The KCaSrTa5O15 photocatalyst with tungsten bronze structure for water splitting and CO2 reduction

Article abstract of DOI:10.1039/c4cp03892d

KCaSrTa₅O₁₅ with tungsten bronze structure and a band gap of 4.1 eV showed activity for water splitting without cocatalysts. The activity was improved by loading the NiO cocatalyst. The apparent quantum yield of optimized NiO-loaded KCaSrTa₅O₁₅ was 2.3% at 254 nm for water splitting. When CO₂ gas was bubbled into the reactant aqueous solution, Ag cocatalyst-loaded KCaSrTa₅O₁₅ produced CO and H₂ as reduction products of CO₂ and H₂O, respectively, and O₂ as an oxidation product of H₂O. The carbon source of CO was confirmed to be CO₂ molecules by using ¹³CO₂. The ratio of the number of electrons to that of holes calculated from the amounts of products (CO, H₂ and O₂) was almost unity. Additionally, the ratio of the turnover number of electrons consumed for CO production to the total number of an Ag atom of the cocatalyst that was the active site for CO₂ reduction was 8.6 at 20 h. These results indicate that water was consumed as an electron donor for this photocatalytic CO₂ reduction in an aqueous medium. Thus, KCaSrTa₅O₁₅ with tungsten bronze structure has arisen as a new photocatalyst that is active for water splitting and CO₂ reduction utilizing water as an electron donor. This journal is

Full text of DOI:10.1039/c4cp03892d

PCCP
View Article Online
View Journal | View Issue
PAPER
The KCaSrTa5O15 photocatalyst with tungsten
bronze structure for water splitting and CO2
reduction†
Cite this: Phys.Chem.Chem.Phys.,
2014, 16, 24417
Tomoaki Takayama,a Kentaro Tanabe,a Kenji Saito,‡a Akihide Iwaseab and
Akihiko Kudo*ab
KCaSrTa5O15 with tungsten bronze structure and a band gap of 4.1 eV showed activity for water splitting
without cocatalysts. The activity was improved by loading the NiO cocatalyst. The apparent quantum
yield of optimized NiO-loaded KCaSrTa5O15 was 2.3% at 254 nm for water splitting. When CO2 gas was
bubbled into the reactant aqueous solution, Ag cocatalyst-loaded KCaSrTa5O15 produced CO and H2 as
reduction products of CO2 and H2O, respectively, and O2 as an oxidation product of H2O. The carbon
source of CO was confirmed to be CO2 molecules by using 13CO2. The ratio of the number of electrons
to that of holes calculated from the amounts of products (CO, H2 and O2) was almost unity. Additionally,
the ratio of the turnover number of electrons consumed for CO production to the total number of an
Ag atom of the cocatalyst that was the active site for CO2 reduction was 8.6 at 20 h. These results
indicate that water was consumed as an electron donor for this photocatalytic CO2 reduction in an
aqueous medium. Thus, KCaSrTa5O15 with tungsten bronze structure has arisen as a new photocatalyst
that is active for water splitting and CO2 reduction utilizing water as an electron donor.
Received 29th August 2014,
Accepted 1st October 2014
DOI: 10.1039/c4cp03892d
www.rsc.org/pccp
reported for the CO2 reduction to form HCOOH, CO, CH3OH and
CH4 in aqueous media without sacrificial reagents.9–16 However,
Introduction
An artificial photosynthesis system has been extensively oxygen evolution in a stoichiometric amount is not observed in
studied to develop systems for CO2 conversion to fuels and many cases. Among them, a ZrO2 photocatalyst produces CO and
chemicals. Photocatalytic CO2 reduction is one of the potential H2 as reduction products and O2 as an oxidation product in a
candidates for the artificial photosynthesis. Homogeneous and stoichiometric amount under UV light irradiation.14 Moreover,
heterogeneous photocatalyst systems for CO2 reduction have the selectivity for CO2 reduction to form CO are enhanced by
15
been studied. The homogeneous photocatalysts including loading a Cu cocatalyst on the ZrO2 photocatalyst. BaLa4Ti4O15
Re-complex and Re–Ru-complex require sacrificial reducing and Zn-doped Ga2O316 are highly active for CO2 reduction using
reagents such as TEOA (Triethanolamine) to reduce CO2 to water as an electron donor when Ag cocatalyst is employed.
CO and HCOOH.1–6 Although a heterogeneous CdS photo- Thus, metal oxide photocatalysts for water splitting can be
catalyst shows activity for CO2 reduction to form CO under applied to CO2 reduction using water as an electron donor if
visible light irradiation, sacrificial reagents are also indispensa- suitable cocatalysts are chosen.
ble.7,8 Heterogeneous metal oxide photocatalysts which possess
We have developed tantalum-based photocatalysts, such as
the ability for O2 evolution by oxidation of water have been NaTaO3 and NaTaO3:A (A = La and Sr) with perovskite structure,
and K2LnTa5O15 (Ln = La, Pr, Nd, Sm, Gd, Tb, Dy and Tm) with
tungsten bronze structure, for highly efficient water splitting
a Department of Applied Chemistry, Faculty of Science, Tokyo University of Science,
under UV light irradiation.17–19 The high activities for the
1-3 Kagurazaka, Shinjuku-ku, Tokyo, 162-8601, Japan.
tantalum-based photocatalysts are mainly due to their high
E-mail: a-kudo@rs.kagu.tus.ac.jp; Fax: +81-3-5261-4631
b Photocatalysis International Research Center, Research Institute for Science and
Technology, Tokyo University of Science, 2641 Yamazaki, Noda-shi, Chiba,
278-8510, Japan
conduction bands formed by Ta 5d orbitals. We have also
reported that a BaLa4Ti4O15 photocatalyst20 with two dimen-
sional anisotropy of crystal structure for water splitting also
shows activity for CO2 reduction using water as an electron
donor by loading a highly dispersed Ag cocatalyst.15 The
selectivity for the CO2 reduction is superior to that for water
reduction to form H2 even in an aqueous medium. On the other
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4cp03892d
‡ Present address: Office for Development of Young Researchers, Research
Planning and Promotion Division, Niigata University, 8050 Ikarashi 2-no-cho,
Nishi-ku, Niigata 950-2181, Japan.
This journal is ©the Owner Societies 2014
Phys. Chem. Chem. Phys., 2014, 16, 24417--24422 | 24417

View Article Online
Paper
PCCP
hand, there is a tantalate group that possesses tungsten bronze Photocatalytic water splitting and CO2 reduction
structure with anisotropy to a c-axis of the crystal structure and
Photocatalytic reactions were conducted using an inner irradia-
tion cell made of quartz with a 400 W high-pressure mercury
lamp. 0.5 g of photocatalyst powder was dispersed in 350 mL of
water. The Ar (99.99%) or CO2 (99.995%) gas was continuously
bubbled at a 30 mL minÀ1 flow rate during the photocatalytic
reaction. Gaseous products of H2, O2 and CO were determined
by GC (Shimadzu: GC-8A with Molecular Sieve 5A, TCD and Ar
carrier, and GC-8A with methanizer, Molecular Sieve 13X, FID
and N2 carrier). 13CO2 (purity: 99.5 atom%) was also employed
for an isotope experiment to confirm the carbon source for
photocatalytic CO2 reduction. 13CO was analyzed using a GC-MS
(Shimadzu: GC-MS Plus 2010, RESTEK: RT-Msieve 5A).
the framework consisting of TaO6 octahedra with corner shar-
ing being similar to the perovskite structure as seen in NaTaO3
of a highly efficient photocatalyst for water splitting. Therefore,
the tantalates with tungsten bronze structure are expected to be
active for water splitting and CO2 reduction. In the present
study, KCaSrTa5O15 with tungsten bronze structure21 was pre-
pared using a solid-state reaction, and their photocatalytic
activities for water splitting and CO2 reduction were investi-
gated. The photocatalyst particles and cocatalysts were char-
acterized using SEM, XPS and DRS.
Results and discussion
Characterization of KCaSrTa5O15
Experimental
Preparation of KCaSrTa5O15
The X-ray diffraction patterns of the materials prepared at
different calcination temperatures were assigned to KCaSr-
Ta5O15 (PDF: 40-351) (Fig. S1, ESI†). Calcination below 1573 K
gave small amounts of Ca2Ta2O7 (PDF: 53-743) and an unknown
compound as impurities (Fig. S1 (a)–(c), ESI†), while the single
phase of highly crystalline KCaSrTa5O15 was obtained by calcining
at 1773 K for 10 h (Fig. S1(d), ESI†). The band gap of KCaSrTa5O15
was estimated to be 4.1 eV from the absorption edges except for
the material prepared at 1173 K as shown in Fig. 1. Primary
particles of KCaSrTa5O15 with an average diameter of 200–300 nm
aggregated, when they were prepared below 1573 K as shown in
Fig. 2(a)–(c). In contrast, the morphology of KCaSrTa5O15 pre-
pared at 1773 K was a rod as shown in Fig. 2(d), reflecting the
tungsten bronze structure in which TaO6 octahedral units were
connected to each other along the c-axis.
KCaSrTa5O15 was prepared by a solid-state reaction as follows.
K2CO3 (Kanto Chemical: 99.0%), CaCO3 (Kanto Chemical: 99.5%),
SrCO3 (Kanto Chemical: 99.9%) and Ta2O5 (Rare Metallic: 99.99%)
were used as starting materials for the solid-state reaction. The
carbonates and oxide were mixed in a molar ratio of K : Ca : Sr : Ta =
1.05:1:1:5 in an alumina mortar. The excess amount of potas-
sium (5 mol%) was to compensate the volatilization.22 The mixed
powder was calcined in air at 1173 K for 1 h and subsequently at
1423 K for 10 h in a platinum crucible. The excess potassium was
washed out with water from the obtained powder.
Various cocatalysts were loaded by impregnation and photo-
deposition methods on the surface of the KCaSrTa5O15 photo-
catalyst. NiO and Ag cocatalysts were loaded by an impregnation
method. Photocatalyst powder was dispersed in aqueous solu-
tions to dissolve Ni(NO3)2 (Wako Pure Chemical: 98.5%) and
AgNO3 (Tanaka Rare Metal) in a porcelain crucible. The slurry
solution was stirred with a glass rod during evaporation using a
hot plate. The obtained powder was calcined in air at 575 K and
723 K for 1 h for loading the NiO and Ag cocatalysts, respectively.
The NiO and Ag cocatalysts were reduced with H2 at 773 K, if
necessary. Ag, Ni, Ru, Rh, Pt, Cu and Au cocatalysts were
photodeposited from aqueous solutions to dissolve suitable
amounts of AgNO3, Ni(NO3)2, RuCl3, RhCl3, H2PtCl6, Cu(NO3)2
and HAuCl4 in situ.
Effect of loading the cocatalyst on water splitting and CO2
reduction
Table 1 shows the activity for water splitting over KCaSrTa5O15
under Ar gas flow. All samples prepared at different tempera-
tures showed the activity for water splitting without a cocatalyst.
Moreover, the activities were drastically enhanced when the NiO
KCaSrTa5O15 powders prepared by a solid-state reaction at
different temperatures were examined by X-ray diffraction
using Cu Ka radiation (Rigaku: Miniflex). Diffuse reflectance
spectra of these powders were recorded using a UV-vis-NIR
spectrometer (Jasco: UbestV-570) and were converted from
reflection to absorption by the Kubelka–Munk method. Photo-
catalyst powders were observed using a scanning electron
microscope (JEOL: JSM-6700F). Surface species of cocatalysts
on photocatalysts were analyzed by X-ray photoelectron spectro-
scopy (Shimazdu: ESCA-3400; Mg anode). Metallic Ni (Nilaco:
99+%), NiO (Soekawa Chemical: 99.9%) and Ni(OH)2 (Wako
Pure Chemical: 95.0%) were employed as references for the XPS
measurements. Binding energies were corrected using C 1s
(285.0 eV) on a metallic Au foil (84.0 eV).23
Fig. 1 Diffuse reflectance spectra of KCaSrTa5O15 prepared at (a) 1173 K,
(b) 1423 K, (c) 1573 K and (d) 1773 K for 10 h.
24418 | Phys. Chem. Chem. Phys., 2014, 16, 24417--24422
This journal is © the Owner Societies 2014

View Article Online
PCCP
Paper
Fig. 2 SEM images of KCaSrTa5O15 prepared at (a) 1173 K, (b) 1423 K,
(c) 1573 K and (d) 1773 K for 10 h.
Fig. 3 Water splitting over NiO (0.5 wt%)-loaded KCaSrTa5O15 photo-
catalyst. Photocatalyst: 0.5 g, reactant solution: water (350 mL) with Ar gas
flow (30 mL minÀ1), light source: a 400 W high-pressure mercury lamp,
reactor: an inner irradiation cell made of quartz. KCaSrTa5O15 was pre-
pared by a solid-state reaction at 1423 K for 10 h. The NiO cocatalyst was
loaded by an impregnation method.
Table 1 Photocatalytic water splitting over KCaSrTa5O15 prepared at
different temperatures
Activity/mmol hÀ1
Preparation
temperature/K Cocatalyst Loading method
H2
O2
1173
1173
1423
1423
1423
1423
1573
1573
1773
1773
None
NiO
None
NiO
Ni
—
271
180
263
42
700
12
62
47
644
31
95
Table 2 Photocatalytic CO2 reduction over various cocatalysts-loaded
KCaSrTa5O15
Impregnation
—
Impregnation
Photodeposition
Impregnation + H2 red. 115
—
Impregnation
—
Impregnation
639
102
1339
25
Activity/mmol hÀ1
Ni
Cocatalyst
Loading condition
H2
O2
CO
None
NiO
None
NiO
106
1270
66
None
NiO
Ni
Cu
Ru
Rh
Ag
Ag
Ag
Pt
Au
—
116
764
111
216
28
43
53
55
96
62
584
48
398
58
100
13
19
37
28
48
23
0
0
0
Impregnationa
Photodeposition
Photodeposition
Photodeposition
Photodeposition
Photodeposition
Impregnationb
Impregnationb + H2 red.
Photodeposition
Photodeposition
212
Trace
Photocatalyst: 0.5 g, cocatalyst: 0.5 wt%, loading conditions: impreg-
nation (573 K for 1 h in air), impregnation and subsequent H2 reduction
(773 K for 2 h in H2 flow), photodeposition (in situ), reactant solution:
water (350 mL), light source: a 400 W high-pressure mercury lamp,
reactor: an inner irradiation cell made of quartz. KCaSrTa5O15 was
prepared by a solid-state reaction.
0
0
8.1
5.5
1.0
0
269
0
Photocatalyst: 0.5 g, cocatalysts: 0.5 wt%, loading conditions: photo-
deposition (in situ), impregnation (a 573 K for 1 h in air, 723 K for 1 h
cocatalyst was loaded on KCaSrTa5O15. In contrast, the activities
of KCaSrTa5O15 were not enhanced when Ni cocatalysts were
loaded by photodeposition, and impregnation and H2 reduction.
NiO-loaded KCaSrTa5O15 prepared at 1423 K showed the highest
activity. Although the initial activity of this photocatalyst was
high, the rates of H2 and O2 evolution decreased at the initial
stage as shown in Fig. 3. H2 and O2 steadily evolved in a
b
in air), impregnation and subsequent H2 reduction (773 K for 2 h in H2
flow), reactant solution: CO2 dissolved in water (350 mL) under 1 atm,
light source: a 400 W high-pressure mercury lamp, reactor: an inner
irradiation cell made of quartz. KCaSrTa5O15 was prepared by a solid-
state reaction at 1423 K for 10 h.
stoichiometric amount after the deactivation. The apparent the Ag-loaded KCaSrTa5O15 photocatalyst produced CO from
quantum yield was 2.3% at 254 nm for water splitting (Fig. S2, CO2 as a reduction product regardless of loading methods
ESI†). Thus, KCaSrTa5O15 with tungsten bronze structure has of the Ag cocatalyst. In particular, the high activity for CO2
arisen as a new photocatalyst for water splitting.
reduction was observed, when the Ag cocatalyst was loaded by
The KCaSrTa5O15 photocatalyst was applied to CO2 reduction photodeposition and impregnation methods as shown in Table 2
as shown in Table 2. The pristine KCaSrTa5O15 produced only H2 and Fig. S3 (ESI†). The fact that CO2 was reduced on the
and O2 without any reduction products of CO2. This indicates Ag-loaded KCaSrTa5O15, but not on the pristine KCaSrTa5O15,
that there were no active sites for CO2 reduction on the surface of indicates that the Ag cocatalyst works as a reduction site for CO2.
the KCaSrTa5O15 photocatalyst. Therefore, various cocatalysts Metallic Ag is a good electrocatalyst for reduction of CO2 to CO.26
were loaded to introduce active sites. Water splitting activity of The process of CO2 reduction to CO on the Ag cocatalyst would be
KCaSrTa5O15 was enhanced when effective cocatalysts such as similar to that on the Ag electrocatalyst27,28 as observed for
15
16
NiO and Au17,24,25 for water splitting were loaded as observed for BaLa4Ti4O15 and Zn-dopeÀd Ga2O3 photocatalysts. A CO2
the BaLa4Ti4O15 photocatalyst.15 No CO2 reduction proceeded, molecule is reduced to CO2 adsorbed on a Ag cocatalyst by Àa
ꢀ
ꢀ
when Ni, Ru, Rh, Pd, Pt and Au were loaded. When the Cu photoexcited electron. Although the redox potential of CO2
cocatalyst was loaded, a small amount of CO evolved. In contrast, formation is À1.9 V,27 the redox potential should become more
This journal is ©the Owner Societies 2014
Phys. Chem. Chem. Phys., 2014, 16, 24417--24422 | 24419

View Article Online
Paper
PCCP
Characterization of cocatalysts loaded on KCaSrTa5O15
NiO and Ag cocatalysts loaded on KCaSrTa5O15 before and after
photocatalytic water splitting and CO2 reduction were analyzed
by scanning electron microscopy (SEM), X-ray photoelectron
spectroscopy (XPS) and diffuse reflectance spectroscopy (DRS)
in order to clarify the active states of these cocatalysts.
The particle size and morphology of NiO and Ni cocatalysts
loaded on KCaSrTa5O15 photocatalysts were observed using
SEM before and after photocatalytic water splitting (Fig. S5,
ESI†). The particle size of NiO loaded by an impregnation
method was about 10 nm (Fig. S5(a), ESI†). Small particle sizes
of metallic Ni remained after H2 reduction, though a part of
metallic Ni sintered (Fig. S5(c), ESI†). The metallic Ni aggre-
gated after the photocatalytic water splitting, and nano-
particles of Ni were hardly observed (Fig. S5(d), ESI†). The
shape of Ni loaded by a photodeposition method was not
spherical being clearly different from that loaded by an impreg-
nation method (Fig. S5(e), ESI†). Thus, the particle size and the
shape of NiO and/or Ni cocatalysts after photocatalytic water
splitting depended on the loading methods.
Fig. 4 CO2 reduction over Ag (0.5 wt%)-loaded KCaSrTa5O15 photo-
catalyst in an aqueous medium. Photocatalyst: 0.5 g, reactant solution:
water (350 mL) with CO2 gas flow (30 mL minÀ1), light source: a 400 W
high-pressure mercury lamp, reactor: an inner irradiation cell made of
quartz. KCaSrTa5O15 was prepared by a solid-state reaction at 1423 K for
10 h. The Ag cocatalyst was loaded by a photodeposition.
À
ꢀ
positive due to stabilization by adsorption. The adsorbed CO2
Fig. 5 shows DRS of NiO and Ni-loaded KCaSrTa5O15 before
is reacted with a H+ ion in water to form an adsorbed –COOH. and after photocatalytic water splitting. Non-loaded KCaSr-
The adsorbed –COOH is subsequently reduced to become CO Ta5O15 was white and possessed 302 nm of an absorption edge
and OHÀ by another electron.27,28
(Fig. 5(a)). A color of KCaSrTa5O15 loaded with NiO by an
H2, O2 and CO evolved steadily with a reaction time on the impregnation method was gray and the background of the
optimized Ag-loaded KCaSrTa5O15 photocatalyst as shown in DRS arose at visible and near IR regions (Fig. 5(b)). A color of
Fig. 4. The turnover number of electrons reacted for CO Ni/KCaSrTa5O15 obtained by an impregnation and subsequent
evolution to the number of the Ag atoms in the cocatalyst H2 reduction was pale brown and gave a broad absorption band
calculated using eqn (1) was 8.6 at 20 h, and the ratio of with a peak at around 400 nm (Fig. 5(d)). All Ni or NiO-loaded
electrons to holes calculated using eqn (2) was unity. These photocatalysts were dark purple after photocatalytic water
results indicate that the CO2 reduction over Ag-loaded KCaSr- splitting giving broad absorption bands in the visible light
Ta5O15 proceeded photocatalytically and water was used as an region (Fig. 5(c), (e) and (f)). These results suggest that the
electron donor.
condition of nickel cocatalysts was similar to each other during
photocatalytic water splitting regardless of the loading methods.
TONCO = (The number of electrons consumed for CO
2
formation)/(The total number of a Ag atom in cocatalyst on
KCaSrTa5O15
)
(1)
eÀ/h+ = (The number of electrons consumed for H2 and CO
formation)/(The numbers of holes consumed for O2 formation)
(2)
It has been reported that methane forms from not CO2 but
an organic contamination adsorbed on the photocatalyst sur-
face.29 Therefore, an isotope experiment using 13CO2 was
carried out to clarify the carbon source of CO formed. When
the 13CO2 gas was analyzed using a GC-MS and using a MS-5A
column, no peaks due to 12CO and 13CO were detected, indicat-
ing that the 13CO2 gas contained negligible amounts of 12CO
and 13CO. In contrast, photocatalytic reduction of 13CO2 over
Ag-loaded KCaSrTa5O15 photocatalyst gave 13CO and not 12CO
(Fig. S4, ESI†). Additionally, the ratio of electrons to holes
calculated from the products was about unity also in this
isotopic experiment. Therefore, it was proven that CO was
produced from CO2 molecules over the Ag-loaded KCaSrTa5O15
photocatalyst.
Fig. 5 Diffuse reflectance spectra of (a) pristine KCaSrTa5O15 and (b)–(f)
NiO (0.5 wt%) and Ni (0.5 wt%)-loaded KCaSrTa5O15 by various methods
before and after water splitting. (b) Before and (c) after water splitting for
the samples prepared by an impregnation method, (d) before and (e) after
water splitting for the samples prepared by impregnation and subsequent
H2 reduction, (f) after water splitting for the samples prepared by a
photodeposition. KCaSrTa5O15 was prepared by a solid-state reaction at
1423 K for 10 h.
24420 | Phys. Chem. Chem. Phys., 2014, 16, 24417--24422
This journal is © the Owner Societies 2014

View Article Online
PCCP
Paper
The dark purple color did not quickly change after being exposed cocatalyst loaded on the photocatalysts with wide band gaps has
to air.
Fig. 6 shows XPS of Ni 2p of NiO and Ni cocatalysts loaded
a unique character.
The impregnation method for loading the nickel cocatalyst
on KCaSrTa5O15 before and after photocatalytic water splitting. was more effective than impregnation and subsequent H2
Binding energies of standard samples of metallic Ni, NiO and reduction, and photodeposition methods for water splitting over
Ni(OH)2 agreed with those of previous reports.30 The surface of the KCaSrTa5O15 photocatalyst as shown in Table 1. SEM and XPS
the standard NiO powder was covered with surface hydroxyl measurements suggested that most active KCaSrTa5O15 photo-
groups. NiO loaded on KCaSrTa5O15 by an impregnation catalysts possessed the cocatalyst of fine NiO particles covered
method also gave peaks due to NiO and Ni(OH)2 (Fig. 6(a)). with surface nickel hydroxide.
Ni/KCaSrTa5O15 obtained by an impregnation and subsequent
SEM images of Ag loaded KCaSrTa5O15 photocatalysts were
H2 reduction gave XPS signals due to metallic Ni with NiO also observed before and after photocatalytic CO2 reduction
(Fig. 6(c)). All samples showed XPS signals mainly due to Ni(OH)2 (Fig. S7, ESI†). The impregnation method gave a Ag cocatalyst
after photocatalytic water splitting (Fig. 6(b), (d) and (e)). Ni(OH)2 with a particle size of about 10 nm before and after H2 reduction
is generally green, not dark purple as mentioned above. DRS of (Fig. S7(a) and (c), ESI†). The Ag cocatalyst aggregated after
the samples after photocatalytic water splitting was different photocatalytic CO2 reduction (Fig. S7(b) and (d), ESI†). The growth
from that of Ni(OH)2 (Fig. S6, ESI†). Therefore, the XPS signals of some Ag particles reached up to 50–100 nm. The particle size of
due to Ni(OH)2 observed for the samples after photocatalytic Ag after photocatalytic CO2 reduction was similar to that obtained
water splitting indicate the existence of not bulky Ni(OH)2 but by a photodeposition method (Fig. S7(e) and (f), ESI†).
the surface nickel hydroxide as observed for a NiOx/SrTiO3
Nano-sized metallic Ag particles generally give a surface
photocatalyst.30 It has been reported that the color of the NiO/ plasmonic absorption band in visible light region.33 Therefore,
InBO4 photocatalyst also changed to dark purple after photo- DRS of samples before and after photocatalytic CO2 reduction
catalytic water splitting.31 An ultra fine NiO cocatalyst loaded on was obtained as shown in Fig. 7. The Ag cocatalyst obtained by
a NaTaO3:La photocatalyst gives visible light absorption bands H2 reduction was orange in color and gave a characteristic
which are different from that of Ni(OH)2.32 Thus, an active NiO surface plasmonic absorption spectrum (Fig. 7(c)), while such a
spectrum was not observed for the Ag cocatalyst obtained by the
impregnation method (Fig. 7(a)). All samples after photocatalytic
CO2 reduction gave similar absorption spectra to the sample
obtained by H2 reduction (Fig. 7(b), (d), (e) and (f)), although the
intensities of the absorption spectra decreased after the photo-
catalytic reduction of CO2. These results suggest dissolution and
re-deposition of Ag during the photocatalytic reaction giving
aggregated Ag particles as observed by SEM. The Ag cocatalyst
prepared by the impregnation method after the photocatalytic
reduction of CO2 also gave the similar absorption spectrum
Fig. 6 X-ray photoelectron spectra of Ni 2p of NiO (0.5 wt%) and Ni
(0.5 wt%)-loaded KCaSrTa5O15. (a) before and (b) after water splitting for
the samples prepared by an impregnation method, (c) before and (d) after
water splitting for the samples prepared by an impregnation and sub-
sequent H2 reduction, (e) after water splitting for the sample prepared by a
photodeposition. Standard sample of (f) Ni(OH)2, (g) NiO, (h) metallic Ni
foil. Reference data30 are indicated as dashed lines respectively: Ni metal
(853.1 eV), NiO (854.5 eV), Ni(OH)2 (856.6 eV).
Fig. 7 Diffuse reflectance spectra of Ag (0.5 wt%)-loaded KCaSrTa5O15 by
various methods before and after CO2 reduction. (a) Before and (b) after
CO2 reduction for the samples prepared by an impregnation method, (c)
before and (d) after CO2 reduction for the samples prepared by an
impregnation and subsequent H2 reduction, after CO2 reduction for the
samples prepared by a photodeposition at (e) 2 h and (f) 20 h, and (g)
metallic and bulky Ag. KCaSrTa5O15 was prepared by a solid-state reaction
at 1423 K for 10 h.
This journal is ©the Owner Societies 2014
Phys. Chem. Chem. Phys., 2014, 16, 24417--24422 | 24421

View Article Online
Paper
PCCP
6 Y. Tamaki, K. Watanabe, K. Koike, H. Inoue, T. Morimoto
and O. Ishitani, Faraday Discuss., 2012, 155, 115.
7 S. M. Aliwi and K. F. Al-Jubori, Sol. Energy Mater., 1989,
18, 223.
indicating that the Ag cocatalyst was reduced to fine metallic
particles by photogenerated electrons. These results indicate
that the metallic Ag giving the surface plasmonic absorption
spectrum is an active site for the photocatalytic CO2 reduction.
8 M. Kanemoto, K. Ishihara, Y. Wada, T. Sakata, H. Mori and
S. Yanagida, Chem. Lett., 1992, 835.
Conclusions
ˇ
9 K. Koci, L. Obalova and Z. Lacny, Chem. Pap., 2008, 62, 1.
10 S. N. Habisreutinger, L. Schmidt-Mende and J. K. Stolarczyk,
Angew. Chem., Int. Ed., 2013, 52, 7372.
11 M. Tahir and N. S. Amin, Energy Convers. Manage., 2013,
76, 194.
12 S. Navalon, A. Dhakshinamoorthy, M. Alvaro and H. Garcia,
ChemSusChem, 2013, 6, 562.
13 J. Mao, K. Li and T. Peng, Catal. Sci. Technol., 2013, 3, 2481.
14 K. Sayama and H. Arakawa, J. Phys. Chem., 1993, 97, 531.
15 K. Iizuka, T. Wato, Y. Miseki, K. Saito and A. Kudo, J. Am.
Chem. Soc., 2011, 133, 20863.
KCaSrTa5O15 (BG = 4.1 eV) with tungsten bronze structure has
arisen as a new photocatalyst for water splitting and CO2
reduction under UV light irradiation. KCaSrTa5O15 showed activity
for water splitting without a cocatalyst. In addition, NiO was an
effective cocatalyst. The NiO cocatalyst activated during water
splitting was dark purple and possessed a Ni hydroxide-shell–
NiO-core structure as an active site for water reduction. The
apparent quantum yield of optimized NiO-loaded KCaSrTa5O15
was 2.3% at 254 nm for water splitting. On the other hand, the
Ag-loaded KCaSrTa5O15 photocatalyst was active for CO2 reduction
to CO in an aqueous medium. The Ag cocatalyst activated during
CO2 reduction gave a surface plasmonic absorption band in the
visible light region suggesting that fine metallic Ag particles are
the active sites for CO2 reduction. The isotope experiment using
13CO2 revealed that the carbon source of produced CO was CO2
molecules. Moreover, a stoichiometric amount of O2 evolved and
16 K. Teramura, Z. Wang, S. Hosokawa, Y. Sakata and
T. Tanaka, Chem. – Eur. J., 2014, 20, 9906.
17 A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253.
18 H. Kato and A. Kudo, Catal. Today, 2003, 78, 561.
19 A. Kudo, H. Okutomi and H. Kato, Chem. Lett., 2000, 1212.
20 Y. Miseki, H. Kato and A. Kudo, Energy Environ. Sci., 2009,
2, 306.
TONCO calculated from products was larger than 1 indicating that
2
21 D. Nelson, B. Scheetz and C. Abate, The International Centre
for Diffraction Data, #40-351.
CO2 reduction to CO photocatalytically proceeded using water as
an electron donor.
22 H. Kato and A. Kudo, J. Phys. Chem. B, 2001, 105, 4285.
23 M. P. Seah, Surf. Interface Anal., 1989, 14, 488.
24 A. Iwase, H. Kato and A. Kudo, Appl. Catal., B, 2013, 89, 136.
25 Y. Negishi, M. Mizuno, M. Hirayama, M. Omatoi,
T. Takayama, A. Iwase and A. Kudo, Nanoscale, 2013,
5, 7188.
26 Y. Hori, K. Kikuchi and S. Suzuki, Chem. Lett., 1985, 1695.
27 Y. Hori, H. Wakebe, T. Tsukamoto and O. Koga, Electrochim.
Acta, 1994, 39, 1833.
28 T. Hatsukade, K. P. Kuhl, E. R. Cave, D. N Abram and
T. F. Jaramillo, Phys. Chem. Chem. Phys., 2014, 16, 13814.
29 O. Ishitani, C. Inoue, Y. Suzuki and T. Ibusuki, J. Photochem.
Photobiol., A, 1993, 72, 269.
30 K. Domen, A. Kudo, T. Onishi, N. Kosugi and H. Kuroda,
J. Phys. Chem., 1986, 90, 292.
Acknowledgements
This work was supported by a Grant in Aid (No. 24246131) from
the Ministry of Education, Sports, Science & Technology in Japan
and the ENEOS hydrogen foundation.
References
1 J. Hawecker, J. M. Lehn and R. Ziessel, J. Chem. Soc., Chem.
Commun., 1983, 536.
2 J. Hawecker, J. M. Lehn and R. Ziessel, Helv. Chim. Acta,
1986, 69, 1990.
3 H. Takeda, K. Koike, H. Inoue and O. Ishitani, J. Am. Chem.
Soc., 2008, 130, 2023.
4 H. Takeda and O. Ishitani, Coord. Chem. Rev., 2010, 254, 346.
31 Q. Jia, Y. Miseki, K. Saito, H. Kobayashi and A. Kudo, Bull.
Chem. Soc. Jpn., 2010, 83, 1275.
5 T. Yui, Y. Tamaki, K. Sekizawa and O. Ishitani, in Photo- 32 H. Kato, K. Asakura and A. Kudo, J. Am. Chem. Soc., 2003,
catalysis, Top. Curr. Chem., ed. C. A. Bignozzi, Springer,
125, 3082.
2011, vol. 303, pp. 151–184.
33 K. Matsubara and T. Tatsuma, Adv. Mater., 2007, 19, 2802.
24422 | Phys. Chem. Chem. Phys., 2014, 16, 24417--24422
This journal is © the Owner Societies 2014