Next Article in Journal
Use of Gas Chromatography-Mass Spectrometry Techniques (GC-MS, GC-MS/MS and GC-QTOF) for the Characterization of Photooxidation and Autoxidation Products of Lipids of Autotrophic Organisms in Environmental Samples
Next Article in Special Issue
Enhancing the Performance of Evolutionary Algorithm by Differential Evolution for Optimizing Distillation Sequence
Previous Article in Journal
Oxazolidine Nitroxide Transformation in a Coordination Sphere of the Ln3+ Ions
Previous Article in Special Issue
Efficient Estimation of Permeate Flux of Asymmetric Ceramic Membranes for Vacuum Membrane Distillation
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Impact of Impure Gas on CO2 Capture from Flue Gas Using Carbon Nanotubes: A Molecular Simulation Study

1
Key Laboratory of Low-Grade Energy Utilization Technologies and Systems, Ministry of Education, School of Energy and Power Engineering, Chongqing University, Chongqing 400044, China
2
State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing 210023, China
3
State Key Laboratory of Materials-Oriented Chemical Engineering, College of Chemical Engineering, Nanjing Tech University, Nanjing 211816, China
*
Authors to whom correspondence should be addressed.
Molecules 2022, 27(5), 1627; https://doi.org/10.3390/molecules27051627
Submission received: 14 December 2021 / Revised: 26 January 2022 / Accepted: 31 January 2022 / Published: 1 March 2022
(This article belongs to the Special Issue Exploration of the Separation Processes in Nanomaterials)

Abstract

:
We used a grand canonical Monte Carlo simulation to study the influence of impurities including water vapor, SO2, and O2 in the flue gas on the adsorption of CO2/N2 mixture in carbon nanotubes (CNTs) and carboxyl doped CNT arrays. In the presence of single impure gas, SO2 yielded the most inhibitions on CO2 adsorption, while the influence of water only occurred at low pressure limit (0.1 bar), where a one-dimensional chain of hydrogen-bonded molecules was formed. Further, O2 was found to hardly affect the adsorption and separation of CO2. With three impurities in flue gas, SO2 still played a major role to suppress the adsorption of CO2 by reducing the adsorption amount significantly. This was mainly because SO2 had a stronger interaction with carbon walls in comparison with CO2. The presence of three impurities in flue gas enhanced the adsorption complexity due to the interactions between different species. Modified by hydrophilic carboxyl groups, a large amount of H2O occupied the adsorption space outside the tube in the carbon nanotube arrays, and SO2 produced competitive adsorption for CO2 in the tube. Both of the two effects inhibited the adsorption of CO2, but improved the selectivity of CO2/N2, and the competition between the two determined the adsorption distribution of CO2 inside and outside the tube. In addition, it was found that (7, 7) CNT always maintained the best CO2/N2 adsorption and separation performance in the presence of impurity gas, for both the cases of single CNT and CNT array.

1. Introduction

Carbon capture and storage (CCS) [1] technologies have been extensively developed to minimized the influence of CO2 emission on the global warming effect. Among the separation techniques, adsorption separation [2] is regarded as a promising solution for its low cost and high efficiency. In this connection, a host of conventional and emerging nanoporous materials have been invented and explored, including zeolites, activated carbons, metal-organic frameworks (MOFs), and carbon nanotubes (CNTs) [3,4,5,6,7,8]. Particularly, CNTs possess large specific surface areas (greater than 1000 m2/g) with strong adsorptive affinities, which could be paired with the superior transport properties to further facilitate the adsorption potentials of [3,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24] CNTs for CO2 capture.
In our previous study, grand canonical Monte Carlo (GCMC) simulations were conducted to investigate the adsorption of CO2 in the internal space of individual single CNTs in the presence of pre-adsorbed water [3]. It was found that the pre-loaded water provided additional H2O–CO2 interactions to facilitate the adsorption of CO2, by taking up the adsorption site available for CO2. Similarly, as reported by Yu et al. [1], the presence of SO2 in the gas phase exerted a negative effect on the adsorption of CO2 for CO2/N2/SO2 mixture in HKUST-1 at ambient temperature. By comparison, the presence of O2 exerted little effect on the adsorption of CO2 in HKUST-1. The main components of flue gases generated by coal-fired power plants include N2 (about 73–77%), CO2 (15–16%), H2O (5–7%), O2 (about 3–4%) [12], and trace amounts of SO2, etc. [25,26]. Therefore, the impurity gases, such as H2O, O2, and SO2 are expected to impose a significant influence on the adsorption and separation of CO2 from flue gas using CNTs [1,3,27,28,29,30].
In practice, oxidation defects often occurred during the acidic/oxidative purification of carbon nanotubes [31], where oxygen-containing functional groups (mainly carbonyl and carboxyl) could be grafted to the defect sites [32]. The oxygen-containing functional groups, such as carboxyl and hydroxyl groups, are hydrophilic, so it could significantly enhance the adsorption of water vapor contained in flue gas, which thus imposes strong influence on the adsorption in CNTs for the rest components in flue gas [33,34]. Further, instead of single carbon nanotube, carbon nanotube bundles were generally obtained during the synthesis procedure. Therefore, to explore the influence of impurity gases on the adsorption and separation of CO2 from flue gas in a practical manner, the adsorption of gas mixtures (CO2/N2/X, X denotes the impurity gases, H2O, SO2, and O2) in the functionalized CNT bundles are required. To the best of our knowledge, the adsorption behavior of impurity gases in the functionalized CNT bundles is still unknown. Hence, the effects of three impurity gases on the separation of CO2 in CNT also have not been systematically studied. Different from binary mixture, there are more complex interactions between three impurity gases, the cooperative impact on CO2 adsorption has hardly been studied. In addition, little is known about how the cooperative effects between adsorbate-CNT interaction and interaction between impurity and adsorbate affect CO2/N2 selectivity. Discussions related to these and other related issues will be obtained in detail in this work. Furthermore, the influence on the optimum diameter of CNTs for separating CO2/N2 is not reported yet.
In this work, GCMC and density functional theory (DFT) simulations were conducted to investigate the adsorption separation of CO2 from flue gases using carbon nanotubes in the presence of impurity species (H2O, O2 and SO2), in order to fundamentally reveal the impacts of impurity gases on the adsorption behaviors and separation performance of CO2. DFT calculations were specifically conducted to add the carboxyl groups to the vacant oxidation defects of CNTs. Both the adsorption of gas mixtures in single carbon nanotubes and carbon nanotube bundles with functional groups were systematically considered. The separation of SO2/N2 mixtures also was investigated in CNTs. As both adsorption capacity and selectivity determine the performance of the adsorbents, the performance coefficient of functionalized CNT bundles was used to comprehensively evaluate the CO2 separation potential using CNTs.

2. Simulation Details

2.1. Molecular Models

In our simulations, CO2 was modeled by EMP2 model of Harris and Yung [35]. N2 and O2 were treated as a rigid three-site model with two LJ sites carrying negative charges to represent the N/O atoms, associated with a dummy particle located at center of mass (COM) being used to carry the positive charges to maintain the electrostatic neutralization of molecule [36]. H2O was represented by the SPC/E model, which treated H2O as a rigid molecule with a positive charges on H atoms and negative partial charges on the O atom [37]. SO2 is modeled as a three-site model as well, with a charged LJ particle being assigned for each atom [38]. In addition, the Steele parameters were used to represent the carbon atoms in CNTs. All of the configurational parameters [13], LJ parameters, and partial charges of these guest molecules and the CNTs are summarized in Table 1. The adsorption configuration of gas molecules in four CNTs, the optimized structure of CNT unit cell with defects, and the constructed CNT array can be seen in Figure 1a,b. The interactions of adsorbate–adsorbent and adsorbate–adsorbate are described by the dispersion and electrostatic terms, given by
u i j ( α , β ) = 4 i j ( α , β ) [ ( σ i j ( α , β ) r i j ( α , β ) ) 12 ( σ i j ( α , β ) r i j ( α , β ) ) 6 ] + 1 4 π ε 0 q i q j r i j ( α , β )
where r i j ( α , β ) is the distance between the atom i and j of molecules α and β . The LJ size parameter σ i j and well depth parameter ε i j for the interactions between different species were estimated using Lorentz–Berthelot mixing rules [39], and the Dot Line Method was used to modify the long range electrostatic interactions in CNTs [40,41].

2.2. Grand Canonical Monte Carlo Simulations

To gain insights of the effect of impure gases on the adsorptive separation of CO2 from flue gas using CNTs, three impurity gases, SO2, H2O, and O2, were used to conduct simulations for the adsorption in four different CNTs ((6, 6), (7, 7), (10, 10) and (12, 12)), having the diameters of 0.81 to 1.63 nm, were considered. Initially, the adsorption of binary mixture CO2/N2 in different CNTs was examined to find out the optimized pore size of the CNT for CO2 separation. Afterwards, ternary mixtures, including CO2/N2/SO2, CO2/N2/O2, and CO2/N2/H2O were used to determine the effects of individual single impurity on the separation performance of CNTs. Eventually, the simulations for the adsorption of quinary mixture, CO2/N2/SO2/H2O/O2 was conducted to reveal the effect of the co-existing impure gases on the CO2 separation, and the optimal CNT pore size for CO2 separation in practice. In all the simulations, the molar ratio of CO2/N2 mixture was fixed at 16/84 in the bulk phase, while the partial pressure of H2O in the ternary mixtures being set as at its saturation pressure of 3.537 kPa, at 300 K. In addition, the mole fraction of SO2 and O2 in the ternary mixture was set as 0.08% and 4%, respectively. However, for the quinary mixture, the mole fractions of each gas species were defined as: 16 CO2: 4 O2: 3.16 H2O: 0.08 SO2 [17], which were chosen to mimic the practical composition of flue gases.
GCMC simulations were conducted to measure the adsorption and separation of CO2 from flue gas in consideration of the effects of impurities, the adsorbate chemical potential μ , system volume V , and temperature T were maintained constant during simulations. Three Monte Carlo trial moves including the displacement, insertion, and deletion with corresponding probabilities of 0.4, 0.3, and 0.3 were implemented. The fugacities of the components in the bulk phases were calculated using the Peng–Robinson equation of state [42] (PR EOS) for mixtures. For the binary mixture, 1 × 107 configurations were used to equilibrate the system, which was supplemented by another 5 × 107 configurations for statistical analysis. For the ternary mixtures, the configurations used for equilibration and statistics become 1 × 108 and 2 × 108, respectively. For the quinary mixture, 3 × 108and 6 × 108 configurations were used for equilibration and measuring the isotherm measurement. The equilibrium selectivity, S i j , was calculated according to
S i / j = ( x i y i ) / ( x j y j )
where, x i and y i were the molar fractions of component i in the adsorbed and bulk phases, respectively.
Four kinds of CNTs with different diameters were doped with carboxyl groups to form CNT bundles. Firstly, the original unit cell of carbon nanotube model was established. Carbon atoms were randomly deleted to produce a vacant defect, where each vacant defect contained three sp3 hybrid carbon atoms. The carboxyl group was randomly grafted to one of the SP3 hybrid carbon atoms, and hydrogen atoms were added to the other two carbon atoms to saturate the free valence. After the three vacancy defects were modified, the cell was randomly rotated and spliced three times to form a supercell, to derive the original structure of functionalized CNTs. Then, the density functional theory (DFT) was used to optimize the structure to derive the best geometry. The DFT calculation was conducted in Vienna ab initio simulation package (VASP) software package, where Perdew Burke ernzerhof (PBE) [43] was used as the exchange correlation function and a plane-wave cutoff energy was set to be 550 eV. The optimized structure was used to construct 2 × 2 carbon nanotube arrays, where the inter-tube distance was maintained at 0.6 nm. The simulation box containing CNT bundles has dimensions of 38 × 38 × 50 Å, and the periodic boundary conditions were applied in the x and y directions.

3. Results and Discussion

3.1. Effect of Pore Size on the Adsorption of CO2/N2 Mixture in CNTs

The adsorption of CO2/N2 mixture (with a mole ratio of 16/84 in the gas phase) in the CNTs at 300 K is conducted to derive the optimal diameter for CO2 capture, where the pore diameters varies from 0.81 to 1.63 nm. Figure 2 depicts the adsorption isotherms of CO2/N2 and the corresponding CO2/N2 selectivity at 300 K. As suggested, within the diameter range, all the adsorption isotherms of CO2 and N2 could be represented by type I according to the IUPAC classification. It is seen that the adsorption of CO2 and the CO2/N2 selectivity in the (6, 6) CNT with a diameter of 0.81 nm achieves their maxima below 1.0 bar. However, for the pressure range from 1.0 to 5.0 bar, the (7, 7) CNT with a diameter of 0.95 nm exhibits superior performance on separation CO2/N2 in comparison with the performance in the rest, in which both the adsorption of CO2 and the CO2/N2 selectivity are the highest. In the larger (10, 10) and (12, 12) CNTs, although the adsorption amount of CO2 monotonically increases with pressure, which is consistently higher than the result in the small CNT, the CO2/N2 selectivity is dramatically reduced compared with the value in the (6, 6) and (7, 7) CNTs. Consequently, the enlarged CNT diameter promotes the adsorption capacity of CO2 and N2 simultaneously, while reducing the CO2/N2 selectivity for the weak adsorbate–adsorbent interactions. Considering the superior adsorption amount of CO2 and significantly higher CO2/N2 selectivity, the (6, 6) and (7, 7) CNTs can provide great potential on CO2 separation from flue gas.
It is understood that the adsorption of CO2/N2 mixture in the CNTs is determined by the competition effect between the adsorbate–adsorbent interactions and the entropic effect. Figure 3 illustrates the variation of CO2–CNT and N2–CNT interactions with pressures in the CNTs with different pore sizes, where the detailed calculating procedure was provided in our previous study [3]. Although both the CO2–CNT and N2–CNT interactions decrease with the pore size of CNTs, the dependency of interactions on the pore size is stronger for CO2. Accordingly, the preferential adsorption of CO2 over N2 is suppressed in the larger CNTs, leading to the reduced CO2/N2 selectivity. In consideration of the nominal diameter, d C N T , of the (6, 6) CNT is 0.81 nm, the effective diameter for CO2 molecules rotating inside the (6, 6) CNT can be approximately measured as deff = dCNTσO−C = 0.49 nm, where σ O C = 0.32 nm is determined according to (σo + σC)/2, using the LJ size parameters of carbon atoms ( σ C ) of the CNT and oxygen atom ( σ O ) of the CO2 molecule. As the molecule size of CO2 molecule (0.5331 nm) in the axial direction is larger and that for N2 molecule (0.441 nm), CO2 molecules in our simulations are found to distribute almost in parallel to the axis of the (6, 6) CNT, showing strong rotational restrictions. However, the rotational freedom of N2 is negligibly affected. In addition, random distributions of CO2 molecules are observed in the (7, 7) CNT with a diameter of 0.95 nm, suggesting that the dramatically enhanced entropic effect is responsible for the reduced CO2/N2 selectivity in the (6, 6) CNT, compared to the (7, 7) CNT.

3.2. Effect of Single Impurity on the Adsorption of CO2/N2 Mixtures in CNTs

The adsorption of ternary mixtures, CO2/N2/X in CNTs at 300 K, with X denoting a specific impure gas including H2O, SO2, and O2, is further investigated. It is found that insignificant impact of impurities on the separation of CO2 is found in the (10, 10) and (12, 12) CNTs in all the cases, so all the simulation results for the (10, 10) and (12, 12) CNTs are provided in Figure S1 in the Supporting Information, and the results for the (6, 6) and (7, 7) CNTs are explored. The results for the adsorption of CO2 and CO2/N2 selectivity in these two CNTs are plotted in Figure 4. The adsorption curves of three impurities are shown in Figure S2.
To quantify the inhibition effect of impurity gas on the adsorption of CO2, an inhibition coefficient is defined as
I = ( a b a i m ) / a b × 100 %
where a b and a i m represents the adsorbed amounts of CO2 for the binary CO2/N2 mixture and for the ternary CO2/N2/X mixture, respectively. As suggested, for the impure gas SO2, the inhibition coefficient in the (6, 6) CNT reaches up to 50.5%, 59.6%, and 61.9%, under the pressure of 0.1, 1.0, and 12.5 bar, respectively. Similarly, the inhibition coefficients in the (7, 7) CNT corresponds to 12.9%, 31.2%, and 28.1% under the same condition. However, as seen in Figure 4c, the impact of H2O on the adsorption of CO2 is significant at low pressure (0.1 bar), which yields an inhibition coefficient of 64.5%. When the pressure is increased to above 0.1 bar, the inhibition coefficient of H2O sharply reduces to be negligible. It is interesting to find that both the adsorption of CO2 and the CO2/N2 selectivity is barely affected by the presence of O2 in the gas phase.
Figure 5a–c depicts the interactions of CO2–CNT and of impurity gas X-CNT in the (6, 6) and (7, 7) CNTs. As given in Figure 5, it is evident that SO2 has much stronger adsorption affinity with the nanotube wall than CO2, so strong adsorptive competition between SO2 and CO2 occurs, associated with the adsorption space being favorably occupied by SO2 molecules. Meanwhile, the interactions between CO2 molecules and the nanotube wall becomes weaker due to the introduction of SO2, so it is safe to conclude that the competitive adsorption and the weakened CO2–CNT interactions are responsible for negative impacts on the adsorption of CO2. Similar to the decreased adsorption of CO2, the adsorption of N2 also becomes smaller in the presence of SO2 (see Figure S3 in Supplementary Materials).
In addition, although both the adsorption amounts of CO2 and N2 are decreased by the presence of SO2 in the (6, 6) and (7, 7) CNTs, only a slight decrease in the CO2/N2 selectivity is found for (6, 6) CNT and the CO2/N2 selectivity is even enhanced in the (7, 7) CNT. To explain this phenomenon, the adsorbate–adsorbate interaction energies are estimated as a function of pressure for SO2–CO2 and SO2–N2 in Figure 5d. It is seen that CO2 molecules are strongly attracted by the adsorbed SO2 molecules in the (6, 6) and (7, 7) CNTs, whereas N2 molecules suffer the strong repulsions from SO2 molecules. As the additional CO2–SO2 interactions actually facilitate the selective adsorption of CO2 over N2, the CO2/N2 selectivity is enhanced by SO2 in the (7, 7) CNT. However, the adsorbed SO2 also enhances the entropic effect for CO2 adsorbing in the (6, 6) CNT, further restricting the rotation freedom of CO2 molecules, but this entropic effect exerts insignificant effect on the rotation of N2 molecules. Although the adsorption of CO2 is energetically favorable in the (6, 6) CNT in the presence of SO2, the strengthened entropic effect has completely dominated over the energetic effect, thereby leading to the dramatically reduced CO2 adsorption. The adsorption reduction arising from the dominant entropic effect is more significant for N2 due to its unfavorable energetic field exerted by SO2. Therefore, the CO2/N2 selectivity is reduced in the presence of SO2 in the (6, 6) CNT.
Figure 4c indicates that, at the rather low pressure <0.1 bar (water vapor is at its saturation pressure, under a mole fraction of ~35.64%), noticeable adsorption of water vapor is found in the (6, 6) CNT, where considerable adsorption space is occupied. As depicted in Figure 6, the adsorption of water vapor decreases rapidly as a consequence of the competitive adsorption of CO2 and N2, where the corresponding partial pressures are enhanced at higher pressures. The inset of Figure 6 depicts the molecular configuration of water adsorbed in the (6, 6) CNT. As reported in the previous study, negligible adsorption of water was observed in the CNTs until the partial pressure of water vapor reached a critical pressure, where water molecules filled the CNT immediately and completely once the partial pressure was above the critical pressure [12,44]. It is shown that the critical pressure for the (6, 6) CNT is around the saturation pressure of water at 300 K, which is increased to 1.75 times of the saturation pressure in the (7, 7) CNT. Based on this reason, negligible adsorption of water is observed in the (7, 7) CNT within the pressure range investigated, while the effect of water vapor is only significant at rather low pressure in the (6, 6) CNT.
Figure 4e,f depicts the adsorption isotherms for CO2 and CO2/N2 selectivity in the presence of O2 in the (6, 6) and (7, 7) CNTs, where both the adsorbed amounts and the CO2/N2 selectivity are hardly affected. This result can be explained by the analysis of the interaction energy between guest molecules and CNTs. As given in Figure 5c, the interactions of O2–CNT are much stronger than the counterparts of N2–CNT, so the competitive adsorption occurs between O2 and N2, leading to an enhanced CO2/N2 selectivity. However, the concentration of O2 in the gas phase is only 4%, far below the mole concentration of N2, 84% of N2. Therefore, no significant decreases in adsorption of N2 occurred, which is also applicable to the result for CO2. A similar result is found in ZIF-68: the presence of O2 has a negligible effect on CO2 adsorption [12].
Apparently, the presence of impurity gas generally imposes a negative effect on the adsorption of CO2, particularly in the rather small CNTs. However, the CO2/N2 selectivity demonstrates a complex dependency on the impure gases, which can be enhanced, reduced, or nearly unaffected. Meanwhile, both the adsorption of CO2 and the CO2/N2 selectivity remain almost unaffected in the larger (10, 10) and (12, 12) CNTs, making it difficult to predict the optimal CNT with the highest separation performance. Therefore, it is necessary to introduce the performance coefficient, λ e , which comprehensively evaluates the effect of the CO2 adsorption and the CO2/N2 selectivity on the separation performance, by following
λ e = exp [ ln ( α 1 M t M p ) + ln ( α 2 S t S p ) ]
where M t and S t denote the adsorption of CO2 and the CO2/N2 selectivity for the CNT of interest at the target pressure, respectively, while M p and S p represent the adsorption of CO2 and the CO2/N2 selectivity for the standard case, respectively, which are chosen as the adsorption of CO2 and the CO2/N2 selectivity of the (7, 7) CNT at 300 K and 1.0 bar. α 1 and α 2 are the weight factor s, which are set as 1.0 in this work.
Figure 7 illustrates the variation of the performance coefficient versus pressure in the (6, 6) CNT and (7, 7) CNT. As suggested, the performance coefficient is slightly increased in the (7, 7) CNT, while it becomes significantly decreased in the (6, 6) CNT. It is seen that SO2 exhibits the most influential impact on the adsorption of CO2 among the three impure gases considered. More specifically, the presence of SO2 dramatically reduces the performance coefficient in the (6, 6) CNT, which is 180% lower than the results for CO2/N2 mixture. This is caused by the strong competitive adsorption between SO2 and CO2. For the impurities of H2O and O2, the changes in performance coefficient are generally negligible, except for the results of CO2/N2/H2O mixture at 1 bar. Based on the above results, it is readily derived that the influence of impurities on the CO2 adsorption in CNTs followed the pattern: SO2 > H2O > O2. Figure 7 indicates that, in the presence of impurities, the (6, 6) CNT still provides better performances for CO2 capture than other CNTs when the pressures are below 0.5 bar, while the (7, 7) CNT exhibits the superior performance at higher pressures.
Additionally, we explored the adsorptive separation performance of CNTs for capturing SO2 from the CO2/N2/SO2 mixture by measuring the isotherm curve of SO2 and the SO2/N2 selectivity, which are depicted in Figure 8. It should be pointed out that the (6, 6) CNT with a diameter of 0.81 nm exhibits outstanding performance for separation of SO2/N2, in which the maximum adsorbed amounts of SO2 and the highest selectivity are achieved among the CNTs considered. More specifically, the SO2/N2 selectivities are unprecedentedly high, reaching 16,796, 13,965 and 7892 at the pressures of 0.1, 1.0 and 12.5 bar at 300 K, in the (6, 6) CNT.

3.3. Impacts of Impurities on CO2 Capture in Functionalized CNT Arrays

From the previous simulation results, it is evident that SO2, as a polar molecule, yielded the strongest interaction with CNT, exerting the greatest impact on CO2/N2 adsorption and separation. As there are more complex interactions between impurities, it is interesting to explore the cooperative impact on the adsorption of CO2 in this part. Due to the hydrophobicity of carbon nanotubes, the adsorption of water molecules is weak, and a small amount of H2O barely affects the adsorption and separation of CO2/N2. In order to further explore the effect of H2O on CO2/N2 adsorption, the hydrophilic carboxyl modified CNT is studied. In order to keep the same number of carboxyl groups distributed on the unit cell of CNT with different diameters, the mass fraction of carboxyl group doping is about 5.01–9.64%. After structure optimization by DFT, a 2 × 2 carbon nanotube array is constructed. When the tube spacing is set at 0.6 nm, GCMC is used to simulate the gas adsorption in carbon nanotube arrays with different diameters, using a fixed temperature and gas composition. After simulation, the adsorption configurations inside and outside the carbon nanotubes are calculated, respectively.
Figure 9 depicts the adsorption curves of CO2 and N2 and CO2/N2 selectivity mixed with impurity gases in four kinds of carbon nanotube arrays with tube spacing of 0.6 nm and temperature of 300 K. For CO2/N2 mixture, the optimal diameter of CNT bundle for adsorption separation of CO2 is found in the (6, 6) CNT, which is different from the result based on single CNT. This is because (6, 6) CNT not only has the strongest adsorbate CNT interaction, but also can provide additional adsorption space between tubes, so the adsorption capacity becomes enhanced. Under the combined effect of the two factors, (6, 6) CNT array has the best adsorption capacity and CO2/N2 selectivity under 10 bar. At higher pressure, due to the limited adsorption space, the adsorption capacity becomes lower than that for the (7, 7) CNT array. Compared with the binary mixture, the adsorption capacity of CO2 and N2 in quinary mixture is severely inhibited, especially in the small diameter (6, 6) CNT array, but the adsorption capacity of CO2 and N2 in the (7, 7) CNT array is the highest below 1 bar. In the (10, 10) and (12, 12) CNT arrays with large diameters, the adsorption capacity of CO2 increases almost linearly with the pressure, which becomes dominant when the pressure is greater than 1 bar. In addition, the CO2/N2 selectivity of the quinary mixture is increased. In particular, for (7, 7) CNT arrays, the adsorption capacity of CO2 and N2 decreased by 2.28 times and 4.45 times at 1 bar, respectively, but the selectivity increased by 1.95 times. This is because the inhibition effect is stronger for N2 (nonpolar molecule), in comparison with CO2. In addition, the selectivity of CO2/N2 in the quinary mixture is increased. By calculating the performance coefficient, as shown in Figure 10, it is found that (7, 7) CNT array always maintains the best adsorption separation performance, except some results at a very low pressure of 0.1 bar.
In order to explore the inhibition mechanism in the CNT array with a small diameter, the adsorption ratio inside and outside the CNT (amount adsorbed inside the CNT/adsorption amount outside the CNT) is calculated. According to Figure 11 plotted the ratio of internal and external adsorption capacity for binary and quinary mixtures. As suggested, in the binary mixture, CO2 and N2 tend to be trapped by the outside of the tube in the small diameter, except some measurements at the pressure below 1 bar. This is due to the strong interaction between adsorbate and CNT in the small diameter below 1 bar. With increase in sorbate loading, the adsorption space in the tube is limited, so a large amount of adsorbate is captured by the outside of the tube. However, the interaction between adsorbate and CNT is weak in CNT with large diameter, so CO2 molecules tend to be adsorbed outside the tube. In the quinary mixture, the adsorption distribution of CO2 molecules is more complex. In the (12, 12) CNT array, CO2 molecules begin to be adsorbed mainly in the tube, which is distributed uniformly outside the tube with pressure. With the increase in the pressure, the pressure in the tube becomes dominant.
The isothermal curves of water molecules and SO2 in the modified CNTs are plotted in Figure 12, where the adsorption capacity of water molecules after carboxyl modification is greatly improved. The adsorption is mainly distributed between tubes, while the adsorption capacity inside tubes is almost zero. According to the molecular snapshot of water molecules adsorbed in (7, 7) CNT array in Figure 13, a large number of water molecules are adsorbed and aggregated between tubes to form chain structures, but the adsorption of water molecules in tubes is hardly observed. At the same time, the adsorption capacity of water molecules decreases with the increase in tube diameter. By calculating the mass fraction of doping carboxyl, it is found that the carboxyl content is an important factor to affect the adsorption capacity of water molecules. As the diameter of the tube increased, the mass fraction of carboxyl group decreases, leading to the decrease in the adsorption capacity of water molecules. The presence of water molecules promotes the adsorption of SO2 in the small-diameter nanotube arrays. In Figure 14, the results for interaction energy of H2O–SO2 indicate in the small-diameter (6, 6) and (7, 7) CNTs, SO2 is subject to stronger H2O–SO2 interaction than CO2–H2O, thereby enhancing the adsorption of SO2.
In the modified CNTs, carboxyl group has little effect on the adsorption of adsorbate molecules. By simulating the adsorption of quinary mixture in a single carboxyl modified CNT, the results show that the adsorption capacity of various adsorbents is reduced, in comparison with the simulation results for unmodified CNTs. This is due to the introduction of defect groups (or the lack of carbon atoms) which weaken the interaction between the adsorbate molecules and the wall of small-diameter CNTs, so the adsorption capacity is reduced. The introduction of carboxyl group barely promotes the adsorption and separation coefficient of adsorbate molecules in the carbon tubes, suggesting that H2O plays an important role in the adsorption capacity and distribution of CO2.
The adsorption of CO2 and N2 in the quinary mixture outside the tube is seriously inhibited, but the inhibition or promotion for adsorption inside the tube varies with nanotubes with different diameters. As the carboxyl functional group hardly exerts a positive effect on the adsorption of CO2 molecules in the tube, the adsorption of CO2 molecules in the tube is mainly affected by the interaction with other adsorbate molecules. Due to the large amount of adsorbed water molecules between the small nanotubes, the adsorption of CO2 molecules mainly occurs in the tube. However, at low pressures, the adsorption of SO2 in the tube is enhanced due to the presence of H2O. Meanwhile, the adsorption of CO2 in the tube is strongly inhibited by the intensive competitive adsorption, so CO2 adsorption mainly occurs outside the tube at low pressures. According to the previous simulation results of CO2/N2/SO2 mixture in a single CNT, SO2 has little effect on the adsorption of CO2 in a large diameter tube, so CO2 is mainly adsorbed in the tube at low pressure. With the increase in pressure, the adsorption amount of H2O outside the tube decreases, where the inhibition effect weakens, so CO2 molecules begin to adsorb outside the tube, and are finally evenly distributed inside and outside the tube. In addition, the adsorption enthalpy of CO2 is increased by the attraction of H2O–CO2 in the tube, where the adsorption space is abundant in the large diameter tube, so the adsorption of CO2 increases.
As derived from the previous analysis, SO2 can enhance the selectivity of CO2/N2 in the small diameter. In addition, CO2 is subject to stronger interaction from H2O than N2, so the presence of water can also promote the CO2/N2 selectivity. The selectivity of CO2/N2 in the small diameter is increased by the combination of the two impure gases. In particular, at 1 bar, the CO2/N2 selectivity of (6, 6) CNT array increases from 30.4 to 53.8, while an increase from 30.7 to 59.9 are found for (7, 7) CNT array. The growth ratio corresponds to 1.77 and 1.95 times, respectively. As the adsorption space in (6, 6) CNT array is very small, the derived adsorption capacity of CO2 is also very limited due to the competitive adsorption of H2O and SO2. For (7, 7) CNT array, the adsorption space is promoted, so the adsorption capacity of CO2 in (7, 7) CNT array becomes higher than that in (6, 6) CNTs. As the inhibition of CO2 in (7, 7) CNTs is weaker than that in (6, 6) CNT array, the selectivity of CO2/N2 is higher. To sum up, the adsorption of H2O molecules mainly occurs between tubes, thereby inhibiting the adsorption of CO2 between tubes, while SO2 molecules compete with CO2 molecules in tubes to induce the inhibition effect. The competition between the two effects determines the adsorption distribution of CO2 inside and outside the tube. In addition, the interaction of H2O and SO2 improves the selectivity of CO2/N2, and the (7, 7) CNT array maintains the best CO2/N2 adsorption and separation performance except the results at low pressure of 0.1 bar.

4. Conclusions

In this work, a grand canonical Monte Carlo simulation is used to investigate the influence of impurity gases, including water, SO2, and O2, on the adsorption of CO2 in singe CNTs and functionalized CNT bundles. Initially, the effect of pore size of CNT on the adsorption of CO2/N2 mixture is examined, and it is revealed that the adsorption capacity had a strong dependence on the CNT diameter. Further, the influence of single impure gas on the adsorption of CO2 in CNTs is explored. By calculating inhibition coefficient to evaluate the influence on the adsorption of CO2, results indicate that SO2 is the most influential impure to affect the adsorption of CO2/N2. By introducing SO2, the interaction of CO2-CNT became weaker. Meanwhile, SO2 could compete with CO2 for the adsorption site, which exerts a negative effect on the adsorption of CO2, so the adsorption amount of CO2 has a significant decrease. Furthermore, the (6, 6) CNT exhibits superior performance for adsorption separation of SO2/N2. As for H2O, due to the partial pressure decreases sharply with pressure, decrease on the adsorption of CO2 only occurs noticeably bellow 0.1 bar. The existence of O2 hardly changes the adsorption amounts andthe CO2/N2 selectivity. Moreover, the performance coefficient is calculated to evaluate the adsorptive separation of CO2 comprehensively. It is shown that SO2 was the most influential impure gas to affect the adsorptive separation of CO2 from flue mixture. Eventually, the coexisting influence of three impure gases is also investigated. The performance coefficient is also calculated for the complex correlation with the diameter; however, it is hardly affected by the complex interaction among adsorbates. Among our simulations, the (7, 7) CNT yields the superior performance for CO2 adsorption and separation, where both the maximum uptakes and the highest selectivity occurs to the ambient temperature and pressure.

Supplementary Materials

The following supporting information can be downloaded online. Figure S1: Adsorption isotherms for CO2 in the presence of impurities, (a) SO2, (c) H2O, and (e) O2, and the corresponding CO2/N2 selectivity (b, d and f), in the (10, 10) and (12, 12) CNTs. Figure S2: Isotherm curves with pressure for (a) SO2 in CO2/N2/SO2, (b) H2O in CO2/N2/H2O, and (c) O2 in CO2/N2/O2, in (6, 6), (7, 7), (10, 10) and (12, 12) CNTs at temperature of 300 K. Figure S3: The adsorption of N2 in the presence of impurities which are (a, b) SO2, (c, d) H2O and (e, f) O2. The leaf side is these mixtures in the (6, 6) and (7, 7) CNTs, and the right side is that in (10, 10) and (12, 12) CNTs. Figure S4: Variation interaction energy of X-CNT which X represents SO2, H2O and O2 with pressure in the (10, 10) (a) and (12, 12) (b) CNTs.

Author Contributions

Conceptualization, S.L. and Y.S.; methodology, Y.S.; software, Y.S.; validation, Y.S., S.L. and X.G.; investigation, S.L.; writing—original draft preparation, Y.S.; writing—review and editing, X.G.; visualization, Y.S.; supervision, S.L.; project administration, S.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Fundamental Research Funds for the Central University (grant number 2021CDJQY-029), State Key Laboratory of Pollution Control and Resource Reuse for the Open Fund support (grant number PCRRF19038) and China Postdoctoral Science Foundation (grant number 2021M690175).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding authors.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Yu, J.; Ma, Y.; Balbuena, P.B. Evaluation of the impact of H2O, O2, and SO2 on postcombustion CO2 capture in metal-organic frameworks. Langmuir 2012, 28, 8064–8071. [Google Scholar] [CrossRef]
  2. Younas, M.; Sohail, M.; Leong, L.K.; Bashir, M.J.; Sumathi, S. Feasibility of CO2 adsorption by solid adsorbents: A review on low-temperature systems. Int. J. Environ. Sci. Technol. 2016, 13, 1839–1860. [Google Scholar] [CrossRef] [Green Version]
  3. Liu, L.; Bhatia, S.K. Molecular Simulation of CO2 Adsorption in the Presence of Water in Single-Walled Carbon Nanotubes. J. Phys. Chem. C 2013, 117, 13479–13491. [Google Scholar] [CrossRef]
  4. Rende, D.; Ozgur, L.; Baysal, N.; Ozisik, R. A Computational Study on Carbon Dioxide Storage in Single Walled Carbon Nanotubes. J. Comput. Theor. Nanosci. 2012, 9, 1658–1666. [Google Scholar] [CrossRef]
  5. Siriwardane, R.V.; Shen, M.S.; Fisher, E.P.; Poston, J.A. Adsorption of CO2 on Molecular Sieves and Activated Carbon. Energy Fuels 2001, 15, 279–284. [Google Scholar] [CrossRef]
  6. Cheng, Y.; Huang, Q.; Eic, M.; Balcom, B.J. CO2 dynamic adsorption/desorption on zeolite 5A studied by 13C magnetic resonance imaging. Langmuir 2005, 21, 4376–4381. [Google Scholar] [CrossRef]
  7. Gao, W.; Butler, D.; Tomasko, D.L. High-pressure adsorption of CO2 on NaY zeolite and model prediction of adsorption isotherms. Langmuir 2004, 20, 8083–8089. [Google Scholar] [CrossRef]
  8. Millward, A.R.; Yaghi, O.M. Metal-Organic Frameworks with Exceptionally High Capacity for Storage of Carbon Dioxide at Room Temperature. Am. Chem. Soc. 2005, 127, 17998–17999. [Google Scholar]
  9. Chen, H.; Johnson, J.K.; Sholl, D.S. Transport Diffusion of Gases Is Rapid in Flexible Carbon Nanotubes. J. Phys. Chem. B 2006, 110, 1971–1975. [Google Scholar] [CrossRef]
  10. Skoulidas, A.I.; Ackerman, D.M.; Johnson, J.K.; Sholl, D.S. Rapid transport of gases in carbon nanotubes. Phys. Rev. Lett. 2002, 89, 185901. [Google Scholar] [CrossRef]
  11. Liu, L.; Nicholson, D.; Bhatia, S.K. Adsorption of CH4 and CH4/CO2 mixtures in carbon nanotubes and disordered carbons: A molecular simulation study. Chem. Eng. Sci. 2015, 121, 268–278. [Google Scholar] [CrossRef] [Green Version]
  12. Liu, L.; Nicholson, D.; Bhatia, S.K. Impact of H2O on CO2 Separation from Natural Gas: Comparison of Carbon Nanotubes and Disordered Carbon. J. Phys. Chem. C 2014, 119, 407–419. [Google Scholar] [CrossRef]
  13. Liu, L.; Nicholson, D.; Bhatia, S.K. Exceptionally high performance of charged carbon nanotube arrays for CO2 separation from flue gas. Carbon 2017, 125, 245–257. [Google Scholar] [CrossRef] [Green Version]
  14. Liu, L.; Hu, C.; Nicholson, D.; Bhatia, S.K. Inhibitory Effect of Adsorbed Water on the Transport of Methane in Carbon Nanotubes. Langmuir 2017, 33, 6280–6291. [Google Scholar] [CrossRef]
  15. Lin, Z.; Liu, C.; Liu, L.; He, D.; Wang, X.; Zhou, X.; Zhang, Y. Unprecedentedly high selective adsorption of Xe/Kr mixtures in carbon nanotubes: A molecular simulation study. Chem. Eng. J. 2020, 393, 124744. [Google Scholar] [CrossRef]
  16. Liu, L.; Nicholson, D.; Bhatia, S.K. Effects of Flange Adsorption Affinity and Membrane Porosity on Interfacial Resistance in Carbon Nanotube Membranes. ACS Appl. Mater. Interfaces 2018, 10, 34706–34717. [Google Scholar] [CrossRef]
  17. Liu, L.; Nicholson, D.; Bhatia, S.K. Interfacial Resistance and Length-Dependent Transport Diffusivities in Carbon Nanotubes. J. Phys. Chem. C 2016, 120, 26363–26373. [Google Scholar] [CrossRef]
  18. Liu, L.; Bhatia, S.K. Influence of Morphology on Transport Properties and Interfacial Resistance in Nanoporous Carbons. J. Phys. Chem. C 2019, 123, 21050–21058. [Google Scholar] [CrossRef]
  19. Wang, Q.; Liu, L.; Liu, C.; Song, J.; Gao, X. Size effect in determining the water diffusion rate in carbon nanotubes. J. Mol. Liq. 2021, 334, 116034. [Google Scholar] [CrossRef]
  20. Song, J.; Liu, L.; Li, Q.; Liu, C.; Song, F. Entrance resistance of water transport into carbon nanotubes: Insights from molecular dynamics simulations. J. Mol. Liq. 2021, 331, 115739. [Google Scholar] [CrossRef]
  21. Lin, Z.; Liu, L.; Liu, C.; Liu, Y. Optimal Performance of Nanoporous Carbons on Adsorptive Separation of CO2 from Flue Gas. Energy Fuels 2021, 35, 8069–8080. [Google Scholar] [CrossRef]
  22. Song, J.; Liu, L.; Liu, C.; Gao, X. Interfacial resistance of gas transport through rigid and flexible zeolites. Sep. Purif. Technol. 2021, 278, 119529. [Google Scholar] [CrossRef]
  23. Wang, Q.; Liu, L.; Han, L.; Liu, C.; Liu, Y. Exchange dynamics of molecules at the fluid-solid interface determining the diffusion rate in nanopores. J. Mol. Liq. 2021, 335, 116030. [Google Scholar] [CrossRef]
  24. Lin, Z.; Liu, L.; Liu, C.; Gao, X. Adsorptive separation of Xe/Kr using nanoporous carbons in the presence of I2 and CH3I. Sep. Purif. Technol. 2021, 275, 119161. [Google Scholar] [CrossRef]
  25. Lee, K.B.; Sircar, S. Removal and recovery of compressed CO2 from flue gas by a novel thermal swing chemisorption process. AIChE J. 2008, 54, 2293–2302. [Google Scholar] [CrossRef]
  26. Liu, J.; Wei, Y.; Zhao, Y. Trace Carbon Dioxide Capture by Metal–Organic Frameworks. ACS Sustain. Chem. Eng. 2018, 7, 82–93. [Google Scholar] [CrossRef]
  27. Chang, S.-C.; Chien, S.-Y.; Chen, C.-L.; Chen, C.o.-K. Analyzing adsorption characteristics of CO2, N2 and H2O in MCM-41 silica by molecular simulation. Appl. Surf. Sci. 2015, 331, 225–233. [Google Scholar] [CrossRef]
  28. Kowalczyk, P. Molecular insight into the high selectivity of double-walled carbon nanotubes. Phys. Chem. Chem. Phys. 2012, 14, 2784–2790. [Google Scholar] [CrossRef]
  29. Li, G.; Xiao, P.; Webley, P.A.; Zhang, J.; Singh, R. Competition of CO2/H2O in adsorption based CO2 capture. Energy Procedia 2009, 1, 1123–1130. [Google Scholar] [CrossRef] [Green Version]
  30. Jia, W.; Murad, S. Separation of gas mixtures using a range of zeolite membranes: A molecular-dynamics study. J. Chem. Phys. 2005, 122, 234708. [Google Scholar] [CrossRef]
  31. Zhang, J.; Zou, H.L.; Qing, Q.; Yang, Y.L.; Li, Q.W.; Liu, Z.F.; Guo, X.Y.; Du, Z.L. Effect of chemical oxidation on the structure of single-walled carbon nanotubes. J. Phys. Chem. B 2003, 107, 3712–3718. [Google Scholar] [CrossRef]
  32. Hu, H.; Zhao, B.; Itkis, M.E.; Haddon, R.C. Nitric acid purification of single-walled carbon nanotubes. J. Phys. Chem. B 2003, 107, 13838–13842. [Google Scholar] [CrossRef]
  33. Daley, M.A.; Mangun, C.L.; DeBarrb, J.A.; Riha, S.; Lizzio, A.A.; Donnals, G.L.; Economy, J. Adsorption of SO2 onto oxidized and heat-treated activated carbon fibers (ACFS). Carbon 1997, 35, 411–417. [Google Scholar] [CrossRef]
  34. Picaud, S.; Collignon, B.; Hoang, P.N.; Rayez, J.C. Adsorption of water molecules on partially oxidized graphite surfaces: A molecular dynamics study of the competition between OH and COOH sites. Phys. Chem. Chem. Phys. 2008, 10, 6998–7009. [Google Scholar] [CrossRef] [PubMed]
  35. Harris, J.G.; Yung, K.H. Carbon Dioxide’s Liquid-Vapor Coexistence Curve and Critical Properties As Predicted by a Simple Molecular Model. J. Phys. Chem. 1995, 99, 12021–12024. [Google Scholar] [CrossRef]
  36. Potoff, J.J.; Siepmann, J.I. Vapor–liquid equilibria of mixtures containing alkanes, carbon dioxide, and nitrogen. AIChE J. 2001, 47, 1676–1682. [Google Scholar] [CrossRef]
  37. Berendsen, H.J.C.; Grigera, J.R.; Straatsma, T.P. The Missing Term in Effective Pair Potentials. J. Phys. Chem. C 1987, 91, 6269. [Google Scholar] [CrossRef]
  38. Sokolić, F.; Guissani, Y.; Guillot, B. Molecular dynamics simulations of thermodynamic and structural properties of liquid SO2. Mol. Phys. 2006, 56, 239–253. [Google Scholar]
  39. Maitland, G.; Rigby, M.; Smith, E.; Wakeham, W.; Henderson, D. Intermolecular Forces: Their Origin and Determination. Phys. Today 1983, 36, 57–58. [Google Scholar]
  40. Tang, Y.W.; Chan, K.Y. The Dot and Line Method: A Long Range Correction to Coulomb Interaction in a Cylindrical Pore. Mol. Simul. 2004, 30, 63–70. [Google Scholar] [CrossRef]
  41. Zhang, Q.; Chan, K.-Y.; Quirke, N. Molecular dynamics simulation of water confined in a nanopore of amorphous silica. Mol. Simul. 2009, 35, 1215–1223. [Google Scholar] [CrossRef]
  42. Peng, D.-Y.; Robinson, D.B. A New Two-Constant Equation of State. Ind. Eng. Chem. Fundam. 1976, 15, 59–64. [Google Scholar] [CrossRef]
  43. John, P.; Perdew, K.B. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865. [Google Scholar]
  44. Liu, Y.; Liu, J.; Lin, Y.S.; Chang, M. Effects of Water Vapor and Trace Gas Impurities in Flue Gas on CO2/N2 Separation Using ZIF-68. J. Phys. Chem. C 2014, 118, 6744–6751. [Google Scholar] [CrossRef]
Figure 1. Snapshots of the adsorption of CO2/N2 mixture in four CNTs in the presence of impurities, at 1.0 bar and 300 K, where the blue and cyan spheres used for N2 molecules, while the red and cyan spheres were for CO2 molecules, and O2 molecules were marked as the red and yellow spheres (e.g., Red = oxygen, yellow = sulfur, cyan = carbon, blue = nitrogen) (a). The optimized structure of CNT unit cell with defects, and the constructed 2 × 2 CNT array (b).
Figure 1. Snapshots of the adsorption of CO2/N2 mixture in four CNTs in the presence of impurities, at 1.0 bar and 300 K, where the blue and cyan spheres used for N2 molecules, while the red and cyan spheres were for CO2 molecules, and O2 molecules were marked as the red and yellow spheres (e.g., Red = oxygen, yellow = sulfur, cyan = carbon, blue = nitrogen) (a). The optimized structure of CNT unit cell with defects, and the constructed 2 × 2 CNT array (b).
Molecules 27 01627 g001
Figure 2. Adsorption isotherms of (a) CO2 and (b) N2, and (c) the variation of the corresponding CO2/N2 selectivity with pressure in different CNTs, at 300 K.
Figure 2. Adsorption isotherms of (a) CO2 and (b) N2, and (c) the variation of the corresponding CO2/N2 selectivity with pressure in different CNTs, at 300 K.
Molecules 27 01627 g002
Figure 3. Variation of the interaction energies of (a) CO2-CNT and (b) N2-CNT with pressure, at 300 K.
Figure 3. Variation of the interaction energies of (a) CO2-CNT and (b) N2-CNT with pressure, at 300 K.
Molecules 27 01627 g003
Figure 4. Adsorption isotherms for CO2 in the presence of impurities, (a) SO2, (c) H2O, and (e) O2, and (b,d,f) the corresponding CO2/N2 selectivity in the (6, 6) and (7, 7) CNTs, respectively.
Figure 4. Adsorption isotherms for CO2 in the presence of impurities, (a) SO2, (c) H2O, and (e) O2, and (b,d,f) the corresponding CO2/N2 selectivity in the (6, 6) and (7, 7) CNTs, respectively.
Molecules 27 01627 g004
Figure 5. Variation of the interaction energies for CO2–CNT and impurity–CNT including (a) SO2–CNT, (b) H2O–CNT and (c) O2–CNT in the (6, 6) and (7, 7) CNTs with pressure. (d) The interaction energies of CO2–SO2 and N2–SO2 in the (6, 6) and (7, 7) CNTs for the CO2/N2/SO2 mixtures, determined from GCMC simulations at 300 K.
Figure 5. Variation of the interaction energies for CO2–CNT and impurity–CNT including (a) SO2–CNT, (b) H2O–CNT and (c) O2–CNT in the (6, 6) and (7, 7) CNTs with pressure. (d) The interaction energies of CO2–SO2 and N2–SO2 in the (6, 6) and (7, 7) CNTs for the CO2/N2/SO2 mixtures, determined from GCMC simulations at 300 K.
Molecules 27 01627 g005
Figure 6. Variation of the H2O mole ratio with total pressure of CO2/N2/H2O mixture, with the partial pressure of H2O fixed at the saturation vapor pressure. The inset depicted a snapshot of the distribution of water in (6, 6) CNT at 0.1 bar and 300 K, in which a one-dimensional chain was evidently obtained.
Figure 6. Variation of the H2O mole ratio with total pressure of CO2/N2/H2O mixture, with the partial pressure of H2O fixed at the saturation vapor pressure. The inset depicted a snapshot of the distribution of water in (6, 6) CNT at 0.1 bar and 300 K, in which a one-dimensional chain was evidently obtained.
Molecules 27 01627 g006
Figure 7. Variation of the performance coefficients of different CNTs in the presence of SO2 (a), H2O (b), and O2 (c), relative to the adsorption of binary CO2/N2 mixture (CO2/N2 is 16/84) in the (7, 7) CNT at 1.0 bar and 300 K.
Figure 7. Variation of the performance coefficients of different CNTs in the presence of SO2 (a), H2O (b), and O2 (c), relative to the adsorption of binary CO2/N2 mixture (CO2/N2 is 16/84) in the (7, 7) CNT at 1.0 bar and 300 K.
Molecules 27 01627 g007aMolecules 27 01627 g007b
Figure 8. Adsorption of (a) SO2 and (b) SO2/N2 selectivity for the CO2/N2/SO2 in CNTs with diameter varied from 0.807 to 1.626 nm at 300 K.
Figure 8. Adsorption of (a) SO2 and (b) SO2/N2 selectivity for the CO2/N2/SO2 in CNTs with diameter varied from 0.807 to 1.626 nm at 300 K.
Molecules 27 01627 g008
Figure 9. Adsorption isothermal curves of CO2 in (a) CO2/N2 mixture and (b) quinary mixture, as well as the corresponding CO2/N2 selectivities for (c) CO2/N2 mixture and (d) quinary mixture.
Figure 9. Adsorption isothermal curves of CO2 in (a) CO2/N2 mixture and (b) quinary mixture, as well as the corresponding CO2/N2 selectivities for (c) CO2/N2 mixture and (d) quinary mixture.
Molecules 27 01627 g009aMolecules 27 01627 g009b
Figure 10. Performance coefficients of CO2/N2 adsorption and separation of quinary mixtures in modified CNTs with different diameters.
Figure 10. Performance coefficients of CO2/N2 adsorption and separation of quinary mixtures in modified CNTs with different diameters.
Molecules 27 01627 g010
Figure 11. The ratio of adsorption capacity of (a,b) CO2 and (c,d) N2 in binary and quinary mixtures inside and outside the CNT arrays, with four different diameters.
Figure 11. The ratio of adsorption capacity of (a,b) CO2 and (c,d) N2 in binary and quinary mixtures inside and outside the CNT arrays, with four different diameters.
Molecules 27 01627 g011
Figure 12. Isothermal adsorption curves of water molecules (a) and SO2 (b) inside and outside the tube in unmodified and modified CNT array.
Figure 12. Isothermal adsorption curves of water molecules (a) and SO2 (b) inside and outside the tube in unmodified and modified CNT array.
Molecules 27 01627 g012
Figure 13. Molecular snapshot of (7, 7) CNT array in cross section (a) and axial direction (b) at 1.0 bar, 300 K.
Figure 13. Molecular snapshot of (7, 7) CNT array in cross section (a) and axial direction (b) at 1.0 bar, 300 K.
Molecules 27 01627 g013
Figure 14. Interaction energy of CO2-H2O (a), H2O-H2O (b), N2-H2O (c) and H2O-CO2 (d) in CNT arrays with four diameters.
Figure 14. Interaction energy of CO2-H2O (a), H2O-H2O (b), N2-H2O (c) and H2O-CO2 (d) in CNT arrays with four diameters.
Molecules 27 01627 g014aMolecules 27 01627 g014b
Table 1. Lennard–Jones parameters, partial charges, and configurational parameters of adsorbates and CNT [1].
Table 1. Lennard–Jones parameters, partial charges, and configurational parameters of adsorbates and CNT [1].
MoleculeSiteLJ ParametersMolecular Model
ε / k B ( K ) σ ( n m ) X(nm)Y(nm)Z(nm)Charge (e)
CNTC28.00.34 0.000
C, H35.2200.355−0.115
C, RCOOH52.8400.3750.520
O(C), RCOOH105.680.296−0.440
O(H),RCOOH85.5500.300−0.530
H, RCOOH0.000150.0000.450
H, RC1000.2420.115
CO2C27.00.2800.00.00.00.70
O79.00.305±0.11490.00.0−0.35
N2N36.00.331±0.0550.00.0−0.482
COM000.00.00.00.964
H2OO78.20.3170.00.00.0−0.848
H00±0.07650.00.05870.424
O2O54.40.305±0.06040.00.0−0.112
COM000.00.00.00.224
SO2O57.40.301±0.12350.00.0−0.235
S145.90.3620.00.00.00.471
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Su, Y.; Liu, S.; Gao, X. Impact of Impure Gas on CO2 Capture from Flue Gas Using Carbon Nanotubes: A Molecular Simulation Study. Molecules 2022, 27, 1627. https://doi.org/10.3390/molecules27051627

AMA Style

Su Y, Liu S, Gao X. Impact of Impure Gas on CO2 Capture from Flue Gas Using Carbon Nanotubes: A Molecular Simulation Study. Molecules. 2022; 27(5):1627. https://doi.org/10.3390/molecules27051627

Chicago/Turabian Style

Su, Yiru, Siyao Liu, and Xuechao Gao. 2022. "Impact of Impure Gas on CO2 Capture from Flue Gas Using Carbon Nanotubes: A Molecular Simulation Study" Molecules 27, no. 5: 1627. https://doi.org/10.3390/molecules27051627

Article Metrics

Back to TopTop