Accepted Manuscript Zirconium Phenylphosphonate-anchored Methyltrioxorhenium as Novel Heterogeneous Catalyst for Epoxidation of Cyclohexene Sha He, Xin Liu, Hongyue Zhao, Yue Zhu, Fazhi Zhang PII: DOI: Reference:

S0021-9797(14)00622-5 http://dx.doi.org/10.1016/j.jcis.2014.08.065 YJCIS 19800

To appear in:

Journal of Colloid and Interface Science

Received Date: Accepted Date:

11 April 2014 29 August 2014

Please cite this article as: S. He, X. Liu, H. Zhao, Y. Zhu, F. Zhang, Zirconium Phenylphosphonate-anchored Methyltrioxorhenium as Novel Heterogeneous Catalyst for Epoxidation of Cyclohexene, Journal of Colloid and Interface Science (2014), doi: http://dx.doi.org/10.1016/j.jcis.2014.08.065

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Zirconium Phenylphosphonate-anchored Methyltrioxorhenium as Novel Heterogeneous Catalyst for Epoxidation of Cyclohexene

Sha He†, Xin Liu†, Hongyue Zhao, Yue Zhu, and Fazhi Zhang* State Key Laboratory of Chemical Resource Engineering Beijing University of Chemical Technology Beijing 100029 (China)

* Corresponding author. Tel: +86 10 64425105; Fax: +86 10 64425385; E-mail: [email protected] (F.Z. Zhang) †

These authors contributed equally to this work.

1

Abstract: Epoxidation of olefins to epoxides is widely recognized as an important unit process in the manufacture of fine chemicals and intermediates. Developing an environmentally benign heterogeneous catalytic system for olefin epoxidation with high activity and selectivity is still a challenge in this research field. Herein, we report our

attempts

to

synthesize

novel

zirconium

phenylphosphonate-anchored

methyltrioxorhenium (MTO/ZrPP) heterogeneous catalysts by a conventional impregnation method and evaluate their catalytic performance for epoxidation of cyclohexene using urea-hydrogen peroxide adduct (UHP) as oxidant without the addition of base ligands. The MTO/ZrPP catalyst samples are characterized by powder X-ray diffraction (XRD), Fourier transform infrared (FT-IR), inductively coupled plasma emission spectrometry (ICP-ES), high resolution transmission electron microscopy (HRTEM), X-ray photoelectron spectroscopy

(XPS), and

solid-state 1H magic-angle spinning nuclear magnetic resonance (1H MAS NMR) techniques. Meanwhile, the density functional theory (DFT) calculation is carried out to further understand the structure feature and interactions of the MTO/ZrPP catalyst. It is revealed that MTO is anchored on support surface by the favored hydrogen-bonding interaction between two oxo ligands of MTO and two H atoms from the adjacent phenyls of ZrPP. MTO/ZrPP catalyst displays excellent catalytic activity for cyclohexene epoxidation. Moreover, only cyclohexene oxide production can be obtained under the employed reaction conditions.

2

Keywords: epoxidation of cyclohexene; methyltrioxorhenium; metal phosphates and phosphonates; density functional theory.

1. Introduction Epoxidation of olefins to epoxides is a fundamentally important procedure in both laboratory and commercial synthetic chemistry for the epoxides are widely used in industry for manufacturing various types of important products, such as perfume materials, epoxy resins, and drugs [1],[2]. Many high-valent metal oxo complexes have been developed to convert olefins to epoxides using several stoichiometric oxidizing agents (e.g., O2, H2O2 and tert-BuOOH) [3],[4]. Among the catalysts applied for the epoxidation reaction, methyltrioxorhenium (abbr. MTO; (CH3)ReO3) has been paid more and more attention since Herrmann et al. discovered that MTO is an efficient homogeneous catalyst for the epoxidation of alkenes with H2O2 as an oxidant [5]-[7]. However, the epoxidation suffers from low yield for the MTO/H2O2 catalysis system because of the formation of 1,2-diol product (through ring opening of the epoxides) as byproduct catalyzed by Lewis acidic rhenium center in the presence of H2O [8],[9]. Several efficient methods have been developed to avoid the side reaction. For example, urea hydrogen peroxide adduct (UHP) with helical channel supposed to serve as a host confined stabilizing environment for MTO was adopted as oxidant in the organic reaction medium, e.g., dichloromethane (CH2Cl2) [10],[11]. Meanwhile, the direct addition of excess pyridine [8], aromatic N-bases [[12],[13], and certain MTO-Schiff base adducts or microencapsuled Lewis base adducts 3

(monodentate and bidentate, aromatic and aliphatic nitrogen containing ligands [14]-[16]) could decrease the diol products by neutralizing the acidity of MTO catalyst system. The homogeneous catalyst recycle and product-catalyst separation play a key role in the industrial applications, lots of efforts have been devoted to develop the heterogeneous supported MTO catalysts [17] in recent years. Various inorganic and organic materials, such as acidic metal oxides (niobia [18],[19], zeolites [20]-[24], or alumina [25]-[27]), polymers[28],[29], natural polysaccharides [30] and functionalized silica matrices [31]-[33] have been employed as supports for the immobilization of MTO. Meanwhile, several important methods for improving the catalyst performance have been revealed by surface chemistry and computational studies, for instance, the structural feature and bonding mode of the MTO species on the support surface, the beneficial pyridine effect manifested in both homogeneous and heterogeneous MTO-catalyzed epoxidation, and the modification of support surface with organic or polymeric functional groups. Despite the considerable progress has been made on the heterogeneous MTO systems, it still remains a challenge to synthesize novel supported MTO catalyst for olefin epoxidation with high selectivity and activity for practical applications. Layered metal phosphates and phosphonates compounds have potential applications in the areas such as catalysis, molecular recognition, proton conduction, corrosion resistance, and immobilization of biological materials [34],[35]. Zirconium phosphonates Zr(O3PR)2 (R = organic groups), with each layer consisting of planes of 4

zirconium bridged through phosphonate groups that alternate above and below the Zr atom planes, oriented away from the basal surfaces in a bilayered fashion in the interlayer region, have various surface and structure characteristics which allows zirconium phosphonates with potential opportunities for novel catalysts and precursors or supports of catalysts [36]-[39]. Specifically, zirconium phosphonates have been developed as catalysts for esterification of acetic acid with ethanol [36], synthesis pyrroles by Paal-Knorr condensation [37], oxidation of phenol [38], and as supports for metallocene polymerization catalysts [39]. As crystalline compounds could provide a well-defined uniform support surface, the two-dimensional geometry prevents substrate transport limitations. Moreover, zirconium phosphonates have been proved to be thermally and chemically very robust by thermogravimetric study [40]. Herein, taking advantage of these attractive structure features, we report zirconium phosphonate, particular the zirconium phenylphosphonate (ZrPP), as a carrier material for immobilization MTO by a conventional impregnation method. Several characterization techniques such as XRD, FT-IR, ICP-ES, HRTEM, XPS, and 1H MAS NMR techniques, as well as DFT calculation, are carried out to understand the structural feature of the MTO/ZrPP sample. The catalytic property of MTO/ZrPP sample is evaluated using the probe reaction of cyclohexene epoxidation with UHP as oxidant without the addition of base ligands. Our findings show an example for immobilization of organometallic catalysts on solid surfaces with appropriate structural features that provide a promising way for olefin epoxidation with high activity and selectivity. 5

2. Experimental 2.1 Synthesis of ZrPP support The ZrPP support was synthesized by a method involving separate nucleation and aging steps, which was used to prepare α-ZrP by our group previously [41]. An aqueous solution of concentrated ZrOCl2·8H2O and hydrofluoric acid in deionized water and an aqueous solution of phenylphosphonic acid (C6H5PO(OH)2) in equal volume of propanol and deionized water were simultaneously added to a colloid mill with a rotor speed of 3000 rpm and mixed for 1 min. The mixture solution has molar ratio [C6H5PO(OH)2]/[Zr] = 2 and [F]/[Zr] = 6. The resulting suspension was removed from the colloid mill and aged at 100 °C for 96 h. The ZrPP product was washed with deionized water, isopropyl alcohol, acetone, and ether, and finally placed in a vacuum oven at 40 °C. 2.2 Preparation of MTO/ZrPP catalyst 1.0 g sample of ZrPP support and 10 mL solution of CH2Cl2 were added to a 25 mL round-bottom flask fitted with a reflux condenser. Then, a known amount of MTO with appropriate concentration was added under a flow of nitrogen, which was adjusted to yield catalyst containing 0.6, 3.1, 4.4, 8.1 and 9.9 wt% Re. The actual Re loadings of series of MTO/ZrPP catalysts were determined by ICP-ES analysis. The suspension was stirred for 24 h at room temperature in dark under flowing nitrogen. The solid was collected by filtration, washed with CH2Cl2, and finally dried at 40 °C in a vacuum oven. 6

2.3 Characterization XRD patterns of the catalyst and support samples were collected on a Shimadzu XRD-6000 instrument with Cu Kα radiation in the 2θ range from 3° to 80°. Elemental analyses for Re were carried out by ICP-ES using a Shimadzu ICP-7500 instrument. HRTEM characterizations were performed on JEM-2100F (200 kV). FT-IR spectra were collected on a Bruker Vector 22 infrared spectrophotometer using the KBr disk method with a sample/KBr weight ratio of 1:100. XPS was carried out with an SKL-12 spectrometer equipped with Mg Kα radiation. Solid-state 1H MAS NMR experiment was performed on a Bruker AVANCE 300 NMR spectrometer with a 4-mm broadband MAS probehead operating at 400 MHz for 1H. The catalytic reaction was monitored by a gas chromatography (Shimadzu Technologies, HP-624 capillary column, 30m×0.25mm) with flame ionization detector using nitrogen as carrier gas. Both the injector and detector temperature were 523 K. The products identified by GC-MS using an HP5790 series mass selective detector. The cyclohexene conversion and the turnover number (TON) were calculated as follows: Cyclohexene conv. (%) = [(moles of cyclohexene converted) × 100]/[moles of cyclohexene in feed] TON = (moles of cyclohexene converted)/(moles of Re in the catalyst) 2.4 DFT calculations In the present work, the C25H23O19ReZr6P4 model is adapted for the DFT calculation of MTO/ZrPP catalyst (Scheme 1). To present the effects of exchange and correlation, Becke’s 3-parameter exchange with correlation functional of Lee, Yang, 7

an nd P Parrr (B B3L LYP P) are a useed [422],[443]. Thhe bassis set 6-331G G*** is used forr the N, N O O, C an nd P attom ms, whhile thee R Re and a Zr atoomss arre trea t ated dw with thee efffecctiv ve core c e pootenntiaal (E ECP P) of o LA L NL L2D DZ [444]-[446]. A All atoomss off thhe ZrP PP cluusteer eexceept H atooms of o beenzeenee arre rrelaaxedd in the t caalcuulatiions, mak m kess th he mo m dels spec s cific to t ZrP Z PP. Thhe rep porrtedd coomppouundd iss veeriffiedd ass bein b ng true t e m miniimaa by b tthe abbsennce off neegaativve eig gennvalluess inn thhe vibr v atioonaal frrequuen ncy anaalyssis, an nd all a oof thhe wor w rk is i carr c riedd ouut by y thee Gau G ussiaan 09 progrram ms [47] [ ]. The T e innteraactiionn en nerggy was correcteed forr baasiss seet su uperrpossition erroor (BS ( SSE E) by b thhe cou c unteerpo oisee teechn niquue [48 [ ].

S Sch hem me 11. The T opttim mized sttruccturre of o MTO M O/Z ZrP PP ccataalyst. 8

2.5 Cyclohexene epoxidation The cyclohexene epoxidation reaction over the supported MTO catalyst was carried out under air by contacting 0.1 g catalyst with 9.8 mmol cyclohexene, 19.2 mmol UHP, and 247.2 mmol methanol in a magnetic stirring reactor at room temperature for 6 h. A measured quantity of toluene is added to the products as internal standard. Prior to analysis by GC, the sample was dried over a short column containing MgSO4.

3. Results and discussion 3.1 Characterization of MTO/ZrPP catalyst Fig. 1a displays the characteristic XRD reflection peaks of ZrPP structure with a series of (00l) ones at low angle [49]. The appearance of (002) reflection of ZrPP phase at 5.89° demonstrates a basal interlayer spacing of 14.6 Å [50]. XRD patterns of the resulting MTO/ZrPP catalysts with the lower Re loading 0.6, 3.1, 4.4, and 8.1% are found to be similar to the corresponding ZrPP support, and no characteristic diffraction for MTO is found. With the Re loading increase up to 9.9%, the (006) and (211) peaks of MTO appear (Fig. 1f), it indicate the aggregation of large MTO particles on the surface of ZrPP support. Moreover, the XRD reflections of ZrPP structure for the five supported MTO catalysts do not shift to a lower angle, which suggest that MTO could not be intercalated into the interlayers of ZrPP under the adopted synthesis conditions. 9

Fig. F 1 XRD X D patt p ternns of o Z ZrPP P suuppportt (a)) an nd M MT TO/Z ZrP PP cata c alyssts con c ntainning 0.6 0 (b), ( , 3.1 ((c), 4.4 4 (dd), 8.1 8 (e)), annd 9.9 9 (f) wtt% R Re.. Paattern of o tthe MT TO phaasee is inccludded forr c mpaarisoon. com

T e HRT The H TEM M phot p tograpphs in Fig F g. 2 show thee morp m phoologgy of o tthe supppoort and a d thhe caatalyyst sampples. The T suupport ZrrPP P shhow ws an reggullar lam mellar strructturee and a thhe cry ystaallinity y iss hiigh.. Thhe mor m rphholoogy is sim milaar with w h thhe suuppportt affter MT TO O annchoored onn ZrPP P an nd thee sppacingg off thhe crys c stalllog grapphicc plan p ne of o 1.48 1 8 nnm (innsett, Fig. F 2b b) ag greees well w l wiith thee (0002)) lattticee pllanes of o the t ZrP PP.

10 0

Fig. 2 HRTEM images of pristine ZrPP support (a) and MTO/ZrPP catalyst containing 4.4 wt% Re (b).

Fig. 3 shows the FT-IR spectra of ZrPP support, 8.1% MTO/ZrPP catalyst, and MTO. The sharp intensity band at 3062 and 1438 cm-1 are attributed to the C-H and C=C stretch of the phenyl ring in ZrPP support, respectively [51] Moreover, the 748 and 692 cm-1 bands are assigned to the absorptions of C-P bonds in the ZrPP framework structure, and PO3 vibrations are observed in the range of 1163-1039 cm-1 [52]. It is noted that anchoring results in a rapid decrease in the intensity of C-H band of the phenyl ring at 3062 cm-1. For comparison of the corresponding Re=O and Re-C stretching mode of MTO before and after anchored on ZrPP, the FT-IR spectra from 400 to 1200 cm-1 is also shown in Fig. 3. The 998 and 956 cm-1 are assigned to the symmetric and asymmetric stretching vibration modes of Re=O, respectively [53]. Compared to pure MTO, both Re=O stretching vibrations of MTO/ZrPP catalyst are shifted to a lower frequency. However, no change is observed for Re-C stretching mode which was appeared at 569 cm-1. The above results suggest that MTO may be interacted with ZrPP support preferentially by Re=O band other than Re-C band. 11

Fig. 3 F FT-IIR spe s ectraa off ZrrPP P suuppoort (a), 8..1% %M MTO O/ZrrPP P caatalyyst (b)), annd MT TO (c). T-IR R sppecttra from m 400 4 0 to 1200 cm m-1 is shhow wn for f thee coomppariisonn off The maagniifyiing FT R =O and Re= a d Ree-C C strretcchinng mod m des of MT TO before annd afteer anch a horred on ZrP PP.

F g. 4 Caalcuulateed IR Fig I speectra off MTO M O (aa) and a MT TO//ZrP PP cattalyyst ((b).

DFT calc DF c culaatioon of o M MTO O/Z ZrP PP cata c alysst iss caarrieed outt to caalcuulatee thhe IIR data d a by y o mizzed strructturee off ZrrPP P/M MTO O (C C25H23O199ReeZr6P4 moodeel, Sch S hem me 11). A ussingg thhe optim dooublle H-b H bondd m mod de is addopptedd w with tw wo oxo o liggandds oof MT MTO inte i eracctedd w with h tw wo H ato oms frrom m thhe neig n ghbbouuringg pphennyls oof ZrPP Z P. For F thee thhreee Re= R =O bon b nds off MTO M O, booth distan nce of Re= R =O1 aand Ree=O O2 inte i eracctedd wiith ZrP PP is calc c culaatedd too bee 1..7088 Å, Å 12 2

while the unconstrained Re=O3 shows a small value 1.706 Å. The distance of O1-H1 and O2-H2 is calculated to be 2.635 and 2.616 Å, respectively. The corresponding H-bond energy of these H-O bonds is calculated to be 14.7 kJ/mol, which is close to the calculated H-bond energy of 17 kJ/mol between the oxorhenium ligand of MTO and the terminal surface hydroxyl group of amorphous silica-alumina support [44]. Meanwhile, in the case of Y zeolite as support for encapsulation MTO inside the 12 Å supercages, the favored anchoring interaction was suggested by hydrogen bonding of the oxo ligand of MTO preferentially to supercage cation or proton sites [21]. Furthermore, the IR data of the symmetric and asymmetric stretching vibration modes of Re=O for MTO are calculated to be at 982 and 1007 cm-1 (Fig. 4). The red shift of both Re=O stretching vibrations happens when MTO is anchored on ZrPP support, which is similar to the results observed for the MTO deposited on zeolite or silica-Al2O3 supports via hydrogen-bonding [21],[44]. The DFT calculation results are consistent with the IR observation, and the above result indicates that MTO is anchored on support surface by the favored hydrogen-bonding interaction between two oxo ligands of MTO and two H atoms from the neighbouring phenyls of ZrPP. For the purpose of further proving the grafting of MTO on ZrPP support, 1H MAS NMR spectra of ZrPP support and MTO/ZrPP catalyst are conducted and the results are shown in Fig. 5. The peak of 0.2 ppm in the two curves is the background signal of the rotor. Compared with ZrPP support, a new sharp peak at 2.8 ppm appears for the MTO/ZrPP sample, which is assigned to methyl protons of the grafted MTO. This NMR chemical shift falls between the corresponding values reported for bulk MTO 13

(4.1 ppm p m) and a dM MTO O diissoolveed in i ppolaar ssolv ventts (2.2 2-2.66 pppm m) [221]. Itt is woorth nooting t shapee off thhe pea p ak at a 22.8 ppm m rem r mainns surp s prissinggly naarroow aafteer aanchorring g, thaat the su uggeestiing thaat MT MTO is mob m bilee onn thhe sup ppoort surf s facee by b hyd h droggenn-boondiingg too th he su urfacce ph henyyls off ZrrPP P [221]. A Acccordding g too tthe litteraaturre, whhen orrgannom metaalliic co ompplexxes aree coovaalenntly boounddedd too thhe silic s ca surf s facees [54] [ ] orr sttronnglyy bounndeed to sillicaa-aluum minaa suuppoort by Leewiss accid//basse inte i eracction n [444], thhe sign s nificcannt decrreasse of o mo obilityy co ouldd reesuult in i a brroaad ssolidd-statee NMR N R peak p k. IIn add a ditioon, Figg. 5 shhow ws a peeak cennterred at 6.88 pppm forr ZrrPP P supppoort whhichh is attrribuutedd too thhe prot p ton of pheenyyl rin ng of ZrP PP [555], andd thheree iss a sliightt do ownnfieeld shiift to 7.00 pppm forr M MTO O/Z ZrPP P caatalyyst.. Th hat dem monnstrratees decr d rease of o the t eleectrron dennsitty of o H frrom m phhennyl rringg by y thee boondding g off electrronnegaativve O frrom m MTO M O.

Fig. F . 5 1H MA AS NM MR R spectrra oof ZrP Z PP supp portt (aa) an nd 8.1% M MT TO//ZrP PP ccataalysst (bb).

14 4

XPS study is carried out to obtain information on the chemical states of ZrPP support and MTO/ZrPP catalyst (Fig. 6). For ZrPP support, the XPS signal of O 1s with binding energies (BEs) around 532.0 eV is attributed to the O-P from ZrPP [56]. Meanwhile, two O 1s signals with BEs around 532.4 and 531.6 eV could be distinguished for MTO/ZrPP catalyst. The former could be attributed to the O-P, which show a shift to a higher BEs (0.4 eV) due to the hydrogen-bonding interaction after anchoring. This shift is related to the electron density of the electron from the outer space of the element. Because of the strong electronegativity (the O atom of MTO), the lone pair electrons on the support migrate to MTO and decrease the electron density of the P atom from ZrPP, leading to the strengthen of the bondage of extranuclear electron from P nucleus and then the binding energy shift to a higher place. Meanwhile, the O 1s signal at 531.6 eV could be assigned to the O=Re [30]. In addition, two Re 4f signals of MTO/ZrPP located at 46.4 and 43.6 eV represent the Re7+ and Re5+/4+ species, respectively [30]. The appearance of Re5+/4+ species may be resulted from a partial reduction of MTO during XPS experiments for the supported MTO sample.

15

Fig.. 6 XPS specctraa off ZrP PP supppoort andd 8.1 1% MT TO/ZrrPP cattalyyst.

16 6

Table 1 Reaction results of epoxidation of cyclohexene over MTO/ZrPP catalysta. entry

catalyst

Re loading

oxidant

(wt%)

a

time

conv.

(h)

(%)

TON

selectivity to

ref.

epoxide (%)h

1

-

-

UHP

6

0.4

-

100

this work

2

ZrPP

-

UHP

6

1.2

-

100

this work

3

MTO/ZrPP

0.6

UHP

6

7.8

238.5

100

this work

4

MTO/ZrPP

3.1

UHP

6

41.9

245.3

100

this work

5

MTO/ZrPP

4.4

UHP

6

60.4

254.4

100

this work

6

MTO/ZrPP

8.1

UHP

6

92.2

208.5

100

this work

7

MTO/ZrPP

9.9

UHP

6

86.3

160.0

100

this work

8

MTO/Nb2O5b

1.2-2.0

UHP

0.5

100

75

100

[18], Table 1, entry 1

9

MTO/zeolitec

3.4

85% H2O2

10

47

51.4

95

[20], Table 2, entry 1

10

MTO/polymerd

9.6

30% H2O2

1

94

15.9

95

[28], Table 4, entry 3

11

MTO/hybrid silicae

3.7

30% H2O2

3

99.9

43.2

86.4

[31], Table 1, entry 1

12

MTO/hybrid silicaf

0.17

unhydrous H2O2

-

-

103

76

[32]

13

MTO/hybrid silicag

1.4

30% H2O2

4

83.8

15.6

62.1

[33], Table 1, entry 5

Reaction conditions: catalyst 0.1 g, cyclohexene 9.8 mmol, UHP 19.2 mmol, methanol 247.2 mmol, room temperature, time 6 h. 17

b

The supported MTO/Nb2O5 catalyst was prepared in either of two ways, by direct sublimation and by placing a toluene solution of MTO in contact with the niobia.

c

The epoxidation reaction was carried out by mixing zeolite NaY, 85% H2O2 oxidant, alkene substrate, and MTO.

d

Poly(4-vinylpyridine) 25% cross-linked with divinylbenzene was used as polymeric support for preparation of the heterogeneous MTO catalyst.

e

The epoxidation reaction was carried out by mixing silica tethered with poly(ethylene oxide) and poly(propylene oxide), 30% H2O2 oxidant, alkene substrate, and MTO.

f

The heterogenization of MTO inside the porous systems of hybrid silica matrixes through a sol-gel method using 1,4-bis(triethoxysilyl)benzene as a co-condensation agent and 4-((3-triethoxysilyl)propylamino)pyridine hydrochloride as a hydrolysable ligand.

g

Silica tethered with γ-(2,2’-dipyridy1)-aminopropyl polysiloxane was used as hybrid support for preparation of the heterogeneous MTO catalyst.

h

The by-product is cyclohexane-1,2-diol.

18

The catalytic performances of the MTO/ZrPP catalysts with different Re loading are evaluated for the epoxidation of cyclohexene using UHP as oxidant (Table 1). In the cases of the absence of catalyst sample and the adopting of pristine ZrPP as catalyst, a very low conversion of cyclohexene reactant (less than 2%) is obtained. With the increase of MTO loading, the catalyst shows a rapid increase in the cyclohexene conversion, which is first increased from 7.8% to 92.2% as the Re loading raises from 0.6 % to 8.1 %. However, further increase of Re loading to 9.9 % results in a slight decrease in cyclohexene conversion, which may be caused by the aggregation of large MTO particles on the ZrPP surface suggested by XRD (Fig. 1). Meanwhile, the decrease of the TON values for the supported MTO catalysts with a larger Re loading could be seen. ZrPP acts as the role of catalyst support and improves the dispersion of MTO species on the surface. The high dispersion of MTO on ZrPP surface may be beneficial for the enhanced catalytic activity for epoxidation reaction. Table 1 exhibits the catalytic performances of several representative heterogeneous MTO catalysts for the epoxidation of cyclohexene reported in the literatures. It is worth noting that our catalytic system (entry 3-7) exhibits a relatively higher TON values and only cyclohexene oxide could be detected as the reaction product by GC for all the samples with UHP as oxidant. It has been revealed that UHP is an effective oxidant in the olefin epoxidation over MTO with [17] or without [10],[11] support materials. By using MTO/Nb2O5 catalyst with UHP as oxidant, a 100% selectivity to epoxide was also obtained (entry 8 [18]). Little (