Nanostructured maghemite-supported silver catalysts for styrene epoxidation

Article abstract of DOI:10.1016/S1872-2067(10)60183-0

The supported silver catalyst Ag/KOH-γ-Fe₂O₃ was prepared by simple impregnation and liquid reduction and then structurally characterized by X-ray diffraction, transmission electron microscopy, and X-ray photoelectron spectroscopy. T

Full text of DOI:10.1016/S1872-2067(10)60183-0

CHINESE JOURNAL OF CATALYSIS
Volume 32, Issue 3, 2011
Online English edition of the Chinese language journal
RESEARCH PAPER
Cite this article as: Chin J Catal, 2011, 32: 428–435.
Nanostructured Maghemite-Supported Silver Catalysts for
Styrene Epoxidation
PAN Zhenyan1, HUA Li1, QIAO Yunxiang1, YANG Hanmin2, ZHAO Xiuge1, FENG Bo1, ZHU Wenwen1,
HOU Zhenshan1,*
1Key Laboratory for Advanced Materials, Research Institute of Industrial Catalysis, East China University of Science and Technology, Shanghai
200237, China
2Key Laboratory of Catalysis and Materials Science of the State Ethnic Affairs Commission & Ministry of Education, Hubei Province,
South-Central University for Nationalities, Wuhan 430074, Hubei, China
Abstract: The supported silver catalyst Ag/KOH-Ȗ-Fe2O3 was prepared by simple impregnation and liquid reduction and then structurally
characterized by X-ray diffraction, transmission electron microscopy, and X-ray photoelectron spectroscopy. The supported silver cata-
lysts were highly efficient for the epoxidation of styrene using tert-butyl hydroperoxide as the oxidant and ethyl acetate as the reaction
medium. The addition of KOH was found to increase the catalytic activity and selectivity significantly. Additionally, the magnetically re-
coverable Ȗ-Fe2O3, as a support, allowed for an easy separation and recycling of the catalyst after the reaction.
Key words: Ȗ-iron oxide; epoxidation; styrene; silver; nanoparticle; magnetic separation
The epoxidation of styrene is of considerable academic and
industrial interest. Styrene oxide is a crucial and useful inter-
mediate for fine chemicals and pharmaceuticals [1,2]. Al-
though homogeneous catalysts always have high activity in
epoxidation reactions, the conventional homogeneous catalyst
systems have serious drawbacks such as catalyst recovery and
product separation being very difficult [3,4]. Therefore, cata-
lyst systems with high activity and strong stability are of in-
terest. Additionally, catalysts should be able to be separated
and reused easily.
with TBHP as oxidant and toluene as solvent. In addition,
Co-Y-ZrO2 [2], Ag/Į-Al2O3 [6], Au/CNTs [7], Au-HAP [8],
CuO/Ga2O3 [9], Mn-Ti-Al-MCM-41 [10], and Co/TS-1 [11]
catalysts have also been reported and used for styrene epoxi-
dation. Among the reported catalysts, the Ag-based heteroge-
neous catalysts are especially efficient in the epoxidation re-
action.
Iron oxides are environmentally friendly and relatively
cheap catalyst supports. There are many different forms of iron
oxides such as Į-Fe2O3, Ȗ-Fe2O3, Fe3O4, and FeO [12]. Gener-
ally, both Ȗ-Fe2O3 and Fe3O4 have ferromagnetic properties.
Nevertheless, Ȗ-Fe2O3 as a catalyst support has many advan-
tages. Firstly, Ȗ-Fe2O3 is very stable during the reaction while
Fe3O4 is gradually oxidized to Ȗ-Fe2O3 in the presence of an
oxidant [13,14]. Secondly, the ferromagnetic property of
Ȗ-Fe2O3 makes the isolation and recycling of the catalyst more
convenient. Although pure Ȗ-Fe2O3 often results in low cata-
lytic activity during selective oxidations under mild reaction
conditions [3,15], it is very attractive as a magnetically sepa-
Supported heterogeneous catalysts have separation and re-
cycling advantages compared to homogenous catalysts. Vari-
ous heterogeneous catalysts have been used for this reaction.
Liu et al. [5] studied styrene epoxidation using the Ag-Ȗ-ZrP
catalyst with tert-butyl hydroperoxide (TBHP) as an oxidant
and acetonitrile as the reaction medium, and they achieved a
high yield of epoxide (89.8%). Zhang et al. [1] reported that the
Ag-Fe3O4 nanocomposite gives high activity, selectivity and it
is a magnetically separable catalyst after styrene epoxidation
Received 25 October 2010. Accepted 8 December 2010.
*Corresponding author. Tel: +86-21-64251686; Fax: +86-21-64253372; E-mail: houzhenshan@ecust.edu.cn
Foundation item: Supported by the National Natural Science Foundation of China (20773037, 21073058), the Research Fund for the Doctoral Program of Higher
Education of China (20100074110014), and the Key Laboratory of Catalysis and Materials Science of the State Ethnic Affairs Commission and Ministry of Education,
Hubei Province, South-Central University for Nationalities (CHCL10001), China.
Copyright © 2011, Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier BV. All rights reserved.
DOI: 10.1016/S1872-2067(10)60183-0

PAN Zhenyan et al. / Chinese Journal of Catalysis, 2011, 32: 428–435
rable catalyst support [16–18].
1.1.2 Preparation of KOH-Ȗ-Fe2O3
Although various supported catalysts have been used for
styrene epoxidation, previous catalytic systems required toxic
organic solvents (toluene, acetonitrile, dichloromethane,
1,2-dichloroethane, and other toxic organic solvents) and,
additionally, filtration or centrifugation for the separation of
the catalyst. This may result in a mechanic loss of catalyst. In
addition, alkali promoters such as K2CO3 and NaHCO3 are
always added to obtain high activity and selectivity in these
oxidation reactions [19,20] but these bases are difficult to reuse
and they generate waste after the reaction. To eliminate these
problems, a supported-base catalyst has been proposed and
used accordingly [21–25]. Therefore, a clean separation and
cost effective regeneration of the catalyst and purification of
products can be achieved.
Ȗ-Fe2O3 (2.00 g) was impregnated with KOH (0.008 g,
SCRC, AR grade), dissolved in 1 ml of distilled water in a
beaker and then stirred at room temperature for 24 h. After-
wards, the mixture was dried under vacuum at 80 °C for 2 h
followed by calcination at 300 °C for 2 h. Catalysts with dif-
ferent KOH loadings were also prepared.
1.1.3 Preparation of 10%Ag/0.4%KOH-Ȗ-Fe2O3
KOH-Ȗ-Fe2O3 (0.200 g), polyvinylpyrrolidone (PVP, 0.100
g, SCRC, AR grade), and AgNO3 (0.032 g, (SCRC, AR grade)
were dispersed in 1 ml of distilled water in a beaker. The
mixture was stirred at 80 °C for 8 h until all the silver ions in
the solution were adsorbed onto the support (No residual silver
ions were detected in the aqueous phase by the sodium chlo-
ride (SCRC, AR grade) test. Cooling the solution to 0 °C, a
NaBH4 (0.5 mol/L, 2 ml, SCRC, AR grade) aqueous solution
was added to the mixture and stirred for 1 h. The materials were
then washed three times with distilled water and ethanol. The
washed catalyst particles were dried under vacuum at 80 °C for
2 h followed by aging for 3 h at 80 °C under nitrogen. The
content of potassium hydroxide in the catalyst was about 0.4%
as determined by ICP-AES, which confirms that the loss of
potassium hydroxide was negligible during catalyst prepara-
tion. The preparation of the magnetically recoverable catalysts
Ag/KOH-Ȗ-Fe2O3 is shown in Scheme 1.
We used silver as a catalytically active component and then
prepared a heterogeneous epoxidation catalyst by combining
the advantages of a magnetic support with a supported-base
promoter. In the first step, the magnetic Ȗ-Fe2O3 nanoparticles
were modified by solid base KOH and then silver was immo-
bilized on the base-modified magnetic support by simple im-
pregnation and consequent liquid reduction. Subsequently,
styrene epoxidation was carried out over the prepared magnetic
catalyst with low toxic ethyl acetate as the reaction solvent. We
found that the as-obtained catalyst was very efficient for sty-
rene epoxidation. Finally, the recyclability of the ferromagnetic
nanocatalyst was also examined.
1 Experimental
1.2 Characterization of the catalysts
1.1 Preparation of the catalysts
1.1.1 Preparation of magnetic Ȗ-Fe2O3 particles
The X-ray diffraction (XRD) pattern of the catalysts was
analyzed by powder XRD D/max 2550 VB/PC using Cu KĮ1
radiation (Ȝ = 0.154056 nm). The sample was scanned over a
2ș range from 10° to 80° at a scanning rate of 0.02°/s. Trans-
mission electron microscopy (TEM) was performed with a
JEOL JEM 2100T HRTEM instrument operating at 200 kV.
X-ray photoelectron spectroscopy (XPS) measurements were
performed on a Thermo ESCALAB 250 spectrometer using
Nonmonochro Al KĮ radiation. The binding energies were
calibrated with the C 1s level of adventitious carbon (284.8 eV)
as the internal standard. The Ag and K contents were analyzed
by ICP-AES analysis on a Varian 710-ES instrument. A
Thermo Nicolet Nexus 670 was used for FT-IR characteriza-
tion.
FeCl3·6H2O (5.00 g, 18.5 mmol, Sinopharm Chemical Re-
agent Co., Ltd., Shanghai (SCRC), AR grade) and
(NH4)2SO4·FeSO4·6H2O (5.44 g, 13.9 mmol, SCRC, AR grade)
were dissolved in 200 ml of distilled water in a beaker.
NH3·H2O (SCRC, AR grade) was then rapidly added to the
mixture under nitrogen and the pH of the solution was adjusted
to above 9. The whole mixture was stirred at 30 °C for a further
2 h. The products were black precipitates. After 3 h of aging the
black precipitates were washed with water and alcohol several
times and subsequently dried at 80 °C for 6 h followed by
calcination at 300 °C for 2 h.
+
+
+
+
+
+
ꢀ
K
K
K
K
K
K
$g
AgNO3(aq) $gꢀ
$gꢀ
reduction
KOH
+
$gꢀ
+
+
+
+
K
K
K
+
K
K
K
$gꢀ
$gꢀ
+
calcination
+
+$gꢀ
K
+
+
+
K
K
K
K
K
Ag/KOH-J-Fe2O3
J-Fe2O3
KOH-J-Fe2O3
Scheme 1. Synthetic route for the magnetically recoverable catalyst Ag/KOH-Ȗ-Fe2O3.

PAN Zhenyan et al. / Chinese Journal of Catalysis, 2011, 32: 428–435
1.3 Catalytic tests
J-Fe2O3
Ag
The catalyst (0.020 g), styrene (0.104 g, 1 mmol, Shanghai
Ling Feng Chemical Reagent Co., Ltd., AR grade), and ethyl
acetate (1.792 g, 2 ml, SCRC, AR grade) were placed in a 25
ml Schlenk glass flask under a N2 atmosphere and then stirred
at room temperature for 30 min. TBHP (0.390 g, 3 mmol,
SCRC, AR grade) was added to the reaction mixture, which
then proceeded for the required time at 80 °C. After the reac-
tion the catalyst was separated simply by a magnet and
washed with ethyl acetate followed by drying in vacuum. The
catalyst was then conveniently used for a next run. The reac-
tion products were analyzed by gas chromatography using an
FID detector.
(3)
(2)
(1)
10
20
30
40
2T/( o
50
60
70
80
)
Fig. 1. XRD patterns of Ȗ-Fe2O3 (1), 10%Ag/0.4%KOH-Ȗ-Fe2O3 before
(2) and after five runs (3).
2 Results and discussion
shows the XRD pattern of the 10%Ag/0.4%KOH-Ȗ-Fe2O3
composites after five runs. There was not much change in the
XRD pattern compared to the fresh catalyst. The lower iron
oxide diffraction was most likely due to the low crystallinity of
iron oxide and the heavy atom effect arising from the silver
atoms [26].
2.1 Characterization results
The formation of metallic Ag and Ȗ-Fe2O3 in the
Ag/KOH-Ȗ-Fe2O3 catalyst was confirmed by powder XRD
(Fig. 1). A single phase of Ȗ-Fe2O3 was prepared successfully.
The XRD pattern of the catalyst (Fig. 1(1)) shows that the
peaks matched well with standard Ȗ-Fe2O3 reflections (JCPDS
25-1402). Figure 1(2) shows the XRD pattern of the
10%Ag/0.4%KOH-Ȗ-Fe2O3 composite. Apart from the char-
acteristic diffraction peaks of the Ȗ-Fe2O3 reflections the clear
diffraction peaks at 2ș = 38.12°, 44.26°, 64.48°, and 77.36° are
assigned to the (111), (200), (220), and (311) crystal planes of
silver, respectively. These are in good agreement with the
standard Ag(0) reflections (JCPDS 65-2871). Figure 1(3)
Figure 2 shows TEM photographs and the size distribution
of the fresh and reused nanoparticles. The nanocomposites
consisted of nearly spherical nanoparticles with an average
diameter of 7–9 nm. The overall shape and size distribution of
the particles did not show obvious changes even after five runs
for the magnetically separable nanocatalyst.
The surface composition and oxidation state of the nano-
catalyst were also analyzed by XPS. Figure 3 shows repre-
sentative XPS spectra of the nanocatalyst. Elemental analysis
confirmed the presence of Ag, Fe, K, C, and O. The observed
(a)
(c)
40
30
20
10
0
3
4
5
6
7
8
9
10 11 12 13 14 15
Particle diameter (nm)
(b)
(d)
40
30
20
10
0
3
4
5
6
7
8
9
10 11 12 13 14 15
Particle diameter (nm)
Fig. 2. TEM images (a and b) and particle size distribution (c and d) of 10%Ag/0.4%KOH-Ȗ-Fe2O3 before (a and c) and after five runs (b and d).

PAN Zhenyan et al. / Chinese Journal of Catalysis, 2011, 32: 428–435
present at a binding energy of 367.84 eV. This was close to the
values of Ag+ in Ag2O [29,30]. Figure 5(b) shows the XPS
spectrum of the freshly reduced Ag/KOH-Ȗ-Fe2O3 catalyst and
the main peaks (Ag 3d5/2) are present at binding energies of
368.09 and 367.60 eV, respectively. This data indicates that the
reduced catalyst contained a higher amount of metallic silver
and less ionic silverin Ag2O [31–33]. The XRD data (Fig. 1(2))
did not indicate the presence of Ag2O but the XPS results did
indicate the possible existence of a few atomic layers of Ag+
species on the nanoparticle surfaces [34]. Figure 5(c) shows the
XPS spectrum of the Ag/KOH-Ȗ-Fe2O3 catalyst after five runs
and the main peaks (Ag 3d5/2) were present at binding energies
of 368.26 and 367.79 eV. Although the XPS results show that a
small amount of Ag2O was still present on the surface of the
nanocatalyst the XRD data (Fig. 1(3)) showed that Ag2O was
not present on the surface of the nanocatalyst. Additionally, the
main peak at 368.26 eV (Ag 3d5/2) for the recycled catalyst
shifted 0.17 eV higher in terms of binding energy compared to
the fresh catalyst. Because a more positive charge density on
typical transition metal atoms results in higher binding ener-
gies, the Ag nanoparticles were more positive after five runs,
which might be due to the oxidant in the reaction system.
The detailed XPS parameters of all the samples are summa-
rized in Table 1. The XPS spectra also contained a K 2p sig-
nal and the K 2p3/2 peak was present at a binding energy of 292
eV [35,36]. The peak intensity and binding energy of the K 2p
peak did not change after these reactions (Table 1).
Fe 2p
O 1s
K 2p
Ag 3d
C 1s
Fe 3p
1000
800
600
400
200
0
Binding energy (eV)
Fig. 3. The XPS spectrum of the 10%Ag/0.4%KOH-Ȗ-Fe2O3 composite.
XPS spectra of C 1s (284.8 eV) and O 1s (530.4 eV) came
from adventitious carbon and iron oxides. In addition, a rela-
tively weak Fe 3p line at 56.1 eV was also detected (Fig. 3).
No other metallic signals were detected in the XPS spectra.
Figure 4 shows the Fe 2p spectra of the different samples.
All the Fe 2p spectra show a typical iron oxide structure with
broad main peaks (Fe 2p3/2 and Fe 2p1/2). The characteristic
doublet of Fe 2p3/2 and 2p1/2 were observed as photoelectron
peaks at 710.5 and 724.1 eV, respectively. The Fe 2p data were
in good agreement with the values reported for Ȗ-Fe2O3 in the
literature [27,28]. These spectra show that the Fe 2p signal
intensity and binding energy of composites (1) and (2) re-
mained unchanged. After reusing Ag/KOH-Ȗ-Fe2O3 five times
there was also no remarkable change in the Fe 2p peak inten-
sity or the binding energy (Fig. 4(3)). These results show that
the addition of KOH or Ag did not affect the structure of
Ȗ-Fe2O3 and Ȗ-Fe2O3 was very stable during catalytic recycling.
To determine the oxidation state of the Ag component the
XPS technique was used. Figure 5 shows the Ag 3d spectra of
different samples and this could be resolved into two spin orbit
components, Ag 3d5/2 and Ag 3d3/2. Figure 5(a) gives the XPS
signal of the catalyst Ag2O/ KOH-Ȗ-Fe2O3, which was not
pre-reduced by NaBH4, and the main peak for Ag 3d5/2 was
(a)
(b)
724.1
710.5
Fe 2p
(3)
(2)
(1)
(c)
378 376 374 372 370 368 366 364 362
Binding energy (eV)
740 735 730 725 720 715 710 705 700
Binding energy (eV)
Fig. 4. Representative Fe 2p XPS spectra of the composites. (1)
Ag/Ȗ-Fe2O3; (2) 10%Ag/0.4%KOH-Ȗ-Fe2O3; (3) 10%Ag/0.4%KOH-Ȗ-
Fe2O3 after five runs.
Fig. 5. Representative Ag 3d XPS spectra of the nanocomposites. (a)
The catalyst was not reduced by NaBH4; (b) 10%Ag/0.4%KOH-Ȗ-Fe2O3;
(c) 10%Ag/0.4%KOH-Ȗ-Fe2O3 after five runs.

PAN Zhenyan et al. / Chinese Journal of Catalysis, 2011, 32: 428–435
oxidation gave a very low conversion (16.4%) in the absence of
Table 1 XPS results of the catalysts
Binding energy (eV)
Fe 2p3/2
a catalyst (Table 2, entry 1), which demonstrates the necessity
for a catalyst in this system.
Catalyst
Ag 3d5/2
367.84
367.60
368.09
367.79
368.26
K 2p3/2
292.20
292.15
a,b
Ag2O/0.4%KOH-Ȗ-Fe2O3
710.43
710.48
2.2.1 Effect of K loading
b
Ag/0.4%KOH-Ȗ-Fe2O3
b,c
The effects of different KOH loadings on the conversion of
styrene and the selectivity for SO were also examined (Table 2,
entries 5–7). When 0.4%KOH/Ȗ-Fe2O3 was used as the cata-
lyst, the selectivities for SO and BZ were 53.6% and 46.4%,
respectively (Table 2, entry 3). Obviously, the Ȗ-Fe2O3 modi-
fied by a solid base KOH improved the selectivity for SO.
Moreover, it can be seen that with an increase in the amount of
KOH loading in the supported-silver catalyst both the conver-
sion of styrene and the selectivity for SO increased corre-
spondingly and a maximum was found for 0.4% KOH (Table 2,
entries 5 and 6). However, when the supported amounts of
KOH exceeded 0.4%, the conversion and selectivity dropped to
75.5% and 82.7% (Table 2, entry 7), respectively.
Ag/0.39%KOH-Ȗ-Fe2O3
710.46
292.20
aThe unreduced catalyst was used.
bThe potassium hydroxide content was determined by ICP-AES.
c10%Ag/0.4%KOH-Ȗ-Fe2O3 was recycled five times.
2.2 Catalytic activity for epoxidation with the magnetic
silver catalyst
The transformation of styrene to styrene oxide (SO) upon
catalysis by the Ȗ-Fe2O3, KOH-Ȗ-Fe2O3, Ag/Ȗ-Fe2O3, and
Ag/KOH-Ȗ-Fe2O3 catalysts was investigated and the results are
listed in Table 2. We found that only benzaldehyde (BZ) was
formed as a main by-product. The content of Ag in the
Ag/KOH-Ȗ-Fe2O3 catalyst has been optimized and we show
that a catalyst loading of 10% Ag gives the most active catalyst.
Therefore, catalysts loaded with 10% Ag were used in the
following studies.
2.2.2 Effect of solvents
As shown in Table 3, the epoxidation of styrene in different
media was investigated. It was clear that the properties of the
solvents greatly affected the conversion and selectivity. Ben-
zoic acid (BA) was the main product obtained under sol-
vent-free conditions after a 15 h reaction (Table 3, entry 2). The
selectivity for SO, BZ, and BA is almost the same (36.4%,
30.9%, and 32.7%, respectively) with a reaction time of 7 h
(Table 3, entry 1). This implied that SO and BZ can be further
oxidized to BA possibly because of the high local concentra-
tion of SO, BZ, and TBHP around the catalyst under sol-
vent-free conditions. When the reaction was performed in an
aqueous medium, with an increase in the reaction time the
selectivity for SO decreased while that for the aldehyde in-
creased (not shown). After a 15 h reaction time, BZ was the
main product with a moderate conversion (Table 3, entry 3).
The change in selectivity for BA and phenylethylene glycol
When single Ȗ-Fe2O3 was used as the catalyst the selectivity
for SO and BZ was 47.8% and 52.2%, respectively (Table 2,
entry 2). The selectivity for SO was improved to 81.9% when
using the 10%Ag/Ȗ-Fe2O3 catalyst (Table 2, entry 4), which
confirmed clearly that both Ag and Ȗ-Fe2O3 components played
a synergic role in the catalytic epoxidation. It has been reported
previously that Ȗ-Fe2O3 provides reactive oxygen species
[37,38] and the active species may easily transfer to the sur-
face of the proximate Ag particles in the nanocatalyst. As a
consequence, this process facilitated the epoxidation of the
styrene molecules adsorbed on the Ag surface. Styrene ep-
Table 2 Epoxidation of styrene with different catalysts in ethyl acetate
O
Table 3 Epoxidation of styrene with the 10%Ag/0.4%KOH-Ȗ-Fe2O3
catalyst in different solvents
Conversion Selectivity (mol%)
Entry
Catalyst
Conversion
(%)
Selectivity (mol%)
(%)
16.4
80.7
78.5
70.0
79.9
89.6
75.5
SO
BZ
Entry
Solvent
SO
36.4
4.8
BZ
BA
32.7
84.9
12.5
0
Enol
0
1
None
41.8
47.8
53.6
81.9
83.0
89.7
82.7
58.2
52.2
46.4
18.1
17.0
10.3
17.3
1a
2
none
none
48.6
30.9
10.3
72.9
14.4
10.3
2
3a
Ȗ-Fe2O3
74.1
0
0.4%KOH-Ȗ-Fe2O3
10%Ag/Ȗ-Fe2O3
3
water
47.8
12.1
85.6
89.7
2.4
0
4
5a
6a
7a
10%Ag/0.1%KOH-Ȗ-Fe2O3
10%Ag/0.4%KOH-Ȗ-Fe2O3
10%Ag/0.91%KOH-Ȗ-Fe2O
4
ethanol
ethyl
37.9
5
89.6
0
0
acetate
acetonitrile
6
90.5
89.6
10.4
0
0
Reaction conditions: styrene 1 mmol, catalyst 20 mg Ag 18.5 ȝmol, ethyl
acetate 1.792 g (2 ml), TBHP 3 mmol, 80 ºC, 15 h, N2 atmosphere. Con-
version and selectivity were determined by GC analysis.
Reaction conditions: styrene 1 mmol, catalyst 20 mg (Ag 18.5 ȝmol),
solvent 2 ml, TBHP 3 mmol, 80 ºC, 15 h (a7 h), N2 atmosphere. Conver-
sion and selectivity were determined by GC analysis.
aThe contents of potassium hydroxide were determined by ICP-AES.

PAN Zhenyan et al. / Chinese Journal of Catalysis, 2011, 32: 428–435
was found to be quite small. The observed variation in product
O
selectivity suggests that SO undergoes C–C bond breakage to
produce BZ as the main product in the water medium. Solvents
like ethanol (SCRC, AR grade), ethyl acetate, and acetonitrile
(Shanghai Ling Feng Chemical Reagent Co., Ltd., AR grade)
(Table 3, entries 4–6) are more favorable for the formation of
SO. The conversion of styrene in ethyl acetate and acetonitrile
was higher and reached 89.6% and 90.5%, respectively. Be-
cause ethyl acetate is a more environmentally benign solvent it
was used as a reaction medium in the following catalytic reac-
tions.
O
OH
TBHP
O
aqueous phase
or no solvent
Scheme 2. Reaction scheme for the epoxidation of styrene.
Styrene conversion
SO selectivity
100
80
60
40
20
0
2.2.3 Reaction kinetics and catalyst recycling
The catalytic performance of the epoxidation of styrene
versus reaction time was studied and this is shown in Fig. 6.
The conversion of styrene increased continuously with time
and was close to full conversion after 24 h. The yield of sty-
rene epoxide was comparable with many of the heterogeneous
catalytic systems reported earlier, even in other organic sol-
vents [39,40]. Interestingly, both the selectivity for SO (ca.
89%) and BZ (ca. 10%) were low and constant during the
whole process. This data shows that the SO and BZ were
formed simultaneously in the reaction. The selectivity for SO
was quite high and the formation of BZ or BA was suppressed
considerably in ethyl acetate compared with that in aqueous
medium or without the use of a solvent. Our proposed reaction
mechanism is shown in Scheme 2.
1
2
3
4
5
Catalyst recycle
Fig. 7. Reuse of the 10%Ag/0.4%KOH-Ȗ-Fe2O3 catalyst for the epoxi-
dation of styrene. Reaction conditions: styrene 1 mmol, catalyst 20 mg
(Ag 18.5 ȝmol), ethyl acetate 1.792 g (2 ml), TBHP 3 mmol, 80 ºC, 24 h,
N2 atmosphere.
89.6% conversion for the second run and 80.1% conversion
for the fifth run. Because the charge density, the overall shape
and the size distribution of the catalytically active silver
nanoparticles did not change significantly after consecutive
runs as discussed above a gradual decrease in the conversion of
styrene occurred. This was possibly because of the dissolution
of silver or potassium leaching into the reaction mixture, es-
pecially in the liquid phase oxidation reactions. To address this
possibility, the isolated products obtained after magnetic
separation were analyzed by ICP-AES. We found that the Ag
concentration in the reaction effluent was only 0.3 × 10–6 and
the content of KOH in the reused Ag/KOH-Ȗ-Fe2O3 catalyst
remained unchanged (0.39%). This indicates that the leaching
of Ag or K during the reaction was negligible. Based on the
analysis above, the strong adsorption of the reaction products
onto the surface of the catalyst could be the reason for the
slow deactivation in the catalytic runs. This is also shown by
the C–H stretching bands (2961.2 and 2926.7 cm–1) present in
the IR spectra of the reused catalyst by comparison with the
fresh catalyst. Because of the high affinity of Ag and KOH for
the magnetic support, a highly efficient magnetic separation
can be achieved for this catalytic system.
100
80
60
40
20
0
100
80
60
40
20
0
Styrene conversion
SO selectivity
BZ selectivity
0
5
10
15
20
25
Reaction time (h)
Fig. 6. Catalytic performance of the 10%Ag/0.4%KOH-Ȗ-Fe2O3 catalyst
as a function of reaction time. Reaction conditions: styrene 1 mmol, cata-
lyst 20 mg (Ag 18.5 ȝmol), ethyl acetate 1.792 g (2 ml), TBHP 3 mmol, 80
ºC, N2 atmosphere.
Recycling experiments were also undertaken for the epoxi-
dation of styrene using the Ag/KOH-Ȗ-Fe2O3 catalyst (Fig. 7).
After the first reaction, the catalyst was recovered easily using
a permanent magnet and the products were simply collected by
decantation. The recovered catalyst was then washed with ethyl
acetate (1 ml × 3) and dried under vacuum. The catalyst was
reused in successive epoxidations by the addition of fresh
solvent, TBHP, and substrate. The catalytic system gave
2.2.4 Epoxidation reaction of different substrates
In the next step, the generality of the oxidative reaction was
investigated by varying the alkene substrates and comparing
their conversions and product selectivities as listed in Table 4.
The 10%Ag/0.4%KOH-Ȗ-Fe2O3 catalyst showed much higher

PAN Zhenyan et al. / Chinese Journal of Catalysis, 2011, 32: 428–435
2007, 8: 1556
Table
4
Epoxidation of various substrates with the 10%Ag/
10 Wang G J, Liu Zh W, Liu Y W, Liu G Q, Xu M X, Wang L.
Chin J Catal, 2008, 29: 1159
0.4%KOH-Ȗ-Fe2O3 catalyst in ethyl acetate
Selectivity (mol%)
Epoxides Aldehyde
Entry Substrate Conversion (%)
11 Jiang J, Li R, Wang H L, Zheng Y F, Chen H N, Ma J T. Catal
Lett, 2008, 120: 221
Enols
—
1
2
3
4
styrene
cyclohexene
cyclooctene
1-octene
89.6
26.6
28.5
14.2
89.7
30.1
91.4
93.8
10.3
—
12 Chou K S, Lee S J. Colloid Surf A, 2009, 336: 23
13 Rane K S, Verenkar V M S, Pednekar R M, Sawant P Y. J Mater
Sci Mater Electron, 1999, 10: 121
69.9
8.6
—
—
6.2
14 Basak S, Rane K S, Biswas P. Chem Mater, 2008, 20: 4906
15 Shi F, Tse M K, Pohl M M, Radnik J, Brückner A, Zhang S M,
Beller M. J Mol Catal A, 2008, 292: 28
Reaction conditions: substrate 1 mmol, catalyst 20 mg (Ag 18.5 ȝmol),
ethyl acetate 1.792 g (2 ml), TBHP 3 mmol, 80 °C, 15 h, N2 atmosphere.
Conversion and selectivity were determined by GC analysis
16 Zhang Y, Li Z, Sun W, Xia C G. Catal Commun, 2008, 10: 237
17 Zhang Y, Xia C G. Appl Catal A, 2009, 366: 141
18 Liu H Q, Liang M H, Xiao C, Zheng N, Feng X H, Liu Y, Xie J
L, Wang Y. J Mol Catal A, 2009, 308: 79
activity for styrene than the cycloolefins and the linear alkene.
The styrene, cyclooctene (Alfa Aesar, AR grade), and
1-octene (Alfa Aesar, AR grade) were converted into corre-
sponding epoxides with excellent selectivity (Table 4, entries
1, 3, and 4) but the oxidation of cyclohexene (SCRC, AR
grade) resulted in the production of enols (Table 4, entry 2).
19 Oliveira R L, Kiyohara P K, Rossi L M. Green Chem, 2010, 12:
144
20 Qi B, Lu X H, Zhou D, Xia Q H, Tang Z R, Fang S Y, Pang T,
Dong Y L. J Mol Catal A, 2010, 322: 73
21 Wilson K, Hardacre C, Lee A F, Montero J M, Shellard L. Green
Chem, 2008, 10: 654
3 Conclusions
22 Li Y, Zhao X Q, Wang Y J. Chin J Catal, 2004, 25: 633
23 Macala G S, Matson T D, Johnson C L, Lewis R S, Iretskii A V,
Ford P C. ChemSusChem, 2009, 2: 215
A supported silver catalyst was successfully prepared by the
reduction of silver salts and they were supported on the mag-
netically recoverable Ȗ-Fe2O3. The magnetically recyclable
catalyst system was highly efficient in catalyzing the epoxida-
tion of styrene with ethyl acetate as the reaction medium. The
reduced metallic Ag component and the supported-base pro-
moter played important roles in improving activity and selec-
tivity during styrene epoxidation. Obviously, this catalyst
system is green, simple, easily separable, and recyclable.
24 Calvino-Casilda V, Martin-Aranda R M, Lopez-Peinado A J,
Sobczak I, Ziolek M. Catal Today, 2009, 142: 278
25 Sharma S K, Parikh P A, Jasra R V. J Mol Catal A, 2010, 317: 27
26 Teng X W, Black D, Watkins N J, Gao Y L, Yang H. Nano Lett,
2003, 3: 261
27 Liu X M, Li Y S. Mater Sci Eng C, 2009, 29: 1128
28 Pereira C, Pereira A M, Quaresma P, Tavares P B, Pereira E,
Araújo J P, Freire C. Dalton Trans, 2010, 39: 2842
29 Lutzenkirchen-Hecht D, Strehblow H H. Surf Interface Anal,
2009, 41: 820
Acknowledgments
The authors would like to express their thanks to Prof.
Walter Leitner for the helpful suggestion and Dr. Nils
Theyssen for assistance.
30 You X F, Chen F, Zhang J L, Anpo M. Catal Lett, 2005, 102: 247
31 Sangpour P, Babapour A, Akhavana O, Moshfegh A Z. Surf In-
terface Anal, 2009, 41: 157
32 Mackova A, Malinsky P, Bocan J, Svorcik V, Pavlik J, Stryhal
Z, Sajdl P. Phys Stat Sol C, 2008, 5: 964
References
33 Zhang D H, Li H B, Li G D, Chen J S. Dalton Trans, 2009, 47:
10527
1
Zhang D H, Li G D, Li J X, Chen J S. Chem Commun, 2008:
3414
34 Schnippering M, Carrara M, Foelske A, Kötz R, Fermín D J.
Phys Chem Chem Phys, 2007, 9: 725
2
3
Tyagi B, Shaik B, Bajaj H C. Catal Commun, 2009, 11: 114
Garade A C, Bharadwaj M, Bhagwat S V, Athawale A A, Rode
C V. Catal Commun, 2009, 10: 485
35 Miyakoshi A, Ueno A, Ichikawa M. Appl Catal A, 2001, 219:
249
36 Hope G A, Woods R, Buckley A N, White J M, McLean J. Inorg
Chim Acta, 2010, 363: 935
4
5
Dioumaev V K, Bullock R M. Nature, 2003, 424: 530
Liu J H, Wang F, Gu Z G, Xu X L. Catal Commun, 2009, 10:
868
37 Schubert M M, Hackenberg S, Van Veen A C, Muhler M, Plzak
V, Behm R J. J Catal, 2001, 197: 113
6
Chimentão R J, Kirm I, Medina F, Rodríguez X, Cesteros Y,
Salagre P, Sueiras J E, Fierro J L G. Appl Surf Sci, 2005, 252: 793
Liu J H, Wang F, Xu T, Gu Z G. Catal Lett, 2010, 134: 51
Liu Y M, Tsunoyama H, Akita T, Tsukuda T. Chem Commun,
2010, 46: 550
38 Liu H C, Kozlov A I, Kozlova A P, Shido T, Iwasawa Y. Phys
Chem Chem Phys, 1999, 1: 2851
7
8
39 Yang Y, Guan J Q, Qiu P P, Kan Q B. Transition Met Chem,
2010, 35: 263
9
Choudhary V R, Jha R, Chaudhari N K, Jana P. Catal Commun,
40 Lu X N, Yuan Y Z. Appl Catal A, 2009, 365: 180