Infrared Spectroscopic Investigation of the Acidity ... - ACS Publications

May 26, 2011 - inGAP Center of Research Based Innovation, Department of Chemistry, ... Norway. bS Supporting Information. 'INTRODUCTION. Zeolites find...
0 downloads 0 Views 2MB Size
ARTICLE pubs.acs.org/JPCC

Infrared Spectroscopic Investigation of the Acidity and Availability of the Surface Hydroxyls of Three-Dimensional 12-Ring Zeotype H-ITQ-7 Ragnhild Skorpa,† Stian Forselv,† Karina Mathisen,† Silvia Bordiga,‡ Morten Bjørgen,*,† and Stian Svelle§ †

Department of Chemistry, Norwegian University of Science and Technology, Høgskoleringen 5, 7491 Trondheim, Norway Dipartimento di Chimica IFM-NIS Centre of Excellence, Universita di Torino, Via P. Giuria 7, 10125 Torino, Italy § inGAP Center of Research Based Innovation, Department of Chemistry, University of Oslo, P.O. Box 1033, Blindern, N-0315 Oslo, Norway ‡

bS Supporting Information ABSTRACT: Zeotype H-ITQ-7 is among the very few microporous materials comprising a three-dimensional large pore channel based topology other than the well studied zeolite H-beta. Thus, H-ITQ-7 is of substantial interest as an acid catalyst for petrochemical processing of bulky molecules where H-beta is employed today. A more exact fundamental understanding of the availability of the surface sites to bulky species and the acidic properties of this little studied zeotype is highly desirable. Quantitative knowledge of this nature constitutes a prerequisite for potential rational process design based on fundamental insights. To illuminate these aspects, we have employed infrared spectroscopy in combination with adsorption of nonpolymerizing, sterically demanding probe molecules (1,2,4,5-tetramethylbenzene and hexamethylbenzene) that undergo infrared visible symmetry changes upon protonation. This approach can be viewed as a general approach to obtain detailed information about proton transfer reactions and the acidic character of such porous solids, in particular when the experimentally based interpretations are supported by spectra obtained from density functional theory (DFT) calculations. On the basis of the degree of protonation and the persistence of the cations within H-ITQ-7, it can be established that the Brønsted acidic strength of H-ITQ-7 is slightly weaker than that reported for H-beta. Moreover, information regarding the microstructure of the H-ITQ-7 particles has been obtained, as it appears that a large fraction of the surface hydroxyl groups are located in internal crystal voids inaccessible to bulky molecules.

’ INTRODUCTION Zeolites find widespread applications as catalysts in the petrochemical and refining industries. Structurally, zeolites may be viewed as crystalline materials based on a three-dimensional network of TO4 tetrahedra, where T is Si or Al, connected by sharing oxygen atoms at each tetrahedral corner. Each substitution of Al for Si leads to the introduction of a charge deficiency that may be compensated for in a manner leading to the formation of strongly acidic Brønsted sites. It is customary to distinguish between small, medium, and large pore zeolites, depending on the number of T-atoms defining the circumference of the pores, that is, 8-ring, 10-ring, or 12-ring materials. The term zeotype is used to describe topologically similar materials comprising T-atoms other than Si and Al. To date, the International Zeolite Association has assigned their well-known three letter codes to nearly 200 unique framework types. Despite this multitude of topologies, three-dimensional 12-ring channel structures are very scarce. The most prominent representative member of this subgroup is the disordered beta (*BEA) zeolite, reported in 1967.1 Zeolite beta is employed commercially, and one defining characteristic of this material is the ability to catalyze reactions involving bulky molecules, such as alkylbenzenes. The widely employed zeolite Y (FAU) is also a three-dimensional 12ring material, but the pore architecture is cage-based and the r 2011 American Chemical Society

properties differ significantly from channel based H-beta topology. The structure of the ITQ-7 (ISV) microporous material was reported in 1999, and this was only the second three-dimensional 12-ring channel based topology.2 Brønsted acidic H-ITQ-7 is by necessity an aluminogermanosilicate, as Ge incorporation is required to stabilize an Al containing, and thus Brønsted acidic, framework.35 A limited number of studies have demonstrated that H-ITQ-7 is an active catalyst in reactions requiring Brønsted acidity.4,69 Despite being crucial in such reactions, the acidic properties of H-ITQ-7 are not completely understood. Infrared spectroscopy (FTIR) is a useful tool to investigate the surface properties of solids, in particular when the interactions with probe molecules and the surface sites are studied. In a previous study, FTIR was employed in combination with several probe molecules of fairly weak basicity, such as CO, H2O, and CH3OH.9 However, when these probe molecules are employed, proton transfer is observed only indirectly for H2O and CH3OH upon the formation of clusters of adsorbates and by the appearance of the Fermi resonance profile. Thus, the exact basicity (or proton affinity, PA) of the adsorbed species is difficult to Received: March 14, 2011 Revised: May 3, 2011 Published: May 26, 2011 12090

dx.doi.org/10.1021/jp202400f | J. Phys. Chem. C 2011, 115, 12090–12094

The Journal of Physical Chemistry C evaluate and the extraction of quantitative information with respect to proton transfer, the key step in acid catalysis, is hampered by complex interpretation of the spectra. Also, these probe molecules have an insufficiently large kinetic diameter to evaluate the steric aspect or shape selective properties of zeolites and zeotypes. Thus, a more detailed assessment of both these issues, that is, protonation capabilities and shape selective properties/accessibility of the various surface sites, is important to further understand the properties of this material of significant potential industrial relevance. To this end, we have combined density functional theory (DFT) with FTIR spectroscopy employing sterically demanding probe molecules of fairly high basicity, 1,2,4,5-tetramethylbenzene (durene) and hexamethylbenzene (hexaMB). The PAs of these molecules are 818 and 860 kJ mol1, respectively. These particular molecular probes, as opposed to, for example, pentamethylbenzene, have been chosen due to the presence of diagnostic vibrational modes which become IR active by the reduction of molecular symmetry induced upon protonation, that is, changes in point groups. Also, they are among the very few probe molecules of high PA which do not oligomerize in zeolites.10 The commonly employed pyridine is disqualified in this context, due to the high PA (930 kJ mol1) that leads to immediate protonation without discriminating between the various surface sites. Moreover, durene is the largest molecule that may show significant diffusion rates through the channels of 10-ring zeolites, whereas the kinetic diameter of hexaMB is commensurate with the pore openings of 12-ring zeolite H-beta. Finally, derivatives of such species, that is protonated methylbenzenes, have been suggested to function as reaction intermediates in several important processes such as the conversion of methanol to hydrocarbons11,12 and transalkylation reactions.13 The spectra presented in this study convey specific information regarding the acid strength of the Brønsted sites responsible for protonation, the accessibility of the various surface hydroxyls, silanol, and germanol groups, in addition to the Brønsted sites, and the ability of these bulky species to diffuse through the channels of H-ITQ-7. These insights will be compared to relevant data available for the structurally comparable H-beta zeolite. Moreover, this study provides a demonstration of a fairly straightforward, but uncommonly employed, approach to characterize the increasingly populated family of mesoporous zeotypes and zeotype materials comprising large and extra large pores.

’ EXPERIMENTAL SECTION The H-ITQ-7 sample was prepared according to procedures described by Leiva et al.3 The full synthetic procedures including involved template preparation and regular characterization of the sample are given in ref 9. Briefly, the particle size range was 0.20.5 μm, the BET surface area measured immediately after template removal was 542 m2 g1, and the elemental analysis by ICP-MS gave (Si þ Ge)/Al = 103. Powder X-ray diffraction with Pawley fit confirmed phase purity. For the FTIR measurements, the zeolite samples were compressed into self-supporting wafers. Before adsorption, the wafers were heated to 550 °C for 3 h under high vacuum to remove water and other adsorbed impurities. Subsequent to the pretreatment, the IR cell was directly attached to a reservoir containing solid durene/hexaMB, and the connection to the hydrocarbon reservoir was opened. To

ARTICLE

Scheme 1. Simplified Representations and Scaled B3LYP/ 6-311þG** Vibrational Frequencies of the Two Relevant Modes Sensitive to Protonation for Benzene To Yield [Benzene þ H]þ a

a

Top: ν1. Bottom: ν2.

maximize the coverage by the hydrocarbons, the entire accessory was treated at 65 °C for 2 h for durene and an additional 30 min at 120 °C followed by three days at room temperature for hexaMB. To remove physisorbed hydrocarbons, the IR cell was connected to a liquid N2 cold trap and evacuated. The IR spectra were obtained at ambient temperature in transmission mode using a Bruker Vertex 70 spectrometer, equipped with an MCT detector, at 1 cm1 resolution. DFT calculations were carried out using the Gaussian 03 program package. Structure optimization and vibrational analysis were carried out at the B3LYP/6-311þG** level of theory. A scaling factor of 0.978 was employed.14,15

’ RESULTS AND DISCUSSION The vibrational spectra of benzene, durene, and hexaMB and their protonated counterparts were calculated using DFT in order to facilitate and put on firmer ground the assignments of the experimental spectra. Benzene is the simplest representative of these aromatics, and the form of the two vibrational modes of relevance for this work remains virtually the same for methyl substituted benzenes.16 Thus, our principal discussion at this point will be based on the two ring modes (denoted ν1 and ν2) of benzene shown in Scheme 1. Animations of the ring modes are given for hexaMB/[hexaMB þ H]þ and durene/[durene þ H]þ in the Supporting Information. Both neutral and protonated benzene display two complex ν(CC) ring modes (ν1 and ν2). Notably, these are well separated from other IR active modes. For neutral benzene, these modes have the same vibrational frequencies. However, due to restrictions imposed by the gross selection rule for IR transitions, that is, that a vibrational transition must be accompanied by a change in the dipole moment; these modes are silent in the infrared spectrum for the neutral species. For protonated benzene, however, due to the reduction of molecular symmetry upon protonation, both bands are IR active, although much less so for ν2 (see below). Also, the vibrational frequency of ν2 is significantly red-shifted, whereas ν1 is largely unaltered. The situation is similar for the methyl substituted benzenes, as seen in Table 1. Upon protonation of durene, both modes become IR active although low intensity is predicted for ν2, as was the case also for protonated benzene. For neutral hexaMB, the frequencies of 12091

dx.doi.org/10.1021/jp202400f |J. Phys. Chem. C 2011, 115, 12090–12094

The Journal of Physical Chemistry C

ARTICLE

Table 1. Calculated (B3LYP/6-311þG**) Dipole Moments (Debye), Point Groups, and Calculated Frequencies (cm1) and IR Intensities (in parentheses) of the Two Ring Modes Sensitive to Protonation dipole moment point group

ν1 (DFT)

ν2 (DFT) 1597 (0)

benzene

0

D6h

1597 (0)

[benzene þ H]þ

0.88

C2v

1602 (76)

1536 (1)

durene

0

D2h

1564 (0)

1621 (0)

[durene þ H]þ

1.22

C2v

1576 (56)

1562 (4)

hexaMB

0

D3d

1575 (0)

1575 (0)

[hexaMB þ H]þ

0.40

Cs

1589 (142) 1519 (51)

Figure 1. FTIR spectrum of dehydrated H-ITQ-7 and assignments of the three most prominent bands in the ν(OH) region.

ν1 and ν2 are practically indistinguishable, but they become separated by 70 cm1 when protonated. The ν1 mode is, as for the other protonated species, predicted to be more intense, but the intensity of ν2 is no longer negligible. Thus, the appearance of absorptions near 1600 cm1, representing ν1 and ν2, in the IR spectra of durene and hexaMB in acidic media will provide evidence for the formation of the corresponding protonated species. Armed with the insight provided by DFT, we turn our attention to the experimental study of durene and hexaMB adsorption on H-ITQ-7. The FTIR spectrum of H-ITQ-7 dehydrated in vacuum at 550 °C for 3 h is shown in Figure 1. The ν(OH) region is dominated by three absorptions, which previously have been assigned to silanol groups (3746 cm1), germanol groups (3677 cm1), and Brønsted sites (∼3620 cm1) as indicated. The broad Brønsted band has been suggested to be composed of at least two components.3,9,17 It may be noted that the sample appears to be virtually free from framework defects giving rise to a silanol nest, which are usually seen as very broad contribution centered around 3450 cm1. The spectra recorded upon adsorption of durene and hexaMB on H-ITQ-7 are shown in Figure 2. The black spectra in Figure 2

Figure 2. Top panels (a): The ν(OH)/ν(CH) regions (left) and the δ(CH)/ν(CC) regions (right) upon adsorption of durene on H-ITQ-7. Black spectra, maximum loading; gray spectra, after desorption by evacuation. Bottom panels (b): As above for the adsorption of hexaMB; note the magnification of the bottom right panel. Dashed spectra represent durene (top panels) and hexaMB (bottom panels) in KBr (arbitrary scale).

are difference plots corresponding to maximum coverage by the hydrocarbons, whereas the gray spectra were recorded after subsequent evacuation. Thus, negative peaks correspond to the consumption of the original surface hydroxyls seen in Figure 1, whereas positive peaks arise from either surface sites perturbed by the adsorbates or from internal vibrations in the adsorbates themselves. Also included are the spectra of the two probes in KBr, displayed as dashed curves (arbitrary scale). Both durene and hexaMB in KBr give rise to absorptions in the ν(CH) (left panels) and δ(CH)/ν(CC) (right panels) regions. It is noteworthy that no signals are discernible in the 1700 1550 cm1 range, in agreement with the DFT calculations, which predict that the modes in this spectral region are IR silent for the neutral species. Turning to the black spectra, obtained at high loading of the molecular probes in the zeotype, it should be noted that durene and hexamethylbenzene give rise to spectra that are very similar to those obtained in KBr, in both the ν(CH) and δ(CH)/ν(CC) regions. This observation is significant, as it implies that there are no major geometric perturbations of the internal vibrational modes of the probe molecules caused by for example 12092

dx.doi.org/10.1021/jp202400f |J. Phys. Chem. C 2011, 115, 12090–12094

The Journal of Physical Chemistry C confinement effects upon adsorption in H-ITQ-7. However, the principal difference between the spectra obtained of the species diluted in KBr and those obtained when adsorbed in the zeotype is seen the high frequency part of the ν(CC) regions (right panels, Figure 2). Clearly, distinct absorptions are seen at 1600 cm1 for durene and 1617 cm1 for hexaMB. These absorptions can only be assigned to the IR active ν(CC) ring modes described in Scheme 1 above. As mentioned above, possible changes in molecular geometry due to confinement effects are very unlikely and can be ruled out as a cause of these bands. Gratifyingly, the DFT calculations reproduce the vibrational frequencies of these particular vibrational modes rather well, especially with respect to the observed difference in frequency seen for the two probe molecules. The appearance of these bands constitutes evidence of proton transfer from the zeotype to the molecular probes, leading to the formation of [durene þ H]þ and [hexaMB þ H]þ, thereby inducing the symmetry reduction that activates the vibrational modes that are diagnostic of the protonated species. We will now address the ν(OH) region of the spectra, where perturbations of the surface sites of the zeotype as a consequence of interactions with the probe molecules are seen. The black spectra corresponding to high loadings will be discussed first. For both durene and hexaMB, two distinct negative peaks are seen at high frequencies, corresponding to the consumption of silanol (3746 cm1) and germanol groups (3677 cm1). Both these groups of surface hydroxyls are virtually completely consumed upon adsorption of durene, as determined by comparison to the spectrum of the dehydrated material (compare Figures 1 and 2). This means that these sites are fully accessible to durene. However, for hexaMB, the negative signals related to the disappearance of both the silanol and germanol groups are less intense. Clearly, hexaMB cannot access all of these surface hydroxyls, despite the harsher adsorption procedure employed for hexaMB (see Experimental Section). Integration suggests that about half of the silanols are accessible to hexaMB. This is a rather unexpected result, as these hydroxyls are usually assumed to be located on the external surface of the zeolite particles. For silanol groups, which are well studied, a sharp peak at about 3745 cm1 can safely be assumed to represent silanol species not involved in any hydrogen bonding, and silanols involved in internal defects give rise to clearly distinguishable absorptions at lower wave numbers, as described above. The inaccessibility of a sizable fraction of the isolated silanols of H-ITQ-7 is in contrast to similar investigations of H-beta, for which all silanols seen at 3745 cm1 disappear upon adsorption of hexaMB.10,18 A plausible explanation for this discrepancy is that the fairly large H-ITQ-7 particles might contain internal voids, for example along intergrowth boundaries, that are sufficiently large so that the silanol groups located on these internal surfaces appear to be fully isolated. Holm et al. employed 2,6-dimethylpyridine to investigate the textural properties of an alkaline treated mesoporous H-ZSM-5 and found similar spectral features.19 Concurrent to the disappearance of the silanol/germanol absorptions, intense and broad peaks appear, centered at 3550 cm1 for durene and 3530 cm1 for hexaMB. These signals correspond to overlapping contributions from silanol (mostly, as these are in clear excess) and germanol hydroxyls interacting with physisorbed durene or hexaMB molecules. The intensities correlate well to the extent of silanol consumption for the two probe molecules, despite the approximate nature of comparisons of signals at different wavenumbers.16 Also, the magnitude of the shift in vibrational frequency of the silanol groups is 196 and

ARTICLE

216 cm1 for durene and hexaMB, respectively. Notably, the shift correlates well with the PAs of the two probe molecules: a larger shift (that is stronger perturbation) is expected for a more basic molecule due to a stronger acid/base interaction. Similar shifts of 180 and 207 cm1 were observed upon adsorption of the same probes on H-beta.10,18 Also, and as expected, these shifts are somewhat higher compared to the shifts reported when slightly less basic probes, such as furan (PA = 803 kJ mol1), are adsorbed on H-beta, H-ZSM-5, and H-mordenite.20 For both probes, negative contributions are discernible at about 3620 cm1 as a consequence of the erosion of the Brønsted bands upon interaction with the hydrocarbons. This feature is less distinct for durene, most likely due to the more intense signal corresponding to perturbed silanol groups. However, it does appear that the majority of the Brønsted sites are accessible to both probes, in contrast to the observations made for the silanol groups. For durene, a weak yet distinct absorption with a maximum around 3240 cm1 is seen. This can only be ascribed to durene physisorbed onto the Brønsted sites, that is unprotonated durene, leading to a substantial red shift of the ν(OH) mode of these strongly acidic hydroxyls by 380 cm1. Notably, no band corresponding to the ν(OH) mode of perturbed Brønsted sites is discernible for hexaMB. Hence, durene does appear to reside to a significantly lesser extent on the protonated form, as inferred from the presence of the band at 3240 cm1 representing durene physisorbed on Brønsted sites. This is related to the lower PA of durene. The appearance of durene physisorbed onto the Brønsted sites was not seen in similar studies of the H-beta zeolite,10 and indicates a slightly lower acid strength of H-ITQ-7 compared to H-beta. It is of relevance to compare the shift in the Brønsted ν(OH) band upon adsorption of durene to those seen when furan is adsorbed in other zeolites. Interestingly, a substantially larger shift in the Brønsted ν(OH) frequency (∼500 cm1) is seen for furan H-beta, H-ZSM-5, and H-mordenite. However, a shift in a range similar to that reported herein for durene in H-ITQ-7 is suggested by interpolation of previously reported data for the H-Y zeolite.20 This further implies that the acid strength of the Brønsted sites of H-ITQ-7 is lower than that measured for H-beta, H-ZSM-5, and H-mordenite and more in line with that of H-Y and also H-SAPO-34, for which a slightly lower acid strength compared to most other zeolites is well documented.20,21 To facilitate the evaluation of the mobility and persistence of the various adsorbed species, the IR cell was evacuated subsequent to hydrocarbon adsorption (gray spectra). For durene, all signals representing adsorbed species are strongly diminished upon evacuation. This is clearly seen for the internal modes of the probe molecule in the ν(CH) and δ(CH)/ν(CC) regions. The silanols (3746 cm1) and germanols (3677 cm1) are to a large extent restored, accompanied by a reduction in the intensity of the signal corresponding to the perturbed counterparts at 3550 cm1. The development of the Brønsted band (3620 cm1) is more difficult to evaluate, due to the large change in the partly overlapping contributions of the perturbed silanols. However, it is clearly seen that the band at 1600 cm1 representing protonated durene is reduced upon evacuation whereas the concentration of durene physisorbed to Brønsted sites (3240 cm1) is reduced to a lesser extent. This is in compliance with a situation where the protonation equilibrium of durene favors the unprotonated species, as mentioned above. The effects of evacuation are significantly less pronounced for hexaMB compared to durene: The spectra show that hexaMB is clearly 12093

dx.doi.org/10.1021/jp202400f |J. Phys. Chem. C 2011, 115, 12090–12094

The Journal of Physical Chemistry C more persistent within the pores of H-ITQ-7. The contrast is especially large for the silanol and germanol species, which are virtually unaltered after evacuation. It is reasonable to link this to the restricted accessibility of these sites for hexaMB, as discussed above. However, as was the case for durene, the concentration of protonated hexaMB is reduced, as seen from the reduction in the band at 1617 cm1 and the corresponding partial reappearance (seen as a less distinct signal in difference plots) of the Brønsted band (3620 cm1). The gray spectra in Figure 2 clearly illustrate that both protonated durene and hexaMB are removed to a large extent from the H-ITQ-7 surface upon evacuation. This lends further support to the notion of a lower acid strength of H-ITQ-7 compared to H-beta, for which these cations were found to be fully persistent even after evacuation at 150 °C.18

’ CONCLUSIONS We have further demonstrated the usefulness of employing IR spectroscopy with probe molecules that undergo IR activation symmetry changes upon adsorption as a tool to characterize proton transfer within microporous solid acids. The adsorption of 1,2,4,5-tetramethylbenzene (durene) and hexamethylbenzene (hexaMB) has been carried out, due to their fairly high basicity. This ensures a strong interaction with the surface hydroxyls. Also, the absence of further reactions such as oligomerization avoids the complications often encountered when employing other molecular probes of similar basicity. Importantly, confident interpretation of the experimental spectra has been facilitated by DFT calculations. Specifically, this work serves to extend the characterization of H-ITQ-7, primarily with respect to the catalytically important acid site strength and availability. In conclusion, it is clear that both durene and hexaMB are protonated by the Brønsted acidic sites of H-ITQ-7. This has previously been observed also for zeolite H-beta and proves that H-ITQ-7 interacts strongly with hydrocarbons having proton affinities above 800 kJ mol1. This result is consistent with a generally strong Brønsted acidity of H-ITQ-7. However, three further observations indicate that the acid strength of H-ITQ-7 is lower than that of H-beta: (1) Durene is less extensively protonated in H-ITQ-7 compared to hexaMB. (2) The shift in the Brønsted ν(OH) frequency upon physisorption of durene is conspicuously low. (3) Both protonated hexaMB and durene in particular are to a significant extent deprotonated and desorbed from H-ITQ-7 upon evacuation. Apparently, the acid strength of H-ITQ-7 is more comparable to that of H-Y and H-SAPO-34, which are known to be among the less strongly acidic zeolites/ zeotypes. Finally, a sizable fraction of the isolated silanols is inaccessible for hexaMB, interpreted as crystal defects blocking diffusion pathways. ’ ASSOCIATED CONTENT

bS

Supporting Information. Animations (.avi) of the key vibrational modes for durene and hexamethylbenzene and their protonated counterparts. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Fax: (þ) 47 735 50877. Phone: (þ) 47 735 94846.

ARTICLE

’ ACKNOWLEDGMENT Thanks are due to the Research Council of Norway for grant of computer time through the NOTUR project (account NN4683K). This publication forms a part of the inGAP Center of Research-based Innovation, which receives financial support from the Research Council of Norway under Contract 174893. ’ REFERENCES (1) Wadlinger, R.; Kerr, G.; Rosinksi, E. U.S. Patent 3 308 069, 1967. (2) Villaescusa, L. A.; Barrett, P. A.; Camblor, M. A. Angew. Chem., Int. Ed. 1999, 38, 1997–2000. (3) Leiva, S.; Sabater, M. J.; Valencia, S.; Sastre, G.; Fornes, V.; Rey, F.; Corma, A. C. R. Chim. 2005, 8, 369–378. (4) Corma, A.; Diaz-Cabanas, M. J.; Fornes, V. Angew. Chem., Int. Ed. 2000, 39, 2346–2349. (5) Corma, A.; Diaz-Cabanas, M. J.; Domine, M. E.; Rey, F. Chem. Commun. 2000, 1725–1726. (6) Corma, A.; Costa-Vaya, V. I.; Diaz-Cabanas, M. J.; Llopis, F. J. J. Catal. 2002, 207, 46–56. (7) Botella, P.; Corma, A.; Sastre, G. J. Catal. 2001, 197, 81–90. (8) Corma, A.; Martinez-Triguero, J.; Martinez, C. J. Catal. 2001, 197, 151–159. (9) Skorpa, R.; Bordiga, S.; Bleken, F.; Olsbye, U.; Arstad, B.; Tolchard, J.; Mathisen, K.; Svelle, S.; Bjørgen, M. Microporous Mesoporous Mater. 2011, 141, 146–156. (10) Bjørgen, M.; Bonino, F.; Arstad, B.; Kolboe, S.; Lillerud, K. P.; Zecchina, A.; Bordiga, S. ChemPhysChem 2005, 6, 232–235. (11) Bjørgen, M.; Svelle, S.; Joensen, F.; Nerlov, J.; Kolboe, S.; Bonino, F.; Palumbo, L.; Bordiga., S.; Olsbye, U. J. Catal. 2007, 249, 193–205. (12) Svelle, S.; Joensen, F.; Nerlov, J.; Olsbye, U.; Lillerud, K. P.; Kolboe, S.; Bjørgen, M. J. Am. Chem. Soc. 2006, 128, 14770–14771. (13) Svelle, S.; Olsbye, U.; Lillerud, K. P.; Kolboe, S.; Bjørgen, M. J. Am. Chem. Soc. 2006, 128, 5618–5619. (14) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; 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 03, Revision C.02; Gaussian, Inc.: Wallingford, CT, 2004. (15) Douberly, G. E.; Ricks, A. M.; Schleyer, P. v. R.; Duncan, M. A. J. Phys. Chem. A 2008, 112, 4869–4864. (16) Colthup, N. B.; Daly, L. H.; Wiberly, S. E.; Introduction to Infrared and Raman Spectroscopy, 3rd ed.; Academic Press: Boston, 1990. (17) Sastre, G.; Fornes, V.; Corma, A. J. Phys. Chem. B 2002, 106, 701–708. (18) Bjørgen, M.; Bonino, F.; Kolboe, S.; Lillerud, K. P.; Zecchina, A.; Bordiga, S. J. Am. Chem. Soc. 2003, 125, 15863–15868. (19) Holm, M. S.; Svelle, S.; Joensen, F.; Christensen, C. H.; Bordiga, S.; Bjørgen, M. Appl. Catal., A 2009, 356, 23–30. (20) Paze, C.; Bordiga, S.; Lamberti, C.; Salvalaggia, M.; Zecchina, A.; Bellussi, G. J. Phys. Chem. B 1997, 101, 4740–4751. (21) Bordiga, S.; Regli, L.; Cocina, D.; Lamberti, C.; Bjørgen, M.; Lillerud, K. P. J. Phys. Chem. B 2005, 109, 2779–2784. 12094

dx.doi.org/10.1021/jp202400f |J. Phys. Chem. C 2011, 115, 12090–12094