English

• As a kind of unconventional energy with high calorific value, coal-bed methane (CBM) is a kind of natural gas mainly composed of methane and nitrogen [1,2]. For the utilization of coal-bed methane, CBM with the high methane concentration can be directly extracted and utilized [3]. On the contrast, a large number of CBM resources with low methane concentration (< 30% CH₄) are often directly vented to the atmosphere, which not only causes resource waste but also aggravates the global greenhouse effect [4–7]. Therefore, the separation of CH₄ from N₂ for low-grade CBM purification is considered to be economical and environment-friendly.

  Owing to their similar physical properties, the separation of CH₄/N₂ is widely regarded as one of the most challenging technical issues [8–11]. Currently, traditional industrial technology to separate CH₄ over N₂ from CBM is cryogenic distillation, which is highly energy-consuming and uneconomical [12,13]. On the contrary, adsorbent-based gas separations have exhibited great prospects due to their low energy consumption and simple operation [14]. To date, several types of porous materials have been investigated and used for separating CBM, including activated carbons [15–18], zeolites [19] and MOFs [20–24]. In this regard, activated carbons and zeolites often shows some drawbacks in practical applications including lack of functional adsorption sites, poor separation selectivity, and low uptake capacity [25]. Although MOFs display excellent methane adsorption capacity and CH₄/N₂ separation selectivity because of the existence of unsaturated metal centers and suitable pore size, they often suffer from the poor chemical and moisture stabilities [26–28]. Therefore, it is of great significance to develop highly efficient and stable adsorbents for selective separation of CH₄/N₂.

  Covalent organic frameworks (COFs), constructed by covalent bonds, are a type of promising porous materials due to their large specific surface area [29], low mass density [30], excellent chemical/thermal stability [31,32] and structural modularity [33,34]. Therefore, COFs is a kind of suitable candidate material for the separation of CH₄/N₂. Unfortunately, current COFs were rarely reported [35] for methane purification which may be attributed to the lack of efficient constructional strategies. The existing strategy of constructing COF adsorbent for CH₄/N₂ separation is adjusting the pore size of COFs. Therefore, it is necessary to develop new strategy for constructing COF based adsorbents for CH₄/N₂ separation. Based on the fact that CH₄ (25.93 × 10⁻²⁵ cm³) has a greater polarizability than N₂ (17.40 × 10⁻²⁵ cm³), we reason that enhancing the polarity of the host framework to form multiple interactions between CH₄ molecules and the skeleton would benefit the CH₄/N₂ separation. Therefore, we proposed a dual polarization strategy by introducing both triazine and polyfluoride sites onto framework to construct polar pores, which would thus enhance the interaction between CH₄ molecules and the host framework and thereby achieving the high-efficient separation of CH₄ from N₂.

  To verify our strategy, four COFs including CTF-1, CTF-2, and their perfluorinated counterparts (F-CTF-1 and F-CTF-2) were chosen as the platform adsorbents for CBM purification (Scheme 1). Both F-CTF-1 and F-CTF-2 with polar pores exhibited higher CH₄ uptake and CH₄/N₂ selectivity than that of CTF-1 and CTF-2. Especially, F-CTF-2 exhibited high CH₄ uptake capacity (21.7 cm³/g) and superior CH₄/N₂ selectivity (5.28), which is 1.76 and 1.42 times higher than that of CTF-2. Simulated calculations revealed that the higher CH₄ uptake and CH₄/N₂ selectivity of F-CTF-2 were attributed to the existence of dual polarization sites on COF pores. These results thus highlight the advantage of the dual polarization strategy to construct CH₄ preferred adsorbent for CH₄/N₂ separation. Additionally, F-CTF-2 exhibits excellent chemical and thermal stabilities, making it among the promising adsorbents for CBM purification.

  Scheme 1

  

  Scheme 1.  Illustration of the dual polarization strategy.

  DownLoad: Full-Size Img PowerPoint

  All of the COF materials were prepared at 500 ℃ under typical ionothermal conditions using ZnCl₂ as catalyst with a ZnCl₂/monomer molar ratio of 10 [36]. The obtained products were abbreviated as CTF-1, CTF-2, F-CTF-1, F-CTF-2, respectively. The successful synthesis of CTFs and F-CTFs was confirmed by Fourier transform-infrared (FT-IR), X-ray photoelectron spectroscopy (XPS), solid-state ¹³C NMR measurements. As shown in the FT-IR spectra (Fig. 1a and Fig. S1 in Supporting information), the characteristic vibrational band for the terminal cyano group at ~2248 cm⁻¹ disappeared after polymerization, representing the complete polymerization has been achieved. Meanwhile, the new characteristic vibrations at ~1569 cm⁻¹ (–C=N– stretching vibration) and 1373 cm⁻¹ (–C–N= stretching vibration) indicated the successful formation of the triazine rings [36,37]. Furthermore, X-ray photoelectron spectroscopy (XPS) was used to verify the structural features of the CTFs and F-CTFs. As displayed in Figs. S2a and b (Supporting information), the C 1s XPS spectrum of each of the samples can be resolved into two peaks. The strongest peak (284.9 eV) was assigned to the sp² hybrid C species in the benzene rings, while the peak with higher binding energy (286.2 eV) was assigned to the sp² hybrid C species in the triazine rings [38]. In the N 1s XPS spectra of CTFs and F-CTFs, the peaks at ~398.0 eV also revealed the formation of the triazine rings (Fig. 1b and Fig. S2c in Supporting information) [39]. Solid-state ¹³C NMR spectra of F-CTFs showed the characteristic peak at ~161 ppm corresponding to the sp²-hybridized carbon atoms in the triazine ring [40], which further revealed the formation of CTFs (Fig. S3 in Supporting information). Additionally, the morphology of F-CTFs was analyzed by the SEM and TEM. The layered structures and porosity were observed. The energy dispersive spectroscopy (EDS) elemental mappings (Fig. S4 in Supporting information) indicated a homogeneous distribution of C, N and F in F-CTFs. Finally, the powder X-ray diffraction (XRD) patterns of CTFs and F-CTFs showed an amorphous character (Fig. S5 in Supporting information).

  Figure 1

  

  Figure 1.  (a) FT-IR, (b) N 1s XPS spectra of CTF-2 and F-CTF-2.

  DownLoad: Full-Size Img PowerPoint

  The porosity property of the model COF materials was examined by N₂ adsorption isotherms at 77 K (Fig. 2a). All of the samples displayed rapid N₂ uptake at low relative pressures (P/P₀ < 0.04), indicating the existence of microporous structures. Then, the hysteresis loops appeared around the relative pressure of P/P₀ = 0.4, showing the presence of mesopores. The formation of mesoporous structure may be caused by the defects in CTFs [41]. The Brunauer-Emmett-Teller (BET) specific surface areas of CTF-1 and CTF-2 were calculated to be 1239 and 1599 m²/g. And that were 1392 and 1812 m²/g for F-CTF-1 and F-CTF-2, respectively. In addition, the pore size distributions were calculated by the density functional theory model (Fig. S6 in Supporting information). Compared with CTF-1 and CTF-2, F-CTF-1 and F-CTF-2 possessed more ultramicropore structures at 5.9 Å and 5.0 Å, which favors small-size gas molecule separation.

  Figure 2

  

  Figure 2.  (a) Nitrogen sorption isotherms at 77 K. (b) Single-component adsorption isotherms of CH₄ and N₂ for CTF-2 and F-CTF-2 at 298 K. (c) Q_st curves for F-CTF-1 and F-CTF-2. (d) IAST selectivity of CH₄/N₂ (1:1, v/v and 1:9, v/v) for CTF-2 and F-CTF-2 at 298 K.

  DownLoad: Full-Size Img PowerPoint

  In order to verify the effectiveness of the dual polarization strategy, the single-component adsorption of CH₄ and N₂ was separately measured at 273 and 298 K (Fig. 2b, Figs. S7a and b in Supporting information). The adsorption capacities of F-CTF-1 and F-CTF-2 for CH₄ at 273 K and 1 bar were 25.2 cm³/g and 33.7 cm³/g, respectively. While the CH₄ uptake for CTF-1 and CTF-2 were a little lower with the capacity of 23.3 cm³/g and 20.2 cm³/g, respectively, under the identical conditions. Consistently, the CH₄ uptake of F-CTF-1 and F-CTF-2 at 298 K and 1 bar were also higher than their counterparts of CTF-1 (16.3 vs. 14.9 cm³/g) and CTF-2 (21.7 vs. 12.3 cm³/g), respectively. Apparently, all F-CTFs absorbed more CH₄ than their non-fluorizated counterparts in the full pressure region at 298 K or 273 K. Then, the virial equation was used to calculate isosteric enthalpy of adsorption (Q_st) for all samples based on the isotherms collected at 273 K and 298 K (Fig. 2c and Fig. S7c in Supporting information). The fitted parameters were provided in Tables S1–S4 (Supporting information). The initial Q_st values of CH₄ for F-CTF-1 and F-CTF-2 were 21.1 and 19.4 kJ/mol, respectively, higher than those for CTF-1 and CTF-2 (17.6 and 16.7 kJ/mol, respectively). As descripted, both the CH₄ uptake and Q_st follow the trend: CTFs < F-CTFs. This phenomenon is due to the introduction of polar sites on F-CTFs pores, which induce a stronger guest-framework interaction. To explore the separation potential of adsorbents, the CH₄/N₂ separation abilities of these COFs were evaluated by the ideal adsorbed solution theory (IAST) calculations (Tables S5 and S6 in Supporting information). Consistent with the trend of CH₄ uptake for CTFs and F-CTFs, the adsorption selectivities of the CH₄/N₂ (1:1, v/v) mixture at 298 K and 1 bar for F-CTF-1 and F-CTF-2 were 5.20 and 5.28, respectively (Fig. 2d and Fig. S7d in Supporting information), higher than that of CTF-1 (4.59) and CTF-2 (3.73). Notably, the selectivity of F-CTFs are higher than that of many benchmark adsorbents, such as 5A zeolite (1.7) [42], BPL (2.6) [43] and MIL-100(Cr) (3.0) [44], indicating the successful practice of dual polarization strategy.

  Considering the practical application, the stability of adsorbent is another key criterion to evaluate the industrial application potential for the purification of CBM. As an optimized candidate, the chemical/thermal stabilities for F-CTF-2 were investigated. F-CTF-2 can remain structurally stable in harsh conditions as verified by the consistent FT-IR results (Fig. S8a in Supporting information) and almost the same CH₄ uptake capacity after being treated in 6 mol/L HCl and 6 mol/L NaOH solution for 72 h, respectively (Fig. S8b in Supporting information). TGA test showed that F-CTF-2 possessed a high thermal stability up to 450 ℃ (Fig. S9 in Supporting information). Moreover, F-CTF-2 exhibited superior recyclability with no visible decay after 10 cycles (Fig. S10 in Supporting information). It is worth noting that the comprehensive applicability of F-CTF-2 has exceeded that of most other benchmarks, such as zeolites [19], MOFs [1], Schiff-based COFs [45], which often suffer from the potential decomposition of porous structures under such harsh conditions (Fig. 3) [46,47]. Therefore, these results thus rank F-CTF-2 among the best porous adsorbents for CBM purification.

  Figure 3

  

  Figure 3.  Comparison of separation performance on F-CTF-2 and other benchmarks (*chemical stability was not studied).

  DownLoad: Full-Size Img PowerPoint

  The adsorption behavior of CH₄ and N₂ was computed using grand canonical Monte Carlo simulations to reveal the adsorption mechanism. Considering its superior CH₄ adsorption capacity and CH₄/N₂ selectivity, F-CTF-2 was chosen as the simulation model. As illustrated in Fig. 4a, CH₄ interacted with the fluorine binding sites of F-CTF-2 via multiple C–H···F hydrogen bond interactions, in which the H···F distance ranged from 2.98 Å to 4.08 Å. Meanwhile, CH₄ interacted with the triazine ring sites via multiple C–H···N hydrogen bond interactions, and the H···N distance differed from 3.83 Å to 4.41 Å. In comparison, there only existed weak Van der Waals interactions between N₂ and the F-CTF-2 host frameworks with the N···F distance from 4.14 Å to 4.58 Å (Fig. 4b). This is consistent with the calculated binding energy of 15.30 kJ/mol between F-CTF-2 and CH₄, and a low binding energy of 5.00 kJ/mol between F-CTF-2 and N₂. The stronger interactions between the dual polarization sites in skeleton and methane molecules could thus account for high methane uptake and selectivity for F-CTF-2.

  Figure 4

  

  Figure 4.  Schematic adsorption sites for (a) CH₄ and (b) N₂ obtained from theoretical calculations. Color code: C, dark gray; H, white; N, blue; F, green.

  DownLoad: Full-Size Img PowerPoint

  To further confirm the practical separation performance on the CBM with a low CH₄ concentration, breakthrough experiment of F-CTF-2 for CH₄/N₂ (1:9, v/v) was purged into a packed column with a total inlet flow rate of 10 mL/min at 298 K (Fig. 5). During the experiment, N₂ firstly penetrated the column before CH₄ because of weaker interactions with the adsorbents. Subsequently, CH₄ broke through the column after 20 min, which corresponds to a time window of 20 min for high-purity CH₄ collection. This separation experiment demonstrated that F-CTF-2 can efficiently purify low concentration CBM, indicating the practicability of the dual polarization strategy.

  Figure 5

  

  Figure 5.  Experimental breakthrough curve for F-CTF-2 at 298 K and 1 bar.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, we developed a dual polarization strategy to construct robust adsorbents for coal-bed methane purification. By introducing triazine and polyfluoride sites onto COF-based materials, the CH₄ uptake and CH₄/N₂ selectivity were significantly enhanced. This was attributed to the enhanced affinity between CH₄ molecules and the skeleton. Theoretical calculation revealed that the surface of F-CTF-2 could have strong interactions with CH₄ via multiple C-H···F and C-H···N hydrogen bonds, which then caused good CH₄ adsorption capacity. Thereby our work not only proposed a new approach to prepare high uptake CH₄ adsorbents but also provided insight of the development of COFs as a new type of adsorbents for low concentration coal-bed methane purification.

  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 National Key R & D Program of China (No. 2022YFA1503300), National Natural Science Foundation of China (Nos. 21978138, 22035003), the Fundamental Research Funds for the Central Universities (Nankai University), and the Haihe Laboratory of Sustainable Chemical Transformations (No. YYJC202101).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.109180.

  
  
• 1. [1]

     S.M. Wang, M. Shivanna, Q.Y. Yang, Angew. Chem. Int. Ed. 61 (2022) e202201017. doi: 10.1002/anie.202201017

  2. [2]

     T.Y. Li, Y.S. Wang, M.Y. Zhang, et al. Chem. Eng. J. 462 (2023) 142118. doi: 10.1016/j.cej.2023.142118

  3. [3]

     Y. Liu, Q. Xu, L. Chen, et al. Nano Res. 15 (2022) 7695–7702. doi: 10.1007/s12274-022-4352-0

  4. [4]

     C. Zhang, Y. Chen, H. Wu, et al. Chem. Eng. J. 435 (2022) 133876. doi: 10.1016/j.cej.2021.133876

  5. [5]

     Q. Shi, J. Wang, H. Shang, et al. Sep. Purif. Technol. 230 (2020) 115850. doi: 10.1016/j.seppur.2019.115850

  6. [6]

     M. Chang, F. Wang, Y. Wei, et al. AIChE J. 68 (2022) e17794. doi: 10.1002/aic.17794

  7. [7]

     J. Liu, X. Tang, X. Liang, et al. AIChE J. 68 (2022) e17589. doi: 10.1002/aic.17589

  8. [8]

     F. Zhang, H. Shang, B. Zhai, et al. AIChE J. 69 (2023) e18079. doi: 10.1002/aic.18079

  9. [9]

     M. Chang, T. Yan, Y. Wei, et al. ACS Appl. Mater. Interfaces 14 (2022) 25374–25384. doi: 10.1021/acsami.2c03619

  10. [10]

      X.W. Liu, Y.M. Gu, T.J. Sun, et al. Ind. Eng. Chem. Res. 58 (2019) 20392–20400. doi: 10.1021/acs.iecr.9b03566

  11. [11]

      M. Chang, J. Ren, Y. Wei, et al. Chem. Mater. 35 (2023) 4286–4296. doi: 10.1021/acs.chemmater.3c00350

  12. [12]

      D. Lv, Y. Wu, J. Chen, et al. AIChE J. 66 (2020) e16287. doi: 10.1002/aic.16287

  13. [13]

      V.I. Águeda Maté, J.A. Delgado Dobladez, S. Álvarez-Torrellas, et al. Chem. Eng. J. 361 (2019) 1007–1018. doi: 10.1016/j.cej.2018.12.154

  14. [14]

      M. Chang, Y. Zhao, D. Liu, et al. Sustain. Energy Fuels 4 (2020) 138–142. doi: 10.1039/c9se00838a

  15. [15]

      S. Xu, W.C. Li, C.T. Wang, et al. Angew. Chem. Int. Ed. 60 (2021) 6339–6343. doi: 10.1002/anie.202014231

  16. [16]

      P. Zhang, J. Wang, W. Fan, et al. Chem. Eng. J. 375 (2019) 121931. doi: 10.1016/j.cej.2019.121931

  17. [17]

      Z. Dong, B. Li, H. Shang, et al. AIChE J. 67 (2021) e17281. doi: 10.1002/aic.17281

  18. [18]

      S.M. Wang, P.C. Wu, J.W. Fu, Q.Y. Yang, Sep. Purif. Technol. 274 (2021) 119121. doi: 10.1016/j.seppur.2021.119121

  19. [19]

      X. Tang, Y. Wang, M. Wei, et al. Sep. Purif. Technol. 318 (2023) 124003. doi: 10.1016/j.seppur.2023.124003

  20. [20]

      Z. Niu, X. Cui, T. Pham, et al. Angew. Chem. Int. Ed. 58 (2019) 10138–10141. doi: 10.1002/anie.201904507

  21. [21]

      S. Qadir, D. Li, Y. Gu, et al. Chem. Eng. J. 408 (2021) 127238. doi: 10.1016/j.cej.2020.127238

  22. [22]

      M. Chang, J. Ren, Q. Yang, D. Liu, Chem. Eng. J. 408 (2021) 127294. doi: 10.1016/j.cej.2020.127294

  23. [23]

      Y. Chen, Y. Wang, Y. Wang, et al. AIChE J. 68 (2022) e17819. doi: 10.1002/aic.17819

  24. [24]

      D. Lv, R. Shi, Y. Chen, et al. Ind. Eng. Chem. Res. 57 (2018) 12215–12224. doi: 10.1021/acs.iecr.8b02596

  25. [25]

      Y. Chen, H. Wu, Y. Yuan, et al. Chem. Eng. J. 385 (2020) 123836. doi: 10.1016/j.cej.2019.123836

  26. [26]

      T. Li, X. Jia, H. Chen, et al. ACS Appl. Mater. Interfaces 14 (2022) 15830–15839. doi: 10.1021/acsami.2c01156

  27. [27]

      X. Zhao, P. Pachfule, A. Thomas, Chem. Soc. Rev. 50 (2021) 6871–6913. doi: 10.1039/d0cs01569e

  28. [28]

      J.N. Chang, Q. Li, Y. Yan, et al. Angew. Chem. Int. Ed. 61 (2022) e202209289. doi: 10.1002/anie.202209289

  29. [29]

      X.H. Han, J.Q. Chu, W.Z. Wang, et al. Chin. Chem. Lett. 33 (2022) 2464–2468. doi: 10.1016/j.cclet.2021.11.066

  30. [30]

      X. Feng, X. Ding, D. Jiang, Chem. Soc. Rev. 41 (2012) 6010–6022. doi: 10.1039/c2cs35157a

  31. [31]

      J. Li, J. Wang, F. Shui, et al. Chin. Chem. Lett. 34 (2023) 107917. doi: 10.1016/j.cclet.2022.107917

  32. [32]

      J.N. Chang, Q. Li, J.W. Shi, et al. Angew. Chem. Int. Ed. 62 (2023) e202218868. doi: 10.1002/anie.202218868

  33. [33]

      T.Y. Yu, Q. Niu, Y. Chen, et al. J. Am. Chem. Soc. 145 (2023) 8860–8870. doi: 10.1021/jacs.2c12313

  34. [34]

      M.X. Wu, Y.W. Yang, Chin. Chem. Lett. 28 (2017) 1135–1143. doi: 10.1016/j.cclet.2017.03.026

  35. [35]

      K.X. Yao, Y. Chen, Y. Lu, Y. Zhao, Y. Ding, Carbon 122 (2017) 258–265. doi: 10.1016/j.carbon.2017.06.073

  36. [36]

      P. Kuhn, M. Antonietti, A. Thomas, Angew. Chem. Int. Ed. 47 (2008) 3450–3453. doi: 10.1002/anie.200705710

  37. [37]

      X. Suo, F. Zhang, Z. Yang, et al. Angew. Chem. Int. Ed. 60 (2021) 25688–25694. doi: 10.1002/anie.202109342

  38. [38]

      G. Wang, K. Leus, H.S. Jena, et al. J. Mater. Chem. A 6 (2018) 6370–6375. doi: 10.1039/c7ta08913a

  39. [39]

      Z.A. Lan, M. Wu, Z. Fang, et al. Angew. Chem. Int. Ed. 61 (2022) e202201482. doi: 10.1002/anie.202201482

  40. [40]

      X. Zhu, C. Tian, S.M. Mahurin, et al. J. Am. Chem. Soc. 134 (2012) 10478–10484. doi: 10.1021/ja304879c

  41. [41]

      M. Liu, Q. Huang, S. Wang, Angew. Chem. Int. Ed. 57 (2018) 11968–11972. doi: 10.1002/anie.201806664

  42. [42]

      J.A.C. Silva, A. Ferreira, P.A.P. Mendes, et al. Ind. Eng. Chem. Res. 54 (2015) 6390–6399. doi: 10.1021/acs.iecr.5b01608

  43. [43]

      S. Sircar, J. Chem. Soc. Faraday Trans. I. 80 (1984) 1101–1111. doi: 10.1039/f19848001101

  44. [44]

      L. Li, J. Yang, J. Li, Y. Chen, J. Li, Microporous Mesoporous Mater. 198 (2014) 236–246. doi: 10.1016/j.micromeso.2014.07.041

  45. [45]

      L. Liao, X. Guan, H. Zheng, et al. Chem. Sci. 13 (2022) 9305–9309. doi: 10.1039/d2sc02365b

  46. [46]

      A. Dong, D. Chen, Q. Li, et al. Small 19 (2023) 2201550. doi: 10.1002/smll.202201550

  47. [47]

      X. Guan, H. Li, Y. Ma, et al. Nat. Chem. 11 (2019) 587–594. doi: 10.1038/s41557-019-0238-5

• 

  Scheme 1  Illustration of the dual polarization strategy.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  (a) FT-IR, (b) N 1s XPS spectra of CTF-2 and F-CTF-2.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Nitrogen sorption isotherms at 77 K. (b) Single-component adsorption isotherms of CH₄ and N₂ for CTF-2 and F-CTF-2 at 298 K. (c) Q_st curves for F-CTF-1 and F-CTF-2. (d) IAST selectivity of CH₄/N₂ (1:1, v/v and 1:9, v/v) for CTF-2 and F-CTF-2 at 298 K.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Comparison of separation performance on F-CTF-2 and other benchmarks (*chemical stability was not studied).

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Schematic adsorption sites for (a) CH₄ and (b) N₂ obtained from theoretical calculations. Color code: C, dark gray; H, white; N, blue; F, green.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Experimental breakthrough curve for F-CTF-2 at 298 K and 1 bar.

  下载: 全尺寸图片 幻灯片
•