of water and vacuum filtered. Then, it was fibrillated by five passes in a colloidal mill at 1% consistency. Finally, the gel-like suspensions were stored at 4 °C. MFC from GVL and control pulps were obtained following the optimal conditions achieved in a previous study [21]. Briefly, a 1 wt.% pulp suspension was first disintegrated for 5 min. The pulp was then fibrillated in a simple disk refiner (0.0254 mm disk gap) until recirculation ceased due to increased viscosity. The resulting suspension was cooled to room temperature and stored at 4 °C in sealed plastic bottles.

2.4. NFC/MFC Characterization

The degree of polymerization (DP) for pulps and MFC was calculated using the Marx–Figini equation [22]:

$$\left[\mathrm{ɳ}\right]=2.28\times {\mathrm{D}\mathrm{P}}^{0.76}$$

(1)

For NFC, it was calculated using the Mark−Houwink−Sakurada equation [23]:

$$\left[\mathrm{ɳ}\right]=0.57\times \mathrm{D}\mathrm{P}$$

(2)

where ɳ is the intrinsic viscosity obtained according to ISO5351/1 [18], DP is the degree of polymerization, and 0.76, 0.57, and 2.28 are the constants when we consider unsubstituted cellulose in cupriethylenediamine and estimated DP.

The intrinsic viscosities for DP determination were obtained according to ISO5351/1 [18]. X-ray diffraction sample spectra were determined using a RIGAKU SmartLab SE X-ray (Rigaku Holdings Corp, Tokyo, Japan) diffractometer with a monochromatic source of CuKα1 radiation between 10 and 60° at a scan speed of 10° s⁻¹. The crystallinity index (CI) was calculated by the Segal method [24]. Fourier-transform infrared spectroscopy (FTIR) was carried out on 0.5 wt.% NFC and MFC air-dried films with a Thermo Mattson Nicolet (ThermoFisher, Waltham, MA, USA) FTIR spectrometer at 400–4000 cm⁻¹ and 4 cm⁻¹ resolution. Transmittance at 800 nm and turbidity for NFC and MFC were measured at 0.1 wt.% using a UV–Vis Shimadzu spectrophotometer (Shimadzu Corporation, Kyoto, Japan) and Hanna HI 98703 (HANNA instruments, Padua, Italy) turbidimeter, respectively. Dynamic viscosity in water was determined at 0.5 wt.% and 23 °C using a Brookfield LVDV-I+ (Brookfield Engineering, Middleboro, MA, USA) rotational viscometer, the carboxylic group rate by conductometric titration, and cationic demand at 0.04 wt.% by polyDADMAC titration [25]. All analyses were performed in duplicate.

Finally, the samples were freeze-dried for 24 h and sputter-coated using chrome. Images from MFC were obtained from films using a ZEISS GeminiSEM 460 (ZEISS, Jena, Germany) scanning electron microscope (UNaM SEM Laboratory) using an accelerating voltage of 5 kV and magnification of 4000×.

Statistical analysis was performed using Statgraphics Centurion XV, with significance set at p < 0.05.

3. Results and Discussion

3.1. Pulps Characterization

The GVL pulping yield for sugarcane bagasse in this larger-scale sample reached a value of 53.0%, like that previously obtained (53.5%) in a smaller-scale study [3]. The percentage of delignification varied significantly between scales (p < 0.05) (74.9% vs. 82.7% in this study and the laboratory scale, respectively). Figure 2 shows some of the measured pulp properties, as well as the basic chemical composition.

Using the same delignification conditions, ISO Brightness, intrinsic viscosity, xylan, glucan, and total lignin content showed statistically significant differences between scales (p < 0.05), probably due to scaling-up phenomena such as heat and mass transfer [26,27]. The exception was the crystallinity index. During pretreatment, three main types of amorphous components were removed: lignin, hemicelluloses, and disordered cellulose fragments (lignin, xylans, and glucans in Figure 2). Consequently, the overall mass balance revealed only minor differences in the residual crystalline fraction of the pulps. Although both volumes are lab-scale, the scale change is a fundamental step for moving forward to pilot-scale.

The GVL pulping showed a high degree of delignification, comparable to processes that consume large amounts of inorganic or organic acids [28,29], and achieved higher levels than those obtained with NaOH and phosphoric acid when applied to sugarcane bagasse [30]. Delignification was comparable to a sequence consisting of hydrothermal treatment, NaOH/AQ pulping, and oxygen delignification, yielding a pulp with 47.6% ISO brightness and an intrinsic viscosity of 850 mg/L [31]. Additionally, the ISO brightness achieved was comparable to that obtained using pulping methods with variable NaOH loads in conventional soda/AQ pulping (ISO brightness between 30 and 46.3%) [32]. GVL system treatment could avoid an additional delignification stage, reducing the contaminating load of bleaching sequences [33]. Furthermore, numerous strategies have been proposed for recovering pulping reagents and lignin to assimilate a kraft circuit [34]. The sugarcane bagasse pulping process is frequently performed in an alkaline medium using NaOH and anthraquinone soda, giving better responses than acid treatments. The suitability of alkaline treatment, combined with biopulping or catalyzed with anthraquinone, has been extensively demonstrated, achieving high degrees of delignification while preserving pulp viscosity [32,35,36].

3.2. Purely Mechanical Fibrillation of GVL Pulps for MFC Production

Using the same microfibrillation conditions, GVL MFC required more microfibrillation time than control MFC to achieve the same dynamic viscosity, suggesting that the difference in the xylan and lignin contents could affect the microfibrillation (

Table 1. Microfibrillation time and final rheological properties for MFC suspensions.

Sample	Microfibrillation Time (min)	Dynamic Viscosity (Pa.s) *
GVL MFC	75	22.1 ± 0.47
Control MFC	20	22.5 ± 0.88

* At a share rate of 0.50 s⁻¹.

).

The xylan content affects intrinsic fiber properties, such as swelling and fibrillation capacity [37]. Previous studies have reported that an increase in xylan content facilitates mechanical fibrillation and reduces energy consumption [38,39]. Xylan reduction hinders fiber bonding, resulting in weaker mechanical properties [40]. Similarly, significant removal of xylans in pine and eucalyptus pulp promoted higher fiber cleavage and less external fibrillation when pulps passed through the MASUKO grinder [41].

The effect of residual lignin content on micro- and nanofibrillation has been widely studied. However, the conclusions vary between different authors. Several studies on alkaline organosolv pulps from sugarcane bagasse [42,43] and kraft pulps [44] reported a plasticizing effect of lignin, which facilitates mechanical fibrillation and reduces energy consumption. Conversely, in other studies, a more cementing effect was observed, reducing fibrillation and requiring more energy to fibrillate, e.g., in spruce pulps obtained by SO₂ fractionation [45] or from neutral sulfite from Pinus sylvestris [46].

The MFC suspensions obtained by the disk refiner behave similarly to a non-Newtonian gel-like fluid (Figure 3), where an increase in shear speed decreases the viscosity of the suspension. This property is relevant to the application since MFC needs to be pumped or pipe-recirculated at the mill [47].

The MFC and NFC chemical structures are rich in OH groups, which generate numerous hydrogen bonds, producing gel-like suspensions that exhibit shear-thinning behavior [48]. This response has also been reported for other types of nanocellulose and for bacterial cellulose, where viscosity decreases with increasing shear rate. The magnitude of this change depends on concentration, temperature, and the specific nanocellulose morphology [47,49].

At rest, MFC forms a three-dimensional gel network that resists deformation. When the applied shear rate exceeds the yield stress, the network breaks and the suspension begins to flow; at higher shear rates, the individual fibrils align along the flow direction, further enhancing shear thinning [50]. In addition to this general trend, several authors have described distinct regions in the viscosity–shear-rate relationship for nanocellulose suspensions [51,52,53].

Schenker et al. [52] identified three zones: zone 1, a transition zone, and zone 2. They noted that these three regions are present in the behavior of all types of nanocellulose, although it depends on the nanocellulose type; CNC differs from mechanically derived MFC/NFC and from TEMPO-CNF. The authors also reported an increase in viscosity within the transition zone (6 s⁻¹, corresponding to 60 rpm), attributed to vortex formation [52]. Similarly, Iotti et al. [53] described three distinct behaviors when applying shear to MFC suspensions. In the first zone, a low shear rate aligns the fibrillar network, producing shear-thinning behavior. In the transition zone, the shear rate continues to increase, but the viscosity rises before dropping again once the network reorients [53]. Thus, these three zones reflect three rheological responses: initial pseudoplastic behavior, shear-thickening behavior in the transition region, and a return to pseudoplastic behavior [51].

In this study, a slight increase was also observed for MFCs at high shear rate. Previously, Karpinnen et al. [54] evaluated the viscosity of MFC suspensions at different concentrations (0.10–2% w/w) and reported that, at low concentrations, viscosity increased under high shear, a behavior they attributed to turbulence eddies [54]. Their work also incorporated image analysis to elucidate further turbulence effects. They found that floc size increased rapidly in the transition region and that size distribution broadened substantially. Visual inspection revealed gaps without fibers within the uniform network, resulting in a small plateau in the flow curve [54].

As already mentioned, the nanocellulose type strongly defines this behavior. This was evident when different commercially bleached conifer-pulp CNFs were evaluated: CNFs with a higher fibrillation degree and greater surface area formed more stable, intertwined networks capable of withstanding higher stresses, displaying a less linear response with a pronounced viscosity increase. In contrast, CNFs with lower fibrillation showed a more linear behavior without abrupt changes in the transition zone. Another study demonstrated the influence of lignin, which can yield structures with lower surface charge, promoting agglomeration and restricted mobility [14].

The factors that control viscosity include consistency, particle size, aspect ratio, and surface charge. Therefore, the size and aspect ratio difference could be appreciable when the samples obtained by mechanical treatment are compared at the same consistency. The apparent viscosities of both pulps were similar, suggesting that they attained comparable levels of fibrillation. The rheological properties, characterized by high viscosities at rest or under low shear rates and a significant decrease with increasing shear stress, determine its suitability as a stabilizer, emulsifier, and rheological modifier [55], making it suitable for use in paints, adhesives, cosmetics, and 3D printing. For example, a higher rest viscosity and a significant reduction in viscosity at low shear stresses would be ideal for cosmetic applications. On the contrary, low resting viscosity and minimal changes upon agitation may be advantageous for bulk paper applications, as lower viscosity facilitates pumping and recirculation within the mill [47]. Figure 4 shows MFC SEM images of the control GVL pulp.

The figure shows the higher fibrillation of samples obtained from GVL pulp. A large agglomeration of fibers with smaller sizes is observed, compared to the MFC obtained from bagasse pulp, where residual fibers are seen, along with less completely fibrillated material. The SEM images also show that MFCs tend to self-agglomerate, mainly due to the presence of OH groups on the fiber surface, which form numerous hydrogen bonds with water. When the sample dries, these water molecules evaporate, causing the union between the microfibril surfaces [56].

UV–Vis transmittance and turbidity were measured in 0.1% MFC suspensions as indicators of fibrillation level [57]. The UV–Vis transmittance at 600 nm was 34.8% and 37.2%, and the turbidity was 710 NTU and 651 NTU for GVL-MFC and the control, respectively. The transmittance values achieved were lower than those obtained for a commercial sugarcane bagasse pulp after 2 h of grinding in a MASUKO equipment [58]. However, the GVL-MFC transmittance value exceeded that of a previous study, in which bleached sugarcane pulp was fibrillated using a PFI mill refiner and high-pressure homogenizer [39].

3.3. Surface Oxidation of GVL Fibers for NFC Production

Table 2. GVL and control NFC suspension properties.

	GVL NFC	Control NFC
Carboxylic groups (µeq/g)	944 ± 1.80	1245 ± 6.10
Cationic demand (µeq/g)	2977	2981
Transmittance at 600 nm (%)	59.8 ± 0.60	46.9 ± 0.10
Turbidity (NTU)	49.4 ± 0.30	82.7 ± 0.10
Viscosity at 0.51 s−1 (Pa.s)	1701 ± 63.8	982 ± 149
Viscosity at 5.10 s−1 (Pa.s)	181 ± 0.67	106 ± 0.96

summarizes the characteristics of the obtained NFC. The carboxylic groups showed the highest value in NFC obtained from the control pulp. However, despite the higher carboxylic group value, fibrillation was observed to be lower. This higher level of carboxylic groups, therefore, could be due to the contribution of COO− groups from the hemicelluloses like arabinoxylans, rather than to the effective oxidation of OH in C6 of cellulose [59]. In general, the presence of xylans also hinders chemical accessibility through a barrier mechanism, reducing the reaction rate of the TEMPO-mediated oxidation [60]. Since xylans significantly influence the reaction, their reduction before the TEMPO oxidation has been recommended [61]. The increase in xylans decreases the reaction rate during the oxidation stage.

The dynamic viscosity (Figure 3) showed high values at rest and a significant viscosity reduction with increasing shear stress. This abrupt rheological change could be advantageous for 3D structure fabrication, providing dimensional stability at rest while allowing continuous flow during printing. However, it still needs the addition of cross-linking substances for subsequent stabilization after drying [62]. Compared to MFC, TEMPO-oxidized NFC shows higher viscosity due to surface charge, aspect ratio, and fibril uniformity [5]. The surface charge affects ionic strength, leading to stronger networks [63]. This effect was also observed when varying carboxymethylation and mechanical fibrillation times with the MASUKO mill were applied to kraft pulps; increased viscosities were observed at greater carboxymethylation, demonstrating the effect of surface charge [64].

3.4. DP, Crystallinity, and FTIR for MFC and NFC

TEMPO oxidation breaks the cellulose chain, as observed in the degree of polymerization of the samples obtained (Figure 5). The reduction was previously observed for loads of up to 10 mmol of NaClO per gram of dry pulp, where the DP was reduced to values below 400. The process involves not only oxidation of the C6 hydroxyl group but also chain cleavage at C1 [23]. As previously mentioned, xylans could decrease the rate of OH oxidation in C6 but not chain breakage, which could explain the lower DP in the control sample. It has previously been demonstrated that oxidation and chain scission depend mainly on the NaClO load and reaction time applied [65].

FTIR spectra of the fingerprint region of MFCs and CNFs are shown in Figure 6. The FTIR spectra are similar to those of the nanocellulose obtained from GVL pulp of aspen chips (Populus grandidentata) [5] and bamboo [13]. In NFC samples, stretching vibration at 1610 cm⁻¹ confirms sodium carboxylate groups at the C6 position of cellulose molecules (COO⁻Na⁺) oxidized during the TEMPO oxidation reaction [66]. This confirms oxidation in NFC samples; however, it does not occur in the MFC, where only mechanical treatment was performed. Also, at 1725 cm⁻¹ peaks were observed corresponding to condensation reactions in the holocellulose fraction that generate ester carbonyls (C=O) [67]. The peak was observed in both MFC and GVL NFC but not in the NFC control sample. Also, MFC showed the absorption peak at around 1635 cm⁻¹, associated with the bending vibration of absorbed water, due to the accessible hydroxyl groups [13].

These peaks were absent from the bleached sample and became undetectable after TEMPO oxidation [68]. The OH bending appears around 1370 cm⁻¹, and the skeletal COC pyranose ring vibrates at 1160 cm⁻¹ [69]. Finally, the β-glycosidic linkages and the OH out-of-plane bending are noticed at 896 cm⁻¹ and 665 cm⁻¹ [70]. The presence of peaks associated with β-(1,4) glycosidic linkage demonstrated that the treatment proceeded only on the cellulose surface, maintaining the crystal intact [5]. This can also be observed in the XRD patterns of the MFCs and NFCs compared with the pulps (Figure 7).

XRD patterns displayed typical cellulose I diffractograms. The pulps showed relevant peaks associated with crystalline and amorphous domains [71]. In pulps, three peaks with defined intensities were observed. The 002 crystalline plane (2θ ≈ 22–23°) corresponds to the primary axial stacking reflection of cellulose I chains, a second zone reflects (110)/(1–10) around 14–16° that corresponds to the lateral arrangement of the chains in cellulose I, and a third zone around 35° is associated with the 004 diffraction plane that shows a certain crystalline periodicity along the longitudinal axis [72,73,74]. In both MFC and NFC, decreases were observed at all peaks, suggesting changes in crystallinity arrangements. The peak associated with the 004–reflection plane disappeared in NFC and MFC, indicating a loss of periodicity in the diffracted 002 planes. A previous study observed a similar loss of crystallinity when a sequence of mechanical cycles was applied [75].

Crystallinity indexes for GVL and control decreased for NFC (62.9% and 65.5%) and MFC (61.2% and 58.2%), compared to the pulps (75.6% and 77.6%). The CI values were similar to those reached when sugarcane bagasse was treated with a sequence of NaCO₂ and KOH, followed by mechanical fibrillation in a high-pressure homogenizer [76]. Also, the values are in the range that TEMPO-NFC obtained from bleached kraft (D-EP-D bleaching) and GVL (GVL/water 7/3, 0.1 M sulfuric acid, 125 °C, and 3 h) pulps obtained from aspen (Populus grandidentata) wood chips [5].

The properties evaluated in both NFC and MFC allow us to estimate their potential behavior in applications. The NFC samples exhibited low DP values, higher viscosities, and numerous oxidized functional groups. The combination of TEMPO oxidation and mechanical fibrillation allows for smaller sizes and a more defined size distribution [18,37]. Furthermore, chemical action allows for greater removal of amorphous components. Therefore, it is expected to exhibit higher tensile strength but lower flexibility than MFC. However, the agglomeration effect is greater in NFC, so an adequate dispersion is paramount to ensure these mechanical properties. The mechanical properties of MFC and the agglomeration effect of NFC have been previously observed with other types of nanocellulose when used as a paper reinforcing agent [77,78], in composite materials, and even in biobased films [79].

4. Conclusions

GVL pulping demonstrates strong potential as an environmentally friendly alternative to conventional soda or kraft processes, achieving high delignification with solvents that can be efficiently recovered and recycled. Although residual lignin remains in the pulp, the values are comparable to those obtained with multi-stage alkaline–oxygen sequences, suggesting that GVL treatment could simplify bleaching and reduce chemical consumption. The fibrillation performance of GVL pulps is particularly promising: disc refining yielded micrometric fibrils with minimal energy input, while TEMPO oxidation combined with colloidal milling produced nanofibers with diameters in the nanometric range. These NFC suspensions exhibited distinctive rheological properties that make them attractive for high-value applications requiring viscosity control, including 3D printing, cosmetics, and functional coatings. Overall, this work demonstrates not only the technical feasibility of sugarcane bagasse GVL pulping for nanocellulose production but also its capacity to integrate sustainability, process efficiency, and product performance in line with the emerging bioeconomy demands.