Microporous carbonaceous adsorbents for CO₂ separation via selective adsorption

Abstract

Selective adsorption of CO₂ has important implications for many energy and environment-related processes, which require the separation of CO₂ from other gases (e.g. N₂ and CH₄) with high uptakes and selectivity. The development of high-performance adsorbents is one of the most promising solutions to the success of these processes. The present review is focused on the state-of-the-art of carbon-based (carbonaceous) adsorbents, covering microporous inorganic carbons and microporous organic polymers, with emphasis on the correlation between their textural and compositional properties and their CO₂ adsorption/separation performance. Special attention is given to the most recently developed materials that were not covered in previous reviews. We summarize various effective strategies (N-doping, surface functionalization, extra-framework ions, molecular design, and pore size engineering) for enhancing the CO₂ adsorption capacity and selectivity of carbonaceous adsorbents. Our discussion focuses on CO₂/N₂ separation and CO₂/CH₄ separation, while including an introduction to the methods and criteria used for evaluating the performance of the adsorbents. Critical issues and challenges regarding the development of high-performance adsorbents as well as some overlooked facts and misconceptions are also discussed, with the aim of providing important insights into the design of novel carbonaceous porous materials for various selective adsorption based applications.

---

Yunfeng Zhao

Yunfeng Zhao received his BS (2004) and PhD degree (2009) from Jilin University, China. After three-years postdoctoral experience at the King Abdullah University of Science and Technology (KAUST) in the Kingdom of Saudi Arabia, he joined Soochow University as an associate professor in 2012. He is now working in Prof. Yu Han's group at KAUST. His research interests mainly focus on the design, synthesis and applications of novel nanoporous materials.

Xin Liu

Xin Liu received his PhD degree at Dalian University of Technology in 2006. After that, he served as a lecturer at that institute and was then promoted to associate professor. In these years, he also worked as a visiting scholar with a CSC fellowship at Rensselaer Polytechnic Institute (RPI, USA) in 2010, and as a postdoc to Prof. Yu Han at King Abdullah University of Science and Technology (KAUST) from 2010 to 2012. His research includes theoretical design of microporous and mesoporous materials for energy and environmental applications.

Yu Han

Yu Han received his PhD degree from Jilin University, China, in 2003. He worked in the Institute of Bioengineering and Nanotechnology, Singapore, as a scientist from 2003 to 2008. He joined King Abdullah University of Science and Technology (KAUST) in the Kingdom of Saudi Arabia in 2009 as an assistant professor, and was promoted to an associate professor in 2011. His research is mainly focused on nanoporous materials and their applications in adsorption, separation, and catalysis. His research interests also include electron microscopy and plasmonic nanocrystals.

---

1. Introduction

The development of new porous materials for carbon dioxide (CO₂) separation by selective adsorption is a rapidly growing research field because of its importance to a number of energy and environment-related applications.^1,2 As generally accepted, the greenhouse gas effect that causes global warming is largely associated with the CO₂ emissions from the burning of fossil fuels in power plants and combustion engines.² It is therefore critical to develop effective methods for CO₂ capture and sequestration (CCS) from post-combustion effluents, such as power plant flue gases that typically comprise 3–15% CO₂ and more than 70% nitrogen (N₂) under a pressure of ∼1 bar.³ That is, a CCS process for flue gases requires the separation of CO₂ from N₂. Other important separation processes involving CO₂ include natural gas upgrading and landfill gas purification. Raw natural gas is mainly composed of methane (CH₄), typically 75–90%, with impurities including CO₂ (∼8%), N₂ (∼5%), and heavier hydrocarbons, and an overall pressure of ∼5 bar, while landfill gas typically contains CH₄ and CO₂ with an approximate ratio of 1 : 1 and a total pressure of ∼1 bar. The separation of CO₂ from CH₄ is therefore an essential step in natural gas and landfill gas upgrading processes.⁴

The most widely adopted approach to CO₂ capture is absorption using aqueous amine solutions, which suffers from some drawbacks including severe corrosion of the equipment and substantial amounts of energy required for the regeneration of the amine solutions.^5,6 Adsorption using porous materials is a promising alternative to the current absorption technology because it has low energy requirements and its process is clean.^7–10 The requirements for adsorbents vary with the type of application, depending on the compositions and pressure of the gas mixture. For example, ideal adsorbents for CO₂ removal from flue gas should exhibit large CO₂ adsorption capacity and high CO₂/N₂ selectivity under typical flue gas conditions, e.g., low CO₂ partial pressures (<0.2 bar). Likewise, for natural gas and landfill gas purification, the adsorbents should show large CO₂ adsorption capacity and high CO₂/CH₄ selectivity at CO₂ pressure of ∼0.5 bar. Thus, unlike CO₂ storage that emphasizes the CO₂ uptake at high pressures (>10 bar), CO₂ separation from flue gas, natural gas, or landfill gas requires the adsorbents to be highly adsorptive and selective at relatively low pressures. It is well established that at low pressures, the adsorption behaviour is essentially dominated by adsorbent/adsorbate interactions. Therefore, specific and strong affinity to CO₂ is a more important factor for the adsorbent than large surface area or high pore volume. In addition to the adsorption capacity and selectivity, there are some other important attributes that should be taken into account for adsorbent evaluation, such as the adsorption kinetics, energy consumption for regeneration, tolerance to moisture or other impurity gases, and long-term stability in the adsorption/separation performance.¹¹

Various porous materials have been tested as adsorbents for CO₂ separation and capture, including activated carbons,^12,13 zeolites,^14,15 mesoporous silicas,¹⁶ metal–organic frameworks (MOFs),^7,17–20 etc. These materials have their own advantages and limitations, and they perform differently for different applications, depending on the composition and pressure of the gas mixture. Zeolites with low Si/Al ratios (e.g., zeolite 13X and zeolite 5A) are able to uptake significant amounts of CO₂ at low pressures (∼2.5 mmol g⁻¹ at 298 K and 0.1 bar), owing to the presence of a large number of extra-framework alkali cations (e.g. Li⁺, Na⁺, K⁺, Cs⁺) that promote CO₂ adsorption via electrostatic (charge-quadrupole) interactions.^21–23 However, zeolites suffer from drastically decreased CO₂ adsorption capacities if the gas stream contains moisture, because the cations would preferentially interact with water molecules due to the stronger charge-dipole interactions.²⁴ MOFs have extremely large surface areas and pore volumes, which enable them to adsorb a large amount of CO₂ at high pressures (>10 bar). In the low pressure range, however, the majority of MOFs exhibit unfavorable adsorption isotherms for CO₂, corresponding to low CO₂ selectivity and uptake (<1.0 mmol g⁻¹ at 298 K and 0.1 bar). Hence, most MOFs seem to be more suitable for CO₂ storage than for separation.^25–28 Synthesizing MOFs with open (unsaturated coordinated) metal sites or polar functional groups has proved to be effective for enhancing their affinity to CO₂, whereas the competitive adsorption of water and long-term stability remain problems associated with MOFs.^29–33 Mesoporous silicas have weak CO₂ adsorption ability at low pressures despite their high pore volumes, because the filling of mesopores by capillary condensation requires a high relative pressure (P/P₀ > 0.4) that corresponds an absolute CO₂ pressure of >20 bar at ambient temperatures. Surface functionalization with alkylamines endowed mesoporous silicas with remarkably enhanced low-pressure CO₂ uptake (∼1.9 mmol g⁻¹ at 298 K and 0.1 bar) and selectivity by chemisorption,^34–36 which, however, in turn leads to difficult and energy-intensive regeneration of the adsorbents, similar to the case of CO₂ absorption using amine solution.

Among numerous solid adsorbents, activated carbons (ACs) hold the greatest potential for commercial use, due to their easy availability, low cost, and superior thermal/chemical stabilities. Pure carbon adsorbents generally have weak affinity to CO₂,³⁷ and consequently, like MOFs, they have rather low CO₂ adsorption capacity at low pressures (∼0.5 mmol g⁻¹ at 298 K and 0.1 bar), despite their large surface areas.^38,39 Therefore, ACs are not suitable for flue gas treatment or natural gas/landfill gas purification unless their low-pressure CO₂ uptake is significantly enhanced by doping or surface-modification. In recent years, a new class of porous materials, porous organic polymers (POPs), were developed, which can be classified into different categories, such as crystalline covalent organic frameworks (COFs),^40–45 porous aromatic frameworks (PAFs),^46–48 triazine-based organic frameworks (CTFs),^49–53 conjugated microporous polymers (CMPs),^54,55 hyper-crosslinked polymers (HCPs).^56–58 Unlike MOFs whose frameworks are extended through organic linker interconnecting metal centers through coordination bonds, POPs have pure organic framework cross-linked by covalent bonds and thus excellent stabilities. On the other hand, POPs have more structural and compositional flexibility than pure carbon materials (e.g. ACs and templated carbons), as their textural properties and surface functionalities can be easily tuned by designed synthesis, e.g., by varying the monomers and controlling the polymerization conditions. These features make POPs promising adsorbents for CO₂ separation.^48,59–64 However, although remarkable progress has been made in synthesis, the investigation of POPs for CO₂ separation is still at the early stage. Most current studies of POPs pay little attention to the low-pressure adsorption performance, but compare different materials using the adsorption capacity at 1 bar as a criterion. Moreover, the POPs are only evaluated by single-gas equilibrium adsorption, and their capabilities to separate gas mixture under kinetic conditions have rarely been investigated. More importantly, the CO₂ adsorption/separation performance of POPs remains poor in comparison with zeolites, MOFs, and microporous carbon materials.

In the last decade, significant efforts have been expended on the investigation of various porous materials for CO₂ separation by selective adsorption, which have spawned a vast literature. The present review is focused on carbon-based (carbonaceous) adsorbents including inorganic ACs and organic POPs, as many systematic and comprehensive review articles for zeolites and MOFs are already published. We review the progress of carbonaceous adsorbents with emphasis on the correlation between their textural and compositional property and their CO₂ adsorption/separation performance, and discuss different strategies for enhancing the CO₂ adsorption capacity and selectivity. The discussion here mainly focuses on flue gas treatment (CO₂/N₂ separation), whereas most strategies and conclusions may also apply to natural gas upgrading and landfill gas purification (CO₂/CH₄ separation). In addition, we introduce common evaluation methods and criteria for the adsorbents used for these applications.

2. Design and evaluation of adsorbents for low-pressure CO₂ adsorption and separation

2.1. Design of adsorbents

The development of high-performance adsorbents for CO₂ adsorption and separation requires a rational design that takes both thermodynamic and kinetic factors into account. The design also depends on the desired property. For example, when the CO₂ adsorption capacity is the major concern, the sorbent should have a large number of ultramicropores (<0.7 nm),⁶⁵ because as compared to larger pores, ultramicropores hold stronger adsorption potential and dominate low-pressure adsorption through pore-filling processes, that is, it is the ultramicropore surface area rather than the overall surface area that determines the CO₂ adsorption capacity at low pressures.⁶⁶ In addition to the generation of ultramicropores, the adsorbent/adsorbate interaction and thus the low-pressure CO₂ uptake can be increased by functionalizing the pore surface to enhance its affinity to CO₂ molecules. However, the overall adsorbent/adsorbate interaction should not be too strong, because that would result in irreversible desorption, i.e., difficulty in the adsorbent regeneration. The interaction strength is usually represented by the isosteric heat of CO₂ adsorption (Q_st), which can be calculated from the CO₂ adsorption isotherms collected at different temperatures using the Clausius–Clapeyron equation. The Q_st can be adjusted by engineering the pore size and the surface functionality of the adsorbent, and as generally accepted, the optimal values for flue gas separation are in the range of 30–50 kJ mol⁻¹.¹¹

When the separation of CO₂ from a gas mixture is concerned, the selectivity of CO₂/N₂ (for flue gas) or CO₂/CH₄ (for natural gas and landfill gas) is of equal, or even more, importance than the CO₂ adsorption capacity. To have a high selectivity, the adsorbent should exhibit distinguish adsorption behaviors to the two gases (favorable for CO₂ and unfavorable for N₂ or CH₄). Both CO₂ and N₂ are linear symmetric molecules with no net dipole moments. However, due to the polar C=O bonds, CO₂ has a much larger quadrupole moment than does N₂ (13.4 × 10⁻⁴⁰ cm² vs. 4.7 × 10⁻⁴⁰ cm²). CH₄ is essentially non-polar without even quadrupole moment due to its tetrahedral molecular geometry (Fig. 1).⁶⁷ It is therefore conceivable that functionalizing adsorbents with polar groups or extra-framework cations would lead to preferential adsorption of CO₂ over N₂ or CH₄ through the stronger dipole (or charge)-quadrupole interaction. This strategy has been extensively used in designing adsorbents for flue gas treatment. The commonly used methods to introduce polarity in carbon-based adsorbents include elemental doping in the framework, pore surface modification, and incorporation of extra-framework cations. These methods are elaborated in Section III with specific examples.

Fig. 1 Electrostatic potential representations of CO₂, N₂ and CH₄ mapped against the iso-electron density value of 0.005 au.

From the kinetics perspective, the CO₂ separation can be optimized by the “molecular sieve” effect. In principle, an adsorbent having a well-defined pore size between the kinetic diameters of two gas molecules (CO₂/N₂ or CO₂/CH₄) would give infinitely high selectivity. Given the very close kinetic diameters of CO₂ (0.330 nm), N₂ (0.364 nm) and CH₄ (0.380 nm)⁶⁸ preparing adsorbents that can act as a “molecular sieve” is difficult, especially in the cases of carbonaceous adsorbents, which are usually amorphous materials with poorly defined pore structures and relatively broad pore size distributions.

However, even though ideal “molecular sieve” effect is difficult to achieve, marked diffusion selectivity would emerge when the adsorbent has uniform pores with sizes approaching the diameter of the large gas molecule in the gas pairs to be separated.⁶⁹ Note that despite the larger kinetic diameter and lower polarity of CH₄ relative to N₂, the CO₂/CH₄ selectivity is lower than CO₂/N₂ selectivity for most adsorbents, because CH₄ has a higher critical temperature (191 K vs. 126 K)^70–72 and polarizability (26 × 10⁻²⁵ cm³ vs. 17.6 × 10⁻²⁵ cm³)^73–75 than does N₂.

Overall, an optimal adsorbent for CO₂ separation via selective adsorption should possess a large amount of ultramicropores, a moderately strong affinity to CO₂, preferential adsorption of CO₂ over N₂ or CH₄, and preferably a suitable pore size to provide kinetic selectivity. The fact that actual gases always contain water (moisture) should also be taken into consideration for the design of adsorbents. Unlike CO₂ molecules that have only quadrupole moments, water molecules have dipole moments. As a consequence, water molecules are preferentially adsorbed if the separation essentially replies on electrostatic interaction, causing a marked decrease in the CO₂ adsorption capacity as well as in the selectivity. The interference from water on CO₂ adsorption is actually a severe issue in flue gas treatment, but it has been ignored in most fundamental studies in which adsorbents were only evaluated with dry CO₂. Retaining the high CO₂ adsorption capability in the presence of water requires new design strategies for the adsorbent. Recently, a perfluorinated covalent triazine-based framework was fabricated for selective CO₂ capture from N₂ with moisture, where the strong polarity of the C–F bonds promoted CO₂ adsorption via electrostatic interactions, and at the same time the hydrophobic nature of the C–F bonds effectively inhibited the competitive adsorption of water. This work represents an excellent example that highly selective and water-tolerant CO₂ adsorption can be achieved by a designed synthesis of adsorbent.⁵³

2.2. Evaluation of adsorbents

As exemplified by flue gas separation, the parameters for the performance evaluation of adsorbents include the CO₂ adsorption capacity, CO₂/N₂ selectivity, adsorption kinetics, tolerance to water, ease of regeneration and lifetime, etc., among which the first two are the most basic and important. The equilibrium adsorption capacity at a specific pressure and temperature can be directly determined from the single-gas physical sorption isotherms. From the single-gas adsorption results, the CO₂/N₂ selectivity can be predicted by using different theories and models. The most widely used is the ideal adsorption solution theory (IAST), as its effectiveness has been proved in many systems.^76–79 With the IAST method, one can estimate the separation performance of an adsorbent simply from the single-gas adsorption isotherms, without the need of special facilities to perform the mixed-gas separation measurement. However, it is possible that the theoretically predicted selectivity sometimes deviates much from that of the real mixed-gas separation. This is because the IAST method is based on the equilibrium adsorption results, whereas the real separation is carried out under kinetic flow conditions, where the gas adsorption equilibrium cannot be fully established in most cases. Therefore, mixed-gas separation is usually necessary for evaluating new adsorbents, which can be conducted in laboratory by a simple “breakthrough” experiment. In recently years, the mixed-gas breakthrough measurement has gradually become a standard process in the field of MOFs,^77,78,80 but it has seldom been applied for carbon-based adsorbents, especially for POPs,⁸¹ because of the short history of this new class of porous material. In the followings, we briefly describe the IAST method and the mixed-gas breakthrough experiment.

(i) The ideal adsorption solution theory (IAST) method. The ideal adsorption solution theory was first established by Myer and Prausnitz to calculate the adsorption equilibria for components in a mixed-gas by using only pure-component (single gas) equilibrium adsorption data.⁷⁶ In this method, the adsorbed phase is treated as an ideal solution that has established equilibrium with the gas phase, and therefore the basic thermodynamic equations for a real fluid can be used for the adsorbed phase. To write these equations for the adsorbed phase, it is necessary only to substitute spreading pressure π for pressure P and to substitute area A for volume V.

To apply the IAST method for predicting the selectivity in mix-gas adsorption, the experimental single-component isotherms for each adsorbate should first be fitted (Fig. 2), for example, by using the dual-site Langmuir–Freundlich equation:

Fig. 2 Illustration of the IAST method, using the adsorption of CO₂ and N₂ on a newly developed (unpublished) microporous carbon as an example. (a) Fitting the experimental single-component adsorption isotherms with the dual-site Langmuir–Freundlich model. (b) The calculated IAST selectivities for a range of total pressure (0–110 kPa) and a fixed mole ratio of CO₂/N₂ (1 : 9).

In this equation, n is the adsorbed amount of the gas per mass of absorbent (in mol kg⁻¹) as the function of P that is the pressure of the gas (in kPa); n_m1 and n_m2 are the saturation uptakes (in mol kg⁻¹) for sites 1 and 2, b₁ and b₂ are the affinity coefficients (in kPa⁻¹) for sites 1 and 2; and t₁ and t₂ represent the deviations from the ideal homogeneous surface for sites 1 and 2.

The fitted isotherms are then used to calculate the selectivity for component i relative to component j,

where x_i and x_j are the mole fractions of components i and j, respectively, in the adsorbed phase, and y_i and y_j are the mole fractions of components i and j, respectively, in the gas phase. The calculations are on the basis of the following equations, which hold true according to the IAST theory.

First, the partial pressure of an adsorbed component is given by the product of its molar fraction in the adsorbed phase and the pressure which it would exert as a pure adsorbed component at the same temperature and spreading pressure as those of the mixture:

Py_i = P^°_i(π)x_i (2)

Py_j = P^°_j(π)x_j (3)

On the other hand, the spreading pressure for sorbates i and j can be calculated from their respective adsorption isotherms using the following equations:

where A represents the specific surface area of the adsorbent, R is the ideal gas constant, T is the temperature.

Furthermore, for the case of adsorption of a binary gas mixture, each component has the same spreading pressure, and the sum of the mole fractions of the two components is 1 in both the gas phase and the adsorbed phase. Therefore, we have the following equations:

π^°_i = π^°_j (6)

x_i + x_j = 1 (7)

y_i + y_j = 1 (8)

The above seven eqn (2)–(8) are seven independent equations with nine unknowns. With two quantities being specified, particularly P and y_i, all of the rest unknowns can be solved. Utilization of the aforementioned equations would yield the following equilibrium expression for adsorbates i and j:

This equation was solved for x_i using numerical analysis and root exploration for a range of pressures (P) at a specified y_i value so as to obtain the selectivities (vs. the total pressures) for adsorbate i relative to adsorbate j (Fig. 2).

(ii) Mixed-gas breakthrough experiment. The mixed-gas breakthrough experiment can be conducted in a lab-scale fix-bed reactor. The adsorbent powders are packed in a column residing in an oven that controls the temperature of the adsorbent bed. The mole fractions of CO₂ and N₂ in the mixed gas are adjusted by controlling the flow rate of each gas. A separate path of inert gas (e.g. He) is used to purge the adsorbent for activation or regeneration. The effluent from the adsorbent bed is monitored by a mass spectrometer or a gas chromatography to determine the breakthrough time point of each gas.

The absolute adsorbed amount of gas i (q_i) is calculated from the breakthrough curve by the equation:

where F_i = influent flow rate of the specific gas (cm³ min⁻¹); t₀ = adsorption time (min); V_dead = dead volume of the system (cm³); F_e = effluent flow rate of the specific gas (cm³ min⁻¹); m = mass of the sorbent (g).

The separation factor (corresponding to the “selectivity”) of the breakthrough experiment is defined as α = (q₁/y₁)/(q₂/y₂), where y_i is the mole fraction of gas i in the gas mixture.

3. Carbonaceous microporous adsorbents for CO₂ adsorption

Here we define carbonaceous adsorbents to include (i) inorganic porous carbons, which feature a high-temperature carbonization (pyrolysis) process during the preparation and have little oxygen and hydrogen remaining in the framework, and (ii) porous organic polymers, while we exclude the organic–inorganic hybrid porous materials, e.g. MOFs, from this review.

3.1. Inorganic porous carbon materials

Inorganic porous carbon materials mainly include activated carbons (ACs)^82,83 and templated carbon materials.^69,84 For the latter, the commonly used templates are microporous zeolites and mesoporous silica materials. The templated carbons may also be activated after the preparation, in order to further increase the surface area and porosity by creating more micropores.^69,85 As discussed earlier, pristine carbon materials have weak affinity to CO₂ with adsorption heat of less than 25 kJ mol⁻¹,³⁷ leading to low CO₂ adsorption capacity as well as low CO₂/N₂ selectivity (∼5)^38,86 or CO₂/CH₄ selectivity (<3)⁸⁷ at low pressures. Various strategies have been reported to enhance the low-pressure CO₂ uptake of inorganic carbon adsorbents, mainly by compositional modification (heteroatom doping, functional groups grafting, and incorporation of extra-framework cations) and pore structure engineering. Remarkable success has been achieved for the CO₂/N₂ separation; however, the CO₂ separation from CH₄ by using carbon materials has rarely been investigated. Representative examples of modified inorganic carbon adsorbents are summarized in Table 1, along with their textural/compositional properties and performances in CO₂ adsorption and separation.

Table 1 Summary of representative inorganic carbon adsorbents with their textural and CO₂ adsorption properties

Materials	BET m^2 g^−1	Comments	CO2 capacity@273 K, mmol g^−1	CO2 capacity@298 K, mmol g^−1		IAST CO2/N2 (1/9) selectivity, 298 K	Qst^c kJ mol^−1	Ref.
			1 bar	1 bar	0.1 bar^a
a The values were estimated from the published isotherms in case they were not specified in the literature.b By initial slopes method.c 273 K, CO2/N2 (15/85).
AC Norit R1	3000	N: 0	—	2.23	0.5	<5	—	86
3C-1000N	2104	N: 3.73	—	3.46	0.86	—	—	88
RFL-500	467	N: 1.92	—	3.13	1.18	—	—	89
HCM-DAH-1-900-1	1392	N: <0.5	4.9	3.3	0.89	17	26.7	90
CP-2-600	1700	N: 10.14	6.2	3.9	0.87	5.3	31.5	91
IBN9-NC1-A	1181	N: 12.91	6.7	4.50	1.10	27	36.1	69
CRHC221-DES	666	—	—	3.29	1.08	4.3	39.9	92
AG-2-700	1940	N: 1.47	7.4	4.4	0.68	—	25	93
AC-2-635	1381	N: 4.59	6.0	3.86	1.01	21	30.4	94
SCEMC	729	S: 6.56	—	2.46	0.52	—	59	95
a-SG6	1396	S: 4.5	—	4.5	1.30	51^b	—	96
KNC-A-K	614	N: 10.5, K: 8.6	5.0	4.04	1.62	48	59.3	97
ACM-5	2501	N: 1.8	11.51	5.14	0.86	—	65.2	98
N-TC-EMC	2559	N: 7	—	4	1.01	14	50	84
CEM-750	3360	N: 5.2	6.92	4.38	0.80	9.5	36	99
N-HCSs	767	N: 14.8	—	2.67	1.12	29	—	100
PAF-1-450	1191	N: 0	4.5	—	—	209^c	27.8	60
FCTF-1-600	1535	N: 15.4	5.53	3.41	0.68	19	32	81

---

3.1.1. Functionalization of porous carbons.
(i) Nitrogen doped carbon materials. Doping with heteroatoms is a common strategy to promote the CO₂ adsorption of carbonaceous adsorbents. Nitrogen (N) is the mostly used heteroatom for doping, which can be incorporated in the carbon framework by either post-treatment or direct synthesis. For example, treating porous carbons at 873–1273 K in NH₃ atmosphere effectively introduced N into the framework of a commercial AC (CWZ-35), as confirmed by Fourier transform infrared spectra, and a remarkable increase (∼40%) in CO₂ uptake was achieved by this treatment.¹⁰¹ Similar treatment has been applied to an ordered mesoporous carbon material, giving rise to a 3.73 wt% N loading and an enhanced CO₂ adsorption capacity (3.46 mmol g⁻¹ at 25 °C and 1 bar).¹⁰² In addition to the difficulty in handling gaseous ammonia, the limitation of this method is that it can only incorporate a small amount of N (<4 wt%) into the carbon adsorbent.

In order to better control as well as to increase the N loading, direct synthesis methods were developed, which employed various N-containing compounds as the synthetic precursors. Depending on the chosen precursor and carbonization conditions, the N content in the final adsorbent can be tuned in a wide range. For example, Lu et al. synthesized a N-doped porous carbon (RFL-500) by using phenol formaldehyde resin as carbon source and L-lysine as both the polymerization catalyst and the nitrogen source.¹⁰³ RFL-500 has 1.92 wt% nitrogen and a CO₂ adsorption capacity of 3.13 mmol g⁻¹ at 298 K and 1 bar. The same group later synthesized porous N-doped carbon monoliths through the copolymerization of various diamines (nitrogen source) with phenols and aldehydes, followed by carbonization. The monoliths prepared in this way feature highly interconnected macroporosity and mesoporosity (implying fast adsorption kinetics), excellent CO₂ capture and separation performance, remarkable mechanical strength, and facile adsorbent regenaragtion.⁹⁰ Notably, the N content in the carbon monolith is rather low (<0.5 wt%), which might account for its relative low CO₂ uptake (∼3.0 mmol g⁻¹ at 298 K and 1 bar) and CO₂/N₂ selectivity (separation factor α = 13–28, as determined in the breakthrough experiments) in comparison with some later developed N-containing carbon adsorbents.^{69,91,104,105}

Apparently, precursors with higher N contents would result in high N loading in the final materials, if the N species could be largely maintained in the carbonization and activation processes. On the basis of this consideration, people intentionally chose high N-content precursors for the synthesis, while applying low carbonization/activation temperature (<973 K) to avoid the elimination of N species. Fuertes et al. prepared a porous carbon (CP-2-600) through the pyrolysis of polypyrrole at 873 K, which has as much as 10.14 wt% nitrogen and accordingly a high CO₂ capacity of 3.9 mmol g⁻¹ at 298 K and 1 atm.⁹¹ Similarly, Han et al. prepared IBN9-NC1-A from a specially selected precursor p-diaminobenzene (containing 25 wt% N) with mild carbonization and activation processes at 873 K. IBN9-NC1-A exhibits an excellent performance for CO₂ adsorption. Specifically, its equilibrium CO₂ adsorption capacity at 298 K reaches up to 1.75 mmol g⁻¹ 0.2 bar and 4.50 mmol g⁻¹ at 1 bar, and its CO₂/N₂ selectivity is 42 according to the IAST method. These results are mainly attributed to its unprecedented nitrogen doping level (13 wt%).⁶⁹

Gutierrez et al. employed deep eutectic solvents (DESs) composed of resorcinol, 3-hydroxypyridine and choline chloride to synthesize hierarchically structured N-doped carbons (CRHC221-DES), where DESs acted as both structure-directing agent and carbon/nitrogen source.⁹² The resulting materials have rather high N contents even when a high carbonization temperature was applied (>10 wt% for 873 K and >5 wt% for 1073 K). It is interesting to note that although the sample prepared at 873 K has a higher N content than its counterpart prepared at 1073 K, the latter exhibited superior CO₂ adsorption capacity (3.29 mmol g⁻¹ vs. 2.45 mmol g⁻¹ at 298 K and 1 bar), due to its larger surface area associated with micropores. This result clearly demonstrated that the N content is not the sole factor that influences the CO₂ uptake in N-doped carbons while the amount of ultramicropores also plays a very important role in promoting CO₂ adsorption. The carbons prepared with this method gave a relatively low CO₂/N₂ selectivity of ∼5 as determined by IAST.

Direct pyrolysis of N-containing natural polymers or biomass is a low-cost approach to prepare N-doped carbon adsorbents (AG-2-700). Sevilla et al. synthesized highly porous N-doped carbon materials (N contents of 1.1–4.7 wt%) with apparent surface areas in the range of 1300–2400 m² g⁻¹ from mixtures of algae and glucose. Significantly, the majority of the porosity of these materials is contributed from uniform micropores with diameters less than 1 nm. These carbons present large CO₂ adsorption capacities up to 4.5 mmol g⁻¹ at 298 K and 1 bar.⁹³ The authors correlated the CO₂ capture capacity with the volume of narrow micropores, and proposed that various N species in the carbon, such as pyridinic-N, pyridonic/pyrrolic-N and quaternary-N, are insignificant in enhancing the CO₂ adsorption capacity, which contradicts the general opinion. Shi et al. directly converted Chitosan to N-doped microporous carbons by a single-step preparation that combined the carbonization and the activation processes. The sample prepared under optimized conditions (AC-2-635) showed a CO₂ uptake of 3.86 mmol g⁻¹ at 298 K and 1 bar, as well as a marked CO₂/N₂ selectivity of ∼21.⁹⁴

Despite the diverse strategies and the great success in preparing N-dope carbon adsorbents, there remain some questions to be addressed.

First, although these adsorbents were supposed to be used for flue gas treatment, they were only evaluated by their CO₂ adsorption capacity at 298 K and 1 bar, while the more important criteria for practical flue gas separation, i.e. the low pressure (<0.2 bar) CO₂ uptake and the CO₂/N₂ selectivity at typical flue gas temperatures (>323 K), were ignored in most studies.

Second, how the doping with nitrogen influences the long-term stability of microporous carbons in the textural properties has not been carefully investigated. This question should not be ignored especially when the doping level is high. Our preliminary results reveal that the textural properties (surface area and pore structure) as well as the CO₂ adsorption capacity of a N-doped AC (containing ∼10 wt% N) decreased by ∼50% after being stored under ambient conditions for half a year.¹¹³ This problem may not be general for all N-doped carbons synthesized by different methods. However, it deserves more careful studies to correlate the structural stability of adsorbent and the type/amount of N species incorporated.

Third, depending on the synthetic strategy, the doped N in the carbon framework may have many different forms, whereas their identification and quantification is not straightforward. X-ray photoelectron spectroscopy (XPS) is the major tool to identify the N species, but the assignments of XPS peaks are inconsistent in different studies, due to the complexity of such materials (the possibility of many different N species co-existing in poorly define carbon frameworks). This fact makes it difficult to investigate the exact contribution of different N species in promoting CO₂ adsorption. We summarize in Fig. 3 the assignments of XPS peak positions, which often appear in the N 1s XPS spectra of N-doped carbon materials, to various typical N species, according to the literature.

Fig. 3 Various nitrogen species in N-doped carbons and the corresponding binding energies as determined by XPS.

Last, the essence of N species–CO₂ interaction, i.e. the mechanism of enhanced CO₂ adsorption by N-doping, remains controversial. It is conventionally assumed that the N species generated in carbon by doping are essentially basic groups and therefore they can promote the adsorption of acidic CO₂ gas.⁸⁹ However, the one-to-one base–acid interaction cannot explain the fact that a small amount of doped nitrogen can significantly increase the CO₂ adsorption capacity. Recently, Qiao et al. indicated through quantum chemical calculations that the doped nitrogen species can facilitate hydrogen-bonding interactions between the surrounding C–H groups and CO₂ molecules.¹⁰⁴ This one-to-many model better explains the observed superior CO₂ uptake of low N-content ACs. More recently, Han et al. correlated the polarizing capabilities of various functional groups including typical N species in carbons with their enhancement effects on CO₂ adsorption and proposed that such effects are essentially based on electrostatic interactions.⁹⁷ Interestingly, they indicated that most previous studies overlooked the presence of HCl in N-doped ACs and did not discuss the effect of HCl on CO₂ adsorption. Their argument is briefed as follows: after a standard activation procedure with KOH, the adsorbent is usually washed with HCl solution and water to completely remove the residual KOH, metal K and other impurities; it is therefore reasonable to expect that some HCl molecules would remain in N-doped carbon materials, considering their strong interaction with basic nitrogen groups. They found that HCl molecules anchored in N-doped carbon actually have a promotion effect on CO₂ adsorption through electrostatic interactions, which contradicts the conventional wisdom that the neutralization of basic sites by acids diminishes the adsorption of acidic CO₂. This study suggests that any modification that increases the polarity of the carbon framework (not necessary to be basic groups) would favour CO₂ adsorption, and provides an alternative and complementary explanation for the enhancement effect of N-doping on CO₂ adsorption.

(ii) Sulphur doped carbon materials. Sulphur (S) has also been incorporated in porous carbons for CO₂ adsorption. Xia et al. synthesized a S-doped microporous carbon (SCEMC) by using zeolite EMC-2 as a hard template and 2-thiophenementhanol as both carbon and sulphur sources.⁹⁵ SCEMC contains 6.5% of S and has a specific surface area of 729 m² g⁻¹. Notably, SCEMC gives an impressive isosteric heat of CO₂ adsorption (Q_st = 59 kJ mol⁻¹) at very low CO₂ coverage, indicating strong interactions between S species and CO₂. However, the Q_st value decreases quickly to <40 kJ mol⁻¹ with the increase of surface coverage, which suggests a limited number of strong adsorption sites. Consequently, its overall CO₂ adsorption capacity is not outstanding (2.46 mmol g⁻¹ at 298 K and 1 bar). Kim et al. prepared a S-doped microporous carbon material (a-SG6) by the chemical activation of a reduced-graphene-oxide/poly-thiophene composite material, which exhibits a large CO₂ adsorption capacity (4.5 mmol g⁻¹ at 298 K and 1 bar), impressive CO₂ selectivities over N₂, CH₄ and H₂, as well as stable adsorption performance during multiple adsorption–desorption recycles. In addition to the S-doping, the large surface area (1396 m² g⁻¹) and optimal pore size (0.6 nm) of this material also account for its excellent CO₂ adsorption property.⁹⁶ These results suggest that S-doping is of comparable (or even more) effectiveness to N-doping for increasing in CO₂ adsorption of microporous carbon materials. In comparison with the extensively investigated N-doping, there are very few studies on the S-doping of carbon adsorbents, possibly because of the long-held viewpoint that acid–base interactions between N-containing basic groups and acidic CO₂ are required to promote CO₂ adsorption. The promising results given by S-doped carbons (S-doping does not generate basic groups) along with other experimental or theoretical findings challenge this viewpoint.⁹⁷
(iii) Extra-framework functionalization of porous carbons. In addition to the doping of heteroatoms in the carbon framework, the CO₂/adsorbent affinity can also be enhanced through extra-framework functionalization of porous carbons.

Given the specific reactions of various amines (primary, secondary, and tertiary amines) with CO₂ and their wide use for CO₂ capture, it is natural to think that the porous carbon adsorbents can be functionalized with amines to enhance their CO₂ adsorption capabilities. However, unlike silica-based materials whose surface silanol groups provide grafting sites for easy functionalization, amorphous carbon materials have ill-defined surface groups of low density, which makes it difficult to graft functional groups on their surfaces with covalent bonds. Hence, physical treatment of carbon materials with solution containing amines was used as an alternative strategy of functionalization. For example, fly ash carbon was treated with a 3-chloropropylamine–hydrochloride solution and potassium hydroxide, and improved CO₂ capture capacity was observed after the treatment.¹¹⁴ However, the use of amines implies difficult adsorbent regeneration due to its chemisorption nature. Moreover, as the amine groups are physically mixed with the carbon sorbents, they would block the porous structure of the carbon materials, leading to limited total capacities.

It is worth noting that despite their well-recognized important role in enhancing CO₂ uptake in the fields of zeolites²² and MOFs,^23,115–120 extra-framework cations have never been introduced to carbon sorbents for the same purpose. On the contrary, the conventional activation procedure includes the use of KOH to etch the carbon material at elevated temperatures, followed by thorough washing with HCl solution to completely remove the residual KOH, metal K and other impurities. The intentional removal of K⁺ from the activated carbon may actually cause unexpected loss in its CO₂ adsorption performance. On the basis of these considerations, Han et al. recently modified the conventional activation procedure simply by washing the KOH-activated carbons with ethanol/water rather than the commonly used HCl solution.⁹⁷ In this way, the remnant K and KOH can be removed while K⁺ cations are retained in the activated carbons. They combined this strategy with N-doping to prepare an activated carbon, and the resulting material (KNC-A-K) demonstrated exceptional CO₂ adsorption capabilities and CO₂/N₂ selectivity, especially at low pressures. Specifically, KNC-A-K exhibited CO₂ uptake of 1.62 mmol g⁻¹ at 298 K and 0.1 bar, far exceeding the CO₂ adsorption capability of most reported carbon material to date. The IAST CO₂/N₂ selectivity of KNC-A-K was 48, which also significantly surpasses the selectivity of conventional carbon materials. The high mixed-gas separation performance and excellent reusability of KNC-A-K were also well demonstrated in the column breakthrough experiments. As a consequence of the combination of extra-framework cations and N-doping, KNC-A-K showed a Q_st value of >50 kJ mol⁻¹, which is much higher than that of pure carbon adsorbents (<20 kJ mol⁻¹) and N-doped carbons (∼30 kJ mol⁻¹). In this work, the interactions of various possible polar species in the carbon adsorbent with CO₂ were also investigated by theoretical calculations. The results showed that these species have different degrees of promotion effect for the CO₂ adsorption through electrostatic interactions, among which K⁺ ions play the most remarkable role (Fig. 4).⁹⁷

Fig. 4 (a) Electrostatic potential surfaces of carbon clusters terminated with different functional groups: –PyN, –NH₂, –OK, –PyN·HCl, –NH₂·HCl, and –OH. (b) Optimized configurations of CO₂ adsorption on carbon clusters with different polar groups (cyan, C; white, H; blue, N; red, O; light green, K) and the corresponding contour plots of the differential charge density. The contour value is ±0.001 au. The purple and lime regions represent the charge accumulation and charge depletion regions, respectively. Reprinted with permission from ref. 97. Copyright 2012, American Chemical Society.

3.1.2. Pore structure engineering of carbon materials. Besides the modification and functionalization of the carbon materials by dopants, surface groups, and extra-framework cations, great efforts have also been made to engineer their pore structures, in order to control the pore sizes, pore geometries, pore connectivities, surface areas, pore volumes, etc. The porous structure of the carbon material can be fabricated by chemical/physical activation, or by “templating” synthesis, or by using a porous precursor for carbonization. It is desired to have a hierarchical pore structure with a large amount of uniform ultramicropores and a high interconnectivity, which would favour the CO₂ adsorption and separation from both thermodynamic and kinetic perspectives.
(i) The activation of carbon materials. “Activation” is an effective route to achieving large surface areas and high porosities in carbon materials, which can be performed in two ways: physical activation and chemical activation. Actually, the two activation methods both involve physical and chemical processes, whereas the difference lies in the activating agents used, which are various oxidizing gases (e.g., air, O₂, CO₂, H₂O, or their mixtures) for physical activation and chemical compounds (e.g., KOH, H₃PO₄, or ZnCl₂) for chemical activation. The activation mainly includes the following processes: (i) etching of the carbon framework by the redox reactions between various activating reagents with carbon, generating micropores; (ii) gasification of the carbon by the generated gases, such as H₂, H₂O, CO, and CO₂, further developing the porosity; (iii) when the carbon is chemically activated by strong bases, the formed alkali metal (e.g., K) can intercalate into the carbon lattice, the removal of which leaves behind additional porosity.

The textural properties of the activated carbon are highly dependent on the activation conditions (i.e., activation temperature, time, and the types and amounts of the activating reagent). Extremely high specific surface area (1000–4000 m² g⁻¹) can be achieved by activation. However, as discussed earlier, the overall surface area of the sorbent is not crucial for the CO₂ adsorption capacity at low pressures. The more important factors are the surface area of ultramicropores and the surface property of the carbon sorbents. To maximize the content of ultramicropores in the sorbent, the activation conditions should be carefully optimized. Harsh activation conditions generally give large surface areas, but in the meanwhile, resulting in the loss of ultramicropores and the elimination of heteroatoms from the carbon framework. Therefore, mild conditions (<923 K) are usually chosen for activating carbon sorbents, especially N-doped ones, when they are prepared for CO₂ adsorption.^{69,91,97,104,121} An exception was reported by Uyama et al., who prepared highly porous N-doped activated carbon monoliths (ACMs) by physical activation (using CO₂) of mesoporous polyacrylonitrile monoliths at 900–1000 °C. Despite a low content of N (1.8 wt%) and a small amount of ultra-micropores due to the high activation temperature, the optimized sorbent (ACM-5) exhibited exceptionally high CO₂ uptake: 5.14 mmol g⁻¹ at ambient pressure and temperature and 11.51 mmol g⁻¹ at ambient pressure and 273 K (Fig. 5).⁹⁸ The authors simply attributed the unprecedented CO₂ adsorption capacity of ACM-5 as well as its high Q_st to its large surface area, which is not reasonable in our view. Similarly high CO₂ adsorption capacity has not been observed in any other carbonaceous sorbents.

Fig. 5 (a) CO₂ adsorption isotherms of ACM-3 and ACM-5 at 273 and 298 K. (b) Isosteric heat of CO₂ adsorption (Q_st) as a function of the adsorbed amount of CO₂ for the two adsorbents.⁹⁸

(ii) “Templating synthesis” of porous carbons. Porous carbon materials can also be synthesized through a “hard-templating” route, where carbon precursors are polymerized and carbonized in the pore systems of a porous silica material that is then selectively removed by a NaOH or HF solution. Both microporous zeolites and mesoporous silica materials have been used as the template for fabricating porous carbon structures.^69,106 Since the carbon material replicates the porous structure of the template (it can be seen as a negative structural replica of the template), it possesses significant porosity in its as-synthesized form even without activation. If need be, activation is applied to further increase the porosity and surface area.⁶⁹

Zeolite-templated carbons have large surface areas mainly arising from micropores with narrow size distributions, and are therefore of great potential for CO₂ adsorption. Different zeolites have been used to fabricate microporous N-doped carbons. For example, Zhuo et al. used zeolite NaY as a hard template and furfuryl alcohol/acetonitrile as precursors to prepare N-doped microporous carbons. The maximum CO₂ adsorption capacity (∼2.5 mmol g⁻¹) was achieved by controlling the carbonization conditions to maximize the N content and the micropore volume.¹⁰⁶ Using zeolite EMC-2 as a hard template, Xia et al. prepared a N-doped zeolite-template carbon (CEM-750) with acetonitrile. CEM-750 has a very large surface area of 3360 m² g⁻¹, 85% of which is contributed by micropores, and its nitrogen content is as high as 5.2% from XPS analysis. As a consequence, it exhibits a high CO₂ adsorption capacity of 4.4 mmol g⁻¹ at 298 K and 1 bar.⁹⁹ It is worth pointing out that zeolite-templated carbons are usually prepared by chemical evaporation deposition (CVD) method to facilitate the transfer of carbon precursors in the small pores of zeolites, which limits the large-scale production of these materials. When mesoporous silicas are employed as template, easy and low-cost solution methods can be used to synthesize porous carbons, but further activation process is usually needed to generate sufficient microporosity for CO₂ adsorption.⁶⁹ Besides zeolites and mesoporous silicas, colloidal silica particles prepared by the Stöber method has also been used as template to synthesize N-doped carbon materials for CO₂ adsorption, where resorcinol–formaldehyde resin acted as the carbon precursor and melamine acted as the nitrogen source. Upon the removal of the silica by etching, N-doped hollow carbon spheres (N-HCSs) with uniform size of 220 nm were obtained. This material possesses a high nitrogen loading content of 14.8 wt% and exhibits a considerable performance for CO₂ capture with a capacity of 2.67 mmol g⁻¹ and a high CO₂/N₂ selectivity.¹⁰⁰

The self-assembly of amphiphilic molecules (surfactants or block copolymers) has also been utilized to fabricate ordered mesoporous carbon materials and this synthetic strategy is referred to as the “soft templating” route. Yuan et al. synthesized ordered mesoporous carbon materials by using resorcinol and formaldehyde as the carbon precursors and tri-block copolymer Pluronic F127 as the template. The resulting materials were further treated in ammonia flow at the temperature of 723–1273 K to dope nitrogen into the carbon frameworks. One of the final materials exhibited a CO₂ adsorption capacity of 3.46 mmol g⁻¹.¹⁰² Zhao et al. reported a simpler one-pot method to synthesize N-doped ordered mesoporous carbons, where dicyandiamide (nitrogen source), resol molecules (carbon precursor), and Pluronic F127 template co-assemble via hydrogen bonding and electrostatic interactions to form ordered porous structures. The obtained carbon materials possess tunable pore sizes (3.1–17.6 nm) and surface areas (494–586 m² g⁻¹), high N contents (up to 13.1 wt%), and CO₂ adsorption capacities of 2.8–3.2 mmol g⁻¹ at 298 K and 1.0 bar.¹¹¹

Dai et al. prepared nitrogen-enriched ordered mesoporous carbons by pyrolysis of the soft-templated polymeric composites in ammonia. This approach takes advantage of the preferential reaction and/or replacement of oxygen with nitrogen species that are generated by decomposition of ammonia at elevated temperatures. It combines carbonization, nitrogen functionalization, and activation into one simple process, generating N-doped carbons with large surface areas (up to 1400 m² g⁻¹), and high nitrogen contents (up to 9.3 wt%).¹⁰⁷ However, the CO₂ adsorption performance of these materials was not evaluated. Lu et al. synthesized porous carbon nanosheets with variable thicknesses by using graphene oxide sheets as the structure directing agent. Although the resulting materials show impressive CO₂ uptakes per unit surface area (3.54 CO₂ molecules per nm²), high CO₂/N₂ selectivity, and excellent moisture tolerance, they suffer from their low surface areas and thus give unremarkable overall CO₂ adsorption capacities (<3 mmol g⁻¹ at 298 K and 1.0 bar).¹²²

(iii) Preparation of porous carbons from porous solid precursors. A new preparative strategy was recently proposed, which employs preformed microporous carbon-containing materials, such as metal–organic frameworks (MOFs) and porous organic polymers (POPs) as precursors to prepare microporous carbons via direct pyrolysis. The advantage of this strategy is that the development of porosity in the carbon material is on the basis of the microporous structure of the precursor, which can be precisely controlled and well defined. A good example was recently reported by Srinivas and co-workers. They carbonized a series of MOFs to yield porous carbons with large surface areas (up to 2734 m² g⁻¹) and exceptionally high pore volumes (up to 5.53 cm³ g⁻¹). These materials have hierarchically porous structures, simultaneously containing micropores, mesopores, and macropores. The micropores are mainly formed by structural inheritance from the MOF precursor, whereas meso- and macro-pores are generated from defects in the crystals. The resulting materials (HPCs) show a significant amount of CO₂ adsorption at high pressure (>27 mmol g⁻¹ at 30 bar and 300 K), and thus great potential for CO₂ storage applications.¹²³ However, they only show ordinary low-pressure CO₂ adsorption capacities, which is reasonable considering that they are comprise of nearly pure carbon.

Qiu et al. combined the carbonization and chemical activation processes to convert large-surface-area PAF-1 to microporous carbon. Despite being a pure carbon material, the activated PAF-1 shows high CO₂ adsorption ability at both high-pressures and low-pressure regions, possibly because it contains a significant amount of ultra-micropores of 0.6 nm.¹²⁴ In order to further narrow down the pore size of the PAF-derived carbon, an extra carbon source was pre-introduced into a PAF followed by carbonization, which afforded a new microporous carbon material with a smaller pore size of 0.54 nm, thus facilitating a high CO₂ uptake capacity of 4.1 mmol g⁻¹ at 295 K and 1 bar.¹²⁵

Han et al. synthesized a perfluorinated covalent triazine-based framework (FCTF-1), which exhibits excellent CO₂ adsorption and separation properties due to the incorporation of polar and hydrophobic C–F bonds in the material (refer to Section 3.2.2 for more detailed discussions). A mild pyrolysis at 873 K was applied to convert FCTF-1 to a carbon material (FCTF-1-600). It is interesting to note that during this process, a significant amount of ultramicropores were generated in the material due to the elimination of F groups. Since these ultramicropores have pore sizes (0.46 nm) approaching the diameter of a N₂ molecule (0.364 nm), they could offer marked diffusion selectivity in separating CO₂ and N₂ under non-equilibrium conditions (i.e., more diffusion constraints for N₂ relative to CO₂). This is confirmed by the fact that FCTF-1-600 exhibited a very high CO₂/N₂ selectivity (152) in the breakthrough experiment although its IAST selectivity was only 19. This work well demonstrated the important effect of the pore size of the adsorbent on the gas separation under kinetic flow conditions.⁵³

3.1.3. Summary. Simultaneous optimization in both the framework composition and the pore structure is needed for a rational design of a carbon sorbent for the applications of CO₂ adsorption and separation. Although diverse synthetic strategies have been used to prepare porous carbon sorbents, it is generally accepted that the adsorbant/adsorbate interaction and the surface area of ultramicropores (instead of the BET surface area) are the two most important factors for the low-pressure CO₂ adsorption and separation. It is quite often to see that there is a trade-off between the two factors. For example, an activation process can produce higher porosity in the carbon, while it sometimes also leads to the elimination of surface functionality. When dynamic separation of CO₂ from a mixed gas is concerned, the pore size of the sorbent is another important factor that would influence both the adsorption kinetics and the separation selectivity. To achieve the selectivity predicted from equilibrium adsorption isotherms, fast adsorption kinetics is required. On the other hand, uniform pores with critical sizes to offer the “molecular sieve” effect are highly desirable for gas separation. As introduced above, some carbon sorbents (e.g., mesoporous silica-templated carbons and MOF-derived carbons) have hierarchically porous structure, containing significant meso- or/and macro-porosity in addition to the microporosity. Although the CO₂ adsorption capacity is mainly contributed by the micropores, the presence of large pores in the sorbent would allow for fast adsorption kinetics. Unfortunately, this potential benefit has not been carefully investigated (nor even been pointed out) in the current studies. Meanwhile, for the majority of the reported AC sorbents, their CO₂/N₂ and CO₂/CH₄ selectivities are determined from single-gas adsorption using the IAST method, whereas their mixed-gas separation performances under kinetic flow conditions are rarely evaluated possibly due to the lack of facilities. As a consequence, the effect of the pore size on diffusion selectivity has been overlooked in most current studies.

3.2. Porous Organic Polymers (POPs)

3.2.1. A brief introduction to POPs. We here use the term POPs to generally refer to the metal-free organic porous structures linked by covalent bonds, which can be crystalline or amorphous. According to the synthetic methods and the structural/compositional features of the materials, POPs can be classified into different families, such as Covalent Organic Frameworks (COFs), Covalent Triazine Frameworks (CTFs), Porous Aromatic Frameworks (PAFs), and Conjugated Microporous Polymers (CMPs).

As a class of the lightest materials, POPs have attracted great research attention in the last decade. Some representatives are briefed as follows. COFs are firstly synthesized by Omar Yaghi in 2005, and their development opened up a new research direction of porous materials.⁴⁰ COFs feature crystalline structures, which are fabricated on the basis of the judicious choice of building blocks and the proper use of reversible condensation reactions. The synthesis of COFs is based on the molecular dehydration reaction, in which three boronic acid molecules converge to form a planar six-membered B₃O₃ (boroxine) ring with the elimination of three water molecules. Both two-dimensional and three-dimensional crystalline COFs have been synthesized. COFs usually possess large surface areas and pore volumes, and accordingly excellent gas storage performance. For example, one three-dimensional COF, COF-102, has a surface area of ∼3500 m² g⁻¹, and very high gas uptake capacities (1180 mg g⁻¹ for CO₂ at 35 bar and 298 K).⁴² Porous Aromatic Frameworks (PAFs) were firstly reported by Qiu's group.^46,47 A representative material in this family is PAF-1, which is composed of carbon and hydrogen only and shows extremely high surface area (S_BET = 5460 m² g⁻¹). PAF-1 was synthesized by using tetrakis(4-bromophenyl) methane as the tetrahedral building unit, and the phenyl rings were coupled using the nickel (0)-catalyzed Yamamoto-type Ullmann cross-coupling reaction. Conjugated microporous polymers (CMPs) are a class of organic porous polymers that combine p-conjugated skeletons with permanent nanopores.¹⁵¹ The most characteristic feature of CMPs is the rather broad diversity of usable building blocks ranging from simple phenyl units to extended arenes, heterocyclic aromatic units, and large macrocycles, which allows for easy tuning and optimization of their textural and compositional properties for different applications.^54,152 In general, being mainly comprised of light elements (C, H, O, B, N etc.), most POPs have large surface areas and high pore volumes (e.g. COF-102 and PAF-1), and are therefore potentially useful for gas storage. On the other hand, however, POPs have lower intrinsic affinities to CO₂, in comparison with MOFs, zeolites, and inorganic carbon materials, because their organic and metal-free frameworks do not contain strong adsorption sites. Therefore, modifications and special design of POPs are necessary to enhance their CO₂ adsorption, and similar to the cases of inorganic carbons, incorporating heteroatoms, functional groups, or metal ions is the main strategy for this purpose. Representative examples of various POP materials along with their CO₂ adsorption and separation data are summarized in Table 2.

Table 2 Summary of representative organic POPs with their textural and CO₂ adsorption properties

Materials	BET m^2 g^−1	CO2 capacity@273 K, mmol g^−1	CO2 capacity@298 K, mmol g^−1		CO2/N2 (1 : 9) selectivity		CO2/CH4(1 : 1) selectivity		Qst kJ mol^−1		Ref.
		1 bar	1 bar	0.1 bar^a	273 K	298 K	273 K	298 K	CO2	CH4
a The values were estimated from the published isotherms in case they were not specified in the literature.b 293 K.c CO2 : N2 ratio of 0.15 : 0.85 at 1 bar.d By initial slops method.e Data of 0.15 bar.f CO2 : N2 ratio of 0.05 : 0.95.g 0.15 bar; 295 K.
PAF-1	5600	2.05	1.09		—	—	—	—	15.6	14	48
PAF-20	702	1.16	0.61		—	—	—	—	30.5	—	126
CTF-1	746	2.47	1.41	0.21	—	20	—	—	27.4	—	81
FCTF-1	701	4.67	3.21	0.92	—	31	—	—	32.9	—	81
fl-CTF-350	1235	4.28	2.29	0.62	—	23	—	—	32.7	—	127
NPTN-2	1558	3.18	1.56	0.24	22	—	—	—	37	—	128
NOP-21	565	2.79	—	—	81	—	—	—	37	—	129
PCTF-5	1180	2.59	1.5^b	0.42	32	—	7	—	27	20.5	130
NOP-6	720	1.31	0.48	0.03	38.7^c	—	—	—	29.2	—	131
TPI-1	809	2.45	1.25	0.23	—	30.9	—	—	34	—	132
IBTP	328	1.77	1.14	0.37	51^d	—	6.3^d	—	37.8	20.7	64
PSN-1	1045	3.41	—	—	—	—	—	—	31.4	—	133
PSN-3	865	3.1	2.0	0.38	88^d	69^d	16^d	10^d	36.7		134
PAN-1	925	3.4	2.1	0.64^e	35^c	—	7	—	33.3		72
PAN-2	1242	4.0	2.9	1.02^e	50^c	—	10	—	39.3		72
SNW-1	821	—	2.08	0.49	—	50^d	—	15	35	25	63
Fe-POP-1	875	4.30	—	—	—	—	—	—	—	—	135
HPE-CMP	662	3.58	1.70	0.39	23.8	—	7.9	—	33.2	—	136
TB-MOP	694	4.05	2.57	0.59	45.2^d	50.6^d	—	—	29.5	—	62
BILP-4	1135	5.34	3.59	0.75	79^d	32^d	10^d	7^d	28.7	13	137
CPOP-1	2220	4.82	—	—	25	—	33	—	—	—	138
Cz-POF-3	1927	4.77	3.05	—	20^c	—	5.6^f	—	27.8	20.2	75
Azo-COP-1	635	2.44	1.48	0.34	—	96.6	—	—	29.3	—	139
CMP-1–COOH	522	1.60	0.94	0.16	—	—	—	—	32.6	—	140
PPN-6–CH2DETA	555	—	4.3	2.27^g	—	329	—	—	56	—	141
2	525	1.50	0.95	0.78^g	—	155	—	—	50	—	142
BINOL-HCP-4	1015	—	2.27	—	—	22.7^d	—	—	29.8	—	143
APOP-1-F	724	3.07	2.02	0.63	43.4^d	31.8^d	6.4	6.3	33.3	25	144
PPN-6-SO3Li	1186	—	3.7	1.25^g	—	414	—	—	35.6	—	145
PAF-18-OLi	981	—	2.02	0.49	—	45	—	—	29.5	—	146
POF1B	917	—	2.14	0.45	—	—	—	—	—	—	147
PAF-26–COOK	430	2.5	—	—	—	50^c	—	8.6	32.6	23	148
PAF-26–COOMg	572	2.7	—	—	—	73^c	—	8.4	30	21.5	148
F-PAF-50	580	1.02	—	—	—	—	3.02	—	—	—	149
2I-PAF-50	91	0.29	—	—	—	—	29.8	—	—	—	149
C	1237	—	2.20	0.33	—	19.8	—	—	33.7	—	150

---

3.2.2. POPs with nitrogen–carbon hybrid framework. POPs with nitrogen–carbon hybrid frameworks are of greater potential as sorbents for CO₂ adsorption, considering the polarity and alkalinity associated with the N species. Covalent triazine frameworks (CTFs) constitute one class of such materials, which are synthesized by dynamic trimerization reaction of aromatic nitriles in ionothermal conditions (molten zinc chloride at 673 K). CTF-1 is a typical example, which has a crystalline structure and a narrow pore distribution (Fig. 6).¹⁵³ CTF-1 does not show a high CO₂ uptake (1.41 mmol g⁻¹ at 298 K and 1 bar), possibly due to its large pore size as well as to the weak alkalinity of the triazine groups. However, it indeed gives an impressive CO₂/N₂ IAST selectivity of 20.⁵³

Fig. 6 Reaction schemes and ideal structures of CTFs synthesized through the trimerization of (a) terephthalonitrile (CTF-1) and (b) tetrafluoroterephthalo-nitrile (FCTF-1).

In order to optimize the selective CO₂ capture performance of CTF-1, Han et al. designed and synthesized a perfluorinated CTF-1 (denoted as FCTF-1) by using tetrafluoroterephthalonitrile (all the hydrogen atoms of terephthalonitrile replaced by fluorine atoms) as the monomer (Fig. 6). The incorporation of fluorine (F) groups played multiple roles in improving the framework's CO₂ adsorption and separation capabilities. Thermodynamically, the introduction of polar C–F bonds promoted the CO₂ uptake via electrostatic interaction (4.67 mmol g⁻¹ at 273 K and 1 bar), especially at low pressures (1.76 mmol g⁻¹ at 0.1 bar). In the meanwhile, incorporating F groups reduced the effective pore diameter down to ultra-micropores region, which offered high gas adsorption potential as well as kinetic (diffusion) selectivity for CO₂/N₂ separation. Consequently, it was interesting to see that in mixed-gas breakthrough experiments, FCTF-1 exhibited an exceptional CO₂/N₂ selectivity of 77 under kinetic flow conditions, much higher than the IAST selectivity (31) derived from single-gas equilibrium adsorption. Moreover, the (i) covalent triazine frameworks (CTFs) hydrophobic nature of the C–F bonds endows FCTF-1 with excellent tolerance to water and its CO₂ capture performance remained excellent when there was moisture in the gas mixture. In addition, the moderate adsorbate–adsorbent interaction (Q_st: 30–35 kJ mol⁻¹) allowed FCTF-1 to be fully regenerated by pressure swing adsorption processes.⁵³

A fluorene-based covalent triazine framework (fl-CTF, Fig. 7a) was synthesized at different temperatures (623–873 K). The material synthesized at 623 K has a high fraction of micropores with a surface area of 1235 m² g⁻¹, and it therefore shows the highest CO₂ sorption capacity (4.28 mmol g⁻¹ at 273 K and 2.29 mmol g⁻¹ at 298 K). Its CO₂/N₂ IAST selectivity is 23. However, it did not exhibit high CO₂ adsorption capacity at low pressures (0.5 mmol g⁻¹ at 298 K and 0.1 bar), and its mixed-gas separation performance was not evaluated.¹²⁷

Fig. 7 The synthetic schemes of (a) various triazine-based frameworks and (b) PCTF-5.

Other triazine-based frameworks include NPTN-1 fabricated from 3,6-dicyanocarbazole, NPTN-2 from 3,6-dicyanodibenzofuran, and NPTN-3 from 3,6-dicyanodibenzothiophene (Fig. 7a). Their CO₂ adsorption capacities are 1.84, 1.56 and 1.34 mmol g⁻¹, respectively.¹²⁸ A series of carbazole-modified triazine-based frameworks were also synthesized, where the pore size and surface properties were tuned at the same time by using pre-functionalized monomers (Fig. 7a). The phenyl-anchored framework (NOP-21) exhibited a CO₂ adsorption capacity of 12.3 wt% (2.79 mmol g⁻¹) at 273 K and 1 bar, while the ethyl acetate-appended framework (NOP-20) gave a high CO₂/N₂ selectivity (81) at the same condition, which was attributed to the uniform ultra-micropores.¹²⁹ Adamantane substituted with two to four 4-cyanophenyl groups was also used to prepare a new series of robust Porous Covalent Triazine-based Framework (PCTF) materials by the trimerization reaction (Fig. 7b). PCTF shows high CO₂ selectivity over N₂ and CH₄ (CO₂/N₂: 41; CO₂/CH₄: 7, at 273 K), as determined by the IAST method. The Q_st values of PCTFs for CO₂ are >25 kJ mol⁻¹ over the entire adsorption range, which are well above the heat of liquefaction of bulk CO₂ or the isosteric enthalpy of adsorption for CO₂ on activated carbons.¹³⁰

Besides the trimerization reaction of nitriles, covalent triazine-based frameworks can also be prepared by using triazine-containing monomers. The polymerization of 2,4,6-trichloro-1,3,5-triazine with benzene, biphenyl, and terphenyl by using aluminum chloride as the catalyst renders three porous polymers (2, 3, and 4 in Fig. 8a). Their surface areas are in the range of 558 to 1266 m² g⁻¹, depending on the aromatic linker length. At ambient pressure and temperature, these polymers exhibit CO₂ uptakes of 1.69–2.28 mmol g⁻¹ with the CO₂/N₂ selectivity of 20–49, but the dynamic flue gas separation selectivity was not evaluated.¹⁵⁴ Likewise, a series of porous polymers (denoted as NOP-1–6) was synthesized via a methane–sulfonic acid-catalysed Friedel–Crafts reaction of 2,4,6-trichloro-1,3,5-triazine with various building blocks (Fig. 8a). Among them, NOP-3 with a specific surface area of 894 m² g⁻¹ exhibited the highest CO₂ uptake of 2.50 mmol g⁻¹ at 273 K and 1.0 bar, while NOP-6 showed a high CO₂/N₂ selectivity of 38.7 under the same conditions. Their performances at room temperature were not evaluated.¹³¹

Fig. 8 Synthetic routes to the CTF-based porous polymer by using triazine-contained precursors.

Another series of CTF materials is the triazine-based porous polyimides (TPIs), which were synthesized by a condensation reaction of 2,4,6-tris(4-aminophenyl)-1,3,5-triazine and various dianhydride building blocks in m-cresol (Fig. 8b). As revealed by argon sorption isotherms, the BET specific surface area of this class of material can reach ∼800 m² g⁻¹ and their pore diameters range from 0.4 to 3 nm. The highest uptake values for CO₂ (2.45 mmol g⁻¹ at 273 K and 1 bar) were observed for TPI-1 and TPI-2, while the highest binding selectivity (56) for CO₂ over N₂ at 298 K was observed for TPI-7. The high degree of functionalization led to high CO₂ adsorption heats for TPIs between 29 kJ mol⁻¹ (TPI-6) and 34 kJ mol⁻¹ (TPI-1).¹³²

(ii) Schiff base. Schiff-base chemistry has been one of the most frequently used approaches to construct organic porous materials. This route features the low cost of synthesis and high product yield, while the reversibility of the imine bond formation allows for the potential ‘error-correction’ behaviour during the synthesis.¹⁵⁵ One example is shown in Fig. 9, in which a benzothiazole linked nanoporous polymer, imine benzothiazole polymer (IBTP), was synthesized via a Schiff-base condensation reaction. IBTP shows a high Q_st for CO₂ (37.8 kJ mol⁻¹), reflecting a strong affinity to CO₂, and accordingly gives a good CO₂ selectivity over N₂ (51) and CH₄ (6.3). However, its CO₂ uptake was only 1.8 mmol g⁻¹ at 273 K and 1 bar.⁶⁴ Other Schiff-based networks include PSN-1 and PSN-2, which have rigid and bulky tetraphenyladmantane as the net node, interconnected by phenyls with C=N groups linked at meta- and para-positions, respectively (Fig. 9).¹³³ PSN-1 has a BET surface area of 1045 m² g⁻¹ with a pore size of around 0.7 nm, whereas the BET surface area of PSN-2 is only 376 m² g⁻¹ with the major pore of 2.2 nm. Compared to PSN-2, PSN-1 can uptake significantly higher CO₂ (3.41 vs. 1.34 mmol g⁻¹, at 273 K and 1 bar). Notably, despite the very similar chemical structure and the same nitrogen content in the two materials, the Q_st value of PSN-1 is larger than that of PSN-2 (31.4 kJ mol⁻¹ vs. 30.0 kJ mol⁻¹), indicating that the ultra-small pore channel in PSN-1 is also advantageous for the capture of CO₂. PSN-3 is constructed from 1,3,5,7-tetrakis(4-formylphenyl)adamantane and 1,3,5,7-tetrakis(4-aminophenyl)adamantine (Fig. 9). It has a BET surface area of 865 m² g⁻¹ with a major pore size around 0.6 nm. The CO₂ uptake of PSN-3 is 3.1 mmol g⁻¹ at 273 K and 1 bar, while the CO₂/N₂ selectivity and CO₂/CH₄ selectivity are 88 and 16, respectively. However, it should be pointed out that such high selectivities are determined by from the initial slopes of the adsorption isotherms for different gases instead of the more generally accepted IAST method.¹³⁴ Tetraphenyladamantane-based polyaminals (PAN-1 and PAN-2) have been successfully synthesized through polycondensation reaction (Fig. 9). In the case of PAN-1, only one amino was involved in the aminal reaction, while the reactions of other two amino groups with aldehydes lead to imine bonds. In the case of PAN-2, all the three amino groups react with aldehydes to yield aminal linkages. The PAN-2 has the smaller pore size, larger surface area and higher pore volume than does PAN-1. At 273 K and 1 bar, PAN-2 displays a heat of adsorption of 39.3 kJ mol⁻¹ and a CO₂ capacity of 4 mmol g⁻¹. At 0.15 bar, the CO₂ uptake of PAN-2 is as high as 1.97 mmol g⁻¹. Moreover, its IAST selectivities for CO₂/N₂ and CO₂/CH₄ reach to 50 and 10 at 273 K, respectively.⁷²

Fig. 9 Basic structural units of IBTP, PSNs and PANs.

A porous polymer (SNW-1) containing both Schiff base and triazine groups was recently reported.⁶³ SNW-1 has a high adsorption capacity for CO₂ with an uptake of 2.23 mmol g⁻¹ at 298 K and 1 bar. More importantly, SNW-1 can be fabricated into a membrane for the separation of mixed-gases, which exhibits a specific selectivity toward CO₂ : CO₂/CH₄ selectivity of 15 and CO₂/N₂ selectivity of 50. The presence of uniform ultramicropores of 0.5 nm and N–H moieties in the skeleton was considered to account for the excellent CO₂ adsorption properties of SNW-1. The high selectivity was explained as the consequence of the preferential adsorption of CO₂ over N₂ and CH₄.

(iii) Other N-containing porous polymers. Porphyrin-containing porous frameworks constitute another important family in POPs. An extended aromatic substitution reaction between pyrrole and aromatic dialdehydes in the presence of small amount of Fe(III) gives rise to porous materials (Fe-POPs) (Fig. 10), which possess surface areas associated with micropores (750–875 m² g⁻¹) and show excellent CO₂ capture ability (up to 4.30 mmol g⁻¹ at 273 K and 1 bar). It was proposed that the basic porphyrin subunits in the polymeric network promote the adsorption of acidic CO₂ molecules.¹³⁵ Porphyrin-based conjugated microporous polymer (HP_E-CMP) with rich nitrogen sites in the skeleton has also been synthesized by alkyne–alkyne homocoupling reaction (Fig. 10). For HP_E-CMP, a BET specific surface area up to 662 m² g⁻¹ and a pore volume of 0.55 cm³ g⁻¹ have been achieved. At 273 K, HP_E-CMP displays a CO₂ uptake capacity of 3.58 mmol g⁻¹ at 1 bar, CO₂/N₂ selectivity of 23.8, and CO₂/CH₄ selectivity of 7.9.¹³⁶

Fig. 10 Chemical structures of Fe-POPs and HP_E-CMP.

There are some other N-containing porous organic polymers with notable CO₂ adsorption or separation properties, which are synthesized by using different reactions and building blocks. For example, Tröger's-base-derived microporous organic polymer (TB-MOP) with a surface area of 694 m² g⁻¹ and good thermal stability was synthesized from a one-pot metal-free polymerization reaction between dimethoxymethane and triaminotriptycene (Fig. 11). The CO₂ uptake can reach up to 4.05 mmol g⁻¹ and 2.57 mmol g⁻¹ at 273 K and 298 K under 1 bar, respectively. Moreover, it shows a high selectivity toward CO₂ over N₂ (50.6) at 298 K.⁶² A series of porous benzimidazole-linked polymers (BILPs) were synthesized by the condensation reaction between aryl-o-diamine and aryl-aldehyde building blocks. The highest selectivity based on initial slope calculations (not the IAST selectivity) at 298 K was observed for BILP-2: CO₂/N₂ (71) and CO₂/CH₄ (7), while the highest CO₂ uptake was given by BILP-4: 3.59 mmol g⁻¹ at 298 K and 1 bar.¹³⁷ Oxidative coupling polymerization was used to generate a microporous polycarbazole net (Fig. 12). The resulting material, CPOP-1, exhibits a specific surface area up to 2220 m² g⁻¹, and a CO₂ uptake capacity of 4.82 mmol g⁻¹ at 273 K and 1 bar. It is worth highlighting that CPOP-1 shows an extraordinarily high CO₂/CH₄ selectivity (33), far exceeding the CO₂/CH₄ selectivity of most reported POP adsorbents, which are usually less than 10. However, the origin of such unusual selectivity was not discussed and the method used for the selectivity calculation was not elaborated in the paper.²⁷

Fig. 11 Chemical structures of various N-containing porous organic polymers.

Fig. 12 Chemical structures of CPOP-1 and Cz-POFs.

Similar carbazolic porous organic frameworks (Cz-POFs) were synthesized by using bulky, dendritic building blocks with high connectivity (Fig. 12). Specifically, Cz-POF-1 and Cz-POF-3 possess high surface areas of 2065 and 1927 m² g⁻¹, respectively. At 1 bar and 273 K, Cz-POF-3 exhibits the highest CO₂ uptake of 4.77 mmol g⁻¹. However, it is interesting to note that the CO₂/CH₄ selectivities of Cz-POFs (4.4–7.1) are much lower than the claimed CO₂/CH₄ selectivity for CPOP-1 (33, as discussed above), given their comparable surface areas and similar structural moieties in the framework (Fig. 12).⁷⁵

Among various N-containing POPs (azo-COPs), one series of materials reported by Yavuz et al. deserves to be highlighted. They are azo-bridged aromatic polymers synthesized by catalyst-free direct coupling of aromatic nitro and amine moieties under basic conditions (Fig. 13). What is unique with this work is that unlike almost all other carbon-based adsorbents, the azo-covalent organic polymers exhibit an unprecedented increase in CO₂/N₂ selectivity with increasing temperature. Considering that the temperatures of flue gases are higher than 323 K, a high CO₂/N₂ selectivity at elevated temperature is very important for the real separation application. It was revealed that azo groups reject N₂, thus making the framework N₂-phobic. Monte Carlo simulations suggest that the origin of the N₂-phobicity of the azo-group is the entropic loss of N₂ gas molecules upon binding, although the adsorption is enthalpically favourable. It was suggested that incorporation azo units in the adsorbent would favour any gas separations that require the efficient exclusion of N₂ gas.¹³⁹ The problem with these materials is that they show rather low CO₂ adsorption capacity. However, this work indeed provides new idea for the separation of N₂ and is particularly useful for the design of adsorbent to separate high-temperature gas mixture. It is anticipated that introducing azo group in other adsorbents with intrinsically high CO₂ capacity would effectively enhance their CO₂/N₂ selectivities.

Fig. 13 Synthesis route to azo-COPs.

3.2.3. POPs containing functional groups. Great efforts have also been made to promote the CO₂ adsorption in POPs by specially designed building blocks with various functional groups. Cooper et al. investigated the CO₂ adsorption properties of a series of conjugated microporous polymer (CMP) networks (Fig. 14), which incorporate a range of chemical functionalities including carboxylic acids, amines, hydroxyl groups, and methyl groups. The carboxylic acid functionalised network, rather than its amine analogue (note that aniline is a weak base), shows the highest Q_st for CO₂ (32.6 kJ mol⁻¹) at low coverage. However, the CO₂ capacities of CMPs are not impressive (0.9–1.2 mmol g⁻¹ at 298 K and 1 bar), in comparison with other adsorbents, and the CO₂ selectivity of these materials was not investigated.¹⁴⁰

Fig. 14 Chemical structures of CMPs.

Post-synthesis functionalization of POPs has also been employed to improve their CO₂ adsorption. The extremely large surface area allow PAF-1 to act as a platform for functionalization because it can retain sufficient surface area, which may be even higher than that of most other as-synthesized POPs, after the functionalization. The good thermal stability is another important feature that motivates people to use PAF-1 for modification. Various strategies that have been employed to functionalize PAF-1 for gas adsorption are illustrated in Fig. 15.

Fig. 15 The scheme represents the functionalization processes of PAF-1 (PPN-6).

Primary amine-functionalized PAF was prepared by using phthalimide-containing monomer, followed by a post-synthesis deprotection step. The resulting material, PAF-1–CH₂NH₂, can uptake as high as 4.37 mmol g⁻¹ CO₂ at 273 K and 1 bar, significantly surpassing that of PAF-1.¹⁵⁶ Zhou's group grafted various polyamines onto PPN-6 (PAF-1), giving rise to materials with excellent CO₂ adsorption characteristics at low pressures. Although PPN-6–CH₂DETA has the lowest surface area, it exhibits the highest CO₂ uptake capacity of all the polyaminetethered PPNs. At 295 K, PPN-6–CH₂DETA's CO₂ uptake is 3.0 mmol g⁻¹ at 0.15 bar and 3.6 mmol g⁻¹ at 1 bar. Because of strong affinity ability to CO₂, PPN-6–CH₂DETA also possesses high selectivity (400) under flue gas condition.¹⁵⁷ Lately, PPN-6–CH₂DETA was used to capture CO₂ from air containing 400 ppm of CO₂, and it exhibited an extraordinarily high CO₂/N₂ selectivity (3.6 × 10¹⁰) according to the IAST calculation.¹⁵⁸ Eddaoudi's group reported the synthesis of a POP with exposed aldehyde groups, which can act as active sites for the subsequent grafting of amines (2 in Fig. 16). The substitution of aldehyde groups by ethylenediamine moieties leads to (i) a substantial increase in Q_st (from 33 to 50 kJ mol⁻¹ at low coverage), reflecting an enhanced adsorbent–adsorbate interaction; and (ii) enhanced qualitative CO₂/N₂ selectivity (155) and CO₂/CH₄ selectivity (around 50). However, its CO₂ adsorption capacity was only about 1 mmol g⁻¹ at 298 K and 1 bar.¹⁴²

Fig. 16 Schematic representation of the amine functionalization process.

Hydroxyl-containing microporous organic polymers, including some binaphthol (BINOL) networks, have been synthesized by Friedel–Crafts alkylation. In comparison with the hydrophobic analogues, these BINOL networks show higher CO₂ capture capacities under dry conditions but lower capacities in the presence of water vapour. These results suggest that idealized measurements may give a poor indication of performance under more realistic carbon capture conditions.¹⁴⁴

A series of aminal-linked porous organic polymers (APOPs) has been constructed by condensation reactions between diaminotriazine-based tectonics and benzaldehyde (Fig. 17).¹⁴⁴ This synthetic strategy allows the fabrication of a platform which incorporates functional groups in nitrogen-rich porous structure to tune the affinity between gas molecules and the frameworks. APOPs have moderate specific surface areas (724–1402 m² g⁻¹). Among the prepared APOPs, APOP-3 that has the highest surface area and a rigid tetrahedral network shows the highest measured CO₂ uptake (4.54 mmol g⁻¹ at 273 K and 1 bar). APOP-1 also exhibits a noteworthy CO₂ capacity of 4.26 mmol g⁻¹ under the same conditions. At 298 K and 1 bar, APOP-1 displayed highest CO₂ capacity (2.69 mmol g⁻¹). The incorporation of –F, –OH and –ONa into the framework was found to markedly increase the Q_st for CO₂ adsorption. At 298 K and 1 bar, APOP-1–F displayed the highest IAST selectivity for both CO₂/N₂ (31.8) and CO₂/CH₄ (6.3) gas pairs. However, their CO₂ adsorption capacities were decreased relative to the unmodified APOP-1.

Fig. 17 Synthesis route to various APOPs.

3.2.4. POPs with extra-framework ions. The role of extra-framework cations in promoting the CO₂ adsorption of adsorbents has been long recognized and various cations have been incorporated in zeolites,²³ MOFs,^115–118 and recently in a carbon material⁹⁷ for selective CO₂ adsorption.

The same strategy has been employed in POP-based adsorbents as well. A porous polymer network (PPN) grafted with sulfonic acid (PPN-6–SO₃H) and its lithium salt (PPN-6–SO₃Li) exhibit significant increases in Q_st of CO₂ adsorption and CO₂-uptake capacities. IAST calculations for a CO₂–N₂ mixture (15 : 85, 1 bar) at 295 K revealed their exceptionally high adsorption selectivity for CO₂ over N₂ (155 for PPN-6–SO₃H and 414 for PPN-6–SO₃Li). The authors emphasized the ultrahigh physicochemical stability of these PPN adsorbents, and thus their great potential for practical applications in post combustion capture of CO₂.¹⁴⁵ Likewise, PPN-6–SO₃NH₄ was demonstrated to have exceptionally high CO₂ capacity at low pressures and CO₂/N₂ selectivity. According to the “initial slope ratio” method, PPN-6–SO₃NH₄ has high selectivity towards CO₂ over other gases: CO₂/N₂ (196), CO₂/CH₄ (40), CO₂/H₂ (1722) and CO₂/CO (109).¹⁵⁹

A hydroxyl-containing PAF, PAF-18–OH, and its lithiated derivative PAF-18–OLi, have been successfully synthesized. In comparison with PAF-18–OH, PAF-18–OLi displays significantly enhanced CO₂ adsorption capacity (3.27 mmol g⁻¹ vs. 2.5 mmol g⁻¹ at 273 K and 1 bar) and CO₂/N₂ selectivity (129 vs. 34 at the same conditions) for simulated post-combustion flue gas mixtures (85% N₂ and 15% CO₂).¹⁶⁰ The same group reported a new porous aromatic framework decorated with carboxyl groups (PAF-26–COOH) and its modification with various light metals including Li, Na, K and Mg via a post-metalation approach. The cation-modified PAF-26 materials have strong affinity to CO₂ and CH₄, as reflected by their high Q_st values toward CO₂ (35 kJ mol⁻¹) and CH₄ (24 kJ mol⁻¹). After modification, the surface areas were decreased from 717 m² g⁻¹ to 430–590 m² g⁻¹. Accordingly, the absolute adsorption capacity for CO₂ (∼2.7 mmol g⁻¹ at 273 K and 1 bar for PAF-26–COOMg) was rather low in comparison with other cation-modified POPs and it is even lower than that of most of unmodified POPs (Table 2). After normalized by the total pore volume, the cation-modified materials (ca. PAF-26–COOK) demonstrated higher CO₂ capacity (572 mg cm⁻³) that that of parent PAF-26–COOH (286 mg cm⁻³). Meanwhile, these materials show low adsorption capacity for N₂. The highest IAST CO₂/N₂ selectivity (73) is given by PAF-26–COOMg at 298 K. In addition, marked CH₄/N₂ selectivities (∼9) were observed on PAF-26–COONa, K and Mg. However, it was noticed that after being exposed in wet air for three days, PAF-26–COOMg lost ∼50% of its original CO₂ adsorption capacity.¹⁴⁸

Besides various metallic cations, anions can also be used to modulate the gas adsorption property both thermodynamically and kinetically. Recently, Zhu et al. reported a family of quaternary pyridinium-type porous aromatic frameworks (X-PAF-50) with adjustable channel diameters. The fine tuning of the channel diameter in the range of 0.34–0.7 nm has been achieved by carefully choosing counter ions (F⁻, Cl⁻, Br⁻, and I⁻) as steric hindrance via a facile ion exchange approach (Fig. 18). These materials demonstrated reversible sorption–desorption. Their effective pore diameters, specific surface areas, and gas adsorption capacities decrease with increasing the size of the counter ions (Fig. 18).¹⁴⁹ The high levels of CO₂ uptake over CH₄ was attributed to that the dipole–quadrupole interactions enhance the affinity for CO₂. 2I-PAF-50 gives a remarkably high CO₂/CH₄ initial slop selectivity (29.8), which, however, is at the expense of surface area (91 m² g⁻¹) as well as the adsorption capacity (0.30 mmol g ⁻¹ for CO₂).

Fig. 18 Schematic structure and pore diameters of X-PAF-50.

4. Challenges and outlook

A large number of novel carbonaceous microporous materials have been synthesized with the aim to act as adsorbents for CO₂ separation from the flue gases, natural gases, and landfill gases. However, the majority of these studies focused on the structural diversity, surface functionality, and synthetic routes of the new materials, without considering the requirements and challenges associated with the real separation process. In many studies of POPs, for example, the materials were evaluated with the CO₂ adsorption capacity at 1 bar, while the low-pressure (0.1–0.2 bar for flue gases; ∼0.5 bar for natural gases and landfill gases) performances were completely ignored. Also, promoting CO₂ adsorption via electrostatic dipole–quadrupole interaction remains the major strategy for the design of new adsorbents. However, this strategy is problematic when moisture is involved as in the cases of real flue gases because the polar sites would preferentially adsorb water over CO₂, but little attention has been paid to this problem in the literature. Therefore, new strategies that enable the adsorbents to retain the high CO₂ uptake and selectivity in the presence of moisture are yet to be developed. Fabricating polar but hydrophobic porous frameworks by fluorination seems to be a promising solution to this problem. Meanwhile, most current researches only resort to the surface functionalization of the adsorbent to enhance the CO₂ selectivity over N₂ and CH₄. We expect that more works will be carried out to optimize the separation from the kinetic aspect, e.g., through fine control of the pore size and/or pore geometry. Preliminary results have shown that with optimal pore sizes of the adsorbent, the CO₂/N₂ selectivity measured in the breakthrough experiment can be much higher than that derived from the equilibrium adsorption isotherms, suggesting remarkable kinetic diffusion selectivity. In comparison with inorganic ACs that are essentially amorphous materials, certain types of POPs can be made in the form of crystalline materials and thus more easily engineered in the pore structure through designed synthesis. Moreover, recent studies on carbonaceous adsorbents rarely discussed the adsorption kinetics, despite its important implications for the real separation process. We suggest that fast adsorption rate can be achieved by using hierarchically structured adsorbents (that is, the materials contain mesopores/macropores in addition to micropores), where the micropores provide CO₂ adsorption sites while the larger pores facilitate molecular diffusion. The hierarchical pore structures can be obtained by either specially designed direct synthesis (e.g., templating synthesis) or post-treatment methods. The long-term stability and regeneration capability of carbonaceous adsorbents should also be carefully investigated. Our recent study revealed that although doping the ACs with nitrogen can promote its CO₂ adsorption, high nitrogen content may result in instability of the microporous structure. If this is a general phenomenon, we should reassess the usability of this strategy from an application perspective.

Overall, carbonaceous adsorbents hold greater potential than other types of adsorbents for commercial use, owing to their ease of synthesis, high stabilities, and low costs. In the last decade, significant progress has been made in the design and synthesis of new carbonaceous adsorbents with focus on the applications of CO₂ separation from flue gases and natural gases. Meanwhile, there remain many challenges to be addressed, which are mainly associated with their performance under the real separation conditions, adsorption kinetics, long-term stability, and regeneration ability.

Acknowledgements

This publication is based on the research work supported by the KAUST Office of Competitive Research Funds (OCRF) under Awards no. URF/1/1672-01-01 and FCC/1/1972-06-01.

Notes and references

1. D. W. Keith, Science, 2009, 325, 1654–1655.
2. M. R. Raupach, G. Marland, P. Ciais, C. Le Quere, J. G. Canadell, G. Klepper and C. B. Field, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 10288–10293.
3. K. B. Lee and S. Sircar, AIChE J., 2008, 54, 2293–2302.
4. R. W. Baker and K. Lokhandwala, Ind. Eng. Chem. Res., 2008, 47, 2109–2121.
5. G. F. Versteeg, L. A. J. Van Dijck and W. P. M. Van Swaaij, Chem. Eng. Commun., 1996, 144, 113–158.
6. C. Gouedard, D. Picq, F. Launay and P. L. Carrette, Int. J. Greenhouse Gas Control, 2012, 10, 244–270.
7. G. Ferey, C. Serre, T. Devic, G. Maurin, H. Jobic, P. L. Llewellyn, G. De Weireld, A. Vimont, M. Daturi and J. S. Chang, Chem. Soc. Rev., 2011, 40, 550–562.
8. F. Chang, J. Zhou, P. Chen, Y. Chen, H. Jia, S. M. I. Saad, Y. Gao, X. Cao and T. Zheng, Asia-Pac. J. Chem. Eng., 2013, 8, 618–626.
9. N. Gargiulo, F. Pepe and D. Caputo, J. Nanosci. Nanotechnol., 2014, 14, 1811–1822.
10. G. Sneddon, A. Greenaway and H. H. P. Yiu, Adv. Energy Mater., 2014, 4, 1301873.
11. Y.-S. Bae and R. Q. Snurr, Angew. Chem., Int. Ed., 2011, 50, 11586–11596.
12. P. Z. Li and Y. Zhao, Chem.–Asian J., 2013, 8, 1680–1691.
13. Z. Chen, S. Deng, H. Wei, B. Wang, J. Huang and G. Yu, Front. Environ. Sci. Eng., 2013, 7, 326–340.
14. O. Cheung and N. Hedin, RSC Adv., 2014, 4, 14480–14494.
15. D. Bonenfant, M. Kharoune, P. Niquette, M. Mimeault and R. Hausler, Sci. Technol. Adv. Mater., 2008, 9, 013007.
16. T.-L. Chew, A. L. Ahmad and S. Bhatia, Adv. Colloid Interface Sci., 2010, 153, 43–57.
17. J. Liu, P. K. Thallapally, B. P. McGrail, D. R. Brown and J. Liu, Chem. Soc. Rev., 2012, 41, 2308–2322.
18. S. Chaemchuen, N. A. Kabir, K. Zhou and F. Verpoort, Chem. Soc. Rev., 2013, 42, 9304–9332.
19. Z. Zhang, Z.-Z. Yao, S. Xiang and B. Chen, Energy Environ. Sci., 2014, 7, 2868–2899.
20. J. Duan, M. Higuchi, R. Krishna, T. Kiyonaga, Y. Tsutsumi, Y. Sato, Y. Kubota, M. Takata and S. Kitagawa, Chem. Sci., 2014, 5, 660–666.
21. S.-T. Yang, J. Kim and W.-S. Ahn, Microporous Mesoporous Mater., 2010, 135, 90–94.
22. Q. Liu, A. Mace, Z. Bacsik, J. Sun, A. Laaksonen and N. Hedin, Chem. Commun., 2010, 46, 4502–4504.
23. F. Akhtar, Q. Liu, N. Hedin and L. Bergstrom, Energy Environ. Sci., 2012, 5, 7664–7673.
24. J. Merel, M. Clausse and F. Meunier, Ind. Eng. Chem. Res., 2007, 47, 209–215.
25. A. R. Millward and O. M. Yaghi, J. Am. Chem. Soc., 2005, 127, 17998–17999.
26. K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm, T. H. Bae and J. R. Long, Chem. Rev., 2012, 112, 724–781.
27. J.-R. Li, Y. Ma, M. C. McCarthy, J. Sculley, J. Yu, H.-K. Jeong, P. B. Balbuena and H.-C. Zhou, Coord. Chem. Rev., 2011, 255, 1791–1823.
28. A. Phan, C. J. Doonan, F. J. Uribe-Romo, C. B. Knobler, M. O'Keeffe and O. M. Yaghi, Acc. Chem. Res., 2010, 43, 58–67.
29. D. Britt, H. Furukawa, B. Wang, T. G. Glover and O. M. Yaghi, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 20637–20640.
30. S. R. Caskey, A. G. Wong-Foy and A. J. Matzger, J. Am. Chem. Soc., 2008, 130, 10870–10871.
31. A. O. Z. R. Yazaydın, R. Q. Snurr, T.-H. Park, K. Koh, J. Liu, M. D. LeVan, A. I. Benin, P. Jakubczak, M. Lanuza, D. B. Galloway, J. J. Low and R. R. Willis, J. Am. Chem. Soc., 2009, 131, 18198–18199.
32. R. Banerjee, H. Furukawa, D. Britt, C. Knobler, M. O'Keeffe and O. M. Yaghi, J. Am. Chem. Soc., 2009, 131, 3875–3877.
33. J. An, S. J. Geib and N. L. Rosi, J. Am. Chem. Soc., 2010, 132, 38–39.
34. S.-T. Yang, J.-Y. Kim, J. Kim and W.-S. Ahn, Fuel, 2012, 97, 435–442.
35. R. Sanz, G. Calleja, A. Arencibia and E. S. Sanz-Pérez, Appl. Surf. Sci., 2010, 256, 5323–5328.
36. H. Y. Huang, R. T. Yang, D. Chinn and C. L. Munson, Ind. Eng. Chem. Res., 2002, 42, 2427–2433.
37. B. Guo, L. Chang and K. Xie, J. Nat. Gas Chem., 2006, 15, 223–229.
38. S. Himeno, T. Komatsu and S. Fujita, J. Chem. Eng. Data, 2005, 50, 369–376.
39. M. Sevilla and A. B. Fuertes, J. Colloid Interface Sci., 2012, 366, 147–154.
40. A. P. Cote, A. I. Benin, N. W. Ockwig, M. O'Keeffe, A. J. Matzger and O. M. Yaghi, Science, 2005, 310, 1166–1170.
41. H. M. El-Kaderi, J. R. Hunt, J. L. Mendoza-Cortes, A. P. Cote, R. E. Taylor, M. O'Keeffe and O. M. Yaghi, Science, 2007, 316, 268–272.
42. H. Furukawa and O. M. Yaghi, J. Am. Chem. Soc., 2009, 131, 8875–8883.
43. S.-Y. Ding and W. Wang, Chem. Soc. Rev., 2013, 42, 548–568.
44. Q. Fang, Z. Zhuang, S. Gu, R. B. Kaspar, J. Zheng, J. Wang, S. Qiu and Y. Yan, Nat. Commun., 2014, 5, 4503.
45. H. Xu and D. Jiang, Nat. Chem., 2014, 6, 564–566.
46. T. Ben, H. Ren, S. Ma, D. Cao, J. Lan, X. Jing, W. Wang, J. Xu, F. Deng, J. M. Simmons, S. Qiu and G. Zhu, Angew. Chem., Int. Ed., 2009, 48, 9457–9460.
47. H. Ren, T. Ben, E. Wang, X. Jing, M. Xue, B. Liu, Y. Cui, S. Qiu and G. Zhu, Chem. Commun., 2010, 46, 291–293.
48. T. Ben, C. Pei, D. Zhang, J. Xu, F. Deng, X. Jing and S. Qiu, Energy Environ. Sci., 2011, 4, 3991–3999.
49. P. Kuhn, M. Antonietti and A. Thomas, Angew. Chem., Int. Ed., 2008, 47, 3450–3453.
50. S. Ren, M. J. Bojdys, R. Dawson, A. Laybourn, Y. Z. Khimyak, D. J. Adams and A. I. Cooper, Adv. Mater., 2012, 24, 2357–2361.
51. J. Roeser, K. Kailasam and A. Thomas, ChemSusChem, 2012, 5, 1793–1799.
52. S. Ren, R. Dawson, A. Laybourn, J.-X. Jiang, Y. Khimyak, D. J. Adams and A. I. Cooper, Polym. Chem., 2012, 3, 928–934.
53. Y. F. Zhao, K. X. Yao, B. Y. Teng, T. Zhang and Y. Han, Energy Environ. Sci., 2013, 6, 3684–3692.
54. A. I. Cooper, Adv. Mater., 2009, 21, 1291–1295.
55. F. Xu, X. Chen, Z. Tang, D. Wu, R. Fu and D. Jiang, Chem. Commun., 2014, 50, 4788–4790.
56. S. Xu, L. Liang, B. Li, Y. Luo, C. Liu and B. Tan, Prog. Chem., 2011, 23, 2085–2094.
57. L. Monnereau, M. Nieger, T. Muller and S. Braese, Adv. Funct. Mater., 2014, 24, 1054–1058.
58. R. Dawson, T. Ratvijitvech, M. Corker, A. Laybourn, Y. Z. Khimyak, A. I. Cooper and D. J. Adams, Polym. Chem., 2012, 3, 2034–2038.
59. R. Dawson, E. Stockel, J. R. Holst, D. J. Adams and A. I. Cooper, Energy Environ. Sci., 2011, 4, 4239–4245.
60. T. Ben, Y. Li, L. Zhu, D. Zhang, D. Cao, Z. Xiang, X. Yao and S. Qiu, Energy Environ. Sci., 2012, 5, 8370–8376.
61. X. Zhu, C. Tian, S. M. Mahurin, S.-H. Chai, C. Wang, S. Brown, G. M. Veith, H. Luo, H. Liu and S. Dai, J. Am. Chem. Soc., 2012, 134, 10478–10484.
62. X. Zhu, D.-T. Chi-Linh, C. R. Murdock, K. M. Nelson, C. Tian, S. Brown, S. M. Mahurin, D. M. Jenkins, J. Hu, B. Zhao, H. Liu and S. Dai, ACS Macro Lett., 2013, 2, 660–663.
63. X. Gao, X. Zou, H. Ma, S. Meng and G. Zhu, Adv. Mater., 2014, 26, 3644–3648.
64. V. S. P. K. Neti, X. Wu, P. Peng, S. Deng and L. Echegoyen, RSC Adv., 2014, 4, 9669–9672.
65. E. J. Ra, T. H. Kim, W. J. Yu, K. H. An and Y. H. Lee, Chem. Commun., 2010, 46, 1320–1322.
66. M. Sevilla and A. B. Fuertes, Energy Environ. Sci., 2011, 4, 1765.
67. D. M. D'Alessandro, B. Smit and J. R. Long, Angew. Chem., Int. Ed., 2010, 49, 6058–6082.
68. D. W. Breck, Zeolite molecular sieves: structure, chemistry, and use, Wiley, 1973.
69. Y. F. Zhao, L. Zhao, K. X. Yao, Y. Yang, Q. Zhang and Y. Han, J. Mater. Chem., 2012, 22, 19726–19731.
70. G. Li and Z. Wang, Macromolecules, 2013, 46, 3058–3066.
71. G. Li, B. Zhang, J. Yan and Z. Wang, J. Mater. Chem. A, 2014, 2, 18881–18888.
72. G. Li, B. Zhang, J. Yan and Z. Wang, Macromolecules, 2014, 47, 6664–6670.
73. R. T. Yang, in Adsorbents: Fundamentals and Applications, John Wiley & Sons, Inc., 2003.
74. Y.-S. Bae, O. K. Farha, J. T. Hupp and R. Q. Snurr, J. Mater. Chem., 2009, 19, 2131.
75. X. Zhang, J. Lu and J. Zhang, Chem. Mater., 2014, 26, 4023–4029.
76. A. L. Myers and J. M. Prausnitz, AIChE J., 1965, 11, 121–127.
77. J. A. Mason, K. Sumida, Z. R. Herm, R. Krishna and J. R. Long, Energy Environ. Sci., 2011, 4, 3030–3040.
78. H. Wang, K. X. Yao, Z. J. Zhang, J. Jagiello, Q. H. Gong, Y. Han and J. Li, Chem. Sci., 2014, 5, 620–624.
79. B. Li, Y. Zhang, R. Krishna, K. Yao, Y. Han, Z. Wu, D. Ma, Z. Shi, T. Pham, B. Space, J. Liu, P. K. Thallapally, J. Liu, M. Chrzanowski and S. Ma, J. Am. Chem. Soc., 2014, 136, 8654–8660.
80. A. M. Fracaroli, H. Furukawa, M. Suzuki, M. Dodd, S. Okajima, F. Gandara, J. A. Reimer and O. M. Yaghi, J. Am. Chem. Soc., 2014, 136, 8863–8866.
81. Y. Zhao, K. X. Yao, B. Teng, T. Zhang and Y. Han, Energy Environ. Sci., 2013, 6, 3684–3692.
82. S. Biniak, G. Szymanski, J. Siedlewski and A. Swiatkowski, Carbon, 1997, 35, 1799–1810.
83. J. L. Figueiredo, M. F. R. Pereira, M. M. A. Freitas and J. J. M. Orfao, Carbon, 1999, 37, 1379–1389.
84. L. Wang and R. T. Yang, J. Phys. Chem. C, 2012, 116, 1099–1106.
85. M. Choi and R. Ryoo, J. Mater. Chem., 2007, 17, 4204–4209.
86. F. Dreisbach, R. Staudt and J. U. Keller, Adsorption, 1999, 5, 215–227.
87. P. Billemont, B. Coasne and G. De Weireld, Adsorption, 2014, 20, 453–463.
88. L. Liu, Q. F. Deng, T. Y. Ma, X. Z. Lin, X. X. Hou, Y. P. Liu and Z. Y. Yuan, J. Mater. Chem., 2011, 21, 16001–16009.
89. G. P. Hao, W. C. Li, D. Qian and A. H. Lu, Adv. Mater., 2010, 22, 853–857.
90. G. P. Hao, W. C. Li, D. Qian, G. H. Wang, W. P. Zhang, T. Zhang, A. Q. Wang, F. Schuth, H. J. Bongard and A. H. Lu, J. Am. Chem. Soc., 2011, 133, 11378–11388.
91. M. Sevilla, P. Valle-Vigón and A. B. Fuertes, Adv. Funct. Mater., 2011, 21, 2781–2787.
92. M. C. Gutierrez, D. Carriazo, C. O. Ania, J. B. Parra, M. L. Ferrer and F. del Monte, Energy Environ. Sci., 2011, 4, 3535–3544.
93. M. Sevilla, C. Falco, M. M. Titirici and A. B. Fuertes, RSC Adv., 2012, 2, 12792–12797.
94. X. Q. Fan, L. X. Zhang, G. B. Zhang, Z. Shu and J. L. Shi, Carbon, 2013, 61, 423–430.
95. Y. D. Xia, Y. Q. Zhu and Y. Tang, Carbon, 2012, 50, 5543–5553.
96. H. Seema, K. C. Kemp, N. H. Le, S. W. Park, V. Chandra, J. W. Lee and K. S. Kim, Carbon, 2014, 66, 320–326.
97. Y. Zhao, X. Liu, K. X. Yao, L. Zhao and Y. Han, Chem. Mater., 2012, 24, 4725–4734.
98. M. Nandi, K. Okada, A. Dutta, A. Bhaumik, J. Maruyama, D. Derks and H. Uyama, Chem. Commun., 2012, 48, 10283–10285.
99. Y. D. Xia, R. Mokaya, G. S. Walker and Y. Q. Zhu, Adv. Energy Mater., 2011, 1, 678–683.
100. S. S. Feng, W. Li, Q. Shi, Y. H. Li, J. C. Chen, Y. Ling, A. M. Asiri and D. Y. Zhao, Chem. Commun., 2014, 50, 329–331.
101. J. Przepiórski, M. Skrodzewicz and A. W. Morawski, Appl. Surf. Sci., 2004, 225, 235–242.
102. L. Liu, Q.-F. Deng, T.-Y. Ma, X.-Z. Lin, X.-X. Hou, Y.-P. Liu and Z.-Y. Yuan, J. Mater. Chem., 2011, 21, 16001–16009.
103. G. P. Hao, W. C. Li, D. Qian and A. H. Lu, Adv. Mater., 2010, 22, 853–857.
104. W. Xing, C. Liu, Z. Y. Zhou, L. Zhang, J. Zhou, S. P. Zhuo, Z. F. Yan, H. Gao, G. Q. Wang and S. Z. Qiao, Energy Environ. Sci., 2012, 5, 7323–7327.
105. X. Zhu, P. C. Hillesheim, S. M. Mahurin, C. M. Wang, C. C. Tian, S. Brown, H. M. Luo, G. M. Veith, K. S. Han, E. W. Hagaman, H. L. Liu and S. Dai, ChemSusChem, 2012, 5, 1912–1917.
106. J. Zhou, W. Li, Z. S. Zhang, W. Xing and S. P. Zhuo, RSC Adv., 2012, 2, 161–167.
107. X. Q. Wang, C. G. Liu, D. Neff, P. F. Fulvio, R. T. Mayes, A. Zhamu, Q. Fang, G. R. Chen, H. M. Meyer, B. Z. Jang and S. Dai, J. Mater. Chem. A, 2013, 1, 7920–7926.
108. J. C. Wang, I. Senkovska, M. Oschatz, M. R. Lohe, L. Borchardt, A. Heerwig, Q. Liu and S. Kaskel, ACS Appl. Mater. Interfaces, 2013, 5, 3160–3167.
109. X. Y. Ma, M. H. Cao and C. W. Hu, J. Mater. Chem. A, 2013, 1, 913–918.
110. C. Pevida, M. G. Plaza, B. Arias, J. Fermoso, F. Rubiera and J. J. Pis, Appl. Surf. Sci., 2008, 254, 7165–7172.
111. J. Wei, D. D. Zhou, Z. K. Sun, Y. H. Deng, Y. Y. Xia and D. Y. Zhao, Adv. Funct. Mater., 2013, 23, 2322–2328.
112. P. Nowicki, R. Pietrzak and H. Wachowska, Energy Fuels, 2010, 24, 1197–1206.
113. U. results.
114. M. L. Gray, Y. Soong, K. J. Champagne, J. Baltrus, R. W. Stevens, P. Toochinda and S. S. C. Chuang, Sep. Purif. Technol., 2004, 35, 31–36.
115. Y.-S. Bae, B. G. Hauser, O. K. Farha, J. T. Hupp and R. Q. Snurr, Microporous Mesoporous Mater., 2011, 141, 231–235.
116. Q. Fang, G. Zhu, M. Xue, J. Sun, Y. Wei, S. Qiu and R. Xu, Angew. Chem., Int. Ed., 2005, 44, 3845–3848.
117. M. Dincă, W. S. Han, Y. Liu, A. Dailly, C. M. Brown and J. R. Long, Angew. Chem., Int. Ed., 2007, 46, 1419–1422.
118. R. Babarao and J. Jiang, J. Am. Chem. Soc., 2009, 131, 11417–11425.
119. W. Lu, D. Yuan, J. Sculley, D. Zhao, R. Krishna and H.-C. Zhou, J. Am. Chem. Soc., 2011, 133, 18126–18129.
120. O. K. Farha, Y. S. Bae, B. G. Hauser, A. M. Spokoyny, R. Q. Snurr, C. A. Mirkin and J. T. Hupp, Chem. Commun., 2010, 46, 1056–1058.
121. V. Chandra, S. U. Yu, S. H. Kim, Y. S. Yoon, D. Y. Kim, A. H. Kwon, M. Meyyappan and K. S. Kim, Chem. Commun., 2012, 48, 735–737.
122. G. P. Hao, Z. Y. Jin, Q. Sun, X. Q. Zhang, J. T. Zhang and A. H. Lu, Energy Environ. Sci., 2013, 6, 3740–3747.
123. G. Srinivas, V. Krungleviciute, Z.-X. Guo and T. Yildirim, Energy Environ. Sci., 2014, 7, 335–342.
124. Y. Li, T. Ben, B. Zhang, Y. Fu and S. Qiu, Sci. Rep., 2013, 3, 2420.
125. Y. Zhang, B. Li, K. Williams, W. Y. Gao and S. Ma, Chem. Commun., 2013, 49, 10269–10271.
126. Z. Yan, H. Ren, H. Ma, R. Yuan, Y. Yuan, X. Zou, F. Sun and G. Zhu, Microporous Mesoporous Mater., 2013, 173, 92–98.
127. S. Hug, M. B. Mesch, H. Oh, N. Popp, M. Hirscher, J. Senker and B. V. Lotsch, J. Mater. Chem. A, 2014, 2, 5928.
128. S. Wu, Y. Liu, G. Yu, J. Guan, C. Pan, Y. Du, X. Xiong and Z. Wang, Macromolecules, 2014, 47, 2875–2882.
129. Y. Liu, S. Wu, G. Wang, G. Yu, J. Guan, C. Pan and Z. Wang, J. Mater. Chem. A, 2014, 2, 7795.
130. A. Bhunia, I. Boldog, A. Moeller and C. Janiak, J. Mater. Chem. A, 2013, 1, 14990–14999.
131. S. Xiong, X. Fu, L. Xiang, G. Yu, J. Guan, Z. Wang, Y. Du, X. Xiong and C. Pan, Polym. Chem., 2014, 5, 3424–3431.
132. M. R. Liebl and J. Senker, Chem. Mater., 2013, 25, 970–980.
133. G. Li, B. Zhang, J. Yan and Z. Wang, Chem. Commun., 2014, 50, 1897–1899.
134. G. Li, B. Zhang and Z. Wang, Macromol. Rapid Commun., 2014, 35, 971–975.
135. A. Modak, M. Nandi, J. Mondal and A. Bhaumik, Chem. Commun., 2012, 48, 248–250.
136. X. Liu, A. Sigen, Y. Zhang, X. Luo, H. Xia, H. Li and Y. Mu, RSC Adv., 2014, 4, 6447–6453.
137. M. G. Rabbani and H. M. El-Kaderi, Chem. Mater., 2012, 24, 1511–1517.
138. Q. Chen, M. Luo, P. Hammershøj, D. Zhou, Y. Han, B. W. Laursen, C.-G. Yan and B.-H. Han, J. Am. Chem. Soc., 2012, 134, 6084–6087.
139. H. A. Patel, S. H. Je, J. Park, D. P. Chen, Y. Jung, C. T. Yavuz and A. Coskun, Nat. Commun., 2013, 4, 1357.
140. R. Dawson, D. J. Adams and A. I. Cooper, Chem. Sci., 2011, 2, 1173–1177.
141. W. Lu, J. P. Sculley, D. Yuan, R. Krishna, Z. Wei and H. C. Zhou, Angew. Chem., Int. Ed., 2012, 51, 7480–7484.
142. V. Guillerm, L. J. Weselinski, M. Alkordi, M. I. Mohideen, Y. Belmabkhout, A. J. Cairns and M. Eddaoudi, Chem. Commun., 2014, 50, 1937–1940.
143. R. Dawson, L. A. Stevens, T. C. Drage, C. E. Snape, M. W. Smith, D. J. Adams and A. I. Cooper, J. Am. Chem. Soc., 2012, 134, 10741–10744.
144. W.-C. Song, X.-K. Xu, Q. Chen, Z.-Z. Zhuang and X.-H. Bu, Polym. Chem., 2013, 4, 4690–4696.
145. W. Lu, D. Yuan, J. Sculley, D. Zhao, R. Krishna and H. C. Zhou, J. Am. Chem. Soc., 2011, 133, 18126–18129.
146. H. Ma, H. Ren, X. Zou, F. Sun, Z. Yan, K. Cai, D. Wang and G. Zhu, J. Mater. Chem. A, 2013, 1, 752–758.
147. A. P. Katsoulidis and M. G. Kanatzidis, Chem. Mater., 2011, 23, 1818–1824.
148. H. Ma, H. Ren, X. Zou, S. Meng, F. Sun and G. Zhu, Polym. Chem., 2014, 5, 144–152.
149. Y. Yuan, F. Sun, L. Li, P. Cui and G. Zhu, Nat. Commun., 2014, 5, 4260.
150. R. Dawson, E. Stoeckel, J. R. Holst, D. J. Adams and A. I. Cooper, Energy Environ. Sci., 2011, 4, 4239–4245.
151. J.-X. Jiang, F. Su, A. Trewin, C. D. Wood, N. L. Campbell, H. Niu, C. Dickinson, A. Y. Ganin, M. J. Rosseinsky, Y. Z. Khimyak and A. I. Cooper, Angew. Chem., Int. Ed., 2007, 46, 8574–8578.
152. Y. Xu, S. Jin, H. Xu, A. Nagai and D. Jiang, Chem. Soc. Rev., 2013, 42, 8012–8031.
153. P. Kuhn, M. Antonietti and A. Thomas, Angew. Chem., Int. Ed., 2008, 47, 3450–3453.
154. H. Lim, M. C. Cha and J. Y. Chang, Macromol. Chem. Phys., 2012, 213, 1385–1390.
155. Y. Jin, Y. Zhu and W. Zhang, CrystEngComm, 2013, 15, 1484–1499.
156. S. J. Garibay, M. H. Weston, J. E. Mondloch, Y. J. Colon, O. K. Farha, J. T. Hupp and S. T. Nguyen, CrystEngComm, 2013, 15, 1515–1519.
157. W. Lu, J. P. Sculley, D. Yuan, R. Krishna, Z. Wei and H.-C. Zhou, Angew. Chem., Int. Ed., 2012, 51, 7480–7484.
158. W. Lu, J. P. Sculley, D. Yuan, R. Krishna and H.-C. Zhou, J. Phys. Chem. C, 2013, 117, 4057–4061.
159. W. Lu, W. M. Verdegaal, J. Yu, P. B. Balbuena, H.-K. Jeong and H.-C. Zhou, Energy Environ. Sci., 2013, 6, 3559–3564.
160. H. Ma, H. Ren, X. Zou, F. Sun, Z. Yan, K. Cai, D. Wang and G. Zhu, J. Mater. Chem. A, 2013, 1, 752.

---