Photocatalytic reduction of carbon dioxide by hydrous hydrazine over Au-Cu alloy nanoparticles supported on SrTiO3/TiO2 coaxial nanotube arrays

Article abstract of DOI:10.1002/anie.201409183

Efficient photocatalytic conversion of CO₂ into CO and hydrocarbons by hydrous hydrazine (N₂H₄?·H₂O) is achieved on SrTiO₃/TiO₂ coaxial nanotube arrays loaded with Au-Cu bimetallic alloy nanoparticles. The synergetic catalytic effect by the Au-Cu alloy nanoparticles and the fast electron-transfer in SrTiO₃/TiO₂ coaxial nanoarchitecture are the main reasons for the efficiency, while N₂H₄?·H₂O as the H source and electron donor provides a reducing atmosphere to protect the surface Cu atoms from oxidation, therefore maintaining the alloying effect which is the basis for the high photocatalytic activity and stability. This approach opens a feasible route to enhance the photocatalytic efficiency, which also benefits the development of photocatalysts and co-catalysts.

Full text of DOI:10.1002/anie.201409183

Angewandte
Chemie
DOI: 10.1002/anie.201409183
CO Photoreduction
2
Photocatalytic Reduction of Carbon Dioxide by Hydrous Hydrazine
over Au–Cu Alloy Nanoparticles Supported on SrTiO /TiO Coaxial
3
2
Nanotube Arrays**
Qing Kang, Tao Wang, Peng Li, Lequan Liu, Kun Chang, Mu Li, and Jinhua Ye*
Abstract: Efficient photocatalytic conversion of CO into CO
their high conduction band position, they are expected to
2
and hydrocarbons by hydrous hydrazine (N H ·H O) is
have enough energy to participate in the multielectron and
2
4
2
achieved on SrTiO /TiO2 coaxial nanotube arrays loaded
multiproton processes for reducing CO . Scheme S1 in the
3
2
with Au–Cu bimetallic alloy nanoparticles. The synergetic
catalytic effect by the Au–Cu alloy nanoparticles and the fast
electron-transfer in SrTiO /TiO coaxial nanoarchitecture are
Supporting Information shows the conduction band, valence-
band potentials, and band-gap energies of STO and TiO2
relative to the redox potentials at pH 7 of compounds
involved in CO2 reduction. Although the STO and TiO2
cannot provide sufficient potential to transfer a single photo-
3
2
the main reasons for the efficiency, while N H ·H O as the H
2
4
2
source and electron donor provides a reducing atmosphere to
protect the surface Cu atoms from oxidation, therefore
maintaining the alloying effect which is the basis for the high
photocatalytic activity and stability. This approach opens
a feasible route to enhance the photocatalytic efficiency,
which also benefits the development of photocatalysts and
co-catalysts.
[
4a]
generated electron to a free CO molecule (À1.9 VNHE), the
2
conduction band of STO and TiO are much higher than the
2
electrochemical reduction potentials of CO into formic acid,
2
carbon monoxide, formaldehyde, methanol, and methane.
This means that the multielectron and multiproton reactions
are feasible. Moreover, benefiting from the favorable energy-
band position, developing STO/TiO2 heterostructure for
better charge separation is an effective strategy to improve
the photocatalytic activity. Previous work reported that the
H
ydrocarbon fuels are currently the most important source
of energy due to their ready availability, stability, and high
[
1]
energy density. Solar-energy-driven conversion of CO into
more contact between STO and TiO , the better the photo-
2
2
[
6]
hydrocarbon fuels can simultaneously generate chemical fuels
catalytic activity. Therefore, close contact between STO and
TiO2 is necessary when designing a heterostructure with
a good photocatalytic performance.
[
2]
to meet energy demand and mitigate rising CO levels. The
2
utilization of the clean and renewable solar power resource is,
on a long-term basis, an essential component of solutions to
One-dimensional (1D) nanomaterials with various mor-
phologies, such as nanotube, nanowire, and nanorod, owing to
their large surface areas, have been demonstrated to facilitate
the electron transport and to minimize the loss of charge
[
2]
address growing global energy demand. Since Halmann
discovered the photoelectrochemical reduction of CO into
2
[3]
organic compounds in 1978, a growing interest in the
development of semiconductor photocatalysts has evolved.
To date, over 130 kinds of photocatalysts have been inves-
[
7]
carriers at grain boundaries. Among the 1D nanostructures,
highly ordered vertical TiO nanotube arrays (NTAs) have
2
[
4]
tigated to catalyze the CO2 reduction. Among them,
strontium titanate (SrTiO , STO) and titania (TiO ) are two
shown the significant photocatalytic activity of CO reduc-
2
[
2,8]
tion.
However, maintaining the long-time stability and
3
2
[
4a,5]
of the most investigated photocatalytic materials.
Since
high activity of the catalyst, especially the co-catalyst, is still
a great challenge. Noble-metal Pt nanoparticles (NPs) are
[9]
easily poisoned by CO during the catalytic process and for
[*] Dr. Q. Kang, Dr. T. Wang, Dr. P. Li, Dr. L. Liu, Dr. K. Chang, Prof. J. Ye
[10]
non-noble metal NPs, changes of surface states is the main
International Center for Materials Nanoarchitectonics (MANA), and
Environmental Remediation Materials Unit
National Institute for Materials Science (NIMS)
reason for the co-catalyst deactivation.
Herein, we develop a new approach that is able to achieve
1-1 Namiki, Tsukuba, Ibaraki 305-0044 (Japan)
high-rate UV/Vis-light-driven conversion of diluted CO into
2
CO and hydrocarbons in which STO/TiO coaxial nanotube
M. Li, Prof. J. Ye
2
Graduate School of Chemical Science and Engineering
Hokkaido University, Sapporo 060-0814 (Japan)
E-mail: Jinhua.YE@nims.go.jp
arrays loaded with an optimized combination of Au–Cu
bimetallic NPs are used as the photocatalyst. Under UV/Vis-
À2 À1
light illumination, a CO production rate of 138.6 ppmcm
h
À1 À1
Prof. J. Ye
(
3.77 mmolg h ) and total hydrocarbon production rate of
TU-NIMS Joint Research Center, School of Material Science and
Engineering, Tianjin University
À2 À1
À1 À1
2
6.68 ppmcm
h
(725.4 mmolg h ) are obtained on
Au Cu@SrTiO /TiO nanotube arrays by using diluted CO
2
3
3
2
92 Weijin Road, Tianjin (P.R. China)
(33.3% in Ar). Generally the highest rate of production (e.g.
[
**] This work received financial support from the World Premier
International Research Center Initiative (WPI Initiative) on Materials
Nanoarchitectonics (MANA), MEXT (Japan), and National Basic
Research Program of China (973 Program, 2014CB239301).
methane) in previous reports does not exceed tens of mmol
[4a,11]
per hour of illumination per gram of photocatalyst.
The
key improvement is the effectiveness of our photocatalyst
owing to following strategies: 1) employing high surface area
nanotube array architectures, with holes in the tube walls to
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201409183.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
enhance the gas diffusion and increase the contact between
photogenerated charge carriers and surface species; 2) devel-
TiO is amorphous, the XRD pattern does not show the TiO
2 2
peaks in the fresh STO/TiO sample. The top-view and cross-
2
oping STO/TiO heterostructures to facilitate the photogen-
sectional energy dispersive X-ray spectroscopy (EDX) map-
2
erated charge separation; 3) distributing noble bimetallic
alloy NPs co-catalysts along the nanotube arrays to help the
redox process; and 4) choosing hydrous hydrazine
pings (Figure S2) verify that the Au Cu NPs are uniformly
3
distributed on the surface of the STO/TiO nanotube arrays
2
and the Au Cu coating exists throughout the length of the
3
(
N H ·H O) as H source and electron donor to provide
nanotube arrays. Furthermore, the cross-section SEM image
2
4
2
a reductive atmosphere for keeping the alloying effect.
and EDX mappings confirm that the STO/TiO is a coaxial
2
For fabrication of the STO/TiO coaxial nanotube arrays,
nanotube structure. After annealing, the amorphous TiO are
2
2
TiO nanotube arrays were first prepared by anodic oxidation.
annealed to anatase, which is confirmed by the XRD pattern
(Figure 1B). Figure 1C shows a typical TEM image of the
Au Cu@STO/TiO nanotube arrays. It can be seen that small
2
The amorphous TiO was converted into STO by a hydro-
2
thermal method. Then Au–Cu bimetallic alloy NPs with
different Au/Cu ratios were deposited on the STO/TiO2
nanotube arrays by a microwave-assisted solvothermal
method. The carbon residue was removed by annealing. Full
details are given in the Supporting Information.
3
2
alloy NPs are uniformly dispersed on the support. The
average diameter of the Au Cu, Au Cu , and AuCu NPs is
3
2
2
3
calculated to be 5.14, 6.62, and 7.78 nm, respectively, indicat-
ing that the ratio of Au/Cu affects the size of the bimetal alloy
NPs (Figure S3). HRTEM image (Figure 1D) shows the
The morphologies of the nanotube array photocatalysts
are shown in SEM images (Figure 1A and Supporting
Au Cu bimetallic NPs is indexed as a face-centered cubic (fcc)
3
cuboctahedron structure. The charac-
terization of the alloy phase is sup-
ported by the fast Fourier transform
(FFT) pattern (Figure 1D, Inset).
The lattice constants calculated from
the lattice spacing of the alloy NPs
(
a = 0.396 nm) is between the lattice
constants for standard Au (JCPDS
4-0784, a = 0.4079 nm) and Cu
JCPDS 04-0836, a = 0.3615 nm),
0
(
and is similar to the value calculated
based on Vegardꢀs law (a =
[10]
0
.3955 nm).
The result suggests
that the Au and Cu components in
the alloy are homogeneously mixed.
As shown in Figure S4, the diffuse-
reflectance UV/Vis spectrum of the
Au@STO/TiO2 nanotube arrays
shows a distinctive surface plasmon
resonance (SPR) band at 570 nm. Cu
alloying decreases the SPR intensity
with a red shift of the band, as is the
case for colloidal Au–Cu alloy par-
[
10,12]
ticles.
Catalytic activity was tested for
Figure 1. A) SEM image, the white arrows indicate the holes in the tube wall; B) XRD pattern of
the photoreduction of CO . The reac-
2
STO/TiO nanotube arrays; C) TEM image of Au Cu@STO/TiO2 nanotube arrays; D) HRTEM
2
3
tions were performed by hanging
a piece of photocatalyst film in
image and FFT pattern of Au Cu nanoparticles.
3
1
0 mL N H ·H O or H O under
2 4 2 2
Information Figure S1). The amorphous TiO2 nanotube
arrays obtained by anodic oxidation are approximately
a diluted CO (33.3% in Ar) atmosphere in the dark or
2
under a Xe-lamp illumination. Figure 2A shows the amount
of CH generated on the Au-Cu@STO/TiO nanotube arrays
1
10 nm in diameter and 30 mm in length. After the hydro-
4
2
thermal process, a part of TiO changes to STO, causing the
by using N H ·H O or H O as the H source and electron
2 4 2 2
2
wall thickness to increase and thus the corresponding
diameter decreases to 100 nm. In addition, the length
decreases to 20 mm. Closer observations in Figure 1A shows
there are plenty holes (as indicated by arrows) in the tube
wall, which are beneficial for the gas flow diffusion and
facilitate the contact between the charge carriers and the
surface species. The X-ray diffraction (XRD) pattern in
Figure 1B confirms the existence of STO. Since unchanged
donor. The Au–Cu alloy co-catalysts exhibit much higher
[
10,13]
activity than Au and Cu owing to the alloying effect.
The
increase in the fraction of Au in bimetallic alloy NPs enhances
the photocatalytic activity, and the Au Cu@STO/TiO nano-
3
2
tube arrays give the largest CH evolution rate of
4
À2 À1
À1 À1
15.49 ppmcm
h
(421.2 mmolg h ). In contrast, the
Au@STO/TiO nanotube arrays and Cu@STO/TiO nanotube
2
2
arrays show lower activity of CH4 evolution. The most
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
[
15]
and it offering the advantage of CO-free H production.
2
[
5,16]
However, in contrast to the broad application of H O
and
2
[
5,17]
H2
as reductants, a study using N H ·H O in a photo-
2
4
2
catalytic CO reduction system has not been reported. The
2
decomposition of hydrazine to H and N is confirmed in our
2
2
experiments. Figure 3A shows a time-dependent change in
the amount of total hydrocarbon, CH , CO, H , and N
4
2
2
2
formed by photoreduction of CO on the Au Cu@STO/TiO
2
3
nanotube arrays catalyst under UV/Vis light illumination. The
increases of released gas are plotted as a function of
illumination
time.
A
CO
production
rate
of
À2 À1
À1 À1
À2 À1
1
38.6 ppmcm
h
(3.77 mmolg h ) and total hydrocarbon
À1 À1
production of 26.68 ppmcm
h
(725.4 mmolg h ) are
obtained under 6 h illumination. In the hydrocarbon produc-
tion, CH4 is the main product with an evolution rate of
À2 À1
À1 À1
1
5.49 ppmcm
h
(421.2 mmolg h ), which is 60% of the
total hydrocarbon products. The other hydrocarbons are
C H , C H , and C H , which are identified by GC (Fig-
2
6
2
4
3
6
ure S6). The corresponding quantum efficiency for the photo-
reduction is 2.51%. After five cycles measurement during
a 34 h test, the CH gas-evolution rate decreases from 15.49 to
4
À2 À1
1
3.57 ppmcm h , which is still 87.6% of its original activity
(
Figure 3B). In addition, H , N , and all the reduction
2 2
products increase continuously with increasing illumination
time. This result indicates that the N H ·H O provides
2
4
2
a strongly reductive atmosphere to maintain the effectiveness
of the photocatalyst, which promises a long stability of the
photocatalyst. This opens a feasible route to choose a reduc-
tant for CO reduction, which also aids the development of
2
the photocatalysts and co-catalysts.
To demonstrate the superiority of the nanoarchitecture,
several control experiments were carried out and the result is
shown in Figure 3C. The amounts of hydrocarbons formed on
Figure 2. A) Amount of CH formed during the photoreduction of CO
4
2
on Au-Cu@STO/TiO nanotube arrays relative to the Au fraction in
2
bimetallic alloy nanoparticles. B) ESR spectra of Au Cu@STO/TiO
3
2
nanotube arrays after the photoreduction reaction with N H ·H O and
2
4
2
STO/TiO coaxial nanotube arrays are much higher than that
2
H O as H source and electron donor.
2
on the STO nanotube arrays and TiO nanotube arrays. This
2
result suggests that the STO/TiO heterostructure can effec-
2
tively separate the photogenerated carriers, promoting the
photocatalytic activity. This conclusion is further supported
by the photoelectrochemical (PEC) performance evaluation
(Figure S7). Further loading with Au–Cu bimetallic alloy NPs
causes a sharp increase in the reduction products. This
demonstrates that the noble Au–Cu NPs help the redox
process. In particular, CO becomes the main product of the
reduction. The amount of hydrocarbons productions are in
interesting feature of the alloy co-catalyst is the significant
activity enhancement observed under N H ·H O protection.
2
4
2
Although using H O as H source and electron donor, the Au-
2
Cu NPs co-catalyst shows a similar alloy effect, it undergoes
a rapid deactivation during the photocatalytic reaction (Fig-
ure S5), which is caused by surface oxidation of Cu atoms with
photogenerated holes and oxygen species (such as ·OH,
[
10,14]
2
O2)
result demonstrates that N H ·H O maintains the Au–Cu
and elimination of the alloying effects. The present
tens of ppm/cm (which are multiplied by a factor of 10 in the
Figure 3C). In contrast, the Au + Cu @STO/TiO plate-like
2
4
2
3
1
2
alloying effect and promotes CO2 reduction without co-
catalyst deactivation, which is confirmed by ESR analysis
catalyst prepared by a step-by-step deposition of Au and Cu
metal onto the support exhibits much lower photoactivity.
This means the nanotube array architecture can greatly
increase the available surface area and provide specific
geometric pathways for charge transport. The photocatalytic
activity of Au Cu@STO/TiO plate is 3.5 times higher than
(
Figure 2B). After the photoreduction process in H O vapor,
2
+
2
an intense signal, which is assigned to Cu , is observed at
a g value of approximately 2.06,
[10,14]
indicating that surface
Cu atoms are oxidized. On the contrary, exposure of the
Au Cu@STO/TiO nanotube array sample to N H ·H O leads
3
2
that of Au + Cu @STO/TiO plate which indicates that the
3
2
2
4
2
3
1
2
2
+
to complete disappearance of the Cu signal. This suggests
that the oxidized surface Cu atoms are reduced by N H ·H O,
homogeneously mixed Au–Cu NPs are necessary for the alloy
effect. For Au–Cu@STO/TiO2 nanotube arrays, the alloy
effect plays an important role in improving the efficiency and
determining the selectivity. As a result of the alloy effect
caused by the homogeneously mixed Au–Cu atoms, the
electron densities on Au–Cu alloy NPs are much higher than
2
4
2
which guarantees a high photocatalytic activity for CO2
reduction.
N H ·H O is widely considered as a promising hydrogen-
2
4
2
storage material owing to its high content of hydrogen (7.9%)
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[
20,23]
hydrocarbons.
In our
case, Au–Cu alloy favors CO
desorption from the Au atom
and boosts the formation of
hydrogenation species on the
Cu atom. This might be the
main reason why the Au–Cu
alloy effect works for hydro-
carbon species but not for the
CO formation. The inferior
activity of Cu@STO/TiO2
nanotube arrays is probably
caused by the low electron-
[
10,12–14]
donating ability of Cu,
which results in a lower elec-
tron density in the Cu@STO/
TiO nanotube arrays than in
2
the Au@STO/TiO2 nanotube
arrays.
Figure 3D shows the CH4
evolution rate when the refer-
ence experiments were car-
ried out. When the experi-
ment was performed in the
absence of N H ·H O, cata-
2
4
2
lyst, or light illumination,
only a negligible amount of
CH4 was detected. Because
there is trace CH4 in the
natural air, the CH4 in the
Figure 3. A) Gas-evolution rates as a function of illumination time. B) Cycling measurements of CH4
generation on Au Cu@STO/TiO nanotube arrays, each cycle is shown with a different symbol. C) Photo-
3
2
catalytic reduction products formed on various catalysts. D) CH evolution in reference experiments. Inset:
4
1
3
Gas chromatogram and mass spectrum of CH (m/z 17, 16, and 15) produced over Au Cu@STO/TiO
4
3
2
nanotube arrays.
above-mentioned
experi-
ments is considered as con-
tamination from the air during
[10,13]
the single Au and Cu NPs.
Moreover, the alloy NPs can
samplings. Nevertheless, when the diluted CO (33.3% in Ar)
2
effectively restrain the reverse transportation of the accumu-
lated photogenerated electrons to the semiconductors.
was replaced by pure Ar gas, a small amount of CH was
4
[18]
found which should be generated from the photoreduction of
[24]
Hence, the Au–Cu alloy co-catalysts exhibit much higher
activity than Au and Cu. The increase in the fraction of Au in
bimetallic alloy NPs enhances the hydrocarbons production.
However, unlike the hydrocarbons, the CO production
increasing with the Au fraction does not follow the alloy
the remaining CO on the sample surface. In a normal-
2
1
3
condition experiment using labeled CO as the substrate
2
1
3
lead to the formation of CH (Inset in Figure 3D), showing
4
that the CH formed is indeed the product of CO reduction.
4
2
Consequently, all the above-mentioned reference experi-
ments confirm that the N H ·H O offers H atoms, CO
effect. According to previous report, CO is first deoxygen-
2
2
4
2
2
ated to form CO, and then gradually reduced to CH and
supplies as carbon source, and the photocatalyst provides
the redox potentials for the whole reaction to finally produce
4
[4a]
other hydrocarbon fuels.
Since the activation energy
associated with the CO desorption is much lower on Au
CO, CH , and other hydrocarbons.
4
À1 [19]
À1 [20]
(
E = 38 kJmol )
than that on Cu (E = 67 kJmol ),
To investigate the mechanism of the CO photoreduction,
a
a
2
desorption of CO from the Au surface is kinetically favored.
So the Au–Cu alloy co-catalysts with higher Au fraction show
higher selectivity for CO formation. Considering the work
in situ FTIR experiments were carried out. As shown in
À1
Figure 4, a series adsorption bands around 1330–1590 cm
À1
and 1600–1890 cm appear under the UV/Vis irradiation,
[
21]
function of Cu (4.65 eV) is less than Au (5.1 eV),
the
and their intensities increase with increasing irradiation time.
À1
photogenerated electrons prefer to move from the semi-
The characteristic adsorption bands 1335–1560 cm and
À1
conductors (STO, TiO ) to the Au atoms through the Schottky
1610–1750 cm are due to the bending vibration of CÀH,
2
[
22]
contact.
The interfacial charge transfer of the electrons
the stretching vibration of C=O and the asymmetric stretch-
ing of OÀC=O, which belong to the intermediate products,
such as aldehydes, carboxylic acids, and bidentate carbo-
from Au atoms to the electron acceptors (such as CO , CO) is
2
much slower than the charge transfer between the inner,
[18]
[25]
À1
homogeneously mixed alloy, metal atoms. These accumu-
lated electrons on the Au atoms may rapidly transfer to the
neighboring Cu atoms. Generally, Cu has been widely shown
to be a highly active catalyst for CO reduction to form
nates. The weak band at 1339 cm is from the bending
.
[25]
vibration of CÀH in CH4 Since it is difficult to adsorb CH4
molecules on the surface of the sample, the intensity of this
adsorption band is rather low. According to the in situ FTIR
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Keywords: Au-Cu cocatalyst · CO conversion ·
2
.
nanotube arrays · photocatalysis · solar fuel
[
1] S. C. Roy, O. K. Varghese, M. Paulose, C. A. Grimes, Acs Nano
010, 4, 1259 – 1278.
2
[
2] X. Zhang, F. Han, B. Shi, S. Farsinezhad, G. P. Dechaine, K.
Shankar, Angew. Chem. Int. Ed. 2012, 51, 11778 – 11782; Angew.
Chem. 2012, 124, 11948 – 11952.
[
3] M. Halmann, Nature 1978, 275, 115 – 116.
[
4] a) S. N. Habisreutinger, L. Schmidt-Mende, J. K. Stolarczyk,
Angew. Chem. Int. Ed. 2013, 52, 7372 – 7408; Angew. Chem.
2013, 125, 7516 – 7557; b) K. F. Li, X. Q. An, K. H. Park, M.
Khraisheh, J. W. Tang, Catal. Today 2014, 224, 3 – 12.
5] Y. Izumi, Coord. Chem. Rev. 2013, 257, 171 – 186.
[
Figure 4. In situ FTIR spectra for CO photoreduction on
[6] J. Zhang, J. H. Bang, C. Tang, P. V. Kamat, ACS nano 2009, 4,
387 – 395.
2
Au Cu@STO/TiO nanotube arrays under full-arc Xe-lamp irradiation
3
2
for different time.
[7] a) X. Wang, Z. Li, J. Shi, Y. Yu, Chem. Rev. 2014, 114, 9346 –
9384; b) Y. Li, X. Y. Yang, Y. Feng, Z. Y. Yuan, B. L. Su, Crit.
Rev. Solid State Mater. Sci. 2012, 37, 1 – 74.
[
8] a) X. Feng, J. D. Sloppy, T. J. LaTempa, M. Paulose, S. Komar-
neni, N. Bao, C. A. Grimes, J. Mater. Chem. 2011, 21, 13429 –
13433; b) O. K. Varghese, M. Paulose, T. J. LaTempa, C. A.
Grimes, Nano Lett. 2009, 9, 731 – 737.
[
9] R. M. Rioux, R. Komor, H. Song, J. D. Hoefelmeyer, M. Grass,
K. Niesz, P. D. Yang, G. A. Somorjai, J. Catal. 2008, 254, 1 – 11.
[
10] Y. Sugano, Y. Shiraishi, D. Tsukamoto, S. Ichikawa, S. Tanaka, T.
Hirai, Angew. Chem. Int. Ed. 2013, 52, 5295 – 5299; Angew.
Chem. 2013, 125, 5403 – 5407.
[
[
[
[
[
11] R. K. de Richter, T. Z. Ming, S. Caillol, Renewable Sustainable
Energy Rev. 2013, 19, 82 – 106.
12] N. E. Motl, E. Ewusi-Annan, I. T. Sines, L. Jensen, R. E. Schaak,
J. Phys. Chem. C 2010, 114, 19263 – 19269.
13] W. Li, A. Wang, X. Liu, T. Zhang, Appl. Catal. A 2012, 433, 146 –
1
51.
14] X. Liu, A. Wang, L. Li, T. Zhang, C. Y. Mou, J. F. Lee, J. Catal.
011, 278, 288 – 296.
2
15] a) L. He, Y. Huang, A. Wang, X. Wang, X. Chen, J. J. Delgado, T.
Zhang, Angew. Chem. Int. Ed. 2012, 51, 3699 – 3702; Angew.
Chem. 2012, 124, 3759 – 3762; b) H. L. Jiang, S. K. Singh, J. M.
Yan, X. B. Zhang, Q. Xu, ChemSusChem 2010, 3, 541 – 549.
Scheme 1. The proposed mechanism for the photoreduction of CO .
ads=adsorbed, h =hole.
2
+
[16] Q. Zhai, S. Xie, W. Fan, Q. Zhang, Y. Wang, W. Deng, Y. Wang,
Angew. Chem. Int. Ed. 2013, 52, 5776 – 5779; Angew. Chem.
2
013, 125, 5888 – 5891.
spectra results and the breaking of CÀO bonds and creating of
new CÀH bonds they imply, we proposed the mechanism for
[
17] F. Sastre, A. Puga, L. Liu, A. Corma, H. Garcꢁa, J. Am. Chem.
Soc. 2014, 136, 6798.
18] R. Su, R. Tiruvalam, A. J. Logsdail, Q. He, C. A. Downing, M. T.
Jensen, N. Dimitratos, L. Kesavan, P. P. Wells, R. Bechstein, ACS
Nano 2014, 8, 3490 – 3497.
the CO photoreduction, as shown in Scheme 1, which is in
2
[
[
4a]
agreement with the glyoxal pathway.
In summary, the results reported herein highlight the
exploration of easily prepared STO/TiO coaxial nanotube
arrays. High CO and hydrocarbon generation rates were
achieved by loading Au–Cu bimetallic alloy NPs onto these
[19] J. Gottfried, K. Schmidt, S. Schroeder, K. Christmann, Surf. Sci.
003, 536, 206 – 224.
20] I. Bçnicke, W. Kirstein, S. Spinzig, F. Thieme, Surf. Sci. 1994, 313,
31 – 238.
21] S. Trasatti, J. Electroanal. Chem. 1972, 39, 163 – 184.
[22] R. Marschall, Adv. Funct. Mater. 2014, 24, 2421 – 2440.
[23] M. D. Porosoff, J. G. G. Chen, J. Catal. 2013, 301, 30 – 37.
[24] a) N. M. Dimitrijevic, B. K. Vijayan, O. G. Poluektov, T. Rajh,
K. A. Gray, H. He, P. Zapol, J. Am. Chem. Soc. 2011, 133, 3964 –
2
2
[
[
2
STO/TiO coaxial nanotube arrays as effective catalysts for
2
the photoreduction of CO . By varying the fraction of one
2
component in this bimetallic alloy system, a Au Cu@STO/
3
TiO nanotube array has been found that is the most reactive
2
photocatalyst in this family to generate hydrocarbons from
diluted CO . N H ·H O, as the H source and electron donor,
3971; b) H. Zhou, J. Guo, P. Li, T. Fan, D. Zhang, J. Ye, Sci. Rep.
2
2
4
2
2013, 3, 1 – 9.
provides a reductive atmosphere for maintaining the alloying
effect. This opens a feasible route to enhance the photo-
catalytic efficiency, which also aids the development of
photocatalysts and co-catalysts.
[
25] T. Wang, X. Meng, P. Li, S. Ouyang, K. Chang, G. Liu, Z. Mei, J.
Ye, Nano Energy 2014, 9, 50 – 60.
Received: September 17, 2014
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
CO Photoreduction
2
Q. Kang, T. Wang, P. Li, L. Liu, K. Chang,
M. Li, J. Ye*
&&&& — &&&&
Photocatalytic Reduction of Carbon
Dioxide by Hydrous Hydrazine over Au–
Cu Alloy Nanoparticles Supported on
SrTiO /TiO Coaxial Nanotube Arrays
3
2
Alloy, alloy: Au–Cu alloy nanoparticles
NPs) supported on SrTiO /TiO coaxial
nanotube arrays are an efficient photo-
a reductive atmosphere for stabilizing the
(
alloy NPs during the reaction. A CO yield
3
2
À1 À1
of 3.77 mmolg
h
and total hydrocar-
À1 À1
catalyst for the conversion of CO into CO
bon of 725.4 mmolg
h
was realized on
2
and hydrocarbons. N H ·H O provides
Au Cu@SrTiO /TiO nanotube arrays.
3 3 2
2
4
2
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!