Efficient Hydrogenolysis of Lignin into Aromatic Monomers over N-Doped Carbon Supported Co and Dual-Phase Mo_xC Nanoparticles

Abstract

The key to selectively cleaving C–O bonds in lignin to produce high-value aromatic chemicals lies in the development of efficient and stable catalysts. In this study, a heterostructured catalyst with N-doped carbon-supported Co and dual-phase Mo_xC nanoparticles was prepared via the in situ pyrolysis of a Co–Mo–N precursor. The dual-phase α-MoC/β-Mo₂C heterostructure is adjusted by varying the Co:Mo ratio to affect the structure and electronic properties of the catalyst. The heterostructures bring about enhanced electron transfer from Co to Mo, which promotes hydrogen dissociation over the Co sites, significantly improving the catalyst’s hydrogenolysis activity and stability. The optimal catalyst with Co₁Mo_xC@NC exhibits excellent hydrogenolysis activity; under the optimal reaction conditions (260 °C, 1 MPa H₂, 3 h), the yield of aromatic monomers reaches 28.5%. Such prominent performance not only benefits from the numerous α-MoC/β-Mo₂C hetero-interfaces that offer abundant active sites for hydrogen dissociation, but also should be ascribed to the strong synergistic effect between Co and Mo.

1. Introduction

Lignin, a natural phenolic polymer abundantly present in plant cell walls, is characterized by its structural heterogeneity and diversity. As the second most abundant natural macromolecule after cellulose, lignin is enriched with aromatic groups and contains not only abundant benzene rings, but also various oxygen-containing functional groups (e.g., -OH, -OCH₃, and -CO) [1,2,3]. These unique chemical properties render lignin a promising alternative to fossil resources such as coal and petroleum [4,5]. However, although large amounts of lignin are generated annually worldwide, its efficient utilization remains significantly limited, with most industrial lignin being used only for low-value combustion. Consequently, achieving the efficient catalytic conversion of lignin remains a significant challenge in the field of biomass resource utilization [6,7].

Lignin, a complex natural aromatic macromolecule, consists primarily of three monolignols—p-coumaryl alcohol (H), coniferyl alcohol (G), and sinapyl alcohol (S)—which are linked through C–O bonds (e.g., β–O–4, α–O–4, and 4–O–5) or C–C bonds (including β–1, β–5, and β–β) [8]. Notably, although lignin exhibits diverse bonding patterns, the β–O–4 bond is documented as the most abundant, accounting for over 50% of the total interunit linkages [9]. This structural characteristic makes the selective cleavage of C–O bonds a crucial route for the targeted conversion of lignin into high-value chemicals and liquid fuels. Lignin depolymerization methods are generally classified into acid/base catalysis, oxidative depolymerization, and catalytic hydrogenolysis [10,11,12]. Among these approaches, catalytic hydrogenolysis is considered one of the most promising techniques owing to its low oxygen content in depolymerized products and high conversion efficiency [4].

Noble metals are well known for their outstanding hydrogenation activity in lignin hydrogenolysis due to their superior H₂ dissociation ability [13,14]. However, their widespread application is constrained by limited natural reserves and high costs [15,16]. In contrast, transition metals such as Ni, Co, and Mo are extensively employed in hydrogenolysis and hydrodeoxygenation, owing to their abundant reserves and low cost [17,18], yet their catalytic activity remains lower than that of noble metals.

In recent years, transition metal carbides (TMC), particularly Mo-based carbides, have been extensively studied in catalysis due to their unique electronic structures and surface chemical properties, which are known to exhibit noble-metal-like properties. Typically, Mo-based carbide catalysts can be classified into two phases: β-Mo₂C and α-MoC. The β-Mo₂C phase, characterized by higher thermal stability and more active sites, has been shown to possess a superior C–O bond cleavage ability since H₂ can be efficiently activated on its surface [19]. In contrast, α-MoC is distinguished by its abundant acidic sites, which play a crucial role in significantly promoting the alkylation of aromatic monomers [20]. Wu et al. [21] notably synthesized a mixed-phase Mo_xC catalyst via an impregnation method. Studies have shown that the coexistence of α-MoC and β-Mo₂C phases in the catalyst leads to higher aromatic hydrocarbon yields compared to single-phase catalysts. This finding suggests that synergistic effects between mixed phases may be leveraged to optimize catalytic performance.

Among catalyst supports, carbon-based materials are extensively used in lignin hydrogenolysis due to their large specific surface area, abundant porous structures, and excellent thermal resistance [8,22]. However, weak interactions between traditional carbon supports and active metals, combined with the lack of effective metal anchoring sites, are known to readily cause metal nanoparticle aggregation, thereby reducing catalytic performance. This issue has been effectively addressed by the emergence of nitrogen-doped carbon supports (NPs). The introduction of N atoms has been shown not only to stabilize and disperse metal nanoparticles (NPs), but also to modify the electronic structure of the carbon support, enhancing the electron transfer capability between NPs and the support [22,23,24]. This doping has been proven to significantly enhance metal–support interactions, consequently improving the catalyst’s stability and activity. Hence, we prepared a heterostructured catalyst with N-doped carbon-supported Co and dual-phase Mo_xC nanoparticles by the in situ pyrolysis of a Co-Mo-N precursor [25,26,27]. The aromatic monomer yield reached 28.5% for the hydrogenolysis of lignin. Such prominent performance is attributed to the increase in electron-rich Mo sites and the numerous α-MoC/β-Mo₂C hetero-interfaces.

2. Results and Discussion

2.1. Structure of CoMo_xC Catalysts

In this study, nitrogen-doped carbon-supported Co_yMo_xC catalysts were successfully synthesized through the in situ pyrolysis of a mixture of glucose, melamine, and metal salts. The typical preparation process is shown in Figure 1a. The morphology of the Co₁Mo_xC@NC catalyst was thoroughly investigated using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As shown in Figure 1b,c, the catalyst exhibits a graphene-like carbon nanosheet morphology with abundant wrinkles and porous structures [28]. This is because melamine completely decomposes into free radicals and volatile gases, which results in the formation of wrinkled graphene-like carbon nanosheets [29,30,31]. As shown in Figure 1d, the metal nanoparticles in the catalyst are well-dispersed. The STEM image (Figure 1e) and EDS elemental mapping (Figure 1f and Figure S1a) further confirm the uniform distribution of metal particles without noticeable aggregation [32]. Additionally, based on the STEM image, the average particle size of the metal NPs is determined to be 8.32 nm. The HRTEM images (Figure 1g,h) show lattice spacings of approximately 0.228 nm, 0.204 nm, and 0.205 nm, corresponding to the (200) plane of α-MoC, the (211) plane of β-Mo₂C, and the (111) plane of Co, respectively. This indicates that, due to the catalytic influence of Co, a heterointerface between α-MoC and β-Mo₂C was formed, and the NPs were tightly encapsulated by carbon layers [25]. As shown in the SEM (Figure 1i) and TEM images (Figure 1j), the Co_1.5Mo_xC@NC catalyst retains its carbon nanosheet structure. However, localized metal aggregation is observed in specific regions (Figure 1k). The STEM image (Figure 1) and EDS mapping (Figure 1m and Figure S1b) indicate a substantial increase in the size of the metal nanoparticles, with an average size of 14.38 nm.

The morphological structure of the MoC@NC catalyst was found to be similar to that of Co₁Mo_xC@NC, as shown in the SEM (Figure S2a) and TEM images (Figure S2b), both exhibiting a graphene-like carbon nanosheet structure. In the HRTEM image (Figure S2c), the lattice planes of graphitic carbon are clearly observed. To investigate the crystal phase structure of the catalyst, XRD analysis was conducted. As shown in Figure 2a, the crystal phases of the active metal and the NC support are distinctly identified. The broad peak at 2θ = 25.8° corresponds to the C (002) plane, while the three diffraction peaks at 2θ = 36.4°, 42.3°, and 61.4° are attributed to the (111), (200), and (311) planes of α-MoC (MoC, JCPDS 89-2868), respectively [33]. In the XRD pattern of MoC@NC, only a weak (111) plane diffraction peak is observed, with no other crystal planes detected, indicating that the formed α-MoC nanoparticles are small in size and uniformly dispersed. This observation is further supported by the STEM image (Figure S2d), which shows that the average particle size of α-MoC NPs is 2.2 nm. When a small amount of Co (JCPDS 15-080) is introduced (Figure 2a and Figure S3), the metal diffraction peak of the red line of Co_0.2Mo_xC@NC catalyst was significantly weakened or even completely disappeared. This phenomenon suggested that a strong interaction between Co and Mo_xC had been established, which significantly enhanced the dispersion of the metal [32,34]. This finding aligns well with the results in Figure S2h, where the average particle size is only 1.39 nm. As the Co content is further increased, diffraction peaks corresponding to the (111) plane (2θ = 44.2°) and the (200) plane (2θ = 51.5°) of Co gradually appear [35]. Notable changes in the crystal phase structure of Mo_xC are observed, with diffraction peaks corresponding to the (211) plane (2θ = 36.5°) and the (400) plane (2θ = 61.5°) of β-Mo₂C emerging. This indicates that the introduction of Co facilitates the formation of a dual-phase heterostructure consisting of α-MoC and β-Mo₂C [25,34].

To investigate the specific surface area and pore structure characteristics of the catalysts, BET analysis was conducted to characterize the N₂ adsorption–desorption isotherms and pore size distribution of various catalysts (Figure S4a,b). For both the single-metal (Co@NC, and MoC@NC) and bimetallic (Co_yMo_xC@NC) catalysts, we see typical Type IV adsorption–desorption isotherms with H4-type hysteresis loops, indicating the presence of mesoporous structures [36,37]. The materials exhibit a broad pore size distribution, in the range of 2–40 nm (with average maxima at about 9–14 nm), which facilitates the effective mass transfer of lignin fragments [38]. Compared to Co@NC and MoC@NC catalysts, Co_yMo_xC@NC possesses a larger specific surface area (Table S1), likely due to the synergistic interaction between Co and Mo, which promotes metal dispersion and prevents aggregation [39]. However, the introduction of excessive Co could lead to pore blockage or coverage, leading to a reduction in specific surface area and pore volume [40]. Notably, the adsorption–desorption isotherms and pore size distribution of the catalysts before and after Co introduction are almost identical, indicating that the mesoporous structure of the catalyst remains unchanged [41,42].

Raman spectroscopy was performed at an excitation wavelength of 785 nm for the characterization of the catalysts (Figure S5). All catalysts display typical D-band and G-band absorption peaks at 1350 cm⁻¹ and 1480 cm⁻¹, respectively. The D-band corresponds to amorphous carbon or defective graphite structures (sp³), while the G-band represents graphitic carbon (sp²). Therefore, a higher intensity ratio of the D-band to the G-band (I_D/I_G) indicates a greater number of defect structures in the catalyst [20]. The experimental results show that the I_D/I_G ratio of the catalysts is approximately 1.00, demonstrating that N doping effectively introduces a significant number of defects into the catalyst [43]. Additionally, it is evident that the introduction of Co further increases the number of defect sites, suggesting that Co is successfully anchored onto the NC carrier [37].

The electronic structure and chemical composition of the catalyst surface were thoroughly investigated via XPS analysis. The survey spectrum of Co_yMo_xC@NC (Figure S6) reveals the presence of Co, Mo, N, C, and O on the catalyst surface. The C 1s spectrum (Figure S7) is deconvoluted into five peaks corresponding to C–Mo (283.8 eV), C–C (284.8 eV), C–N (285.5 eV), C–O (286.2 eV), and C=C (289.0 eV) [34,44,45]. The spectrums of N 1s, Co 2p and Mo 3d were shown in Figure 3. The red line represents original data of different element. The dash line represents the binding energy of Co and Mo, respectively. The N 1s spectrum (Figure 3a and Figure S8) displays three deconvoluted peaks assigned to Mo–N (394.2 eV), Pyridinic-N/Co–N (398.4 eV), and Graphitic-N (401.3 eV) [46,47]. The presence of C–N and Pyridinic-N confirms the effective incorporation of N atoms into the carbon support. Moreover, Co₁Mo_xC@NC exhibits the highest proportion of Pyridinic-N or Co–N (56%) (Table S2). Since Pyridinic-N serves as a metal coordination site, its increased content likely facilitated electron transfer between metal NPs and the carbon support, thus markedly improving the catalytic performance [48]. The Co 2p_3/2 spectrum (Figure 3b and Figure S9) shows four deconvoluted peaks corresponding to Co⁰ (778.2 eV), Co₃O₄ (781.3 eV), Co–N (781.7 eV), and a satellite peak of Co 2p_3/2 (785.5 eV) [17,49,50]. In the Mo 3d spectrum (Figure 3c and Figure S10), four doublets are identified, indicating the presence of four valence states of Mo species on the catalyst surface. Peaks at 228.6/231.8 eV and 229.2/232.5 eV are attributed to β-Mo₂C (Mo²⁺) and α-MoC (Mo³⁺), respectively [51,52]. Mo⁴⁺ and Mo⁶⁺ species are assigned to MoO₂ and MoO₃, which likely formed due to unavoidable surface oxidation [53,54,55]. Notably, compared to Co@NC, the Co 2p spectrum of Co_yMo_xC@NC undergoes a pronounced positive shift, while the Mo 3d spectrum shows a negative shift. In the Co 2p spectrum (Figure 2b) and Mo 3d spectrum (Figure 2c) of Co₁Mo_xC@NC, clear shifts in the binding energies of Co and Mo are observed. This phenomenon may be attributed to the lower electronegativity of Co (1.88) compared to Mo (2.16), leading to electron transfer from Co to Mo_xC. Such electronic interactions likely alter the electronic structure of the metal particles and the support, which consequently improve the catalyst’s stability. Additionally, the electron-deficient Co may have promoted the dissociative adsorption of hydrogen, further improving the catalyst’s hydrogenolysis activity [35,47,56].

2.2. Catalysts Evaluation for the C–O Bond Cleavage

As previously noted, the β–O–4 bond is the most abundant C–O bond in lignin (≥40%). Therefore, to evaluate the hydrogenolysis activity of the catalyst, 2-phenoxy-1-phenylethanol (PPE), a widely used β–O–4 model compound, is taken as the reaction substrate to evaluate the catalytic performance (

Table 1. Hydrolysis of PPE on different catalysts.

Entry	Catalysts	Conversion (%)	Product Yield (%)
			2	3	4
1	Blank	0	0	0	0
2	NC	0	0	0	0
3	MoC@NC	68.4	37.8	29.2	17.7
4	Co0.2MoxC@NC	84.3	43.8	38.5	0
5	Co0.8MoxC@NC	90.4	42.3	36.1	0
6	Co1MoxC@NC	99.9	49.9	41.2	0
7	Co1.5MoxC@NC	92.6	44.5	36.2	0
8	Co2MoxC@NC	87.9	39.5	32.1	0
9	Co@NC	84.8	32.8	28.6	0

Reaction conditions: 0.1 g PPE, 0.04 g catalyst, 20 mL isopropanol, 220 °C, 3 h, 1 MPa H₂.

). Notably, in the absence of a catalyst or with only the NC carrier, PPE remained intact with no detectable depolymerization products (entries 1–2). In contrast, catalysts loaded with active metals (entries 3–9) demonstrated varying degrees of depolymerization activity, indicating that the active metals play a crucial role in the lignin hydrogenolysis reaction. Although MoC@NC (entry 3) and Co@NC (entry 9) exhibited some hydrogenation activity, their depolymerization efficiency remained below expectations. The bimetallic CoMo_xC@NC (entries 4–8) exhibited a much higher conversion and phenol yield. The experimental results reveal that the synergistic interaction between Co and Mo substantially boosted the catalytic efficiency. Notably, Co₁Mo_xC@NC exhibited the highest hydrogenation efficiency, achieving the complete conversion of PPE with yields of ethylbenzene and phenol reaching 49.9% and 41.2%, respectively. However, as the Co content increased, a decline in depolymerization efficiency was observed, likely due to the aggregation of excess Co, which negatively impacted the hydrogenation activity of the catalyst [57].

2.3. Reusability of the Catalyst

The stability and reusability of a catalyst are key criteria in evaluating its potential for large-scale industrial applications. To evaluate the stability of the Co₁Mo_xC@NC catalyst, cycling stability tests were performed at 200 °C and 1 MPa H₂ for 3 h per cycle. The results reveal that the PPE conversion rate exhibited no significant decrease after five cycles (Figure 4a). To gain deeper insights into the crystal phase and morphological changes of the catalyst before and after the reaction, systematic physicochemical characterizations were performed using XRD, XPS, BET, and SEM. The XRD patterns (Figure 4b) reveal that the heterogeneous structure and particle size of the recovered catalyst were largely retained relative to the fresh catalyst. XPS analysis indicates that the binding energies of Co 2p (Figure 4c) and Mo 3d (Figure 4d) in the recovered catalyst were nearly identical to those of the fresh catalyst, suggesting that the electronic structure of the catalyst remained stable during the reaction. N₂ adsorption–desorption isotherms (Figure 4e) and pore size distribution (Figure 4f) reveal that while the specific surface area of the catalyst slightly decreased after five cycles, its mesoporous structure was well preserved. Additionally, SEM (Figure 4g) and TEM images (Figure 4h) further confirm that the porous wrinkled graphene-like carbon nanosheet structure of the catalyst was well preserved with no noticeable degradation. These findings collectively indicate that the Co₁Mo_xC@NC catalyst exhibits excellent cycling stability.

In Co-based catalysts, the hydrogenolysis of PPE typically begins with a dehydrogenation process, leading to the swift generation of intermediates with reduced bond energies, such as 2-phenoxyacetophenone (228 KJ mol⁻¹) [58]. However, there were no detectable intermediates, including 2-phenoxyacetophenone or acetophenone. Instead, the depolymerization products exhibited significant yields of ethylbenzene and phenol, accompanied by small amounts of styrene. Based on these experimental results, the following depolymerization pathway is proposed (illustrated in Figure 4i). Driven by the electronic synergy between Co and Mo_xC, along with the involvement of active hydrogen (H*), the catalytic reaction preferentially cleaves the C_β–O bond [59], generating the intermediates 1-phenylethanol and phenol. Subsequently, 1-phenylethanol rapidly undergoes dehydroxylation to form styrene, which is further hydrogenated to produce the final target product, ethylbenzene (Route 1) [60]. In contrast, if the reaction initially involves the cleavage of the C_α–OH bond, it would produce phenethoxybenzene with a higher bond energy (275.3 KJ mol⁻¹) (Route 2), thereby hindering subsequent hydrogenation reactions [17,61]. Therefore, the primary reaction pathway is the direct cleavage of the C_β–O bond.

2.4. Hydrogenolysis of Birch Lignin

Based on the selective C–O bond cleavage capability demonstrated by the Co_yMo_xC@NC catalyst in lignin model compounds, this study extended its investigation to the catalytic depolymerization of birch lignin under conditions of 260 °C and 1 MPa H₂ for 3 h (as shown in Figure 5a and Table S3). The depolymerized products were analyzed using GC-MS, revealing that the blank control group produced only 5.7% aromatic monomers. The monomer yields obtained with the MoC@NC and Co@NC catalysts were 11% and 17.4%, respectively. Importantly, the Co₁Mo_xC@NC catalyst demonstrated superior catalytic performance, achieving an aromatic monomer yield of 28.5%, surpassing that of even the commercial Ru/C catalyst (20.1%). These results indicate that the synergistic interaction between Co and Mo_xC is pivotal in enabling efficient C–O bond cleavage in birch lignin and producing high-value aromatic monomers. The comparison of hydrogenolysis activity using the Co₁Mo_xC@NC catalyst with other reported metal-based catalysts is shown in Figure S11. Although the reported noble metals (Pd, Pt, Ru) exhibit outstanding catalytic activity, their high cost restricts their large-scale application. In this study, non-precious metals (Co and Mo) are used in the hydrogenolysis of lignin due to their low cost and moderate hydrogenation activity. Furthermore, the heterostructured catalyst with N-doped carbon-supported Co and dual-phase Mo_xC nanoparticles exhibits higher hydrogenolysis activity and a significant increase in monomer yield/selectivity compared to the individual catalysts. Therefore, the Co₁Mo_xC@NC catalysts demonstrate enormous potential for large-scale lignin depolymerization.

Furthermore, this study systematically investigated the influences of crucial reaction parameters such as temperature, gas pressure, and reaction time on lignin depolymerization. As shown in Figure 5b, under an argon atmosphere (1 MPa Ar) without H₂, the yield of monomers significantly decreased to 9.4%, demonstrating the essential role of H₂ in the hydrogenolysis reaction. Additionally, it was found that elevated reaction temperatures (Figure 5c) and prolonged reaction times (Figure 5d) both led to the repolymerization of the products, thereby reducing the yield of aromatic monomers from lignin. These findings provide valuable insights into optimizing reaction conditions to maximize monomer production and simultaneously mitigate unwanted side reactions.

3. Material and Methods

3.1. Materials

The deionized water used in the experiment was obtained from a purified water system. The following chemicals were purchased from Shanghai Macklin Biochemical Co., Ltd. (Shanghai, China). cobalt nitrate hexahydrate (Co(NO₃)₂·6H₂O, 99%), ammonium molybdate ((NH₄)₂MoO₄, 98%), melamine (C₃H₆N₆, 99%), glucose anhydrous (C₆H₁₂O₆, 99%), 2-phenoxy-1-phenylethanol (C₁₄H₁₄O₂, 98%), n-decane (C₁₀H₂₂, 99%) and dodecane (CH₃(CH₂)₁₀CH₃, 99%). Methanol, ethanol, isopropanol, and sodium hydroxide were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All chemicals used were of high purity and were used without further purification.

3.2. Catalyst Preparation

The MoC@NC catalyst was synthesized via a two-step pyrolysis method. The synthesis procedure was conducted as follows: 1 g of anhydrous glucose and 10 g of melamine were dissolved in 40 mL of deionized water and stirred thoroughly for 2 h. Subsequently, 0.2 g of ammonium molybdate was added, and the mixture was stirred for an additional 2 h. The mixture was then dried in an oil bath at 80 °C to eliminate residual moisture, and the resulting solid was transferred to an oven and dried overnight at 80 °C (approximately 12 h). After drying, the solid mixture was ground into a fine powder, placed in a ceramic boat, and loaded into a tube furnace. Under an argon atmosphere, the temperature was raised to 600 °C at a heating rate of 2.5 °C/min, maintained for 1 h, then increased to 800 °C at the same rate and held for another hour. After natural cooling to room temperature, the product was labeled as MoC@NC for subsequent testing.

The preparation method for Co_yMo_xC@NC catalysts followed a similar procedure to that of MoC@NC (where y represents the mass ratio of cobalt nitrate hexahydrate to ammonium molybdate, with y = 0.2, 0.8, 1, 1.5 and 2). The typical synthesis steps were as follows: 1 g of anhydrous glucose and 10 g of melamine were dissolved in 40 mL of deionized water and stirred thoroughly for 2 h. Then, 0.2 g of ammonium molybdate and a specified amount of cobalt nitrate hexahydrate were added, and the mixture was stirred for another 2 h. After drying in an oil bath and oven, the precursor was ground into a fine powder and placed in a tube furnace for pyrolysis following the aforementioned temperature protocol. After natural cooling to room temperature, the product was labeled as Co_yMo_xC@NC. For comparison, a catalyst with 0.4 g of cobalt nitrate hexahydrate was synthesized using the same method and labeled as Co@NC.

3.3. Catalyst Characterization

X-ray diffraction (XRD) analysis was performed using an Ultima IV instrument manufactured by Rigaku, Japan, with Cu Kα radiation (λ = 1.541 Å). The scan range was set to 10–80°, with a scanning speed of 5°/min. The results were compared against standard powder XRD cards compiled by the Joint Committee on Powder Diffraction Standards (JCPDS). Scanning electron microscopy (SEM) was performed using a MIRA LMS instrument manufactured by TESCAN, Czech Republic, with an accelerating voltage of 3 kV for morphology imaging and 15 kV for mapping and energy-dispersive spectroscopy. Transmission electron microscopy (TEM) was performed on a Tecnai G2 F20S-Twin instrument manufactured by FEI, including high-resolution transmission electron microscopy (HRTEM), scanning transmission electron microscopy (STEM), and energy-dispersive X-ray spectroscopy (EDS). X-ray photoelectron spectroscopy (XPS) was performed on a Scientific K-Alpha instrument manufactured by Thermo, Waltham, USA, with a pass energy of 150 eV for survey scans and 50 eV for narrow scans, using a monochromatic Al X-ray source (maximum energy: 1486.6 eV). Raman spectroscopy was employed to evaluate the degree of graphitization of the samples, with measurements performed using a Scientific LabRAM HR Evolution instrument from HORIBA, Tokyo, Japan (excitation wavelength: 785 nm). The specific surface area was determined by the BET method and the pore size distribution was determined by BJH using N₂ adsorption data obtained on a JW-BK132F static nitrogen adsorption instrument.

3.4. Catalyst Activity Tests

All experiments were performed in a 50 mL high-pressure reactor, with the stirring speed consistently maintained at 600 rpm. The experiments were conducted under controlled conditions, and the specific operational procedure was as follows: Initially, 0.1 g of 2-phenoxy-1-phenylethanol, 0.4 g of the catalyst, 20 mL of isopropanol, and 20 μL of dodecane (as an internal standard) were added to the reactor. Subsequently, the reactor’s airtightness was checked, and after confirming no leakage, H₂ was introduced for three evacuation cycles to remove residual air. Next, a predetermined amount of H₂ was introduced, and the reaction conditions were adjusted accordingly. After the reaction, the high-pressure reactor was rapidly cooled to room temperature using an ice water bath. The reaction products were then separated from the catalyst, which was subsequently recovered. The recovered catalyst was thoroughly washed with ethanol and deionized water to remove residual reactants and products, then dried under vacuum at 50 °C for reuse. The depolymerization products were analyzed using a gas chromatography–mass spectrometry (GC-MS) system (Agilent 5975-7890A, equipped with an HP-5 quartz capillary column, 30 m × 0.25 mm × 0.20 μm) from Agilent Technologies, Santa Clara, USA. Quantitative analysis was conducted via gas chromatography, and the substrate conversion rate and monomer yield were calculated using the following equations,

$$\mathrm{Conversion} (\%): {\displaystyle \frac{{\mathrm{W}}_{\mathrm{i}1}-{\mathrm{W}}_{\mathrm{r}1}}{{\mathrm{W}}_{\mathrm{i}1}}}\times 100\%$$

$$\mathrm{Yield} \mathrm{of} \mathrm{monomer} (\%): {\displaystyle \frac{{\mathrm{W}}_{\mathrm{a}1}}{{\mathrm{W}}_{\mathrm{i}1}}}\times 100\%$$

${\mathrm{W}}_{\mathrm{i}1}$ is the weight of the lignin model compound; ${\mathrm{W}}_{\mathrm{r}1}$ is the weight of the remaining lignin model compound after the reaction; ${\mathrm{W}}_{\mathrm{a}1}$ is the weight of the aromatic products after the reaction.

The lignin degradation reaction was performed in a fashion similar to that of the lignin model compound. Specifically, 0.3 g of birch lignin, 0.06 g of catalyst, 20 mL of isopropanol, and 20 μL of dodecane were added to a high-pressure reactor, following the same procedure. The lignin conversion and aromatic monomer yield were calculated using the following formulae.

$$\mathrm{Conversion} (\%): {\displaystyle \frac{{\mathrm{W}}_{\mathrm{i}2}- {\mathrm{W}}_{\mathrm{r}2}}{{\mathrm{W}}_{\mathrm{i}2}}}\times 100\%$$

$$\mathrm{Yield} \mathrm{of} \mathrm{aromatic} \mathrm{monomers} (\%): {\displaystyle \frac{{\mathrm{W}}_{\mathrm{a}2}}{{\mathrm{W}}_{\mathrm{i}2}}}\times 100\%$$

${\mathrm{W}}_{\mathrm{i}2}$ is the weight of lignin; ${\mathrm{W}}_{\mathrm{r}2}$ is the weight of remaining lignin after the reaction; ${\mathrm{W}}_{\mathrm{a}2}$ is the weight of aromatic products after the reaction.

4. Conclusions

In conclusion, we have developed Co–Mo heterostructures comprising N-doped carbon and dual-phase Mo_xC nanoparticles as bimetallic catalysts for the hydrogenolysis of lignin. The experimental results indicate that the addition of Co not only facilitates the formation of dual-phase α-MoC/β-Mo₂C in molybdenum carbide, but also constructs Mo–N and Co–N bonds, creating heterointerfaces between N-doped carbon and Co and Mo_xC nanoparticles. The optimal catalyst, Co₁Mo_xC@NC, exhibits excellent hydrogenolysis activity, achieving the nearly complete conversion of PPE under reaction conditions of 220 °C, 1 MPa H₂, and 3 h. After five reaction cycles, the catalyst retained its catalytic activity with excellent stability. When applied to real lignin degradation, the aromatic monomer yield reached 28.5% under optimal reaction conditions (260 °C, 1 MPa H₂, 3 h). Co simultaneously promotes C–O bond cleavage in lignin by increasing electron-rich Mo sites and enhancing the numerous α-MoC/β-Mo₂C heterointerfaces. This study provides a feasible strategy for lignin degradation and its conversion into high-value-added chemicals.