English

• The development of high-performance stationary phases plays a crucial role in chromatographic separation [1-4]. Both the mass transfer of analytes and the thermodynamic interactions between stationary phases and analytes should be taken into consideration when constructing high-performance stationary phases for gas chromatography (GC) [5-8]. The fast mass transfer usually results in low diffusion resistance, achieving high column efficiency [9,10]. Meanwhile, the prerequisite for the separation is that the stationary phases can provide sufficiently distinguishable thermodynamic interactions with different analytes [11-13]. It has been reported that the similar thermodynamic adsorption behavior to compounds results in close retention times, which leads to the low-resolution performance of chromatographic columns [14]. Besides, changing the thermodynamic interactions between stationary phases with analytes can adjust the elution order of specific components [15,16]. Therefore, it is critical to modulate the thermodynamic interactions of stationary phases to different substances for chromatographic separation.

  Traditional stationary phases offer simplex thermodynamic interactions for separating analytes, such as van der Waals forces or polarity interaction. However, this simplicity could pose challenges in differentiating isomers with similar physicochemical properties [17]. Metal-organic frameworks (MOFs) possess the advantages of high thermal stability, designable structure, tunable pore size, as well as diverse functionalities, showing great potential as stationary phases for chromatographic separations [18-20,17,21]. On the one hand, the suitable pore size of MOFs is conducive to the mass transfer of analytes [22,23]. On the other hand, MOFs can be functionalized with various functional groups on metal nodes or organic ligands, which can introduce multiple interaction sites for the separation of isomers [24,25]. Previous studies have reported that adjusting the interaction sites of MOFs has an obvious influence on the separation performance of chromatographic columns. For example, by changing the metal sites of MIL-100 from Cr clusters to Fe clusters, the separation performance of the MOF stationary phase for the separating of alkanes was enhanced [26]. The reason was that MIL-100(Cr) provided excessive interactions with alkanes, resulting in tailing chromatographic peaks, while MIL-100(Fe) weakened the interactions between MOF and CH groups of alkanes. In contrast, the modification of pyridine on the metal sites of MIL-101(Cr) resulted in decreased separation capability in separating xylene isomers [27]. This modification weakened the differentiation of polarity interactions with xylenes, leading to low separation ability. Thus, the efficient development of high-performance MOF-based stationary phases needs the reasonable modulation of interaction sites without bringing in too much diffusion resistance.

  Here, two types of MOFs, MIL-125 and MIL-143 were selected as models to investigate the effect of thermodynamic interactions between MOFs and analytes on GC separation performance (Scheme 1). MIL-125 was functionalized with the amino group by replacing 1,4-dicarboxybenzene (H₂BDC) with 2-aminoterephthalic acid (H₂BDC-NH₂) in the synthesis process, obtaining MIL-125-NH₂ [28,29]. While MIL-143-BTB (BTB = 1,3,5-tri(4-carboxyphenyl)benzene) was modified by using ligands with triazine core rather than benzene, resulting in MIL-143-TATB (TATB = 2,4,6-tris(4-carboxyphenyl)-1,3,5-triazine) [30,31]. The four MOFs were coated into the inner wall of capillary columns for gas chromatographic separation of various analytes. MIL-125-NH₂ and MIL-143-TATB coated columns exhibited higher separation ability than pristine MIL-125 and MIL-143-BTB, respectively. Besides, the MIL-125-NH₂ coated column presented superior resolutions of p-xylene/m-xylene and p-xylene/o-xylene than other reported stationary phases. The mass transfer resistance evaluated by GC revealed that the kinetic factor was not the dominant element for separation improvement. The thermodynamic experiments indicated that compared with MIL-125, MIL-125-NH₂ had stronger and differential interactions with the styrene/ethylbenzene, which led to stronger retention behavior and higher selectivity. Moreover, these enhanced thermodynamic interactions also caused reversed elution sequences of benzene/cyclohexane and methylbenzene/methyl-cyclohexane on MIL-125 and MIL-125-NH₂ coated columns. The same increase of thermodynamic interactions also existed in MIL-143-TATB. The MIL-143-TATB coated column provided stronger interactions with xylene isomers than the MIL-143-BTB coated column. We concluded that both the amino group in MIL-125-NH₂ and the triazine core in MIL-143-TATB caused additional interactions with analytes, provided distinctly different thermodynamic interactions with different analytes, and then improved separation performance. This discovery provides a guide to enhancing the separation selectivity of MOF-based stationary phases.

  Scheme 1

  

  Scheme 1.  Schematic diagram of improving separation ability of MOF-stationary phases by increasing the thermodynamic differentiation.

  DownLoad: Full-Size Img PowerPoint

  All four MOFs were synthesized according to previous reports. MIL-125 was built up from Ti₈O₈(OH)₄ cluster and H₂BDC organic ligand, with the octahedral cage of 12.0 Å and the tetrahedral cage of 6.0 Å (Fig. 1). The powder X-ray diffraction (PXRD) patterns of MIL-125 and MIL-125-NH₂ matched well with the simulated one, which confirmed the successful synthesis (Fig. 1b). Scanning electron microscopy (SEM) images displayed that MIL-125 exhibited the morphology of circular plates (Fig. S1 in Supporting information). By modulating the proportion of solvents, the morphology of MIL-125-NH₂ was changed from circular plates to tetragonal plates, and octahedron crystals (Fig. S1). MIL-143 was composed of two sets of tetrahedrons [30], where Fe clusters located at the vertexes and four organic ligands coordinated at the faces or edges (Fig. 2). The PXRD patterns of MIL-143-BTB and MIL-143-TATB also proved their successful synthesis (Fig. 2b). The SEM images of MIL-143-BTB and MIL-143-TATB showed irregular polyhedral morphology (Fig. S1).

  Figure 1

  

  Figure 1.  (a) Representation of octahedral cage (12.0 Å) and tetrahedral cage (6.0 Å) in MIL-125. (b) PXRD patterns of synthesized MIL-125, MIL-125-NH₂, and simulated MIL-125. (c) FTIR spectra of MIL-125 and MIL-125-NH₂. (d) The N₂ adsorption-desorption isotherms measured at 77 K and (e) pore size distribution of MIL-125 and MIL-125-NH₂.

  DownLoad: Full-Size Img PowerPoint

  Figure 2

  

  Figure 2.  (a) Representation of large tetrahedron (11.0 Å) and small tetrahedron (9.0 Å) in MIL-143-BTB. (b) PXRD patterns of synthesized MIL-143-BTB, MIL-143-TATB, and simulated MIL-143-BTB. The patterns were collected at BL17B1 beamline, Shanghai Synchrotron Radiation Facility (SSRF) (λ = 1.24 Å). (c) TGA plots of MIL-143-BTB and MIL-143-TATB. Experiment condition: temperature ramp from room temperature to 800 ℃ at 10 ℃/min under the oxygen atmosphere.

  DownLoad: Full-Size Img PowerPoint

  The Fourier transform infrared (FTIR) spectra of MIL-125 and MIL-125-NH₂ exhibited typical vibrational bands in the 1400–1600 cm⁻¹ region of carboxylate groups in the Ti-based MOFs (Fig. 1c). The apparent bands at 1254 cm⁻¹ and 1338 cm⁻¹ in MIL-125-NH₂ represented the C-N stretching vibrations of aromatic amines, and the new band at 1624 cm⁻¹ corresponded to the N-H bending vibration [32,33]. Water contact angle experiments were performed to examine the hydrophilic/hydrophobic properties of synthesized MIL-125 and MIL-125-NH₂. The water contact angle of MIL-125-NH₂ (40.1^o) was smaller than that of MIL-125 (57.7^o), indicating the more hydrophilic character of MIL-125-NH₂ (Fig. S2 in Supporting information).

  To characterize the porosities of MOFs, the N₂ adsorption-desorption isotherms were recorded at 77 K. As shown in Fig. 1d and Fig. S3 (Supporting information), all the isotherms displayed the type-Ⅰ behavior, implying the existence of micropores. The Brunauer-Emmett-Teller (BET) surface areas of MIL-125 and MIL-125-NH₂ were 1142 and 957 m²/g, respectively (Table S1 in Supporting information). Besides, the main pore size calculated by the H-K (Saito-Foley) method of MIL-125 and MIL-125-NH₂ was 10.1 Å and 9.6 Å, respectively (Fig. 1e). The slightly decreased pore size of MIL-125-NH₂ was attributed to the amino group in the organic ligands. The main pore sizes of both MIL-143-BTB and MIL-143-TATB were mainly concentrated around 7.0 and 11.0 Å calculated by the DFT method (Fig. S3b in Supporting information).

  To investigate the thermal stability of the four MOFs, thermogravimetric analysis (TGA) under the oxygen atmosphere was performed. As shown in Fig. S4 (Supporting information), MIL-125 and MIL-125-NH₂ were stable up to 300 ℃. The first weight loss of around 90 ℃ was attributed to the removal of water and MeOH, and the weight loss ratios were around 16% and 18% for MIL-125 and MIL-125-NH₂, respectively. Besides, MIL-125-NH₂ possessed a two-step decomposition process due to the more thermolabile BDC-NH₂ ligands [34-36]. The weight loss during 440–550 ℃ was attributed to the continued fragmentation of organic ligands [34]. From room temperature to 300 ℃, the weight loss of MIL-143-BTB and MIL-143-TATB were approximately 20% and 12%, respectively (Fig. 2c). The collapse of MIL-143-BTB and MIL-143-TATB occurred above 320 ℃, indicating good thermal stability. The triazine core of the organic linker had little effect on the thermal stability of the MIL-143 network.

  After the full characterization, all four MOFs were dynamically coated into the inner wall of capillary columns. After coating, the four columns were aged at 250 ℃ for 240 min using a temperature program (see Supporting information for details). The successful coating of MOFs onto the capillary columns was proved by SEM images with the cross-section view of these columns at different segments (Figs. S5–S9 in Supporting information). It could be seen that the coating and aging process did not change the morphologies of these MOFs. MIL-125 and MIL-125-NH₂ were also coated on the quartz substrates under the same conditions of column coating and heated under the temperature program as column aging. The PXRD patterns of these treated MOFs still maintained good crystallinity (Figs. S10 and S11 in Supporting information), further indicating the coating and aging process did not affect the crystallinity of MOFs.

  Before evaluating the separation performance of four MOFs with different structures, the MIL-125-NH₂ with different morphologies (circular plates, tetragonal plates, and octahedron crystals) were coated on the columns, respectively, to illustrate that the morphology of MOFs with isotropic three-dimensional topology did not affect the separation performance (Fig. S12 in Supporting information). Thus, the subsequent discussion focused on the influence of MOF structure on separation. Then, various groups of analytes were utilized to evaluate the separation ability of four MOFs with different structures (Fig. 3 and Figs. S13–S16 in Supporting information). Separation conditions for these analytes were given in Table 1 and Tables S2 and S3 (Supporting information). On the whole, the two nitrogenous MOFs, MIL-125-NH₂ and MIL-143-TATB manifested better separation performance than corresponding MIL-125 and MIL-143-BTB. As shown in Figs. 3a and b, compared with MIL-125, the MIL-125-NH₂ coated columns exhibited higher separation ability for xylene isomers and styrene/ethylbenzene. For example, the separation resolution (R_s) of styrene/ethylbenzene was 4.58 on the MIL-125-NH₂ coated column, larger than that on the MIL-125 coated column (R_s = 1.09, Table S4 in Supporting information). Besides, the R of p-xylene/m-xylene (pX/mX) and p-xylene/o-xylene (pX/oX) on the MIL-125-NH₂ coated columns were remarkably higher than other reported stationary phases and the commercial columns (Fig. 3i) [27,37,9,7]. Furthermore, MIL-125 and MIL-125-NH₂ coated columns presented reversed selectivity of benzene/cyclohexane and methylbenzene/methyl-cyclohexane (Fig. 3c and Figs. S13 and S14). Similarly, the MIL-143-TATB coated column presented great separation performance toward substituted benzene isomers, alkane isomers, and the mixture of alkene/alkane/cycloalkane (Fig. S16). For example, the R_s for 1-octene/cyclooctane was as high as 35.92 (Table S5 in Supporting information). However, the MIL-143-BTB coated column had virtually no separation ability with obvious peak tailing phenomena (Fig. 3 and Fig. S15). The enhanced performance of nitrogenous MOFs possibly resulted from the additional interaction between analytes and N atoms in nitrogenous MOFs.

  Figure 3

  

  Figure 3.  Gas chromatogram on MIL-125 and MIL-125-NH₂ coated capillary columns at the N₂ flow of 0.5 mL/min for the separation of (a) xylene isomers, (b) styrene/ethylbenzene, and (c) benzene/cyclohexane. Gas chromatogram on MIL-143-BTB and MIL-143-TATB coated capillary columns at the N₂ flow of 0.5 mL/min for the separation of (d) xylene isomers, (e) C₆H₁₄ isomers, and (f) the mixture of 1-hexene, n-hexane, and cyclohexane. The numbers on the left represented the starting time of the elution of the first chromatographic peaks. Comparison of separation resolution on (g) MIL-125 and MIL-125-NH₂ coated columns and (h) MIL-143-BTB and MIL-143-TATB coated columns. (i) Comparison of the separation resolution of pX/oX and pX/mX on different stationary phases.

  DownLoad: Full-Size Img PowerPoint

  Table 1

  

  Table 1.  Separation conditions of different isomers on MIL-125, MIL-125-NH₂, and MIL-143-TATB coated columns.

  DownLoad: CSV

  Analyte	MIL-125 column				MIL-125-NH2 column				MIL-143-TATB column
  	Split ratio	Ta	tb		Split ratio	T	t		Split ratio	T	t
  Xylene	20:1	150-250@20	4.36		20:1	150-250@20	5.23		10:1	120-250@20	6.34
  Chlorotoluene	20:1	180-250@20	3.39		20:1	200-250@20	4.05		10:1	160-250@20	4.83
  C6H14	20:1	140-250@20	2.59		10:1	140-250@20	2.57		20:1	120-250@20	3.80
  C8H18	5:1	180-250@20	2.97		30:1	180-250@20	3.32		20:1	160-250@20	3.70
  a T (℃) was the temperature program of isomers in MOF-based columns, i.e., “140-250@20” implied that the temperature is maintained at 140 ℃ for 1 min and then reaches 250 ℃ with the rate of 20 ℃/min. b t (min) was the retention time of the first eluted analyte. Other separation conditions of MOF-based columns were given in Table S2 and S3 (Supporting information).

  Furthermore, MIL-125, MIL-125-NH_2, and MIL-143-TATB coated columns exhibited good repeatability within six times injections (Figs. S17–S19 in Supporting information) and good stability after storage for one month (Figs. S20 and S21 in Supporting information). The column efficiency of these columns was determined by using n-octane as the target (Fig. S22 in Supporting information). Corresponding to their minimum HETP, the column efficiency was 2128 plates/m for the MIL-125 coated column, 2703 plates/m for the MIL-125-NH₂ coated column, 3846 plates/m for the MIL-143-TATB coated column, and 252 plates/m for the MIL-143-BTB coated column, respectively (Table S6 in Supporting information). Then, the McReynolds constants of the four columns were measured and listed in Table S7 (Supporting information). The detection conditions were carefully optimized, but only benzene was eluted from the four columns. The analytes including n-butanol, 2-pentanone, nitropropane, and pyridine did not elute from these columns. We supposed that the metal sites of MOFs induced strong metal affinity for these analytes, resulting in analytes being firmly adsorbed in MOFs.

  To reveal the mechanism of improved separation performance of MIL-125-NH₂ and MIL-143-TATB, both the kinetic diffusion and thermodynamic adsorption experiments were accomplished. The kinetic parameter, the resistance to mass transfer coefficient (C_s), was calculated according to the Golay equation. And the thermodynamic parameters, including adsorption enthalpy (ΔH), adsorption entropy (ΔS), and Gibbs free energy (ΔG) were calculated according to the van't Hoff equation (see Supporting information for detail).

  The reason for improved R_s of styrene/ethylbenzene on the MIL-125-NH₂ coated column was first investigated. On the one hand, the C_s values of ethylbenzene on the two columns were similar, while the C_s value of styrene on the MIL-125-NH₂ coated column was larger than that on the MIL-125 coated column (Figs. 4a and b and Table S8 in Supporting information), which resulted in a slight peak tailing phenomenon of styrene on the MIL-125-NH₂ coated column. Typically, larger C_s values lead to poor separation results, but the MIL-125-NH₂ coated column exhibited higher separation performance than MIL-125, indicating kinetic diffusion was not the driving force here. On the other hand, the van't Hoff plots showed linear relations (Fig. 4c), which indicated that the mechanism of interactions between MOFs and analytes did not change over the measured temperature range [38,39,14]. All the ΔG values were negative, suggesting that the mass transfer of analytes from the mobile phase to the MOF-based stationary phases was a spontaneous process (Table S9 in Supporting information). The ΔH value of styrene on the MIL-125 coated column was similar to that of ethylbenzene, implying that MIL-125 provided approximate interactions to styrene and ethylbenzene (Fig. 4d). In contrast, the MIL-125-NH₂ coated column increased the thermodynamic differentiation of styrene (-64.96 ± 0.31 kJ/mol) and ethylbenzene (-61.96 ± 0.48 kJ/mol). The more negative ΔS value of styrene on the MIL-125-NH₂ coated column also proved this conclusion (Table S9 in Supporting Information). The increased differentiation of thermodynamic interactions led to a higher separation resolution of styrene and ethylbenzene on the MIL-125-NH₂ coated column.

  Figure 4

  

  Figure 4.  (a) The Golay plots of styrene and ethylbenzene on MIL-125 and MIL-125-NH₂ coated columns at 503.15 K. (b) The calculated C_s constants of styrene and ethylbenzene on MIL-125 and MIL-125-NH₂ coated columns. (c) The van't Hoff plots of styrene and ethylbenzene on MIL-125 and MIL-125-NH₂ coated columns. (d) The calculated ΔH values of styrene and ethylbenzene on MIL-125 and MIL-125-NH₂ coated columns. (e) The van't Hoff plots of xylene isomers on MIL-143-BTB and MIL-143-TATB coated columns. (f) The calculated ΔH values of xylene isomers on MIL-143-BTB and MIL-143-TATB coated columns.

  DownLoad: Full-Size Img PowerPoint

  The mechanism of reversed elution order of benzene/cyclohexane on MIL-125 and MIL-125-NH₂ was also investigated by thermodynamic adsorption experiments (Fig. S23 in Supporting information). The ΔH value of cyclohexane was more negative than that of benzene on the MIL-125 coated column, but the opposite was true on the MIL-125-NH₂ coated column (Table S9 in Supporting information). This indicated that MIL-125 had a stronger interaction with cyclohexane, while MIL-125-NH₂ provided a stronger interaction with benzene. Thus, the two columns presented reversed elution order. We supposed that the amino group in organic ligands of MIL-125-NH₂ introduced additional N-H···π interactions towards benzene, which resulted in stronger retention of benzene in the MIL-125-NH₂ coated column. The increased thermodynamic interaction was also observed on the MIL-143-TATB coated column. The ΔH values of xylene isomers on the MIL-143-TATB coated column were more negative than those on the MIL-143-BTB coated column (Figs. 4e and f), indicating that MIL-143-TATB provided stronger thermodynamic interactions with analytes. In addition, the order of enthalpies of xylene isomers was p-xylene < m-xylene < o-xylene on the MIL-143-TATB coated column, which was consistent with the elution order of xylene isomers (Fig. 3d). This obvious difference in adsorption enthalpy proved that MIL-143-TATB provided distinguished interactions with xylene isomers, which led to good separation results. However, ΔH values of xylene isomers on the MIL-143-BTB coated column were similar, resulting in low separation ability. MIL-143-BTB and MIL-143-TATB possessed the same structure but different organic ligands. Hence, we believed that the triazine core of the TATB ligand was the factor for increasing thermodynamic interactions with analytes, and then improving the separation ability of the MIL-143-TATB coated column. All the results proved that the increase of thermodynamic differentiation between MOFs and analytes was the key to improving the separation ability of MOF-based stationary phases.

  In summary, two types of MOF, MIL-125/MIL-125-NH₂, and MIL-143-BTB/MIL-143-TATB were synthesized, characterized, and applied as stationary phases for GC separations of various isomers. Compared with the MIL-125 and MIL-143-BTB, MIL-125-NH₂ and MIL-143-TATB exhibited better separation performance. Besides, the MIL-125-NH₂ coated column also exhibited exceedingly higher capability for the separation of xylene isomers than other reported stationary phases and commercial columns. Both the amino groups and triazine core in organic ligands strengthened the thermodynamic interactions between MOFs and analytes, which resulted in enhanced separation ability This work focuses on the design of high-performance MOF-based stationary phases with modulable thermodynamic interactions, which provides an innovative approach to improving the resolution of stationary phases.

  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.

  CRediT authorship contribution statement

  Sha-Sha Meng: Writing – original draft, Investigation, Data curation. Xiao-Yi Fu: Investigation, Data curation. Hai-Yue Wei: Investigation, Data curation. Ming Xu: Writing – review & editing, Validation, Project administration. Zhi-Yuan Gu: Writing – review & editing, Project administration, Conceptualization.

  Acknowledgments

  This work is supported by the National Natural Science Foundation of China (Nos. 22174067, 22204078, 22374077, and 22474059), the Natural Science Foundation of Jiangsu Province of China (No. BK20220370), Jiangsu Provincial Department of Education (No. 22KJB150009), Jiangsu Association for Science and Technology (No. TJ-2023-076), and the Priority Academic Program Development of Jiangsu Higher Education Institutions. This work is carried out with the support of BL11B, BL14W1, and BL17B at the Shanghai Synchrotron Radiation Facility (SSRF) and BL01B at the National Synchrotron Radiation Laboratory (NSRL).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110720.

  
  
• 1. [1]

     Q. Zheng, J. Huang, Y. He, et al., ACS Appl. Mater. Interfaces 14 (2022) 9754–9762. doi: 10.1021/acsami.1c20989

  2. [2]

     X.D. Cheng, Z. Zhang, Y.P. Li, Anal. Chim. Acta 1265 (2023) 341337. doi: 10.1016/j.aca.2023.341337

  3. [3]

     B. Huang, K. Li, Q.Y. Ma, et al., Anal. Chem. 95 (2023) 13289–13296. doi: 10.1021/acs.analchem.3c02438

  4. [4]

     H. Li, X. Wang, C. Shi, et al., Chin. Chem. Lett. 34 (2023) 107606. doi: 10.1016/j.cclet.2022.06.029

  5. [5]

     T. Böhle, F. Mertens, Microporous Mesoporous Mater. 216 (2015) 82–91. doi: 10.1016/j.micromeso.2015.03.019

  6. [6]

     A.S. Münch, F.O.R.L. Mertens, Microporous Mesoporous Mater. 270 (2018) 180–188. doi: 10.1016/j.micromeso.2018.05.012

  7. [7]

     M. Xu, S.S. Meng, P. Cai, et al., ACS Cent. Sci. 8 (2022) 184–191. doi: 10.1021/acscentsci.1c01344

  8. [8]

     W.C. Deng, H.L. Qian, C. Yang, et al., ACS Appl. Mater. Interfaces 15 (2023) 54977–54985. doi: 10.1021/acsami.3c13853

  9. [9]

     S.S. Meng, T. Han, Y.H. Gu, et al., Anal. Chem. 94 (2022) 14251–14256. doi: 10.1021/acs.analchem.2c02575

  10. [10]

      Z.H. Wang, C. Yang, T. Liu, et al., Anal. Chem. 95 (2023) 8145–8149. doi: 10.1021/acs.analchem.3c01550

  11. [11]

      F. Aldaeus, Y. Thewalim, A. Colmsjö, Anal. Bioanal. Chem. 389 (2007) 941–950. doi: 10.1007/s00216-007-1528-0

  12. [12]

      C.F. Poole, S.K. Poole, J. Chromatogr. A 1184 (2008) 254–280. doi: 10.1016/j.chroma.2007.07.028

  13. [13]

      C. Chen, H. Guan, H. Li, et al., Angew. Chem. Int. Ed. 61 (2022) e202201646. doi: 10.1002/anie.202201646

  14. [14]

      A. Natraj, W. Ji, J. Xin, et al., J. Am. Chem. Soc. 144 (2022) 19813–19824. doi: 10.1021/jacs.2c07166

  15. [15]

      C.T. He, L. Jiang, Z.M. Ye, et al., J. Am. Chem. Soc. 137 (2015) 7217–7223. doi: 10.1021/jacs.5b03727

  16. [16]

      Y. Gu, J.J. Zheng, K.I. Otake, et al., Angew. Chem. Int. Ed. 60 (2021) 11688–11694. doi: 10.1002/anie.202016673

  17. [17]

      W.Q. Tang, J.Y. Xu, Z.Y. Gu, Chem. Asian J. 14 (2019) 3462–3473. doi: 10.1002/asia.201900738

  18. [18]

      S.M. Xie, L.M. Yuan, J. Sep. Sci. 40 (2017) 124–137. doi: 10.1002/jssc.201600808

  19. [19]

      J. Zhang, Z. Chen, J. Chromatogr. A 1530 (2017) 1–18. doi: 10.1016/j.chroma.2017.10.065

  20. [20]

      X. Zhao, Y. Wang, D.S. Li, et al., Adv. Mater. 30 (2018) 1705189. doi: 10.1002/adma.201705189

  21. [21]

      S.K. Firooz, D.W. Armstrong, Anal. Chim. Acta 1234 (2022) 340208. doi: 10.1016/j.aca.2022.340208

  22. [22]

      A.C. Forse, M.I. Gonzalez, R.L. Siegelman, et al., J. Am. Chem. Soc. 140 (2018) 1663–1673. doi: 10.1021/jacs.7b09453

  23. [23]

      H. Yang, J.J. Liu, W.Q. Tang, et al., Anal. Chem. 95 (2023) 18760–18766. doi: 10.1021/acs.analchem.3c03283

  24. [24]

      W.T. Kou, C.X. Yang, X.P. Yan, J. Mater. Chem. A 6 (2018) 17861–17866. doi: 10.1039/c8ta06804f

  25. [25]

      Q. Zhang, Y. Cui, G. Qian, Coord. Chem. Rev. 378 (2019) 310–332. doi: 10.1016/j.ccr.2017.10.028

  26. [26]

      L. Fan, X.P. Yan, Talanta 99 (2012) 944–950. doi: 10.1016/j.talanta.2012.07.063

  27. [27]

      Z.Y. Gu, X.P. Yan, Angew. Chem. Int. Ed. 49 (2010) 1477–1480. doi: 10.1002/anie.200906560

  28. [28]

      Y. Zhang, G. Li, L. Kong, et al., Fuel 219 (2018) 103–110. doi: 10.1109/icamechs.2018.8507116

  29. [29]

      F. Guo, M. Yang, R.X. Li, et al., ACS Catal. 12 (2022) 9486–9493. doi: 10.1021/acscatal.2c02789

  30. [30]

      H. Chevreau, T. Devic, F. Salles, et al., Angew. Chem. Int. Ed. 52 (2013) 5056–5060. doi: 10.1002/anie.201300057

  31. [31]

      E. Virmani, O. Beyer, U. Lüning, et al., Mater. Chem. Front. 1 (2017) 1965–1974. doi: 10.1039/C7QM00263G

  32. [32]

      M.A. Moreira, M.P.S. Santos, C.G. Silva, et al., Adsorption 24 (2018) 715–724. doi: 10.1007/s10450-018-9976-8

  33. [33]

      X.Y. Ji, Y.Y. Wang, Y. Li, et al., Nano Res. 15 (2022) 6045–6053. doi: 10.1007/s12274-022-4301-y

  34. [34]

      M.Z. Hussain, M. Bahri, W.R. Heinz, et al., Microporous Mesoporous Mater. 316 (2021) 110957. doi: 10.1016/j.micromeso.2021.110957

  35. [35]

      J. Jin, P. Li, D.H. Chun, et al., Adv. Funct. Mater. 31 (2021) 2102511. doi: 10.1002/adfm.202102511

  36. [36]

      S. Naghdi, A. Cherevan, A. Giesriegl, et al., Nat. Commun. 13 (2022) 282. doi: 10.1038/s41467-021-27775-7

  37. [37]

      Z.R. Tao, J.X. Wu, Y.J. Zhao, et al., Nat. Commun. 10 (2019) 2911. doi: 10.1038/s41467-019-10971-x

  38. [38]

      Z.L. Fang, S.R. Zheng, J.B. Tan, et al., J. Chromatogr. A 1285 (2013) 132–138. doi: 10.1016/j.chroma.2013.02.024

  39. [39]

      T.T. Ma, C. Yang, H.L. Qian, et al., J. Chromatogr. A 1673 (2022) 463085. doi: 10.1016/j.chroma.2022.463085

• 

  Scheme 1  Schematic diagram of improving separation ability of MOF-stationary phases by increasing the thermodynamic differentiation.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  (a) Representation of octahedral cage (12.0 Å) and tetrahedral cage (6.0 Å) in MIL-125. (b) PXRD patterns of synthesized MIL-125, MIL-125-NH₂, and simulated MIL-125. (c) FTIR spectra of MIL-125 and MIL-125-NH₂. (d) The N₂ adsorption-desorption isotherms measured at 77 K and (e) pore size distribution of MIL-125 and MIL-125-NH₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Representation of large tetrahedron (11.0 Å) and small tetrahedron (9.0 Å) in MIL-143-BTB. (b) PXRD patterns of synthesized MIL-143-BTB, MIL-143-TATB, and simulated MIL-143-BTB. The patterns were collected at BL17B1 beamline, Shanghai Synchrotron Radiation Facility (SSRF) (λ = 1.24 Å). (c) TGA plots of MIL-143-BTB and MIL-143-TATB. Experiment condition: temperature ramp from room temperature to 800 ℃ at 10 ℃/min under the oxygen atmosphere.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Gas chromatogram on MIL-125 and MIL-125-NH₂ coated capillary columns at the N₂ flow of 0.5 mL/min for the separation of (a) xylene isomers, (b) styrene/ethylbenzene, and (c) benzene/cyclohexane. Gas chromatogram on MIL-143-BTB and MIL-143-TATB coated capillary columns at the N₂ flow of 0.5 mL/min for the separation of (d) xylene isomers, (e) C₆H₁₄ isomers, and (f) the mixture of 1-hexene, n-hexane, and cyclohexane. The numbers on the left represented the starting time of the elution of the first chromatographic peaks. Comparison of separation resolution on (g) MIL-125 and MIL-125-NH₂ coated columns and (h) MIL-143-BTB and MIL-143-TATB coated columns. (i) Comparison of the separation resolution of pX/oX and pX/mX on different stationary phases.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) The Golay plots of styrene and ethylbenzene on MIL-125 and MIL-125-NH₂ coated columns at 503.15 K. (b) The calculated C_s constants of styrene and ethylbenzene on MIL-125 and MIL-125-NH₂ coated columns. (c) The van't Hoff plots of styrene and ethylbenzene on MIL-125 and MIL-125-NH₂ coated columns. (d) The calculated ΔH values of styrene and ethylbenzene on MIL-125 and MIL-125-NH₂ coated columns. (e) The van't Hoff plots of xylene isomers on MIL-143-BTB and MIL-143-TATB coated columns. (f) The calculated ΔH values of xylene isomers on MIL-143-BTB and MIL-143-TATB coated columns.

  下载: 全尺寸图片 幻灯片

  Table 1.  Separation conditions of different isomers on MIL-125, MIL-125-NH₂, and MIL-143-TATB coated columns.

  Analyte	MIL-125 column				MIL-125-NH2 column				MIL-143-TATB column
  	Split ratio	Ta	tb		Split ratio	T	t		Split ratio	T	t
  Xylene	20:1	150-250@20	4.36		20:1	150-250@20	5.23		10:1	120-250@20	6.34
  Chlorotoluene	20:1	180-250@20	3.39		20:1	200-250@20	4.05		10:1	160-250@20	4.83
  C6H14	20:1	140-250@20	2.59		10:1	140-250@20	2.57		20:1	120-250@20	3.80
  C8H18	5:1	180-250@20	2.97		30:1	180-250@20	3.32		20:1	160-250@20	3.70
  a T (℃) was the temperature program of isomers in MOF-based columns, i.e., “140-250@20” implied that the temperature is maintained at 140 ℃ for 1 min and then reaches 250 ℃ with the rate of 20 ℃/min. b t (min) was the retention time of the first eluted analyte. Other separation conditions of MOF-based columns were given in Table S2 and S3 (Supporting information).

  下载: 导出CSV
•