Article

Cite This: Langmuir 2019, 35, 5455−5462 pubs.acs.org/Langmuir

Doping of Nb5+ Species at the Au−TiO2 Interface for Plasmonic

Photocatalysis Enhancement
Yasuhiro Shiraishi,*,† Jun Imai,† Naoki Yasumoto,† Hirokatsu Sakamoto,† Shunsuke Tanaka,§
Satoshi Ichikawa,‡ and Takayuki Hirai†
†
Research Center for Solar Energy Chemistry, and Division of Chemical Engineering, Graduate School of Engineering Science and
‡
Institute for NanoScience Design, Osaka University, Toyonaka 560-8531, Japan
§
Department of Chemical, Energy, and Environmental Engineering, Kansai University, Suita 564-8680, Japan
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

ABSTRACT: Au nanoparticles loaded on semiconductor TiO2 absorb visible light due to

their surface plasmon resonance (SPR) and inject the photogenerated hot electrons (ehot−)
Downloaded via UNIV FED DE PERNAMBUCO on February 16, 2022 at 19:00:17 (UTC).

into the conduction band of TiO2. The separated charges promote oxidation and reduction
reactions. The step that determines the rate of the plasmonic photocatalysis on the Au/TiO2
system is the ehot− injection through the Schottky barrier created at the Au−TiO2 interface. In
the present work, niobium (Nb5+) oxide species were doped at the Au−TiO2 interface by
loading Nb5+ onto the TiO2 surface followed by deposition of Au particles (2 wt % of TiO2).
Visible light irradiation of the Au/Nb5+/TiO2 catalysts promotes aerobic oxidation of alcohols
with much higher efficiency than that of undoped Au/TiO2. Lewis acidity of the Nb5+ species located at the interface cancels the
negative charges of Au and creates a barrier with a narrower depletion layer, promoting tunneling ehot− injection. Efficiency of
the ehot− injection depends on the amount of Nb5+ doped. Loading small amounts of Nb5+ (∼0.1 wt % of TiO2) creates
mononuclear NbO4 species and shows large activity enhancement. In contrast, loading larger amounts of Nb5+ creates
aggregated polynuclear Nb2O5 species. They decrease the electron density of Au particles and weaken their SPR absorption.
This suppresses the ehot− generation on the Au particles and decreases the activity of plasmonic photocatalysis.

■ INTRODUCTION
Surface plasmon resonance (SPR), a resonant oscillation of
Scheme 1. Proposed Mechanism for SPR-Driven (a)
Overbarrier and (b) Tunneling Hot Electron Transfera
metal surface electrons by incident photons, is one attractive
feature of Au nanoparticles (AuNPs).1 AuNPs strongly absorb
visible light (λ > 400 nm). Application of this feature to
photocatalysis has therefore attracted attention for solar-driven
chemical conversions.2,3 For this purpose, semiconductor TiO2
loaded with AuNPs (Au/TiO2) has extensively been studied.4
Generally, the plasmonic photocatalysis is considered to be
initiated by the hot electron (ehot−) injection.5,6 Oscillation of
sp band electrons of AuNPs by visible light (Scheme 1)
produces ehot−. Some of the ehot− in the conduction band (CB)
are injected into the TiO2 CB, whereas hot holes (δ+) remain
on the AuNPs.7 These eCB− and δ+ promote reduction and
oxidation reactions, respectively.
The step that determines the rate of plasmonic photo-
catalysis is the ehot− injection from the photoactivated AuNPs
to the TiO2 CB.3 The Schottky barrier is created at the Au−
TiO2 interface due to the difference of their Fermi levels.8,9
The overbarrier path (Scheme 1a) is the mainly accepted
a
EF and ϕB denote the Fermi level and Schottky barrier height,
respectively.
mechanism for the ehot− injection. In this case, the ehot− with an
energy higher than the barrier height are injected, where high
overbarrier energy is the main limitation.10 Several methods
have therefore been proposed to enhance the ehot− injection by clarified yet, the tunneling ehot− injection is the key to creating
lowering the height of the Schottky barrier.11,12 In contrast, the highly active plasmonic photocatalysts.
tunneling path has been reported in some Au/TiO 2
systems.13,14 As shown in Scheme 1b, the ehot− are injected Received: December 7, 2018
into the TiO2 CB by penetrating migration with much lower Revised: March 5, 2019
energy.7 Although its triggering mechanism has not been Published: March 27, 2019

© 2019 American Chemical Society 5455 DOI: 10.1021/acs.langmuir.8b04075

Langmuir 2019, 35, 5455−5462
Langmuir
Figure 1. Typical TEM images of the respective catalysts.
Recently, we have clarified one of the factors triggering the
tunneling ehot− injection.15 We made AuNPs on the anatase
particles of the Degussa P25 TiO2.16,17 Visible light irradiation
of the Au/P25_anatase catalyst promotes aerobic oxidation
with high quantum yield (7.7% at 550 nm) via the tunneling
ehot− injection, whereas other commercially available TiO2
loaded with AuNPs shows low quantum yields (<1%). The
tunneling ehot− injection on the Au/P25_anatase catalyst can
be explained by the surface morphology of the P25_anatase
particles. As shown in Scheme 1a, TiO2 usually consists of 5-
fold-coordinated Ti4+ (Ti5c). In contrast, as shown in Scheme
1b, the P25_anatase particles contain a large number of 4-fold-
coordinated Ti4+ (Ti4c) on the surface due to its unique
synthesis based on the flame pyrolysis of TiCl4. The Lewis
acidic Ti4c species cancel the negative charge of AuNPs and
make the depletion layer of the barrier much narrower,
triggering tunneling ehot− migration. The “acidic species” at the
Au−TiO2 interface is the key for this phenomenon.
The purpose of this work is the creation of a new method to
enhance the activity of plasmonic photocatalysis of Au/TiO2
based on the acidic species. We doped Lewis acidic niobium
(Nb5+) oxide species18 at the interface for neutralization of the
AuNPs negative charges. Nb5+ was loaded onto TiO2, and
AuNPs were then loaded onto the obtained Nb5+/TiO2
particles. Visible light irradiation of the Au/Nb5+/TiO2
catalysts promotes aerobic oxidation with much higher
efficiency than undoped Au/TiO2 by the tunneling ehot−
injection. We clarified that the injection efficiency depends
on the amount of Nb5+ doped. Doping small amounts of Nb5+
shows large activity enhancement. Doping larger amounts of
5456
Article
Nb5+, however, decreases the activity due to the suppression of
the ehot− photogeneration on the AuNPs.
■
RESULTS AND DISCUSSION
Preparation and Characterization of the Catalyst. The
ST-21 anatase TiO2 powder supplied from Ishihara Sangyo
Co., Ltd. (average diameter, 20 nm; Brunauer−Emmett−
Teller surface area, 50 m2 g−1) was used for the catalyst
preparation. Nb5+x/TiO2 powders loaded with x wt % Nb5+ (=
Nb5+/TiO2 × 100) were prepared by the impregnation of
NbCl5 in water. Au/Nb5+x/TiO2 powders with 2 wt % AuNPs
[=Au/(Nb5+x/TiO2) × 100] were made by the deposition−
precipitation technique,16,19 where stirring the Nb5+x/TiO2
powders in water with HAuCl4·4H2O followed by calcination
afforded red-purple powders.
Figure 1 shows the transmission electron microscopy
(TEM) images of Au/TiO2, Au/Nb5+0.1/TiO2, and Au/
Nb5+0.5/TiO2 catalysts. All of the catalysts show spherical
AuNPs strongly attached to the TiO2 surface. As shown in
Figure 2, the average diameters of AuNPs on these catalysts
determined by the TEM observations are similar (∼4 nm). As
shown in Figure 3, diffuse-reflectance (DR) UV−vis spectrum
of Au/Nb5+0.1/TiO2 shows strong SPR absorption of AuNPs
similar to that of undoped Au/TiO2.15 In contrast, Au/
Nb5+0.5/TiO2 shows decreased SPR absorbance. This indicates
that doping larger amounts of Nb5+ at the Au−TiO2 interface
scarcely affect the size of AuNPs, but decreases the SPR
absorbance.
Photocatalysis. Figure 4a summarizes the amounts of
acetophenone formed by aerobic oxidation of 1-phenylethanol
at 298 K for 12 h in the dark (black) or under λ > 450 nm light
DOI: 10.1021/acs.langmuir.8b04075
Langmuir 2019, 35, 5455−5462
Langmuir
Figure 2. Size distribution of AuNPs on the respective catalysts.
Figure 3. DR UV−vis spectra of the respective catalysts.
Figure 4. (a) Amounts of acetophenone produced by the reaction on
the respective catalysts (black) in the dark or (yellow) under λ > 450
nm light irradiation (light intensity at 450−800 nm, 16.8 mW cm−2).
(b) Change in the amounts of acetophenone formed on Au/Nb5+0.1/
TiO2 in the dark or under photoirradiation, where 10 mg of catalyst
was used. For all runs, acetophenone was produced solely with >99%
selectivity.
irradiation by a Xe lamp (yellow). Bare TiO2 and Nb5+0.1/TiO2
scarcely produce acetophenone (<0.1 μmol) even under visible
5457
Article
light. Au/TiO2 gives 15 μmol of acetophenone under visible
light, but Au/Nb5+x/TiO2 gives much larger amounts of
acetophenone, suggesting that Nb5+ doping enhances the
activity. The highest activity is achieved with the catalyst
doped with 0.1 wt % Nb5+. The Au/Nb5+0.1/TiO2 catalyst
produces 63 μmol of acetophenone, which is 4 times of that
formed on Au/TiO2, although its activity in the dark is much
lower. However, doping more than 0.1 wt % Nb5+ decreases
the activity. It is noted that AuNPs loaded on Nb2O5 (Au/
Nb2O5) scarcely promotes the reaction. In addition, the
Nb5+0.1/Au/TiO2 catalyst, prepared by impregnation of Nb5+
onto the Au/TiO2 catalyst, shows almost the same activity as
Au/TiO2. These data indicate that the doping of appropriate
amount of Nb5+ at the Au−TiO2 interface is necessary for the
activity enhancement.
Apparent quantum yields for this reaction on Au/Nb5+0.1/
TiO2 determined by monochromatic light irradiation15,16 are
3.1% at 450 nm, 4.4% at 550 nm, and 2.7% at 700 nm, where
550 nm irradiation exhibits the highest yield. The results agree
well with the SPR absorbance of the catalyst shown by a red
line in Figure 3, suggesting that SPR activation of AuNPs
initiates the reaction. Figure 4b shows the time-dependent
change in the amounts of acetophenone formed during the
reaction on Au/Nb5+0.1/TiO2. The activity is almost constant
during photoirradiation, indicating that the catalyst stably
promotes the reaction. In addition, the catalyst successfully
oxidizes several types of benzylic alcohols to the corresponding
aldehydes and ketones with high >95% selectivity (Table 1).
Furthermore, inductively coupled plasma (ICP) analysis of the
solution recovered after photoreaction did not detect leaching
of Au and Nb species, indicating that the species are stably
adhered on the surface even after the reaction.
ehot− Injection. Properties of the ehot− injection from the
photoactivated AuNPs to TiO2 were studied by photocurrent
response analysis of the catalysts supported on a fluorine tin
oxide electrode, where the ehot− photogenerated on the AuNPs
are injected into the TiO2 CB and transferred to the
electrode.20 As shown in Figure 5, the photocurrent density
on Au/Nb5+0.1/TiO2 is much larger than that on Au/TiO2 and
Au/Nb5+0.5/TiO2, where the order is consistent with the
photocatalytic activity (Figure 4a). It is noted that bare TiO2,
Nb5+0.1/TiO2, and Au/Nb2O5 exhibit very low current density.
The results also agree with the activity data (Figure 4a). These
data indicate that the doping of relatively low amounts of Nb5+
(∼0.1 wt %) at the Au−TiO2 interface enhances the injection
of ehot− from the photoactivated AuNPs to TiO2 CB and,
hence, enhances plasmonic photocatalysis.
The properties of ehot− injection were further studied using
electrochemical impedance spectroscopy (EIS), where the eCB−
injected to TiO2 behave as the charge carriers.21,22 Figure 6
shows the Nyquist plots of the catalysts in the dark or under
photoirradiation. The inset of this figure shows the proposed
equivalent circuit model,23 where RS is the Ohmic resistance,
RCT is the electron transfer resistance, and CDL is the double
layer capacitance. The EIS fitting parameters are shown in
Table 2. In the dark (closed keys), Au/Nb5+0.1/TiO2 shows
lower RCT than Au/TiO2 and Au/Nb5+0.5/TiO2. Visible light
irradiation of the catalysts (open keys) decreases RCT,
indicating that the ehot− photogenerated on AuNPs are indeed
injected into TiO2 CB. Au/Nb5+0.1/TiO2 shows much lower
RCT than Au/TiO2 and Au/Nb5+0.5/TiO2. The results agree
well with the photocatalytic activity (Figure 4a) and the
photoresponse data (Figure 5). These findings again suggest
DOI: 10.1021/acs.langmuir.8b04075
Langmuir 2019, 35, 5455−5462
Langmuir
Table 1. Aerobic Photooxidation of Various Types of
Benzylic Alcohols on Au/Nb5+0.1/TiO2a
a
Conditions: toluene (5 mL), Au/Nb0.15+/TiO2 (20 mg), substrate (5
mM), O2 (1 atm), time (24 h), λ > 450 nm, temperature (298 K).
b
=[substrate converted]/[initial amount of the substrate] × 100.
c
=[product formed]/[substrate converted] × 100.
Figure 5. Photocurrent response of the catalysts measured at 298 K in
0.5 M Na2SO4 under λ > 450 nm light at a bias of 0.1 V.
that doping small amounts of Nb5+ at the Au−TiO2 interface
enhances the ehot− injection from photoactivated AuNPs to
TiO2 CB. This thus promotes efficient charge separation and
exhibits activity enhancement of plasmonic photocatalysis. It is
well known that, in the heterojunction system, the width of the
depletion layer is inversely proportional to the double layer
capacitance (CDL).23 As shown in Table 2, Au/Nb5+0.1/TiO2
shows much larger CDL than Au/TiO2 and Au/Nb5+0.5/TiO2,
indicating that the depletion layer of Au/Nb5+0.1/TiO2 is much
narrower.
X-ray photoelectron spectroscopy (XPS) was used to clarify
the electronic properties of AuNPs. Figure 7a shows the charts
5458
Article
Figure 6. Nyquist plots of the catalysts measured at 298 K in 0.5 M
Na2SO4 at a bias of 2 V, (closed keys) in the dark and (open keys)
under λ > 450 nm irradiation.
Table 2. EIS Fitting Parameters from the Equivalent Circuit
of the Respective Catalystsa
substrate light RS/Ω RCT/kΩ CDL/μF
Au/TiO2 + 35.1 6.01 0.210
− 32.9 6.05 0.209
Au/Nb5+0.1/TiO2 + 23.9 3.04 0.331
− 23.5 5.16 0.309
Au/Nb5+0.5/TiO2 + 55.2 5.96 0.134
− 53.4 6.47 0.123
a
The data are from Figure 6.
of the catalysts at the Au 4f level. Au/TiO2 shows Au peaks at
lower binding energies than Au/Nb2O5. Fermi levels of metal
oxide semiconductors lie at a more positive position than those
of Au.24,25 This leads to an electron donation from metal oxide
semiconductors to AuNPs for the Fermi level balancing and
creates the Schottky barrier (Scheme 1). The Fermi level of
Nb2O5 lies at a more negative position than that of TiO2.26
This results in the donation of smaller amounts of electrons to
AuNPs and creates AuNPs with lower electron density. As a
result of this, Au/Nb2O5 shows very weak SPR absorption
(Figure 3) and scarcely produces ehot−, resulting in almost no
photocatalytic activity (Figure 4a).
As shown in Figure 7a, Au/Nb5+0.1/TiO2 shows Au peaks at
the binding energy similar to that of Au/TiO2, indicating that
the doped Nb5+ scarcely affects the electron density of AuNPs.
In contrast, Au/Nb5+0.5/TiO2 shows Au peaks at higher
binding energies. As shown in Figure 3, Au/Nb5+0.5/TiO2
shows SPR absorption much weaker than Au/TiO2 and Au/
Nb5+0.1/TiO2. These data suggest that doping excess amounts
of Nb5+ leads to binding between Au and Nb species and
decreases the electron density of AuNPs. As a result of this,
Au/Nb5+0.5/TiO2 shows weaker SPR absorption. This there-
fore suppresses the ehot− photogeneration and, hence, shows
decreased activity of plasmonic photocatalysis (Figure 4a). As
shown in Figure 7b, XPS charts of the catalysts at the Ti 2p
level revealed that the positions of Ti peaks are scarcely
affected by the amounts of Nb5+ doped, probably due to much
smaller amounts of Nb5+ than that of surface Ti species. These
data indicate that small amounts of Nb5+ (∼0.1 wt %) scarcely
affect the electron density of AuNPs and maintain strong SRP
absorption. In contrast, doping excess amount of Nb5+
decreases the electron density of AuNPs, resulting in weaker
SPR absorption and decreased photocatalytic activity.
DOI: 10.1021/acs.langmuir.8b04075
Langmuir 2019, 35, 5455−5462
Langmuir Article

Figure 7. XPS charts of the respective catalysts at (a) Au 4f, (b) Ti

2p, and (c) Nb 3d levels.

Structure of Doped Nb5+ Species. To clarify the

structure of the doped Nb5+ species, diffuse-reflectance infrared
Fourier transform spectroscopy (DRIFTS) was used at 100 K
with carbon monoxide (CO) as a probe molecule.18,27 As
shown in Figure 8a, the DRIFTS chart of bare TiO2 shows a
strong band at 2177 cm−1 assigned to CO adsorbed on the Ti5c
species and a weak band at 2156 cm−1 assigned to CO
adsorbed on the −OH groups of the metal oxide surface.28 In
this case, a component at 2183 cm−1 assigned to CO adsorbed
on the Ti4c species does not appear. This indicates that the
Lewis acidic Ti4c species15 do not exist on the surface of the
present ST-21 TiO2 powders. Figure 8. DRIFTS charts of CO adsorbed on (a) TiO2, (b) Nb5+0.1/
As shown in Figure 8b, the DRIFTS chart of Nb5+0.1/TiO2 TiO2, (c) Au/Nb5+0.1/TiO2, (d) Nb5+0.5/TiO2, or (e) Au/Nb5+0.1/
TiO2 at 100 K. The black line is the obtained chart, and the gray line
shows an additional component at 2174 cm−1 assigned to CO is the sum of the respective components.
adsorbed on the 4-fold-coordinated Nb5+ species (Nb4c).18
This suggests that, as shown in Scheme 2a, highly dispersed
mononuclear NbO4 species exist on the TiO2 surface.29,30 As nuclear NbO4 species exist at the Au−TiO2 interface (Scheme
shown in Figure 8c, Au/Nb5+0.1/TiO2 shows a new component 2a).
at 2128 cm−1 assigned to CO adsorbed on the AuNPs As shown in Figure 8d, the DRIFTS chart of Nb5+0.5/TiO2
surface,31 but it shows almost no Nb4c component (2174 shows the components at 2177 and 2174 cm−1 assigned to CO
cm−1). This suggests that almost all of the surface NbO4 adsorbed on the Ti5c and Nb4c species, respectively, as is the
species are covered by the loaded AuNPs. These findings case for Nb5+0.1/TiO2 (Figure 8b). In this case, a new strong
indicate that, on the Au/Nb5+0.1/TiO2 catalyst, the mono- component is observed at 2181 cm−1, which is assigned to CO
5459 DOI: 10.1021/acs.langmuir.8b04075
Langmuir 2019, 35, 5455−5462
Langmuir
Scheme 2. Proposed Au−TiO2 Interface for (a) Au/Nb5+0.1/
TiO2, and (b) Au/Nb5+0.5/TiO2
adsorbed on the 5-fold-coordinated Nb5+ species (Nb5c).18
This indicates that, as shown in Scheme 2b, loading larger
amounts of Nb5+ leads to aggregation of the NbO4 species and
creates aggregated Nb2O5 species. As shown in Figure 8e, Au/
Nb5+0.5/TiO2 shows almost no component at 2181 cm−1,
suggesting that the aggregated Nb2O5 are covered by the
loaded AuNPs. This indicates that, as shown in Scheme 2b, the
aggregated Nb2O5 species exist at the Au−TiO2 interface.
The formation of mononuclear NbO4 and aggregated Nb2O5
species on the respective Au/Nb5+0.1/TiO2 and Au/Nb5+0.5/
TiO2 catalysts is confirmed by the XPS charts at the Nb 3d
level (Figure 7c). The Nb peaks appear at higher binding
energy with an increase in the amount of Nb5+ doped (Au/
Nb5+0.1/TiO2 < Au/Nb5+0.5/TiO2 < Au/Nb2O5) owing to the
formation of aggregated Nb2O5 species.32,33 The data strongly
support the formation of mononuclear NbO4 and aggregated
Nb2O5 at the Au−TiO2 interface on the respective Au/
Nb5+0.1/TiO2 and Au/Nb5+0.5/TiO2 catalysts.
An interesting phenomenon observed in the DRIFTS charts
is the almost complete coverage of the NbO4 or Nb2O5 species
by the loaded AuNPs. The selective formation of AuNPs on
the Nb species may originate from the deposition−
precipitation method for the preparation of AuNPs,19 as
depicted in Scheme 3. At a neutral pH (∼7), the Au precursor
(HAuCl4) exists mainly as Au(OH)3 in an equilibrium with
Au(OH)4−, whereas neutral Ti−OH species form mainly on
the TiO2 surface.34,35 In contrast, the Nb species contain
polarized NbO groups.18 Electrostatic interaction between
Au(OH)3 and the NbO groups may therefore create Au
species adsorbed onto the Nb species, behaving as an Au
growth site. Thermal reduction of Au species followed by their
migration may lead to a growth of AuNPs there, resulting in
the selective formation of AuNPs around the Nb species.
5460
Article
Scheme 3. Proposed Mechanism for Selective Formation of
AuNPs on the Nb Oxide Species
Although we cannot provide direct evidence for the above
mechanism at the present stage, almost complete coverage of
the Nb species by AuNPs (Figure 8) indicates that the
interaction between the Au species and the Lewis acidic Nb
species proposed in Scheme 3 may probably be involved in this
phenomenon.
Mechanism for ehot− Injection Enhancement. The high
ehot− injection efficiency on the Au/Nb5+0.1/TiO2 catalyst
originates from the tunneling mechanism, where the ehot−
penetrate through the Schottky barrier.13−15 As shown in
Scheme 1a, on undoped Au/TiO2, AuNPs are charged
negatively by electron transfer from TiO2. The positively
charged wide depletion layer is created at the TiO2 surface and
suppresses the tunneling ehot− injection.36,37 On the Au/
Nb5+0.1/TiO2 catalyst, doping small amounts of Nb5+ creates
the mononuclear NbO4 (Nb4c) species at the Au−TiO2
interface (Scheme 2a). The Lewis acidity of the species18
makes the surface charge of TiO2 more positive and cancels
the negative charge of AuNPs. This creates a narrow depletion
layer, as confirmed by the larger CDL in the EIS analysis (Table
2) and, hence, facilitates tunneling ehot− injection. In contrast,
on the Au/Nb5+0.5/TiO2 catalyst, doping larger amounts of
Nb5+ creates aggregated Nb2O5 (Nb5c) species (Scheme 2b).
This leads to binding between AuNPs and the Nb2O5 species
and decreases the electron density of AuNPs. As a result of
this, the catalyst exhibits weaker SPR absorption (Figure 3)
and shows decreased activity of plasmonic photocatalysis
(Figure 4a).
■
CONCLUSIONS
We fabricated Au/Nb5+/TiO2 plasmonic photocatalysts doped
with Nb5+ species at the Au−TiO2 interface and found that
they efficiently catalyze aerobic oxidation of alcohols under
visible light. The doped Lewis acidic species cancel the
negative charges of AuNPs, creating a narrower depletion layer.
This triggers tunneling ehot− injection and enhances the activity
of plasmonic photocatalysis. The highest ehot− injection
efficiency is achieved by doping of relatively small amounts
of Nb5+ (∼0.1 wt %), where mononuclear NbO4 species exist
at the Au−TiO2 interface. Apparent quantum yield for the
photocatalysis (4.4% at 550 nm) is lower than that of our Au/
P25_anatase catalyst (7.7% at 550 nm),15 which employs
anatase particles of Degussa P25 TiO2 containing a large
number of surface Lewis acidic 4-fold-coordinated Ti4+ (Ti4c)
DOI: 10.1021/acs.langmuir.8b04075
Langmuir 2019, 35, 5455−5462
Langmuir
species. This indicates that the efficiency for tunneling ehot−
migration in the present system is lower than that in the Au/
P25_anatase system. The results presented here, however,
reveal that doping acidic species at the Au−TiO2 interface is an
efficient way for the enhancement of plasmonic photocatalysis.
So far, several strategies have been proposed for plasmonic
photocatalysis such as Ohmic contact between the plasmonic
nanoparticles and the semiconductor,38 transfer of hot holes
(δ+) formed on the plasmonic metal to the semiconductor,39
and electric field enhancement around the plasmonic metal.40
The doping acidic species at the Au−semiconductor interface
may become a new powerful strategy for plasmonic photo-
catalysis enhancement.
■
EXPERIMENTAL SECTION
General. All reagents were purchased from Wako, Tokyo Kasei,
and Sigma-Aldrich and used as received. Anatase ST-21 TiO2 was
kindly supplied by Ishihara Sangyo Co., Ltd. (Japan).
Catalyst Preparation. Nb5+x/TiO2 [x (wt %) = Nb5+/TiO2 ×
100; x = 0.01, 0.1, 0.3, and 0.5] was prepared as follows. TiO2 (0.5 g)
was stirred in 1 M HCl (10 mL) at 363 K for 3 h to remove the
impurities. The powders were washed with water and dried in vacuo
at room temperature for 12 h. The powders (0.2 g) and NbCl5 (1.2,
11.9, 35.6, or 59.3 mg) were suspended in water (20 mL), and their
pH values were adjusted to ∼12 with NaOH pellets. After stirring, the
suspension for 1 h at 363 K, the powders were recovered by
centrifugation, washed with water, and dried in vacuo at room
temperature for 12 h. Inductively coupled plasma (ICP) analysis
indicated that no Nb5+ remained in the resulting solution, suggesting
that all of the Nb5+ added were loaded on TiO2. Au/Nb5+x/TiO2
catalysts were prepared as follows. Nb5+x/TiO2 (0.15 g) was added to
water (50 mL) containing HAuCl4·4H2O (6.4 mg). Its pH was
adjusted to ∼7 with 0.1 mM NaOH, and the solution was stirred at
353 K for 3 h. The powders were washed with water and dried in
vacuo for 12 h. Then, the powders were calcined in air with a heating
rate of 2 K min−1 and a holding time of 2 h at 673 K. ICP analysis
indicated that no Au and Nb species remained in the resulting
solutions after the Au loading, suggesting that all of the Au species
added were successfully loaded onto the solids without leaching of the
Nb species. The mole fractions of the Nb and Au species on the
catalysts were therefore calculated using their feed amounts. The Au/
Nb/Ti mole ratios on the Au/Nb5+0.1/TiO2 and Au/Nb5+0.5/TiO2
catalysts were therefore determined to be 0.805:0.085:99.110 and
0.805:0.425:98.770, respectively.
Photocatalysis. The catalyst was added to a Pyrex tube (ϕ, 10
mm; capacity, 20 mL) containing toluene with alcohol, and the tube
was sealed with a rubber septum cap. The catalyst was dispersed by
ultrasonication, and O2 was bubbled for 5 min. The tube was left in a
water bath and photoirradiated at λ > 450 nm with magnetic stirring
by a 2 kW Xe lamp with a glass filter.41,42 After the reactions, the
catalyst was recovered by centrifugation and the solution was analyzed
by GC-FID.
Analysis. Electrochemical analysis was performed in a three-
electrode cell using 0.5 M Na2SO4 as an electrolyte with a Pt wire and
an Ag/AgCl electrode as the counter and reference electrodes,
respectively. The working electrode was prepared according to a
previously described procedure.15 EIS measurements were carried out
in the frequency range of 10 mHz to 100 kHz.43 DRIFTS charts were
obtained on a FT/IR-610 system (JASCP Corp.).44 The sample (30
mg) was added to the cell, evacuated (<1 Pa) at 423 K for 3 h, and
cooled to 100 K. CO (21 μmol, 100 Pa) was then added and left for 1
h. DR UV−vis and XPS analyses were performed according to the
procedure described previously.45 TEM observations were performed
using an FEI Tecnai G2 20ST analytical electron microscope (200
kV).46
5461
Article
■ AUTHOR INFORMATION
Corresponding Author
*E-mail: shiraish@cheng.es.osaka-u.ac.jp.
ORCID
Yasuhiro Shiraishi: 0000-0003-1812-0644
Shunsuke Tanaka: 0000-0001-5157-3317
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
This work was supported by the Grant-in-Aid for Scientific
Research (No. 18H01779) from the Ministry of Education,
Culture, Sports, Science and Technology, Japan (MEXT).
■ REFERENCES
(1) Brongersma, M. L.; Halas, N. J.; Nordlander, P. Plasmon-
Induced Hot Carrier Science and Technology. Nat. Nanotechnol.
2015, 10, 25−34.
(2) Wang, P.; Huang, B.; Ying, D.; Whangbo, M. H. Plasmonic
Photocatalysts: Harvesting Visible Light with Noble Metal Nano-
particles. Phys. Chem. Chem. Phys. 2012, 14, 9813−9825.
(3) Linic, S.; Christopher, P.; Ingram, D. B. Plasmonic-Metal
Nanostructures for Efficient Conversion of Solar to Chemical Energy.
Nat. Mater. 2011, 10, 911−921.
(4) Primo, A.; Corma, A.; Garcia, H. Titania Supported Gold
Nanoparticles as Photocatalyst. Phys. Chem. Chem. Phys. 2011, 13,
886−910.
(5) Tian, Y.; Tatsuma, T. Mechanisms and Applications of Plasmon-
Induced Charge Separation at TiO2 Films Loaded with Gold
Nanoparticles. J. Am. Chem. Soc. 2005, 127, 7632−7637.
(6) Furube, A.; Du, L.; Hara, K.; Katoh, R.; Tachiya. Ultrafast
Plasmon-Induced Electron Transfer from Gold Nanodots into TiO2
Nanoparticles. J. Am. Chem. Soc. 2007, 129, 14852−14853.
(7) Govorov, A. O.; Zhang, H.; Gun’ko, Y. K. Theory of
Photoinjection of Hot Plasmonic Carriers from Metal Nanostructures
into Semiconductors and Surface Molecules. J. Phys. Chem. C 2013,
117, 16616−16631.
(8) Schottky, W. Halbleitertheorie der Sperrschicht. Naturwissen-
schaften 1938, 26, 843.
(9) Mott, N. F. Note on the Contact between a Metal and an
Insulator or Semi-Conductor. Math. Proc. Cambridge Philos. Soc. 1938,
34, 568−572.
(10) Zuo, Z.; Cui, G.; Shi, Y.; Liu, Y.; Ji, G. Gold-Thickness-
Dependent Schottky Barrier Height for Charge Transfer in Metal-
Assisted Chemical Etching of Silicon. Nanoscale Res. Lett. 2013, 8,
973.
(11) Moon, S. Y.; Song, H. C.; Gwag, E. H.; Nedrygailov, I. I.; Lee,
C.; Kim, J. J.; Doh, W. H.; Park, J. Y. Plasmonic Hot Carrier-Driven
Oxygen Evolution Reaction on Au Nanoparticles/TiO2 Nanotube
Arrays. Nanoscale 2018, 10, 22180−22188.
(12) Lee, H.; Lee, H.; Park, J. Y. Direct Imaging of Surface Plasmon-
Driven Hot Electron Flux on the Au Nanoprism/TiO2. Nano Lett.
2019, 19, 891−896.
(13) Mubeen, S.; Hernandez-Sosa, G.; Moses, D.; Lee, J.; Moskovits,
M. Plasmonic Photosensitization of a Wide Band Gap Semi-
conductor: Converting Plasmons to Charge Carriers. Nano Lett.
2011, 11, 5548−5552.
(14) Moskovits, M. Hot Electrons Cross Boundaries. Science 2011,
332, 676−677.
(15) Shiraishi, Y.; Yasumoto, N.; Imai, J.; Sakamoto, H.; Tanaka, S.;
Ichikawa, S.; et al. Quantum Tunneling Injection of Hot Electrons in
Au/TiO2 Plasmonic Photocatalysts. Nanoscale 2017, 9, 8349−8361.
(16) Tsukamoto, D.; Shiraishi, Y.; Sugano, Y.; Ichikawa, S.; Tanaka,
S.; Hirai, T. Gold Nanoparticles Located at the Interface of Anatase/
Rutile TiO2 Particles as Active Plasmonic Photocatalysts for Aerobic
Oxidation. J. Am. Chem. Soc. 2012, 134, 6309−6315.
DOI: 10.1021/acs.langmuir.8b04075
Langmuir 2019, 35, 5455−5462
Langmuir Article

(17) Hirakawa, H.; Katayama, M.; Shiraishi, Y.; Sakamoto, H.; A.; Shalaev, V. M. Broadband Hot-Electron Collection for Solar
Wang, K.; Ohtani, B.; Ichikawa, S.; Tanaka, S.; Hirai, T. One-Pot Water Splitting with Plasmonic Titanium Nitride. Adv. Optical Mater.
Synthesis of Imines from Nitroaromatics and Alcohols by Tandem 2017, 5, No. 1601031.
Photocatalytic and Catalytic Reactions on Degussa (Evonik) P25 (39) DuChene, J. S.; Tagliabue, G.; Welch, A. J.; Cheng, W.-H.;
Titanium Dioxide. ACS Appl. Mater. Interfaces 2015, 7, 3797−3806. Atwater, H. A. Hot Hole Collection and Photoelectrochemical CO2
(18) Nakajima, K.; Baba, Y.; Noma, R.; Kitano, M.; Kondo, N. J.; Reduction with Plasmonic Au/p-GaN Photocathodes. Nano Lett.
Hayashi, S.; Hara, M. Nb2O5 nH2O as a Heterogeneous Catalyst with 2018, 18, 2545−2550.
Water-Tolerant Lewis Acid Sites. J. Am. Chem. Soc. 2011, 133, 4224− (40) Hayashido, Y.; Naya, S.-I.; Tada, H. Local Electric Field-
4227. Enhanced Plasmonic Photocatalyst: Formation of Ag Cluster-
(19) Haruta, M. Size- and Support-dependency in the catalysis of Incorporated AgBr Nanoparticles on TiO2. J. Phys. Chem. C 2016,
gold. Catal. Today 1997, 36, 153−166. 120, 19663−19669.
(20) Mali, S. S.; Shim, C. S.; Kim, H.; Patil, P. S.; Hong, C. K. In Situ (41) Shiraishi, Y.; Sugano, Y.; Tanaka, S.; Hirai, T. One-Pot
Processed Gold Nanoparticle-Embedded TiO2 Nanofibers Enabling Synthesis of Benzimidazoles by Simultaneous Photocatalytic and
Plasmonic Perovskite Solar Cells to Exceed 14% Conversion Catalytic Reactions on Pt@TiO2 Nanoparticles. Angew. Chem., Int. Ed.
Efficiency. Nanoscale 2016, 8, 2664−2677. 2010, 49, 1656−1660.
(21) Low, J.; Yu, J.; Li, Q.; Cheng, B. Enhanced Visible-Light (42) Sugano, Y.; Shiraishi, Y.; Tsukamoto, D.; Ichikawa, S.; Tanaka,
Photocatalyt17c Activity of Plasmonic Ag and Graphene Co-Modified S.; Hirai, T. Supported Au−Cu Bimetallic Alloy Nanoparticles: An
Bi2WO6 Nanosheets. Phys. Chem. Chem. Phys. 2014, 16, 1111−1120. Aerobic Oxidation Catalyst with Regenerable Activity by Visible-Light
(22) Liu, H.; Cheng, S.; Wu, M.; Wu, H.; Zhang, J.; Li, W.; Cao, C. Irradiation. Angew. Chem., Int. Ed. 2013, 52, 5295−5299.
Photoelectrocatalytic Degradation of Sulfosalicylic Acid and Its (43) Kofuji, Y.; Isobe, Y.; Shiraishi, Y.; Sakamoto, H.; Tanaka, S.;
Electrochemical Impedance Spectroscopy Investigation. J. Phys. Ichikawa, S.; Hirai, T. Carbon Nitride−Aromatic Diimide−Graphene
Chem. A 2000, 104, 7016−7020. Nanohybrids: Metal-Free Photocatalysts for Solar-to-Hydrogen
(23) Van Zeghbroeck, B. V. Principles of Semiconductor Devices and Peroxide Energy Conversion with 0.2% Efficiency. J. Am. Chem. Soc.
Heterojunctions, Prentice Hall: New Jersey, 2010. 2016, 138, 10019−10025.
(24) Bhanu, U.; Islam, M. R.; Tetard, L.; Khondaker, S. I. (44) Hirakawa, H.; Hashimoto, M.; Shiraishi, Y.; Hirai, T.
Photoluminescence Quenching in Gold−MoS2 Hybrid Nanoflakes. Photocatalytic Conversion of Nitrogen to Ammonia with Water on
Sci. Rep. 2014, 4, No. 5575. Surface Oxygen Vacancies of Titanium Dioxide. J. Am. Chem. Soc.
(25) Imanishi, A.; Tsuji, E.; Nakato, Y. Dependence of the Work 2017, 139, 10929−10936.
Function of TiO2 (Rutile) on Crystal Faces, Studied by a Scanning (45) Sakamoto, H.; Imai, J.; Shiraishi, Y.; Tanaka, S.; Ichikawa, S.;
Auger Microprobe. J. Phys. Chem. C 2007, 111, 2128−2132. Hirai, T. Photocatalytic Dehalogenation of Aromatic Halides on
(26) Xu, Y.; Schoonen, M. A. A. The Absolute Energy Positions of Ta2O5-Supported Pt-Pd Bimetallic Alloy Nanoparticles Activated by
Conduction and Valence Bands of Selected Semiconducting Minerals. Visible Light. ACS Catal. 2017, 7, 5194−5201.
Am. Mineral. 2000, 85, 543−556. (46) Shiraishi, Y.; Tanaka, K.; Shirakawa, E.; Sugano, Y.; Ichikawa,
(27) Hadjiivanov, K.; Lamotte, J.; Lavalley, J. C. FTIR Study of Low- S.; Tanaka, S.; Hirai, T. Light-Triggered Self-Assembly of Gold
Temperature CO Adsorption on Pure and Ammonia-Precovered Nanoparticles Based on Photoisomerization of Spirothiopyran. Angew.
TiO2 (Anatase). Langmuir 1997, 13, 3374−3381. Chem., Int. Ed. 2013, 52, 8304−8308.
(28) Deiana, C.; Fois, E.; Coluccia, S.; Martra, G. Surface Structure
of TiO2 P25 Nanoparticles: Infrared Study of Hydroxy Groups on
Coordinative Defect Sites. J. Phys. Chem. C 2010, 114, 21531−21538.
(29) Burcham, L. J.; Datka, J.; Wachs, I. E. In Situ Vibrational
Spectroscopy Studies of Supported Niobium Oxide Catalysts. J. Phys.
Chem. B 1999, 103, 6015−6024.
(30) Pittman, R. M.; Bell, A. T. Raman Studies of the Structure of
Nb2O5/TiO2. J. Phys. Chem. 1993, 97, 12178−12185.
(31) Green, I. X.; Tang, W.; Neurock, M.; Yates, J. T., Jr.
Spectroscopic Observation of Dual Catalytic Sites During Oxidation
of CO on a Au/TiO2 Catalyst. Science 2011, 333, 736−739.
(32) King, B. R.; Patel, H. C.; Gulino, D. A.; Tatarchuk, B. J. Kinetic
measurements of oxygen dissolution into niobium substrates: In situ
X-ray photoelectron spectroscopy studies. Thin Solid Films 1990, 192,
351−369.
(33) Morris, D.; Dou, Y.; Rebane, J.; Mitchell, C. E. J.; Egdell, R. G.;
Law, D. S. L.; Vittadini, A.; Casarin, M. Photoemission and STM
study of the electronic structure of Nb-doped TiO2. Phys. Rev. B 2000,
61, 13445−13457.
(34) Moreau, F.; Bond, G. C.; Taylor, A. O. Gold on Titania
Catalysts for the Oxidation of Carbon Monoxide: Control of pH
During Preparation with Various Gold Contents. J. Catal. 2005, 231,
105−114.
(35) Moreau, F.; Bond, G. C. Preparation and Reactivation of Au/
TiO2 Catalysts. Catal. Today 2007, 122, 260−265.
(36) Chen, J.; Zhang, L.; Z. Lam, Z.; Tao, H. B.; Zeng, Z.; Yang, H.
B.; Luo, J.; Ma, L.; Li, B.; Zheng, J.; Jia, S.; Wang, Z.; Zhu, Z.; Liu, B.
Tunneling Interlayer for Efficient Transport of Charges in Metal
Oxide Electrodes. J. Am. Chem. Soc. 2016, 138, 3183−3189.
(37) Padovani, F. A.; Stratton, R. Field and Thermionic-Field
Emission in Schottky Barriers. Solid-State Electron. 1966, 9, 695−707.
(38) Naldoni, A.; Guler, U.; Wang, Z.; Marelli, M.; Malara, F.;
Meng, X.; Besteiro, L. V.; Govorov, A. O.; Kildishev, A. V.; Boltasseva,

5462 DOI: 10.1021/acs.langmuir.8b04075

Langmuir 2019, 35, 5455−5462