Heterocycle-modified 12-tungstophosphoric acid as heterogeneous catalyst for epoxidation of propylene with hydrogen peroxide

Article abstract of DOI:10.1016/j.molcata.2013.06.014

12-Phosphotungstic acid (PWA)-based complex catalysts for propylene epoxidation were prepared by modification with nitrogen-containing heterocycles such as imidazole, pyrazole and 1,2,4-triazole. Heterocycles played an important role in promotion of the c

Full text of DOI:10.1016/j.molcata.2013.06.014

Journal of Molecular Catalysis A: Chemical 378 (2013) 232–237
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Heterocycle-modified 12-tungstophosphoric acid as heterogeneous
catalyst for epoxidation of propylene with hydrogen peroxide
a
a
b,∗
a,∗∗
Hyun Jae Kim , YuKwon Jeon , Joo-Il Park , Yong-Gun Shul
a
Department of Chemical and Biomolecular Engineering, Yonsei University, Seoul, Republic of Korea
Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Fukuoka 816-8580, Japan
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 8 May 2013
Received in revised form 21 June 2013
Accepted 22 June 2013
Available online 1 July 2013
12-Phosphotungstic acid (PWA)-based complex catalysts for propylene epoxidation were prepared by
modification with nitrogen-containing heterocycles such as imidazole, pyrazole and 1,2,4-triazole. Hete-
rocycles played an important role in promotion of the catalytic activity and selectivity to propylene oxide,
mostly due to the strong electronic interactions between heterocycles and terminal tungsten-oxygen on
PWA. Propylene epoxidation over the complex catalyst seriously depended on the pressure and the tem-
perature, favoring high pressure and low temperature for improved conversion and yields, respectively.
Keywords:
◦
The complex catalysts maintained good thermal stability up to 450 C as well as recycling performance for
Reaction engineering
Propylene epoxidation
Heteropoly acid
Catalyst selectivity
Multiphase reactions
Stability
5
reaction runs without any significant loss of catalyst, suggesting that they are promising heterogeneous
catalysts on propylene epoxidation.
© 2013 Elsevier B.V. All rights reserved.
1
. Introduction
unique redox property, HPAs are more active than the conven-
tional solid acid catalysts in selective oxidation and hydroxylation
[7]. Even though the mechanism of HPAs-catalyzed reaction is
not yet fully understood, it has been suggested that phospho-
tungstic acid reacts with hydrogen peroxide to produce small
peroxotungstophosphates formed on polynuclear sites such as
Owing to wide usage of epoxides as raw material for some
resins, paints, surfactants and intermediates in organic synthesis,
propylene epoxidation is regarded as one of the important reac-
tion in the field of chemical industry producing propylene oxide
over 4.5 million tons annually [1]. In general, there are two classi-
cal processes of propylene epoxidation, chlorohydrin process and
hydroperoxide process. The chlorohydrin process is fairly simple,
requiring only two reaction steps, but it causes to serious environ-
mental problem producing large outputs of chloride-laden sewage
3−
{PO [WO(O ) ] } , which catalyze oxidation [8]. However, one
4
2 2 4
of the major disadvantages is the selectivity to propylene oxide;
large amount of co-products yields 2–4 times higher than that
of propylene oxide [6]. Furthermore, since these reaction systems
are homogeneous catalysis, there are several problems that should
overcome for application: low thermal sustainability and high solu-
bility of HPA in polar solvents is a major obstacle to catalyst/product
separation and catalyst recycling.
[
2–5]. The recently built plants adopted the latter, considering envi-
ronmental aspect. Since they utilize hydroperoxide having high
content of active oxygen species as oxidants, the environment-
friendly by-products such as H O are produced. This process usually
adopts transition-metal catalysts including titanosilicates, zeolite
supported metal catalysts [6]. Lately, in several industrial processes,
Several groups have conducted research in order to compen-
sate these inherent defects of HPA catalysts. It was reported
that partial or total substitution of proton to other cations in
Keggin polyoxometalates could enhance thermal stability of HPA
without any structural change of primary Keggin backbone in
heteropoly anion [9,10]. Organic-modified polyoxometalates were
also extensively highlighted as heterogeneous hybrid catalysts. As
the redox properties of the inorganic cluster of HPA were modi-
fied by organic ␲-electron, high yield of product could be achieved
by applying tetrabutylammonium-modified or pyridine-modified
HPA catalysts in epoxidation of olefins or benzene hydroxylation,
respectively [1,11]. However, it is still ambiguous how the organic
2
heteropoly acids (HPAs) with Keggin structure such as H PW12O40
3
has been paid attention as a substitute for traditional transition-
metal catalyst. Due to its strong acidity, uniform acidic sites, and
∗
Corresponding author. Tel.: +81 92 583 7633; fax: +81 92 583 7640.
Corresponding author. Tel: +82 2 2123 2758, fax: +82 2 312 6401.
E-mail addresses: parkjoo9@asem.kyushu-u.ac.jp (J.-I. Park),
∗
∗
shulyg@yonsei.ac.kr (Y.-G. Shul).
1
381-1169/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2013.06.014

H.J. Kim et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 232–237
233
␲
-bond affects the catalytic activity and what factor should be con-
sidered in choosing hetero-compound for the modified complex
catalyst.
In this work, we attempt to modify heteropoly acids with two
or three nitrogen atoms-containing heterocycles and apply them
as the heterogeneous catalyst for epoxidation of propylene. In
preparation of catalysts, imidazole, pyrazole or 1,2,4-triazole were
utilized for modification of 12-phosphotungstic acids in order to
investigate the effect of heterocycles on the complex catalysts. Fur-
thermore, the potential of the complex catalyst was assessed as the
heterogeneous catalyst for propylene epoxidation by focusing on
regioselectivity in various reaction conditions, thermal stabilities
and recyclabilities.
2
. Experimental
2.1. Catalyst preparation and characterization
Fig. 1. XRD patterns of raw PWA and imidazole-modified PWA by varying imidazole
contents.
1
2-Phosphotungstic acid (PWA, H PW12O40, >98.0%; Kanto
3
Chemical) was used as heteropoly acids in this study. The
heterocycle-modified PWA composite catalysts were prepared by
ion-exchange with nitrogen containing heterocycle compound
such as imidazole (>98.0%, TCI Chemical), pyrazole (>98.0%, TCI
Chemical) and 1,2,4-triazole (>98.0%, TCI Chemical). 5 M of aque-
ous solution of heterocycle was added dropwise to 0.1 M aqueous
solution of PWA with designated volumetric ratios. The mixture
was stirred for 24 h and then was centrifuged at 1200 rpm to sep-
arate the precipitates from solution. The precipitate was washed
CO2 into methane and detect in FID with the detection limit of
several ppm level. GC calibration for propylene and CO2 were per-
formed using mass flow controllers (MKS) with the balanced in He
gas. PO, acetone, IPA and propanol were calibrated by vaporizing
known quantities of liquids in a heated, evacuated 2000 cm3 stain-
less steel tank and using He as a carrier gas. All calibration data
showed linear 5-point plots (R2 of at least 0.996) with peak area as
the basis for determining the conversion and yields.
◦
with deionized water thrice and dried at 100 C overnight. The
The conversion and selectivity was calculated from the propyl-
ene oxide data as shown below.
obtained samples were denoted as ([Heterocycle]x/PWA), where
x is the molar ratio of heterocycle to PWA. In order to investigate
the thermal stability of the catalyst, the catalyst was placed into the
oven and oven was slowly heated up to the designated temperature
1
− XC3H8
Conversion (%) =
× 100
× 100
X
0,C3H8
◦
◦
(
200–500 C) with a heating rate of 2 C/min.
XPO
X-ray diffractometer (XRD, Rigaku D/MAX-II A) with monochro-
Selectivity (%) = ꢀ
Xproduct
mated Cu-K␣ radiation source at 30 kV and 30 mA were used to
investigate the crystal structure change by addition of hetero-
cycles. X-ray diffraction patterns were obtained in a scan range
where X0,C H , X
, XPO and Xproduct are the moles of initial C H ,
3 8
C
H
8
3
8
3
remaining C H8 after reaction, propylene oxide and each product,
3
◦
◦
of 2ꢀ = 5–60 with a scan rate of 4 /min. In order to clarify the
interaction between PWA and heterocycles, infrared spectra were
obtained from Fourier transformed infra red spectroscopy (FT-IR,
Digilab Excalibur series) at universal attenuated total reflection
respectively.
The catalyst recycle test was performed to observe the ther-
mal and chemical stability of complex catalysts. After each cycle,
the spent catalyst was collected, washed by acetone, dried under
vacuum and then weighed to check the loss of the catalyst during
reaction and separation of the catalyst from product. There was no
change in the weight of the weight of the used catalyst, indicating
that the prepared catalyst was stable under the reaction conditions.
−
1
(
ATR) mode in the range of 550–4000 cm . The thermogravimetric
analysis (TGA, Rigaku Thermo plus TG8120) was conducted under
◦
air atmosphere with a heating rate of 2 C/min in the range of
◦
2
5–800 C.
2.2. Catalytic reaction
3. Results and discussion
Epoxidation of propylene was carried out in a micro-batch reac-
3.1. Characterization of heterocycles-modified heteropoly acid
catalysts
tor. 30 mg of the catalyst, 0.2 mL of hydrogen peroxide (35% in H O,
2
Aldrich) and 0.3 mL of methanol as medium were introduced into
the stainless steel reactor having 10 mL of inner volume and then
the reactor was charged with the 0.5 MPa of propylene. The reactor
was vertically positioned and agitated by the temperature control-
lable shaker for 8 h of the reaction. To minimize the loss of product
Fig. 1 shows the comparison of XRD patterns among the raw
PWA and imidazole-modified PWAs as a function of the ratio of
imidazole to PWA (x = 1 and 3). From the XRD pattern, it was
confirmed that the raw PWA is the hexa-hydrated form, the
characteristic peaks of the body centered cubic structure [12].
It was found that the crystal structures of imidazole-modified
PWA depended on the contents of imidazole. The XRD pattern of
◦
samples (gases and liquids), the quenching (1 C water bath) has
been performed after reaction. When the temperature of reactor
◦
reached at 5 C, gas phases have been collected in PVDF bag (0.1 L)
through degassing line attached to reactor. The conversion of feed
and products (gases and liquids) were determined by analysis of
gas chromatography (GC; Hewlett-Packard 6890, USA) equipped
with flame ionization detector (FID) and DB 1701 capillary column
(imidazole) /PWA exhibit totally different from that of the raw
3
PWA, while (imidazole) /PWA contains both patterns of the raw
1
PWA and (imidazole) /PWA. It can be deduced that the imidazole
3
content at (imidazole) /PWA was not enough to change the crystal
1
(
J&W Science). CO in the gases has been specially analyzed by using
structure of PWA, but at (imidazole) /PWA, the surface was entirely
2
3
methanizer-FID with Carboxene 1006 column, which can convert
modified by imidazole addition.

2
34
H.J. Kim et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 232–237
Fig. 2. XRD patterns of raw PWA and PWAs modified with various heterocycles.
Fig. 4. XRD patterns of raw PWA and (imidazole)3/PWA by various heat treatment
temperatures.
XRD patterns of the modified PWAs with various heterocycles
are compared in Fig. 2. No matter what heterocyclic compounds
added, all crystal structures of modified PWAs were transformed.
However, unlike others, the XRD pattern of the raw PWA still
slightly appeared on pyrazole-modified PWA, suggesting that only
pyrazole was not fully interacted on the surface of PWA clusters
even though the ratio of heterocycle to PWA was set to 3. One
may explain that since pyrazole has inductive effect due to two
nitrogen atoms in adjacent positions, electron density of pyrazole
is more localized, showing weaker resonant stabilization over the
ring compared to others. Therefore, even though some amount of
pyrazole molecules are anchored onto PWA, it may cause repulsive
force to hinder other pyrazole molecule to adsorb on PWA. On the
other hand, both imidazole and 1,2,4-triazole are amphoteric and
functioned as both acid and base by delocalized electron pairs in
their resonance structures [13].
heterocycles and P O. Since P O bands are located at the center
of the structure, the heterocycle molecules are difficult to pene-
trate the structure of PWA. When comparing the stretching mode
of W
O W intra-bridges between corner-sharing WO6, there was
−
1
no peak shift observed on (pyrazole)3/PWA, 879.5 cm , but slight
−
1
shifts to 883.4 cm were detected on both (imidazole)3/PWA and
(triazole)3/PWA. In the case of the stretching mode of W Oterminal,
all modified PWA catalysts showed significant shift toward higher
frequencies, suggesting that there is a strong electronic interaction
between terminal W O cluster in PWA and heterocycles. From the
spectra results, it can be speculated that heterocycle is predom-
inantly anchored onto PWA by donation of lone-pair electron of
nitrogen, i.e. the conjugation of ␲ electrons of heterocycles into
the inorganic framework of PWA. It is also previously reported that
for organic compound modified hybrid HPA catalysts, there exists
a strong electronic interaction between the metal oxygen cluster
and the organic segment. The similar results with pyridine-PMoO
were previously reported in elsewhere [15]. It was suggested that
nitrogen-containing heterocycle molecules such as pyridine are
mainly placed in the interstices between polyanions and broaden
the interstitial spacing. Therefore, our results are supportive on
that the organic ␲ electrons may extend their conjugation to the
inorganic framework and thus dramatically modify the acidity and
For further investigation of electronic interaction of the com-
plex catalysts, FT-IR spectra of PWA and (heterocycle) /PWA were
3
adopted and depicted in Fig. 3. It was identified that all catalysts
exhibited three IR vibration peaks assigned to Keggin-structured
heteropolyacid; the stretching mode of W
O Wedge, W O W
intra-bridges between corner sharing WO6 octahedra, W Oterminal
−
1
and P O in PWA (780, 879.5, 956.7 and 1072.4 cm , respec-
tively [14]), which reveals that the Keggin structures of these
catalysts are well maintained after modification with heterocyclic
compounds. No shift was observed on P O bands over all com-
plex catalysts, suggesting that there is no interaction between
redox property of the cluster [16–18]. Hence, the shift in the W
O
band of PWA might be mainly attributed to the interaction of
the cationic heterocycle species with polyanions. Likewise, small
amount of imidazole and triazole molecules might be also anchored
onto W
cent nitrogen atoms as exo-bidentate ligands might be difficult to
interact with W W bridges by steric hindrance [19].
O W intra bridges. However, pyrazole having two adja-
O
3.2. Thermal stability of complex catalysts
In order to investigate thermal stability of the complex cat-
alysts, (imidazole) /PWA complex catalysts were heat-treated at
3
high temperature. As depicted in Fig. 4, (imidazole) /PWA treated
3
◦
up to 400 C still shows the same XRD pattern of as-prepared
one treated at 100 C. However, the complex catalyst treated at
00 C shows similar to that of the raw PWA rather than others.
◦
◦
5
Fig. 5 presents the comparison between TGA results of raw PWA
and (imidazole) /PWA. The raw PWA exhibits a typical TGA curve
3
with four steps of weight loss: elimination of (1) H O molecule
2
◦
physisorbed on the surface in 25–145 C, (2) 6 mol of H O at six-
hydrated PWA (H3PW12O40·6H2O) in 145–230 C, (3) removal of
2
◦
Fig. 3. FT-IR spectra of raw PWA and PWAs modified with various heterocycles.

H.J. Kim et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 232–237
235
Fig. 5. TGA data of raw PWA and (imidazole)3/PWA.
1
2
.5 mol of structural H O from anhydrous PWA (H PW12O40) in
2 3
◦
30–500 C and (4) decomposition of PW12O38.5 into WO3 and
◦
P O5 over 500 C. The calculated value of removed H O molecules
2
2
from TGA result is comparable to theoretical value, showing 5.99
and 1.04 (6.0 and 1.5, theoretically). However, (imidazole) /PWA
3
◦
complex catalyst started to decompose around 470 C without
any previous decomposition step. Hence, combining TGA results
in Fig. 5 with XRD results in Fig. 4, it can be deduced that
◦
(
imidazole) /PWA was thermally stable up to 400 C, maintaining
3
almost the same crystal structure as the prepared one, but over
◦
5
00 C, imidazole anchored on PWA was decomposed and thus the
crystal structure of (imidazole) /PWA was destroyed.
3
Fig. 7. Conversion of propylene and selectivity to PO and propanol over
(
imidazole)3/PWA by varying (above) temperatures and (below) propylene
3.3. Catalytic activity
pressures.
Propylene epoxidation over the complex catalysts were per-
formed in various conditions and the conversions and the
selectivities were listed in Table 1. As shown in Fig. 6 plotted from
the data in Table 1, the conversion of propylene as well as the
selectivity to propylene oxide increased as the imidazole contents
in complex catalysts increased, which is an obvious promotion
effect of heterocycles on the modified PWA complex catalysts. The
conversion of propylene and the selectivity to propylene oxide
unlike the other by-product showed almost constant yield. It is
also noteworthy that although PWA itself was not selective toward
propylene oxide in the reaction, the modified PWA with imidaz-
ole results in enhancement of catalytic activity and the selectivity
to propylene oxide. This result is quite supportive to Mizuno’s
research that complexation of PWA with appropriate cation forms
micro-structures, which lead to the regioselectivity [2].
over (imidazole) /PWA are 97.0% and 60.4%, respectively, improv-
Dependencies of the propylene conversion and the selectivity
to propylene oxide on pressure and temperature are depicted in
Fig. 7. The conversion of propylene increased significantly from
97.0% to 99.9% as the reaction temperature increased from 313 to
353 K. At 313 K, propylene oxide was the main product with 60.4%
yield. However, as the reaction temperature increased, the selec-
tivity to propylene oxide decreased, and propanol, the by-product
of the reaction and water produced by decomposition of hydrogen
peroxide were dominant. Even though the reaction is endother-
mic, and thus prefers low reaction temperature, the main reason of
deactivation at elevated temperature should be the produced water
molecules, which might be strongly interacted with the catalytic
active sites to inhibit epoxidation [21].
3
ing (over 50%) the selectivity to propylene oxide compared with
that over the raw PWA. It is interesting that propanol yield, the
major by-product probably due to anti-Markovnikov addition by
the peroxide effect [20], decreased as imidazole contents increased,
The graph also exhibits dependence of propylene pressure
charged in the reactor on propylene epoxidation. The conversion
dramatically increased as the pressure of propylene increased from
0
.1 to 0.5 MPa, but the differences of conversion and selectivity
between 0.5 and 1.0 MPa were negligible. At 0.1 MPa, the selectivity
to propylene oxide was 29.7% and the main product was propanol,
showing 49.1% of selectivity. However, at 0.5 MPa, both conver-
sion and selectivity to propylene oxide are almost twice compared
with those at 0.1 MPa and propanol production drastically dropped
down to half of that at 0.1 MPa. It can be deduced that epoxida-
tion of propylene is governed by the pressure of propylene, which
Fig. 6. Conversion of propylene and yield of PO over imidazole-modified PWA with
different imidazole contents.

2
36
H.J. Kim et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 232–237
Table 1
Reaction condition, conversion and selectivity of propylene epoxidation over heterocycle-modified PWA.
Catalyst
PWA
T (K)
P (MPa)
Conv. (%)
PO (%)
Acetone (%)
IPA (%)
Propanol (%)
C3 C6 a (%)
CO2 (%)
313
313
313
313
333
353
313
313
313
313
0.5
0.5
0.5
0.5
0.5
0.5
0.1
1.0
0.5
0.5
92.1
92.1
95.0
97.0
99.7
99.9
51.7
99.2
96.1
98.1
40.1
42.1
56.2
60.4
50.1
25.3
29.7
57.2
55.2
59.9
10.5
10.5
7.7
11.5
12.2
14.4
15.5
12.1
12.7
12.0
7.4
5.2
3.1
2.2
3.4
3.9
3.7
1.9
4.9
5.2
32.2
35.1
27.2
20.5
29.3
49.6
49.1
24.8
22.9
21.8
3.5
4.0
1.0
4.0
3.1
1.1
2.0
1.5
2.6
2.7
1.9
3.1
0.3
1.4
2.9
5.7
0
2.5
2.7
2.0
(Imidazole)1/PWA
(Imidazole)2/PWA
(Imidazole)3/PWA
(Imidazole)3/PWA
(Imidazole)3/PWA
(Imidazole)3/PWA
(Imidazole)3/PWA
(Pyrazole)3/PWA
(Triazole)3/PWA
a
Traced amounts of by-products: C3 (acrolein, propionaldehyde, propylene glycol), C4 (methoxy 2-propanol), C5 (2-pentanol) and C6 (2-hexanol).
is mostly likely that the amount of propylene dissolved in the
methanol medium depends on the pressure of propylene charged
in the reactor. Therefore, both conversion and selectivity depends
on pressure in the range below 0.5 MPa. Meanwhile, at higher pres-
sure above 0.5 MPa, propylene dissolved in the medium might be
fully saturated and the reaction is not dependent on the pressure
of propylene.
Fig. 8 shows the conversion of propylene and the selectivity of
main products, propylene oxide over the complex catalysts modi-
fied with various heterocycles. All complex catalysts showed higher
propylene conversion and selectivity than the untreated PWA.
The selectivity toward propylene oxide increased in the order of
(
imidazole) /PWA ≈ (triazole) /PWA > (pyrazole) /PWA > untreated
3
3
3
PWA. It was previously reported that the lattice spacing between
the polyanions in the HPA crystal can be expanded by interaction of
the heterocycle molecules with both surface and the bulk acid sites,
which makes the pseudo-liquid phase behavior of heteropolyacid
more obvious [15]. As seen from these results together with the
discussion of XRD and FT-IR results, it is inferred that the location
of nitrogen atoms of heterocycles also affect the lattice spacing of
the complex catalyst, thereby influence the catalytic reaction.
The catalyst recycle test was performed to investigate whether
the complex catalyst is thermally and chemically durable dur-
Fig. 9. Conversion and selectivity on spent catalyst recycling.
and selectivity to propylene oxide at every run were almost con-
sistent compared with those at the first run, suggesting that
(
imidazole) /PWA catalyst maintains its catalytic activity for con-
3
secutive runs.
Fig. 10 shows the comparison of FT-IR spectra of
ing the consecutive runs. The complex catalyst, (imidazole) /PWA,
3
(
imidazole) /PWA between before reaction and after 5 runs.
3
was vacuum-distilled and reused 5 times. It is interesting to note
that the most amount of complex catalyst was collected without
significant loss after drying under vacuum. This result suggests
that imidazole-modified complex catalyst is not soluble in water
and methanol medium, which means that the hydrophilic and
water soluble nature of PWA was changed to hydrophobic by
imidazole-modification. Fig. 9 shows the conversion and selectiv-
It is also notable that both spectra of catalysts before and after
reaction were almost identical. This is also well agreed with the
reaction results, suggesting that the modified catalyst is thermally
and chemically stable, thus is recyclable and compensates the
defect of the raw PWA catalyst.
ity over (imidazole) /PWA for five runs. All results of conversion
3
Fig. 8. Conversion of propylene and selectivity to PO over various heterocycle-
doped PWA.
Fig. 10. FT-IR spectra of (imidazole)3/PWA catalyst before and after reaction.

H.J. Kim et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 232–237
237
4
. Conclusions
References
[
[
[
1] K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi, S. Hikichi, N. Mizuno, Science
00 (2003) 964–966.
2] N. Mizuno, K. Yamaguchi, K. Kamata, Coord. Chem. Rev. 249 (2005) 1944–
1956.
3] N. Mizuno, K. Yamaguchi, K. Kamata, Y. Nakagawa, in: S.T. Oyama (Ed.), Mech-
anisms in Homogeneous and Heterogeneous Epoxidation Catalysis, Elsevier,
Amsterdam, 2008, pp. 155–176.
In this study, heterocycle-modified PWA complex catalysts were
3
adopted as the hybrid catalysts for epoxidation of propylene with
hydrogen peroxide as oxidant. The morphology of the modified
complex catalysts was significantly influenced by the kinds of het-
erocycles, the number of nitrogen on heterocycles and the ratio of
heterocycle to PWA. The higher regioselectivity as well as catalytic
activity was observed on epoxidation over all modified catalysts.
Especially, at the optimized reaction pressure and temperature,
imidazole and 1,2,4-triazole modified PWA catalyst exhibited the
highest selectivity to propylene oxide with 60.4% and conver-
sion with 98.1%, respectively. Modification of heterocycles also
[4] R. Stobaugh, V. Calarco, R. Morris, L. Stroud, Hydrocarbon Process. 52 (1973)
9–108.
[
[
9
5] B.S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457–2474.
6] T. Nijhuis, M. Makkee, J. Moulijn, Ind. Eng. Chem. Res. 45 (2006) 3447–
3459.
[7] J.-M. Bregeault, Dalton Trans. (2003) 3289–3302.
[
8] D.C. Duncan, R.C. Chambers, E. Hecht, C.L. Hill, J. Am. Chem. Soc. 117 (1995)
81–691.
6
◦
increased thermal stability of the catalysts up to 450 C, mainly due
[
9] H. Benaissa, P.N. Davey, Y.Z. Khimyak, I.V. Kozhevnikov, J. Catal. 253 (2008)
244–252.
to strong electronic interaction between heterocycle and terminal
[
[
[
10] I.V. Kozhevnikov, J. Mol. Catal. A. Chem. 262 (2007) 86–92.
11] Y. Leng, H. Ge, C. Zhou, J. Wang, Chem. Eng. J. 145 (2008) 335–339.
12] L.R. Pizzio, M.N. Blanco, Appl. Catal. A: Gen. 255 (2003) 265–277.
[13] K. Potts, Chem. Rev. 61 (1961) 87–127.
14] C. Paze, S. Bordiga, A. Zecchina, Langmuir 16 (2000) 8139–8144.
W
O bonds of PWA. The modified PWA catalysts can be recycled,
preventing significant loss of catalyst from polar solvent produced
in epoxidation. Therefore, the heterocycle-modified PWA complex
catalysts have the potential as sustainable heterogeneous catalysts
for epoxidation of propylene, overcoming the inherent problems
of PWA: low regioselectivity, low thermal stability and difficulty of
regeneration.
[
[
15] I.K. Song, M.S. Kaba, M.A. Barteau, J. Phys. Chem. 100 (1996) 17528–
1
7534.
[16] J. Kang, B. Xu, Z. Peng, X. Zhu, Y. Wei, D.R. Powell, Angew. Chem. Int. 44 (2005)
902–6905.
17] M. Lu, Y. Wei, B. Xu, C.F.-C. Cheung, Z. Peng, D.R. Powell, Angew. Chem. Int. 41
2002) 1566–1568.
[18] J. Wang, L. Yan, G. Qian, G. Lv, G. Li, J. Suo, X. Wang, React. Kinet. Catal. Lett. 91
1) (2007) 111–118.
6
[
(
Acknowledgements
(
[
[
19] S. Trofimenko, Chem. Rev. 72 (1972) 497–509.
20] F. Minisci, Acc. Chem. Res. 8 (1975) 165–171.
We are very thankful to the previous members of our laboratory,
Dong Ho Seo, Jin Sol Kim, Sang-Sun Park and Sung Won Choi for
their helpful discussion.
[21] J. Kaur, I.V. Kozhevnikov, Catal. Commun. 5 (2004) 709–713.