Pretreated multiwalled carbon nanotube adsorbents with amine-grafting for removal of carbon dioxide in confined spaces

Abstract

Three different methods, including thermal treatment, treatment with HNO₃ and O₂ oxidation, were used to pretreat multiwalled carbon nanotubes (MWCNTs) before grafting with N-(2-aminoethyl)-3-aminopropyltrimethoxysilane (AEAPS). The types and contents of the O-containing groups generated by the various pretreatment methods were quantified and the corresponding amine-grafting reactions investigated. The alkoxyl groups of AEAPS react with the O-containing groups on the pretreated MWCNTs through silylation reactions in which there is no carboxyl acids induced by O₂ gas oxidation. The grafted primary amino groups can be accessible to capture the maximum amount of CO₂. A dynamic fixed-bed system was used to characterize the adsorption behavior of low concentrations of CO₂. The adsorption/desorption operations of 10 repeated cycles were investigated to verify the sustained excellent performance. The highest CO₂ adsorption capacity of 0.64 mmol g⁻¹ was achieved by the O₂-oxidized MWCNTs with AEAPS-grafting, which is almost 7.1 times the CO₂ adsorption capacity of the oxidized MWCNTs without amine-grafting. This indicates that O₂ gas oxidation is simple to operate and highly efficient for the pretreatment of MWCNTs. The adsorbent obtained has a high capacity, high thermal stability, high tolerance to moisture and low regeneration cost and shows promise for the direct capture of low concentrations of CO₂ in confined spaces.

---

1. Introduction

Carbon capture and sequestration technologies have received increasing attention in recent years^1,2 as a means of reducing the impacts of the increasing concentration of CO₂ in the atmosphere on the global climate. It is estimated that approximately one-third of global carbon emissions are emitted from distributed sources, such as transportation vehicles, where low concentrations of CO₂ (about 400 ppm) need to be directly captured from the ambient air (“air capture”).³ Capture technologies for the high concentrations of CO₂ generated from large point sources, such as coal-fired power plants (about 10% CO₂), are relatively mature.^4,5 In addition, CO₂ capture is also required in confined spaces, such as submarines, space shuttles and space stations, where CO₂ is emitted by humans during breathing, from mechanical equipment and during the oxidation of various materials.^6,7 High concentrations of CO₂ in such enclosed spaces can threaten the survival of humans.^8,9

The various approaches used to capture CO₂ from a mixture of gases include absorption, cryogenic distillation, adsorption and membrane separation.^10–13 Chemical absorption processes using lithium hydroxide or liquid amines are mature commercial technologies and have a high capacity for CO₂ capture. However, there are drawbacks to these techniques, such as high operational costs, container corrosion, their high energy intensity, and viscosity and foaming issues.¹⁴ Cryogenic distillation and membrane separation processes generally have low efficiencies and high costs, especially for capture from ambient air.¹⁰ Adsorption using porous solid sorbents, including carbon materials,^15,16 polymers,⁶ silica materials^17–20 and metal–organic frameworks,²¹ is more promising because these materials require less energy for regeneration.²²

In addition to physical adsorbents, amine-functionalized adsorbents are attractive because they have active alkaline sites that can be used for the adsorption of CO₂.^15,23 These chemical adsorbents are typically prepared by physically immobilizing liquid amines or chemically grafting amino groups onto the surface of a porous support.²⁴ A higher loading of amine can be achieved by an impregnation method; however, these adsorbents exhibit low thermal stability and poor distribution of the amine.^25–28 The CO₂ adsorption capacity decreases gradually with increasing numbers of adsorption/desorption cycles as a result of the leaching of amines and the efficiency of these adsorbents is low.²⁹ Intensive research efforts have been devoted to chemically grafting amines onto silica-based solid supports such as SBA-16,¹⁷ SBA-15,^30,31 and MCM-41.^29,32 Carbon nanotubes (CNTs), however, are more suitable as supports because of their tolerance to moisture compared with silica-based adsorbents.^22,33–35 It has been shown that CNT-based adsorbents are light in weight, have high surface areas and high thermal and chemical stabilities. In addition, their chemical reactivity can be controlled.^36–38 Studies of their thermodynamics and regeneration have shown that amine-grafted CNTs are potential cost-effective sorbents for CO₂ capture.^33,34

The introduction of O-containing groups onto supports by chemical pretreatment is an important method of chemical grafting that can be used to covalently link amine functional groups. The most commonly used pretreatment methods are acid oxidation, UV photo-oxidation, thermal treatment, and plasma and gas phase oxidation.^37,39,40 Velasco-Santos et al.⁴⁰ used an arc-discharge method to oxidize carbon nanotube surfaces to improve the chemical bonding with different polymer materials. They showed that the trisilanol groups on organosilanes can react with the hydroxyl groups on the pre-oxidized nanotube surface and therefore the organosilanes will be attached to the surface of the nanotubes.⁴⁰ However, the arc-discharge treatment needs to be conducted at high voltages and the high costs of the power supply have restricted its use in industrial applications. Acid treatment gives relatively high yields of functional groups, including carboxylic acid, phenol, carbonyl, carboxylic anhydride and lactone groups.³⁴ However, acid treatment is difficult and time-consuming with respect to the required rinsing and drying operations. Gas phase oxidation is more convenient and effective, particularly as it is contaminant-free, and O-containing groups including phenol, carbonyl/quinone and lactone groups can be generated.^37,41 Thermal treatment can selectively remove some functional groups, such as carboxylic acid. An intensive study of the correlation between O-containing groups on the oxidized CNTs and the efficiency of the silylation reaction was conducted by Gaspar et al.³⁷ The highest efficiency was achieved with oxidation by HNO₃ with subsequent thermal treatment; this method gave a high content of O-containing groups, including phenol groups, but without carboxylic acids. However, these workers did not concentrate on the study of CO₂ capture.

Some pretreatment methods, including thermal, acid and NaOH treatments, have been used only to purify CNTs before amine-grafting^22,35,42 and the tuning of the O-containing functional groups induced by the oxidative methods on amine-grafting for CO₂ capture has seldom been investigated. In the work reported here, multiwalled CNTs (MWCNTs) were first pretreated by acid oxidation, thermal treatment or gas phase oxidation and then N-(2-aminoethyl)-3-aminopropyltrimethoxysilane (AEAPS) was used to tether the pretreated MWCNTs. We investigated the effects of the type and content of the O-containing groups on the corresponding amine-grafting reaction. The amine-grafting efficiencies for CO₂ capture and the mechanisms for the adsorption of CO₂ were also studied. The textural properties, surface chemical composition and morphologies of the resulting modified MWCNTs were characterized by N₂ adsorption isotherms, thermogravimetry, X-ray photoelectron spectrometry, Fourier transform infrared spectrometry, elemental analysis and scanning electron microscopy. The amine-grafted MWCNTs obtained were then used for removal of CO₂ from confined spaces, where the representative CO₂ level is about 2 vol%.⁸ The effects of the operating temperature and humidity of the gas mixture on the adsorption behavior of CO₂ and the repeatability of the procedure were studied.

2. Experimental section

2.1 Preparation of amine-grafted MWCNTs

2.1.1 Materials. All the reagents and solvents were used without further purification. Commercial MWCNTs (TNM2, 95 wt% purity, Chengdu Organic Chemicals Co. Ltd) were prepared through the catalytic decomposition of natural gas over a Co-based catalyst. The outer diameter of the MWCNTs was in the range 8–15 nm and the length was about 50 μm. The chemicals used in the oxidation and silylation reactions were: HNO₃ (65–68 wt%, Gaojing), H₂SO₄ (98 wt%, Gaojing), anhydrous toluene (AR, Guoyao) and AEAPS (>95 wt%, Aladdin).

2.1.2 Chemical pretreatment. The pristine MWCNTs were pretreated by thermal treatment, acid oxidation and oxidation in the gas phase; the samples obtained are denoted as t-MWCNTs, h-MWCNTs and o-MWCNTs, respectively.
Thermal treatment. The pristine MWCNTs were thermally treated in a fused-silica tube with N₂ gas (100 mL min⁻¹) at 300 °C for 1 h and then cooled to room temperature.
Acid oxidation. About 4 g of pristine MWCNTs were refluxed in 300 mL of 68 wt% HNO₃ at 80 °C for 8 h and then cooled to room temperature. The samples were then filtered and washed with deionized water until the pH of the filtrate became neutral. The samples obtained were then dried overnight in a vacuum at 60 °C. When the raw MWCNTs were treated by mixture of HNO₃ and H₂SO₄ (1 : 3 v/v) at 80 °C for 4 h, the obtained samples were denoted as m-MWCNTs.
Gas oxidation. The pristine MWCNTs were heated to 500 °C at 4 °C min⁻¹ in a 5 vol% O₂/N₂ gas mixture (100 mL min⁻¹) and kept at 500 °C for 3 h. The samples were then cooled to room temperature under an N₂ atmosphere.

2.1.3 Amine-grafting of MWCNTs. AEAPS was tethered onto the surface of the pretreated MWCNTs through a silylation reaction.³⁷ First, 5 mL of AEAPS were dispersed in 95 mL of anhydrous toluene and continuously stirred for 5 min. About 2 g of pretreated MWCNTs were then added and refluxed at 110 °C for 24 h under Ar protection. The amine-functionalized MWCNTs were then filtered and rinsed with toluene. The amine-grafted MWCNTs obtained were dried at 80 °C for 8 h. A simple notation was used for the obtained samples: o-MWCNTs–AEAPS represents the samples pretreated by O₂/N₂ oxidation and subsequently grafted with AEAPS.

2.2 Characterization of adsorbents

The materials were characterized texturally by N₂ physical adsorption/desorption isotherms (TriStar 3000, Micrometrics) at −196 °C. The samples were vacuum-dried at 120 °C for 3 h before determination. The Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods were used to calculate the surface area and average pore diameter, respectively, from the adsorption branch of the isotherms. The total pore volume was calculated from the amount of N₂ adsorbed at P/P₀ = 0.98 and the micro surface areas and volume were calculated by the t-plot method. The surface morphological features were observed by field-emission scanning electron microscopy (SU-70 microscope, Hitachi) at 3 kV. The thermal stabilities were obtained by thermogravimetry and differential scanning calorimetry (SDT Q600, TA Instruments) in an N₂ atmosphere at a flow-rate of 120 mL min⁻¹. The functional groups present below the corresponding decomposition temperature were determined quantitatively. The sample was heated from room temperature to 900 °C at a heating rate of 10 °C min⁻¹ in an N₂ atmosphere. X-ray photoelectron spectrometry (Amicus spectrometer, Shimadzu) using a monochromatized Al Kα radiation source (1486.7 eV) and Fourier transform infrared spectrometry (Nicolet 6700 instrument) were used to study and identify the surface chemical properties, particularly the determination of the amino groups. The elemental content of the adsorbents was quantitatively determined using a Flash EA 1112 instrument (Thermo Finnigan).

2.3 Adsorption experiments

The experimental studies of the adsorption of CO₂ were conducted in a dynamic fixed-bed system (Fig. 1). The glass adsorption column packed with 1 g of adsorbent was placed in a temperature-controlled box (AT-950, Science Instrument, China) to maintain a constant temperature. The internal diameter and total length of the column were approximately 1 and 20 cm, respectively. A mixture of pure N₂ and CO₂ was blended in a glass bottle and the influent CO₂ concentration was kept at 2 vol%. Mass flow controllers (D07-19C, Sevenstar Electronics, China) were used to control the flow-rate of the influent mixture gas at 50 mL min⁻¹. The influent gas mixture was passed through the adsorption column and CO₂ was adsorbed and removed. Once the CO₂ adsorption process had reached equilibrium, the saturated adsorbents were then desorbed at an adsorption column temperature of 75 °C and 5 kPa vacuum. After 5 min, the concentration of desorbed CO₂ in the effluent gas decreased to zero, indicating that the adsorption ability of the adsorbents had been recovered and that they were available for the next round of the adsorption study.

Fig. 1 Diagram of the adsorption experimental setup. (1) Nitrogen, (2) carbon dioxide, (3) mass flow controller, (4) gas blending bottle, (5) water bubbler, (6) water-bath, (7) gas chromatograph, (8) adsorption column, (9) temperature control box.

Gas chromatography using a chromatographic column (3.0 mm × 4.0 mm, Proapak QS) and a thermal conductivity detector (GC9790, Fuli, China) was used to automatically detect the CO₂ concentration of the influent and effluent gas streams. The operating temperatures of the injector, column and detector were set at 100 °C, 140 °C, and 150 °C, respectively. The moisture content of the influent gas mixture was generated from a water bubbler in which the stream of N₂ was bubbled through deionized water and the amount of water carried was controlled by adjusting the temperature of the water-bath. A higher temperature resulted in a higher water vapor content. The water content was calculated by the following equation.

where Q₁ and Q₂ are the flow-rates of N₂ and the influent gas mixture, respectively. p* represents the saturated vapor pressure of water at a particular temperature and p is the atmospheric pressure. A similar installation was used by Ye et al.²⁸ and a maximum 7% water content can be achieved. The calculated water content represents the percentage input in the influent gas mixture rather than the real equilibrium content. A water fog was formed when it reached >3%. As the length of the transport pipe between the water bubbler and the adsorption column is short, the water fog can be quickly transferred into the adsorption column and adsorbed by the adsorbents.

The CO₂ adsorption capacity (q, mmol g⁻¹) at a certain time (t, min) was calculated as follows:

where m is the weight of the pristine adsorbents (g) and Q is the flow-rate of the influent gas (mL min⁻¹). C_in and C_out represent the influent and effluent CO₂ concentrations (vol%/vol%), respectively. T represents the operating temperature in K. The molar volume of gas, V_m, is 22.4 L mol⁻¹ at the standard temperature T₀ = 273 K.

The calculation of the amine-grafting efficiencies for CO₂ capture (η, CO₂/N molar ratio) recommended by Sculley and Zhou,⁴³ which indicates how many effective amino groups per total weight of adsorbent are accessible to capture CO₂, was evaluated:

where q_e is the CO₂ equilibrium adsorption capacity (mmol g⁻¹) and C_N is the N content in the modified MWCNTs (mmol g⁻¹) determined by elemental analysis.

3. Results and discussion

3.1 CO₂ adsorption performance of modified MWCNTs

Fig. 2 shows the CO₂ equilibrium adsorption capacity of the modified MWCNTs. The pre-oxidative MWCNTs showed a dramatic enhancement in their CO₂ equilibrium adsorption capacity after AEAPS-grafting for each pretreatment method. This may be attributed to the presence of sufficient effective amino groups on the surfaces of the AEAPS-grafted MWCNTs to capture a greater amount of CO₂. The CO₂ adsorption capacity of the o-MWCNTs–AEAPS was 0.64 mmol g⁻¹, which is almost 7.1 times that of the o-MWCNTs (0.09 mmol g⁻¹) before AEAPS-grafting. The CO₂ adsorption capacities of the t-MWCNTs–AEAPS and h-MWCNTs–AEAPS were 0.38 mmol g⁻¹ and 0.36 mmol g⁻¹, respectively, which are both significantly improved compared with the MWCNTs without AEAPS-grafting. However, it can be seen that oxidation with O₂ produces higher CO₂ adsorption capacities after AEAPS-grafting than the thermal and HNO₃ pretreatment methods.

Fig. 2 CO₂ adsorption capacity of the modified MWCNTs at 25 °C with a C_in value of 2 vol%.

Various pretreatment methods have been reported previously for the purification of MWCNTs for amine-grafting to improve the adsorption of CO₂. For example, Su et al.⁴² reported amine-modified MWCNTs pretreated by NaOH that had a CO₂ adsorption capacity of 0.984 mmol g⁻¹ at 20 °C for 15 vol% CO₂. Gui et al.³⁵ used H₂SO₄/HNO₃ (3 : 1 vol/vol) as a pretreatment for the amine-grafting of MWCNTs and the CO₂ uptake at 60 °C for 5% CO₂ was 1.71 mg g⁻¹. The relatively higher adsorption capacity was mainly attributed to the high level of treated CO₂. However, few comparative studies of different pretreatments on AEAPS-grafting for CO₂ capture at low concentrations have been reported, in particular the potential advantages of oxidation by O₂.

3.2 Characterization of the modified MWCNTs

3.2.1 Direct observation of the pretreated MWCNTs. Fig. 3 shows a photograph of 0.5 g samples of MWCNTs with different pretreatments; the variations in the bulk volumes can be observed. The bulk volumes of MWCNTs after the thermal and O₂ pretreatments remained unchanged compared with the pristine MWCNTs. However, the volume of the MWCNTs shrank significantly after the acid treatment, especially after treatment with mixed H₂SO₄/HNO₃. This severe agglomeration may be attributed to the enhanced interactions with the hydrogen bonds of the carboxylic acids generated by the acid treatment.⁴⁴ This adversely affected the diffusion of CO₂ and the corresponding adsorption capacity was therefore low. It is therefore obvious that conventional acid pretreatment is not suitable for the amine-grafting MWCNTs for use in CO₂ adsorption.

Fig. 3 Photograph of the MWCNTs after different pretreatments. (a) Pristine MWCNTs; (b) t-MWCNTS; (c) o-MWCNTs; (d) h-MWCNTs; and (e) m-MWCNTs.

3.2.2 Textural properties and morphology.
Textural properties. Fig. 4 shows the N₂ adsorption/desorption isotherms of the pretreated and AEAPS-modified MWCNTs. The IV shape and H1 type of hysteresis loop according to the IUPAC classification indicate that all the samples exhibit a mesoporous structure with cylindrical pores, in spite of being pretreated and amine-grafted.³¹ The sharp increment in the adsorption isotherms at high P/P₀ values and the adsorption hysteresis loop may be caused by capillary condensation in the mesoporous pores. The o-MWCNTs had the highest N₂ adsorption capacity among the pretreated MWCNTs (Fig. 4a), indicating that the highest porosities are obtained after O₂ oxidation. However, no significant changes were seen in the N₂ adsorption isotherms after the thermal and HNO₃ pretreatments. Table 1 gives the physical properties obtained and provides detailed information about the variations between the different pretreatments. The specific surface area (S_BET) and pore volume (V_p) of the o-MWCNTs are 151.8 m² g⁻¹ and 0.394 cm³ g⁻¹, respectively, an increase of almost 18.8% and 15.9% compared with the pristine MWCNTs. The S_BET and V_p values of t-MWCNTs and h-MWCNTs show only slight changes, as expected. High specific surface areas and pore volumes are beneficial for amine-grafting and CO₂ transport in MWCNTs. Table 1 also shows that amine-grafting leads to a drastic decrease in the surface area compared with the pretreated precursor MWCNTs as a result of the introduction of the organic amines. The S_BET value of t-MWCNTs–AEAPS, h-MWCNTs–AEAPS and o-MWCNTs–AEAPS are 102.74 m² g⁻¹, 66.22 m² g⁻¹, and 67.05 m² g⁻¹, respectively, which is a reduction of almost 19.1%, 47.6%, and 55.8%, respectively. The grafted amines are the main contributors to the partial or complete blockage of the pores and result in the decrease in the S_BET and V_p values.³⁷ However, the S_BET value of the modified MWCNTs did not agree with the CO₂ adsorption capacity obtained. The S_BET value of t-MWCNTs–AEAPS is relatively higher, indicating that the AEAPS loading may be lower. The average pore diameters all increased after amine-grafting as a result of the coverage of the well-distributed and uniform AEAPS. The average pore diameters of the t-MWCNTs–AEAPS, h-MWCNTs–AEAPS and o-MWCNTs–AEAPS are 13.3 nm, 11.8 nm, and 11.1 nm, respectively, increased by almost 11.3%, 5.9%, and 6.63%, respectively. The thermal treatment resulted in higher average pore diameters because it removed only some of the original oxygen groups from the pristine MWCNTs and the AEAPS were subsequently mainly grafted onto the sidewalls of the nanotubes. HNO₃ or O₂ oxidation can occur on the end-caps as well as on the sidewalls of the nanotubes, generating more O-containing groups.

Fig. 4 N₂ adsorption (solid lines)/desorption (broken lines) isotherms of (a) pretreated and (b) AEAPS-grafted MWCNTs.

Table 1 Textural and structural properties of the modified MWCNTs

Sample	SBET (m^2 g^−1)	Vp (cm^3 g^−1)	Average pore diameter (nm)
MWCNTs	127.8	0.340	10.6
t-MWCNTs	126.9	0.379	11.9
h-MWCNTs	126.3	0.353	11.2
o-MWCNTs	151.8	0.394	10.4
t-MWCNTs–AEAPS	102.7	0.341	13.3
h-MWCNTs–AEAPS	66.2	0.196	11.8
o-CMWNTs–AEAPS	67.1	0.186	11.1

---

Morphology. Fig. 5 shows the morphologies of the pretreated MWCNTs and corresponding AEAPS-grafted MWCNTs. There was no obvious change between the t-MWCNTs and pristine MWCNTs (Fig. 5a and its inset, respectively). The surfaces of the h-MWCNTs (Fig. 5b) and o-MWCNTs (Fig. 5c) appear clear because most of the impurities were removed by the HNO₃ or O₂ treatment. As a result of the aggressive treatment of the O₂ oxidation, the open and cutoff structures of the MWCNTs can be seen more readily (highlighted by the circles in Fig. 5c). With respect to the corresponding AEAPS-grafted samples, it is observed that AEAPS was successfully tethered onto the nanotube surfaces. The AEAPS loading on the t-MWCNTs was relatively scattered and low (Fig. 5d). The h-MWCNTs–AEAPS shown in Fig. 5e tends to significant cross-linking. This may be related to the limited diffusion of CO₂ to the amino groups deep inside the pores and possibly to blocking of the pore entrances. The CO₂ adsorption capacities obtained in these instances were low. The nanotube density of o-MWCNTs–AEAPS was large as a result of the coverage of the well-distributed and uniform AEAPS (Fig. 5f). This shows that O₂ oxidation is suitable for AEAPS-grafting and that the uniform distribution of amine groups obtained results in the capture of more CO₂.

Fig. 5 Morphologies of the modified MWCNTs: (a) t-MWCNTs, with pristine MWCNTs shown in the inset; (b) h-MWCNTs; (c) o-MWCNTs; (d) t-MWCNTs–AEAPS; (e) h-MWCNTs–AEAPS; and (f) o-MWCNTs–AEAPS.

3.2.3 Compositional analysis. The functional groups on the AEAPS-grafted MWCNTs were characterized by FTIR analysis (Fig. 6). The characteristic bands for the AEAPS-grafted MWCNTs show no significant differences. The band at 1580 cm⁻¹ is assigned to the typical stretching vibration of C=C bonds in MWCNTs and the band at 1200 cm⁻¹ is associated with the scissoring vibration of C–C bonds and the stretching vibration of C–O bonds.⁴⁵ The broad band at 3440 cm⁻¹ results from the presence of free and associated hydroxyl groups as a result of intercalated water and structural hydroxyl groups (–COOH and –COH). The occurrence of bands at 2920 cm⁻¹ and 2850 cm⁻¹ verifies the presence of C–H bonds from the alkyl and alkoxyl groups in aminosilane. In particular, the band at 1116 cm⁻¹ is assigned to the asymmetrical stretching vibration of the Si–O–R bond, accompanied by the bending vibration of Si–C and Si–OH bonds. The weak band at 1456 cm⁻¹ is due to the bending vibration of the N–H bond in aminosilane.⁴⁶ This demonstrates that aminosilane has been successfully grafted onto the pretreated MWCNTs by the silylation reactions.

Fig. 6 FTIR spectra of the AEAPS-grafted MWCNTs.

Table 2 presents the quantitative results of the X-ray photoelectron spectrometry analysis of the obtained samples. The highest O content of 7.8% was achieved by HNO₃ pretreatment. Elemental Si and N appeared in all the samples, indicating that AEAPS was successfully grafted onto the pretreated MWCNTs. The N contents of the t-MWCNTs–AEAPS, h-MWCNTs–AEAPS, and o-MWCNTs–AEAPS are 6.15%, 8.33% and 7.28%, respectively. However, the high N content is not a linear function with respect to the CO₂ adsorption capacity. This is because the CO₂ adsorption capacity is dependent on the effective amine groups, especially on the primary amine groups.³⁰ The amine-grafting for o-MWCNTs–AEAPS is more effective because the CO₂ adsorption obtained is higher than for the other pretreatment methods.

Table 2 Surface atomic percentages of pretreated and amine-modified MWCNTs

Sample	C 1s	O 1s	N 1s	Si
MWCNTs	95.13	4.23	0.64	—
t-MWCNTs	97.26	2.74	—	—
h-MWCNTs	91.74	7.8	0.46	—
o-MWCNTs	97.26	2.74	—	—
t-MWCNTs–AEAPS	80.59	9.43	6.15	3.84
h-MWCNTs–AEAPS	72.09	14.42	8.33	5.16
o-MWCNTs–AEAPS	76.63	11.42	7.28	4.67

---

Fig. 7 shows the thermogravimetric curves of the pretreated and AEAPS-grafted MWCNTs. There is a weight loss of less than 2% from room temperature to about 110 °C for all the samples, which is mainly a result of the evaporation of the adsorbed water. The h-MWCNTs show the highest percentage total weight loss among the three pretreated samples, indicating that the amount of O-containing groups introduced by HNO₃ treatment is the highest. The weight of the pristine MWCNTs gradually decreases over the whole heating process, verifying that some of the O-containing groups are present on the surface.^36,37 The differences in the thermal stabilities based on the decomposition temperature can provide some qualitative information about the O-containing functional groups.^37,47 For example, carboxylic acids groups can be selectively removed by thermal treatment at 300 °C. The weight loss of the o-MWCNTs at high temperature is mainly attributed to the decomposition of phenols and carbonyl/quinone groups as well as lactone groups. With respect to the AEAPS-grafted samples, the highest weight loss was obtained by h-MWCNTs–AEAPS, corroborating the result that the AEAPS loading was high in these samples. The high yield of the silylation reaction may be attributed to the large amount of O-containing groups in the pretreated MWCNTs.

Fig. 7 Thermogravimetric curves of the pretreated and AEAPS-grafted MWCNTs.

The types and contents of the O-containing functional groups on the pretreated MWCNTs can be simply quantified on the basis of the decomposition temperature and the corresponding differential weight loss.^37,41 Typical functional groups such as carboxylic acids, phenols and carbonyls/quinones can be related to decomposition temperatures of about 200–450 °C, 600–750 °C and 700–950 °C, respectively.³⁷ The contributions of other groups such as carboxylic anhydrides and lactones were relatively low and can be omitted. We assumed here that the weight loss between 700 °C and 750 °C was equal for phenols and carbonyls/quinones. Table 3 summarizes the types and contents of the O-containing groups in the pretreated MWCNTs. The total contents of O-containing groups in t-MWCNTs, h-MWCNTs, and o-MWCNTs are 2.6%, 5.0% and 3.3%, respectively. For the HNO₃ pretreatment, the content of carboxylic acids is 2.4%, a total of 48% of the O-containing groups. However, no carboxylic acids existed in the o-MWCNTs because the O₂ pretreatment was conducted at 500 °C; phenols and carbonyl/quinone groups were dominant. This shows that the high N loading obtained can be attributed to the high carboxylic acid content. The absence of carboxylic acids in o-MWCNTs was the basis of the effective AEAPS-grafting, which results in the high CO₂ adsorption capacity of the o-MWCNTs–AEAPS.

Table 3 Types and contents of O-containing groups in the pretreated MWCNTs

Sample	Functional groups	Decomposition temperature (°C)	Δ (%)
t-MWCNTs	Carboxylic acids	200–450	0.6
	Phenols	600–750	0.8
	Carbonyl/quinone groups	700–950	1.2
		Total	2.6
h-MWCNTs	Carboxylic acids	200–450	2.4
	Phenols	600–750	1.3
	Carbonyl/quinone groups	700–950	1.3
		Total	5.0
o-MWCNTs	Phenols	600–750	1.5
	Carbonyl/quinone groups	700–950	1.8
		Total	3.3

---

3.3 Amine-grafting reactions in MWCNTs used for CO₂ capture

3.3.1 Amine-grafting reaction. Fig. 8 illustrates the AEAPS-grafting reaction with the pretreated MWCNTs. Priority is given to retaining the primary amine groups for CO₂ capture, rather than consuming these during the amine-grafting. The alkoxyl groups in AEAPS react with the O-containing groups on the pretreated MWCNTs via a silylation reaction in which there is no influence on the amine groups in the compounds. Aminosilane is therefore a better amino compound for amine-grafting to improve the adsorption of CO₂ than an aliphatic amine. The grafted primary amino groups should be accessible to capture the maximum amount of CO₂.

Fig. 8 Schematic diagram of the AEAPS-grafting reaction with pretreated MWCNTs.

The primary amino groups react with CO₂ in a 1 : 1 molar ratio and show a higher capacity for CO₂ adsorption.⁴⁸ The low CO₂ adsorption capacities obtained for the t-MWCNTs–AEAPS and h-MWCNTs–AEAPS are mainly attributed to the consumption of the primary amino groups by possible side-reactions, including the NH₂–silicon polymerization and amide reactions of amines as (Fig. 9). AEAPS is a bifunctional organosilane and the N of the –NH₂ group is nucleophilic. The oxygen of the carboxylic acids generated by acid oxidation shows strong electron-withdrawing properties and the carbon connected to the oxygen therefore has a small δ⁺ value and promotes the nucleophilic attack of the NH₂ groups (Fig. 9a). The toluene solvent is a good water-carrying agent and may also promote the amide reaction. Fig. 9b shows the direct attack of the nucleophilic amino groups on the silicon center in the presence of acidic carboxylic acids, where the methoxyl group acts as a leaving group under acid-catalyzed conditions. Polymerization caused by excessive AEAPS-grafting on the MWCNTs surface would block the pores and limit the diffusion of CO₂ into the internal pores.³⁴ Although the highest loading of amino groups anchored to the pretreated MWCNTs was obtained by HNO₃ pretreatment, some NH₂ groups were hindered and were not free to act as active adsorption sites for CO₂. Thus the adsorption capacity obtained was low.

Fig. 9 Possible side-reactions in AEAPS-grafting in the presence of carboxylic acids: (a) amide reaction and (b) NH₂–silicon polymerization.

3.3.2 Amine-grafting efficiency for CO₂ capture. Table 4 shows the amine efficiencies for the capture of CO₂ (η, CO₂/N, mol mol⁻¹) for AEAPS-grafted MWCNTs; some previously reported results for other amine-modified adsorbents are listed for comparison. The amine efficiency (CO₂/N) of h-MWCNTs–AEAPS is 0.13 mol mol⁻¹, which is only 54.2% of that of o-MWCNTs–AEAPS, although the N content is high. The low amine efficiency is partly a result of the reduction of the primary amino groups by side-reactions. The N contents of the AEAPS-grafted adsorbents determined by elemental analysis range from 1.81 to 2.77 mmol g⁻¹, which are much lower than those of other adsorbents obtained by the impregnation of samples with polyethyleneimine and grafting with aminoethylaminoethylyaminopropyltrimethoxysilane. However, the amine efficiencies for the capture of low concentrations of CO₂ are similar. The adsorption capacity for high concentrations of CO₂ at 15 vol% is 1.31 mmol g⁻¹ and the corresponding value of η is 0.49, which is close to the theoretical maximum value of 0.5. The amine efficiency obtained is much higher than that of the other adsorbents listed. The well-distributed amino groups on the support surface obtained by chemical grafting methods promote the amine efficiency of CO₂ adsorption compared with amino groups obtained by impregnation.

Table 4 Summary of the effect of amine-grafted MWCNTs adsorbents on CO₂ adsorption^a

Support material	Amine type	Method	CN (mmol g^−1)	Adsorption capacity		Operating conditions		CO2/N (mol mol^−1)	Reference
				mg g^−1	mmol g^−1	Cin (%)	T (°C)
a Cin = influent concentration of CO2; tri = aminoethylaminoethylyaminopropyltrimethoxysilane; PEI = polyethyleneimine; * calculated values.
t-MWCNTs	AEAPS	Grafting	1.81	—	0.38	2	25	0.21	This work
h-MWCNTs			2.77	—	0.36	2		0.13
o-MWCNTs			2.66	—	0.64	2		0.24
o-MWCNTs			2.67	—	1.31	15		0.49
MCM-41	Tri	Anhydrous grafting	5.75*	—	1.04	5	25	0.181	Harlick et al.^3
PE–MCM-41		Anhydrous grafting	6.10*	—	1.55			0.254
PE–MCM-41		Water-aided grafting	7.98*	—	2.65			0.332
HMS-C	45% PEI	Impregnation	10.19*	94	2.14*	100	75	0.21	Chen et al.^49
	70% PEI		16.33*	108	2.45*			0.15
HMS-T90	60% PEI		13.93*	184	4.18*			0.3
	70% PEI		16.30*	165	3.75*			0.23
SBA-15	30% TEPA + 20% DEA	Impregnation	—	—	—	100	75	0.4	Yue et al.^50

---

3.4 Optimization of CO₂ capture on AEAPS-grafted MWCNTs

3.4.1 Effect of operating temperature. Fig. 10 shows the effect of operating temperature on CO₂ adsorption. The CO₂ adsorption capacities of the o-MWCNTs–AEAPS maintained their high level with increasing operating temperatures. The adsorption capacity showed a modest increase from 0.61 to 0.64 mmol g⁻¹ from 15 to 30 °C. This is because the increased temperature promoted the reaction between CO₂ and the amino groups and improved the diffusion of CO₂ into the internal pores of the adsorbent. A slight decrease occurred when the temperature increased to 40 °C, indicating the exothermic nature of the chemisorption adsorption process. The adsorption capacities of the pristine MWCNTs with increasing operating temperature are also presented for comparison. The gradually decreasing adsorption capacities show that physical adsorption may be dominant.²⁸ Therefore a temperature range of 17–25 °C in a confined space is suitable for the applications. As room temperature in confined spaces is usually set at 25 °C, this temperature was chosen for the subsequent adsorption experiment.

Fig. 10 Effect of operating temperature on CO₂ adsorption (C₀ = 2 vol%, Q = 50 mL min⁻¹, dry gas mixture).

3.4.2 Effect of moisture in the influent gas stream. The humidity in confined spaces is high if several people live or stay in the very limited space for a long time. Not only water vapor, but also water fog are present in the atmosphere. The effects of a high water content on the CO₂ adsorption process in confined spaces was therefore investigated. Moisture in the gas mixture plays an important role in CO₂ adsorption. The adsorption under humid conditions was hindered for many hydrophilic adsorbents. Fig. 11 shows the CO₂ adsorption capacities of o-MWCNTs–AEAPS for different water contents in gas streams at 25 °C. The capacity remarkably improved to 0.72 mmol g⁻¹ with an increase in the water content to 4 vol%, which is almost 13% higher than that in dry gas. However the capacity slightly decreased to 0.71 mmol g⁻¹ at 7 vol% water content, which may be caused by the competitive adsorption of H₂O and CO₂ at the adsorption sites. The capacity remained stable with further increases in the water content. This shows that the MWCNTs had a high tolerance to moisture compared with silica-based adsorbents.^22,33–35 As the dissolved CO₂ in the surface-adsorbed water cannot be seen, the promotion effects were attributed to the formation of bicarbonate ions (HCO₃⁻), either directly or indirectly, when the amino groups reacted with CO₂ at a molar ratio of 1 : 1 with the assistance of H₂O.³² In contrast, two molecular amino groups were required to bond one molecule of CO₂ under anhydrous conditions. Therefore the promoted amine efficiency resulted in a high capacity for CO₂ adsorption.

Fig. 11 Effect of water content on CO₂ adsorption for o-MWCNTs–AEAPS (C₀ = 2 vol%, T = 25 °C, Q = 50 mL min⁻¹).

3.4.3 Adsorption performance in repeated operations. The excellent sustained adsorption performance of the adsorbents on multi-run adsorption/desorption operations is valuable in practical applications. The adsorption index was calculated based on the ratio of the adsorption capacity obtained relative to the first-run adsorption capacity. Fig. 12 shows the variation in adsorption performance on the basis of the adsorption index for 10 repeated cycles. The adsorption performance showed little fluctuation compared with the original capacity and no significant reduction was obtained. This indicates that the adsorption capacity of the adsorbent can be readily recovered. The desorption process was highly efficient where the operation was conducted at 75 °C with a 5 kPa vacuum for 5 min. The regeneration temperature of common solid amine sorbents has been reported to be typically from 50 to 100 °C.⁴⁸ The relatively low temperature in this work indicates the low energy cost required. Similarly, Su et al.⁴² carried out a desorption process for an APTS-grafted CNT adsorbent at 120 °C with a vacuum of 0.145 atm (14.7 kPa) and a regeneration time of 5 min was obtained. Although a similar time was obtained in this work, a lower temperature was applied and one-third of the vacuum value was required. It is estimated that half of the total energy cost can be saved. The adsorbent showed a high thermal stability and low regeneration cost and it is promising for use in direct CO₂ capture, especially at low concentrations of CO₂.

Fig. 12 Adsorption performance for 10 repeated cycles (C₀ = 2 vol%, T = 25 °C, Q = 50 mL min⁻¹).

4. Conclusion

The effects of the type and content of O-containing groups induced on MWCNTs by different pretreatments were investigated. The alkoxyl groups of AEAPS react with the O-containing groups on the pretreated MWCNTs through a silylation reaction. The obtained adsorbents were used to remove CO₂ in confined spaces at ambient temperature. A high silylation efficiency can result in high amine loading as a result of the high content of O-containing groups on the pretreated MWCNTs; however, this was not proportional to the high CO₂ adsorption capacity. The presence of carboxylic acids generated by HNO₃ treatment resulted in side-reactions, including the amide reaction and NH₂–silicon polymerization, which consumed the primary amine groups and resulted in a lower capacity. The MWCNTs adsorbents pretreated with O₂ and AEAPS-grafting showed a high CO₂ adsorption capacity of 0.64 mmol g⁻¹, which is almost 7.1 times that of o-MWCNTs. The amine-grafting efficiencies at 2 vol% and 15 vol% are 0.24 mol CO₂/mol N and 0.49 mol CO₂/mol N, respectively; the latter is close to the theoretical maximum value of 0.5. The capacity maintained this high level at operating temperatures of up to 40 °C. The adsorbent can be readily recovered and no significant reduction in adsorption was observed after 10 repeated adsorption/desorption cycles. If a lower desorption temperature was applied and one-third of the vacuum value, then this would significantly reduce the energy cost. This indicates that O₂ gas oxidation is both simple in operation and highly efficient for the pretreatment of MWCNTs if subsequent grafting with aminosilane is used. The adsorbent obtained has a high capacity for CO₂ adsorption, a high thermal stability, a high tolerance to moisture and a low regeneration cost, and is promising for the direct capture of CO₂ in confined spaces.

Acknowledgements

This work was partially supported by the National Natural Science Foundation of China (NSFC) (nos 21076188 and 21406197), the Natural Science Foundation of Zhejiang Province (LY12E06001) and the Postdoctoral Research Foundation of Zhejiang Province (BSH1302042).

References

1. Y. Kuwahara, D.-Y. Kang, J. R. Copeland, N. A. Brunelli, S. A. Didas, P. Bollini, C. Sievers, T. Kamegawa, H. Yamashita and C. W. Jones, Dramatic Enhancement of CO₂ Uptake by Poly(ethyleneimine) Using Zirconosilicate Supports, J. Am. Chem. Soc., 2012, 134(26), 10757–10760.
2. S. Choi, J. H. Drese and C. W. Jones, Adsorbent Materials for Carbon Dioxide Capture from Large Anthropogenic Point Sources, ChemSusChem, 2009, 2(9), 796–854.
3. P. J. E. Harlick and A. Sayari, Applications of Pore-Expanded Mesoporous Silica. 5. Triamine Grafted Material with Exceptional CO₂ Dynamic and Equilibrium Adsorption Performance, Ind. Eng. Chem. Res., 2006, 46(2), 446–458.
4. C. M. White, B. R. Strazisar, E. J. Granite, J. S. Hoffman and H. W. Pennline, Separation and Capture of CO₂ from Large Stationary Sources and Sequestration in Geological Formations – Coalbeds and Deep Saline Aquifers, J. Air Waste Manage. Assoc., 2003, 53(6), 645–715.
5. A. Samanta, A. Zhao, G. K. H. Shimizu, P. Sarkar and R. Gupta, Post-Combustion CO₂ Capture Using Solid Sorbents: A Review, Ind. Eng. Chem. Res., 2011, 51(4), 1438–1463.
6. S. Satyapal, T. Filburn, J. Trela and J. Strange, Performance and Properties of a Solid Amine Sorbent for Carbon Dioxide Removal in Space Life Support Applications, Energy Fuels, 2001, 15(2), 250–255.
7. R. Serna-Guerrero and A. Sayari, Modeling adsorption of CO₂ on amine-functionalized mesoporous silica. 2: kinetics and breakthrough curves, Chem. Eng. J., 2010, 161(1–2), 182–190.
8. P. Moloney, C. Huffman, O. Gorelik, P. Nikolaev, S. Arepalli, R. Allada, M. Springer and L. Yowell, Advanced life support for space exploration: air revitalization using amine coated single wall carbon nanotubes, Mater. Res. Soc. Symp. Proc., 2005, 2005, 59–64.
9. R. Allada, P. Moloney, M. Anderson, F. Smith, S. Arepalli, L. Yowell, J. Chattopadhyay, K. Shah, W. E. Billups and T. Filburn, Nanoscale materials for human spaceflight applications: regenerable carbon dioxide removal using single-wall carbon nanotubes, in SAE Technical Papers, Norfolk, VA, United States, 2006.
10. S. Choi, J. H. Drese, P. M. Eisenberger and C. W. Jones, Application of Amine-Tethered Solid Sorbents for Direct CO₂ Capture from the Ambient Air, Environ. Sci. Technol., 2011, 45(6), 2420–2427.
11. Z. Chen, N. Ren, A. Wang, Z. Zhang and Y. Shi, A novel application of TPAD-MBR system to the pilot treatment of chemical synthesis-based pharmaceutical wastewater, Water Res., 2008, 42(13), 3385–3392.
12. W. Lu, J. P. Sculley, D. Yuan, R. Krishna, Z. Wei and H.-C. Zhou, Polyamine-Tethered Porous Polymer Networks for Carbon Dioxide Capture from Flue Gas, Angew. Chem., Int. Ed., 2012, 51(30), 7480–7484.
13. V. Manovic and E. J. Anthony, Steam reactivation of spent CaO-based sorbent for multiple CO₂ capture cycles, Environ. Sci. Technol., 2007, 41(4), 1420–1425.
14. E. B. Rinker, S. S. Ashour and O. C. Sandall, Absorption of Carbon Dioxide into Aqueous Blends of Diethanolamine and Methyldiethanolamine, Ind. Eng. Chem. Res., 2000, 39(11), 4346–4356.
15. M. L. Gray, Y. Soong, K. J. Champagne, J. Baltrus, R. W. Stevens Jr, P. Toochinda and S. S. C. Chuang, CO₂ capture by amine-enriched fly ash carbon sorbents, Sep. Purif. Technol., 2004, 35(1), 31–36.
16. H. An, B. Feng and S. Su, CO₂ capture capacities of activated carbon fibre-phenolic resin composites, Carbon, 2009, 47(10), 2396–2405.
17. J. Wei, J. Shi, H. Pan, Q. Su, J. Zhu and Y. Shi, Thermal and hydrothermal stability of amino-functionalized SBA-16 and promotion of hydrophobicity by silylation, Microporous Mesoporous Mater., 2009, 117(3), 596–602.
18. J.-E. Park, H.-K. Youn, S.-T. Yang and W.-S. Ahn, CO₂ capture and MWCNTs synthesis using mesoporous silica and zeolite 13X collectively prepared from bottom ash, Catal. Today, 2012, 190(1), 15–22.
19. Y. Le, D. Guo, B. Cheng and J. Yu, Amine-functionalized monodispersed porous silica microspheres with enhanced CO₂ adsorption performance and good cyclic stability, J. Colloid Interface Sci., 2013, 408, 173–180.
20. J. Yu, Y. Le and B. Cheng, Fabrication and CO₂ adsorption performance of bimodal porous silica hollow spheres with amine-modified surfaces, RSC Adv., 2012, 2(17), 6784–6791.
21. L. Bastin, P. S. Barcia, E. J. Hurtado, J. A. C. Silva, A. E. Rodrigues and B. Chen, A Microporous Metal–Organic Framework for Separation of CO₂/N₂ and CO₂/CH₄ by Fixed-Bed Adsorption, J. Phys. Chem. C, 2008, 112(5), 1575–1581.
22. C. Lu, H. Bai, B. Wu, F. Su and J. F. Hwang, Comparative Study of CO₂ Capture by Carbon Nanotubes, Activated Carbons, and Zeolites, Energy Fuels, 2008, 22(5), 3050–3056.
23. O. Leal, C. Bolívar, C. Ovalles, J. J. García and Y. Espidel, Reversible adsorption of carbon dioxide on amine surface-bonded silica gel, Inorg. Chim. Acta, 1995, 240(1–2), 183–189.
24. Y. Liu, J. Shi, J. Chen, Q. Ye, H. Pan, Z. Shao and Y. Shi, Dynamic performance of CO₂ adsorption with tetraethylenepentamine-loaded KIT-6, Microporous Mesoporous Mater., 2010, 134(1–3), 16–21.
25. J. C. Hicks, J. H. Drese, D. J. Fauth, M. L. Gray, G. Qi and C. W. Jones, Designing Adsorbents for CO₂ Capture from Flue Gas-Hyperbranched Aminosilicas Capable of Capturing CO₂ Reversibly, J. Am. Chem. Soc., 2008, 130(10), 2902–2903.
26. J. Liu, Y. Liu, Z. Wu, X. Chen, H. Wang and X. Weng, Polyethyleneimine functionalized protonated titanate nanotubes as superior carbon dioxide adsorbents, J. Colloid Interface Sci., 2012, 386(1), 392–397.
27. M. G. Plaza, C. Pevida, A. Arenillas, F. Rubiera and J. J. Pis, CO₂ capture by adsorption with nitrogen enriched carbons, Fuel, 2007, 86(14), 2204–2212.
28. Q. Ye, J. Jiang, C. Wang, Y. Liu, H. Pan and Y. Shi, Adsorption of Low-Concentration Carbon Dioxide on Amine-Modified Carbon Nanotubes at Ambient Temperature, Energy Fuels, 2012, 26(4), 2497–2504.
29. M. B. Yue, L. B. Sun, Y. Cao, Y. Wang, Z. J. Wang and J. H. Zhu, Efficient CO₂ capturer derived from as-synthesized MCM-41 modified with amine, Chem.–Eur. J., 2008, 14(11), 3442–3451.
30. Y. G. Ko, S. S. Shin and U. S. Choi, Primary, secondary, and tertiary amines for CO₂ capture: designing for mesoporous CO₂ adsorbents, J. Colloid Interface Sci., 2011, 361(2), 594–602.
31. N. Hiyoshi, K. Yogo and T. Yashima, Adsorption characteristics of carbon dioxide on organically functionalized SBA-15, Microporous Mesoporous Mater., 2005, 84(1–3), 357–365.
32. X. Xu, C. Song, B. G. Miller and A. W. Scaroni, Influence of Moisture on CO₂ Separation from Gas Mixture by a Nanoporous Adsorbent Based on Polyethylenimine-Modified Molecular Sieve MCM-41, Ind. Eng. Chem. Res., 2005, 44(21), 8113–8119.
33. S.-C. Hsu, C. Lu, F. Su, W. Zeng and W. Chen, Thermodynamics and regeneration studies of CO₂ adsorption on multiwalled carbon nanotubes, Chem. Eng. Sci., 2010, 65(4), 1354–1361.
34. F. Su, C. Lu and H. S. Chen, Adsorption, desorption, and thermodynamic studies of CO₂ with high-amine-loaded multiwalled carbon nanotubes, Langmuir, 2011, 27(13), 8090–8098.
35. M. M. Gui, Y. X. Yap, S. P. Chai and A. R. Mohamed, Multi-walled carbon nanotubes modified with (3-aminopropyl)triethoxysilane for effective carbon dioxide adsorption, Int. J. Greenhouse Gas Control, 2013, 14, 65–73.
36. Z.-Z. Zhu, Z. Wang and H.-L. Li, Functional multi-walled carbon nanotube/polyaniline composite films as supports of platinum for formic acid electrooxidation, Appl. Surf. Sci., 2008, 254(10), 2934–2940.
37. H. Gaspar, C. Pereira, S. L. H. Rebelo, M. F. R. Pereira, J. L. Figueiredo and C. Freire, Understanding the silylation reaction of multi-walled carbon nanotubes, Carbon, 2011, 49(11), 3441–3453.
38. C. Baleizão, Vanadyl salen complexes covalently anchored to single-wall carbon nanotubes as heterogeneous catalysts for the cyanosilylation of aldehydes, J. Catal., 2004, 221(1), 77–84.
39. V. Datsyuk, M. Kalyva, K. Papagelis, J. Parthenios, D. Tasis, A. Siokou, I. Kallitsis and C. Galiotis, Chemical oxidation of multiwalled carbon nanotubes, Carbon, 2008, 46(6), 833–840.
40. C. Velasco-Santos, A. L. Martínez-Hernández, M. Lozada-Cassou, A. Alvarez-Castillo and V. M. Castaño, Chemical functionalization of carbon nanotubes through an organosilane, Nanotechnology, 2002, 13(4), 495.
41. A. G. Gonçalves, J. L. Figueiredo, J. J. M. Órfão and M. F. R. Pereira, Influence of the surface chemistry of multi-walled carbon nanotubes on their activity as ozonation catalysts, Carbon, 2010, 48(15), 4369–4381.
42. F. Su, C. Lu, W. Cnen, H. Bai and J. F. Hwang, Capture of CO₂ from flue gas via multiwalled carbon nanotubes, Sci. Total. Environ., 2009, 407(8), 3017–3023.
43. J. P. Sculley and H.-C. Zhou, Enhancing Amine-Supported Materials for Ambient Air Capture, Angew. Chem., Int. Ed., 2012, 51(51), 12660–12661.
44. J. Liu, A. G. Rinzler, H. Dai, J. H. Hafner, R. K. Bradley, P. J. Boul, A. Lu, T. Iverson, K. Shelimov, C. B. Huffman, F. Rodriguez-Macias, Y.-S. Shon, T. R. Lee, D. T. Colbert and R. E. Smalley, Fullerene Pipes, Science, 1998, 280(5367), 1253–1256.
45. R. Sitko, E. Turek, B. Zawisza, E. Malicka, E. Talik, J. Heimann, A. Gagor, B. Feist and R. Wrzalik, Adsorption of divalent metal ions from aqueous solutions using graphene oxide, Dalton Trans., 2013, 42(16), 5682–5689.
46. E. R. Pérez, J. R. Garcia, D. R. Cardoso, B. R. McGarvey, E. A. Batista, U. P. Rodrigues-Filho, W. Vielstich and D. W. Franco, In situ FT-IR and ex situ EPR analysis for the study of the electroreduction of carbon dioxide in N,N-dimethylformamide on a gold interface, J. Electroanal. Chem., 2005, 578(1), 87–94.
47. H. F. Gorgulho, F. Gonçalves, M. F. R. Pereira and J. L. Figueiredo, Synthesis and characterization of nitrogen-doped carbon xerogels, Carbon, 2009, 47(8), 2032–2039.
48. F. Zheng, D. N. Tran, B. J. Busche, G. E. Fryxell, R. S. Addleman, T. S. Zemanian and C. L. Aardahl, Ethylenediamine-Modified SBA-15 as Regenerable CO₂ Sorbent, Ind. Eng. Chem. Res., 2005, 44(9), 3099–3105.
49. C. Chen, W.-J. Son, K.-S. You, J.-W. Ahn and W.-S. Ahn, Carbon dioxide capture using amine-impregnated HMS having textural mesoporosity, Chem. Eng. J., 2010, 161(1–2), 46–52.
50. M. B. Yue, L. B. Sun, Y. Cao, Z. J. Wang, Y. Wang, Q. Yu and J. H. Zhu, Promoting the CO₂ adsorption in the amine-containing SBA-15 by hydroxyl group, Microporous Mesoporous Mater., 2008, 114(1–3), 74–81.

---