Fe or Ni Catalytic Hydrothermal Depolymerization with Ethanol for Efficient Anaerobic Digestion of Corn Stover

Abstract

This study investigated the enhancement of anaerobic digestion (AD) performance of corn stover (CS) through Fe/Ni catalytic hydrothermal depolymerization with ethanol. The CS depolymerization process was conducted using Fe/C, Ni/C, Fe/CNT and Ni/CNT catalysts in combination with ethanol or water/ethanol solvents. The results revealed that the depolymerization with catalyst-ethanol (DC-E) effectively disrupted the physical encapsulation of cellulose by lignin. It also showed that the Ni/CNT catalyst in ethanol significantly promoted β-O-4 bond cleavage in lignin, achieving a lignin conversion rate of 48.5% and 2.7 g/L total phenol concentration (TPC). The water/ethanol (9:1) system effectively degraded hemicellulose (53.6% conversion) while retaining over 90% cellulose for AD. Structural analysis revealed that DC disrupted cellulose hydrogen bonds, reducing crystallinity index (CrI decreased from 38.4% to 32.6%) and increasing cellulose accessibility to 909 mg/g (2.6 times higher than untreated CS). The efficient depolymerization of CS obviously improved the biodegradability of cellulose and hemicellulose, contributing to the increase of biomethane production. Biomethane yield (BY) of E-Ni/CNT was 18.1% and 27.6% higher than that of E-HP and the control group, respectively. These findings indicated that ethanol-promoted catalytic depolymerization of CS can enhance the performance of AD.

1. Introduction

The transition from fossil fuels to renewable energy is crucial for addressing environmental and energy challenges [1]. Corn stover (CS), a major lignocellulosic biomass, exhibits great potential for biomethane production via anaerobic digestion (AD) due to its global abundance and renewability [2]. However, lignin binds cellulose and hemicellulose through hydrogen and covalent bonds (e.g., β-O-4), forming a stable structure that limits microbial access [3]. Effective depolymerization is essential to break this structure barrier and enhance cellulose accessibility.

A number of depolymerization methods have been established including chemical, physical, biological, physicochemical and combinations of these methods to reduce the recalcitrant structure of lignocellulose. Hydrothermal depolymerization (HP) operates under water-mediated autohydrolysis at 160–240 °C, offering advantages such as corrosion resistance and solvent-free operation [4]. However, HP is always carried out at high temperatures and pressures while dissolving large amounts of hemicellulose, which could result in excessive costs and loss of the carbon source needed in AD. The addition of catalysts to the HP process is an emerging method for depolymerization of lignocellulosic biomass to achieve depolymerization conversion of lignin to improve cellulose digestibility [5].

Metal-loaded catalysts have shown promising performance in biomass depolymerization through the selective generation of specific phenolic monomers. Ni-based catalysts can efficiently promote lignin depolymerization, especially for the selective cleavage of β-O-4 bonds without the need for an external source of hydrogen from solvents or gases [6]. Klein et al. [7] demonstrated that 10 wt% Ni/C catalyst at 200 °C for 6 h converts birch lignin into dihydroeugenol (10%) and 2,6-dimethoxy-4-propylphenol (18%). Li et al. [8] reacted lignin obtained by enzymatic degradation of corn kernels in isopropyl alcohol over Ni/C catalyst at 260 °C for 5 h. The results showed a significant increase in the percentage of lignin degradation to 80.1%. Similarly, the yield of phenolic monomers increased from 5.27% to 9.63%. Fe/C catalysts containing iron-based active components have a dual function: to promote structural modification of biomass and to guide sugar conversion. Cheng et al. [9] found that FeCl₃ promotes the solubilization of low molecular weight lignin and the breaking of lignin–carbohydrate bonds of 2-O-Ac-β-D-xylopyranosides in Moso bamboo.

At the same time, the depolymerization efficiency and product distribution depend strongly on the solvent used. In lignocellulose depolymerization, ethanol is both a good solvent for lignin fragments and a hydrogen donor solvent. Huang et al. [10] used CuMgAlO to depolymerize soda lignin in ethanol solvent, which resulted in a high monomer yield (23 wt%) without coke formation. Ethanol can also hinder the recombination of lignin thus improving the accessibility of cellulose and hemicellulose [5]. Lu et al. [11] promoted the partial degradation of lignocellulose by FeCl₃-catalyzed ethanol treatment, and nanofibrils yield increased to 67%.

Current research predominantly focused on characterizing depolymerized monomers, while over depolymerization induces premature conversion of cellulose and hemicellulose into C5/C6 monosaccharides, leading to undesirable carbon loss of AD process. Conversely, insufficient depolymerization severely restricts cellulose biodegradability, thereby compromising subsequent bioconversion processes. This study developed an ethanol-enhanced catalytic depolymerization strategy using Fe/Ni-loaded carbon-based materials to improve cellulose accessibility in CS.

The objectives of this study were to (1) investigate the effects of ethanol-promoted hydrothermal depolymerization with Fe/C, Ni/C, Fe/CNT and Ni/CNT as catalysts on the chemical composition and structural properties of the solid fraction of CS; (2) analyze the composition and changes of the liquid fraction after CS depolymerization and analyze possible DC conversion pathways; and (3) investigate the effects of DC-E and DC-0.1E depolymerization of CS on biogas yield and substance conversion.

2. Results and Discussion

2.1. Solid Fraction Transformation Analysis

2.1.1. Chemical Composition Changes

Depolymerization could effectively promote the dissolution of hemicellulose, the conversion of lignin and the degradation of components such as pectin, protein and extracts on the surface of CS [12]. The hemicellulose and lignin conversion rate and cellulose retention rate after DC was shown in Figure 1. The lignin conversion rate of the DC-E group (18.9–48.5%) was significantly higher than that of the depolymerization with catalysts-10% ethanol (DC-0.1E) group (9.9–11.2%), which may be related to the fact that ethanol, under the action of the catalysts during the DC process, also act as a hydrogen donor and solvent and participates in the reductive cleavage of the C-O and C-C bonds, which promoted the lignin reductive cleavage of complex molecular structures [13]; whereas, the hemicellulose conversion in the DC-0.1E group (41.2–53.6%) was significantly higher than that in the DC-E group (10.8–17.8%). Renders et al. [14] found similar results due to the fact that the ethanol/water mixtures can depolymerize the hemicellulose fraction better than pure alcohols/pure water. In addition, cellulose was well retained (more than 90%) in all test groups, preserving more digestible substrate for subsequent AD process, which would facilitate biomethane conversion and production [1].

2.1.2. Structural Modification

FTIR

FTIR revealed changes in chemical bonding and functional groups during DC process. While the overall spectral profiles remained comparable before and after DC (Figure 2a), characteristic peak attenuation indicated structural alterations. The hydroxyl stretching vibration at 3414 cm⁻¹ showed marked intensity reduction in DC samples, suggesting partial disruption of hydrogen bonding networks in cellulose macromolecules [15], consistent with observed improvements in accessibility and CrI. The diminished absorption at 2921 cm⁻¹ (C-H/C-H₂ stretching) reflected carbohydrate decomposition following depolymerization [16,17]. A critical observation emerged at 1723 cm⁻¹, where the C=O stretching vibration (indicative of ester linkages between lignin and hemicellulose) became undetectable in E-Ni/CNT and Ni/CNT samples [18], confirming cellulose exposure through lignocellulosic bond cleavage. This was related to the fact that the more abundant active sites of Ni/CNT can promote hydrodeoxygenation and cracking reactions, in agreement with the results of Zhou et al. [19]. This aligned with the high conversion rates of lignin and hemicellulose. Spectral deconvolution further demonstrated structural decomposition of hemicellulose, evidenced by reduced intensities at 1514 cm⁻¹ (C-H deformation), 1428 cm⁻¹ (C-H bending), and critical polysaccharide signatures: 1054 cm⁻¹ (C-O-C stretching) and 898 cm⁻¹ (glycosidic linkage vibration) [20], correlating with increased soluble sugars and VFAs production. In the E-Ni/CNT sample, the 1245 cm⁻¹ peak (guaiacyl C-O stretching) almost disappeared, which was consistent with the 48.5% lignin conversion. That was related to the fact that ethanol, through its α-OH group, act as a hydrogen donor and synergistically interacts with Ni/CNT to promote the cleavage of β-O-4 through the proton-coupled electron transfer mechanism [21]. The achieved lignin conversion efficiency exceeded that of the Ni-Ru/Al₂ O₃ catalyst (41.5% lignin conversion at 220 °C) prepared by Wang et al. [6]. These observations suggest that DC has the potential to disrupt the natural anti-degradation structure within CS, thus allowing cellulose exposure to be fully converted and utilized in subsequent AD.

XRD

The hydrogen-bonding network within cellulose conferred CS with stable crystalline structures that resist anaerobic microbial conversion. Crystallinity index (CrI), a critical parameter influencing enzymatic hydrolysis efficiency [22], quantitatively reflected cellulose crystallinity alterations. Ni/CNT catalyst addition induced significant CrI reduction from 38.4% to 32.6% (

Table 1. The CrI of CS before and after depolymerization.

	E-Ni/CNT	E-HP	Ni/CNT	HP	Control
CrI	36.3	34.8	32.6	36.0	38.4

), attributable to hydrogen bond cleavage that degraded both crystalline and amorphous cellulose domains [23]. This crystallinity suppression created preferential enzymatic access to biodegradable amorphous regions, thereby enhancing biomethane production. In contrast, the crystallinity of E-Ni/CNT was higher than that of E-HP, due to the fact that ethanol-mediated Ni-CNT catalysis preferentially destroys the amorphous regions, thus creating an optimally structured substrate for anaerobic conversion [24]. This phenomenon underscored the ability of the catalyst to modify the cellulose supramolecular structure through targeted crystalline domain deconstruction [21].

Surface Lignin Area and Hydrophobicity

While the majority of lignin remained in the solid phase during depolymerization, its structural modifications induced ultrastructural deformations that altered surface properties [25]. As illustrated in Figure 2b, these changes enhanced lignin physical blocking capacity and non-productive enzyme adsorption, ultimately impeding biomethane conversion during AD [26]. The DC-0.1E group exhibited drastic lignin surface area reduction (172–216 L/g), representing 94.1–144.0% decrease compared to the control (418 L/g). The DC-E group showed minimal surface area variation (313–318 L/g, 31.4–33.8% reduction) despite higher lignin conversion, potentially attributable to lignin migration and subsequent spherical particle formation on fiber surfaces. Hydrophobicity analysis revealed critical structural alterations, with contact angle measurements decreasing from 1.5 L/g to 0.33–0.6 L/g post-depolymerization. This hydrophilicity enhancement likely stemmed from Ni/CNT-catalyzed exposure of polar functional groups (e.g., phenolic hydroxyls), leading to a decrease in hydrophobicity, which could enhance the instability of lignocellulosic structure and create accessible substrates for enzymes [27].

2.1.3. Accessibility Improvement

The DC process not only modified lignocellulosic components but also deconstructed chemical bonding networks within biomass substrates. This structural modification enhanced microbial accessibility while preserving the skeletal matrix framework [4]. Particularly in AD of CS, cellulose accessibility (Figure 3a) has been identified as the critical determinant governing enzymatic degradation efficiency [28]. The results demonstrated that DC-E significantly improved cellulose accessibility, with depolymerization groups (481–921 mg/g) showing 1.4–2.7 times higher cellulose accessibility than that of the control group (346 mg/g). The cellulose accessibility of the Ni-based catalyst groups was higher than that of the Fe-based one, which was attributed to the fact that Ni catalysts have Lewis acidic sites, which can effectively reduce the activation energy of C-O bonds. The synergistic effect between ethanol solvent and Ni catalysts substantially enhanced the breakdown of ether and hydrogen bonds among lignocellulosic constituents [29], thereby increasing cellulose exposure. The E-Ni/CNT group achieved exceptional accessibility of 909 mg/g, which was 1.3 times higher than that of the Ni/CNT group (724 mg/g), correlating with its superior lignin conversion efficiency. The cellulose accessibility was significantly higher (707.8 mg/g) than that of HP rape straw at 220 °C for 80 min by Zhu et al. [27].

SEM characterization (Figure 3b–f) corroborated these findings. The untreated CS exhibited intact ultrastructure with smooth surfaces and ordered fiber bundle arrangements (Figure 3d). HP induced structural disintegration, manifesting as surface cracks and fragmented particles (Figure 3c,f). Ni/CNT catalyst implementation caused more pronounced structural damage with extensive surface erosion (Figure 3b,e), creating additional substrate–enzyme interaction sites. This enhanced cellulose–cellulase interfacial contact area ultimately contributes to improved biogas production.

2.2. Liquid Fraction Characteristics

2.2.1. RS and pH

Depolymerization selectively hydrolyzes glycosidic bonds in hemicellulose and cellulose xylan chains, liberating RS (primarily oligosaccharides and monosaccharides) that facilitate rapid microbial utilization during early AD [2]. As depicted in Figure 4a, the DC-0.1E system achieved substantially higher RS concentrations than DC-E groups. DC-E system showed limited sugar accumulation (0.7–0.9 g/L), possibly indicating ethanol-enhanced sugar conversion pathways [5] requiring further mechanistic investigation. The RS concentration in the Fe/C group reached 10.5 g/L, which is 1.1 times and 7.5 times that in the HP group (9.7 g/L) and the control group (1.4 g/L), respectively. The elevated RS availability directly contributed to the earlier onset of peak biogas production in DC-0.1E groups, consistent with metabolic prioritization mechanisms reported by Xu [30] and Li et al. [31]. Furthermore, Figure 4b revealed a strong linear inverse correlation (R² = 0.9921) between RS concentration and pH in DC-0.1E systems, corroborating the acid-catalyzed hydrolysis dynamics described by Zhu et al. [27].

2.2.2. VFAs

The depolymerization process facilitated hemicellulose deacetylation, converting acetyl groups into VFAs whose accumulation primarily drove pH depression, albeit with delayed pH response relative to VFAs generation kinetics [32]. As evidenced in Figure 4c, the DC-E system demonstrated substantially elevated VFAs concentrations (13.8–16.9 g/L), representing 2.9–5.3 times and 29.8–36.5 times higher compared to DC-0.1E (3.2–4.8 g/L) and control groups (0.5 g/L), respectively. This aligned with Section 3.1 analyses, suggesting that the solvent of ethanol effectively promoted the conversion of hemicellulose and the conversion of sugar to acid [33]. In addition, Fe/Ni-loaded carbon catalysts induced distinct product selectivity: isovaleric acid dominated VFAs profiles in DC-E groups (90.6–92.3%), contrasting sharply with valeric acid predominance in E-HP systems (95.7%). This metabolic divergence suggested the catalysts modulate reaction pathways through intermediate stabilization, while ethanol solvent exerted selective solvation effects favoring isovalerate biosynthesis. The DC-0.1E system exhibited acetic acid dominance (77.7–87.0%), consistent with canonical AD sequences where valerate undergoes acetolactic conversion prior to methanogenesis—mechanistically explaining its earlier biogas production peak. The pH of the DC-0.1E group (3.93–4.17) was higher than that of the HP group (3.73), which indicated that the addition of catalyst could effectively stabilize the reaction intermediates and inhibit the occurrence of side reactions [34].

2.2.3. TPC

Lignin, primarily composed of alkylphenolic heteropolymers derived from phenylpropane units interconnected via C-C and C-O bonds, exhibited depolymerization extent quantifiable through TPC monitoring [21]. As illustrated in Figure 4d, the TPC of the DC-E group after depolymerization was significantly higher than that of the DC-0.1E group, in which the TPC of the E-Fe/CNT group reached 2.7 g/L, which was 34 times higher than that of the control group, and far exceeded the threshold of TPC for AD [35]. Therefore, solid–liquid separation prior to AD was essential to use only the solid fraction for subsequent operations. The DC-0.1E group was slightly higher than the previous study [3], suggesting that 0.1% acid in combination with water/ethanol solvent promotes lignin depolymerization well.

GC-MS analysis elucidated substrate depolymerization pathways, revealing significant product profile variations across treatment groups (Figure 4e). The E-Ni/CNT system demonstrated superior product abundance and relative composition, with 4-ethoxyphenol (23.4%), vanillin (9.3%) and ethyl benzoate (31.3%) predominating—characteristic lignin-derived phenolic compounds consistent with findings in TPC. This confirmed ethanol’s synergistic role in facilitating lignin-to-phenol conversion through solvation effects. Ni/CNT systems exhibited enhanced 4-ethoxyphenol and vanillin production compared to HP, attributable to Ni-based catalysts effectively reducing the activation energy of the C-O bond during depolymerization to promote the conversion of lignin [36]. The catalytic proficiency in selective bond cleavage demonstrates nickel-based materials’ superiority in directing lignin valorization pathways toward high-value aromatic products.

2.3. Possible Depolymerization Conversion Pathways

The Ni/CNT catalyst comprises nickel nanoparticles anchored on CNTs, a configuration that provided abundant active sites while enhancing structural stability and electron transfer capacity [37]. As depicted in Figure 5, the depolymerization mechanism involves three coordinated pathways. The β-1,4 glycosidic bond in cellulose is broken and first hydrolyzed to cellulosic disaccharides and oligofructose, whereas the Ni/CNT catalyst can accelerate the hydrolysis process of cellulose by providing acidic sites for protonation to promote the β-1,4 glycosidic bond breakage [38]. The generated glucose Ni/CNT catalyst with the action of the active site is further dehydrated and converted to short-chain carboxylic acids (87.0% selectivity of acetic acid, see Section 2.2.2) by the inverse hydroxyl aldehyde condensation pathway, and accompanied by esterification reactions to produce compounds such as ethyl acetate (24.5–27.0%). The depolymerization of hemicellulose, on the other hand, exhibits a multiphase character: it depolymerases under hydrothermal conditions to low-carbon sugars (pentose and hexose) and soluble hemicellulose, whereas the dehydration of pentose occurs catalyzed by H₃O⁺ generated by solvent auto-ionization, followed by its further conversion to C2–C5 short-chain acids via a Ni/CNT-catalyzed decarboxylation-hydrogenation tandem reaction [39]. Ethanol inhibits side reactions by stabilizing the carbon positive intermediate state (via hydrogen bonding of -OH) [40]. For the lignin fraction, it was first degraded to lignin oligomers under hydrothermal conditions. It was shown that Ni/CNT has the ability to protonate ether bonds in lignin and cleave lignin, while the catalyst acidic site induces isomerization and dehydration of the molecule [41]. FTIR and GC-MS analyses confirmed the heterolytic dissociation of the β-O-4 bond on the surface of Ni/CNT to generate phenoxy radical intermediates, which were further converted by hydrolysis-rearrangement reaction to high value-added phenolic products (4 ethoxy phenol, vanillin and ethyl benzoate, etc.) [42].The Ni/CNT catalyst not only enhanced the conversion of reducing sugars but also improved the selectivity of the target products, which was related to the inhibition of polycondensation side-reactions by the active sites of the metal and the domain-limited mass transfer on the CNT surface [43]. Ni/CNT catalysts not only enhance the reaction rate of hydrothermal depolymerization but also improve the selectivity of the target products.

2.4. AD Performances

2.4.1. Daily Biogas Production

Figure 6a,b illustrated daily biogas production (DBP) profiles across experimental groups. DC-E systems exhibited their first production peak on day two post solid–liquid separation, whereas DC-0.1E systems demonstrated accelerated process initiation with an earlier day one peak, attributable to abundant bioavailable monomers (sugars and acids) in hydrolysates [44]. In addition, Fe/CNT achieved 343 mL, representing a 68.4% increase over control (203 mL), correlating with RS content (8.8 g/L). Nickel-containing systems showed delayed first-phase production (138–264 mL), consistent with microbial adaptation requirements to nickel-rich environments, as documented by Abdelsalam et al. [45]. All DC groups reached secondary production peaks during 6–9 d, 1–4 days earlier than controls. The E-Fe/CNT system achieved 406 mL secondary peak, surpassing E-HP (305 mL) and Fe/CNT (358 mL) by 33.0% and 13.5%, respectively, likely due to enhanced lignin conversion and cellulose accessibility from bond cleavage [44]. DC-E systems consistently outperformed DC-0.1E counterparts in secondary phase production, potentially reflecting differential hemicellulose utilization patterns during pretreatment. These catalytic divergences align with established metal-specific functions: iron’s redox activity accelerates initial hydrolysis, while nickel’s role in methanogen cofactors (though requiring microbial acclimation) enhances system stability and ultimate yield [46]. The DBP demonstrated DC dual benefits—rapid initial conversion of labile fractions followed by sustained lignocellulose utilization.

2.4.2. Biomethane Yield

The analysis demonstrated that DC groups universally enhanced substrate conversion efficiency and biomethane yield (BY). BY as a critical parameter to assess AD performance and substrate utilization, revealed significant catalytic variations. As illustrated in Figure 6c, Ni/CNT systems exhibited superior methanogenic performance across solvent environments, with aqueous/ethanol systems achieving peak BY of 192 mL/g VS—representing 24.9% and 32.8% enhancements over HP and control groups, respectively. Ethanol-mediated DC showed slightly reduced efficiency (184 mL/g VS), corresponding to 18.1% and 27.6% improvements versus E-HP and controls, likely attributable to carbon loss during ethanol-phase processing. Fe-based catalysts demonstrated comparable efficacy, with BY of Fe/C systems increased 21.1% and 28.9% relative to control and HP groups, respectively. The acid–alcohol promoted depolymerization can effectively promote the conversion of biomethane.

2.4.3. Substrate Conversion

Biogas production directly correlates with substrate bioconversion efficiency through anaerobic microbial metabolism. As a lignocellulosic substrate, CS requires effective depolymerization for optimal conversion. TS and VS conversion rates quantitatively reflect biodegradation performance, as shown in Figure 6d. For the ethanol group, E-Ni/CNT achieved the highest TS (70.3%) and VS (70.6%) conversion rates, significantly surpassing control (51.1% and 53.0%) and E-HP groups (53.9% and 63.6%). In DC-0.1E systems, Ni/CNT demonstrated superior VS conversion (71.7%), indicating enhanced substrate bioavailability through catalytic depolymerization.

Figure 6e revealed differential polysaccharide conversion patterns. DC-E and DC-0.1E systems demonstrated superior cellulose (70.9–86.6%) and hemicellulose (43.7–84.1%) conversion compared to HP and control groups. The DC-0.1E group hemicellulose conversion was lower than that of the DC-E group, which was related to the loss of hemicellulose from the DC-0.1E group during depolymerization. The highest cellulose conversion was obtained for Ni/CNT, and the highest hemicellulose conversion was obtained for E-Ni/CNT, which were higher than those of the HP group and the control group. This suggested that the addition of catalysts in the depolymerization can significantly increase the conversion of cellulose and hemicellulose, thus promoting an increase in biomethane production.

Post-digestion hydrolysate phenolic profiles (Figure 6f) showed ethanol systems maintained lower phenolic levels (21–32 mg/L) through effective solid–liquid separation; whereas, it was 103–119 mg in the DC-0.1E group, which was significantly higher than the DC-E group and the control group. The total phenol content of the Fe/CNT group reached 119 mg, which was 3.7 and 5.8 times higher than that of the control group (32 mg) and the E-Fe/CNT group (21 mg), respectively. This phenolic accumulation inversely correlates with methane yield, indicating that the presence of total phenol in the solution affects the yield of biomethane.

3. Materials and Methods

3.1. Materials

Field-collected CS from Yanqing District, Beijing was stored in zip lock bags under dark, anhydrous conditions prior to experiments. The inoculum obtained from the Fengtai Circular Economy Industrial Park (Beijing) had undergone dark stabilization at ambient temperature (25 ± 2 °C) for 7 days, followed by supernatant decantation. The characteristics of the raw materials were detailed in Table S1.

The carbon supports employed in this study comprised activated carbon (AC, 100–200 mesh, 75–150 μm) and multiwalled carbon nanotubes (CNTs, 8–15 nm). Prior to metal loading, all supports were thermally activated in an oven at 105 °C for 12 h. Metal precursors (FeCl₃·6H₂O and NiCl₂·6H₂O) were dissolved in deionized water to prepare 1.0 M solutions. Through incipient wetness impregnation, 10 wt% metal loading was achieved by dropwise addition of precursor solutions to carbon matrices [47,48,49]. The impregnated catalysts underwent freeze-drying (−50 °C, 10 Pa) for 12 h followed by programmed thermal treatment in a tube furnace under N₂ atmosphere (100 mL/min). Based on previous studies conducted by our group, the calcination temperatures for Fe/C, Ni/C, Fe/CNT and Ni/CNT were 600, 800, 800 and 800 °C, respectively [3]. The XRD and XPS plots of the catalysts showed characteristic peaks corresponding to Ni and Fe metal particles, respectively (Figures S1 and S2).

3.2. DC Methods

CS was used as the raw material, 4% catalyst (TS) was added and solvent containing 0.1% H₂SO₄ (ethanol or water/ethanol 9/1, v/v) was added, stirred homogeneously and then placed in a high temperature and pressure reactor at a temperature of 140 °C for 20 min, then cooled down quickly to room temperature after the end of the reaction and quickly transferred to a blue cap bottle. At the end of the depolymerization, the hydrolysis products and the changes of solid components in each test group were determined. The catalysts were selected as Fe/C-600, Ni/C-800, Fe/CNT-800 and Ni/CNT-800 according to the research results of the group [3]. Hydrothermal groups (E-HP and HP) were also set up as controls. The experimental design was shown in Table S2.

3.3. Batch AD Experiment

The sample of 3.2 was subjected to batch AD test (35 ± 1 °C). Due to the high concentration of inhibitors after DC-E, solid–liquid separation was carried out after depolymerization and the solid fraction was taken for AD. The water/ethanol as solvent group was directly subjected to AD after the depolymerization treatment. In the AD test, the load of raw material was 50 g TS/L, the inoculum was 30 g TS/L, the fermentation bottle was 250 mL blue cap bottle with 200 mL effective volume and the fermentation time was 40 d. Three parallels were set up in each group, and the results were averaged.

To avoid the effect of ethanol in the depolymerization solvent, AD with the same volume of water/ethanol was set up as a control. The biogas production was recorded daily during anaerobic digestion and its gas composition was determined; and the resulting gas production was converted to the volume under standard conditions by the ideal gas equation of state (PV = nRT). The pH, alkalinity, VFAs, TS and VS of the fermentation materials were determined at the end of anaerobic digestion.

3.4. Analytical Methods

The daily biogas production was determined using the drainage method. Biomethane concentration was analyzed by GC-2014 (Shimadzu, Kyoto, Japan).

X-ray photoelectron spectroscopy (XPS) was conducted using a thermo scientific K-Alpha instrument (Thermo Fisher Scientific, Waltham, MA, USA). X-ray diffraction (XRD) analysis was conducted using a Rigaku D/Max-2500 X-ray diffractometer (Rigaku Corporation, Tokyo, Japan). Cellulose, hemicellulose and lignin contents were quantified by the method of Lu et al. [21]. The microstructure of CS was examined by SEM (Hitachi, S-4700, Tokyo, Japan). Functional group analysis was performed using a Fourier transform infrared spectrometer (Nicolet 6700, Thermo Fisher, Waltham, MA, USA) and KBr particle compression. Cellulose accessibility, surface lignin area and hydrophobicity were quantified by the method of Zhu et al. [27]. The crystallinity index (CrI) of the different samples was calculated from the reflected intensity using Equation (1) as described in the literature [50]:

$$\mathrm{C}\mathrm{r}\mathrm{I}\left(\%\right)={\displaystyle \frac{{\mathrm{I}}_{002}-{\mathrm{I}}_{\mathrm{a}\mathrm{m}}}{{\mathrm{I}}_{002}}}\times 100\%$$

(1)

where I₀₀₂ represents the diffraction intensity of the 002 plane and I_am represents a diffraction angle of 18.0°.

The pH values in this study were measured using a pH meter (Five Easy Plus, Mettler Toledo, Greifensee, Switzerland). The concentration of total volatile fatty acids (VFAs) was determined using GC-2014 (Shimadzu, Kyoto, Japan). The reducing sugar content (RS) of the samples was quantified by DNS method [51]. Total phenolic content (TPC) was assessed using the Folin phenol reagent method [52].

Data analyses were performed in Microsoft Excel 2019 (17.0, Microsoft, Redmond, WA, USA, 2018), Spearman correlation data analyses were performed in SPSS 26.0 (IBM, Armonk, NY, USA, 2019), and all graphs were generated using Origin 2024 software.

4. Conclusions

This study demonstrated the effectiveness of the DC-E strategy for enhancing the AD performance of CS. The synergistic interaction of ethanol and catalysts (particularly Ni/CNT) significantly altered the chemical composition and structural properties of CS, leading to improved substrate accessibility. The Ni/CNT catalyst exhibited superior catalytic activity in ethanol, achieving a lignin conversion rate of 48.5%. The water/ethanol (9:1) system efficiently degraded hemicellulose (53.6% conversion) while preserving >90% cellulose, providing an ideal substrate for AD. Structural modifications, including hydrogen bond disruption (CrI reduction from 38.4% to 32.6%) and increased cellulose accessibility (909 mg/g), significantly improved enzymatic hydrolysis and microbial utilization. The E-Ni/CNT system achieved the highest BY (184 mL/g VS), 27.6% higher than the control group, attributed to targeted lignin–carbohydrate bond cleavage and enhanced substrate bioavailability.