Continuous polycrystalline ZIF-8 membrane supported on CO₂-selective mixed matrix supports for CO₂/CH₄ separation

Abstract

Thin and compact ZIF-8 membranes were successfully synthesized on PES-ZIF-8 mixed matrix supports via secondary growth. An enhancement in the overall CO₂/CH₄ selectivity was demonstrated by incorporating both a selective ZIF-8 layer and a PES-ZIF-8 mixed matrix support in a membrane.

---

The development of energy efficient and environmentally friendly separation processes has become the current research trend for dealing with global issues such as natural gas purification and CO₂ capture. With lower energy costs and fewer negative environmental impacts, membrane separation technology is a promising alternative as compared to conventional industrial processes.¹ Extensive research has been performed during the last three decades² to enhance both permeability and selectivity of polymeric membranes, due to a trade-off between permeability and selectivity suffered by conventional polymeric membrane, as encapsulated in the Robeson upper bound relationship in 1991,³ and revisited in 2008.⁴

To overcome the trade-off limitations of polymeric membrane, the idea of mixed matrix membranes (MMMs) with selective fillers deposited into polymeric membranes is expected to complement the permeability and selectivity of the overall membrane by combining the advantages of both phases. Although MMMs^5–7 have been reported to exhibit attractive separation properties, unresolved issues of poor adhesion between the filler phase and polymer continuous phase resulting in deteriorating selectivity have impeded their successful utilization in industrial applications. ZIF-8 (zeolitic imidazolate framework-8), with sodalite topology connected from zinc(II) cations and 2-methylimidazole anions and possessing a pore cavity of 1.16 nm and pore aperture of 0.34 nm, has attracted attention as a filler for mixed matrix membranes^8–10 due to its good compatibility with polymers combined with its molecular sieving effect.

For a supported metal organic framework (MOF) membrane, the compatibility between the selective layer and support is vital in determining the integrity of the overall membrane and the permeation stability. Several published studies related to ZIF-8 membranes were directed towards the growth of continuous and thin ZIF-8 layers on porous supports where the used supports were predominantly an inorganic material such as Al₂O₃ (ref. 11–14) or TiO₂.¹⁵ Nonetheless, continuous and defect-free membranes cannot be formed on ceramics without support modification or seeding steps^16–18 due to weak interactions of MOFs on the inorganic supports. The fabrication of MOF composite membranes would be more facile and straightforward using porous polymer as the support layers, particularly due to the rapid preparation method of the polymer support, easy scaleup as compared to inorganic supports, and better adhesion of MOFs with polymers due to favourable interactions between the polymer support and organic ligands of MOFs.

Consequently, several researchers have put their efforts towards synthesizing ZIF membranes on polymeric supports. For instance, Brown et al.¹⁹ reported the synthesis of ZIF-90 membranes on Torlon hollow fiber supports, achieving CO₂/N₂ and CO₂/CH₄ selectivities of 3.5 and 1.5, respectively, well above their corresponding Knudsen selectivities (0.8 and 0.6). Ge et al.²⁰ fabricated a thin and continuous ZIF-8 layer on a porous polyethersulfone (PES) substrate, showing good affinity of both ZIF-8 and PES materials. Their PES-supported ZIF-8 membrane successfully achieved H₂/Ar, H₂/O₂, H₂/N₂, and H₂/CH₄ ideal separation factors of 9.7, 10.8, 9.9, and 10.7, respectively. Barankova et al.²¹ synthesized ZIF-8 membranes on a mixed matrix support coated with polyetherimide and zinc oxide nanoparticles, where the zinc oxide particles served as a secondary metal source for the growth of ZIF-8 membrane. Their membrane exhibited H₂/C₃H₈ ideal selectivity of 22.4, which exceeded the corresponding Knudsen selectivity by a factor of 4. Recently, Cacho-Bailo et al.²² prepared ZIF-8 membrane on a porous polysulfone (PSF) support, where the ZIF-8 comprised over 75 wt% of the overall membrane, indicating an almost self-supported MOF membrane. Their permeation tests of gas mixture separations displayed H₂/CH₄ and H₂/N₂ selectivities of 10.5 and 12.4, respectively.

To the best of our knowledge, the performance of a ZIF-8 membrane supported on mixed matrix supports (PES/ZIF-8) for CO₂/CH₄ gas separation has not yet been reported. Due to its high hydrothermal stability, ZIF-8 was selected as one of the most stable and promising MOFs for gas separation applications. Herein, we demonstrate the synthesis of dense ZIF-8 membrane supported on tailor-made polyethersulfone (PES)/ZIF-8 mixed matrix supports for CO₂/CH₄ separation. We prepared continuous ZIF-8 layers on neat PES, 5 wt% and 10 wt% ZIF-8 loadings in PES supports, as shown in Fig. 1, where PES was chosen as the polymer matrix due to its feasibility in gas separations.^23,24 Improved performance can be expected by fabricating a dense ZIF-8 layer on the mixed matrix support as compared to the polymer support via a short time for secondary seeded growth. It is speculated that the ZIF-8 crystals embedded in the PES polymer as mixed matrix supports have a positive effect in the overall CO₂/CH₄ selectivity predominantly due to its CO₂ adsorption effect.²⁵ One side of the membrane is covered with a continuous selective ZIF-8 layer, while on the other side of the membrane, the PES-ZIF-8 mixed matrix will act as the membrane support as well as engage in the overall CO₂/CH₄ separation performance.

Fig. 1 Schematic illustration of ZIF-8 membranes fabricated on neat PES, 5% ZIF-8-PES, and 10% ZIF-8-PES mixed matrix supports for CO₂/CH₄ separations.

As a vital role in CO₂/CH₄ separation, the top layer ZIF-8 membrane should be continuous and dense to avoid gas leakage leading to insignificant selectivity. Several researchers^25–27 have reported that the presence of sodium formate in the precursor solution enhances the intergrowth of ZIF-8 crystals, mainly due to the higher deprotonation of 2-methylimidazole leading to the growth of ZIF-8 crystals in all directions. In this work, ZIF-8 membranes were prepared by secondary seeded growth on a mixed matrix support using a synthesis precursor with a molar ratio of 1ZnCl₂–5.8Hmim (2-methylimidazole)–4HCOONa (sodium formate)–180methanol. In addition, the seeding step is essential for determining the quality of the final ZIF-8 membrane, which is created by the seeds on the support acting as a base layer for the formation of gap-free ZIF-8 during secondary growth. We adopted the seeding technique recommended by Liu et al.²⁸ by performing rub seeding followed by dip coating in 1% ZIF-8 solution (refer to the ESI†). The ZIF-8 seeds will be fixed on the surface of the porous supports after seeding to provide nucleation sites for the growth of ZIF-8 crystals into a dense membrane during the solvothermal process, ensuring the adhesivity between the ZIF-8 layer and the supports.

Fig. 2 shows the FESEM (field emission scanning electron microscopy) cross-sectional micrographs of the ZIF-8 membranes after secondary seeded growth. Continuous ZIF-8 membranes were fabricated on sponge-like PES and PES-ZIF-8 mixed matrix supports. No visible interface between the ZIF-8 layer and supports (Fig. 2a–c) can be seen in the SEM images, revealing that both ZIF-8 and the mixed matrix supports are compatible and well-adhered together, indicating the successful fabrication of asymmetric membranes. The cross-sections also clearly show the continuous and homogeneous morphology of ZIF-8 across their thickness of approximately 6.5–8 μm after 6 hours of growth. Our composite membranes with mixed matrix supports not only have a greater degree of flexibility as compared to conventional inorganic supports, but with ZIF-8 crystals fixed within the PES polymer, will also play an important role in enriching the CO₂ adsorption²⁵ participating in CO₂/CH₄ separation. The top view of the secondary growth ZIF-8 layer is shown in Fig. 3. In the lower magnification view as shown in Fig. 3a, it can be observed that the layer is very dense and compact with no visible cracks present. Increased magnification of the FESEM images shown in Fig. 3b presents a well inter-grown and continuous ZIF-8 membrane. In addition, the shape and intergrowth of the crystals are comparable to those of ZIF-8 membranes synthesized on inorganic supports.

Fig. 2 FESEM cross-sectional micrographs of continuous ZIF-8 membranes supported on (a) neat PES, (b) 5 wt% ZIF-8-PES, and (c) 10 wt% ZIF-8-PES, mixed matrix support.

Fig. 3 FESEM micrographs showing the surface morphologies of the mixed matrix-supported ZIF-8 membranes with (a) low magnification and (b) high magnification.

Fig. 4 shows the X-ray diffraction (XRD) patterns of the ZIF-8 membrane supported on neat PES, 5 wt%, and 10 wt% PES-ZIF-8 mixed matrix supports. The high crystallinity of the synthesized ZIF-8 membranes was verified by powder X-ray diffraction (XRD) analysis, with values that were comparable to those in the literature.^29,30 The peaks appearing at 2θ = 7.3, 10.3, 12.6, 14.5, 16.3, and 18.0° were attributed to ZIF-8 peaks, implying that high quality ZIF-8 crystals in the absence of the amorphous phase were prepared. The ZIF-8 crystals exhibited a pore volume and BET surface area of 0.603 cm³ g⁻¹ and 1107 m² g⁻¹, respectively, comparable to previously published values.^8,14,31 These crystals were used as seeds for the synthesis of membranes via secondary growth as well as the fillers in the PES mixed matrix support. Both CO₂ and CH₄ adsorption on the ZIF-8 crystals was investigated to evaluate the strength of adhesion between the gas molecules, as reported in our previous work.²⁵ The volume of CO₂ (2.4 mmol CO₂ per g ZIF-8) adsorbed across the ZIF-8 crystals was larger than that of CH₄ (1.1 mmol CH₄ per g ZIF-8), and was mainly caused by the narrow bottleneck structure of the ZIF-8 pore aperture, where the ZIF-8 crystalline structure is constituted of large cavities (1.16 nm) connected through small apertures (0.34 nm).³² In addition, it could also be attributed to the polar nature of the carbon–oxygen bond in CO₂, which is able to interact strongly with 2-methylimidazole molecules or ZIF-8 crystal sites.¹¹

Fig. 4 X-ray diffraction patterns of ZIF-8 membranes supported on (a) neat PES, (b) 5 wt% ZIF-8-PES MMM, and (c) 10 wt% ZIF-8-PES MMM. (The symbol * denotes ZIF-8 peaks.)

For the mixed matrix supports, we obtained both improved permeance and selectivity by introducing ZIF-8 crystals into the PES continuous phase demonstrated via single gas permeations, as shown in Table 1. An increase in both CO₂ and CH₄ permeance was observed on both 5 wt% ZIF-8-PES and 10 wt% ZIF-8-PES, respectively, as compared to the neat PES membrane. The increase in permeance could be reasoned by the interaction between polymer chains and ZIF-8 crystals, where the ZIF-8 crystals will interrupt the chain packing of the polymer matrix leading to an increase in free volume among the polymer chains, resulting in higher gas permeance. Although these mixed matrix membranes have shown improvement in terms of their permeance, nonetheless, their CO₂/CH₄ selectivity remained intact without exhibiting a trade-off. The uninterrupted CO₂/CH₄ selectivity as a result of further increments in ZIF-8 loadings is most likely due to the increased presence of selective particles in the matrix supports, which allows for more selective adsorption of CO₂ on the ZIF-8 layer as compared to CH₄ gas, leading to an increase in CO₂/CH₄ selectivity.

Table 1 CO₂ and CH₄ gas separation performances of mixed matrix supports and mixed matrix-supported ZIF-8 membranes^a

Membranes	CO2 permeance (10^−9 mol m^−2 s Pa)	CH4 permeance (10^−10 mol m^−2 s Pa)	Ideal selectivity
a The symbol * refers to secondary growth of the ZIF-8 layers supported on mixed matrix supports.
Neat PES	8.36	9.29	9.0
5 wt% ZIF-8-PES	8.86	9.54	9.3
10 wt% ZIF-8-PES	9.64	9.98	9.7
*Neat PES	7.06	4.80	14.7
*5 wt% ZIF-8-PES	7.47	4.91	15.2
*10 wt% ZIF-8-PES	8.05	5.13	15.7

---

Membranes	CO2 permeance (10^−9 mol m^−2 s Pa)	CH4 permeance (10^−10 mol m^−2 s Pa)	Binary selectivity
*Neat PES	6.55	4.75	13.8
*5 wt% ZIF-8-PES	6.80	4.85	14.0
*10 wt% ZIF-8-PES	7.34	5.03	14.6

---

Looking at Table 1, the 5 wt% and 10 wt% ZIF-8-PES mixed matrix supports show a 6% and 15.3% increase in CO₂ permeance as compared to the neat PES, respectively, while the 5 wt% and 10 wt% ZIF-8-PES mixed matrix membranes show a 2.7% and 7.4% increase in CH₄ permeance as compared to that of neat PES, respectively. Therefore, it can be deduced that the unaffected selectivity was contributed by the larger CO₂ permeance as a result of CO₂ adsorption on ZIF-8 crystals. On the contrary, although it has been attempted to increase the ZIF-8 loadings in PES polymer to 20 wt%, unfortunately, their poor mechanical properties with unselective voids were not sufficient to endure gas permeance measurements, which hampered the efforts of gas separations. Moreover, secondary growth of the ZIF-8 layer on 20 wt% ZIF-8-PES was affirmed to be non-feasible, leading to cracking of the membranes into pieces due to its brittleness during solvothermal growth.

For secondary growth of the ZIF-8 layer on the mixed matrix supports, their CO₂/CH₄ selectivities are indeed enhanced to a greater extent (approximately 63%), where both CO₂ and CH₄ have shown an average of 16% and 49% reduction in their permeances, respectively, as compared to only mixed matrix supports. It is astonishing that the additional ZIF-8 layer does not impose a significant resistance to its gas permeations, possibly due to the inherent pore flexibility of the MOF. Reported ZIF-8 membranes have generally shown moderate CO₂/CH₄ selectivity in the range of 3–7,^14,33–35 possibly due to their flexible framework structure.³⁶ The remarkable improvement in CO₂/CH₄ selectivities from approximately 9 to 15 after fabrication of the ZIF-8 layer on mixed matrix supports could be ascribed to the combined molecular sieving and CO₂ adsorption of the ZIF-8 layer, as well as the vital role that is played in defect abatement by diminishing the existing defect flow through the skin layer of the mixed matrix membranes. Nevertheless, these factors improved the overall selectivity owing to the molecular sieving effect, corroborating the advantages of the ZIF-8 layer on mixed matrix membranes. A rough estimate of improvement on the CO₂/CH₄ selectivity with top layer ZIF-8 membranes is approximately 5–6, whereby the overall good CO₂/CH₄ selectivity of the membrane was also contributed by the CO₂-selective support rather than solely the ZIF-8 layers.

Although the improvement in CO₂/CH₄ selectivities of the mixed matrix supports was not as compelling as compared to its gas permeability, which increases substantially with increasing ZIF-8 loadings, the conceptual theory of fabricating an additional selective ZIF-8 layer on the mixed matrix membranes that could stimulate the overall CO₂/CH₄ selectivities would serve as an innovative approach for future research. Presuming that the additional selective layer will be fabricated on existing high-performance mixed matrix membranes,^7,37,38 it definitely will have a positive impact on the gas permeation properties of the overall membrane.

The secondary growth of ZIF-8 membranes on mixed matrix supports was further evaluated by single gas permeation up to 2 bar pressure difference, as shown in Fig. 5. It is self-evident that the neat PES, 5 wt%, and 10 wt% supported ZIF-8 show a modest decrease in CO₂ permeance with increasing pressure difference. With ZIFs reported to be a promising CO₂ storage material exhibiting high CO₂ adsorption,^39,40 the decrease in CO₂ permeance with increasing pressure difference could be reasoned by the rapid filling up of CO₂ on the ZIF-8 crystal adsorption sites, which further leads to a decrease in the jumping frequency of CO₂ molecules to their neighbouring sites in the ZIF-8 structure.⁴¹ Therefore, in our study, a slight decrease in CO₂ permeance was observed in driving the decrease in CO₂ permeance of the neat PES-ZIF-8-supported ZIF-8 membrane. However, it has been generally reported that an increase in pressure difference on polymeric membranes usually entails membrane compaction and a reduction of free volume within the polymer, leading to reduced gas mobility and a decrease in permeance. As a control experiment, the CO₂ permeance on neat PES was also measured up to a pressure difference of 2 bar. However, it shows a slight permeance increase, which excludes the possibility of a permeance reduction due to the nature of the polymeric support, with a pressure difference of up to 2 bar.

Fig. 5 CO₂ and CH₄ single gas permeations with the effect of a pressure gradient up to 2 bar on the mixed matrix-supported ZIF-8 membranes.

Owing to their superior separation performances, mixed gas permeation tests of CO₂/CH₄ (50/50 mol%) were performed on mixed matrix-supported ZIF-8 membranes at room temperature and 1 bar pressure difference. In comparison to single gas permeation data, both CO₂ and CH₄ presented a decrease in permeance and selectivity for all supported ZIF-8 membranes. CO₂ shows an average of 8.4% decrease in permeance while CH₄ shows an average of 1.3% decrease in permeance as compared to their single permeation data, respectively. Perhaps the decrease was due to the competitive adsorption and permeation between the two penetrants, where the slowly permeating CH₄ largely hinders the CO₂ permeation rate. Notwithstanding the interactions between both gases, their CO₂/CH₄ selectivities were still kept at approximately 14, marking the evidence of their feasible application in mixed gas separations.

The separation performances of these membranes have also been included in the Robeson plot shown in Fig. 6. Average thicknesses of 100 μm and 7 μm were used to calculate the permeability in Barrers for mixed matrix supports and top layer ZIF-8 membranes, respectively. These data points are close to the present upper bound for polymeric membranes, and slightly lower than those of thermally rearranged polymeric membranes.

Fig. 6 Robeson plot for CO₂–CH₄ mixtures.⁴ The and symbols represent the performance of the MMM-supported ZIF-8 membranes upon single gas and binary gas permeation, respectively, while the symbol represents the performance of the MMM supports.

Conclusions

For the first time, an approach to simultaneously enhance both the permeability and selectivity were demonstrated by fabrication of ZIF-8 membranes via secondary seeded growth on ZIF-8-PES mixed matrix supports. The natural compatibility properties of both ZIF-8 with PES polymer are assets to successfully prepare robust CO₂ selective hybrid membranes. It was demonstrated that the enhancement in CO₂/CH₄ selectivity was reinforced by fabricating ZIF-8 membranes on mixed matrix supports, predominantly due to both molecular sieving and CO₂ adsorption properties. Positively, the novel ZIF-8 membranes supported on ZIF-8-PES mixed matrix supports still exhibit superior CO₂/CH₄ separation despite being subjected under binary gas separations. This could possibly make them promising candidate in membrane applications such as CO₂ capture or natural gas purifications.

Acknowledgements

We would like to thank the Ministry of Higher Education Malaysia through the Long-term Research Grant Scheme (LRGS) (A/C number 2110226-113-00) for the financial support given.

Notes and references

1. Z. Y. Yeo, T. L. Chew, P. W. Zhu, A. R. Mohamed and S.-P. Chai, J. Nat. Gas Chem., 2012, 21, 282–298.
2. W. J. Koros and R. Mahajan, J. Membr. Sci., 2000, 175, 181–196.
3. L. M. Robeson, J. Membr. Sci., 1991, 62, 165–185.
4. L. M. Robeson, J. Membr. Sci., 2008, 320, 390–400.
5. V. Nafisi and M.-B. Hägg, J. Membr. Sci., 2014, 459, 244–255.
6. L. Hao, P. Li, T. Yang and T.-S. Chung, J. Membr. Sci., 2013, 436, 221–231.
7. S. N. Wijenayake, N. P. Panapitiya, S. H. Versteeg, C. N. Nguyen, S. Goel, K. J. Balkus, I. H. Musselman and J. P. Ferraris, Ind. Eng. Chem. Res., 2013, 52, 6991–7001.
8. Q. Song, S. K. Nataraj, M. V. Roussenova, J. C. Tan, D. J. Hughes, W. Li, P. Bourgoin, M. A. Alam, A. K. Cheetham, S. A. Al-Muhtaseb and E. Sivaniah, Energy Environ. Sci., 2012, 5, 8359–8369.
9. C. Zhang, Y. Dai, J. R. Johnson, O. Karvan and W. J. Koros, J. Membr. Sci., 2012, 389, 34–42.
10. Y. Dai, J. R. Johnson, O. Karvan, D. S. Sholl and W. J. Koros, J. Membr. Sci., 2012, 401–402, 76–82.
11. G. Xu, J. Yao, K. Wang, L. He, P. A. Webley, C.-s. Chen and H. Wang, J. Membr. Sci., 2011, 385–386, 187–193.
12. Y. Pan, T. Li, G. Lestari and Z. Lai, J. Membr. Sci., 2012, 390, 93–98.
13. K. Tao, C. Kong and L. Chen, Chem. Eng. J., 2013, 220, 1–5.
14. S. R. Venna and M. A. Carreon, J. Am. Chem. Soc., 2010, 132, 76–78.
15. H. Bux, F. Liang, Y. Li, J. Cravillon, M. Wiebcke and J. Caro, J. Am. Chem. Soc., 2009, 131, 16000–16001.
16. J. Nan, X. Dong, W. Wang, W. Jin and N. Xu, Langmuir, 2011, 27, 4309–4312.
17. Y. Hu, X. Dong, J. Nan, W. Jin, X. Ren, N. Xu and Y. M. Lee, Chem. Commun., 2011, 47, 737–739.
18. Z. Xie, J. Yang, J. Wang, J. Bai, H. Yin, B. Yuan, J. Lu, Y. Zhang, L. Zhou and C. Duan, Chem. Commun., 2012, 48, 5977–5979.
19. A. J. Brown, J. R. Johnson, M. E. Lydon, W. J. Koros, C. W. Jones and S. Nair, Angew. Chem., Int. Ed., 2012, 51, 10615–10618.
20. L. Ge, W. Zhou, A. Du and Z. Zhu, J. Phys. Chem. C, 2012, 116, 13264–13270.
21. E. Barankova, N. Pradeep and K.-V. Peinemann, Chem. Commun., 2013, 49, 9419–9421.
22. F. Cacho-Bailo, B. Seoane, C. Téllez and J. Coronas, J. Membr. Sci., 2014, 464, 119–126.
23. S. Saedi, S. S. Madaeni, F. Seidi, A. A. Shamsabadi and S. Laki, Int. J. Greenhouse Gas Control, 2013, 19, 126–137.
24. S. Saedi, S. S. Madaeni and A. A. Shamsabadi, Chem. Eng. Res. Des., 2014 DOI:10.1016/j.cherd.2014.02.010.
25. Z. Y. Yeo, P. W. Zhu, A. R. Mohamed and S.-P. Chai, CrystEngComm, 2014, 16, 3072–3075.
26. M. C. McCarthy, V. Varela-Guerrero, G. V. Barnett and H.-K. Jeong, Langmuir, 2010, 26, 14636–14641.
27. M. Shah, H. T. Kwon, V. Tran, S. Sachdeva and H.-K. Jeong, Microporous Mesoporous Mater., 2013, 165, 63–69.
28. Y. Liu, Z. Yang, C. Yu, X. Gu and N. Xu, Microporous Mesoporous Mater., 2011, 143, 348–356.
29. H. T. Kwon and H.-K. Jeong, J. Am. Chem. Soc., 2013, 135, 10763–10768.
30. Y. Pan and Z. Lai, Chem. Commun., 2011, 47, 10275–10277.
31. D. Fairen-Jimenez, S. A. Moggach, M. T. Wharmby, P. A. Wright, S. Parsons and T. Düren, J. Am. Chem. Soc., 2011, 133, 8900–8902.
32. K. S. Park, Z. Ni, A. P. Côté, J. Y. Choi, R. Huang, F. J. Uribe-Romo, H. K. Chae, M. O'Keeffe and O. M. Yaghi, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 10186–10191.
33. A. Huang, Q. Liu, N. Wang and J. Caro, J. Mater. Chem. A, 2014 10.1039/c4ta00299g.
34. Y. Pan, B. Wang and Z. Lai, J. Membr. Sci., 2012, 421–422, 292–298.
35. Z. Y. Yeo, S.-P. Chai, P. W. Zhu and A. R. Mohamed, Microporous Mesoporous Mater, 2014, 196, 79–88,  DOI:10.1016/j.micromeso.2014.05.002.
36. H. Bux, A. Feldhoff, J. Cravillon, M. Wiebcke, Y.-S. Li and J. Caro, Chem. Mater., 2011, 23, 2262–2269.
37. Y. Dai, J. R. Johnson, O. Karvan, D. S. Sholl and W. J. Koros, J. Membr. Sci., 2012, 401–402, 76–82.
38. M. J. C. Ordoñez, K. J. Balkus Jr, J. P. Ferraris and I. H. Musselman, J. Membr. Sci., 2010, 361, 28–37.
39. E. Haldoupis, T. Watanabe, S. Nair and D. S. Sholl, ChemPhysChem, 2012, 13, 3449–3452.
40. W. Morris, B. Leung, H. Furukawa, O. K. Yaghi, N. He, H. Hayashi, Y. Houndonougbo, M. Asta, B. B. Laird and O. M. Yaghi, J. Am. Chem. Soc., 2010, 132, 11006–11008.
41. J. van den Bergh, S. Ban, T. J. H. Vlugt and F. Kapteijn, J. Phys. Chem. C, 2009, 113, 17840–17850.

---

Footnote

† Electronic Supplementary Information (ESI) available. See DOI: 10.1039/c4ra09547b

---