Revealing the potential application of chiral covalent organic frameworks in CO₂ adsorption and separation

Abstract

The adsorption and separation abilities of gases (including CO₂, CH₄, N₂, and H₂) in a series of chiral COFs (CCOFs) were studied using the Grand canonical Monte Carlo (GCMC) method, periodic density functional theory (DFT) (PBE functional) calculations and accurate dispersion corrected double-hybrid DHDF-D3 to explore the potential application of recently synthesized chiral COFs (Han et al., J. Am. Chem. Soc., 2018, 140, 892–895 and Han et al., J. Am. Chem. Soc., 2017, 139, 8693–8697). In contrast to four classical structures (including IRMOF-1, UMCM-1, COF-5, and ZIF-8), CCOF6 shows a better CO₂ adsorption capacity at 298 K in a wide range of pressure from 0 to 100 kPa. More importantly, the selectivity of CO₂ over N₂, CH₄ and H₂ in CCOF6 is obviously greater than that of other classic porous materials. GCMC simulation demonstrates that the CO₂ molecules prefer the smaller B1 channel in CCOF6 at pressure lower than 10 kPa; however, there is no obvious preferential adsorption site in CCOF5. To better understand the mechanism of CO₂ adsorption in CCOFs, the interaction energy between CO₂ and CCOFs was calculated by the double-hybrid DFT-D3, confirming the preferred adsorption sites in CCOF6 due to the proper narrow channel and the charge transfer from CO₂ to CCOF6. Our well established theoretical calculation reveals that CCOF6, with promising CO₂ adsorption and selectivity properties, has the potential to be an excellent material for CO₂ adsorption and separation.

---

Introduction

Covalent organic frameworks (COFs)^1,2 are novel porous crystalline materials^3–6 composed of light elements such as carbon, hydrogen, oxygen, nitrogen and boron linked by covalent bonds. They have a great potential for application in the fields of catalysis,^7,8 adsorption,^9–11 separation,^12,13etc.^14–16 owing to their high stability, low density, large surface area, and tunable pore sizes. Previous studies have reported that COFs exhibit outstanding abilities in terms of gas adsorption and separation. Furukawa and Yaghi found that COFs showed excellent adsorption capacity for carbon dioxide, hydrogen, and methane among the most common carbon materials.¹⁷ A perfluorinated covalent triazine-based framework (FTCF-1) demonstrated not only exceptionally high CO₂ storage capacity but also ideal separation of CO₂ from N₂ reported by Zhao et al..¹⁸ Baldwin et al. discovered that 3D DBA-COFs expressed high uptake for ethane and ethylene gas.¹⁹

Recently, Han et al.^20,21 synthetized a new series of novel chiral COFs (CCOFs) which feature an enantioselective process. Furthermore, the CCOFs show permanent porosity and high stability in water and most organic solvents, which draw our intensive attention, and spark our interest in exploring their adsorption and separation properties. It is vitally important to search for the potential application of already synthesized, or even the hypothetical materials, with the aim to identify promising materials and reveal the structure–property relations with the aid of the molecular modeling method.²² Our group performed several interesting modeling studies on novel porous materials. For example, Pang et al. displayed that thiophene-based CPTs were suitable for the application of H₂ storage and its purification.²³ Zhou et al. reported that CAU-17 and functionalized CAU-17 can selectively adsorb CH₄ over H₂.²⁴ Guan et al. established a screening method to analyze the content of activated pores in COFs and this approach can be expanded to be automated for high-throughput screening of a lot of porous crystalline materials, especially for predicting their adsorption potential.²⁵ Zhang et al. considered that spiroborate-linked ICOFs were likely five-fold interpenetrating models and ICOFs are promising CH₄ storage materials.²⁶

In this work, we explored the CO₂, CH₄, N₂, and H₂ adsorption capacity of CCOFs and investigated the CH₄/H₂, CH₄/N₂, CO₂/N₂, CO₂/CH₄ and CO₂/H₂ separating performance in CCOF4, CCOF5, and CCOF6 by the combination of GCMC simulations and DFT calculations. Due to the appropriate narrow channel and the charge transfer between CO₂ and CCOF6, CCOF6 shows the preferred CO₂ adsorption sites at low pressure. We find that CCOF6 is a good candidate for CO₂ adsorption and separation by comparing it with other CCOFs and 4 classic porous materials.

Models and computational methods

CCOF structures

The models of a series of CCOFs were taken from the reference^20,21 and are shown in Fig. 1. The detailed structural parameters of the unit cells in CCOFs are listed in Table S1 (ESI†). Fig. 1a, c, and e display the unit cell of CCOF4, CCOF5, and CCOF6, respectively. And Fig. 1b, d and f display the supercell of CCOF4, CCOF5, and CCOF6 in the form of 2 × 2 × 3, 2 × 2 × 2, and 2 × 2 × 2 unit cell, respectively, as a modeling box. CCOF4 shows a 1D channel, and then CCOF5 and CCOF6 show 3D channels, which are represented in the green surface. CCOF5 and CCOF6 have the same topology of the framework. The A1 channel in CCOF5 is surrounded by benzene rings, while the B1 channel in CCOF6 contains ketones with the exposed oxygen. The structures of the typical porous materials such as IRMOF-1,²⁷ UMCM-1,²⁸ COF-5,¹ and ZIF-8²⁹ were taken from the literature for comparison.

Fig. 1 The structures of (a and b) CCOF4, (c and d) CCOF5, and (e and f) CCOF6, and H atoms were omitted for clarity.

Potential models

To describe the adsorption properties of various gas molecules in the concerned porous materials, the following models were adopted. CO₂ was considered as a three-site rigid linear molecule, and its quadrupole moment was described by a partial charge model.³⁰ For CH₄, N₂, and H₂ molecules, we used a spherical united-atom model.^31–33 The above potential models have been employed well in previous work.³⁴ Given that the Dreiding force field³⁵ could successfully provide accurate prediction of gas adsorption and separation,³⁶ it was applied to the potential parameters of framework atoms. All the LJ interaction parameters between adsorbates and adsorbents were calculated by Lorentz–Berthelot mixing rules, as shown in eqn (1) and (2):³⁷

ε_ij = (ε_iiε_ij)^1/2 (1)

σ_ij = (σ_iiσ_ij)/2 (2)

All the potential parameters with partial charges are listed in Table 1. As for the charge of the frameworks, we made use of the QEq³⁸ method based on Ewald summation.

Table 1 Lennard-Jones potential parameters for adsorbates and adsorbents

Species	Site	ε/kB (K)	σ (Å)	q (e)
CH4	CH4	148.000	3.730
H2	H2	36.700	2.958
CO2	C	27.000	2.800	+0.70
	O	79.000	3.050	−0.35
N2	N2	94.950	3.549
	C	47.900	3.470
	H	7.660	2.850
Adsorbents	O	48.158	3.033
	N	39.007	3.263
	B	47.806	3.580
	Zn	27.718	4.450

---

GCMC simulations

All GCMC simulations were calculated by using the MUSIC code.³⁹ The cutoff radius was set to 12.0 Å, and the porous frameworks were expended into the supercell greater than 24.0 Å along each dimension. There were 4 × 10⁷ trial moves during each simulation, including the first 2 × 10⁷ steps for equilibrium and the rest 2 × 10⁷ steps for data collection. The experimental data usually provide excess uptake N_ex, however GCMC simulations provide absolute uptake N_abs. Therefore, we should convert the absolute uptake to excess uptake for contrast as given in eqn (3)

N_ex = N_abs − ρ_gV_g (3)

where ρ_g is the density of bulk gas which is calculated from the PREOS,⁴⁰ and V_g is the effective volume of the adsorbent. V_g is obtained using the following eqn (4)⁴¹

V_g = RN_mT/Pm_m (4)

where T, R, and P represent the temperature, gas constant, and pressure, respectively. And N_m is the number of adsorbent probe molecules per molar mass m_m of the framework, which is obtained by GCMC simulation of non-adsorbed helium in the absorbent at low pressure and room temperature.⁴²

To reflect the intensity of force between gas molecules and the adsorbed phase, Q_st is expressed by eqn (5)

Q_st = RT − (∂U/∂N)_T,V (5)

where R is the universal gas constant, T is the temperature, N is the number of adsorbed particles and U is the energy of the sorbate in the adsorbed state which consists of adsorbate–adsorbate and adsorbate–adsorbent interactions.

The adsorption selectivity refers to the ability of an adsorbate to preferentially adsorb certain substances due to its pore topology and pore-surface character. The adsorption selectivity of compound i relative to compound j in a mixture can be calculated using eqn (6)

S_ads(i/j) = (x_i/x_j)/(y_i/y_j) (6)

where x_i and y_i are the mole fractions of compound i in the adsorbed phase and the bulk phase, respectively.

DFT calculations

It is critical to obtain the interaction energy (E_int) between gas molecules and frameworks using DFT calculations⁴³ so as to further study the adsorption mechanism. The interaction energy is calculated using eqn (7)

E_int = E_total − E_a − E_b (7)

where E_total is the total energy of the gas molecule and the framework, E_a is the energy of the gas molecule alone, and E_b is the energy of the framework alone.

The potential energy surface of CO₂ in the channel of CCOFs was scanned using periodic DFT at the PBE-D2^44,45/DNP⁴⁶ level by Dmol3 module of Materials Studio software, where the two oxygen atoms of carbon dioxide were relaxed. In addition, a high accuracy method at the B2PLYP-D3^47,48/def2-tZVPP⁴⁹ level was employed to evaluate the interaction and NBO charge⁵⁰ transfer between CO₂ and the efficiency fragment cut from the pore surface of the CCOFs using the Gaussian 09 package.⁵¹

Results and discussion

Verification of the force field

Fig. 2 shows the comparison of the predicted adsorption isotherms of N₂ in (a) CCOF4, (b) CCOF5, and (c) CCOF6 at 77 K with the experimental data.^16,17 The deviations for these N₂ adsorption isotherms in the three materials are acceptable as perfect crystals are employed in the simulation, while pore blocking, crystal defects, and impurity influence the isotherms in actual measurements.^21,47 As a result, the Dreiding force field is adopted in the GCMC simulations.

Fig. 2 Adsorption isotherms of N₂ in (a) CCOF4, (b) CCOF5, and (c) CCOF6 at 77 K. The black lines represent the experimental data,^16,17 and red lines represent simulation results.

Adsorption of pure gases

In order to evaluate the abilities of the CCOFs for pure gas adsorption (including N₂, H₂, CH₄, and CO₂), the comparison of adsorption isotherms with several classic porous materials (including COF-5, IRMOF-1, ZIF-8 and UMCM-1) is shown in Fig. 3. From Fig. 3a, we can determine that the N₂ loading of CCOF5 and CCOF6 is slightly higher than that in other structures. In terms of H₂ adsorption, the CCOF6 structure with the most H₂ loading (0.5 cc g⁻¹ at a pressure of 100 kPa) in the CCOFs is only ranked third in all the selected structures, as shown in Fig. 3b. Impressively, CCOF5 and CCOF6 show extremely high CH₄ and CO₂ uptakes when compared with other frameworks. The high gas uptake can be attributed to the strong interactions between the gas molecules and the CCOF6. And, it is very clear that the CO₂ adsorption capacity of CCOF6 is much higher than that in all structures. Moreover, the excess uptake of N₂, H₂, CH₄, and CO₂ in CCOF6 is in the order of H₂ (0.50 cc g⁻¹) < N₂ (6.68 cc g⁻¹) < CH₄ (30.42 cc g⁻¹) < CO₂ (116.36 cc g⁻¹) at 100 kPa. That is to say, the excess uptake of CO₂ is significantly greater than other pure gas uptake. As shown in Fig. 3d, the CO₂ uptake is larger than 110 cc g⁻¹, which is much higher than that of the other porous materials including CCOFs and classic materials. As a result, CCOF6 exhibits a large amount of CO₂ adsorption and is expected to be a promising CO₂ adsorbent.

Fig. 3 Comparison of adsorption isotherms of pure (a) N₂, (b) H₂, (c) CH₄, and (d) CO₂ in CCOFs with several typical frameworks at 298 K.

To better understand the pure gas adsorption in porous materials, we calculated the isosteric heats of various gases as shown in Fig. 4. It can be seen that the CO₂ isosteric heat in CCOF6 is the largest within the selected pressure range as shown in Fig. 4d. This proves that CCOF6 has strong ability to adsorb CO₂. Overall, it can be found that the isosteric heats of pure gases in CCOF6 follow the order of H₂ (6.88 kJ mol⁻¹) < N₂ (14.74 kJ mol⁻¹) < CH₄ (21.03 kJ mol⁻¹) < CO₂ (33.36 kJ mol⁻¹) at 100 kPa, which is consistent with the adsorption capacities of pure gases under the same conditions. And the result revealed that the interaction between CO₂ and the frameworks is stronger than that of other gases. Apparently, the biased affinity of CCOF6 for different gases demonstrates that CCOF6 may be a promising candidate for the CO₂ separation.

Fig. 4 Isosteric heats of (a) N₂, (b) H₂, (c) CH₄, and (d) CO₂ in CCOFs and some typical frameworks at 298 K.

Separation of binary gases

Fig. 5 shows the adsorption selectivity of the five mixtures in the selected classic frameworks and CCOFs at 298 K. To meet the practical applications of all kinds of gases, the different mole ratios of CH₄/H₂, CH₄/N₂, CO₂/N₂, CO₂/CH₄, and CO₂/H₂ selected are 50 : 50, 50 : 50, 15 : 85, 15 : 85, and 20 : 80, respectively. As shown in Fig. 5a and b, the selectivity of CH₄/H₂ and CH₄/N₂ in CCOF5 and CCOF6 is slightly higher than that of others and these values remain nearly constant with increasing pressure. Furthermore, it can be obviously found that the selectivity of CO₂/N₂, CO₂/CH₄, and CO₂/H₂ in CCOF6 and CCOF5 is greater than that in other frameworks from Fig. 5c, d and e. In other words, CCOF5 and CCOF6 have a great ability for CO₂ separation over H₂, N₂, and CH₄. In particular, the selectivity of CO₂/N₂, CO₂/CH₄ and CO₂/H₂ in CCOF6 is much higher than that of other materials including CCOF5. Meanwhile, it is observed that the selectivity of CO₂/N₂ and CO₂/CH₄ decreases with the increasing pressure, and the selectivity of CO₂/H₂ initially decreases with increasing pressure and then reaches a platform after 70 kPa. This phenomenon for the decline at first can be explained by the following. At low pressure, CO₂ molecules preferentially occupied the CCOF6 channels. With the increase of pressure, there are fewer and fewer locations for adsorption of CO₂. Consequently, the selectivity decreases accordingly. Overall, the selectivity for binary gases in CCOF6 is in the order CH₄/N₂ (4.67) < CO₂/CH₄ (14.35) < CH₄/H₂ (44.18) < CO₂/N₂ (73.60) < CO₂/H₂ (536.26) at 100 kPa. In a word, these results indicate that CCOF6 could be a promising candidate for CO₂ separation over N₂, CH₄, and H₂.

Fig. 5 Adsorption selectivity for the binary mixtures at 298 K. (a) CH₄/H₂, (b) CH₄/N₂, (c) CO₂/N₂, (d) CO₂/CH₄ and (e) CO₂/H₂.

Preferential adsorption sites

In efforts to study the CO₂ adsorption mechanism, we provided the snapshots of CO₂ in CCOF5 and CCOF6 at 298 K at different pressures, as shown in Fig. 6. At a pressure of 1 kPa, the B1 channels adsorb more CO₂ molecules than B2 channels in CCOF6, whereas the amount of CO₂ molecules adsorbed by the A1 and A2 channels of CCOF5 is similar. And so, the B1 channel is the preferential adsorption site of CO₂ in CCOF6, while CCOF5 does not possess the obvious preferential adsorption site of CO₂. As the pressure increases, the amount of adsorption per channel becomes more and more. It can be seen from Fig. 6b that the adsorption amount of CO₂ is greatly increased from 1 kPa to 10 kPa, which is consistent with the results of the CO₂ adsorption isotherm of CCOF6 in Fig. 3d. In addition, H₂, N₂, and CH₄ are randomly distributed in each of the channels of CCOF5 and CCOF6 at low pressures from Fig. S1 and S2 (ESI†), so there is no obvious preferential adsorption site for H₂, N₂, and CH₄. It can be concluded that there is an excellent CO₂ separation of CCOF6 in binary gases.

Fig. 6 Snapshots of CO₂ in CCOF5 (a) and CCOF6 (b) at T = 298 K at different pressures.

Interaction between gases and frameworks

With the aim of deeply investigating the microscopy mechanism, periodic DFT calculations at the PBE-D2/DNP level were performed to describe the interaction between gases and frameworks. The profile of E_int of CO₂ in two planes with z = 0.5 and y = 0.5 of CCOF5 and CCOF6 is shown in Fig. 7. The darker red color is found, and the stronger interaction between the gases and CCOFs is shown in Fig. 7. Fig. 7a and c clearly show that the A1 and A2 channels of CCOF5 show a similar maximum interaction energy to CO₂ with the E_int in the range of −4.539 ∼−5.577 kcal mol⁻¹, while the maximum interaction energy between CO₂ molecules and the B1 channels (−9.893 kcal mol⁻¹) is higher than the maximum interaction energy between the CO₂ molecules and the B2 channels (−5.477 kcal mol⁻¹). In Fig. 7a and c, the maximum interaction of CO₂ in the narrow B1 channels of CCOF6 (−9.893 kcal mol⁻¹) is greater than the maximum interaction in the A1 channels of CCOF5 (−4.539 kcal mol⁻¹), and we can also find that the maximum interaction of CO₂ in the narrow B1 channels of CCOF6 (−9.584 kcal mol⁻¹) is larger than the maximum interaction of CO₂ in the A1 channels of CCOF5 (−5.438 kcal mol⁻¹) from Fig. 7b and d, which are in line with the low-pressure gas distribution from GCMC simulations in Fig. 6. The main difference between the B1 channel of CCOF6 and the A1 channel of CCOF5 is that the B1 channel of CCOF6 has no oxygen atoms, but the A1 channel of CCOF5 has. The bulky oxygen atoms in the B1 channel of CCOF6 can narrow the channel to fit CO₂ molecules, resulting in higher interaction energies between them. Moreover, the exposed oxygen atoms may provide induction force to the CO₂ molecules, leading to the enhancement of interaction between them. As a result, an accurate calculation and detailed discussion on the interaction between CO₂ and the key fragment of channels should be performed.

Fig. 7 The profiles of E_int of CO₂ in two planes with z = 0.5 and y = 0.5 of CCOF5 (a and b) and CCOF6 (c and d) in the display style of 2 × 2 × 1 (a and c) and 2 × 1 × 2 (b and d) on the basis of the unit cell. The points are centers of C atoms of CO₂ molecules and the distance of each C atom is 0.5 Å. The colour represents the framework–gas interaction energy. All H atoms are omitted for the sake of clarity.

The accurate method at the B2PLYP-D3/def2-TZVPP level was used to calculate the interaction energy between the key fragment of CCOFs and CO₂. The most stable ones are shown in Fig. 8, and the metastable structures and interaction energies are shown in Fig. S3 (ESI†). The distance of the CO₂ molecule to the segment of CCOF5 is somewhat longer than that to the segment of CCOF6, whilst the maximum interaction energy of CO₂ to the segment of CCOF5 (−5.265 kcal mol⁻¹) is smaller than that to the segment of CCOF6 (−6.161 kcal mol⁻¹). It can be concluded that CCOF6 and CO₂ have stronger interaction. Furthermore, the NBO charges of the skeletons interacting with the CO₂ molecule shown in Fig. 8 were calculated to explore the effect of charge on the interaction at the B2PLYP-D3/def2-TZVPP level. As listed in Table 2, the C–CO₂ charge transferred from free CO₂ to the one adsorbed in CCOF6 is 0.02103 e, which is greater than the corresponding one in CCOF5 (0.01516 e). And the charge reduction of two oxygen atoms of CO₂ from free CO₂ to the one adsorbed in CCOF6 (0.02300 e) is more than that of CCOF5 (0.01544 e). Meanwhile, the total charge transfer of CO₂ upon adsorption in CCOF6 (−0.00197 e) is more remarkable than that upon adsorption in CCOF5 (−0.00028 e). This illustrates that the large charge transfer of CO₂ in CCOF6 over CCOF5 is a reason for the greater interaction energy of CCOF6 with CO₂ than CCOF5. Very interestingly, the transferred charge of O17 in CCOF6 upon CO₂ adsorption is the same as the transferred charge of N3 in CCOF5 with the amount of 0.0203 e, which is the maximum transferred charge in the CCOF fragments. One should note that, the charge of the O17 atom in CCOF6 is larger than the charge of the N3 atom in CCOF5, which will result in a greater electrostatic interaction of the O17 atom to CO₂ in CCOF6 than the N3 atom to CO₂ in CCOF5. It is indicated that the strong electrostatic interaction between the additional O atom of CCOF6 and CO₂ is a cause for the increasing interaction energy. All in all, CCOF6 has a closer distance, more charge transfer, and larger interaction with CO₂, which endows CCOF6 with promising CO₂ adsorption and separation properties when compared with CCOF5.

Fig. 8 The conformations and interaction energies of the most stable dimers of the segment of CCOF5-CO₂ (a) and CCOF6-CO₂ (b) with the distances shown in angstroms.

Table 2 (a) Calculated partial NBO charges (e) in isolated CCOF5, CO₂, and CCOF5 interacting with CO₂. (b) Calculated partial NBO charges (e) in isolated CCOF6, CO₂, and CCOF6 interacting with CO₂

(a)
	CCOF5	CO2	CCOF5 + CO2
N3	−0.4452		−0.4655
C26		1.0908	1.1059
O27		−0.5454	−0.5496
O28		−0.5454	−0.5566

---

(b)
	CCOF6	CO2	CCOF6 + CO2
O17	−0.6284		−0.6487
C27		1.0908	1.1118
O28		−0.5454	−0.5455
O29		−0.5454	−0.5682

---

Conclusions

In order to explore the potential application of the newly synthesized CCOFs, we have studied the adsorption of pure gases such as CO₂, CH₄, N₂, and H₂ and the selectivity of CH₄/H₂, CH₄/N₂, CO₂/N₂, CO₂/CH₄ and CO₂/H₂ in the selected CCOFs by GCMC simulations and DFT calculations. GCMC simulations show that CCOF6 in the string of CCOFs has a high uptake amount of CO₂ and a good selectivity of CO₂/N₂, CO₂/CH₄ and CO₂/H₂. Moreover, the stronger affinity of CCOF6 with CO₂ also indicates that CCOF6 has a strong adsorption capacity and selectivity for CO₂. From the GCMC simulations of CO₂ in CCOF5 and CCOF6, the B2 channel of CCOF6 is the preferential adsorption site, and there is no obvious preferential adsorption site in CCOF5. Furthermore, accurate double-hybrid B2PLYP-D3 was applied to study the interaction energy between CO₂ and CCOFs. It is found that the maximum interaction energy and charge transfer between CO₂ and CCOF6 are larger than those of CCOF5, and CCOF6 does have a preferential site. In summary, it can be concluded that the CO₂ adsorption and selectivity of CCOF6 are beyond those of CCOF5 and several classic porous materials, making CCOF6 a potential candidate for CO₂ adsorption and separation.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This project was sponsored by the NSFC (21873018 and 21573036), the Science and Technology Development Planning of Jilin Province (20180520001JH), and the Fundamental Research Funds for the Central Universities (2412019FZ014).

Notes and references

1. A. P. Côté, A. I. Benin, N. W. Ockwig, M. Keeffe, A. J. Matzger and O. M. Yaghi, Science, 2005, 310, 1166.
2. H. M. El-Kaderi, J. R. Hunt, J. L. Mendoza-Cortés, A. P. Côté, R. E. Taylor, M. Keeffe and O. M. Yaghi, Science, 2007, 316, 268.
3. L. Zhou, Y. Zhou, M. Li, P. Chen and Y. Wang, Langmuir, 2000, 16, 5955–5959.
4. K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm, T.-H. Bae and J. R. Long, Chem. Rev., 2012, 112, 724–781.
5. N. Wang, N. Zhang, J. Yan and S. Yao, Int. J. Mol. Sci., 2018, 34, 45–48.
6. J. Shi, S. Zhang and J. Du, J. Mol. Sci., 2017, 33, 503–508.
7. G. Lin, H. Ding, R. Chen, Z. Peng, B. Wang and C. Wang, J. Am. Chem. Soc., 2017, 139, 8705–8709.
8. S. Y. Ding, J. Gao, Q. Wang, Y. Zhang, W. G. Song, C. Y. Su and W. Wang, J. Am. Chem. Soc., 2011, 133, 19816–19822.
9. Y. Zeng, R. Zou and Y. Zhao, Adv. Mater., 2016, 28, 2855–2873.
10. X. Guan, Y. Ma, H. Li, Y. Yusran, M. Xue, Q. Fang, Y. Yan, V. Valtchev and S. Qiu, J. Am. Chem. Soc., 2018, 140, 4494–4498.
11. W. Li, Y. Pang and J. Zhang, Int. J. Mol. Sci., 2013, 29, 469–479.
12. M. Tong, Q. Yang and C. Zhong, Microporous Mesoporous Mater., 2015, 210, 142–148.
13. Z. Kang, Y. Peng, Y. Qian, D. Yuan, M. A. Addicoat, T. Heine, Z. Hu, L. Tee, Z. Guo and D. Zhao, Chem. Mater., 2016, 28, 1277–1285.
14. C. J. Doonan, D. J. Tranchemontagne, T. G. Glover, J. R. Hunt and O. M. Yaghi, Nat. Chem., 2010, 2, 235.
15. S. Kandambeth, A. Mallick, B. Lukose, M. V. Mane, T. Heine and R. Banerjee, J. Am. Chem. Soc., 2012, 134, 19524–19527.
16. Q. Fang, Z. Zhuang, S. Gu, R. B. Kaspar, J. Zheng, J. Wang, S. Qiu and Y. Yan, Nat. Commun., 2014, 5, 4503.
17. H. Furukawa and O. M. Yaghi, J. Am. Chem. Soc., 2009, 131, 8875–8883.
18. Y. Zhao, K. X. Yao, B. Teng, T. Zhang and Y. Han, Energy Environ. Sci., 2013, 6, 3684–3692.
19. L. A. Baldwin, J. W. Crowe, D. A. Pyles and P. L. McGrier, J. Am. Chem. Soc., 2016, 138, 15134–15137.
20. X. Han, J. Huang, C. Yuan, Y. Liu and Y. Cui, J. Am. Chem. Soc., 2018, 140, 892–895.
21. X. Han, Q. Xia, J. Huang, Y. Liu, C. Tan and Y. Cui, J. Am. Chem. Soc., 2017, 139, 8693–8697.
22. Y. J. Colón and R. Q. Snurr, Chem. Soc. Rev., 2014, 43, 5735–5749.
23. Y. Pang, W. Li and J. Zhang, J. Phys. Chem. C, 2016, 120, 4329–4336.
24. B. Zhou, W. Li and J. Zhang, J. Phys. Chem. C, 2017, 121, 20197–20204.
25. Y. Guan, W. Li, H. Wang and J. Zhang, Chemistry, 2019, 25, 2303–2312.
26. X. Zhang, W. Li, Y. Guan, B. Zhou and J. Zhang, Chem. – Eur. J., 2019, 25, 6569–6574.
27. N. Lock, Y. Wu, M. Christensen, L. J. Cameron, V. K. Peterson, A. J. Bridgeman, C. J. Kepert and B. B. Iversen, J. Phys. Chem. C, 2010, 114, 16181–16186.
28. K. Koh, A. G. Wong-Foy and A. J. Matzger, Angew. Chem., Int. Ed., 2008, 47, 677–680.
29. 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.
30. J. Potoff Jeffrey and J. I. Siepmann, AIChE J., 2004, 47, 1676–1682.
31. M. G. Martin and J. I. Siepmann, J. Phys. Chem. B, 1998, 102, 2569–2577.
32. F. Darkrim and D. Levesque, J. Chem. Phys., 1998, 109, 4981–4984.
33. Q. Wang, H. Wang, S. Peng, X. Peng and D. Cao, J. Phys. Chem. C, 2014, 118, 10221–10229.
34. Z. Yang, X. Peng and D. Cao, J. Phys. Chem. C, 2013, 117, 8353–8364.
35. S. L. Mayo, B. D. Olafson and W. A. Goddard, J. Chem. Phys., 1990, 94, 8897–8909.
36. G. Garberoglio and R. Vallauri, Microporous Mesoporous Mater., 2008, 116, 540–547.
37. G. Maitland, M. Rigby, E. Smith, W. Wakeham and D. Henderson, Intermolecular Forces: Their Origin and Determination, Clarendon Press, 1981.
38. A. K. Rappe and W. A. Goddard, J. Chem. Phys., 1991, 95, 3358–3363.
39. A. Gupta, S. Chempath, M. J. Sanborn, L. A. Clark and R. Q. Snurr, Mol. Simul., 2003, 29, 29–46.
40. J. Cho, S. Rho, S. Park, J. Lee and H. Kim, Fluid Phase Equilib., 1998, 144, 69–75.
41. O. Talu and L. Myers Alan, AIChE J., 2004, 47, 1160–1168.
42. S. E. Wenzel, M. Fischer, F. Hoffmann and M. Fröba, Inorg. Chem., 2009, 48, 6559–6565.
43. Z. Liu, K. Zhang, Y. Wu and H. Xi, Appl. Surf. Sci., 2018, 440, 351–358.
44. J. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
45. S. Grimme, J. Comput. Chem., 2006, 27, 1787–1799.
46. B. Delley, J. Chem. Phys., 2000, 113, 7756–7764.
47. S. Grimme, J. Chem. Phys., 2006, 124, 034108.
48. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104.
49. F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297–3305.
50. A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 88, 899–926.
51. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Gaussian, Inc., Wallingford, CT, 2009.

---

Footnote

† Electronic supplementary information (ESI) available: Structural parameters of the unit cells in CCOFs, snapshots of CH₄, N₂, and H₂ in CCOF5 and CCOF6 at T = 298 K at different pressures. Other confirmations and interaction energies between CO₂ and the segment of CCOF5 and CCOF6. See DOI: 10.1039/c9nj05172d

---