How Keggin-Type Polyoxometalates Self-Organize into Crystals

Sep 29, 2009 - P. Mothé-Esteves†, M. Maciel Pereira†, J. Arichi‡ and B. Louis*‡. † Universidade Federal do Rio de Janeiro, Centro de Tecnol...
0 downloads 0 Views 4MB Size
DOI: 10.1021/cg900984z

How Keggin-Type Polyoxometalates Self-Organize into Crystals

2010, Vol. 10 371–378

P. Mothe-Esteves,† M. Maciel Pereira,† J. Arichi,‡ and B. Louis*,‡ †

Universidade Federal do Rio de Janeiro, Centro de Tecnologia, Instituto de Quimica, eriaux, Cidade Universitaria, Rio de Janeiro, 21949-900 RJ, Brazil, and ‡Laboratoire des Mat Surfaces et Proc ed es pour la Catalyse (LMSPC), European Laboratory for Catalysis and Surface Science (ELCASS), UMR 7515 du CNRS Universit e de Strasbourg, 25 rue Becquerel, 67087 Strasbourg Cedex 2, France Received August 18, 2009; Revised Manuscript Received September 9, 2009

ABSTRACT: Vanadium-containing polyoxometalates (POM) were synthesized by two distinct procedures (hydrothermal and under acidic conditions) and were thoroughly characterized by X-ray diffraction, Fourier transform infrared spectroscopy, Brunauer-Emmett-Teller (BET) method, UV-vis, 31P and 51V magic angle spinning NMR, scanning electron microscopy, and extended X-ray absorption fine structure spectroscopy-X-ray absorption near-edge spectroscopy (EXAFS-XANES). POM crystals were successfully grown under acidic atmosphere, thus forming a self-assembled structure constituted by cubic crystals that are between 20 and 80 μm in size. EXAFS-XANES experiments and molecular orbital calculations upon the Keggin unit cell allowed the determination of both the nature of the interactions and the assembly of the polyanions. A growth model was therefore developed based on the aggregation of Keggin units, ranging from one heteropolyanion unit to a crystal. Herein, we report a bottom-up approach which can serve as a powerful tool to assemble complex nanostructures via self-assembly, as well as to identify the building blocks present in solution during the assembly process. Finally, these auto-organized cubic vanadomolybdate crystals exhibit the highest catalytic activity and selectivity in the liquid phase partial oxidation of 2-methylnaphtalene into K3 vitamin.

1. Introduction Spontaneous self-organization of matter demonstrates the capacity of nature to structure itself when specific conditions are present. The design and control of chemical systems over multiple-length scales from the molecule to the crystal thus represent one of the greatest challenges of science,1-3 in particular, the ability to understand how molecules organize and thus build supramolecular architectures to produce functional nanosystems and nanomachines.1,2 Supramolecular chemistry deals with the association of two or more chemical species, based on molecular noncovalent interactions.4,5 These interactions often confer to the supramolecular structure new chemical and physical properties.6 The nano-objects of supramolecular chemistry are both defined by the nature of the individual building blocks, and by the type of interactions which hold them together. The driving force of the assembly process can be hydrogen bonding, electrostatic forces, van der Waals forces, or metal-ion coordination.4-9 Heterogeneous catalysis can also take advantage of the peculiar properties of self-assembled catalysts.10-12 These materials can be useful for the conception of new structured catalysts and reactors.10,13 Indeed, self-assembled catalysts can avoid the use of an inert binder to pack isolated crystals in catalytic beds, thus to ensure improved mass transfer. One can therefore reasonably hope to achieve a higher catalyst activity which might be accompanied by a rise in selectivity. Because of their simple synthesis procedure and thermal stability, polyoxometalates (POM) having the Keggin structure14 are often used in acid catalysis.15-23 These acid POMs are molecular clusters formed by different metal oxides with the general formula H3PM12O40. Heteropolyacids are ionic

crystals in the solid state, consisting of large polyanions (primary structure: PM12O403-), counter cations, and water of crystallization. This creates the so-called secondary structure.20 In addition, the arrangement of the particles, their morphology, and pore structure generate the tertiary structure.20 The aim of the present study is to produce an organized tertiary structure of V-containing POM, while developing a suitable synthesis procedure. Since noncovalent interactions between elementary cells often act as the driving force to assemble the particles of inorganic materials,11 we have focused our work on the understanding of the assembly process by experiments and calculations. The self-organization of pure dodecatungstophosphoric acid was recently achieved by Mizuno8 while modifying the countercation. Our present approach tends to induce a selforganization of mixed vanadomolybdate polyoxometalates (H3þnPVnMo12-nO40). Indeed, the substitution of one molybdenum atom by one vanadium atom can modify the crystal growth, and thus can act as a driving force for the process. 2. Experimental Section

*To whom correspondence should be addressed. E-mail: blouis@chimie. u-strasbg.fr. Tel: þ33390242760. Fax: þ33390242761.

2.1. Preparation Procedures of POM. 12-Molybdophosphoric acid (H3PMo12O40), furnished by Fluka, was used without further purification. Vanadium containing POM were prepared either via hydrothermal treatment or via simple acidification of an aqueous solution containing an alkali metal salt of molybdate and vanadate compounds, and subsequent isolation of the material via ether extraction. Heteropolyvanadates (HPV) have therefore been synthesized following the initial procedure described by Tsigdinos and Hallada,24 by mixing aqueous solutions of sodium molybdate dihydrate (Na2MoO4 3 2H2O, Fluka, 99.5 wt %), disodium hydrogen phosphate dihydrate (Na2HPO4 3 2H2O, Fluka, 99 wt %), and sodium metavanadate (NaVO3, Fluka, 98 wt %). Concentrated sulfuric acid was added dropwise under vigorous stirring, to acidify the solution. The red-colored solution was cooled to 273 K, and kept 2 h at this temperature before being extracted by diethyl ether. The

r 2009 American Chemical Society

Published on Web 09/29/2009

pubs.acs.org/crystal

372

Crystal Growth & Design, Vol. 10, No. 1, 2010

Mothe-Esteves et al.

Figure 1. FT-IR spectrum of H5V2Mo10O40 polyoxometalate obtained after exposure under sulfuric acid vapors for 6 days. resulting HPV-etherate complex was either dried at 333 K overnight to remove the solvent, or placed in a desiccator under sulfuric acid vapors for several days to allow a slow crystallization of the POM. The compound PVMo11 has also been prepared via a hydrothermal process,25,26 where a stoichiometric mixture of 0.98 g (0.01 mol) of phosphoric acid, 0.91 g (0.005 mol) of vanadium pentoxide, and 14.4 g (0.11 mol) of molybdenum trioxide was suspended in 150 mL of distilled water. The mixture was stirred for 3 h at 353 K, and then cooled to room temperature. After removal of insoluble molybdates and vanadates, the HPA solution was evaporated and dried at 338 K for 48 h yielding orange anhydrous PVMo11 solid. The PV2Mo10 material was synthesized following the same hydrothermal procedure. 2.2. Characterizations. X-ray diffraction (XRD) patterns were acquired on a D8 Advance Bruker AXS powder diffractometer (θ/2θ) using monochromatized Cu KR radiation in the range of 2θ from 5° to 60°. Fourier transform infrared (FT-IR) measurements were performed with a Perkin-Elmer spectrum BX apparatus, using KBr pellets. The specific surface areas (SSA) were measured by N2 adsorption-desorption at 77 K using a Micromeritics TriStar apparatus (BET method). Scanning electron microscopy (SEM) was operated on a JEOL FEG 6700F microscope working at 9 kV accelerating voltage. Energy dispersive X-ray (EDX) spectra were acquired to determine the composition of the material. The EDX spectra were acquired using 20 kV primary electron energy. Quantification was done using the standard-less ZAF correction method in the Genesis software from EDX. Thermogravimetric analyses (TGA) were carried out on a SETARAM equipment at a heating rate of 5°/min from room temperature up to 700 °C, under Ar flow. Solid-state 31P NMR experiments were performed on a Bruker DSX-400 spectrometer (B0 = 9.4 T) at 161.9 MHz. The magic angle spinning (MAS) NMR experiments were recorded on a standard double bearing probe with a 4 mm diameter ZrO2 rotor, with spinning frequencies ranging from 3.5 to 10 kHz. 31P MAS NMR spectra were acquired with a π/2 pulse duration of 3.5 ms, and a recycle time of 193 s to avoid saturation. 51V MAS NMR was

performed on a Bruker Avance III 400WB spectrometer at a spinning frequency of 28 kHz, with a 2.5 mm probe. V2O5 was used as a reference with a chemical shift of -610 ppm. X-ray absorption near-edge spectroscopy (XANES) and extended X-ray absorption fine structure spectroscopy (EXAFS) measurements were carried out at the vanadium K-edge (5465 eV) in fluorescence mode at room temperature in the XAFS1 beamline of Brazilian Synchrotron Light Laboratory (LNLS) using a Si (111) crystal monochromator. Calibrations of the spectrometer were made before, between, and after scanning vanadium catalysts using vanadium-metal foil. 2.3. Calculations. Molecular orbital calculations were performed using the semiempirical PM6 method27 available at package MOPAC2007.28 Geometries were fully optimized at C1 symmetry and considering a gradient norm of 0.1 kcal/A˚. The obtained geometries were characterized as minima in the potential energy surface by the analysis of the Hessian of the energy showing the absence of imaginary frequencies. 2.4. Catalytic Partial Oxidation of Aromatics. To evaluate their potential as catalysts, the different POM materials were tested in liquid-phase oxidation of 2-methylnaphtalene at 333 K for 2 h. The operating conditions were as follows: 10 mL of acetonitrile (solvent), 0.1 g of catalyst, 0.01 mol of aromatic substrate, 0.05 mol of H2O2 (30 wt % in water). The products were analyzed by gas chromatography (HP 5890 Series II) with a capillary column (PONA, 50m) and FID detector.

3. Characterizations of POM Materials 3.1. Vibration, Electronic Spectroscopies, BET and X-ray Diffraction Patterns. Figure 1 shows the FT-IR spectrum of H5PV2Mo10O40 material obtained after exposure under sulfuric acid vapors during 6 days. The four infrared bands can be attributed to an R- or β-Keggin structure, in agreement with the literature, as follows: 1057 cm-1 (νas P-O), 958 cm-1 (νas MdO, with M: Mo, V), 866 cm-1 (νas M-O-M

Article

Crystal Growth & Design, Vol. 10, No. 1, 2010

373

Figure 2. XRD patterns of V1 and V2 polyoxometalates.

interoctahedral), 784 cm-1 (νas M-O-M intraoctahedral).29-31 The bands located at 3300-3500 cm-1 and 1610-1650 cm-1 correspond to the presence of water in the sphere of coordination (secondary structure). H5PV2Mo10O40 POM presents an absorption band at 305 nm (figure not shown). PM12 R-isomer of Keggin structures display O f M ligand-to-metal charge transfer bands (LMCT) between 200 and 400 nm,29,32 corresponding to Mo6þ in distorted octahedral position.33 After adding hydrogen peroxide (for catalyst evaluation) to the material, a new band was observed around 323 nm which corresponds to the formation of V peroxo species and thus to the presence of V5þ species.33 The chemical composition of V1 and V2 POM materials was already confirmed by elemental analysis and potentiometric titrations of acidic hydrogens in the initial work from Tsigdinos and Hallada.24 Both EDX elemental analysis and 31P and 51V MAS NMR spectra (see coming sections), further evidenced the formation of the Keggin structure containing the appropriate number of vanadium atoms. The BET surface areas of POM were estimated between 5 and 17 m2/g, in agreement with earlier investigations.26,34 Pore size analyses showed that the distributions were broad, ranging from about 10 to 20 A˚. Moffat demonstrated that these micropores result from the translation and rotation of the Keggin anions.21 Figure 2 shows the XRD patterns of the different V1 and V2 heteropolyvanadates. The typical reflexion at 2θ = 9° is a typical feature of the Keggin structure.18,21 Despite a usual decrease in stability while introducing a second V-atom in the Keggin unit,19 the relative intensity was kept elevated on the powder diffraction patterns for the materials obtained via the classical procedure.24 However, the materials prepared by the hydrothermal method exhibited a reduced crystallinity. 3.2. SEM-EDX Analysis. The morphology of H4PVMo11O40 and H5PV2Mo10O40 polyoxometalates was investigated by SEM. Figure 3 shows the morphology of V1 material synthesized by hydrothermal treatment at 353 K. Whereas H4PVMo11O40 material prepared under hydrothermal conditions exhibit no crystalline microstructure but rather particles with a random morphology (Figure 3), H4PVMo11O40 and H5PV2Mo10O40 POM recrystallized

Figure 3. SEM image of V1 POM material obtained by the hydrothermal method.

Figure 4. SEM image of self-organized H5PV2Mo10O40 material.

under sulfuric acid vapors produced organized microscopically orange crystals of tenths of micrometers in size. Figure 4 shows the micrograph of V2 material, being constituted by

374

Crystal Growth & Design, Vol. 10, No. 1, 2010

cubic crystals of 70 μm in size. A deeper focus on the cracks present at the surface of the crystals shows the formation of smaller cubic crystals 100 nm in size, which tend to self-aggregate to produce larger cubic crystals (Figure 5). It is noteworthy that these peculiar tertiary structures formed by the assembly of several Keggin units have been influenced by the synthesis conditions, and hence the crystallization process. Further measurements and calculations tend to explain the growth model of these autoassembled POM materials (see section 4). Figure 6 presents the EDX mapping of P, Mo, O, and V elements within the crystal that further support an appropriate chemical composition of V2 Keggin-type material. It is noteworthy that SEM mapping confirms a homogeneous distribution of P,

Figure 5. SEM image showing the cubic building blocks present inside the cracks present at the surface of H5PV2Mo10O40 crystals.

Mothe-Esteves et al.

Mo, V, and O elements with appropriate PV2Mo10O40 stoichiometry. 3.3. 31P and 51V MAS NMR. Figure 7 shows the 31P MAS NMR spectra of the different POM either prepared by classical procedure and drying (classic), or hydrothermal (HT), or crystallization (AA). It appears that as-synthesized samples via hydrothermal treatment,25 or classical procedure then dried in oven24 exhibit large signals, thus indicating the presence of heteropolyanions with 1 or 2 V-atoms, present in the Keggin unit, represented by an array of positional isomers.18,23 These multiple-line 31P spectra were also confirmed by the 51V MAS NMR spectra35,36 (figure not shown). At present, we can conclude that these procedures led to the formation of a mixture of mono- and disubstituted species as already reported elsewhere.18,31,37 On the contrary, the 31P MAS NMR spectrum for the 11-molybdo-1-vanadophosphate crystals gave a single peak at -3.8 ppm, thus supporting the high level of organization of Keggin subunits, essentially into R-isomer.38 Crystals of 10-molybdo-2-vanadophosphate POM exhibit the same narrow signal centered at -3.8 ppm. Moreover, a second signal arose at -3.5 ppm corresponding to the presence of a minor isomer. Again the high level of organization was observed around central phosphorus atom. Figure 8 presents the 51V MAS NMR spectrum of autoorganized H5PV2Mo10O40 crystals (Figures 4-6). Besides a small shoulder at -560 ppm, an intense peak centered at -548 ppm can be seen that confirms the formation of a predominant V-species. According to Huang et al., the symmetry of such signal indicates the presence of two magnetically nonequivalent positions for vanadium36 in the crystal lattice, possibly spherical and square pyramidal. V2 polyoxometalate crystals appear to be constructed by the self-assembly of Keggin units, having the same chemical composition and similar configuration. Likewise, this

Figure 6. EDX mapping of Mo, O, P, and V elements within the V2 POM crystal.

Article

Crystal Growth & Design, Vol. 10, No. 1, 2010

Figure 7.

31

Figure 8. material.

51

375

P MAS NMR spectra of the different V-containing POM materials.

V MAS NMR spectrum of V2 POM self-organized

demonstrates that the presence of the same Keggin isomer is essential for the growth of a self-assembled structure, since no crystals could be observed with other synthesis procedures. Hence, the formation of one predominant isomer leads to a high level of organization, and consequently creates the driving force of the assembly process. Indeed, this is essential for the formation of nanocrystals and their further packing. 4. How Turn Keggin Units into Crystals? The mechanism explaining POM crystals organization and aggregation degree is not fully understood yet, due to the complexity of modeling the aggregate crystal growth.39 Recently, an outstanding contribution from Mizuno et al. shed the light on dodecatungstophosphate POM crystals formation by changing the synthesis temperature and the nature of countercations.8 Semiempirical PM6 calculations were performed herein to investigate the stability of different V-containing POM isomers. These calculations tend to support the formation of one stable isomer, with two V-atoms separated by one

Figure 9. PM6 optimized geometry of Keggin unit containing two V-atoms. Table 1 Keggin POM [PMo12O40]3[VPMo11O40]4[V2PMo10O40]5[V2PMo10O40]5[V2PMo10O40]5[V2PMo10O40]5-

isomer

A B C D

heat of formation [kcal]

relative stability [kcal/mol]

-2220.3 -2127.3 -1962.7 -1968.9 -1964.8 -1965.5

þ6.2 0 þ4.0 þ3.4

Mo atom and oxo ligands (Figure 9). Among the monomeric anionic species, this is the more stable structure. Table 1 contains the relative PM6 energy of more stable isomers. Geometry information on each isomer can be found on the Supporting Information. Their relative energies somehow reflect the balance and internal charge-charge repulsion and chemical bonding. These building units can

376

Crystal Growth & Design, Vol. 10, No. 1, 2010

Mothe-Esteves et al.

Figure 10. (a) Optimized geometry for a hydrogen-bonded trimer form of the POM. (b) Optimized geometry for a condensed trimer form of the POM.

Figure 11. XANES profiles for AA, HT, and Classic catalysts.

bind to each other through hydrogen bonding, cation bonding (e.g., H3Oþ), or by dehydration (oxygen bridged structures). In order to better understand the condensation of these Keggin units, dimerization of two Keggin structures with protons as countercations was investigated at the PM6 level. The dimerization is estimated to be favored by -42.6 kcal/ mol, indicating that hydrogen bonding in such systems is really a driving force for reversible autoassembling. The insertion of vanadium in the structure increases even more the hydrogen bonding ability of the Keggin units, leading to even more favorable condensation. It appears that the trimerization can occur either by hydrogen bonding that eventually leads to oxygen-bridged Keggin units, or via covalent V-O-V bonding. Figure 10 presents the optimized PM6 geometry of the two possible forms of trimers for these Keggin units. EXAFS-XANES experiments further confirmed the presence of distorted V-atoms as observed on XANES profiles with the signal at 5470 eV which supports the existence of an inversion center (Figure 11). In addition, the EXAFS profiles for different POM materials (Figure 12) give signals at a bond

Figure 12. EXAFS profiles for V1 and V2 AA POM and classic V1 catalysts. For comparison, the profile of V2O5 is also given.

distance of 1.60 and 3.65 A˚, respectively. These distances are both in agreement with previous studies from Ressler and Timpe,40 and with PM6 calculations (Figure 9) and can be ascribed to the V-O bond for the former and V-O-Mo for the latter with its second neighbor, in V1 and V2 AA samples. In addition, a signal located around 4.9 A˚ can be observed for these V1 and V2 self-assembled materials (Figure 12). This distance corresponds to the V-O-(H)-M inter-Keggin distance calculated for the hydrogen-bonded trimer (Figure 10a), thus supporting a noncovalent assembly process for the Keggin units. Finally, thermogravimetric (TGA) experiments were performed for V2 hydrothermally and V2 AA as-prepared materials to evaluate their water loss. Whereas the former sample loss 5.3% in weight, the latter observed a loss of 11.8% between 50 and 250 °C. This tends to confirm that the V2 AA was formed by H-bonding between water molecules and V-Keggin units. A covalent bonding between these Keggin units can therefore be barely accepted (Figure 10b), and hence support a self-assembly process via noncovalent interactions.

Article

Crystal Growth & Design, Vol. 10, No. 1, 2010

377

These findings are in line with those from Mizuno et al.8 who studied the self-organization of dodecatungstophosphate acid into either spherical or cubic crystal morphology. Herein, we report the primary self-organization of mixed vanadomolybdate POM. The substitution of molybdenum atoms by vanadium modifies the crystal growth, thus acting as a driving force for the process. 5. Catalytic Activity: Do POM Crystals Behave As True Heterogeneous Catalysts? Figure 13 shows the activity of self-assembled POM in the partial oxidation of 2-methylnaphtalene (2MN), to produce valuable K3 vitamin after 24 h. Whereas the 2MN conversion was similar among all Keggin-type polyoxometalates, the more selective catalyst was H5PV2Mo10O40 autoassembled crystalline material. This further confirms the importance of having an appropriate tertiary structure of POM in catalysis. Since several vitamins are currently synthesized by relatively nasty processes,41 we focused our study in the development of an alternative “green route” using hydrogen peroxide as in homogeneous processes.42 We have then tried to enhance the yield toward K3 vitamin while enhancing the ratio catalyst/ 2MN reactant by a factor of 4. The series so-called “optimized V2” presents an impressive 74% conversion of 2MN together with a high selectivity to desired naphtoquinone (72%), leading to a 53% yield in K3 vitamin. In order to settle a viable process, the reusability of the catalyst has to be investigated.43 The following question “does any leaching of metal from the catalyst to the solution take place?” frequently arises in systems containing H2O2. Leaching was therefore investigated to see whether POM behaves as a true heterogeneous catalyst. Recycling experiments with self-assembled H4PVMo11O40 and H5PV2Mo10O40 materials were performed. It appears that no significant loss of activity could be observed with V1 POM crystals, thus indicating its potential reusability. The aromatic substrate conversion decreased about 30% (relative value) during the second run with the V2 material (under optimized conditions), thus raising the question of leaching. While a recycling experiment without a significant loss of activity is not a sufficient proof of heterogeneity (H4PVMo11O40), a loss in activity does not necessary imply the leaching of active species (case of H5PV2Mo10O40). We have therefore collected at the reaction temperature (to avoid an eventual soluble vanadium to be readsorbed upon cooling) the V2 crystals, after 150 min and allowed the filtrate to react further. Hopefully, we found that upon hot filtration, the mother liquor did not react further. We were therefore able to conclude that even if part of the metal leaches it does not produce an active homogeneous catalyst,42 and hence the reaction remains heterogeneous. Whereas V1 and to some extent V2 Keggin self-organized materials behave similarly, a drastic decrease in activity was observed with V2 POM either obtained by hydrothermal or classical procedure. This is due to Keggin structure decomposition as already observed by Nomiya.44 EXAFS profiles confirmed the decomposition of these “non-crystalline” solids after reaction. Moreover, the profile became similar to the one observed for V2O5 (Figure 12). These aged catalysts therefore decomposed into vanadia upon reaction. The higher stability toward decomposition of crystalline self-assembled V2 material can therefore rely on its pseudovanadia core formed upon Keggin units assembly (Figure 10). This relative stability toward leaching can also be explained by the large size of organized POM crystals (tenths of micrometers) as reported

Figure 13. Catalytic activity of POM materials in the 2-methylnaphtalene oxidation and selectivity toward K3 vitamin.

by Sheldon et al. for chromium species.43,45 Indeed, the amount of Cr leached was decreased from 34% to less than 0.5% while increasing the particle size from 5 to 75 μm.45 The fact that self-assembled POM catalysts exhibit a higher selectivity and sometimes an improved activity26 means that the nature of the interaction created by Keggin units packing into nanoparticles is a key issue in catalyst performance. 6. Conclusion This study reports for the first time the microscopic arrangement of vanadium-substituted Keggin-type polyoxometalates into a three-dimensional crystalline structure. The mechanism explaining such organization and aggregation degree was investigated and both EXAFS and solid-state NMR experiments, as well as PM6 calculations confirmed a favored dimer formation (two Keggin units), followed by the formation of a trimer (three Keggin oxoanions), and evidenced the driving force of the process, being the localization of V-atoms within the Keggin unit, which guide reversible Hbond formation. Eventually these structures can further condensate into oxygen-bridged structures. Herein, we report the primary self-organization of mixed vanadomolybdate POM. The substitution of molybdenum atoms by vanadium modifies the crystal growth, thus acting as a driving force for the process. These organized V-containing POM appeared to be versatile catalysts in the liquid-phase oxidation of 2-methylnaphtalene, since their activity could be (at least partially) maintained during a second run. The occurrence of homogeneous catalysis has been excluded; a true and active heterogeneous catalyst has therefore been prepared. The further challenge will be success in combining macroscopic with microscopic and molecular design for appropriate applications. Acknowledgment. The authors heartfully thank Prof. Octavio Ceva Antunes, their guide, for steering them along the right way, keeping them on the tracks, doing his best to make sure their work will contribute scientifically. Dr. Jean-Philippe Tessonnier and Gisela Weinberg (FHI Berlin), Dr. Fabien Aussenac and Dr. Hulot (Bruker, Wissembourg) are acknowledged for their technical assistance. P.M.E. and M.M.P. acknowledge CNPq and FAPERJ. Supporting Information Available: The different geometries of Keggin units, containing two V-atoms, obtained by PM6 calculations, are presented. This material is available free of charge via the Internet at http://pubs.acs.org.

378

Crystal Growth & Design, Vol. 10, No. 1, 2010

References (1) Ramanan, A.; Whittingham, M. S. Cryst. Growth Des. 2006, 6, 2419. (2) Stein, A.; Keller, S. W.; Mallouk, T. E. Science 1993, 259, 1558. (3) Zaworotko, M. J. Cryst. Growth Des. 2007, 7, 4. (4) Lehn, J. M. Science 1985, 227, 849. (5) Lehn, J. M. Angew. Chem., Int. Ed. 1988, 27, 89. (6) Lehn, J. M. Angew. Chem. 1988, 100, 91. (7) Barton, T. J.; Bull, L. M.; Klemperer, W. G.; Loy, D. A.; McEnaney, B.; Misono, M.; Monson, P. A.; Pez, G.; Scherer, G. W.; Vartuli, J. C.; Yogui, O. M. Chem. Mater. 1999, 11, 2633. (8) Okamoto, K.; Uchida, S.; Ito, T.; Mizuno, N. J. Am. Chem. Soc. 2007, 129, 7378. (9) Whitesides, G. M.; Grzybowski, B. Science 2002, 295, 7378. (10) Ivanova, S.; Louis, B.; Ledoux, M. J.; Pham-Huu, C. J. Am. Chem. Soc. 2007, 129, 3383. (11) Pileni, M. P. Acc. Chem. Res. 2007, 40, 685. (12) Hu, Y.; Mei, T.; Guo, J.; White, T. Inorg. Chem. 2007, 46, 11031. (13) Ivanova, S.; Vanhaecke, E.; Louis, B.; Libs, S.; Ledoux, M. J.; Rigolet, S.; Marichal, C.; Pham, C.; Luck, F.; Pham-Huu, C. ChemSusChem 2008, 1, 851. (14) Keggin, J. F. Proc. R. Soc. A 1934, 144, 75. (15) Corma, A. Chem. Rev. 1995, 95, 559. (16) Hartmann, M. Angew. Chem., Int. Ed. Engl. 2004, 43, 5880. (17) Izumi, Y.; Urabe, K.; Onaka, M. Zeolite, Clay, and Heteropolyacid in Organic Reactions; VCH Weinheim: Tokyo, 1992. (18) Kozhevnikov, I. V. Chem. Rev. 1998, 98, 171. (19) Lee, K. Y.; Misono, M. Heteropoly Compounds. In Handbook of Heterogeneous Catalysis; Wiley: New York, 1997; Vol. 1, p 118. (20) Misono, M.; Okahura, T. Chemtech 1993, 23. (21) Moffat, J. B.; McMonagle, J. B.; Taylor, D. Solid State Ionics 1988, 26, 101. (22) Neumann, R. Prog. Inorg. Chem. 1998, 47, 317. (23) Pope, M. T. Heteropoly and Isopoly Oxometalates; Springer: Berlin, 1983. (24) Tsigdinos, G. A.; Hallada, C. J. Inorg. Chem. 1968, 7, 437.

Mothe-Esteves et al. (25) (26) (27) (28) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40) (41) (42) (43) (44) (45)

Kern, F.; Ruf, S.; Emig, G. Appl. Catal., A 1997, 150, 143. Arichi, J.; Eternot, M.; Louis, B. Catal. Today 2008, 138, 117. Stewart, J. J. P. J. Mol. Modeling 2007, 13, 1173. Stewart, J. MOPAC 7.1; http://mopac2007.net, 15210 Paddington Circle, Colorado Springs, CO 80921, 2007. Bonfim, R. d. P. F.; Moura, L. C. d.; Pizzala, H.; Caldarelli, S.; Paul, S.; Eon, J. G.; Mentre, O.; Capron, M.; Delevoye, L.; Payen, E. Inorg. Chem. 2007, 46, 7371. Inumaru, K.; Ono, A.; Kubo, H.; Misono, M. J. Chem. Soc., Faraday Trans. 1998, 94, 1765. Nomiya, K.; Yagishita, K.; Nemoto, Y.; Kamataki, T. J. Mol. Catal. A 1997, 126, 43. Eon, J. G.; Olivier, R.; Volta, J. C. J. Catal. 1994, 145, 318. Alekar, N. A.; Indira, V.; Halligudi, S. B.; Srinivas, D.; Gopinathan, S.; Gopinathan, C. J. Mol. Catal. A. 2000, 164, 181. Dias, J. A.; Osegovic, J. P.; Drago, R. S. J. Catal. 1999, 183, 83. Huang, W.; Schopfer, M.; Zhang, C.; Howell, R. C.; Todaro, L.; Gee, B. A.; Francesconi, L. C.; Polenova, T. J. Am. Chem. Soc. 2007, 130, 481. Huang, W.; Todaro, L.; Yap, G. P. A.; Beer, R.; Francesconi, L. C.; Polenova, T. J. Am. Chem. Soc. 2004, 126, 11564. Nomiya, K.; Nemoto, Y.; Hasegawa, T.; Matsuoka, S. J. Mol. Catal A 2000, 152, 55. Chen, J.; Gao, S.; Xu, J. Catal. Commun. 2008, 9, 728. Leroy, F.; Miro, P.; Poblet, J. M.; Bo, C.; Avalos, J. B. J. Phys. Chem. B 2008, 112, 8591. Ressler, T.; Timpe, O. J. Catal. 2007, 247, 231. Bonrath, W.; Eggersdorfer, M.; Netscher, T. Catal. Today 2007, 121, 45. Adam, W.; Herrmann, W. A.; Lin, J.; Saha-Moeller, C. R.; Fischer, R. W.; Correia, J. D. G. Angew. Chem., Int. Ed. 1994, 33, 2745. Sheldon, R. A.; Wallau, M.; Arends, I. W. C. E.; Schuchardt, U. Acc. Chem. Res. 1998, 31, 485. Nomiya, K.; Yanagibayashi, H.; Nozaki, C.; Kondoh, K.; Hiramatsu, E.; Shimizu, Y. J. Mol. Catal. A 1996, 114, 181. Lempers, H. E. B.; Sheldon, R. A. J. Catal. 1998, 175, 62.