Hydrophilic pore-blocked metal–organic frameworks: a simple route to a highly selective CH₄/N₂ separation

Abstract

The separation of gas mixtures with similar thermodynamic and transport properties is a challenging issue. For that, various types of adsorbents and membranes have been introduced. To date, solving this difficult problem remains an urgent core task in the energy and environment fields. Herein, we introduce a simple method to achieve highly selective separation of CH₄/N₂. To demonstrate our concept, a CH₄/N₂ mixture was separated by Cu₃BTC₂ filled with water, with the water blocking the hydrophilic pore to reject the inclusion of gas molecules having weak affinity for water, and with the empty hydrophobic pores acting as gas storage sites. This led to high equilibrium selectivity of the CH₄/N₂ mixture, at 24.7, which is 6 times higher than the untreated Cu₃BTC₂ itself. Moreover, the formation of methane hydrate in the mesopore of MOF was observed.

---

Introduction

Pores, confined space, which allow molecular passage and facilitate occupation, are meaningful structures in various science disciplines. They are observed in various systems ranging from cell membrane to rocks. Materials incorporating pores are called porous materials, and these have provided unprecedented opportunities to develop emerging technologies such as catalysts, fuel cells, ion exchange mechanisms, molecular storage systems and separation mechanisms. In particular, porous materials reveal tremendous potential in the field of molecular separation in relation to the synthesis of new porous structures. They also offer technological advances when seeking to manipulate pore properties such as the pore size, size distribution, shape and surface functionality. Membranes and porous materials, which are receiving considerable attention at present, separate molecular mixtures based on the difference in the permeability of molecules through the incorporated pores, and these are expected to provide energy-efficient processes in industry. Such methods have been successfully applied in industrial areas such as hydrogen recovery (from H₂/N₂, H₂/CO₂, and H₂/CH₄), air separation (N₂/O₂), and CO₂ separation (from CH₄, CO, N₂ or hydrocarbons).^1–5 Microporous materials, such as zeolites and metal organic frameworks, also show notable performance levels when used in gas separation applications. The separation process is induced by the difference in the interaction between the gases with the functionality incorporated in the materials. Recently, Long et al. reported the successful separation of olefin-paraffin mixtures having similar molecular sizes and volatility levels using an MOF with open iron(II) coordination sites,⁶ as well as hexane isomers using an MOF with triangular channels.⁷ Regarding the separation of gas mixtures, the introduction of functionalities, such as open metal sites,^8,9 a functional group attached to a linker^10,11 and a target-designed cavity known as a ‘single molecular trap’^12,13 in porous material, leads to high selectivity for CO₂.

Despite the bright advances and good results in gas separation using porous materials, realizing the separation of gas mixtures having similar physical and transport properties is not an easy task. Their low separation efficiency remains a difficult and urgent issue.

We note that there are two possible routes to overcome the current limitations and achieve a highly selective molecular capturing process. The first route would be to design and synthesize new porous materials via chemical modification, as considerably demonstrated by a number of porous materials. The second route would be to use the physical nature of diverse gas storage compounds, favorably offering facile scale-up capabilities and good cost-effectiveness. We tried dramatically enhancing the affinity between target molecules and a surface or confined space via the second route to facilitate highly selective gas separation.

Here, we introduce a simple physical treatment of nanoporous materials composed of both hydrophilic and hydrophobic pores to invoke a highly selective gas capture and rejection mechanism. The insertion of hydrophilic molecules causes the hydrophilic pores to be occupied, while the hydrophobic pores remain empty and store hydrophobic molecules, and vice versa. This is spontaneously achieved by instigating contact between an inserted molecule and appropriate porous materials. The inserted molecules change the pore structure and properties; thus, tuning the affinity for specific gases within the pore networks.

Experimental section

Cu₃BTC₂ was synthesized by slightly modifying a previously reported solvothermal method.¹⁴ Copper nitrate trihydrate (4.35 g, Sigma Aldrich) and trimesic acid (2.1 g, Sigma Aldrich) were dissolved in a solvent that consisted of ethanol (60 ml, Junsei Chemical Co.) and deionized water (60 ml). The as-prepared solution was mixed for 12 hours and sonicated for 30 min then placed in a stainless reactor and allowed to react at 120 °C for 12 hours. The as-synthesized crystalline solid was washed with ethanol 3 times a day for 2 days then dried at room temperature for 12 hours and at 180 °C for 12 hours. To prepare HPB-Cu₃BTC₂, dehydrated Cu₃BTC₂ was placed in a vacuum oven with a beaker containing water and evacuated to 4 torr, where the relative humidity was 100%. HPB-Cu₃BTC₂ (1.5 g) was loaded into a high-pressure cell (6 ml), pressurized at room temperature and then kept at −30 °C for 1 day. To measure its storage capacity, the pressurized cell was vented at −30 °C and kept at room temperature to dissociate the stored gas. The volume of the emitted gas was measured by the water displacement method described in Fig. S1.† The CH₄/N₂ gas, purchased from Special gas, Korea, was pressurized at −30 °C for separation. The dissociated gas was injected into a GC equipped with a TCD sensor, as described in Fig. S2.† LT-XRD, HRPD, Raman and NMR analyses were performed after precooling the reactor with liquid nitrogen. The samples were finely ground to particles smaller than 200 μm at 77 K, and placed in a pre-cooled sample loader. Diffraction was performed as follows: (1) low Temperature XRD (D/MAX, Rigaku) with CuKα radiation (λ = 1.5406 Å) at a generator voltage of 40 kV and generator current of 300 mA, and (2) HRPD using the synchrotron at the Pohang Accelerator Laboratory (λ = 1.54950 Å). A Raman analysis was performed using a high-resolution dispersive Raman microscope (Horiba Jobin Yvon LabRAM HR UV/Vis/NIR) with 514.53 nm light emitted from an Ar-ion laser (30 mW) at −180 °C with following measurement conditions: 1800 grating, D1 filter and 1000 hole. All the Raman measurements were performed with Cu₃BTC₂ single crystal, as shown in Fig. S3.† The sample pressurized with ¹³C CH₄ was analyzed with a 400 MHz ¹³C CP-MAS Solid NMR (BRUKER) at 4 K spinning rate and at −40 °C.

Results and discussions

To demonstrate our concept and its effectiveness and simplicity, we carried out the separation of CH₄/N₂ mixtures. To date, CH₄/N₂ separation has been mainly done by the means of cryogenic distillation, which requires considerable amounts of energy to cool CH₄. To overcome the technical and economic barriers of energy and cost aspects, versatile separation technologies based on absorption, adsorption, and membranes were extensively adopted;³ however, the poor specific gas selectivity associated with these methods sets strict limits on their practical use. Under these circumstances, we adopted our approach to the use of Cu₃BTC₂ (BTC: 1,3,5-benzentricarboxylate) metal organic framework (MOF), in which the ‘hydrophilic pores are blocked’ (HPB) with water molecules and the hydrophobic pores are empty. This material was selected because it is one of the most characterized MOFs, which showed maintenance of porous structure in our working condition, and it contains both hydrophilic and hydrophobic pores together. In addition, it is already commercially available.

Cu₃BTC₂ contains three types of pores 5, 10.6 and 12.4 Å of pore sizes, respectively.¹⁵ The smallest pore is surrounded by four benzene rings and exhibits hydrophobicity. The other two pores are hydrophilic and can be filled with water molecules. The basic structure and pore properties of Cu₃BTC₂ were first reported by Williams et al.¹⁶ For our work, the MOFs were synthesized and well characterized by following a previously reported procedure.¹⁴ XRD and BET analyses revealed high crystallinity and surface area (1091 m² g⁻¹), which guarantees the successful synthesis (see ESI, Fig. S4† and pore analysis result). The synthesized Cu₃BTC₂ was saturated with water by vaporization, by which Cu₃BTC₂ containing 36 wt% of water was prepared and used during the entire experiment (Fig. S5†). The complete water-filling of the hydrophilic Cu₃BTC₂ pores, blocks gas inclusion and physically causes a significant reduction in active storage sites for gas molecules in hydrophilic pores and makes the remaining pores to be available for the inclusion of the target molecules.

From here, we report the results from our attempt to demonstrate the potential of HPB for separating CH₄ from a gaseous CH₄/N₂ mixture with remarkably high selectivity. Fig. 1A represents the amount of captured CH₄ and N₂ per 1 g of hydrophilic pore-blocked Cu₃BTC₂ (HPB-Cu₃BTC₂), according to system pressure. The amount of CH₄ stored in the MOF matrix rapidly increased to 50 bar; however, at higher pressures, the rate of increase slowed and reached 1.5 mmol g⁻¹ at 100 bar. In contrast, the stored N₂ showed a slight tendency to increase with 0.13 mmol g⁻¹ at 100 bar. Consider that the CH₄ stored in dried Cu₃BTC₂ is reported to be 11.6 mmol g⁻¹.¹⁷ Of course, the water-filling of the hydrophilic pores leads to considerable reduction of total stored CH₄ as the water-filling completely blocks the occupation by gaseous molecules in large pores. The overall scheme of competitive CH₄ selection versus N₂ rejection is depicted in Fig. 1B. We analyzed the composition of the gas captured in HPB-Cu₃BTC₂ after direct exposure to an equimolar mixture of CH₄/N₂. Fig. 1C shows the separation performance of HPB-Cu₃BTC₂ with applied pressure at −30 °C (see Table S1† for specific data). Due to the separation, CH₄ was enriched up to 95.8 mol% from a 50 mol% feed stream with the equilibrium selectivity of 24.7 at 80 bar, calculated from (Y_CH4/Y_N2) × (X_N2/X_CH4), where X and Y indicate the mol fractions of each gas in the gaseous and material phase at equilibrium, respectively. In addition, the CH₄/N₂ (90/10) feed stream can be enriched to more than 99% CH₄. The highest equilibrium selectivity of HPB-Cu₃BTC₂ is considerably higher than both experimental (about 3, Fig. S6†) and simulation results (about 4) of Cu₃BTC₂ itself.¹⁸ In addition, we would like to point out that the storage and separation test was performed at −30 °C to prevent the emission of stored gas from the solid phase and perform precise analysis. It is noteworthy to perform analysis at a wide temperature and pressure range in the future.

Fig. 1 Gas storage and separation of HPB-Cu₃BTC₂. (A) The stored amount of CH₄ (black) and N₂ (red) in relation to pressure. The line is fitted from experimental data. Each point is average value after 3 times measurement and the error bars indicate standard deviation within ±0.03. (B) Schematic description of gas separation mechanism in HPB-Cu₃BTC₂. (C) Composition of stored gases after exposure to equimolar (left) and 90/10 (right) mixtures of CH₄/N₂ at various pressures.

To elucidate the origin of CH₄ storage and the high selectivity of this method, we first conducted structural analysis by high-resolution powder diffraction with synchrotron radiation at the Pohang Accelerator Laboratory. The diffraction results for HPB-Cu₃BTC₂ before and after CH₄ exposure at 100 bar are shown in Fig. 2. A few new peaks were detected and merged at 27.2° and 28.3° corresponding to the (320) and (321) diffractions of crystalline structure I CH₄ hydrate. Only a small fraction of the water in mesopores and on the Cu₃BTC₂ surface is likely to combine with CH₄ and form stable methane hydrate. The CH₄ inclusions in Cu₃BTC₂ caused lattice expansion, but there was no notable change in peak intensities. The lattice parameter of Cu₃BTC₂ calculated from the (222) peak of highest intensity, increased from 26.3314 Å to 26.3651 Å, providing more stable confined spaces for capturing guest molecules.

Fig. 2 High resolution powder diffraction (HRPD) of HPB-Cu₃BTC₂ (red) and CH₄-included HPB-Cu₃BTC₂ (black) at 100 bar. The subset is the enlarged view of the circled region. Two new peaks are indexed to CH₄ hydrate.

The HPB-Cu₃BTC₂ samples pressurized at 100 bar of CH₄ were analyzed using Raman and ¹³C CP-MAS Solid NMR, and the results are shown in Fig. 3A and B. For these samples, the Raman spectra were collected at −180 °C and ambient pressure. Three peaks corresponding to the C–H symmetry stretching vibrational mode of CH₄ were observed at 2896.8, 2902.8 and 2913.8 cm⁻¹, which correspond to the shifts by −20.2, −14.2 and −3.2 cm⁻¹ from free CH₄ molecules. The last two peaks belong to the CH₄ that was captured in structure I CH₄ hydrate. These peaks exhibited wavenumber deviations of −1 to about 2 from that of bulk gas hydrate. These deviations resulted from the effect of Cu₃BTC₂ on neighbouring sI CH₄ hydrate. Excluding the possible influence of the appearance of a new hydrate phase, the highest peak was confirmed to result from CH₄ inclusion in the HPB-Cu₃BTC₂ framework. The earlier Raman analysis of CH₄ adsorption onto various MOFs indicated that the higher shift is caused by the stronger interaction of CH₄ with the MOFs. The shift of −20.2 from the HPB-Cu₃BTC₂ experiments is considerably higher than 7.6–10.1 cm⁻¹ of IRMOFs.¹⁹ The NMR spectrum in Fig. 3B shows four peaks at −4.3, −6.7, −11.2 and −12.7 ppm, where a chemical shift at −11.2 represents gaseous CH₄ that was weakly dissociated during the measurements. That CH₄ itself might experience three distinctive inclusion states, which was also revealed by the Raman spectra.

Fig. 3 Spectroscopic observations of CH₄ inclusion phenomena. (A) Raman and (B) ¹³C NMR spectra of CH₄ captured in HPB-Cu₃BTC₂.

The de-shielded small peaks at −4.3 and −6.7 ppm were assigned to CH₄ in small and large cages of sI CH₄ hydrate, respectively.²⁰ The carbon nucleus of the CH₄ molecule included in the HPB-Cu₃BTC₂ resulted in the up-field shift of −11.2 to −12.7 ppm. It is known that the interaction between C–H and the π-electrons of the benzene ring induces a strong shielding effect for the carbon of CH₄.²¹ The CH₄ actively participates in the formation of two coexisting phases: hydrate and Cu₃BTC₂, among which the amount of CH₄ in Cu₃BTC₂ is considerably greater than that of hydrate. This was well confirmed by the Raman spectra. At this stage, the question arises as to how CH₄ molecules occupy small empty hydrophobic pores in HPB-Cu₃BTC₂. The directly vaporized water can readily access the large hydrophilic pores of two distinctive sizes (Fig. 4A and B), and this would not allow capture of both CH₄ and N₂. In contrast, the small pores surrounded by the four benzene rings (strongly hydrophobic) remain vacant because they strongly repel water; thus, forming stable niches where CH₄ molecules can be stored.²² Because there was no noticeable change in the HRPD peak patterns, we can speculate that the inclusion of CH₄ does not disturb the ordering of the water confined in the pore, nor affect close interaction between the guest methane and pore water. In addition, the sharp NMR peak also confirms the inclusion of CH₄ in small hydrophobic pores, as shown in Fig. 4C. The geometrical aspect of the hydrophobic pores causes the methane molecules to be securely positioned. This is because the benzene rings of the hydrophobic pore form a tetrahedral arrangement with respect to the center of the pore. We conjecture that the interaction of the four (C–H)s with the surrounding four π electrons causes the higher Raman shift and nuclear shielding observed in the NMR.

Fig. 4 Structural changes of Cu₃BTC₂. (A) and (B) Two distinctive hydrophilic large pores before and after water saturation. (C) Hydrophobic small pore after CH₄ inclusion. Yellow and blue spheres represent CH₄ molecules and water clusters, respectively. (D) Lattice structure of both CH₄ and water included Cu₃BTC₂. (E) Raman spectra of water in hexagonal ice phase (blue) and in HPB-Cu₃BTC₂ (black).

For more direct verification, we prepared two Cu₃BTC₂ samples saturated with tetrahydrofuran (THF) and water. Here, THF was specially chosen because it has a polar functional group that is known to form structure II hydrate.²³ THF is not able to enter the small pores, but it was acceptable that it efficiently fitted into large pores with water. Cu₃BTC₂ was immersed in solutions of 1 and 10 mol% THF, followed by a filtering process to remove excess solution from the outside of the MOF crystal surface. Then, it was exposed to 100 bar of CH₄. The as-prepared samples were analyzed by Raman spectroscopy, and the results are given in Fig. S7.† THF inclusion slightly shifts the CH₄ wavenumber to a higher frequency, but there was no change in the peak intensity. This confirmed that the storage of CH₄ molecules in the hydrophobic pores was not affected by the neighbouring THF molecules in hydrophilic pores.

In addition to the stronger interaction of hydrophobic pore with CH₄ over N₂, the water confined in Cu₃BTC₂ also contributes to higher methane selectivity by inducing a more favourable environment for the diffusion of CH₄ over N₂. It is known that the confining environment provides favorable condition for the formation of hydrogen bonded water clusters, whose structure is like solid ice, and it can be stable even at room temperature.^24–28 This type of hydrogen bonded water structure can be formed in the hydrophilic pore of Cu₃BTC₂. To verify how the confined water molecules exist in the pores of Cu₃BTC₂, we performed a Raman analysis on the OH-stretching vibration of water, which provides information about its dynamics. In general, liquid water shows a broad O–H stretching peak from 3100–3600 cm⁻¹ due to the overtone of four different vibration modes (I to IV).²⁹ As water transforms to a solid phase at low temperature or becomes confined in nano-sized space, a sharp peak corresponding to the mode V appears (∼3100 cm⁻¹), which indicates the formation of tetrahedral hydrogen bonding between the molecules.^30–33 Therefore, the Mode-V peak is an indicator of the solid water phase. In Fig. 4E, the water confined in Cu₃BTC₂ shows a high sharp V-peak at room temperature, which confirms that the water is fixed and ordered by hydrogen bonding even at room temperature. This strongly blocks the hydrophilic pores of MOFs. To cross-check this interpretation, we used differential scanning calorimetry (DSC) for HPB-Cu₃BTC₂ (Fig. S8†). During a cooling and heating cycle from 20 to −40 °C, the first-order structure transition of water to solid ice was not observed.

In the situation that the pore is filled with water, two pathways can be expected for the gas diffusion in to the material. One is gas diffusion through confined water and the other one is through the interface between the hydrophobic surface of MOFs and confined water. Considering that the diffusion coefficient of gas in the tetrahedrally ordered solid water phase is extremely smaller than liquid water, e.g., the diffusion coefficient of CH₄ in hexagonal ice is 5 orders of magnitude less than that of liquid water,^34,35 the gas diffusion through confined water is not significant. Furthermore, CH₄ and N₂ gas do not strongly interact with water. Therefore, the interaction between the hydrophobic surface and gas molecule is an important factor for the storage and separation. As previous reports revealed, polarizability effects the adsorption and separation of gas; the gas with higher polarizability can have stronger interaction with the hydrophobic surface, vice versa.^36–39 For separating CH₄/N₂, the polarizability of CH₄ and N₂ is 2.448 ų and 1.710 ų, respectively.³ CH₄ with higher polarizability have higher chance to interact with the surface of MOFs and diffuse in the HPB-Cu₃BTC₂, as shown in Fig. 5. The specific description about the pore structure of water-filled hydrophilic pore can be found elsewhere: small free voids exist between the hydrophobic wall and water cluster due to a bridging effect.¹⁴

Fig. 5 Schematic description of the diffusion of N₂ (left) and CH₄ (right) in the water-filled pore of Cu₃BTC₂, respectively.

To verify our mechanism for higher CH₄ selectivity than N₂, we observed the change in the CH₄/N₂ selectivity of Cu₃BTC₂ in terms of water content in MOF, as shown in Fig. S9.† With the presence of water, CH₄/N₂ selectivity slightly increased; however, significant high selectivity was achieved at sufficiently high water content to block the hydrophilic pore of Cu₃BTC₂. This result confirms that our mechanism for higher CH₄ selectivity over N₂ is reasonable.

In addition, we measured the storage and dissociation kinetics, as shown in Fig. 6. The initial pressure of 81 bar in the reactor containing HPB-Cu₃BTC₂ rapidly decreased to 77.4 bar within 10 min with high CH₄ absorption, and then slowly decreased to 76.5 bar.

Fig. 6 (A) Adsorption and rate of CH₄ in HPB-Cu₃BTC₂ at −30 °C. (B) Raman spectra of CH₄-included HPB-Cu₃BTC₂ at ambient pressure with increasing temperature. (C) Dissociation rate of stored CH₄.

The expulsion of CH₄ from the cage, according to temperature change, could be observed using the XRD and Raman spectroscopy. The Raman peaks of CH₄ rapidly disappeared between −40 °C and −20 °C due to its dissociation, as shown in Fig. 6B, and XRD showed a dramatic structural change between −60 and −40 °C due to the emission of gas and water from pores, as shown in Fig. S10.† A slight deviation of dissociation temperatures can likely be attributed to the difference in pressure during the Raman spectroscopy (1 bar) and XRD (vacuum). We observed that 80% of the stored CH₄ was rapidly dissociated within 5 min, but that additional heating was needed for the complete recovery (Fig. 6C).

To check the stability of Cu₃BTC₂ in our working condition, we observed separation capability during 10 times of repeating test over 5 days, which showed the maintenance of separation performance (Fig. S11†). Subsequently, we recovered HPB-Cu₃BTC₂ and performed XRD measurement, which showed sharp diffraction peaks (Fig. S12†) and confirmed the maintenance of structure during gas separation. Moreover, the mass of HPB-Cu₃BTC₂ after exposure to 10 times of dry CH₄/N₂ mixture is not changed, which means the water confined in Cu₃BTC₂ is not removed in our working condition, even after 10 times of contact with dry gas.

Considering the features of HPB-Cu₃BTC₂, such as (1) strong dependency of CH₄ storage capacity on pressure, (2) small storage capacity of N₂, and (3) recovery of stored gas and regeneration of material by just depressurization; this material could be used for pressure swing adsorption. Even though additional heating is required for the complete recovery of adsorbed CH₄, the energy consumed for additional temperature swing could reduce the efficiency of this technology.

Despite of beneficial features of this method, additional research should follow in the near future for its actual use in the natural gas fields, e.g., sensitivities toward other gases, such as H₂S and CO₂, existing in natural gases.

Conclusions

The present outcomes clearly demonstrate that the extraordinary selectivity for CH₄ over N₂ is achieved by the simple water saturation of Cu₃BTC₂. Furthermore, the methane gas can readily be recovered by pressurizing and depressurizing cycles (e.g., common pressure swing adsorption), leading to applications for natural gas processing, methane production from landfill gas and N₂ removal. Moreover, the ability to switch from hydrophilic to hydrophobic pores by the selection of the most suitable guest filler is expected to facilitate the extension of HPB to include the chemical fields of zeolites, other MOFs and silica materials. Moreover, the formation of methane hydrate in the mesopore of Cu₃BTC₂ suggests the possible approach to combine MOF and gas hydrate technologies to develop enhanced gas storage and separation methods. These new options should lead to further opportunities for the application of porous materials on gas separation.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF Grant no. 2013-059738) funded by the Korea government (MSIP) and by the WCU program (R-31-10055-0). We gratefully acknowledge the Pohang Accelerator Laboratory (Beamline 9B-HRPD) and the Korea Basic Science Institute for allowing us to use solid-state NMR.

References

1. J. D. Figueroa, T. Fout, S. Plasynski, H. McIlvried and R. D. Srivastava, Int. J. Greenhouse Gas Control, 2008, 2, 9–20.
2. D. M. D'Alessandro, B. Smit and J. R. Long, Angew. Chem., Int. Ed., 2010, 49, 6058–6082.
3. T. E. Rufford, S. Smart, G. C. Y. Watson, B. F. Graham, J. Boxall, J. C. D. da Costa and E. F. May, J. Pet. Sci. Eng., 2012, 94–95, 123–154.
4. H. W. Kim, H. W. Yoon, S. M. Yoon, B. M. Yoo, B. K. Ahn, Y. H. Cho, H. J. Shin, H. Yang, U. Paik, S. Kwon, J. Y. Choi and H. B. Park, Science, 2013, 342, 91–95.
5. H. Li, Z. N. Song, X. J. Zhang, Y. Huang, S. G. Li, Y. T. Mao, H. J. Ploehn, Y. Bao and M. Yu, Science, 2013, 342, 95–98.
6. E. D. Bloch, W. L. Queen, R. Krishna, J. M. Zadrozny, C. M. Brown and J. R. Long, Science, 2012, 335, 1606–1610.
7. Z. R. Herm, B. M. Wiers, J. A. Mason, J. M. van Baten, M. R. Hudson, P. Zajdel, C. M. Brown, N. Masciocchi, R. Krishna and J. R. Long, Science, 2013, 340, 960–964.
8. D. Britt, H. Furukawa, B. Wang, T. G. Glover and O. M. Yaghi, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 20637–20640.
9. A. Mallick, S. Saha, P. Pachfule, S. Roy and R. Banerjee, J. Mater. Chem., 2010, 20, 9073–9080.
10. S. Couck, J. F. M. Denayer, G. V. Baron, T. Remy, J. Gascon and F. Kapteijn, J. Am. Chem. Soc., 2009, 131, 6326–6327.
11. A. Torrisi, R. G. Bell and C. Mellot-Draznieks, Cryst. Growth Des., 2010, 10, 2839–2841.
12. M. Wriedt, J. P. Sculley, A. A. Yakovenko, Y. G. Ma, G. J. Halder, P. B. Balbuena and H. C. Zhou, Angew. Chem., Int. Ed., 2012, 51, 9804–9808.
13. J. R. Li, J. M. Yu, W. G. Lu, L. B. Sun, J. Sculley, P. B. Balbuena and H. C. Zhou, Nat. Commun., 2013, 4, 1538.
14. P. Kusgens, M. Rose, I. Senkovska, H. Frode, A. Henschel, S. Siegle and S. Kaskel, Microporous Mesoporous Mater., 2009, 120, 325–330.
15. P. J. Ryan, O. K. Farha, L. J. Broadbelt, R. Q. Snurr and Y.-S. Bae20120073438 A1, 2012.
16. S. S.-Y. Chui, S. M.-F. Lo, J. P. H. Charmant, A. G. Orpen and I. D. Williams, Science, 1999, 283, 1148–1150.
17. I. Senkovska and S. Kaskel, Microporous Mesoporous Mater., 2008, 112, 108–115.
18. B. Liu and B. Smit, Langmuir, 2009, 25, 5918–5926.
19. D. Y. Siberio-Pérez, A. G. Wong-Foy, O. M. Yaghi and A. J. Matzger, Chem. Mater., 2007, 19, 3681–3685.
20. E. D. Sloan and C. A. Koh, in Clathrate hydrates of natural gases, CRC Press, USA, 2008, p. 353.
21. J. A. Platts and K. Gkionis, Phys. Chem. Chem. Phys., 2009, 11, 10331–10339.
22. H. Wu, J. M. Simmons, Y. Liu, C. M. Brown, X. S. Wang, S. Ma, V. K. Peterson, P. D. Southon, C. J. Kepert, H. C. Zhou, T. Yildirim and W. Zhou, Chem.–Eur. J., 2010, 16, 5205–5214.
23. L. J. Florusse, C. J. Peters, J. Schoonman, K. C. Hester, C. A. Koh, S. F. Dec, K. N. Marsh and E. D. Sloan, Science, 2004, 306, 469–471.
24. M. L. Wei, C. He, W. J. Hua, C. Y. Duan, S. H. Li and Q. J. Meng, J. Am. Chem. Soc., 2006, 128, 13318–13319.
25. Y. Wang, T. A. Okamura, W. Y. Sun and N. Ueyama, Cryst. Growth Des., 2008, 8, 802–804.
26. M. L. Cao, J. J. Wu, H. J. Mo and B. H. Ye, J. Am. Chem. Soc., 2009, 131, 3458–3459.
27. C. M. Jin, Z. Zhu, Z. F. Chen, Y. J. Hu and X. G. Meng, Cryst. Growth Des., 2010, 10, 2054–2056.
28. Z. Yang, S. G. Hua, W. J. Hua and S. H. Li, J. Phys. Chem. B, 2011, 115, 8249–8256.
29. Q. Sun, Vib. Spectrosc., 2009, 51, 213–217.
30. C. G. Venkatesh, S. A. Rice and J. B. Bates, J. Chem. Phys., 1975, 63, 1065–1071.
31. T. C. Sivakumar, S. A. Rice and M. G. Sceats, J. Chem. Phys., 1978, 69, 3468–3476.
32. F. Mallamace, M. Broccio, C. Corsaro, A. Faraone, D. Majolino, V. Venuti, L. Liu, C. Y. Mou and S. H. Chen, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 424–428.
33. M. Erko, G. H. Findenegg, N. Cade, A. G. Michette and O. Paris, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 84, 104205.
34. A. Hori and T. Hondoh, Can. J. Phys., 2003, 81, 251–259.
35. M. Takahata, Y. Kashiwaya and K. Ishii, Mater. Trans., 2010, 51, 727–734.
36. S. L. Suib, Science, 2003, 302, 1335–1336.
37. S. T. Meek, S. L. Teich-McGoldrick, J. J. Perry, J. A. Greathouse and M. D. Allendorf, J. Phys. Chem. C, 2012, 116, 19765–19772.
38. J. F. Yang, J. M. Li, W. Wang, L. B. Li and J. P. Li, Ind. Eng. Chem. Res., 2013, 52, 17856–17864.
39. P. S. Yaremov, N. D. Scherban and V. G. Ilyin, Theor. Exp. Chem., 2013, 48, 394–400.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra12967a

---