Li-Doped and Nondoped Covalent Organic Borosilicate Framework for

Jan 29, 2010 - reach 4.39 wt % and 25.86 g/L at T ) 298 K and p ) 100 bar, respectively, ... calculation and grand canonical Monte Carlo (GCMC) simula...
0 downloads 0 Views 3MB Size
Download PDF
3108

J. Phys. Chem. C 2010, 114, 3108–3114

Li-Doped and Nondoped Covalent Organic Borosilicate Framework for Hydrogen Storage Jianhui Lan, Dapeng Cao,* and Wenchuan Wang* DiVision of Molecular and Materials Simulation, Key Lab for Nanomaterials, Ministry of Education, College of Chemical Engineering, Beijing UniVersity of Chemical Technology, Beijing 100029, China ReceiVed: NoVember 9, 2009; ReVised Manuscript ReceiVed: January 7, 2010

A multiscale theoretical method, which combines the first-principles calculation and grand canonical Monte Carlo (GCMC) simulation, is used to investigate the adsorption capacities of hydrogen in nondoped and Li-doped covalent organic borosilicate frameworks (COF-202). Our simulations indicate that the total gravimetric and volumetric hydrogen uptakes of COF-202 reach 7.83 wt % and 44.37 g/L at T ) 77 K and p ) 100 bar, respectively. To enhance the hydrogen storage capacity of COF-202, the doping of Li atoms in COF-202 is studied systematically. First, the first-principles calculations are performed to investigate the possible adsorption sites and the quantity of Li atoms doped in COF-202. Our results prove that, for a single Li atom, the top of the phenyl groups in COF-202 is the most favorable adsorption site; for coadsorption of two Li atoms, with one adsorbed at the top site of a phenyl group and the other at its neighboring interstitial site between the phenyl group and the B-O-Si linkage is the most favorable adsorption mode. Our GCMC simulations predict that the total gravimetric and volumetric uptakes of hydrogen in the Li-doped COF-202 reach 4.39 wt % and 25.86 g/L at T ) 298 K and p ) 100 bar, respectively, where the weight percent of Li equals to 7.90 wt %. This suggests that the Li-doped COF-202 is one of the most promising candidates for hydrogen storage at room temperature. 1. Introduction Hydrogen storage in porous materials is an important issue in hydrogen economics. Schlapbach1 proposed that one of the key factors to store hydrogen efficiently is the high BET specific surface area (SSA). Since some metal organic frameworks (MOFs) (for example, MOF-177,2 SBET ) 4746 m2/g; MOF-5,3 SBET ) 3800 m2/g) exhibit extremely high BET SSA, these high BET MOFs were considered to be a good candidate for hydrogen. Definitely, MOF-1772 and MOF-53 also show the maximum hydrogen excess uptakes about 7% at 77 K. However, these materials still exhibit poor performance at room temperatures. Recently, a series of newly synthesized three-dimensional covalent organic frameworks (3D COFs)4 have attracted a lot of attention because of their extremely high surface areas and free volumes as well as low densities. These porous materials have been considered as the most promising candidates for hydrogen storage. In the past two years, the storage of hydrogen in the newly synthesized 3D COFs has been extensively investigated. Garberoglio5 studied the adsorptions of hydrogen and methane in the 3D COFs by using Dreiding and Universal force fields (UFF), and predicted that the 3D COFs display evidently superior hydrogen storage capacities than MOFs. Klontzas et al.6 studied the adsorption of hydrogen in the 3D COFs by their simulation method and reported that the gravimetric uptake of hydrogen in the 3D COFs is two times larger than that of the best MOFs while the volumetric uptake still remains comparable to MOFs. In particular, the gravimetric uptake of hydrogen in COF-108 reaches 21 and 4.5 wt % at p ) 100 bar and T ) 77 and 298 K, respectively. Han et al.7 studied the storage of hydrogen in the 3D COFs by their simulation method and found that the best candidates for hydrogen storage are COF-105 and * To whom correspondence should be addressed. E-mail: (D.C.) caodp@ mail.buct.edu.cn; (W.W.) [email protected].

COF-108, each of which produces a maximum excess H2 uptake of about 10 wt % at 77 K. In our previous works8 the hydrogen storage capacities of the 3D COFs were also studied by using a multiscale theoretical method that combines the first-principles calculation and grand canonical Monte Carlo (GCMC) simulation.9 Our results show that the 3D COFs have high hydrogen storage capacities at cryogenic temperature, for example, COF105 and COF-108 have H2 uptakes of 18.05 and 17.80 wt % at 77 K and 100 bar, respectively, due to their extremely high surface areas and free volumes. These predictions are in good agreement with previous works.4,6,7 The above investigations suggest that the 3D COFs are the most promising candidates for hydrogen storage at present. The Li-doping surface modification technique becomes a novel research focus in hydrogen storage studies recently. Goddard and co-workers10,11 suggested that doping MOFs with electropositive Li is a good strategy to enhance the hydrogen storage performance. The Li-doped MOF was also prepared experimentally by chemical reduction method.11,12 It was found that the Li-doping nearly doubles the hydrogen capacity of the MOF studied. Mavrandonakis et al.12 put forward that positively charged Li cations doped on MOFs can bind with H2 strongly, leading to a significant enhancement of hydrogen adsorption in MOFs. The adsorption behavior of hydrogen in the Li-doped 3D COFs was also studied in our previous work.8 Our predictions reveal that the Li-doping method can enhance the hydrogen storage capacities of the COFs significantly. For example, the gravimetric storage capacity of hydrogen in the Li-doped COFs reaches 6.7 wt % at T ) 298 K and p ) 100 bar, which is in agreement with the predictions from the firstprinciples calculations.13 Most recently, a new porous covalent organic borosilicate framework (named as COF-202) was synthesized by linking organic units with strong covalent bonds found in Pyrex (borosilicate glass, B-O and Si-O) (see Figure 1).13 COF-202

10.1021/jp9106525  2010 American Chemical Society Published on Web 01/29/2010

Li-Doped and Nondoped COF-202 for H2 Storage

J. Phys. Chem. C, Vol. 114, No. 7, 2010 3109

Figure 1. The unit cell of COF-202. (a) [100] direction; [b] [111] direction. White, green, pink, red, yellow, and violet denote H, C, B, O, Si, and Li, respectively.

TABLE 1: Unit Cell Parameters, Mass, Density and Free Volumes of 3D COF-20213 materials

a)b)c (Å)

mass (g/mol)

density (g/cm3)

free volume (%)a

COF-202

30.1051

8578.12

0.52

64.02

a The free volume is the accessible volume of H2 within one unit cell. It is accessible if the potential energy of the interaction between H2 and the solid framework is less than 104 K.

possesses comparably high porosity and thermal stability compared to those previously reported 3D COFs.4 The evaluated BET SSA and pore volume of this material are 2690 m2/g and 1.09 cm3/g, respectively. The advantage of COF-202 suggests that it may also be a good candidate for hydrogen storage. Accordingly, in this work, we intend to use our multiscale theoretical method to investigate the adsorption of hydrogen in COF-202. As the Li-doping method is proved to be an effective strategy to enhance the hydrogen uptake, our focus is therefore placed on adsorption behavior of hydrogen in the Li-doped COF202 at room temperature. 2. Computational Details The structural information of COF-202 derived from experimental data is listed in Table 1.13 A multiscale research strategy was adopted here to predict the adsorption of hydrogen in COF202 and its Li-doped compounds. First of all, the interaction energies between H2 and the host materials were obtained by the first-principles calculations. Then, the force field parameters for the interaction between H2 and the host materials were obtained by fitting the first-principles results to the Morse potential (see eq 2). Finally, the GCMC simulations were performed to evaluate the hydrogen isotherms of COF-202 and its Li-doped compounds by using the fitted force field parameters as input. 2.1. First-Principles Calculations. As shown in Figure 1, COF-202 is mainly composed of six-membered hydrocarbon rings and B-O-Si linkages. Because of the large size of the unit cell of COF-202, three cluster models were adopted to represent its atom types in order to reduce the first-principles computational cost. Figure 2 shows the selected cluster models used to represent COF-202. In Figure 2, the C6H6 cluster is used to represent the hydrocarbon rings of COF-202, the C5H12 cluster is used to represent the sp3 hybridized C atoms, while the B-O-Si linkage is used to represent other atom types such as B, O, and Si. The interaction energies between H2 and the cluster

Figure 2. Cluster models used to represent the atom types in COF202. (a) C6H6; (b) C5H12; (c) H-terminated B-O-Si linkage. White, green, pink, red, and yellow denote H, C, B, O, and Si, respectively.

models were calculated in the framework of the widely used Møller-Plesset (MP2) method with cc-PVTZ basis set. All the first-principles calculations were implemented by Gaussian 03 program package.14 Using the calculated interaction energies, the force field parameters for the interaction between H2 and COF-202 can be obtained (see the details in our previous work8). To study the Li-doping, we first investigated the possible adsorption sites of single and multiple Li atoms in COF-202 systematically. To gain reliable results, a large cluster model was adopted here to represent COF-202 (see the detail in Section 3.2). The geometry optimizations were performed by the widely used B3LYP/6-31G* method, during which the methyl and phenyl terminals bound to the four-coordinated C atoms were frozen to remain the constraints from the 3D crystal lattice. In addition, the binding energies and the Mulliken charge were calculated at the theoretical level of PW91/6-311G** method. The PW91 exchange-correction functional has been widely used in previous works15-17 due to its good performance in energy

3110

J. Phys. Chem. C, Vol. 114, No. 7, 2010

Lan et al.

TABLE 2: van der Waals Force Field Parameters for the Nonbond Interaction between H2 and COF-202 As Well As Its Li Dopant, Derived from the First-Principles Calculations in This Worka parameter atom typesb H_A---H_A H_---H_A C_R---H_A B_2---H_A O_2---H_A C_3---H_A Si3---H_A Li---H_A

c

D (kcal/mol)

re (Å)

γ

0.0182 0.0124 0.1120 0.0328 0.0690 0.0700 0.1021 2.2512

3.5698 3.3001 3.1800 3.4400 3.1503 3.1800 3.5190 2.0543

10.7094 11.0027 10.5000 11.0004 11.0003 12.0062 12.0461 6.9421

a H2 is treated as a diatomic molecule and H_A denotes H in a H2 molecule. b The first two characters correspond to the chemical symbol; an underscore appears in the second column if the symbol has one letter. The third column describes the hybridization or geometry: 1 ) linear, 2 ) trigonal, R ) resonant, 3 ) tetrahedral. c The force field for H_A and H_A is derived from the literature of Han et al.10

TABLE 3: The First-Principles and Morse Force Field Data on the Binding Energies of H2 with the Selected Three Cluster Models Used Here H2 ∆E (kcal/mol)

C 6 H6

C5H12

B-O-Si linkage

first-principles force field

–1.156 –1.155

–0.256 –0.251

–0.512 –0.506

Nexc ) Ntot - F(T, P)Vfree

prediction. The binding energy between the doped Li atoms and COF-202 is defined as

BE ) E(nLi/host) - E(host) - nE(Li)

(1)

Here, the variable of n denotes the number of Li atoms deposited on the host material. By the above analysis, the reasonable amounts of Li atoms doped in COF-202 can be determined theoretically. A series of single point energies for the coadsorption system of H2 and Li-doped COF-202 were subsequently calculated by using the cluster model method at the theoretical level of PW91/6-311G**, where the basis set superposition error correction is included. On the basis of these calculations, the force field parameters for the interaction between H2 and the doped Li atom can be achieved in next section. 2.2. Fitting of Force Fields. To bridge the first-principles calculation and GCMC simulation, the fitting of the force field parameters for the interaction between H2 and COF-202 were carried out by using the following Morse potential

Uij(rij) ) D[x2 - 2x],

( ( ))

x ) exp -

γ rij -1 2 re

results indicate that in COF-202 the borosilicate unit also contributes to the hydrogen storage properties, though its binding energy with H2 is much lower than that of phenyls. In contrast, the B-O-Si linkage in COF-202 increases the framework density greatly, compared to other 3D COFs.4 The force field parameters for the interaction between H2 and the doped Li atom were also fitted to the Morse potential, and the detail is presented in Section 3.2. 2.3. GCMC Simulation. Using the first-principles-based force field parameters as input, the GCMC simulations were performed to obtain the adsorption isotherms of hydrogen in COF-202 and its Li-doped compound. In our GCMC simulation, the temperature, volume and chemical potential were specified in advance. Moreover, the Widom’s test particle insertion method in a NVT ensemble was used to get the relationship between chemical potential and pressure, as described in our previous work.9 To eliminate the boundary effect, the periodic boundary conditions were applied in all three dimensions, and a 2 × 2 × 2 supercell was used for the simulations. The cutoff radius was set as 5 times the collision diameter. For each state point, the GCMC simulation consisted of 1 × 107 steps to guarantee equilibration, and the following 1 × 107 steps were used to sample the desired thermodynamics properties, such as the adsorption amount and isosteric heat of hydrogen. The total amount of H2 molecules adsorbed in per unit cell Ntot was converted to excess adsorption using the following formula

(2)

where rij is the interaction distance in Å. D, γ, and re denote the well depth, the stiffness (force constant), and the equilibrium bond distance, respectively. Here, the H2 molecule was treated as a diatom one. The obtained force field parameters for the interaction between H2 and COF-202 were listed in Table 2, respectively (see details in our previous work8). For comparison, Table 3 lists the calculated binding energies of H2 with the selected three cluster models derived from our first-principles calculations and the fitted Morse force fields, respectively. Our

(3)

where F(T,P) represents the density at the given temperature and pressure, and Vfree is the free volume for adsorption. Vfree was calculated as the volume within one unit cell where the potential energy of interaction of a hydrogen molecule with the solid framework is less than 104 K. Table 1 lists the calculated values of the free volume of COF-202. 2.4. Isosteric Heats. A thermodynamic quantity of interest in adsorption studies is the isosteric heat, which is the released heat for each molecule added to the adsorbed phase. By using the fluctuation theory, the isosteric heat is calculated from the following equation18

qiso )

〈U〉〈N〉 - 〈UN〉 + kBT 〈N2〉 - 〈N〉〈N〉

(4)

where 〈 · · · 〉 denotes the ensemble average, N is the number of particles, and U is the configuration energy of the system. 3. Results and Discussion 3.1. Adsorption Isotherms of Hydrogen in COF-202. Figure 3a,b shows the total and excess gravimeteric and volumetric adsorption isotherms of H2 in COF-202 at T ) 77 K, respectively, where the excess amount is defined as the total adsorption amount minus the amount of hydrogen in the free volume in the bulk phase. For comparison, the experimental adsorption isotherms of hydrogen in MOF-17719 were also inserted. As displayed in Figure 3a, COF-202 is superior to MOF-177 in both total and excess gravimetric H2 capacities at p < 20 bar, while at higher pressures MOF-177 exhibits more excellent hydrogen storage performance than COF-202. This observation can be attrbuted to the differences of pore size and pore volume between COF-202 and MOF-177. Compared to the average pore size of 17 Å in MOF-177, COF-202 has an

Li-Doped and Nondoped COF-202 for H2 Storage

J. Phys. Chem. C, Vol. 114, No. 7, 2010 3111

Figure 3. Computed H2 adsorption isotherms of COF-202 at T ) 77 and 298 K. (a) Total and excess gravimetric isotherms at T ) 77 K, where the experimental H2 isotherms of MOF-177 are also presented for comparison.17 (b) Total and excess volumetric isotherms at T ) 77 K. (c) Total and excess gravimetric isotherms at T ) 298 K. (d) Total and excess volumetric isotherms at T ) 298 K.

smaller average pore size of 11 Å, which leads to more overlap of the potential fields from opposing pore walls and thus presents a stronger affinity toward hydrogen. Therefore, at p < 20 bar, COF-202 shows higher hydrogen storage capacity than MOF177. However, at p > 20 bar, the MOF exceeds COF-202 in hydrogen storage due to its larger pore volume (the pore volumes of COF-202 and MOF-177 are 1.09 and 1.69 cm3/g, respectively). The total gravimetric hydrogen uptakes of COF202 are 6.56 and 7.83 wt % at p ) 20 and 100 bar, respectively. The excess adsorption isotherm of H2 indicates that COF-202 exhibits an optimum H2 gravimetric uptake of 6.05 wt % at 77 K and p ) 30 bar. Impressively, Figure 3b shows that COF202 posscesses superior hydrogen volumetric storage, which is comparable with the previously reported MOFs19 and 3D COFs.8 At p ) 20 and 100 bar, the total volumetric uptakes of hydrogen in COF-202 reach 36.69 and 44.37 g/L, respectively. Moreover, the excess hydrogen volumetric uptake exhibits a maximum of 33.64 g/L at 77 K and p ) 30 bar. Therefore, it can be concluded that COF-202 has the promising hydrogen volumetric capacity at 77 K, which nearly reaches the volumetric target (45 g/L) proposed by the DOE for mobile applications. In the light of the above encouraging results, we further simulated hydrogen adsorption in COF-202 at room temperature within the pressure range up to 100 bar for possibly practical uses. Figure 3c,d shows the total and excess gravimeteric and volumetric adsorption isotherms of H2 in COF-202 at T ) 298 K. It is found that the total H2 gravimetric uptake in COF-202 only reaches 1.52 wt % while the excess amount decreases to 0.70 wt % at T ) 298 K and p ) 100 bar. In Figure 3d, the total and excess volumetric uptakes of H2 in COF-202 are 8.08 and 3.70 g/L at T ) 298 K and p ) 100 bar, respectively. These results prove that the hydrogen storage capacity of COF-202 at room temperature is still away from the 2010 target of 6 wt % proposed by US DOE.

3.2. Adsorption Isotherms of Hydrogen in the Li-Doped COF-202. Unfortunately, our results show that COF-202 is still poor for hydrogen storage at room temperature. Mulfort and Hupp11 have prepared the Li-doped MOF by chemical reduction method experimentally and found that the Li-doping nearly doubles the hydrogen capacity of the MOF studied. Therefore, to enhance the hydrogen storage capacity of COF-202, we further studied the doping of COF-202 with Li atoms and predicted the hydrogen storage performance of the Li-doped COF-202 at room temperature. To determine the reasonable amount of Li-doping, all the possible adsorption sites for single and multiple Li atoms in COF-202 were first investigated. Figure 4 shows the optimized adsorption sites for Li atoms in COF-202, where the distances between the doped Li atoms and the host material were presented accordingly. In addition, the calculated binding energies and the Mulliken charge loaded by the adsorbed Li atoms were also presented for analysis. Figure 4a,b gives the optimized adsorption sites for a single Li atom in COF-202. After optimization, only two adsorption sites for a single Li atom were obtained, (a) the top site on the hydrocarbon ring (TS) and (b) the interstitial site between the hydrocarbon ring and the B-O-Si linkage (IS). Our first-principles calculations reveal that the Li atom doped in COF-202 prefers to be adsorbed at the TS with a binding energy of -23.76 kcal/mol, whereas at the IS the binding energy is just -16.15 kcal/mol. The Mulliken and natural population analysis show that there exists clear charge transfer from the doped Li atoms to the host material. For example, the charge transferred from a single Li atom adsorbed at the TS is about 0.424 |e|, while at the IS is about 0.381 |e|. It can also be found from our calculations that when a Li atom is placed on the hollow site of the B-O-Si linkage, it will migrate to the neighboring IS after geometry optimizations, as shown in Figure 4b. Figure 4c-g presents the

3112

J. Phys. Chem. C, Vol. 114, No. 7, 2010

Lan et al.

Figure 4. Optimized adsorption sites for Li atoms in COF-202. The charges loaded by Li and the binding energies are also presented, respectively. Write, green, red, pink, and violet represent H, C, O, B, and Li atoms, respectively.

optimized coadsorption sites for two Li atoms in COF-202. Figure 4c shows the most favorable coadsorption mode for two Li atoms in COF-202 in which the Li atoms are located at the TS and IS on one side, respectively. In the optimum coadsorption mode, the calculated coadsorption energy is about -51.51 kcal/mol, and the charges loaded by Li atoms are 0.309 |e| and 0.539|e|, respectively. In contrast, when these Li atoms are coadsorbed in other modes shown in Figure 4d-f, the coadsorption energies were calculated to be -50.30, -45.06, and -45.06 kcal/mol, respectively. As is well-known, only positively charged Li atoms can enhance the hydrogen adsorption significantly. The strong affinity of positively charged Li atoms to H2 is due to the formation of a dative bond between the electrons of the H2 σ bond and the empty Li 2s orbital.12,20 In this work, we considered both the adsorption of single and two Li atoms on each phenyl group of COF-202 in the following GCMC simulations (see Figure 1). For convenience, we define these two cases as case 1 and 2, respectively. In case 1 and 2, the weight percents of Li reach 3.95 and 7.90 wt %, and the molar ratios of C/Li are 8.92:1 and 4.46:1, respectively. To obtain the interaction between H2 and the doped Li cations, a relatively large cluster model was constructed to represent COF-202 (see Figure 5). Then the interaction between the Li cation and H2 was derived from high quality first-principles calculation, PW91/6-311g**, with the basis set superposition error correction. Finally, the force field parameters for the Li-H2 interaction were obtained by fitting the interaction energies

Figure 5. The potential energies as a function of the distance between Li and the mass center of H2 derived from our first-principles calculations and force-field fitting, respectively. The color scheme is the same as that in Figure 4.

between H2 and the doped Li atom to the Morse potential (see eq 2). The fitted force field parameters for H2 interaction with Li cations were listed in Table 2. Figure 5 displays the Li-H2 interaction potential energies obtained from both the firstprinciples calculation and the fitted force field. Apparently, the potential energies from our force fields are in good agreement with those from first-principles calculations. According to our results, the binding energy between H2 and a Li cation doped in COF-202 is about 4.61 kcal/mol, which is comparable to the previously reported 4.0 kcal/mol calculated by X3LYP/6311G(d, p) method for the interaction of H2 with a Li cation doped in MOFs.10 For the interaction between H2 and naked Li

Li-Doped and Nondoped COF-202 for H2 Storage

Figure 6. Computed H2 adsorption isotherms of the Li-doped COF202 at T ) 298 K. (a) Total and excess gravimetric isotherms. (b) Total and excess volumetric isotherms. The H2 isotherms of the nondoped COF-202 are also presented for comparison. The values 1 and 2 denote the Li weight percent of 3.95 and 7.90 wt %, respectively.

cations, it is found in previous work21 that the binding energy is about 5.92 kcal/mol obtained by MP4/6-311G(d, p), close to the experimental value of 6.49 kcal/mol.22 By comparison we can see that the binding energy between H2 and Li cation doped in porous materials is lower than that of H2 and naked Li cations. On the basis of these data, it can be found that PW91 functional gives relatively accurate results for the alkali doped systems, although dispersion interactions are not captured by this functional.23 To achieve high-capacity adsorbents, we predicted the adsorption isotherms of hydrogen in the Li-doped COF-202 at T ) 298 K. Figure 6a shows the total and excess H2 gravimetric capacities of Li-modified COF-202, where the capacities of hydrogen in the nondoped COF-202 are also inserted for comparison. As the weight percent of Li equals to 3.95 wt %, the calculated total H2 gravimetric uptake of the Li-doped COF202 reaches 3.32 wt % at T ) 298 K and p ) 100 bar, while this value increases to 4.39 wt % as the quantity of Li dopant is doubled. In contrast, the total H2 gravimetric uptake of the nondoped COF-202 is just 1.52 wt % at T ) 298 K and p ) 100 bar. Obviously, the increment of hydrogen stored in COF202 reaches 2.87 wt % as the quantity of Li dopant equals to 7.90 wt %, compared to the nondoped material. Recently, the hydrogen uptakes of Li-doped COF-102, -103, -105, and -108 are also predicted in our previous work.8 By comparison it is found that the Li-doped COF-202 shows inferior hydrogen storage performance to the previously reported Li-doped COFs. For example, the Li-doped COF-105 shows the hydrogen uptake of about 6.84 wt % at T ) 298 and 100 bar. The above difference is mainly attributed to the higher density (0.52 g/cm3) and lower free volume (64.02 wt %) of COF-202 than COF105. Moreover, the excess H2 uptakes of the Li-doped COF202 are 2.56 and 3.68 wt % for cases 1 and 2, respectively, significantly higher than 0.70 wt % of the nondoped COF-202 at T ) 298 K and p ) 100 bar. It can be observed from the excess H2 uptakes that, each Li atom can adsorb about two H2 molecules at T ) 298 K and p ) 100 bar at the low Li-doping rate. However, at the high Li-doping rate, the average number of H2 adsorbed by per Li decreases to 1.6 due to the steric effect, which is originated from the close distribution of Li atoms on each phenyl group. A comparison of the volumetric capacities between the Li-doped and nondoped COF-202 is shown in Figure 6b. Our results indicate that the volumetric storage capacities of H2 in the Li-doped COF-202 are also significantly enhanced compared to the nondoped COF-202. For example, the total volumetric uptakes of the Li-doped COF-202 are 18.63 and 25.86 g/L for cases 1 and 2 at T ) 298 K and p ) 100 bar

J. Phys. Chem. C, Vol. 114, No. 7, 2010 3113

Figure 7. Computed isosteric heats of H2 adsorption in the nondoped and Li-doped COF-202 at T ) 298 K. The values 1 and 2 denote the Li weight percent of 3.95 and 7.90 wt %, respectively. The data for COF-202 are also presented for comparison.

(8.08 g/L for the nondoped COF-202), respectively. Besides, the excess H2 volumetric uptakes of the Li-doped COF-202 reaches 14.25 and 21.48 g/L at 298 K and p ) 100 bar for the two Li-doping rates studied, much higher than 2.84 g/L of the nondoped COF-202. On the basis of the above analysis, it is concluded that the Li-doping method does enhance the H2 capacity of COF-202 significantly both in gravimetric and volumetric units. The Li-doped COF-202 is one of the promising candidates for hydrogen storage at room temperature. The isosteric heat of H2 adsorption in COF-202 was also calculated and presented in Figure 7. It is found that the Lidoping in COF-202 increases the isosteric heat of H2 evidently due to the strong affinity of the doped Li cation to H2. For the nondoped COF-202, the calculated maximum isosteric heat of H2 is about 1.47 kcal/mol at T ) 298 K. However, for the Lidoped COF-202, the calculated maximum isosteric heats of H2 are approximately 5.40 and 7.24 kcal/mol at T ) 298 K for the Li-doping rates of 3.95 and 7.90 wt %, respectively, and these value are comparable to that in the Li-doped 3D COFs. The isosteric heat of H2 in the Li-doped COF-202 is approximately within the range between physisorption and chemisorption, so that the hydrogen stored in the host material could be released reversibly at room temperature without the need of higher temperatures. 4. Conclusions A multiscale theoretical method has been used to predict the adsorption of hydrogen in a recently prepared covalent organic framework material, COF-202. In this multiscale theoretical method, the first-principles calculations were adopted to evaluate the interaction between H2 and COF-202. By fitting the firstprinciples calculation results to a Morse potential, we obtained the force field parameters for the interaction between H2 and COF-202. The GCMC simulations were then performed to predict the hydrogen adsorption isotherms of COF-202 by using the fitted force field parameters as input. Our calculations show that the recently prepared COF-202 exhibits a superior hydrogen storage capability at T ) 77 K, especially in volumetric capacity. At 77 K and p ) 100 bar, the total hydrogen volumetric uptake of COF-202 reaches 44.37 g/L, indicating that COF-202 is very promising as a potential hydrogen adsorbent. To meet the practical application in hydrogen storage, we suggest using the Li-doping method to modify the hydrogen storage performance of microporous materials. As the quantity of Li doped in COF-202 equals to 7.90 wt %, the total gravimetric and volumetric uptakes of hydrogen of the Li-doped COF-202 reach 4.39 wt % and 25.86

3114

J. Phys. Chem. C, Vol. 114, No. 7, 2010

g/L at 298 K and p ) 100 bar, respectively, showing significant increase compared to those for the nondoped COF-202. To the best of our knowledge, the Li-doped COF-202 material is one of promising candidates for hydrogen storage at room temperature nowadays. Acknowledgment. This work is supported by NSF of China (20776005, 20736002), National Basic Research Program of China (2007CB209706), Beijing Novel Program (2006B17), NCET Program (NCET-06-0095) from the MOE of China, Chemical Grid Program, Chinese Universities Scientific Fund (No. ZD0901) and Excellent Talent of BUCT. References and Notes (1) Schlapbach, L.; Zuttel, A. Nature 2001, 414, 353. (2) Eddaoudi, M.; Kim, J.; Rosi, N. L.; Vodak, D. T.; Wachter, J.; O’Keeffe, M.; Yaghi, O. M. Science 2002, 295, 469. (3) Kaye, S. S.; Dailly, A.; Yaghi, O. M.; Long, J. R. J. Am. Chem. Soc. 2007, 129, 14176. (4) El-Kaderi, H. M.; Hunt, J. R.; Mendoza-Cortes, J. L.; Cote, A. P.; Taylor, R. E.; O’Keeffe, M.; Yaghi, O. M. Science 2007, 316, 268. (5) Garberoglio, G. Langmuir 2007, 23, 12154. (6) Klontzas, E.; Tylianakis, E.; Froudakis, G. E. J. Phys. Chem. C 2008, 112, 9095. (7) Han, S. S.; Furukawa, H.; Yaghi, O. M.; Goddard, W. A., III J. Am. Chem. Soc. 2008, 130, 11580. (8) Cao, D.; Lan, J.; Wang, W.; Smit, B. Angew. Chem., Int. Ed. 2009, 48, 4730. (9) (a) Lan, J.; Cheng, D.; Cao, D.; Wang, W. J. Phys. Chem. C 2008, 112, 5598. (b) Xiang, Z.; Lan, J.; Cao, D.; Shao, X.; Wang, W.; Broom, D. P. J. Phys. Chem. C 2009, 113, 15106. (c) Lan, J.; Cao, D. P.; Wang, W. C. ACS Nano 2009, 3, 3294. (10) Han, S. S.; Goddard, W. A., III J. Am. Chem. Soc. 2007, 129, 8422. (11) Mulfort, K. L.; Hupp, J. T. J. Am. Chem. Soc. 2007, 129, 9604.

Lan et al. (12) Mavrandonakis, A.; Tylianakis, E.; Stubos, A. K.; Froudakis, G. E. J. Phys. Chem. C 2008, 112, 7290. (13) Hunt, J. R.; Doonan, C. J.; LeVangie, J. D.; Cote, A. P.; Yaghi, O. M. J. Am. Chem. Soc. 2008, 130, 11872. (14) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision B.02; Gaussian, Inc.: Wallingford, CT. (15) Alapati, S. V.; Johnson, J. K.; Sholl, D. S. J. Phys. Chem. C 2008, 112, 5258. (16) Zhou, Z.; Zhao, J.; Chen, Z.; Gao, X.; Yan, T.; Wen, B.; Schleyer, P. v. R. J. Phys. Chem. B 2006, 110, 13363. (17) Mpourmpakis, G.; Froudakis, G. E.; Lithoxoos, G. P.; Samios, J. Nano Lett. 2006, 6, 1581. (18) Nicholson, D.; Parsonage, N. Computer simulation and statistical mechanics of adsorption; Academic Press: London, 1982. (19) Wong-Foy, A. G.; Matzger, A. J.; Yaghi, O. M. J. Am. Chem. Soc. 2006, 128, 3494. (20) Vitillo, J. G.; Damin, A.; Zecchina, A.; Ricchiardi, G. J. Chem. Phys. 2005, 122, 114311. (21) Barbatti, M.; Jalbert, G.; Nascimento, M. A. C. J. Chem. Phys. 2001, 114, 2213. (22) Wu, C. H. J. Chem. Phys. 1979, 71, 783. (23) Er, S.; de Wijs, G. A.; Brocks, G. J. Phys. Chem. C 2009, 113 (43), 18962.

JP9106525