Selective oxidation of propylene to propylene oxide over silver-supported tungsten oxide nanostructure with molecular oxygen

Article abstract of DOI:10.1021/cs5004454

Propylene oxide (PO) is a versatile chemical intermediate, and by volume it is among the top 50 chemicals produced in the world. The catalytic conversion of propylene to PO by molecular oxygen with minimum waste production is of high significance from an academic as well as an industrial point of view. We have developed a new synthesis strategy to prepare 2-5 nm metallic silver nanoparticles (AgNPs) supported on tungsten oxide (WO₃) nanorods with diameters between 30 and 40 nm, in the presence of cationic surfactant (cetyltrimethylammonium bromide: CTAB), capping agent (polyvinylpyrrolidone: PVP), and hydrazine. The synergy between the surface AgNPs and WO₃ nanorods facilitates the dissociation of molecular oxygen on the metallic Ag surface to produce silver oxide, which then transfers its oxygen to the propylene to form PO selectively. The catalyst exhibits a PO production rate of 6.1 × 10^-2 mol g_cat^-1h^-1, which is almost comparable with the industrial ethylene-to-ethylene oxide production rate.

Full text of DOI:10.1021/cs5004454

Letter
pubs.acs.org/acscatalysis
Selective Oxidation of Propylene to Propylene Oxide over Silver-
Supported Tungsten Oxide Nanostructure with Molecular Oxygen
Shilpi Ghosh, Shankha S. Acharyya, Ritesh Tiwari, Bipul Sarkar, Rajib K. Singha, Chandrashekar Pendem,
Takehiko Sasaki,† and Rajaram Bal*
Catalytic Conversion & Processes Division, CSIR-Indian Institute of Petroleum, Dehradun-248005, India
S
* Supporting Information
ABSTRACT: Propylene oxide (PO) is a versatile chemical intermediate, and by
volume it is among the top 50 chemicals produced in the world. The catalytic
conversion of propylene to PO by molecular oxygen with minimum waste
production is of high significance from an academic as well as an industrial point
of view. We have developed a new synthesis strategy to prepare 2−5 nm metallic
silver nanoparticles (AgNPs) supported on tungsten oxide (WO3) nanorods with
diameters between 30 and 40 nm, in the presence of cationic surfactant
(cetyltrimethylammonium bromide: CTAB), capping agent (polyvinylpyrroli-
done: PVP), and hydrazine. The synergy between the surface AgNPs and WO3
nanorods facilitates the dissociation of molecular oxygen on the metallic Ag
surface to produce silver oxide, which then transfers its oxygen to the propylene
to form PO selectively. The catalyst exhibits a PO production rate of 6.1 × 10−2
mol gcat−1 h−1, which is almost comparable with the industrial ethylene-to-
ethylene oxide production rate.
KEYWORDS: selective oxidation, propylene oxidation, propylene oxide, molecular oxygen, silver nanoparticles, tungsten oxide nanorod
ropylene oxide (PO) is one of the most important
produce propylene oxide directly from propylene using solely
Psynthetic intermediates produced in the industry to make environmentally benign molecular oxygen.
It has been reported in the literature that supported Ag
various commodity chemicals, such as polyeurethane foams,
propylene glycol, polypropylene glycol, propylene carbonate,
etc., and currently, its production exceeds 10 million tons per
annum.1 Ninety percent of the world production of PO is
produced either through a chlorohydrin process or by an
organic hydroperoxide process (Halcon method).2 Apart from
acute environmental problems, the chlorohydrin process is also
associated with equipment corrosion due to the usage of
significant amount of chlorine. In addition, the Halcon process
involves autoxidation of ethylbenzene or isobutene to produce
alkylhydroperoxide, which finally acts as an oxidant to produce
PO, accompanied by substantial formation of unwanted
byproduct, peroxycarboxylate.2 Thus, the selective oxidation
of propylene to PO, in an economically viable and more
environmentally friendly means, without producing minimal
waste, is of paramount importance.
The selective oxidation of propylene to PO is regarded as the
Holy Grail in catalysis.3 The main difficulty arising for PO
production is the presence of labile allylic H atom, which is
more susceptible to combustion through C−H cleavage.4
Haruta and co-workers reported selective oxidation of
propylene to PO mediated by 2−4 nm gold nanoparticles on
titania in the presence of O2 and H2.5 Although the selectivity
for PO is high, the use of H2 suffers from its high cost and low
efficiency, and the selectivity is greatly dependent on the size
and morphology of the gold nanoparticles.5,6 There is no
reported catalyst to date that can be used industrially to
catalyst activates molecular oxygen, and the activity of the
catalyst largely depends on the size of the Ag particles.7
Although supported silver catalyst has been employed
successfully for the epoxidation of ethylene to ethylene oxide
on a commercial scale, successful implementation of silver in
the propylene epoxidation reaction has not been reported so far
because of either low conversion or poor PO selectivity.8
Recently, Lei et al. reported that AgNPs supported on Al2O3
with a narrow particle size distribution is a prerequisite for high
catalytic activity of this reaction.9 Here, we have designed a new
synthesis strategy to prepare metallic AgNPs supported on
WO3 nanorod. In recent years, significant efforts have been
devoted to the controlled synthesis of nanostructured silver
particles, but most of the preparation methods are energy-
intensive, require substantial heat treatment, and produce larger
particles.10 Supported WO3 catalysts have been shown to be
catalytically active for selective oxidation of propylene to
propylene oxide in a H2O2 medium.11 Although there have
been several reports of preparation of AgNPs, to date, the
controlled synthesis of dispersed ultrasmall AgNPs (<5 nm)
supported on WO3 nanorods (Ag/WO3) has not been
reported.
Received: December 16, 2013
Revised: May 31, 2014
© XXXX American Chemical Society
2169
dx.doi.org/10.1021/cs5004454 | ACS Catal. 2014, 4, 2169−2174

ACS Catalysis
Letter
Herein, we report the preparation of 2−5 nm AgNPs
supported on WO3 nanorods with diameters of ∼30−40 nm for
selective conversion of propylene to propylene oxide using only
molecular O2. The Ag/WO3 catalyst exhibits ∼16% propylene
conversion with PO selectivity of 83% in a continuous flow
process, where a PO production rate of 6.1 × 10−2 mol gcat
−1
h−1 was achieved in a steady state condition.
The Ag/WO3 catalyst was prepared in aqueous medium in
the presence of surfactant cetyltrimethylammonium bromide
(CTAB), polyvinylpyrrolidone (PVP), and hydrazine, where
tungstic acid and AgNO3 were used as the W and Ag sources,
respectively. It is a well established phenomena that Ag, W, and
CTAB exist in the form of Ag+, WO42−, CTA+ in aqueous
alkaline solution (pH >7).12 Therefore, a kind of cooperative
self-assembly between the cationic part of the surfactant,
CTAB, and anionic species can be formed via electrostatic
interaction.12 When the CTAB concentration become high
(more than critical micelle concentration), cationic CTA+
2−
molecules start adsorbing onto the surface of the WO4 ions
and form spherical micelles.13
The curvature of an ionic micelle can be tuned from a
spherical to a rodlike micelle by adding certain additives.14 In
our preparation method, the spherical micelles of CTAB−
WO41− were transformed into rod-shaped micelles as a result of
attractive electrostatic interaction between the CTAB-stabilized
negatively charged tungstate ions and the positively charged
silver ions, where Ag+ ions trigger symmetry-breaking
anisotropic growth through selective adsorption onto particular
crystal planes of tungsten. Thus, the decrease in surface energy
of the WO3 seed crystals in one direction results in the
formation of a 1-D rodlike micellar structure, and this acts as
the nucleating agent for the growth of the WO3 nanorod. We
observed that the amount of silver ions present in the solution
has a pronounced influence on the nanorod growth.15 Ag
loading between 4 and 15 wt % leads to predominantly 1-D
growth, and above 15 wt % Ag loading, different morphologies
were observed (Figures 1, 2 and Supporting Information (SI)
Figure 2. TEM images of (a) Ag/WO3 catalyst (Ag loading 4.8 wt %),
(b) AgNPs size distribution, (c) high-resolution image, and (d) lattice
fringes.
30 and 40 nm can be prepared by mixing a particular amount of
Ag, W, CTAB, hydrazine, and PVP. The rod-like morphology
of WO3 helps to increase the dispersion of metallic Ag, which
ultimately facilitates easy access of oxygen to adsorb on it to
form Ag2O, which is the surface active oxygen species.17
Metallic AgNPs with size (2−5 nm) supported on WO3 with
diameter 30−40 nm can be imaged by SEM and TEM (Figures
1, 2). The lattice fringe with a d-spacing of 0.38 nm
corresponds to the [020] plane of WO3, and the distance of
0.20 nm corresponding to the [200] plane of Ag (Figure 2b)
can also be seen. It should be noted that the pure WO3 phase
was formed without forming any substoichiometric WO3−x
phase. The W 4f5/2 and 4f7/2 spectra attributed to the binding
energies 38.06 and 35.9 eV, respectively, suggest that the
tungsten in the tungsten oxide nanostructure exists as W6+.18
The absence of a binding energy peak at 34.8 eV confirms that
there is no substoichiometric WO3−x present in the sample (SI
Figure S4).18 The X-ray diffraction peaks at 2θ values of 23.2°,
23.5°, 24.3°, 33.2°, and 34.2° depict the formation of WO3
monoclinic structure (JCPDS No. 43-1035; space group, P21/
n), and the peaks at 38.1° (111), 44.3° (200), 64.4° (220), and
77.4° (311) indicate the presence of metallic silver in the
catalyst (JCPDS No. 04-0783) (Figure 3). The XPS spectrum
of the fresh catalyst showed a Ag 3d5/2 binding energy value of
368.6 eV and a Ag 3d7/2 binding energy value of 374.6 eV
(Figure 4). It is reported that the Ag 3d binding energy value
changes with the particle size.9,19 We can say that in our
prepared Ag/WO3 catalyst, because the Ag NPs are very small
in size (2−5 nm), they exhibit a higher Ag 3d 5/2 binding
energy value of 368.6 eV compared with the metallic Ag
powder (368.2 eV) generally used as reference. Hence, we can
say that the Ag 3d binding energy in our catalyst corresponds to
metallic silver.
Figure 1. SEM images of fresh Ag/WO3 catalyst (4.8 wt % Ag
loading): (a) lower magnification, (b) higher magnification.
Figures S1, S2). It was found that PVP changes the size of the
AgNPs rather than changing the size of the tungsten oxide
nanorods. It was also observed that in the absence of PVP, Ag
particles with nonuniform size (>5 nm) were formed (SI Figure
S3). We believe that PVP coordinates easily with the Ag+
surface with its carbonyl end and prevents aggregation of Ag
particles. Generally, hydrazine reduces Ag+ through formation
of a complex Ag (N2H4) +.16 Being a reducing agent, hydrazine
reduces Ag+ to Ag onto the surface of WO42−, thus discarding
the formation of AgBr, and we observed only metallic silver
deposition on the WO3 surface. Thus, well-defined 2−5 nm
AgNPs supported on WO3 nanorods with diameters between
The metallic state of the Ag was also confirmed by EXAFS, in
which the Ag−Ag bond length of 2.87 Å with a coordination
number (C.N.) of 6.3 was observed (Supporting Information,
Table S1, Figure S5a). The association of CTA+ and PVP with
2170
dx.doi.org/10.1021/cs5004454 | ACS Catal. 2014, 4, 2169−2174

ACS Catalysis
Letter
vacant π* molecular orbital of adsorbed O2.23 In a silver-
catalyzed system, high oxygen pressure would affect the
activation energy for dissociative chemisorption of O2 on the
silver surface and accelerate the conversion of the silver surface
to a silver oxide overlayer, as a result of which a loosely bonded
Ag−O species is formed.24 The XPS spectrum of our spent
catalyst showed a Ag 3d5/2 binding energy value of 368.2 eV, a
negative binding energy shift of 0.4 eV from the fresh one
(368.6 eV) (Figure 4b). It is known in the literature that the
binding energy of Ag+ is lower than the metallic Ag and that
the binding energy difference of the Ag 3d in a metallic and in
an oxidized state is rather small.9,25,26 As a continuation of the
earlier reports, we can say that the binding energy shift of 0.4
eV for the spent catalyst (368.2 eV) indicates that metallic
silver nanoparticles are partially oxidized after the reaction.26,27
The EXAFS analysis clearly indicates that after the reaction,
silver is partially oxidized (SI Table S1, Figure S5b). The O 1s
binding energy of the fresh catalyst exhibits a peak at 530.7 eV
due to the WO in WO3,18 and the O 1s spectrum of the
spent catalyst showed two peaks, the peak at 530.8 eV and the
additional small peak at 532.5 eV (Supporting Information
Figures S8, S9). The binding energy value at 532.5 eV may be
attributed to the oxygen species attached with Ag.27 Ag K-edge
extended X-ray absorption fine structure (EXAFS) analysis of
the catalyst revealed significant changes in the silver species
during the course of the propylene epoxidation reaction and
showed that a Ag−O bond with a bond length of 2.349 Å is
present in the spent catalyst (Supporting Information, Table
S1, Figure S5b). This observation indicates that metallic Ag is
being transformed to Ag2O during the reaction.
Figure 3. XRD patterns of (a) Ag(0), (b) Ag(I) oxide, (c) W(VI)
oxide, (d) fresh Ag/WO3, and (e) spent Ag/WO3.
We assume that the weakly associated Ag−O bond at 2.349
Å is the active oxygen species for propylene epoxidation
reaction. We did a pulse experiment to check whether lattice
oxygen is taking part during oxidation of propylene to
propylene oxide transformation. When the pulse of propylene
was used over the fresh Ag/WO3 catalyst, no propylene oxide
formation was detected. This clearly indicated that surface
oxygen does not take part during the reaction. We then
pretreated our catalyst with O2 and measured the XPS of the
catalyst and found Ag2O was formed (from the binding energy
shift of 0.3 eV, Supporting Information Figure S10). Then a
propylene pulse was used over O2 pretreated catalyst, and we
noticed the formation of propylene oxide, so we conclude that
during the reaction, metallic Ag was transformed to Ag2O,
which then converted propylene to propylene oxide. SI Figure
S10 showed the Ag 3d XPS spectra after the O2 treatment. XPS
spectra were taken for the fresh catalyst after the pretreatment
with O2 and it was found that there is a negative binding energy
shift (368.3 eV), as compared with the fresh catalyst (368.6
eV). In the fresh Ag/WO3 catalyst, the highly intense band at
804 cm−1 in the Raman spectrum corresponds to the symmetric
stretching vibration and the band at 708 cm−1 corresponds to
the asymmetric stretching vibration of O−W−O bond, which
coincides with bare WO3 (Figure 5a, Figure 5b).18 For the
spent catalyst, the Raman spectrum (Figure 5c) showed a
rather weak band around 524 cm−1, which is attributed to Ag2O
species, owing to the Raman forbidden nature of Ag2O, since
oxygen atoms in silver oxide are located at inversion sites.28
The intense peak at 936 cm−1 could be assigned to the γ (O
O) stretching mode of adsorbed molecular oxygen on
silver,23,28 and the peak at 367 cm−1 manifested due to δ
(W−O−W) bending mode of bridging oxygens.18 Thus, the
Figure 4. Ag 3d core level spectra: (a) fresh Ag/WO3 and (b) spent
Ag/WO3.
the WO42− and Ag+ crystals were confirmed by TGA and FTIR
analyses (Supporting Information, Figures S6, S7). The TGA
curve depicts a decomposition peak around 400 °C, which is
consistent with the decomposition temperature of CTAB. A
total 13.9% mass loss suggests that surfactant molecules remain
strongly associated with the WO42− and Ag+ crystals for the as-
synthesized sample. Moreover, the morphology is not affected
by the removal of surfactant during calcination.
To get further support for the association of PVP and CTAB,
FTIR was taken for the calcined and uncalcined samples. The
FTIR spectra of the uncalcined sample exhibit the presence of
CN, CO, N+−CH3, and C−H functional groups. The
peaks between 700 and 900 cm−1 can be assigned to bridging
O−W−O.20 The vibrational band at 1628 cm−1 may be
assignable to CO stretching, which is consistent with the
presence of PVP.21 A sharp, intense peak at 1408 cm−1 may be
attributed to the C−N bond of the CTA+ moiety.22 Thus, the
presence of CN, CO, and N+−CH3 confirms that both
CTAB and PVP is involved in the nanostructure growth
process and control of the shape and size of the Ag
nanoparticles and WO3 nanorods.
The Ag/WO3 nanostructure catalyst was studied for the
propylene epoxidation reaction with molecular oxygen in a
fixed-bed, high-pressure microreactor. The catalyst showed
15.5% propylene conversion with 83% selectivity toward
propylene oxide at 250 °C and at 2 MPa pressure. It is a
well-known phenomenon that the activation of molecular
oxygen occurs by transferring charge density from metal to the
2171
dx.doi.org/10.1021/cs5004454 | ACS Catal. 2014, 4, 2169−2174

ACS Catalysis
Letter
8.3% Ag loading showed a propylene conversion of 21% with
67% PO selectivity. The conversion increased to 26% with 58%
PO selectivity when Ag loading was 14.6% (Table 1). We found
that the higher Ag loading resulted in an increase in the average
size of the AgNPs (15−20 nm with 8.3 wt % Ag loading, 30−50
nm with 14.6 wt % Ag loading) (Supporting Information,
Figures S1, S2). It is very clear from the experiment that very
small Ag NPs (<5 nm) facilitate the activation of molecular
oxygen and selectively transform propylene to propylene oxide.
The effect of pressure is shown in SI Figure S12. We
observed that pressure has a direct effect on the activation of
molecular oxygen as well as the selectivity of PO. In a silver-
catalyzed system, high oxygen pressure affects the activation
energy for dissociative chemisorption of O2 on the silver surface
and accelerates the conversion of the silver surface to a silver
oxide overlayer, as a result of which loosely bonded Ag−O
species are formed.24 The catalyst does not exhibit any activity
at atmospheric pressure. At 1 MPa pressure, the conversion of
propylene was ∼7%, and the selectivity to PO was 39%, but
with increasing pressure to 2 MPa, a significant increase in PO
selectivity of 83% was observed. However, with additional
pressure (3−4 MPa), although conversion increased, a
significant lowering of the PO selectivity (55%) was observed
as a result of the formation of carbon dioxide. This result is in
conformity with that of silver-catalyzed ethylene epoxidation
reaction, in which high oxygen pressure boosts the selectivity
toward ethylene oxide.24 We believe that Ag−O bond, with
bond length of 2.349 Å that is slightly more than the normal
Ag2O (bond length 2.25 Å), could easily donate its labile
oxygen atom to the CC bond of propylene to form PO
selectively. For this reason, our catalyst exhibited much better
performance than that of other silver-catalyzed systems
reported earlier (Supporting Information Table S2).
The effect of time on-stream is shown in SI Figure S13. It
was observed that although the conversion was constant until
18 h, the PO selectivity decreases slightly after 18 h, from 83%
to 79%. The plausible mechanistic pathway is shown in SI
Scheme S1. We believe that initially, an oxygen molecule
dissociates over the metallic Ag and a Ag2O species with a Ag−
O bond length of 2.349 Å is formed. This oxygen associated
with Ag enables the formation of a four-membered cyclic
transition state, with the propylene double bond followed by
the insertion of oxygen into the CC bond of propylene to
form PO. This study is in conformity with the earlier report in
which the direct abstraction of the O atom from Ag3O to the
double bond of propylene took place according to the
Langmuir−Hinshelwood mechanism.29
Figure 5. Raman spectra of (a) W (VI) Oxide, (b) fresh Ag/WO3, and
(c) spent Ag/WO3.
Raman analysis is in full agreement with the EXAFS and XPS
analysis.
The Ag/WO3 nanostructure catalyst was inactive toward
selective oxidation of propylene to propylene oxide with
molecular oxygen at atmospheric pressure and a temperature
below 150 °C, but the catalyst showed activity when both the
pressure and temperature were increased. It is worth
mentioning that PO is the dominant product during the
reaction conditions over Ag/WO3 (Ag loading was 4.8 wt %).
Acrolein, acrylic acid, CO2, and other oxygenated products
were detected as side products (Table 1).
The effect of temperature was carried out with the Ag/WO3
catalyst; the result is presented in SI Figure S11. It was
observed that when the temperature is increased from 200 to
400 °C, conversion of propylene to PO increases from 9% to
27%, but a decreasing trend in PO selectivity (91% to 54%) was
observed. The catalyst was inactive at a temperature below 150
°C. A PO selectivity of ∼83% was observed after 6 h of reaction
time at 250 °C using 4.8 wt % Ag loading catalyst. With
increasing temperature, the selectivity to PO decreases, most
probably because of the decomposition of propylene because
we detected the formation of acetaldehyde, acetic acid, and
formaldehyde as the main side products. A similar variation of
propylene to PO conversion with different Ag loadings on the
WO3 surface was also observed (Table 1).
It was found that with increasing silver loading, although the
conversion increases, the PO selectivity decreases. The catalyst
with 4.8 wt % Ag loading exhibited a propylene conversion of
15.5% with 83% propylene oxide selectivity. The catalyst with
a
Table 1. Catalytic Activities in Propylene Epoxidation Reaction
d
selectivity (%)
c
propylene conversion
(%)
propylene
oxide
b
e
entry
catalyst
loading of [Ag] (wt %)
acrolein CO2 other(s) PO production rate (mol POgcat−1 h−1
)
f
1
Ag/WO3
Ag/WO3
Ag/WO3
4.8
8.3
15.5
21
83
67
58
8
13
17
2
5
7
7
15
18
6.1 × 10−2
3.7 × 10−2
2.3 × 10−2
g
2
h
3
14.6
26
a
Reaction conditions: catalyst 0.30 g, reaction temperature = 250 °C, pressure = 2 MPa, feed gas C3H6 /O2/Ar = 4/2/94, gas hourly space velocity
b
(GHSV) = 16 000 mL gcat−1 h−1, time on stream = 6 h. Ag loadings estimated by inductively coupled plasma (ICP) atomic emission spectroscopy
(AES) analysis. Conversion of propylene = [moles of propylene reacted/initial moles of propylene used] × 100. PO selectivity = (moles of PO
formed/mol of propene converted) × 100. Others include acrylic acid, acetic acid, acetaldehyde, and hydrocarbon. Ag particles size 2−5 nm. Ag
particles size 15−20 nm. Ag particles size 30−50 nm.
c
d
e
f
g
h
2172
dx.doi.org/10.1021/cs5004454 | ACS Catal. 2014, 4, 2169−2174

ACS Catalysis
Letter
To find out whether isolated WO3 nanorods are necessary
for the propylene oxidation reaction, we prepared several other
metal oxide supports and studied their catalytic activity (SI
Table S3). The activities of the Ag catalyst supported on
different group VI oxides, such as Cr2O3 and MoO3, were found
to be very low as compared with the Ag/WO3 nanostructured
catalysts. We also prepared CuO particles supported on WO3
and found that the catalyst exhibits ∼4% propylene conversion
with 60% PO selectivity. Although the Cu catalyst showed good
PO selectivity, the catalyst deactivation after 3 h may be due to
the sintering of the Cu particles. Ag/WO3 prepared by the
conventional impregnation method showed irregular and larger
particle sizes, leading to negligible (<5%) PO selectivity. Ag or
WO3 alone also does not exhibit any activity. The study
revealed that using metallic AgNPs with 2−5 nm on WO3
nanorods of 30−40 nm diameter is the key parameter for the
catalytic activity, in which the rod-shaped tungsten oxide
facilitates uniform dispersion of ultrasmall (<5 nm) AgNPs,
which facilitates the dissociation of molecular oxygen to form
Ag2O. The synergistic effect between AgNPs and WO3
nanorods plays the crucial role for the formation of PO, and
the support tungsten oxide not only stops coalescence and
agglomeration of the AgNPs but also highly supports PO
formation. The Ag/WO3 catalyst does not change its shape and
size during propylene epoxidation reaction, as confirmed from
TEM analysis (Supporting Information, Figure S14).
In summary, we explored an easy and facile synthesis strategy
to prepare silver nanoparticles with size <5 nm supported on
WO3 nanorods with diameters of 30−40 nm using surfactant
cetyltrimethylammonium bromide (CTAB), capping agent
polyvinylpyrrolidone (PVP), and hydrazine. The metallic
AgNPs supported on the WO3 nanorods activated molecular
oxygen, which can produce propylene oxide with very high
selectivity directly from propylene without any additional
reducing agent. The catalyst showed a propylene conversion of
15.5% with 83% PO selectivity at 250 °C using molecular
oxygen. Using AgNPs of 2−5 nm and WO3 nanorods with 30−
40 nm diameters is the determining factor for propylene
oxidation reaction. We believe that this promising catalyst may
become a potential catalyst for other various oxidation
reactions.
thank ASD, IIP, for analytical services. XAFS measurements
were performed at KEK-IMSS-PF with the approval of the
Photon Factory Advisory Committee (project 2010G109). R.B.
acknowledges CSIR for funding Five Year Plan Project (12
FYP), CSC-0125.
REFERENCES
■
(1) (a) Cavani, F.; Teles, J. H. ChemSusChem 2009, 2, 508−534.
(b) Nijhuis, T. A.; Musch, S.; Makkee, M.; Moulijn, J. A. Appl. Catal.,
A 2000, 196, 217−224.
(2) (a) Zuwei, X.; Ning, Z.; Yu, S.; Kunlan, L. Science 2001, 292,
1139−1141. (b) Nijhuis, T. A.; Makkee, M.; Moulijn, J. A.;
Weckhuysen, B. M. Ind. Eng. Chem. Res. 2006, 45, 3447−3459.
(c) Weissermel, K.; Arpe, H. J. Industrial Organic Chemistry, 4th ed.;
Wiley-VCH: Weinheim, Germany, 2003, p 146.
(3) McCoy, M. Chem. Eng. News 2001, 79, 19−20.
(4) Vaughan, O. P. H.; Kyriakou, G.; Macleod, N.; Tikhov, M.;
Lambert, R. M. J. Catal. 2005, 236, 401−404.
(5) Hayashi, T.; Tanaka, K.; Haruta, M. J. Catal. 1998, 178, 566−575.
(6) (a) Qi, C.; Okumura, M.; Akita, T.; Haruta, M. Appl. Catal., A
2004, 263, 19−26. (b) Haruta, M.; Date, M. Appl. Catal. 2001, 222,
427−437.
(7) (a) Qu, Z.; Cheng, M.; Huang, W.; Bao, X. J. Catal. 2005, 229,
446−458. (b) Van Santen, R. A.; Kuipers, H. P. C. E. Adv. Catal. 1987,
35, 265−321. (c) Lu, J.; Bravo-Suarez, J. J.; Atsushi Takahashi, A.;
́
Haruta, M.; Oyama, S. T. J. Catal. 2005, 232, 85−95. (d) Molina, L.
M.; Lee, S.; Sell, K.; Barcaro, G.; Fortunelli, A.; Lee, B.; Seifert, S.;
Winans, R. E.; Elam, J. W.; Pellin, M. J.; Barke, I.; Oeynhausen, V. V.;
Lei, Y.; Meyer, R. J.; Alonso, J. A.; Rodríguez, A. F.; Kleibert, A.;
Giorgio, S.; Henry, C. R.; Meiwes-Broer, K. H.; Vajda, S. Catal. Today
2011, 160, 116−130.
(8) (a) Yao, W.; Guo, Y. L.; Liu, X. H.; Guo, Y.; Wang, Y. Q.; Wang,
Y. S.; Zhang, Z. G.; Lu, G. Z. Catal. Lett. 2007, 119, 185−190.
(b) Serafin, J. G.; Liu, A. C.; Seyedmonir, S. R. J. Mol. Catal. A: Chem.
1998, 131, 157−168. (c) Barteau, M. A.; Madix, R. J. J. Am. Chem. Soc.
1983, 105, 344−349. (d) Wang, R. P.; Guo, X. W.; Wang, X. S.; Hao,
J. Q. Catal. Today 2004, 93−95, 217−222.
(9) Lei, Y.; Mehmood, F.; Lee, S.; Greeley, J.; Lee, B.; Seifert, S.;
Winans, R. E.; Elam, J. W.; Meyer, R. J.; Redfern, P. C.; Teschner, D.;
Schlogl, R.; Pellin, M. J.; Curtiss, L. A.; Vajda, S. Science 2010, 328,
̈
224−228.
(10) (a) Li, Y.; Wu, Y.; Ong, B. S. J. Am. Chem. Soc. 2005, 127,
3266−3267. (b) Hiramatsu, H.; Osterloh, F. E. Chem. Mater. 2004, 16,
2509−2511. (c) Kashiwagi, Y.; Yamamoto, M.; Nakamoto, M. J.
Colloid Interface Sci. 2006, 300, 169−175.
ASSOCIATED CONTENT
* Supporting Information
(11) Kamata, K.; Yonehara, K.; Sumida, Y.; Hirata, K.; Nojima, S.;
Mizuno, N. Angew. Chem., Int. Ed. 2011, 50, 12062−12066.
(12) Zhou, Y. X.; Zhang, Q.; Gong, J. Y.; Yu, S. H. J. Phys. Chem. C
2008, 112, 13383−13389.
■
S
Characterization techniques, experimental details, additional
analysis data. This material is available free of charge via the
Internet at http://pubs.acs.org.
(13) De, S.; Mandal, S. Colloids Surf., A 2013, 421, 72−83.
(14) Lamont, R. E.; Ducker, W. A. J. Am. Chem. Soc. 1998, 120,
7602−7607.
AUTHOR INFORMATION
Corresponding Author
*E-mail: raja@iip.res.in.
■
(15) Jana, N. R. Small 2005, 1, 875−882.
(16) Tatarchuk, V. V.; Sergievskaya, A. P.; Korda, T. M.; Druzhinina,
I. A.; Zaikovsky, V. I. Chem. Mater. 2013, 25, 3570−3579.
(17) Liang, X. L.; Dong, X.; Lin, G. D.; Zhang, H. B. Appl. Catal., B
2009, 88, 315−322.
Present Address
†(T.S.) Department of Complexity Science and Engineering,
Graduate School of Frontier Sciences, The University of
Tokyo, Kashiwanoha, Kashiwa-shi, Chiba 277−8561, Japan.
(18) Breedon, M.; Spizzirri, P.; Taylor, M.; Plessis, J.; McCulloch, D.;
Zhu, J.; Yu, L.; Hu, Z.; Rix, C.; Wlodarski, W.; Kalantar-zadeh, K. Cryst.
Growth Des. 2010, 10, 430−439.
(19) Lim, D. C.; Lopez-Salido, I.; Kim, Y. D. Surf. Sci. 2005, 598, 96−
103.
Notes
The authors declare no competing financial interest.
(20) Phuruangrat, A.; Ham, D. H.; Hong, S. J.; Thongtema, S.; Lee, J.
S. J. Mater. Chem. 2010, 20, 1683−1690.
ACKNOWLEDGMENTS
■
(21) Sun, H.; He, J.; Wang, J.; Zhang, S. Y.; Liu, C.; Sritharan, T.;
Mhaisalkar, S.; Han, M. Y.; Wang, D.; Chen, H. J. Am. Chem. Soc.
2013, 135, 9099−9110.
S.G. thanks UGC, S.S.A. thanks CSIR, and B.S. and R.K.S.
thank UGC, India for the fellowship. The Director, CSIR-IIP, is
acknowledged for his help and encouragement. The authors
(22) Sau, T. K.; Murphy, C. J. Langmuir 2005, 21, 2923−2929.
2173
dx.doi.org/10.1021/cs5004454 | ACS Catal. 2014, 4, 2169−2174

ACS Catalysis
Letter
(23) Pulido, A.; Concepcion, P.; Boronat, M.; Corma, A. J. Catal.
2012, 292, 138−147.
(24) Van Santen, R. A.; Neurock, M. Molecular Heterogeneous
Catalysis: A Conceptual and Computational Approach; Wiley-VCH:
Weinheim, 2006; pp 39−40.
(25) Serezhkina, S. V.; Tyavlovskaya, E. A.; Shevchenko, G. P.;
Rakhmanov, S. K. J. Non-Cryst. Solids 2005, 351, 35−40.
(26) Ferraria, A. M.; Carapeto, A. P.; Botelho do Rego, A. M. Vaccum
2012, 86, 1988−1991.
(27) Jeong, S. J.; Lim, D. C.; Boo, J. H.; Lee, S. B.; Hwang, H. N.;
Hwang, C. C.; Kim, Y. D. Appl. Catal., A 2007, 320, 152−158.
(28) Pettinger, B.; Bao, X.; Wilcock, I.; Muhler, M.; Schlogl, R.; Ertl,
G. Angew. Chem., Int. Ed. 1994, 33, 85−86.
(29) 53rd Annual Report on Research 2008 (46131-G5 by Randall
Mayer).
2174
dx.doi.org/10.1021/cs5004454 | ACS Catal. 2014, 4, 2169−2174