English

• In recent years, ferrocene has been used as a functional ligand to modify various photoelectric materials due to its inherent good photoelectric response, excellent redox properties, and superior multi-electron transfer capabilities [1,2]. The modified compounds typically exhibit a variety of structural types, extended light absorption, and rapid charge transfer rates [3,4], which can be applied in photocatalytic [5-7] electrocatalytic [8,9], and photo-electrocatalytic reactions [10], showing excellent catalytic activity. Research shows that the primary reason for the improved performance of these materials is the strong metal-ligand charge transfer effect between the metal atoms and ligands after coordination with ferrocene [11-13]. However, to our knowledge, the reported ferrocene-modified photocatalytic materials exhibit limited or negligible light absorption in the near-infrared (NIR) region, which greatly limits the utilization of light in photocatalytic reactions.

  NIR light constitutes about 50% of the sunlight, which is comparable to the ultraviolet-visible (UV–vis) range [14]. Due to the unique advantages of NIR light such as excellent penetration ability and low phototoxicity [15-18], it has a broad range of applications in biomedicine [17], spectroscopy [19], photodetectors [20], and photocatalysis [21-27]. Especially in the field of photocatalysis, effectively harnessing the NIR region can maximize the utilization of solar energy and improve light efficiency [22,23,25]. Additionally, the lower energy of NIR light minimizes disturbances to the structure of photocatalysts and the reaction medium in the reaction system [17]. However, precisely because of the low energy of NIR light, it cannot provide the energy required to excite most photocatalysts, which greatly limits its application in photocatalytic reactions [28]. Based on the reported ferrocene-modified catalysts, we found that the type of metal coordinated with ferrocene in the structure of the catalyst can significantly influence the optical properties of the modified materials. We speculate that by selecting the appropriate coordinating metal atoms, the light absorption of ferrocene-modified materials can not only be extended to the NIR region but also combined with the rapid multi-electron transfer characteristics of ferrocene itself, thus effectively improving the photocatalytic reaction activity. However, this understanding necessitates the modeling of catalysts for further validation.

  Based on the aforementioned considerations, we designed and synthesized two isomorphic iron-oxo clusters modified with 1, 1-ferrocene dicarboxylic acid (Fcdc) through a simple solvothermal method, named Fe₁₁-Fcdc (Fe₁₁(µ₄-O)₄(µ-O)(Fcdc)₈(µ₂-OCH₃)₃·DMF), and BiFe₁₀-Fcdc (BiFe₁₀(µ₄-O)₆(µ₂-O)(µ-OH)(Fcdc)₈(µ₃-OCH₃)₂·H₂O). BiFe₁₀-Fcdc was synthesized through hetero-metal doping, which involves replacing one iron atom in the structure of Fe₁₁-Fcdc with a bismuth (Bi) atom. Experiments show that these two clusters exhibit strong light absorption in the NIR region, along with good capabilities for photo-generated charge transfer and separation. It is worth noting that Fe₁₁-Fcdc shows markedly higher absorption than BiFe₁₀-Fcdc in the NIR region. Based on these, the oxidative coupling reaction of benzylamine (BA) was employed as a model reaction to investigate the photocatalytic performance of the two clusters. The results show that both photocatalysts exhibit high catalytic activity with a high conversion of substrate (> 94%) under full-spectrum light. Notably, under NIR light, the catalytic performance of Fe₁₁-Fcdc (82.3%) is better than that of BiFe₁₀-Fcdc (41.0%), which may be attributed to the stronger metal-ligand charge transfer effect between the metal atoms and ligands in Fe₁₁-Fcdc. Importantly, this is the first report on the use of ferrocene-modified photocatalytic materials to catalyze the conversion of organic molecules in the NIR region.

  The black Fe₁₁-Fcdc and BiFe₁₀-Fcdc crystals were synthesized by the one-step solvothermal synthesis method (Figs. S1 and S2 in Supporting information). Single-crystal X-ray diffraction (SCXRD) analysis reveals that Fe₁₁-Fcdc and BiFe₁₀-Fcdc both crystallize in the triclinic space group P 1 ¯ (Table S1 in Supporting information). The molecule of Fe₁₁-Fcdc consists of an undeca-nuclear Fe-oxo core, eight Fcdc ligands, three µ₂–OCH₃, one terminal-coordinated N,N-dimethylformamide (DMF) molecule, and a free DMF molecule present (Fig. 1 and Fig. S3 in Supporting information). The Fe atoms in the undeca-nuclear Fe-oxo core are connected by five µ₄-O, one µ₃-O and three µ₂-OCH₃ (Fig. 1a). As shown in the Fig. S3a, except for the penta-coordinated Fe9, Fe10, and Fe11, the rest of the Fe atoms are hexa-coordinated. The structure of BiFe₁₀-Fcdc is essentially similar to that of Fe₁₁-Fcdc (Fig. 1b), but the difference is that Bi11 replaces the original position of Fe11, which is coordinated with eight O atoms to form a dodecahedral structure (Figs. 1c and d, Figs. S3a and S4a in Supporting information). At the same time, the Bi atom is connected to Fe atoms through four µ₄-O, one µ₂-O and two µ₃-OCH₃. Additionally, Fe4 and Fe5 are connected by a µ₂-O, and the terminal O atom coordinated to Fe1 and Fe8 come from hydroxide and water molecules, respectively (Fig. S4a). It is worth noting that the terminal-coordinated solvent molecule in BiFe₁₀-Fcdc and Fe₁₁-Fcdc can easily dissociate, potentially becoming catalytically active sites for absorbing activated small molecules during the catalytic process (Figs. 1e and f).

  Figure 1

  

  Figure 1.  Molecular structures of Fe₁₁-Fcdc and BiFe₁₀-Fcdc. (a, b) The structures of two Fe-oxo cores. (c, d) The polyhedron structures of two Fe-oxo cores. (e, f) The entire structures of Fe₁₁-Fcdc and BiFe₁₀-Fcdc. All hydrogen atoms are omitted for clarity.

  DownLoad: Full-Size Img PowerPoint

  As shown in Figs. S5 and S6 (Supporting information), the powder X-ray diffraction (PXRD) patterns are highly consistent with the simulated patterns derived from SCXRD data, which indicates that Fe₁₁-Fcdc and BiFe₁₀-Fcdc possess high crystallinity and purity. Subsequently, the solvent stability of the two clusters was assessed. After both materials were soaked in CH₃CN for 24 h, the PXRD patterns still matched well with the experimentally obtained PXRD patterns, indicating that these two clusters have good stability in CH₃CN, which provides the possibility for heterogeneous catalysis (Figs. S7 and S8 in Supporting information). The Fourier-transform infrared spectroscopy (FT-IR) spectrum shows distinct peaks at 1489, 915, and 513 cm^-1, corresponding to the δ(C=C), ν(O-H), and ν(Fe-C) vibrations of Fcdc [29], respectively, which confirms the successful modification of Fcdc into structures of the two clusters (Figs. S9 and S10 in Supporting information). Additionally, thermogravimetric analysis (TGA) was performed to evaluate the thermal stabilities of the two clusters. As shown in Figs. S11 and S12 (Supporting information), both Fe₁₁-Fcdc and BiFe₁₀-Fcdc can maintain their basic structural frameworks before ca. 250 ℃ under an N₂ atmosphere, which provides a good foundation for further performance applications.

  As the photosensitivity of ferrocene, the UV–vis diffuse reflectance spectroscopy was conducted to evaluate the light absorption capabilities of Fe₁₁-Fcdc and BiFe₁₀-Fcdc. Fig. 2a shows that both Fe₁₁-Fcdc and BiFe₁₀-Fcdc possess strong light absorption capabilities and a wide absorption range, which can be extended into the NIR region. Significantly, Fe₁₁-Fcdc exhibits superior light absorption ability in the NIR region. Based on the UV–vis diffuse reflectance spectra, the optical band gaps (E_g) of Fe₁₁-Fcdc and BiFe₁₀-Fcdc were further calculated using the Kubelka-Munk function, which are about 1.98 and 2.57 eV, respectively (Fig. 2b). Subsequently, to evaluate the band structure of the compounds, the Mott-Schottky electrochemical measurements were conducted to determine their lowest unoccupied molecular orbital (LUMO). As shown in Figs. S13 and S14 (Supporting information), the Mott-Schottky plots of Fe₁₁-Fcdc and BiFe₁₀-Fcdc exhibit a positive slope, indicating that they exhibit characteristics similar to n-type semiconductors. Thus, their flat band potential (E_fb) is −0.58 V and −0.41 V vs. Ag/AgCl, respectively, and the LUMO positions of Fe₁₁-Fcdc and BiFe₁₀-Fcdc are estimated to be −0.38 V and −0.21 V vs. NHE, respectively, due to the bottom of LUMO nearly equals to the E_fb in the n-type semiconductors [30]. Based on the equation E_HOMO = E_LUMO − E_g [31,32], the corresponding highest occupied molecular orbital (HOMO) positions of Fe₁₁-Fcdc and BiFe₁₀-Fcdc are ascertained to be 1.60 V and 2.36 V vs. NHE, respectively, which is consistent with the results of ultraviolet photoelectron spectroscopy (UPS) (Figs. 2c and d). Obviously, these two clusters have more positive HOMO levels, which makes them suitable for applications in certain oxidation reactions, such as the oxidation of amines (Fig. 2e). The photocurrent response was subsequently used to characterize the separation efficiency of photo-induced carriers. As shown in Fig. 2f, the photocurrent response intensity of Fe₁₁-Fcdc is significantly stronger than that of BiFe₁₀-Fcdc, indicating that Fe₁₁-Fcdc has a higher electron migration efficiency and would provide more available surface carriers in the photocatalytic process [33-35].

  Figure 2

  

  Figure 2.  (a) UV-visible diffuse reflectance spectra of Fe₁₁-Fcdc and BiFe₁₀-Fcdc. (b) Tauc plots of Fe₁₁-Fcdc and BiFe₁₀-Fcdc. (c) UPS spectra of Fe₁₁-Fcdc. (d) UPS spectra of BiFe₁₀-Fcdc. (e) The energy band structures diagram for Fe₁₁-Fcdc and BiFe₁₀-Fcdc. (f) Transient photocurrent curves of Fe₁₁-Fcdc and BiFe₁₀-Fcdc measured in 0.5 mol/L Na₂SO₄ aqueous solution.

  DownLoad: Full-Size Img PowerPoint

  Based on the aforementioned excellent photophysical properties, the performance of photocatalytic oxidative coupling reaction of BA with the two clusters was investigated, respectively. The qualitative analysis of products was determined by gas chromatography-mass spectrometry (GC-MS), and the conversion of substrate and selectivity of products were determined by gas chromatography (GC) (Figs. S15-S17 in Supporting information). As shown in Fig. 3a and Fig. S18 (Supporting information), under full-spectrum light irradiation for 18 h, both Fe₁₁-Fcdc and BiFe₁₀-Fcdc achieved high conversion of the substrate (97.0% and 97.7%) and selectivity of the product (90.3% and 91.5%). The exceptional light absorption capability of the two clusters in the NIR region prompts an investigation into their performance in catalyzing the oxidation of BA to N-benzylidenebenzylamine (BDA) under NIR light. The experimental results show that under NIR light, Fe₁₁-Fcdc can also demonstrate excellent performance, which can reach a conversion of 82.3% with a selectivity of 80.2%. However, under the same conditions, BiFe₁₀-Fcdc shows poor catalytic performance with a conversion of 41.0% and a selectivity of 73.4%. To explore the relationship between the conversion of substrate and time under NIR light conditions, the reaction solution was extracted and analyzed every 3 h (Fig. S19 in Supporting information). The results show that Fe₁₁-Fcdc, as a photocatalyst, has better reaction activity than BiFe₁₀-Fcdc under NIR light. This difference can be explained by the varying strengths of light absorption in the NIR region and the interfacial charge transfer rate of the two photocatalysts. Moreover, electrochemical impedance spectroscopy (EIS) was performed to investigate the impedance characteristics of these two photocatalysts. In Fig. S20 (Supporting information), Fe₁₁-Fcdc has a smaller electronic transmission resistance, indicating that it has a faster interfacial charge transport rate than BiFe₁₀-Fcdc, which is consistent with the results of the photocurrent response tests. To further investigate the utilization of full-spectrum light by the two clusters, the photocatalytic performance under UV–vis spectrum light was further evaluated. As displayed in Fig. 3a and Fig. S18, both clusters can achieve high conversion of the substrate (94.5% and 96.3%) and good selectivity of the product (89.2% and 90.7%) under UV–vis spectrum light to realize a great utilization of light.

  Figure 3

  

  Figure 3.  (a) Conversion of BA under light sources with different wavelengths. (b) Conversion of BA and selectivity of product for oxidative coupling reaction with Fe₁₁-Fcdc in 6 cycles experiments under full-spectrum light. (c) Control experiments of the oxidative coupling reaction of BA under NIR light. (d) EPR spectra of Fe₁₁-Fcdc under irradiated and in the dark in the presence of PBN.

  DownLoad: Full-Size Img PowerPoint

  Recycling experiments were conducted to demonstrate the tolerance of the photocatalysts in the reaction. After 6 cycles of experiments, Fe₁₁-Fcdc consistently maintains stable and efficient reaction activity (Fig. 3b), proving that Fe₁₁-Fcdc maintains good stability during the catalytic process. Besides, PXRD and FT-IR characterizations of tested photocatalysts were performed after the photocatalytic reaction to further verify the stability of photocatalysts. Figs. S21-S24 (Supporting information) show that the PXRD patterns and IR spectra of the catalyst after the reaction match well with that before the reaction, which confirms that Fe₁₁-Fcdc and BiFe₁₀-Fcdc maintain good structural integrity during the photocatalytic reaction. X-ray photoelectron spectroscopy (XPS) analysis confirms that the valence states of the metal elements remained consistent before and after the reaction (Figs. S25 and S26 in Supporting information). Subsequently, to determine the indispensable reaction conditions, a series of control experiments were conducted. As shown in Table S4 (Supporting information) and Fig. 3c, photocatalyst, light, and tert‑butyl hydroperoxide (TBHP) are the basic conditions for the oxidation coupling reaction of BA, otherwise, only a small or negligible amount of conversion can be observed.

  To study the mechanism of the reaction, electron paramagnetic resonance (EPR) tests were conducted at room temperature with N-tert-butyl-α-phenylnitrone (PBN) as the trapping agent to identify the reactive radicals involved in the reaction. The yielded clear EPR peaks for the PBN-spin adducts indicate that TBHP may generate tBuO^•/tBuOO^• radical during the reaction, which could be the active species in the reaction (Fig. 3d) [36-38]. We also conducted quenching experiments to distinguish the contributions of specific active species. As shown in Fig. S27 (Supporting information), the addition of p-benzoquinone (BQ) and 1, 4-diazabicyclo[2.2.2]octane (DABCO) results in only a marginal reduction in reaction conversion, ruling out the presence of O₂ ^•- and ¹O₂ radicals during the photocatalytic process. Subsequently, butylated hydroxytoluene (BHT) is used as a scavenger for tBuO^•/tBuOO^• radical and added to the reaction. The significant decrease in the conversion of benzylamine (BA) indicates that tBuO^•/tBuOO^• radical plays a crucial role in the amine coupling reaction. Moreover, holes are likely involved in the oxidative amine coupling process because the addition of KI as a hole scavenger during the photocatalytic reaction results in a reduced conversion to BDA (60.8%). Furthermore, when CCl₄ was introduced as an electron scavenger, the conversion of BA to BDA decreased to 51.5%, indicating that the transfer of photogenerated electrons plays a significant role in the reaction (Table S4).

  To gain a deeper understanding of the reaction mechanism involved in the photocatalytic conversion of BA to BDA, the in-situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) measurement was conducted (Fig. 4) [39,40]. Under a dark condition, no new IR band/peak is observed at 0 min after achieving adsorption equilibrium for 20 min, indicating the photocatalytic nature of the conversion of BA to BDA. Upon light stimulation, the new IR peak appears at 1670 cm^-1 attributed to the C=N stretching vibration of the benzyl imine intermediate. Over time, the product BDA generates, as evidenced by the appearance of the C=N stretching vibration (1652 cm^-1) for BDA. This phenomenon suggests that BDA can form via the pathway that the benzyl imine intermediate is nucleophilically attacked by BA molecules. Simultaneously, the peak at 1557 cm^-1 for N—H bending vibration gradually intensifies, indicating the formation of NH₃ alongside the production of BDA. Additionally, we observe the characteristic IR peak of the C=O bond at 1699 cm^-1, indicating the presence of another pathway involving a benzaldehyde intermediate for the photocatalytic conversion of BA to BDA in this photocatalytic system.

  Figure 4

  

  Figure 4.  In-situ DRIFTSR on cluster complexes under dark and irradiation conditions using BA as the reactants.

  DownLoad: Full-Size Img PowerPoint

  Based on all the foregoing discoveries, a possible mechanism for the oxidative coupling of BA is proposed. As shown in Fig. 5, upon light irradiation, Fe₁₁-Fcdc is photoexcited to generate electrons and holes. The adsorbed BA molecules are oxidized by photo-generated holes to form BA radical cations (Ⅰ), while TBHP reacts with photo-generated electrons to produce tBuO^•/tBuOO^•. In the process of generating the final coupling products from BA radical cations (Ⅰ), there are two distinct pathways [41-43]. In path A, intermediate Ⅰ reacts with tBuO^•/tBuOO^•, further transforming into intermediate Ⅱ. Then, Ⅱ can be readily attacked by another free BA molecule. After the release of ammonia, the final product BDA (Ⅴ) is produced. In path B, Ⅲ is hydrolyzed to obtain benzaldehyde (Ⅴ), which is then condensed with another BA to produce the coupling product

  Figure 5

  

  Figure 5.  Proposed reaction mechanisms for photocatalytic oxidative amine coupling reaction by Fe₁₁-Fcdc.

  DownLoad: Full-Size Img PowerPoint

  To demonstrate the generality of the reaction, the substrate scope was investigated (Scheme 1). Various benzylamine derivatives were tested as substrates under the same conditions. Benzylamines with different substituents (such as Me-, MeO-, F-, Cl-, and Br-) can be effectively converted into the desired product and the conversion of substrate follows the trend of electron-donating groups being superior to electron-withdrawing groups. When the substrates contain the electron-donating group, p-Me, and p-OMe, the conversion of amines can reach 80.6% and 86.2%, respectively. However, when the substrates contain the electron-withdrawing groups p-F, p-Cl, and p-Br, the conversion of amines is determined to be 75.7%, 72.6%, and 71.5%, respectively (Figs. S28-S33 in Supporting information). This may be due to the electron-withdrawing group, after resonance with the benzene ring, induces more positive charge on the carbon atom adjacent to the nitrogen atom in the amine, making it difficult for the hydrogen atom on the carbon to depart, thereby resulting in a decrease in conversion rate [44-47].

  Scheme 1

  

  Scheme 1.  Heterogeneous photocatalytic oxidative coupling of diverse amines. Reagents: substrate (0.1 mmol), photocatalyst (20.0 mg), solvent (1 mL CH₃CN), TBHP (0.27 mmol), 300 W Xenon lamp (200 mW/cm²) with a cutoff filter (> 800 nm).

  DownLoad: Full-Size Img PowerPoint

  In summary, we synthesized two stable iron-oxo clusters modified with ferrocene ligands (Fe₁₁-Fcdc and BiFe₁₀-Fcdc), which contain different metal atoms (Fe or Bi) at the center of the cluster and exhibit excellent full-spectrum utilization efficiency. When they were used as photocatalysts for photo-induced oxidative coupling reaction of BA, high conversion of 97% and great selectivity over 90% can be achieved under full-spectrum light. Especially, compared to BiFe₁₀-Fcdc, Fe₁₁-Fcdc demonstrates superior photocatalytic activity under NIR light, achieving a conversion of 82.3% and a product selectivity of 80.2%. The excellent photocatalytic performance under NIR may be attributed to the stronger metal-ligand charge transfer effect between the metal atoms and ligands in Fe₁₁-Fcdc. This study broadens the application of crystalline materials in photocatalytic reactions in the NIR region.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Li-Ling He: Writing – original draft, Investigation, Data curation. Sheng-Nan Sun: Writing – original draft, Project administration, Investigation. Jing-Wen Shi: Software, Investigation, Data curation. Jiang Liu: Writing – review & editing, Supervision, Funding acquisition. Ning Li: Writing – review & editing, Supervision, Funding acquisition.

  Acknowledgments

  This work was supported by the National Key R & D Program of China (No. 2023YFA1507201), the National Natural Science Foundation of China (No. 22201046), Guangzhou Basic and Applied Basic Research Fund Project (No. 2025A04J0092), Young Top Talents of Pearl River Talent Program of Guangdong Province (No. 2021QN02L617), Guangdong Basic and Applied Basic Research Foundation (No. 2023B1515120060).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111111.

  
  
• 1. [1]

     F. Jäkle, J.B. Sheridan, Angew. Chem. Int. Ed. 47 (2008) 7587-7587. doi: 10.1002/anie.200885616

  2. [2]

     M.G. Walawalkar, P. Pandey, R. Murugavel, Angew. Chem. Int. Ed. 60 (2021) 12632–12635. doi: 10.1002/anie.202101770

  3. [3]

     J. Liang, X. Gao, B. Guo, et al., Angew. Chem. Int. Ed. 60 (2021) 12770–12774. doi: 10.1002/anie.202101878

  4. [4]

     H. Zhang, S. Si, G. Zhai, et al., Appl. Catal. B: Environ. 337 (2023) 122909. doi: 10.1016/j.apcatb.2023.122909

  5. [5]

     L. Xia, W. Zhou, Y. Xu, et al., Chem. Eng. J. 451 (2023) 138747. doi: 10.1016/j.cej.2022.138747

  6. [6]

     M. Xu, D. Li, K. Sun, et al., Angew. Chem. Int. Ed. 60 (2021) 16372–16376. doi: 10.1002/anie.202104219

  7. [7]

     J.J. Liu, N. Li, J.W. Sun, et al., ACS Catal. 11 (2021) 4510–4519. doi: 10.1021/acscatal.0c04495

  8. [8]

     G. Deng, H. Yun, Y. Chen, et al., Angew. Chem. Int. Ed. (2024) e202418264.

  9. [9]

     Y. Tan, G. Sun, T. Jiang, et al., Angew. Chem. Int. Ed. 63 (2024) e202317471. doi: 10.1002/anie.202317471

  10. [10]

      J.J. Liu, S.N. Sun, J. Liu, et al., J. Am. Chem. Soc. 145 (2023) 6112–6122. doi: 10.1021/jacs.2c11509

  11. [11]

      H. Imahori, D.M. Guldi, K. Tamaki, et al., J. Am. Chem. Soc. 123 (2001) 6617–6628. doi: 10.1021/ja004123v

  12. [12]

      R.E. Lehmann, J.K. Kochi, J. Am. Chem. Soc. 113 (1991) 501–512. doi: 10.1021/ja00002a018

  13. [13]

      C. Du, J. Sheng, F. Zhong, et al., Proc. Natl. Acad. Sci. U. S. A. 121 (2024) e2315956121. doi: 10.1073/pnas.2315956121

  14. [14]

      W. Gao, B. Tian, W. Zhang, et al., Appl. Catal. B: Environ. 257 (2019) 117908. doi: 10.1016/j.apcatb.2019.117908

  15. [15]

      L. Li, J. Wang, Q. Zhang, et al., ACS Catal. 14 (2024) 3041–3048. doi: 10.1021/acscatal.3c05441

  16. [16]

      W. Liang, C. Nie, J. Du, et al., Nat. Photonics17 (2023) 346–353. doi: 10.1038/s41566-023-01156-6

  17. [17]

      K. Long, W. Lv, Z. Wang, et al., Nat. Commun. 14 (2023) 8112. doi: 10.1038/s41467-023-43805-y

  18. [18]

      X.D. Su, Q. Liu, X.N. Li, et al., Chem. Catal. 3 (2023) 100710.

  19. [19]

      A. Fassio, E.G. Fernández, E.A. Restaino, A. La Manna, D. Cozzolino, Comput. Electron. Agric. 67 (2009) 59–63. doi: 10.1016/j.compag.2009.03.001

  20. [20]

      C. Hu, L. Li, G. Shen, Chin. J. Chem. 39 (2021) 2141–2146. doi: 10.1002/cjoc.202100229

  21. [21]

      J. Li, C. Zhang, T. Bao, et al., Adv. Mater. 37 (2025) 2416210. doi: 10.1002/adma.202416210

  22. [22]

      X. Li, L. Li, G. Chen, et al., Nat. Commun. 14 (2023) 4034. doi: 10.1038/s41467-023-39666-0

  23. [23]

      Z. Lian, F. Wu, J. Zi, et al., J. Am. Chem. Soc. 145 (2023) 15482–15487. doi: 10.1021/jacs.3c03990

  24. [24]

      B. Su, Y. Kong, S. Wang, et al., J. Am. Chem. Soc. 145 (2023) 27415–27423. doi: 10.1021/jacs.3c08311

  25. [25]

      Y. Xu, M. Fan, W. Yang, et al., Adv. Mater. 33 (2021) 2101455. doi: 10.1002/adma.202101455

  26. [26]

      S. Yang, W.J. Byun, F. Zhao, et al., Adv. Mater. 36 (2024) 2312616. doi: 10.1002/adma.202312616

  27. [27]

      R. Feng, W. Lei, G. Liu, M. Liu, Adv. Mater. 30 (2018) 1804770. doi: 10.1002/adma.201804770

  28. [28]

      L. Wang, R. Sa, Y. Wei, et al., Angew. Chem. Int. Ed. 61 (2022) e202204561. doi: 10.1002/anie.202204561

  29. [29]

      T.N. Kumar, S. Sivabalan, N. Chandrasekaran, K.L.N. Phani, J. Mater. Chem. B2 (2014) 6081–6088. doi: 10.1039/C4TB00823E

  30. [30]

      R. Beranek, Adv. Phys. Chem. 2011 (2011) 786759. doi: 10.1155/2011/786759

  31. [31]

      L.G. Julius, D.T. Mapolelo, T.B. Demissie, et al., Appl. Organomet. Chem. 38 (2024) e7341. doi: 10.1002/aoc.7341

  32. [32]

      K. Wu, X.Y. Liu, P.W. Cheng, et al., J. Am. Chem. Soc. 145 (2023) 18931–18938. doi: 10.1021/jacs.3c05585

  33. [33]

      J. Chen, B. An, Y. Chen, et al., J. Am. Chem. Soc. 145 (2023) 19225–19231. doi: 10.1021/jacs.3c03943

  34. [34]

      G.C. Guo, J.P. Zhao, S. Guo, et al., Angew. Chem. Int. Ed. 63 (2024) e202402374. doi: 10.1002/anie.202402374

  35. [35]

      M. Zhang, P. Huang, J.P. Liao, et al., Angew. Chem. Int. Ed. 62 (2023) e202311999. doi: 10.1002/anie.202311999

  36. [36]

      S. Daliran, A. Santiago-Portillo, S. Navalón, et al., J. Colloid Interf. Sci. 532 (2018) 700–710. doi: 10.1016/j.jcis.2018.07.140

  37. [37]

      S. Jung, J. Kim, S. Hong, Adv. Synth. Catal. 359 (2017) 3945–3949. doi: 10.1002/adsc.201701072

  38. [38]

      Y. Wang, P. Li, J. Wang, et al., Catal. Sci. Technol. 11 (2021) 4429–4438. doi: 10.1039/d1cy00328c

  39. [39]

      W. Liu, Y. Wang, H. Huang, et al., J. Am. Chem. Soc. 145 (2023) 7181–7189. doi: 10.1021/jacs.2c12182

  40. [40]

      H. Chen, C. Liu, W. Guo, et al., Catal. Sci. Technol. 12 (2022) 1812–1823. doi: 10.1039/d1cy02344f

  41. [41]

      S. Li, L. Li, Y. Li, et al., ACS Catal. 10 (2020) 8717–8726. doi: 10.1021/acscatal.0c01242

  42. [42]

      C. Xu, H. Liu, D. Li, J.H. Su, H.L. Jiang, Chem. Sci. 9 (2018) 3152–3158. doi: 10.1039/c7sc05296k

  43. [43]

      F.J. Zhao, G. Zhang, Z. Ju, Y.X. Tan, D. Yuan, Inorg. Chem. 59 (2020) 3297–3303. doi: 10.1021/acs.inorgchem.9b03743

  44. [44]

      Z.L. Chen, M.H. Huang, J. Mater. Chem. A11 (2023) 22198–22205. doi: 10.1039/d3ta04844f

  45. [45]

      G. He, Y. Lai, Y. Guo, et al., ACS Appl. Mater. Interfaces14 (2022) 53724–53735. doi: 10.1021/acsami.2c14554

  46. [46]

      S. Liu, Q. Su, W. Qi, et al., Catal. Sci. Technol. 12 (2022) 2837–2845. doi: 10.1039/d2cy00167e

  47. [47]

      F. Su, S.C. Mathew, L. Möhlmann, et al., Angew. Chem. Int. Ed. 50 (2011) 657–660. doi: 10.1002/anie.201004365

• 

  Figure 1  Molecular structures of Fe₁₁-Fcdc and BiFe₁₀-Fcdc. (a, b) The structures of two Fe-oxo cores. (c, d) The polyhedron structures of two Fe-oxo cores. (e, f) The entire structures of Fe₁₁-Fcdc and BiFe₁₀-Fcdc. All hydrogen atoms are omitted for clarity.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) UV-visible diffuse reflectance spectra of Fe₁₁-Fcdc and BiFe₁₀-Fcdc. (b) Tauc plots of Fe₁₁-Fcdc and BiFe₁₀-Fcdc. (c) UPS spectra of Fe₁₁-Fcdc. (d) UPS spectra of BiFe₁₀-Fcdc. (e) The energy band structures diagram for Fe₁₁-Fcdc and BiFe₁₀-Fcdc. (f) Transient photocurrent curves of Fe₁₁-Fcdc and BiFe₁₀-Fcdc measured in 0.5 mol/L Na₂SO₄ aqueous solution.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Conversion of BA under light sources with different wavelengths. (b) Conversion of BA and selectivity of product for oxidative coupling reaction with Fe₁₁-Fcdc in 6 cycles experiments under full-spectrum light. (c) Control experiments of the oxidative coupling reaction of BA under NIR light. (d) EPR spectra of Fe₁₁-Fcdc under irradiated and in the dark in the presence of PBN.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  In-situ DRIFTSR on cluster complexes under dark and irradiation conditions using BA as the reactants.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Proposed reaction mechanisms for photocatalytic oxidative amine coupling reaction by Fe₁₁-Fcdc.

  下载: 全尺寸图片 幻灯片
  

  Scheme 1  Heterogeneous photocatalytic oxidative coupling of diverse amines. Reagents: substrate (0.1 mmol), photocatalyst (20.0 mg), solvent (1 mL CH₃CN), TBHP (0.27 mmol), 300 W Xenon lamp (200 mW/cm²) with a cutoff filter (> 800 nm).

  下载: 全尺寸图片 幻灯片
•