Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2020 Sep 18;124(41):22577–22590. doi: 10.1021/acs.jpcc.0c07062

Computational Selection of High-Performing Covalent Organic Frameworks for Adsorption and Membrane-Based CO2/H2 Separation

Gokhan Onder Aksu 1, Hilal Daglar 1, Cigdem Altintas 1, Seda Keskin 1,*
PMCID: PMC7591139  PMID: 33133330

Abstract

graphic file with name jp0c07062_0008.jpg

Covalent organic frameworks (COFs) have high potential in gas separation technologies because of their porous structures, large surface areas, and good stabilities. The number of synthesized COFs already reached several hundreds, but only a handful of materials were tested as adsorbents and/or membranes. We used a high-throughput computational screening approach to uncover adsorption-based and membrane-based CO2/H2 separation potentials of 288 COFs, representing the highest number of experimentally synthesized COFs studied to date for precombustion CO2 capture. Grand canonical Monte Carlo (GCMC) simulations were performed to assess CO2/H2 mixture separation performances of COFs for five different cyclic adsorption processes: pressure swing adsorption, vacuum swing adsorption, temperature swing adsorption (TSA), pressure−temperature swing adsorption (PTSA), and vacuum−temperature swing adsorption (VTSA). The results showed that many COFs outperform traditional zeolites in terms of CO2 selectivities and working capacities and PTSA is the best process leading to the highest adsorbent performance scores. Combining GCMC and molecular dynamics (MD) simulations, CO2 and H2 permeabilities and selectivities of COF membranes were calculated. The majority of COF membranes surpass Robeson’s upper bound because of their higher H2 permeabilities compared to polymers, indicating that the usage of COFs has enormous potential to replace current materials in membrane-based H2/CO2 separation processes. Performance analysis based on the structural properties showed that COFs with narrow pores [the largest cavity diameter (LCD) < 15 Å] and low porosities (ϕ < 0.75) are the top adsorbents for selective separation of CO2 from H2, whereas materials with large pores (LCD > 20 Å) and high porosities (ϕ > 0.85) are generally the best COF membranes for selective separation of H2 from CO2. These results will help to speed up the engineering of new COFs with desired structural properties to achieve high-performance CO2/H2 separations.

1. Introduction

Separation of CO2 from H2 is important for various processes such as petroleum refining and production of hydrogen, methanol, or ammonia.1 For example, in the reaction of synthesis gas with steam, pure H2 and electricity are produced, while precombustion CO2 capture is necessary to remove CO2 from synthesis gas.2,3 Amine scrubbing is widely used for CO2 capture, but it is economically not feasible because of the high cost of solvent regeneration.4 Adsorption-based gas separation processes, such as pressure swing adsorption (PSA), vacuum swing adsorption (VSA), temperature swing adsorption (TSA), pressure−temperature swing adsorption (PTSA), and vacuum−temperature swing adsorption (VTSA), using porous materials are alternatives for CO2 capture.5−12 Zeolites and activated carbons are widely used as adsorbents for CO2 capture from syngas, but they generally suffer from low working capacities.13 Membrane-based separation is another option for CO2 capture and H2 purification.14 Polymers are widely used, but their performances are restricted by the trade-off between gas permeability and selectivity.15−17 Zeolite membranes offer higher CO2/H2 selectivities and permeabilities than polymers,18 but they have high manufacturing costs and limited tunability of the pore size.19 Therefore, it is highly required to discover new materials for adsorption and membrane-based separation of CO2/H2 mixtures.

Metal–organic frameworks (MOFs), composed of metal nodes connected with organic linkers, are considered as promising materials for various gas separations.20−23 Covalent organic frameworks (COFs) consist of lighter elements (H, C, N, O, and B) and organic linkers that are covalently bonded.24 Compared to MOFs and other porous materials, most of the COFs have high porosities, good thermal stabilities, lower densities, and larger surface areas.25 Therefore, COFs can also be considered as potential adsorbents and membranes for gas separation applications.24,26 Experimental and computational work on COFs mostly concentrated on single-component and binary/ternary mixture separation on a small set of materials.27−41 For example, molecular simulations showed that CO2/CH4 selectivities of five COFs are higher than those of zeolites (DD3R, MFI), nanoporous carbon (C168 schwarzite) and a prototype MOF (IRMOF-1) based on the pure gas adsorption isotherm data.37 A total of 46 COFs were studied to determine their CO2/H2, CH4/H2, and CO2/CH4 separation performances under PSA conditions, and COFs with high working capacities and selectivities were found to be comparable with zeolites and MOFs.42 Earlier computational studies showed that the BTP-COF membrane has a higher CO2 permeability than zeolite membranes such as MFI and CHA but a lower CO2/H2 selectivity.43 COF-6 was computed to have higher CH4/H2 selectivity compared to some MOF membranes,33 and it was predicted to have higher CO2 and H2 permeabilities than some polymers, zeolites, and MOF membranes for CO2/CH4 and H2/CH4 separations.44

The number of synthesized COFs has been continuously growing, and it is not practical to assess gas separation performances of a large number of COFs using purely experimental methods. High-throughput computational screening of materials is very useful to examine gas separation performances of a very large number of materials to guide the future experimental studies to the best candidates. Development of COF databases, computation-ready experimental COFs (CoRE COFs)45−47 and clean, uniform, and refined with automatic tracking from experimental database (CURATED) COFs,48 facilitated computational screening of experimentally synthesized COFs for various applications.42,45−51 The CoRE COF database was screened for CH4 deliverable capacity46 and H2 storage.49 Computational screening of 187 CoRE COFs for adsorption-based noble gas separations under PSA and VSA conditions revealed that COFs can have high adsorption selectivities for Kr/Ar, Xe/Kr, and Rn/Xe and high working capacities for Kr and Xe.45 A total of 298 CoRE COFs were investigated for CO2/CH4 separation,47 and it was found that −F and −Cl functional groups increased the membrane selectivities up to 2 orders of magnitude and carried several COF membranes over Robeson’s upper bound17 which was defined for polymeric membranes. A total of 295 CoRE COFs were recently screened for CO2/N2 separation, and it was concluded that many COF adsorbents can compete with MOFs in CO2 capture from flue gas.51 A total of 296 CURATED COFs were also screened with a workflow algorithm to evaluate their postcombustion CO2 capture performances, and the COF with the lowest parasitic energy was identified for the PTSA process.48 Hypothetical (computer-generated but not yet synthesized) COFs were studied for CH4 storage.52,53 A total of ∼69,000 hypothetical COF structures (hCOFs) were screened for CO2/N2 separation under PTSA conditions, and many hCOFs were found to have higher CO2 working capacities compared to CURATED COFs.54 As can be seen from this literature review, COFs have not been screened so far for adsorption- and membrane-based CO2/H2 separation. We recently showed that a large number of MOFs can outperform zeolites and polymers in CO2/H2 separation processes.55 Similarity of COFs with MOFs in terms of structural properties makes COFs also potential adsorbents and membranes for CO2/H2 separation. Therefore, a comprehensive work aiming to unlock CO2/H2 separation performances of experimentally synthesized COFs is strongly needed to guide the selection of the best COF adsorbents and membranes for future studies.

In this work, we screened the CoRE COF database47 to distinguish the best COF adsorbents and membranes for CO2/H2 separation. Grand canonical Monte Carlo (GCMC) simulations were performed to compute adsorption-based CO2 separation performances of COFs for the CO2/H2: 15/85 mixture under five different process conditions, PSA, VSA, TSA, PTSA, and VTSA. The selectivities, working capacities, adsorbent performance scores (APSs), and regenerabilities were computed for all COF adsorbents at each process condition and compared to elucidate how cyclic adsorption processes affect gas separation performances of COFs. The top 10 promising COFs were identified for each separation process based on the combination of different evaluation metrics. The effect of using different generic force fields in molecular simulation of COFs on the calculated adsorbent performance metrics was also examined. Molecular dynamics (MD) simulations were then performed to predict the performances of COF membranes for H2/CO2 separation. The selectivities and permeabilities of COF membranes were compared with those of zeolites and polymers, and the top 10 COF membranes were identified. Finally, the relationship between selectivities and structural features of COFs was studied to describe the optimum structural properties for highly selective adsorbents and membranes. Our results will provide a molecular-level understanding of how CO2 and H2 adsorb and diffuse in COFs and guide the selection of high-performing COFs for five different swing adsorption processes in addition to the membrane-based precombustion CO2 capture process.

2. Computational Details

We used the third version of the CoRE COF database which consists of 309 COFs.47 First, structural features such as pore-limiting diameter (PLD), the largest cavity diameter (LCD), accessible surface area (Sacc), density (ρ), and porosity (ϕ) of materials were estimated using Zeo++ software (version 0.3).56Sacc was calculated using a probe that has the same kinetic diameter as the N2 molecule (3.64 Å), and we eliminated the COFs with null accessible surface area. Then, the COF database was narrowed down to materials with PLDs larger than the kinetic diameter of the CO2 molecule, 3.3 Å, to allow the passage of both CO2 and H2 molecules in the framework. We note that the atomic coordinates of HAT-NTBA-COF were corrected following the literature.48,57 Overall, we ended up with 288 diverse COFs illustrating different chemical and structural features.

GCMC simulations were performed using RASPA simulation software58 (version 2.0.36) to obtain the CO2 and H2 uptakes in COFs. In order to represent the industrial precombustion gas mixture, the mole fractions of CO2 and H2 in the bulk mixture were set as 0.15 and 0.85, respectively.59 Simulations were conducted at two different temperatures (298 and 393 K) and three different pressures (0.1, 1, and 10 bar) to compute the CO2 and H2 uptakes of COFs for five different processes (PSA, VSA, TSA, PTSA, and VTSA), as given in detail in Table 1. Pressure was converted to fugacity by using the Peng–Robinson equation of state.58 Lennard-Jones (LJ) potential was used to specify dispersion interactions between all atoms, and Coulomb potential was used to define the electrostatic interactions between charged atoms. A single-site spherical model with the LJ 12-6 potential60 was used to model H2, while CO2 was modeled as a three-site linear, rigid molecule with a C–O bond length of 1.16 Å and partial point charges located at the center of each site.61 In our previous works, it was shown that both the universal force field (UFF) and Dreiding force field can be used for gas adsorption simulations of MOFs and COFs to reproduce the experimental gas uptakes.33,55 We computed the single-component adsorption isotherms of CO2 in three COFs (COF-5, COF-6, and COF-10) using both the UFF and Dreiding and compared our results with the available experimental data of these COFs.27Figure S1 in the Supporting Information shows that computational CO2 uptakes are in better agreement with the experimental results when the Dreiding force field was used, and therefore, we used Dreiding for simulations of all COFs. Because CO2 has a quadrupole moment, the Coulomb potential was used to compute CO2–CO2 and CO2–COF electrostatic interactions. The partial point charges of the atoms of COFs were assigned using the charge equilibration method (Qeq) as implemented in RASPA.62 The long-range electrostatic interactions between CO2 molecules and COF atoms were calculated using the Ewald summation.63 Lorentz–Berthelot mixing rules were applied to define pairwise interactions between unlike atoms. A total of 30,000 cycles were used for GCMC simulations, while 10,000 (20,000) cycles were used for initialization (taking ensemble averages). Here, we want to note that each cycle consists of a minimum of 20 and a maximum of N steps, where N implies the number of adsorbed molecules at each cycle.58 Translation, regrow, reinsertion, swap, and identity exchange moves were used in binary mixture simulations, while for CO2 molecules, the rotation move was also considered. 14 Å was defined as the cutoff radius for intermolecular interactions as it has been previously used in high-throughput screening of COFs,45−47 and each dimension of the simulation box was expanded to at least 28 Å following the literature.64

Table 1. Adsorption–Desorption Conditions of Processes.

process adsorption pressure and temperature desorption pressure and temperature references
PSA 10 bar, 298 K 1 bar, 298 K (11)
VSA 1 bar, 298 K 0.1 bar, 298 K (11)
TSA 1 bar, 298 K 1 bar, 393 K (7)
PTSA 10 bar, 298 K 1 bar, 393 K (8,103)
VTSA 1 bar, 298 K 0.1 bar, 393 K (104,105)
Open in a new tab

To compute the gas permeabilities, we calculated the self-diffusivities (Dself) of CO2 and H2 in COFs by performing MD simulations. The gas uptakes acquired from binary mixture GCMC simulations at 10 bar and 298 K were used for the mixture MD simulations. COFs were assumed to be rigid during MD simulations to reduce the computational cost. A total of 106 cycles were used for both initialization and equilibration. MD simulations were carried out for 5 × 106 cycles in the NVT ensemble using a time step of 1 fs. A Nosé–Hoover thermostat65 was used in MD simulations to keep the temperature constant. We obtained the mean-square displacement of gas molecules using the modified order-N algorithm as implemented in RASPA58 which utilizes blocking averages to reduce statistical errors,64 and the self-diffusivities were calculated by Einstein’s relation.66 To reduce uncertainties, simulations were performed twice.

We utilized various material performance evaluation metrics as listed in Table 2 to assess adsorption- and membrane-based gas separation performances of COFs. One of the most important metrics to identify a good adsorbent is adsorption selectivity (Sads), and it was computed by dividing the uptake of the strongly adsorbed component (CO2) by the uptake of the weakly adsorbed component (H2), normalized by dividing with the bulk mixture composition, and represented as Sads,CO2/H2 in this work. Working capacity (ΔN) is important to determine the amount of gas that can be captured at each adsorption cycle. In this work, ΔNCO2 was calculated by taking the difference of CO2 loadings under adsorption and desorption conditions given in Table 1. To approximate the performance of COFs for the PTSA and VTSA processes, GCMC simulation results obtained from three different conditions (both pressure and temperature difference) were used.67 The APS was obtained by multiplying the selectivity and working capacity.68 Percent regenerability (R %) is the ratio of working capacity to the gas uptake at the adsorption pressure.69 The performances of COF membranes were evaluated using the gas permeabilities (Pi) and membrane selectivities (Smem). The CO2 and H2 permeabilities were calculated by using the results obtained from GCMC and MD simulations. Membrane selectivity was defined as the ratio of permeability of the more permeable gas species to permeability of the less permeable one, as provided in Table 2.70 The permeability of H2 is greater than the permeability of CO2 in COFs; therefore, membrane selectivities were reported as H2/CO2 (Smem,H2/CO2). The steps of our computational work are provided in Scheme S1 of the Supporting Information.

Table 2. Performance Metrics Computed To Evaluate COF Adsorbents and COF Membranesa.

metric formula
mixture adsorption selectivity
graphic file with name jp0c07062_m001.jpg
working capacity ΔNCO2 = Nads,CO2 – Ndes,CO2
adsorbent performance score
graphic file with name jp0c07062_m002.jpg
percent regenerability
graphic file with name jp0c07062_m003.jpg
mixture permeability
graphic file with name jp0c07062_m004.jpg
mixture diffusion selectivity
graphic file with name jp0c07062_m005.jpg
mixture membrane selectivity
graphic file with name jp0c07062_m006.jpg
Open in a new tab
a

Nads,i: gas uptake under adsorption conditions (mol/kg), Ndes,i: gas uptake under desorption conditions (mol/kg), Dimix: self-diffusivity of gas species in the mixture (m2/s), y: bulk composition of the gas mixture, ci: molar concentration of adsorbed gas species (mol/m3), fi: partial pressure of gas species in the mixture (Pa). 1 barrer = 3.348 × 10–16 mol·m/(m2·s·Pa).

3. Results and Discussion

3.1. COF Adsorbents

We evaluated COF adsorbents for five different adsorption-based separation processes. In the PSA and VSA processes, adsorbed gases are desorbed by lowering the pressure at constant temperature. TSA is based on the temperature difference between adsorption and desorption steps, and gas molecules are desorbed upon heating, while pressure is kept constant. The combination of PSA or VSA with TSA, named PTSA and VTSA, is useful to have a good regeneration of adsorbents. We compared the performances of COF adsorbents under all these process conditions to elucidate the optimum process condition for high-efficiency CO2/H2 separation using COFs. The most promising adsorbents should provide both high Sads,CO2/H2 and high ΔNCO2. Figure 1 shows the relation between Sads,CO2/H2 and ΔNCO2 for different processes, and the general trend for all processes is that COFs with high Sads,CO2/H2 suffer from low ΔNCO2 and vice versa. COFs which offer moderate Sads,CO2/H2 and ΔNCO2 have APSs between 10 and 100 mol/kg. According to Figure 1a,b, APSs of COFs under PSA and PTSA conditions show a similar trend, while Figure 1c–e shows that APSs under VSA, TSA, and VTSA conditions are more alike. Therefore, we discussed these processes separately as under (i) PSA and PTSA and (ii) VSA, TSA, and VTSA conditions.

Figure 1.

Figure 1

Open in a new tab

Relation between Sads,CO2/H2, ΔNCO2, and APS of COFs for the CO2/H2: 15/85 mixture under (a) PSA, (b) PTSA, (c) VSA, (d) TSA, and (e) VTSA conditions. Black, blue, and red data show COFs with APS < 10 (mol/kg), 10 < APS < 100 (mol/kg), and APS > 100 (mol/kg), respectively.

Because the adsorption conditions are the same for the PSA and PTSA processes, as provided in Table 1, the Sads,CO2/H2 values are the same and computed as 2.38–428. The ΔNCO2 values of COFs under PSA conditions (0.21–6.82 mol/kg) are lower than those under PTSA conditions (0.23–8.61 mol/kg). This is expected because increasing the temperature under desorption conditions results in an increase in the amount of desorbed gas, which increases ΔNCO2.71 Black, blue, and red points represent COFs offering low (<10 mol/kg), moderate (between 10 and 100 mol/kg), and high (>100 mol/kg) APSs, respectively. A total of 32 (40) out of 288 COFs offer the highest APSs (>100 mol/kg) under PSA (PTSA) conditions. COFs with high APSs (>100 mol/kg) under PSA (PTSA) conditions have Sads,CO2/H2 between 30.1 and 428 (22.9–428) and ΔNCO2 between 0.95 and 6.82 (1.74–8.61) mol/kg. Similar to the APS, a high R % is also required for efficient adsorption-based gas separation processes. We previously showed that MOFs with high APSs generally suffer from low R %.55Figure S2a shows that most COFs (244 out of 288) have R % > 85%, but those with high APSs (>100 mol/kg) have a low R % under PSA conditions. On the other hand, Figure S2b shows that all COFs have R % > 85% as the lowest R % was computed as 95% under PTSA conditions. This is due to the combination of temperature and pressure swing in PTSA which leads to higher ΔNCO2 and a higher R % compared to the PSA process.

The most promising COF adsorbents offering R % > 85% together with the highest APSs were identified for each process condition. The top 10 COFs for PSA and PTSA conditions are highlighted in Figure S2a,b and listed in Tables S1 and S2, with their APS, R %, Sads,CO2/H2, and structural properties. TPE-COF-I72 was identified as the best COF for the PSA process. Because of its narrower pores (LCD: 7.82 Å, PLD: 7.17 Å) and lower porosity (0.60) compared to other COFs, adsorption of CO2 molecules is favored in TPE-COF-I, resulting in enhanced selectivity. TpPa-F473 is another COF among the top 10 materials offering high performance for CO2/H2 separation under PSA conditions. Sads,CO2/H2 and ΔNCO2 were computed as 64.7 and 2.03 mol/kg, respectively, for this COF which is functionalized with fluorine (F) atoms. These functional groups contribute to the electrostatic interactions and enhanced CO2 selectivity. To represent this, we repeated GCMC simulations by neglecting CO2–COF electrostatic interactions for all COFs and then compared their selectivities, as shown in Figure S3. Neglecting the framework charges significantly underestimates the selectivities of many COFs at both 1 and 10 bar, as shown in Figure S3a,b, respectively. When the Coulomb interactions were not considered, Sads,CO2/H2 of TpPa-F4 dramatically decreased from 64.7 to 11.3 at 10 bar because of the decrease in CO2 adsorption. Similarly, when the framework charges were absent, another top COF functionalized with F atoms, EB-COF:F,74 was computed to have a lower Sads,CO2/H2 (15.1) than the one calculated by considering electrostatic interactions (61.6). We also provided snapshots for adsorption of CO2 and H2 molecules in EB-COF:F with and without electrostatic interactions under PSA conditions, as shown in Figure S4. CO2 molecules mostly gather close to the pore walls because of their stronger intermolecular interactions with the framework atoms than H2 molecules. These results showed that CO2–COF electrostatic interactions play a major role in determining selectivities and hence identification of the top COFs.

Interestingly, when the top COFs of the PSA and PTSA processes were compared, no common material was found. For the PTSA process, NPN-275 was found to be the best COF with an Sads,CO2/H2 of 413 and a ΔNCO2 of 5.31 mol/kg. The other two top COFs, NPN-1 and NPN-3, also have high Sads,CO2/H2, which can be attributed to their narrow pore sizes (between 4.11 and 6.14 Å).42,75 Under PSA conditions, NPN-2 has the highest APS among all COFs, but it is not one of the best COFs because of its low R %. This is because of the desorption condition of PSA (1 bar, 298 K) at which CO2 molecules cannot desorb enough, leading to a low ΔNCO2 and a low R %. When we examined the performances of all the top COF adsorbents of the PTSA process under PSA conditions, we noticed that they cannot be categorized as the top materials for PSA because of their low R %. Although COFs offer the same selectivity under both PSA and PTSA conditions, they have higher ΔNCO2 and R % under PTSA conditions compared to those under PSA conditions. Therefore, COFs offering high Sads,CO2/H2 become prominent as the top adsorbents for CO2/H2 separation under PTSA conditions. We also compared the selectivities and working capacities of COFs that we examined in this work with those of zeolites, MOFs, and hypothetical MOFs (hMOFs) taken from the literature, as listed in Table 3. The results revealed that many COFs have better/similar performances than/to zeolites and hMOFs under PSA conditions, whereas MOFs generally have higher selectivities than COFs.55,68,76,77

Table 3. Comparison of Molecular Simulation Results for Zeolites, MOFs, and hMOFs for Syngas Separations.

material Sads,CO2/H2 condition ΔNCO2 (mol/kg) references
NaX 1257 CO2/H2: 15/85, 300 K 1.26a (77)
NaY 540 CO2/H2: 15/85, 300 K 2.69a (77)
AFX 505.5 CO2/H2: 15/85, 300 K 3.42a (77)
CHA 117.4 CO2/H2: 15/85, 300 K 2.69a (77)
GME 97.2 CO2/H2: 15/85, 300 K 2.25a (77)
MOR 134.1 CO2/H2: 15/85, 300 K 0.84a (77)
MFI 95.4 CO2/H2: 15/85, 300 K 1.25a (77)
FAU 17 CO2/H2: 15/85, 300 K 0.67a (77)
MOFs 2.43 to 2.4 × 104 CO2/H2: 15/85, 298 K 1 × 10–3 to 11.28a (55)
MOFs 2.47 to 8.4 × 104 CO2/H2: 15/85, 298 K 7 × 10–3 to 4.60b (55)
hMOFs 1.62–717.3 CO2/H2: 20/80, 313 K 7 × 10–3 to 7.78c (68)
hMOFse 0.001–337.5 CO2/H2: 40/60, 313 K 0.76–38.77d (76)
Open in a new tab
a

Working capacity is reported between 10 and 1 bar.

b

Working capacity is reported between 1 and 0.1 bar.

c

Working capacity is reported between 20 and 1 bar.

d

Working capacity is reported between 40 and 1 bar.

e

Predicted by the gradient-boosted trees regression method.

Figure 1c–e shows the relationship between Sads,CO2/H2 and ΔNCO2 under VSA, TSA, and VTSA conditions, respectively. These processes have the same adsorption condition (1 bar, 298 K), and therefore, COFs have the same Sads,CO2/H2 range, 2.39–625. The ΔNCO2 values of COFs under VSA (TSA) conditions were computed to be between 0.02 and 1.74 mol/kg (0.02–1.86 mol/kg). Slightly higher ΔNCO2 values were observed under VTSA conditions, 0.02–2.05 mol/kg. All COFs offering APS > 100 mol/kg under VSA conditions have APS > 100 mol/kg under TSA and VTSA conditions as well. The relation between R % and APSs of COFs under VSA, TSA, and VTSA conditions is shown in Figure S2c–e, respectively. COFs with high APSs have low R % under VSA conditions. However, a different trend was observed under TSA conditions such that COFs with high APSs also have high R % and vice versa. The residual adsorbed amount of CO2 under desorption conditions (Ndes,CO2) is the only term varying for these two processes because the desorption conditions of VSA (0.1 bar, 298 K) and TSA (1 bar, 393 K) are different, while the adsorption condition is the same (1 bar, 298 K). Therefore, we investigated the trends of Ndes,CO2 under the desorption conditions of TSA and VSA, as shown in Figure S5a. For most of the COFs, the temperature difference between adsorption and desorption conditions of the TSA process leads to higher Ndes,CO2 and lower ΔNCO2 values compared to Ndes,CO2 values obtained with the pressure change of the VSA process. In other words, we observed that more CO2 is still adsorbed at 1 bar and 393 K than at 0.1 bar and 298 K for most of the COFs. This causes lower ΔNCO2 and a lower R % under TSA conditions compared to those under VSA conditions and shows that pressure swing is more effective than temperature swing for the desorption of CO2 molecules. As a result, the number of COFs with R % > 85% under VSA conditions is more than that under TSA conditions, as shown in Figure S5b. On the other hand, all COFs have high R % values (>95%) under VTSA conditions, similar to PTSA.

The number of COFs with R % > 85% is 262 (53) under VSA (TSA) conditions. The top 10 COFs with the highest APSs while R % > 85% are shown by red points in Figure S2c–e for VSA, TSA, and VTSA conditions, respectively. These top adsorbents are also listed in Tables S3–S5. There are four common top materials for these processes, NPN-2,75 COF-42-gra,78 FLT-COF-1-staggered,79 and COF-JLU-3.80 We observed that staggered modes in COF-42-gra and FLT-COF-1-staggered favor the adsorption of CO2 molecules. The eclipsed modes of these two COFs, COF-42-bnn and FLT-COF-1-eclipsed, have lower Sads,CO2/H2 and CO2 uptakes than the staggered ones because narrower pores of staggered modes favor the confinement of CO2 molecules compared to the eclipsed mode of the same COFs. Postsynthetic modifications are very important to increase the productivity of COFs for separation applications.81 3D-COOH-COF is another top COF for the TSA and VTSA processes, and it is a postsynthetic carboxylated version of 3D-OH-COF.82 When we compared their adsorption selectivities, Sads,CO2/H2 of 3D-COOH-COF (625) is much larger than Sads,CO2/H2 of its predecessor (27.6), and it can be attributed to the additional adsorption sites provided by the carboxylate groups of 3D-COOH-COF. Four materials, NPN-2, COF-42-gra, FLT-COF-1-staggered, and COF-JLU-3 among 288 COFs, were found to be promising for all processes except the PSA process. As mentioned before, the selectivities were underestimated when Coulombic interactions between COFs and CO2 molecules were neglected, and the selectivities of the top materials for the VSA, TSA, and VTSA processes such as 3D-COOH-COF and FLT-COF-1-staggered also decreased at 1 bar as shown in Figure S3a.

The distribution and ranges of Sads,CO2/H2, ΔNCO2, APSs, and R % under PSA, VSA, TSA, PTSA, and VTSA conditions are provided in Figure 2 and Table 4. Figure 2a illustrates that COFs tend to have higher Sads,CO2/H2 (>50) under VSA, TSA, and VTSA conditions compared to those under PSA and PTSA conditions. On the other hand, the ΔNCO2 values of COFs under PSA and PTSA conditions are much larger than those under VSA, TSA, and VTSA conditions as shown in Figure 2b. Figure 2c shows that there are more COFs with mediocre (>10 mol/kg) and high APSs (>100 mol/kg) under PSA and PTSA conditions compared to VSA, TSA, and VTSA conditions. In other words, among the five different cyclic adsorption processes we considered, COFs perform as the best in the PSA and PTSA processes. It is important to note that the main factor determining the APS trends of different processes is ΔNCO2. For example, COF-42-gra has a high APS under PTSA conditions (877 mol/kg), which is almost triple of its corresponding APS under VTSA conditions (298 mol/kg). Sads,CO2/H2 of this COF under VTSA conditions (190) is larger than the one calculated under PTSA conditions (134). Because ΔNCO2 of this COF under PTSA conditions (6.52 mol/kg) is much larger than its ΔNCO2 under VTSA conditions (1.57 mol/kg), COF-42-gra has a higher APS and performs better under PTSA conditions compared to VTSA conditions. Overall, the highest APSs were obtained under PSA and PTSA conditions for adsorption-based CO2/H2 mixture separation using COFs. Figure 2d shows that all COFs have high R % under PTSA and VTSA conditions. As discussed before, the addition of temperature swing to the pressure swing process enhances CO2 desorption and leads to high R %. However, another important factor is that materials used in temperature swing processes should have high thermal stabilities because of high operating temperatures. We examined that COFs listed as our top materials for the TSA, PTSA, and VTSA processes have very good thermal stabilities up to 500 °C except NPN-group COFs which have thermal stabilities around 140 °C.75,78−80,82−88

Figure 2.

Figure 2

Open in a new tab

Distribution of (a) Sads,CO2/H2, (b) ΔNCO2, (c) APS, and (d) R % for each adsorption-based separation process.

Table 4. Ranges of Performance Metrics Computed with Molecular Simulation Results for 288 COFs under PSA, VSA, TSA, PTSA, and VTSA Conditions.

condition Sads,CO2/H2 ΔNCO2 (mol/kg) APS (mol/kg) R (%)
PSA 2.38–428 0.21–6.82 1.10–1511 52.0–91.1
VSA 2.39–625 0.02–1.74 0.11–801 76.3–90.8
TSA 2.39–625 0.02–1.86 0.08–857 50.3–92.5
PTSA 2.38–428 0.23–8.61 1.17–2196 94.8–98.9
VTSA 2.39–625 0.02–2.05 0.12–936 95.6–99.2
Open in a new tab

As previously mentioned, molecular simulations using Dreiding agreed better with the reported experimental gas uptakes. We further investigated how using a different generic force field for the COFs can affect the simulation results. GCMC simulations were repeated using UFF parameters, and the adsorbent performance metrics were recalculated for all COFs under VSA conditions. Our group previously reported that the quantitative values of performance metrics obtained by using the UFF or Dreiding can be significantly different; however, ranking of materials may not significantly change.89 Thus, we specifically focused on the effect of using Dreiding or UFF parameters on ranking of COF adsorbents and computed Spearman’s ranking correlation coefficient (SRCC) that compares the rankings of COFs as shown in Table S6. The results show that rankings of COFs based on Sads,CO2/H2, ΔNCO2, and APSs are highly similar, resulting in SRCC > 0.95. However, SRCC of material ranking based on R % was calculated as 0.53, indicating that ranking of COFs based on R % would be very dissimilar when a different generic force field was used. In order to better understand the effect of force field on the identification of the most promising COFs, we compared the R % values obtained from GCMC simulations using Dreiding and the UFF, as shown in Figure 3. A total of 8 of the top 10 materials are the same when either of the force fields was used; however, if we used the UFF, some COFs would not be distinguished as the top. For instance, PyTTA-BFBIm-iCOF was identified as a top COF for the VSA process when simulations were performed with Dreiding. However, it was not among the top COFs when the UFF was used (Table S7) because its R % was computed to be <85%. Overall, these results showed that force field selection can be important if the COFs are ranked based on R %.

Figure 3.

Figure 3

Open in a new tab

Comparison of the top COFs identified from simulations using Dreiding or the UFF for the VSA process. Red and blue data represent the top materials identified from simulations using Dreiding or the UFF, respectively, whereas green data represent the common top materials identified from simulations using Dreiding or UFF parameters.

Adsorption-based gas separation has multiple requirements including low cost, high purity, and cyclic usage of the adsorbent, and real-time experimental techniques, such as utilization of fixed-bed adsorption units, generally focus on the working capacity and the selectivity of the material.90 To identify the most promising adsorbents among many candidates for adsorption-based gas separation using high-throughput computational screening, adsorbent performance evaluation metrics such as adsorption selectivity, working capacity, and regenerability are generally used.69,90−92 High selectivity values help to identify the adsorbents which can provide high-purity separation, while high working capacity and high regenerability values indicate that the adsorbent can be used several times with high productivity. Because of the differences between experimental and computational setups (fluctuations in the temperature or pressure, impurity of the gas feed, defects in the adsorbent material, or diffusional limitations), one can expect that calculated adsorbent performance evaluation metrics represent the approximate performance of the adsorbent and they may not perfectly reflect the real-time experimental performance. The aim of our work was to highlight the most promising COF adsorbents among several hundred adsorbents via utilizing these simulated metrics, and a more detailed analysis can be made on the top material candidates. Finally, it is important to note that the method that we used to assign partial atomic charges to framework atoms can be critical for the calculated values of adsorbent performance evaluation metrics. In our recent work, we showed that although the ranking of materials does not change significantly, the choice of the charge method can overestimate/underestimate the performance values of materials.51,93 Therefore, it would be a good practice to test the performances of the most promising COF adsorbents by using high-quality partial charges, such as density-derived electrostatic and chemical (DDEC) charges assigned to CURATED COFs, before experimental efforts.

3.2. COF Membranes

Membranes are alternatively used for CO2/H2 mixture separation because of different molecular sizes of H2 (2.9 Å) and CO2 (3.3 Å) and their molar compositions in the industrial gas streams. We investigated the membrane performances of COFs for H2/CO2 separation at 10 bar and 298 K. Because H2 molecules are lighter than CO2 molecules, they diffuse faster than CO2 through the membrane pores;94 thus, we concentrated on the H2/CO2 selectivity of COF membranes. Membrane selectivity is a product of adsorption and diffusion selectivity; Figure 4 shows the relations between adsorption, diffusion, and membrane selectivities of COFs. The adsorption selectivities for H2 (Sads,H2/CO2) were found to be lower than unity (between 2.3 × 10–3 and 0.42) because of the stronger affinity of COFs to CO2 than to H2. On the other hand, weakly adsorbed and light H2 molecules diffuse faster than CO2 through COF membranes, resulting in H2 diffusion selectivities (Sdiff,H2/CO2) higher than unity, in the range of 2.4–86.9. Figure 4 shows that 88 COFs having H2/CO2 adsorption selectivities between 0.03 and 0.42 became H2-selective membranes (Smem,H2/CO2 > 2) because of their high H2/CO2 diffusion selectivities (9–86.9). For these COFs, H2 diffusion selectivity dominated adsorption selectivity. When we examined 20 COF membranes offering the highest Smem,H2/CO2 (in the range of 3.07–4.74), 19 of them have Sdiff,H2/CO2 in the range of 11.3–18.7. Although it has a moderate Sdiff,H2/CO2 (11.3), CCOF-295 has the highest Smem,H2/CO2 (4.74) among all COFs. The main factor is that it has the highest Sads,H2/CO2 (0.42) because the large pores (an LCD of 22.90 Å) allow the passage and adsorption of both molecules. COF-LZU8 has the highest Sdiff,H2/CO2 (86.9) among all COF membranes. It can be attributed to its functional groups96 which probably hinder the diffusion of CO2 molecules and favor the diffusion of H2 molecules. Low Sads,H2/CO2 of COF-LZU8 led to an Smem,H2/CO2 (3.88) slightly lower than that of CCOF-2. A total of 82 COF membranes exhibited a nonselective behavior because their Smem,H2/CO2 values were found in the range of 0.7–1.5, and neither Sads,H2/CO2 nor Sdiff,H2/CO2 led to high Smem,H2/CO2 for these COFs. Usage of these COFs as adsorbents is more appropriate as they have good Sads,CO2/H2 values, up to ∼40. Another result of Figure 4 is that 9 COFs with the highest CO2 adsorption selectivities (Sads,CO2/H2 > 90) under PSA and PTSA conditions were identified as CO2-selective membranes (Smem,CO2/H2 > 3). The CO2 uptake and Sads,CO2/H2 of these COFs were computed in the range of 1.77–6.64 mol/kg and 90–428 at 10 bar and 298 K, respectively. High CO2 uptake in the pores of COFs causes steric hindrance and blocks the passage of H2 molecules through pores because these COFs have narrow pore sizes (4.11–14.93 Å). As a result, these COFs have limited Sdiff,H2/CO2, and they act as CO2-selective membranes.

Figure 4.

Figure 4

Open in a new tab

Comparison of adsorption, diffusion, and membrane selectivities of COFs computed at 10 bar and 298 K. The black dashed line symbolizes that Smem,H2/CO2 is equal to 1. Red, green, and blue data points represent COF membranes offering high, moderate, and low H2/CO2 diffusion selectivities, respectively.

Figure 5 shows the relation between H2 selectivity and permeability of COF membranes. Robeson’s upper bound17 is also depicted to compare the separation performances of COF membranes with polymeric membranes. The Smem,H2/CO2 values of COFs were calculated in the range of 5.91 × 10–3 to 4.74, and their H2 permeabilities were computed to range from 599 to 1.5 × 106 barrer. The Smem,H2/CO2 and PH2 values of MOF membranes were previously reported in the ranges of 2 × 10–5 to 6.34 and 229 to 1.7 × 106 barrer, respectively, under infinite dilution conditions.55 Although conditions are different since we studied separation of the H2/CO2 mixture at a practical pressure in this work, a comparison of COF membranes with MOFs showed that they have similar performances. Compared to the membrane-based separation performances of polymers (Smem,H2/CO2: 0.1–99.6, PH2: 4.57 × 10–2 to 2.26 × 104 barrer), COFs have a narrower range of Smem,H2/CO2 values but much higher PH2 values. Almost all COFs, 268 out of 288, surpass Robeson’s upper bound, suggesting that COF membranes can perform better than polymeric membranes for H2/CO2 separation. COF membranes also have better performances than various zeolite membranes (NaX, NaY, TON, MFI, CHA, DDR, ERI, and TSC) which have an Smem,H2/CO2 of 0.02–1.02 and a PH2 in the range of 2.26 × 103−4.19 × 104 barrer at 10 bar and 300 K.77 A total of 146 COFs shown by blue in Figure 5 had high permeabilities (PH2 > 3 × 105 barrer), and the top 10 COF membranes were selected to have PH2 > 3 × 105 barrer with the highest Smem,H2/CO2 values as shown with red symbols. Structural features (LCD, PLD, ϕ, and Sacc) and adsorption, diffusion, and membrane selectivities of the top 10 COF membranes are listed in Table S8. All top COF membranes have low CO2 adsorption selectivities (<5) and mediocre H2 diffusion selectivities (11–15). This is expected because adsorption favors CO2, while diffusion favors H2. To obtain high PH2 and Smem,H2/CO2, H2 diffusion selectivities should dominate the CO2 adsorption selectivities. We also note that the most promising COF membranes generally have large pores (LCDs and PLDs > 20 Å) and very porous structures (ϕ > 0.85), as discussed in detail below.

Figure 5.

Figure 5

Open in a new tab

Relationship between H2/CO2 membrane selectivity and H2 permeability of COF membranes. The red dashed line symbolizes that Smem,H2/CO2 is equal to 1, whereas the red solid line indicates the Robeson upper bound for H2/CO2 separation.

3.3. Structure–Performance Relations

The analyses relating structural features to performance metrics help us to reveal which adsorbents/membranes tend to have exceptional performances according to structural parameters and also give molecular insight to experimentalists for synthesizing efficient materials. Therefore, we investigated the relationships between selectivities of COF adsorbents/membranes and their structural properties, ϕ and LCD. We would like to note that all COF structures were assumed to be rigid and stable in the form that was obtained from the database. At low pressure, highly selective COF adsorbents (Sads,CO2/H2 > 90) have a large range of pore sizes (4 Å < LCD < 25 Å) and porosities (0.44 < ϕ < 0.86) as shown in Figure 6a. Because COFs that we examined have larger pores than the size of CO2 and H2, we did not observe any molecular sieving effect. The CO2/H2 selectivities of COFs are mainly determined by the interactions of gas molecules with the adsorption sites of COFs at low pressures rather than the structural features. Contribution of CO2–COF electrostatic interactions to selectivities was also shown at low pressure, as shown in Figure 6b, by switching off the electrostatic interactions between CO2 and the COF. Sads,CO2/H2 of 6 (15) COFs dropped from the range of 90–638 (25–90) to the range of 25–90 (2–25). Although the Coulombic interactions are important especially for some of the highly selective COFs, van der Waals interactions between COF atoms and gases are the main factor that determines Sads,CO2/H2 of most COFs at low pressures. For example, the contribution of van der Waals interactions to host-adsorbate energy is >90% for the majority of COFs (196 among 288 COFs) at 0.1 bar. Because CO2 molecules have three interaction sites, they adsorb more strongly in COFs than H2 molecules which are modeled as single-site molecules. This results in CO2-selective COFs regardless of their pore sizes and porosities at 0.1 bar.

Figure 6.

Figure 6

Open in a new tab

Relationship between porosities, pore sizes, and CO2/H2 adsorption selectivities of COFs at (a) 0.1 bar, (b) 0.1 bar without Coulomb interactions (WoutCl), (c) 1 bar, and (d) 10 bar at 298 K. (e) Relation between porosities (ϕ), LCDs, and H2/CO2 membrane selectivities (Smem,H2/CO2) of COFs. Blue, green, and red points symbolize COFs with low, moderate, and high selectivities, respectively.

Figure 6c,d shows how Sads,CO2/H2 changes with respect to LCD and ϕ at 1 and 10 bar. A significant result is that the number of COFs with high Sads,CO2/H2 at 1 and 10 bar is lower compared to that at 0.1 bar. When pressure increases, adsorbed H2 molecules increase more than the increase in the adsorbed CO2 molecules in pores; therefore, lower Sads,CO2/H2 values are obtained at 1 and 10 bar compared to 0.1 bar. COFs with high Sads,CO2/H2 at 10 bar generally have narrow pores (LCD < 10 Å) and low porosities (ϕ < 0.65). This is because COFs with large pores provide more adsorption space available for both gases, whereas in COFs with narrow pores, CO2 molecules can dominate the available adsorption sites. To get a better understanding of the pressure effect on the CO2 and H2 adsorption, we performed GCMC simulations at an additional pressure (5 bar) and different temperatures (338 and 423 K). The Sads,CO2/H2 values were computed for each temperature–pressure combination, and the results are shown in Figure S6. The highest number of COFs having Sads,CO2/H2 > 100 was obtained at 0.1 bar and 298 K. When the pressure was increased from 0.1 to 1 bar or from 1 to 10 bar at a constant temperature, the Sads,CO2/H2 values dropped. When the pressure was constant and the temperature was increased, low Sads,CO2/H2 values were obtained, and the effect of pressure on the selectivities was found to be negligible at high temperatures (>298 K). As a result, COFs have high Sads,CO2/H2 at low pressures and low temperatures.

Apart from computable structural properties, heat of adsorption (Qst) is also known to be important in determining the gas separation performances of materials. Yang et al.97 proposed an equation, ln S = −0.8558 + ΔQst0/(R × T), to relate adsorption selectivity to the heats of adsorption at infinite dilution. Here, S is the selectivity, ΔQst represents the difference between isosteric heats of adsorption of gases at infinite dilution, R is the ideal gas constant, and T is the temperature. Motivated from this model, we showed the relation between our simulated Sads,CO2/H2 and ΔQst0 values, as shown in Figure S7, where Yang’s model was demonstrated with a red line. We noticed that the Sads,CO2/H2 values of COFs have a wider distribution and they deviate more from the S values of Yang’s model at 1 and 10 bar compared to those at 0.1 bar. This is expected because the model is based on the pore–gas interactions at infinite dilution, and at high pressures, gas–gas interactions play an important role on the adsorption selectivity. Another important pattern is that the agreement between Sads,CO2/H2 and S values was good/moderate when ΔQst ≤ 15 kJ/mol at all pressures, but S significantly overestimated the Sads,CO2/H2 values of COFs for which ΔQst0 > 15 kJ/mol. Overall, for the COFs with ΔQst ≤ 15 kJ/mol, the model can accurately predict selectivities, at 0.1, 1, and 10 bar.

To better understand the correlation between ΔQst0 and pore properties of COFs, we also investigated the relationship between pore sizes and ΔQst, as shown in Figure S8, which was colored according to Sads,CO2/H2. The ΔQst0 values of COFs were computed in the range of 4.7–29.4 kJ/mol. As expected, COFs with small pores (LCD < 15 Å) have high ΔQst values. This trend was previously reported for H2 storage of MOFs in the literature.98 However, in Figure S8, there are many highly selective COFs with large pores and with ΔQst0 > 15 kJ/mol, especially at 0.1 bar. We examined these COFs in more detail to understand what causes the high ΔQst values. For instance, CPF-group COFs99 were computed to have a ΔQst0 of 26 kJ/mol, although they have large pores (LCD > 20 Å). This group of COFs include porphyrin subunits, and it was reported in the literature that these groups favor the CO2 adsorption.100 We realized that several of the COFs with high Sads,CO2/H2, ΔQst ≥ 15 kJ/mol, and large pores (LCD > 15 Å) also have porphyrin subunits.86,101,102 Therefore, high ΔQst0 of these large-pored COFs can be attributed to the existence of subunits favoring adsorption of CO2 such as porphyrins. As Yang’s model correlates the selectivity of adsorbents to the pore size and confinement of the gas molecule in the pores at infinite dilution, it deviates for materials with specific interactions such as the ones having strong electrostatic interactions.97 Therefore, overestimation of the selectivities of some COFs by the model can be attributed to the existence of specific sites that lead to strong adsorption of CO2 molecules.

Finally, structure–performance relations of COF membranes were studied. The relationship between LCD, ϕ, and Smem,H2/CO2 of COFs, which was computed for the H2/CO2: 85/15 mixture, is illustrated in Figure 6e. H2-selective COF membranes mostly have PLDs and LCDs > 20 Å and ϕ > 0.8. As mentioned before in the discussion of Figure 6d, we observed that COFs with narrow pores and low porosities tend to become CO2-selective materials. CO2 molecules were more strongly confined into the narrow pores, and they block the passage of H2 molecules. Therefore, low H2/CO2 membrane selectivities were obtained for these COFs at 10 bar and 298 K. We also observed that COFs consisting of large pores (LCD > 20 Å) and having highly porous (ϕ > 0.8) structures usually have low Sads,CO2/H2 (<25) at 10 bar and 298 K as both gas molecules can easily adsorb. As a result, H2 diffusion dominates CO2 adsorption in these COFs and COFs with highly porous structures tend to be H2-selective membranes.

4. Conclusions

In this work, we uncovered the CO2/H2 mixture separation performances of 288 COFs using a high-throughput computational screening approach for several different adsorption-based processes, VSA, PSA, TSA, VTSA, and PTSA. We showed that while VSA, TSA, and VTSA are attractive processes in terms of offering high adsorption selectivities, the PSA and PTSA processes offer higher working capacities for CO2/H2 separation. Considering the combination of these performance metrics of COFs, the highest APSs accompanied with high R % values for COF adsorbents were obtained under PTSA conditions, revealing the effectiveness of combining pressure and temperature swing for adsorption-based CO2 separation processes. Many COFs were found to have higher adsorption selectivities (2.4–428) and working capacities (0.21–6.82 mol/kg) compared to zeolites, indicating that COFs can replace traditional porous materials in adsorption-based CO2/H2 mixture separation. Structure–performance relations revealed that COFs with low porosities (ϕ < 0.65) and narrow pores (LCD < 10 Å) lead to high Sads,CO2/H2 because these structural features enhance the confinement of CO2 molecules. We also examined membrane-based CO2/H2 mixture separation performances of 288 COFs at 10 bar and 298 K, and the majority of COF membranes (268) exceeded Robeson’s upper bound because of their higher PH2 (599 to 1.51 × 106 barrer) than the conventional polymer membranes. Promising COF membranes for selective separation of H2 from CO2 were computed to have high porosities (ϕ > 0.85) and large pores (LCD > 20 Å). These results indicated that COFs are strong candidates to replace zeolites as adsorbents and to replace polymers as membranes for CO2/H2 separation. As a newly emerged material group, COFs are on the verge of becoming promising options for precombustion CO2 capture, and the results of this high-throughput computational screening study will complement the experimental efforts for selecting the best COFs for adsorption and membrane-based CO2/H2 separation.

Acknowledgments

S.K. acknowledges ERC-2017-Starting Grant. This study has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (ERC-2017-Starting Grant, grant agreement no. 756489-COSMOS).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jpcc.0c07062.

  • Flowchart diagram of the screening procedure for COF sets; comparison of experimental data and simulation results for single-component adsorption isotherms of CO2 for COFs; R %–APS relations of COFs; adsorption selectivities of COFs with and without the Coulomb interactions; snapshots of EB-COF:F with adsorbed CO2 and H2 molecules; comparison of CO2 adsorption uptakes and R % for the VSA and TSA processes; comparison of adsorption selectivities under different conditions; relation between isosteric heats of adsorption and CO2/H2 adsorption selectivities of COFs; relation between pore sizes and isosteric heats of adsorption of COFs; lists of the top-performing COFs; and correlation coefficients for the performance evaluation metrics (PDF)

The authors declare no competing financial interest.

Supplementary Material

jp0c07062_si_001.pdf (1.2MB, pdf)

References

  1. Zito P. F.; Caravella A.; Brunetti A.; Drioli E.; Barbieri G. CO2/H2 Selectivity Prediction of NaY, DD3R, and Silicalite Zeolite Membranes. Ind. Eng. Chem. Res. 2018, 57, 11431–11438. 10.1021/acs.iecr.8b02707. [DOI] [Google Scholar]
  2. Theo W. L.; Lim J. S.; Hashim H.; Mustaffa A. A.; Ho W. S. Review of Pre-Combustion Capture and Ionic Liquid in Carbon Capture and Storage. Appl. Energy 2016, 183, 1633–1663. 10.1016/j.apenergy.2016.09.103. [DOI] [Google Scholar]
  3. Davison J.; Arienti S.; Cotone P.; Mancuso L. Co-Production of Hydrogen and Electricity with CO2 Capture. Int. J. Greenhouse Gas Control 2010, 4, 125–130. 10.1016/j.ijggc.2009.10.007. [DOI] [Google Scholar]
  4. Kenarsari S. D.; Yang D.; Jiang G.; Zhang S.; Wang J.; Russell A. G.; Wei Q.; Fan M. Review of Recent Advances in Carbon Dioxide Separation and Capture. RSC Adv. 2013, 3, 22739–22773. 10.1039/c3ra43965h. [DOI] [Google Scholar]
  5. Mason J. A.; Sumida K.; Herm Z. R.; Krishna R.; Long J. R. Evaluating Metal-Organic Frameworks for Post-Combustion Carbon Dioxide Capture via Temperature Swing Adsorption. Energy Environ. Sci. 2011, 4, 3030–3040. 10.1039/c1ee01720a. [DOI] [Google Scholar]
  6. Sircar S.; Golden T. C. Purification of Hydrogen by Pressure Swing Adsorption. Sep. Sci. Technol. 2000, 35, 667–687. 10.1081/ss-100100183. [DOI] [Google Scholar]
  7. Ben-Mansour R.; Qasem N. A. A. Management An Efficient Temperature Swing Adsorption (TSA) Process for Separating CO2 from CO2/N2 Mixture Using Mg-MOF-74. Energy Convers. Manag. 2018, 156, 10–24. 10.1016/j.enconman.2017.11.010. [DOI] [Google Scholar]
  8. García S.; Gil M. V.; Pis J. J.; Rubiera F.; Pevida C. Cyclic Operation of a Fixed-Bed Pressure and Temperature Swing Process for CO2 Capture: Experimental and Statistical Analysis. Int. J. Greenhouse Gas Control 2013, 12, 35–43. 10.1016/j.ijggc.2012.10.018. [DOI] [Google Scholar]
  9. Xiao P.; Wilson S.; Xiao G.; Singh R.; Webley P. Novel Adsorption Processes for Carbon Dioxide Capture within a IGCC Process. Energy Procedia 2009, 1, 631–638. 10.1016/j.egypro.2009.01.083. [DOI] [Google Scholar]
  10. Wang L.; Liu Z.; Li P.; Yu J.; Rodrigues A. E. Experimental and Modeling Investigation on Post-Combustion Carbon Dioxide Capture Using Zeolite 13X-APG by Hybrid VTSA Process. Chem. Eng. J. 2012, 197, 151–161. 10.1016/j.cej.2012.05.017. [DOI] [Google Scholar]
  11. Hufton J. R.; Mayorga S.; Sircar S. Sorption-Enhanced Reaction Process for Hydrogen Production. AIChE J. 1999, 45, 248–256. 10.1002/aic.690450205. [DOI] [Google Scholar]
  12. Kamakoti P.; Leta D. P.; Deckman H. W.; Ravikovitch P. I.; Anderson T. N.. Pressure-Temperature Swing Adsorption Process. U.S. Patent 8,784,534B2, 2014.
  13. Yu J.; Xie L.-H.; Li J.-R.; Ma Y.; Seminario J. M.; Balbuena P. B. CO2 Capture and Separations Using MOFs: Computational and Experimental Studies. Chem. Rev. 2017, 117, 9674–9754. 10.1021/acs.chemrev.6b00626. [DOI] [PubMed] [Google Scholar]
  14. Scholes C. A.; Smith K. H.; Kentish S. E.; Stevens G. W. CO2 Capture from Pre-Combustion Processes—Strategies for Membrane Gas Separation. Int. J. Greenhouse Gas Control 2010, 4, 739–755. 10.1016/j.ijggc.2010.04.001. [DOI] [Google Scholar]
  15. Yampolskii Y. Polymeric Gas Separation Membranes. Macromolecules 2012, 45, 3298–3311. 10.1021/ma300213b. [DOI] [Google Scholar]
  16. Sanders D. F.; Smith Z. P.; Guo R.; Robeson L. M.; McGrath J. E.; Paul D. R.; Freeman B. D. Energy-Efficient Polymeric Gas Separation Membranes for a Sustainable Future: A Review. Polymer 2013, 54, 4729–4761. 10.1016/j.polymer.2013.05.075. [DOI] [Google Scholar]
  17. Robeson L. M. The Upper Bound Revisited. J. Membr. Sci. 2008, 320, 390–400. 10.1016/j.memsci.2008.04.030. [DOI] [Google Scholar]
  18. Korelskiy D.; Ye P.; Fouladvand S.; Karimi S.; Sjöberg E.; Hedlund J. Efficient Ceramic Zeolite Membranes for CO2/H2 Separation. J. Mater. Chem. A 2015, 3, 12500–12506. 10.1039/c5ta02152a. [DOI] [Google Scholar]
  19. Caro J.; Noack M.. Zeolite Membranes–Status and Prospective. Advances in Nanoporous Materials; Elsevier: Oxford, 2010; Vol. 1, pp 1–96. [Google Scholar]
  20. Li J.-R.; Sculley J.; Zhou H.-C. Metal-Organic Frameworks for Separations. Chem. Rev. 2012, 112, 869–932. 10.1021/cr200190s. [DOI] [PubMed] [Google Scholar]
  21. Kang Z.; Fan L.; Sun D. Recent Advances and Challenges of Metal–Organic Framework Membranes for Gas Separation. J. Mater. Chem. A 2017, 5, 10073–10091. 10.1039/c7ta01142c. [DOI] [Google Scholar]
  22. Banerjee D.; Simon C. M.; Plonka A. M.; Motkuri R. K.; Liu J.; Chen X.; Smit B.; Parise J. B.; Haranczyk M.; Thallapally P. K. Metal–Organic Framework with Optimally Selective Xenon Adsorption and Separation. Nat. Commun. 2016, 7, 11831. 10.1038/ncomms11831. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lin R.-B.; Li L.; Zhou H.-L.; Wu H.; He C.; Li S.; Krishna R.; Li J.; Zhou W.; Chen B. Molecular Sieving of Ethylene from Ethane Using a Rigid Metal–Organic Framework. Nat. Mater. 2018, 17, 1128–1133. 10.1038/s41563-018-0206-2. [DOI] [PubMed] [Google Scholar]
  24. Cote A. P.; Benin A. I.; Ockwig N. W.; O’Keeffe M.; Matzger A. J.; Yaghi O. M. Porous, Crystalline, Covalent Organic Frameworks. Science 2005, 310, 1166–1170. 10.1126/science.1120411. [DOI] [PubMed] [Google Scholar]
  25. El-Kaderi H. M.; Hunt J. R.; Mendoza-Cortes J. L.; Cote A. P.; Taylor R. E.; O’Keeffe M.; Yaghi O. M. Designed Synthesis of 3D Covalent Organic Frameworks. Science 2007, 316, 268–272. 10.1126/science.1139915. [DOI] [PubMed] [Google Scholar]
  26. Zeng Y.; Zou R.; Zhao Y. Covalent Organic Frameworks for CO2 Capture. Adv. Mater. 2016, 28, 2855–2873. 10.1002/adma.201505004. [DOI] [PubMed] [Google Scholar]
  27. Furukawa H.; Yaghi O. M. Storage of Hydrogen, Methane, and Carbon Dioxide in Highly Porous Covalent Organic Frameworks for Clean Energy Applications. J. Am. Chem. Soc. 2009, 131, 8875–8883. 10.1021/ja9015765. [DOI] [PubMed] [Google Scholar]
  28. Yu J.-T.; Chen Z.; Sun J.; Huang Z.-T.; Zheng Q.-Y. Cyclotricatechylene Based Porous Crystalline Material: Synthesis and Applications in Gas Storage. J. Mater. Chem. 2012, 22, 5369–5373. 10.1039/c2jm15159f. [DOI] [Google Scholar]
  29. Li Y.; Yang R. T. Hydrogen Storage in Metal-Organic and Covalent-Organic Frameworks by Spillover. AIChE J. 2008, 54, 269–279. 10.1002/aic.11362. [DOI] [Google Scholar]
  30. Rabbani M. G.; Sekizkardes A. K.; Kahveci Z.; Reich T. E.; Ding R.; El-Kaderi H. M. A 2D Mesoporous Imine-Linked Covalent Organic Framework for High Pressure Gas Storage Applications. Chem.—Eur. J. 2013, 19, 3324–3328. 10.1002/chem.201203753. [DOI] [PubMed] [Google Scholar]
  31. Han S. S.; Furukawa H.; Yaghi O. M.; Goddard W. A. Covalent Organic Frameworks as Exceptional Hydrogen Storage Materials. J. Am. Chem. Soc. 2008, 130, 11580–11581. 10.1021/ja803247y. [DOI] [PubMed] [Google Scholar]
  32. Garberoglio G. Computer Simulation of the Adsorption of Light Gases in Covalent Organic Frameworks. Langmuir 2007, 23, 12154–12158. 10.1021/la701736m. [DOI] [PubMed] [Google Scholar]
  33. Keskin S. Adsorption, Diffusion, and Separation of CH4/H2 Mixtures in Covalent Organic Frameworks: Molecular Simulations and Theoretical Predictions. J. Phys. Chem. C 2012, 116, 1772–1779. 10.1021/jp209804x. [DOI] [Google Scholar]
  34. Babarao R.; Jiang J. Exceptionally High CO2 Storage in Covalent-Organic Frameworks: Atomistic Simulation Study. Energy Environ. Sci. 2008, 1, 139–143. 10.1039/b805473h. [DOI] [Google Scholar]
  35. Yang Q.; Zhong C. Molecular Simulation Study of the Stepped Behaviors of Gas Adsorption in Two-Dimensional Covalent Organic Frameworks. Langmuir 2009, 25, 2302–2308. 10.1021/la8035902. [DOI] [PubMed] [Google Scholar]
  36. Zhang M.; Zheng R.; Ma Y.; Chen R.; Sun X.; Sun X. N-Rich Covalent Organic Frameworks with Different Pore Size for High-Pressure CO2 Adsorption. Microporous Mesoporous Mater. 2019, 285, 70–79. 10.1016/j.micromeso.2019.04.021. [DOI] [Google Scholar]
  37. Zeng H.; Liu Y.; Liu H. Adsorption and Diffusion of CO2 and CH4 in Covalent Organic Frameworks: an MC/MD Simulation Study. Mol. Simul. 2018, 44, 1244–1251. 10.1080/08927022.2018.1481959. [DOI] [Google Scholar]
  38. Wang L.; Dong B.; Ge R.; Jiang F.; Xiong J.; Gao Y.; Xu J. A Thiadiazole-Functionalized Covalent Organic Framework for Efficient CO2 Capture and Separation. Microporous Mesoporous Mater. 2016, 224, 95–99. 10.1016/j.micromeso.2015.11.030. [DOI] [Google Scholar]
  39. Liu Y.; Liu D.; Yang Q.; Zhong C.; Mi J. Comparative Study of Separation Performance of COFs and MOFs for CH4/CO2/H2 Mixtures. Ind. Eng. Chem. Res. 2010, 49, 2902–2906. 10.1021/ie901488f. [DOI] [Google Scholar]
  40. Vicent-Luna J. M.; Luna-Triguero A.; Calero S. Storage and Separation of Carbon Dioxide and Methane in Hydrated Covalent Organic Frameworks. J. Phys. Chem. C 2016, 120, 23756–23762. 10.1021/acs.jpcc.6b05233. [DOI] [Google Scholar]
  41. Ge Y.; Zhou H.; Ji Y.; Ding L.; Cheng Y.; Wang R.; Yang S.; Liu Y.; Wu X.; Li Y. Understanding Water Adsorption and the Impact on CO2 Capture in Chemically Stable Covalent Organic Frameworks. J. Phys. Chem. C 2018, 122, 27495–27506. 10.1021/acs.jpcc.8b09033. [DOI] [Google Scholar]
  42. Tong M.; Yang Q.; Zhong C. Computational Screening of Covalent Organic Frameworks for CH4/H2, CO2/H2 and CO2/CH4 Separations. Microporous Mesoporous Mater. 2015, 210, 142–148. 10.1016/j.micromeso.2015.02.034. [DOI] [Google Scholar]
  43. Krishna R.; van Baten J. M. Maxwell–Stefan Modeling of Slowing-Down Effects in Mixed Gas Permeation across Porous Membranes. J. Membr. Sci. 2011, 383, 289–300. 10.1016/j.memsci.2011.08.067. [DOI] [Google Scholar]
  44. Tong M.; Zhang Y.; Yan T.; Chen G.; Long Z.; Qin Z.; Yang Q.; Zhong C. Computational Insights on the Role of Nanochannel Environment in the CO2/CH4 and H2/CH4 Separation Using Restacked Covalent Organic Framework Membranes. J. Phys. Chem. C 2019, 123, 22949–22958. 10.1021/acs.jpcc.9b05183. [DOI] [Google Scholar]
  45. Tong M.; Lan Y.; Yang Q.; Zhong C. Exploring the Structure-Property Relationships of Covalent Organic Frameworks for Noble Gas Separations. Chem. Eng. Sci. 2017, 168, 456–464. 10.1016/j.ces.2017.05.004. [DOI] [Google Scholar]
  46. Tong M.; Lan Y.; Qin Z.; Zhong C. Computation-Ready, Experimental Covalent Organic Framework for Methane Delivery: Screening and Material Design. J. Phys. Chem. C 2018, 122, 13009–13016. 10.1021/acs.jpcc.8b04742. [DOI] [Google Scholar]
  47. Yan T.; Lan Y.; Tong M.; Zhong C. Screening and Design of Covalent Organic Framework Membranes for CO2/CH4 Separation. ACS Sustainable Chem. Eng. 2018, 7, 1220–1227. 10.1021/acssuschemeng.8b04858. [DOI] [Google Scholar]
  48. Ongari D.; Yakutovich A. V.; Talirz L.; Smit B. Building a Consistent and Reproducible Database for Adsorption Evaluation in Covalent-Organic Frameworks. ACS Cent. Sci. 2019, 5, 1663–1675. 10.1021/acscentsci.9b00619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Gülçay E.; Findikci İ. E. Computational Screening of Covalent Organic Frameworks for Hydrogen Storage. J. Turk. Chem. Soc., Sect. A 2020, 7, 65–76. 10.18596/jotcsa.565460. [DOI] [Google Scholar]
  50. Gulcay E.; Erucar I. Molecular Simulations of COFs, IRMOFs and ZIFs for Adsorption-based Separation of Carbon Tetrachloride from Air. J. Mol. Graphics Modell. 2019, 86, 84–94. 10.1016/j.jmgm.2018.10.006. [DOI] [PubMed] [Google Scholar]
  51. Altundal O. F.; Altintas C.; Keskin S. Can COFs Replace MOFs in Flue Gas Separation? High-Throughput Computational Screening of COFs for CO2/N2 Separation. J. Mater. Chem. A 2020, 8, 14609–14623. 10.1039/d0ta04574h. [DOI] [Google Scholar]
  52. Mercado R.; Fu R.-S.; Yakutovich A. V.; Talirz L.; Haranczyk M.; Smit B. In Silico Design of 2D and 3D Covalent Organic Frameworks for Methane Storage Applications. Chem. Mater. 2018, 30, 5069–5086. 10.1021/acs.chemmater.8b01425. [DOI] [Google Scholar]
  53. Martin R. L.; Simon C. M.; Medasani B.; Britt D. K.; Smit B.; Haranczyk M. In Silico Design of Three-Dimensional Porous Covalent Organic Frameworks via Known Synthesis Routes and Commercially Available Species. J. Phys. Chem. C 2014, 118, 23790–23802. 10.1021/jp507152j. [DOI] [Google Scholar]
  54. Deeg K. S.; Damasceno Borges D.; Ongari D.; Rampal N.; Talirz L.; Yakutovich A. V.; Huck J. M.; Smit B. In Silico Discovery of Covalent Organic Frameworks for Carbon Capture. ACS Appl. Mater. Interfaces 2020, 12, 21559–21568. 10.1021/acsami.0c01659. [DOI] [PubMed] [Google Scholar]
  55. Avci G.; Velioglu S.; Keskin S. High-Throughput Screening of MOF Adsorbents and Membranes for H2 Purification and CO2 Capture. ACS Appl. Mater. Interfaces 2018, 10, 33693–33706. 10.1021/acsami.8b12746. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Willems T. F.; Rycroft C. H.; Kazi M.; Meza J. C.; Haranczyk M. Algorithms and Tools for High-Throughput Geometry-Based Analysis of Crystalline Porous Materials. Microporous Mesoporous Mater. 2012, 149, 134–141. 10.1016/j.micromeso.2011.08.020. [DOI] [Google Scholar]
  57. Xu S.-Q.; Liang R.-R.; Zhan T.-G.; Qi Q.-Y.; Zhao X. Construction of 2D Covalent Organic Frameworks by Taking Advantage of the Variable Orientation of Imine Bonds. Chem. Commun. 2017, 53, 2431–2434. 10.1039/c6cc09906h. [DOI] [PubMed] [Google Scholar]
  58. Dubbeldam D.; Calero S.; Ellis D. E.; Snurr R. Q. RASPA: Molecular Simulation Software for Adsorption and Diffusion in Flexible Nanoporous Materials. Mol. Simul. 2016, 42, 81–101. 10.1080/08927022.2015.1010082. [DOI] [Google Scholar]
  59. Lin S.; Harada M.; Suzuki Y.; Hatano H. Hydrogen Production from Coal by Separating Carbon Dioxide During Gasification. Fuel 2002, 81, 2079–2085. 10.1016/s0016-2361(02)00187-4. [DOI] [Google Scholar]
  60. Buch V. Path Integral Simulations of Mixed Para-D2 and Ortho-D2 Clusters: The Orientational Effects. J. Chem. Phys. 1994, 100, 7610–7629. 10.1063/1.466854. [DOI] [Google Scholar]
  61. Potoff J. J.; Siepmann J. I. Vapor-Liquid Equilibria of Mixtures Containing Alkanes, Carbon Dioxide, and Nitrogen. AIChE J. 2001, 47, 1676–1682. 10.1002/aic.690470719. [DOI] [Google Scholar]
  62. Wilmer C. E.; Snurr R. Q. Towards Rapid Computational Screening of Metal-Organic Frameworks for Carbon Dioxide Capture: Calculation of Framework Charges via Charge Equilibration. Chem. Eng. J. 2011, 171, 775–781. 10.1016/j.cej.2010.10.035. [DOI] [Google Scholar]
  63. Ewald P. P. Evaluation of Optical and Electrostatic Lattice Potentials. Ann. Phys. 1921, 369, 253–287. 10.1002/andp.19213690304. [DOI] [Google Scholar]
  64. Frenkel D.; Smit B.. Understanding Molecular Simulation: From Algorithms to Applications; Elsevier: Amsterdam, 2001; Vol. 1. [Google Scholar]
  65. Evans D. J.; Holian B. L. The Nose-Hoover Thermostat. J. Chem. Phys. 1985, 83, 4069–4074. 10.1063/1.449071. [DOI] [Google Scholar]
  66. Einstein A.Investigations on the Theory of the Brownian Movement; Courier Corporation: New York, 1956. [Google Scholar]
  67. Fang H.; Kulkarni A.; Kamakoti P.; Awati R.; Ravikovitch P. I.; Sholl D. S. Identification of High-CO2-Capacity Cationic Zeolites by Accurate. Chem. Mater. 2016, 28, 3887–3896. 10.1021/acs.chemmater.6b01132. [DOI] [Google Scholar]
  68. Chung Y. G.; Gómez-Gualdrón D. A.; Li P.; Leperi K. T.; Deria P.; Zhang H.; Vermeulen N. A.; Stoddart J. F.; You F.; Hupp J. T. In Silico Discovery of Metal-Organic Frameworks for Precombustion CO2 Capture Using a Genetic Algorithm. Sci. Adv. 2016, 2, e1600909 10.1126/sciadv.1600909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Bae Y.-S.; Snurr R. Q. Development and Evaluation of Porous Materials for Carbon Dioxide Separation and Capture. Angew. Chem., Int. Ed. 2011, 50, 11586–11596. 10.1002/anie.201101891. [DOI] [PubMed] [Google Scholar]
  70. Krishna R.; van Baten J. M. In Silico Screening of Zeolite Membranes for CO2 Capture. J. Membr. Sci. 2010, 360, 323–333. 10.1016/j.memsci.2010.05.032. [DOI] [Google Scholar]
  71. Raganati F.; Chirone R.; Ammendola P. CO2 Capture by Temperature Swing Adsorption: Working Capacity as Affected by Temperature and CO2 Partial Pressure. Ind. Eng. Chem. Res. 2020, 59, 3593–3605. 10.1021/acs.iecr.9b04901. [DOI] [Google Scholar]
  72. Gao Q.; Li X.; Ning G.-H.; Xu H.-S.; Liu C.; Tian B.; Tang W.; Loh K. P. Covalent Organic Framework with Frustrated Bonding Network for Enhanced Carbon Dioxide Storage. Chem. Mater. 2018, 30, 1762–1768. 10.1021/acs.chemmater.8b00117. [DOI] [Google Scholar]
  73. Chandra S.; Kandambeth S.; Biswal B. P.; Lukose B.; Kunjir S. M.; Chaudhary M.; Babarao R.; Heine T.; Banerjee R. Chemically Stable Multilayered Covalent Organic Nanosheets from Covalent Organic Frameworks via Mechanical Delamination. J. Am. Chem. Soc. 2013, 135, 17853–17861. 10.1021/ja408121p. [DOI] [PubMed] [Google Scholar]
  74. Ma H.; Liu B.; Li B.; Zhang L.; Li Y.-G.; Tan H.-Q.; Zang H.-Y.; Zhu G. Cationic Covalent Organic Frameworks: A Simple Platform of Anionic Exchange for Porosity Tuning and Proton Conduction. J. Am. Chem. Soc. 2016, 138, 5897–5903. 10.1021/jacs.5b13490. [DOI] [PubMed] [Google Scholar]
  75. Beaudoin D.; Maris T.; Wuest J. D. Constructing Monocrystalline Covalent Organic Networks by Polymerization. Nat. Chem. 2013, 5, 830–834. 10.1038/nchem.1730. [DOI] [PubMed] [Google Scholar]
  76. Dureckova H.; Krykunov M.; Aghaji M. Z.; Woo T. K. Robust Machine Learning Models for Predicting High CO2 Working Capacity and CO2/H2 Selectivity of Gas Adsorption in Metal Organic Frameworks for Precombustion Carbon Capture. J. Phys. Chem. C 2019, 123, 4133–4139. 10.1021/acs.jpcc.8b10644. [DOI] [Google Scholar]
  77. Krishna R.; van Baten J. M. In Silico Screening of Metal–Organic Frameworks in Separation Applications. Phys. Chem. Chem. Phys. 2011, 13, 10593–10616. 10.1039/c1cp20282k. [DOI] [PubMed] [Google Scholar]
  78. Uribe-Romo F. J.; Doonan C. J.; Furukawa H.; Oisaki K.; Yaghi O. M. Crystalline Covalent Organic Frameworks with Hydrazone Linkages. J. Am. Chem. Soc. 2011, 133, 11478–11481. 10.1021/ja204728y. [DOI] [PubMed] [Google Scholar]
  79. Thompson C. M.; Occhialini G.; McCandless G. T.; Alahakoon S. B.; Cameron V.; Nielsen S. O.; Smaldone R. A. Computational and Experimental Studies on the Effects of Monomer Planarity on Covalent Organic Framework Formation. J. Am. Chem. Soc. 2017, 139, 10506–10513. 10.1021/jacs.7b05555. [DOI] [PubMed] [Google Scholar]
  80. Li Z.; Zhang Y.; Xia H.; Mu Y.; Liu X. A Robust and Luminescent Covalent Organic Framework as a Highly Sensitive and Selective Sensor for the Detection of Cu2+ Ions. Chem. Commun. 2016, 52, 6613–6616. 10.1039/c6cc01476c. [DOI] [PubMed] [Google Scholar]
  81. Segura J. L.; Royuela S.; Mar Ramos M. Post-Synthetic Modification of Covalent Organic Frameworks. Chem. Soc. Rev. 2019, 48, 3903–3945. 10.1039/c8cs00978c. [DOI] [PubMed] [Google Scholar]
  82. Lu Q.; Ma Y.; Li H.; Guan X.; Yusran Y.; Xue M.; Fang Q.; Yan Y.; Qiu S.; Valtchev V. Postsynthetic Functionalization of Three-Dimensional Covalent Organic Frameworks for Selective Extraction of Lanthanide Ions. Angew. Chem., Int. Ed. 2018, 130, 6150–6156. 10.1002/ange.201712246. [DOI] [PubMed] [Google Scholar]
  83. Zhao S.; Dong B.; Ge R.; Wang C.; Song X.; Ma W.; Wang Y.; Hao C.; Guo X.; Gao Y. Channel-Wall Functionalization in Covalent Organic Frameworks for the Enhancement of CO2 Uptake and CO2/N2 Selectivity. RSC Adv. 2016, 6, 38774–38781. 10.1039/c6ra04859e. [DOI] [Google Scholar]
  84. Huang N.; Wang P.; Addicoat M. A.; Heine T.; Jiang D. Ionic Covalent Organic Frameworks: Design of a Charged Interface Aligned on 1D Channel Walls and Its Unusual Electrostatic Functions. Angew. Chem., Int. Ed. 2017, 56, 4982–4986. 10.1002/anie.201611542. [DOI] [PubMed] [Google Scholar]
  85. He S.; Rong Q.; Niu H.; Cai Y. Construction of a Superior Visible-Light-Driven Photocatalyst Based on a C3N4 Active Centre-Photoelectron Shift Platform-Electron Withdrawing Unit Triadic Structure Covalent Organic Framework. Chem. Commun. 2017, 53, 9636–9639. 10.1039/c7cc04515h. [DOI] [PubMed] [Google Scholar]
  86. Zhang C.; Zhang S.; Yan Y.; Xia F.; Huang A.; Xian Y. Highly Fluorescent Polyimide Covalent Organic Nanosheets as Sensing Probes for the Detection of 2,4,6-trinitrophenol. ACS Appl. Mater. Interfaces 2017, 9, 13415–13421. 10.1021/acsami.6b16423. [DOI] [PubMed] [Google Scholar]
  87. Nath B.; Li W.-H.; Huang J.-H.; Wang G.-E.; Fu Z.-h.; Yao M.-S.; Xu G. A New Azodioxy-Linked Porphyrin-Based Semiconductive Covalent Organic Framework with I2 Doping-Enhanced Photoconductivity. CrystEngComm 2016, 18, 4259–4263. 10.1039/c6ce00168h. [DOI] [Google Scholar]
  88. Neti V. S. P. K.; Wu X.; Deng S.; Echegoyen L. Synthesis of a Phthalocyanine and Porphyrin 2D Covalent Organic Framework. CrystEngComm 2013, 15, 6892–6895. 10.1039/c3ce40706c. [DOI] [Google Scholar]
  89. Dokur D.; Keskin S. Effects of Force Field Selection on the Computational Ranking of MOFs for CO2 Separations. Ind. Eng. Chem. Res. 2018, 57, 2298–2309. 10.1021/acs.iecr.7b04792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  90. Krishna R. Screening Metal–Organic Frameworks for Mixture Separations in Fixed-Bed Adsorbers Using a Combined Selectivity/Capacity Metric. RSC Adv. 2017, 7, 35724–35737. 10.1039/c7ra07363a. [DOI] [Google Scholar]
  91. Park J.; Rubiera Landa H. O.; Kawajiri Y.; Realff M. J.; Lively R. P.; Sholl D. S. How Well Do Approximate Models of Adsorption-Based CO2 Capture Processes Predict Results of Detailed Process Models?. Ind. Eng. Chem. Res. 2019, 59, 7097–7108. 10.1021/acs.iecr.9b05363. [DOI] [Google Scholar]
  92. Daglar H.; Keskin S. Recent Advances, Opportunities, and Challenges in High-Throughput Computational Screening of MOFs for Gas Separations. Coord. Chem. Rev. 2020, 422, 213470. 10.1016/j.ccr.2020.213470. [DOI] [Google Scholar]
  93. Altintas C.; Keskin S. Role of Partial Charge Assignment Methods in High-throughput Screening of MOF Adsorbents and Membranes for CO2/CH4 Separation. Mol. Syst. Des. Eng. 2020, 5, 532–543. 10.1039/c9me00163h. [DOI] [Google Scholar]
  94. Baker R. W.Membrane Technology and Applications; John Wiley & Sons: California, 2012. [Google Scholar]
  95. Wang X.; Han X.; Zhang J.; Wu X.; Liu Y.; Cui Y. Homochiral 2D Porous Covalent Organic Frameworks for Heterogeneous Asymmetric Catalysis. J. Am. Chem. Soc. 2016, 138, 12332–12335. 10.1021/jacs.6b07714. [DOI] [PubMed] [Google Scholar]
  96. Ding S.-Y.; Dong M.; Wang Y.-W.; Chen Y.-T.; Wang H.-Z.; Su C.-Y.; Wang W. Thioether-Based Fluorescent Covalent Organic Framework for Selective Detection and Facile Removal of Mercury(II). J. Am. Chem. Soc. 2016, 138, 3031–3037. 10.1021/jacs.5b10754. [DOI] [PubMed] [Google Scholar]
  97. Yang Z.; Peng X.; Cao D. Carbon Dioxide Capture by PAFs and an Efficient Strategy to Fast Screen Porous Materials for Gas Separation. J. Phys. Chem. C 2013, 117, 8353–8364. 10.1021/jp402488r. [DOI] [Google Scholar]
  98. Frost H.; Snurr R. Q. Design Requirements for Metal-organic Frameworks as Hydrogen Storage Materials. J. Phys. Chem. C 2007, 111, 18794–18803. 10.1021/jp076657p. [DOI] [Google Scholar]
  99. Zhang W.; Jiang P.; Wang Y.; Zhang J.; Zhang P. Bottom-Up Approach to Engineer Two Covalent Porphyrinic Frameworks as Effective Catalysts for Selective Oxidation. Catal. Sci. Technol. 2015, 5, 101–104. 10.1039/c4cy00969j. [DOI] [Google Scholar]
  100. Kumar S.; Wani M. Y.; Arranja C. T.; e Silva J. d. A.; Avula B.; Sobral A. J. F. N. Porphyrins as Nanoreactors in the Carbon Dioxide Capture and Conversion: A Review. J. Mater. Chem. A 2015, 3, 19615–19637. 10.1039/c5ta05082k. [DOI] [Google Scholar]
  101. Huang N.; Krishna R.; Jiang D. Tailor-Made Pore Surface Engineering in Covalent Organic Frameworks: Systematic Functionalization for Performance Screening. J. Am. Chem. Soc. 2015, 137, 7079–7082. 10.1021/jacs.5b04300. [DOI] [PubMed] [Google Scholar]
  102. Yang H.; Zhang S.; Han L.; Zhang Z.; Xue Z.; Gao J.; Li Y.; Huang C.; Yi Y.; Liu H.; Li Y. High Conductive Two-Dimensional Covalent Organic Framework for Lithium Storage with Large Capacity. ACS Appl. Mater. Interfaces 2016, 8, 5366–5375. 10.1021/acsami.5b12370. [DOI] [PubMed] [Google Scholar]
  103. Zhao R.; Zhao L.; Deng S.; Song C.; He J.; Shao Y.; Li S. A Comparative Study on CO2 Capture Performance of Vacuum-Pressure Swing Adsorption and Pressure-Temperature Swing Adsorption Based on Carbon Pump Cycle. Energy 2017, 137, 495–509. 10.1016/j.energy.2017.01.158. [DOI] [Google Scholar]
  104. Su F.; Lu C. CO2 Capture from Gas Stream by Zeolite 13X Using a Dual-Column Temperature/Vacuum Swing Adsorption. Energy Environ. Sci. 2012, 5, 9021–9027. 10.1039/c2ee22647b. [DOI] [Google Scholar]
  105. Su F.; Lu C.; Chung A.-J.; Liao C.-H. CO2 Capture with Amine-Loaded Carbon Nanotubes via a Dual-Column Temperature/Vacuum Swing Adsorption. Appl. Energy 2014, 113, 706–712. 10.1016/j.apenergy.2013.08.001. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

jp0c07062_si_001.pdf (1.2MB, pdf)

Articles from The Journal of Physical Chemistry. C, Nanomaterials and Interfaces are provided here courtesy of American Chemical Society

RESOURCES