English

• Metal-organic frameworks (MOFs), composed of metal ions or metal clusters as nodes and multi-topic organic ligands as linkers, have attracted much attention due to their fascinating structural diversities and wide potential applications in the areas of gas adsorption and separation, proton conduction, dye separation, catalysis, chiral resolution, fluorescence, and so on [1-11]. Yet despite all that, the instability of most of MOFs, such as the collapse of frameworks under high temperature conditions due to guest removal, has been a catastrophic shortcoming for their applications. Thus, continuous efforts have been devoted to create highly stable MOFs during the past decades, however, such materials are still limited to date.

  Up to data, many strategies to build up stable MOFs have been explored, including 1) the strengthening of coordination bonds between metal ions and organic ligands by the use of ligands with higher pKa or/and highly charged metal cations (such as Cr³⁺, Al³⁺, Zr⁴⁺, Sc³⁺, etc.); 2) the introduction of hydrophobic functional groups or polymers for constructing water-stable MOFs via in situ synthesis or post-synthesis modification; 3) the employment of multi-nuclear metal clusters as building units for synthesizing stable cluster-based MOFs [12-21].

  In MOF chemistry, carboxylic ligands are one kind of the most common ligands, which have built a vast library of porous MOFs. However, only a small part of them are thermally stable above 500 ℃, such as UiO-66, UiO(bpdc) and MIL-53·Al [22-24]. In contrast, phosphonate ligands usually can form relatively stable metal-phosphonate frameworks [25], while, most of phosphonatebased frameworks are not or less porous [26]. So, the phosphonocarboxylate ligands with integrated functional groups of phosphonates and carboxylates could be expected to form highly stable MOFs with high porosity.

  Here, we report a rare porous, chiral, and neutral 3D Znphosphonocarboxylate framework Zn₄(μ₃-OH)₂(DMF)(TPO)₂ (1), H₃TPO = tris-(4-carboxylphenyl)phosphineoxide, DMF = N, N-dimethylformamide) with a very high thermal stability up to at least 500 ℃. Interestingly, the similar reaction to 1 but in the absence of additional water can produce a new porous, chiral but anionic 3D Zn-phosphonocarboxylate framework [(CH₃)₂NH₂] [Zn₃(μ₃-OH)(TPO)₂] (2). While, upon 2 is soaked in water, it can undergo phase rebuilding to generate another porous, chiral, and neutral 3D Zn-phosphonocarboxylate framework Zn₄(μ₃-OH)₂(H₂O)₂(TPO)₂ (3) in a single-crystal-to-single-crystal mode. Moreover, the similar reaction to 1 but by the replacement of Zn²⁺ with Co²⁺ can lead to a porous, neutral but achiral 3D Cophosphonocarboxylate framework Co₄(μ₃-OH)₂H₂O)₄(TPO)₂ (4). Surprisingly, although compounds 1-4 crystallize in different space groups and have different secondary building units (SBUs), compositions, porosities, or framework charge properties, their frameworks represent the same type of CaF₂-type topology. Note that bulk samples of 1-3 are racemic mixtures and thus do not exhibit any chiral features.

  Compound 1 crystallizes in the orthorhombic chiral space group P2₁2₁2, obtained by the reaction of Zn(Ac)₂·2H₂O and H₃TPO in a DMF/CH₃CN/H₂O mix-solvent at 100 ℃ for 5 days. X-ray diffraction analysis shows structure 1 is constructed from tetranuclear [Zn₄(OH)₂(DMF)]⁶⁺ SBUs (denoted as Zn₄-A, Fig. 1) joined together by square-pyramid-like TPO^3- ligands, giving a cluster-based 3D porous metal-phosphonocarboxylate framework with 1D double-walled rhombic channels (ca. 5 ×7 Å) along the c-axis direction. A side view of the 1D channels indicates that they consist of square cages with a diameter of 11 Å formed by eight Zn₄-A SBUs crosslinked by TPO^3- ligands. The Zn₄-A in a boat-type conformation contains four crystallographically independent Zn²⁺ ions linked together by two μ₃-OH^- hydroxyl groups. The four independent Zn²⁺ ions show two kinds of coordination environments. Both Zn1 and Zn2 are 4-coordinated by four O atoms from three carboxylate groups and one μ₃-OH^- to give tetrahedral configurations. While, Zn3 and Zn4 adopt 6-coordinated octahedral geometry defined by six O atoms from three carboxylate groups, two μ₃-OH^-, and one P = O moiety or DMF molecule. In 1, each Zn₄-A SBU coordinates with eight TPO^3- ligands, while each TPO^3- ligand bonds to four Zn₄-A SBUs (Figs. S1a and b in supporting informaiton). As a result, 1 shows a neutral 3D framework that can be simplified as a 4, 8-connected network with CaF₂-type topology (Fig. S1). PLATON calculation indicates that the guest-accessible volume per unit cell is 2264 ų (35.9% of the total unit cell volume).

  Figure 1

  

  Figure 1.  The synthesis method and structures (Zn₄-A, Zn₃, Zn₄-B and Co₄ SBUs; square cages and 3D frameworks with 1D double-walled rhombic channels) of 1-4. Color code: Zn, cyan; Co, blue violet; P, purple; O, red; N, blue; C, gray. The cavities in square cages are shown by yellow balls. Symmetry code: (A) -x, 1-y, -z. All hydrogen atoms and [(CH₃)₂NH₂]⁺ counterions are omitted for clarity

  DownLoad: Full-Size Img PowerPoint

  Fascinatingly, compound 1 exhibits an outstanding high thermal stability. Variable-temperature PXRD patterns indicate that the crystallinity of 1 can be kept up to at least 500 ℃ (Fig. 2a), which is comparable to the highest thermal stable MOFs such as ZIF-8, FJU-66, UiO-66, UiO(bpdc) and MIL-53·Al [27, 28]. Note that highly thermal stable MOFs are usually based on trivalent or tetravalent metal ions (e.g., UiO-66, UiO(bpdc) and MIL-53·Al). Compound 1 shows a rare bivalent-metal-based porous MOF with thermal stability over 500 ℃. The high thermal stability of 1 should be attributed to the combination of its many intriguing structural features including double-walled porous structure with strong π... π interactions between TPO^3- ligands, OH^- bridging multinuclear SBUs and their high coordination number with multitopic TPO^3- ligands, and neutral framework. Additionally, 1 also can maintain its crystallinity in a variety of common organic solvents such as MeOH, EtOH, cyclohexane, and so on (Fig. 2b). Notably, though 1 was synthesized in the presence of H₂O, it is unstable in water.

  Figure 2

  

  Figure 2.  The PXRD patterns under (a) thermal and (b) immersing in a variety of common organic solvents of 1.

  DownLoad: Full-Size Img PowerPoint

  It is interesting to find that the reaction of 1 without water produced a new crystalline porous material 2. Different from 1, compound 2 crystallizes in another orthorhombic chiral space group P222₁ and shows a 3D porous framework based on trinuclear [Zn₃(OH)]⁵⁺ (Zn₃) clusters as SBUs (Fig. 1). The triangular Zn₃ consists of one 5-coordinated (Zn1) and two 4-coordinated (Zn2 and Zn3) Zn²⁺ cations, indicating the coordination mode of Zn₃ with TPO^3- is distinct with that in 1.

  Nevertheless, each Zn₃ in 2 also coordinates with eight TPO^3- ligands and each TPO^3- ligand bonds to four Zn₃ (Fig. S2 in Supporting information), also forming a CaF₂-type 3D porous framework with similar guest-accessible volume (34.9% of the total unit cell volume) to that of 1. It is worth noting that the removal of one Zn atom from Zn₄-A to Zn₃ makes the neutral framework 1 transform into an anionic framework 2.

  Variable-temperature PXRD tests and TG analysis confirm that 2 can maintain its crystallinity over 300 ℃ (Figs. S3 and S4 in Supporting informaiton), showing 2 also has relatively high thermal stability. Nevertheless, compared with 1, 2 exhibits obviously lower thermal stability, which should be due to the removal of cationic organic guests [(CH₃)₂NH₂]⁺ over 300 ℃, and thus leading to its framework collapse. Similar to 1, 2 can also retain its structure in various common organic solvents but water.

  Further, PXRD measurements reveal that, if the sample 2 is soaked in deionized water for 3 h at room temperature, a new crystalline phase 3 can be found to coexist with 2. While, if the sample 2 is soaked in water more than 10 h, both 2 and 3 will turn into amorphous structures. Interestingly, single-crystal X-ray studies show 3 crystallizes in space group P2₁2₁2 and has almost the same 3D structure as 1 though their unit cells are obviously different. One remarkably structural difference between them lies in that the tetranuclear [Zn₄(OH)₂H₂O)₂]⁶⁺ SBU (Zn₄-B, Fig. 1) in 3 can be derived from Zn₄-A by replacing its terminal DMF molecule with one water molecule. The results indicate the Zn₃-based anionic framework 2 can realign to regenerate Zn₄-based neutral framework 3 in the presence of H₂O.

  Finally, if the starting chemical Zn(Ac)₂·2H₂O is replaced by Co(Ac)₂·4H₂O, the similar reaction to 1 can yield compound 4. Different from 1-3, solid 4 crystallizes in the monoclinic achiral space group P2₁c. Structural analysis shows the SBU in 4 is a chairtype tetranuclear [Co₄(OH)₂H₂O)₄]⁶⁺ (Co4) cluster (Fig. 1), differing from the boat-type SBUs Zn₄-A and Zn₄-B in 1 and 3. Nevertheless, it is unexpected to find that 4 has the same 3D CaF₂- type framework as 1-3, which is built by 8-connected Co₄ SBUs linked by 4-connected TPO^3- ligands (Fig. S2).

  The same framework topology and similar components of 1, 2 and 4 make us better compare and understand the influence of different structural features on their thermal stabilities. PLATON calculation indicates that 4 has the most porous framework (ca. 50.6% guest-accessible volume per unit cell) among 1-4. Due to the highest porosity, the thermal stability (ca. 250 ℃, Figs. S3 and S4) of the neutral framework 4 is much lower than that of the neutral framework 1, and even lower than that of the anionic framework 2. Additionally, from 1, 2 to 4, the TPO^3- ligand bonds to more and more metal ions from 5 (in 1), 6-7 (in 2), to 7 (in 4) (Fig. S5 in Supporting information). However, the thermal stabilities of their frameworks actually decrease gradually from 1, 2 to 4. The above results show that the porosities and charge properties of frameworks have more significant effect on the thermal stability of these materials than the bonding interactions between inorganic nodes and organic linkers. Especially, the framework porosities have a key effect on their thermal stabilities. For the frameworks with similar porosities, the neutral frameworks would have better thermal stability than the charged frameworks.

  Gas adsorption measurements were run to characterize the porosities of 1, 2, and 4. The N₂ sorption studies of all compounds at 77 K show a reversible type-Ⅰ isotherm typical of permanent microporous materials (Fig. 3a). The Langmuir/BET surface areas of 1, 2, and 4 are 829/977, 622/690, and 1525/1614 m²/g, respectively. Their corresponding median pore sizes of 6.4, 7.3, and 8.6 Å were also calculated (Fig. 3b). Further, compounds 1, 2, and 4 exhibit high CO₂ uptake capacity of 82.9/49.2, 75.07/52.08 and 104.3/ 54.7 cm³/g at 273/298 K and 1 atm (Fig. 3c), respectively. Notably, although numerous MOFs have been known, only a small part of them were reported to have a CO₂ uptake of more than 50 cm³/g at room temperature (298 K) and 1 atm.

  Figure 3

  

  Figure 3.  Gas sorptions of 1, 2 and 4. a) N₂ sorption cures at 77 K; b) pore distributions; c) CO₂-adsorption performances and d) CO₂ adsorption heats (Q_st)

  DownLoad: Full-Size Img PowerPoint

  The CO₂ adsorption heats (Q_st) of 1, 2, and 4 were calculated by the Virial method, and their values at zero coverage were respectively estimated to be 22.3, 24.2 and 17.9 kJ/mol (Fig. 3d). These Q_st values are much lower than those of many classical porous materials, such as MIL-53(Cr) (32 kJ/mol), HKUST-1 (30 kJ/mol), JUC-132 (30 kJ/mol), and JLU-Liu (30 kJ/mol) [29-32]. The high CO₂ uptakes with low Q_st values of 1, 2, and 4 are desired properties for the application in CO₂ storage and separation. Additionally, in contrast, 1, 2, and 4 have very low N₂/H₂ uptake capacities of 1.5/3.4, 2.5/0.1, and 1.1/3.6 cm³/g (Fig. S6 in Supporting information) at 273 K and 1 atm, respectively, indicating these materials have high selectivity for adsorption of CO₂ relative to both N₂ and H₂.

  In summary, a rare porous metal-phosphonocarboxylate framework 1 with high thermal stability has been prepared. Fascinatingly, by synthetic control, three new porous metalphosphonocarboxylate frameworks 2-4 derived from 1 can be obtained, all of which show relatively high thermal stability in MOFs. A unique phenomenon is that the four compounds 1-4 have different structural features but the same CaF₂-type topology, providing a nice case to study the relationship between their structures and thermal stability. The results show the design and synthesis of phosphonocarboxylate-based neutral MOFs combined with reasonable porosity could give rise to highly thermally stable MOF materials.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundations of China (Nos. 21303018, 21371033, and 21401195), the Natural Science Foundation for Young Scholars of Fujian Province (No. 2015J05041), and Projects from State Key Laboratory of Structural Chemistry of China (Nos. 20150001 and 20160020).

  Appendix A. Supplementary data

  Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.cclet.2017.09.060.

• 1. [1]

     W.G. Lu, Z.W. Wei, Z.Y. Gu, et al., Chem. Soc. Rev. 43(2014) 5561-5593. doi: 10.1039/C4CS00003J

  2. [2]

     K. Sumida, D.L. Rogow, J.A. Mason, et al., Chem. Rev. 112(2012) 724-781. doi: 10.1021/cr2003272

  3. [3]

     B. Wang, H. Yang, Y.B. Xie, et al., Chin. Chem. Lett. 27(2016) 502-506. doi: 10.1016/j.cclet.2015.12.034

  4. [4]

     Q.G. Zhai, C.Y. Mao, X. Zhao, et al., Angew. Chem. Int. Ed. 54(2015) 7886-7890. doi: 10.1002/anie.201503095

  5. [5]

     X. Zhao, X.H. Bu, T. Wu, et al., Nat. Commun. 4(2013) 2344-2352. doi: 10.1038/ncomms3344

  6. [6]

     G. Cai, H.L. Jiang, Angew. Chem. Int. Ed. 56(2017) 563-567. doi: 10.1002/anie.201610914

  7. [7]

     Y.Z. Chen, Z.U. Wang, H. Wang, et al., J. Am. Chem. Soc. 139(2017) 2035-2044. doi: 10.1021/jacs.6b12074

  8. [8]

     X.L. Luo, Y. Cao, T. Wang, et al., J. Am. Chem. Soc. 138(2016) 786-789. doi: 10.1021/jacs.5b12516

  9. [9]

     X.T. Rao, T. Song, J.K. Gao, et al., J. Am. Chem. Soc. 135(2013) 15559-15564. doi: 10.1021/ja407219k

  10. [10]

      T. Wang, Q.H. Liu, Y. Gao, et al., Chin. Chem. Lett. 27(2016) 497-501. doi: 10.1016/j.cclet.2016.01.011

  11. [11]

      G.X. Wen, Y.P. Wu, W.W. Dong, et al., Inorg. Chem. 55(2016) 10114-10117. doi: 10.1021/acs.inorgchem.6b01876

  12. [12]

      G. Mouchaham, L. Cooper, N. Guillou, et al., Angew. Chem. Int. Ed. 54(2015) 13297-13301. doi: 10.1002/anie.201507058

  13. [13]

      K.C. Wang, X.L. Lv, D.W. Feng, et al., J. Am. Chem. Soc. 138(2016) 914-919. doi: 10.1021/jacs.5b10881

  14. [14]

      S.H. Lo, C.H. Chien, Y.L. Lai, et al., J. Mater. Chem. A 1(2013) 324-329. doi: 10.1039/C2TA00030J

  15. [15]

      H. Reinsch, M.A. Veen, B. Gil, et al., Chem. Mater. 25(2013) 17-26. doi: 10.1021/cm3025445

  16. [16]

      I.A. Ibarra, S.H. Yang, X. Lin, et al., Chem. Commun. 47(2011) 8304-8306. doi: 10.1039/c1cc11168j

  17. [17]

      K.S. Park, Z. Ni, A.P. Côté, et al., Proc. Nal. Acad. Soc. 103(2006) 10186. doi: 10.1073/pnas.0602439103

  18. [18]

      D.Y. Ma, Y.W. Li, Z. Li, Chem. Commun. 47(2011) 7377-7379. doi: 10.1039/c1cc11752a

  19. [19]

      Z.J. Zhang, H.T.H. Nguyen, S.A. Miller, et al., J. Am. Chem. Soc. 138(2016) 920-925. doi: 10.1021/jacs.5b11034

  20. [20]

      A.W. Peters, Z.Y. Li, O.K. Farha, J.T. Hupp, ACS Appl. Mater. Interfaces 8(2016) 20675-20681. doi: 10.1021/acsami.6b04729

  21. [21]

      S.H. Hong, M. Oh, M. Park, et al., Chem. Commun. 36(2009) 5397-5399.

  22. [22]

      M. Kandiah, M.H. Nilsen, S. Usseglio, et al., Chem. Mater. 22(2010) 6632-6640. doi: 10.1021/cm102601v

  23. [23]

      L. Li, S. Tang, C. Wang, et al., Chem. Commun. 50(2014) 2304-2307. doi: 10.1039/c3cc48275h

  24. [24]

      I.J. Kang, N.A. Khan, E. Haque, S.H. Jhung, Chem. Eur. J. 17(2011) 6437-6442. doi: 10.1002/chem.201100316

  25. [25]

      M. Taddei, F. Costantino, R. Vivani, Eur. J. Inorg. Chem. 27(2016) 4300-4309.

  26. [26]

      G.K.H. Shimizu, R. Vaidhyanathan, J.M. Taylor, Chem. Soc. Rev. 38(2009) 1430-1449. doi: 10.1039/b802423p

  27. [27]

      K.S. Park, Z. Ni, A.P. Côté, et al., PNAS 103(2006) 10186-10191. doi: 10.1073/pnas.0602439103

  28. [28]

      Z.Y. Li, Z.J. Zhang, Y.X. Ye, et al., J. Mater. Chem. A 5(2017) 7816-7824. doi: 10.1039/C6TA11242K

  29. [29]

      A. Demessence, D.M. D'Alessandro, M.L. Foo, J.R. Long, J. Am. Chem. Soc. 131(2009) 8784-8786. doi: 10.1021/ja903411w

  30. [30]

      J.S. Choi, W.J. Son, J. Kim, W.S. Ahn, Microporous Mesoporous Mater.116(2008) 727-731. doi: 10.1016/j.micromeso.2008.04.033

  31. [31]

      S. Bourrelly, P.L. Llewellyn, C. Serre, et al., J. Am. Chem. Soc. 127(2005) 13519-13521. doi: 10.1021/ja054668v

  32. [32]

      Z. Liang, M. Marshall, A.L. Chaffee, Energy Procedia 1(2009) 1265-1271. doi: 10.1016/j.egypro.2009.01.166

• 

  Figure 1  The synthesis method and structures (Zn₄-A, Zn₃, Zn₄-B and Co₄ SBUs; square cages and 3D frameworks with 1D double-walled rhombic channels) of 1-4. Color code: Zn, cyan; Co, blue violet; P, purple; O, red; N, blue; C, gray. The cavities in square cages are shown by yellow balls. Symmetry code: (A) -x, 1-y, -z. All hydrogen atoms and [(CH₃)₂NH₂]⁺ counterions are omitted for clarity

  下载: 全尺寸图片 幻灯片
  

  Figure 2  The PXRD patterns under (a) thermal and (b) immersing in a variety of common organic solvents of 1.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Gas sorptions of 1, 2 and 4. a) N₂ sorption cures at 77 K; b) pore distributions; c) CO₂-adsorption performances and d) CO₂ adsorption heats (Q_st)

  下载: 全尺寸图片 幻灯片
•