Ab-Initio Study of the Adsorption and Separation of NOx and SOx

View Sections. ACS2GO © 2018. ← → → ←. loading. To add this web app to the home screen open the browser option menu and tap on Add to hom...
0 downloads 0 Views 4MB Size
Download PDF
ARTICLE pubs.acs.org/JPCC

Ab-Initio Study of the Adsorption and Separation of NOx and SOx Gases in Functionalized IRMOF Ligands Konstantinos A. Fioretos, George M. Psofogiannakis, and George E. Froudakis* Department of Chemistry, University of Crete, PO Box 2208, Voutes, Heraklion Crete, 710 03 Greece ABSTRACT: The interaction of NOx, SOx, and other gases with benzene and the monosubstituted benzene derivatives, aniline, phenol, and toluene, were studied by the MP2 postHartree Fock method. The aromatic molecules were used as models for metal organic frameworks’ ligands in order to examine the effect of the substituents on the adsorption and separation of these gases. It was found that polar substituents, as exemplified by the case of the amino ( NH2) group, increase the contribution of electrostatic dipole dipole interactions with polar molecules (particularly NO2 and SO2). This leads to stronger interaction energies and altered minimum-energy structures. Furthermore, we calculated the interaction energy of a variety of gases with the amino-substituted ring in order to examine the possibilities for gas-separation applications. The interaction energies increased in the order H2 < N2 < CH4 < NO < NO2 < CO2 < SO2 < SO3. The forces responsible for the adsorption of these gases were studied with the aid of charge analyses and electrostatic potential maps.

’ INTRODUCTION Metal organic frameworks1,2 (MOFs) are currently under intense study for gas storage, gas separation and purification, catalysis, and medical and biological applications.3 5 The adsorptive separation of NOx (NO, NO2) and SOx (SO2, SO3) gases from industrial gas mixtures has been clearly identified as a potential application of MOFs.6 Purification of flue-gas streams by removal of NOx and SOx gases is crucially important due to the well-known adverse environmental and health effects of the emission of these gases in the atmosphere. Nevertheless, there has been no systematic investigation of the storage and separation properties of MOFs with respect to the NOx and SOx gases. Another possible application of nitric oxide (NO) storage in MOFs has been identified in the delivery of NO for medical therapies, where the MOF can act as a nontoxic gas carrier.7 Several MOFs have already been identified that possess excellent characteristics for NO delivery for biomedical applications.5,8 With these applications in mind, we have set out a project to systematically identify MOFs that can potentially be useful in the adsorptive storage of NOx and SOx gases as well as in their separation from other industrial gases that are typically present in industrial gas streams, such as N 2 , CO 2 , and CH 4 . Because of the polar nature of NO, NO2, and SO2, one strategy that we decided to study is the utilization of functional groups in the MOF ligands in order to enhance the electrostatic interaction between the adsorbing molecules and the MOF framework. A variety of functional groups can be present in the final structure of metal organic frameworks. Some functionalities can be inserted in the MOF structure as a result of the solvothermal complexation of the metal ions with organic linkers that contain the functional groups prior to synthesis.9 Nevertheless, a large variety of functional groups can be inserted into r 2011 American Chemical Society

the organic linker after the synthesis by utilizing developed postsynthetic modification methods.10,11 Through these chemical functionalization routes, synthesis of MOFs has now reached the point of being able to introduce a variety of functional groups on the same MOF linkers. For example, Deng et al.12 synthesized such complex MOFs containing 1,4-benzenedicarboxylate organic linkers modified with NH2, Br, (Cl)2, NO2, (CH3)2, C4H4, (OC3H5)2, and (OC7H7)2 groups. The functionalization of the ligands has proved to be an efficient strategy for enhancing the storage and separation capacities of MOFs for a variety of gases due to enhanced interaction of the functional groups with the adsorbing molecules.9,13 Many successful attempts have been made to model computationally the adsorption of gases in MOFs. These modeling attempts usually proceed hierarchically: At the most fundamental level, the interaction of one or more adsorbing molecules with the MOF structure is studied using ab initio methodology. At a higher level, force fields can be utilized in order to quantify the gas adsorption capacity of MOFs at selected thermodynamic conditions, often via Monte Carlo simulations. Molecular dynamic simulations are also being employed to study the diffusion and separation of gases within the MOF pores. These methodologies have become routine in the study of gas adsorption in MOFs. Nevertheless, quite justifiably, the greatest majority of these studies have concentrated on the interaction of MOFs with H2, CO2, and CH4. However, systematic computational investigation of the interaction of MOFs with NOx and SOx gases is also crucial because of the potentially important industrial and biological applications in the storage and separation of these Received: September 1, 2011 Revised: November 7, 2011 Published: November 08, 2011 24906

dx.doi.org/10.1021/jp2084378 | J. Phys. Chem. C 2011, 115, 24906–24914

The Journal of Physical Chemistry C gases. As a first step toward this end, we have used ab initio methodology to study the effect of various functional groups of IRMOF ligands in the adsorption and separation of these gases. In the present work, we used ab initio methods in order to systematically accomplish the following objectives: (1) to study the interaction between a variety of MOF ligands and the molecules NO and NO2, (2) to elucidate the effects of the ligand functionalities on the adsorption energies and structures of NO and NO2, and (3) to study the energetics of adsorption of the SOx gases (SO2, SO3) and various other gases (H2, CH4, N2, CO2) on amino-functionalized MOF ligands in order to discover the possibilities for NOx and SOx separation and purification applications. As a first step in our quest, we were only interested in studying the interaction of these gases with the ligands and not with the full material framework. This choice is justified by our primary objective, which was to study the influence of the functional groups on the adsorption energies in order to test the hypothesis of enhanced interactions with the polar molecules. As we do not attempt to model any particular MOF, we chose monosubstituted benzene molecules (aniline, toluene, phenol) as models for the ligands. The effect of NH2, OH, and CH3 substituents on the aromatic ligand rings was examined, based on the fact that MOFs that contain these substituents have already been synthesized. A similar approach was previously followed in the study of interactions of CO2 with substituted benzenes.14,15

’ COMPUTATIONAL METHODS The computational work consisted mainly of geometry optimizations, calculation of interaction energies, calculation of the basis-set superposition error, analysis of atomic charges, and generation of electrostatic potential maps. In all optimizations, many different starting geometries were attempted as in many cases the interacting molecules have several local minima. The MP2 method was used in the optimizations and energy calculations.16,17 The choice of post-Hartree Fock Møller Plesset perturbation method was based on its ability to include dispersion interactions systematically with no extreme computational cost. The dispersion interactions are dominant in nearly all cases where weak physisorption is involved. In contrast, DFT methods do not treat dispersion interactions adequately. The basis sets used for selected interactions were the 6-31++G** and 6-311++G** in order to include the effects of diffuse and polarization functions on split-valence basis sets.18 We verified that the two basis sets give very similar results. The interaction energies were calculated as the energy difference between the dimer composed of the interacting monomers and the sum of energies of the separated monomers. All results were corrected for basis set superposition error (BSSE) using the counterpoise method as applied to the optimized structures. The BSSE correction was found to be very significant in the calculations and reduced the uncorrected interaction energies by more than 50% in most cases. Atomic charges were calculated using the CHELPG method19 on the MP2-optimized structures. This method works by optimizing atomic charges in order to reproduce the electrostatic potential of the molecule. As such, the fitted atomic charges are particularly effective in the qualitative description of electrostatic interactions among interacting molecules. Electrostatic potential maps were generated from the calculated electron densities. The maps are colored, identifying

ARTICLE

regions of low and high electrostatic potential, corresponding to electron-rich and electron-poor regions, respectively. All calculations were performed using the Gaussian03 program.20

’ RESULTS AND DISCUSSION Interaction of NO and NO2 with Benzene, Toluene, Phenol, and Aniline. In the first part of this work, the interaction of

the NO and NO2 molecules with the aromatic molecules benzene (C6H6), toluene (C6H5 CH3), phenol (C6H5 OH), and aniline (C6H5 NH2) was systematically examined. These molecules served as a simplified model for IRMOF linkers and the purpose was to examine the effect of the methyl, hydroxyl, and amino substituent on the interaction energy of the aromatic ring with the NOx gases. Figure 1 shows the MP2-optimized structures of the interacting molecules with NO and NO2 as well as the calculated atomic charges, generated with the CHELPG method. In order to explain the interactions qualitatively, it is instructive to compare the calculated charges shown in Figure 2 for benzene, toluene, phenol, aniline, NO, and NO2. The interaction of NO with benzene is typical of dispersion interaction. The NO molecule lies at a distance of ∼3 Å away from the ring plane. The interaction of NO with toluene and phenol is similar in nature to the case of benzene as evidenced by the similar structures in Figure 1a c and is dominated by dispersion forces. NO adsorption on aniline results in a different structure (Figure 1d). The interacting geometry is clearly adopted due to the contribution of electrostatic dipole dipole interaction between the polar NO molecule (Figure 2e) and the polarized C N bond of aniline (Figure 2d). The structure in Figure 1d shows that the NO molecule is positioned such that the dipole moment vector is parallel to the C N bond and with reverse direction of polarity, resulting in dipole dipole electrostatic attraction. Although both toluene and phenol also have a polar bond to carbon (Figure 2b,c), the smaller dispersion interaction and larger exchange repulsion (with the protruding CH3 group) do not favor a similar geometry in the case of NO adsorption. The interaction of NO2 with the aromatic molecules is shown in structures 1e to 1h. The interaction with benzene and toluene is resulting in similar structures (Figure 1e,f). The interaction of NO2 with toluene is enhanced by interaction with the one of the methyl hydrogen atoms. It is shown in Figure 1f that one of the O atoms is coordinated toward one of the H atoms of the methyl group that is 2.89 Å away. This H atom has additional induced positive charge in comparison to the other H atoms. Thus, in addition to dispersion interactions, an electrostatic component to the interaction increases the binding. For both phenol and aniline (Figure 1g,h) the location of the molecule is determined by the increased contribution of the dipole dipole type of interaction with the C O and C N bonds. As NO2 is a stronger dipole than NO (Figure 2e,f), NO2 adopts this structure when it interacts either with phenol or aniline, whereas NO adopts the structure only with aniline (Figure 1d). The calculated interaction energies of the aromatic molecules with NO are shown in Figure 3. These were calculated at the MP2/6-311++G** level of theory and were corrected for BSSE. The results indicate that only the CH3 substituent resulted in an increase of the magnitude of interaction with NO. The methyl group of toluene increases the binding of NO by 0.32 kcal/mol (1.34 kJ/mol). This enhancement is in agreement with the additional polarization of the NO molecule (Figure 1b) caused 24907

dx.doi.org/10.1021/jp2084378 |J. Phys. Chem. C 2011, 115, 24906–24914

The Journal of Physical Chemistry C

ARTICLE

Figure 1. MP2-optimized structures and atomic charges (CHELPG) of benzene, toluene, phenol, and aniline interacting with NO and NO2. Green = N, red = O, and blue = H.

24908

dx.doi.org/10.1021/jp2084378 |J. Phys. Chem. C 2011, 115, 24906–24914

The Journal of Physical Chemistry C

ARTICLE

Figure 2. CHELPG atomic charges for benzene, toluene, phenol, aniline, NO, and NO2.

Figure 3. Calculated Interaction energies with MP2/6-311++G** between NO and benzene, toluene, phenol, and aniline.

Figure 4. Calculated interaction energies with MP2/6-311++G** between NO2 and benzene, toluene, phenol, and aniline.

by the methyl group. The other substituents have a much smaller effect. Evidently, for aniline interacting with NO, the location of the NO molecule is such that there is an additional electrostatic contribution that compensates the loss of dispersion interaction associated with smaller interaction with the π-system, but the net effect is not altering the interaction energy. The calculated BSSE-corrected interaction energies of the aromatic molecules with NO2 at the MP2/6-311++G** level are

shown in Figure 4. The results indicate that the incorporation of the CH3 and NH2 substituents in the structure increases the interaction energy of the molecule with NO2 by 0.35 kcal/mol (1.36 kJ/mol) and 0.59 kcal/mol (2.47 kJ/mol), respectively, compared to the benzene ring. Thus, the interaction with the NH2 group enhances the binding by ∼35%. This increase is attributed to the electrostatic interactions at the specific binding location. 24909

dx.doi.org/10.1021/jp2084378 |J. Phys. Chem. C 2011, 115, 24906–24914

The Journal of Physical Chemistry C From a practical standpoint, there is an appreciable increase in the interaction energies with NOx gases that can be caused by functionalization of the aromatic ring with selected functional groups. These results could be valuable in the following cases: (a) For NO2, the incorporation of CH3 and NH2 substituents in the structure increases the strength of interaction appreciably in comparison to the nonsubstituted ring. For NO, only the incorporation of the CH3 substituent in the structure increases the strength appreciably as well. Thus, insertion of these functionalities could be useful in adsorption storage applications of NO and NO2 in MOFs. The incorporation of more than one substituent in the same aromatic ring could further enhance the interaction, as more adsorbed molecules would experience the increased specific dipole dipole interaction simultaneously. (b) The NH2 functionality has a much greater effect on the adsorption of NO2 than on the adsorption of NO due to the greater electrostatic attraction with the more polar NO2 molecule. Figures 3 and 4 show that the benzene NO interaction energy is 1.49 kcal/mol, whereas the benzene NO2 interaction energy is 1.67 kcal/mol. The interaction energy difference (0.18 kcal/mol = 0.75 kJ/mol) does not allow for selective adsorption eparation of NO/NO2 mixtures. When aniline is used instead, the aniline NO interaction energy is 1.46 kcal/mol, whereas the aniline NO2 interaction energy is 2.26 kcal/mol. The 0.8 kcal/ mol (3.35 kJ/mol) interaction energy difference could be useful for effective separation of NO/NO2 mixtures by selective adsorption of NO2 in a MOF structure that contains NH2 groups. The increased interaction strength of aniline with NO2, compared to NO, is a consequence of increased electrostatic dipole dipole interaction with the more polar NO2 molecule. (c) The above conclusion would be expected to be more general. In particular, the amino group was shown to specifically enhance the interaction with the more polar molecule NO2 due to dipole dipole specific interactions. Thus, the amino group would be helpful for the separation of strongly polarized molecules from nonpolar molecules or less polar molecules that would experience a smaller electrostatic contribution due to the amino group. Interactions of H2, CH4, N2, CO2, NO, NO2, SO2, and SO3 with Aniline. The results of the previous analysis suggested that the amino substituent acts to increase the interaction strength with NO2 due to a specific electrostatic interaction. This result prompted us to examine systematically the interaction of an amino-substituted benzene ring (aniline) with a variety of gases, both polar and nonpolar. Aniline served as a model for NH2functionalized MOF linkers. The choice of the amino group as the substituent, partly justified by its effect on increasing the interaction with NO2 and partly by the existence of experimental NH2-functionalization strategies of MOFs, also served as an example of the effect of an electron-donating polar substituent. This study was directed toward gas separation applications in mind. In particular, the goal was to obtain an estimate of the relative interaction strengths of aniline with a variety of gases. Large differences in interaction strengths among the various gases with aniline would indicate the possibility of adsorptionbased separation within the framework of NH2-functionalized MOFs. The series of molecules contained many gases that appear in various common industrial separations. The optimizations and charge analysis were performed at the MP2/6-31++G** level of theory, corrected for BSSE, that was judged to be sufficiently accurate.

ARTICLE

Figure 5 shows the minimum-energy structures of H2, CH4, N2, CO2, SO2, and SO3 interacting with aniline (NO and NO2 are shown in Figure 1). The pictures also show the atomic charges as calculated with the CHELPG method on the MP2optimized structures. The preferred configurations of H2, CH4, and N2 are on top of the ring, slightly toward the substituted side of the ring. CO2 prefers a location above the N atom of the amino group such that there exists an electrostatic interaction between the electron-deficient C atom of CO2 and the lone pair of the N atom of the amino group. SO2 is directed with the dipole moment vectors parallel to the C N bond of aniline, in a binding mode similar to NO and NO2 (Figure 1). SO3 is directed such that a weak S N chemical bond is formed. The qualitative aspects of these interactions will be discussed below. The interaction energies between aniline and the various molecules are shown in Figure 6 in the form of a bar chart. The nonpolar gases H2, CH4, and N2 interact weakly with aniline, while the strength of interaction of polar gases NO, NO2, and SO2 follows the order of the magnitude of their dipole moment. The strength of interaction of aniline with SO3 is by far the greatest. Attributes of the minimum-energy structures and relative strength of interactions can be qualitatively explained. H2, CH4, and N2 do not have permanent dipole moments. It is therefore expected that the contribution of dispersion energy in the interaction of these molecules with aniline will be the determining factor. In fact, for the case of H2, CH4, and N2 the charge analysis indicated no significant polarization and the location of the molecules is what would be expected on the basis of maximization of the dispersion interactions with the aromatic ring of aniline. For those interactions where the dispersion contribution predominates it is known that the strength is increasing with increasing the molecular mass of the interacting molecules. Large atoms and molecules tend to have greater polarizability so that it is easier to induce momentary dipoles and the larger number of electrons increases the dispersion interaction between the electron clouds. A rough indicator of the relative extent of dispersion forces can be given by counting the number of electrons. Figure 7 shows a plot of the MP2/6-31++G**-calculated interaction energy of the nonpolar molecules with aniline versus the total number of electrons of the molecules. The result indicates an approximately linear correlation that proves the expected effect of molecular size on the interaction energy of the molecules with aniline. The minimum-energy structure and strength of interaction of the polar molecules NO, NO2, and SO2 are quite different from the nonpolar molecules. For all three polar molecules, NO, NO2, and SO2, the preferred location that maximizes the attractive strength is with the molecular dipole moment vector positioned parallel to the C N bond of aniline (Figures 1d, 1h, and 5e). A dipole dipole type of interaction is evident when the NO, NO2, and SO2 molecules are located in parallel and with reverse dipole vector direction to the C N bond. The charges in Figures 1d, 1h, and 5e clearly indicate the dipole dipole electrostatic nature of the interaction. The relative strength of the interaction is expected to correlate to an indicator of the polarity of the molecule, such as the dipole moment. Figure 8 shows a plot of the MP2/6-31++G**-calculated interaction energy of the polar molecules with aniline versus the dipole moment of the molecule, calculated with the same level of theory. The result indicates an approximately linear correlation that proves the expected 24910

dx.doi.org/10.1021/jp2084378 |J. Phys. Chem. C 2011, 115, 24906–24914

The Journal of Physical Chemistry C

ARTICLE

Figure 5. MP2/6-31++G** optimized structures of a variety of gases (H2, CH4, N2, CO2, SO2, SO3) interacting with aniline. Atomic charges as calculated with the CHELPG method are also shown.

Figure 6. The bar chart depicts graphically the interaction energies between aniline and H2, CH4, N2, NO, NO2, CO2, SO2, and SO3, calculated at the MP2/6-31++G** level of theory.

relative strengths. It does not, of course, prove that a strictly linear relationship exists. The interaction between aniline and SO3 is of altogether different nature. The large strength of binding (14.7 kcal/mol) points to orbital interactions and chemical bonding. This is also supported by the charge analysis that shows a net charge transfer

Figure 7. MP2-calculated energy of interaction between aniline and H2, CH4, and N2 versus the total number of electrons in H2, CH4, and N2.

of 0.33 electrons from aniline to the SO3 group as well as the proximity of the N and S atoms (2.08 Å), suggesting a weak chemical bonding. In fact, the structure resembles the zwitterion in the crystal of sulfamic acid (NH3SO3). The bond formation is based on the acidity of SO3 that is electrophilic and accepts the lone-pair electrons of the amino group. In this structure, the charge analysis indicates that there is electron donation, from the 24911

dx.doi.org/10.1021/jp2084378 |J. Phys. Chem. C 2011, 115, 24906–24914

The Journal of Physical Chemistry C

ARTICLE

Figure 10. Electrostatic potential surfaces of NO2 (left) and SO2 (right). Figure 8. MP2-calculated energy of interaction between aniline and NO, NO2, and SO2 versus the calculated dipole moment of the three molecules.

Figure 11. Electrostatic potential surface of the optimized structure of SO2 interacting with aniline.

Figure 9. Electrostatic potential map of aniline. Regions in red are electron-rich, whereas blue regions are electron-deficient.

N atom of the NH3 group to the SO3 group, which is largely responsible for the bonding. From a practical viewpoint, it is evident from the relative magnitude of the interactions that the amino group attached to an aromatic ring within a metal organic framework has the potential to aid in separation of a variety of gas mixtures. In particular, it is evident from Figure 6 that the aromatic ring interacts quite stronger with SOx gases than with H2, CH4, and N2. The same is true for NOx and CO2, albeit to a lesser extent. Figure 6 can act as a rough guide to the usefulness of the aminosubstituted ring in the highly diverse range of separation applications using MOFs that one could imagine. Of course, the diffusivity of these gases in the pores of any particular MOF would be also additionally affected by the details of the pore structure, aperture windows, and the collective interactions of the diffusing molecules with the pore walls. Nevertheless, the most obvious conclusion that this analysis affords is summarized in Figure 6. In particular, the amino-substituted ring shows great potential for desulfurization (DeSOx) applications, and further studies are warranted. Qualitative Characteristics of the Interactions. Further information on the qualitative aspects of the interaction of aniline with NO2, CO2, SO2, and SO3 were obtained by constructing representative electrostatic potential surfaces in order to rationalize electrostatic contributions of the minimum-energy structures.

Figure 12. Electrostatic potential surfaces of aniline (left), CO2 (middle), and aniline interacting with CO2 (right).

In each case, the electrostatic potential, as calculated from the multipole molecular moments, was mapped onto the molecular electron density, and the results were visualized as colored maps. The calculations were performed at the MP2/6-31++G** level. The resulting map is color-coded such that electron-rich regions are shown red and electron-poor regions are shown as blue. The electrostatic potential surface corresponding to aniline is shown in Figure 9. The map shows the nature of the dipole of the C N bond. The electron-rich region is situated close to the N atom and the electron-deficient region closer to the C atom of the C N bond. The electron-rich region of the lone pair of the N atom extends further out above the molecular plane on the opposite side of the N H bonds (H atoms are facing down in Figure 9). The H atoms of the NH2 group are electrondeficient. Figure 10 shows the electrostatic potential surfaces of NO2 and SO2. These molecules have permament dipole moments. Greater electron density is localized around the O atoms in these molecules. Examination of the electrostatic potential maps of 24912

dx.doi.org/10.1021/jp2084378 |J. Phys. Chem. C 2011, 115, 24906–24914

The Journal of Physical Chemistry C

ARTICLE

Figure 13. Left: HOMO of aniline. Center: LUMO of SO3. Right: electrostatic potential map of aniline interacting with SO3.

aniline and NO2, SO2 reveals that the observed molecular orientation that maximizes dipole dipole electrostatic interactions is indeed expected. Figure 11 shows that the SO2 molecule is oriented above the C N bond with the electron-deficient region around the S atom interacting with the electron-rich region above the N atom of the amino group. Simultaneously, the electron-rich O atoms can interact with the electron-deficient region around the C atom of the C N bond. This mode of binding is very similar for NO2 and SO2, the difference being only in the strength of the interaction. The optimized structure and electrostatic potential map of CO2 interacting with aniline are shown in Figure 12 (right). Figure 12 (center) shows the electrostatic potential of CO2, where the electronegative O atoms are electron-rich and the C atom is electron-deficient, so that the molecule possesses quadrupole moment. The electron-deficient C atom of CO2 interacts with the electron-rich area of the lone pair electrons of the amino group’s N atom. This mode of interaction between aniline and CO2 has been previously discussed and quantified via DFT,15 but the electrostatic potential maps presented here depict accurately the nature of the interaction. Finally, the interaction of aniline with SO3 results in chemical bond formation between the N and S atoms as evidenced by the short bond length and electron donation to the SO3 group. Figure 13 (right) shows the electrostatic potential surface of the interacting molecules, clearly showing electron density in the bonding region, concentrated mainly on the sides of this region of this region. The chemical nature of the bond can be rationalized by examining the frontier orbitals of the two molecules. Figure 13 (left) depicts the highest occupied molecular orbital (HOMO) of aniline, calculated at the MP2/6-31++G** level, while Figure 13 (center) depicts the lowest unoccupied molecular orbital (LUMO) of SO3. The front lobes of the HOMO of aniline that extend above the molecular plane have the same symmetry as the lobes of the LUMO of SO3 that are situated far below (and above) the molecule’s surface in Figure 13 (center). Constructive interference among these orbitals can be expected. Electrons from the HOMO of aniline can be promoted to the new molecular bonding orbitals.

’ CONCLUSIONS In summary, it was found in this work that the adsorption and separation of NOx and SOx gases in metal organic frameworks can be enhanced by the incorporation of polar substituents in the MOF ligands. The amino ( NH2) substituent can be particularly beneficial. The interaction of strongly polar molecules, such as NO2 and SO2, with the amino-substituted aromatic rings is characterized by the contribution of electrostatic dipole dipole forces that result in enhanced adsorption strength. In these cases,

the interaction energy at the optimized structure correlates with the dipole moment of the molecule. Thus, amino groups that are incorporated in the MOF structure can induce selective adsorption of polar molecules, such as NO2 and SO2, from gas mixtures with nonpolar gases, such as H2, CH4, and N2. Furthermore, they can aid in the separation of mixtures of polar molecules, as the amino group will selectively enhance the adsorption strength of the more polar molecules. For example, in a mixture of NO and NO2, the amino groups have the effect of enhancing the interaction with NO2 selectively. Furthermore, a strong interaction of NH2substituted rings with SO2 was predicted (4.6 kcal/mol). Finally, SO3 forms a weak chemical bond with NH2-substituted rings, characterized by donation of electronic density from the amino group to the sulfur trioxide molecule. Aniline binds SO3 with 14.7 kcal/mol bond strength.

’ AUTHOR INFORMATION Corresponding Author

*Phone: (+302810) 545055. Fax: (+302810) 545001. E-mail: [email protected].

’ ACKNOWLEDGMENT Financial support for G.P. from a European Union MarieCurie International Reintegration Grant (IRG) is greatly appreciated. ’ REFERENCES (1) Yaghi, O. M.; Li, G.; Li, H. Nature 1995, 378, 703. (2) Li, H.; Eddaoudi, M.; O’Keefe, M.; Yaghi, O. M. Nature 1999, 402, 276. (3) Czaja, A. U.; Trukhanand, N.; M€uller, U. Chem. Soc. Rev. 2009, 38, 1284. (4) Ma, S.; Zhou, H.-C. Chem. Commun. 2010, 46, 44. (5) McKinlay, A. C.; Morris, R. E.; Horcajada, P.; Ferey, G.; Gref, R.; Couvreur, P.; Serre, C. Angew. Chem., Int. Ed. 2010, 49, 6260. (6) Li, J. R.; Kuppler, R. J.; Zhou, H-C. Chem. Soc. Rev. 2009, 38, 1477. (7) Keefer, L. K. Nature Mater. 2003, 2, 357. (8) McKinlay, A. C.; Xiao, B.; Wragg, D. S.; Wheatley, P. S.; Megson, I. L.; Morris, R. E. J. Am. Chem. Soc. 2008, 130, 10440. (9) Zhao, Y.; Wu, H.; Emge, T. J.; Gong, Q.; Nijem, N.; Chabal, Y. J.; Kong, L.; Langreth, D. C.; Liu, H.; Zeng, H.; Li, J. Chem.—Eur. J. 2011, 17, 5101. (10) Tanabe, K. K.; Wang, Z.; Cohen, M. S. J. Am. Chem. Soc. 2008, 130, 8508. (11) Cohen, S. M. Chem. Sci. 2010, 1, 32. (12) Deng, H.; Doonan, C. J.; Furukawa, H.; Ferreira, R. B.; Towne, J.; Knobler, C. B.; Wang, B.; Yaghi, O. M. Science 2010, 327, 846. (13) Morris, W.; Leung, B.; Furukawa, H.; Yaghi, O. K.; He, N.; Hayashi, H.; Houndonoungo, Y.; Asta, M.; Laird, B. B.; Yaghi, O. M. J. Am. Chem. Soc. 2010, 132, 11006. 24913

dx.doi.org/10.1021/jp2084378 |J. Phys. Chem. C 2011, 115, 24906–24914

The Journal of Physical Chemistry C

ARTICLE

(14) Torrisi, A.; Mellot-Draznieks, C.; Bell, R. G. J. Chem. Phys. 2009, 130, 194703. (15) Torrisi, A.; Mellot-Draznieks, C.; Bell, R. G. J. Chem. Phys. 2010, 132, 044705. (16) Head-Gordon, M.; Pople, J. A.; Frisch, M. J. Chem. Phys. Lett. 1988, 153, 503. (17) Raghavachari, K.; Frisch, M. J.; Pople, J. A. Chem. Phys. Lett. 1980, 72, 4244. (18) Szabo, A.; Ostlund, N. S. Modern Quantum Chemistry; McGraw-Hill: New York, 1989. (19) Breneman, C. M.; Wiberg, K. B. J. Comput. Chem. 1990, 11, 361. (20) Gaussian 03, Revision C.02: Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian, Inc.: Wallingford, CT, 2004.

24914

dx.doi.org/10.1021/jp2084378 |J. Phys. Chem. C 2011, 115, 24906–24914