Two Co(II)-Based MOFs Constructed from Resorcin[4]Arene Ligand: Syntheses, Structures, and Heterogeneous Catalyst for Conversion of CO2

Two Co(II)-based metal–organic frameworks (MOFs) with open channels, [(CH3)2NH2]2[Co5L(H2O)8]·4H2O (1) and [Co6L(DMF)2(H2O)8]·2H2O (2), were synthesized using resorcin[4]arene ligand (H12L). Compounds 1 and 2 exhibit different 3D microporous framework structures: 1 possesses two kinds of open channels parallel to the a-axis (ca. 5.0 × 5.0 Å) and the b-axis (ca. 4.0 × 6.0 Å), and 2 is an open framework with a window size of 5.6 × 5.6 Å. The activated crystal 1 involves many Lewis acid sites; thus, 1 shows prominent activity and recyclability for the reaction of carbon dioxide coupled with epoxides. Most strikingly, catalyst 1 can be reused for five successive cycles and provides outstanding catalytic activity.

Metal-organic frameworks (MOFs), as a kind of functional material, have attracted tremendous interest due to their large surface area, tunable pore structure, and rich active Lewis acid sites [33][34][35]. MOFs have a high adsorption capacity for CO 2 , which can increase the concentration of CO 2 around the catalytic active sites. Additionally, framework pores can provide confined space for CO 2 reaction [36,37]. Organic linkers play a crucial part in the synthesis of MOFs with a variety of pore size and chemical environments [38][39][40]. For this application, resorcin [4]arene is especially attractive because of its multiple coordinate sites and tunable structure. Many elegant structures have been obtained using functionalized resorcin [4]arene ligands [41][42][43][44][45][46][47][48].
Herein, we selected a functionalized-resorcin [4]arene (H 12 L) as a ligand, with twelve carboxylate groups in one ligand, so it has multiple possible coordination modes with metal ions.

Materials and Methods
All the raw materials were obtained commercially. The method through which the H12L ligand was synthesized is consistent with the literature [49]. The PXRD patterns of 1 and 2 were collected using Cu Kα radiation (λ = 0.154 nm) on a Rigaku Dmax 2000 X-ray diffractometer. 1 H NMR spectra were captured on a Bruker 600 MHz spectrometer in CDCl3 or DMSO-d6. TGA data were obtained using a TGA5500 analyzer (5 °C min −1 , 25-600 °C, N2 flow). The C, H, and N elemental analyses were performed using a Vario MACRO cube analyzer. IR spectra were collected on a Thermo Scientific Nicolet 10. The CO2 gas sorption was performed on V-Sorb 2800S.

Materials and Methods
All the raw materials were obtained commercially. The method through which the H 12 L ligand was synthesized is consistent with the literature [49]. The PXRD patterns of 1 and 2 were collected using Cu Kα radiation (λ = 0.154 nm) on a Rigaku Dmax 2000 X-ray diffractometer. 1 H NMR spectra were captured on a Bruker 600 MHz spectrometer in CDCl 3 or DMSO-d6. TGA data were obtained using a TGA5500 analyzer (5 • C min −1 , 25-600 • C, N 2 flow). The C, H, and N elemental analyses were performed using a Vario MACRO cube analyzer. IR spectra were collected on a Thermo Scientific Nicolet 10. The CO 2 gas sorption was performed on V-Sorb 2800S.  3.90; N, 1. 26. Found: C, 43.99; H, 3.86; N, 1.22. IR

Coupling of CO 2 with Epoxides
To obtain the activated sample, catalyst 1 was immersed in acetone for 12 hours and then dried at 60 • C for 10 hours under vacuum. The reactions were executed in a 15 mL flask, the reaction system was refreshed with CO 2 three times, and then the CO 2 pressure was maintained at 1 atm. Epoxide (5 mmol), catalyst 1 (30 mg, 0.0014 mmol), and n-Bu 4 NBr (0.16 g, 0.50 mmol) were mixed in the flask, and then stirred at 80 • C for 8 hours. The conversion of the reactions were calculated by 1 H NMR.

X-ray Crystallography
Diffraction data for compounds 1 and 2 were recorded at room temperature using an Oxford Diffraction Gemini R CCD diffractometer with Mo Kα radiation (λ = 0.71073 Å). The structures of 1 and 2 were solved by direct methods (SHELXS-2014) and refined on F 2 by full-matrix least-squares using the SHELXS-2014 [50][51][52]. The solvent molecules were highly disordered, so the produced diffused electron densities were removed using the SQUEEZE program in PLATON [53]. Based on the TGA, electron diffraction density, and elemental analysis results, the solvent molecules were directly merged into the final molecular formula. The reflection peaks of hydrogen atoms on the solvent molecules were too weak to assign, so they were directly enclosed in the final molecular formula. Non-H atoms were refined anisotropically. Crystallographic data for 1 (CCDC 2078907) and 2 (CCDC 2078908) are summarized in Table 1, Tables S1 and S2.  10.5320(6) 11.2564(6) b (Å) 13.2619(7) 15.9785(9) c (Å) 18.1118(10) 16  Compound 1 crystallizes in the triclinic system with space group P-1. Because of the disordered solvents, the SQUEEZE program in PLATON was used during the refinement. There are twelve water molecules and two [(CH 3 ) 2 NH 2 ] + cations, produced by the decomposition and protonation of DMF, in the structure [54,55], which was calculated by elemental analysis, TGA, and electron diffraction density. The asymmetric structure of 1 is composed of two and a half CoII cations (Co1, Co2, and Co3), half a L 12− ligand, and four coordinated water molecules ( Figure 1a). All the Co II cations adopt a six-coordinate mode but different coordination environments. Co1 is coordinated with four water molecules and two O atoms from two L 12− ligands; the occupancy of Co1 is 0. 5. Co2 is linked with six O atoms from four L 12− ligands. Co3 is surrounded by two water molecules and but different coordination environments. Co1 is coordinated with four water molecules and two O atoms from two L 12− ligands; the occupancy of Co1 is 0. 5. Co2 is linked with six O atoms from four L 12− ligands. Co3 is surrounded by two water molecules and four O atoms from three L 12− ligands. As shown in Figure 1b, each L 12− ligand bridges sixteen Co II cations. In this manner, 1 shows a three-dimensional structure. As displayed in Figure  1c

Structure of [Co6L(DMF)2(H2O)8]·2H2O (2)
The crystal 2 belongs to the triclinic system with space group P-1. The SQUEEZE function was used to remove the disordered solvents. The asymmetric structure of 2 comprises three CoII cations (Co1, Co2, and Co3), half a L 12− ligand, and four coordinated water molecules (Figure 2a). Compared with Co2 and Co3, Co1 shows different coordination spheres: Co1 is coordinated with one coordinated water molecule and five O atoms from five L 12− ligands; Co2 and Co3 both adopt a six-coordinate mode with one coordinated water molecule, one DMF molecule, and four O atoms from three L 12− ligands. The Co-O bond lengths vary from 1.993 (4) The crystal 2 belongs to the triclinic system with space group P-1. The SQUEEZE function was used to remove the disordered solvents. The asymmetric structure of 2 comprises three CoII cations (Co1, Co2, and Co3), half a L 12− ligand, and four coordinated water molecules ( Figure 2a). Compared with Co2 and Co3, Co1 shows different coordination spheres: Co1 is coordinated with one coordinated water molecule and five O atoms from five L 12− ligands; Co2 and Co3 both adopt a six-coordinate mode with one coordinated water molecule, one DMF molecule, and four O atoms from three L 12− ligands. The Co-O bond lengths vary from 1.993 (4)   into an open framework. The window size is 5.6 × 5.6 Å along the a axis (Figure 2c,d). The solvent-accessible volume of compound 2 is ca. 20.0% based on the PLATON calculation.
The 3D structures of 2 down the a-axis (c) and caxis (d).

Characterization of the Crystal Structure of 1 and 2
The TGA of compounds 1 and 2 was conducted under a N2 atmosphere. As displayed in Figure 3a, the TGA of compound 1 indicated that the weight loss before 240 °C is due to the DMF molecules and water molecules, and the framework begins to collapse after 240 °C. The TGA of compound 2 shows that the weight loss before 300 °C belongs to the DMF molecules and water molecules, then the weight loss from 300 °C is attributable to the framework decomposition. The PXRD pattern of 1 is consistent with the simulated one, which indicates that 1 is stable in air. Some characteristic peaks disappeared in the PXRD pattern of 2, which may have occurred due to an optimum growth orientation being chosen. The CO2 adsorption was performed at 273 K ( Figure S4), and the CO2 uptake capacity was found to be ca. 0.48 mmol/g.
The 3D structures of 2 down the a-axis (c) and c-axis (d).

Characterization of the Crystal Structure of 1 and 2
The TGA of compounds 1 and 2 was conducted under a N 2 atmosphere. As displayed in Figure 3a, the TGA of compound 1 indicated that the weight loss before 240 • C is due to the DMF molecules and water molecules, and the framework begins to collapse after 240 • C. The TGA of compound 2 shows that the weight loss before 300 • C belongs to the DMF molecules and water molecules, then the weight loss from 300 • C is attributable to the framework decomposition. The PXRD pattern of 1 is consistent with the simulated one, which indicates that 1 is stable in air. Some characteristic peaks disappeared in the PXRD pattern of 2, which may have occurred due to an optimum growth orientation being chosen. The CO 2 adsorption was performed at 273 K ( Figure S4), and the CO 2 uptake capacity was found to be ca. 0.48 mmol/g.

Coupling of CO2 with Epoxides.
Given the high-density Lewis acid sites and high yield of compound 1, the heterogeneous catalytic performance of 1 was investigated for the coupling reaction of CO2 with

Coupling of CO 2 with Epoxides
Given the high-density Lewis acid sites and high yield of compound 1, the heterogeneous catalytic performance of 1 was investigated for the coupling reaction of CO 2 with epoxides. As shown in Scheme 2 and Table 2, gylcidylphenylether was selected as a typical substrate to obtain the optimum reaction conditions. Firstly, the reaction between the gylcidylphenylether and CO 2 was performed in the presence of activated catalyst 1 (10 mg) and n-Bu 4 NBr (0.16 g) at 80 • C for 1 hour; the conversion was only 24% (entry 1, Figure  S1a). Thus, the catalyst amount was increased from 10 to 20 and 30 mg, and the corresponding conversions were increased from 24% to 26% and 32%, respectively (entries 2 and 3, Figure S1b,c). The conversion was only 23% when the n-Bu 4 NBr was absent, which indicted that n-Bu 4 NBr is an important co-catalyst (entry 4, Figure S1d). When the reactions were performed at 25 and 50 • C, the conversions were 0% and 12%, respectively (entries 5 and 6, Figure S1e,f). To improve the conversion, the reaction time was extended to 2, 4, 6, and 8 hours; the corresponding conversions were 51%, 80%, 83%, and 98%, respectively (entries 7-10, Figure S1g-j). These experimental results showed that the optimum reaction conditions are 30 mg catalyst 1, 0.16 g n-Bu 4 NBr, and 1 atm CO 2 reaction at 80 • C for 8 h.

Coupling of CO2 with Epoxides.
Given the high-density Lewis acid sites and high yield of compound 1, the heterogeneous catalytic performance of 1 was investigated for the coupling reaction of CO2 with epoxides. As shown in Scheme 2 and Table 2, gylcidylphenylether was selected as a typical substrate to obtain the optimum reaction conditions. Firstly, the reaction between the gylcidylphenylether and CO2 was performed in the presence of activated catalyst 1 (10 mg) and n-Bu4NBr (0.16 g) at 80 °C for 1 hour; the conversion was only 24% (entry 1, Figure S1a). Thus, the catalyst amount was increased from 10 to 20 and 30 mg, and the corresponding conversions were increased from 24% to 26% and 32%, respectively (entries 2 and 3, Figure S1b,c). The conversion was only 23% when the n-Bu4NBr was absent, which indicted that n-Bu4NBr is an important co-catalyst (entry 4, Figure S1d). When the reactions were performed at 25 and 50 °C, the conversions were 0% and 12%, respectively (entries 5 and 6, Figure S1e,f). To improve the conversion, the reaction time was extended to 2, 4, 6, and 8 hours; the corresponding conversions were 51%, 80%, 83%, and 98%, respectively (entries 7-10, Figure S1g-j). These experimental results showed that the optimum reaction conditions are 30 mg catalyst 1, 0.16 g n-Bu4NBr, and 1 atm CO2 reaction at 80 °C for 8 hours.   Different epoxides were selected as the substrates to further examine the applicability of catalyst 1. The conversions of epichlorohydrin 2-ethyloxirane, 2-butyloxirane, 2-(butoxymethyl)oxirane, 1,2-epoxyethylbenzene and benzylglycidylether were 99%, 99%, 99%, 95%, 75%, and 93%, respectively (entries 1-6 in Table 3, Figure S2a-f). The reaction rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figure S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction.

Entry
Epoxides Products Conversion (%) b 1 99%, 95%, 75%, and 93%, respectively (entries 1-6 in Table 3, Figure S2a-f). The reaction rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). 99%, 95%, 75%, and 93%, respectively (entries 1-6 in Table 3, Figure S2a-f). The reaction rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). 99 2 99%, 95%, 75%, and 93%, respectively (entries 1-6 in Table 3, Figure S2a-f). The reaction rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). 99%, 95%, 75%, and 93%, respectively (entries 1-6 in Table 3, Figure S2a-f). The reaction rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). 99 3 99%, 95%, 75%, and 93%, respectively (entries 1-6 in Table 3, Figure S2a-f). The reaction rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). 99%, 95%, 75%, and 93%, respectively (entries 1-6 in Table 3, Figure S2a-f). The reaction rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). 99 4 99%, 95%, 75%, and 93%, respectively (entries 1-6 in Table 3, Figure S2a-f). The reaction rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). 99%, 95%, 75%, and 93%, respectively (entries 1-6 in Table 3, Figure S2a-f). The reaction rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). 95 5 rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). 75 6 rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). rate did not decrease when the length of the alkyl chain increased; these experimental results showed that the substrates do not enter the channel but react on the surface of the catalyst (entries 2-4 in Table 3, Figure S2b-d). Compared with other substrates, the reaction of 1,2-epoxyethylbenzene is relatively slow, which may due to the steric-hinderance effect (entries 5 in Table 3, Figures S2e). These results confirmed that the epoxides with variable alkyl chains or aromatic rings are all suitable substrates for the reaction. To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figures 4 and S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure  S5). To further explore the recyclability and stability of 1, recycling experiments were performed using gylcidylphenylether as the substrate. After the reaction, 1 was recovered from the mixture by centrifugation and filtration and then washed with dichloromethane three times. The catalyst 1 recovered from the reaction was reused for five consecutive cycles with conversions of more than 90% (Figure 4 and Figure S3a-c). This result proved that 1 is a heterogeneous and recyclable catalyst. The PXRD pattern of activated 1 is not consistent with the simulated one, which may due to the single-crystal to single-crystal transition [56]. However, the PXRD pattern of activated 1 after five recycles is consistent with the activated sample, which indicated that 1 is stable during the reaction process ( Figure S5).

Conclusions
In summary, we synthesized two microporous MOFs (1 and 2) using Co(II) cations and functionalized resorcin [4]arene. The synthesized compounds 1 and 2 were characterized by single-crystal X-ray diffraction analysis, PXRD, IR, TGA, and elemental analysis. The activated catalyst 1 possesses a large number of unsaturated coordination Co II cations; thus, compound 1 is a promising heterogeneous catalyst for the CO2 conversion reaction. Most strikingly, 1 can be easily recovered and reused for five consecutive circles with high

Conclusions
In summary, we synthesized two microporous MOFs (1 and 2) using Co(II) cations and functionalized resorcin [4]arene. The synthesized compounds 1 and 2 were characterized Crystals 2021, 11, 574 8 of 10 by single-crystal X-ray diffraction analysis, PXRD, IR, TGA, and elemental analysis. The activated catalyst 1 possesses a large number of unsaturated coordination Co II cations; thus, compound 1 is a promising heterogeneous catalyst for the CO 2 conversion reaction. Most strikingly, 1 can be easily recovered and reused for five consecutive circles with high catalytic activity.