Photocatalytic reduction of CO2 by CuxO nanocluster loaded SrTiO3 nanorod thin film

Article abstract of DOI:10.1016/j.cplett.2016.06.062

Photocatalytic carbon dioxide (CO₂) conversion into carbon monoxide (CO) using H₂O as an electron donor was achieved by the strontium titanate (SrTiO₃: STO) nanorod thin films loaded with amorphous copper oxide (Cu_xO) nanoclusters. The loading of the Cu_xO-cocatalysts onto STO nanorods clearly improved the photocatalytic activity compared to bare STO nanorods. The Cu_xO-cocatalysts are composed of abundant and non-toxic elements, and can be loaded by using a simple and economical method. Our findings demonstrate that Cu_xO nanoclusters function as a general cocatalyst and can be used in combination with various semiconductors to construct low-cost and efficient photocatalytic CO₂ reduction systems.

Full text of DOI:10.1016/j.cplett.2016.06.062

Accepted Manuscript
Photocatalytic Reduction of CO2 by CuxO Nanocluster loaded SrTiO3 Nanorod
Thin Film
Shusaku Shoji, Ge Yin, Masami Nishikawa, Daiki Atarashi, Etsuo Sakai,
Masahiro Miyauchi
PII:
S0009-2614(16)30457-2
DOI:
http://dx.doi.org/10.1016/j.cplett.2016.06.062
CPLETT 33971
Reference:
Chemical Physics Letters
To appear in:
Received Date:
Accepted Date:
22 May 2016
22 June 2016
Please cite this article as: S. Shoji, G. Yin, M. Nishikawa, D. Atarashi, E. Sakai, M. Miyauchi, Photocatalytic
Reduction of CO2 by CuxO Nanocluster loaded SrTiO3 Nanorod Thin Film, Chemical Physics Letters (2016), doi:
http://dx.doi.org/10.1016/j.cplett.2016.06.062
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Shoji Page 1
Photocatalytic Reduction of CO2 by CuxO Nanocluster loaded SrTiO3
Nanorod Thin Film
Shusaku Shoji, † Ge Yin, † Masami Nishikawa, ‡ Daiki Atarashi, † Etsuo Sakai, † Masahiro
Miyauchi†*
† Department of Materials Science and Engineering, School of Materials and Chemical
Technology, Tokyo Institute of Technology, 2-12-1 Ookayama, Meguro-ku, Tokyo 152-8552,
Japan
‡ Department of Materials Science and Technology, Nagaoka University of Technology, 1603-1,
Kamitomioka-machi, Nagaoka, Niigata 940-2188, Japan
*Corresponding author.
Prof. Masahiro Miyauchi
Email: mmiyauchi@ceram.titech.ac.jp
Tel:+81-3-5734-2527
[Insert Running title of <72 characters]

[First Authors Last Name] Page 2
Abstract
Photocatalytic carbon dioxide (CO2) conversion into carbon monoxide (CO) using H2O as an
electron donor was achieved by the strontium titanate (SrTiO3: STO) nanorod thin films loaded
with amorphous copper oxide (CuxO) nanoclusters. The loading of the CuxO-cocatalysts onto
STO nanorods clearly improved the photocatalytic activity compared to bare STO nanorods. The
CuxO-cocatalysts are composed of abundant and non-toxic elements, and can be loaded by using
a simple and economical method. Our findings demonstrate that CuxO nanoclusters function as a
general cocatalyst and can be used in combination with various semiconductors to construct low-
cost and efficient photocatalytic CO2 reduction systems.
Keywords: photocatalyst, nanorods, cocatalyst, CuxO nanocluster, Impregnation method, CO2
reduction, ubiquitous elements, ESR analysis, Isotope labeling analysis.
[Insert Running title of <72 characters]

[First Authors Last Name] Page 3
Introduction
Photocatalytic carbon dioxide reduction, in which CO2 is converted into organic compounds,
such as methane, methanol, formic acid, and formaldehyde, has the potential to greatly contribute
to reducing global warming and fossil fuel dependence.[1,2] In a photocatalytic CO2 reduction
system, photoexcited electrons in the conduction band reduce CO2, while photogenerated holes in
the valence band oxidize the electron donor. Since Halmann[3] reported that p-type gallium
phosphide functions as a photocatalyst for CO2 reduction into methanol, formic acid, or
formaldehyde, various semiconductor photocatalysts have been studied, including TiO2, ZnO,
CdS, GaP, SiC,[4] Zn-doped Ga2O3,[5] and SrTiO3.[6] In artificial photosynthesis systems
designed for the reduction of CO2 for solar fuel production, water molecules should ideally serve
as the electron donor, similar to photosystems I and II in natural plants. Further, the CO2 reduction
reaction competes with the reduction of protons to produce molecular hydrogen. However,
because the redox potential of H+/H2 (-0.41 V vs NHE at pH 7) is lower than that of CO/CO2 (-
0.53 V vs NHE at pH 7)[2], the proton reduction reaction is thermodynamically favored.
Therefore, increasing the selectivity of CO2 reduction products over hydrogen generation is
critical for the development of efficient photocatalytic systems for CO2 reduction.
One approach for improving the products selectivity of semiconductor photocatalysts is the
loading of cocatalysts to drive oxidation or reduction selectivity. For example, Sato et al.[7,8]
efficiently reduced CO2 using a combination of ruthenium complex cocatalyst and either N-doped
Ta2O5, GaP or InP semiconductors. In addition to complex cocatalysts, metal nanoparticles also
exhibit high selectivity for CO2 reduction, as demonstrated for silver (Ag)-loaded ALa4Ti4O15 (A=
Ca, Sr, or Ba) and Ag cluster-loaded Ga2O3.[9] Au-Cu alloy nanoparticles were also reported to
function as a cocatalyst when loaded onto SrTiO3/TiO2 photocatalyst.[10] In contrast to noble
metal organic complexes and particles, copper and copper oxide are ubiquitous in the
environment and are nontoxic. Further, copper compounds have balanced chemisorption-
[Insert Running title of <72 characters]

[First Authors Last Name] Page 4
desorption properties for CO2 and CO, [11,12] and therefore have high CO2 reduction
activity[13,14] and high selectivity of products, such as carbon monoxide and formic acid
(HCOOH).[15,16] In addition to the requirement for high CO2/CO chemisorption-desorption,
cocatalysts must accumulate photogenerated electrons at specific reaction sites to drive efficient
CO2 photoreduction, as CO2 reduction proceeds via a multi-electron process. For example, two
electrons are required for CO and HCOOH production, and six electrons are needed for CH3OH
generation. Recently, our group reported that a copper oxide nanocluster cocatalyst drives the
multi-electron reduction of both oxygen to produce hydrogen peroxide[17,18] and CO2 to
generate CO.[19] For copper oxide cocatalysts to drive CO2 photoreduction, light-harvesting
semiconductors with a high conduction band, such as niobate nanosheets, are required, as the
photoreduction of CO2 requires excited electrons with a higher potential energy than that needed
for hydrogen production.[19] During the CO2 reduction reaction, copper oxide consists of Cu(I)
and Cu(II) valence states and is designated as CuxO (x=1 or 2).
Herein, we examined the potential of strontium titanate (SrTiO3: STO) to function as a light
harvester with loading of CuxO cocatalysts, as the STO has high chemical stability and sufficient
negative conduction band for CO2 reduction.[20] In addition, compared to reported light
harvesters, including niobate materials, STO has a relatively narrow band gap and unique crystal
structure. As the band gap of STO (3.2 eV) is narrower than that of niobate (3.56 eV), STO is
expected more effectively utilize solar energy than niobate. Further, the crystal phase of STO is a
perovskite structure (ABO3), which enables doping and the maintenance of charge balance by co-
doping into A and B sites.[21,22] In the present research, we used STO nanorod thin films[23] as
a UV-light harvester due to its high crystallinity and increased surface area for the loading
-of
CuxO cocatalyst. The photocatalytic CO2 reduction activity of CuxO cocatalyst loaded STO
nanorod (CuxO-STO) thin films was evaluated under UV-light irradiation in the present study. In
addition, excited species in CuxO-STO were investigated by electron spin resonance (ESR)
[Insert Running title of <72 characters]

[First Authors Last Name] Page 5
analysis to examine the charge separation process. Further, isotope labeling experiments with
18
13CO2 and H2 O isotopes were used to confirm that the generated CO originated from CO2 and
that the electron donor was water.
Materials and Methods
Synthesis of STO nanorod thin films
STO nanorod thin films were synthesized by the annealing of strontium ion-exchanged
titanate nanotube thin films.[23] Titanate nanotube thin film was prepared by the hydrothermal
treatment of titanium (Ti) foil (220.1 cm, Nilaco Co.).[24] Briefly, Ti foil was immersed in 30
mL of 10 M aqueous NaOH solution and allowed to react in a 100-mL Teflon-lined autoclave at
413 K for 3 h. After cooling to room temperature, the titanate nanotube thin film was washed
with distilled water and immersed in a 0.1 M HNO3 aqueous solution for 1 day to protonate the
interlayer of the titanate nanotube thin film. The protonated film was immersed in a saturated
aqueous solution of 2 M strontium acetate for 1 day to exchange the protons in the interlayer of
the titanate nanotubes with strontium ions (Sr2+). The ion-exchanged sample was then annealed
at 923 K for 2 h in air. The samples were washed with a 0.1 M HNO3 aqueous solution to
remove strontium carbonate from the nanotube thin film surface. For the measurement of optical
absorption of STO nanorod excluding the signal from metal Ti substrate, powder form of STO
nanorod was also synthesized with the same manner to the fabrication of STO nanorod thin film.
Titanate nanotube powder was prepared by the hydrothermal treatment of TiO2 (P-25) powder in
10 M NaOH aqueous solution for 1 day.
Loading of CuxO nanocluster cocatalyst onto STO nanorod thin films
[Insert Running title of <72 characters]

[First Authors Last Name] Page 6
CuxO nanocluster cocatalyst was loaded onto the STO nanorod thin films using an
impregnation method.[25] Briefly, the STO nanorod thin film was immersed into 60 mL of a
CuCl2 aqueous solution in a glass beaker. The CuxO nanoclusters were then loaded onto the STO
films through initiation of a hydrolysis reaction in the aqueous medium at 363 K for 1 h. The
fraction weight of Cu was set to be 0 to 0.005 wt% against H2O. The synthesized CuxO-STO
nanorod thin films were washed with distilled water and dried at 378 K in air. CuxO nanoclusters
were also loaded onto the STO nanorod powder by the similar manner.
Characterization
Crystal structures of the obtained samples were determined by X-ray diffraction (XRD;
Rigaku, SmartLab) using Cu-kα X-rays (λ=1.5418 Å). The incident angle was fixed at 1.2
degrees and the 2 scan angle ranged from 5 to 60 degrees. The sample morphologies were
observed by scanning electron microscopy (SEM; JSM-7500F, JEOL, Ltd.) and transmission
electron microscopy (TEM; JEM-2010F, JEOL, Ltd.). Chemical compositions of the sample
were investigated using an energy dispersive X-ray spectroscope (EDX: PHI5000) equipped with
a TEM apparatus. The valence number of copper nanoclusters was determined by measuring
electron spin resonance (ESR; EMX Plus, Bruker) under an N2 atmosphere. The ESR results
were recorded as differential spectra between the bare STO nanorod and CuxO-STO thin films.
X-ray photoelectron spectroscopy (XPS; model ESCA-5500MT, Perkin Elmer Instruments,
Japan) analysis was conducted to investigate the Cu-2p core level using standard Mg Kα X-rays.
Photocatalytic CO2 reduction
[Insert Running title of <72 characters]

[First Authors Last Name] Page 7
The prepared CuxO-STO film was added to an electrolyte solution (20 mL) consisting of 0.5
M KHCO3 and adjusted to pH=12 using aqueous NaOH, in a sealed quartz glass reactor (500
mL). To purge air in the reactor, CO2 or Ar gasses were bubbled into the reactor through a
Teflon tube at a flow rate of 100 mL/min for 30 min. Carbonic acid ions were formed in an
aqueous media by the addition of KHCO3 and bubbling of CO2. In the case of Ar bubbling
conditions, 0.5 M KCl aqueous solution, which was adjusted to pH= 12, was used as the
electrolyte. Products in the headspace of the reactor were evaluated using a gas chromatograph
(flame ionization detector (FID)) equipped with a methanizer unit under UV light irradiation
(Hg-Xe lamp, LA-410UV; Hayashi Watch Works, Ltd.). An isotope experiment was also
conducted in a quartz glass cell (115 cm) to verify that the generated O2 and CO originated
from the photocatalytic reduction of CO2 with water as an electron donor. A 0.52.00.1 cm
section of CuxO-STO film was immersed into a 0.5 M KH13CO3 (Isotec-Sigma-Aldrich) aqueous
18
H2 O (Isotec-Sigma-Aldrich) solution in the quartz glass cell. The headspace and electrolyte
were saturated with 13CO2 (Isotec-Sigma-Aldrich) gas for 5 min bubbling, and the cell was then
irradiated with UV light. Products in the headspace of the cell were evaluated by gas
chromatography-mass spectrometry (GC-MS; Shimadzu). Before performing the photocatalytic
CO2 reduction measurements, samples were irradiated under UV light in air to remove surface
organic contaminants.
Results and Discussion
Structural and morphological properties
Figure 1 shows the XRD patterns of the prepared CuxO-STO thin films. Hydrothermal
treatment of Ti foil converted the polycrystalline metal titanium to a layered hydrogen titanate
[Insert Running title of <72 characters]

[First Authors Last Name] Page 8
crystal structure.[24] After the titanate film was immersed in a saturated aqueous solution of
strontium acetate and annealed in air, the film adopted the perovskite structure of STO.
Annealing treatment on the Sr2+ ion-exchanged titanate nanotubes generated STO, as perovskite-
type STO is composed of corner-sheared TiO6 octahedrons, which have a similar lattice structure
as the edges of titanate nanosheets.
We measured the optical absorption properties of CuxO-STO nanorod and STO nanorod
powders, which were prepared by the same manner with the thin film fabrication, to exclude the
optical signal from metal titanium substrate. The UV-Vis spectra recorded by a diffuse
reflectance method are shown in our supporting information (Figure S1). The band-to-band
transition of the STO nanorod started around 380 nm, and the absorption at 430 nm of CuxO-
STO nanorod is attributable to interfacial charge transfer (IFCT).[26] The broad absorption
around 580-800 nm is assigned to the d-d transition of the Cu(II) species.[27] The bandgap of the
STO nanorod was estimated to be 3.2 eV, as calculated from the intersection of the plotted line
of square root of absorption coefficient versus photon energy with the abscissa axis.
Figure 2 shows SEM and TEM images of the STO nanorod and CuxO-STO nanorod thin
films. The SEM image revealed that the film surface was porous and numerous rods were
uniformly covered on a titanium foil substrate. The TEM image revealed that amorphous
nanoclusters of 2-3 nm in size were attached and highly dispersed on the STO nanorods. Energy
dispersive X-ray (EDX) spectroscopy analysis was conducted at the point on an STO nanorod
that was considered to be a CuxO nanocluster cocatalyst (Figure 2b). And as a result of the EDX
spectrum, a signal corresponding to Cu was clearly observed at the position of small nanoclusters
(Figure 2c). The EDX signal of molybdenum (Mo) was attributed to a contaminant from the
sample grid holder.
[Insert Running title of <72 characters]

[First Authors Last Name] Page 9
To determine the chemical state of the CuxO nanocluster cocatalyst and the behaviour of the
photo-excited electrons, ESR was measured before and after UV irradiation of the CuxO-STO
film (Figure 3). We also measured the ESR spectrum of bare STO nanorod thin film, and did not
detect any obvious peaks corresponding to copper ions or any changes to the spectrum after UV
irradiation. In contrast, the CuxO-STO film exhibited a broad ESR peak, which was assigned to
Cu2+ species, around 345 mT under dark conditions, and this peak decreased in intensity after
UV irradiation (Figure 3). As ESR detects unpaired electrons, the decrease of the Cu2+ peak
under UV irradiation indicates that the valence number of the Cu2+ species, which contain nine
electrons in the d-orbital, partially changed to Cu+, which is inactive for ESR. Thus, under UV-
light irradiation, photo-excited electrons of STO were injected into the CuxO nanocluster
cocatalyst. In the ESR analysis, trapped holes were not detected. It is assumed that the holes in
the trapping site were inhomogeneous and that the ESR signal became broad due to its nanorod
structure. We also measured XPS spectrum for the CuxO-STO film (supporting information,
Figure S2). Two peaks of Cu-2p core level were observed, whereas these peaks were not
appeared on the bare STO nanorod film. The peak at 933 eV corresponds to Cu 2p3/2, while that
at 953 eV corresponds to Cu 2p1/2.[28] If the sample contained Cu2+ species, satellite peaks
would be appeared at 942 and 962 eV. [29-31] However, these satellite peaks were not observed
in our sample, thus the copper species mainly consist of Cu1+ species, resulting from the
bombardment effect by X-ray irradiation under high vacuum.[17] Our previous X-ray absorption
near edge structure (XANES) spectra revealed the as-prepared CuxO nanocluster mainly
composed of Cu2+ species,[17] thus the CuxO nanoclusters in the present study also consist of
amorphous CuO at the as-prepared condition.
[Insert Running title of <72 characters]

[First Authors Last Name] Page 10
Photocatalytic CO2 reduction
Figure 4 shows the time course of photocatalytic CO generation for bare STO and CuxO-
STO film under UV irradiation with Ar or CO2 bubbling conditions. The amount of CuxO in the
film was optimized to be 0.0005 wt%, as described below. The CO generation rate was highest
for CuxO-STO under CO2 bubbling conditions and was markedly reduced under Ar bubbling
conditions, whereas that of bare STO was limited under both CO2 and Ar bubbling conditions.
The small amount of CO production detected under Ar bubbling was attributed to the oxidation
of organic contaminants adsorbed on the film surface, even though the surface of the samples
was cleaned by pre-irradiation treatment in air. To prove the photocatalytic CO generation under
CO2 bubbling condition, we have conducted isotope trace experiment as described later. We also
evaluated the hydrogen (H2) production under UV irradiation (see our supporting information,
Figure S3). Although small amount of H2 was detected, main product by our photocatalyst was
CO, indicating its high selectivity for CO2 photoreduction. It is noteworthy that the loading of
CuxO onto STO exhibited 20 % improvement of selectivity for CO2 photoreduction.
Figure 5 shows the influence of the amount of loaded CuxO nanocluster cocatalyst on the
photocatalytic CO generation activity of STO nanorod thin films. Based on the analysis, the
optimum amount of CuxO for efficient CO production was determined to be 0.0005 wt% in the
solution used for CuxO loading. Below optimum condition, the amount of CuxO is insufficient,
while those are overloaded above optimum condition. The trend of photocatalytic activity versus
the amount of CuxO nanoclusters will be explained in our discussion section below.
We next conducted isotope tracer analysis on the optimized CuxO-STO film (Figure 6). The
peaks in the mass chromatography spectra at m/z = 29 and 36 were assigned to 13CO and 18O2,
respectively. These results confirmed that the CO generated by the CuxO-STO film originated
[Insert Running title of <72 characters]

[First Authors Last Name] Page 11
from CO2, and that O2 molecules were produced through water oxidation. Similar to
photosystems I and II in natural plants, our CuxO-STO film system generates fuel from stable
molecules like CO2 and water in an uphill reaction under photon irradiation.
Discussion on photocatalytic process
Here, we discuss the cocatalytic effect of the CuxO nanoclusters for CO2 reduction.
According to previous electrochemical studies, copper elements effectively promoted CO2
electro-reduction under cathodic bias. For example, Kortlever et al.[32] reported that CO2
reduction proceeded by two alternate pathways, one involving the generation of formate species
and the other leading to the production of CO species, ultimately resulting in the formation of
various final products, including methane and hydrocarbon. We monitored formic acid
production in the electrolyte solution, and found that only a negligible amount of formic acid was
produced. Further, no significant differences in formic acid production were detected between
the CO2 and Ar bubbling conditions. Azuma et al.[14] reported that copper catalyzed the
reduction of CO2 molecules to CO species rather than formate. This finding is consistent with
our present results that the CuxO nanoclusters have high selectivity for CO generation and do not
promote hydrogen or formate formation.
Next, we discuss the photoexcitation process of the CuxO-STO film. There are three
expected excitation processes as follows,
1. interband transition in STO and electron injection from the conduction band of STO to Cu2+
2. d-d transition in Cu2+ species
3. interfacial charge transfer (IFCT) from the valence band of STO valence band to Cu2+ (Inter
facial charge transfer [18]
[Insert Running title of <72 characters]

[First Authors Last Name] Page 12
The bandgap of STO was 3.2 eV and its optical absorption was appeared under 380 nm as
shown in our supporting information (Figure S1). On the other hand, the absorption of Cu2+ by d-
d transition was appeared above 580 nm, while that by IFCT was appeared around 400 to 430 nm
(Figure S1). We have investigated visible light activity under a Xe lamp with a UV cutoff filter,
which could excite both d-d transition and IFCT. However, any CO was not produced under
visible light irradiation. These results indicate that the d-d transition and IFCT do not contribute
to the CO2 photoreduction. Therefore, our photocatalytic CO2 reduction is driven by the
interband transition of STO and following electron injection from the conduction band of STO
into CuxO clusters, which act as reactive sites for CO2 reduction. Also, our ESR analysis under
UV irradiation indicated that electrons are injected from the conduction band of STO into the
CuxO nanoclusters.
Although the conduction band of STO is higher than the redox potential of CO2/CO, the
redox potential of Cu(II)/Cu(I) (+0.16 V vs NHE at pH 0)[25,33] is lower than that of CO2/CO.
Despite the lower redox potential of Cu(II)/Cu(I), our experimental results revealed that CO2 was
reduced to CO by the UV irradiation of CuxO-STO film. The efficiency of the CO2
photoreduction reaction is due to a negative shift of the Fermi level of the CuxO nanocluster
resulting from the injection of excited electrons from the conduction band of STO. Therefore, the
mixed valence states of the Cu species in CuxO was critical for the ability of this cocatalyst to
promote CO2 reduction. Yin et al.[19] reported that the initial chemical composition of CuxO
clusters loaded onto niobate nanosheets was CuO, and that a proportion of the Cu(II) species in
the clusters were converted to Cu(I) during the initial 3 h of UV-light irradiation. After 5 h of
irradiation, Cu(II) and Cu(I) were balanced and CO2 photoreduction was efficient and stable. As
[Insert Running title of <72 characters]

[First Authors Last Name] Page 13
observed in Figure 5, the size or amount of the CuxO nanoclusters also influenced the upshift in
the CuxO Fermi level following electron injection.
Conclusion
We have demonstrated the applicability of modifying SrTiO3 with CuxO nanoclusters as a
cocatalyst for CO2 photoreduction. The CuxO-STO film exhibited high activity and selectivity
for CO evolution under UV irradiation. The isotope tracer analysis confirmed that water served
as an electron donor and that CO molecules were produced by the reduction of CO2 under UV
irradiation. Similar to photosystems in natural plants, fuel molecules were successfully generated
by the CuxO-STO film from CO2 and water in response to photon irradiation. The key underlying
factors for the efficient multi-electron reduction of CO2 by this system are the high conduction
band of STO and the loading of highly dispersed CuxO clusters onto the STO surface. Further, as
the CuxO is composed of non-toxic and abundant elements, it has the potential to be an effective
cocatalyst onto various light-harvesting materials to achieve CO2 photo-reduction.
Acknowledgements
This research was supported by a grant from the Advanced Catalytic Transformation
Program for Carbon Utilization (ACT-C) of the Japan Science and Technology Agency (JST),
and JSPS KAKENHI Grant Number 26410234. We thank Mr. K. Hori at the Center for
Advanced Materials Analysis of the Tokyo Institute of Technology for helping with the TEM
characterization. We also thank Mr. H. Hara at the Bruker Corporation for measuring ESR
spectra.
[Insert Running title of <72 characters]

[First Authors Last Name] Page 14
ReferencesPrimary Sources
Secondary Sources
Uncategorized References
[1]
[2]
J. Hansen, L. Nazarenko, R. Ruedy, M. Sato, J. Willis, A. Del Genio, D. Koch, A. Lacis,
K. Lo, S. Menon, T. Novakov, J. Perlwitz, G. Russell, G.A. Schmidt, N. Tausnev,
Science 308 (2005) 1431.
S.N. Habisreutinger, L. Schmidt-Mende, J.K. Stolarczyk, Angewandte Chemie
(International ed. in English) 52 (2013) 7372.
[3]
[4]
[5]
M. Halmann, Nature 275 (1978) 115.
T. Inoue, A. Fujishima, S. Konishi, K. Honda, Nature 277 (1979) 637.
K. Teramura, Z. Wang, S. Hosokawa, Y. Sakata, T. Tanaka, Chemistry (Weinheim an der
Bergstrasse, Germany) 20 (2014) 9906.
[6]
[7]
M. Halmann, M. Ulman, B. Aurian-Blajeni, Solar Energy 31 (1983) 429.
S. Sato, T. Arai, T. Morikawa, K. Uemura, T.M. Suzuki, H. Tanaka, T. Kajino, Journal of
the American Chemical Society 133 (2011) 15240.
[8]
[9]
S. Sato, T. Morikawa, S. Saeki, T. Kajino, T. Motohiro, Angewandte Chemie
(International ed. in English) 49 (2010) 5101.
M. Yamamoto, T. Yoshida, N. Yamamoto, T. Nomoto, Y. Yamamoto, S. Yagi, H.
Yoshida, Journal of Materials Chemistry A 3 (2015) 16810.
[10] Ş. Neaţu, J.A. Maciá-Agulló, P. Concepción, H. Garcia, Journal of the American
Chemical Society 136 (2014) 15969.
[11] R.G. Copperthwaite, P.R. Davies, M.A. Morris, M.W. Roberts, R.A. Ryder, Catalysis
Letters 1 11.
[12] K.P. Kuhl, T. Hatsukade, E.R. Cave, D.N. Abram, J. Kibsgaard, T.F. Jaramillo, Journal
of the American Chemical Society 136 (2014) 14107.
[13] Y. Hori, A. Murata, R. Takahashi, Journal of the Chemical Society, Faraday Transactions
1: Physical Chemistry in Condensed Phases 85 (1989) 2309.
[14] M. Azuma, K. Hashimoto, M. Hiramoto, M. Watanabe, T. Sakata, Journal of The
Electrochemical Society 137 (1990) 1772.
[15] K. Hara, A. Kudo, T. Sakata, Journal of Electroanalytical Chemistry 391 (1995) 141.
[16] R. Reske, H. Mistry, F. Behafarid, B. Roldan Cuenya, P. Strasser, Journal of the
American Chemical Society 136 (2014) 6978.
[17] X. Qiu, M. Miyauchi, K. Sunada, M. Minoshima, M. Liu, Y. Lu, D. Li, Y. Shimodaira,
Y. Hosogi, Y. Kuroda, K. Hashimoto, ACS Nano 6 (2011) 1609.
[18] M. Miyauchi, H. Irie, M. Liu, X. Qiu, H. Yu, K. Sunada, K. Hashimoto, The journal of
physical chemistry letters 7 (2016) 75.
[19] G. Yin, M. Nishikawa, Y. Nosaka, N. Srinivasan, D. Atarashi, E. Sakai, M. Miyauchi,
ACS nano 9 (2015) 2111.
[20] D.E. Scaife, Solar Energy 25 (1980) 41.
[21] M. Miyauchi, M. Takashio, H. Tobimatsu, Langmuir 20 (2004) 232.
[Insert Running title of <72 characters]

[First Authors Last Name] Page 15
[22] R. Konta, T. Ishii, H. Kato, A. Kudo, The Journal of Physical Chemistry B 108 (2004)
8992.
[23] M. Miyauchi, The Journal of Physical Chemistry C 111 (2007) 12440.
[24] M. Miyauchi, H. Tokudome, Y. Toda, T. Kamiya, H. Hosono, Applied Physics Letters 89
(2006) 043114.
[25] M. Liu, X. Qiu, M. Miyauchi, K. Hashimoto, Chemistry of Materials 23 (2011) 5282.
[26] X. Qiu, M. Miyauchi, K. Sunada, M. Minoshima, M. Liu, Y. Lu, D. Li, Y. Shimodaira,
Y. Hosogi, Y. Kuroda, K. Hashimoto, ACS nano 6 (2012) 1609.
[27] N.L. Dias Filho, Microchimica Acta 130 233.
[28] V. Hayez, A. Franquet, A. Hubin, H. Terryn, Surf Interface Anal 36 (2004) 876.
[29] J.C. Otamiri, S.L.T. Andersson, A. Andersson, Applied Catalysis 65 (1990) 159.
[30] M.C. Biesinger, B.P. Payne, A.P. Grosvenor, L.W.M. Lau, A.R. Gerson, R.S.C. Smart,
Applied Surface Science 257 (2011) 2717.
[31] S. Poulston, P.M. Parlett, P. Stone, M. Bowker, Surf Interface Anal 24 (1996) 811.
[32] R. Kortlever, J. Shen, K.J.P. Schouten, F. Calle-Vallejo, M.T.M. Koper, The Journal of
Physical Chemistry Letters 6 (2015) 4073.
[33] H. Irie, S. Miura, K. Kamiya, K. Hashimoto, Chemical Physics Letters 457 (2008) 202.
Figure Captions
Figure. 1 XRD patterns of Ti foil (a), titanate nanotube thin film (b), and STO nanorod thin film
(c). Triangles, circles, and diamonds indicate metal titanium, layered hydrogen titanate, and
strontium titanate crystals, respectively.
Figure. 2 Structural analyses of CuxO-STO film. SEM image of the surface of STO nanorod thin
film observed on the surface (a). TEM image of a CuxO nanocluster cocatalyst-loaded STO
nanorod (b). Panel (c) shows the EDX spectrum at the point marked by a circle in the TEM
image (b).
Figure. 3 ESR spectra of the CuxO-STO film under dark and UV-irradiated conditions.
[Insert Running title of <72 characters]

[First Authors Last Name] Page 16
Figure. 4 Time course of photocatalytic carbon monoxide (CO) generation by CuxO-STO and
bare STO nanorod thin films.
Figure. 5 Relationship between photocatalytic CO generation rate and concentration of Cu2+
solution used for CuxO loading.
Figure. 6 Mass chromatography spectra of 13CO (m/z=29) and 18O2 (m/z=36) generated during
18
the photocatalytic reduction of 13CO2 by the optimized CuxO-STO film in the presence of H2 O.
[Insert Running title of <72 characters]

[First Authors Last Name] Page 17
Figure 1
[Insert Running title of <72 characters]

[First Authors Last Name] Page 18
Figure 2
[Insert Running title of <72 characters]

[First Authors Last Name] Page 19
Figure 3
[Insert Running title of <72 characters]

[First Authors Last Name] Page 20
Figure 4
[Insert Running title of <72 characters]

[First Authors Last Name] Page 21
Figure 5
[Insert Running title of <72 characters]

[First Authors Last Name] Page 22
Figure 6
[Insert Running title of <72 characters]

[First Authors Last Name] Page 23
Graphical abstract
H2O
O2
CuxO-SrTiO3
CO
CO2
[Insert Running title of <72 characters]

[First Authors Last Name] Page 24
Highlight
・CO2 photoreduction was achieved by the CuxO-cocatalysts loaded STO nanorod thin film.
・CuxO-cocatalysts were loaded by using a simple and economical impregnation method.
・Photocatalytic reaction mechanism was discussed by using electron spin resonance and
isotope labeling analysis.
[Insert Running title of <72 characters]