Applied Surface Science 256 (2010) 3346–3351

Enhanced catalytic performances by surface silylation of Cu(II) Schiffbase-containing SBA-15 in epoxidation of styrene with H2O2

Ying Yang, Jingqi Guan, Pengpeng Qiu, Qiubin Kan *

College of Chemistry, Jilin University, Jiefang Road 2519, Changchun 130023, Jilin, PR China

A R T I C L E I N F O

Article history:

Received 15 November 2009

Received in revised form 9 December 2009

Accepted 9 December 2009

Available online 16 December 2009

Keywords:

Copper(II) Schiff base complex

Silylation

Hydrophobicity

Epoxidation of styrene

A B S T R A C T

Schiff base functionalized SBA-15 mesoporous materials were synthesized by post-grafting of

salicylaldehyde onto silylated and non-silylated amino-modified SBA-15 and followed by the

introduction of Cu(II) ions via a ligand exchange reaction. Both hybrid materials prepared were

characterized by XRD, FT-IR, UV–vis spectroscopy, N2 adsorption/desorption, TG/DTA and ICP-AES

techniques and comparatively examined as catalysts in epoxidation of styrene with 30 wt.% aqueous

hydrogen peroxide as oxidant. It was found that the silylated material was more active and selective to

styrene oxide than the non-silylated one in CH3CN. The considerably improved activity (86.1%) and

styrene oxide selectivity (95.2%) were achieved after 30 min when adding sodium hydroxide to maintain

a pH of 7.5–8.0 in reaction medium. Moreover, the silylated catalyst showed good recoverability and

relatively high stability against leaching of active copper species. These superior effects were attributed

to the high hydrophobic character of the solid surface produced by the silanol neutralization.

� 2009 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Applied Surface Science

journal homepage: www.e lsev ier .com/ locate /apsusc

1. Introduction

The epoxidation of olefins is of great interest due to theimportance of epoxides in the manufacture of both bulk and finechemicals [1,2]. Previously, much attention has been drawn ondeveloping novel catalytic processes based on titanosilicalite(TS-1) and Al-free Ti-beta catalysts [3]. Along with the Ti-containing mesoporous molecular sieves (i.e. Ti-MCM-41, Ti-MCM-48), these Ti-based catalysts are active for aqueous H2O2

oxidation of alkanes and alkenes [4,5]. However, the yields ofepoxide are relatively low as a consequence of the lower H2O2

selectivities. It is assumed that the low epoxide selectivity andsevere deactivation of Ti-containing catalysts are caused by thepoison of catalytically active Ti species by H2O adsorbed on thesurface with relatively high hydrophilicity derived from a largenumber of silanol groups. Therefore, a lot of researches have beenpursued into the trimethylsilylation of silica materials toincrease the surface hydrophobicity. For example, Cagnoli etal. [6] reported that silylation of Ti-MCM-41 could improve thecatalytic activity and selectivity in the limonene oxidation withH2O2. Tatsumi and coworkers [7] also pointed out that thesilylated Ti-MCM-41 (48) showed remarkably higher catalyticactivity in the oxidation of substrates with various molecule sizes(from C6 to C12) with H2O2 compared to non-silylated samples.

* Corresponding author. Tel.: +86 431 88499140; fax: +86 431 88499140.

E-mail address: qkan@mail.jlu.edu.cn (Q. Kan).

0169-4332/$ – see front matter � 2009 Elsevier B.V. All rights reserved.

doi:10.1016/j.apsusc.2009.12.032

Recently, silylation of metal complex-containing mesoporousmaterials also seems to be effective. In this field, worthy ofspecial mention is the work of Jia et al. [8] silylating oxodiperoxomolybdenum complex-containing MCM-41 by trimethylsilylchlorine (TMCS) and the catalytic activity was improved incyclooctene epoxidation with TBHP.

Schiff base transition metal complexes have been extensivelystudied because of their remarkable electronic tunability andpotential use as catalysts in a wide range of epoxidation reactions.However, most heterogenized metal Schiff base complexesshowed low selectivity to epoxides and poor stability in liquid-phase oxidations [9,10]. A typical e.g. is Cu(II) Schiff base-containing MCM-41 (Cu-MCM-41) material, which was highlyefficient for epoxidation of olefins, reaching 97% conversion ofstyrene and 86% yield of epoxide after 24 h with TBHP. However,only 34% conversion and 6% yield were reached when using H2O2

as the oxidant and rapid deactivation was observed in recycledruns [11]. The question is the same as the originally stated Ti-containing catalyst in epoxidation of olefins by H2O2. Thus, somestrategies, such as tunability of the electronic properties of Schiffbases [12], using TBHP or O2 instead of H2O2 in reaction medium[8,13], have been adopted to improve epoxide selectivity andcatalyst stability. As far as we know, there has been no report onimproving the catalytic performance of heterogenized metal Schiffbase complex by surface silylation of support in epoxidation ofstyrene with H2O2.

Herein, we first report a Cu(II) Schiff base-containing silylatedSBA-15 hybrid material synthesized by a post-grafting route.

Y. Yang et al. / Applied Surface Science 256 (2010) 3346–3351 3347

Considering the required more severe conditions and, in the caseof chloride, resulted in loss of copper from the complex, thesilylation procedure was carried out by hexamethyldisilazane(HMDS) instead of commonly used TMCS. Both the silylated andthe non-silylated catalysts were examined as catalysts inepoxidation of styrene with 30 wt.% aqueous hydrogen peroxideas oxidant.

2. Experimental

2.1. Silylated Cu-SBA-15 synthesis

The mesoporous support SBA-15 (1.0 g), prepared by aliterature method [14], was activated by heating at 120 8C for2 h and used to prepare amino-functionalized SBA-15 using aprocedure described in our early work [15]. Analysis found for APS-SBA-15: C, 5.37; H, 1.41; N, 1.39. FTIR (KBr pellets, cm�1): 1510(NH2), 1000–1130 (Si–O–Si), 960 (Si–OH), 687 (N–H).

Silylated APS-SBA-15 was prepared by a post-grafting method.In a typical synthesis, 4.0 g of HMDS was added dropwise to asuspension of 2.0 g APS-SBA-15 in anhydrous hexane. The mixturewas stirred under N2 atmosphere at room temperature for 24 h.The resulting solid was filtered, washed with hexane and driedunder vacuum. This solid (0.3 g) was then refluxed withsalicylaldehyde (6.9 mmol, 0.84 g) in ethanol under N2 atmo-sphere at 80 8C for 3 h. The resulting yellowish solid was thencollected by filtration and was dried under vacuum. Finally, thesilylated Cu-SBA-15 was prepared by dissolving Cu(NO3)2�3H2O(0.17 mmol) in methanol and by stirring the above said yellowishsolid (0.122 g) in suspension at room temperature for 12 h. Thegreen solid thus formed during stirring was filtered, washed withmethanol using Soxhlet and dried under vacuum (Scheme 1). ICP-AES result showed copper content in silylated Cu-SBA-15 is ca.0.122 mmol/g.

Scheme 1. Preparation of Cu(II) Schiff

2.2. Cu-SBA-15 synthesis

Cu-SBA-15 was synthesized according to the procedurereported by the literature [11] for comparison. ICP-AES resultshowed copper content in Cu-SBA-15 catalyst is ca. 0.130 mmol/g.

2.3. Characterization

Powder XRD was collected with a Rigaku X-ray diffractometerwith nickel filtered CuKa radiation (l = 1.5418 A). The sampleswere scanned in the range 2u = 0.4–5.08 and in steps of 28/min. N2

adsorption/desorption isotherms were recorded at �196 8C with aMicromeritics ASAP 2020. Before measurements, the samples wereoutgassed at 120 8C for 12 h. The specific surface area wascalculated by using the Brunauer–Emmett–Teller (BET) methodand the pore size distributions were measured by using Barrett–Joyner–Halenda (BJH) analyse from the desorption branch of theisotherms. The infrared spectra (IR) of samples were recorded inKBr disks using a NICOLET impact 410 spectrometer. UV–visspectra were recorded on a Perkin Elmer UV-vis spectrophotome-ter Lambda 20 using barium sulfate as the standard. TG/DTA wascarried out on Shimadzu DTG-60 instrument. Microanalyses for C,H, N were performed at the Perkin Elmer 2400. Metal content wasestimated by inductively coupled plasma atomic emissionspectroscopy (ICP-AES) analysis conducted on a Perkin Elmeremission spectrometer.

2.4. Catalytic reactions

The catalytic reactions were carried out in a glass batch reactorsupplied with a magnetic stirrer and backflow condenser.Typically, 5 mmol of styrene along with 5 ml of solvent (CH3CN)and 25 mg of catalyst were added to the flask. The reaction wasstarted by adding 30% H2O2 (15 mmol, 3 equiv.) at reaction

base complex on silylated SBA-15.

Fig. 2. FT-IR spectra of (a) SBA-15, (b) Cu-SBA-15 and (c) silylated Cu-SBA-15.

Y. Yang et al. / Applied Surface Science 256 (2010) 3346–33513348

temperature. After carrying out the reaction, the catalyst wasfiltered, washed with CH3CN, dried at 100 8C overnight and reuseddirectly without further purification. The liquid organic productswere quantified by using a gas chromatography (Shimadzu, GC-8A)equipped with a flame detector and an HP-5 capillary column. Theliquid organic products were identified by comparison withauthentic samples and GC–MS coupling.

3. Results and discussion

3.1. XRD

The powder XRD patterns for SBA-15, Cu-SBA-15 and silylatedCu-SBA-15 are depicted in Fig. 1. The SBA-15 sample shows threepeaks, indexed as the (1 0 0), (1 1 0) and (2 0 0) diffraction peaksassociated with typical two-dimensional hexagonal symmetry ofthe SBA-15 material (Fig. 1a) [16]. For the hybrid material Cu-SBA-15, the relative intensity of the prominent diffraction peak (1 0 0)decreased after introduction of bulky organometallic groups. Theintensity reduction may be mainly due to contrast matchingbetween the silicate framework and organic moieties which arelocated inside the channels of SBA-15 [8]. In the X-ray diffraction ofthe silylated sample, the peaks corresponding to the (1 1 0) and(2 0 0) diffractions further decreased in intensity, suggesting thatthe partial loss of the hexagonal symmetry of the sample [17].

3.2. Spectroscopic characterization

HMDS treatment of amino-modified SBA-15 material intro-duced IR bands consistent with trimethylsilylation. Compared tothe pure silicious SBA-15 and Cu-SBA-15, the silylation broughtabout new bands to the IR spectra that can be attributed to SiMe3

groups, especially the very distinctive band at 1250–1260 cm�1

and the CH3 rocking modes at ca. 840 and 755 cm�1 (Fig. 2). Thesebands can be regarded as important evidence for the alkylsilylgroup (SiMe3) replacing the hydrogen in silanol [18]. Simulta-neously, the increased C–H stretches observed at 2963 and2907 cm�1 and the decreased Si–OH vibrations at 960 cm�1 inthe silylated Cu-SBA-15 indirectly confirmed the presence oftrimethylsilyl groups on the surface of the copper(II) complex-containing SBA-15 material [18,19]. Moreover, a very weak band at574 cm�1 is due to the presence of ordered silica structure. Thisband decreased more significantly in silylated samples, confirmingthe results of the X-ray diffractograms and suggesting the bandsdue to organic moieties appeared in the spectrum of the silylated

Fig. 1. XRD patterns of (a) SBA-15, (b) Cu-SBA-15 and (c) silylated Cu-SBA-15.

sample [17]. New bands in the region 1600–1300 cm�1 furtherdemonstrate this speculation (Fig. 2b and c).

The UV–vis spectra of both the silylated and the non-silylatedsamples displayed a peak at ca. 390 nm typical of the metal ligandband and a broad band at ca. 630 nm associated with d–dtransitions (not shown). This was similar to related metal salencompounds described in the literature [20], indicating thesuccessful anchoring of Cu(II) Schiff base complex on SBA-15matrix. However, in comparison with Cu-SBA-15, the sharper bandat 390 nm and the stronger d–d transitions in the silylated Cu-SBA-15 confirmed that Si–OH groups were transformed into Si–O–SiMe3 and the coordination environment was optimized by surfacesilylation. Similar results were also obtained by Guidotti et al. [21].

3.3. N2 adsorption/desorption studies

The N2 adsorption/desorption isotherms of SBA-15 and thehybrid materials are shown in Fig. 3. Obviously, the silylated Cu-SBA-15 hybrid materials maintain the characteristics of type IVisotherm with a sharp inflection due to the capillary condensationin the uniform mesopore channels and show an H1 type hysteresisloop, suggesting the channels of the SBA-15 support remain

Fig. 3. N2 adsorption/desorption isotherms and pore size distribution profiles

(inset) of (a) SBA-15, (b) Cu-SBA-15 and (c) silylated Cu-SBA-15.

Table 1Textural properties of samples.

Materials SBET (m2 g�1) Vp (cm3 g�1) Dpa (nm)

SBA-15 892 1.01 6.35

Cu-SBA-15 403 0.67 6.14

Silylated Cu-SBA-15 360 0.60 5.38

a Calculated from the desorption branch.

Fig. 5. Styrene conversion (solid lines), benzaldehyde selectivity (dash lines) and

styrene oxide selectivity (dot lines) obtained in epoxidation of styrene with

silylated Cu-SBA-15 (&) and Cu-SBA-15 ($). Reaction conditions: styrene 5 mmol,

catalyst 25 mg, H2O2 15 mmol, CH3CN 5 ml and temperature 80 8C.

Y. Yang et al. / Applied Surface Science 256 (2010) 3346–3351 3349

accessible after surface silylation. However, the textural propertiesof the mesoporous SBA-15 were significantly affected. In thesilylated sample, the decrease in P/P0 where the inflection pointwas found is directly proportional to pore narrowing. Thereduction was already expected because hydroxyl groups weresubstituted by bulkier trimethylsilyl groups. This is the reason fordiminished pore volumes and surface area in silylated samplecompared with the non-silylated one (Table 1). It is in agreementwith the previous reports that both the surface area and the porevolume decreased as the silanol groups inside the pore of Ti-MCM-41 were trimethylsilylated [22]. Moreover, the uniformity of thepore size distribution was kept after silylation as suggested byusing the BJH formalism (inset of Fig. 3). There was a decrease ofca. 8 A in the pore size upon silylation, a value very close to thatexpected taking into account the steric volume of the trimethylsiylgroups anchored on the surface of pores of the Ti-MCM-48structure [5].

3.4. TG/DTA studies

TG/DTA analysis results of Cu-SBA-15 and silylated Cu-SBA-15are depicted in Fig. 4. The silylated sample showed a very littleweight loss at temperatures below 150 8C (1.2 wt.%), typicallyattributed to weakly adsorbed water [6]. The non-silylated sample,on the contrary, showed a weight loss of about 3.0 wt.% below150 8C. This behavior indicates a poor capacity for wateradsorption in the former one. On the other hand, a step of about2.0 wt.% weight loss was clearly observed in the silylated sample at450 8C, that is a temperature at which trimethylsilyl groupsdecomposed leading to surface silanols [5], indicating that thesilylation process indeed took place on the surface of SBA-15channels. Moreover, the exothermic peaks ranged 250–400 8C,assigned to the Schiff base ‘‘burnt’’ [23], shifted to lower values inthe silylated Cu-SBA-15 samples, suggesting that the thermalstability of Cu(II) Schiff base-containing SBA-15 material waschanged by surface silylation.

Fig. 4. TG and DTA curves of silylated Cu-SBA-15 (solid lines) and Cu-SBA-15 (dash

lines).

3.5. Oxidation of styrene

The comparisons with respect to the catalytic performance ofthe silylated and non-silylated samples strongly indicate that theintroduction of trimethylsilyl groups leads to a significantimprovement in catalytic activity. Fig. 5 shows the results fromthe epoxidation of styrene using H2O2 as oxidant and silylated ornon-silylated Cu-SBA-15 as catalyst. It is clear that the silylatedmaterial was more active (97.2% conversion after 8 h, TOF 199 h�1)than the non-silylated one (89.0% after 8 h, TOF 170 h�1). This canbe explained by the good interaction of the polar epoxide with thepolar surface of the non-silylated Cu-SBA-15 compared with itspoor interaction with the apolar reaction medium. Thus theepoxide remains in the Cu-SBA-15 channels, blocking the activecatalytic sites. In the silylated Cu-SBA-15, the epoxide is moreeasily removed from the channels, thus increasing the activity ofthis material. Considering the similarity of dispersion degree andloading of active species for both catalysts, the different catalyticperformance could be related to their different textural propertiesexcept for surface properties. The introduction of trimethylsilylgroups did increase the steric hindrance demonstrated bysignificantly decreased surface area and pore size of the silylatedCu-SBA-15 material. But in our study, it is not a factor affecting thecatalytic properties due to larger pore size of the silylated sampleas compared with the less molecule size of styrene and epoxide. Itis different from the fact that catalytic activity can be enhanced byincreasing surface area of transition metal nanoclusters inheterogeneous catalysis reported by Ozkar [24]. Clearly, theenhanced activity could be reasonably attributed to the hydro-phobic surface properties due to the introduction of trimethylsilylregulators.

The benzaldehyde is dominant in epoxidation of styrene withH2O2 in combination with CH3CN as the solvent in most cases,since further oxidation of styrene oxide formed in the first step by anucleophilic attack of H2O2 on styrene oxide was followed bycleavage of the intermediate hydroperoxistyrene to form higherselectivity to benzaldehyde [9]. The silylated Cu-SBA-15 is not anexception. As illustrated in Fig. 5, the selectivity to benzaldehyderanged 56.5–78.6% for both catalysts. However, the silylatedmaterial was more selective to styrene oxide than the non-silylated one. The selectivity to epoxide was 17.3% for the silylatedCu-SBA-15, 13.5% for the non-silylated Cu-SBA-15 after 4 h, withcorresponding initial turnover frequencies 352 and 322 h�1. Thesedata reinforce the earlier-stated assumption that the epoxide

Fig. 6. (&:) Kinetic profiles for the oxidation of styrene with H2O2. (~:)

Heterogeneous reaction check by continuing the reaction after removing the

catalyst after 15 min. Reaction conditions: 0.05 g catalyst, 10 mmol styrene,

30 mmol H2O2, 10 ml CH3CN, pH = 7.5–8.0: (a) silylated Cu-SBA-15 and (b) non-

silylated Cu-SBA-15.

Y. Yang et al. / Applied Surface Science 256 (2010) 3346–33513350

remains inside the channels of the non-silylated Cu-SBA-15,facilitating further transformation and diminishing selectivity. Onthe other hand, probably, silylation promotes the attacking of theC55C in olefin to produce more epoxide and the hydrophobicsurface dose not favor water attachment for epoxirane openingreactions [17]. The relatively improved styrene oxide selectivityimplies that the surface silylation of Cu(II) Schiff base-containingSBA-15 can improve the epoxide selectivity to some extent butcannot change the reaction mechanism or the product distribu-tion accurately, since reaction medium plays a more significantrole.

Previously, it was reported that the selectivity to styrene oxidenever went above 10% in epoxidation of styrene with H2O2 whenTi-containing catalysts were used [3–5]. It is probably that styreneoxide was converted to benzaldehyde in mildly acidic H2O2

medium. It was the same case with the Cu(II) Schiff base-containing SBA-15 catalyzed reactions. According to somereports, except for surface silylation, the addition of some alkalinespecies (e.g. NaOH, KOH, NaHCO3) into the reaction system wasbeneficial to epoxide formation [25–27]. In our study, aqueous1.0 M sodium hydroxide was added to maintain a pH of 7.5–8.0. Itwas found that the conversion reached maximal value of 86.1%after 30 min and remained almost the same after 120 and 480 min,suggesting that the reaction started instantaneously and balancedrapidly [2]. The selectivity to styrene oxide was significantlyimproved (95.2%, 30 min) and remained high even after 480 min,indicating alkaline environment maintained by NaOH buffereffectively prohibited the process of the nucleophilic attack ofH2O2 on styrene oxide and a completely different pathway wasresponsible for styrene oxide formation [25]. Therefore, thereaction time of all recycling experiments conducted under basewas set to 30 min.

Recoverability studies showed that the silylated Cu-SBA-15could be successfully recycled for three times without significantdecrease of activity or selectivity under test conditions, while theactivity of the non-silylated Cu-SBA-15 decreased by 30.4% of theinitial value after a single recycling (Table 2). To test for leaching,we filtered the catalyst at the reaction temperature (80 8C), e.g. thesilylated Cu-SBA-15 after 15 min (53.4% styrene conversion). Atthis time, half the volume was filtered and the resulting clearsolution let to react. The percentage of leaching was estimated bycomparing the time-conversion plot of the twin reactions withand without solid. It was found that after the hot filtration, themother liquor reacted further at roughly the same rate as thatobserved when the Cu-SBA-15 catalyst was not filtered (Fig. 6b)which leads to the inevitable conclusion that all of the observedactivity can be attributed to homogeneous Cu(II) leached from thehydrophilic Cu-SBA-15 by reaction with H2O2. While little further

Table 2Recycling properties of silylated and non-silylated catalysts.

Catalysts Cycle Styrene

conversion (%)a

Product selectivity

(mol.%)b

So Bza Diol

Silylated Cu-SBA-15 1 86.1 95.2 2.5 1.3

2 79.4 93.7 2.0 1.2

3 74.8 93.1 2.8 1.5

4 64.5 91.5 4.6 3.0

Cu-SBA-15 1 68.5 84.2 11.6 4.2

2 38.1 83.7 3.5 6.7

3 32.6 84.4 3.0 8.5

4 22.2 60.7 18.0 21.3

a Reaction conditions: the molar ratio of H2O2/styrene = 3: 1, CH3CN 5 ml,

pH = 7.5–8.0, temperature 80 8C, duration 30 min.b So: styrene oxide, Bza: benzaldehyde, diol: 1-phenylethane-1,2-diol.

reaction was observed for the silylated Cu-SBA-15 (Fig. 6a),suggesting only trace of active copper species leached from thesilylated Cu-SBA-15 and implying the large majority of thecatalysis is carried out by truly heterogeneous copper catalyst.The superior selectivity, recyclability and stability are probablyattributed to the more hydrophobic character of the silylatedmaterial that prevents the adsorption of water, the opening of theoxirane rings that generated glycols and the consequent catalystdeactivation.

4. Conclusions

Cu(II) Schiff base-containing SBA-15 material was successfullysilylated and confirmed by various techniques and examined ascatalyst for the epoxidation of styrene. It was found that thesilylated material shows higher catalytic activity, selectivity tostyrene oxide and superior stability to the non-silylated sample.The excellent performances of the silylated catalyst wereattributed to its more hydrophobic character by surface silylation.

Acknowledgements

Financial assistance from the National Basic Research Programof China (2004CB217804) and the National Natural ScienceFoundation of China (20673046) is gratefully acknowledged. Wealso greatly appreciate the suggestions from editor and refereesconcerning improvement to this paper.

References

[1] G. Grigoropoulou, J.H. Clark, J.A. Elings, Green Chem. 5 (2003) 1.[2] B.S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457.[3] M.B. D’Amore, S. Schwarz, Chem. Commun. (1999) 121.[4] M.A. Camblor, A. Corma, P. Esteve, A. Martınez, S. Valencia, Chem. Commun.

(1997) 795.[5] M.L. Pena, V. Dellarocca, F. Rey, A. Corma, S. Coluccia, L. Marchese, Micropor.

Mesopor. Mater. 44–45 (2001) 345.[6] M.V. Cagnoli, S.G. Casuscelli, A.M. Alvarez, J.F. Bengoa, N.G. Gallegos, M.E. Crivello,

E.R. Herrero, S.G. Marchetti, Catal. Today 107–108 (2005) 397.[7] N. Igarashi, K. Hashimoto, T. Tatsumi, Micropor. Mesopor. Mater. 104 (2007) 269.[8] M.J. Jia, A. Seifert, W.R. Thiel, Chem. Mater. 15 (2003) 2174.[9] M.R. Maurya, M. Kumar, U. Kumar, J. Mol. Catal. A: Chem. 273 (2007) 133.

[10] M.R. Maurya, U. Kumar, P. Manikandan, Eur. J. Inorg. Chem. (2007) 2303.[11] S. Jana, B. Dutta, R. Bera, S. Koner, Langmuir 23 (2007) 2492.[12] S. Mobebbi, A.I. Sarvestani, Transit. Met. Chem. 31 (2006) 749.[13] D.M. Boghaei, S. Mohebi, J. Mol. Catal. A: Chem. 179 (2002) 41.

Y. Yang et al. / Applied Surface Science 256 (2010) 3346–3351 3351

[14] D.Y. Zhao, Q.S. Huo, J.L. Feng, B.F. Chmelka, G.D. Stucky, J. Am. Chem. Soc. 120(1998) 6024.

[15] H.S. Wang, J.H. Huang, S.J. Wu, C. Xu, L.H. Xing, L. Xu, Q.B. Kan, Mater. Lett. 60(2006) 2662.

[16] N.K.K. Raj, S.S. Deshpande, R.H. Ingle, T. Raja, P. Manikandan, Catal. Lett. 98 (2004)217.

[17] J.M.R. Gallo, H.O. Pastore, U. Schuchardt, J. Catal. 243 (2006) 57.[18] J. Bu, H.-K. Rhee, Catal. Lett. 65 (2000) 141.[19] R.R. Sever, R. Alcala, J.M. Dumesic, T.W. Root, Micropor. Mesopor. Mater. 66 (2003)

53.[20] S. Mukherjee, S. Samanta, B.C. Roy, A. Bhaumik, Appl. Catal. A: Gen. 301 (2006) 79.

[21] M. Guidotti, I. Batonneau-Gener, E. Gianotti, L. Marchese, S. Mignard, R. Psaro, M.Sgobba, N. Ravasio, Micropor. Mesopor. Mater. 111 (2008) 39.

[22] J. Bu, H.-K. Rhee, Catal. Lett. 66 (2000) 245.[23] C. Jin, W.B. Fan, Y.J. Jia, B.B. Fan, J.H. Ma, R.F. Li, J. Mol. Catal. A: Chem. 249 (2006)

23.[24] S. Ozkar, Appl. Surf. Sci. 256 (2009) 1272.[25] C.V. Rode, U.N. Nehete, M.K. Dongare, Catal. Commun. 4 (2003) 365.[26] W. Adam, P.B. Rao, H.G. Degen, C.R. Saha-Mller, J. Am. Chem. Soc. 122 (2000)

5654.[27] X.H. Lu, Q.H. Xia, H.J. Zhan, H.X. Yuan, C.P. Ye, K.X. Su, G. Xu, J. Mol. Catal. A: Chem.

250 (2006) 62.