Styrene epoxidation over carbon nanotube-supported gold catalysts

Article abstract of DOI:10.1007/s10562-009-0230-6

Carbon nanotube supported gold catalysts prepared by deposition- precipitation with urea (DP urea) were characterized by various techniques. Its catalytic activity was examined for the oxidation of styrene using t-butylhydroperoxide as oxidant. This syste

Full text of DOI:10.1007/s10562-009-0230-6

Catal Lett (2010) 134:51–55
DOI 10.1007/s10562-009-0230-6
Styrene Epoxidation over Carbon Nanotube-Supported Gold
Catalysts
Junhua Liu
Zhenggui Gu
•
Fang Wang
•
Tan Xu
•
Received: 23 October 2009 / Accepted: 16 November 2009 / Published online: 25 November 2009
Ó Springer Science+Business Media, LLC 2009
Abstract Carbon nanotube supported gold catalysts pre-
pared by deposition–precipitation with urea (DP urea) were
characterized by various techniques. Its catalytic activity
was examined for the oxidation of styrene using t-butyl-
hydroperoxide as oxidant. This system showed good
epoxide selectivity. The other factors, such as solvent,
reaction time, concentrations of oxidant and catalyst, have
also been investigated and reaction conditions are opti-
mized. It is a novel highly active/selective and reusable
heterogeneous catalyst for styrene epoxidation.
Consequently, there is a growing interest in the synthesis of
fine chemicals via this versatile reaction.
The gold-based catalysts have demonstrated a very
interesting and promising activity, and different types of
gold-based homogeneous and heterogeneous catalysts in
the form of metal complex or nano-particles have been
developed for the oxidation of olefins. Choudhary et al.
[2–4] and subsequently Yin et al. [5] have shown that gold
nanoparticles (Au–NPs) supported on a range of oxides
are active for the epoxidation of styrene using tertiary
butylhydroperoxide (TBHP) in greater than stoichiome-
tric amounts. Selectivities to styrene oxide of ca. 50%
are readily achieved using this approach.
Keywords Carbon nanotubes (CNTs) Á Epoxidation Á
Gold Á Styrene
Generally, the catalytic properties of heterogeneous gold
catalysts strongly depend on the particle size. The use of
gold as a catalyst requires careful preparation of the gold
particles focusing on achieving a very small particle size.
Recent trends in this area have fallen into two broad cat-
egories: (1) modulation of nano-size gold particles pre-
pared by different preparation methods [6, 7]; (2)
application of new nano material as carrier to control the
nano-size of gold particles. Much attention was paid to the
latter. Some new catalytic materials, such as meso–micro
sieves [8, 9], nanotubes [10] and polymers [11], have been
used to prepare gold-based catalysts. However, carbon
nanotubes (CNTs) were extensively studied due to their
unique chemical and mechanical properties for different
potential applications. CNTs are resistant to acid/base
media and precious metals can easily be recovered by
burning off the support. The material also provides high
dispersion of the metals. Herein, we present a strategy for
the design of high-performance heterogeneous catalysts
utilizing the carbon nanotube supported Au–NPs by
deposition–precipitation (DP) method, which exhibits high
styrene conversion and epoxide selectivity.
1
Introduction
Olefin epoxidation, especially styrene, is a subject of great
interest from both academic and industrial points of view;
the oxirane group is highly reactive which makes epoxides
an essential group of organic intermediates [1].
J. Liu (&) Á T. Xu Á Z. Gu
College of Chemistry and Environmental Science,
Nanjing Normal University, 210097 Nanjing, China
e-mail: liujh2010@yahoo.cn
F. Wang (&)
College of Sciences, Nanjing University of Technology,
210009 Nanjing, China
e-mail: qufuwangfang@yahoo.com.cn
123

5
2
J. Liu et al.
2
Experimental
(3 ml), the reaction was started by immersing the flask in
an oil bath kept at the reaction temperature (82 °C). The
resulting mixture was then refluxed for 10 h. After filtra-
tion and extraction, the filtrate was concentrated by rotary
evaporator then analyzed by gas chromatography, the
nature of the products was also determined by GC–MS.
2
.1 Materials
CNTs were purchased from Chengdu Times Nano Com-
pany, TBHP(70% mass concentration in water) and H O2
2
(
50% mass concentration in water) were purchased from
Beijing Chemical Reagents Company. Other chemicals and
reagents used in the present study were of analytical grade
and purchased from Sinopharm Chemical Reagent Com-
pany and used without any further purification.
3 Results and Discussion
3.1 Catalyst Characterization
2
.2 Catalyst Preparation
SEM micrographs show the morphologies of CNTs and
Au/CNTs catalyst (Fig. 1a, b). The diameter of CNTs is
distributed between 20 and 30 nm. A TEM image of CNTs
with 0.3 wt% deposited Au–NPs is shown in Fig. 1c. One
bundle of nanotube is covered with ca. 5 nm Au–NPs with
a very high particle density. The progressive decomposi-
tion of urea in solution at temperatures above 60 °C
Au/CNTs catalysts were prepared by DP with urea, in the
so-called standard preparation procedures: 1 g CNTs was
added into 25 ml of an aqueous HAuCl solution
4
-
4.2 9 10 M) and 0.63 g urea (0.42 M). The initial pH
3
(
was about 2. The suspension thermostated at 80 °C, was
vigorously stirred for 8 h (pH increases) and then centri-
fuged, washed and dried (Au content: 0.3 wt%).
-
releases OH ions, which gradually increase the medium
pH. This method facilitates the slow precipitation of
hydroxides onto the support, and avoids a brutal and local
increase of pH, which could induce precipitation in solu-
tion. The small size of the gold particles for the Au/CNTs
sample prepared by DP urea implies that an interaction
occurs between the carbon nanotube and the gold precipi-
tate, which leads to a high dispersion of Au–NPs. Thus,
their channels will not be blocked and the Au–NPs may not
sinter under reaction conditions, which can provide an
excellent catalytic activity for styrene epoxidation.
2
.3 Catalyst Characterization
The morphologies of obtained carbon nanotube and Au/CNTs
catalyst were examined by transmission electron microscope
(
(
TEM) (JEM-1200EX) and scanning electron microscope
SEM) (JSM-5600 LV). The crystals of the catalysts were
’
studied by X-ray diffraction (XRD) (X Pert PRO). Surface
composition was determined by X-ray photoelectron spec-
troscopy (XPS), using an ESCALAB 210 XPS system with a
Mg Ka source. BET surface areas were obtained by physi-
sorption of nitrogen at 77 K using a Micromeritics
ASAP2010 instrument. Elemental analysis was determined
by the Panalytical Company with a Magix PW 2403 XRF
Spectrometer.
The CNTs and Au/CNTs catalyst were also character-
ized by XRD (Fig. 2). It could be observed that the typical
peak located at a 2h value of about 26.0° was referred to
(0 0 2) reflection of CNTs supports, and the characteristic
peaks assigned to CNTs become weaker after loading Au.
The other featured four peaks were indexed to the (1 1 1),
(2 0 0), (2 2 0) and (3 1 1) reflection of face centered
2
.4 Styrene Epoxidation
cubic (fcc) crystalline Au (JCPDS, Card no. 7440-57-5),
which indicated the formation of gold crystal loaded on
CNTs.
In a typical run, a mixture of catalyst (0.05 g), substrate
4 mmol) and acetonitrile (5 ml) was stirred in a 25 ml
(
To obtain information on the surface state of gold in
Au/CNTs catalysts, XPS study was carried out, as can be
seen from Fig. 3. Two distinct peaks appeared at 84.6 and
round-bottom flask equipped with a condenser at room
temperature for 30 min. After the addition of oxidant
Fig. 1 SEM images of a CNTs
b Au/CNTs catalyst and c TEM
image of Au/CNTs catalyst
1
23

Styrene Epoxidation over Carbon Nanotube-Supported Gold Catalysts
53
in uncalcined samples than in calcined ones and proposed
that oxidized gold species are the most active sites for low-
temperature CO oxidation for Au/Fe O catalysts. In our
(
111)
∗
∗
Au-NPs
CNTs
2
3
∆
(
200)
∗
case, the Au/CNTs catalysts were obtained just by desic-
cation, without any high-temperature process, so there
should be no the formation of metallic gold particles.
Combining the XPS data, we concluded that the ionic
Au(III) formed the true active sites, in agreement with the
hypothesis proposed by Visco.
(
002)
∆
(220)
∗
(311)
∗
(
b)
∆
(
002)
(
a)
3.2 Catalytic Test
The oxidations of styrene to styrene oxide (SO) on a
series of Au/CNTs catalysts were investigated. Phenyl-
acetaldehyde (PA) and benzaldehyde (BA) have been the
main by-products during the oxidation of styrene on these
catalysts.
1
0
20
30
40
50
60
70
80
2
θ
Fig. 2 XRD patterns of a CNTs; b Au/CNTs
The effect of solvents on styrene epoxidation was
examined and the results are shown in Table 1. The
Au/CNTs catalyst exhibited the highest catalytic activ-
ity and selectivity for styrene oxide in acetonitrile. 1,
Au 4f 7/2
Au 4f 5/2
2-dichloroethane also was found to be a favorable solvent
for this reaction (Table 1, entry 2), but its styrene con-
version and epoxide selectivity were lower than that in
acetonitrile. The reaction, however, showed low activity in
toluene or N,N-dimethylacetamide (DMA) despite of their
moderate epoxide selectivity (Table 1, entries 3–4). The
results showed that the conversion and epoxide selectivity
was relative to the polarity of solvents: the epoxidation
yields increased gradually with increasing polarity of sol-
vents except DMA (the polarity of solvent was determined
by dielectric constant e: acetonitrile (e: 37.5) [ dichloro-
ethane (e: 10.4) [ DMA (e: 6.4) [ toluene (e: 2.4)). Dif-
ferent oxidants, such as air and hydrogen peroxide have
also been used for styrene epoxidation over this catalyst
9
6
94
92
90
88
86
84
82
80
78
76
Binding Energy (eV)
Fig. 3 XPS of Au 4f region of Au/CNTs catalyst
(Table 1, entries 5–6), although very poor yields of styrene
oxide were obtained.
8
8.3 eV, which were close to the Au 4f binding energies of
The effect of the amount of catalyst on styrene epoxi-
dation was thereafter examined (Table 2). When the sub-
strate/catalyst ratio was between 80:1 and 120:1, the
reaction rate decreased considerably with the reduction in
catalyst concentration; however, increasing the substrate/
catalyst ratio from 20:1 to 80:1 had little effect on the
reaction rate of this reaction. According to this study,
whatever the substrate/catalyst ratio, no significant effect
was found with respect to the epoxide selectivity, in other
words, the epoxide selectivity of this oxidation reaction is
independent of the catalyst concentration.
oxidized Au(III) [12]. A lot of efforts have been spent to
study the nature of the ‘‘active sites’’ in the oxidative
reactions over gold-catalysts. Some authors suggest that the
activity of gold supported catalysts depends on the size of
the gold particles with a maximum of activity usually
observed for a particle size of 3 nm; some argue that the
effect of the oxidation state of gold is at least as important
as that of the particle size of metallic gold. However, the
state of the gold in active catalysts is unclear, most of them
have accepted that metallic gold generated by high calci-
nation temperature favors a high catalytic activity [13].
Hoflund and co-workers claimed that part of the gold
supported on a-Fe O and Co O is nonmetallic, and that it
The change of styrene epoxidation was monitored and
plotted with respect to the time. The results are shown in
Fig. 4. It was seen that the conversion of styrene increased
continuously when the reaction time increased and was close
to complete conversion ultimately after 22 h. The epoxide
2
3
3 4
may be responsible for the low-temperature activity [14],
while Visco et al. [15] found much larger catalytic activity
123

5
4
J. Liu et al.
Table 1 Oxidations of styrene on Au/CNTs catalyst using different solvents/oxidants
b
a
b
Entry
Solvent
Time (h)
Conversion (%)
Selectivity (%)
TON
SO
BZ
PA
9.4
1
2
3
4
Acetonitrile
10
10
10
10
10
10
67.2
48.5
45.0
6.9
77.5
71.1
68.1
61.4
74.7
–
13.1
2733
1811
1609
223
90
Dichloroethane
Toluene
15.5
20.9
38.6
10.0
13.4
11.0
–
N,N-dimethylacetamide
Acetonitrile
c
5
2.3
15.3
–
d
6
Acetonitrile
10.2
100
0
Reaction conditions Au/CNTs catalyst 0.05 g, styrene 4 mmol, solvent 5 mL, TBHP 3 mL, 82 °C
a
Determined by GC
b
TON = moles of styrene oxides/moles of gold in the catalyst
c
Air bubbling, without TBHP
d
2 2
H O as oxidant, without TBHP
Table 2 Styrene epoxidation at different concentrations of Au/CNTs catalyst
a
b
b
Entry
Substrate/Au–CNTs (mmol/g)
Time (h)
Conversion (%)
Selectivity (%)
TON
SO
BZ
PA
1
2
3
4
5
20/1
10
10
10
10
10
73.1
71.8
67.2
51.6
42.7
74.5
75.6
77.5
76.1
74.6
12.8
13.6
13.1
11.8
13.2
12.7
10.8
9.4
715
1781
2733
2578
2509
50/1
80/1
100/1
120/1
12.1
12.2
Reaction conditions Au/CNTs catalyst, CH
a
Determined by GC
3
CN 5 mL, TBHP 3 mL, 82 °C
b
TON = moles of styrene oxides/moles of gold in the catalyst
higher catalytic activity and epoxide selectivity over Au/
CNTs catalyst are obtained among these gold-contained
catalysts to the best of our knowledge. Furthermore, when
the reaction time was 22 h, the recycled catalyst could be
used for epoxidation of styrene without changes in conver-
sion and selectivity values, producing a 67.2% yield for the
second use and a 66.7% yield for the third use. It was con-
firmed by induced coupled plasma techniques (ICP) that the
gold content of recycled Au/CNTs catalyst was almost the
same as that of the fresh ones.
The effect of oxidant concentrations on styrene oxida-
tion was plotted in Fig. 5. An increase of oxidant con-
centration enhanced the rate of styrene oxidation, and
epoxide selectivity was increased firstly then decreased
with increased concentration of oxidant. The conversion of
styrene was 55.4% and the epoxide selectivity was 59.8%
at 2 ml oxidant, then a maximum styrene conversion
Fig. 4 Styrene epoxidation on Au/CNTs catalyst at different time.
Reaction conditions: Au/CNTs catalyst 0.05 g, styrene 4 mmol,
CH
3
CN 5 mL, TBHP 3 mL, 82 °C
(78.0%) and high epoxide selectivity (77.3%) were
obtained at 3 ml. The catalytic activity was found to
increase slightly and the epoxide selectivity decreased
gradually when the oxidant exceeded 3 ml, so an optimal
concentration of oxidant should be 3 ml.
selectivity was nearly steady at all times (72.1–77.5%).
Although there have been several reports on the develop-
ment of heterogeneous gold catalysts for this reaction, the
1
23

Styrene Epoxidation over Carbon Nanotube-Supported Gold Catalysts
55
species 3 interacts with styrene to form a p-bonded tran-
sient species 4. The transfer of oxygen to olefinic bond in a
rate-determining step forms a metalloepoxy intermediate
species 5, which dissociate to form products, while
regenerating the active species.
4
Conclusions
Au/CNTs catalyst is active in catalyzing the oxidation of
styrene with relative good conversion and high epoxide
selectivity using TBHP as oxidant. This study confirms that
the key factors governing the activity and selectivity of
Au/CNTs are highly dispersed Au(III) species. Further
work is in progress for the oxidation of other olefins, and
details of the experiments will be given in a forthcoming
paper.
Fig. 5 Styrene epoxidation on Au/CNTs catalyst with different
concentrations of oxidant. Reaction conditions: Au/CNTs catalyst
0
.05 g, styrene 4 mmol, CH
3
CN 5 mL, 82 °C, 14 h
Acknowledgments This work was supported by the Natural Sci-
ence Foundation of Jiangsu province for corporational doctor project
(
2
No. BK2009472), and the Leading Academic Discipline Program,
11 Project for Nanjing Normal University (the 3rd phase).
References
1
. Ullmann’s Encyclopedia of Industrial Chemistry (1998) 6th edn
Electronic Release). Wiley/VCH, New York
(
2
3
. Choudhary VR, Durnbre DK (2009) Catal Commun 10:1738
. Patil NS, Uphade BS, Jana P, Bharagava SK, Choudhary VR
Scheme 1
(
2004) J Catal 223:236
3
.3 Mechanism Investigation
4. Patil NS, Uphade BS, Jana P, Sonawane RS, Bharagava SK,
Choudhary VR (2004) Catal Lett 94:89
5
6
. Yin D, Qin L, Liu J, Jin CLY (2005) J Mol Catal A Chem 240:40
. Eustis S, Hsu H-Y, El-Sayed MA (2005) J Phys Chem B
Mechanistic issues most often addressed have been the
roles of oxygen and of promoters in olefin epoxidation
reactions. It is typically reported that atomic, rather than
molecular, adsorbed oxygen reacts with ethylene [16],
styrene [17], and norbornene [18] in the rate-determining
step. Moreover, in our experiments, the data showed that
the key factors governing the activity and selectivity of
catalyst are highly dispersed Au(III) species. Based on the
above description, a possible mechanism was proposed
1
7. Liu Y, Lin L, Chiu W (2004) J Phys Chem B 108:19237
09:4811
8
. K o´ nya Z, Puntes VF, Kiricsi I, Zhu J, Ager JW, Ko MK, Frei H,
Alivisatos P, Somorjai GA (2003) Chem Mater 15:1242
. Hu J, Chen L, Zhu K, Suchopar A, Richards R (2007) Catal
Today 122:277
9
10. Bavykin DV, Lapkin AA, Plucinski PK, Murciano LT, Friedrich
JM, Walsh FC (2006) Top Catal 39:151
1. Tsunoyama H, Sakurai H, Tsukuda T (2006) Chem Phys Lett
29:528
2. Cattaruzza E, Battaglin G, Canton P, Sada C (2005) J Non-Cryst
Solids 351:1932
3. Zanella R, Giorgio S, Shin CH, Henry CR, Louis C (2004) J Catal
1
1
1
1
4
(
Scheme 1). Initial complex formation between metal cat-
alyst and TBHP renders the peroxidic oxygen more elec-
trophilic and hence more labile to attack by an olefinic
double bond.
2
22:357
4. Epling WE, Hoflund GB, Weaver J, Tsubota S, Haruta M (1996)
J Phys Chem 100:9929
15. Visco AM, Neri F, Neri G, Milone C, Galvagno S (1999) Phys
Chem Chem Phys 1:2869
6. Grant RB, Lambert RM (1985) J Catal 92:364
7. Hawker S, Mukoid C, Badyal JPS, Lambert RM (1989) Surf Sci
219:L615
18. Roberts JT, Madix RJ (1988) J Am Chem Soc 110:8540
The free radical mechanism involves the interaction of
TBHP with the Au(III) species 1 in catalyst to form a
hydroperoxy species 2 in a preequilibrium step. Species 2,
being unstable, undergoes one-electron reduction with a
loss of water molecule and rearranges to give the Au(III)–
O ad (peroxo) species 3. In another preequilibrium step,
1
1
123