Optimization of Cellulose Recovery Using Deep Eutectic Solvent Fractionation: A Response Surface Method Approach

Abstract

Lignocellulosic biomass is a crucial renewable energy source for producing biofuels and valuable compounds, making it an attractive alternative to fossil resources. In this study, an environmentally friendly method was developed for cellulose fractionation from sugarcane bagasse using deep eutectic solvents (DESs), focusing on achieving high cellulose purity and specific physicochemical properties. The effects of different parameters were investigated by comparing four DESs: choline chloride–lactic acid (ChCl-LA), choline chloride–glycerol (ChCl-G), choline chloride–urea (ChCl-U), and choline chloride–polyalcohol (ChCl-P), under various reaction temperatures and times. The fractionation process was conducted under standard conditions at a temperature of 100 °C for 120 min with a 1:1 molar ratio. The results indicated that all DESs produced comparable cellulose recovery, ranging from 91.83% to 97.07%. A relatively high cellulose recovery was observed in the presence of ChCl-LA, at 95.47%. In addition, ChCl-LA demonstrated the highest efficiency in removing hemicellulose and lignin, at 95.36% and 93.38%, respectively, and high recovery yields of 70.45% for hemicellulose, and 70.66% for the lignin fraction. The fractionation conditions were further optimized using response surface methodology (RSM), achieving a ChCl-LA ratio of 1:2 v/v at 120 °C for 120 min. This resulted in impressive yields: 97.86% cellulose recovery, 96.50% hemicellulose removal, 74.40% hemicellulose recovery, 77.3% lignin recovery, and 71.5% lignin yield from sugarcane bagasse. These results closely match the predicted values, emphasizing the effectiveness of the process and its potential for economic application in lignocellulosic biorefinery operations.

1. Introduction

Fossil fuels, including coal, oil, and natural gas, remain the primary source of energy for many countries and industries worldwide, despite growing efforts to transition to cleaner and more sustainable forms of energy. These non-renewable resources, formed from the remains of ancient plants and animals millions of years ago, have significantly contributed to the advancement and well-being of human society by supporting higher living standards [1]. However, the reliance on fossil fuels has also led to severe environmental consequences, including urban air pollution, climate change, and water pollution, all of which pose significant risks to the biophysical environment, biodiversity, ecosystems, and human health [2,3].

Considering this challenge, the shift towards renewable energy sources has become increasingly urgent. Transitioning to renewable energy sources, such as solar, hydroelectric, geothermal, wind, and hydrogen, offers a sustainable alternative to fossil fuels. Additionally, biomass energy is particularly important as it not only reduces dependence on fossil fuels but also helps mitigate the negative impacts associated with fossil fuel consumption. By focusing on biomass energy, we can contribute to a more sustainable energy future while addressing the pressing environmental issues linked to fossil fuel use [4].

Lignocellulosic materials can help reduce environmental problems because lignocellulosic materials from agricultural residual waste have immense potential as replacements for fossil fuels and can be found in the natural environment [5]. Currently, lignocellulosic biomass is considered a zero-carbon feedstock that is renewable whilst also helping to alleviate global warming and waste problems. Thailand produces an enormous amount of agricultural by-products annually. The utilization of underused agricultural residues in biorefineries provides a promising approach for value addition to agricultural waste while solving the problem of their disposal. Lignocellulosic biomass refers to any organic matter derived from the agricultural or forestry sectors, e.g., corn as agricultural products, rice husk as agricultural waste, woodchips as industrial waste, and used paper as household waste. Biomass refers to renewable (sustainable) resources, which have compositions similar to fossil fuels (contain C and H). Furthermore, the products derived from biomass exhibit similarities to those derived from petroleum. Lignocellulosic biomass is a composite material made up of multiple structures [6]. The composition consists mainly of three polymers: cellulose, hemicelluloses, and lignin, which are interconnected with small amounts of acids, salts, and minerals. Consequently, there is a growing interest in the advancement of the sustainable biorefinery sector to transform underutilized lignocellulosic materials into a wide range of products, such as fuels, chemicals, and materials, through the application of interdisciplinary technologies [7].

The green solvent fractionation process is a promising fractionation technology for separating and recovering lignocellulose-derived components effectively. Solvents are often used in large quantities, especially when different variants of solvents are required for different processes [8]. The use of deep eutectic solvents (DESs) in biorefinery applications has gained significant interest. The key features of DESs are their biodegradability, the strong versatility of their physical–chemical properties, and their low vapor pressure, which is determined by the initial components [9]. Diverse applications of DESs have been seen in biomass processing, including the dissolving of biopolymers [10], extraction of phenolic compounds [11], and fractionation of lignocellulosic biomass for various purposes in biorefinement [12,13,14]. In a previous study, Thi and Lee (2019) [15] conducted research to examine the performance of DESs in the fractionating of oil palm empty fruit bunches. The study examined three types of deep eutectic solvents: choline chloride–lactic acid, choline chloride–urea, and choline chloride–glycerol. The efficiency of the pretreatment was assessed by the examination of cellulose digestibility and the subsequent analysis of structural and morphological alterations. The best yield of reducing sugar, amounting to 20.7%, was obtained using ChCl-LA at a molar ratio of 1:2. The high yield was attributed to the significant delignification ability of ChCl-LA and its efficacy in eliminating hemicellulose, which improved the digestibility of cellulose during enzymatic hydrolysis. The remarkable capacity of ChCl-LA to disrupt cellulose, hemicellulose, and lignin, thereby exposing the cellulose fraction to enzymatic hydrolysis, was further confirmed using FT-IR and SEM analyses. Lin et al. (2020) [16] conducted a study where they used various molar ratios of choline chloride–lactic acid to fractionate bamboo residues using DESs. The enzymatic digestibility of the processed bamboo residues was examined and correlated with the changes in their physical characteristics. The highest enzymatic hydrolysis yield was obtained at 76.9% when it underwent pretreatment with a mixture of choline chloride and lactic acid at a ratio of 1:4 at a temperature of 130 °C. In their study, Hong et al. (2020) [17] examined the extraction of lignin and hemicelluloses from non-wood biomass of luffa sponge. They used an acidic deep eutectic solvent made up of choline chloride and oxalic acid dihydrate. The result was a residue high in cellulose. Under the most favorable reaction circumstances (at a temperature of 90 °C for a duration of 150 min), solid portions with a cellulose content of 76.4% by weight (originally 51.8% by weight) and a residual lignin content of 10.7% by weight (initially 17.8% by weight) were obtained. Gan et al. (2024) [18] developed a sustainable biorefinery process using microwave-assisted deep eutectic solvents to efficiently separate and upgrade sugarcane bagasse. The pretreatment at 135 °C for 20 min retained 95.2% of cellulose, while effectively removing 84.1% of hemicellulose and 88.3% of lignin. For optimization, the relationships between key variables in the DES pretreatment of lignocellulosic biomass should be considered. Xu et al. (2020) [19] highlighted temperature and reaction severity as crucial factors for effective pretreatment, with DES properties related to hydrogen bonding being critical for achieving high delignification and glucan recovery rates. Hence, the selection of solvents employed in biomass processing will partially dictate the economic feasibility and ecological sustainability of a biorefinery [20]. In this research, the main focus is on the utilization of cellulose. Cellulose offers several advantages due to its biocompatibility, biodegradability, renewability, strong mechanical properties, environmental friendliness, and standing as one of the safest chemicals on earth. Cellulose is a biogenic material that is used in many ways to exploit sustainable energy sources and address ecological issues. Moreover, cellulose is widely applied for many applications, such as membranes, separators, food preservation, air filters, antibacterial products, and coatings [20].

The objective of this work is to develop a green solvent-based method for the fractionation of cellulose from sugarcane bagasse using DESs. The focus is on achieving high cellulose purity and optimizing its physio-chemical properties for use as a precursor in the production of bio-based chemicals or value-added applications. To achieve this, various characterization techniques, including FTIR, SEM, XRD, BET, and XRF, were employed to qualitatively and quantitatively assess the separated cellulose. Additionally, the fractionation of hemicellulose and lignin was analyzed as co-products, contributing to the overall economic feasibility of the biorefinery process. The responses on the fractions of hemicellulose and lignin were further determined. The high recovery of these co-products emphasizes the potential of this approach for optimizing biorefinery operations and advancing sustainable materials science.

2. Materials and Methods

2.1. Materials

The raw sugarcane bagasse was acquired from Ban So, Phayao, Thailand. The sample was subjected to a drying process in a hot air oven at a temperature of 70 °C for a duration of 24 h using a hot air oven. Subsequently, the dehydrated sugarcane bagasse was pulverized and filtered to achieve a particle size ranging from 0.5 to 1.0 mm. Size reduction was carried out by a pin mill (Swentech (Thailand) Ltd., Bangkok, Thailand).), equipped with sieve sizes 20–120 mesh and motor power of 4 kw/380 V for a maximum rotation speed of 4500 rpm. The ultimate moisture content of the milled sugarcane bagasse was 7%, determined by calculating the reduction in weight after subjecting the sample to drying in an oven at 105 °C until a consistent weight was attained after a duration of 5 h. The chemical composition of the biomass was determined using a standard NREL method [20]. The desiccated biomass was placed in airtight plastic bags and kept at ambient temperature for further investigation.

2.2. Cellulose Fractionation from Sugarcane Bagasse Using DES Fractionation

Reactions were performed in a 600 mL stainless steel high-pressure reactor equipped with a mixing system and a thermocouple for internal temperature measurement (Parr Reactor 4560, Parr Instrument, Moline, IL, USA). The standard reaction contained 10 g of sugarcane bagasse and 100 g of the DES. Four different types of HBD (carboxylic acid, polyols and alcohol, amides and amines, and phenolics) in ChCl-based DESs were used as starting solvents. The reaction was heated to the desired temperature (90–140 °C) for a specified reaction time (90–150 min) in a temperature-controlled jacket with stirring at 100 rpm. After treatment, the reaction mixture was centrifuged at 10,000 rpm for 10 min to separate the solid and liquid phases. Then, the solid residues were rinsed with an ethanol/water mixture. Afterwards, the filtrate containing water and ethanol was recovered and recycled by using a rotary evaporator at 40 °C and 60 °C for 2 h each. The neutral deionized water was added to the resultant dark brown liquid (lignin-rich fraction) to precipitate lignin. Subsequently, the lignin was collected by centrifugation, then washed with the water/ethanol mixture to eliminate impurities. Finally, the recovered lignin fractions were freeze-dried for 48 h and preserved in a desiccator. The responses were determined as follows in Equations (1)–(8). The flow chart of this research is shown in Figure 1.

$$\mathrm{P}\mathrm{u}\mathrm{l}\mathrm{p} \mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\mathrm{\%}\right)={\displaystyle \frac{Total weight of remaining solid pulp after reaction \left(\mathrm{mg}\right)}{Total weight of starting material \left(\mathrm{mg}\right)}}\times 100$$

(1)

$$\mathrm{C}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{u}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{e} \mathrm{r}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{y}\left(\mathrm{\%}\right)={\displaystyle \frac{(cellulose remaining in solid pulp (\mathrm{mg}\left)\right)}{cellulose content in raw material \left(\mathrm{mg}\right)}}\times 100$$

(2)

$$\mathrm{L}\mathrm{i}\mathrm{g}\mathrm{n}\mathrm{i}\mathrm{n} \mathrm{r}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l}\left(\mathrm{\%}\right)={\displaystyle \frac{(lignin content in raw material \left(mg\right)-lignin remaining in solid pulp (\mathrm{mg}\left)\right)}{lignin content in raw material \left(\mathrm{mg}\right)}}\times 100$$

(3)

$$\mathrm{L}\mathrm{i}\mathrm{g}\mathrm{n}\mathrm{i}\mathrm{n} \mathrm{r}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{y}\left(\mathrm{\%}\right)={\displaystyle \frac{(recovered lignin after precipitation in liquid fraction (\mathrm{mg}\left)\right)}{amount of lignin removal \left(\mathrm{mg}\right)}}\times 100$$

(4)

$$\mathrm{L}\mathrm{i}\mathrm{g}\mathrm{n}\mathrm{i}\mathrm{n} \mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\mathrm{\%}\right)={\displaystyle \frac{(recovered lignin after precipitation in liquid fraction (\mathrm{mg}\left)\right) }{lignin content in raw material \left(\mathrm{mg}\right)}}\times 100$$

(5)

$$\mathrm{H}\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{u}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{e} \mathrm{r}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l}\left(\mathrm{\%}\right)={\displaystyle \frac{\left(hemicellulose content in raw material-hemicellulose remaining in solid pulp \left(\mathrm{mg}\right)\right)}{hemicellulose content in raw material \left(\mathrm{mg}\right)}}\times 100$$

(6)

$$\mathrm{H}\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{u}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{e} \mathrm{r}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{y}\left(\mathrm{\%}\right)={\displaystyle \frac{\left(recovered hemicellulose as pentose sugars in liquid fraction \left(\mathrm{mg}\right)\right)}{Total pentose sugars based on hemicellulose removal \left(\mathrm{mg}\right)}}\times 100$$

(7)

$$\mathrm{H}\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{u}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{e} \mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\mathrm{\%}\right)={\displaystyle \frac{\left(recovered hemicellulose as pentose sugars in liquid fraction \left(\mathrm{mg}\right)\right)}{Total pentose sugars content in raw material \left(\mathrm{mg}\right)}}\times 100$$

(8)

According to the equations, pulp yield is defined as the weight of the separated solid compared with the initial weight of solid loading in the pretreatment process. Cellulose recovery is defined as the weight of cellulose in the remaining solid compared with the cellulose content in the raw material. Lignin removal is defined as the percentage of lignin content in the remaining solid compared with the lignin content in the raw material. Lignin recovery is determined based on the weight of lignin recovered in the liquid fraction compared with the weight of lignin removal after the pretreatment process. Lignin yield is determined based on the weight of lignin recovered in the liquid fraction compared with the lignin content in the raw material. Hemicellulose removal is defined as the percentage of hemicellulose content in the remaining solid compared with the lignin content in the raw material. Hemicellulose recovery is determined based on the weight of hemicellulose recovered in the liquid fraction compared with the weight of lignin removal after the pretreatment process. Hemicellulose yield is determined based on the weight of lignin recovered in the liquid fraction compared with the lignin content in the raw material.

2.3. Statistical Design and Optimization of Cellulose Fractionation Using RSM

The experimental design and statistical analysis were performed using the response surface design approach with the use of Design Expert software (version 10.0.1). The design included three independent variables: reaction temperature (X₁, 100–140 °C), reaction duration (X₂, 90–150 min), and choline chloride-to-lactic acid ratio (X₃, 2:1, 1:1, 1:2), each assessed at three levels (−1, 0, 1). Fifteen experiments were conducted, with three replications at the center point, all processed in a random sequence. The variation for each factor was divided into offset, linear, interaction, and quadratic components. The analysis of all parameters was conducted using a second-order polynomial as follows in Equation (9).

Y = β₀ + β₁X₁ + β₂X₂ + β₃X₃ + β₁₂X₁X₂
     + β₁₃X₁X₃ + β₂₃X₂X₃ + β₁₁X²₁ + β₂₂X²₂ + β₃₃X²₃

(9)

In this equation, Y represents the predicted response; X₁, X₂, and X₃ are the independent variables; β₀ is a constant; β₁, β₂, and β₃ are the linear coefficients; β₁₂, β₁₃, and β₂₃ are the interaction coefficients; and β₁₁, β₂₂, and β₃₃ are the quadratic coefficients. The resulting quadratic polynomial model was used to generate 3D surface plots, illustrating the relationships between the independent variables and the response.

2.4. Quantitative Analysis of Sugars and Their Breakdown Products in the Liquid Fraction

The soluble product profiles in the aqueous fraction were examined using a high-performance liquid chromatograph (LDC Model 4100, Shimadzu, Kyoto, Japan) that was fitted with a refractive index detector and a UV–visible spectroscopy (UV-Vis) detector and an Aminex HPX-87H column (Bio-Rad, Hercules, CA, USA) operating at 65 °C with 5 mM H₂SO₄ as the mobile phase at a flow rate of 0.5 mL/min. The quantity of oligosaccharides was assessed using the NREL technique [20].

2.5. Analysis of the Cellulose Component

The chemical compositions of the solid fractions were determined according to standard NREL method [20].

The Brunauer–Emmett–Teller (BET) method was employed to measure the total surface area of both the raw material and the fractionated sugarcane bagasse. Nitrogen adsorption/desorption isotherms were utilized to assess the surface area and pore volume of the samples. This analysis was performed using a surface area analyzer (TriStar II 3020, Micromeritics Co., Norcross, GA, USA).

The microstructure of the raw material and fractionated sugarcane bagasse was examined using scanning electron microscopy (SEM) with a JSM-6301F scanning electron microscope (JEOL, Tokyo, Japan). The specimens were desiccated and coated with gold for examination. The investigation was conducted using an electron beam with an energy of 5 kilovolts.

Each sample was prepared for XRF analysis as follows: Firstly, the sample was dried at 70 °C for removing the moisture. Then, the sample was cut with a Retsch ZM200 to an average size of 200 mesh. Then, 2.5 g of the sample was mixed with wax binder to form a pellet. Last, the pellet was pressed in a cylindrical press machine. The sample was analyzed by using an S4 Pioneer (Bruker AXS, Karlsruhe, Germany) instrument. This condition was set at 60 kV/50 mA, range 0.2–2 A, total resolution 3–100 eV, and typical measurement time 2–10 s per element.

Fourier transform infrared spectroscopy (FTIR) was used to analyze the chemical composition of the raw and fractionated sugarcane bagasse samples with a PerkinElmer instrument (Waltham, MA, USA). The samples were prepared using the KBr pellet method, and the spectra were recorded over a range of 4000 to 400 cm⁻¹, with a resolution of 4 cm⁻¹ and 32 scans. The peaks from the raw and fractionated materials were compared to standard functional group peaks.

X-ray diffraction (XRD) was used to assess the crystallinity of both the raw and fractionated sugarcane bagasse. The investigation was performed using an X’ Pert PRO diffractometer manufactured by PA-Nalytical in Almelo, The Netherlands. The specimens were subjected to a scanning process within a 2θ range of 10°–30°, with a constant increment of 0.02°. This scanning was carried out using a voltage of 500 kV and a current of 30 mA. The crystallinity index was determined by using the following equation:

$$\mathrm{C}\mathrm{r}\mathrm{I}={\displaystyle \frac{{\mathrm{I}}_{002}-{\mathrm{I}}_{\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{r}\mathrm{p}\mathrm{h}\mathrm{o}\mathrm{u}\mathrm{s}}}{{\mathrm{I}}_{002}}}\times 100$$

(10)

The variable I₀₀₂ reflects the intensity of the crystalline component (cellulose) of the biomass at a specific angle of 2ϴ = 22.4. Conversely, I_amorphous represents the intensity of the peak of the amorphous component (including cellulose, hemicellulose, and lignin) at a different angle of 2ϴ = 18.0.

3. Results and Discussion

3.1. Chemical Composition of Raw Sugarcane Bagasse

The chemical composition, with standard deviation, of raw sugarcane bagasse is shown in

Table 1. Chemical composition of raw sugarcane bagasse in this study.

Composition	Raw Sugarcane Bagasse
Cellulose	35.12% ± 0.43%
Hemicellulose	24.36% ± 0.91%
Acid-insoluble lignin	22.78% ± 0.57%
Acid-soluble lignin	5.03% ± 0.18%
Ash	5.82 ± 0.22%
Other	6.89 ± 0.36%

Note: Chemical composition is based on a dry basis.

. The raw sugarcane bagasse contained 35.12 ± 0.43% cellulose, 24.36 ± 0.91% hemicellulose, 22.78 ± 0.57% acid-insoluble lignin, 5.03 ± 0.18% acid-soluble lignin, 5.82 ± 0.22% ash, and 6.89 ± 0.36% other components on a dry weight basis. Sugarcane bagasse mostly consists of cellulose, which is a homopolymer made up of glucose. Hemicellulose, a substantial component, is a complex polymer consisting of both C₆ and C₅ sugars, such as xylose, arabinose, and galactose. The lignin fraction is a complex macromolecule composed of several phenolic chemicals. It was found that the percentage of cellulose, hemicellulose, and lignin was in the same range as in previous studies [21,22].

3.2. Effects of Different DESs on the Fraction of Cellulose

The reaction conditions were initially studied in batch processing to determine the optimal DES pretreatment for cellulose recovery. Four different DESs were evaluated: (ChCl-LA), (ChCl-G), (ChCl-U), and (ChCl-P). The experiments were conducted under standard conditions, with the reactions performed at 100 °C for 120 min and a molar ratio of 1:1. The results, as shown in Figure 2, illustrate the chemical composition of fractionated sugarcane bagasse after treatment with each DES. It is evident from the figure that the use of green solvents was effective in extracting hemicellulose and lignin fractions, while retaining a significant portion of the cellulose fraction, as indicated by the high recovery rates. The cellulose recovery across the different DESs ranged from 91.83% to 97.07%, demonstrating consistent performance in retaining cellulose. Among the solvents, ChCl-LA was particularly effective in achieving high cellulose recovery, as well as significant removal of hemicellulose and lignin. This is reflected in the compositional changes observed in the treated sugarcane bagasse. The use of ChCl-G, ChCl-U, and ChCl-P also yielded substantial cellulose recovery, though with slight variations in the removal efficiency of hemicellulose and lignin.

The co-products in the liquid phase, particularly those from the hemicellulose-rich and lignin-rich fractions, were analyzed following pretreatment with different DESs -vents. The product profiles in liquid fraction are illustrated in Figure 3. The results show the concentration of derived sugars and by-products in the liquid fraction after the pretreatment of sugarcane bagasse with four different DESs: ChCl-LA, ChCl-G, ChCl-U, and ChCl-P. These results corresponded with the responses of hemicellulose recovery and hemicellulose yield. The analysis revealed that the carbohydrate fraction, primarily composed of hemicellulose, was extensively degraded in the liquid phase, resulting in the formation of various sugars and by-products. The sugars included both polysaccharides and monosaccharides, while the by-products consisted of acetic acid, formic acid, levulinic acid, hydroxymethylfurfural (HMF), and furfural (FF). Among the DESs tested, ChCl-LA showed a notable performance, with a high recovery and yield of the hemicellulose fraction, at 73.26% and 70.45%, respectively. This indicates that ChCl-LA not only effectively fractionated hemicellulose but also facilitated the conversion of hemicellulose into soluble products. The other DESs, ChCl-G, ChCl-U, and ChCl-P also produced significant amounts of derived sugars and by-products, although with some variation in the concentrations of individual components. For ChCl-LA, the presence of higher concentrations of C5 sugars, along with lower concen-trations of by-products, highlights the effectiveness of these ChCl-LA solvents for further utilization into valuable co-products.

The results of lignin fractionation using different DESs are depicted in Figure 4, which shows the percentage of lignin recovery and yield following the pretreatment of sugarcane bagasse with four different DESs: ChCl-LA, ChCl-G, ChCl-U, and ChCl-P. The data reveal that the ChCl-LA solvent achieved the highest lignin recovery, with a value of 76.42%, indicating its superior ability to extract lignin from sugarcane bagasse during pretreatment. The other DESs also demonstrated effective lignin recovery, with ChCl-P, ChCl-U, and ChCl-G showing recovery percentages of 72.63%, 68.46%, and 66.29%, respectively. In addition to lignin recovery, the yield of lignin obtained through these pretreatment processes was evaluated. The lignin yield showed a consistent trend across the different DES solvents, with slight variations, depending on the solvent used. The analysis also suggests that the choice of DES has a notable impact on the lignin fraction. ChCl-LA, in particular, not only excelled in lignin recovery but also appeared to influence the fractionation process in a way that resulted in a more efficient breakdown of the sugarcane bagasse into its respective components [22].

3.3. Influence of Independent Factors on Cellulose Recovery, Hemicellulose Removal, Hemicellulose Recovery, Lignin Recovery, and Lignin Yield

The optimization of the deep eutectic solvent (DES) fractionation process, as summarized in

Table 2. Model summary statistics for the DES fractionation process under optimal conditions.

		Factors	Responses (%)
Run no.	T (°C)	Time (min)	ChCl-LA (% v/v)	Cellulose Recovery (%) a	Hemicellulose Removal (%)	Hemicellulose Recovery (%)	Lignin Recovery (%)	Lignin Yield (%)
1	120	120	49.99	97.35	96.30	74.3	77.31	71.28
2	140	120	66.66	90.75	90.76	70.10	68.27	64.75
3	120	150	66.66	88.05	91.37	70.71	69.90	65.09
4	100	150	49.99	91.49	90.62	72.33	75.69	70.12
5	120	90	33.33	90.64	91.86	70.40	75.11	70.88
6	100	120	33.33	94.81	91.78	72.19	72.74	69.10
7	100	120	66.66	90.39	90.54	72.10	73.60	67.31
8	120	120	49.99	97.42	96.51	74.41	77.33	71.03
9	120	120	49.99	97.86	95.35	74.11	77.26	71.54
10	140	90	49.99	89.53	90.29	70.12	76.21	70.50
11	120	150	33.33	90.89	93.50	71.46	72.58	68.14
12	120	90	66.66	87.3	91.70	70.65	73.34	67.75
13	140	120	33.33	90.0	92.32	70.59	73.59	68.98
14	140	150	49.99	88.18	91.45	68.90	67.90	63.01
15	100	90	49.99	89.42	90.37	70.21	73.11	68.00

^a Based on relative content of cellulose in remaining pulp.

, reveals critical insights into the effect of temperature and reaction time, and how the ChCl-LA ratio influences the recovery and yield of cellulose, hemicellulose, and lignin from pretreatment of sugarcane bagasse. The experimental design, employing response surface methodology (RSM), allowed for a systematic investigation of these factors across a range of conditions: temperatures of 90 °C to 140 °C, reaction times of 90 to 150 min, and ChCl-LA ratios of 2:1, 1:1, and 1:2. The results demonstrated significant variability in the responses, highlighting the importance of these parameters in optimizing the DES fractionation process (Table S1). The graphical representations in Figure 5A–E depict the response surfaces, highlighting the optimal regions for each response and validating the experimental design. It has been noted that the purpose of this study is to maximize the recovery of the cellulose fraction while effectively removing the hemicellulose and lignin fractions based on the composition in raw material.

The cellulose recovery ranged from 88.0% to 97.8%, with the highest recoveries generally observed at lower ChCl-LA ratios (2:1) and higher temperatures (120 °C to 140 °C). Interestingly, the cellulose recovery decreased significantly when the temperature was increased to 140 °C with a higher ChCl-LA ratio (1:2), particularly at longer reaction times (150 min). This decline can be attributed to the potential degradation of cellulose under more severe conditions, where excessive heat and an increased acidic condition from a higher LA concentration might lead to the breakdown of cellulose into smaller oligosaccharides or even monosaccharides, which are not recovered in the solid fraction. This observation is good agreement with previous studied [15].

The removal of hemicellulose was consistently high across all runs, ranging from 90.2% to 96.5%, indicating that the DES system effectively solubilized hemicellulose under various conditions. Hemicellulose recovery in the liquid fraction, however, varied more widely, from 68.9% to 74.4%. This suggests that while hemicellulose is efficiently removed from the solid phase, its complete recovery is challenging, particularly at higher temperatures and longer reaction times, which may lead to the further hydrolysis of hemicellulose into sugars, reducing its presence in the recovered solid.

Lignin recovery and yield also showed significant dependence on the process conditions. Lignin recovery ranged from 67.9% to 77.3%, with the highest recoveries generally observed at lower ChCl-LA ratios (2:1) and higher temperatures (120 °C to 140 °C). Similarly to cellulose, lignin yield decreased at higher temperatures and ChCl-LA ratios, particularly at 140 °C and a 1:2 ratio. This decrease in yield could be due to lignin depolymerization or dissolution into the liquid phase, making it less recoverable in the liquid fraction.

The optimization of deep eutectic solvent (DES) fractionation conditions was systematically investigated across a range of temperatures (90–140 °C), reaction times (90–150 min), and ChCl-LA ratios (2:1, 1:1, 1:2). All variable options were considered to maximize cellulose recovery, hemicellulose removal, hemicellulose recovery, lignin recovery, and lignin yield. The calculated regression equation for the optimization of DES fractionation conditions showed that cellulose recovery (Y1, %), hemicellulose removal (Y2, %), hemicellulose recovery (Y3, %), lignin recovery (Y4, %), and lignin yield (Y5, %) depend on temperature (X₁, °C), reaction time (X₂, min), and ChCl-LA ratio (X₃, % v/v). These equations illustrate the complex interplay between the independent variables, with the quadratic terms indicating non-linear effects on the responses. The models were validated using Design Expert software, which confirmed their statistical significance and predictive accuracy. The statistical analysis of the models, as indicated by the high R-squared values of 0.9934 for cellulose recovery, 0.9873 for hemicellulose removal, 0.9987 for hemicellulose recovery, 0.9999 for lignin recovery, and 0.9937 for lignin yield, demonstrates the robustness and reliability of the regression equations. The significance of the quadratic models (p < 0.005) further underscores the precision of the optimization process. The regression Equations (11)–(15) provided the following predictive models for each response:

Cellulose recovery (Y1, %) = −101.36 + 1.6167X₁ + 1.5194X₂ + 0.6006X₃ − 0.0014X₁X₂ − 0.0038X₁X₃ − 0.00025X₂X₃ − 0.0070X₁² − 0.0056X₂² − 0.0116X₃²

(11)

Hemicellulose removal (Y2, %) = −68.382 + 1.8150X₁ + 0.6297X₂ + 0.6980X₃ − 0.0003X₁X₂ − 0.0002X₁X₃ − 0.0009X₂X₃ − 0.0076X₁² − 0.0025X₂² − 0.0058X₃²

(12)

Hemicellulose recovery (Y3, %) = −54.158 + 1.1705X₁ + 0.7768X₂ + 0.5583X₃ − 0.0013X₁X₂ − 0.0003X₁X₃ − 0.0005X₂X₃ − 0.0043X₁² − 0.0024X₂² − 0.0047X₃²

(13)

Lignin recovery (Y4, %) = −141.2 + 2.146X₁ + 0.9713X₂ + 1.5777X₃ − 0.0045X₁X₂ − 0.0046X₁X₃ − 0.0004X₂X₃ − 0.0059X₁² − 0.0018X₂² − 0.0103X₃²

(14)

Lignin yield (Y5, %) = −90.369 + 1.668X₁ + 0.8264X₂ + 0.7877X₃ − 0.0040X₁X₂ − 0.0018X₁X₃ − 0.0004X₂X₃ − 0.0047X₁² − 0.0016X₂² − 0.0066X₃²

(15)

The predicted values of cellulose recovery, hemicellulose removal, hemicellulose recovery, lignin recovery, and lignin yield show the highest values of 97.8%, 96.5%, 74.4%, 77.3%, and 71.5% of the predicted values, respectively. Under the optimum conditions of DES fractionation at a temperature of 120 °C, a reaction time of 120 min, and a ChCl-LA ratio of 1:2% v/v, the maximum cellulose recovery of the DES fractionation process equal to 97.86% was predicted. In addition, the experimental results under optimal conditions meet all the following five criteria: >80% of cellulose recovery, >90% of hemicellulose removal, >70% of hemicellulose recovery, >70% of lignin recovery, and >60% of lignin yield under optimal conditions, which compared with the typical performance of the fractionation process. Afterward, the specifics of the discussion were considered.

The results obtained from the deep eutectic solvent (DES) fractionation process in this study demonstrate notable alignment with previous research findings, particularly those investigating the effectiveness of DES systems in lignocellulosic biomass pretreatment. For instance, Thi and Lee (2019) [15] reported similar trends where cellulose recovery diminished at elevated temperatures and higher acid concentrations due to increased depolymerization, consistent with the observed decrease in cellulose recovery at 140 °C and a 1:2 ChCl-LA ratio in the present study. Furthermore, the high efficiency of hemicellulose removal observed here, with values consistently above 90%, parallels findings by Ma et al. (2021) [23], who demonstrated that DES systems could achieve substantial hemicellulose solubilization, particularly under conditions of increased temperature and acidity. However, our study extends this understanding by highlighting the challenges in hemicellulose recovery from the liquid phase, which varied between 68.9% and 74.4%, possibly due to further hydrolysis into soluble sugars. This is a phenomenon similarly noted by Huo et al. (2023) [24] in their exploration of DES pretreatment efficacy. Lignin recovery in this study, ranging from 67.9% to 77.3%, is also consistent with the results reported by Francisco et al. (2018) [25], who found that DES systems, particularly those incorporating choline chloride with lactic acid, were effective in lignin solubilization and recovery. Consistent with a prior study, the presence of acid in DES pretreatment showed a comparable performance to that of acidified ChCl in terms of lignin and xylan removal [26]. The observed decrease in lignin yield at higher ChCl-LA ratios and temperatures can be attributed to lignin depolymerization and dissolution into the liquid phase, which aligns with the depolymerization patterns described by Hong et al. (2020) [17] when using DES pretreatment of poplar.

3.4. Characterization of Raw Materials and Fractionated Sugarcane Bagasse after DES Fractionation

The FT-IR spectroscopy can be used to describe the structure of raw sugarcane bagasse and fractionated sugarcane bagasse during ChCl-LA fractionation under optimal conditions.

Table 3. The FTIR spectrum for raw sugarcane bagasse and the fractionated sugarcane bagasse during ChCl-LA fractionation under optimal conditions.

Raw Sugarcane Bagasse	Fractionated Sugarcane Bagasse	Functional
3419	3384	O-H stretching
2937	2924	C–H stretching in methoxyl, methyl and methylene groups
1696	-	syringol
1602	1604	Aromatic ring vibrations and C=O stretching (S > G)
1513	1514	aromatic skeletal vibrations in lignin
1462	1462	CH2 bending
1423	1424	asymmetric stretching vibrations C=O
-	1265	C-O stretching vibration
1263	-	C-O stretching of the acetyl group
1126	1123	Aromatic in-plane C–H bending (typical for S units)
1031	1032	Holocellulose and lignin: C–O stretching
835	833	p-hydroxyphenyl units (H unit)

presents the FT-IR spectrum of raw sugarcane bagasse and after DES fractionation. The fingerprint region (3419 and 3384 cm⁻¹) of raw sugarcane bagasse and after DES fractionation corresponds to the hydroxyl group (O-H) of phenolic and aliphatic compounds [27]. The assigned peak related to the main functional group in the cellulose fraction. The intensity at 2937 and 2924 cm⁻¹ corresponded to the C–H stretching of methoxyl, methyl and methylene groups [28]. The band around 1697 cm⁻¹ is the syringol group of raw sugarcane bagasse. Similarly, the presence of two bands in the range of aromatic ring vibrations and C=O stretching (S > G) at 1602 and 1604 cm⁻¹ for raw sugarcane bagasse and fractionated sugarcane bagasse, respectively, indicates the structural characteristics associated with lignin [29]. The vibrations observed at the wavenumber range of 1513–1514 cm⁻¹ are associated with the aromatic skeletal vibrations present in lignin [30]. The absorbance observed at a wavenumber of 1462 cm⁻¹ is indicative of a bending motion in the CH₂ group [31]. The band seen at 1423–1424 cm⁻¹ corresponds to the asymmetric stretching vibrations of the C=O bond [32]. C-O stretching vibrations were centered 1265 cm⁻¹ [33]. Similarly, the band around 1263 cm⁻¹ is attributed to the C-O stretching of the acetyl group of raw sugarcane bagasse, corresponding to structures in hemicellulose and lignin [34]. This is consistent with the observation that the absorption band disappears following the DES pretreatment. The signal at 1126 and 1123 cm⁻¹ corresponds to aromatic in-plane C–H bending (typical for S units) [29]. The adsorption at 1031 and 1032 cm⁻¹ corresponds to the holocellulose and lignin: C–O stretching. In addition, the band at 835 and 833 cm⁻¹ of the raw sugarcane bagasse, and after the DES fractionation process, is attributed to the stretch vibration of p-hydroxyphenyl units (H unit) in lignin unit [35].

Scanning electron microscopy (SEM) offers a detailed analysis of the surface morphology of sugarcane bagasse, providing insights into the structural changes induced by the deep eutectic solvent (DES) fractionation process. The SEM images presented in Figure 6 highlight the distinct differences between raw sugarcane bagasse (Figure 6A) and sugarcane bagasse fractionated under optimal DES conditions (Figure 6B). In the untreated raw sugarcane bagasse, the SEM images showed a relatively smooth, compact, and flat surface, characteristic of the intact lignocellulosic structure. The fibers appear densely packed with limited porosity, indicating the presence of an intact lignin and hemicellulose matrix that effectively binds the cellulose fibers together. This dense structure is typically resistant to enzymatic hydrolysis and other chemical treatments, necessitating effective pretreatment methods to disrupt the complex matrix. Post-DES fractionation, the SEM images of the treated sugarcane bagasse show significant morphological changes. The surface of the cellulose fibers is notably rougher and more porous compared to the untreated sample. The increased roughness and porosity are indicative of substantial disruption to the lignocellulosic structure, resulting from the degradation and removal of lignin and hemicellulose. These changes suggest that the ChCl-LA treatment effectively breaks down the hemicellulose–lignin network, exposing more cellulose fibers and increasing the overall surface area available for subsequent enzymatic hydrolysis or chemical processing [36].

X-ray diffraction (XRD) analysis was employed to investigate the crystallinity index (CrI) of raw sugarcane bagasse and fractionated sugarcane bagasse during ChCl-LA fractionation under optimal conditions (Figure 7). The crystallinity index (CrI) of the raw sugarcane bagasse was measured at 62.43%, indicating a moderate degree of crystalline cellulose interwoven with amorphous regions consisting of hemicellulose and lignin. This level of crystallinity is typical for untreated lignocellulosic biomass, where the structural integrity is maintained by the presence of lignin and hemicellulose. Following the DES treatment under optimal conditions, the CrI of the sugarcane bagasse decreased to 58.78%. The diffraction patterns confirm these observations, with a noticeable broadening and reduction in the intensity of the characteristic cellulose peaks, particularly around 2θ values of 22°. This reduction in crystallinity suggests that the DES fractionation process not only removes amorphous hemicellulose and lignin but also disrupts the crystalline regions of cellulose [15].

Based on the BET results shown in

Table 4. BET surface area and crystallinity index of raw sugarcane bagasse and fractionated sugarcane bagasse during ChCl-LA fractionation under optimal conditions.

Biomass	BET Area (m2/g)
Raw material	3.53 ± 0.11
Fractionated sugarcane bagasse	7.82 ± 0.08

, the surface areas of the raw and frac-tionated bagasse were 3.53 m²/g and 7.82 m²/g, respectively. This indicates a significant increase in the surface area of the solids following the pretreatment process. The observed increase is consistent with previous findings, which reported that fractionation process leads to a substantial enhancement of the cellulose structure’s surface area [27].

X-ray fluorescence spectroscopy serves as a valuable tool for discerning the elemental composition of sugarcane bagasse samples. In this investigation, the chemical makeup of sugarcane bagasse samples pre- and post-DES fractionation under optimized conditions was scrutinized. XRF analysis, as detailed in

Table 5. Elemental composition of raw sugarcane bagasse and fractionated sugarcane bagasse during ChCl-LA fractionation under optimal conditions by XRF analysis.

Composition	Raw Sugarcane Bagasse (%)	Fractionated Sugarcane Bagasse (%)
Si	0.568 ± 0.01	0.45 ± 0.02
Fe	0.498 ± 0.01	0.17 ± 0.01
Ca	0.369 ± 0.02	0.16 ± 0.00
Na	0.133 ± 0.01	-
AI	0.077 ± 0.01	0.059 ± 0.01
Mn	0.042 ± 0.00	-
Zn	0.037 ± 0.01	0.032 ± 0.00
Cu	0.036 ± 0.01	0.036 ± 0.00
Ti	0.035 ± 0.00	-
Mg	0.030 ± 0.00	0.015 ± 0.00
K	0.006 ± 0.01	0.050 ± 0.00
Compton	0.03	0.02
Rayleigh	-	-
Sum (%)	2.08	1.29

, delineated the presence of several chemical components in the sugarcane bagasse samples prior to DES fractionation: calcium (Ca, 0.369 ± 1.2), sodium (Na, 0.133 ± 0.5), aluminum (Al, 0.077 ± 0.11), manganese (Mn, 0.042 ± 0.84), zinc (Zn, 0.037 ± 0.22), copper (Cu, 0.036 ± 0.67), titanium (Ti, 0.035 ± 0.68), magnesium (Mg, 0.030 ± 0.67), and potassium (K, 0.006 ± 0.66), collectively constituting up to 2.8% of the sample prior to DES fractionation. Remarkably, the cellulose recovered subsequently to the ChCl-LA fractionation process exhibited a noteworthy decrease in chemical composition, amounting to 1.29%. Moreover, observations indicated that certain metals post-DES fractionation, such as Na, Mn, and Ti, were not found. This absence suggests potential destruction, degradation, or leaching of these elements during the DES separation process, leading to a significant reduction in the concentrations of various chemical components [37].

4. Conclusions

This study successfully developed a green solvent-based method using ChCl-LA for efficient cellulose fractionation from sugarcane bagasse, achieving a high cellulose recovery of 97.86% under optimized conditions. The process also enabled significant recovery of hemicellulose and lignin, contributing to the economic feasibility of biorefinery operations. The structural changes in the isolated cellulose highlight its potential for diverse applications in sustainable across multiple industries, including bio-based materials, pharmaceuticals, food, and environmental solutions. While DESs like ChCl-LA are environmentally friendly, further research is needed to assess their long-term environmental impact and scalability. Future work should focus on the development of more cost-effective DESs, explore large-scale applications, and investigate the recyclability and lifecycle analysis of these solvents to ensure their sustainable use in industrial processes.