Understanding Adsorption of Highly Polar Vapors on Mesoporous MIL

Mar 26, 2013 - The adsorption of polar water and methanol vapors on the meso- and microporous metal–organic frameworks (MOFs) MIL-100(Cr) and MIL-10...
0 downloads 13 Views 3MB Size
Article pubs.acs.org/JPCC

Understanding Adsorption of Highly Polar Vapors on Mesoporous MIL-100(Cr) and MIL-101(Cr): Experiments and Molecular Simulations Martijn F. De Lange,*,†,‡ Juan-Jose Gutierrez-Sevillano,§ Said Hamad,§ Thijs J. H. Vlugt,‡ Sofia Calero,§ Jorge Gascon,*,† and Freek Kapteijn† †

Catalysis Engineering, Chemical Engineering Department, Delft University of Technology, Julianalaan 136, 2628 BL Delft, The Netherlands ‡ Process & Energy Laboratory, Delft University of Technology, Leeghwaterstraat 44, 2628CA Delft, The Netherlands § Department of Physical, Chemical, and Natural Systems, University Pablo de Olavide, Carretera de Utrera km 1, 41013 Seville, Spain S Supporting Information *

ABSTRACT: The adsorption of polar water and methanol vapors on the meso- and microporous metal−organic frameworks (MOFs) MIL-100(Cr) and MIL-101(Cr) has been studied by a combined experimental and simulation approach. The computational effort for these MOFs with large unit cells was reduced by using primitive unit cells that were 4 times smaller. Our results demonstrate that both adsorbate− adsorbent and adsorbate−adsorbate interactions control the adsorption process. At low loadings, before all coordinatively unsaturated chromium sites are occupied, the MOF structure determines the shape of the isotherm, and the molecular model for the polar sorbate is less important. A clear difference was found between fully fluorinated and hydroxylated MIL-101 structures for both methanol and water, demonstrating that partial charges on Cr drive the initial shape of the isotherm. At higher loadings, adsorbate−adsorbate interactions become much more important, and the choice of the water model, in particular, is crucial for the agreement between experimental and simulation results. The simplest SPC/E model reproduces the experimental results with the best accuracy, in contrast to more advanced models such as TIP5PEw, which can be explained by the slightly stronger Coulombic interactions predicted by the former. For methanol, the general TraPPE force field performs well. A composite type IV isotherm for methanol and a composite type V isotherm for water, according to the IUPAC classification, were found. The heats of adsorption are in line with these conclusions. To the best of our knowledge, this effect of the sorbate model has not been observed in adsorption in microporous materials, and it highlights the complexity behind molecular simulations in periodic mesostructured materials.



INTRODUCTION

functionalization, grafting, and encapsulation. These materials exhibit very high adsorption capacities, specific surface areas, and pore volumes. Their porosity is much higher than that of their inorganic counterpart zeolites (up to 90% higher). Their thermostability is sometimes unexpectedly high, reaching temperatures above 400 °C. Obviously, MOFs have attracted much attention: the majority of studies have dealt with the synthesis of new structures,3 and most applications have focused on adsorption/separation,4 storage,5 encapsulation,6 and catalysis.7 Among the hundreds of known MOF structures, materials such as MIL-1018 and MIL-1008b (MIL stands for material from Institut Lavoisier) offer tremendous possibilities for material engineers. These hybrid solid are built from supertetrahedra (ST) building units, which are formed by rigid terephthalic or trimesic acid linkers and trimeric chromium(III)

Synthetic crystalline micro- and mesoporous materials have been extensively researched during the past few decades.1 Several unique aspects of these materials are responsible for their success: Because they have a very high and tunable adsorption capacity, active sites of different strengths can be generated in the frameworks. The sizes of their channels and cavities fall within the range of many molecules of interest, and many materials present excellent ion-exchange capabilities and exciting electronic properties, ranging from insulators to conductors and semiconductors.2 In addition, owing to their periodic nature, nanostructured materials are excellent playgrounds for scientists, because macroscopic events can be explained on the basis of interactions at the molecular level. Among the different classes, metal−organic frameworks (MOFs) bridge micro- and mesoporous materials and present unprecedented topological richness. The combination of organic and inorganic building blocks offers an almost infinite number of combinations; enormous flexibility in pore size, shape, and structure; and unlimited opportunities for © 2013 American Chemical Society

Received: December 28, 2012 Revised: March 26, 2013 Published: March 26, 2013 7613

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622

The Journal of Physical Chemistry C

Article

autoclave. The autoclave was then heated in an oven at 493 K for 4 days under static conditions. The resulting solid was filtered off, washed with deionized water and then acetone, and finally dried overnight at 433 K and stored under an air atmosphere. Nitrogen physisorption measurements were performed at 77 K using a Quantachrome Autosorb-6B unit gas adsorption analyzer. Vapor adsorption isotherms were measured using a Quantachrome Autosorb 1C volumetric adsorption analyzer equipped with a vapor-dosing system. An equilibration time of 600 s was used for all measurements. All samples were outgassed for 16 h under a vacuum at 473 K before adsorption analysis, for both nitrogen and vapor measurements. Pressures were converted to fugacities using the Peng−Robinson equation of state, which is valid for the low pressures used in this work.12 The isosteric heat of adsorption was estimated from isotherm measurements at 303 and 313 K. Crystallinity was assessed using X-ray diffraction (XRD) on a Bruker-AXS D5005 diffractometer (Co Kα radiation).

oxide octahedral clusters. The resulting solids contain two types of quasi-spherical mesoporous cages, limited by 12 pentagonal faces for the smaller cages and 16 faces for the larger cages. The former so-called medium cavities are accessible through 1.2-nm (MIL-101) or 0.5-nm (MIL-100) pentagonal windows, whereas the large cavities are communicated through the same pentagonal windows and 1.5-nm (MIL-101) or 0.9-nm (MIL100) hexagonal windows. Since the discovery of the two structures, numerous publications have reported on their excellent stability and on several prospective applications.9 Because a very large number of MOFs have been synthesized to date and many more are possible, the role of molecular simulations becomes even more important to screen the properties of new materials, to gain microscopic insight, and to elucidate the underlying physics behind molecular interactions upon the adsorption of different adsorbates. Adsorptive behavior of porous materials is indeed a key feature, because it is important not only for gas storage or separation but also for other applications such as catalysis or nanomedicine. To date, most simulation studies of metal−organic frameworks have focused on the adsorption and transport properties of small gases (mainly CO2 and CH4) in microporous MOFs.10 When it comes to micromesoporous structures such as MIL-101 and MIL-100, because of their unit-cell complexity and large computational requirements, the number of simulation works is even lower.10k,11 In addition, very little attention has been devoted to studying the interaction of polar vapors such as water with MOFs, which are of the utmost importance for stability for the “real-life” application of these materials. The latter is mostly due to the complexity of describing the adsorption of such polar adsorbates, where molecule−molecule interactions play a major role. In this work, we make a quantum leap in understanding the adsorption of highly polar vapors on micromesostructured materials with the MIL-100 and MIL-101 topologies. Adsorption of water and methanol was studied on both structures by a combined experimental and simulation approach: The main adsorption sites, mechanism of adsorption, and the role of adsorbate−adsorbent and adsorbate−adsorbate interactions have been identified.



SIMULATION METHODS Both MIL-100(Cr) and MIL-101(Cr) were modeled as rigid structures, interacting with adsorbing molecules through van der Waals and Coulombic interactions. The assumption of rigidity is often made when simulating adsorption in MOFs13 and is justified when pore dimensions exceed the kinetic diameters of the adsorbing molecules and when the structure does not undergo any significant adsorbent-induced deformation. The van der Waals interactions were described by Lennard-Jones potentials, for which the parameters were taken from the DREIDING force field,14 except those for chromium atoms, for which UFF (universal force field) was used.15 The partial charges needed to describe the Coulombic interactions were taken from Yazaydin16 for the fluorinated structures. For the hydroxylated structure, the atomic charges were calculated with the code DMol317 as implemented in the Accelrys software package Materials Studio18 using the PW91 exchangecorrelation functional19 with the double numerical plus polarization (DNP) basis set. The cluster approach was employed to obtain the partial charges. The atom-centered charges were those that best fit the electrostatic potential (ESP) of the cluster shown in Figure S.1 (Supporting Information), which is a model of a chromium trimer in a mesoporous cage of MIL-101. It is well-known that fluorine is involved in the terminal bond of the trimeric chromium species and partly substitutes the terminal water molecules attached to chromium in MIL-100 and MIL-101. Although it is not yet fully clear, it seems that the use of fluorine provides a strong interaction with the chromium octahedral motif and supports effective nucleus formation of MIL-101 during the hydrothermal reaction. To study the effect of fluorine on the adsorption of polar vapors, we employed the fully fluorinated structure of MIL-10116 as a starting point and compared its adsorption properties with those of a structure in which all fluorine atoms were exchanged by OH groups. It should be stressed that, in the real structure, a mixture of fluorinated and hydroxylated Cr trimers will be found. The full set of applied parameters are listed in Table S.1 (Supporting Information). The original MIL-101(Cr) and MIL-100(Cr) unit cells were simplified to primitive unit cells using Materials Studio,18 by making use of symmetry. The volumes of these primitive cells are only one-fourth of the volumes of the original cubic cells (Table S.2, Supporting Information), but the unit cells are no longer cubic. As the



EXPERIMENTAL METHODS All chemicals were obtained from Sigma-Aldrich and were used without further purification. Synthesis of MIL-101(Cr) was performed as previously reported in the literature.8 Specifically, 1.63 g of chromium(III) nitrate [Cr(NO3)3·9H2O, 97%], 0.7 g of terephthalic acid [C6H4-1,4-(CO2H)2, 97%], 0.20 g of hydrofluoric acid (HF, 40%), and 20 g of distilled water were added to a Teflon container that was inserted in a stainless steel autoclave. The autoclave was heated for 8 h at 493 K in an oven under static conditions. After synthesis, the solid product was filtered from the synthesis solution. Then, a solvothermal treatment was performed using ethanol (EtOH, 95%) at 353 K for 24 h. The resulting solid was exchanged in a 1 M solution of ammonium fluoride (NH4F) at 343 K for 24 h and was immediately filtered off and washed with hot water. The solid was finally dried overnight at 433 K and stored under an air atmosphere. Synthesis of MIL-100(Cr) was performed in accordance with the literature as well.8b A mixture of 0.5 g of chromium(VI) oxide (CrO3), 1.05 g of trimesic acid [C6H3-1,3,5-(CO2H)3, 97%], 0.2 g hydrofluoric acid (HF), and 24 g of H2O was added to a Teflon container that was inserted in a stainless steel 7614

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622

The Journal of Physical Chemistry C

Article

ensemble. In this ensemble, the chemical potential, volume, and temperature are kept constant, allowing the number of adsorbing molecules to fluctuate. The chemical potential was fixed by fixing the fugacity. Monte Carlo moves included rotation, translation, and (re)insertion of adsorbing molecules. The heat of adsorption at zero coverage was calculated using the Widom particle insertion method in the canonical ensemble (fixed number of particles, volume, and temperature).30 The heat of adsorption at nonzero coverage was determined using the method developed by Vlugt et al.31 Radial distribution functions were obtained by collecting and averaging the atomic positions of all atoms over more than 100 simulation cycles, after equilibration of the simulation box.

initially reported cells are relatively large (that of MIL-101 occupies a volume of 903 Å3), the use of primitive cells considerably decreases the computational requirements while still describing the MOF structures equally well. Methanol was described using the transferable TraPPE force field for polar hydrocarbons20 (Table S.3, Supporting Information), which accurately describes vapor−liquid coexistence and is commonly used to describe methanol adsorption in ordered porous materials.21 For simulations concerning the adsorption of water, there is no consensus on which molecular representation to use. Multiple models are available in the literature, none of which are able to describe several arbitrarily chosen properties of water simultaneously.22 In this work, three distinct water models were used to investigate the effect of water−water interactions during adsorption in mesoporous frameworks. Tip5PEw23 was selected because it accurately reproduces the liquid density and has been used previously for adsorption in microporous materials.21a,24 SPC/E25 was selected because is a relatively simple representation of water and therefore requires less computational power. Also, this model was used to obtain a reasonable description of experimental adsorption data in microporous zeolites.26 Finally, TIP4PEw27 was used for water adsorption because of its good reproduction of bulk liquid properties.28 All three models consider water as a rigid, nonpolarizable molecule. Although these models might be less accurate in describing water properties, they require less computational efforts, a necessity considering the large unit-cell size of the periodic systems under study. All three models assume a single Lennard-Jones interaction site, the oxygen, with similar parametrizations. The main difference among these models is the incorporation of electrostatic interactions (Figure 1, Table 1). TIP5PEw uses two dummy atoms to distribute the



RESULTS AND DISCUSSION The methanol adsorption results on MIL-100(Cr) are presented in Figure 2. The shape of the isotherm found

Figure 2. Simulated (■) and experimentally measured (○) adsorption of methanol on MIL-100(Cr) at 303 K and experimental results scaled by 5/4 (+).

experimentally is reproduced with fair accuracy by the simulations, although the loading obtained in the simulations is higher. This is not surprising, as the simulation uses a perfect crystal whereas the synthesized material is prone to have imperfections to a certain extent. If one scales the simulated results, common practice when combining simulations with experiments, to take into account the inaccessible porosity in the synthesized material,32 very good agreement is found at higher fugacities. At lower fugacities, the simulations appear to overpredict the adsorption. This might indicate that the employed partial charges are slightly too high. Polar compounds are generally sensitive to small changes in partial charges.24 In Figure 3, the experimental methanol adsorption results are compared with simulations using both the fluorinated and hydroxylated structure models of MIL-101. The results for both structures agree to a large extent with the scaled experimental

Figure 1. Representation of the water models used. Assigned values of the model parameters are presented in Table 1.

negative charge, TIP4PEw uses a single dummy atom, and SPC/E has the negative charge located on the oxygen atom. All mixed-pair potentials were calculated using Lorentz−Berthelot mixing rules. Lennard-Jones potentials were cut off and shifted at a radius of 12 Å. For electrostatics, an Ewald summation29 with a relative precision of 10−6 was used for truncation of the Coulombic potentials at a radius of one-half the length of the simulation box. Adsorption isotherms were calculated using classical Monte Carlo simulations in the grand canonical Table 1. Parameters of the Water Models Used model

εOkb−1 (K)

σO (Å)

qO (e)

qH (e)

qM (e)

qL (e)

lO−H (Å)

lO−M (Å)

lO−L (Å)

ΘH−O−H (deg)

ΘH−O−M (deg)

ΘL−O−L (deg)

SPC/E25 TIP4PEw27 TIP5PEw23

78.2 81.899 89.633

3.1656 3.16435 3.097

−0.8476 − −

0.4238 0.52422 0.241

− −1.04844 −

− − −0.241

1 0.9572 0.9572

− 0.125 −

− − 0.7

109.47 104.52 104.52

− 52.26 −

− − 109.47

7615

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622

The Journal of Physical Chemistry C

Article

Figure 4. Simulated adsorption of water on fluorinated MIL-101(Cr) for two different fugacities at 303 K for SPC/E (black), TIP4PEw (gray), and TIP5PEw (white).

Figure 3. Simulated adsorption for fluorinated (■) and hydroxylated (▲) structures and experimentally measured (○) adsorption of methanol on MIL-101(Cr) at 303 K and experimental results scaled by 5 /4 (+).

Table 2. Dipole Moments of Water Models Used Compared to Experimentally Found Dipoles

results. At fugacities above 102 Pa, the simulation results for both structures are very similar. Below this fugacity, there is a discrepancy between the two simulation results. The difference can most likely be attributed to the partial charges on the coordinatively unsaturated chromium atoms. In the fluorinated structure, these charges are higher (see Table S.1, Supporting Information), hence the higher adsorption at very low fugacities herein. The absence of notable discrepancies between results for these two structures above 102 Pa shows that the effect of charge is diminished when sufficient methanol molecules have adsorbed. The loading at this fugacity corresponds to ∼140 molecules per unit cell, the total number of coordinatively unsaturated chromium sites present in the structure. If one were to extrapolate the experimentally found adsorption isotherm to lower fugacity, it would appear to correspond better to the hydroxylated structure. However, no definitive qualitative discrimination can be made between the two structural representations of MIL-101 because of the lack of reliable experimental data at very low fugacities. The shape of the methanol isotherms on both MIL-100 and MIL-101 can be characterized by a combination of two IUPAC type IV isotherms, one corresponding to medium cages and one to large cages.33 This is visible when the fugacity is depicted linearly, as is done in Figure S.2 (Supporting Information). As mentioned in the previous section, to describe water adsorption in MIL-100 and MIL-101, three distinct water models were employed: SPC/E, TIP4PEw, and TIP5PEw. To assess the individual performances of these models, simulations were performed using the different models for two intermediate fugacities in MIL-101(Cr). The results, presented in Figure 4, clearly indicate large differences in loading for the three different water models. At both fugacities, the loading is highest for the SPC/E model and lowest for TIP5PEw. As the Lennard-Jones interactions are quite similar for the different models (see Table 1), the difference is most likely due to the difference in Coulombic interactions. If one compares the dipole moments of the three models (see Table 2), it becomes apparent that a small increase in dipole moment leads to a significant increase in adsorption. On another note, these water models all underpredict the dipole of liquid water and overpredict that of water vapor. This is a known problem,22 which could be circumvented by using polarizable models. However, considering the tremendous increase in computa-

dipole moment (D) model SPC/E TIP4PEw TIP5PEw actual40 ice liquid gas

2.35 2.32 2.29 3.09 2.95 1.85

tional expenditure of these models, doing so would be computationally infeasible. In the remainder of this work, the SPC/E and TIP5PEw water models were used to investigate the effect of Coulombic interactions over a broad range of fugacities in both structures under study. Figure 5 presents the simulated adsorption results for both SPC/E and TIP5PEw, supplemented with experimental literature data for MIL-100(Cr). At low fugacities, there is no notable difference between the two models: both slightly overpredict the experimentally found adsorption. A striking difference is visible when comparing the fugacities at which the two models predict a large step in uptake. For TIP5PEw, this

Figure 5. Simulated adsorption on MIL-100(Cr) using TIP5PEw (■) and SPC/E (◀) water models at 303 K. Experimental isotherms from Chang et al. (open triangles)35 and Akyiama et al. (298 K, ◊).36. 7616

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622

The Journal of Physical Chemistry C

Article

impermeable parts that only effectively dilute the part of the material that is available for adsorption or as regions where the formation of small amounts of nonporous phases (not noticeable by powder XRD) takes place. This assumption would justify the comparison with the scaled experimental results. Identical to the findings for methanol on MIL-101(Cr), at loadings below 140 molecules per unit cell, there is a discrepancy between the results obtained from the hydroxylated and fluorinated structures. This can again be attributed to the difference in partial charges on the coordinatively unsaturated chromium sites. Above this loading threshold, the results for the hydroxylated and fluorinated structures are very similar. These conclusions hold individually for both the SPC/E and TIP5PEw models. Again, the correspondence between the hydroxylated and fluorinated structures suggest that the effect of the framework charges is negligible. As for MIL-100(Cr), for MIL-101(Cr), the location of the uptake step is predicted to occur at fugacities 1 order of magnitude higher than those found experimentally by the TIP5Ew model. In contrast, again as for MIL-100(Cr), the SPC/E model provides good agreement with the experimental data. The large difference in adsorption observed as a function of adsorbent polarity seems to contradict the statement that the results for the hydroxylated and fluorinated structures are similar, as one of the key differences between these structures is in the partial charges employed. That this is, in fact, a paradox and not a contradiction lies in the fact that adsorbate−adsorbate interactions dominate the adsorption process once all of the coordinatively unsaturated chromium sites are occupied. This is plausible considering the achieved loading. Up to 5000 molecules of water can be present in a reduced unit cell of MIL-101(Cr), which contains only ∼210 chromium sites, of which ∼140 are coordinatively unsaturated. The increased dipole for the SPC/E model, with respect to TIP5PEw, clearly affects the affinity between water molecules. As a result, the step in adsorption occurs at a lower fugacity for SPC/E. The shapes of the water isotherms for both MIL-100 and MIL-101 can be described by a combination of two IUPAC type V isotherms, one corresponding to medium cages and one to large cages.33 The radial distribution functions at 102 Pa (see Figures S.3−S.6, Supporting Information) show the relative hydrophobicity of the structural fluor groups, compared to the structural hydroxyl group. The ratio between the number of water molecules in close proximity to the unsaturated chromium sites and the number of water molecules close to the coordinated chromium sites is larger in fluorinated MIL-101 (see Figures S.3 and S.4, Supporting Information). Water is located far from the structural fluor groups (Figure S.5, Supporting Information), as evidenced by the fact that the second peak in the radial distribution corresponding to the distance between this fluor group and the oxygen atom of water is larger. Furthermore, these distributions indicate that the oxygen atom of the hydroxyl group and the hydrogen atom of water undergo a clear interaction in the hydroxylated structure (Figure S.6, Supporting Information). The importance of adsorbate−adsorbate interactions in adsorption in these structures can also be observed in the heat of adsorption of water. As shown in Figure 8 for MIL101(Cr), the heat of adsorption drops very quickly from ∼80 kJ mol−1 at very low loadings to just above the heat of evaporation of water when the loading is only slightly increased. This clearly indicates that water−water interactions control adsorption after the chromium sites have been occupied. As expected, the heats

step occurs at a fugacity 1 order of magnitude higher than those found experimentally, whereas the fugacity for this step predicted by SPC/E is very close to the experimental values. The difference can again be attributed to the different polarity for water in these models. For both fluorinated and hydroxylated MIL-101(Cr), the simulated adsorption results for TIP5Ew and SPC/E, supplemented with experimental and literature data, are shown in Figure 6 (with the low-fugacity regime expanded in

Figure 6. Simulated adsorption for fluorinated MIL-101(Cr) using the TIP5PEw (■) and SPC/E (◀) water models and for hydroxylated MIL-101(Cr) using the TIP5PEw (▲) and SPC/E (▶) water models at 303 K. Experimental data from Ehrenmann et al. [298 K, volumetric (○) and gravimetric (□)],37 Küsgens et al. (298 K, △),38 Akiyama et al (298 K, ◊),9a Chang et al. (open triangles),35 and our experiments (▽).

Figure 7. Low-fugacity regime of Figure 6.

Figure 7). The experimental results in the literature all show a characteristic step in water uptake at fugacities of ∼103 Pa. The quantity adsorbed at saturation, however, varies. This discrepancy can be related to material quality. For MIL101(Cr), it is known that the synthesis conditions strongly influence the quality of the resulting material (specifically, some chromium oxide can precipitate together with the MOF).34 Because the only noticeable difference between the different experimental isotherms is in the saturation loading, it seems plausible that the imperfections in these materials do not play a role during adsorption. These impurities can be thought of as 7617

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622

The Journal of Physical Chemistry C

Article

tant than in the case of MIL-101(Cr). The heats of adsorption of methanol in MIL-100 and MIL-101, shown in Figures 10

Figure 8. Simulated heat of adsorption of water for fluorinated MIL101(Cr) for the TIP5PEw (■) and SPC/E (◀) water models and for hydroxylated MIL-101(Cr) using the SPC/E (▶) water model at 303 K. Also included are isosteric heat of adsorption measurements from Chang et al. (open triangles)35 and heat of adsorption estimates from Küsgens et al. (△)38 and Akiyama et al (◊).9a The dashed line indicates the enthalpy of evaporation of water.39 Error bars indicate the 95% confidence interval.

Figure 10. Simulated heat of adsorption of methanol on MIL-100(Cr) (■) and estimated isosteric heat of adsorption (○). The dashed line indicates the enthalpy of evaporation of methanol.39 Error bars indicate the 95% confidence interval.

of adsorption obtained by simulations using the SPC/E model are closer to the values observed experimentally than those obtained using the TIP5PEw model. The heats of evaporation of bulk water at room temperature predicted by these two models, 48.9 and 43.4 kJ mol−1 for SPC/E22 and TIP5PEw,23 respectively, explain the difference in the heat of adsorption. In addition, before all coordinatively saturated chromium sites are occupied, the heat of adsorption observed for the hydroxylated structure is lower than that for the fluorinated one. For MIL100(Cr), for which the water adsorption results are presented in Figure 9, the heat of adsorption decreases more gradually as

Figure 11. Simulated heat of adsorption of methanol on fluorinated (■) and hydroxylated (▲) MIL-101(Cr) and estimated isosteric heat of adsorption (○). The dashed line indicates the enthalpy of evaporation of methanol.39 Error bars indicate the 95% confidence interval.

and 11, respectively, exhibit trends similar to those for water: The heat of adsorption decreases from high values to values close to the heat of evaporation upon a moderate increase in loading. The experimentally calculated heats of adsorption are in very good agreement with those found in the simulations. Figure 12 shows the locations of water molecules in both the medium and large cages as a function of loading (corresponding results for methanol are shown in Figure S.7, Supporting Information), and Figure 13 shows water molecules located close to a supertetrahedron (for methanol, see Figure S.8, Supporting Information). From these results in combination with the adsorption results, an adsorption process can be deduced. At low loadings, below ∼140 molecules per unit cell, the adsorbate molecules are located only next to the coordinatively unsaturated chromium atoms. As the loading increases, adsorbate molecules start clustering around the molecules that were already present at these chromium sites,

Figure 9. Simulated heat of adsorption of water on MIL-100(Cr) using the TIP5PEw (■) and SPC/E (◀) water models at 303 K. Also included are heat of adsorption measurements on MIL-100(Fe) from Chang et al. (open triangles).35 The dashed line indicates the enthalpy of evaporation of water.39 Error bars indicate the 95% confidence interval.

a function of loading. These results demonstrate that structure−adsorbate interactions are of relatively more importance for MIL-100(Cr) at low to intermediate loadings: Because the size of the MIL-100(Cr) cavities is smaller, fewer water molecules are adsorbed at saturation, and hence, structure−adsorbate interactions become much more impor7618

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622

The Journal of Physical Chemistry C

Article

adsorption process. At low loadings, before all coordinatively unsaturated chromium sites have been occupied, the structure determines the shape of the isotherm, and the water model is less important. A clear difference was found between fully fluorinated and hydroxylated MIL-101 structures for both methanol and water, demonstrating that the Cr partial charges drive the initial shape of the isotherm. At higher loadings, adsorbate−adsorbate interactions become much more important, and the choice of water model determines the agreement between experimental and simulated results. In this sense, the simplest SPC/E model reproduces the experimental results with the best accuracy in contrast to more advanced methods such as TIP5PEw because of the slightly higher Coulombic interactions predicted by the former. A composite type IV isotherm for methanol and a composite type V isotherm for water were found, according to the IUPAC classification. The heat of adsorption results are in line with these conclusions. To the best of our knowledge, this effect of the sorbate model has not been observed in adsorption in microporous materials, and it highlights the complexity behind molecular simulations in periodic mesostructured materials.



ASSOCIATED CONTENT

S Supporting Information *

Radial distribution functions of water (SPC/E) in both fluorinated and hydroxylated MIL-101, characterization of the used samples of MIL-100 and MIL-100, simulated results on the normal pressure scale, unit-cell parameters of both the cubic and primitive unit cells, and force-field parameters of methanol and all structures used. This material is available free of charge via the Internet at http://pubs.acs.org.



Figure 12. Water (using the SPC/E model) located in (left) medium and (right) large cages as a function of pressure. The top, middle, and bottom structures are for 1 Pa, 15 kPa, and 30 kPa, respectively.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (M.F.D.L.), j.gascon@tudelft. nl (J.G.). Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS Financial support from the Advanced Dutch Energy Materials program of the Dutch Ministry of Economic Affairs, Agriculture, and Innovation is gratefully acknowledged. This work was also sponsored by the Stichting Nationale Computerfaciliteiten (National Computing Facilities Foundation, NCF, of The Netherlands) for the use of supercomputing facilities, with financial support from the Nederlandse Organisatie voor Wetenschappelijk onderzoek (Netherlands Organization for Scientific Research, NWO). We thank the European Research Council for support through an ERC Starting Grant (S.C.).

Figure 13. Water (SPC/E) located close to a supertetrahedron at (left) 1 Pa and (right) 15 kPa.



thus filling the windows of the cavities. When the loading is increased further, the adsorbate molecules start filling the cavities completely. Overall, this leads to a composite type IV isotherm for methanol and type V isotherm for water, according to the IUPAC classification.33

REFERENCES

(1) Schuth, F.; Schmidt, W. Microporous and Mesoporous Materials. Adv. Mater. 2002, 14 (9), 629−638. (2) (a) Corma, A. State of the Art and Future Challenges of Zeolites As Catalysts. J. Catal. 2003, 216 (1−2), 298−312. (b) Corma, A.; Garcia, H. Zeolite-Based Photocatalysts. Chem. Commun. 2004, 13, 1443−1459. (3) Tranchemontagne, D. J.; Mendoza-Cortes, J. L.; O’Keeffe, M.; Yaghi, O. M. Secondary Building Units, Nets and Bonding in the Chemistry of Metal−Organic Frameworks. Chem. Soc. Rev. 2009, 38 (5), 1257−1283.



CONCLUSIONS The adsorption of polar vapors on mesoporous MOFs has been studied by a combination of experimental and simulation techniques. Our results demonstrate that both adsorbate− adsorbent and adsorbate−adsorbate interactions control the 7619

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622

The Journal of Physical Chemistry C

Article

Using Operando IR Spectroscopy. Chem.Eur. J. 2012, 18 (38), 11959−11967. (10) (a) Arici, M.; Yesilel, O. Z.; Keskin, S.; Tas, M. A ThreeDimensional Silver(I) Framework Assembled from 3,3′-Thiodipropionate: Synthesis, Structure and Molecular Simulations for Hydrogen Gas Adsorption. Polyhedron 2012, 45 (1), 103−106. (b) Atci, E.; Keskin, S. Atomically Detailed Models for Transport of Gas Mixtures in ZIF Membranes and ZIF/Polymer Composite Membranes. Ind. Eng. Chem. Res. 2012, 51 (7), 3091−3100. (c) Calero, S.; Jose Gutierrez-Sevillano, J.; Garcia-Perez, E. Effect of the Molecular Interactions on the Separation of Nonpolar Mixtures Using Cu-BTC Metal−Organic Framework. Microporous Mesoporous Mater. 2013, 165, 79−83. (d) Calero, S.; Martin-Calvo, A.; Hamad, S.; Garcia-Perez, E. On the Performance of Cu-BTC Metal Organic Framework for Carbon Tetrachloride Gas Removal. Chem. Commun. 2011, 47 (1), 508−510. (e) Castillo, J. M.; Vlugt, T. J. H.; Calero, S. Understanding Water Adsorption in Cu-BTC Metal−Organic Frameworks. J. Phys. Chem. C 2008, 112 (41), 15934−15939. (f) Chmelik, C.; Kaerger, J.; Wiebcke, M.; Caro, J.; van Baten, J. M.; Krishna, R. Adsorption and Diffusion of Alkanes in CuBTC Crystals Investigated Using Infra-red Microscopy and Molecular Simulations. Microporous Mesoporous Mater. 2009, 117 (1−2), 22−32. (g) Dickey, A. N.; Yazaydin, A. O.; Willis, R. R.; Snurr, R. Q. Screening CO2/N2 Selectivity in Metal− Organic Frameworks Using Monte Carlo Simulations and Ideal Adsorbed Solution Theory. Can. J. Chem. Eng. 2012, 90 (4), 825−832. (h) Erucar, I.; Keskin, S. Screening Metal−Organic Framework-Based Mixed-Matrix Membranes for CO2/CH4 Separations. Ind. Eng. Chem. Res. 2011, 50 (22), 12606−12616. (i) Erucar, I.; Keskin, S. Computational Screening of Metal Organic Frameworks for Mixed Matrix Membrane Applications. J. Membr. Sci. 2012, 407, 221−230. (j) Haldoupis, E.; Watanabe, T.; Nair, S.; Sholl, D. S. Quantifying Large Effects of Framework Flexibility on Diffusion in MOFs: CH4 and CO2 in ZIF-8. ChemPhysChem 2012, 13 (15), 3449−3452. (k) Hamon, L.; Heymans, N.; Llewellyn, P. L.; Guillerm, V.; Ghoufi, A.; Vaesen, S.; Maurin, G.; Serre, C.; De Weireld, G.; Pirngruber, G. D. Separation of CO2−CH4 Mixtures in the Mesoporous MIL-100(Cr) MOF: Experimental and Modelling Approaches. Dalton Trans. 2012, 41 (14), 4052−4059. (l) Huang, H.; Zhang, W.; Liu, D.; Liu, B.; Chen, G.; Zhong, C. Effect of Temperature on Gas Adsorption and Separation in ZIF-8: A Combined Experimental and Molecular Simulation Study. Chem. Eng. Sci. 2011, 66 (23), 6297−6305. (m) Lamia, N.; Jorge, M.; Granato, M. A.; Almeida Paz, F. A.; Chevreau, H.; Rodrigues, A. E. Adsorption of Propane, Propylene and Isobutane on a Metal−Organic Framework: Molecular Simulation and Experiment. Chem. Eng. Sci. 2009, 64 (14), 3246−3259. (n) Lyubchyk, A.; Esteves, I. A. A. C.; Cruz, F. J. A. L.; Mota, J. P. B. Experimental and Theoretical Studies of Supercritical Methane Adsorption in the MIL-53(Al) Metal Organic Framework. J. Phys. Chem. C 2011, 115 (42), 20628−20638. (o) Martin-Calvo, A.; Lahoz-Martin, F. D.; Calero, S. Understanding Carbon Monoxide Capture Using Metal− Organic Frameworks. J. Phys. Chem. C 2012, 116 (11), 6655−6663. (p) Petit, C.; Huang, L.; Jagiello, J.; Kenvin, J.; Gubbins, K. E.; Bandosz, T. J. Toward Understanding Reactive Adsorption of Ammonia on Cu-MOF/Graphite Oxide Nanocomposites. Langmuir 2011, 27 (21), 13043−13051. (q) Ramsahye, N. A.; Thuy Khuong, T.; Bourrelly, S.; Yang, Q.; Devic, T.; Maurin, G.; Horcajada, P.; Llewellyn, P. L.; Yot, P.; Serre, C.; Filinchuk, Y.; Fajula, F.; Férey, G.; Trens, P. Influence of the Organic Ligand Functionalization on the Breathing of the Porous Iron Terephthalate Metal Organic Framework Type Material upon Hydrocarbon Adsorption. J. Phys. Chem. C 2011, 115 (38), 18683−18695. (r) Rosenbach, N., Jr.; Ghoufi, A.; Deroche, I.; Llewellyn, P. L.; Devic, T.; Bourrelly, S.; Serre, C.; Férey, G.; Maurin, G. Adsorption of Light Hydrocarbons in the Flexible MIL53(Cr) and Rigid MIL-47(V) Metal−Organic Frameworks: A Combination of Molecular Simulations and Microcalorimetry/ Gravimetry Measurements. Phys. Chem. Chem. Phys. 2010, 12 (24), 6428−6437. (s) Salles, F.; Bourrelly, S.; Jobic, H.; Devic, T.; Guillerm, V.; Llewellyn, P.; Serre, C.; Férey, G.; Maurin, G. Molecular Insight into the Adsorption and Diffusion of Water in the Versatile

(4) (a) Li, J. R.; Kuppler, R. J.; Zhou, H. C. Selective Gas Adsorption and Separation in Metal−Organic Frameworks. Chem. Soc. Rev. 2009, 38 (5), 1477−1504. (b) Couck, S.; Denayer, J. F. M.; Baron, G. V.; Remy, T.; Gascon, J.; Kapteijn, F. An Amine-Functionalized MIL-53 Metal−Organic Framework with Large Separation Power for CO2 and CH4. J. Am. Chem. Soc. 2009, 131 (18), 6326−6327. (c) Gücüyener, C.; van den Bergh, J.; Gascon, J.; Kapteijn, F. Ethane/Ethene Separation Turned on Its Head: Selective Ethane Adsorption on the Metal−Organic Framework ZIF-7 through a Gate-Opening Mechanism. J. Am. Chem. Soc. 2010, 132 (50), 17704−17706. (5) Murray, L. J.; Dinca, M.; Long, J. R. Hydrogen Storage in Metal− Organic Frameworks. Chem. Soc. Rev. 2009, 38 (5), 1294−1314. (6) Juan-Alcañ iz, J.; Gascon, J.; Kapteijn, F. Metal−Organic Frameworks As Scaffolds for the Encapsulation of Active Species: State of the Art and Future Perspectives. J. Mater. Chem. 2012, 22 (20), 10102−10118. (7) Gascon, J.; Aktay, U.; Hernandez-Alonso, M. D.; van Klink, G. P. M.; Kapteijn, F. Amino-Based Metal−Organic Frameworks As Stable, Highly Active Basic Catalysts. J. Catal. 2009, 261 (1), 75−87. (8) (a) Férey, G.; Mellot-Draznieks, C.; Serre, C.; Millange, F.; Dutour, J.; Surblé, S.; Margiolaki, I. A Chromium Terephthalate-Based Solid with Unusually Large Pore Volumes and Surface Area. Science 2005, 309 (5743), 2040−2042. (b) Llewellyn, P. L.; Bourrelly, S.; Serre, C.; Vimont, A.; Daturi, M.; Hamon, L.; De Weireld, G.; Chang, J.-S.; Hong, D.-Y.; Kyu Hwang, Y.; Jhung, S. H.; Férey, G. High Uptakes of CO2 and CH4 in Mesoporous Metal−Organic Frameworks MIL-100 and MIL-101. Langmuir 2008, 24 (14), 7245−7250. (c) Juan-Alcañiz, J.; Ramos-Fernandez, E. V.; Lafont, U.; Gascon, J.; Kapteijn, F. Building MOF Bottles around Phosphotungstic Acid Ships: One-Pot Synthesis of Bi-Functional Polyoxometalate-MIL-101 Catalysts. J. Catal. 2010, 269 (1), 229−241. (9) (a) Akiyama, G.; Matsuda, R.; Sato, H.; Hori, A.; Takata, M.; Kitagawa, S. Effect of Functional Groups in MIL-101 on Water Sorption Behavior. Microporous Mesoporous Mater. 2012, 157, 89−93. (b) Jiang, D.; Keenan, L. L.; Burrows, A. D.; Edler, K. J. Synthesis and Post-Synthetic Modification of MIL-101(Cr)-NH2 via a Tandem Diazotisation Process. Chem. Commun. 2012, 48 (99), 12053−12055. (c) Juan-Alcañiz, J.; Goesten, M.; Martinez-Joaristi, A.; Stavitski, E.; Petukhov, A. V.; Gascon, J.; Kapteijn, F. Live Encapsulation of a Keggin Polyanion in NH2-MIL-101(Al) Observed by in Situ Time Resolved X-ray Scattering. Chem. Commun. 2011, 47 (30), 8578− 8580. (d) Klyamkin, S. N.; Berdonosova, E. A.; Kogan, E. V.; Kovalenko, K. A.; Dybtsev, D. N.; Fedin, V. P. Influence of MIL-101 Doping by Ionic Clusters on Hydrogen Storage Performance up to 1900 bar. Chem.Asian J. 2011, 6 (7), 1854−1859. (e) RamosFernandez, E. V.; Pieters, C.; van der Linden, B.; Juan-Alcañiz, J.; Serra-Crespo, P.; Verhoeven, M. W. G. M.; Niemantsverdriet, H.; Gascon, J.; Kapteijn, F. Highly Dispersed Platinum in Metal Organic Framework NH2-MIL-101(Al) Containing Phosphotungstic Acid Characterization and Catalytic Performance. J. Catal. 2012, 289, 42− 52. (f) Yang, C.-X.; Yan, X.-P. Metal−Organic Framework MIL101(Cr) for High-Performance Liquid Chromatographic Separation of Substituted Aromatics. Anal. Chem. 2011, 83 (18), 7144−7150. (g) Dhakshinamoorthy, A.; Alvaro, M.; Horcajada, P.; Gibson, E.; Vishnuvarthan, M.; Vimont, A.; Greneche, J.-M.; Serre, C.; Daturi, M.; Garcia, H. Comparison of Porous Iron Trimesates Basolite F300 and MIL-100(Fe) As Heterogeneous Catalysts for Lewis Acid and Oxidation Reactions: Roles of Structural Defects and Stability. ACS Catal. 2012, 2 (10), 2060−2065. (h) Juan-Alcañiz, J.; Goesten, M. G.; Ramos-Fernandez, E. V.; Gascon, J.; Kapteijn, F. Towards Efficient Polyoxometalate Encapsulation in MIL-100(Cr): Influence of Synthesis Conditions. New J. Chem. 2012, 36 (4), 977−987. (i) Soubeyrand-Lenoir, E.; Vagner, C.; Yoon, J. W.; Bazin, P.; Ragon, F.; Hwang, Y. K.; Serre, C.; Chang, J.-S.; Llewellyn, P. L. How Water Fosters a Remarkable 5-Fold Increase in Low-Pressure CO2 Uptake within Mesoporous MIL-100(Fe). J. Am. Chem. Soc. 2012, 134 (24), 10174−10181. (j) Wuttke, S.; Bazin, P.; Vimont, A.; Serre, C.; Seo, Y.-K.; Hwang, Y. K.; Chang, J.-S.; Férey, G.; Daturi, M. Discovering the Active Sites for C3 Separation in MIL-100(Fe) by 7620

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622

The Journal of Physical Chemistry C

Article

Hydrophilic/Hydrophobic Flexible MIL-53(Cr) MOF. J. Phys. Chem. C 2011, 115 (21), 10764−10776. (t) Salles, F.; Jobic, H.; Devic, T.; Llewellyn, P. L.; Serre, C.; Férey, G.; Maurin, G. Self and Transport Diffusivity of CO2 in the Metal−Organic Framework MIL-47(V) Explored by Quasi-elastic Neutron Scattering Experiments and Molecular Dynamics Simulations. ACS Nano 2010, 4 (1), 143−152. (u) Sarmiento-Perez, R. A.; Marleny Rodriguez-Albelo, L.; Gomez, A.; Autie-Perez, M.; Lewis, D. W.; Rabdel Ruiz-Salvador, A. Surprising Role of the BDC Organic Ligand in the Adsorption of CO2 by MOF-5. Microporous Mesoporous Mater. 2012, 163, 186−191. (v) Wu, D.; Wang, C.; Liu, B.; Liu, D.; Yang, Q.; Zhong, C. Large-Scale Computational Screening of Metal−Organic Frameworks for CH4/ H2 Separation. AIChE J. 2012, 58 (7), 2078−2084. (w) Yang, Q.; Jobic, H.; Salles, F.; Kolokolov, D.; Guillerm, V.; Serre, C.; Maurin, G. Probing the Dynamics of CO2 and CH4 within the Porous Zirconium Terephthalate UiO-66(Zr): A Synergic Combination of Neutron Scattering Measurements and Molecular Simulations. Chem.Eur. J. 2011, 17 (32), 8882−8889. (x) Yang, Q.; Vaesen, S.; Vishnuvarthan, M.; Ragon, F.; Serre, C.; Vimont, A.; Daturi, M.; De Weireld, G.; Maurin, G. Probing the Adsorption Performance of the Hybrid Porous MIL-68(Al): A Synergic Combination of Experimental and Modelling Tools. J. Mater. Chem. 2012, 22 (20), 10210−10220. (y) Zhuang, W.; Yuan, D.; Liu, D.; Zhong, C.; Li, J.-R.; Zhou, H.-C. Robust Metal− Organic Framework with an Octatopic Ligand for Gas Adsorption and Separation: Combined Characterization by Experiments and Molecular Simulation. Chem. Mater. 2012, 24 (1), 18−25. (z) Dubbeldam, D.; Krishna, R.; Calero, S.; Yazaydın, A. Ö . Computer-Assisted Screening of Ordered Crystalline Nanoporous Adsorbents for Separation of Alkane Isomers. Angew. Chem., Int. Ed. 2012, 51 (47), 11867−11871. (11) Chen, Y. F.; Babarao, R.; Sandler, S. I.; Jiang, J. W. Metal− Organic Framework MIL-101 for Adsorption and Effect of Terminal Water Molecules: From Quantum Mechanics to Molecular Simulation. Langmuir 2010, 26 (11), 8743−8750. (12) Poling, B. E.; Prausnitz, J. M.; O’Connel, J. P., The Properties of Gases and Liquids, 5th ed.; McGraw-Hill: New York, 2001. (13) (a) Duren, T.; Bae, Y.-S.; Snurr, R. Q. Using Molecular Simulation to Characterise Metal−Organic Frameworks for Adsorption Applications. Chem. Soc. Rev. 2009, 38 (5), 1237−1247. (b) 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− Organic Framework Materials. Ind. Eng. Chem. Res. 2008, 48 (5), 2355−2371. (c) Li, J.-R.; Ma, Y.; McCarthy, M. C.; Sculley, J.; Yu, J.; Jeong, H.-K.; Balbuena, P. B.; Zhou, H.-C. Carbon Dioxide CaptureRelated Gas Adsorption and Separation in Metal−Organic Frameworks. Coord. Chem. Rev. 2011, 255 (15−16), 1791−1823. (14) Mayo, S. L.; Olafson, B. D.; Goddard, W. A. DREIDING: A Generic Force Field for Molecular Simulations. J. Phys. Chem. 1990, 94 (26), 8897−8909. (15) Rappe, A. K.; Casewit, C. J.; Colwell, K. S.; Goddard, W. A.; Skiff, W. M. UFF, a Full Periodic Table Force Field for Molecular Mechanics and Molecular Dynamics Simulations. J. Am. Chem. Soc. 1992, 114 (25), 10024−10035. (16) Yazaydin, A. Ö . Personal communication, 2011. (17) Delley, B. From Molecules to Solids with the DMol3 Approach. J. Phys. Chem. 2000, 113 (18), 7756−7764. (18) Lai, S. M.; Ng, C. P.; Martin-Aranda, R.; Yeung, K. L. Knoevenagel Condensation Reaction in Zeolite Membrane Microreactor. Microporous Mesoporous Mater. 2003, 66 (2−3), 239−252. (19) Perdew, J. P. Generalized Gradient Approximations for Exchange and Correlation: A Look Backward and Forward. Physica B: Phys. Condens. Matter 1991, 172 (1−2), 1−6. (20) Chen, B.; Potoff, J. J.; Siepmann, J. I. Monte Carlo Calculations for Alcohols and Their Mixtures with Alkanes. Transferable Potentials for Phase Equilibria. 5. United-Atom Description of Primary, Secondary, and Tertiary Alcohols. J. Phys. Chem. B 2001, 105 (15), 3093−3104.

(21) (a) Kuhn, J.; Castillo-Sanchez, J. M.; Gascon, J.; Calero, S.; Dubbeldam, D.; Vlugt, T. J. H.; Kapteijn, F.; Gross, J. Adsorption and Diffusion of Water, Methanol, and Ethanol in All-Silica DD3R: Experiments and Simulation. J. Phys. Chem. C 2009, 113 (32), 14290− 14301. (b) Nalaparaju, A.; Zhao, X. S.; Jiang, J. W. Molecular Understanding for the Adsorption of Water and Alcohols in Hydrophilic and Hydrophobic Zeolitic Metal−Organic Frameworks. J. Phys. Chem. C 2010, 114 (26), 11542−11550. (c) Lively, R. P.; Dose, M. E.; Thompson, J. A.; McCool, B. A.; Chance, R. R.; Koros, W. J. Ethanol and Water Adsorption in Methanol-Derived ZIF-71. Chem. Commun. 2011, 47 (30), 8667−8669. (d) Zang, J.; Chempath, S.; Konduri, S.; Nair, S.; Sholl, D. S. Flexibility of Ordered Surface Hydroxyls Influences the Adsorption of Molecules in Single-Walled Aluminosilicate Nanotubes. J. Phys. Chem. Lett. 2010, 1 (8), 1235− 1240. (22) Guillot, B. A Reappraisal of What We Have Learnt during Three Decades of Computer Simulations on Water. J. Mol. Liq. 2002, 101, 219−260. (23) Rick, S. W. A Reoptimization of the Five-Site Water Potential (TIP5P) for Use with Ewald Sums. J. Chem. Phys. 2004, 120 (13), 6085−6093. (24) Castillo, J. M.; Vlugt, T. J. H.; Calero, S. Understanding Water Adsorption in Cu−BTC Metal−Organic Frameworks. J. Phys. Chem. C 2008, 112 (41), 15934−15939. (25) Berendsen, H. J. C.; Grigera, J. R.; Straatsma, T. P. The Missing Term in Effective Pair Potentials. J. Phys. Chem. 1987, 91 (24), 6269− 6271. (26) (a) Henninger, S.; Schmidt, F.; Henning, H. M. Characterisation and Improvement of Sorption Materials with Molecular Modeling for the Use in Heat Transformation Applications. Adsorption 2011, 1−11. (b) Csányi, É.; Ható, Z.; Kristóf, T. Molecular Simulation of Water Removal from Simple Gases with Zeolite NaA. J. Mol. Model. 2012, 18 (6), 2349−2356. (27) Horn, H. W.; Swope, W. C.; Pitera, J. W.; Madura, J. D.; Dick, T. J.; Hura, G. L.; Head-Gordon, T. Development of an Improved Four-Site Water Model for Biomolecular Simulations: TIP4P-Ew. J. Phys. Chem. 2004, 120 (20), 9665−9678. (28) Paranthaman, S.; Coudert, F.-X.; Fuchs, A. H. Water Adsorption in Hydrophobic MOF Channels. Phys. Chem. Chem. Phys. 2010, 12 (28), 8123−8129. (29) Ewald, P. P. Die Berechnung optischer und elektrostatischer Gitterpotentiale. Ann. Phys. 1921, 369 (3), 253−287. (30) Myers, A. L.; Monson, P. A. Adsorption in Porous Materials at High Pressure: Theory and Experiment. Langmuir 2002, 18 (26), 10261−10273. (31) Vlugt, T. J. H.; García-Pérez, E.; Dubbeldam, D.; Ban, S.; Calero, S. Computing the Heat of Adsorption using Molecular Simulations: The Effect of Strong Coulombic Interactions. J. Chem. Theory Comput. 2008, 4 (7), 1107−1118. (32) (a) García-Pérez, E.; Gascón, J.; Morales-Flórez, V. c.; Castillo, J. M.; Kapteijn, F.; Calero, S. Identification of Adsorption Sites in CuBTC by Experimentation and Molecular Simulation. Langmuir 2009, 25 (3), 1725−1731. (b) 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 (1−2), 152−161. (c) Surblé, S.; Millange, F.; Serre, C.; Düren, T.; Latroche, M.; Bourrelly, S.; Llewellyn, P. L.; Férey, G. Synthesis of MIL-102, a Chromium Carboxylate Metal−Organic Framework, with Gas Sorption Analysis. J. Am. Chem. Soc. 2006, 128 (46), 14889−14896. (33) Rouquerol, F.; Rouquerol, J.; Sing, K. Adsorption by Powders & Porous Solids; Academic Press: London, 1999. (34) Hong, D.-Y.; Hwang, Y. K.; Serre, C.; Férey, G.; Chang, J.-S. Porous Chromium Terephthalate MIL-101 with Coordinatively Unsaturated Sites: Surface Functionalization, Encapsulation, Sorption and Catalysis. Adv. Funct. Mater. 2009, 19 (10), 1537−1552. (35) Chang, J.-S. Porous Metal(III) Carboxylates as Multifunctional Adsorbents and Catalytic Materials. Presented at the iCeMS-ERATO Symposium, Kyoto, Japan, Jul 26, 2012. 7621

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622

The Journal of Physical Chemistry C

Article

(36) Akiyama, G.; Matsuda, R.; Kitagawa, S. Highly Porous and Stable Coordination Polymers as Water Sorption Materials. Chem. Lett. 2010, 39 (4), 360−361. (37) Ehrenmann, J.; Henninger, S. K.; Janiak, C. Water Adsorption Characteristics of MIL-101 for Heat-Transformation Applications of MOFs. Eur. J. Inorg. Chem. 2011, 2011 (4), 471−474. (38) Kusgens, P.; Rose, M.; Senkovska, I.; Frode, H.; Henschel, A.; Siegle, S.; Kaskel, S. Characterization of Metal−Organic Frameworks by Water Adsorption. Microporous Mesoporous Mater. 2009, 120 (3), 325−330. (39) Green, D. W.; Perry, R. H. Perry’s Chemical Engineers’ Handbook, 8th ed.; McGraw-Hill: New York, 2008. (40) Guideline on the Use of Fundamental Physical Constants and Basic Constants of Water; International Association for the Properties of Water and Steam (IAPWS): Gaithersburg, MD, 2001.

7622

dx.doi.org/10.1021/jp3128077 | J. Phys. Chem. C 2013, 117, 7613−7622