Gold nanoparticles located at the interface of anatase/rutile TiO 2 particles as active plasmonic photocatalysts for aerobic oxidation

Article abstract of DOI:10.1021/ja2120647

Visible-light irradiation (λ > 450 nm) of gold nanoparticles loaded on a mixture of anatase/rutile TiO₂ particles (Degussa, P25) promotes efficient aerobic oxidation at room temperature. The photocatalytic activity critically depends on the catalyst architecture: Au particles with <5 nm diameter located at the interface of anatase/rutile TiO₂ particles behave as the active sites for reaction. This photocatalysis is promoted via plasmon activation of the Au particles by visible light followed by consecutive electron transfer in the Au/rutile/anatase contact site. The activated Au particles transfer their conduction electrons to rutile and then to adjacent anatase TiO₂. This catalyzes the oxidation of substrates by the positively charged Au particles along with reduction of O₂ by the conduction band electrons on the surface of anatase TiO₂. This plasmonic photocatalysis is successfully promoted by sunlight exposure and enables efficient and selective aerobic oxidation of alcohols at ambient temperature.

Full text of DOI:10.1021/ja2120647

Article
pubs.acs.org/JACS
Gold Nanoparticles Located at the Interface of Anatase/Rutile TiO2
Particles as Active Plasmonic Photocatalysts for Aerobic Oxidation
Daijiro Tsukamoto,† Yasuhiro Shiraishi,*,† Yoshitsune Sugano,† Satoshi Ichikawa,‡ Shunsuke Tanaka,§
and Takayuki Hirai†
†Research Center for Solar Energy Chemistry and Division of Chemical Engineering, Graduate School of Engineering Science,
Osaka University, Toyonaka 560-8531, Japan
‡Institute for NanoScience Design, Osaka University, Toyonaka 560-8531, Japan
§Department of Chemical, Energy and Environmental Engineering, Kansai University, Suita 564-8680, Japan
S
* Supporting Information
ABSTRACT: Visible-light irradiation (λ > 450 nm) of gold nano-
particles loaded on a mixture of anatase/rutile TiO2 particles
(Degussa, P25) promotes efficient aerobic oxidation at room tem-
perature. The photocatalytic activity critically depends on the cata-
lyst architecture: Au particles with <5 nm diameter located at the
interface of anatase/rutile TiO2 particles behave as the active sites
for reaction. This photocatalysis is promoted via plasmon activation
of the Au particles by visible light followed by consecutive electron
transfer in the Au/rutile/anatase contact site. The activated Au
particles transfer their conduction electrons to rutile and then to
adjacent anatase TiO2. This catalyzes the oxidation of substrates by the positively charged Au particles along with reduction of O2
by the conduction band electrons on the surface of anatase TiO2. This plasmonic photocatalysis is successfully promoted by
sunlight exposure and enables efficient and selective aerobic oxidation of alcohols at ambient temperature.
positively charged Au particles receive e− from the electron
INTRODUCTION
Gold nanoparticles on solid supports have been extensively
■
donor (Fe2+).
This e− transfer implies that, as depicted in Scheme 1,
studied as promising heterogeneous catalysts for a variety of
plasmon activation of Au particles on semiconductor particles
oxidation reactions. In particular, aerobic oxidation with O2 as
an oxidant has attracted a great deal of attention from the view-
point of green chemistry.1−4 Several types of substrates, such as
Scheme 1. Proposed Mechanism for Visible-Light-Driven
Aerobic Oxidation by Au Particles Supported on
Semiconductor Particles
alcohols,4−9 aldehydes,10 and hydrocarbons,11,12 are success-
fully oxidized at ∼393 K by Au particles on neutral supports
such as TiO2 or basic supports such as CeO2. Extensive studies
have revealed that the size of Au particles is critical for their
catalytic activity; Au particles with a diameter (dAu) of 2−5 nm show
especially high activity for oxidation of CO and propylene.13−15
The application of Au particles to optical materials for bio-
logical imaging/sensing has also attracted much attention be-
cause of their strong light absorption in the visible region, which
arises from a resonant oscillation of free electrons coupled by
light, known as localized surface plasmon resonance (SPR).16
The application of SPR to photovoltaic cells has also been
studied for light-to-energy conversion. Tian et al.17 reported that
visible-light irradiation (λ > 420 nm) of Au particles on a semi-
conductor TiO2 film coated on an indium tin oxide electrode
generates an anodic photocurrent in the presence of Fe2+. This
photoelectrochemical response is explained by an electron
transfer mechanism similar to that for dye-sensitized TiO2:18
collective oscillation of electrons on the Au particles induced
by the incident light transfers conduction electrons (e−)
from the Au particles to the TiO2 conduction band, and the
by visible light generates positive charges on the Au particles
and conduction-band e− on the semiconductor particles. This
means that oxidation and reduction reactions occur simulta-
neously, as with conventional semiconductor catalysts activated
Received: December 26, 2011
Published: March 22, 2012
© 2012 American Chemical Society
6309
dx.doi.org/10.1021/ja2120647 | J. Am. Chem. Soc. 2012, 134, 6309−6315

Journal of the American Chemical Society
Article
by UV light.19 This plasmonic photocatalyst, if employed for
aerobic oxidation, might oxidize a substrate (electron donor)
on the Au surface, while e− is consumed by the reduction of O2
(electron acceptor) on the semiconductor surface. This may
promote aerobic oxidation at room temperature under sunlight
irradiation,20 enabling green organic transformations.
This scenario has motivated the application of plasmonic
photocatalysis to aerobic oxidation. Kowalska et al.21 studied
the activity of Au particles (dAu = 12−60 nm) loaded on TiO2
for aerobic oxidation of 2-propanol at room temperature under
visible-light irradiation (λ > 450 nm). They reported that while
no reaction occurs in the dark, visible-light irradiation produces
acetone and larger Au particles loaded on rutile TiO2 show
better catalytic activity. Tanaka et al.22 studied aerobic oxidation
of benzylic alcohols with Au particles (dAu = 13−70 nm) loaded
on CeO2 at room temperature under irradiation with green
light (λmax = 530 nm). They also found that while no reaction
occurs in the dark, light irradiation produces benzaldehyde and
larger Au particles (dAu > 30 nm) show better activity. These
studies, however, did not take into consideration the size
dependence of the catalytic activity of Au particles for aerobic
oxidation in the dark;13−15 small Au particles (dAu < 5 nm) are
active, but large ones are inactive. This means that in the above
reports,21,22 the photoreaction was carried out using catalysts
with inherently low activity in the dark; therefore, it is unclear
whether the photoreaction is actually advantageous over the
dark reaction.
Figure 1. (a) Diffuse reflectance UV−vis spectra of the catalysts. The
spectra for other catalysts are summarized in Figure S2 in the
Supporting Information. (b) Action spectrum for aerobic oxidation of
1 on Au2(DP673)/P25. The apparent quantum yield for the formation
of 2 was calculated using the equation ΦAQY = (Yvis − Ydark)/(number
of photons entering the reaction vessel) × 100%, where Yvis and Ydark
are the amounts of 2 formed under light irradiation and dark
conditions, respectively.
(1 mmol) and catalyst (20 mg) under 1 atm O2. The solution
temperature was kept rigorously at 298
0.5 K. Figure 2
Herein we report an active plasmonic photocatalyst that
definitely shows enhanced activity relative to the reaction in the
dark at room temperature. The catalyst architecture is critical
for the activity: small Au particles (dAu < 5 nm) loaded on a
mixture of anatase/rutile TiO2 particles (Degussa, P25) are
necessary. The place where the Au particles are loaded is also
important: Au particles located at the anatase/rutile interface
behave as the active sites, facilitating efficient e− transfer to
TiO2 and successful aerobic oxidation under irradiation with
sunlight.
RESULTS AND DISCUSSION
■
Catalyst Preparation. Au/TiO2 catalysts were prepared by
a deposition−precipitation (DP) method13,15 using anatase
( Japan Reference Catalyst JRC-TIO-1; average particle size,
21 nm; BET surface area, 80 m2 g−1), rutile ( JRC-TIO-6;
15 nm; 104 m2 g−1), and P25 ( JRC-TIO-4; 24 nm; 57 m2 g−1;
anatase/rutile ≈ 83/17) TiO2, which were supplied by the
Catalyst Society of Japan. The respective TiO2 particles were
stirred in water (pH 7) with HAuCl4 at 353 K and calcined in
air at 473−873 K, affording purple powders of Aux(DPy)/TiO2,
where x is the amount (in wt %) of Au loaded [x wt % = Au/
(Au + TiO2) × 100%] and y is the calcination temperature
(in K). The transmission electron microscopy (TEM) images
of the catalysts show hemispherical Au particles (Figure S1 in
the Supporting formation). The average diameters of the Au
particles on the Au2(DP673)/anatase, Au2(DP673)/rutile, and
Au2(DP673)/P25 catalysts are dAu = 4.6, 4.2, and 3.7 nm,
respectively. Diffuse-reflectance UV−vis spectra of the catalysts
(Figure 1a) show a strong absorption band at ca. 550 nm
assigned to the SPR of the Au particles.20
Figure 2. Amounts of 2 formed during aerobic oxidation of 1 with
respective catalysts (black) in the dark or (white) under visible-light
irradiation (λ > 450 nm; light intensity at 450−800 nm, 16.8 mW
cm−2). The dAu values for the catalysts are given in parentheses. The
detection limit of 2 was 0.02 μmol (4 μM), and the range of
calibration was 0.004−200 mM.
summarizes the amounts of 2 formed after 12 h of reaction in
the dark (black bars) or under visible-light irradiation (Xe lamp,
λ > 450 nm, white bars). Notably, both reaction conditions
selectively produced 2, where the mass balance of 1 and 2
was >99% and no byproducts were detected. In addition, the
Catalytic Activity. The efficacy of the Au/P25 catalyst
for photoreaction was evidenced by the aerobic oxidation of
1-phenylethanol (1), which is often used as substrate for aero-
bic oxidation,7−9 to give acetophenone (2). The reaction was
performed by stirring a toluene solution (5 mL) containing 1
6310
dx.doi.org/10.1021/ja2120647 | J. Am. Chem. Soc. 2012, 134, 6309−6315

Journal of the American Chemical Society
Article
absence of O2 scarcely promoted the reaction. With bare P25,
no reaction occurred in the dark, and light irradiation produced
<10 μmol of 2. In contrast, Au2(DP673)/P25 in the dark
produced 17 μmol of 2 as a result of the high activity of the
small Au particles.13−15 Light irradiation further enhanced the
reaction: 74 μmol of 2 was produced, which is more than 4 times
that obtained from the dark reaction. This suggests that visible-
light irradiation of the Au/P25 catalyst indeed enhances the
reaction.
An anatase or rutile TiO2 support was ineffective for reac-
tion: the Au2(DP673)/anatase and Au2(DP673)/rutile catalysts
produced only ca. 10 μmol of 2 even under light irradiation.
Large Au particles were also ineffective for reaction: the
Au2(photo)/anatase, Au2(photo)/rutile, and Au2(photo)/P25
catalysts, which were prepared by a photodeposition method,21
contained Au particles with dAu = 45.3, 26.5, and 20.2 nm,
respectively,23 and were inactive under both sets of reac-
tion conditions. A semiconductor CeO2 support ( JRC-CEO-3;
20 nm; 82 m2 g−1) was also ineffective. The dark activity of
Au2(DP673)/CeO2 (dAu = 3.9 nm) was very high because of the
enhanced dehydrogenation of 1 on the basic CeO2 surface,5 but
the activity was scarcely enhanced by light irradiation. In
addition, Au2(photo)/CeO2 (dAu = 44.4 nm) was inactive
under both sets of conditions. These data clearly suggest that
small Au particles loaded on P25 TiO2 specifically show high
catalytic activity under visible-light irradiation. As shown in
Figure 1a, the SPR intensities of the catalysts are inconsistent
with their activities, although early reports suggested the asso-
ciation.21,22 These data indicate that the combination of small
Au particles and the mixed anatase/rutile TiO2 support is
crucial for high activity.
Figure 3. ESR spectra of the catalysts. The catalysts were treated with
20 Torr O2 (gray) in the dark or (black) under visible-light irradiation
at 298 K or (green) in the dark at 363 K. After evacuation, the samples
were measured at 77 K. The g = 1.997 signal (spectrum b) is assigned
to e− at the lattice trapping site of TiO2.26
probably interacts with a residual positive charge on the Au
particles, as shown in Figure 3b, and produces the peroxo-type
anion. It must be noted that, as shown in Figure 3b (green), the
Au/P25 catalyst, when treated with O2 at 363 K in the dark, did
not show the O−O− signal. This indicates that light-to-heat
conversion,28 even if it occurs on the photoactivated Au par-
ticles, does not promote O2 reduction. These findings strongly
support the e− transfer from photoactivated Au particles to TiO2
and subsequent O2 reduction on the TiO2 surface (Scheme 1).
Effects of Size and Location of the Au Particles. The
amount of Au loaded strongly affects the activity of the
plasmonic photocatalysis. Figure 4a shows the results of aerobic
oxidation using Aux(DP673)/P25 catalysts with different Au
loadings [x wt % = Au/(Au + TiO2) × 100%]. The activity
increased with the Au loading (∼2 wt %) but decreased at
higher loadings (>3 wt %); the activity of the Au5 catalyst under
visible-light irradiation was similar to the dark activity. This
suggests that Au loadings of 2−3 wt % show the highest
catalytic activity. The dAu values for the Au0.5, Au2, Au3, and Au5
catalysts were 2.4, 3.7, 4.9, and 7.8 nm, respectively, implying
that small Au particles with dAu < 5 nm are active for reaction.
To clarify further the effect of the Au particle size on the
activity, Au2(DPy)/P25 catalysts prepared at different calcina-
tion temperatures y (in K) were used for the reaction. As shown
Electron Transfer from Plasmon-Activated Au Particles.
The enhanced aerobic oxidation on Au/P25 by visible-light
irradiation is initiated by the plasmon activation of Au par-
ticles.17 Action spectrum analysis confirmed this (Figure 1b):
the SPR band of Au2(DP673)/P25 correlates well with the
apparent quantum yield (ΦAQY) for the formation of 2. As
shown in Scheme 1, the photoactivated Au particles transfer e−
to TiO2, and e− is consumed by the reduction of O2 there. This
was confirmed by electron spin resonance (ESR) analysis of the
catalysts at 77 K after treatment with O2 at room tempera-
ture in the dark or under visible-light irradiation. As shown in
Figure 3a (gray), bare P25 treated in the dark was ESR-silent.
In contrast, visible-light irradiation (black) created very weak
−
signals attributable to superoxide-type oxygen anions (O2 ;
gxx = 2.003, gyy = 2.009, gzz = 2,025),24 which is formed by the
reduction of O2 on the photoexcited TiO2 surface.25,26 As
shown in Figure 3b, Au2(DP673)/P25 in the dark was also ESR-
silent, but visible-light irradiation created very strong signals
(gxx′ = 2.002, gyy′ = 2.004, gzz′ = 2.008) attributable to peroxo-
type oxygen anions (O−O−).27 In contrast, as shown in Figure
3c,d, the Au2(DP673)/anatase and Au2(DP673)/rutile catalysts,
which were inactive for photoreaction (Figure 2), scarcely showed
these signals, although their dAu values are similar to that for
Au2(DP673)/P25. This suggests that the O−O− species do not
form on the Au surface but do form on the P25 TiO2 surface and
that, as shown in Scheme 1, plasmon activation of Au particles
transfers e− to P25 TiO2 and promotes O2 reduction there.
in Figure 4b, the dAu values for the DP473, DP573, DP673, DP773
,
and DP873 catalysts were 3.3, 3.4, 3.7, 7.6, and 8.3 nm,
respectively. The DP773 and DP873 catalysts with larger Au
particles showed much lower activities than DP673, suggesting
that Au particles with dAu > 5 nm are indeed inactive. In
contrast, dAu for DP473 was similar to that for DP673, but the
activity was much lower. This indicates that the calcination
temperature as well as the size of the Au particles affects the
activity for the plasmonic reaction.
The formation of O−O− species, not superoxide (O2 ), is
−
rationalized by O2 reduction at the Au/TiO2 interface. The
structure of the oxygen anion depends strongly on its local
environment.24 In the present case, the e− transfer might occur
at the Au/TiO2 interface. The formed oxygen anion therefore
The active sites for the plasmonic reaction on Au/P25 are
the Au particles (dAu < 5 nm) located at the anatase/rutile
6311
dx.doi.org/10.1021/ja2120647 | J. Am. Chem. Soc. 2012, 134, 6309−6315

Journal of the American Chemical Society
Article
upon calcination at 473 K, Au particles form mainly on the
rutile surface because a number of oxygen vacancies act as the
crystal nucleation sites.31 In contrast, calcination at higher
temperature promotes the migration of Au particles, and they
are stabilized at the anatase/rutile interface (Figure 5b).29 The
TEM image of Au2(DP473)/P25 (Figure 5a) shows Au particles
located on the surface of the TiO2 particles, but that of
Au2(DP673)/P25 (Figure 5b) shows Au particles located at the
interfaces of the TiO2 particles. As shown in Figure 6, the
Figure 6. HRTEM images of Au2(DP673)/P25. More images are
shown in Figure S4 in the Supporting Information.
Figure 4. Amounts of 2 formed during oxidation of 1 at 298 K with
(a) Aux(DP673)/P25 and (b) Au2(DPy)/P25 catalysts. The reaction
conditions were identical to those in Figure 2.
HRTEM images of Au2(DP673)/P25 clearly show the Au par-
ticles located at the anatase/rutile interface and well-defined
contact surfaces for the Au/anatase/rutile phases. X-ray photo-
electron spectroscopy (XPS) of the catalysts further confirmed
this. As shown in Figure 7, the binding energy for the Au 4f7/2
interfaces. The difference in the activities of catalysts prepared
at the different calcination temperatures 473−673 K (Figure
4b) is explained by the migration of Au particles on the P25
TiO2 surface during calcination, as clarified by Akita et al.29
During the preparation of Au particles on P25 TiO2 by the DP
method, calcination at ≥423 K fully reduces the Au species and
produces Au particles.30 As schematically shown in Figure 5a,
Figure 7. XPS results for the catalysts.
orbital of Au2(DP673)/rutile (82.58 eV) is much lower than that
of Au2(DP673)/anatase (82.71 eV) because of the strong
binding of Au with the oxygen vacancies on the rutile surface.31
The binding energy of Au2(DP473)/P25 (82.60 eV) is similar to
that of Au2(DP673)/rutile (82.58 eV), but that of Au2(DP673)/
P25 (82.65 eV) is much higher. These data indicate that, as
reported,29 Au particles prepared at higher calcination temper-
atures are mainly located at the anatase/rutile interfaces and
behave as the active sites for the plasmonic reaction.
Mechanism for Plasmonic Catalysis. Au2(DP673)/
anatase and Au2(DP673)/rutile, which are inactive for the plasmo-
nic reaction, show almost no ESR signals for O−O− species
(Figure 3c and d). This indicates that O2 is scarcely reduced
on these TiO2 surface. As reported,32 O2 reduction by the
conduction-band e− on the rutile surface is inefficient because
of the low affinity between the surface and O2.33 This may result
in inefficient O2 reduction on the rutile catalyst. In contrast,
Figure 5. Effect of calcination temperature on the location and size of Au
particles on P25 TiO229 and typical TEM images for Au2(DPy)/P25 cata-
lysts. More images are shown in Figure S3 in the Supporting Information.
6312
dx.doi.org/10.1021/ja2120647 | J. Am. Chem. Soc. 2012, 134, 6309−6315

Journal of the American Chemical Society
Article
anatase is active for O2 reduction.34 The almost complete
absence of O2 reduction on the anatase catalyst is probably due
to the weak binding between Au and anatase, as evidenced by the
XPS analysis (Figure 7). This may suppress e− transfer from
photoactivated Au particles to anatase. Efficient O2 reduction on
the Au/P25 catalyst is explained by the following mechanism, as
summarized in Scheme 2: (a) plasmon activation of Au particles
Scheme 2. Proposed Mechanism for Aerobic Oxidation of
Alcohol on the Au/P25 Catalyst under Visible-Light
Irradiation
Figure 8. Effect of reaction temperature on the amount of 2 formed
during oxidation of 1 with Au2(DP673)/P25. The reaction conditions
were identical to those in Figure 2.
visible-light irradiation, the highest activity was obtained at 298 K,
and the activity decreased at lower or higher temperature.
Two factors may affect the plasmonic catalysis. The activity
decrease at higher temperature is due to the positive ECB shift
of TiO2, as usually observed for dye-sensitized TiO2:41 e− trans-
fer within TiO2 decreases with increasing in temperature at
293−333 K. In the present case, this positive ECB shift probably
suppresses the e− transfer from rutile to anatase and/or the O2
reduction on the anatase surface, resulting in the activity de-
crease at higher temperature. This means that lower tempera-
ture is favorable for plasmonic catalysis. The activity, however,
decreases for T < 298 K. This is probably due to the decrease in
electron conductivity of Au particles with decreasing temper-
ature (<333 K).42 This may suppress the e− injection from
Au particles to TiO2 and decrease the catalytic activity. These
two factors affect the activity of plasmonic catalysis at lower or
higher temperature.
These data suggest that visible-light irradiation at around
room temperature are the conditions suitable for maximizing
the activity of plasmonic catalysis. Sunlight exposure is a typical
case. Table 1 summarizes the results of aerobic oxidation of
alcohols using the Au2(DP673)/P25 catalyst under sunlight,
where the solution temperature during exposure was 300−303 K.
Exposure to sunlight successfully promoted selective oxidation of
alcohols to carbonyl compounds in high yields (79−99%), and
the yields were higher than those obtained under dark con-
ditions at 303 K. These results suggest that the plasmonic cata-
lysis is successfully activated by sunlight and promotes aerobic
oxidation.
transfers e− to the tightly bound rutile; (b) e− is then transferred
to well-conjugated anatase; and, (c) O2 is reduced there by e−
(formation of O−O− species). The e− transfer from rutile to
anatase is thermodynamically unfavorable because the con-
duction band potential (ECB) of rutile (−0.05 V vs NHE) is
more positive than the ECB of anatase (−0.25 V).35 However, it is
allowed by a negative shift of the rutile ECB (Scheme 2b);34,36,37
the e− injection from Au particles to rutile accumulates e− in its
conduction band and results in the negative shift of ECB. This
promotes consecutive e− transfer from the Au particles to rutile
and then to anatase and efficient O2 reduction on the anatase
surface. Therefore, the Au/anatase/rutile joint active site
promotes the plasmonic reaction.
As shown in Figure 4b, the Au2(DPy)/P25 catalysts prepared
at higher calcination temperatures (>673 K) show decreased
activity. As shown in Figure 5c, calcination at >673 K produces
larger Au particles via the sintering of particles.29 The formation
of larger Au particles decreases the number of joint active sites,
thus resulting in an activity decrease.38 Aerobic oxidation at the
joint active site is explained as in Scheme 2. The O−O− species
(c) probably attracts the H atom of the alcohol while the
positive charge on Au adsorbs the alcohol, producing a Au−
alcoholate species (d).39,40 Subsequent removal of the H atom
from this species (e) produces the carbonyl product.
CONCLUSION
■
We have clarified that Au nanoparticles with diameters of
<5 nm located at the interface of anatase/rutile TiO2 particles
behave as active sites for plasmonic photocatalysis under
visible-light irradiation. The joint active site consisting of Au/
anatase/rutile phases facilitates smooth e− transfer from the
photoactivated Au particles to TiO2 and efficient O2 reduction.
Sunlight activation of the catalyst successfully promotes aerobic
oxidation of alcohols at ambient temperature. The catalyst
architecture clarified here for active plasmonic photocatalysis
may contribute to the design of more efficient catalysts driven
by visible light and may open a new strategy for selective
organic transformations by sunlight.
Aerobic Oxidation by Exposure to Sunlight. The reac-
tion temperature strongly affects the activity of the plasmo-
nic reaction. Figure 8 shows the results of aerobic oxidation
with Au2(DP673)/P25 at different temperatures. The dark ac-
tivity increased with increasing temperature.8 In contrast, under
EXPERIMENTAL SECTION
Preparation of Catalysts. Aux(DPy)/P25 catalysts [x = 0.5, 1, 2,
3, 4, 5; y = 473, 573, 673, 773, 873] were prepared by the following
■
6313
dx.doi.org/10.1021/ja2120647 | J. Am. Chem. Soc. 2012, 134, 6309−6315

Journal of the American Chemical Society
Article
The catalyst was dispersed well by ultrasonication for 5 min, and O2
was bubbled through the solution for 5 min. The tube was immersed
Table 1. Effect of Sunlight Exposure on the Aerobic
Oxidation of Alcohols Using the Au2(DP673)/P25 Catalyst
a
in a temperature-controlled water bath (298
0.5 K) and
photoirradiated at λ > 450 nm with magnetic stirring using a 2 kW
Xe lamp (USHIO Inc.)44 filtered through a CS3-72 glass (Kopp Glass
Inc.); the light intensity at 450−800 nm was 16.8 mW cm−2 (Figure
S6 in the Supporting Information). Sunlight reactions were performed
on Oct 4, 2011, from 10:00 to 14:00 at the top of the laboratory
building (north latitude 34.7°, east longitude 135.5°; see Figure S7 in
the Supporting Information). The light intensity at 300−800 nm was
11.5 mW cm−2 (Figure S6 in the Supporting Information). The
highest solution temperature during exposure was 303 K, and the dark
experiments were carried out at 303 K. After the reactions, the catalyst
was recovered by centrifugation, and the solution was analyzed by
GC−FID (Shimadzu GC-2040) equipped with a DB-17 capillary
column (Agilent Technologies, 30 m × 0.250 mm × 0.25 μm). The
substrate and product concentrations were calibrated with authentic
samples. The analysis was performed at least three times, and the
errors were 0.2%. Products were identified using a GC−MS system
(Shimadzu GCMS-QP5050A).
Action Spectrum Analysis. The photoreaction was carried out
with a toluene solution (2 mL) containing 1 (0.4 mmol) and the
Au2(DP673)/P25 catalyst (8 mg) within a Pyrex glass tube. After
ultrasonication and O2 bubbling, the solution was photoirradiated
using a 2 kW Xe lamp whose light was monochromatized by band-pass
glass filters (Asahi Techno Glass Co.). The full-width at half-maximum
(fwhm) of the light was 11−16 nm. The number of photon entering
the reaction vessel was determined with a spectroradiometer (USR-40,
USHIO Inc.).
ESR Measurements. The spectra were recorded in the X-band
using a Bruker EMX-10/12 spectrometer with a 100 kHz magnetic
field modulation at a microwave power of 10.0 mW.45 The magnetic
field was calibrated using 1,1′-diphenyl-2-picrylhydrazyl (DPPH) as a
standard. The catalyst (20 mg) was placed in a quartz ESR tube, which
was evacuated at 423 K for 3 h and cooled to room temperature. O2
(20 Torr) was introduced into the tube and kept for 10 min. The tube
was photoirradiated at 298 K using a Xe lamp at λ >450 nm. The tube
was then evacuated for 10 min to remove excess O2 and analyzed at 77 K.
Analysis. Total amounts of Au in the catalysts were determined
using an inductively coupled argon plasma atomic emission spectrom-
eter (ICAP-AES; SII Nanotechnology, SPS 7800) after dissolution of
the catalysts in aqua regia. TEM observations were carried out using
an FEI Tecnai G2 20ST analytical electron microscope operated
at 200 kV. XPS analysis was performed using a JEOL JPS-9000MX
spectrometer with Mg Kα radiation as the energy source. Diffuse-
reflectance UV−vis spectra were measured on an UV−vis spectropho-
tometer ( Jasco V-550 with Integrated Sphere Apparatus ISV-469)
using BaSO4 as a reference. XRF analysis was carried out on an X-ray
fluorescence spectrometer (SEA2110, Seiko Instruments Inc.).
a
Reaction conditions: toluene (5 mL), alcohol (10 μmol), catalyst
(20 mg), O2 (1 atm), exposure time (4 h). The light intensity at 300−
800 nm was 11.5 mW cm−2, which involves only ca. 2% <400 nm light.
b
The solution temperature during exposure was 300−303 K. The dark
c
reaction (−) was performed at 303 K. (Alcohol converted)/(initial
d
amount of alcohol) × 100%. (Product formed)/(initial amount of
e
f
alcohol) × 100%. Benzoic acid (0.4 μmol) was formed. p-Methoxy-
benzoic acid (0.4 μmol) was formed. Catalyst (10 mg). 1-Napthoic
acid (0.5 μmol) was formed.
g
h
procedure: P25 (1.0 g) was added to water (50 mL) containing
HAuCl4·4H2O (11.2, 22.9, 45.8, 68.7, 91.6, or 114.5 mg). The solution
pH was adjusted to ca. 7 with 1 mM NaOH, and the solution was
stirred at 353 K for 3 h. The particles were recovered by centrifugation,
washed thoroughly with water, and dried at 353 K for 12 h. The
powders were calcined under air with a heating rate of 2 K min−1 and a
holding time of 2 h at the designated temperature. Au2(DP673)/
anatase, Au2(DP673)/rutile, and Au2(DP673)/CeO2 were prepared in a
similar manner. X-ray fluorescence (XRF) analysis indicated that only
a trace amount of Cl− (<10 ppm) remained in the catalysts.
Au2(photo)/P25 was prepared as follows: TiO2 (1.0 g) was added
to a water/MeOH mixture (1:1 v/v, 50 mL) containing HAuCl4·4H2O
(45.8 mg) within a Pyrex glass tube (φ, 25 mm; capacity, 100 mL),
and the tube was sealed with a rubber septum cap. The particles were
dispersed by ultrasonication for 5 min, and Ar gas was bubbled through
the solution for 15 min. The tube was photoirradiated at λ > 320 nm
for 2 h at 298 K with magnetic stirring using a 300 W Hg lamp
(Eikohsha Co.)43 and then filtered through a 5 wt % CuSO4 solution.
The powders were recovered by centrifugation, washed thoroughly
with water, and dried at 353 K for 12 h. Au2(photo)/anatase and
Au2(photo)/rutile were prepared in a similar manner. Au2(photo)/
CeO2 was prepared by photoirradiation (2 h, 298 K) of CeO2 (1.0 g)
suspended in water (50 mL) containing HAuCl4·4H2O (45.8 mg) and
citric acid monohydrate (70.5 mg). XRF analysis indicated that only a
trace amount of Cl− (<10 ppm) remained in the catalysts.
ASSOCIATED CONTENT
■
S
* Supporting Information
Catalyst properties (Table S1), TEM images and size distribu-
tion of Au particles (Figure S1), diffuse-reflectance UV−vis spectra
of catalysts (Figure S2), typical TEM images of Au2(DP)y/P25
(Figure S3), HRTEM images of Au2(DP673)/P25 (Figure S4),
XRD patterns of catalysts (Figure 5), spectral irradiance of the
light source (Figure S6), and an image of the sunlight reaction
(Figure S7). This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
shiraish@cheng.es.osaka-u.ac.jp
Reaction Procedure. The catalyst (20 mg) was added to toluene
(5 mL) containing the alcohol within a Pyrex glass tube (φ, 10 mm;
capacity, 20 mL), and the tube was sealed with a rubber septum cap.
Notes
The authors declare no competing financial interest.
6314
dx.doi.org/10.1021/ja2120647 | J. Am. Chem. Soc. 2012, 134, 6309−6315

Journal of the American Chemical Society
Article
(30) Zanella, R.; Giorgio, S.; Shin, C.-H.; Henry, C. R.; Louis, C.
J. Catal. 2004, 222, 357−367.
ACKNOWLEDGMENTS
■
This work was supported by a Grant-in-Aid for Scientific
Research (23360349) from the Ministry of Education, Culture,
Sports, Science and Technology (MEXT), Japan. D.T. thanks
the Japan Society for Promotion of Science ( JSPS) for a
Research Fellowship for Young Scientists.
(31) Chen, M.; Goodman, W. Acc. Chem. Res. 2006, 39, 739−746.
(32) Ohno, T.; Sarukawa, K.; Matsumura, M. J. Phys. Chem. B 2001,
105, 2417−2420.
(33) Smith, W. R.; Forg, D. G. J. Phys. Chem. 1965, 69, 3587−3592.
(34) Ohno, T.; Tokieda, K.; Higashida, S.; Matsumura, M. Appl.
Catal., A 2003, 244, 383−391.
(35) Maruska, H. P.; Ghosh, A. K. Sol. Energy 1978, 20, 443−458.
(36) Dunn, W. W.; Aikawa, Y.; Bard, A. J. J. Am. Chem. Soc. 1981,
103, 3456−3459.
REFERENCES
■
(1) Haruta, M. Nature 2005, 437, 1098−1099.
(2) Herzing, A. A.; Kiely, C. J.; Carley, A. F.; Landon, P.; Hutchings,
G. J. Science 2008, 321, 1331−1335.
(37) Fujii, M.; Kawai, T.; Kawai, S. Chem. Phys. Lett. 1984, 106, 517−
522.
(38) The anatase-to-rutile phase transition of P25 upon calcination at
higher temperature is not the major factor causing decreased photo-
catalytic activity. As shown in Figure S5 in the Supporting Information),
XRD patterns indicated the anatase/rutile ratios of Au2(DP673)/P25 and
Au2(DP773)/P25 catalysts to be 83:17, which is similar to that of pure
P25, although the increased amount of rutile in Au2(DP873)/P25 as a
result of the phase transition changed the ratio to 73:27. As shown in
Figure 4b, Au2(DP773)/P25 showed lower activity than Au2(DP673)/P25,
even though their anatase/rutile ratios are similar. This indicates that the
phase transition of P25 is not the major factor causing the lower
photocatalytic activity of the catalysts prepared at higher calcination
temperatures.
(3) Pina, C. D.; Falletta, E.; Prati, L.; Rossi, M. Chem. Soc. Rev. 2008,
37, 2077−2095.
(4) Gajan, D.; Guillois, K.; Delicher
̀
e, P.; Basset, J. M.; Candy, J. P.;
Caps, V.; Coper
131, 14667−14669.
(5) Abad, A.; Concepcion, P.; Corma, A.; Garcia, H. Angew. Chem.,
Int. Ed. 2005, 44, 4066−4069.
́
et, C.; Lesage, A.; Emsley, L. J. Am. Chem. Soc. 2009,
(6) Pratti, L.; Rossi, M. J. Catal. 1998, 176, 552−560.
(7) Ishida, T.; Nagaoka, M.; Akita, T.; Haruta, M. Chem.Eur. J.
2008, 14, 8456−8460.
(8) Abad, A.; Corma, A.; Garcia, H. Chem.Eur. J. 2008, 14, 212−
222.
(39) It is well-known that in the aerobic oxidation of alcohols with
supported Au catalysts under dark conditions, the interaction between
the alcohol and the surface of the support is very important for
oxidation activity. In that case, basic supports such as CeO2 accelerate
the deprotonation of alcohols and promote the formation of Au−
alcoholate species (see refs 5 and 8). As shown in Table S1 in the
Supporting Information, the point of zero charge of Au2(DP673)/P25
was determined to be 5.9, suggesting that the P25 surface is weakly
acidic and less active for the deprotonation of alcohols. This means
that the interaction between the alcohol and the P25 surface scarcely
affects the plasmonic catalysis.
(40) A similar mechanism for the formation of Au−alcoholate species
was proposed by Ishida et al. (ref 7). They reported that Au clusters
supported on porous coordination polymers (PCPs) showed very high
activity for aerobic oxidation of alcohols under dark conditions, even
though the PCP support is weakly acidic. This high activity was
explained by successful O2 activation on the Au clusters due to the
electronic interaction with the PCP support. This produces the
peroxo-type oxygen anion and positive charge on the Au clusters and
promotes the formation of Au−alcoholate species, similar to our
system (c → d in Scheme 2).
(9) Mitsudome, T.; Noujima, A.; Mizugaki, T.; Jitsukawa, K.; Kaneda, K.
Adv. Synth. Catal. 2009, 351, 1890−1896.
(10) Corma, A.; Domine, M. E. Chem. Commun. 2005, 4042−4044.
(11) Hughes, M. D.; Xu, Y.-J.; Jenkins, P.; McMorn, P.; Landon, P.;
Enache, D. I.; Carley, A. F.; Attard, G. A.; Hutchings, G. J.; King, F.;
Stitt, E. H.; Johnston, P.; Griffin, K.; Kiely, C. J. Nature 2005, 437,
1132−1135.
(12) Turner, M.; Golovko, V. B.; Vaughan, O. P. H.; Abdulkin, P.;
Berenguer-Murcia, A.; Tikhov, M. S.; Johnson, B. F. G.; Lambert, R. M.
Nature 2008, 454, 981−984.
(13) Haruta, M. Catal. Today 1997, 36, 153−166.
(14) Valden, M.; Lai, X.; Goodman, D. W. Science 1998, 281, 1647−
1650.
(15) Haruta, M.; Date, M. Appl. Catal., A 2001, 222, 427−437.
(16) Jain, P. K.; Huang, X.; El-Sayed, I. H.; El-Sayed, M. A. Acc.
Chem. Res. 2008, 41, 1578−1586.
(17) Tian, Y.; Tatsuma, T. J. Am. Chem. Soc. 2005, 127, 7632−7637.
(18) Gratzel, M. Nature 2001, 414, 338−344.
̈
(19) Linsebigler, A. L.; Lu, G.; Yates, J. T. Jr. Chem. Rev. 1995, 95,
735−738.
(20) Primo, A.; Corma, A.; Garcia, H. Phys. Chem. Chem. Phys. 2011,
13, 886−910.
(41) O’Regan, B. C.; Durrant, J. R. J. Phys. Chem. B 2006, 110, 8544−
8547.
(21) Kowalska, E.; Mahaney, O. O. P.; Abe, R.; Ohtani, B. Phys.
Chem. Chem. Phys. 2010, 12, 2344−2355.
(42) Terrill, R. H.; Postlewaite, T. A.; Chen, C.; Poon, C.-D.; Terzis,
A.; Chen, A.; Hutchison, J. E.; Clark, M. R.; Wignall, G.; Londono, J. D.;
Superfine, R.; Falvo, M.; Johnson, C. S. Jr.; Samulski, E. T.; Murray,
R. W. J. Am. Chem. Soc. 1995, 117, 12537−12548.
(22) Tanaka, A.; Hashimoto, K.; Kominami, H. Chem. Commun.
2011, 47, 10446−10448.
(23) As shown in Figure S1 in the Supporting Information, the
photodeposition method creates Au particles with a wide size
distribution (ca. 40 nm), as observed for related materials (see refs
20−22). In contrast, the DP method creates Au particles with size
distributions of ca. 5 nm (see refs 13 and 15).
(24) Anteliff, K. L.; Murphy, D. M.; Griffiths, E.; Giamello, E. Phys.
Chem. Chem. Phys. 2003, 5, 4306−4316.
(43) Shiraishi, Y.; Saito, N.; Hirai, T. J. Am. Chem. Soc. 2005, 127,
8304−8306.
(44) Shiraishi, Y.; Sugano, Y.; Tanaka, S.; Hirai, T. Angew. Chem., Int.
Ed. 2010, 49, 1656−1660.
(45) Shiraishi, Y.; Saito, N.; Hirai, T. J. Am. Chem. Soc. 2005, 127,
12820−12822.
(25) Anpo, M.; Che, M.; Fubini, B.; Garrone, E.; Giamello, E.;
Paganini, M. C. Top. Catal. 1999, 8, 189−198.
(26) Rajh, T.; Ostafin, A. E.; Micic, O. I.; Tiede, D. M.; Thurnauer,
M. C. J. Phys. Chem. 1996, 100, 4538−4545.
(27) Chowdhury, B.; Bravo-Suarez, J. J.; Mimura, N.; Lu, J.; Bando,
K. K.; Tsubota, S.; Haruta, M. J. Phys. Chem. B 2006, 110, 22995−
22999.
(28) Chen, X.; Zhu, H.-Y.; Zhao, J.-C.; Zheng, Z.-F.; Gao, X.-P.
Angew. Chem., Int. Ed. 2008, 47, 5353−5356.
(29) Akita, T.; Lu, P.; Ichikawa, S.; Tanaka, K.; Haruta, M. Surf.
Interface Anal. 2001, 31, 73−78.
6315
dx.doi.org/10.1021/ja2120647 | J. Am. Chem. Soc. 2012, 134, 6309−6315