English

• Metal–organic frameworks (MOFs) are an emerging class of porous crystalline materials constructed from inorganic nodes (metal ions/clusters) with organic linkers through coordination bonds [1-3]. Due to their unique properties of diverse composition, large surface area, and ease of functionalization, MOFs have interesting applications in gas adsorption and separation, energy storage, molecular sensing, optoelectronics, heterogeneous catalysis, and so on [4-9].

  In recent years, environmental pollution caused by the consumption of fossil energy and emissions from the chemical industry has become a global concern [10, 11]. For example, excessive emissions of greenhouse gases, especially carbon dioxide (CO₂), are considered to be the main reason of the extreme climate in recent years, such as melting glaciers and global warming [12-16]. In addition, various toxic organic dyes or intermediates produced by the chemical industry are commonly detected in the groundwater and arable land, with negative effects on human health through the biological chain [17-19]. To ameliorate these issues, assorted MOF-based materials with highly porous structures and functional groups have been designed and prepared, and some success has been achieved through absorption, conversion, or degradation [20-22]. A key factor determining the performance of MOFs during their usage is the appropriate interaction between waste molecules and metal sites in the framework [23, 24]. In particular, MOFs with unsaturated metal sites have shown excellent performance in the field of adsorption and catalysis due to the unique structural and electronic characteristics [25-29]. Unfortunately, the metal nodes in most MOFs structures are usually blocked by organic linkers, leaving no coordination active sites available for the binding and activation of substrates, which limits their practical application.

  Fortunately, the high structural designability and easy functionalization of MOFs make it an ideal platform in a variety of applications [30, 31]. From the structural viewpoint of MOFs, metal clusters can provide more coordination sites during the organization of the framework and therefore generate unsaturated metal sites [32-34], although it is challengeable due to the flexible coordination modes. In this work, two new Co-based MOFs were solvothermal synthesized and structural characterized. Interestingly, the same metal nodes and organic ligands assembled but different solvents gave two MOFs with distinct topologies. Their potential usage in the CO₂ selective adsorption and chemical transformation as well as dye adsorption was evaluated in detail.

  Single-crystal diffraction analysis reveals that HL-8 crystallizes in the orthorhombic system, space group Pna2₁ (Table S1 in Supporting information). As exhibited in Fig. S1a (Supporting information), there exists one crystallographically independent Co(Ⅱ) center, which is coordinated by four carboxylate O atoms from two L^3– ligands and two imidazole N atoms from two dibp ligands, showing a tetrahedral geometry. From Figs. S1b–d (Supporting information), each HL^2– ligand as a µ₂-bridging linker with µ₂-ƞ¹: ƞ¹ coordination mode connects Co(Ⅱ) forming a 1D chain structure and each didp ligand linking two Co(Ⅱ) also from another chain structure. Finally, the two kinds of chains weave together to generate the final 3D framework of HL-8 (Fig. S1e in Supporting information). Due to the stemming of the uncoordinated -COOH groups, HL-8 reveals a low porosity, and the solvent accessible volume of HL-8 is 306.1 ų per 3275.6 ų unit cell volume (9.3% of the total crystal volume). Moreover, if the Co(Ⅱ) center is viewed as 4-connected node, the structure of HL-8 can be simplified as 4-connected sqc5 type topology with the point symbol {6⁵·8} (Fig. S1f in Supporting information).

  Different from HL-8, HL-9 crystallizes in the monoclinic P2₁/n space group (Table S1) and has a novel three-dimensional (3D) porous structure. In the structure, there exist two crystallographically independent Co atoms (Fig. S1a). Co1 is coordinated by three O atoms from carboxylates of different L^3– ligands, two μ₃-OH entities, one N atom from dibp ligand, while Co2 is coordinated by three O atoms from carboxylates of different L^3– ligands, one μ₃-OH entities, one N atom from dibp ligand and one O from a terminal H₂O. Two μ₃-OH and six -COO^– groups connecting four Co atoms form a Co₄(OH)₂(COO)₆ cluster (Fig. 1A), and each L^3– ligand in a µ₆-ƞ²: ƞ²: ƞ² coordination manner links three Co₄ clusters forming a 3D framework with 1D rhombic channels along a axis (Figs. 1B and C). Each dibp ligand connects to two Co₄ clusters and inserts to the rhombic channels generating the final framework structures of HL-9 (Fig. 1D) with a total solvent-accessible volume of 34.2% estimated by PLATON [34]. Interestingly, due to the insertion of dibp ligands, the framework of HL-9 is split into finite cage-like voids (Figs. 1E and F). Topologically, L^3– ligand can be seen as a 3-connected linker and Co₄ cluster as an 8-connected node, the 3D framework of HL-9 can thus be simplified as a novel (3,8)-connected network with the point symbol of {4·5²}₂{4²·5¹⁰·6¹²·7·8³} (Fig. S2b in Supporting information).

  Figure 1

  

  Figure 1.  Structures of HL-9: (A) The Co₄ cluster formed by μ₃-OH and -COO^– groups. (B) The coordination mode of the -COO^– groups in L^3– ligands. (C) The porous framework with one-dimensional (1D) rhombic channels formed by the connection of Co₄ cluster and L^3– ligands. (D) The final structures of HL-9 with dibp ligands splitting the rhombic channels. (E, F) Formation of the cage structures inside the framework.

  DownLoad: Full-Size Img PowerPoint

  The synthesized samples were further characterized by powder X-ray diffraction (PXRD) and thermogravimetric analysis (TG) to confirm the purity of crystal phases as well as solvent stability in potential applications. The characteristic peaks of PXRD patterns for both HL-8 and HL-9 match well with corresponding simulated ones, as demonstrated in Fig. S3 (Supporting information), confirming the high purity of the crystals. In addition, the position and intensity of each peak were almost unchanged after soaking both compounds in water and different dye solvents, proving their crystallinity was maintained. The high chemical stability is of vital importance for their practical usage [35, 36]. TG curves under nitrogen atmosphere indicated that HL-8 and HL-9 can be thermal stable up to 366 ℃ and 405 ℃ (Fig. S4 in Supporting information), respectively, confirming their excellent thermal stabilities and wide application potential under mild conditions.

  In view of the interesting gas adsorption behavior of MOF-based materials, the gas adsorption properties of the as-synthesized compounds were investigated. As shown in Fig. S5A (Supporting information), negligible adsorption for nitrogen, methane, and carbon dioxide was obtained over HL-8, which should be caused by its low porosity and solvent accessible volume. As for HL-9, inferior nitrogen and methane uptake is also observed (Fig. S5B in Supporting information). However, as displayed in Fig. 2A, HL-9 exhibited superior CO₂ adsorption behavior under low pressure (1 atm). The amount of CO₂ uptake on HL-9 is 72.7 cm³/g, which is 52 times that of HL-8 (1.4 cm³/g). The calculated isosteric heats of CO₂ adsorption (Q_st) at zero coverage was calculated in a range of 27.2−36.5 kJ/mol, by fitting the CO₂ isotherms at 273 and 298 K with the virial method, which reflects a good interaction between the CO₂ molecular and HL-9 (Fig. S6 in Supporting information). Based on the results in previous works [37, 38], the unique CO₂ adsorption behavior on HL-9 should be correlated to the existence of Co₄ clusters and cage-like voids in the framework, which may generate interaction with CO₂ molecules and therefore result in high adsorption performances.

  Figure 2

  

  Figure 2.  (A) CO₂ sorption isotherms of HL-8 and HL-9 at 273 K. (B) Catalytic performance in CO₂ cycloaddition with epoxides over HL-8 and HL-9 and (C) recyclability of HL-9 in CO₂ cycloaddition process.

  DownLoad: Full-Size Img PowerPoint

  Taking advantage of the excellent CO₂ adsorption property and relatively high thermal/solution stability, the potential usage of the obtained compounds in CO₂ catalytic transformation was further explored. Till now, a variety of CO₂ harvesting and chemical fixation methods have been explored by chemists and some successes have been achieved in converting CO₂ into valuable energy and/or chemicals [39]. Among the chemical conversion routes, the synthesis of value-added epoxy carbonates through cycloaddition of CO₂ with epoxides was widely studied by researchers due to its high reactivity under mild conditions as well as the wide usage of the final products as solvents, organic intermediates, and electrolytes [37, 40-42]. The efficient adsorption and subsequent activation of CO₂ is the key to the above transformation.

  Fig. 2B and Table S3 (Supporting information) summarized the catalytic performance of the different catalysts in cycloaddition of CO₂ with various epoxides. No product was obtained in the absence of any catalysts or with only ligands of HL-9 as catalyst. Traditional tetrabutylammonium bromide (TBAB) catalyst also exhibited inferior activity with low epoxy carbonates yields, which was in agreement with the previous literature results [42, 43]. Interestingly, the addition of the Co-containing compounds greatly improves the conversion of reactants under the same conditions, confirming the synergism of the compounds and TBAB in the catalytic process. The yield of propylene carbonate is 93.2% and 70.1% for HL-9 and HL-8, respectively, confirming the superior performance of HL-9 under mild conditions. This value is comparable to most literature-reported MOFs (Table S3). Notable, a 91.7% yield of propylene carbonate was also obtained when using homogeneous Co(NO₃)₂ as a catalyst, confirming the Co species is the main active site for CO₂ catalytic transformation. In view of the recovery and reuse of catalysts, the heterogeneous nature of HL-9 is advantageous for practical application.

  In addition, HL-9 also exhibited size-selective catalytic behavior in the title reaction. As shown in Fig. 2B, a decrease in the yield of epoxy carbonate was observed when we increase the molecular size of epoxides substrates (the yield is 93.2%, 70.7%, 55.8%, and 21.3% for 4-methyl-1,3-dioxolan-2-one, 4-(chloromethyl)-1,3-dioxolan-2-one, 4-ethyl-1,3-dioxolan-2-one and 4-((allyloxy)methyl)-1,3-dioxolan-2-one, respectively), which should be correlated with the limited diffusion of larger epoxides in the finite cage-like voids and channels of the framework. Additionally, cyclic catalysis of HL-9 maintained a substrate yield higher than 90% in 3 cycles (Fig. 2C), confirming its usage potential in practical applications.

  Inspired by the unique size-selective behavior of HL-9 in catalytic transformation, the dye selective capture ability was also evaluated using three selected dyes with different sizes: methyl orange (MO), methylene blue (MB), and rhodamine B (RhB). Similar to gas adsorption results, as shown in Fig. 3A and Fig. S7 (Supporting information), HL-8 displayed neglected adsorption for all dyes. In contrast, as displayed in Fig. 3B, the orange solution of MO faded to colorless in only 35 min with the addition of HL-9 under same conditions. The significant decrease in color intensity was consistent with the UV-vis spectroscopy results, confirming its outstanding performance in rapid adsorption of MO. The saturated capacity and apparent rate constants (k) of MO adsorption over HL-9 were 72.4 mg/g and 44.0 × 10⁻³ min⁻¹, respectively (Fig. S8 in Supporting information). In addition, we also investigated the recyclability of HL-9 in MO adsorption. It is interesting that, as shown in Fig. 3C, HL-9 could be regenerated by releasing MO in N, N-dimethylacetamide (DMA) solution and reused at least three times with > 95% MO adsorption efficiency (Fig. 3D). The shift in the position of the absorption peak of MO is caused by the different solvents. Besides, the PXRD patterns of RhB@HL-9, MB@HL-9, and MO-released HL-9 after the 3^rd cycle still remained the same as the as-synthesized samples (Fig. S3B in Supporting information), confirming its high stability during the adsorption process.

  Figure 3

  

  Figure 3.  Adsorption performance and color changes during adsorption of MO in the presence of (A) HL-8 and (B) HL-9. (C) The UV-vis spectra of MO solution desorbed in different solvents and (D) reusability of MO adsorption over HL-9. MO selective adsorption performance from the mixture solution of (E) MO + RhB and (F) MO + MB.

  DownLoad: Full-Size Img PowerPoint

  As for RhB and MB solution, only a slight change in color and UV-vis absorbance was observed after adding HL-9 for 180 min (Fig. S9 in Supporting information). By comparing the open sizes of the cage-like voids of HL-9 and the molecular sizes of MO and RhB, it is suitable for HL-9 to adsorb and sieve MO molecule with smaller size and narrower shape than RhB (Fig. S10 in Supporting information). The above results unveiled its potential usage in selective adsorption of MO from the mixture of other dye solutions. Therefore, the competitive adsorption experiments over HL-9 were further performed. As shown in the inset of Figs. 3E and F, the colors rapidly changed from pink for MO + RhB to purple for RhB, and from light green for MO + MB to green for MB, respectively, after immersion of HL-9 in the mixed dye solutions. Correspondingly, all UV-vis spectra revealed that the absorption peaks of MO decreased and gradually disappeared within 35 min without obvious change in the absorbance of RhB or MB. The outstanding absorption performance of HL-9 highlights its potential application for size-selective separation towards the MO dyes in wastewater.

  In summary, two new Co-containing MOFs were synthesized with high thermal/solution stability. Benefited from the Co₄ cluster nodes and suitable pore in the framework of HL-9, superior performance and interesting size selectivity were achieved in the applications of CO₂ adsorption/catalytic transformation and dye molecular adsorption. This work underlines the importance of parameter regulation during MOF synthesis, which may be a promising direction for the construction of novel MOFs with multifunctional performance.

  Declaration of competing interest

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

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (Nos. 21902022, 21601028, 81903501 and 22061019), Qingchuang Science and Technology Plan of Shandong Province (No. 2021KJ054), the Natural Science Foundation of Shandong Province (Nos. ZR2018LB018, ZR2019QB026 and ZR2020KB014) and Scientific Research Foundation of Dezhou University (Nos. 30101905, 30102708 and 30102701).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2022.03.091.

  
  
• 1. [1]

     H. Furukawa, K.E. Cordova, M. O'Keeffe, O.M. Yaghi, Science 341 (2013) 1230444. doi: 10.1126/science.1230444

  2. [2]

     X.J. Kong, J.R. Li, Engineering 7 (2021) 1115–1139. doi: 10.1016/j.eng.2021.07.001

  3. [3]

     G. Huang, L. Yang, Q. Yin, et al., Angew. Chem. Int. Ed. 59 (2020) 4385–4390. doi: 10.1002/anie.201916649

  4. [4]

     L. Li, Z.B. Fang, W. Deng, et al., CCS Chem. 3 (2021) 2839–2849.

  5. [5]

     Y. Zhou, L. Qin, M.K. Wu, L. Han, Cryst. Growth Des. 18 (2018) 5738–5744. doi: 10.1021/acs.cgd.8b00895

  6. [6]

     F. Bi, X. Zhang, J. Chen, Y. Yang, Y. Wang, Appl. Catal. B: Environ. 269 (2020) 118767. doi: 10.1016/j.apcatb.2020.118767

  7. [7]

     D.S. Zhang, Q. Gao, Z. Chang, et al., Adv. Mater. 30 (2018) 1804715. doi: 10.1002/adma.201804715

  8. [8]

     J.Q. Liu, Z.D. Luo, Y. Pan, et al., Coord. Chem. Rev. 406 (2020) 213145. doi: 10.1016/j.ccr.2019.213145

  9. [9]

     W. Du, Y.L. Bai, Z. Yang, et al., Chin. Chem. Lett. 31 (2020) 2309–2313. doi: 10.1016/j.cclet.2020.04.017

  10. [10]

      L. He, M.X. Li, F. Chen, et al., J. Hazard. Mater. 417 (2021) 126113. doi: 10.1016/j.jhazmat.2021.126113

  11. [11]

      J. He, Y. Xu, Z. Xiong, et al., Chemosphere 256 (2020) 127056. doi: 10.1016/j.chemosphere.2020.127056

  12. [12]

      H. Hu, D. Zhang, H. Liu, et al., Chin. Chem. Lett. 32 (2021) 557–560. doi: 10.1016/j.cclet.2020.02.028

  13. [13]

      B. Han, X. Ou, Z. Deng, et al., Angew. Chem. Int. Ed. 57 (2018) 16811–16815. doi: 10.1002/anie.201811545

  14. [14]

      Y. Wang, Y. Pu, D. Yuan, et al., ACS App. Mater. Interfaces 11 (2019) 44058–44068. doi: 10.1021/acsami.9b12886

  15. [15]

      X. Zhang, Y. Jin, G. Wang, et al., J. Solid State Chem. 296 (2021) 121979. doi: 10.1016/j.jssc.2021.121979

  16. [16]

      G. Si, X. Kong, T. He, et al., Chin. Chem. Lett. 32 (2021) 918–922. doi: 10.1016/j.cclet.2020.07.023

  17. [17]

      H. Bi, Z. Yin, X. Cao, et al., Adv. Mater. 25 (2013) 5916–5921. doi: 10.1002/adma.201302435

  18. [18]

      L. Geng, G. Li, X. Zhang, et al., J. Solid State Chem. 296 (2021) 121960. doi: 10.1016/j.jssc.2021.121960

  19. [19]

      Q. Wei, B.D. Ge, J. Zhang, et al., Chem. Asian J. 14 (2019) 269–277. doi: 10.1002/asia.201801555

  20. [20]

      Y. Peng, Y. Li, Y. Ban, et al., Science 346 (2014) 1356–1359. doi: 10.1126/science.1254227

  21. [21]

      L. Cao, T. Wang, C. Wang, Chin. J. Chem. 36 (2018) 754–764. doi: 10.1002/cjoc.201800144

  22. [22]

      A.X. Tian, M.L. Yang, Y.B. Fu, J. Ying, X.L. Wang, Inorg. Chem. 58 (2019) 4190–4200. doi: 10.1021/acs.inorgchem.8b03248

  23. [23]

      X. Zhang, H. Cheng, H. Zhang, Adv. Mater. 29 (2017) 1701704. doi: 10.1002/adma.201701704

  24. [24]

      X. Wang, T. Zhang, Y. Li, et al., Inorg. Chem. 59 (2020) 17583–17590. doi: 10.1021/acs.inorgchem.0c02798

  25. [25]

      A. Liu, X. Peng, Q. Jin, et al., ACS Appl. Mater. Interfaces 11 (2019) 4686–4700. doi: 10.1021/acsami.8b20447

  26. [26]

      M. Položij, M. Rubeš, J. Čejka, P. Nachtigall, ChemCatChem 6 (2014) 2821–2824. doi: 10.1002/cctc.201402411

  27. [27]

      S. Aslam, J. Zeng, F. Subhan, et al., J. Colloid Interface Sci. 505 (2017) 186–195. doi: 10.1016/j.jcis.2017.05.090

  28. [28]

      L. Geng, W. Zhou, X. Wang, et al., J. Alloy. Compd. 888 (2021) 161494. doi: 10.1016/j.jallcom.2021.161494

  29. [29]

      R. Mohammadian, M.K. Alavijeh, N. Kamyar, M.M. Amini, A. Shaabani, Polyhedron 156 (2018) 174–187. doi: 10.1016/j.poly.2018.09.029

  30. [30]

      L. Jiao, J. Wang, H.L. Jiang, Acc. Mater. Res. 2 (2021) 327–339. doi: 10.1021/accountsmr.1c00009

  31. [31]

      C. Gu, J. Li, G. Yang, et al., Chin. Chem. Lett. 31 (2020) 2263–2267. doi: 10.1016/j.cclet.2020.02.044

  32. [32]

      M. Peng, C. Dong, R. Gao, et al., ACS Cent. Sci. 7 (2021) 262–273. doi: 10.1021/acscentsci.0c01486

  33. [33]

      Y. Bai, Y. Dou, L.H. Xie, et al., Chem. Soc. Rev. 45 (2016) 2327–2367. doi: 10.1039/C5CS00837A

  34. [34]

      X. Zhang, Y.Z. Zhang, Y.Q. Jin, et al., Inorg. Chem. 59 (2020) 11728–11735. doi: 10.1021/acs.inorgchem.0c01611

  35. [35]

      Y.Z. Zhang, T. He, X.J. Kong, et al., ACS Mater. Lett. 1 (2019) 20–24. doi: 10.1021/acsmaterialslett.9b00021

  36. [36]

      H.Z. Li, D.Y. Du, Y. Sun, F. Wang, J. Zhang, Dalton Trans. 49 (2020) 4690–4693. doi: 10.1039/D0DT00791A

  37. [37]

      J. Li, W.J. Li, S.C. Xu, et al., Inorg. Chem. Commun. 106 (2019) 70–75. doi: 10.1016/j.inoche.2019.05.031

  38. [38]

      R. Dalapati, B. Sakthivel, A. Dhakshinamoorthy, et al., CrystEngComm 18 (2016) 7855–7864. doi: 10.1039/C6CE01704E

  39. [39]

      W.G. Cui, G.Y. Zhang, T.L. Hu, X.H. Bu, Coord. Chem. Rev. 387 (2019) 79–120. doi: 10.1016/j.ccr.2019.02.001

  40. [40]

      H. Chen, T. Hu, L. Fan, X. Zhang, Inorg. Chem. 60 (2021) 1028–1036. doi: 10.1021/acs.inorgchem.0c03134

  41. [41]

      Y.J. Chen, X. Huang, Y. Chen, et al., CCS Chem. 1 (2019) 561–570. doi: 10.31635/ccschem.019.20190028

  42. [42]

      X. Zhang, R.H. Zhang, H. Hu, et al., Dalton Trans. 49 (2020) 2058–2062. doi: 10.1039/C9DT04736K

  43. [43]

      J.J. Lv, F.L. Cao, M.L. Wang, et al., Polyhedron 195 (2021) 114982. doi: 10.1016/j.poly.2020.114982

• 

  Figure 1  Structures of HL-9: (A) The Co₄ cluster formed by μ₃-OH and -COO^– groups. (B) The coordination mode of the -COO^– groups in L^3– ligands. (C) The porous framework with one-dimensional (1D) rhombic channels formed by the connection of Co₄ cluster and L^3– ligands. (D) The final structures of HL-9 with dibp ligands splitting the rhombic channels. (E, F) Formation of the cage structures inside the framework.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (A) CO₂ sorption isotherms of HL-8 and HL-9 at 273 K. (B) Catalytic performance in CO₂ cycloaddition with epoxides over HL-8 and HL-9 and (C) recyclability of HL-9 in CO₂ cycloaddition process.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Adsorption performance and color changes during adsorption of MO in the presence of (A) HL-8 and (B) HL-9. (C) The UV-vis spectra of MO solution desorbed in different solvents and (D) reusability of MO adsorption over HL-9. MO selective adsorption performance from the mixture solution of (E) MO + RhB and (F) MO + MB.

  下载: 全尺寸图片 幻灯片
•