Spirobifluorene-Based Three-Dimensional Covalent Organic Frameworks with Rigid Topological Channels as Efficient Heterogeneous Catalyst
2020; Chinese Chemical Society; Volume: 3; Issue: 4 Linguagem: Inglês
10.31635/ccschem.020.202000493
ISSN2096-5745
AutoresYamei Liu, Chenyu Wu, Qingzhu Sun, Fan Hu, Qingyan Pan, Jing Sun, Yinghua Jin, Zhibo Li, Wei Zhang, Yingjie Zhao,
Tópico(s)Conducting polymers and applications
ResumoOpen AccessCCS ChemistryRESEARCH ARTICLE1 Apr 2021Spirobifluorene-Based Three-Dimensional Covalent Organic Frameworks with Rigid Topological Channels as Efficient Heterogeneous Catalyst Yamei Liu†, Chenyu Wu†, Qingzhu Sun, Fan Hu, Qingyan Pan, Jing Sun, Yinghua Jin, Zhibo Li, Wei Zhang and Yingjie Zhao Yamei Liu† College of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042 , Chenyu Wu† College of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042 School of Chemical Engineering, The University of New South Wales, Sydney, NSW 2052 , Qingzhu Sun College of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042 , Fan Hu College of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042 , Qingyan Pan College of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042 , Jing Sun College of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042 , Yinghua Jin Department of Chemistry, University of Colorado, Boulder, CO 80309 , Zhibo Li *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] College of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042 , Wei Zhang *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Department of Chemistry, University of Colorado, Boulder, CO 80309 and Yingjie Zhao *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] College of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042 https://doi.org/10.31635/ccschem.020.202000493 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Although the past decade has witnessed tremendous progress in the development of covalent organic frameworks (COFs), three-dimensional (3D) COFs fabrications and characterizations are still much less studied compared with two-dimensional (2D) COFs, due to their complicated topology structures caused by interpenetration and limited choices of node-building blocks. In this work, we constructed a novel 3D COF ( SP-3D-COF-BPY) successfully using orthogonal dual-planar spirobifluorene and bipyridine as building blocks. Also, we demonstrated the application of 3D COF-supported Pd(II) catalyst in heterogeneous catalysis. A sevenfold interpenetrated dia structure was revealed for SP-3D-COF-BPY by powder X-ray diffraction (PXRD) in conjunction with structural simulation and Pawley refinement. The bipyridine-linked frameworks bear one-dimensional (1D) unobstructed rigid channels and offer a high density of discrete coordination sites for chelating Pd(II) species. Very high loading of Pd(II) (∼15 wt %) was achieved in the asymmetric (as)-prepared Pd(II)@3D-COF-BYP. The generated highly ordered porous channels and easily accessible catalytic sites, such as COF-supported Pd(II) complex, serves as a highly active and stable microporous heterogeneous catalyst in Suzuki–Miyaura coupling reactions. The observed catalytic efficiency was high and the catalyst could be conveniently reused multiple times without any noticeable decay in catalytic performance. Download figure Download PowerPoint Introduction Covalent organic frameworks (COFs) are characteristic topological covalent structures with enormous internal surface area resulting from their nanoporous structures.1–6 Over the past decade, COFs have displayed a large variety of applications.7–12 Besides their traditional advantages in gas storage and separation, COFs recently revealed promising potentials as heterogeneous catalysts, given their highly ordered porous structure and a chemically modifiable internal surface.11,13–16 Owing to the well-defined topological arrangement of active sites, COFs could combine both advantages of small molecule catalysts and traditional heterogeneous catalysts. Indeed, traditional heterogeneous catalysts usually suffer from limited active area, recyclability, stability, leaching and deactivation issues.17–22 By contrast, COF-based heterogeneous catalysts benefit from tremendous internal surface areas, which endow them with much more discrete active sites. More importantly, the highly ordered structures with defined channels facilitate the diffusion of reactants and the precise loading of active metal catalysts. Besides, the visualization of the structure makes it possible to obtain a detailed insight into the relationship between the structure and catalytic activity. Thus, rationally designed COFs could be an excellent alternative to heterogeneous catalysts.23–26 Three-dimensional (3D) COFs exclusively constructed from covalent bonds exhibit superior structural stability and possess unique permanent pores across the topological frameworks.8,15,27–32 Such a high specific surface area and permanent porosity are readily accessible to substrates, making 3D COFs attractive candidates as heterogeneous catalysts, although they have been limited to only a few examples owing to their synthetic challenges.33 One successful example was reported by Qiu's group,34 in which they used dual-linker (DL)-COFs as an acid–base catalyst for one-pot cascade reactions with ∼90% yields. The lack of choices in favorable 3D-building blocks and their relatively complex 3D structures restrict the development of 3D COFs as heterogeneous catalysts.12,35–37 We have employed spirobifluorene ( SP) as a novel type of tetragonal-disphenoid 3D-building block28,38 to construct a new series of 3D COFs ( SP-3D-COFs). Due to the high rigidity and orthogonality of the two intersected fluorene groups in SP, the SP-3D-COFs possess uniform unobstructed one-dimensional (1D) channels. This could be particularly beneficial for heterogeneous catalysis, as these unobstructed channels throughout SP-3D-COFs are more favorable for anchoring noble metal ions, thus maximizing the accessibility of the active sites, compared with the traditional tetraphenylmethane-based 3D COFs. With this consideration, we introduced the 2,2′-bipyridine ( BPY) units (commonly used as metal chelating ligands) as linkers for designing a new SP-3D-COF ( SP-3D-COF-BPY). The characterization of the product by powder X-ray diffraction (PXRD) analysis revealed distinct features of unobstructed rigid channels and confirmed a sevenfold interpenetrated ( dia-c7) structure of SP-3D-COF-BPY. Subsequently, by a simple one-step soaking process at room temperature, the discrete bipyridine units could chelate Pd(II) ions. Due to the high accessibility of the active sites through the channels, the Pd(II)@SP-3D-COF-BPY exhibited superior catalytic efficiency over those with similar structures such as the Pd(II)-loaded SP-3D-COF-BPH (without bipyridine as the chelation group). We also demonstrated the superiority of Pd(II) ions, compared with reduced palladium nanoparticles (PdNPs) when each of them was deposited in the same COF in catalyzing Suzuki–Miyaura coupling reactions. In short, Pd(II)@SP-3D-COF-BPY as a highly ordered nanoporous catalyst demonstrated substantially improved performance and stability in catalyzing Suzuki–Miyaura coupling reactions than its other counterparts. Experimental Methods Materials Organic solvents including dichloromethane, ethanol, petroleum ether, acetone, tetrahydrofuran (THF), mesitylene, benzyl alcohol, acetic acid, N,N-dimethylformamide (DMF), N,N-dimethylsulfoxide (DMSO), 9,9′-spirobifluorene, 4-phthalaldehyde, 1-iodo-4-nitrobenzene, phenylboronic acid, 4-bromobiphenyl, 1-bromo-4-iodobenzene, 1-bromo-4-nitrobenzene, 2-bromonaphthalene, 4-iodoanisole, 4-iodotoluene, palladium diacetate, 3-pyridinecarboxaldehyde, and isoamyl nitrite were purchased from Adamas-Beta (Shanghai, China) and used as received. Fuming nitric acid and acetic anhydride were purchased from Sinopharm (Beijing, China) and used as received. Palladium on activated carbon was purchased from Acros Organics (Chile, China) and used as received. Hydrogen stored in a high-pressure gas cylinder was ordered from Dehai Gas (Qingdao, China). Synthesis of SP-3D-COF-BPY The synthesis of SP-3D-COF-BPY is illustrated in Scheme 1. The orthorhombic building block A (9,9′-spirobi[fluorene]-3,3′,6,6′-tetraamine) was synthesized as described previously.39 Compound D containing bipyridine was used as the linker. The synthesis of SP-3D-COF-BPY through the imine condensation of A (15 mg, 0.04 mmol) and D [3,3′-bipyridine]-6,6′-dicarbaldehyde (17 mg, 0.08 mmol) was performed in a mixture of phenylmethanol (134 μL), mesitylene (266 μL), and 6 M acetic acid (40 μL) in a decompressive glass tube at 120 °C for 3 days. Subsequently, the crystalline precipitates were isolated via filtering, followed by washing with THF before it was dried at 80 °C under vacuum for 12 h. SP-3D-COF-BPY was obtained as a yellow crystalline powder, which appeared insoluble in most common solvents. Scheme 1 | (a) Synthesis of SP-3D-COF-BPY, SP-3D-COF-BPH, and the corresponding model compound C. (b) Schematic representation of Pd(II) or PdNPs impregnation to the SP-3D-COF-BPY. Download figure Download PowerPoint Synthesis of Pd(II)@SP-3D-COF-BPY Palladium acetate (5.0 mg, 0.022 mmol) and SP-3D-COF-BPY (5 mg) were dispersed in 2 mL of dichloromethane. The suspension was stirred for 3 days at room temperature. The residue was isolated by filtration then washed with dichloromethane. The resulting powder was subjected to Soxhlet extraction for 1 day in dichloromethane and dried under vacuum at 80 °C for 12 h to yield Pd(II)@SP-3D-COF-BPY as a brown powder. Synthesis of [email protected] A well-dispersed suspension of SP-3D-COF-BPY (4 mg) in methanol (3 mL) was mixed with a solution of K2PdCl4 (4 mg, 0.012 mmol) in H2O (2 mL) for 3 h at room temperature. The mixture was brought to dryness under vacuum with stirring to deposit metal precursors in SP-3D-COF-BPY support. The residue was dispersed in methanol and H2O (5 mL, MeOH/H2O, v/v = 3∶2). A solution of NaBH4 in methanol (0.25 M, 2 mL) was added dropwise, stirring the mixture for 2 days at room temperature. The product was collected by centrifugation, washed three times with ethanol and dichloromethane, and dried under vacuum until further use. Suzuki–Miyaura coupling reaction catalyzed by Pd(II) @SP-3D-COF-BPY In a typical one-pot of Suzuki–Miyaura coupling reaction, phenylboronic acid (92 mg, 0.75 mmol), aryl halide (0.5 mmol), potassium carbonate (138 mg, 1.0 mmol), and Pd(II)@SP-3D-COF-BPY (2 mg, 0.5 mol %) were dispersed in 1 mL of p-xylene. The suspension was heated to 70 °C for 2 h. The reaction progress was monitored by thin-layer chromatography (TLC) until completion. The resulting product was filtered and purified by chromatographic column. Model compound C For comparison, model compound C was also prepared under the same condition except for that monofunctional compound B (2-pyridinecarboxaldehyde) was used instead of D. 9,9′-Spirobifluorene-3,3′,6,6′-tetraamine (10 mg, 0.026 mmol) and 3-pyridinecarboxaldehyde (28 mg, 0.26 mmol) were added to ethanol (1.5 mL) and chloroform (1.5 mL). The mixture was refluxed overnight and then dropped into petroleum ether. The precipitate was centrifuged and washed with petroleum ether to obtain a light yellow solid (18 mg; 95% yield) product, characterized as follows: Proton nuclear magnetic resonance (1H NMR; 600 MHz, CDCl3, δ): 9.11 (s, 4H), 8.77 (d, J = 61.2 Hz, 4H), 8.64 (s, 4H), 8.35 (d, J = 7.7 Hz, 4H), 7.76 (t, J = 18.1 Hz, 4H), 7.47 (s, 4H),7.07 (d, J = 8.4 Hz, 4H), 6.86 (d, J = 8.4 Hz, 4H). Carbon nuclear magnetic resonance (13C NMR; 101 MHz, CDCl3, δ): 157.3, 154.8, 152.0, 151.7, 150.9, 147.4, 142.4, 135.7, 135.2, 124.7, 121.2, 112.6, 59.5. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS): 733 [M + H]+. Results and Discussion The solid-state 13C crosspolarization magic angle spinning (CP-MAS) NMR spectra spectra of SP-3D-COF-BPY and C showed very similar characteristic peaks indicating similar chemical structures. The resonance signals noted at 156 ppm were attributed to the carbon of the C=N bond ( Supporting Information Figure S1). Moreover, the Fourier transform infrared (FT-IR) spectrum of SP-3D-COF-BPY revealed strong, apparent C=N stretch modes characteristic of imines at 1625 cm−1 ( Supporting Information Figure S2). Thermogravimetric analysis (TGA) showed good thermal stability up to 500 °C ( Supporting Information Figure S3). Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) revealed that the SP-3D-COF-BPY is a bulk crystal with regular continuous edges (Figure 4a and Supporting Information Figure S4). To confirm the proposed rigid channel configuration and the existence of coordination sites within the structure of SP-3D-COF-BPY, we performed PXRD analysis with the help of structural simulation (Figure 1). A tetragonal-disphenoid diamond ( dia) topology lattice with interpenetration was proposed.28 After a detailed comparison between the calculated PXRD profiles from different interpenetrated structures and the experimental profiles ( Supporting Information Figure S5), we finally identified a sevenfold-interpenetrated ( dia-c7) net, highly similar to a previously reported SP-3D-COF-BPH in which a biphenyl ( BPH) unit was used instead of bipyridine ( BPY) (Scheme 1a).28 As shown in Figure 1a, the PXRD pattern of SP-3D-COF-BPY exhibited very intense peaks for the {200} Bragg planes (actual planes {200} at 4.5°, secondary {400} diffraction at 9.0°, and tertiary {600} diffraction at 13.5°). In our simulated unit cell (Figure 1b as stacking of seven unit cells along the interpenetration axis), {200} planes corresponded to the smooth channels' coplanar walls. Undoubtedly, the atom density of these {200} planes was several orders of magnitude higher than any other planes, and hence, exhibited the strongest peak signals. Indeed, this type of pattern (strong peaks solely from one plane group) is very different from those of other 3D COFs and could only be observed in SP-3D-COFs, evidently proving the proposed structure with unobstructed channel configuration. The structural simulation was performed by the Forcite package with geometry optimization at the molecular-mechanics level (Figure 1b). As seen in Figure 1a and inset, the experimental profile (black) correlates well with the predicted profile from the simulated model (green). Pawley refinement revealed minimal differences (Figure 1a, red; Rp = 4.24% and Rwp = 5.02%). Unit cell parameters of SP-3D-COF-BPY were determined from Pawley refinement to be a = 37.9 nm, b = 39.5 nm, c = 6.9 nm, and α = β = γ = 90°, in excellent agreement with the simulated values of a = b = 38.0 nm, c = 7.1 nm, and α = β = γ = 90°. Structures with degrees of interpenetration higher than seven were found to cause structural distortion, and thus, excluded ( Supporting Information Table S1). The degrees of interpenetration —one to seven were compared in detail, exclusively revealing sevenfold as the actual interpenetration mode ( Supporting Information Table S1 and Figure S5). Hence, the structural analysis of the experimental PXRD pattern proved our proposed structure for SP-3D-COF-BPY successfully bearing numerous periodic metal chelation sites and channels. Figure 1 | (a) Indexed experimental (black), Pawley refined (red) PXRD patterns with their difference (blue), and the calculated pattern (green) from dia-c7 net of SP-3D-COF-BPY. Inset: zoomed view of detailed PXRD profile without the primary peak; (b) structural representation of SP-3D-COF-BPY (dia-c7). Top left: ball-and-stick images; top right: the representation of interpenetration; bottom: space-filling model views perpendicular (left) and parallel to 1D channels (right). Download figure Download PowerPoint The sevenfold interpenetration of SP-3D-COF-BPY yielded numerous coordination sites that could chelate metal ions firmly (Scheme 1b). The remarkable advantage of the SP-3D-COF family is that the uniform square pore channel was constructed from rigid orthogonal planar SP units. The channel walls with a high density of discrete coordination sites could efficiently bind metal ions and fully expose the active metal ions to the channel space to collide with substrates (Scheme 1b). Thus, SP-3D-COF-BPY is considered superior in serving as a heterogeneous catalyst to its counterparts due to the unique topological structure. The surface area and porosity of SP-3D-COF-BPY were determined by nitrogen adsorption–desorption analysis at 77 K. As shown in Figure 2a, under low-pressure stage (P/P0 < 0.05), SP-3D-COF-BPY shows type I isotherm with a sharp uptake. The Brunauer–Emmett–Teller (BET) surface area was obtained as SBET = 1945 m2 g−1, which is relatively large and beneficial for the introduction of Pd(II). The total pore volume is Vp = 1.10 cm3 g−1 (calculated at P/P0 = 0.99). The pore size distribution curve clearly shows that the pore-limiting diameter is 1.36 nm (Figure 2b), which agrees with the simulated model (1.56 nm, Figure 1b). Figure 2 | (a) N2 adsorption–desorption isotherms (77 K) of SP-3D-COF-BPY. (b) Pore-size distribution from the quenched solid density functional theory of SP-3D-COF-BPY. Download figure Download PowerPoint The Pd(II) was then introduced into the SP-3D-COF-BPY through a simple wet-chemistry approach (see Supporting Information). Pd(II)@SP-3D-COF-BPY was obtained as an ionic framework, with Pd(II) chelated to the bipyridine moieties and the acetate counter anions located nearby. The metal content adsorbed by the framework was determined to be 14.8 wt % by inductively coupled plasma (ICP) analysis, which is slightly lower than the theoretical value of 17.9%. A comparison of the PXRD patterns of SP-3D-COF-BPY and Pd(II)@SP-3D-COF-BPY indicated that the crystallinity and the periodic structure of Pd(II)@SP-3D-COF-BPY were well kept after introducing Pd(II) ions ( Supporting Information Figure S6). The BET surface area of Pd(II)@SP-3D-COF-BPY decreased to SBET = 640 m2 g−1 due to the introduction of Pd(OAc)2 ( Supporting Information Figure S7), clearly supporting the efficient deposition of Pd(II) in the channels of SP-3D-COF-BPY. The X-ray photoelectron spectroscopy (XPS) was then performed to investigate the oxidation state of the Pd and incorporation with the COF. As shown in Figure 3, the binding energy of Pd 3d3/2 in Pd(II)@SP-3D-COF-BPY (338.4 eV) was very similar to the reference compound Pd(OAc)2 (338.5 eV). The XPS results confirmed the palladium's divalent nature in the COF and efficient coordination with the bipyridine units in the COF. Figure 3 | XPS spectra of the Pd 3d orbital in Pd(II)@SP-3D-COF-BPY (red), Pd(OAc)2 (blue), and [email protected] (green). Download figure Download PowerPoint Notably, no characteristic peak of Pd(0) was observed from the PXRD patterns of Pd(II)@SP-3D-COF-BPY ( Supporting Information Figure S6), which further confirmed the Pd(II) state in Pd(II)@SP-3D-COF-BPY. As for TEM images, it has been reported that the electron beams in TEM would reduce Pd(II) to Pd(0) effectively during analysis,40,41 which also led to the invention of an electron-beam-induced synthetic approach of Pd(0) NPs.42–44 In periodic structures, this phenomenon has been often observed in MOFs.45,46 Recently, Wang's group14 also noticed this in their Pd(II)-containing two-dimensional (2D)-COF system. Indeed, we observed some ultrasmall PdNPs from the TEM of Pd(II)@SP-3D-COF-BPY (Figure 4b). The statistic histogram for randomly selecting more than 100 particles indicated that the Pd NPs had an average size diameter of 1.28 nm with narrow size distributions ( Supporting Information Figure S8). The interplanar spacing of these PdNPs was calculated to be 0.17 nm, consistent with the {100} plane spacing of the Pd cubic close-packed (ccp) structure (Figure 4c). The corresponding selected area electron diffraction (SAED) pattern of diffraction ring revealed the polycrystalline nature of PdNPs and typical diffraction rings for {100} and {110} planes, as well as their secondary diffraction signals, with respect to the crystal structure of Pd NPs. Figure 4 | (a) High-resolution TEM (HR-TEM) image of SP-3D-COF-BPY. (b) HR-TEM image of Pd(II)@SP-3D-COF-BPY. (c) SAED pattern of the Pd(II)@SP-3D-COF-BPY. Insert: HR-TEM image of the Pd(II)@SP-3D-COF-BPY. (d) HR-TEM image of Pd(II)@SP-3D-COF-BPY after five cycles. Download figure Download PowerPoint Recently, COF-supported PdNPs have been commonly reported as heterogeneous catalysts for various coupling reactions.16,47,48 For comparison reasons, we also prepared [email protected] following the literature procedure by reducing Pd(II) ions after their deposition in the COF (for experimental details, see Supporting Information). The XPS results of the [email protected] showed that the binding energies of Pd 3d3/2 and Pd 3d5/2 are located at 341.6 and 336.3 eV, which is 2.1 eV smaller than the Pd(II)@SP-3D-COF-BPY (Figure 3). The metal content adsorbed by the framework was determined to be 11.1 wt % by ICP analysis, slightly lower than that of Pd(II)@SP-3D-COF-BPY. To confirm the critical role of the bipyridine as a chelated metal unit, SP-3D-COF-BPH reported in our previous work was used as a reference (Scheme 1a).28 The structure of SP-3D-COF-BPH is almost identical to SP-3D-COF-BPY except that the linear BPH linker was used in SP-3D-COF-BPH instead of bipyridine. The control experiment was done under precisely the same conditions. As expected, Pd(II) could also be loaded into SP-3D-COF-BPH, given its nanoporous structures. However, the ICP analysis showed that the Pd content was only 3.1 wt %, which is almost fivefold lower than that of SP-3D-COF-BPY (14.8 wt %). Apparently, the bipyridine had a critical effect on the coordination of Pd(II). The XPS experiment also confirmed the divalent nature of the palladium in SP-3D-COF-BPH ( Supporting Information Figure S9). Next, we examined the catalytic activity of Pd(II)@SP-3D-COF-BPY and compared it with Pd(II)@SP-3D-COF-BPH and [email protected]. The Suzuki–Miyaura reaction was chosen as an example.14 The reaction was performed with a catalytic amount of Pd(II)@SP-3D-COF-BPY under typical one-pot Suzuki–Miyaura coupling conditions using a small molecule catalyst Pd(OAc) 2.14,16 Phenylboronic acid (91.5 mg, 0.75 mmol), aryl halide (0.5 mmol), potassium carbonate (138 mg, 1.0 mmol), and Pd(II)@SP-3D-COF-BPY (2 mg, 0.5 mol % effective Pd) were dispersed in 1 mL of p-xylene. The suspension was heated to 70 °C for 2 h. The mixture was centrifuged and Pd(II)@SP-3D-COF-BPY was washed with dichloromethane for the next round. The product was purified by column chromatography over silica gel and calculated the yield (96–97%). Pd(II)@SP-3D-COF-BPY showed excellent catalytic activity with only 0.5 mol % amount of Pd loading under facile conditions (70 °C, 2 h). Various substrates with different substituent groups, such as electron-donating or electron-withdrawing groups with steric-hindrance effectors, were considered. Excellent yields were obtained (96–97%) for all the substrates. For the nitro-substituted substrates, a longer reaction time (12 h) was required to achieve a 95% yield. Even less active bromo-substituted substrates also gave yields up to 98%. A minute catalyst amount (0.1 mol % effective Pd) was then tested in the reaction to determine the catalytic performance further. We found that the yield could still reach 96–98% when the reaction time was extended to 5 h. All these results clearly indicated that Pd(II)@SP-3D-COF-BPY is an excellent catalyst for the Suzuki–Miyaura coupling reaction.14,49,50 The catalytic performance of the reference catalyst Pd(II)@SP-3D-COF-BPH was then investigated. The amount of the catalyst used in the reaction was based on the ICP analysis to make sure that the exact same amount of effective Pd was introduced. As expected, the activity was much lower than Pd(II)@SP-3D-COF-BPY. Under the same conditions (10 mg Pd(II)@SP-3D-COF-BPH with 0.5 mol % effective Pd), the yield reached only ∼20% compared with Pd(II)@SP-3D-COF-BPY (Table 1). In contrast, [email protected] as a catalyst also exhibited lower catalytic activity than that of Pd(II)@SP-3D-COF-BPY. The yield reached ∼60% (Table 1). This superiority of Pd(II)@SP-3D-COF-BPY could be due to the presence of well-ordered metal-binding bipyridine sites (equivalents to ligands) immobilized in the backbone of the COF, which could minimize the agglomeration of the in situ produced Pd(0). Indeed, the "single-atom" catalytic site is well known to be superior to nanoclusters given the same amount of active matter due to the significantly larger number of real "active" sites. To verify the importance of the crystallinity of the SP-3D-COF-BPY, the amorphous sample of the SP-3D-COF-BPY was used as another reference. The catalyst prepared using this amorphous support under the same conditions showed significantly decreased Pd content (6.3% vs. 14.8 wt %, respectively). The product yields in Suzuki–Miyaura coupling reactions decreased by as much as 50%, indicating the significantly lower catalytic performance of the amorphous catalyst under the same Pd loading (0.5 mol % of effective Pd) ( Supporting Information Table S2). These results indicated that the crystallinity of COF's support is critical for Pd loading and catalytic performance. Table 1 | Catalytic Activities of Pd(II)@SP-3D-COF-BPY, Pd(II)@SP-3D-COF-BPH, and [email protected] in the Suzuki–Miyaura Coupling Reactiona Entry R X Time (h) Yield (%)a Yield (%)b Yield (%)c 1 Br 2 98 15 60 2 Br 2 98 13 56 3 Br 2 40d 7 32 4 Br 2 98 14 51 5 I 2 30e 5 18 6 I 2 98 15 58 7 I 2 98 12 43 8f Br 5 97 6 24 a Pd(II)@SP-3D-COF-BPY (2.0 mg, 0.5 mol % effective Pd) as catalyst. b Pd(II)@SP-3D-COF-BPH (10.0 mg, 0.5 mol % effective Pd) as catalyst. c [email protected] (2.6 mg, 0.5 mol % effective Pd) as catalyst. dThe yield reached 98% after 12 h. eThe yield reached 95% after 12 h. f0.1 mol % effective Pd was used. The recyclability and stability of Pd(II)@SP-3D-COF-BPY were then investigated. The catalyst can be easily recycled through centrifugation and washed with solvent. The experiment was performed using 0.5 mol % effective Pd of the catalyst in the reaction of phenylboronic acid and bromobenzene. The results demonstrated that the high catalytic activity of Pd(II)@SP-3D-COF-BPY remained even after five cycles ( Supporting Information Table S3). Highly dispersed PdNPs were again observed from the TEM (Figure 4d). The PXRD patterns of the recycled catalyst showed a slightly increased amorphous solid phase in the range of 2θ ≈ 15–35°. However, the prominent diffraction peaks were mostly maintained, suggesting that the alteration of the crystalline structure was not significant ( Supporting Information Figure S6). The XPS spectra of the recycled Pd(II)@SP-3D-COF-BPY catalyst shows a slight negative shift (0.5 eV) of the Pd(II) peaks compared with the fresh Pd(II)@SP-3D-COF-BPY ( Supporting Information Figure S10). This could be attributed to the partial reduction of Pd(II) to the catalytically active Pd(0) species. The ICP analysis of the recycled Pd(II)@SP-3D-COF-BPY after five cycles showed that the Pd content remained 11.5 wt %. The decrease in Pd loading after multiple recycling (0.6–0.7 wt % loss per cycle) is likely due to the slow leaching of the Pd species, especially those loosely bound or physically adsorbed on the surface and in the pores during the reaction and the subsequent recycling process, which involved stirring in the reaction media for several hours, centrifugation, and repetitive washings. Finally, another control experiment was conducted to evaluate the source of the active Pd(II), either from the Pd(II)@SP-3D-COF-BPY or free Pd(II) detached from the Pd(II)@SP-3D-COF-BPY in the solution. Under exactly the same reaction condition, except for the omission of the substrates, Pd(II)@SP-3D-COF-BPY was heated up to 70 °C for 2 h. The filtrate from the mixture was used to catalyze the reaction of phenylboronic acid and bromobenzene. No conversion was observed. All these results further demonstrate that the Pd(II) was tightly bound in the SP-3D-COF-BPY. Even under harsh conditions of Suzuki–Miyaura coupling reaction, negligible amount
Referência(s)