English

• Photocatalysis for the production of clean hydrogen (H₂) is emerging as a highly promising technology to counter the increasing energy crisis [1–5]. However, developing highly efficient and inexpensive photocatalysts that do not rely on noble metals and exhibit excellent H₂ production performance still poses significant challenges. In this context, metal nanoclusters (NCs) with well-defined compositions and structures have garnered attention in the field of energy conversion [6–9]. Atomically precise metal NCs stand out due to their high catalytic activity and their role as metal model catalysts, which can help reveal the structure-activity relationship. In particular, organic ligand-protected copper nanoclusters, characterized by their atomically accurate structure and uniform sub-nanometer size, have demonstrated catalytic performance in many reactions. These include photo/electrocatalytic CO₂ reduction [10–16], photocatalytic degradation of organic pollutants [17], and organic catalytic reactions [18,19]. Notably, previous research and theoretical calculations have shown that certain copper nanoclusters exhibit semiconductor-like properties, indicating their potential applications in photocatalytic H₂ evolution [20]. However, although the coordination of organic ligands stabilizes the metal cluster core, it also passivates the reactive metal centers, especially in the field of photocatalysis. The presence of a thick organic ligand layer is actually detrimental to the photoactivation process, as well as the separation of photogenerated charges and carrier transfer in the photocatalytic process. Therefore, the use of copper NCs as photocatalysts for photocatalytic H₂ production reactions is relatively limited in the current literature [21,22]. Especially, designing copper NCs as single-component photocatalysts capable of achieving rapid H₂ evolution remains a formidable challenge.

  Rhodamine B (RB) is a common dye molecule known for its strong visible light absorption and structural modifiability [23]. Incorporating well-established rhodamine-based linkers can lead to photoactive coordination polymers (CPs) that exhibit highly effective visible-light-driven production of hydrogen [24]. Despite its popularity and potential, rhodamine has not been fully exploited as a protective ligand for creating photoactive metal nanoclusters. Negatively charged alkynyl ligands exhibit a remarkable propensity for forming robust coordination interactions with metal ions as both σ and π bonds. In particular, the π bond introduces an additional layer of complexity and stability to the metal-ligand interaction. Attaching alkynyl groups to RB can create a photoactive rhodamine-based ligand. This ligand serves as a bulky alkyne substituent and offers various coordination modes, which not only helps stabilize metal clusters kinetically but also introduces unexpected and beneficial functional properties [25].

  Herein, we synthesized a propynyl-modified rhodamine molecule as a dye ligand, denoted as RB-CH (Scheme S1 in Supporting information). Its nuclear magnetic resonance (NMR) hydrogen spectrum is shown in Fig. S1 (Supporting information). And then we reacted it with Cu⁺ ions to form a novel 11-core copper(Ⅰ) alkynyl cluster of [Cu₁₁(RB-C)₉(Et₃N)(EtOH)]·PF₆ (denoted as Cu₁₁) by a facile one-pot synthesis technique (Scheme S2 in Supporting information). In this work, the bulky dye group RB was chosen for the first time, as it has not been previously employed in the alkynyl-protected coinage metal clusters. This combination of the photoactive rhodamine-derived ligand with Cu⁺ ions forms an atomically precise, photoresponsive copper(Ⅰ)-alkynyl nanocluster that efficiently photocatalyzes the reduction of protons into hydrogen. Furthermore, when the Cu₁₁ cluster was deposited onto the surface of TiO₂ nanosheets, an efficient composite semiconductor photocatalyst, Cu₁₁@TiO₂, was formed. The composite exhibits remarkably enhanced photocatalytic activity and outstanding stability during the photocatalytic H₂ evolution reactions.

  The single-crystal X-ray diffraction reveals that Cu₁₁ crystallizes in the trigonal system with space group of P3 (Table S1 in Supporting information). As shown in Fig. 1a, the cation part of Cu₁₁ is protected by nine RB-CH ligands, one triethylamine molecule and one ethanol molecule. Eleven copper atoms form a three-axis propeller configuration with Cu-Cu distances in the range of 2.4223(7)-2.7838(6) Å (Fig. 1b), indicating strong cuprophilic interactions. The nine RB-C⁻ ligands bound to the Cu₁₁ core adopt two types of coordination modes. Six RB-C⁻ ligands are coordinated to copper atoms by alkynyl carbon atoms and carbonyl oxygen atoms with Cu(Ⅰ)-O bond lengths of 2.098 and 2.228 Å. The remaining three RB-C⁻ ligands bind with four copper atoms only by alkynyl carbon atoms. Among these ligands, six alkynyl groups adopt a μ₄-η_σ¹, η_σ¹, η_π², η_π² ligation mode and the remaining three take a μ₃-η_σ¹, η_σ¹, η_π² mode (Figs. 1c–e), which relatively increases the symmetry of metal kernel, adopting a C3 symmetry viewed from the c-axis (Fig. 1b). The Cu-C distances fall in the ranges of 1.784(4)-2.438(5) and 2.125(5)-2.373(4) Å for σ- and π-type bonding, respectively. When using the alkynyl ligand bearing a bulky rhodamine group, steric hindrance effect provides the possibility for the coordination of solvent molecules. One triethylamine and one ethanol molecule are coordinated with two copper atoms in Cu₁₁ core through N and O atoms, respectively. The coordinated solvent molecules may detach from the surface of the copper clusters during catalysis, thereby exposing the metal site as a potential catalytically active center.

  Figure 1

  

  Figure 1.  (a) The overall structure of Cu₁₁. (b) The metal kernel in Cu₁₁ with C3 symmetry. (c-e) The coordination modes of RB-CH ligands in Cu₁₁ cluster. All hydrogen atoms are omitted for clarity, and the carbon atoms of the ethynide group are represented as small black spheres. Cu: brown, O: red, N: blue, C: gray.

  DownLoad: Full-Size Img PowerPoint

  The powder X-ray diffraction (PXRD) patterns confirmed the phase purity of Cu₁₁ (Fig. S2 in Supporting information). The disappearance of the alkyne hydrogen vibration signal in the infrared spectrum indicates that the alkyne group of the ligand has successfully participated in coordination (Fig. S3 in Supporting information). Thermogravimetric analysis indicates that Cu₁₁ clusters can maintain the thermal stability up to 297 ℃ (Fig. S4 in Supporting information). Such excellent thermal stability of Cu₁₁ cluster is very scarce. Up to now, there are no similar reports of atomically precise copper(Ⅰ) clusters in the literature. Analysis of Cu₁₁ by X-ray photoelectron spectroscopy (XPS) showed that the binding energies of the Cu 2p_1/2 and Cu 2p_3/2 levels were 952.7 and 932.8 eV, respectively (Fig. S5 in Supporting information), which suggests that the Cu element in Cu₁₁ cluster exists in the +1 valence state [26].

  To explore the catalytic activity, photocatalytic H₂ evolution experiments with Cu₁₁ were conducted. The optimum conditions of photocatalytic reaction were explored in detail. As shown in Fig. 2a, fluorescein (FL) shows better photocatalytic properties than Eosin Y. Therefore, FL was introduced as a photosensitizer in the photocatalytic system to combine with the rhodamine species of the ligand in Cu₁₁ for light harvesting. And triethylamine (TEA) exhibits a better hole capture capacity than triethanolamine (TEOA). The optimized system contained Cu₁₁ (1.5 mg), FL (10 mg) and triethylamine (TEA, 10% v/v) in a mixed solvent of CH₃COCH₃/H₂O (1:1) at pH 13 (Figs. S6 and S7 in Supporting information). Under optimized conditions, Cu₁₁ was highly active for hydrogen evolution, with an average rate of hydrogen production of 8.13 mmol g⁻¹h⁻¹, and a maximum of 48.8 mmol/g in the first 6 h of irradiation (Fig. S7). Furthermore, no H₂ was detected in the dark or in the absence of Cu₁₁ under the measured conditions, indicating that Cu₁₁ is essential for an efficient photochemical reaction. Notably, Cu₁₁ demonstrates activity even in the absence of FL, resulting in a detected H₂ yield of 3.27 µmol after 6 h of irradiation. After six consecutive cycles of catalytic testing, the photocatalytic hydrogen production performance of the Cu₁₁ exhibits no significant attenuation, indicating that the Cu₁₁ cluster possesses excellent catalytic cycle stability (Fig. 2b).

  Figure 2

  

  Figure 2.  (a) Effect of different photosensitizer and sacrificial agent in the photocatalytic production. (b) The cyclic test performance of Cu₁₁ cluster. (c) Transient photocurrent response of the bare glassy carbon electrode (black), bare glassy carbon electrode + FL (blue), pristine Cu₁₁ (red) and FL + Cu₁₁ (green). (d) Emission spectra of FL, FL + Cu₁₁ and FL + TEA.

  DownLoad: Full-Size Img PowerPoint

  Transient photocurrent response was performed to understand the internal charge separation and transport process. As shown in Fig. 2c, the Cu₁₁ demonstrated a small photocurrent response, which is consistent with the low catalytic activity without the photosensitizer. Notably, the incorporation of the photosensitizer (FL) could significantly boost the photocurrent of Cu₁₁, indicating efficient transport of photo-generated electrons from FL to Cu₁₁, thereby facilitating effective separation and transfer of photogenerated electrons [27–29].

  Fluorescence spectroscopy and time-resolved photoluminescence (TRPL) experiments were performed to explore the photogenerated electron transfer and carrier separation. As illustrated in Fig. 2d and Fig. S8 (Supporting information), addition of Cu₁₁ into a solution of FL in CH₃OCH₃–H₂O (1:1) results in obvious quenching between the Cu₁₁ catalyst and excited FL*, which indicates Cu₁₁ could greatly promote electron transport for improving the charge separation efficiency in this photocatalytic system. The addition of the sacrificial electron donor triethylamine (TEA) resulted in a significant decrease in the emission intensity of fluorescence (Fig. S9 in Supporting information). This suggests that TEA molecules effectively dissipated the majority of the energy from FL* molecules via charge transfer within the practical photocatalytic system. Additionally, the lifetime (8.32 ns) of the excited state of FL decreased to 4.90 ns and 3.96 ns upon addition of Cu₁₁ and TEA, respectively (Fig. S10 in Supporting information), which was due to their suppression of the recombination of photogenerated charge carriers to enhance the electron transfer. The concentration of triethylamine introduced into the photocatalytic reaction system is four orders of magnitude greater than that of the Cu₁₁ catalyst. This significant disparity results in a more pronounced quenching effect on the initial FL solution. Consequently, it can be inferred that the reduction quenching effect of triethylamine on the excited FL is more substantial than the oxidation quenching effect exerted by the Cu₁₁ catalyst. Thus, the photoexcited FL* is reductively quenched by triethylamine to form FL⁻. Subsequent electrons transferred from FL⁻ to Cu₁₁ catalyst encounter protons for effective reduction to produce hydrogen and the plausible mechanism of photocatalytic hydrogen evolution from Cu₁₁ is illustrated in Fig. S11 (Supporting information).

  The above results indicated that the Cu₁₁ cluster exhibited excellent catalytic performance for proton reduction, but it could only be effective in multicomponent catalytic systems containing photosensitizers (FL). Hybrid materials formed by the composite of metal clusters and semiconductors can promote the separation/transfer of photo-generated electrons and holes in semiconductor substrates, accompanied by accelerating proton reduction through metal clusters. Meanwhile, for metal clusters, the introduction of semiconductors can significantly increase the concentration of photo-generated charge carriers, enhance photocatalytic performance, and especially avoid the introduction of organic photosensitizers [30]. Therefore, the TiO₂ nanosheet was synthesized as the efficient photocatalytic semiconductor [31], then the Cu₁₁@TiO₂ hybrid material was obtained by loading discrete Cu₁₁ cluster units on this photosensitive semiconductor substrate. Fig. 3a shows the solid-state UV-vis absorption spectrum of materials. The results demonstrated that the TiO₂ nanosheet only showed significant absorption in the ultraviolet region, while the Cu₁₁ cluster showed a wide absorption range extending to 615 nm. The absorption spectrum of the Cu₁₁@TiO₂ composite showed a significant increase in signal that can be attributable to the Cu₁₁ cluster, indicating the successful loading of copper clusters. Based on the Kubelka–Munk theory [4,32], the band gap energies (E_g) derived from the Tauc plots were evaluated to be 2.66 eV for the Cu₁₁ cluster, 3.56 eV for TiO₂ nanosheet and 3.50 eV for the Cu₁₁@TiO₂ composite (Fig. 3b). Fig. 3c shows the powder X-ray diffraction (PXRD) patterns of Cu₁₁, TiO₂ nanosheets, and Cu₁₁@TiO₂ composite with different loading capacity. The absence of Cu₁₁ diffraction peaks in the PXRD of Cu₁₁@TiO₂ composite indicates that Cu₁₁ clusters loaded on TiO₂ nanosheets do not exist in an aggregated state but in a dispersed state. Moreover, the SEM images showed that the overall morphology of TiO₂ did not change before and after the loading of Cu₁₁ (Fig. S12 in Supporting information). In order to clearly demonstrate the dispersed copper clusters loaded on TiO₂ nanosheets, we characterized the composite material using high-resolution transmission electron microscopy (HRTEM). As shown in Figs. S13 and S15 (Supporting information), TiO₂ exhibited a clear nanosheet stacking state in the Cu₁₁@TiO₂ hybrid material, which maintained the nanosheet morphology of before loading. More importantly, the dispersed copper cluster units are clearly observed in the edge layer region (Fig. 3d), demonstrating the successful preparation of Cu₁₁@TiO₂ composite. However, no signal from Cu₁₁ was detected in the infrared spectrum, which is attributed to the low loading amount of Cu₁₁ (Fig. S14 in Supporting information). Furthermore, the elemental mapping results demonstrated the uniform distribution of Cu element in the whole composite material (Fig. S15), proving the discrete loading of Cu₁₁ clusters onto TiO₂.

  Figure 3

  

  Figure 3.  (a) UV–vis absorption spectra of Cu₁₁, TiO₂ and Cu₁₁@TiO₂. (b) The band gap of Cu₁₁, TiO₂, and Cu₁₁@TiO₂ calculated by UV–vis absorption spectra. (c) PXRD patterns of Cu₁₁, TiO₂, and Cu₁₁@TiO₂ with different loadings. (d) HRTEM images of Cu₁₁@TiO₂. (e) High-resolution Ti 2p XPS spectra of Cu₁₁ and Cu₁₁@TiO₂. (f) High-resolution Cu 2p XPS spectra of Cu₁₁ and Cu₁₁@TiO₂.

  DownLoad: Full-Size Img PowerPoint

  In addition, XPS was conducted to characterize the Cu₁₁@TiO₂ hybrid material. Two O 1s peaks were observed at the binding energies of 529.6 and 531.4 eV in TiO₂ nanosheet. After loading Cu₁₁ clusters, three fitted O 1s peaks were observed at 529.9, 531.7 and 533.3 eV (Fig. S16 in Supporting information), in which the last two peaks were clearly attributed to the signal of the organic ligand of the loaded Cu₁₁ clusters units. Combined with the emerged XPS signal of copper elements in Cu₁₁@TiO₂ material, they jointly prove the successful introduction of copper cluster units. More importantly, the XPS signals of the Cu₁₁@TiO₂ composite showed significant displacement compared to the corresponding individual Cu₁₁ clusters and TiO₂. Especially, the XPS binding energy of O 1s attributed to the Cu₁₁ cluster in Cu₁₁@TiO₂ showed a negative shift compared to that of individual Cu₁₁. The O 1s signal attributed to TiO₂ at low binding energy (529.9 eV) showed a positive shift compared to pure TiO₂ nanosheet (529.6 eV). As shown in Fig. 3e, the Ti 2p spectrum of Cu₁₁@TiO₂ presented two peaks at binding energies of 464.1 eV and 458.2 eV, respectively, both of which showed more positive shifts than those of pure TiO₂ nanosheet. In contrast, the XPS binding energy of Cu 2p_1/2 and Cu 2p_3/2 levels in Cu₁₁@TiO₂ (Fig. 3f) both showed negative shifts of 0.1 eV and 0.2 eV (952.6 eV, 932.6 eV) as compared to that of Cu₁₁ cluster (952.7 eV, 932.8 eV). The opposite tendency of binding energy deviation indicated that there is electron transfer from TiO₂ to Cu₁₁ in the Cu₁₁@TiO₂ hybrid material, which could improve the separation efficiency of photoinduced charge and holes, and promote the electron transfer to copper clusters to trigger proton reduction.

  The photocatalytic hydrogen evolution performance of the Cu₁₁@TiO₂ composite with different loadings of Cu₁₁ was investigated. As shown in Fig. 4a, when TiO₂ was used as the photocatalyst, only a very small amount of hydrogen could be produced in a water and methanol solution after 6 h of irradiation, which implied that it is challenging to fully harness the photocatalytic activity of TiO₂ nanosheets in the absence of co-catalysts. In contrast, the Cu₁₁@TiO₂ composite exhibited significantly enhanced photocatalytic hydrogen evolution activity under the same condition. Specifically, it could reach the optimum hydrogen production of 99.84 µmol with 3.5% loading of Cu₁₁. The photocatalytic efficiency of Cu₁₁@TiO₂ composite is 60 times greater than that of pure TiO₂ nanosheet, indicating a remarkable synergistic catalytic effect between Cu₁₁ clusters and TiO₂ nanosheets. Cyclic tests were conducted to investigate the stability of Cu₁₁@TiO₂ hybrid material. It can be seen from Fig. 4b that the Cu₁₁@TiO₂ catalyst maintained a high catalytic activity after eight catalytic cycles. In order to further verify the stability of Cu₁₁@TiO₂ after the photocatalytic reaction, multiple characterization methods were performed on the post-catalyzed Cu₁₁@TiO₂ material. The SEM and TEM images (Figs. S17 and S18 in Supporting information) indicated that the nanosheet stacking morphology of Cu₁₁@TiO₂ was not disrupted after the photocatalytic reaction. Both the unchanged PXRD and FTIR spectra confirmed the integrity of the structure and composition after the photocatalytic reaction (Fig. S19 in Supporting information). Furthermore, as shown in Fig. S18, the uniformly dispersed copper cluster units could still be clearly seen in the Cu₁₁@TiO₂ composite after photocatalysis reaction, and no obvious particle aggregation was observed, which was consistent with the absence of copper diffraction peaks in the PXRD. These results demonstrated the photocatalytic stability of Cu₁₁@TiO₂.

  Figure 4

  

  Figure 4.  (a) Comparison of hydrogen evolution performance of Cu₁₁@TiO₂ with different Cu₁₁ loadings. (b) The cyclic test performance of Cu₁₁@TiO₂. (c) Transient photocurrent response of Cu₁₁, TiO₂ and Cu₁₁@TiO₂. (d) EIS of TiO₂, Cu₁₁, and Cu₁₁@TiO₂. (e) The schematic diagram of the photocatalytic process of Cu₁₁@TiO₂ hybrid material.

  DownLoad: Full-Size Img PowerPoint

  The efficiency of photocatalysis is not only affected by light capture ability and intrinsic catalytic activity, but also significantly affected by the charge separation efficiency. Transient photocurrent response and electrochemical impedance spectroscopy (EIS) measurements were performed to evaluate the photoelectrochemical properties of materials. As shown in Fig. 4c, all materials showed the obvious light response behavior. The pure TiO₂ nanosheets only exhibited a weak photocurrent response, while the Cu₁₁@TiO₂ generated a stronger photocurrent signal than Cu₁₁ and TiO₂, which proved that the introduction of Cu₁₁ clusters could effectively promote the separation of photo-generated electrons and holes of TiO₂. Meanwhile, as shown in Fig. 4d, Cu₁₁@TiO₂ composite afforded a smaller Nyquist plot diameter, indicating a lower carrier-migration resistance compared to pure TiO₂ nanosheets and thus a more efficient interface charge transfer process. In addition, the steady-state photoluminescence (PL) emission spectra of TiO₂ and Cu₁₁@TiO₂ were analyzed. As depicted in Fig. S20 (Supporting information), a notable reduction of the photoluminescence intensity is observed for Cu₁₁@TiO₂ composite in comparison to pristine TiO₂. This suggests that the intrinsic process of radiative recombination of photoexcited electron-hole pairs within the TiO₂ nanosheets undergoes substantial suppression due to the Cu₁₁ clusters, which promotes the efficiency of photogenerated carrier separation. The proton affinity of the catalysts was evaluated using linear sweep voltammetry (Fig. S21 in Supporting information). At a current density of 0.2 mA/cm², Cu₁₁@TiO₂ exhibited the lowest overpotential, demonstrating its superior proton affinity. These results were consistent with the optimal photocatalytic hydrogen evolution of Cu₁₁@TiO₂ hybrid material. Based on the above results, a schematic diagram of the photocatalytic process of Cu₁₁@TiO₂ hybrid material is displayed in Fig. 4e. Under the irradiation, the TiO₂ semiconductor matrix generated photo-generated electron-hole pairs. The copper cluster units uniformly dispersed on the unfolded nanosheets serve as capture centers, accepting transferred electrons to catalyze proton reduction at the active sites. Simultaneously, the holes in the valence band of TiO₂ nanosheets were trapped by methanol. The introduction of copper cluster cocatalyst greatly facilitated a more efficient separation and migration of photo-generated electrons/holes and rapid charge transfer to facilitate the H₂ production reaction.

  In summary, we reported a remarkable case involving an 11-core copper nanocluster that is protected by a photoactive rhodamine B alkynyl ligand in this work. Owing to the stabilization and sensitization effects rendered by the RB-ligands, the Cu₁₁ cluster showed a commendable capacity for proton reduction within optical photocatalytic systems. Furthermore, the Cu₁₁ clusters were deposited on TiO₂ nanosheets to construct a highly active composite material Cu₁₁@TiO₂, which effectively promoted the separation and migration of photoinduced electrons and holes. Significantly, the dual-component photocatalytic system demonstrated an extraordinarily improved photocatalytic activity and achieved a sixty-fold enhancement when compared to pure TiO₂ nanosheets. This work not only provides a valuable reference for the design of dye ligand-protected metal nanoclusters but also broadens the application scope of the metal cluster-based composites.

  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

  Yan-Ling Li: Writing – review & editing, Visualization, Project administration. Yue Xu: Writing – original draft, Visualization, Investigation, Formal analysis, Data curation. Chen-Hong Wang: Validation, Formal analysis, Data curation. Rui Wang: Writing – review & editing, Supervision, Investigation, Funding acquisition. Shuang-Quan Zang: Writing – review & editing, Funding acquisition, Conceptualization.

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (Nos. 22371263 and U2004193), Natural Science Foundation of Henan Province (No. 232300421225).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111256.

  
  
• 1. [1]

     X.P. Tao, Y. Zhao, S.Y. Wang, C. Li, R.G. Li, Chem. Soc. Rev. 51 (2022) 3561–3608. doi: 10.1039/d1cs01182k

  2. [2]

     Z. Wei, M.L. Liu, Z.J. Zhang, et al., Energy Environ. Sci. 11 (2018) 2581–2589. doi: 10.1039/c8ee01316k

  3. [3]

     P.C. Fan, Y.H. He, J. Pan, et al., Chin. Chem. Lett. 35 (2024) 108513. doi: 10.1016/j.cclet.2023.108513

  4. [4]

     H.L. Wang, J.A. Liu, X.D. Xiao, et al., Chin. Chem. Lett. 34 (2023) 107125. doi: 10.1016/j.cclet.2022.01.018

  5. [5]

     B.K. Zhang, D.B. Wang, J.M. Cao, et al., Chin. Chem. Lett. 35 (2024) 110254. doi: 10.1016/j.cclet.2024.110254

  6. [6]

     R.C. Jin, C.J. Zeng, M. Zhou, Y.X. Chen, Chem. Rev. 116 (2016) 10346–10413. doi: 10.1021/acs.chemrev.5b00703

  7. [7]

     I. Chakraborty, T. Pradeep, Chem. Rev. 117 (2017) 8208–8271. doi: 10.1021/acs.chemrev.6b00769

  8. [8]

     Q.F. Yao, T. Chen, X. Yuan, J.P. Xie, Acc. Chem. Res. 51 (2018) 1338–1348. doi: 10.1021/acs.accounts.8b00065

  9. [9]

     C.L. Chen, R. Chen, L.S. Long, L.S. Zheng, X.J. Kong, Chin. Chem. Lett. 35 (2024) 108795. doi: 10.1016/j.cclet.2023.108795

  10. [10]

      J.P. Dong, Y. Xu, X.G. Zhang, et al., Angew. Chem. Int. Ed. 62 (2023) e202313648. doi: 10.1002/anie.202313648

  11. [11]

      L.J. Liu, Z.Y. Wang, Z.Y. Wang, et al., Angew. Chem. Int. Ed. 61 (2022) e202205626. doi: 10.1002/anie.202205626

  12. [12]

      Q. Tang, Y.J. Lee, D.Y. Li, et al., J. Am. Chem. Soc. 139 (2017) 9728–9736. doi: 10.1021/jacs.7b05591

  13. [13]

      L.L.M. Zhang, W.Y. Wong, Aggregate 4 (2023) e266. doi: 10.1002/agt2.266

  14. [14]

      S. Biswas, S. Das, Y. Negishi, Nanoscale Horiz. 8 (2023) 1509–1522. doi: 10.1039/d3nh00336a

  15. [15]

      Y.P. Liu, J.L. Yu, Y.F. Lun, et al., Adv. Funct. Mater. 33 (2023) 2304184. doi: 10.1002/adfm.202304184

  16. [16]

      Z.F. Xin, Z.B. Yuan, J.J. Liu, et al., Chin. Chem. Lett. 34 (2023) 107458. doi: 10.1016/j.cclet.2022.04.056

  17. [17]

      L.J. Liu, J.W. Zhang, M. Asad, et al., Chem. Commun. 57 (2021) 5586–5589. doi: 10.1039/d1cc01319j

  18. [18]

      T. Jia, J. Ai, X.G. Li, et al., Nat. Commun. 15 (2024) 9551. doi: 10.1038/s41467-024-53950-7

  19. [19]

      M. Qu, F.Q. Zhang, D.H. Wang, et al., Angew. Chem. Int. Ed. 59 (2020) 6507–6512. doi: 10.1002/anie.202001185

  20. [20]

      T.X. Song, X. Cai, Y. Zhu, Nanoscale 16 (2024) 13834–13846. doi: 10.1039/d4nr01835d

  21. [21]

      Y.D. Cao, H.P. Hao, H.S. Liu, et al., Nanoscale 13 (2021) 16182–16188. doi: 10.1039/d1nr04683g

  22. [22]

      Y.D. Cao, D. Yin, M.L. Wang, et al., J. Mater. Chem. A 12 (2024) 12043–12053. doi: 10.1039/d3ta07218e

  23. [23]

      A.U. Petersen, C. Kjær, C. Jensen, M. Brøndsted Nielsen, S. Brøndsted Nielsen, Angew. Chem. Int. Ed. 59 (2020) 20946–20955. doi: 10.1002/anie.202008314

  24. [24]

      X.Y. Dong, M. Zhang, R.B. Pei, et al., Angew. Chem. Int. Ed. 55 (2016) 2073–2077. doi: 10.1002/anie.201509744

  25. [25]

      X.Y. Chang, K.H. Low, J.Y. Wang, J.S. Huang, C.M. Che, Angew. Chem. Int. Ed. 55 (2016) 10312–10316. doi: 10.1002/anie.201604762

  26. [26]

      H.L. Zhu, H.Y. Chen, Y.X. Han, et al., J. Am. Chem. Soc. 144 (2022) 13319–13326. doi: 10.1021/jacs.2c04670

  27. [27]

      C.Q. Ji, W.J. Wang, E.M. El-Sayed, et al., Appl. Catal. B: Environ. 285 (2021) 119782. doi: 10.1016/j.apcatb.2020.119782

  28. [28]

      Y. Xu, Q.G. Dong, J.P. Dong, et al., Chem. Commun. 59 (2023) 3067–3070. doi: 10.1039/d3cc00275f

  29. [29]

      Z. Han, L. Shen, W.W. Brennessel, P.L. Holland, R. Eisenberg, J. Am. Chem. Soc. 135 (2013) 14659–14669. doi: 10.1021/ja405257s

  30. [30]

      X.Q. Liu, J. Locozzia, Y. Wang, et al., Energy Environ. Sci. 10 (2017) 402–434. doi: 10.1039/C6EE02265K

  31. [31]

      G.C. Zuo, Y.T. Wang, W.L. Teo, Q.M. Xian, Y.L. Zhao, Appl. Catal. B: Environ. 291 (2021) 120126. doi: 10.1016/j.apcatb.2021.120126

  32. [32]

      C. Cheng, Y.K. Liu, G. Sheng, et al., Angew. Chem. Int. Ed. 63 (2024) e202403473. doi: 10.1002/anie.202403473

• 

  Figure 1  (a) The overall structure of Cu₁₁. (b) The metal kernel in Cu₁₁ with C3 symmetry. (c-e) The coordination modes of RB-CH ligands in Cu₁₁ cluster. All hydrogen atoms are omitted for clarity, and the carbon atoms of the ethynide group are represented as small black spheres. Cu: brown, O: red, N: blue, C: gray.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Effect of different photosensitizer and sacrificial agent in the photocatalytic production. (b) The cyclic test performance of Cu₁₁ cluster. (c) Transient photocurrent response of the bare glassy carbon electrode (black), bare glassy carbon electrode + FL (blue), pristine Cu₁₁ (red) and FL + Cu₁₁ (green). (d) Emission spectra of FL, FL + Cu₁₁ and FL + TEA.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) UV–vis absorption spectra of Cu₁₁, TiO₂ and Cu₁₁@TiO₂. (b) The band gap of Cu₁₁, TiO₂, and Cu₁₁@TiO₂ calculated by UV–vis absorption spectra. (c) PXRD patterns of Cu₁₁, TiO₂, and Cu₁₁@TiO₂ with different loadings. (d) HRTEM images of Cu₁₁@TiO₂. (e) High-resolution Ti 2p XPS spectra of Cu₁₁ and Cu₁₁@TiO₂. (f) High-resolution Cu 2p XPS spectra of Cu₁₁ and Cu₁₁@TiO₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Comparison of hydrogen evolution performance of Cu₁₁@TiO₂ with different Cu₁₁ loadings. (b) The cyclic test performance of Cu₁₁@TiO₂. (c) Transient photocurrent response of Cu₁₁, TiO₂ and Cu₁₁@TiO₂. (d) EIS of TiO₂, Cu₁₁, and Cu₁₁@TiO₂. (e) The schematic diagram of the photocatalytic process of Cu₁₁@TiO₂ hybrid material.

  下载: 全尺寸图片 幻灯片
•