Skip to main content
NCBI home page
Communications Chemistry logoLink to Communications Chemistry
. 2026 Feb 23;9:141. doi: 10.1038/s42004-026-01947-2

Cobalt single-atom and nano catalysts for efficient transfer hydrodeoxygenation of vanillin with formic acid

Jiayi Li 1,#, Guangling Shi 1,#, Zisheng Xu 1, Lu Wang 2,✉, Maofei Ran 1, Tao Dai 1,2,✉
PMCID: PMC13039161  PMID: 41731031

Abstract

This study presents a Co catalyst with coexisting single-atoms and nanoparticles on N-doped carbon (Co1+Con/N-C). Synthesized via a sacrificial MgO-template method, it achieves 100% vanillin conversion and >99% selectivity to 2-methoxy-4-methylphenol at 160°C using formic acid, outperforming single-component catalysts. Kinetic studies reveal a water-mediated dual hydrogen transfer pathway, lowering the apparent activation energy to 61.7 kJ/mol. Isotopic studies suggest a water-mediated mechanism, wherein water molecules facilitate proton transfer and hydrogen spillover through a hydrogen-bonding network. This process synergizes with formic acid dehydrogenation to enable a dual hydrogen transfer pathway, involving protonation of the aldehyde group and hydride (H⁻) attack. The catalyst maintains 95% conversion over ten cycles, demonstrating high stability for biomass upgrading.

Subject terms: Heterogeneous catalysis, Nanoparticles, Nanoparticles

Synergistic effects between single-atom sites and nanoparticles can enhance the performance of single-atom catalysts. In this work, the authors design a catalyst that incorporates cobalt single atoms and nanoparticles co-anchored on a nitrogen-doped carbon support for the catalytic transfer hydrodeoxygenation of vanillin to 2-methoxy-4-methylphenol (MMP) using formic acid, reporting full vanillin conversion with over 99% selectivity for MMP at 160°C.

Introduction

With the gradual depletion of fossil fuels and the escalating demand for energy, the exploration of renewable energy sources has garnered increasing attention1,2. Bio-oil is considered one of the most promising alternatives to conventional fossil fuels. However, using bio-oil directly as biofuel is challenging due to its high oxygen content, which leads to low energy density and unstable combustion3.4. Hydrodeoxygenation (HDO) is recognized as an energy-efficient and viable method for reducing oxygen content5–7. Vanillin, a typical pyrolysis product of lignin, contains oxygen functional groups8,9. Hydrodeoxygenation of vanillin to 2-methoxy-4-methylphenol (MMP) is a potential approach10–12.

The conventional catalytic HDO of vanillin typically necessitates high-pressure hydrogen gas, which poses significant safety risks and entails substantial energy consumption13,14. Moreover, this process can lead to undesirable reactions, such as demethoxylation. Recently, many efforts have increasingly focused on exploring alternative hydrogen sources to mitigate the conditions15,16. Formic acid has gained attention as a highly promising green hydrogen carrier17. Formic acid can in situ release H₂, thereby obviating the need for high-pressure hydrogen gas. Significant advancements have been achieved in the development of noble metal catalysts for the HDO of vanillin utilizing formic acid as the hydrogen source18–20. Nevertheless, the high cost and limited availability of these noble metals constrain their applicability in large-scale industrial processes. Consequently, the development of efficient and stable non-noble metal catalysts has emerged as a new research focus.

Single-atom catalysts (SACs) have exhibited considerable promise across arange of applications21–23, including energy conversion24,25, environmental remediation26,27, and chemical synthesis28. This potential is attributed to their precisely defined active sites and maximized atomic utilization efficiency. Furthermore, SACs exhibit exceptional activity in the catalytic reaction29. However, SACs may encounter problems in complex reactions, such as limited diversity of active sites, which impede in catalyzing multi-step processes that require distinct active centers30. Recent research suggests that the intrinsic activity of SACs can be enhanced by introducing additional metal nanoparticles (NPs)31,32. The synergistic effects such as electronic interactions, spatial cooperation, and tandem catalysis mechanisms between single-atom sites and NPs can optimize reactant adsorption and activation, while broadening catalytic reaction33–37. Therefore, we systematically designed and synthesized synergistic cobalt based catalysts. These catalysts incorporate precisely engineered cobalt single atoms and cobalt nanoparticles co-anchored on nitrogen-doped carbon support. These catalysts exhibit remarkable efficiency in HDO of vanillin to MMP, utilizing formic acid as the hydrogen source.

Therefore, we systematically designed and synthesized a synergistic cobalt-based catalyst. This catalyst incorporates precisely engineered cobalt single atoms (Co1) and cobalt nanoparticles (Con) co-anchored on a nitrogen-doped carbon support. The catalyst exhibits remarkable efficiency in the hydrodeoxygenation of vanillin to 2-methoxy-4-methylphenol (MMP).

Results and discussions

Preparation of cobalt single atom and nanoparticle catalysts

The Co-N-C catalysts, featuring both single-atom and nano-dispersion, were successfully synthesized using a sacrificial support method. In this process, cobalt (II) acetylacetonate, dopamine, and 1,10-phenanthroline monohydrate were combined in ethanol to form the Co(phen)2(acac)2-PDA/MgO precursor. The resulting mixture was pyrolyzed at varying temperatures (973-1173 K) in a 10% H2/N2 atmosphere for 2 hours, followed by the removal of MgO through nitric acid treatment. The catalysts obtained were characterized by distinct active centers and are denoted as Co1/N-C, Co1+Con/N-C, and Con/N-C, respectively (Fig. 1a). Inductively coupled plasma-atomic emission spectroscopy (ICP-AES) analysis revealed cobalt contents of 5.14%, 4.42%, and 3.78% in these catalysts, as detailed in Table S1. The nitrogen adsorption-desorption isotherms of these catalysts exhibited a characteristic type IV isotherm accompanied by an H1-type hysteresis loop, indicative of a mesoporous architecture (Fig. 1b). The X-ray diffraction (XRD) patterns did not reveal any peaks corresponding to cobalt, implying that the cobalt species were either atomically dispersed or amorphous within the Co1/N-C catalyst (Fig. S1). However, with an increase in pyrolysis temperature, diffraction peaks corresponding to metallic cobalt emerged in the Co1+Con/N-C and Con/N-C catalysts.

Fig. 1. Synthesis and characterization of the catalysts.

Fig. 1

Open in a new tab

a A schematic illustration of the synthesis procedure of Co1/N-C, Co1+Con/N-C, and Con/N-C catalysts. b Nitrogen adsorption-desorption isotherm curves of the catalysts. c Raman spectra of the catalysts. d Atomic force microscopy (AFM) imaging of Co1+Con/N-C catalyst.

The ID/IG ratio increased from 0.43 in Co1/N-C to 0.48 in Con/N-C, and further to 0.58 in Co1+Con/N-C, indicating a higher defect density in the Co1+Con/N-C catalyst, likely due to the pyrolysis process (Fig. 1c). Atomic force microscopy (AFM) imaging of the Co1+Con/N-C catalyst (Fig. 1d) revealed that a 5 nm thick film exhibited a continuous and uniform structure, with a sparse distribution of nanoscale particles on its surface. Consistent with the X-ray diffraction (XRD) results, transmission electron microscopy (TEM) images (Fig. S1) did not reveal any Co nanoparticles or clusters in Co1/N-C catalyst, indicating that the cobalt species are likely highly dispersed as single atoms within the Co1/N-C catalyst.

X-ray photoelectron spectroscopy (XPS) spectra of the three catalysts confirmed the presence of Co, N, and C elements. With increasing pyrolysis temperature, the total nitrogen content in the samples decreased, as nitrogen species are unstable and can decompose into NH3 and other compounds at elevated temperatures. As illustrated in Fig. S3b, the N 1 s spectra can be deconvoluted into four characteristic nitrogen peaks corresponding to pyridinic N (398.4 eV), pyrrolic N (401.1 eV), graphitic N (402.3 eV), and Co–N (399.2 eV). The presence of Co–N bonds across all samples suggests that a portion of the nitrogen atoms is coordinated to single-cobalt sites. An increase in pyrolysis temperature significantly diminishes the content of Co–N species (Fig. S2b), from 31.6% to 13.4%, indicating a potential alteration in the Co–N coordination number and the partial formation of Co-Co bonds. The Co 2p spectra (Fig. S3c) can be deconvoluted into two peaks at 780.8 eV (Co2+ species) and 796.1 eV (Co0 species), which are associated with Co-N and Co nanoparticles, respectively. The Co–N peaks exhibit a positive shift in Con/N-C compared to Co1/N-C and Co1+Con/N-C, attributed to the reduced number of N-C supports. The predominance of Co-N species is further corroborated by the EXAFS experiments (Table S2). This Co²⁺ species, strongly coordinated with four nitrogen atoms within the graphitic layer, demonstrates robustness against aggregation. In addition to nitrogen, cobalt, carbon and residual magnesium (0.06 atomic percent) were also detected on the catalyst surface via X-ray photoelectron spectroscopy (XPS).

The transmission electron microscopy (TEM) and aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (AC-HAADF-STEM) images are depicted in Fig. 2 and S1. The TEM images confirm the presence of Co nanoparticles in the Con/N-C catalyst (Fig. S2c). In contrast, no Co nanoparticles or clusters were detected in the AC-STEM images of the Co1/N-C catalyst (Fig. 2a). A minority of the Co species in the Co1+Con/N-C catalyst existed as ensemble Co clusters (indicated by the blue rectangle) and single Co atoms (indicated by the red circle). The electron microscopy results indicate the coexistence of Co single atoms and Co nanoparticles in the Co1+Con/N-C catalyst (Fig. 2b). Subsequently, synchrotron X-ray absorption near-edge spectroscopy (XANES) was employed to elucidate the electronic structure and local coordination environment of the Co species in the Co1/N-C and Co1+Con/N-C catalysts (Fig. 2c). The XANES results revealed the Co K-edge absorption spectra of Co1/N-C and Co1+Con/N-C, referenced against Co foil and Co3O4, demonstrating that the valence states of Co species in the two catalysts ranged between 0 and +2. This finding is consistent with the Co 2p XPS results. The Fourier-transformed Co K-edge EXAFS analysis revealed a peak at 0.19 nm, corresponding to the Co-N shell in both Co1/N-C and Co1+Con/N-C samples. Additionally, a peak at 0.24 nm, indicative of Co-Co bonding, was observed, suggesting the formation of single Co clusters in the Co1+Con/N-C catalyst (Fig. 2d–g). The coordination numbers for Co-N were determined to be 4 and 2.6, and for Co-Co, 0 and 2.6, in Co1/N-C and Co1+Con/N-C catalysts, respectively (Table S2). These findings suggest that atomically dispersed Co sites and Co nanoparticles are effectively anchored on N-C carriers in the Co1+Con/N-C catalyst, whereas only Co single sites are present in the Co1/N-C catalyst, and only Co nanoparticles are present in the Con/N-C catalyst. These observations indicate that the three investigated catalysts possess distinct chemical environments and electronic structures, which are likely to influence their catalytic performance in the transfer hydrodeoxygenation of vanillin.

Fig. 2. Atomic-scale structure and chemical state analysis of the catalysts.

Fig. 2

Open in a new tab

Atomic-resolution STEM images of a Co1/N-C and b Co1+Con/N-C. c Co K-edge XANES of Co-based catalyst. WT-EXAFS for K-edge for d Co foil, e Co3O4, f Co1/N-C, and g Co1+Con/N-C.

Investigation the catalytic transfer hydrodeoxygenation (CTH) performance over Co1/N-C, Co1+Con/N-C, and Con/N-C catalysts

We assessed the catalytic transfer hydrodeoxygenation (CTH) activities of vanillin to MMP using formic acid as a hydrogen source. The control experiment demonstrated that the reaction did not proceed at 160 °C in the absence of a catalyst. Catalysts such as Mn1/N-C, Cu1/N-C, Ni1/N-C, and Fe1/N-C exhibited lower vanillin conversions, all below 18%. In contrast, the Co1/N-C, Con/N-C, and Co1+Con/N-C catalysts significantly enhanced the reaction, achieving vanillin conversions of 52.5%, 100%, and 80%, respectively, with selectivity towards MMP exceeding 99% (Fig. 3a). Additionally, the impact of reaction solvents on vanillin conversion and product selectivity was investigated over Co1+Con/N-C catalyst. Lower conversions were observed in aprotic and weakly polar solvents (Fig. 3b). Notably, when the solvent was changed from water to methanol, no conversion was observed. The high conversions in water can be attributed to the water stabilizes intermediates through hydrogen bonding, which facilitates the hydrodeoxygenation of vanillin.

Fig. 3. Evaluation of catalytic performance for vanillin hydrodeoxygenation.

Fig. 3

Open in a new tab

Catalytic transfer hydrodeoxygenation (CTH) activities of vanillin to MMP (2-methoxy-4-methylphenol) with formic acid and water over various reaction conditions: a different catalysts, b reaction solvent on vanillin conversion (A) and MMP selectivity (B).

To elucidate the interrelationship between reaction time, reaction temperature, catalyst quantity, and formic acid concentration, the catalytic performance of the Co1+Con/N-C catalyst was systematically examined (Fig. S4). The CTH of vanillin was conducted over a reaction time from 1 to 6 hours at a temperature of 160 °C, utilizing 175 µL of formic acid. As illustrated in Fig. S4a, the conversion of vanillin increased from 26% to nearly 100% as the reaction duration was extended from 1 to 4 h. Upon extending the reaction time to 6 h, no significant changes in conversion or selectivity were observed. When the Co1+Con/N-C catalyst quantity was below 30 mg, the conversion of vanillin remained under 34% (Fig. S4b). Conversely, nearly complete conversion of vanillin was achieved with a catalyst amount of 50 mg. A reduction in reaction temperature to 150 °C resulted in a 41% decrease in vanillin conversion, whereas an increase in temperature to 165 °C did not affect the conversion or selectivity of MMP (Fig. S4c). The influence of varying formic acid concentrations on the CTH of vanillin was also investigated, with the findings presented in Fig. S4d. When the amount of formic acid added was 87.5 µL, the conversion rate of vanillin was only 47%. An increase in formic acid amount from 175 µL to 350 µL resulted in a decline in vanillin conversion from 100 to 86%, and further to 63%. This suggests that excessive formic acid addition inhibits the reaction progression, potentially due to alterations in the system’s pH values.

Subsequently, we examined the impact of pH on the CTH activity of vanillin (Fig. 4). At lower pH levels (below pH value 1.27), vanillin conversion decreased as the concentration of H+ ions increased. This phenomenon may be attributed to the competitive interaction between formic acid decomposition and vanillin hydrodeoxygenation at the catalyst’s active sites, leading to diminished reaction activity. Conversely, under high pH conditions, the reaction demonstrated reduced activity, likely due to the low concentration of H+ ions, which is unfavorable for proton transfer in aqueous media (The pKa value of HCOOH is 3.75). This, in turn, affects the dehydrogenation of formic acid, thereby impeding the hydrodeoxygenation of vanillin. Consequently, optimal vanillin conversion is achieved with an appropriate concentration of H+ ions (pH value 1.27). The H+ ions can associate with water molecules to form hydrated protons, which facilitate proton hopping and hydrogen spillover (The active H+ species participated in the reduction of WO3 via hydrogen spillover, Fig. S5). These processes enable the hydrated protons to attack the carbonyl functional groups (C = O) of vanillin, leading to the formation of a stable intermediate species and accelerating the transfer hydrodeoxygenation of vanillin using formic acid.

Fig. 4. Effect of pH.

Fig. 4

Open in a new tab

Catalytic transfer hydrodeoxygenation of vanillin to MMP with formic acid and water over Co1+Con/N-C under different pH values.

The Arrhenius plots for the transfer hydrodeoxygenation of vanillin and vanillin alcohol over Co1/N-C, Con/N-C and Co1+Con/N-C catalysts were presented in Fig. 5. The apparent activation energy for the transfer hydrodeoxygenation of vanillin to MMP (Fig. 5a) using Co1+Con/N-C (61.7 kJ/mol) was lower than that observed for Co1/N-C (89.3 kJ/mol) and Con/N-C (84.0 kJ/mol). Similarly, the apparent activation energy for the transfer hydrodeoxygenation of HMP to MMP (Fig. 5b) with Co1+Con/N-C (36.5 kJ/mol) was lower than that for Co1/N-C (47.3 kJ/mol) and Con/N-C (57.1 kJ/mol). These findings indicate that the Co1+Con/N-C catalyst exhibits superior performance compared to Co1/N-C and Con/N-C catalysts, as demonstrated by the reduced energy barriers for the conversion of vanillin to MMP and HMP to MMP. Furthermore, analysis of the products from isotope experiments, supported by MS (Fig. 5d) and HNMR (Fig. 5e, f) results, confirmed the presence of MMP. The differential performance observed among the catalysts can be attributed to the synergistic interaction between single-atom and nanoparticles, which significantly influences catalytic activity.

Fig. 5. Dynamic Experiments and Result Analysis.

Fig. 5

Open in a new tab

a CTH activities of vanillin to MMP. Arrhenius plots of CTH of b vanillin and c HMP on Co1/N-C, Co1+Con/N-C, and Con/N-C catalysts. d MS profile of CTH of vanillin with formic acid and deuterated water. NMR spectrum profile of CTH of vanillin with formic acid and e water and f deuterated water.

The aforementioned results underscore the superior catalytic performance of Co1+Con/N-C in CTH of vanillin. However, the roles of formic acid and water in this process remain unclear. To elucidate the reaction pathway, isotope labeling experiments were conducted using the Co1+Con/N-C catalyst. As depicted in Fig. 6, various isotopic combinations of formic acid and water were employed in these experiments. A significant isotopic effect was observed during the hydrolysis of vanillin with formic acid (HCOOH) in deuterated water (D2O), yielding a kinetic isotope effect (KIE) of 2.16 (Fig. 6a). The use of deuterated water appears to decrease the rate of protonation. Furthermore, a significant isotopic effect was observed during the hydrolysis of vanillin using DCOOH and HCOOH as hydrogen donors in water, with a KIE of 2.26 (Fig. 6b). The transfer of the hydride ion (H−) from formic acid is identified as a crucial step in the reaction mechanism. Formic acid serves as a hydrogen donor by releasing hydride ions through decarboxylation, which subsequently attack the carbonyl carbon of the oxonium ion, resulting in the formation of an alcohol intermediate. The use of DCOOH, which impedes the transfer of H−, consequently reduces the reaction rate. Additionally, a pronounced isotopic effect was observed during the hydrolysis of vanillin in D2O and DCOOH, with a KIE of 2.49 (Fig. 6c). This value is lower than the individual effects (2.16 × 2.26 = 4.88), suggesting a synergistic mechanism. According to the Swain−Schaad relationship38, assuming a synergistic double hydrogen transfer mechanism, the theoretical KIE should correspond to the combined effects of D2O and DCOOH. The experimentally observed KIE (2.49) slightly surpasses the calculated value (KIED2O×KIEDCOOH = 2.21), which may be attributed to variations in acidity in D2O. Therefore, the activation of water and the dehydrogenation of formic acid are proposed as the rate-determining steps, distinguishing this mechanism from the typical of conventional transfer hydrodeoxygenation reactions. Water participated the kinetic steps of formic acid activation, reduces the activation barrier for carbonyl hydrogenation through proton hopping and hydrogen spill over via a hydrogen-bonded water network.

Fig. 6. Isotope Experiments.

Fig. 6

Open in a new tab

Isotopic effect of the catalytic transfer hydrodeoxygenation (CTH) of vanillin with various conditions, a H2O + HCOOH, and D2O + HCOOH, b H2O + HCOOH, and H2O + DCOOH, and c H2O + HCOOH, and D2O + DCOOH.

The potassium thiocyanate (KSCN) poisoning experiments were conducted to assesses the active sites of three distinct catalysts (Fig. 7a). The results from the KSCN poisoning tests indicated substantial deactivation across all three catalysts. Specifically, the introduction of 2 equiv. of KSCN resulted in an 86% reduction in activity for the Co1/N-C catalyst, while the Co1+Con/N-C and Con/N-C catalysts experienced activity losses of 42% and 8%, respectively. To further investigates the electronic and steric hindrance effect of various para-substituted benzaldehyde on the catalytic transfer hydrodeoxygenation using formic acid and water, a distinct volcano-type relationship was observed between the logarithm of the conversion rate (log (Conv. X/Conv. H). The results indicated that substrates with both electron-donating and electron-withdrawing para-substituents exhibited lower activity compared to unsubstituted benzaldehyde, likely due to the steric hindrance effects of the substituents (Fig. 7b). Building on this foundation, catalyst stability was evaluated through hot filtration tests (Fig. 7c) and recycling experiments (Fig. 7d). After the catalyst was removed following 30 minutes of reaction, the subsequent reaction progress was monitored under catalyst-free conditions. The hot filtration results showed negligible change in vanillin conversion after catalyst separation. Notably, after five consecutive reaction cycles, vanillin conversion remained at 95% with MMP selectivity exceeding 99%, demonstrating the catalyst’s exceptional reusability and stability in the catalytic transfer hydrogenation of vanillin with formic acid and water.

Fig. 7. Stability Experiments and Related analysis.

Fig. 7

Open in a new tab

a Potassium thiocyanate poisoning experiments over these Co-based catalysts. b Transfer hydrodeoxygenation of other aldehyde compounds. c Hot filtration and d recycling tests for transfer hydrodeoxygenation of vanillin over Co1+Con/N-C catalyst.

Conclusion

In summary, synergistic combination of cobalt single atoms and nanoparticles catalysts were successfully synthesized using a sacrificial support method. These catalysts were employed for the highly efficient CTH of vanillin to MMP, utilizing formic acid and water as the hydrogen source. The synergistic interactions between cobalt single-atom and nanoparticle sites, as indicated by shifts in Co 2p binding energy and reduced Co-N coordination. The mesoporous structure and high defect density of the catalyst enhanced mass transport and accessibility to active sites. The Co1+Con/N-C catalyst exhibited an excellent performance, achieving 99% conversion of vanillin and over 99% selectivity for MMP at 160 °C. The apparent activation energy for the conversion of vanillin to MMP over Co1+Con/N-C was determined to be 61.7 kJ/mol, representing a reduction of over 30% compared to Co1/N-C (89.3 kJ/mol) and Con/N-C (84.0 kJ/mol). The apparent activation energy for the conversion of HMP to MMP using Co1+Con/N-C (36.5 kJ/mol) is lower than that observed for Co1/N-C (47.3 kJ/mol) and Con/N-C (57.1 kJ/mol). Optimization of formic acid concentration and pH indicates that a moderate concentration of H⁺ (pH = 1.27) effectively balances formic acid dehydrogenation and substrate activation. KSCN poisoning experiments confirm the critical role of cobalt single-atom sites in facilitating reactivity, as demonstrated by a 92% reduction in activity for Co1/N-C. Isotopic labeling experiments reveal a dual hydrogen transfer pathway mediated by water, where hydronium ion (H3O⁺) attacks the carbonyl group, forming an oxonium ion intermediate (RCH(OH)⁺), followed by nucleophilic addition of formate-derived hydride ions (H⁻). Hot filtration and recycling tests verify the stability of the Co1+Con catalyst in CTH of vanillin to MMP. Additionally, substrate scope studies demonstrate the catalyst’s versatility in converting biomass-derived aromatic compounds. This study not only introduces an innovative paradigm for the design of non-noble metal single atom and nanoparticle synergistic catalysts but also elucidates the fundamental cooperative mechanism between water and formic acid.

Methods

The preparation and characterization of Co-based catalysts and the performances are detailed in the Supplementary information.

Supplementary information

Transparent Peer Review file (1.5MB, pdf)
Supplementary Information (1.8MB, pdf)
42004_2026_1947_MOESM3_ESM.pdf (87.2KB, pdf)

Description of Additional Supplementary Files

Supplementary Data 1 (9.7KB, rar)
Supplementary Data 2 (60.7KB, rar)
Supplementary Data 3 (27.3KB, rar)
Supplementary Data 4 (7.5KB, rar)
Supplementary Data 5 (1.2MB, rar)

Acknowledgements

This work was supported by the National Natural Science Foundation of China (32402907, 21908183), Scientific and Technological Innovation Team for Qinghai-Tibetan Plateau Research in Southwest Minzu University (Grant No.2024CXTD15), and ‘Weizhou Team’ Special Fund Project in Southwest Minzu University. The authors would like to thank Jin Dong (www.sci-go.com) for the XPS and AC-TEM measurements.

Author contributions

J.L.: investigation, formal analysis; G.S.: data curation, investigation, formal analysis; Z.X.: Conceptualization, Software; L.W.: visualization, writing, funding acquisition, supervision; M.R.: resources; T.D.: conceptualization, funding acquisition, resources, supervision, writing - review & editing.

Peer review

Peer review information

. Communications Chemistry thanks Botao Qiao, Dingsheng Wang and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The data supporting the findings of this study are available within the article and its Supplementary Information files. The raw data of this study are available within the Supplementary Data 1-5. The following Supplementary Data files are provided. Supplementary Data 1: Source data for Fig. 1b. Supplementary Data 2: Source data for Fig. 1c. Supplementary Data 3: Source data for Fig. 2. Supplementary Data 4: Source data for Figs. 3–7, S4. Supplementary Data 5: Source data for Fig. 5e, f. All other relevant raw data can be obtained from the corresponding author upon request.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Jiayi Li, Guangling Shi.

Contributor Information

Lu Wang, Email: Luwangbest@163.com.

Tao Dai, Email: tdaicat@163.com.

Supplementary information

The online version contains supplementary material available at 10.1038/s42004-026-01947-2.

References

  • 1.Manikandan, S. et al. Recent development patterns, utilization and prospective of biofuel production: Emerging nanotechnological intervention for environmental sustainability – A review. Fuel314, 122757 (2022). [Google Scholar]
  • 2.Jeswani, H. K., Chilvers, A. & Azapagic, A. Environmental sustainability of biofuels: a review. Proc. R. Soc. A. 476. 10.1098/rspa.2020.0351 (2020). [DOI] [PMC free article] [PubMed]
  • 3.Zhang, Z., Song, J. & Han, B. Catalytic transformation of lignocellulose into chemicals and fuel products in ionic liquids. Chem. Rev.117, 6834–6880 (2017). [DOI] [PubMed] [Google Scholar]
  • 4.Abu-Omar, M. M. & Ford, P. C. The lignin challenge in catalytic conversion of biomass solids to chemicals and fuels. RSC Sustain.1, 1686–1703 (2023). [Google Scholar]
  • 5.Wang, H. et al. Effects of lignin structure on hydrodeoxygenation reactivity of pine wood lignin to valuable chemicals. ACS Sustain. Chem. Eng.5, 1824–1830 (2017). [Google Scholar]
  • 6.Diao, X. et al. Catalytic hydrodeoxygenation and C–C coupling of lignin and its derivatives into renewable jet-fuel-range cycloalkanes. Green. Chem.26, 11406–11426 (2024). [Google Scholar]
  • 7.Xiang, S. et al. A unique Co@CoO catalyst for hydrogenolysis of biomass-derived 5-hydroxymethylfurfural to 2,5-dimethylfuran. Nat. Commun.13, 3657. 10.1038/s41467-022-31362-9 (2022). [DOI] [PMC free article] [PubMed]
  • 8.Pang, T. et al. Non-carbonized Pd single-atom catalyst supported on lignin-functionalized phenolic resin for potent catalytic transfer hydrogenation of lignin-derived aldehydes. Angew. Chem. Int Ed.64, e202503195 (2025). [DOI] [PubMed] [Google Scholar]
  • 9.Song, J. et al. Gallium-based eutectic alloys as liquid electron donors to ruthenium single-atom catalysts for selective hydrogenation of vanillin. Angew. Chem. Int. Ed.64, 05073. 10.1002/anie.202505073 (2025). [DOI] [PubMed]
  • 10.Li, Z. et al. One-step synthesis of single palladium atoms in WO2.72 with high efficiency in chemoselective hydrodeoxygenation of vanillin. Appl. Catal. B Environ.298, 120535 (2021). [Google Scholar]
  • 11.Wang, K. et al. Enrichment of nanocavities in hollow sandwich catalysts for boosting hydrodeoxygenation of biomass-derived vanillin. Chem. Eng. J.505, 159749 (2025). [Google Scholar]
  • 12.Fan, R. et al. Highly dispersed copper nanoparticles supported on activated carbon as an efficient catalyst for selective reduction of vanillin. Small14, 01953 (2018). [DOI] [PubMed]
  • 13.Liu, J. et al. Hydrodeoxygenation of lignin-derived oxygenated aromatic compounds catalyzed by nano-floral hydrotalcite for the preparation of liquid-phase organic hydrogen carriers. Appl. Catal. B Environ. Energy371, 125285 (2025). [Google Scholar]
  • 14.Santos, J. L. et al. Hydrodeoxygenation of vanillin over noble metal catalyst supported on biochars: Part II: catalytic behaviour. Appl. Catal. B Environ.268, 118425 (2020). [Google Scholar]
  • 15.Li, X. et al. Efficient atomically dispersed co/n-c catalysts for formic acid dehydrogenation and transfer hydrodeoxygenation of vanillin. ChemSusChem17, e202300871 (2024). [DOI] [PubMed] [Google Scholar]
  • 16.Xue, Y. et al. Heteropolyacids promoted MOF-derived high-performance Co(H4)@C-HPW0.25 catalysts for catalytic transfer hydrogenation of vanillin in mild condition. Chem. Eng. J.478, 147456 (2023). [Google Scholar]
  • 17.Valentini, F., Marrocchi, A. & Vaccaro, L. Liquid organic hydrogen carriers (LOHCs) as H-Source for bio-derived fuels and additives production. Adv. Energy Mater.12, 2103362. 10.1002/aenm.202103362 (2022).
  • 18.Wu, P., Zhao, D., Lu, G. & Cai, C. Supported Pd–Au bimetallic nanoparticles as an efficient catalyst for the hydrodeoxygenation of vanillin with formic acid at room temperature. Green. Chem.24, 1096–1102 (2022). [Google Scholar]
  • 19.Xue, Z. et al. Fabrication modulation of lignin-derived carbon nanosphere supported Pd nanoparticle via lignin fractionation for improved catalytic performance in vanillin hydrodeoxygenation. Int. J. Biol. Macromol.258, 128963 (2024). [DOI] [PubMed] [Google Scholar]
  • 20.Zhao, Z. et al. Engineering the electronic structure of Pt for selective hydrogenation of vanillin to vanillyl alcohol and p-cresol. Chem. Eng. Sci.300, 120616 (2024). [Google Scholar]
  • 21.Liang, X., Fu, N., Yao, S., Li, Z. & Li, Y. The progress and outlook of metal single-atom-site catalysis. J. Am. Chem. Soc.144, 18155–18174 (2022). [DOI] [PubMed] [Google Scholar]
  • 22.Wang, A., Li, J. & Zhang, T. Heterogeneous single-atom catalysis. Nat. Rev. Chem.2, 65–81 (2018). [Google Scholar]
  • 23.Sun, K., Shan, H., Neumann, H., Lu, G. P. & Beller, M. Efficient iron single-atom catalysts for selective ammoxidation of alcohols to nitriles. Nat. Commun.13, 27014. 10.1038/s41467-022-29074-1 (2022). [DOI] [PMC free article] [PubMed]
  • 24.Hiragond, C. B., Powar, N. S., Lee, J. & In, S. Single-atom catalysts (SACs) for photocatalytic CO2 reduction with H2O: activity, product selectivity, stability, and surface chemistry. Small18, 2201428. 10.1002/smll.202201428 (2022). [DOI] [PubMed]
  • 25.Wang, X. et al. Phosphorus-doped single atom copper catalyst as a redox mediator in the cathodic reduction of quinazolinones. Angew. Chem. Int. Ed.64, 05085. 10.1002/anie.202505085 (2025). [DOI] [PubMed]
  • 26.Gan, D. et al. Atomically dispersed copper-zinc dual sites anchored on nitrogen-doped porous carbon toward peroxymonosulfate activation for degradation of various organic contaminants. J. Colloid Interface Sci.673, 756–764 (2024). [DOI] [PubMed] [Google Scholar]
  • 27.Xia, Y., Xia, L. & Lin, X. Laccase-based self-amplifying catalytic system enables efficient antibiotic degradation for sustainable environmental remediation. Adv. Sci.10, 00210. 10.1002/advs.202300210 (2023). [DOI] [PMC free article] [PubMed]
  • 28.Li, X., Yang, X., Huang, Y., Zhang, T. & Liu, B. Supported noble-metal single atoms for heterogeneous catalysis. Adv. Mater.31, 02031. 10.1002/adma.201902031 (2019). [DOI] [PubMed]
  • 29.Fan, H. et al. Improving the selectivity of hydrogenation and hydrodeoxygenation for vanillin by using vacancy-coupled Ru–N3 single atoms immobilized on defective boron nitride. J. Mater. Chem. A11, 17560–17569 (2023). [Google Scholar]
  • 30.Li, Z. et al. Emerging ultrahigh-density single-atom catalysts for versatile heterogeneous catalysis applications: redefinition, recent progress, and challenges. Small Structures3, 00041. 10.1002/sstr.202200041 (2022).
  • 31.Shi, Y. et al. Combination of nanoparticles with single-metal sites synergistically boosts co-catalyzed formic acid dehydrogenation. Nat. Commun.15, 52517. 10.1038/s41467-024-52517-w (2024). [DOI] [PMC free article] [PubMed]
  • 32.Gao, Y., Liu, B. & Wang, D. Microenvironment engineering of single/dual-atom catalysts for electrocatalytic application. Adv. Mater.35, 09654. 10.1002/adma.202209654 (2023). [DOI] [PubMed]
  • 33.Wu, X. et al. Engineering the coordination sphere of isolated active sites to explore the intrinsic activity in single-atom catalysts. Nano-Micro Lett. 13, 00668. 10.1007/s40820-021-00668-6 (2021). [DOI] [PMC free article] [PubMed]
  • 34.Chen, R., Chen, S., Wang, L. & Wang, D. Nanoscale metal particle modified single-atom catalyst: synthesis, characterization, and application. Adv. Mater.36, 04713. 10.1002/adma.202304713 (2024). [DOI] [PubMed]
  • 35.Kuai, L. et al. Titania supported synergistic palladium single atoms and nanoparticles for room temperature ketone and aldehydes hydrogenation. Nat. Commun.11, 13941. 10.1038/s41467-019-13941-5 (2020). [DOI] [PMC free article] [PubMed]
  • 36.Luo, G. et al. Advances of synergistic electrocatalysis between single atoms and nanoparticles/clusters. Nano-Micro Lett. 16, 01463 (2024). 10.1007/s40820-024-01463-9. [DOI] [PMC free article] [PubMed]
  • 37.Sun, Z. et al. Concerted catalysis of single atom and nanocluster enhances bio-ethanol activation and dehydrogenation. Nat. Commun.16, 59127. 10.1038/s41467-025-59127-0 (2025). [DOI] [PMC free article] [PubMed]
  • 38.Kohen, A. & Jensen, J. H. Boundary conditions for the Swain-Schaad relationship as a criterion for hydrogen tunneling. J. Am. Chem. Soc.124, 3858–3864 (2002). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Transparent Peer Review file (1.5MB, pdf)
Supplementary Information (1.8MB, pdf)
42004_2026_1947_MOESM3_ESM.pdf (87.2KB, pdf)

Description of Additional Supplementary Files

Supplementary Data 1 (9.7KB, rar)
Supplementary Data 2 (60.7KB, rar)
Supplementary Data 3 (27.3KB, rar)
Supplementary Data 4 (7.5KB, rar)
Supplementary Data 5 (1.2MB, rar)

Data Availability Statement

The data supporting the findings of this study are available within the article and its Supplementary Information files. The raw data of this study are available within the Supplementary Data 1-5. The following Supplementary Data files are provided. Supplementary Data 1: Source data for Fig. 1b. Supplementary Data 2: Source data for Fig. 1c. Supplementary Data 3: Source data for Fig. 2. Supplementary Data 4: Source data for Figs. 3–7, S4. Supplementary Data 5: Source data for Fig. 5e, f. All other relevant raw data can be obtained from the corresponding author upon request.

Articles from Communications Chemistry are provided here courtesy of Nature Publishing Group

RESOURCES