Improving Biogas Separation and Methane Storage with Multilayer

Aug 13, 2012 - Barbara Szczęśniak , Jerzy Choma , Mietek Jaroniec .... K. Christian Kemp , Humaira Seema , Muhammad Saleh , Nhien H. Le , Kandula ...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/est

Improving Biogas Separation and Methane Storage with Multilayer Graphene Nanostructure via Layer Spacing Optimization and Lithium Doping: A Molecular Simulation Investigation Jie-Jie Chen,† Wen-Wei Li,† Xue-Liang Li,*,‡ and Han-Qing Yu*,† †

Department of Chemistry, University of Science and Technology of China, Hefei, 230026, China School of Chemical Engineering, Hefei University of Technology, Hefei, 230009, China



S Supporting Information *

ABSTRACT: Methane is a desirable alternative to conventional fossil fuels, and also a main component of biogas from anaerobic fermentation of organic wastes. However, its relatively lower purity and poor storage by existing adsorbent materials negatively affect its wide application. Thus, efficient, cost-effective, and safe adsorbent materials for methane purification and storage are highly desired. In this study, multilayer graphene nanostructures (MGNs) with optimized structure are investigated as a potential adsorbent for this purpose. The effects of layer distance and Li doping on MGN performance in terms of methane storage and acid gas (H2S and CO2) separation from biogas are examined by molecular simulations. The mechanisms for the interactions between gas molecules and substrates are elucidated by analyzing the binding energy, geometric structures, and charge distribution from the first-principles calculations. The results show that nonhydrocarbons in biogas can be effectively separated using Li-doped MGNs with the optimal layer distance of 0.68 nm, and then the pure methane gas can be stored in MGNs with capacity satisfying the DOE target. This work offers a molecular-level insight into the interactions between gas molecules and MGNs and might provide useful information for development of new materials for methane purification and storage.



effective, and safe systems for CH4 separation and storage.4−7 For example, CH4 is the dominant component of biogas, accounting for 55−65% of the gas volume. Meanwhile, biogas also contains 30−40% carbon dioxide (CO2) and some trace

INTRODUCTION

Methane (CH4) has been recognized as an attractive energy carrier.1 It can be produced from renewable biomass or even wastes, possesses high energy intensity than other hydrocarbon fuels, and is abundantly present in biogas, natural gas, and landfill gas.2,3 In addition, an effective utilization of CH4 would favor a decreased greenhouse gas emission. However, the application scope of CH4 is still limited to date. One major barrier to its vehicular application is the lack of efficient, cost© 2012 American Chemical Society

Received: Revised: Accepted: Published: 10341

May 3, August August August

2012 11, 2012 13, 2012 13, 2012

dx.doi.org/10.1021/es301774g | Environ. Sci. Technol. 2012, 46, 10341−10348

Environmental Science & Technology

Article

components (water, hydrogen sulfide, organic acid, etc).8,9 These nonhydrocarbon fractions have to be separated from biogas before CH4 can be stored and transported for practical application. One viable way to achieve this is through physical adsorption. Thus, adsorbents with a great and reversible gas uptake, high gas selectivity, good chemical and thermal stability, and low cost are highly desired.10,11 In this regard, microporous materials, including porous carbon and silica materials,12−15 metal−organic frameworks (MOFs),16−18 covalent organic frameworks (COFs),19 and porous aromatic frameworks,20 may offer an appealing option. Several microporous materials have been tested as a potential candidate for CH4 purification and storage. Among them, carbon-based materials are considered as an ideal candidate not only for CH4 storage but also for subsequent vehicle applications, attributed to their small weight, high stability, good biocompatibility,21 and environmentally friendly nature. Especially, activated carbon has been widely adopted because of its high adsorption capacity, ease of processing, and controllable pore structure.22,23 For instance, pitch-based activated carbon exhibited a gravimetric adsorption capacity of 150 g CH4/kg C at 298.15 K and up to 1.4 MPa pressure.24 Single-walled carbon nanotubes (SWNTs) composed of cylindrical pores (with 8.2 and 6.8 Å diameter) were found to effectively separate hydrogen−CH4 mixture at room temperatures.13 SWNT arrays with van der Waals (VDW) gap of 0.8 nm were reported to be an excellent candidate for CH4 storage.25 For carbon-like slitshaped pore, Kowalczyk et al.26 investigated the formation of carbon-like slit-shaped pores for hydrogen−CH4 mixture storage through grand canonical Monte Carlo (GCMC) simulation, and found that the optimized slit-like pore size could be obtained at 293 K and a pressure of up to 2 MPa. Chakraborty et al.27 discovered the stable existence of methane hydrates confined in very narrow carbon-like slit-shaped pores, further demonstrating a high capability of such materials for selective CH4 adsorption. Moreover, these studies indicate that the spatial structure of carbon nanomaterials, e.g., slit-pore size, diameter of CNTs, and VDW gap between CNT arrays, may play an important role in CH4 separation and storage. The isosteric heats of adsorption of the components in a gas mixture are crucial for designing gas-separating adsorbents.28 Thus, a comparison between the adsorption thermodynamics of CH4 in carbon adsorbents with those of impurity gases in biogas (CO2 and H2S) indicates that the structure of carbon adsorbents may affect the variety and strength of gas adsorption (Table S1 in Supporting Information). Target gas adsorption could be achieved through adjusting and designing the structure of carbon materials. Recently, another carbon-based material, graphene, is drawing intensive attention. Graphene is a one-atom-thick membrane of carbon atoms packed in a honeycomb lattice with a huge specific surface area (2630 m2/g)29 and great mechanical strength.30 The tunable structure of graphene confers it a high potential to effectively and selectively adsorb energy gases. Moreover, molecular simulations are an effective tool to screen existing and novel structures for a given gas storage and separation task.31 GCMC simulations are able to provide the results that can be compared directly with macroscopic experimental gas adsorption data, and hence offer a computational alternative to laborious experimental work.31 The differences between the results of molecular modeling and experimental measurements can be reduced with a correction model, which approaches the structure of material examined in

the experiments. The comparison results19,32,33 indicate the good agreement between the simulated and measured gas adsorption behaviors. Therefore, computational simulation methods are able to guide experiments by designing the novel materials for gas storage applications. A theoretical calculation34 showed that a multilayer graphene sheet with corrugation structure could potentially reach a gravimetric hydrogen storage capacity of up to 8 wt %. A novel threedimensional (3D) carbon nanostructure35 consisting of graphene layers and CNTs is suggested to be capable of enhancing hydrogen storage proven by a multiscale theoretical investigation. Furthermore, recent studies show that the hydrogen storage capacity of planar graphene monolayer36 and pillared graphene nanostructures35,37 could be significantly promoted by Li doping, due to the strong affinity of the doped Li cation to the hydrogen molecule. Li doping makes it possible for graphene to adsorb functional groups that do not attach to graphene originally. This may open new avenues for investigating the chemistry of the rather unreactive sp2 framework of graphene.38 Furthermore, the first-principles calculations39 indicate that doping Li cation into the COFs can significantly enhance the binding strength between CH4 and the substrate due to the London dispersion and the induced dipole interaction. Thus, 3D carbon nanomaterial based on planar graphene multilayer with appropriate spatial structure and Li doping might be a promising candidate for nonpolar gas storage. Moreover, these nanostructures might also be effective for gas separation. This study aims to assess the feasibility of layer spacing and the effects of Li doping in multilayer graphene sheets for CH4 storage and acid gas separation from biogas model with ternary mixtures at room temperature. The interactions among CH4, CO2, and H2S molecules with raw and Li-doped graphenes are elucidated by the first-principles calculations. Additionally, the gas separation and storage performance of the Li-doped graphene multilayer with different layer spacing are compared with those of raw graphene 3D structure by GCMC simulations. In this way a molecular-level insight into the adsorption phenomena observed macroscopically in porous materials might be offered, and may provide useful information for guiding new adsorbent materials design.



MODEL SYSTEM AND COMPUTATIONAL METHODS Multilayer Graphene Nanostructure Configurations. A supercell of multilayer graphene nanostructures (MGNs) was used in this work. It consisted of five graphene sheets, with the layer distance varing from 0.34 to 2.04 at 0.34-nm increments to obtain six MGN configurations. Table S2 shows the lattice parameters of MGNs, including the cell volume (Vcell) and the graphene layer distance (dGL). For the Li-doped graphene sheet, three different doping sites of Li cation depicted in Figure S1 were investigated to search for the stable configurations. The three sites were (a) the hollow site in the center of the hexagonal ring (H), (b) the top site above the carbon atom (T), and (c) the bridge site above the bond center of the C−C bond (Z). The most favored site for Li doping in graphene sheet with the minimum total energy could be found after full relaxation by the first-principles calculation. In addition, the Li-doped MGN (Li−MGN) was composed of five graphene sheets containing Li cations uniformly distributed at most favored sites. 10342

dx.doi.org/10.1021/es301774g | Environ. Sci. Technol. 2012, 46, 10341−10348

Environmental Science & Technology

Article

Density Functional Theory Calculations. To investigate CH4 storage, high-quality first-principles calculations were performed to obtain the binding energy between CH4 and the graphene layer (GL). Because of the large size of the unit cells in the MGNs, one GL was adopted to reduce the computational time of the first-principles calculations. Firstprinciples density functional theory (DFT) calculations were employed with the generalized gradient approximation using plane-wave basis sets and ultrasoft pseudopotentials,40 as implemented in the CASTEP module41 of Materials Studio. Geometry and electronic nature of the gas molecule adsorbed on the raw and Li-doped graphene sheets were investigated by DFT computation and Hirshfeld charge analysis. The exchange-correlation energy and potential were described selfconsistently using the Perdew, Burke, and Ernzerhof correction.42 The tolerances of the energy, stress, and displacement convergences were 1 × 10−5 eV/atom, 0.05 GPa, and 0.001 Å, respectively. Depending on the unit cell size and shape, the Brillouin zone integration was obtained with variable numbers of k-points generated by the Monkhorst− Pack algorithm.43 GCMC Simulations. The GCMC simulations (constant chemical potential μ, cell volume V, and temperature T) were subsequently implemented to predict the adsorption isotherms of gases in the MGNs and their Li-doped systems using the COMPASS force fields.44 This is the first ab initio force field that enables accurate and simultaneous prediction of gas-phase properties and condensed-phase properties. At each pressure used 1 000 000 GCMC steps in the production stage to attain convergence in each simulation. The electrostatic interactions were evaluated by the Ewald summation45 with an accuracy of 10−3 kcal/mol. The van der Waals interactions were assessed by the atom-based summation using cubic spline truncation with 15.5 Å cutoff distance and 1.0 Å spline width. The GCMC simulation gave the average loading number of adsorbate molecules Nad at specified T, V, and μ. Adsorption selectivities in a ternary mixture of biogas model were predicted to detect whether MGNs and Li−MGNs had the ability to separate the gas mixtures. The isosteric heat Qst, which reflects the strength of forces between adsorbents and gas molecules, could be approximated by28 Q st = RT −

⎛ ∂U ⎞ ⎜ ⎟ ⎝ ∂N ⎠T , V

is discussed in the Supporting Information. The hollow site above the center of the hexagon out of the graphene plane (H site, Figure S1a) with hH = 1.90 Å was found to be the most energetically favorable site for Li doping, with Eb, H = 9.06 eV. Thus, the binding energy of gas molecule and hollow site of the Li-doped graphene layer (Li−GL) can be obtained from the following expression ΔE(gas−Li−GL) = E(Li−GL) + E(single gas) − E(Li−GL with gas)

From the binding energy of the gas molecules in the GL and Li−GL systems as shown in Table 1, it is obvious that the Table 1. Binding Energy (ΔE), the Shortest Distance of Gas Molecule and Adsorbent for GL Systems and Distances As Shown in Figure 2 for Li−GL Systems (D), and the Hirshfeld Charge (CH in e units) of the Gas Molecules system

ΔE (eV)

D (Å)

CH

GL−CH4 GL−CO2 GL−H2S Li−GL−CH4 Li−GL−CO2 Li−GL−H2S

−0.062 3.517 0.016 0.464 2.027 0.923

3.343 3.343 3.343 2.150 1.912 2.427

−0.060 0.030 0.080 0.110 0.180 0.190

binding strength of each gas molecule to the Li-doped substrate is much higher than that to the raw GL, except for the CO2. Figure 1 shows the stable structures of a Li-doped GL with one

(1)

where R and T are the universal gas constant and temperature, and U and N are the total adsorbed energy and number of gas molecules, respectively.



Figure 1. Energy-minimized structures of gas molecules adsorbed on the Li-doped graphene sheet with the view of z-axis direction (above) and y-axis direction (below), including the height of Li cation off the graphene layer, the lattice parameters: a = b = 12.3 Å, c = 6.8 Å, α = β = 90°, and γ = 120°.

RESULTS AND DISCUSSION Binding Energy of Gas Molecules and Graphene Layer. The interactions of one raw GL with the gas molecules (CH4, H2S, and CO2) in biogas were performed by the firstprinciples calculations. The structure of a gas molecule on the raw GL is optimized and the binding energy is defined as

gas molecule adsorbed. The height of the Li cation from the graphene plane (h1, h2, and h3) was a bit larger than that of the initial Li−GL structure (hH = 1.90 Å). The calculated binding energy between a CH4 molecule and the doped Li cation of Li− GL system was 0.464 eV, which is much higher than that with the COFs (about 5.71 kcal/mol)39 and the raw GL systems. This proves that doping Li cation can also enhance the binding of CH4 onto graphene substrate surface. In addition, the distance between Li and the C atom of CH4 (DLi−C) in optimized structure (Figure 1a) was determined to be about

ΔE(gas − GL) = E(GL) + E(single gas) − E(GL with gas)

(3)

(2)

The contribution of the zero-point vibrational energy to the total energy was not considered in this study. For the Li-doped graphene structure, the binding energy (Eb, Table S3) of the Li cation adsorbed on the different sites of GL 10343

dx.doi.org/10.1021/es301774g | Environ. Sci. Technol. 2012, 46, 10341−10348

Environmental Science & Technology

Article

2.150 Å. This indicates that there is no bonding interaction between the doped Li cation and the adsorbed CH4 molecule, because DLi−C is larger than the sum of single-bond covalent radii46,47 (rcov, Table S4) of C and Li. The binding strength of H2S gas molecule to Li−GL was also higher than that to raw GL. However, the stable configuration with CO2 molecule adsorbed on Li shows that the binding energy was about 2.027 eV, which is smaller than that between CO2 and bare graphene. This is because CO2 adsorption on the Li cation might be based on Li−O bonding interaction that partially consumes energy. Therefore, our calculations here demonstrated that the Lidoping method can enhance the binding of gas molecules with the graphene via nonbonding interactions. This makes the sorption of even highly symmetric and nonpolar adsorbates, such as CH4 molecules, become possible. The atoms of gas molecules with a high electronegativity may interact with the doped Li cation by forming chemical bonds between the electrons in the donor and the empty Li 2s orbital. Charge Distribution of the Gas Molecules. The charge distributions of the gas molecules adsorbed on GL and Li−GL from Hirshfeld charge analysis are shown in Figure S2. On the GL, the gas molecules, as free molecules, show a symmetric charge distribution. But, 0.060|e| electrons are transferred from CH4 to the substrate, while 0.030|e| and 0.080|e| are transferred from the graphene sheet to CO2 and H2S, respectively. For the Li−GL substrate, CH4 adsorption on the Li cation perturbs the charge distribution of the CH4 molecule and makes it polarized. This reduces the molecular symmetry, induces dipole moments, and strengthens the adsorption interactions. Additionally, the interaction of one O atom of CO2 with the Li cation leads to a perturbed charge distribution. Meanwhile, the Li doping site moves from the H site to near the Z site (above the C−C bond) in the Li−GL−CO2 system. The electrons of CH4, CO2, and H2S induced by adsorbing at the Li cation are about 0.110| e|, 0.180|e|, and 0.190|e|, respectively, indicating an evident charge transfer from gas molecules to the Li-doped substrate. The Hirshfeld charge analysis confirms that the doped Li cation strengthens the adsorption interactions between gas molecules and substrates. Then, the CH4 storage capacity and acid gas separation are determined for nanostructures constructed by multiple GLs (MGN) or Li−GLs (Li−MGN). The doped Li cations would not simultaneously appear at the hollow sites of the adjacent hexagons in GLs so as to achieve a stable system. CH4 Storage Capacities of MGN and Li−MGN. The CH4 storage capacities of MGN and Li−MGN were evaluated by GCMC simulations. Figure 2 shows the simulated excess adsorption isotherms of CH4 in MGN and Li-MGN with different layer spaces at 298.15 K. In addition, the strength of forces between adsorbent and fluid molecules could be reflected by the isosteric heat Qst of adsorption as a function of pressure. The profiles of Qst vs pressure for MGN and Li− MGN are shown in Figure S3. The calculated results show that the MGNs and Li−MGNs with the layer distance (dGL) of 0.34 nm had no CH4 storage capacity. The nanostructures with dGL of 0.51 nm (0.34 + 0.17) also showed no adsorption of CH4 gas molecules. These results suggest that no gas molecules could be adsorbed at a too narrow layer gap due to the steric effect of the adsorption space. It seems that the layer spacing should be no less than 0.68 nm to allow the adsorption of CH4 molecule. Furthermore, the adsorption amounts of CH4 in MGN and Li− MGN increased rapidly at a low pressure and then became

Figure 2. Comparison of adsorption isotherms of pure CH4 in (a) MGNs, and (b) Li-MGNs with different graphene layer distances in the range of 0.34−2.04 nm at 298.15 K.

almost unchanged at a high pressure at 0.68−2.04 nm. The possible relationships that link the saturation amount with pressure and dGL were derived and are described in the Supporting Information. In addition, the excess CH4 uptake decreased obviously with the broadening dGL at a low pressure, and the disparity became narrow at a high pressure. To improve the vehicular application of CH4, the U.S. Department of Energy (DOE) has set the target of 180 v(STP)/v (standard temperature of 298 K and pressure of 1.01 bar, equivalent volume of CH4 per volume of the adsorbent material) for CH4 storage at 35 bar and ambient temperatures. As presented in Figure 2a, at 35 bar, to maximize CH4 uptakes per volume in raw graphene sheets, the optimal layer spacing needs to be within a range of 0.68−2.04 nm. Apparently, the MGN materials can satisfy the DOE target at lower pressures. The volumes of the adsorbent nanostructures with different dGL are shown in Table S2. The Qst of CH4 in MGN (Figure S3a) was found to decrease with the increasing interlayer spacing, and the decreasing extent of Qst was more significant at a smaller spacing. The values in Figure S3a for MGNs with a small dGL are higher than those for the other carbon adsorbents shown in Table S1, indicating the larger strength of forces between CH4 molecules and GLs. At 35 bar, MGN with dGL of 0.68 nm showed the highest uptake of 310 v(STP)/v and had the highest Qst of 6.5 kcal/mol compared to those with other interlayer distances. However, at 100 bar, the highest uptake reached 352 v(STP)/v with a dGL of 1.02 nm and Qst of 5.1 kcal/mol, while the MGN with 0.68 nm layer spacing had a 10344

dx.doi.org/10.1021/es301774g | Environ. Sci. Technol. 2012, 46, 10341−10348

Environmental Science & Technology

Article

Figure 3. Snapshots and adsorption capacities of CH4 in MGN with a dGL of (a) 0.68 nm and (b) 1.02 nm, and in Li-MGN with the dGL of (c) 0.68 nm and (d) 1.02 nm at 298.15 K and 100 bar, including the three-dimensional view and front view.

for Li−MGN would be necessary. In contrast, for the other adsorbents, Li cations are located on the outer surface for CH4 capture with a high binding energy, thus the storage capacities of MOFs and COFs increased after Li doping. In addition, Li− MGNs exhibit the heats of adsorption in the range of 2.5−6.0 kcal/mol with CH4 loading (Figure S3b), which reduces the total number of adsorbed CH4 molecule compared to the MGN systems. As a consequence, the binding energy between CH4 and adsorbent substrate strengthens after Li doping, but, for multilayer graphene sheets, the effective adsorption space plays a more important role in the gas storage. Adsorption Selectivity for CH4/CO2/H2S Mixtures (Biogas Model). Nonhydrocarbons, CO2, and trace H2S in biogas have to be separated before biogas can be stored and transported for further utilization. The separation capabilities of multilayer graphene sheets are simulated in this work. Figure 4 shows the selectivities of CH4/CO2/H2S adsorption in MGNs and Li−MGNs at 298 K. The ternary gas mixture contained 65% CH4, 30% CO2, and 5% H2S. The selectivity of each adsorbent was reflected by its relative selectivity (S) to CH4, which is defined as8

capacity of only 337 v(STP)/v. This is because that the MGN can store two single layers of CH4 molecules in the gap between two graphene layers when the dGL is increased to 1.02 nm (Figure 3b). Although the number of CH4 molecules increased with the broadening graphene layer gap, the volume of the adsorption substrate also increased and resulted in the reducing unit volume capacity. Moreover, the side-views of one gap snapshots in Figure S4 clearly present no other CH4 molecule between the highlighted ones and the substrate. This indicates that the adsorption mode is single-molecule layer adsorption but not in the same plane. The interaction between CH4 and the graphene sheet became weaker when the MGN adsorbed more than 3 planes of CH4 molecules in one graphene layer gap. Thus, the number of CH4 molecules only increased slightly with the increasing interlayer distance, leading to decreased storage capacity of CH4 per volume. Unlike the other adsorbent materials, such as MOFs,48 COFs,39,49 and pillared graphene,35 the MGNs did not show a significant improvement in CH4 adsorption after Li-doping. As shown in Figure 2b, at 35 bar, the adsorption uptake of CH4 in Li−MGN with 0.68 nm of dGL even decreased by about 17.7% compared to that in MGN. Such a decreased adsorption might be attributed to the fact that the doped Li cations occupy the effective gaps between graphene layers for gas molecules adsorption as shown in Figure 3c, d. Thus, in order to satisfy the DOE target, an interlayer spacing range of 0.68−1.36 nm

Sacid gas/CH4 = 10345

A acid gas BCH4 Bacid gas A CH4

(4)

dx.doi.org/10.1021/es301774g | Environ. Sci. Technol. 2012, 46, 10341−10348

Environmental Science & Technology

Article

increasing pressure, indicating an infinite selectivity of CO2 and H2S to CH4 adsorption. However, the separation capability of the membranes for a small amount of H2S, which is undesirable because of high corrosiveness and toxicity, has rarely been mentioned. For Li−MGN systems, the disparity of CO2 and H2S selectivity remained almost unchanged with an increase in layer spacing. These results suggest that MGN with Li doping and an appropriate interlayer distance could be an excellent material for biogas purification. Therefore, multilayer graphene sheets have two functions, i.e., energy gas storage and acid gas separation. To better achieve the bifunctions, nonhydrocarbons in biogas can be separated with the designed Li−MGNs before storing the main effective gas composition in MGNs with an appropriate layer distance. Furthermore, the effect of the temperature on separation and adsorption capability has been investigated and is shown in Supporting Information. To date, a number of approaches, such as ultrahigh-vacuum conditions,52 electrochemistry,53 and chemical vapor deposition (CVD),54 have been used to synthesize functionalized graphene. These efforts have significantly promoted its potential application. Recently, Wassei et al.54 reported a CVD method for controllable manufacturing of bi- and multilayer graphene. This progress indicates that, with the improvements of manufacturing technology, the MGNs with tunable interlayer spacing and metal doping could be readily prepared for various applications, e.g., CH4 purification and storage. Significance of This Work. Several main influencing factors for CH4 storage and acid gas separation in MGNs and Li−MGNs were identified, including the binding energy between gas molecules and substrates, the optimal layer spacing for CH4 adsorption, and the Li doping strategy. Doping of Li cation was able to enhance the binding of CH4 and H2S molecules with the graphene substrate surface via nonbonding interactions. The CO2 molecule could interact with Li cation through chemical bond attributed to the high electronegativity of O atoms. The results demonstrate that CH4 storage capacity could be enhanced by layer space optimization. These results facilitate a better understanding on the gas adsorption processes by graphene, and provide useful guidance for optimal design of novel graphene derivatives for gas storage and purification.

Figure 4. Selectivity of CO2−CH4 and H2S−CH4 at various layer distances of the MGN (a) and Li−MGN (b) at 298.15 K.

where A and B denote the mole fractions of components in the adsorbed and bulk phases, respectively. Figure S5 shows the adsorption isotherms of the ternary mixture in MGNs and Li−MGNs at 298 K, with the interlayer distance ranging from 0.68 to 2.04 nm. According to the average loading number per cell of gas molecules in substrates, the mole fraction of each component can be calculated and then the values of SCO2/CH4 and SH2S/CH4 can be obtained. Both the MGN and Li−MGN tend to adsorb acid gas molecules from the biogas, even though the amount of CO2 and H2S is obviously less than that of CH4. This may be attributed to relatively high binding energy between acid gas molecules and substrates. The selectivities of these materials are dependent largely on the interlayer spacing and the total pressure of the gases. In this work, the MGNs showed a greater adsorption to H2S−CH4 than to CO2−CH4. As shown in Figure S5, for the MGNs with the dGL of 1.36−2.04 nm, this selectivity was obvious at a total pressure up to 5−10 bar. However, for the Li−MGNs, the priority of acid gas adsorption was observed from the very beginning (1.01 bar) until the end (100 bar). Figure 4a shows that the adsorption selectivity of H2S was much greater than CO2 at a narrow interlayer distance and that the disparity became narrow when dGL was larger than 1.36 nm. The membranes are well-known for biogas separation, which has been described in detail in the Supporting Information. Compared to MGNs, Li−MGNs exhibited a higher selectivity for adsorption of both CO2 and H2S (Figure 4b), and the selectivity of CO2/CH4 in the Li−MGNs with the narrow dGL has exceeded the value of polymeric membranes (∼200)50 and membrane contactor (210).51 Especially for the Li−MGN with a dGL of 0.68 nm, no CH4 adsorption was witnessed despite the



ASSOCIATED CONTENT

* Supporting Information S

Discussion about Li doping sites on graphene sheet, relationships between adsorption amount with pressure and interlayer distance, effect of temperature on the purification and storage capability, thermodynamics of CO2, H2S, and CH4 adsorption on carbon adsorbents, supercell structural parameters of MGNs, covalent radii of Li, C, O, and S, snapshots of CH4 in MGN and Li−MGN with the dGL of 1.36, 1.70, and 2.04 nm, and adsorption isotherms of CH4/CO2/H2S in MGNs and Li− MGNs. This information is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Fax: +86-551-2901450 (X.-L.L); +86-551-3601592 (H.-Q.Y.). E-mail: [email protected] (X.-L.L.); [email protected] (H.Q.Y.). 10346

dx.doi.org/10.1021/es301774g | Environ. Sci. Technol. 2012, 46, 10341−10348

Environmental Science & Technology

Article

Notes

(11) Thornton, A. W.; Nairn, K. M.; Hill, J. M.; Hill, A. J.; Hill, M. R. Metal-organic frameworks impregnated with magnesium-decorated fullerenes for methane and hydrogen storage. J. Am. Chem. Soc. 2009, 131, 10662−10669. (12) Luo, J.; Liu, Y.; Jiang, C.; Chu, W.; Jie, W.; Xie, H. Experimental and modeling study of methane adsorption on activated carbon derived from anthracite. J. Chem. Eng. Data 2011, 56, 4919−4926. (13) Kowalczyk, P.; Brualla, L.; Ż ywociński, A.; Bhatia, S. K. Singlewalled carbon nanotubes: Efficient nanomaterials for separation and on-board vehicle storage of hydrogen and methane mixture at room temperature? J. Phys. Chem. C 2007, 111, 5250−5257. (14) Zhou, L.; Liu, X.; Sun, Y.; Li, J.; Zhou, Y. Methane sorption in ordered mesoporous silica SBA-15 in the presence of water. J. Phys. Chem. B 2005, 109, 22710−22714. (15) Rodríguez-Reinoso, F.; Almansa, C.; Molina-Sabio, M. Contribution to the evaluation of density of methane adsorbed on activated carbon. J. Phys. Chem. B 2005, 109, 20227−20231. (16) Chen, L.; Grajciar, L.; Nachtigall, P.; Düren, T. Accurate prediction of methane adsorption in a metal-organic framework with unsaturated metal sites by direct implementation of an ab initio derived potential energy surface in GCMC simulation. J. Phys. Chem. C 2011, 115, 23074−23080. (17) Guo, Z.; Wu, H.; Srinivas, G.; Zhou, Y.; Xiang, S.; Chen, Z.; Yang, Y.; Zhou, W.; O’Keeffe, M.; Chen, B. A metal-organic framework with optimized open metal sites and pore spaces for high methane storage at room temperature. Angew. Chem., Int. Ed. 2011, 50, 3178−3181. (18) Eddaoudi, M.; Kim, J.; Rosi, N.; Vodak, D.; Wachter, J.; O’Keeffe, M.; Yaghi, O. M. Systematic design of pore size and functionality in isoreticular MOFs and their application in methane storage. Science 2002, 295, 469−472. (19) Mendoza-Cortés, J. L.; Han, S. S.; Furukawa, H.; Yaghi, O. M.; Goddard, W. A. Adsorption mechanism and uptake of methane in covalent organic frameworks: Theory and experiment. J. Phys. Chem. A 2010, 114, 10824−10833. (20) Ben, T.; Pei, C.; Zhang, D.; Xu, J.; Deng, F.; Jing, X.; Qiu, S. Gas storage in porous aromatic frameworks (PAFs). Energy Environ. Sci. 2011, 4, 3991−3999. (21) Nayagam, D. A.; Williams, R. A.; Chen, J.; Magee, K. A.; Irwin, J.; Tan, J.; Innis, P.; Leung, R. T.; Finch, S.; Williams, C. E.; Clark, G. M.; Wallace, G. G. Biocompatibility of immobilized aligned carbon nanotubes. Small 2011, 7, 1035−1042. (22) Wang, Y.; Ercan, C.; Khawajah, A.; Othman, R. Experimental and theoretical study of methane adsorption on granular activated carbons. AIChE J. 2012, 58, 782−788. (23) Sevilla, M.; Foulston, R.; Mokaya, R. Superactivated carbidederived carbons with high hydrogen storage capacity. Energy Environ. Sci. 2010, 3, 223−227. (24) Rahman, K. A.; Loh, W. S.; Yanagi, H.; Chakraborty, A.; Saha, B. B.; Chun, W. G.; Ng, K. C. Experimental adsorption isotherm of methane onto activated carbon at sub- and supercritical temperatures. J. Chem. Eng. Data 2010, 55, 4961−4967. (25) Cao, D.; Zhang, X.; Chen, J.; Wang, W.; Yun, J. Optimization of single-walled carbon nanotube arrays for methane storage at room temperature. J. Phys. Chem. B 2003, 107, 13286−13292. (26) Kowalczyk, P.; Bhatia, S. K. Optimization of slitlike carbon nanopores for storage of hythane fuel at ambient temperatures. J. Phys. Chem. B 2006, 110, 23770−23776. (27) Nath Chakraborty, S.; Gelb, L. D. A monte carlo simulation study of methane clathrate hydrates confined in slit-shaped pores. J. Phys. Chem. B 2012, 116, 2183−2197. (28) Mohr, R.; Rao, M. B. Isosteric heat of adsorption: Theory and experiment. J. Phys. Chem. B 1999, 103, 6539−6546. (29) Rao, C. N.; Sood, A. K.; Subrahmanyam, K. S.; Govindaraj, A. Graphene: The new two-dimensional nanomaterial. Angew. Chem., Int. Ed. 2009, 48, 7752−7777. (30) Radovic, L. R.; Bockrath, B. On the chemical nature of graphene edges: Origin of stability and potential for magnetism in carbon materials. J. Am. Chem. Soc. 2005, 127, 5917−5927.

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank the Fundamental Research Funds for the Central Universities (WK2060190007) for the partial support of this study. The numerical calculations in this paper have been done on the supercomputing system in the Supercomputing Center of University of Science and Technology of China.



NOMENCLATURE MGN multilayer graphene nanostructure dGL graphene layer distance Li−MGN Li-doped multilayer graphene nanostructure GL graphene layer DFT density functional theory GCMC grand canonical Monte Carlo Qst isosteric heat ΔE(gas−GL) binding energy of gas molecule on the raw GL Eb binding energy of the Li cation on GL Li−GL Li-doped graphene layer hH the initial height of the Li cation at the hollow site above the center of the hexagon off the graphene plane h1, h2, and h3 the height of the Li cation off the graphene plane with CH4, CO2, and H2S adsorbed, respectively STP standard temperature and pressure of 298 K and 1.01 bar S relative selectivity to CH4



REFERENCES

(1) Baxter, J.; Bian, Z.; Chen, G.; Danielson, D.; Dresselhaus, M. S.; Fedorov, A. G.; Fisher, T. S.; Jones, C. W.; Maginn, E.; Kortshagen, U.; Manthiram, A.; Nozik, A.; Rolison, D. R.; Sands, T.; Shi, L.; Sholl, D.; Wu, Y. Nanoscale design to enable the revolution in renewable energy. Energy Environ. Sci. 2009, 2, 559−588. (2) Dhingra, R.; Christensen, E. R.; Liu, Y.; Zhong, B.; Wu, C.-F.; Yost, M. G.; Remais, J. V. Greenhouse gas emission reductions from domestic anaerobic digesters linked with sustainable sanitation in rural china. Environ. Sci. Technol. 2011, 45, 2345−2352. (3) Lee, H.-H.; Ahn, S.-H.; Nam, B.-U.; Kim, B.-S.; Lee, G.-W.; Moon, D.; Shin, H. J.; Han, K. W.; Yoon, J.-H. Thermodynamic stability, spectroscopic identification, and gas storage capacity of CO2CH4-N2 mixture gas hydrates: Implications for landfill gas hydrates. Environ. Sci. Technol. 2012, 46, 4184−4190. (4) Saha, D.; Bao, Z.; Jia, F.; Deng, S. Adsorption of CO2, CH4, N2O, and N2 on MOF-5, MOF-177, and Zeolite 5A. Environ. Sci. Technol. 2010, 44, 1820−1826. (5) Mendoza-Cortes, J. L.; Pascal, T. A.; Goddard, W. A. Design of covalent organic frameworks for methane storage. J. Phys. Chem. A 2011, 115, 13852−13857. (6) Düren, T.; Sarkisov, L.; Yaghi, O. M.; Snurr, R. Q. Design of new materials for methane storage. Langmuir 2004, 20, 2683−2689. (7) Morris, R. E.; Wheatley, P. S. Gas storage in nanoporous materials. Angew. Chem., Int. Ed. 2008, 47, 4966−4981. (8) Wang, W.; Peng, X.; Cao, D. Capture of trace sulfur gases from binary mixtures by single-walled carbon nanotube arrays: A molecular simulation study. Environ. Sci. Technol. 2011, 45, 4832−4838. (9) Himeno, S.; Komatsu, T.; Fujita, S. High-pressure adsorption equilibria of methane and carbon dioxide on several activated carbons. J. Chem. Eng. Data 2005, 50, 369−376. (10) Mohanty, P.; Kull, L. D.; Landskron, K. Porous covalent electron-rich organonitridic frameworks as highly selective sorbents for methane and carbon dioxide. Nat. Commun. 2011, 2, 401. 10347

dx.doi.org/10.1021/es301774g | Environ. Sci. Technol. 2012, 46, 10341−10348

Environmental Science & Technology

Article

(31) Xiang, Z.; Cao, D.; Lan, J.; Wang, W.; Broom, D. P. Multiscale simulation and modelling of adsorptive processes for energy gas storage and carbon dioxide capture in porous coordination frameworks. Energy Environ. Sci. 2010, 3, 1469−1487. (32) Yazaydın, A. O. z. r.; Snurr, R. Q.; Park, T.-H.; Koh, K.; Liu, J.; LeVan, M. D.; Benin, A. I.; Jakubczak, P.; Lanuza, M.; Galloway, D. B.; Low, J. J.; Willis, R. R. Screening of metal-organic frameworks for carbon dioxide capture from flue gas using a combined experimental and modeling approach. J. Am. Chem. Soc. 2009, 131, 18198−18199. (33) Shao, X.; Feng, Z.; Xue, R.; Ma, C.; Wang, W.; Peng, X.; Cao, D. Adsorption of CO2, CH4, CO2/N2 and CO2/CH4 in novel activated carbon beads: Preparation, measurements and simulation. AIChE J. 2011, 57, 3042−3051. (34) Tozzini, V.; Pellegrini, V. Reversible hydrogen storage by controlled buckling of graphene layers. J. Phys. Chem. C 2011, 115, 25523−25528. (35) Dimitrakakis, G. K.; Tylianakis, E.; Froudakis, G. E. Pillared graphene: A new 3-D network nanostructure for enhanced hydrogen storage. Nano Lett. 2008, 8, 3166−3170. (36) Cabria, I.; Lopez, M. J.; Alonso, J. A. Enhancement of hydrogen physisorption on graphene and carbon nanotubes by Li doping. J. Chem. Phys. 2005, 123, 204721. (37) Tylianakis, E.; Psofogiannakis, G. M.; Froudakis, G. E. Li-doped pillared graphene oxide: A graphene-based nanostructured material for hydrogen storage. J. Phys. Chem. Lett. 2010, 1, 2459−2464. (38) Denis, P. A. Chemical reactivity of lithium doped monolayer and bilayer graphene. J. Phys. Chem. C 2011, 115, 13392−13398. (39) Lan, J.; Cao, D.; Wang, W. High uptakes of methane in Lidoped 3D covalent organic frameworks. Langmuir 2009, 26, 220−226. (40) Vanderbilt, D. Soft self-consistent pseudopotentials in a generalized eigenvalue formalism. Phys. Rev. B 1990, 41, 7892−7895. (41) Segall, M. D.; Lindan, P. J. D.; Probert, M. J.; Pickard, C. J.; Hasnip, P. J.; Clark, S. J.; Payne, M. C. First-principles simulation: ideas, illustrations and the CASTEP code. J. Phys-Condens. Mat. 2002, 14, 2717−2744. (42) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865−3868. (43) Monkhorst, H. J.; Pack, J. D. Special points for Brillouin-zone integrations. Phys. Rev. B 1976, 13, 5188−5192. (44) Sun, H. COMPASS: An ab initio force-field optimized for condensed-phase applications-Overview with details on alkane and benzene compounds. J. Phys. Chem. B 1998, 102, 7338−7364. (45) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.; Pedersen, L. G. A smooth particle mesh Ewald method. J. Chem. Phys. 1995, 103, 8577−8593. (46) Pyykkö, P.; Atsumi, M. Molecular single-bond covalent radii for elements 1−118. Chem.Eur. J. 2009, 15, 186−197. (47) Cordero, B.; Gomez, V.; Platero-Prats, A. E.; Reves, M.; Echeverria, J.; Cremades, E.; Barragan, F.; Alvarez, S. Covalent radii revisited. Dalton Trans. 2008, 2832−2838. (48) Mulfort, K. L.; Hupp, J. T. Chemical reduction of metal-organic framework materials as a method to enhance gas uptake and binding. J. Am. Chem. Soc. 2007, 129, 9604−9605. (49) Cao, D.; Lan, J.; Wang, W.; Smit, B. Lithium-doped 3D covalent organic frameworks: High-capacity hydrogen storage materials. Angew. Chem., Int. Ed. 2009, 48, 4730−4733. (50) Hussain, A.; Hägg, M.-B. A feasibility study of CO2 capture from flue gas by a facilitated transport membrane. J. Membr. Sci. 2010, 359, 140−148. (51) Shalygin, M. G.; Roizard, D.; Favre, E.; Teplyakov, V. V. CO2 transfer in an aqueous potassium carbonate liquid membrane module with dense polymeric supporting layers: Influence of concentration, circulation flow rate and temperature. J. Membr. Sci. 2008, 318, 317− 326. (52) Hossain, M. Z.; Johns, J. E.; Bevan, K. H.; Karmel, H. J.; Liang, Y. T.; Yoshimoto, S.; Mukai, K.; Koitaya, T.; Yoshinobu, J.; Kawai, M.; Lear, A. M.; Kesmodel, L. L.; Tait, S. L.; Hersam, M. C. Chemically homogeneous and thermally reversible oxidation of epitaxial graphene. Nat. Chem. 2012, 4, 305−309.

(53) Gan, L.; Zhang, D.; Guo, X. Electrochemistry: An efficient way to chemically modify individual monolayers of graphene. Small 2012, 8, 1326−1330. (54) Wassei, J. K.; Mecklenburg, M.; Torres, J. A.; Fowler, J. D.; Regan, B. C.; Kaner, R. B.; Weiller, B. H. Chemical vapor deposition of graphene on copper from methane, ethane and propane: Evidence for bilayer selectivity. Small 2012, 8, 1415−1422.

10348

dx.doi.org/10.1021/es301774g | Environ. Sci. Technol. 2012, 46, 10341−10348