Investigation of Reverse-Hydrogen Spillover on Zeolite-Supported

Aug 13, 2009 - The ONIOM Method and Its Applications. Lung Wa Chung , W. M. C. Sameera , Romain Ramozzi , Alister J. Page , Miho Hatanaka , Galina P. ...
0 downloads 8 Views 2MB Size
Download PDF
16070

J. Phys. Chem. C 2009, 113, 16070–16076

Investigation of Reverse-Hydrogen Spillover on Zeolite-Supported Palladium Tetramer by ONIOM Method Bulumoni Kalita and Ramesh C. Deka* Department of Chemical Sciences, Tezpur UniVersity, Tezpur 784028, Assam, India ReceiVed: February 12, 2009; ReVised Manuscript ReceiVed: June 5, 2009

The present study explores favorable chemical state of zeolite-supported Pd4 clusters using the ONIOM (ourown-N-layered integrated molecular orbital + molecular mechanics) approach, as implemented in the Gaussian03 code. Lowest-energy structures of Pd4 in the gas phase as well as on a zeolite support are found to be in the triplet state. The calculations reveal that reverse hydrogen spillover from bridging OH groups of zeolite support onto Pd4 results in a hydrogenated Pd4H/Zeo((m - 1)H), m ) 1-3 species, which are energetically preferable over the bare zeolite-supported form of Pd4/Zeo(mH). The process of single hydrogen transfer from the zeolite support to the Pd4 cluster is found to be exothermic and more favorable than twoand three-proton transfer processes. 1. Introduction Zeolites are often used as solid catalyst in the petroleum and chemical industries due to their outstanding properties, i.e., Brønsted and Lewis acidity, size-shape-selectivity, and thermal stability.1 Numerous distinct structures of zeolites have been reported in the literature.2 Among these, dealuminated or siliceous FAU zeolites have attracted special attention due to their superior thermal stability and catalytic properties,3 especially their ability to host transition metal clusters. The properties of such encapsulated metal species depend on their position inside the zeolite cavities as well as on the structure and aluminum content of the zeolite. The first step in understanding these properties is to clarify how the interaction with the zeolite framework can modify the structural and electronic properties of supported metal clusters. Over the last 30 years, a growing interest has been witnessed in the preparation and characterization of zeolite-hosted metal clusters.4-9 Metal clusters inside zeolite pores lead to the formation of uniform active sites for catalytic reactions.10 The existence of such active species has been achieved by CVD (chemical vapor deposition), ship-in-bottle, and ion-exchange methods.4,10 Zeolite hosted transition metal clusters form an important class of supported catalysts.4,6,11,12 Hydrogenated metal particles MnHxx+, denoted as “metal-proton adducts” play an important role in the catalytic conversion of methyl-cyclopentane over rhodium and palladium species in zeolite Y.13 In recent years, small palladium clusters have become an emerging field of research due to their wide application in catalytic hydrogenation, oxidation, and reduction of hydrocarbons.14,15 Small clusters of Pd are used in automotive exhaust systems to reduce toxic pollutants such as CO, NO, and hydrocarbons. Partially deactivated Pd-zeolite catalysts are found to be several times more active than conventional oxidation catalysts Pd/Al2O3, Pt/ Al2O3, and the most active oxide CeO · 6Al2O3.16 In exhaust gas treatment, catalytic reduction of nitrogen monoxide with propane takes place on zeolite-supported Pd clusters.17,18 A large number of experimental studies have been put forward to study the incorporation of small palladium clusters inside pores of FAU * Corresponding author. Fax: +91 3712 267005, E-mail address: ramesh@ tezu.ernet.in (R. C. Deka).

zeolites.19,20 The transmission electron microscopy (TEM) and extended X-ray adsorption fine structure (EXAFS) techniques have confirmed the existence of small Pdn clusters (n ) 1-4, 6, 13) inside sodalite cages and supercages of FAU zeolites.19,20 There is also evidence for pure silica-supported palladium clusters.21 Sachtler and Zhang11 proposed that an electrondeficient state of a metal cluster formed in a zeolite cavity can result from its interaction with zeolitic protons. Transition metal atom or clusters may undergo considerable change in electronic structure and thus in their catalytic behavior through the interaction with zeolite framework. Performance and selectivity of the catalyzed processes depend on the nature of metals as well as the acidic sites of zeolites. Metal species can have both electron-deficient and electron-enriched states in zeolite cavities on the basis of experimental data.6 Electron-deficient states arise as a result of interaction of the metal particles with Brønsted acid sites of zeolites,4,11 while basic framework oxygen centers of zeolite can attribute to the electron-enriched metal state.6 Moreover, electron-donor or electron-acceptor species and interchannel cations, present in the zeolite cages, can also affect the electronic structure and properties of the deposited metal particles. To understand the protonation of zeolite-supported small palladium clusters and the effect of protonation on interaction of these clusters with CO probe molecules, Yakovlev et al.22 carried out density functional based investigation on free and protonated Pd4 and Pd6 clusters. From this study, they showed that electron-deficient [PdnH]+ species can be formed in faujasites (e.g., Pd/NaY) as a result of interaction of guest metal particles with zeolitic protons. It was also found that protonation reduces the CO adsorption energy and increases the vibrational frequency of adsorbed CO. Harmsen et al.23 reported a model description of the interaction of a palladium atom with zeolitic Brønsted acid site. They found that the reduction of Pd2+ to Pd0 and 2H+ is strongly exothermic. Recently, Morokuma and co-workers24 have studied the reactivity of zeolite-supported palladium tetramers toward hydrogen molecules. They found that embedding of Pd4 on a zeolite reduces the barrier for H2 addition by 7.53 kJ/mol from that in the gas-phase Pd4 cluster. The complicated phenomenon of metal-support interaction comprises a variety of mechanisms such as spillover, bifunc-

10.1021/jp901313n CCC: $40.75  2009 American Chemical Society Published on Web 08/13/2009

Reverse-H Spillover on Zeolite-Supported Pd Tetramer tional catalysis, and sintering. Theoretical studies reveal that supported metal clusters can easily be contaminated by reactions with surface-active sites such as OH groups.25,26 During the past few years, Ro¨sch and co-workers25-27 have developed a systematic study of reverse hydrogen spillover from the hydroxyl groups of zeolite support onto four-atom and six-atom clusters. In a computational study of Rh6 clusters in close contact with zeolite support, Vayssilov et al.25 reported that reverse spillover of protons of zeolite OH groups onto supported Rh6 cluster results in an oxidation of metal atoms. An extended study on zeolite-supported transition metal clusters M6 of platinum and gold groups (Fe, Co, Ni, Cu, Ru, Rh, Pd, Ag, Os, Ir, Pt, Au) resulted in an energetically favorable reverse hydrogen spillover process for all the metals.26 Petrova et al.27a observed that stepwise reverse hydrogen spillover from bridging OH groups of zeolite onto Ir4 clusters is an exothermic process, releasing a total energy of -421 kJ/mol for the three steps studied. An accurate quantum mechanics/molecular mechanics (QM/MM) approach utilizing a variant of the elastic polarizable environment (EPE) scheme adapted for systems with polar-covalent bonds (covEPE)28,29 has been applied to investigate the favorable chemical state of zeolite-supported M6 metal clusters (M ) Rh, Ir, Au). Unlike the simple finite model approaches, this study has predicted that reverse hydrogen spillover from less and more acidic OH groups onto Au6 is endothermic and exothermic, respectively.27c Mikhailov et al.30 have studied the interaction of Pt6 cluster with different oxides including FAU and ZSM-5 zeolites. In spite of the large number of studies, the reactivity of the Pd4 cluster with hydroxyl groups of the zeolite support has not yet been investigated either experimentally or theoretically to the best of our knowledge. In the present study, we apply QM/ MM approach implemented in the ONIOM scheme of the Gaussian03 program for the first time to investigate the reverse spillover of hydrogen from the hydroxyl groups of faujasite zeolite support onto the Pd4 cluster.

J. Phys. Chem. C, Vol. 113, No. 36, 2009 16071 structure (Figure 1b,e,i), each including one supercage, where the metal species and adsorbates can be trapped inside. One, two, and three Si atoms of the 6T ring of Zeo1, Zeo2, and Zeo3 are isomorphously substituted by one, two, and three Al atoms, respectively. Substitutions of Si by Al in Zeo2(2H) and Zeo3(3H) are made in an alternating sequence according to the Lo¨wenstein rule.34 The extra negative charges generated due to Al substitution are balanced by adding one, two, and three protons to corresponding number of oxygen atoms at O1 crystallographic positions of Zeo1, Zeo2, and Zeo3, respectively. There are six bridging oxygens in the six-ring of the zeolite clusters with alternating bending in and out positions and are denoted throughout as O2 and O1, respectively (Figure 1b,e,i). In our study, the high level consists of a 6T ring facing the supercage of FAU zeolite and the Pd4 cluster, while the rest of the zeolite framework constitutes the low level. We have used the B3LYP35 functional to perform the high-level (QM) calculations in the ONIOM scheme. The 6-31G(d,p) basis set has been employed on all electrons in H, O, Al, and Si, whereas the heavier element Pd has been treated by the LANL2DZ relativistic pseudopotential.36 We have used molecular mechanics force field, UFF, to treat the low level of our system. In an earlier study37 on adsorption in zeolites, it has been observed that the ONIOM2 (B3LYP/6-31G(d,p):UFF) method provides reasonable values corresponding to the experimental prediction. To reduce the errors due to inaccuracies in the force field parameters and to simulate the zeolite pore structure correctly, we have fixed all the outer layer atoms to their respective crystallographic positions. Only the active region of the zeolite and the Pd4 cluster are allowed to relax during structure optimization. At the outer boundary of the MM layer, O atoms have been replaced by H atoms along the Si-O covalent bonds and the Si-H distance is fixed to 1.47 Å. To determine the reactivity of the metal clusters with the OH group of the zeolite fragment, we have calculated the energy of reverse hydrogen spillover (ERS) from the bridging OH groups to the supported Pd4 cluster using the following formula:

2. Computational Details We have carried out embedded-cluster modeling (QM/MM) of Pd4/FAU catalysts by using two-layer ONIOM31 scheme available in the Gaussian03 programs.32 According to the twolayer ONIOM approach, energy calculation can be simplified by treating the active region (i.e., the active Brønsted acidic site of a zeolite catalyst) with a high-level quantum mechanical (Hartree-Fock (HF)) or density functional (DF) level and the extended framework environment with a less expensive level, HF, semiempirical, and molecular mechanics force field methods. The simplified expression of total energy of the whole system within the framework of ONIOM methodology can be given as Real Cluster Cluster EONIOM2 ) ELow + (EHigh - ELow )

where the superscript “Real” means the whole system and the superscript “Cluster” means the active region. Subscripts “High” and “Low” mean high- and low-level calculations used in the ONIOM method. Due to the limitation of computational resources and time consumption, we have treated the QM part more accurately with the DFT method, while the rest of the model is approximated by the molecular mechanics method using a universal force field (UFF).33 In the present work, the zeolite support is represented by three 60T (T ) tetrahedral atoms) model clusters Zeo1, Zeo2, and Zeo3 of faujasite

ΕRS ) {Ε[Ρd4Hn /Ζeo((m - n)Η)] Ε[Pd4 /Ζeo(mΗ)]}/n m, n ) 1 - 3 (1) In the results to follow, the individual energy (Etrans) in stepwise (n ) 1-3) spillover process of each hydrogen transfer is calculated as

Εtrans ) Ε[Pd4Hn /Ζeo((m - n)Η)] Ε[Pd4Ηn-1 /Ζeo((m + 1))-n)Η)] (2) In addition, adsorption energy (Eads) of bare Pd4 cluster on the neutral zeolite fragment has been evaluated as

Εads ) Ε[Pd4 /Ζeo(mΗ) - Ε[Ζeo(mΗ)] - Ε[Pd4]

(3) Negative values of all the energies defined above show that the final state of transformation is more stable than the initial state. 3. Results and Discussion 3.1. Geometry and Energetics. 3.1.1. Gas-Phase Pd4 Clusters. In the present investigation, we have first optimized the structure of a gas-phase Pd4 cluster. A symmetry-unrestricted

16072

J. Phys. Chem. C, Vol. 113, No. 36, 2009

Kalita and Deka

Figure 1. Optimized lowest-energy structures of (a) gas-phase Pd4 cluster; ONIOM2 model clusters of (b) Zeo1(1H); (c) Pd4/Zeo1(1H); (d) Pd4H/ Zeo1; (e) Zeo2(2H); (f) Pd4/Zeo2(2H); (g) Pd4H/Zeo2(1H); (h) Pd4H2/Zeo2; (i) Zeo3(3H); (j) Pd4/Zeo3(3H); (k) Pd4H/Zeo3(2H); (l) Pd4H2/Zeo3(1H); and (m) Pd4H3/Zeo3.

full geometry optimization led bare Pd4 to a distorted tetrahedron (C1 symmetry) with the triplet state as the lowest-energy structure (Figure 1a). Previous investigations38-40 have also established the ground electronic state of isolated Pd4 to be a triplet in a slightly distorted tetrahedral shape. We have computed the Pd-Pd bond distance of 2.61-2.72 Å in the gasphase Pd4 cluster. This value is in agreement with earlier studies.38,39 From our calculations, it is found that the difference in energy between singlet (S) and triplet (T) states (C1 symmetry) of Pd4 is 68.99 kJ/mol and the singlet state has Pd-Pd distances of 2.58-2.82 Å. These values are close to the S-T difference of 70.29 kJ/mol and Pd-Pd distances of 2.58-2.84 Å in the singlet state of Pd4, as observed by Moc et al.38 3.1.2. Zeolite-Supported Bare Pd4 Clusters. The ONIOM2 level optimized structures of faujasite zeolites (Zeo1, Zeo2, and Zeo3) are shown in Figure 1b,e,i, respectively. Important structural parameters of the QM region are summarized in Table 1. Our calculations reveal that Al-O1 distances (1.89 Å in Zeo1, 1.87 Å in Zeo2, and 1.85 Å in Zeo3) are higher than Al-O2 distances (1.73 Å in Zeo1, 1.74 Å in Zeo2, and 1.75 Å in Zeo3), which agrees with the earlier theoretical and experimental observations41,42 that the Al-O distance associated with the acidic proton is significantly longer than the other Al-O distances, by up to 0.2 Å. The values of the Al-O1 distances are found to be close to that calculated from EXAFS multiple scattering studies of H-FAU.43 Al(O1)-Si, Al(O2)-Si, O2-O2,

and O1-H distances in Zeo1, Zeo2, and Zeo3 are in agreement with the observation of Ro¨sch and co-workers.27c Our calculated values of O1-H distances (0.97 Å) agrees well with the ONIOM2 (B3LYP/6-31G(d,p):UFF) result of Kasuriya et al.37 The resulting structures of embedded-cluster models of 60T zeolites have been chosen to study the adsorption complexes of zeolite-supported palladium clusters. The lowest-energy Pd4 cluster is placed orienting its base toward the 6T ring of the optimized zeolite supercages. Orientation of Pd4 cluster is so chosen that the Pdz atoms are coordinated to the oxygen atoms at O2 positions, which do not participate in the bridging OH groups. Labeling of Pdz and Pdt atoms are shown in the lowest energy structures of zeolite-supported bare Pd4 cluster, Pd4/ Zeo1(1H), Pd4/Zeo2(2H), and Pd4/Zeo3(3H), calculated at the ONIOM2 (B3LYP/6-31G(d,p), LANL2DZ:UFF) level (Figure 1c,f,j). To test the spin multiplicity of the Pd4 cluster on the zeolite-support, we have performed calculations for the Pd4/ Zeo3(3H) system in singlet and triplet states. It has been observed that, like isolated Pd4, zeolite-supported Pd4 cluster has a triplet ground state (Table 1). The singlet state of Pd4/ Zeo3(3H) complex lies 79.07 kJ/mol higher in energy than the triplet state. Recent observation has also confirmed the triplet ground state of Pd4 on the zeolite-support.24 Therefore, we have carried out the calculations for Pd4/Zeo1(1H) and Pd4/Zeo2(2H) complexes in triplet states only. Our computed structural characteristics of the lowest-energy Pd4/Zeo(3H) complexes are given in Table 1. It is found that each of Pd4/Zeo1(H), Pd4/

Zeo1(H)

-

6.38 3.44 3.09 1.89 1.73 1.65 1.64 4.06 0.97 -

0.39

-

-

Pd4

energies (kJ/mol)

charges

average distances (Å)

-

2.66 -

-

-

-

Pdz-Pdz

Pdz-Pdt Al-Simax Al(O1)-Si Al(O2)-Si Al-O1 Al-O2 Si-O1 Si-O2 O2-O2 O1-H Pdz-O1 Pdz-O2 Pdz-Si Pdz-Al Pdz-H/Pdt-H

Pdt Pd4 H1/H2/H3

Pd4Hn Eads

ERS Etrans

calculated parameters

-

-0.23 -48.75(-8.75)

-0.19 -0.05 -0.62 0.39

2.61 2.66 6.42 3.43 3.16 1.86 1.77 1.64 1.65 4.13 1.04 3.45 2.48 3.38 3.20 -

2.72

Pd4/ Zeo1(H)

-61.35 -61.35

-0.06 -

-0.04 0.02 -0.10 0.05

2.73 2.73 6.41 3.27 3.22 1.75 1.82 1.64 1.67 4.06 3.40 2.30 3.24 3.08 1.87/1.69

2.72

Pd4H/ Zeo1

-

-

0.39

6.41 3.41 3.08 1.87 1.74 1.67 1.62 3.99 0.97 -

-

Zeo2(2H)

-

0.11 -74.09(-11.30)

-0.21 -0.04 -0.68 0.39/0.39

2.61 2.67 6.48 3.42 3.15 1.85 1.76 1.66 1.64 4.14 1.02 3.35 2.41 3.34 3.28 -

2.72

Pd4/ Zeo2(2H)

-65.16 -65.16

0.24 -

-0.07 0.03 -0.19 0.05/0.38

2.73 2.72 6.44 3.33 3.17 1.80 1.81 1.67 1.66 4.01 0.97 3.41 2.26 3.22 3.19 1.86/1.68

2.71

Pd4H/ Zeo2(H)

-10.16 44.84

0.21 -

2.92 2.79 6.45 3.29 3.24 1.78 1.83 1.64 1.66 4.08 3.28 2.57 3.00 3.49 1.62/1.72, 1.66/1.66 0.02 0.07 0.11 0.02/0.07

2.66

Pd4H2/ Zeo2

complexes

-

-

0.38/0.38/0.38

6.45 3.38 3.07 1.85 1.75 1.71 1.60 3.91 0.97 -

-

Zeo3(3H)

-

-72.65(-20.39)

-0.26 0.01 -0.78 0.39/0.39/0.39

2.61 2.67 6.55 3.41 3.14 1.83 1.76 1.69 1.61 4.17 1.01 3.26 2.36 3.30 3.33 -

2.72

Pd4/ Zeo(3H)

TABLE 1: Important Structural Parameters, Mulliken Charges, and Energetics of Gas-Phase and Zeolite-Supported Pd4 Clusters

-33.32 -33.32

-0.08 0.08 -0.16 -0.05/0.41/ 0.39 0.59 -

2.74 2.73 6.50 3.34 3.15 1.81 1.79 1.66 1.64 3.95 0.97 3.42 2.24 3.22 3.20 1.80/1.67

2.72

Pd4H/ Zeo(2H)

12.40 58.12

2.83 2.76 6.48 3.29 3.19 1.77 1.80 1.63 1.66 3.94 0.97 3.36 2.20 3.24 3.10 1.66/1.71, 1.62/1.70 -0.01 0.13 0.10 0.05/0.03/ 0.38 0.56 -

2.68

Pd4H2/ Zeo(1H)

42.07 101.40

0.33 -

2.84 2.83 6.43 3.25 3.22 1.73 1.79 1.61 1.67 3.97 3.33 2.23 3.35 2.97 1.65/1.68, 1.65/1.70, 1.52/2.64 0.01 0.12 0.15 0.06/0.05/0.07

2.81

Pd4H3/ Zeo

Reverse-H Spillover on Zeolite-Supported Pd Tetramer J. Phys. Chem. C, Vol. 113, No. 36, 2009 16073

16074

J. Phys. Chem. C, Vol. 113, No. 36, 2009

Zeo2(2H), and Pd4/Zeo3(3H) complexes exhibits three shorter Pd-O2 bond distances of 2.31-2.59 Å, 2.32-2.58 Å, and 2.36-2.37 Å (with longer Pd-O1 bond distances of 3.23-3.64 Å, 3.21-3.65 Å, and 3.25-3.27 Å), respectively. These values are shorter in comparison with the study of Morokuma and coworkers.24 Our calculated Pd-O distances are ∼0.15-0.40 Å shorter than the values of 2.74-2.76 Å derived from EXAFS measurements of NaX FAU zeolite-supported Pdn (n ) 2-4) clusters.19a However, this experiment gives 1.5 Pd and 2.1 O neighbors for each Pd atom, which is different from the present study. In spite of this, Pd4/NaX FAU zeolite data are presented here for comparison because of the unavailability of EXAFS structural data for Pd4/H-FAU zeolite system. The Pd-O2 bond distances in the Pd4/Zeo(3H) complex are close to those reported by Vayssillov and Ro¨sch26 for Pd6/Zeo(3H) complex (2.28 Å) and are essentially close to the corresponding distances of Pd3 clusters supported on the R-Al2O3 (0001) surface, optimized without any symmetry constraints.44 It is observed that Si-O1 and Al-O1 bond lengths decrease and Si-O2 and Al-O2 bond lengths increase in the faujasite zeolite support due to the adsorption of Pd4. As seen in Table 1, Pd-Pd bond lengths of of zeolite-supported Pd4 clusters are 2.59-2.76 Å, which are slightly deviated from those in the gas-phase Pd4 cluster. This finding is similar to the observation of Bussai et al.45 that adsorption of a metal cluster in a zeolite cage does not change the structure of the cluster in a significant way compared with a tetrahedral structure in the gas phase. A similar observation has also been reported in the case of purely siliceous FAU zeolite-supported palladium tetramer24 as well as in Pd6 cluster on a constrained six-ring model FAU zeolite cluster.26 Almost unchanged structural parameters along with the preserved tetrahedral shape of the gas-phase Pd4 cluster on the zeolitesupport indicate a weak interaction between the palladium cluster and the support. 3.1.3. Zeolite-Supported Hydrogenated Pd4 Clusters. Figure 1d; g, h; k, l, m shows the ONIOM2 model optimized adsorption complexes of Pd4H/Zeo1; Pd4H/Zeo2(1H), Pd4H2/Zeo2; Pd4H/ Zeo3(2H), Pd4H2/Zeo3(1H), and Pd4H3/Zeo3, respectively, corresponding to one, two, and three proton transfer processes in Zeo1, Zeo2, and Zeo3, respectively. It is found that, in the hydrogenated Pd4 adsorbed zeolite complexes, the transfer of one and two hydrogen atoms got adsorbed in twofold coordination between Pdz and Pdt atoms. However, the third hydrogen in the Pd4H3/Zeo3 complex is coordinated with only a Pdz atom. Characteristic features of these complexes are given in Table 1. In hydrogenated Pd4/Zeo clusters, elongation of the average Pd-Pd bond length is observed by 0.05-0.07 Å compared to that of adsorbed bare Pd4 clusters. Strong interaction of the metal cluster with the zeolite supports results from reverse hydrogen spillover, which is reflected by the shorter Pd-O2 distances compared to the adsorbed bare cluster (Table 1). Bond length values of T-O1 and T-O2 (T ) Si, Al) in hydrogenated Pd4 clusters are found to be less and greater, respectively, in comparison with those in bare Pd4 clusters on zeolite support. 3.2. Charge Analysis. On the basis of finite model as well as embedded cluster studies, it has been already established that most of the bare metal clusters (M4, M6) on zeolite supports are internally polarized, with uneven distribution of charges between Mz and Mt metal atoms. Mulliken charge analysis in the present work reveals a qualitatively similar charge polarization effect of Pd4 on Zeo1, Zeo2, and Zeo3 supports (Table 1). However, unlike earlier observations,26,27a we have found that metal atoms Pdz close to the framework oxygen atoms O2 of a zeolite fragment carry higher negative charges while the apex

Kalita and Deka Pdt atom carries small negative or positive charges. A similar charge distribution has been observed in the case of the Au6 cluster on the zeolite support by Ivanova Shor et al.27c The adsorbed Pd4 clusters in our study are found to carry a total negative charge of -0.62 to -0.78 e, which is larger than the value of -0.18 e reported for adsorbed Pd6 cluster in a finite zeolite model.26 Further, in line with the earlier results,26,27a,c we have observed that Pd4Hn (n ) 1-3) species formed as a result of reverse hydrogen spillover carry positive charges, 0.21-0.59 e (with small negative charge of -0.06 for Pd4H/ Zeo1). It is seen from Table 1 that reverse hydrogen spillover results in the decrease and increase of electron density of metal atoms and transferred hydrogen species, respectively. Therefore, the process of reverse hydrogen spillover is associated with a partial oxidation of the metal clusters. The shift of electron density from metal atoms to hydrogens is larger for Pdz atoms, which can interact with the transferred protons more easily than apex Pdt atoms. Electron density is less polarized in the adsorbed metal hydride species, Pd4Hn (n ) 1-3) compared to adsorbed bare Pd4 species. This redistribution of electron density in zeolite-supported metal clusters may be important for their reactivity and catalytic activity. 3.3. Energies of Adsorption and Reverse Hydrogen Spillover. The calculated energies of adsorption and reverse hydrogen spillover are tabulated in Table 1. It should be noted that the energy values do not include basis-set superposition error (BSSE) corrections. The facility of performing counterpoise corrections in ONIOM calculations of Gaussian03 is not available at this point. However, we have calculated the BSSE using counterpoise method in the fully optimized QM geometries of Pd4 adsorbed zeolite complexes as reported by Joshi et al.46 We have performed pure QM calculations on Pd4 adsorbed in 6T ring of faujasite zeolite clusters. In the first step of these calculations, we have allowed only the terminating Si-H bonds to relax by keeping all other atoms fixed to their respective crystallographic positions. In the subsequent step, the terminating hydrogen atoms are kept fixed, and all other atoms are allowed to relax. The values of BSSE in Pd4/Zeo1(1H), Pd4/ Zeo2(2H), and Pd4/Zeo3(3H), estimated from pure QM calculations are 50.56, 70.00, and 71.41 kJ/mol, respectively, giving positive values of corrected Eads. These are only approximated corrections, as they have been calculated from pure QM calculations on adsorption complexes of Pd4 with small zeolite clusters and not from QM/MM calculations on more realistic Pd4 adsorption zeolite model complexes. Therefore, we have not included the BSSE corrected adsorption energies in Table 1. A BSSE value of 54 kJ/mol has been reported by Moc et al.24 for the adsorption energy of Pd4 cluster on the 24T model cluster of purely siliceous faujasite zeolite. We have calculated the MM contribution (Eads,MM) to the total QM/MM adsorption energy (Eads) as Eads,MM ) Eads - Eads,QM, where Eads,QM is the pure QM adsorption energy. These values are shown in parentheses of Table 1. Adsorption energies of Pd4 on Zeo1, Zeo2, and Zeo3 have been found to be -48.75, -74.09, and -72.65 kJ/mol, respectively, using eq 3. These values are less than the adsorption energy (BSSE uncorrected) of -84.94 kJ/ mol for Pd4 on 24T model cluster of purely siliceous FAU zeolite.24 An earlier study showed that the adsorption energy of the six-atom octahedral Pd6 cluster on a 6T zeolite cluster is -56 kJ/mol.26 It suggests that Pd4 has stronger binding than Pd6 with FAU zeolite. This difference in adsorption energy may be due to the lower coordination saturation of the metal atoms in the free Pd4 cluster, which results in a comparatively stronger interaction with oxygen atoms of the zeolite support. A similar

Reverse-H Spillover on Zeolite-Supported Pd Tetramer observation has been reported in the case of Ir4 and Ir6 clusters.26,27a Our calculations reveal that binding of a Pd tetramer to the zeolite is thus weaker than that to a MgO support (-164.03 kJ/mol).47 In the present study, zeolite-supported Pd4 clusters in the triplet ground state are found to be more stable in the configurations with 1H impurity (Figure 1d,g,k). We have calculated that proton migration from bridging OH group of the Zeo1 support to the Pd4 cluster is exothermic, which is given by ERS value of -61.35 kJ/mol (Table 1). The corresponding process in the singlet state of Pd4/Zeo(3H) is less exothermic with ERS value of -54.28 kJ/mol. Values of ERS corresponding to single proton transfer in Pd4H/Zeo2(1H) and Pd4H/Zeo3(2H) complexes are calculated to be -65.16 and -33.32 kJ/mol, respectively, showing the exothermic nature of these processes. Migration of both protons from the Zeo2 support to Pd4 is found to be less exothermic, with ERS of -10.16 kJ/mol. On the other hand, transfer processes of two and three protons in Pd4H2/ Zeo3(1H) and Pd4H3/Zeo3, respectively, take place at the expense of energies (ERS), 12.40 and 42.07 kJ/mol, respectively. However, in the singlet excited states of Zeo3 adsorption complexes, reverse hydrogen spillover processes for transferring two and three protons are found to be exothermic (ERS ) -4.30 kJ/mol) and endothermic (ERS ) 19.51 kJ/mol), respectively. This observation contradicts the previous finding of exothermicity of three hydrogen transfer process in the case of Pd6/ Zeo(3H) complex with finite models of the zeolite framework.26 Recently, the elaborate embedded cluster model study computed reduced ERS values in comparison to finite model studies. This study showed that zeolite-supported hydrogenated M6H3 (M ) Rh, Ir, Au) species are not always energetically favorable over supported bare M6 clusters.27c In this study, it is also established that the nature of reverse hydrogen spillover depends on the acidity of the corresponding OH groups, and the process proceeds more easily from the most acidic hydroxyl groups. With these results in mind, there is a possibility of a change in the nature of the spillover process in Pd4/Zeo(mH), m ) 1-3, systems with the change of corresponding OH group, which may be a subject of further investigations. However, there is no theoretical study on zeolite-supported metal tetramers with proper modeling of the extended zeolite framework until the present time. On the basis of the present investigation, we can conclude that reverse hydrogen spillover processes of more than one proton transfer are energetically unfavorable for the Pd4 cluster on a faujasite zeolite support. 4. Summary and Conclusions We have studied faujasite supported bare Pd4 clusters and their interaction with OH groups of the support to form hydrogenated Pd4 clusters. The calculations have been carried out using the QM/MM based ONIOM2 (B3LYP:UFF) approach of Gaussian03 to make modeling of the zeolite more realistic. The average Pd-Pd bond length of the gas-phase Pd4 cluster remains almost unchanged in Pd4/Zeo(mH), m ) 1-3, complexes. Elongation of the Pd-Pd bond lengths are found to be ∼0.05-0.07 Å, going from Pd4/Zeo(mH) to Pd4Hn/Zeo((m 1)H) (m, n ) 1-3) complexes. The presence of the support is effective in producing internal charge polarization in bare as well as hydrogenated Pd4 clusters. Pd4 clusters has been observed to undergo partial oxidation due to the proton transfer process; the metal atoms closer to the zeolite oxygens are more oxidized than those located farther away. Taking into account the effects of the zeolite environment properly, it has been found that reverse hydrogen spillover processes for single proton

J. Phys. Chem. C, Vol. 113, No. 36, 2009 16075 transfer from zeolite OH groups to the metal clusters are exothermic with ERS values of -61.35, -65.16, and -33.32 kJ/mol in the case of Zeo1, Zeo2, and Zeo3, respectively. On the other hand, stepwise transfer processes of two protons in Zeo2 and Zeo3 are endothermic with Etrans values of 44.84 and 58.12 kJ/mol, respectively, in the triplet states. The value of Etrans for the third proton transfer in the Pd4H2/Zeo3 complex is 101.40 kJ/mol. In summary, we have observed that, for zeolitesupported tetranuclear palladium clusters, Pd4H/Zeo((m - 1)H), m ) 1-3 in the triplet states is the more stable complex and may be useful as an active species in catalytic processes. Acknowledgment. This work was supported by the Department of Science and Technology, New Delhi. B.K. thanks the Council of Scientific and Industrial Research, New Delhi, for a research fellowship. References and Notes (1) Catlow, C. R. A., Ed.; Modelling of Structure and ReactiVity in Zeolites, Academic Press: London, 1992. (2) (a) Meier, W. M.; Olson, D. H. Atlas of Zeolite Structure Types, 3rd revised ed.; Butterworth-Heinemann: London, 1992; (b) Baerlocher, C.; Meier, W. M.; Olson, D. H., Eds.; Atlas of Zeolite Framework; Elsevier: Amsterdam, 2001. (3) (a) Hriljac, J. A.; Eddy, M. M.; Cheetham, A. K.; Donohue, J. A.; Ray, G. J. J. Solid State Chem. 1993, 106, 66. (b) Colligan, M.; Forster, P. M.; Cheethamm, A. K.; Lee, Y.; Vogt, T.; Hriljac, J. A. J. Am. Chem. Soc. 2004, 126, 12015. (c) Henson, N. J.; Cheetham, A. K.; Stockenhuber, M.; Lercher, J. A. J. Chem. Soc., Faraday Trans. 1998, 94, 3759. (4) Sachtler, W. M. H. Acc. Chem. Res. 1993, 26, 383. (5) Alexeev, O.; Gates, B. C. Top. Catal. 2000, 10, 273. (6) Barthomeuf, D. Catal. ReV. 1996, 38, 521. (7) Hartmann, M.; Bischof, C.; Luan, Z.; Kevan, L. Microporous Mesoporous Mater. 2001, 44-45, 385. (8) Stakheev, A. Yu.; Shpiro, E. S.; Jaeger, N. I.; Schulz-Ekloff, G. Catal. Lett. 1995, 32, 147. (9) Nakano, T.; Takase, T.; Araki, S.; Kamiyama, T.; Nozue, Y.; Ikeda, S. Nucl. Instrum. Methods Phys. Res., Sect. A 2008, in press. (10) Okumura, K.; Yoshimoto, R.; Uruga, T.; Tanida, H.; Kato, K.; Yokota, S.; Niwa, M. J. Phys. Chem. B 2004, 108, 6250. (11) Sachtler, W. M. H.; Zhang, Z. AdV. Catal. 1993, 39, 129. (12) Gates, B. C. Chem. ReV. 1995, 95, 511. (13) McCarthy, T. J.; Lei, G. -D.; Sachtler, W. M. H. J. Catal. 1996, 159, 90. (14) Ciebien, J. F.; Cohen, R. E.; Duran, A. Supramol. Sci. 1998, 5, 31. (15) Bertani, V.; Cavallotti, C.; Masi, M.; Carra`, S. J. Phys. Chem. A 2000, 104, 11390. (16) Slovetskaya, K. I. Russ. Chem. Bull. 2004, 53, 2168. (17) Tanabe, S.; Matsumoto, H. J. Mater. Sci. Lett. 1994, 13, 1540. (18) Matsumoto, H.; Tanabe, S. J. Phys. Chem. 1995, 99, 6951. (19) (a) Moller, K.; Koningsberger, D. C.; Bein, T. J. Phys. Chem. 1989, 93, 6116. (b) Moller, K.; Bein, T. J. Phys. Chem. 1990, 94, 845. (c) Zhang, Z.; Chen, H.; Sheu, L.-L.; Sachtler, W. M. H. J. Catal. 1991, 127, 213. (d) Bai, X.; Sachtler, W. M. H. J. Catal. 1991, 129, 121. (e) Zhang, Z.; Chen, H.; Sachtler, W. M. H. J. Chem. Soc., Faraday Trans. 1991, 87, 1413. (f) Kim, J. G.; Ihm, S. K.; Lee, J. Y.; Ryoo, R. J. Phys. Chem. 1991, 95, 8546. (g) Zhang, Z.; Sachtler, W. M. H. J. Mol. Catal. 1991, 67, 349. (h) Ryoo, R.; Cho, S. J.; Pak, C.; Kim, J. G.; Ihm, S. K.; Lee, J. Y. J. Am. Chem. Soc. 1992, 114, 76. (i) Stakheev, A. Yu.; Sachtler, W. M. H. J. Chem. Soc.,Faraday Trans. 1991, 87, 3703. (j) Beutel, T.; Zhang, Z.; Sachtler, W. M. H.; Kno¨zinger, H. J. Phys. Chem. 1993, 97, 3579. (k) Sordelli, L.; Martra, G.; Psaro, R.; Dossi, C.; Coluccia, S. J. Chem. Soc., Dalton Trans. 1996, 765. (l) Vogel, W.; Kno¨zinger, H.; Carvill, B. T.; Sachtler, W. M. H.; Zhang, Z. C. J. Phys. Chem. B 1998, 102, 1750. (20) Jiang, Y.-X.; Weng, W.-Z.; Si, D.; Sun, S.-G. J. Phys. Chem. B 2005, 109, 7637. (21) Yokoyama, T.; Kimoto, S.; Ohta, T. Physica B 1989, 158, 255. (22) Yakovlev, A. L.; Zhidomirov, G. M.; Neyman, K. M.; Nasluzov, V. A.; Ro¨sch, N. Ber. Bunsen Ges. Phys. Chem. 1996, 100, 413. (23) Harmsen, R.; Bates, S.; van Santen, R. A. Faraday Discuss. 1997, 106, 443. (24) Moc, J.; Musaev, D. G.; Morokuma, K. J. Phys. Chem. A 2008, 112, 5973. (25) Vayssilov, G. N.; Gates, B. C.; Ro¨sch, N. Angew. Chem., Int. Ed. 2003, 42, 1391. (26) Vayssilov, G. N.; Ro¨sch, N. Phys. Chem. Chem. Phys. 2005, 7, 4019.

16076

J. Phys. Chem. C, Vol. 113, No. 36, 2009

(27) (a) Petrova, G. P.; Vayssilov, G. N.; Ro¨sch, N. Chem. Phys. Lett. 2007, 444, 215. (b) Petrova, G. P.; Vayssilov, G. N.; Ro¨sch, N. J. Phys. Chem. C 2007, 111, 14484. (c) Ivanova Shor, E. A.; Nasluzov, V. A.; Shor, A. M.; Vayssilov, G. N.; Ro¨sch, N. J. Phys. Chem. C 2007, 111, 12340. (28) Nasluzov, V. A.; Ivanova Shor, E. A.; Shor, A. M.; Vayssilov, G. N.; Birkenheuer, U.; Ro¨sch, N. J. Phys. Chem. B 2003, 107, 2228. (29) Ivanova Shor, E. A.; Shor, A. M.; Nasluzov, V. A.; Vayssilov, G. N.; Ro¨sch, N. J. Chem. Theory Comput. 2005, 1, 459. (30) (a) Mikhailov, M. N.; Kustov, L. M.; Mordkovich, V. Z. Russ. Chem. Bull., Int. Ed. 2007, 56, 397. (b) Mikhailov, M. N.; Kustov, L. M.; Kazansky, V. B. Catal. Lett. 2008, 120, 8. (31) Dapprich, S.; Komaromi, I.; Byun, K. S.; Morokuma, K.; Frisch, M. J. J. Mol. Struct. THEOCHEM 1999, 461, 1. (32) 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 A.1; Gaussian, Inc.: Pittsburgh, PA, 2003. (33) Rappe, A. K.; Casewit, C. J.; Colwell, K. S.; Goddard, W. A.; Skiff, W. M. J. Am. Chem. Soc. 1992, 114, 10024.

Kalita and Deka (34) Lo¨wenstein, W. Am. Mineral. 1954, 39, 92. (35) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (c) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200. (d) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem. 1994, 98, 11623. (36) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270. (b) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299. (37) Kasuriya, S.; Namuangruk, S.; Treesukol, P.; Tirtowidjojo, M.; Limtrakul, J. J. Catal. 2003, 219, 320. (38) Moc, J.; Musaev, D. G.; Morokuma, K. J. Phys. Chem. A 2000, 104, 11606. (39) Gomes, J. R. B.; Lodziana, Z.; Illas, F. J. Phys. Chem. B 2003, 107, 6411. (40) Zhang, W.; Ge, Q.; Wang, L. J. Chem. Phys. 2003, 118, 5793. (41) Eichler, U.; Bra¨ndle, M.; Sauer, J. J. Phys. Chem. B 1997, 101, 10035. (42) Ehresmann, J. O.; Wang, W.; Herreros, B.; Luigi, D. -P.; Venkatraman, T. N.; Song, W.; Nicholas, J. B.; Haw, J. F. J. Am. Chem. Soc. 2002, 124, 10868. (43) Joyner, R. W.; Smith, A. D.; Stockenhuber, M.; van den Berg, M. W. E. Phys. Chem. Chem. Phys. 2004, 6, 5435. (44) Nasluzov, V. A.; Rivanenkov, V. V.; Shor, A. M.; Neyman, K.; Ro¨sch, N. Chem. Phys. Lett. 2003, 374, 487. (45) Bussai, C.; Kru¨ger, S.; Vayssilov, G. N.; Ro¨sch, N. Phys. Chem. Chem. Phys. 2005, 7, 2656. (46) Joshi, A. M.; Delgass, W. N.; Thomson, K. T. J. Phys. Chem. 2007, 111, 11888. (47) Neyman, K. M.; Ro¨sch, N.; Pacchioni, G. Appl. Catal., A 2000, 191, 3.

JP901313N