English

• 0.   Introduction

  An artificial photocatalytic system inspired by natural enzymes that react under an ambient atmosphere and use moderate solvents and clean energy is a major endeavor in catalytic chemistry^[1-3]. To mimic the selectivity and efficiency of enzymatic systems, chemists have constructed supramolecular systems with defined hydrophobic cavity sites to catalyze specific chemical transformations based on their special microenvironments^[4-6]. Coordination-driven self-assembled metal-organic cages (MOCs) are outstanding candidates for emulating the qualities of enzyme active sites well^[7-12] due to the exquisite structures and high order functions such as catalysis^[13-16], separation^[17], drug delivery^[18-19], and stabilizing activated species^[20-21]. In addition, MOCs can be used as hosts to encapsulate guest molecules^[22], allowing the components to be forced closer within the confined space and efficiently enhanced in electron, energy, or substance transfer, which has also brought about widespread attention^[23-24].

  The photocatalytic system, as a highly selective conversion and clean method for the synthesis of fine chemicals, has been attracting much attention due to mild reaction conditions^[25-26]. More than ever, the selective oxidation of organic compounds by molecule dioxygen (O₂) or air under light conditions is considered to be availably and environmentally friendly and thus of great academic and application significance^[27-28]. Anthraquinones (AQs) as favorable photocatalysts are generally applied in organic reactions^[29]. In biochemistry, AQs which are components of redox proteins and enzymes^[30] have been considered as the charge acceptors and carriers in electron transport reactions, such as natural photosynthesis^[31] and aerobic respiration^[32]. AQs and its derivatives are well known for their remarkable redox properties^[34]. They can be employed as redox catalysts in organic reactions to generate reactive oxygen species (ROS), which have powerful oxidization, including hydroxyl radicals (·OH), superoxide radical (·O₂^-), and singlet oxygen(¹O₂)^[35]. Recently, they have been proven to be effective in oxidizing inert C(sp³)-H bonds^[33]. Apart from the redox properties, AQs also have excellent photocatalytic properties^[35]. With the combined above two characteristics, we can assume that AQs can be photocatalysts for aerobic oxidative reactions.

  Herein, we constructed novel supramolecular structures featuring large hydrophobic cavities and favorable light absorption, which were obtained by the self-assembly of metal ions with well-functionalized and qualifiable AQs-based ligands to increase the level of structural complexity and functional diversity. Benefiting from these characteristics, we have presented an aerobic oxidation system using metal-organic cages as photocatalysts for the highly selective production of relevant carbonyl compounds through the oxidation of toluene and aromatic alcohols using O₂ as the oxidant under mild conditions and without the addition of additives.

  1.   Experimental

  1.1   Materials and physical measurements

  All chemicals were of reagent-grade quality obtained from commercial sources and used as supplied unless otherwise stated. ¹H NMR spectra were measured on a Varian DLG 400M spectrometer. Electrospray ionization mass spectrometry (ESI-MS) was carried out on an HPLC-Q-Tof MS spectrometer. UV-Vis spectra were measured on an HP 8453 spectrometer. Gas chromatography-mass spectrometry was carried out in Agilent 6890N GC. Elemental analysis was performed on the Elementar Vario EL cube. Thermogravimetric analysis (TGA) was performed on SDT Q600 V20.9 Build 20 instrument.

  1.2   Synthesis of the ligands and cages

  Ligands L¹ and L² (L¹=N², N⁷-di((2, 2′-bipyridin)-5-yl)-9, 10-dioxo-9, 10-dihydroanthracene-2, 7-dicarboxamide, L²=N², N⁷-bis(4-((2, 2′-bipyridin)-5-yl) phenyl)-9, 10-dioxo-9, 10-dihydroanthracene-2, 7-dicarboxamide) were synthesized according to established methods by coupling 9, 10-quinone-2, 7-dicarboxylic acid^[36] to 5-amino-2, 2′-bipyridine^[37] and 5-(4-aminophenyl)-2, 2′-bipyridine^[38], respectively. A detailed description of each of these steps followed.

  1.2.1   Synthesis of L¹ and L²

  20 mL thionyl chloride and 9, 10-quinone-2, 7-dicarboxylic acid (0.8 g, 2.7 mmol) were added into a 100 mL round-bottom flask, then three drops of DMF were added. The mixture was heated at 80 ℃ for several hours. The solvent was spun off after the reaction solution was clarified to obtain a yellow-green solid. The obtained solid was dissolved in 80 mL of dry tetrahydrofuran, and 80 mL of tetrahydrofuran solution of 5-amino-2, 2′-bipyridine (1.4 g, 8.1 mmol) containing 3 mL of triethylamine was added dropwise. After the mixture was stirred at room temperature overnight, the temperature was raised to 85 ℃ and the reaction was continued for 10 h. Then the solution was cooled to room temperature and suction filtered. The upper filter cake was washed with dichloromethane, methanol, and water in turn and dried to obtain buff ligand L¹ (1.4 g, 2.3 mmol) with a yield of 85.6%. ¹H NMR (400 MHz, DMSO-d₆): δ 11.07 (s, 2H), 9.10 (s, 2H), 8.84 (d, J=12.0 Hz, 2H), 8.67 (d, J=4.1 Hz, 2H), 8.51 (s, 2H), 8.38 (dd, J=21.6, 7.6 Hz, 8H), 7.93 (s, 2H), 7.43 (s, 2H). Elemental analysis Calcd. for C₃₆H₂₂N₆O₄(%): H, 3.68; C, 71.75; N, 13.95. Found(%): H, 3.91; C, 71.63; N, 12.53.

  The ligand L² was synthesized in a similar way to L¹ except that 5-amino-2, 2′-bipyridine was replaced by 5-(4-aminophenyl)-2, 2′-bipyridine, claybank ligand L² (1.6 g, 2.2 mmol) was obtained with a yield of 79.5%. ¹H NMR (400 MHz, DMSO-d₆): δ 10.92 (s, 2H), 9.07 (s, 2H), 8.85 (d, J=7.6 Hz, 2H), 8.72 (d, J=6.2 Hz, 2H), 8.54 (d, J=8.1 Hz, 2H), 8.50-8.41 (m, 6H), 8.29 (d, J=8.5 Hz, 2H), 8.05-7.96 (m, 6H), 7.90 (d, J=8.7 Hz, 4H), 7.47 (s, 2H). Elemental analysis Calcd. for C₄₈H₃₀N₆O₄(%): H, 4.01; C, 76.38; N, 11.13. Found(%): H, 4.63; C, 75.53; N, 10.16.

  1.2.2   Synthesis of 1-Zn and 2-Zn

  Treating L¹ (180.6 mg, 0.3 mmol) or L² (226.2 mg, 0.3 mmol) with Zn(BF₄)₂ (47.8 mg, 0.2 mmol) in 30 mL DMF at 80 ℃ for 16 h resulted in the formation of a new set of "2+3" type cages 1-Zn and 2-Zn (Scheme 1), respectively in yields of 75.8% and 72.1%, based on the solid dried on vacuum. Elemental analysis Calcd. for Zn₂(C₃₆H₂₂N₆O₄)₃·4BF₄·C₃H₇NO(%): H, 3.12; C, 56.52; N, 11.28. Found(%): H, 3.05; C, 56.34; N, 11.35. Elemental analysis Calcd. for Zn₂(C₄₈H₃₀N₆O₄)₃·4BF₄·C₃H₇NO(%): H, 3.47; C, 62.71; N, 9.45. Found(%): H, 3.54; C, 62.49; N, 9.36. The characterizations of ¹H NMR, ESI-MS spectra, and TGA of the cages are shown in subsequent results and discussion.

  Scheme 1

  

  Scheme 1.  Schematic representation of the synthesis of cages

  下载: 全尺寸图片 幻灯片

  1.2.3   Synthesis of 2-Fe

  In the presence of Ar, treating L² (226.2 mg, 0.3 mmol) with Fe(BF₄)₂·6H₂O (67.5 mg, 0.2 mmol) in 30 mL DMF at 65 ℃ for 12 h resulted in the formation of 2-Fe in yields of 76.9%, based on the solid dried on a vacuum. Elemental analysis Calcd. for Fe₂(C₄₈H₃₀N₆O₄)₃ ·4BF₄·H₂O·C₃H₇NO(%): H, 3.55; C, 62.73; N, 9.46. Found(%): H, 3.53; C, 62.21; N, 9.80. The single crystal X-ray analysis and TGA of 2-Fe are shown in subsequent results and discussion.

  1.3   X-ray crystallography

  A brown-red crystalline product of 2-Fe was obtained by diffusion of diethyl ether into the mixture of ligand L² and Fe(BF₄)₂·6H₂O in DMF solution. X-ray intensities of the complex were collected on a Bruker D8 Venture diffractometer with graphite monochromated Mo Kα (λ=0.071 073 nm) using the SMART and SAINT programs. The structure was solved by direct methods and refined on F ² by full-matrix least-squares methods with SHELXTL-2018. In the structural refinement, except the partly occupied solvent molecules and the disordered parts in the cage, the other non-hydrogen atoms were refined anisotropically and hydrogen atoms within the ligand backbone were fixed geometrically at calculated distances and allowed to ride on the parent non-hydrogen atoms. The highly disordered state of the incorporated molecule solvents meant that lots of them could not be located, and hence in the final refinement, the electron density was treated with the SQUEEZE routine in the PLATON program package. One of the benzene rings, and the BF₄^- counter ions were disordered into two parts with the site occupied factors (s.o.f.) of each part being fixed in suitable value. To assist the stability of refinements, several restraints were applied: (1) geometrical constraints of idealized regular polygons for the disordered benzene ring in the ligand were used, with phenyl ring C—C bond lengths of 0.139 nm, and C—C distances diagonally across the phenyl ring of 0.278 nm; (2) the respective bond distances in the BF₄^- anions were restrained to the idealized geometry; (3) thermal parameters on adjacent atoms the disorder parts and the BF₄^- anions were restrained to be similar. The A alert in checkcif was caused by the weak diffraction intensity of the crystal. The crystal data of complex 2-Fe are summarized in Table 1.

  Table 1

  

  Table 1.  Crystal data and refinement details for complex 2-Fe

  下载: 导出CSV

  Parameter	2-Fe		Parameter	2-Fe
  Formula	Fe2C144H96B4F16N18O15·xsolvent		V / nm3	24.389(6)
  Formula weight	2 777.32		Z	6
  Crystal system	Trigonal		Dc / (g·cm-3)	1.135
  Space group	R3		F(000)	8 532
  a / nm	2.002 1(2)		Reflection measured, independent, observed [I > 2σ(I)]	37 512, 5 624, 2 598
  b / nm	2.002 1(2)		Rint	0.074
  c / nm	7.025 7(8)

  2.   Results and discussion

  2.1   Characterization and crystal structure

  The formation of the cages was characterized by ¹H NMR analysis and ESI-MS. The ¹H NMR spectra of complexes 1-Zn and 2-Zn showed one set of signals after coordination, which suggested the formation of discrete and symmetric assemblies, and most signals of 1-Zn and 2-Zn showed slight downfield-shifted resonance concerning the free ligands, respectively (Fig. 1a and 1b). ESI-MS spectrum was conducted for supporting the formations and stability of these cages in solution. The spectrum of 1-Zn exhibited signals displaying the expected isotopic patterns at m/z 484.59, 675.12, and 1056.18 (Fig. 1c). A simple comparison with the simulation results based on the natural isotopic abundances indicated that the peaks correspond to [Zn₂(L¹)₃(BF₄)_n]^(4-n)+ (n=0-2). In the same way, as shown in Fig. 1d, the main peaks in the ESI-MS spectrum of 2-Zn were observed at m/z 598.64, 827.19, and 1 284.28, which are assigned to [Zn₂(L²)₃(BF₄)_n]^(4-n)+ (n=0-2), respectively. The results revealed the formation of stable Zn₂L₃ species for each complex.

  Figure 1

  

  Figure 1.  ¹H NMR spectra of (a) 1-Zn and L¹, (b) 2-Zn and L² in DMSO-d₆ (298 K, 400 MHz) and ESI-MS spectra of cages (c) 1-Zn and (d) 2-Zn with an inset showing the observed and calculated isotope patterns

  下载: 全尺寸图片 幻灯片

  The TGA showed that solvent molecules in powders dried on a vacuum of 1-Zn, 2-Zn, and 2-Fe could be removed in a temperature range of room temperature to 110, 120, and 110 ℃, respectively (Fig. 2). The results indicated that the complexes 1-Zn and 2-Zn could maintain structures stabilization during the catalytic process at room temperature.

  Figure 2

  

  Figure 2.  TGA curves of (a) 1-Zn, (b) 2-Zn, and (c) 2-Fe

  The data was collected under N₂ followed by a ramp of 10 ℃·min^-1 up to 1 000 ℃.

  下载: 全尺寸图片 幻灯片

  The light-harvesting capability of the complexes was investigated by UV-Vis absorption spectra and fluorescence spectrum. The UV-Vis absorption spectra absorption peaks of 1-Zn were similar to that of L¹, but the maximum absorption peak position of 1-Zn showed a significant redshift, however, the UV-Vis absorption spectra of 2-Zn revealed a new absorption band at 335 nm compared with that of the ligand L² (Fig. 3a and 3b). At an excitation wavelength of 350 nm, 1-Zn exhibited a strong fluorescence emission peak at 432 nm, which occurred a little red shift compared to L¹, while the fluorescence emission peak at 419 nm for 2-Zn revealed a little blue shift compared to that of L² (Fig. 3c and 3d) at an excitation wavelength of 320 nm. Therefore, the above photophysical studies disclosed that the complexes have good photoactivity and can be applied in the photocatalytic field.

  Figure 3

  

  Figure 3.  UV-Vis absorption spectra and fluorescence spectrum of L¹, L², 1-Zn, and 2-Zn

  下载: 全尺寸图片 幻灯片

  Since the quality of single crystals for 1-Zn and 2-Zn was too poor to get structural information, the single crystal analysis of 2-Fe with a similar coordinated configuration was carried out to show the M₂L₃ structural feature of these complexes (Fig. 4). Single-crystal X-ray diffraction study revealed that 2-Fe crystallizes in the trigonal crystal system with the R3 space group. Each L² ligand coordinates with two Fe(Ⅱ) centers through its two chelating 2, 2′-bipyridine groups. And, two Fe(Ⅱ) ions located in the vertexes are linked by three L² ligands to form a C₃ symmetric "2+3" helicate cage. In each of the three ligands, the 2, 2′-bipyridine ring planes are not coplanar with the anthraquinone backbone, and the two planes are approximately perpendicular to each other. The anthraquinone rings are oriented with their edges directed to the center of the cage.

  Figure 4

  

  Figure 4.  Crystal structure of 2-Fe showing the coordination geometries of Fe(Ⅱ) ions and empty sphere (yellow ball) for guest encapsulation

  Solvent molecules and anions are omitted for clarity; from (a) a-axis, (b) b-axis, and (c) c-axis view of the crystal structure of 2-Fe showing cluster geometry.

  下载: 全尺寸图片 幻灯片

  2.2   Photooxidation reaction

  Encouraged by the photo-redox activity of anthraquinone groups and the well-defined cavities, we decided to evaluate the usefulness of cages 1-Zn and 2-Zn as photocatalysts for oxygenation. In a typical experiment, the oxidation reaction employing the photocatalyst 1-Zn/2-Zn (0.1 μmol) and toluene (0.03 mmol) in a Schlenk quartz tube containing 3.0 mL MeCN/H₂O (1∶1, V/V) as the solvent in the presence of O₂ at room temperature gave product benzaldehyde with yields of 16.9% and 12.8%, and the turnover number (TON) values were 50.8 and 38.5 within 3.5 h under 365 nm LED irradiation (Table 2, Entry 1). It is worth noting that benzaldehyde products had more than 99% high selectivity within 3.5 h.

  Table 2

  

  Table 2.  Oxidation reactions of toluene under different conditionsa

  下载: 导出CSV

  Entry	Deviation from standard conditions	Yieldb / %			TON
  		1-Zn	2-Zn		1-Zn	2-Zn
  1	None	16.9	12.8		50.8	38.5
  2	In dark	Trace	Trace		—	—
  3	No catalyst	Trace	Trace		—	—
  4	L1 instead of 1-Zn	Trace			—
  5	L2 instead of 2-Zn	Trace			—
  6	Ar instead of O2	Trace	Trace		—	—
  a Standard reaction conditions: 0.03 mmol substrate, 0.1 μmol photocatalyst, 3 mL solvent MeCN/H2O (1∶1, V/V), 365 nm LED light, O2, and room temperature; b Yields were determined by GC analysis.

  Control experiments indicated that light and the photocatalysts were both of great importance for toluene oxidation (Table 2, Entry 2 and 3). Under the same reaction conditions, the ligands afforded scarcely any product. The results suggested that metal-organic cages had absolute superiority over ligands in toluene oxidation reactions (Table 2, Entry 4 and 5). In the presence of Ar, the reaction was inhibited, indicating that oxygen as the oxidant plays an indispensable role in oxidation (Table 2, Entry 6). The host-guest chemistry of toluene being introduced into 1-Zn and 2-Zn were investigated, respectively. The ESI-MS data of the host-guest assemblies showed peaks at m/z=530.62 and 644.67, assigned to cations [toluene+1-Zn]⁴⁺ and [toluene+2-Zn]⁴⁺, respectively (Fig. 5). The ESI-MS analysis confirmed the 1∶2 host-guest binding stoichiometry. According to ¹H NMR spectra, when adding toluene to cages during the above assemblies in DMSO-d₆, slight upfield-shifted signals of the encapsulated guests were noticed which suggests the possible encapsulation of the toluene guests into the cage hosts (Fig. 6). The results suggest the presence of potential weak interactions between the cages and the guests, and indicate that formation of host-guest complexes between the cages and substrate might be the key factor for the photooxidation of the toluene.

  Figure 5

  

  Figure 5.  ESI-MS spectra of host-guest assemblies (a) [toluene+1-Zn]⁴⁺ and (b) [toluene+2-Zn]⁴⁺ with an inset showing the observed and calculated isotope patterns

  The circle was the peak of the metal-organic cages and the triangle was the peak of the host-guest complexes.

  下载: 全尺寸图片 幻灯片

  Figure 6

  

  Figure 6.  ¹H NMR spectra (400 MHz, 298 K) of 1-Zn, 2-Zn, toluene, toluene+1-Zn and toluene+2-Zn in DMSO-d₆

  下载: 全尺寸图片 幻灯片

  Various scavengers were added to the process to further identify the reaction pathway and reactive active species in this toluene photocatalytic oxidation reaction. The reactions were performed after the addition of benzoquinone (p-BQ, a quencher of superoxide radicals)^[39], isopropanol (IPA, a quencher of hydroxyl radicals)^[39], and TEMPO (a quencher of all free radicals)^[40]. As shown in Fig. 7a, when TEMPO, p-BQ, and IPA were added as quench agents, the reaction yields decreased significantly so superoxide radicals and hydroxyl radicals were both two reactive oxygen species for 1-Zn catalyzing toluene. Nevertheless, the findings in Fig. 7b suggest that the superoxide radicals as the main superoxide radicals should be answerable for the present photocatalytic oxidation toluene of 2-Zn. To determine whether the photooxidation process was an electron transfer (ET) or energy transfer (EnT) pathway, we replaced the solvent with a deuterium reagent which can enhance the lifetime of singlet oxygen^[41-42]. Fig. 7 shows that in deuterium reagents, the yields of 1-Zn and 2-Zn as photocatalysts did not increase. The results confirm the ability of 1-Zn and 2-Zn to activate molecular oxygen through the ET pathway under light irradiation.

  Figure 7

  

  Figure 7.  Oxidation reactions of toluene catalyzed by 1-Zn (a) and 2-Zn (b) in the presence of different scavengers

  Reaction conditions: 0.03 mmol substrate, 0.1 μmol photocatalyst, 0.03 mmol scavenger, 3 mL solvent MeCN/H₂O (1∶1, V/V), 365 nm LED light, O₂, room temperature; Yields were determined by GC analysis.

  下载: 全尺寸图片 幻灯片

  We also focused on the oxidation of aromatic alcohols to the corresponding aldehydes to explore the photocatalytic performance of 1-Zn and 2-Zn under mild conditions. 4-methoxybenzyl alcohol was selected as a model substrate in MeCN/H₂O (1∶1, V/V) under an oxygen atmosphere at room temperature for 3.5 h. The data for the catalytic oxidation of aromatic alcohols displayed in Table 3 with yields of about 70% indicate that cages 1-Zn and 2-Zn exhibit efficient catalytic activity for the oxidation of aromatic alcohols.

  Table 3

  

  Table 3.  Scope of the oxidation of aromatic alcohol substrates

  下载: 导出CSV

  Entry	Substrate	Yield / %
  		1-Zn	2-Zn
  1		70.5	68.0
  2		95.6	81.3
  3		35.6	34.8
  4		40.9	55.2
  5		23.7	31.1
  6		92.2	> 99
  Standard reaction conditions: 0.03 mmol substrate, 0.1 μmol photocatalyst, 3 mL solvent MeCN/H2O (1∶1, V/V), 365 nm LED light, O2, room temperature; Yields were determined by GC analysis.

  As demonstrated in Table 3, some alcohol derivatives modifying electron-donating (Entry 1, 2, and 6) or electron-deficient substituents (Entry 3 and 4) were used as the substrates for the photooxidation reactions by cages 1-Zn and 2-Zn. When the electron-donating groups —OCH₃ and —OCH₂CH₃ were applied, the catalytic oxidation could yield higher than the alcohol derivatives with electron-deficient groups —Cl and —Br. The yields of aromatic alcohol photooxidation reaction were decreased with the gradual change of substituents from electron-donating to electron-withdrawing. The reason could be that the stability of free radicals can be increased by electron-donating groups and the reaction activity of substrates mainly depended on the stability (or electronegativity) of photogenerated free radicals. With the increasing of electron-donating ability of substituents, the stability of free radicals can be gradually improved. As demonstrated before, the electronic properties of the substitutions on the phenyl ring exerted an important effect on the photooxidation reaction.

  To further learn about the underlying mechanism, additional experiments were conducted. As shown in Fig. 8, only when TEMPO or p-BQ was added as scavengers did the reaction yields decline significantly, thus the formation of the superoxide radical that mainly participated in this photocatalytic reaction. In the deuterium reagent, the yields of 1-Zn and 2-Zn were unchanged from those in common reagents, therefore photocatalysts activated molecular oxygen through the ET pathway under light irradiation.

  Figure 8

  

  Figure 8.  Oxidation of 4-methoxybenzyl alcohol catalyzed by 1-Zn (a) and 2-Zn (b) in the presence of different scavengers

  Standard reaction conditions: 0.03 mmol substrate, 0.1 μmol photocatalyst, 0.03 mmol scavenger, 3 mL solvent MeCN/H₂O (1∶1, V/V), 365 nm LED light, O₂, room temperature; Yields were determined by GC analysis.

  下载: 全尺寸图片 幻灯片

  3.   Conclusions

  In summary, anthraquinone-based redox-active cages with confined cavity sizes were constructed for the photooxidation of toluene and aromatic alcohols to assess the photocatalytic activity of 1-Zn and 2-Zn. The supramolecular systems were able to activate oxygen through the ET pathway for the photooxidation process. The aerobic oxidation reactions were not only mild and eco-friendly, occurring in ambient oxygen at room temperature but also with high selectivity and yields without any co-catalysts or additives, providing potential material for the photocatalytic oxidation reaction of organic compounds and on behalf of an ideal photocatalytic platform for economical and green organic synthesis in the academia.

  
  
• 1. [1]

     Brown C J, Toste F D, Bergman R G, Raymond K N. Supramolecular catalysis in metal-ligand cluster hosts[J]. Chem. Rev., 2015, 115(9):  3012-3035. doi: 10.1021/cr4001226

  2. [2]

     Yan K, Fujita M. A speedy marriage in supramolecular catalysis[J]. Science, 2015, 350(6265):  1165-1166. doi: 10.1126/science.aad7245

  3. [3]

     Hooley R J. Taking on the turnover challenge[J]. Nat. Chem., 2016, 8(3):  202-204. doi: 10.1038/nchem.2461

  4. [4]

     杨林林, 景旭, 何成, 段春迎. 氧杂蒽染料修饰钴-硫脲配合物的光解水放氢性能[J]. 无机化学学报, 2017,33,(6): 913-922. YANG L L, JING X, HE C, DUAN C Y. Photocatalytic hydrogen production based on cobalt-thiosemicarbazone complex with the xanthene dye moiety[J]. Chinese J. Inorg. Chem., 2017, 33(6):  913-922.

  5. [5]

     Kaphan D M, Levin M D, Bergman R G, Toste F D. A supramolecular microenvironment strategy for transition metal catalysis[J]. Science., 2015, 350(6265):  1235-1238. doi: 10.1126/science.aad3087

  6. [6]

     赵亮, 楚少帅, 蔡俊凯, 魏建伟, 李亚楠, 段春迎. 修饰NADH模拟物的金属-有机三元环光催化制氢[J]. 无机化学学报, 2021,37,(5): 769-777. ZHAO L, CHU S S, CAI J K, WEI J W, LI Y N, DUAN C Y. Metal-organic triangles with NADH mimics for photocatalytic hydrogen production[J]. Chinese J. Inorg. Chem., 2021, 37(5):  769-777.

  7. [7]

     Sun Q F, Iwasa J, Ogawa D, Ishido Y, Sato S, Ozeki T, Sei Y, Yamaguchi K, Fujita M. Self-assembled M₂₄L₄₈ polyhedra and their sharp structural switch upon subtle ligand variation[J]. Science, 2010, 328(5982):  1144-1147. doi: 10.1126/science.1188605

  8. [8]

     Zhou X P, Wu Y, Li D. Polyhedral metal-imidazolate cages: control of self-assembly and cage to cage transformation[J]. J. Am. Chem. Soc., 2013, 135(43):  16062-16065. doi: 10.1021/ja4092984

  9. [9]

     Cook T R, Stang P J. Recent developments in the preparation and chemistry of metallacycles and metallacages via coordination[J]. Chem. Rev., 2015, 115(15):  7001-7045. doi: 10.1021/cr5005666

  10. [10]

      Hong C M, Bergman R G, Raymond K N, Toste F D. Self-assembled tetrahedral hosts as supramolecular catalysts[J]. Acc. Chem. Res., 2018, 51(10):  2447-2455. doi: 10.1021/acs.accounts.8b00328

  11. [11]

      Zhang D, Ronson T K, Nitschke J R. Functional capsules via subcomponent self-assembly[J]. Acc. Chem. Res., 2018, 51(10):  2423-2436. doi: 10.1021/acs.accounts.8b00303

  12. [12]

      Purba P C, Maity M, Bhattacharyya S, Mukherjee P S. A self-assembled palladium(Ⅱ) barrel for binding of fullerenes and photosensitization ability of the fullerene-encapsulated barrel[J]. Angew. Chem. Int. Ed., 2021, 133(25):  14228-14235. doi: 10.1002/ange.202103822

  13. [13]

      Yoshizawa M, Klosterman J K, Fujita M. Functional molecular flasks: New properties and reactions within discrete, self-assembled hosts[J]. Angew. Chem. Int. Ed., 2009, 48(19):  3418-3438. doi: 10.1002/anie.200805340

  14. [14]

      Gaeta C, Manna P L, Rosa M D, Soriente A, Talotta C, Neri P. Supramolecular catalysis with self-assembled capsules and cages: What happens in confined spaces[J]. ChemCatChem, 2021, 13(7):  1638-1658. doi: 10.1002/cctc.202001570

  15. [15]

      Liu W, Stoddart J F. Emergent behavior in nanoconfined molecular containers[J]. Chem, 2021, 7(4):  919-947. doi: 10.1016/j.chempr.2021.02.016

  16. [16]

      李和川, 李明凤, 何成, 段春迎. 基于非贵金属镍的金属有机大环的光解水放氢性能[J]. 无机化学学报, 2018,34,(1): 11-19. Li H C, Li M F, He C, DUAN C Y. Construction of a noble-metal-free nickel metal-organic macrocycle for photocatalytic hydrogen production[J]. Chinese J. Inorg. Chem., 2018, 34(1):  11-19.

  17. [17]

      Zhang D W, Ronson T K, Zou Y Q, Nitschke J R. Metal-organic cages for molecular separations[J]. Nat. Rev. Chem., 2021, 5(3):  168-182. doi: 10.1038/s41570-020-00246-1

  18. [18]

      Zhu C Y, Pan M, Su C Y. Metal-organic cages for biomedical applications[J]. Isr. J. Chem., 2019, 59(3/4):  209-219.

  19. [19]

      Samanta S K, Isaacs L. Biomedical applications of metal organic polygons and polyhedra (MOPs)[J]. Coord. Chem. Rev., 2020, 410:  213181. doi: 10.1016/j.ccr.2020.213181

  20. [20]

      Breiner B, Clegg J K, Nitschke J R. Reactivity modulation in container molecules[J]. Chem. Sci., 2011, 2(1):  51-56. doi: 10.1039/C0SC00329H

  21. [21]

      Galan A, Ballester P. Stabilization of reactive species by supramolecular encapsulation[J]. Chem. Soc. Rev., 2016, 45(6):  1720-1737. doi: 10.1039/C5CS00861A

  22. [22]

      Ronson T K, Meng W, Nitschke J R. Design principles for the optimization of guest binding in aromatic-paneled Fe₄^ⅡL₆ cages[J]. J. Am. Chem. Soc., 2017, 139(28):  9698-9707. doi: 10.1021/jacs.7b05202

  23. [23]

      Leenders S H, Gramage D R, Bruin B D, Reek J N H. Transition metal catalysis in confined spaces[J]. Chem. Soc. Rev., 2015, 44(2):  433-448. doi: 10.1039/C4CS00192C

  24. [24]

      Zhao L, Jing X, Li X Z, Guo X Y, Zeng L, He C, Duan C Y. Catalytic properties of chemical transformation within the confined pockets of Werner-type capsules[J]. Coord. Chem. Rev., 2019, 378:  151-187. doi: 10.1016/j.ccr.2017.11.005

  25. [25]

      Yang B, Fu Z H, Su A Q, She J L, Chen M K, Tang S P, Hu W W, Zang C, Liu Y C. Influence of tetraalkylammonium cations on quality of decatungstate and its photocatalytic property in visible light-triggered selective oxidation of organic compounds by dioxygens[J]. Appl. Catal. B Environ., 2019, 242:  249-257. doi: 10.1016/j.apcatb.2018.09.099

  26. [26]

      Tang S P, Wu W F, Fu Z H, Zou S, Liu T C, Zhao H H, Kirk S R, Yin D L. Vanadium-substituted tungstophosphoric acids as efficient catalysts for visible-light-driven oxygenation of cyclohexane by dioxygen[J]. ChemCatChem, 2015, 7(17):  2637-2645. doi: 10.1002/cctc.201500314

  27. [27]

      Ji X C, Tan M, Fu M Y, Fu M, Deng G J, Huan H W. Photocatalytic aerobic alpha-thiolation/annulation of carbonyls with mercaptobenzimidazoles[J]. Org. Biomol. Chem., 2019, 17(20):  4979-4983. doi: 10.1039/C9OB00625G

  28. [28]

      Wang Z Z, Ji X C, Zhao J W, Huan H W. Visible-light-mediated photoredox decarbonylative Minisci-type alkylation with aldehydes under ambient air conditions[J]. Green Chem., 2019, 21(2):  5512-5516.

  29. [29]

      Jiang D B, Zhang Q, Yang L, Deng Y E, Yang B, Liu Y C, Zhang C, Fu Z H. Regulating effects of anthraquinone substituents and additives in photo-catalytic oxygenation of p-xylene by molecular oxygen under visible light irradiation[J]. Renew. Energy, 2021, 174:  928-938. doi: 10.1016/j.renene.2021.04.100

  30. [30]

      Kern M, Simon J. Electron transport chains and bioenergetics of respiratory nitrogen metabolism in Wolinella succinogenes and other Epsilonproteobacteria[J]. Biochim. Biophys. Acta-Bioenerg., 2009, 1787(6):  646-656. doi: 10.1016/j.bbabio.2008.12.010

  31. [31]

      Sladic D, Gasic M J. Reactivity and biological activity of the marine sesquiterpene hydroquinone avarol and related compounds from sponges of the order Dictyoceratida[J]. Molecules, 2006, 11(1):  1-33. doi: 10.3390/11010001

  32. [32]

      Lee C H, Guo J C, Chen L X, Mandal B K. Novel zinc phthalocyanine-benzoquinone rigid dyad and its photoinduced electron transfer properties[J]. J. Org. Chem., 2008, 73(21):  8219-8227. doi: 10.1021/jo801293s

  33. [33]

      Zhou M J, Zhang L, Liu G X, Xu C, Huang Z. Site-selective acceptorless dehydrogenation of aliphatics enabled by organophotoredox/ cobalt dual catalysis[J]. J. Am. Chem. Soc., 2021, 143(40):  16470-16485. doi: 10.1021/jacs.1c05479

  34. [34]

      Hao H, Li X, Lang X. Anthraquinones as photoredox active ligands of TiO₂ for selective aerobic oxidation of organic sulfides[J]. Appl. Catal. B-Environ., 2019, 259:  118038. doi: 10.1016/j.apcatb.2019.118038

  35. [35]

      González C J, Vosburg D A, Mora Rodriguez S E, Rodriguez S E M, Vázquez M A, Zepeda L G, Gómez C V, Rivera S L. Anthraquinones: Versatile organic photocatalysts[J]. ChemCatChem, 2020, 12(15):  3811-3827. doi: 10.1002/cctc.202000376

  36. [36]

      Akpinar H, Schlueter J A, Allão Cassaro R A, Friedman J R, Lahti P M. Rigid core anthracene and anthraquinone linked nitronyl and iminoyl nitroxide biradicals[J]. Cryst. Growth Des., 2016, 16(7):  4051-4059. doi: 10.1021/acs.cgd.6b00588

  37. [37]

      Gupta G, Iqbal P, Yin F, Liu J, Palmer R E, Sharma S, Leung C F, Mendes P M. Pt diffusion dynamics for the formation Cr-Pt core-shell nanoparticles[J]. Langmuir, 2015, 31(24):  6917-6923. doi: 10.1021/acs.langmuir.5b01410

  38. [38]

      Mahata K, Frischmann P D, Würthner F. Giant electroactive M₄L₆ tetrahedral host self-assembled with Fe(Ⅱ) vertices and perylene bisimide dye edges[J]. J. Am. Chem. Soc., 2013, 135(41):  15656-15661. doi: 10.1021/ja4083039

  39. [39]

      Li X H, Chen J S, Wang X C, Sun J H, Antonietti M. Metal-free activation of dioxygen by graphene/g-C₃N₄ nanocomposites: Functional dyads for selective oxidation of saturated hydrocarbons[J]. J. Am. Chem. Soc., 2011, 133(21):  8074-8077. doi: 10.1021/ja200997a

  40. [40]

      Cosemans I, Hontis L, Den B D V, Palmaerts A, Wouters J, Cleij T J, Lutsen L, Maes W, Junkers T, Vanderzande D J M. Discovery of an anionic polymerization mechanism for high molecular weight PPV derivatives via the sulfinyl precursor route[J]. Macromolecules, 2011, 44(19):  7610-7616. doi: 10.1021/ma201453s

  41. [41]

      Rineh A, Dolla N K, Ball A R, Magana M, Bremner J B, Hamblin M R, Tegos G P, Kelso M J. Attaching the NorA efflux pump inhibitor INF55 to methylene blue enhances antimicrobial photodynamic inactivation of methicillin-resistant Staphylococcus aureus in vitro and in vivo[J]. ACS Infect. Dis., 2017, 3(10):  756-766. doi: 10.1021/acsinfecdis.7b00095

  42. [42]

      Li Y M, Rizvi S A-E-A, Hu D Q, Sun D W, Gao A H, Li J, Jiang X P. Selective late-stage oxygenation of sulfides with ground-state oxygen by uranyl photocatalysis[J]. Angew. Chem. Int. Ed., 2019, 58(38):  13499-13506. doi: 10.1002/anie.201906080

• 

  Scheme 1  Schematic representation of the synthesis of cages

  下载: 全尺寸图片 幻灯片
  

  Figure 1  ¹H NMR spectra of (a) 1-Zn and L¹, (b) 2-Zn and L² in DMSO-d₆ (298 K, 400 MHz) and ESI-MS spectra of cages (c) 1-Zn and (d) 2-Zn with an inset showing the observed and calculated isotope patterns

  下载: 全尺寸图片 幻灯片
  

  Figure 2  TGA curves of (a) 1-Zn, (b) 2-Zn, and (c) 2-Fe

  The data was collected under N₂ followed by a ramp of 10 ℃·min^-1 up to 1 000 ℃.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  UV-Vis absorption spectra and fluorescence spectrum of L¹, L², 1-Zn, and 2-Zn

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Crystal structure of 2-Fe showing the coordination geometries of Fe(Ⅱ) ions and empty sphere (yellow ball) for guest encapsulation

  Solvent molecules and anions are omitted for clarity; from (a) a-axis, (b) b-axis, and (c) c-axis view of the crystal structure of 2-Fe showing cluster geometry.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  ESI-MS spectra of host-guest assemblies (a) [toluene+1-Zn]⁴⁺ and (b) [toluene+2-Zn]⁴⁺ with an inset showing the observed and calculated isotope patterns

  The circle was the peak of the metal-organic cages and the triangle was the peak of the host-guest complexes.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  ¹H NMR spectra (400 MHz, 298 K) of 1-Zn, 2-Zn, toluene, toluene+1-Zn and toluene+2-Zn in DMSO-d₆

  下载: 全尺寸图片 幻灯片
  

  Figure 7  Oxidation reactions of toluene catalyzed by 1-Zn (a) and 2-Zn (b) in the presence of different scavengers

  Reaction conditions: 0.03 mmol substrate, 0.1 μmol photocatalyst, 0.03 mmol scavenger, 3 mL solvent MeCN/H₂O (1∶1, V/V), 365 nm LED light, O₂, room temperature; Yields were determined by GC analysis.

  下载: 全尺寸图片 幻灯片
  

  Figure 8  Oxidation of 4-methoxybenzyl alcohol catalyzed by 1-Zn (a) and 2-Zn (b) in the presence of different scavengers

  Standard reaction conditions: 0.03 mmol substrate, 0.1 μmol photocatalyst, 0.03 mmol scavenger, 3 mL solvent MeCN/H₂O (1∶1, V/V), 365 nm LED light, O₂, room temperature; Yields were determined by GC analysis.

  下载: 全尺寸图片 幻灯片

  Table 1.  Crystal data and refinement details for complex 2-Fe

  Parameter	2-Fe		Parameter	2-Fe
  Formula	Fe2C144H96B4F16N18O15·xsolvent		V / nm3	24.389(6)
  Formula weight	2 777.32		Z	6
  Crystal system	Trigonal		Dc / (g·cm-3)	1.135
  Space group	R3		F(000)	8 532
  a / nm	2.002 1(2)		Reflection measured, independent, observed [I > 2σ(I)]	37 512, 5 624, 2 598
  b / nm	2.002 1(2)		Rint	0.074
  c / nm	7.025 7(8)

  下载: 导出CSV

  Table 2.  Oxidation reactions of toluene under different conditionsa

  Entry	Deviation from standard conditions	Yieldb / %			TON
  		1-Zn	2-Zn		1-Zn	2-Zn
  1	None	16.9	12.8		50.8	38.5
  2	In dark	Trace	Trace		—	—
  3	No catalyst	Trace	Trace		—	—
  4	L1 instead of 1-Zn	Trace			—
  5	L2 instead of 2-Zn	Trace			—
  6	Ar instead of O2	Trace	Trace		—	—
  a Standard reaction conditions: 0.03 mmol substrate, 0.1 μmol photocatalyst, 3 mL solvent MeCN/H2O (1∶1, V/V), 365 nm LED light, O2, and room temperature; b Yields were determined by GC analysis.

  下载: 导出CSV

  Table 3.  Scope of the oxidation of aromatic alcohol substrates

  Entry	Substrate	Yield / %
  		1-Zn	2-Zn
  1		70.5	68.0
  2		95.6	81.3
  3		35.6	34.8
  4		40.9	55.2
  5		23.7	31.1
  6		92.2	> 99
  Standard reaction conditions: 0.03 mmol substrate, 0.1 μmol photocatalyst, 3 mL solvent MeCN/H2O (1∶1, V/V), 365 nm LED light, O2, room temperature; Yields were determined by GC analysis.

  下载: 导出CSV
•