Heteroatom-doped nanoporous carbon derived from MOF-5 for CO2 capture

Heteroatom-doped nanoporous carbon derived from MOF-5 for CO2 capture

Accepted Manuscript Title: Heteroatom-doped nanoporous carbon derived from MOF-5 for CO2 capture Authors: Xiancheng Ma, Liqing Li, Chen Ruofei, Wang C...

1MB Sizes 0 Downloads 3 Views

Accepted Manuscript Title: Heteroatom-doped nanoporous carbon derived from MOF-5 for CO2 capture Authors: Xiancheng Ma, Liqing Li, Chen Ruofei, Wang Chunhao, Hailong Li, Wang Shaobin PII: DOI: Reference:

S0169-4332(17)33329-9 https://doi.org/10.1016/j.apsusc.2017.11.069 APSUSC 37651

To appear in:

APSUSC

Received date: Revised date: Accepted date:

3-7-2017 7-11-2017 9-11-2017

Please cite this article as: Ma X, Li L, Ruofei C, Chunhao W, Li H, Shaobin W, Heteroatom-doped nanoporous carbon derived from MOF-5 for CO2 capture, Applied Surface Science (2010), https://doi.org/10.1016/j.apsusc.2017.11.069 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Heteroatom-doped nanoporous carbon derived from MOF-5 for CO2 capture

Xiancheng Ma a, Liqing Li a, *, Ruofei Chen a, Chunhao Wang a, Hailong Li a, Shaobin Wang b

School of Energy Science and Engineering, Central South University, Changsha 410083, Hunan, China

b

Department of Chemical Engineering, Curtin University, GPO Box U1987, Perth, WA 6845, Australia

SC R

IP T

a

* To whom correspondence should be addressed:

N

U

TEL: +86-731-8887-9863

M

A

Email: [email protected]

A

CC E

PT

ED

Graphical abstract

Highlights 

Four nanoporous carbons were prepared from a porous metal-organic framework (MOF-5) template and additional carbon source (i.e. urea) by carbonization at different temperatures (600-900 ℃).



The as-obtained sample MUC900 exhibits the highest surface areas (2307 m2·g-1) and pore volumes (2.54

mL·g-1). 

By changing the carbonization temperature it can finely tune the pore volume of the MUCT, which having a uniform pore size of around 4.0 nm.



The detailed interaction mechanism between functional groups and CO2 molecules is elucidated.

ABSTRACT: Four nanoporous carbons (MUCT) were prepared from metal-organic framework

IP T

(MOF-5) template and additional carbon source (i.e. urea) by carbonization at different temperatures

SC R

(600–900 ℃). The results showed that specific surface area of four samples was obtained in the range from 1030 to 2307 m2 g-1. By changing the carbonization temperature it can finely tune the pore volume of the MUCT, which having a uniform pore size of around 4.0 nm. With an increasing

U

carbonization temperature, the micropore surface area of MUCT samples varied slightly, but

A

N

mesopore surface area increased obviously, which had little influence on carbon dioxide (CO2)

M

adsorption capacity. The as-obtained sample MUC900 exhibited the superior CO2 capture capacity of 3.7 mmol g-1 at 0 ℃ (1 atm). First principle calculations were conducted on carbon models with

ED

various functional groups to distinguish heterogeneity and understand carbon surface chemistry for

PT

CO2 adsorption. The interaction between CO2 and N-containing functional groups is mainly weak Lewis acid-base interaction. On the other hand, the pyrrole and amine groups show exceptional

CC E

hydrogen-bonding interaction. The hydroxyls promote the interaction between carbon dioxide and functional groups through hydrogen-bonding interactions and electrostatic potentials, thereby

A

increasing CO2 capture of MUCT.

Keywords: MOF-5; nanoporous carbon; CO2 capture; hydrogen-bonding interactions; acid-base interactions; electrostatic potential

1. Introduction The greenhouse effect that causes global warming is largely associated with carbon dioxide (CO2)[1, 2]. Therefore, there is growing concern about the development of technologies to effectively adsorption or storage large amounts of carbon dioxide. Currently, porous carbon materials are considered as excellent candidates for the adsorbents of CO2 due to the high surface

IP T

area, good chemical and physical stability and controlled pore structure [3, 4]. As a result, a variety of methods, including chemical vapor decomposition (CVD) [5],laser ablation[6] , and templating

SC R

[7-9] as well as physical or chemical activation methods[10, 11], have been explored to prepare carbon and control their pore structures. Among these methods, the template method is an effective

U

method for preparation microporous and mesorporous porous carbon materials [12].

N

Metal-organic frameworks (MOFs) have obtained rapid development over the past decade due

A

to their interesting architectures and topologies, adjustable pore size, and extensive applications such

M

as catalyst, gas separation and storage, drug delivery, and so on [13-16]. Recently, the use of MOF as

ED

a template and / or precursor has rapidly developed nanoporous carbon materials. In general, there are two methods to prepare porous carbon materials using MOFs as templates. One method is to use

PT

MOFs as templates and carbon precursors because such MOFs contain abundant organic

CC E

species[17-19]. The other is to add an additional carbon source that is disseminated into the pore space of the MOFs; such as Furfural (FA) or glucose [12, 20-23]. Two methods have merits in the preparation of carbons with designed nanostructures and functions[24], since additional carbon

A

sources can bring about heteroatoms such as nitrogen, sulfur and oxygen, which can change the properties of porous carbon and hence enhance the CO2 capture performance. MOF-5 (ZnO4(OOCC6H4COO)3) can be prepared at room temperature with a three-dimensional channel structure [25]. Hierarchical carbon with high surface area and larger pore volume of was prepared by carbonization of MOF-5[26]. The disorder micropores have a slower gas diffusion rates,

the presence of mesopores can better promote gas diffusion and transport into micropores by reducing the resistance to diffusion and pathway distance. Therefore, taking into account the diffusion kinetics, hierarchical pore as a highly interconnected system is necessary for fast gas transport. With this in mind, we designed a MOF-5 as a template and urea (nitrogen source) for porous carbon with hierarchical porosity (macro-, meso-, and micropore), and a nitrogen-doping

IP T

framework. The variation of the MUCT samples structure with increasing temperature and their CO2 capture performance is discussed in detail.

SC R

2. Experimental section 2.1 Preparation of MOF-5

U

MOF-5 was synthesized [27] by stirred vigorously of zinc nitrate (1.21 g) and terephthalic acid

N

(0.34 g) in triethylamine (1.6 g) and N,N-dimethylfomamide solution (DMF, 40 mL) for 0.5 h at

A

room temperature. Subsequently, a small amount (0.2 mL) of H2O2 (30 wt.%) aqueous solution was

M

added to the above solution. Finally, after washing with DMF (30 mL), and then dried at 120 ℃, the

ED

MOF-5 was obtained. 2.2 Preparation of carbon materials

PT

Four nanoporous carbon materials were prepared using MOF-5 as template and carbon

CC E

precursor and urea as a nitrogen source, in which urea immersed in the pores of MOF-5. The prepared MOF-5 was degassed at 200 ℃ for 3 h under a nitrogen atmosphere to remove the solvent molecules in the pore channel of MOF-5. The degassed MOF-5 (1 g) was added in 40 mL urea

A

ethanol solution (0.15 mol L-1), stirred vigorously for 3 h at room temperature and resulting mixture was stood overnight, then filtered and washed three times with ethanol. Then the MOF-5/urea composite was placed into a corundum boat and transferred into tube furnace for carbonization. Then, carbonization of the composite material was carried out at 600, 700, 800, and 900 ℃ for 5 h with a heating gradient of 4 ℃ min-1. After carbonization, the samples were washed with HCl

solution (5%) vigorously and subsequently washed with distilled water several times, and dried under vacuum at 120 ℃ for overnight. The products were labeled as MUCT samples, where MUC represents the porous carbon obtained by carbonizing the MOF-5 and urea composite, and T refers to the carbonization temperature. 2.3 Characterization

IP T

Transmission electron microscopy (TEM) images were performed using a Tecnai G2 20S-Twin electron microscope equipped with a cold field emission gun under an acceleration voltage of 200

SC R

kV. X-ray diffraction (XRD, X’ PertPro MPD, PANalytical B.V., NED) patterns were carried out on a PANalytical powder diffractometer using Cu/Ka as the radiation source operated at 40 mA and 40

U

kV. N2 adsorption isotherm was measured by using a SA3100 specific surface area analyzer

N

(Beckman-Coulter Instrument Co., USA) at 77K. The sample was ground to a powder, degassed at

A

100 ° C under vacuum for 5 hours, and then tested. The surface area, pore volume and pore size

M

distribution were calculated from the N2 isotherms by Brunauer-Emmett-Teller (BET) equation[28,

ED

29], 26, Horvath–Kawazoe equation [30] and Barrett−Joyner−Halenda (BJH) method[31]. Fourier transform infrared spectroscopy (FTIR) spectra were measured with a PerkinElmer-2000 FTIR

PT

spectrometer ranging from 4000 to 400 cm-1. The types and amount of functional groups on MUCT

CC E

surfaces were measured using a K-Alpha 1063 XPS analyzer (Thermo Fisher Scientific Inc.). The adsorption of CO2 was measured using an Autosorb-iQ-MP (Quantachrome) static volume

analyzer. The sample (about 50 mg) was degassed at 120 ℃ for several hours before adsorption

A

analysis. CO2 adsorption experiments were carried out at two temperatures of 0 ℃and 25 ℃. 2.4 Computational details All of the first principle calculations were conducted on the basis of density functional theory (DFT) with the use of the DMol3 code [32]. The energy of the functional group and the interaction of carbon dioxide was calculated using the DFT calculation coupled the van der Waals correct

correction (DFT-D)[33]. Perdew, Burke, and Ernzerhof (PBE) within the generalized gradient approximation (GGA-PBE) was selected [34]. The atomic orbit is described using double numeric polarization (DNP) basis set, which is comparable to 6-31G (d, p). The type of core processing is set up using a DFT half-core Pseudopots (DSPP) specifically designed for DMol3 calculations [35]. the real-space orbital global cutoff is 3.7 Å. The convergence threshold parameters for the optimization

IP T

were 10−5 Hartree (energy), 2 × 10−3 Hartree (gradient), and 5 × 10−3 Hartree (displacement), respectively[36]. Therefore, BSSE effects need not be taken into consideration for calculating the



Eads  EsurfaceCO2  Esurface  ECO2

SC R

energies. The CO2 adsorption energy is calculated in the following manner:



U

where Eads, Esurface+CO2, and ECO2 are adsorption energy and total energy of adsorption-adsobate

N

complex, carbon surface, and isolated CO2, respectively. The binding energy of CO2 in

A

group-functionalized surface is a significant parameter for the CO2 adsorption on the

M

group-functionalized surface. An enhancement in the Eads would be very beneficial for capturing

3.Results and discussion

ED

CO2 of the MUCT.

PT

3.1 Textural properties of the prepared MUCT

CC E

Fig. 1 display TEM images of MC900. As shown in Fig. 1a, the shape and size of the MC900 were similar to the shape and size of the parent MOF-5, which clearly demonstrates that the cube porous framework remains well during the carbonization process. The pore size of the porous carbon

A

was observed to be about 4 nm in Fig. 1b. The pore size estimated from the TEM image was close to the pore size determined from the nitrogen adsorption data. XRD patterns of MUC600, MUC700, MUC800, andMUC900 were shown in Fig. 2a. As shown in Fig. 2a, after the heat treatment of the MOF-5/urea composite, the diffraction peak of the impurity cannot be observed, indicating that the carbonaceous material is completely converted. During the

carbonization process, ZnO was reduced Zn by carbon and then vaporized at the temperature of 900℃, and the other residual was removed by the acid wash [37]. The graphitization degree of MUCTs samples was estimated from Raman spectra (Figure 2b). All the samples present two intensive peaks at around 1348 and 1598 cm-1, which can be attributed to D-band (D) and G band (G). D may be associated with the sp3 defective, and G could arise from

IP T

the bond stretching of all sp2-bonded pairs. The intensity ratio of the D band and G band (ID/IG) can reflect the disorder degree of MUCTs. Furthermore, the ID/IG ratio of MUC600, MUC700, MUC800

SC R

and MUC900 is 1.09, 1.10, 1.11 and1.13, respectively, showing that the ID/IG ratio changes slightly with increasing activation temperature.

U

As shown in Fig. 3a, the nitrogen adsorption-desorption isotherm was measured by calculating

N

the pore size distribution and surface area of the four MUCT samples. Calculation results were

A

summarized in Table 1. Nitrogen adsorption-desorption isotherms of the MCTs samples show

M

type-IV curves with a sharp capillary condensation step in the relative pressure range of 0.4-0.6 and

ED

H1-type hysteresis loop, indicative of uniform cylindrical pores. The type of the N2 adsorption isotherms of MUCT samples indicates the presence of micropores, mesopores and macropores. The

PT

pore size distribution of the four MUCT samples calculated from nitrogen adsorption isotherms was

CC E

shown in Fig. 3b. As shown in Fig. 3b, all samples show similar pore size profiles, most of which are centered around 4.0 nm, and pore apertures around 0.7 and 1.7 nm. These carbon materials have high BET surface areas (1030-2307 m2 g-1) and pore volumes (1.10-2.54 mL g-1). Among them,

A

MUC900 obtained the highest specific surface area (2307 m2 g-1). During the carbonation process (≥900 ° C), the evolution of Zn, CO and CO2 is due to the reduction of C and O and the framework ligands of ZnO leaving more defects or hollow carbon networks. This process is a simple self-activation of carbon, similar to a carbonaceous precursor that is activated by physical or chemical activation (KOH or CO2) to obtain porous carbon and is not etched to the apparatus [38-40].

With the increase of carbonization temperature, the micropore structure of MUCT samples changed slightly, but the mesoporous and macroporous structures increased obviously 3.2 Surface chemical properties of MUCTs The evolution of the chemical composition of the obtained porous carbon was also characterized by Fourier-transform infrared (FTIR) spectroscopy. As shown in Fig. 4, the broad peak

IP T

at ca. 3420 cm-1 could be attributed to the N-H or O-H stretching model of amine group and hydroxyl group [41]. The peaks around 2900 cm-1 may be ascribed to the C-H -CH2-NH-CH2- or

SC R

-CH2-NH-CH3 stretching vibration [42]. The bands ca. 1620 cm-1 were related to C=C stretching vibration, which derived from the stretching vibrations of aromatic ring. Two peaks at 1152 and

U

1090 cm-1 were assigned to the C-O or C-C stretching vibration, respectively [43].

N

Since functionalized-groups surface property of porous carbon also plays an important role in

A

achieving high CO2 performance, the surface composition of MUC600, MUC700, MUC800, and

M

MUC900 were analyzed by XPS. The content of oxygen and nitrogen in the material will have a

ED

direct impact on the CO2 uptake. Fig. 5a shows the high resolution O1s photoelectron spectrum for MUCT samples and their deconvolution leads to four separate peaks with binding energies around

PT

532.3 (C=O), 533.5 (C-O), 534.3 (C-OH), and 536.3 eV (H2Oads and CO2ads) labeled as O1, O2, O3,

CC E

and O4 respectively(Table 2). Peak O1 is assigned to C=O of carbonyl or ketone group, O2 to carbonyl oxygen of esters, amides, oxygen of carboxylic groups and anhydrides, O3 to hydroxyl group, and O4 to oxygen in carbon dioxide and water[44]. The fitting of the N1s peaks (Fig. 5b)

A

gave the following binding energies: pyridinic-N (398.6 eV), pyrrolic-N (399.8 eV), amino (400.5 eV), graphitic-N (401 eV) and oxidized-N (402.4 eV) signed as N1, N2, N3, N4, N5 respectively with relative area contribution and corresponding FWHM listed in Table 3. Upon pyrolysis, the change in temperature causes a significant change in the type of nitrogen in the material. For the MUC600, the amide is obtained. With the temperature increasing to 700℃, the amide was

disappeared. These species are transformed into protonated graphitic-N, pyridinic-N, and oxidized-N species. Noticeably, samples pyrolyzed at 600℃ provide high amounts of amide. The amide might serve as anchors for CO2 capture. 3.3 CO2 adsorption capacity CO2 capture capacity of the four samples was investigated at two temperatures (0 and 25 ℃)

IP T

and related data were listed in Table 4. Besides the microporous structure, the presence of functional groups on the carbon surface plays a significant role in determining the CO2 uptake capacity as well.

SC R

Thus, the doping of heteroatoms into the carbon framework is an effective way to enhance CO2 uptake. Many studies have reported that O-doped and N-doped into carbon framework has

U

significant effects on adsorption [39, 45]. N-doped into carbon may increase CO2 adsorption by

N

acid-base interactions. Besides these dispersive interactions, hydrogen bonding of CO2 molecules to

A

the surface of adsorbents has been indicated as a possible retention mechanism by Liu and

M

coworkers who studied CO2 adsorption [46]. Recently, Xing and Ma reported that the effect of

ED

nitrogen species in enhancing CO2 adsorption has been interpreted as hydrogen bonds (N-H…O and O-H…O).[37, 47]. At 1 bar, the porous carbon MUCTs with different functional groups and pore

PT

structure have high CO2 capture in the range of 2.22-2.44 (97.68-107.36 mg g-1) and 3.34-3.71 mmol

CC E

g-1 (146.96-163.24 mg g-1) at 25 and 0 ℃, respectively. It is noteworthy that MUC800 and MUC900 have high specific surface area (1365-2307 m2 g-1) and a pore volume (1.99-2.54 ml g-1) relative to the MUC600 and MUC700, but they show lower CO2 capture of 2.22-2.32 mmol g-1 (97.68-102.08

A

mg g-1) than the MUC600 and MUC700 at 25 ℃ (Table 3). It can be inferred that CO2 adsorption capacities would be improved for the MUCT associated with more functional groups. Lower nitrogen contents and C-OH groups have negative effects on the CO2 capture the mesopore and macropore structure play a less important role in enhancing the CO2 capture. These results clearly show that the CO2 capture is largely dependent on the microporous surface area, nitrogen contents

and C-OH groups of porous carbons. As expected, when the adsorption temperature decreases, the amount of carbon dioxide adsorbed increases (Fig. 6). When decreasing the capture temperature from 25 to 0 ℃, the CO2 capture remarkable increase by 45.5 %, 48.7 %, 50.5 %, and 60.6 % from 2.44, 2.32, 2.22, and 2.31 mmol g-1 to 3.55, 3.45, 3.34, and 3.71 mmol g-1 for MUC600, MUC700, MUC800, and MUC900 at 1 bar, respectively (Table 3). There results showed that the maximum

IP T

CO2 capture capacity for MUC900 may be assigned to the most surface area. At such low pressure, the CO2 capture capacity of the samples at 25 ℃ is in the range of 0.42-0.72 mmol g-1. The amount

SC R

of CO2 adsorbed decrease when decreasing the nitrogen contents and C-OH of porous carbon for MUC600, MUC700, MUC800, and MUC900.

U

The adsorption isothermal data were fitted with well-known two-parameter models, Langmuir

N

[48] and Freundlich [49], which are expressed as: q K p

Langmuir, q  1 max KL p L

A

(1)

Freundlich, q  K f( p)1/ n

M

(2)

ED

where p is the adsorption pressure of CO2 [50], q is the adsorption capacity (mg g-1), qmax is the monolayer capture capacity (mg g-1), KL is Langmuir isotherm constants, Kf (mg g-1) is an empirical

PT

constant and 1/n (dimensionless) is the heterogeneity factor. From Table 5, determination coefficient,

CC E

R2 of both, Langmuir and Freundlich equation, is >0.99, indicating the good agreement with the experimental data for these adsorbents. ( T2  T1 ) Qst / (R TT 1 2 )  In( p2 / p1 )

(3)

A

Isosteric heat of adsorption (Qst) were calculated from CO2 adsorption isotherms measured at

two temperatures using the Clausius-Clapeyron equation (Eqs(3)); where p2 and p1 are the p (CO2) values; T1 and T2 are 0 and 25 ℃, respectively; and R is the universal gas constant. The plots of Qst as a function of the CO2 capture for MUCT are presented in Fig. 7. The isosteric heat of CO2 adsorption were calculated to lie in the range 26.7-20.7, 25.9-19.9, 23.9-18.6, and 20.7-15.9 kJ mol-1

for MUC600, MUC700, MUC800, and MUC900,respectively. Sample MT600, which contains the highest N content exhibits the highest Qst of 26.7-20.7 kJ mol-1 at highest uptake. In addition, Sample MUC900, despite having the highest surface area, has a moderate isothermal adsorption heat of 20.7-15.9 kJ mol-1. These results show that the presence of the N group and the C-OH group plays a significant role in determining the interaction between the group-functionalized surface and the

IP T

CO2 if the N content and C-OH group was high. On the other hand, the sample MUC900, despite having the highest surface area, has a moderate isothermal adsorption heat of 20.7-15.9 kJ mol-1. Our

SC R

data, therefore, demonstrates the importance of C-OH and N functional groups on porous carbon

U

materials in determining CO2 uptake.

N

3.4 The detailed interaction mechanism between functional groups and CO2

A

In order to distinguish which factor of porous carbon was affective for CO2 adsorption, the

M

mesopore and macropore structure as well as N1, N2, N3 and O3 content of the MUCT is

ED

correlation with the amount of CO2 adsorbed. Fig. 8a and Fig. 8b show that the relationship between the CO2 uptake and mesopore and macropore structure. It is found that there is opposite

PT

tread correlating the CO2 uptake and mesopore and macropore structure at 25 ℃ and 0.15 bar, and

CC E

it is slowed by the increasing temperature and reducing pressure. It further implies that mesopore and macropore structure play a less important role in CO2 uptake at ambient pressures. But by reducing the mass transfer resistance and path distance, promote gas transport and diffusion into

A

the micropores play an important role [51]. Fig. 8c and Fig. 8d show that there is no clear tread correlating the CO2 uptake at 1 bar and functional groups (i.e. pyridinic-N, pyrrolic-N, amino, and C-OH), indicating that the presence of functional groups does not govern the CO2 adsorption performance of the MUCT. However, there is a clear trend correlating the CO2 uptake at 0.15 bar ( 25 ℃) and functional groups of the MUCT, resulting in a much higher correlation coefficient of

0.75 and 0.79 than that of 1 bar (25 ℃) correlating with the CO2 uptake (correlation coefficient of 0.45 and 0.42). This indicates that the functional groups might have an important influence on the CO2 sorption behavior at 0.15 bar (25 ℃). O and N groups on the surface of the porous carbon have been validated by FTIR and XPS. The detailed interaction between the functional group and the carbon dioxide molecule is graphical.

IP T

O and N groups on the surface of the porous carbon have been validated by FTIR and XPS. The detailed interaction mechanism between group-functionalized surface and CO2 molecules is

SC R

summarized in Fig. 9. Among the several kinds of functional groups, the highest binding energy of CO2, 21.26 kJ mol-1, is obtained in the pyridine group (Fig. 9a). Upon adsorption, CO2 locates

U

closely to the nitrogen atom of pyridine group due to Lewis acid-base interaction where the lone

N

pair electron of nitrogen atom donates the electronic change to the carbon atom of CO 2 and steric

A

hindrance of the interaction between the functional group and CO2 [52]. The different adsorption

M

energetics and configurations of CO2 on functionalized carbon surface can be understood by

ED

examining their bonding nature with CO2. For the pyrrole and amine group-functionalized surface (Fig. 9b, i), CO2 interacts with the nitrogen and hydrogen atom of the functionalized surfaces.

PT

Here, the highest adsorption energy of pyrrole and amine is 18.60 and 16.45 kJ mol-1 respectively,

CC E

and the horizontal and next to pyrrole and amine group configuration (HN) are 10.82 and 5.72 kJ mol-1 respectively, and the distance between O atom of CO2 and H atom of functional groups is 2.290 and 2.544 Å, respectively (Fig. 9d, e). Therefore, on the pyrrole and amine

A

group-functionalized surface, two types of interaction, the Lewis-base and hydrogen-bonding interaction, play a significant role in determining adsorption behaviors [53]. To confirm the effect of the hydrogen-bonding interaction, nitrogen atom of pyrrole and amine group-functionalized surface was replaced with C-H (Fig. 9g, h). Recalculation of Eads with replacement of nitrogen atom with C-H exhibits a weaker interaction, resulting in Eads=3.97 kJ mol-1. Therefore, for the

pyrrole and amine, the hydrogen-bonding interaction plays a major role in CO2 adsorption behavior. When CO2 is adsorbed on a hydroxyl group-functionalized surface, the highest adsorption energy is 14.88 kJ mol-1 (Fig. 9c). For the horizontal and next to the group configuration (HN), the adsorption energy is 12.11 kJ mol-1 (Fig. 9f). Regarding the interactions of CO2 molecule with this functional group, the interaction distances suggest two scenarios: (i)

IP T

hydrogen-bonding interaction between the hydrogen atom of hydroxyl group and the oxygen atom of CO2 molecule[54]; oxygen contained within functional groups has a partial charge of -0.444 e

SC R

and therefore has the high electrostatic potential among all the surface atoms to donate electrons to the CO2 molecules. The hydroxyl group (HN) is different from other cases and implies that

U

hydrogen-bonding interaction (O-H…O) can occur. The distance between O atom of CO2 and H

N

atom of hydroxyl is 2.277 Å, shorter than that with H atom of N-H.

A

.

M

4. Conclusions

ED

In summary, metal-organic framework (MOF-5) was employed as a template to prepare nanoporous carbons materials, which shows high specific surface areas and superior CO2 uptake

PT

properties. Particularly, the as-obtained sample MUC900 exhibits the highest surface areas (2307

CC E

m2·g-1) and pore volumes (2.54 mL·g-1). By changing the carbonization temperature it can finely tune the pore volume of the MUCT, which having a uniform pore size of around 4.0 nm. With an increasing carbonization temperature, the micropore structure of MUCT samples varied slightly, but

A

mesopore and macropore structure increased obviously, and had little influence on CO2 adsorption capacity. For the MUC900, the highest CO2 adsorption capacity of 3.70 mmol g-1 at 0 ℃ (1 atm) was obtained. The effect of functional groups introduced in carbonaceous surface was investigated for CO2 capture by DFT. Among the several functional groups studied, pyridine group exhibit the highest binding energies of 21.26 kJ mol-1. The interaction between CO2 and N-containing functional

groups mainly consists of weak Lewis acid-based interaction. On the other hand, the pyrrole and amine groups show exceptional hydrogen-bonding interaction. The hydroxyl group increases the interaction between quadrupole CO2 and functional groups by hydrogen-bonding interaction and electrostatic potential, and thus gives rise to an adsorbent with high CO2 capture.

IP T

Acknowledgements

This work was supported by the National Key Technology Support Program (No.2015BAL04B02)

A

CC E

PT

ED

M

A

N

U

SC R

and the National Nature Science Foundation China (No. 21376274).

References

[1] N. Tit, K. Said, N.M. Mahmoud, S. Kouser, Z.H. Yamani, Ab-initio investigation of adsorption of CO and CO2 molecules on graphene: Role of intrinsic defects on gas sensing, Appl. Surf. Sci., 394 (2016) 219-230. [2] C. Chen, S. Bhattacharjee, Trimodal nanoporous silica as a support for amine-based CO2

IP T

adsorbents: Improvement in adsorption capacity and kinetics, Appl. Surf. Sci., 396 (2016).

[3] C. Liang, Z. Li, S. Dai, Mesoporous Carbon Materials: Synthesis and Modification, Angew.

SC R

Chem. Int. Ed., 47 (2008) 3696-3717.

[4] A.H. Lu, F. Schüth, Nanocasting: A Versatile Strategy for Creating Nanostructured Porous

U

Materials, Adv. Mater., 18 (2006) 1793-1805.

N

[5] Y. Xia, Z. Yang, R. Mokaya, CVD Nanocasting Routes to Zeolite-Templated Carbons for

A

Hydrogen Storage, Chem. Vap. Deposition, 16 (2010) 322-328.

M

[6] A. Thess, R. Lee, P. Nikolaev, H. Dai, P. Petit, J. Robert, C. Xu, Y.H. Lee, S.G. Kim, A.G. Rinzler,

ED

Crystalline Ropes of Metallic Carbon Nanotubes, Science, 273 (1996) 483-487. [7] R. Li, X. Ren, X. Feng, X. Li, C. Hu, B. Wang, A highly stable metal- and nitrogen-doped

PT

nanocomposite derived from Zn/Ni-ZIF-8 capable of CO2 capture and separation, Chem. Commun.,

CC E

50 (2014) 6894-6897.

[8] P. Pachfule, B.P. Biswal, R. Banerjee, Control of porosity by using isoreticular zeolitic

A

imidazolate frameworks (IRZIFs) as a template for porous carbon synthesis, Chemistry, 18 (2012) 11399-11408. [9] J. Li, Q. Xu, J. Wang, J. Jiao, Z. Zhang, Controlled Synthesis of Monolithic Hierarchical Porous Materials Using Wood as a Template with Assistance of Supercritical Carbon Dioxide, Ind. Eng. Chem. Res., 47 (2008) 7680-7685. [10] A. Ahmadpour, D.D. Do, The preparation of active carbons from coal by chemical and physical

activation, Carbon, 34 (1996) 471-479. [11] M. Yang, L. Guo, G. Hu, X. Hu, J. Chen, S. Shen, W. Dai, M. Fan, Adsorption of CO 2 by Petroleum Coke Nitrogen-Doped Porous Carbons Synthesized by Combining Ammoxidation with KOH Activation, Ind. Eng. Chem. Res., (2016). [12] J. Hu, H. Wang, Q. Gao, H. Guo, Porous carbons prepared by using metal–organic framework

IP T

as the precursor for supercapacitors, Carbon, 48 (2010) 3599-3606. [13] H. Li, M. Eddaoudi, M. O'Keeffe, O.M. Yaghi, Design and synthesis of an exceptionally stable

SC R

and highly porous metal-organic framework, Nature, 402 (1999) págs. 276-279.

[14] B. Liu, H. Shioyama, T. Akita, Q. Xu, Metal-organic framework as a template for porous carbon

U

synthesis, J. Am. Chem. Soc., 130 (2008) 5390-5391.

N

[15] K.L.M. And, ‡, J.T.H. †, Chemical Reduction of Metal−Organic Framework Materials as a

A

Method to Enhance Gas Uptake and Binding, J. Am. Chem. Soc., 129 (2007) 9604-9605.

M

[16] Y.-V.K. Dr., J.M.T. Dr., D.V. Prof., R. Dabirian, L.W.J. Prof., C.A.v.W. Dr., H.S.J.v.d.Z. Dr.,

ED

Charge Transport in Three-Terminal Molecular Junctions Incorporating Sulfur-End-Functionalized Tercyclohexylidene Spacers †, Angew. Chem. Int. Ed, 45 (2006) 2540-2542.

PT

[17] A.J. Amali, J.K. Sun, Q. Xu, From assembled metal-organic framework nanoparticles to

CC E

hierarchically porous carbon for electrochemical energy storage, Chem. Commun., 50 (2014) 1519-1522.

[18] W. Chaikittisilp, M. Hu, H. Wang, H.S. Huang, T. Fujita, K.C. Wu, L.C. Chen, Y. Yamauchi, K.

A

Ariga, Nanoporous carbons through direct carbonization of a zeolitic imidazolate framework for supercapacitor electrodes, Chem. Commun., 48 (2012) 7259-7261. [19] L. Zhang, Z. Su, F. Jiang, L. Yang, J. Qian, Y. Zhou, W. Li, M. Hong, Highly graphitized nitrogen-doped porous carbon nanopolyhedra derived from ZIF-8 nanocrystals as efficient electrocatalysts for oxygen reduction reactions, Nanoscale, 6 (2014) 6590-6602.

[20] P. Zhang, F. Sun, Z. Xiang, Z. Shen, J. Yun, D. Cao, ZIF-derived in situ nitrogen-doped porous carbons as efficient metal-free electrocatalysts for oxygen reduction reaction, Energy Environ. Sci., 7 (2014) 442-450. [21] A. Aijaz, N. Fujiwara, Q. Xu, From metal-organic framework to nitrogen-decorated nanoporous carbons: high CO2 uptake and efficient catalytic oxygen reduction, J. Am. Chem. Soc., 136 (2014)

IP T

6790-6793. [22] P. Zhang, F. Sun, Z. Shen, D. Cao, ZIF-derived porous carbon: a promising supercapacitor

SC R

electrode material, J. Mater. Chem. A, 2 (2014) 12873.

[23] B. Liu, H. Shioyama, H. Jiang, X. Zhang, Q. Xu, Metal–organic framework (MOF) as a

U

template for syntheses of nanoporous carbons as electrode materials for supercapacitor, Carbon, 48

N

(2010) 456-463.

A

[24] J.-K. Sun, Q. Xu, Functional materials derived from open framework templates/precursors:

M

synthesis and applications, Energy Environ. Sci., 7 (2014) 2071.

ED

[25] J. Gonzalez, R.N. Devi, D.P. Tunstall, P.A. Cox, P.A. Wright, Deuterium NMR studies of framework and guest mobility in the metal–organic framework compound MOF-5, Zn4O(O2CC6H4

PT

CO2)3, Microporous Mesoporous Mater., 84 (2005) 97-104.

CC E

[26] G. Srinivas, V. Krungleviciute, Z.-X. Guo, T. Yildirim, Exceptional CO2 capture in a hierarchically porous carbon with simultaneous high surface area and pore volume, Energy Environ. Sci., 7 (2014) 335-342.

A

[27] L. Huang, H. Wang, J. Chen, Z. Wang, J. Sun, D. Zhao, Y. Yan, Synthesis, morphology control, and properties of porous metal–organic coordination polymers, Microporous Mesoporous Mater., 58 (2003) 105-114. [28] M. Sevilla, A.B. Fuertes, CO2 adsorption by activated templated carbons, J. Colloid & Interface Sci., 366 (2012) 147-154.

[29] K. Xia, X. Tian, S. Fei, K. You, Hierarchical porous graphene-based carbons prepared by carbon dioxide activation and their gas adsorption properties, Int. J. Hydrogen Energy, 39 (2014) 11047-11054. [30] L.S. Cheng, Improved Horvath—Kawazoe equations including spherical pore models for calculating micropore size distribution, Chem. Eng. Sci., 49 (1994) 2599-2609.

IP T

[31] F.C. Wu, R.L. Tseng, Preparation of highly porous carbon from fir wood by KOH etching and CO2 gasification for adsorption of dyes and phenols from water, J. Colloid & Interface Sci, 294

SC R

(2006) 21-30.

[32] L. Yang, L. Jing, C. Ming, C. Zheng, Theoretical studies of CO2 adsorption mechanism on

U

linkers of metal–organic frameworks, Fuel, 95 (2012) 521–527.

N

[33] S. Grimme, Semiempirical GGA-type density functional constructed with a long-range

A

dispersion correction, J. Comput. Chem., 27 (2006) 1787-1799.

M

[34] G. Lim, K.B. Lee, H.C. Ham, Effect of N-Containing Functional Groups on CO2 Adsorption of

ED

Carbonaceous Materials: A Density Functional Theory Approach, J. Phys. Chem. C, 120 (2016). [35] X. Wang, Y. Liu, X. Ma, S.K. Das, M. Ostwal, I. Gadwal, K. Yao, X. Dong, Y. Han, I. Pinnau,

CC E

Mater., 29 (2017).

PT

Soluble Polymers with Intrinsic Porosity for Flue Gas Purification and Natural Gas Upgrading, Adv.

[36] Y. Liu, J. Liu, M. Chang, C. Zheng, Effect of Functionalized Linker on CO 2 Binding in Zeolitic Imidazolate Frameworks: Density Functional Theory Study, J. Phys. Chem. C, 116 (2012)

A

16985-16991.

[37] X. Ma, L. Li, S. Wang, M. Lu, H. Li, W. Ma, T.C. Keener, Ammonia-treated porous carbon derived from ZIF-8 for enhanced CO2 adsorption, Appl. Surf. Sci., 369 (2016) 390-397. [38] G. Srinivas, J. Burress, T. Yildirim, Graphene oxide derived carbons (GODCs): synthesis and gas adsorption properties, Energy. Environ. Sci., 5 (2012) 6453.

[39] M. Nandi, K. Okada, A. Dutta, A. Bhaumik, J. Maruyama, D. Derks, H. Uyama, Unprecedented CO2 uptake over highly porous N-doped activated carbon monoliths prepared by physical activation, Chem. Commun., 48 (2012) 10283-10285. [40] G. Srinivas, V. Krungleviciute, Z.-X. Guo, T. Yildirim, Exceptional CO2 capture in a hierarchically porous carbon with simultaneous high surface area and pore volume, Energy Environ.

IP T

Sci., 7 (2014) 335-342. [41] Z. Zhao, Y. Wang, M. Li, R. Yang, High performance N-doped porous activated carbon based

SC R

on chicken feather for supercapacitors and CO2 capture, Rsc Adv., 5 (2015) 34803-34811.

[42] G.P. Hao, W.C. Li, D. Qian, A.H. Lu, Rapid Synthesis of Nitrogen-Doped Porous Carbon

U

Monolith for CO2 Capture, Carbon, 22 (2010) 853-857.

N

[43] L. Chang, J. Li, X. Duan, W. Liu, Porous carbon derived from Metal-organic framework (MOF)

A

for capacitive deionization electrode, Electrochim. Acta, 176 (2015) 956-964.

M

[44] K. László, E. Tombácz, K. Josepovits, Effect of activation on the surface chemistry of carbons

ED

from polymer precursors, Carbon, 39 (2001) 1217-1228. [45] W. Xing, C. Liu, Z. Zhou, J. Zhou, G. Wang, S. Zhuo, Q. Xue, L. Song, Z. Yan,

PT

Oxygen-containing functional group-facilitated CO2 capture by carbide-derived carbons, Nanoscale

CC E

res. lett., 9 (2014) 1-8.

[46] Y. Liu, J. Liu, W. Yao, W. Cen, H. Wang, X. Weng, Z. Wu, The effects of surface acidity on CO2 adsorption over amine functionalized protonated titanate nanotubes, Rsc Adv., 3 (2013)

A

18803-18810.

[47] W. Xing, C. Liu, Z. Zhou, L. Zhang, J. Zhou, S. Zhuo, Z. Yan, H. Gao, G. Wang, S.Z. Qiao, Superior CO2 uptake of N-doped activated carbon through hydrogen-bonding interaction, Energy Environ. Sci, 5 (2012) 7323-7327. [48] I. Langmuir, THE ADSORPTION OF GASES ON PLANE SURFACES OF GLASS, MICA

AND PLATINUM. J. Am. Chem. Soc. [49] T. Depci, Comparison of activated carbon and iron impregnated activated carbon derived from Gölbaşı lignite to remove cyanide from water, Chem. Eng. J., (2012) 467-478. [50] K.S. Lakhi, S.C. Wang, S. Joseph, B.J. Wood, S.S. Aldeyab, G. Lawrence, J.H. Choy, A. Vinu, Cage type mesoporous carbon nitride with large mesopores for CO2 capture, Catal. Today, 243 (2015)

IP T

209-217. [51] K.M. Nelson, S.M. Mahurin, R.T. Mayes, B. Williamson, C.M. Teague, A.J. Binder, L.

SC R

Baggetto, G.M. Veith, S. Dai, Preparation and CO2 adsorption properties of soft-templated mesoporous carbons derived from chestnut tannin precursors, Microporous Mesoporous Mater., 222

U

(2016) 94-103.

N

[52] G. Lim, K.B. Lee, H.C. Ham, Effect of N-Containing Functional Groups on CO2 Adsorption of

A

Carbonaceous Materials: A Density Functional Theory Approach, J. Phys. Chem. C, 120 (2016)

M

8087-8095.

ED

[53] C. Chen, J. Kim, W.S. Ahn, Efficient carbon dioxide capture over a nitrogen-rich carbon having a hierarchical micro-mesopore structure, Fuel, 95 (2012) 360-364.

PT

[54] W. Ren, R. Saito, L. Gao, F. Zheng, Z. Wu, B. Liu, M. Furukawa, J. Zhao, Z. Chen, H.M. Cheng,

CC E

Edge phonon state of mono- and few-layer graphene nanoribbons observed by surface and

A

interference co-enhanced Raman spectroscopy, Physical Review B, 81 (2010) 1718-1720.

ED

IP T

M

A

N

U

SC R

Fig. 1 Representative TEM image for MUCT samples (MUC900).

PT

Fig. 2 Structural evolution of MUCTs samples by XRD and Raman spectroscopy. (a) XRD patterns

A

CC E

and (b) Raman spectra

0.04

0.24

dV/dD(cm3g-1nm-1)

(b)

-1

dV/dD(cm g nm )

0.20

3 -1

0.16 0.12

MUC-600 MUC-700 MUC-800 MUC-900

0.03 0.02 0.01 0.00

0.08

0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2

Pore diameter(nm) 0.04 0.00 0

2

4

6

8

10

12

14

16

18

20

Pore diameter(nm)

A

CC E

PT

ED

M

A

N

U

SC R

IP T

Fig. 2 (a) Nitrogen adsorption-desorption isotherms and (b) pore size distribution calculated from N2 adsorption isothermals for MUCT samples.

3420

1620 2900

1155 1090

MUC900

Intensity

MUC800

MUC700 MUC600

3500

3000

2500

2000

1500

Wavenumber( cm

-1

1000

500

)

IP T

4000

A

CC E

PT

ED

M

A

N

U

SC R

Fig. 4 FTIR spectra of MUCT samples.

O4 O3 O2 O1

N4 N3

a

MC900

MC900 N5

N2 b N1

MC800

Intensity

Intensity

MC800

MC600

538

MC600

536

534

532

530

528

526

SC R

540

IP T

MC700

MC700

412 410 408 406 404 402 400 398 396 394

Binding energy (eV)

Binding energy (eV)

A

CC E

PT

ED

M

A

N

U

Fig. 5 XPS species fitted O1s spectra (a), and N1s spectra (b) for MUCT samples, respectively.

4

3

b

a -1

Capacity (mmol g )

-1

Capacity (mmol g )

3

2

MUC600 MUC700 MUC800 MUC900

1

2

MUC600 MUC700 MUC800 MUC900

1

0

0 0.0

0.0

0.2

0.4

0.6

0.8

0.2

1.0

0.4

0.6

0.8

1.0

Pressure (bar)

Pressure (bar)

A

CC E

PT

ED

M

A

N

U

SC R

IP T

Fig. 6 CO2 adsorption isotherms for MUCT samples at 0 ℃ (a) and 25 ℃ (b), respectively.

30

20 15

MUC600 MUC700 MUC800 MUC900

10 5 0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

IP T

-1

Qst( kJ mol )

25

2.2

CO2 uptake ( mmol g ) -1

A

CC E

PT

ED

M

A

N

U

SC R

Fig. 7 Isosteric heats of adsorption Qst for the adsorption of CO2.

3.6

3.6

a

2

3.2 -1

R =-0.45755

Capacity (mmol g )

2.8 2.4 2.0

2

R =-0.04538

O

1 bar 25 C O 1 bar 0 C O 0.5 bar 25 C linear fit

1.6 1.2

2

R =0.74236 0.8

b

2

R =-0.25566 2.8 2.4 2.0

2

R =-0.35405

O

1 bar 25 C O 1 bar 0 C O 0.5 bar 25 C linear fit

1.6 1.2

2

R =0.76692

0.8 0.4

0.4 400

600

800

1000

1200

1400

1600

0.8

1800

1.0

1.2

1.4

Smes+Smar 3.6

1.8

2.0

2.2

2.4

c

2

3.2 -1

Capacity (mmol g )

R =-0.44899 -1

2.8 2.4 2.0

O

1 bar 25 C O 1 bar 0 C O 0.5 bar 25 C linear fit

2

R =0.45328

1.6 1.2

0.1

0.2

2.0

0.3

0.4

0.5

O

1 bar 25 C O 1 bar 0 C O 0.5 bar 25 C linear fit

2

R =0.42267

1.6 1.2

R =0.78932

0.8

0.4 0.0

2.4

2

R2=0.75283

0.8

d

2

R =-0.46304 2.8

0.6

0.7

0.4 0.05

0.10

0.15

N1, N2 and N3 content (at. %)

0.20

IP T

3.6

3.2

Capacity (mmol g )

1.6

Vmes+Vmar

SC R

-1

Capacity (mmol g )

3.2

0.25

0.30

0.35

U

O3 content (at. %)

N

Fig. 8 CO2 uptake versus (a) mesopore and macropore surface area, (b) mesopore volume, (c) N1,

A

CC E

PT

ED

M

A

N2 and N3 content, (d) O3 content

IP T SC R U N A

A

CC E

PT

ED

M

Fig. 9 Molecular interaction of CO2 with group-functionalized surface

Table 1. Specific surface area and pore size of the four MUCT samples Samples

Smic m2·g-1 554 581 527 601

Smes+Smar m2·g-1 607 449 838 1706

Vtotal mL·g-1 1.31 1.10 1.79 2.54

Vmic mL·g-1 0.25 0.27 0.24 0.26

Vmes+Vmar mL·g-1 1.06 0.83 1.55 2.28

A

CC E

PT

ED

M

A

N

U

SC R

IP T

MUC-600 MUC-700 MUC-800 MUC-900

SBET m2·g-1 1161 1030 1365 2307

Table 2. Deconvolution results of O1s core level spectra Samples

Total O (%)

MUC600 MUC700 MUC800 MUC900

4.31 3.31 2.68 2.02

A% (FWHM) O1 59.9 (2.58) 61.3 (2.02) 61.2 (1.64) 56.4 (1.14)

O2 25.3 (1.09) 22.1 (0.73) 29.8 (0.80) 29.3 (0.59)

O3 8.2 (0.35) 6.3 (0.21) 4.6 (0.12) 4.2 (0.08)

O4 6.6 (0.28) 10.3 (0.34) 4.5 (0.12) 10.2( 0.21)

Table 3. Deconvolution results of N1s core level spectra

N2 19.2 (0.18) 14.5 (0.10) 9.0 (0.04) -

N3 34.7 (0.33) -

N4 24.6 (0.23) 33.3 (0.23) 66.8 (0.27) 100.0 (0.35)

N A M ED PT CC E A

N5 10.7 (0.10) 28.5 (0.20) 15.8 (0.06) -

IP T

N A% (FWHM) N1 10.7 (0.10) 23.8 (0.17) 9.0 (0.04) -

U

MUC600 MUC700 MUC800 MUC900

Total (%) 0.94 0.70 0.41 0.35

SC R

Samples

Table 4. CO2 capture capacities of the porous carbon at different adsorption temperature CO2 capture capacity (mmol g-1) 0 ℃ 1 bar 25 ℃ 1 bar 3.55±0.06 2.44±0.02 3.45±0.02 2.32±0.01 3.34±0.02 2.22±0.01 3.71±0.03 2.31±0.02

25 ℃ 0.15 bar 0.73±0.01 0.64±0.01 0.55±0.02 0.43±0.01

Table 5. Parameters of the Langmuir and Freundlich fitting results. -1

KL/m g 2.0106 1.1616 1.8807 0.9287 1.2374 0.7399 0.4255 0.1859

Freundlich R Kf/mg g-1 0.9938 3.6120 0.9970 2.4365 0.9953 3.5525 0.9977 2.3315 3.3545 0.9958 2.2075 0.9979 0.9994 3.7102 0.9998 2.3133 2

M ED PT CC E A

1/n 0.5273 0.6362 0.5821 0.6789 0.6227 0.7194 0.8093 0.9019

SC R

3

A

MUC600 0 25 MUC700 0 25 MUC800 0 25 MUC900 0 25

Langmuir qmax/mg g-1 288.87 193.69 221.34 207.65 258.98 223.49 538.99 644.82

U

T/℃

N

Samples

IP T

Samples MUC600 MUC700 MUC800 MUC900

R2 0.9990 0.9994 0.9995 0.9996 0.9997 0.9998 0.9999 0.9999