Ind. Eng. Chem. Res. 2009, 48, 3425–3431
3425
Adsorption and Separation of Noble Gases by IRMOF-1: Grand Canonical Monte Carlo Simulations Jeffery A. Greathouse,*,† Tiffany L. Kinnibrugh,†,‡ and Mark D. Allendorf§ Geochemistry Department, Sandia National Laboratories, Albuquerque, New Mexico 87185-0754, and Microfluidics Department, Sandia National Laboratories, LiVermore, California 94551
The gas storage capacity of metal-organic frameworks (MOFs) is well-known and has been investigated using both experimental and computational methods. Previous Monte Carlo computer simulations of gas adsorption by MOFs have made several questionable approximations regarding framework-framework and framework-adsorbate interactions: potential parameters from general force fields have been used, and framework atoms were fixed at their crystallographic coordinates (rigid framework). We assess the validity of these approximations with grand canonical Monte Carlo simulations for a well-known Zn-based MOF (IRMOF-1), using potential parameters specifically derived for IRMOF-1. Our approach is validated by comparison with experimental results for hydrogen and xenon adsorption at room temperature. The effects of framework flexibility on the adsorption of noble gases and hydrogen are described, as well as the selectivity of IRMOF-1 for xenon versus other noble gases. At both low temperature (78 K) and room temperature, little difference in gas adsorption is seen between the rigid and flexible force fields. Experimental trends of noble gas inflation curves are also matched by the simulation results. Additionally, we show that IRMOF-1 selectively adsorbs Xe atoms in Xe/Kr and Xe/Ar mixtures, and this preference correlates with the trend in van der Waals parameters for the adsorbate atoms. Introduction Metal-organic frameworks (MOFs) are now well-established nanoporous materials and are attracting considerable attention with respect to storage of gases such as H2, CH4, and CO2.1-6 The science of MOF-based chemical separations is less well developed, but their tailorable pore sizes and pore chemistry make them attractive for both membrane- and molecular-sieve separations.4,7-10 Several examples of MOFs demonstrating molecular separation potential have been reported, including separation or preferential adsorption of linear vs branched hydrocarbons,11,12 xylenes,13 benzene,14 alcohols,15,16 CO2,17-19 and H2.18-21 The purification of rare gases is also a problem of industrial interest, since there are both medical and lighting applications of xenon. Mueller et al. measured the volumespecific uptake of rare gases by IRMOF-1, showing that this MOF exhibits preferential adsorption of xenon over the lighter rare gases.22 In addition, they found that Cu-BTC (also known as HKUST-1) can be used to separate xenon from krypton, with a calculated capacity of more than 60 wt %, exceeding highsurface-area carbons. Although chemical intuition built upon expanded knowledge of the geometric aspects of MOF coordination chemistry developed over the past several years has led to an explosion of new MOFs, it is still largely unclear how MOF structure and pore chemistry influence the resulting gas adsorption properties. Theory is now beginning to play a constructive role in guiding and accelerating synthetic efforts. Several recent investigations focused on identifying design criteria for gas storage,1,23 and separations24,25 using MOFs have been published. In particular, Keskin and Sholl (KS) showed that mixture * To whom correspondence should be addressed. E-mail:
[email protected]. † Geochemistry Department, Sandia National Laboratories. ‡ Present Address: Department of Chemistry, Texas A&M University, P. O. Box 30012, College Station, Texas 77842. § Microfluidics Department, Sandia National Laboratories.
effects play a very important role in determining the separation efficiency of MOF-5 but nevertheless were able to combine single-component adsorption isotherms with a thermodynamic theory to predict membrane separation ability.24 This is not always true, however. Karra and Walton (KW) modeled gas sorption in the MOF Cu-BTC25 and predicted from grand canonical Monte Carlo (GCMC) calculations that mixture selectivities are consistent with pure-component gas isotherms. KS obtained good agreement with experimental singlecomponent isotherms using a rigid force field. Computational methods employing force fields are powerful tools for assessing the importance of such effects. In particular, GCMC simulations have been used extensively to predict the gas adsorption properties of MOFs, including those containing aluminum,26,27 zinc,1,23,28-37 copper,38-43 cobalt,44 and vanadium.27 Standard potentials, such as the DREIDING,45 OPLS,46 and UFF47 were used in these atomistic simulations thus far, and in all cases, the positions of the MOF atoms were fixed. Numerous investigations make clear, however, that MOFs are structurally flexible and can exhibit substantial changes in unit cell parameters upon adsorption or desorption of guest molecules. Many examples of structural changes in MOFs upon adsorption of guest species are known,48,49 and flexible structures have been designed and synthesized to enhance selectivity for a particular guest.14,21,50 Recently, Allendorf et al.51 showed that detectable stresses are induced at an interface between a Cu-BTC film and a microcantilever as a result of the adsorption of CO2 and alcohols. This suggests that gas adsorption can induce small structural changes in MOFs, supporting the notion that a flexible force field approach could be useful. Fixed-atom force fields obviously cannot capture these effects. A recent review article points to the need for additional simulations on adsorbate mobility to assess the effects of framework flexibility, which has implications for the ability of MOFs to separate one gas from another.52 To assess the importance of structural flexibility in the modeling of MOF adsorption properties, as well as the potential
10.1021/ie801294n CCC: $40.75 2009 American Chemical Society Published on Web 03/09/2009
3426 Ind. Eng. Chem. Res., Vol. 48, No. 7, 2009
for chemical reactions involving the framework, we developed a nonbonded force field for isoreticular MOFs (IRMOFs).53,54 In addition to correctly predicting the structure of IRMOF-1, molecular dynamics (MD) simulations of the interaction of adsorbed water in IRMOF-1 with this force field predict framework collapse above a critical water concentration, in qualitative agreement with experiment.55,56 We recently reported a detailed validation of the flexible force field concept, demonstrating its ability to accurately predict important IRMOF properties such as vibrational frequencies, the dependence of lattice parameter on temperature and pressure, benzene diffusion, and elastic properties.54 Dubbeldam et al.57 also used a hybrid force field for simulations of several Zn-based IRMOFs, following our approach to allow motion of framework atoms. These authors used structural comparisons and adsorption isotherms for CO2 and CH4 in the parameter fitting process, resulting in a force field that correctly predicts measured adsorption isotherms for these gases.57 Alternative approaches employ a “flexible” but still fully bonded force field.58-60 Schmid et al. parametrized the MM3 force field to account for interactions with the Zn4O clusters in IRMOF-1.58,59 This model successfully predicts the benzene self-diffusion constant in IRMOF-1, yielding a value within ∼30% of that measured by NMR. Huang et al.60 used a similar force field to calculate the phonon thermal conductivity and vibrational power spectra of this MOF using molecular dynamics. Although successful in these respects, these fully bonded force fields cannot be used to probe framework reactivity with respect to either adsorbates or solvent environments. In this paper, we assess the validity of rigid framework approximations to perform grand canonical Monte Carlo simulations of the adsorption of weakly interacting gases by IRMOF-1. The adsorbate gases include the noble gases He, Ar, Kr, and Xe, as well as H2. At both low temperature (78 K) and room temperature, little difference in gas adsorption is seen between the rigid and flexible force fields. The results qualitatively predict the experimental trends of noble gas inflation curves as well as noble gas separation based on gas atomic properties. Methods A hybrid force field is used for IRMOF-1, which includes nonbonded interactions between Zn atoms, inorganic O atoms, and benzene dicarboxylate (BDC) linkers. The original nonbonded parameters54 were derived from the consistent valence force field (CVFF)61 and parameters optimized for the mineral zincite (ZnO), but they were recently modified by Dubbeldam et al. to reproduce CO2 and CH4 adsorption isotherms.57 The initial atom positions for the framework are taken from the experimental crystal structure at 213 K.62 We note that the reported structure contained guest solvent molecules, resulting in a lattice parameter (25.669 Å) that is slightly smaller than that found for the evacuated framework (25.885 Å) at 169 K.62 Unlike the implementation we used for molecular dynamics simulations,54 in which the linker molecules are fully flexible (bond, angle, torsion), here we use a semiflexible approach where the BDC linkers are treated as rigid structures. Test simulations in which linker flexibility was included by a torsional pivot move gave results similar to the semiflexible force field. The nonbonded potential energy Eij between atoms i and j separated by a distance r is given by
Eij )
[( ) ( ) ]
qiqj σij + 4εij r r
12
-
σij r
6
(1)
where geometric mixing rules are used to calculate the interatomic interactions (σij ) (σiσj)1/2 and εij ) (εiεj)1/2 represent van der Waals radii and energy well depths, respectively). Pairwise electrostatic interactions are calculated using atomic charges, q, based on quantum chemical calculations on representative clusters.57 The noble gases and methane molecules are described as spherically symmetric, uncharged Lennard-Jones particles, and the intermolecular interactions are modeled only with van der Waals potential parameters. A rigid three-site model, with the H-H bond length fixed at 0.74 Å, is used for H2.63 A LennardJones core is placed at the center of mass of the molecule with a point charge magnitude -2q and point charge magnitude of q at the positions of the two protons. All interatomic potential parameters are given in Table 1. The grand canonical Monte Carlo (GCMC) code Towhee67 was used at fixed temperature T, volume V, and adsorbate chemical potential, µ, to simulate gas adsorption. A Monte Carlo move consists of one of the following: translation of a molecule, rotation of a molecule, insertion of a new guest molecule, or deletion of an existing guest molecule. Rotation moves are not necessary for spherical particles. For simulations at room temperature, a total of 8 million MC moves are used for the simulations, and the last 5 million moves are used for averaging and analysis. A total of 28 million MC moves are used for the simulations of Ar adsorption at 78 K, and the last 10 million moves are used for averaging and analysis. Results are obtained by a block averaging technique, with each block representing 1/20th of the production period. Insertion and deletion statistics are monitored to ensure microscopic reversibility, and the standard deviations of the adsorbate densities are within (3.0% of the average. The simulation box consists of one unit cell (space group, Fm3m; lattice parameter, 25.699 Å), with 424 atoms. Larger box dimensions (2 × 2 × 2 unit cell repeats) have previously been shown to give similar results as a single unit cell.57 The real-space cutoff for the nonbonded potential energy is 11.0 Å, and the long-range electrostatic interactions are calculated by the Ewald summation68 with a precision of 1.0 × 10-4. The Towhee code requires adsorbate chemical potentials rather than feed gas fugacity as input, so we performed “empty box” GCMC simulations (i.e., a box without the MOF unit cell) to establish appropriate chemical potentials for the target pressures.40 Rather than applying an unrelated equation of state to correlate these two quantities, the same force field parameters used in the adsorption simulations are used to accurately determine the feed gas pressure. The standard deviations of the pressure and densities for the empty box simulations are approximately 1%. For each value of µ, the simulated pressure was compared to the corresponding ideal gas pressure.68 The two pressures did not deviate until at least 10 bar, suggesting reasonable accuracy at low pressures for spherical and polyatomic guest particles. Excess adsorption, Nex, is described as the amount of adsorbate in excess of bulk fluid at the system temperature and pressure in the available void volume. Simulations measure the absolute amount adsorbed, Nabs; therefore, excess adsorption is calculated by69 Nex ) Nabs - VgFg
(2)
where Fg is the bulk gas density determined by the empty box GCMC simulations mentioned previously. The void volume,
Ind. Eng. Chem. Res., Vol. 48, No. 7, 2009 3427 Table 1. Force Field Parameters for Nonbonded Interactions According to Equation 1
IRMOF-1
speciesa
description
Zn Ocent Ocarb Ccarb Cphenyl
zinc inorganic oxygen carboxylate oxygen carboxylate carbon phenyl carbon bonded to Ccarb phenyl carbon bonded to H hydrogen center-of-mass hydrogen
Cphenyl adsorbate
H H2
ε/kB
CH4 Xe Kr Ar He
b
(K) σ (Å)
q (e)
0.42 700.00 70.50 47.00 47.86
2.700 2.980 3.110 3.740 3.470
1.275 -1.500 -0.600 0.475 0.125
47.86
3.470
-0.150
7.65 36.70 0.00 148.00 216.85 163.99 119.80 10.90
2.850 2.958 0.000 3.730 4.100 3.827 3.400 2.640
0.150 -0.936 0.468 0.000 0.000 0.000 0.000 0.000
a
IRMOF-1 and adsorbate parameters were taken from the literature: IRMOF-1,57 H2,63 CH4,64 Xe,65 Kr,65 Ar,66 and He.66 b kB represents the Boltzmann constant.
Figure 1. H2 adsorption isotherm for IRMOF-1 at 298 K. Results from this work are compared with previous simulation results71 using a rigid framework with UFF parameters as well as experiment.74
Vg, is calculated from GCMC simulations of He adsorption and found to be 0.75 of the unit cell volume for IRMOF-1. Upon adsorption or desorption, the enthalpy change of the adsorbate from the bulk phase to the adsorbed phase is described by the isosteric heat of adsorption, Qst. For systems with low guest loading and feed gas pressure, Qst can be calculated from the thermal energy RT and the average potential energy per molecule, 〈V 〉, using the ideal gas approximation29
Table 2. Comparison of Excess H2 Adsorption (mg · g-1) in IRMOF-1 at 298 K
Qst ) RT - 〈V 〉
(3)
Investigation of noble gas selectivity by IRMOF-1 was carried out by GCMC simulations for Xe/Ar and Xe/Kr mixtures at 300 K at total pressures of 1 and 10 bar. The Xe selectivity, SXe, was calculated from the ratios of gas mole fractions in the feed gas phase (y) and the adsorbed phase (x). For Xe/Kr mixtures, SXe becomes70 sXe )
xXe yKr xKr yXe
(4)
Results Several simulation studies have shown that framework flexibility effects can be essential in predicting IRMOF properties such as framework stability,53 guest diffusion,54,58 and mechanical properties.54 Our primary aim in this work is to determine the effect of such flexibility on adsorption properties, and we use the H2 adsorption results to further validate the force field parameters. Figure 1 shows that very similar H2 adsorption isotherms at 298 K result when either the rigid framework or semiflexible framework approaches are used. We present results at room temperature only, so quantum effects between H2 molecules can be ignored. The three-site model for H2 includes atomic charges to enhance interactions with framework atoms. At very high feed gas pressure, framework flexibility results in a slight decrease in loading. Also in Figure 1, our results show a slight underprediction compared to the simulations of Garberoglio et al.,71 who used UFF parameters for framework atoms. In that work, H2 was modeled using a charge-neutral single van der Waals site,72 which showed better agreement with experiment than the three-site H2 model73 at 78 K. Excess H2 loadings are compared with previous simulation and experimental results in Figure 1 and Table 2. Our results are centered between the experimental results of Panella et al.74 and Dailly et al.,75 while the simulation results of Garberoglio
simulation P (bar)
this work
scaled by 58/75
20 35 50 65
1.1 1.9 2.5 3.2
0.85 1.6 1.9 2.5
experiment a
ref 71
ref 74
ref 75
1.4 2.3 3.2 3.9
1.0 1.6 2.2 2.8
1.4 2.5 3.8 4.5
a Simulation results from this work were scaled by the ratio 58/75 to compensate for defects in the experimental samples, as explained in ref. 76.
et al.71 are in better agreement with the results of Dailly et al.75 Unfortunately, the difference in experimental procedures, combined with the well-known problem of consistent synthesis of high surface area IRMOF-1 crystals for adsorption experiments,76,77 prevents us from drawing definitive conclusions concerning the accuracy of the simulation results. Two recent MOF simulation papers have applied a scaling factor to compensate for the difference in accessible volume.78,79 The experimental void volume was estimated to be between 55 and 61%,62 compared to 75% from our simulations. The resulting scaled loadings for H2 are also given in Table 2, but the agreement is not necessarily improved. This scaling factor appears to be more important for liquid- rather than gas-phase guests, because in the former state the framework is in constant contact with guest molecules. Any defects from the synthesis would be amplified in that case. The isosteric heat of adsorption at 298 K was found to be 4.5 kJ · mol-1, which is comparable to other GCMC simulation values reported at both 77 and 298 K.30,71 Good agreement is also found with the H2 binding energy to IRMOF-1 obtained from quantum calculations (4.16 kJ · mol-1).80 The agreement is also excellent with experimental values at 7874 and 298 K75 (both 3.8 kJ · mol-1). The remainder of our results concern the adsorption capacity and separation capabilities of IRMOF-1 with respect to noble gases. Figure 2 compares the effect of framework flexibility on the simulated Ar isotherms at 78 and 298 K. For Ar adsorption at both temperatures, both framework approximations result in an Ar loading of approximately 2125 mg · g-1 which is significantly higher than the experimental value of ≈1500 mg · g-1.62 As has been pointed out in several papers, however, defects from the synthesis procedures often result in an underestimation of the accessible volume,78,79 so the measured excess adsorption would be too low. In fact, after applying the
3428 Ind. Eng. Chem. Res., Vol. 48, No. 7, 2009
Figure 3. Xe adsorption isotherms for IRMOF-1 at 292 K, comparing simulation and experiment.82
Figure 2. Argon adsorption isotherms for IRMOF-1 at 78 K (top) and 298 K (bottom).
58/75 scaling factor from Table 2, the maximum loading from the simulations is 1640 mg · g-1, in much better agreement with experiment. The slight difference in flexibility effects at low loading could be due to the type of intermolecular interactions between guest and framework. In the case of H2, electrostatic interactions are included, and flexibility results in a slight decrease in loading. For weakly interacting guests such as Ar, where there are no electrostatic interactions with the framework, flexibility results in a slight increase in loading. At high loading, in which liquidlike conditions exist in the pore, the decrease in adsorption accompanied by framework flexibility is probably due to packing effects in a more disordered framework. Practical applications of noble gas adsorption and separation by IRMOFs require results near room temperature. Although we have seen little difference between rigid and flexible frameworks for small molecule adsorption, the computational cost of including framework moves is small. We therefore include framework moves for the remainder of our results. In Figure 3 we compare our simulated Xe adsorption isotherm for IRMOF-1 with experimental data collected at 292 K. There is good agreement at low pressure, but at higher pressures the simulation underpredicts the experimental Xe loading. The results for H2 adsorption shown in Table 2 indicate that adsorption simulations can either over- or underpredict experimental results. However, the very good agreement at low Xe loading is a strong validation of our approach. A snapshot from the production simulation (Figure 4) indicates that there is no preferred binding site for Xe atoms at room temperature. This is different behavior than predicted for light gases at 30 K, in which the zinc-oxygen clusters are the preferred binding sites.78 However, Figure 4 is consistent with 129Xe NMR spectra at room
Figure 4. Snapshot showing Xe adsorbed by IRMOF-1 at 292 K (Zn, blue; O, red; C, gray; H, white; Xe, green).
temperature, indicating that Xe atoms occupy all possible adsorption sites within the IRMOF-1 pore.81 The force field parameters used in this study were optimized by fitting to experimental CO2 adsorption data and validated by comparison with CH4 adsorption data, both obtained at room temperature.57 We are confident that this potential parametrization for IRMOF-1 successfully captures binding sites for both electrostatic and van der Waals interactions. The ability of IRMOF-1 to store noble gases at much higher densities than the corresponding “empty containers” has been demonstrated experimentally.22 Corresponding GCMC simulation results are presented in Figure 5 for comparison. The agreement between the empty container compression curves (red lines in Figure 5) and the experimental results22 is excellent, suggesting that the van der Waals parameters correctly describe the gas-gas interactions for compressed Ar, Kr, and Xe. The simulated inflation curves (blue lines in Figure 5) show qualitatively that significant additional amounts of each gas can be stored in the same container filled with IRMOF-1. As usual, the simulation results overpredict the adsorption isotherms, but for this specific comparison two comments are in order. First, the simulations involve perfectly crystalline IRMOF-1, while
Ind. Eng. Chem. Res., Vol. 48, No. 7, 2009 3429
Figure 5. Simulated compression curves for noble gases in empty containers (red), compared with inflation curves (blue) in the presence of IRMOF-1. The blue curves represent total gas adsorption, not excess adsorption, for comparison with experimental results.22
Figure 7. Effect of feed gas attractive potential on Xe selectivity at 300 K and pressures of 1 and 10 bar, for equimolar feed gas mixtures of Xe/Ar and Xe/Kr. The attractive potential energies for Ar, Kr, and Xe are indicated.
substrate, as well as their intermolecular interactions. For IRMOF-1, the separation of Xe/Kr and Xe/Ar mixtures shows a dependence on feed gas properties. Figure 7 compares the Xe selectivity in equimolar Xe/Kr and Xe/Ar mixtures at total feed pressures of 1 and 10 bar. The comparison is made using the van der Waals well depth ε, but a similar trend is seen if the van der Waals diameter σ is used. Xe selectivity is larger in the Xe/Ar mixture because the ratio of εXe/εAr is larger than εXe/εKr. Conclusions
Figure 6. Effect of feed gas composition on Xe selectivity for Xe/Ar mixtures (solid lines) and Xe/Kr mixtures (dashed lines) at 300 K and total feed pressures of 1.0 and 10.0 bar.
the “in situ activated MOF sample” used experimentally22 almost certainly contained defects that would have reduced the surface area. Second, the simulated container has the dimensions of exactly one unit cell of IRMOF-1, while the experimental container could not have been completely filled with IRMOF-1 and therefore contained some empty space. Having validated the flexible force field for IRMOF-1 with adsorption isotherms of pure noble gases, we aim to use the simulations as a predictive tool in noble gas separations. A copper MOF has been shown to preferentially adsorb Xe in a Xe/Kr mixture, which leads to efficient separation.22 According to Figure 6, IRMOF-1 also shows a preferential adsorption for Xe over the smaller noble gases Kr and Ar. The selectivities shown in Figure 6 are largely independent of feed gas composition at 1 bar total pressure. At 10 bar, guest-guest interactions are more dominant, and Xe selectivity is dependent on composition.83 The selectivity for Xe, and the pressure dependence of SXe, is greatly enhanced when Xe is mixed with smaller atoms (Ar) compared with larger atoms (Kr). However, even smaller selectivities seen in the Xe/Kr mixtures (2.5-3.0) lead to an effective separation of the gases. When yXe ) 0.1 (10% of the feed gas is Xe; 90% is Kr), the adsorbed phase consists of 38% Xe and 62% Kr. The optimization of MOFs for gas separation must involve an examination of the atomic properties of both feed gas and
We have examined framework flexibility effects in GCMC simulations of gas adsorption by IRMOF-1. Trends in roomtemperature adsorption isotherms for IRMOF-1 are in good agreement with experiment for the noble gases and H2. The isosteric heat of adsorption for H2 agrees with both experimental and quantum values, suggesting that the IRMOF-1 force field parameters57 correctly model framework-H2 interactions. If the simulation results consistently over- or underpredicted the experimental loadings, some doubt could be cast on the force field parameters. However, the simulated loadings fall within the range of experimental loadings (H2, Table 2) or slightly underpredict experimental loadings (Xe, Figure 3). Additionally, these same force field parameters for IRMOF-1 successfully reproduced the adsorption isotherms at 298 K for CO2 and CH4.57 For Ar adsorption at 78 K, the agreement with experiment is good once the scaling factor is taken into account. Additionally, we showed that IRMOF-1 is effective at separating Xe from either Xe/Ar or Xe/Kr mixtures. There is a clear correlation between Xe selectivity and van der Waals well depth for each gas, but further investigation is required to determine if other gas or framework properties can also explain this trend. For monatomic gases and light gases such as H2, there is a negligible difference in gas loading between rigid and flexible frameworks. It is encouraging that, at least in the case of IRMOF-1, our modeling shows that framework moves represent a minimal increase in computation time with today’s hardware. The use of a flexible framework approach therefore should be feasible for MOFs in many cases, as well as being the most physically realistic approach to modeling these structurally flexible materials. Because the development of flexible forcefields is a time-consuming task, however, the use of standard, rigid force fields is a reasonable strategy for screening the
3430 Ind. Eng. Chem. Res., Vol. 48, No. 7, 2009
adsorption behavior of new or uncharacterized MOFs. This is particularly true for light or weakly interacting molecules, as illustrated by the results presented here. For larger adsorbate molecules, framework flexibility will most certainly affect gas adsorption, as we have seen for benzene diffusion in IRMOF1.54 In addition, gases that interact more strongly with the framework, including electron donors such as water, alcohols, and evidently CO2 (as shown by microcantilever experiments51) will likely require a flexible force field approach to capture all of the relevant effects. A recent GCMC paper on a Cr-MOF showed that large changes in framework structure upon adsorption of CO2 can be successfully modeled using a flexible framework.84 We are currently addressing these observations in our force field development efforts and will report the results in upcoming publications. Acknowledgment This work is supported by Sandia National Laboratories under its LDRD program. Sandia is a multiprogram laboratory operated by Sandia Corp., a Lockheed Martin company, for the U.S. Department of Energy under Contract DE-AC0494AL85000. Literature Cited (1) Frost, H.; Snurr, R. Q. Design Requirements for Metal-Organic Frameworks as Hydrogen Storage Materials. J. Phys. Chem. C 2007, 111, 18794. (2) Hirscher, M.; Panella, B. Hydrogen Storage in Metal-Organic Frameworks. Scr. Mater. 2007, 56, 809. (3) Lin, X.; Jia, J. H.; Hubberstey, P.; Schroder, M.; Champness, N. R. Hydrogen Storage in Metal-Organic Frameworks. Crystengcomm 2007, 9, 438. (4) Maspoch, D.; Ruiz-Molina, D.; Veciana, J. Old Materials with New Tricks: Multifunctional Open-Framework Materials. Chem. Soc. ReV. 2007, 36, 770. (5) Rowsell, J. L. C.; Yaghi, O. M. Strategies for Hydrogen Storage in Metal-Organic Frameworks. Angew. Chem. 2005, 44, 4670. (6) Xu, W.; Tao, Z. L.; Chen, J. Progress of Research on Hydrogen Storage. Prog. Chem. 2006, 18, 200. (7) Mu, C. Z.; Xu, F.; Lei, W. Application of Functional Metal-Organic Framework Materials. Prog. Chem. 2007, 19, 1345. (8) Nunes, S. P. Organic-Inorganic Membranes for Gas Separation. Ann. Chim.-Sci. Mat. 2007, 32, 119. (9) Snurr, R. Q.; Hupp, J. T.; Nguyen, S. T. Prospects for Nanoporous Metal-Organic Materials in Advanced Separations Processes. AIChE J. 2004, 6, 1090. (10) Suh, M. P.; Cheon, Y. E.; Lee, E. Y. Syntheses and Functions of Porous Metallosupramolecular Networks. Coord. Chem. ReV. 2008, 252, 1007. (11) Barcia, P. S.; Zapata, F.; Silva, J. A. C.; Rodrigues, A. E.; Chen, B. L. Kinetic Separation of Hexane Isomers by Fixed-Bed Adsorption with a Microporous Metal-Organic Framework. J. Phys. Chem. B 2007, 111, 6101. (12) Pan, L.; Olson, D. H.; Ciemnolonski, L. R.; Heddy, R.; Li, J. Separation of Hydrocarbons with a Microporous Metal-Organic Framework. Angew. Chem., Int. Ed. 2006, 45, 616. (13) Alaerts, L.; Kirschhock, C. E. A.; Maes, M.; van der Veen, M. A.; Finsy, V.; Depla, A.; Martens, J. A.; Baron, G. V.; Jacobs, P. A.; Denayer, J. E. M.; De Vos, D. E. Selective Adsorption and Separation of Xylene Isomers and Ethylbenzene with the Microporous Vanadium(IV) Terephthalate MIL-47. Angew. Chem., Int. Ed. 2007, 46, 4293. (14) Shimomura, S.; Horike, S.; Matsuda, R.; Kitagawa, S. GuestSpecific Function of a Flexible Undulating Channel in a 7,7,8,8-Tetracyanop-quinodimethane Dimer-Based Porous Coordination Polymer. J. Am. Chem. Soc. 2007, 129, 10990. (15) Pan, L.; Parker, B.; Huang, X. Y.; Olson, D. H.; Lee, J.; Li, J. Zn(tbip) (H2tbip ) 5-tert-Butyl Isophthalic Acid): A Highly Stable GuestFree Microporous Metal Organic Framework with Unique Gas Separation Capability. J. Am. Chem. Soc. 2006, 128, 4180. (16) Takamizawa, S.; Kachi-Terajima, C.; Kohbara, M. A.; Akatsuka, T.; Jin, T. Alcohol-Vapor Inclusion in Single-Crystal Adsorbents
[M2(II)(bza)(4)(pyz)](n) (M ) Rh, Cu): Structural Study and Application to Separation Membranes. Chem. Asian J. 2007, 2, 837. (17) Bastin, L.; Barcia, P. S.; Hurtado, E. J.; Silva, J. A. C.; Rodrigues, A. E.; Chen, B. A Microporous Metal-Organic Framework for Separation of CO2/N2 and CO2/CH4 by Fixed-Bed Adsorption. J. Phys. Chem. C 2008, 112, 1575. (18) Cheon, Y. E.; Suh, M. P. Multifunctional Fourfold Interpenetrating Diamondoid Network: Gas Separation and Fabrication of Palladium Nanoparticles. Chem. Eur. J. 2008, 14, 3961. (19) Dybtsev, D. N.; Chun, H.; Yoon, S. H. Microporous Manganese Formate: A Simple Metal-Organic Porous Material with High Framework Stability and Highly Selective Gas Sorption Properties. J. Am. Chem. Soc. 2004, 126, 32. (20) Dinca, M.; Long, J. R. Strong H2 Binding and Selective Gas Adsorption within the Microporous Coordination Solid Mg3(O2C-C10H6CO2)3. J. Am. Chem. Soc. 2005, 127, 9376. (21) Ma, S. Q.; Sun, D. F.; Wang, X. S.; Zhou, H. C. A Mesh-Adjustable Molecular Sieve for General Use in Gas Separation. Angew. Chem., Int. Ed. 2007, 46, 2458. (22) Mueller, U.; Schubert, M.; Teich, F.; Puetter, H.; Schierle-Arndt, K.; Pastre, J. Metal-Organic FrameworkssProspective Industrial Applications. J. Mater. Chem. 2006, 16, 626. (23) Du¨ren, T.; Sarkisov, L.; Yaghi, O. M.; Snurr, R. Q. Design of New Materials for Methane Storage. Langmuir 2004, 20, 2683. (24) Keskin, S.; Sholl, D. S. Screening Metal-Organic Framework Materials for Membrane-Based Methane/Carbon Dioxide Separations. J. Phys. Chem. C 2007, 111, 14055. (25) Karra, J. R.; Walton, K. S. Effect of Open Metal Sites on Adsorption of Polar and Nonpolar Molecules in Metal-Organic Framework Cu-BTC. Langmuir 2008, 24, 8620. (26) Ramsahye, N. A.; Maurin, G.; Bourrelly, S.; Llewellyn, P.; Loiseau, T.; Ferey, G. Charge Distribution in Metal Organic Framework Materials: Transferability to a Preliminary Molecular Simulation Study of the CO2 Adsorption in the MIL-53 (Al) System. Phys. Chem. Chem. Phys. 2007, 9, 1059. (27) Ramsahye, N. A.; Maurin, G.; Bourrelly, S.; Llewellyn, P. L.; Devic, T.; Serre, C.; Loiseau, T.; Ferey, G. Adsorption of CO2 in Metal Organic Frameworks of Different Metal Centres: Grand Canonical Monte Carlo Simulations Compared to Experiments. Adsorption 2007, 13, 461. (28) Babarao, R.; Hu, Z. Q.; Jiang, J. W.; Chempath, S.; Sandler, S. I. Storage and Separation of CO2 and CH4 in Silicalite, C-168 Schwarzite, and IRMOF-1: A Comparative Study from Monte Carlo Simulation. Langmuir 2007, 23, 659. (29) Du¨ren, T.; Snurr, R. Q. Assessment of Isoreticular Metal-Organic Frameworks for Adsorption Separations: A Molecular Simulation Study of Methane/n-Butane Mixtures. J. Phys. Chem. B 2004, 108, 15703. (30) Frost, H.; Duren, T.; Snurr, R. Q. Effects of Surface Area, Free Volume, and Heat of Adsorption on Hydrogen Uptake in Metal-Organic Frameworks. J. Phys. Chem. B 2006, 110, 9565. (31) Han, S. S.; Goddard, W. A. Lithium-Doped Metal-Organic Frameworks for Reversible H2 Storage at Ambient Temperature. J. Am. Chem. Soc. 2007, 129, 8422. (32) Jhon, Y. H.; Cho, M.; Jeon, H. R.; Park, I.; Chang, R.; Rowsell, J. L. C.; Kim, J. Simulations of Methane Adsorption and Diffusion within Alkoxy-Functionalized IRMOFs Exhibiting Severely Disordered Crystal Structures. J. Phys. Chem. C 2007, 111, 16618. (33) Jiang, J. W.; Sandler, S. I. Monte Carlo Simulation for the Adsorption and Separation of Linear and Branched Alkanes in IRMOF-1. Langmuir 2006, 22, 5702. (34) Jung, D. H.; Kim, D.; Lee, T. B.; Choi, S. B.; Yoon, J. H.; Kim, J.; Choi, K.; Choi, S. H. Grand Canonical Monte Carlo Simulation Study on the Catenation Effect on Hydrogen Adsorption onto the Interpenetrating Metal-Organic Frameworks. J. Phys. Chem. B 2006, 110, 22987. (35) Sagara, T.; Klassen, J.; Ganz, E. Computational Study of Hydrogen Binding by Metal-Organic Framework-5. J. Chem. Phys. 2004, 121, 12543. (36) Walton, K. S.; Snurr, R. Q. Applicability of the BET Method for Determining Surface Areas of Microporous Metal-Organic Frameworks. J. Am. Chem. Soc. 2007, 129, 8552. (37) Wang, S. Y. Comparative Molecular Simulation Study of Methane Adsorption in Metal-Organic Frameworks. Energy Fuels 2007, 21, 953. (38) Krungleviciute, V.; Lask, K.; Heroux, L.; Migone, A. D.; Lee, J. Y.; Li, J.; Skoulidas, A. Argon Adsorption on Cu3(Benzene-1,3,5Tricarboxylate)2(H2O)3 Metal-Organic Framework. Langmuir 2007, 23, 3106. (39) Liu, J. C.; Culp, J. T.; Natesakhawat, S.; Bockrath, B. C.; Zande, B.; Sankar, S. G.; Garberoglio, G.; Johnson, J. K. Experimental and Theoretical Studies of Gas Adsorption in Cu3(BTC)2: An Effective Activation Procedure. J. Phys. Chem. C 2007, 111, 9305.
Ind. Eng. Chem. Res., Vol. 48, No. 7, 2009 3431 (40) Nicholson, T. M.; Bhatia, S. K. Role of Electrostatic Effects in the Pure Component and Binary Adsorption of Ethylene and Ethane in CuTricarboxylate Metal-Organic Frameworks. Adsorpt. Sci. Technol. 2007, 25, 607. (41) Vishnyakov, A.; Ravikovitch, P. I.; Neimark, A. V.; Bulow, M.; Wang, Q. M. Nanopore Structure and Sorption Properties of Cu-BTC Metal-Organic Framework. Nano Lett. 2003, 3, 713. (42) Wang, S. Y.; Yang, Q. Y.; Zhong, C. L. Adsorption and Separation of Binary Mixtures in a Metal-Organic Framework Cu-BTC: A Computational Study. Sep. Purif. Technol. 2008, 60, 30. (43) Yang, Q. Y.; Zhong, C. L. Understanding Hydrogen Adsorption in Metal-Organic Frameworks with Open Metal Sites: A Computational Study. J. Phys. Chem. B 2006, 110, 655. (44) Sears, D. N.; Wasylishen, R. E.; Ueda, T. Grand Canonical Monte Carlo Simulations of the Xe-129 NMR Line Shapes of Xenon Adsorbed in (()-[Co(en)3]Cl3. J. Phys. Chem. B 2006, 110, 11120. (45) Mayo, S. L.; Olafson, B. D.; Goddard, W. A. DREIDING: A Generic Force Field for Molecular Simulations. J. Phys. Chem. 1990, 94, 8897. (46) Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. Development and Testing of the OPLS All-Atom Force Field on Conformational Energetics and Properties of Organic Liquids. J. Am. Chem. Soc. 1996, 118, 11225. (47) Rappe´, A. K.; Casewit, C. J.; Colwell, K. S.; Goddard, W. A., III; Skiff, W. M. UFF, A Full Periodic Table Force Field for Molecular Mechanics and Molecular Dynamics Simulations. J. Am. Chem. Soc. 1992, 114, 10024. (48) Fletcher, A. J.; Thomas, K. M.; Rosseinsky, M. J. Flexibility in Metal-Organic Framework Materials: Impact on Sorption Properties. J. Solid State Chem. 2005, 178, 2491. (49) Uemura, K.; Matsuda, R.; Kitagawa, S. Flexible Microporous Coordination Polymers. J. Solid State Chem. 2005, 178, 2420. (50) Cussen, E. J.; Claridge, J. B.; Rosseinsky, M. J.; Kepert, C. J. Flexible Sorption and Transformation Behavior in a Microporous MetalOrganic Framework. J. Am. Chem. Soc. 2002, 124, 9574. (51) Allendorf, M. D.; Houk, R. J. T.; Andruszkiewicz, L.; Talin, A. A.; Pikarsky, J.; Choudhury, A.; Gall, K. A.; Hesketh, P. J. Stress-Induced Chemical Detection Using Flexible Metal-Organic Frameworks. J. Am. Chem. Soc. 2008, 130, 14404. (52) Keskin, S.; Liu, J.; Rankin, R. B.; Johnson, J. K.; Sholl, D. S. Progress, Opportunities, and Challenges for Applying Atomically Detailed Modeling to Molecular Adsorption and Transport in Metal--rganic Framework Materials. Ind. Eng. Chem. Res. 2009, 48, 2355. (53) Greathouse, J. A.; Allendorf, M. D. The Interaction of Water with MOF-5 Simulated by Molecular Dynamics. J. Am. Chem. Soc. 2006, 128, 10678. (54) Greathouse, J. A.; Allendorf, M. D. Force Field Validation for Molecular Dynamics Simulations of IRMOF-1 and Other Isoreticular Zinc Carboxylate Coordination Polymers. J. Phys. Chem. C 2008, 112, 5795. (55) Huang, L. M.; Wang, H. T.; Chen, J. X.; Wang, Z. B.; Sun, J. Y.; Zhao, D. Y.; Yan, Y. S. Synthesis, Morphology Control, and Properties of Porous Metal-Organic Coordination Polymers. Microporous Mesoporous Mater. 2003, 58, 105. (56) Kaye, S. S.; Dailly, A.; Yaghi, O. M.; Long, J. R. Impact of Preparation and Handling on the Hydrogen Storage Properties of Zn4O(1,4benzenedicarboxylate)3 (MOF-5). J. Am. Chem. Soc. 2007, 129, 14176. (57) Dubbeldam, D.; Walton, K. S.; Ellis, D. E.; Snurr, R. Q. Exceptional Negative Thermal Expansion in Isoreticular Metal-Organic Frameworks. Angew. Chem., Int. Ed. 2007, 46, 4496. (58) Amirjalayer, S.; Tafipolsky, M.; Schmid, R. Molecular Dynamics Simulation of Benzene Diffusion in MOF-5: Importance of Lattice Dynamics. Angew. Chem., Int. Ed. 2007, 46, 463. (59) Tafipolsky, M.; Amirjalayer, S.; Schmid, R. Ab Initio Parametrized MM3 Force Field for the Metal-Organic Framework MOF-5. J. Comput. Chem. 2007, 28, 1169. (60) Huang, B. L.; McGaughey, A. J. H.; Kaviany, M. Thermal Conductivity of Metal-Organic Framework 5 (MOF-5): Part I. Molecular Dynamics Simulations. Int. J. Heat Mass Transfer 2007, 50, 393. (61) Dauber-Osguthorpe, P.; Roberts, V. A.; Osguthorpe, D. J.; Wolff, J.; Genest, M.; Hagler, A. T. Structure and Energetics of Ligand Binding to Proteins: Escherichia Coli Dihydrofolate Reductase-Trimethoprim, a Drug-Receptor System. Proteins: Struct., Funct., Genet. 1988, 4, 31. (62) Li, H.; Eddaoudi, M.; O’Keeffe, M.; Yaghi, O. M. Design and Synthesis of an Exceptionally Stable and Highly Porous Metal-Organic Framework. Nature 1999, 402, 276.
(63) Darkrim, F.; Levesque, D. Monte Carlo Simulations of Hydrogen Adsorption in Single-Walled Carbon Nanotubes. J. Chem. Phys. 1998, 109, 4981. (64) Goodbody, S. J.; Watanabe, K.; Macgowan, D.; Walton, J.; Quirke, N. Molecular Simulation of Methane and Butane in Silicalite. J. Chem. Soc., Faraday Trans. 1991, 87, 1951. (65) Hirschfelder, J. O.; Curtiss, C. F.; Bird, R. B. Molecular Theory of Gases and Liquids; Wiley: New York, 1965. (66) Maitland, G. C.; Rigby, M.; Smith, E. B.; Wakeham, W. A. Intermolecular Forces: Their Origin and Determination; Clarendon Press: Oxford, U.K., 1981. (67) http://towhee.sourceforge.net. (68) Frenkel, D.; Smit, B. Understanding Molecular Simulation: From Algorithms to Applications, 2nd ed.; Academic Press: San Diego, 2002. (69) Myers, A. L.; Monson, P. A. Adsorption in Porous Materials at High Pressure: Theory and Experiment. Langmuir 2002, 18, 10261. (70) Heuchel, M.; Snurr, R. Q.; Buss, E. Adsorption of CH4-CF4 Mixtures in Silicalite: Simulation, Experiment, and Theory. Langmuir 1997, 13, 6795. (71) Garberoglio, G.; Skoulidas, A. I.; Johnson, J. K. Adsorption of Gases in Metal Organic Materials: Comparison of Simulations and Experiments. J. Phys. Chem. B 2005, 109, 13094. (72) Buch, V. Path-Integral Simulations of Mixed para-D2 and orthoD2 ClusterssThe Orientational Effects. J. Chem. Phys. 1994, 100, 7610. (73) Darkrim, F.; Vermesse, J.; Malbrunot, P.; Levesque, D. Monte Carlo Simulations of Nitrogen and Hydrogen Physisorption at High Pressures and Room Temperature. Comparison with Experiments. J. Chem. Phys. 1999, 110, 4020. (74) Panella, B.; Hirscher, M.; Putter, H.; Muller, U. Hydrogen Adsorption in Metal-Organic Frameworks: Cu-MOFs and Zn-MOFs Compared. AdV. Funct. Mater. 2006, 16, 520. (75) Dailly, A.; Vajo, J. J.; Ahn, C. C. Saturation of Hydrogen Sorption in Zn Benzenedicarboxylate and Zn Naphthalenedicarboxylate. J. Phys. Chem. B 2006, 110, 1099. (76) Hafizovic, J.; Bjorgen, M.; Olsbye, U.; Dietzel, P. D. C.; Bordiga, S.; Prestipino, C.; Lamberti, C.; Lillerud, K. P. The Inconsistency in Adsorption Properties and Powder XRD Data of MOF-5 Is Rationalized by Framework Interpenetration and the Presence of Organic and Inorganic Species in the Nanocavities. J. Am. Chem. Soc. 2007, 129, 3612. (77) Tsao, C. S.; Yu, M. S.; Chung, T. Y.; Wu, H. C.; Wang, C. Y.; Chang, K. S.; Chent, H. L. Characterization of Pore Structure in MetalOrganic Framework by Small-Angle X-ray Scattering. J. Am. Chem. Soc. 2007, 129, 15997. (78) Dubbeldam, D.; Frost, H.; Walton, K. S.; Snurr, R. Q. Molecular Simulation of Adsorption Sites of Light Gases in the Metal-Organic Framework IRMOF-1. Fluid Phase Equilib. 2007, 261, 152. (79) Surble, S.; Millange, F.; Serre, C.; Duren, T.; Latroche, M.; Bourrelly, S.; Llewellyn, P. L.; Ferey, G. Synthesis of MIL-102, a Chromium Carboxylate Metal-Organic Framework, with Gas Sorption Analysis. J. Am. Chem. Soc. 2006, 128, 14889. (80) Sagara, T.; Klassen, J.; Ortony, J.; Ganz, E. Binding Energies of Hydrogen Molecules to Isoreticular Metal-Organic Framework Materials. J. Chem. Phys. 2005, 123, 014701. (81) Ooms, K. J.; Wasylishen, R. E. 129Xe NMR Study of Xenon in Iso-reticular Metal-Organic Frameworks. Microporous Mesoporous Mater. 2007, 103, 341. (82) Pawsey, S.; Moudrakovski, I.; Ripmeester, J.; Wang, L. Q.; Exarhos, G. J.; Rowsell, J. L. C.; Yaghi, O. M. Hyperpolarized Xe-129 Nuclear Magnetic Resonance Studies of Isoreticular Metal-Organic Frameworks. J. Phys. Chem. C 2007, 111, 6060. (83) Yang, Q. Y.; Xue, C. Y.; Zhong, C. L.; Chen, J. F. Molecular Simulation of Separation of CO2 from Flue Gases in Cu-BTC Metal-Organic Framework. AIChE J. 2007, 53, 2832. (84) Salles, F.; Ghoufi, A.; Maurin, G.; Bell, R. G.; Mellot-Draznieks, C.; Ferey, G. Molecular Dynamics Simulations of Breathing MOFs: Structural Transformations of MIL-53(Cr) Thermal Activation and CO2 Adsorption. Angew. Chem., Int. Ed. 2008, 47, 8487.
ReceiVed for reView August 25, 2008 ReVised manuscript receiVed December 23, 2008 Accepted February 4, 2009 IE801294N
