Visible-light-induced hydrogen and oxygen formation over Pt/Au/WO 3 photocatalyst utilizing two types of photoabsorption due to surface plasmon resonance and band-gap excitation

Article abstract of DOI:10.1021/ja410230u

Photocatalytic H₂ and O₂ formations under visible light irradiation (λ > 400 nm) are demonstrated using Pt-Au nanopaticles for the reduction site and WO₃ for the oxidation site in solid-state Pt/Au/WO₃.

Full text of DOI:10.1021/ja410230u

Communication
pubs.acs.org/JACS
Visible-Light-Induced Hydrogen and Oxygen Formation over Pt/Au/
WO3 Photocatalyst Utilizing Two Types of Photoabsorption Due to
Surface Plasmon Resonance and Band-Gap Excitation
Atsuhiro Tanaka, Keiji Hashimoto, and Hiroshi Kominami*
Department of Applied Chemistry, Faculty of Science and Engineering, Kinki University, Kowakae, Higashiosaka, Osaka 577-8502,
Japan
S
* Supporting Information
the conduction band of WO3 through SPR, H2 would not be
ABSTRACT: Photocatalytic H2 and O2 formations under
visible light irradiation (λ > 400 nm) are demonstrated
using Pt−Au nanopaticles for the reduction site and WO3
for the oxidation site in solid-state Pt/Au/WO3.
formed because the potential of the conduction band of WO3
(ca. +0.5 VNHE at pH 0) is insufficient for the proton reduction
(0 VNHE at pH 0). Therefore, electron transfer to WO3 should
be avoided if we expect H2 formation in the Au/WO3 system.
In the course of our studies on SPR-type photocatalysts,7b we
reported formation of H2 from various compounds in aqueous
suspensions of Au/TiO2 with co-catalysts under visible light
irradiation. By using the traditional photodeposition (PD)
method and subsequent colloid photodeposition (CPD)10
method, Au/TiO2 with co-catalysts exhibiting stronger photo-
absorption at around 550 nm due to SPR and higher levels of
activity for H2 production were successfully prepared. These
results indicate that introduction of a co-catalyst such as
platinum is important for efficient H2 formation because the co-
catalyst works as a site for H+ reduction.
In this study, we prepared a new type of photocatalyst, i.e.,
Pt/Au/WO3, in which Pt nanoparticles were loaded on Au
particles. To achieve the Pt-on-Au structure, Au particles were
first loaded on WO3 by using CPD, and then Pt particles were
loaded on Au particles by using PD. The Pt/Au/WO3 sample
was used for formation of H2 and O2 under visible light
irradiation in the presence of sacrificial reagents such as biomass
under mediator-free conditions. We also investigated whether
two types of photoabsorption contribute to the formation of H2
and O2 by using an action spectrum, and we discuss the
working mechanism in formation of H2 and O2 over Pt/Au/
WO3.
The contents of Au and Pt were fixed at 3.0 and 0.5 wt%,
respectively, and the sample is designated here as Pt(0.5)/
Au(3.0)/WO3. Figure S1 shows absorption spectra of the WO3
and Pt(0.5)/Au(3.0)/WO3 samples. The bare WO3 sample
exhibited absorption only at λ < 500 nm due to the band-gap
excitation (2.8 eV). In the spectrum of the Pt(0.5)/Au(3.0)/
WO3 sample, strong photoabsorption was observed at around
550 nm, which was attributed to SPR of the supported
AuNPs.2−11
A transmission electron microscope (TEM) image of
Au(3.0)/WO3 prepared by CPD using the Au colloidal solution
is shown in Figure 1a. Gold particles were observed in the
image, indicating that AuNPs were deposited on the WO3
surface by CPD. The average diameter of Au particles in the
oupling of two kinds of photocatalysts responding to
C
visible light has recently attracted much attention as a Z-
scheme process. The Z-scheme process has various advantages:
(1) effective charge separation can be achieved, (2) high
reduction power and deep oxidation power can be simulta-
neously obtained even by using visible light having less energy,
and (3) various photocatalysts can be paired.1 However, since
stable photocatalysts producing H2 under visible light
irradiation are rare, only a few combinations have been
reported. In most cases, mediators connecting two kinds of
photocatalysts are essential. Therefore, an efficient photo-
catalyst for H2 formation under visible light irradiation and a
mediator-free system are needed.
It is known that gold nanoparticles (AuNPs) supported on
semiconductors such as cerium(IV) oxide and titanium(IV)
oxide (TiO2) exhibit strong photoabsorption due to surface
plasmon resonance (SPR).2−11 Recently, supported Au
materials have been used as visible-light-responding photo-
catalysts for various chemical reactions,2 including oxidation of
organic substrates,5 selective oxidation of an aromatic alcohol to
a carbonyl compound,6 H2 formation from alcohols,7 and
reduction of organic compounds.8 We consider an SPR-type
photocatalyst to be a strong candidate for an H2 formation
photocatalyst in the coupling of two kinds of photocatalysts. In
the case of Au/TiO2, TiO2 only works as an electron-
transportation material under visible light irradiation because
TiO2 only responds to UV light, which leads us to the idea that
tungsten(VI) oxide (WO3) can be used as a Au-supporting
material. Two absorptions due to band-gap excitation of WO3
(<450 nm) and SPR of AuNPs (450−600 nm) hardly overlap,
indicating that solar light can be effectively utilized without
conflict between WO3 and AuNPs. In a Au/WO3 sample, an
efficient charge separation can be expected by coupling of the
band-gap excitation of WO3 and the SPR of AuNPs. In this
case, WO3 and supported AuNPs would work as photocatalysts
for O2 and H2 evolution, respectively. Since AuNPs and WO3
are in direct juncture, no mediators are required for the
electron transfer in the Au/WO3 system. However, there is one
problem in the Au/WO3 system: if electrons of Au transform to
Received: October 5, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja410230u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
deactivation, even for long irradiation times. The total amount
of H2 gas evolved reached ca. 210 μmol, exceeding the
stoichiometric amounts of the photocatalysts (Au, 7.5 μmol,
corresponding to 3.0 wt%; WO3, 210 μmol), indicating that the
H2 production observed under the present irradiation
conditions was not the result of a quantitative reagent reaction
between Au and glycerin but the result of a (photo)catalytic
reaction.
The combination of two kinds of light-emitting diodes
(LEDs) has been shown to be useful for evaluating a composite
photocatalyst consisting of two types of photocatalysts
responding to light with different wavelengths.12 To examine
whether Au and WO3 in Pt(0.5)/Au(3.0)/WO3 contribute to
H2 formation under visible light, lights of blue and green LEDs
were used solely or simultaneously as the sources (see Figure
3a).
Figure 1. TEM photographs of (a) Au(3.0)/WO3 and (b) Pt(0.5)/
Au(3.0)/WO3.
sample was determined to be 13 nm (Figure S2), which was in
good agreement with the average diameter of original colloidal
AuNPs before Au loading (Figure S3). In the TEM image of
the Pt(0.5)/Au(3.0)/WO3 sample (Figure 1b), small Pt
particles were observed on a Au particle, indicating that
deposition of Pt on Au particles can be attributed to an increase
in the volume of nanoparticles.
Figure 2 shows time courses of evolution of H2 from glycerin
(10 μmol) in aqueous suspensions of Pt(0.5)/Au(3.0)/WO3
Figure 3. (a) Visible light from blue and green LEDs and (b) time
course of evolution of H2 from 2-propanol in aqueous suspensions of
Pt(0.5)/Au(3.0)/WO3 under visible light irradiation from the blue
and/or green LED.
Light of the blue LED mainly induces band-gap excitation of
WO3 because the blue light hardly overlaps with SPR of
AuNPs. On the other hand, light of the green LED matches
photoabosorption due to SPR of Au particles but does not
induce band-gap excitation of WO3. It was preliminarily
confirmed that formation of H2 was negligible in the
combination of green LED and Pt/WO3 and the combination
of blue LED and Au/WO3 because of mismatching between the
wavelength and the photoabsorption. Figure 3b shows
formation of H2 from 2-propanol in aqueous suspensions of
Pt(0.5)/Au(3.0)/WO3 under visible light irradiation from the
blue and/or green LED. The amount of H2 formed was very
small under irradiation by light from either the blue or green
LED. On the other hand, H2 was formed linearly with
irradiation by light from both the blue and green LEDs, and
apparent quantum efficiency (AQE) was determined to be
0.40%. These results indicate that both lights were essential for
H2 formation; in other words, band-gap excitation of WO3 and
SPR of AuNPs simultaneously contribute to H2 formation.
It is important to examine, for any given photoreaction
system, the wavelength response for the reaction. In a study on
an indium oxide−copper oxide (CuO) photocatalyst, we
examined the action spectrum under additional irradiation by
light only inducing band-gap excitation of CuO.12 To elucidate
the working mechanisms of Pt(0.5)/Au(3.0)/WO3, the action
spectrum was measured in H2 formation from 2-propanol
under additional irradiation by light from the blue or green
LED. Figure 4 shows the dependence of AQE on the
wavelength of incident light of monochromated visible light
from a Xe lamp with light width of 5 nm in the presence and
absence of irradiation by light from the blue LED (5.9 mW
Figure 2. Time courses of evolution of H2 (blue) and CO2 (red) from
glycerin (10 μmol) in aqueous suspensions of Pt(0.5)/Au(3.0)/WO3
under visible light irradiation from a Xe lamp with an L-42 filter.
under visible light irradiation from a Xe lamp with an L-42 filter
under a deaerated condition. Since a large amount of glycerin is
formed as a byproduct in production of soap and biodiesel oil
from fats and oils, effective utilization of glycerin is now keenly
desired. Just after visible light irradiation, H2 and CO2 were
evolved from suspensions of Pt(0.5)/Au(3.0)/WO3. With
increasing photoirradiation time, the amounts of H2 and CO2
increased linearly, and the rates of H2 and CO2 evolution were
determined to be 6.6 and 2.8 μmol h−1, respectively. Since the
amount of CO2 was in good agreement with that of H2, glycerin
was shown to have decomposed stoichiometrically to H2 and
CO2 as shown in eq 1.
C3H5(OH)3 + 3H2O → 3CO2 + 7H2
(1)
No gas was evolved in the dark between 0 and 5 h, indicating
that no thermocatalytic H2 formation occurred over Pt(0.5)/
Au(3.0)/WO3 under the present conditions. To evaluate the
stability of Pt/Au/WO3 in H2 and CO2 production from
glycerin, Pt/Au/WO3 was used repeatedly. Visible light
irradiation to the reaction mixtures after evacuation and
glycerin injection (10 μmol) resulted in evolution of H2 and
CO2 again, and the reaction proceeded with no noticeable
B
dx.doi.org/10.1021/ja410230u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
no change in the photocatalyst occurs because nothing is
deposited on the photocatalyst, in contrast to the use of silver
ions as electron acceptors. Rates of O2 formation from H2O
over various photocatalysts under irradiation by both the blue
and green LEDs are shown in Figure 6.
Figure 4. Dependence of rates of H2 evolution by Pt(0.5)/Au(3.0)/
WO3 on the wavelength of incident light in the presence and absence
of irradiation by light from the blue LED. The line means AQE under
irradiation by light from the blue LED. The subtraction spectrum
obtained from the spectra of Pt(0.5)/Au(3.0)/WO3 and WO3 is also
shown.
Figure 6. Rates of formation of O2 in aqueous suspensions of various
photocatalysts under visible light irradiation from the blue LED in the
presence and absence of the green LED.
cm−2). The value of AQE was determined by using eq 2. The
denominator of eq 2 includes all of the photons irradiated to
the photocatalyst, i.e., photons from both the monochromator
and blue LED.
2 × the amount of H2
the amount of incident photon
For comparison, results under irradiation by light from either
the blue or green LED are also shown. Results for WO3 and
Pt(0.5)/WO3 indicate that Pt worked as a co-catalyst in the
reduction of Cr6+ 13 and that the effect of green light irradiation
on O2 formation was negligible in WO3 and Pt(0.5)/WO3.
Since WO3 is not photoexcited by light from a green LED, the
negligible effect is reasonable. A synergistic effect of the band-
gap excitation by blue light and SPR by green light was
observed in O2 formation over Au(3.0)/WO3 and Pt(0.5)/
Au(3.0)/WO3. Results shown in Figure 6 indicate that the
combination of a WO3 semiconductor, AuNPs, and Pt co-
catalyst is effective for O2 formation and that green light, i.e.,
light with less energy, can be applied for acceleration of O2
formation. In the case of Pt/Au/WO3, AQE for water oxidation
was calculated to be 0.2% under visible light irradiation from
the blue and green LEDs.
AQE =
× 100
(2)
The dependency was similar to the subtraction spectrum
obtained from the spectra of Pt(0.5)/Au(3.0)/WO3 and WO3.
These results indicate that H2 formation was induced by
photoabsorption due to SPR of Au. The large difference in
AQE with and without the blue LED (Figure 4) indicates that
light from the blue LED greatly accelerated H2 formation.
Figure 5 shows action spectra in the presence and absence of
It was clearly shown that both the band-gap excitation of
WO3 and the photoabsorption due to SPR of Au supported on
WO3 contributed to the photocatalytic H2 and O2 formation
under irradiation by light from both the green and blue LEDs.
The higher photocatalytic activities of Pt(0.5)/Au(3.0)/WO3
are explained by an efficient charge separation by successful
coupling of the band-gap excitation of WO3 and the SPR of
AuNPs. Based on the results shown above, the expected
working mechanism for photocatalytic H2 formation in the
presence of electron donors over Pt/Au/WO3 under visible
light irradiation is shown in Scheme 1a.
Figure 5. Dependence of rates of H2 evolution by Pt(0.5)/Au(3.0)/
WO3 on the wavelength of incident light in the presence of irradiation
by light from the blue LED. The line means AQE under irradiation by
light from the green LED. The diffuse reflectance spectrum of WO3 is
also shown.
irradiation by light from the green LED (7.3 mW cm−2). The
dependency was similar to the absorption spectrum of WO3,
indicating that H2 formation was also induced by photo-
absorption due to the band-gap excitation of WO3. The large
gap between AQE around 400 nm in the presence and absence
of the green LED shows that the band-gap excitation of WO3
contributes to H2 formation only when there is SPR due to
AuNPs.
Scheme 1. Expected Reaction Mechanism for Production of
(a) H2 and (b) O2 in Aqueous Suspensions of Pt/Au/WO3
Since a WO3 photocatalyst is one of the most efficient
materials for O2 formation from water under visible light
irradiation, we examined water oxidation over the Pt/Au/WO3
sample. Water oxidation activity of the Pt/Au/WO3 sample was
evaluated in the presence of hexavalent chromium (Cr6+ in
Cr2O72−) as an electron acceptor.11 In this O2 evolution system,
C
dx.doi.org/10.1021/ja410230u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Five processes would occur: (1) the incident photons are
absorbed by Au particles through their SPR excitation,2−11 (2)
electrons are transferred from the Au particles into the Pt
particles, and (3) H+ is reduced by electrons over Pt, resulting
in the formation of H2. On the other hand, (4) the incident
photons are absorbed by WO3, and holes in the valence band of
WO3 oxidize various substrates such as glycerin and 2-propanol,
and (5) electrons in the conduction band of WO3 are
transferred to the electron-deficient Au particles, returning to
the original state. The mechanism may be applied for the
synergy effect of simultaneous irradiation by UV and visible
light on H2 formation over Au/TiO2,9 in which band-gap
excitation of TiO2 and SPR of Au particles are induced by UV
and visible light, respectively.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental details and Figures S1−S5. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
hiro@apch.kindai.ac.jp
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was partly supported by a Grant-in-Aid for Scientific
Research (No. 23560935) from the Ministry of Education,
Culture, Sports, Science, and Technology (MEXT) of Japan.
H.K. and A.T. are grateful for financial support from Iketani
Science and Technology Foundation. A.T. is grateful to the
Japan Society for the Promotion of Science (JSPS) for a
Research Fellowship for young scientists.
Since hydrogen overvoltage of Pt metal (0.01 V14) is smaller
than that of Au metal (0.18 V14), i.e., H2 evolution over Pt is
easier than that over Au, electron transfer from Au to Pt in
process (2) is reasonable. If electrons are transferred to the
conduction band of WO3 in process (2), H2 formation would
not occur any more because of insufficient potential of the
conduction band of WO3 for H+ reduction as mentioned in the
second paragraph. Formation of H2 indicates that electrons of
AuNPs are transferred to Pt particles, not to the conduction
band of WO3. For comparison, another sample, Au/WO3−Pt,
in which Pt particles were loaded on WO3 without alloying Au
particles,7b was prepared (Figure S4) and used for H2 formation
under the same conditions (Figure S5). The Au/WO3−Pt
sample showed negligible activity, indicating that electron
transfer from Au to the conduction band4 of WO3 (or Pt on
WO3) was predominant in this sample. Large differences in H2
formation rate between Pt/Au/WO3 and Au/WO3−Pt (Figure
S5) indicate that keeping the electron potential negative
(preventing electron transfer to WO3) in the Au/WO3 catalyst
system is important for H2 evolution under visible light
irradiation. The value of AQE over Pt/Au/WO3 was, however,
low, suggesting that a part of the electrons is injected to WO3
and finally transferred to the electron-deficient Au particles.
The expected working mechanism for photocatalytic O2
formation in the presence of an electron acceptor over Pt/
Au/WO3 under visible light irradiation is shown in Scheme 1b.
As well as the H2 formation system, effective charge separation
accounts for the higher activity of Pt/Au/WO3.
In summary, by using colloid photodeposition of Au particles
on WO3 followed by photodeposition of Pt particles onto Au
particles, a Pt/Au/WO3 sample was successfully prepared. The
Pt/Au/WO3 sample continuously yielded H2 and CO2 from
glycerin under visible light irradiation. Results for H2 formation
under visible light irradiation from a blue and/or green LED
indicate that both lights were essential for H2 formation. The
action spectrum under additional irradiation by light from the
blue LED or green LED clarified that both the band-gap
excitation of WO3 and SPR of AuNPs simultaneously
contribute to the H2 formation. Pt/Au/WO3 also produced
O2 by oxidation of H2O under visible light irradiation, to which
both the band-gap excitation of WO3 and SPR of AuNPs
simultaneously contributed. The SPR photocatalyst was applied
for the first time to a two-step photoexcitation system through
combination with WO3. There are many variations in
combination of plasmonic photocatalysts and band-gap photo-
catalysts. The results obtained in this study can be widely
applied to design of a new type of photocatalyst utilizing both
SPR and band-gap excitation.
REFERENCES
■
(1) (a) Sayama, K.; Mukasa, K.; Abe, R.; Abe, Y.; Arakawa, H. Chem.
Commun. 2001, 2416−2417. (b) Abe, R.; Takata, T.; Sugihara, H.;
Domen, K. Chem. Commun. 2005, 3829−3831. (c) Abe, R.; Sayama,
K.; Sugihara, H. J. Phys. Chem. B 2005, 109, 16052−16061. (d) Maeda,
K.; Higashi, M.; Lu, D.; Abe, R.; Domen, K. J. Am. Chem. Soc. 2010,
132, 5858−5868. (e) Sasaki, Y.; Kato, H.; Kudo, A. J. Am. Chem. Soc.
2013, 135, 5441−5449. (f) Maeda, K. ACS Catal. 2013, 3, 1486−1503.
(2) (a) Chen, X.; Shen, S.; Guo, L.; Mao, S. S. Chem. Rev. 2010, 110,
́ ́
6503−6570. (b) Navarro, R. M.; Sanchez-Sanchez, M. C.; Alvarez-
Galvan, M. C.; del Valle, F.; Fierro, J. L. G. Energy Environ. Sci. 2009, 2,
35−54. (c) Zhou, X.; Liu, G.; Yu, J.; Fan, W. J. Mater. Chem. 2012, 22,
21337−21354. (d) Linic, S.; Christopher, P.; Ingram, D. B. Nat. Mater.
2011, 10, 911−921. (e) Tatsuma, T. Bull. Chem. Soc. Jpn. 2013, 86, 1−
9.
(3) Tian, Y.; Tatsuma, T. J. Am. Chem. Soc. 2005, 127, 7632−7637.
(4) Furube, A.; Du, L.; Hara, K.; Katoh, R.; Tachiya, M. J. Am. Chem.
Soc. 2007, 129, 14852−14853.
(5) (a) Kowalska, E.; Abe, R.; Ohtani, B. Chem. Commun. 2009,
241−243. (b) Kominami, H.; Tanaka, A.; Hashimoto, K. Chem.
Commun. 2010, 46, 1287−1289.
(6) (a) Naya, S.; Teranishi, M.; Isobe, T.; Tada, H. Chem. Commun.
2010, 46, 815−817. (b) Tanaka, A.; Hashimoto, K.; Kominami, H. J.
Am. Chem. Soc. 2012, 134, 14526−14533.
(7) (a) Yuzawa, H.; Yoshida, T.; Yoshida, H. Appl. Catal., B 2012,
115, 294−302. (b) Tanaka, A.; Sakaguchi, S.; Hashimoto, K.;
Kominami, H. ACS Catal. 2013, 3, 79−85.
(8) Ke, X.; Sarina, S.; Zhao, J.; Zhang, X.; Chang, J.; Zhu, H. Chem.
Commun. 2012, 48, 3509−3511.
(9) (a) Silva, C. G.; Juarez, R.; Marino, T.; Molinari, R.; Garcia, H. J.
Am. Chem. Soc. 2011, 133, 595−602. (b) Mubeen, S.; Lee, J.; Singh,
N.; Kramer, S.; Stucky, G. D.; Moskovits, M. Nat. Nanotechnol. 2013,
8, 247−251. (c) Chem, J.-J.; Wu, J. C. S.; Wu, P. C.; Tsai, D. P. J. Phys.
Chem. C 2011, 115, 210−216.
(10) Tanaka, A.; Ogino, A.; Iwaki, M.; Hashimoto, K.; Ohnuma, A.;
Amano, F.; Ohtani, B.; Kominami, H. Langmuir 2012, 28, 13105−
13111.
(11) Tanaka, A.; Nakanishi, K.; Hamada, R.; Hashimoto, K.;
Kominami, H. ACS Catal. 2013, 3, 1886−1891.
(12) Sasaki, Y.; Tanaka, A.; Hashimoto, K.; Kominami, H. Chem. Lett.
2013, 42, 419−421.
(13) Xu, Y.; Chen, X. Chem. Ind. 1990, 6, 497−498.
(14) Kominami, H.; Furusho, A.; Murakami, S.-Y.; Inoue, H.; Kera,
Y.; Ohtani, B. Catal. Lett. 2001, 76, 31−34.
D
dx.doi.org/10.1021/ja410230u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX