Photocatalytic activities of noble metal ion doped SrTiO3 under visible light irradiation

Article abstract of DOI:10.1021/jp049556p

Mn-, Ru-, Rh-, and Ir-doped SrTiO₃ possessed intense absorption bands in the visible light region due to excitation from the discontinuous levels formed by the dopants to the conduction band of the SrTiO₃ host. Mn- and Ru-doped SrTiO₃ showed photocatalytic activities for O₂ evolution from an aqueous silver nitrate solution while Ru-, Rh-, and Ir-doped SrTiO₃ loaded with Pt cocatalysts produced H₂ from an aqueous methanol solution under visible light irradiation (?? > 440 nm). The Rh(1%)-doped SrTiO₃ photocatalyst loaded with a Pt cocatalyst (0.1 wt %) gave 5.2% of the quantum yield at 420 nm for the H ₂ evolution reaction.

Full text of DOI:10.1021/jp049556p

8992
J. Phys. Chem. B 2004, 108, 8992-8995
Photocatalytic Activities of Noble Metal Ion Doped SrTiO3 under Visible Light Irradiation
Ryoko Konta,† Tatsuya Ishii,† Hideki Kato,† and Akihiko Kudo*,†,‡
Department of Applied Chemistry, Faculty of Science, Science UniVersity of Tokyo, 1-3 Kagurazaka,
Shinjyuku-ku, Tokyo 162-8601, Japan, and Core Research for EVolutional Science and Technology, Japan
Science and Technology Agency (CREST, JST), 4-1-8 Honcho, Kawaguchi-shi, Saitama 332-0012, Japan
ReceiVed: January 31, 2004; In Final Form: April 2, 2004
Mn-, Ru-, Rh-, and Ir-doped SrTiO3 possessed intense absorption bands in the visible light region due to
excitation from the discontinuous levels formed by the dopants to the conduction band of the SrTiO3 host.
Mn- and Ru-doped SrTiO3 showed photocatalytic activities for O2 evolution from an aqueous silver nitrate
solution while Ru-, Rh-, and Ir-doped SrTiO3 loaded with Pt cocatalysts produced H2 from an aqueous methanol
solution under visible light irradiation (λ > 440 nm). The Rh(1%)-doped SrTiO3 photocatalyst loaded with
a Pt cocatalyst (0.1 wt %) gave 5.2% of the quantum yield at 420 nm for the H2 evolution reaction.
1. Introduction
85.5%), and PtO2 (Kojima Chemical; 95%) were mixed with a
small amount of methanol in a ratio according to the composi-
tion of SrTi1-xMxO3, and then the mixtures were calcined in
air at 1173-1573 K for 5-25 h using an alumina crucible
(purity 99.7%). The obtained powders were determined by
powder X-ray diffraction (Rigaku; MiniFlex, Cu KR). Diffuse
reflectance spectra were obtained using a UV-vis-NIR spec-
trometer (Jasco; UbestV-570) and were converted from reflec-
tion to absorbance by the Kubelka-Munk method. Pt cocatalysts
were loaded in situ by a photoreduction method using an
aqueous H2PtCl6 solution (Tanaka Kikinzoku; 37.55% as Pt in
H2PtCl6‚6H2O).
Photocatalytic reactions of H2 evolution from an aqueous
methanol solution (10 vol %) and O2 evolution from an aqueous
silver nitrate solution (0.05 mol L-1) were carried out in a closed
gas circulation system. The photocatalyst powder (0.3 g) was
dispersed by a magnetic stirrer in a reactant solution in the cell
with a top window made of Pyrex. The light source was a 300
W Xe lamp (ILC Technology; CERMAX LX-300F) attached
with cutoff filters (HOYA; L42 and Y44) to control the
wavelength of the incident light. The amounts of evolved gases
were determined using on-line gas chromatography (Shimazu;
GC-8A, MS-5A column, TCD, Ar carrier). The quantum yield
was measured using a 300 W Xe lamp (ILC Technology;
CERMAX LX-300F) attached with a band-pass filter (Kenko)
and a cutoff filter (HOYA). A thermopile (OPHIR; a 3A-P-SH
head and a NOVA energy monitor) was used for the measure-
ment of the number of incident photons. The value obtained by
this experiment is an apparent quantum yield (percent), calcu-
lated by the following equation:
Development of photocatalysts which can efficiently split
water into H2 and O2 under sunlight irradiation has extensively
been explored. The number of photocatalysts which are active
for H2 evolution under visible light irradiation is limited even
for the reaction in the presence of an electron donor. In some
cases, visible light responses of photocatalysts with large band
gaps are brought by doping of metal cations which form electron
donor levels within the band gaps. TiO2 and SrTiO3 are wide
band gap photocatalysts (band gap ) 3.2 eV) which can split
water into H2 and O2 under ultraviolet light irradiation.1-5 There
are some reports that the photocatalysts6-8 and semiconductor
electrodes9-15 of TiO2 and SrTiO3 doped with transition metal
ions show visible light responses. The present authors have
reported that TiO2 and SrTiO3 photocatalysts with Cr and Sb
ions or Cr and Ta ions substituted for Ti sites showed activity
for O2 or H2 evolution from aqueous solutions containing the
sacrificial reagents under visible light irradiation.7,8 On the other
hand, it has been reported that photoelectrodes of SrTiO3 doped
with Rh or Pt in the surface region give a photoresponse to
visible light.14,15
Mn and some noble metals seem to be interesting as dopants
that bring the visible light response to the SrTiO3 photocatalyst
because of their redox properties. In the present study, the effects
of doping of Mn and noble metal cations into the wide band
gap photocatalyst SrTiO3 on the photocatalytic activity were
examined to create a visible absorption center and an active
surface site. The visible light response was discussed with a
proposed electronic structure.
2. Experimental Section
apparent quantum yield (%) )
SrTiO3 powders doped with Mn, Ru, Rh, Pd, Ir, and Pt ions
(denoted as SrTiO3:M(X %)) were synthesized in the following
manner. The starting materials, SrCO3 (Kanto Chemical; 99.9%,
treated in air at 573 K for 1 h before use), TiO2 (Soekawa
Chemical; 99.9%), MnO2 (Nacalai Tesque; 99%), RuO2 (Rare
Metallic; 99.9%), Rh2O3 (Wako Pure Chemical; 95%), PdO
(Wako Pure Chemical; 98.0%), IrO2 (Soekawa Chemical;
[number of reacted electrons or holes]/
[number of incident photons] × 100 )
[number of H2 molecules evolved × 2]/
[number of incident photons] × 100
3. Results and Discussion
3.1. Characterization. XRD patterns of SrTiO3:M(0.5%) (M
) Mn, Ru, Rh, Pd, Ir, and Pt) powders synthesized at 1423 K
showed a single phase of SrTiO3 except SrTiO3:Pd(0.5%), in
which some peaks assigned to metallic palladium were con-
* To whom correspondence should be addressed. E-mail: a-kudo@
rs.kagu.tus.ac.jp.
† Science University of Tokyo.
‡ JST.
10.1021/jp049556p CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/27/2004

Noble Metal Ion Doped SrTiO3 Photocatalysis
J. Phys. Chem. B, Vol. 108, No. 26, 2004 8993
Figure 1. Diffuse reflectance spectra of SrTiO3:M(0.5%). M ) (a)
Mn, (b) Ru, (c) Rh, (d) Pd, (e) Ir, and (f) Pt. A broken line represents
a spectrum of nondoped SrTiO3.
Figure 2. Dependence of H2 evolution activity of Rh-doped SrTiO3
upon the doping amount. Catalyst, 0.3 g; cocatalyst, Pt (0.5 wt %);
reactant solution, 150 mL of 10 vol % aqueous MeOH; light source,
300 W Xe lamp with cutoff filters (λ > 440 nm).
TABLE 1: Photocatalytic Activities of SrTiO3:M(0.5%) for
H2 or O2 Evolution from an Aqueous Solution Containing
Sacrificial Reagents under Visible Light Irradiationa
TABLE 2: H2 Evolution from an Aqueous Methanol
Solution under Visible Light Irradiation of SrTiO3:Rh(1%)
Photocatalysts Prepared under Different Conditionsa
activity/µmol h-1
b
c
calcination condition
M
Eg/eV
H2
O2
temp/K
time/h
H2/µmol h-1
Mn
Ru
Rh
Ir
2.7
1.9
1.7
2.3
0.2
1.7
17.2
8.6
2.7
3.9
0
1173
1273
25
5
10
20
5
10
20
10
9
46
90
58
42
65
24
21
0.4
a Catalyst, 0.3 g; reactant solution, 150 mL; light source, 300 W Xe
lamp with cutoff filter (λ > 440 nm). b From 10 vol % aqueous MeOH;
cocatalyst, Pt (0.5 wt %). c From 0.05 mol L-1 aqueous AgNO3.
1423
1573
firmed. Doping of Pd2+ seemed to be difficult because its ion
radius (0.86 Å) was much larger than that of Ti4+ (0.605 Å).
a Catalyst, 0.3 g; cocatalyst, Pt (0.1 wt %); reactant solution, 150
mL of 10 vol % aqueous MeOH; light source, 300 W Xe lamp with
cutoff filters (λ > 440 nm).
Figure 1 shows diffuse reflectance spectra of SrTiO3:M(0.5%)
(M ) Mn, Ru, Rh, Pd, Ir, and Pt) powders. In SrTiO3:M(0.5%)
(M ) Mn, Ru, Rh, and Ir), absorption bands with shoulders in
the visible light region were observed in addition to the band
gap absorption of SrTiO3. These shapes of the diffuse reflectance
spectra were similar to those of the doped sintered electrode
that was reported by Matsumura et al.14 These shapes were the
characteristics of doped photocatalysts and indicated that
discontinuous levels were formed by the dopants in the
forbidden band. The existence of two absorption bands of a
shoulder around 420 nm and broad bands around 580 and 1000
nm for the SrTiO3:Rh powder indicated that some doping levels
due to different oxidation numbers were formed in the forbidden
band. The diffuse reflectance spectrum of SrTiO3 not doped
but loaded with Rh or rhodium oxides on the surface by an
impregnation method was quite different from that of the present
material, demonstrating that the doped material was obtained
by the present preparation condition.
mainly be due to the surface states created by their dopants
below the conduction bands, resulting in difficulty in H2
evolution.
Among the photocatalysts listed in Table 1, further investiga-
tion was carried out for SrTiO3:Rh, which showed the highest
activity for H2 evolution.
The optimum preparation condition of the SrTiO3:Rh pho-
tocatalyst was examined. Pt/SrTiO3:Rh(1%) showed the highest
activity for H2 evolution from an aqueous methanol solution
under visible light irradiation among SrTiO3 photocatalysts with
different doping amounts of Rh as shown in Figure 2. The
increase in the doping amount enhances visible light absorption
whereas it makes a recombination center between photogener-
ated carriers increase. Therefore, the volcano-type dependence
was obtained.
The optimal calcination condition of SrTiO3:Rh(1%) was
examined as shown in Table 2. Starting materials had to be
calcined more than 20 h at 1173 K or 10 h at 1273 K to get a
single phase. The best preparation temperature was 1273 K.
The suitable calcination time was 10 h for all temperatures. The
preparation temperature and time affected the crystallinity,
particle size, homogeneity of the dopant, and other natures which
were important factors dominating photocatalytic activities.
The SrTiO3:Rh(1%) photocatalyst calcined for 10 h at 1273
K showed the highest activity for H2 evolution from an aqueous
methanol solution under visible light irradiation in the presence
of a Pt cocatalyst as shown in Figure 3. An induction period
was observed at the early stage of the photocatalytic reaction.
The induction period suggests that some reduction process
should have proceeded at the beginning stage instead of water
3.2. Photocatalytic Activity. Table 1 shows hydrogen
evolution from an aqueous methanol solution and oxygen
evolution from an aqueous silver nitrate solution under visible
light irradiation (λ > 440 nm) on SrTiO3:M(0.5%) which
possessed absorption bands in a visible light region. Mn- and
Ru-doped SrTiO3 showed photocatalytic activities for O2
evolution from an aqueous silver nitrate solution while Ru-, Rh-,
and Ir-doped SrTiO3 loaded with Pt cocatalysts produced H2
from an aqueous methanol solution. No O2 evolution activity
of Rh-doped SrTiO3 is due to the lack of an O2 evolution site
as discussed below. In the cases of SrTiO3 doped with Mn and
Ru, the dopants might work as active sites for O2 evolution,
resulting in showing the activities for O2 evolution. In contrast,
the low activities for H2 evolution on these photocatalysts would

8994 J. Phys. Chem. B, Vol. 108, No. 26, 2004
Konta et al.
high oxidation number was reduced by the photoexcited
electrons at the early stage of the reaction. The yellow color of
the reduced SrTiO3:Rh(1%) photocatalyst approximately re-
turned to the original color as soon as the photocatalyst was
exposed to the air. The doped Rh ions were in the condition
that they could oscillate between oxidized and reduced forms.
The yellow powder was also obtained by reduction with H2 at
573 K. However, the color of the thermally reduced powder
did not return to the original color. Moreover, although the
induction period disappeared, the photocatalytic activity (5 µmol
h-1) was much lower than that of the photocatalytically reduced
sample.
The onset of the action spectrum agreed well with the
absorption edge of the diffuse reflectance spectrum of the
reduced photocatalyst rather than with that of the original
photocatalyst. Visible light up to 540 nm can be used. The action
spectrum was similar to those of photocurrent reported by
Matsumura et al.14 and Watanabe et al.15 The quantum yield of
the Pt(0.1 wt %)/SrTiO3:Rh(1%) photocatalyst was 5.2% at 420
nm. This represents a very high value among doped oxide
photocatalysts reported in the literature. When a foreign element
is added into some host photocatalysts, the quantum yields
generally become much lower than those of the undoped
materials because the dopant could provide an electron-hole
recombination site. However, the apparent quantum yield of the
Pt/SrTiO3:Rh(1%) photocatalyst (6% at 365 nm) was almost
the same as that of nondoped Pt/SrTiO3 (6% at 365 nm).
Moreover, the H2 evolution rate of the Pt/SrTiO3:Rh(1%)
photocatalyst (330 µmol h-1) was faster than that of nondoped
Pt/SrTiO3 (166 µmol h-1) even under full arc irradiation. The
Rh doping did not make the photocatalytic activity decrease
even under the band gap excitation. Rh ions also form
recombination centers more or less. However, the similarity in
the activity between Rh-doped and nondoped photocatalysts
indicates that the contribution of Rh dopant to the formation of
a recombination center is less than that of other transition metal
dopants such as iron ions. In general, transition metal ions of
which the valency is easily changed between relatively stable
oxidation numbers tend to form a recombination center. In the
present photocatalyst, most Rh ions are trivalent under the
working conditions for H2 evolution as mentioned below. The
trivalency is definitely the most stable oxidation number among
Rh ions in oxides, suggesting the stability of Rh3+. Therefore,
the possibility that Rh ions form a recombination center would
be less than that for other ions such as iron ions of which di-
and trivalencies are stable.
Some oxide, oxynitride, and oxysulfide photocatalysts which
are active for H2 evolution under visible light irradiation have
been reported.16-18 However, their activities are not high. Pure
oxide photocatalysts with a high activity for H2 evolution under
visible light irradiation has not been reported so far. In this
regard, the Rh-doped SrTiO3 powder is a unique photocatalyst
for H2 evolution.
3.3. Mechanism. The derived schematic mechanism for the
photocatalytic H2 evolution on Pt/SrTiO3:Rh is illustrated in
Figure 5. The visible light absorption bands in the diffuse
reflectance spectrum suggested the existence of two different
species of the doped Rh at least. They can be assigned to be
Rh3+ and a rhodium species with a higher oxidation number
than Rh3+ such as Rh5+. These Rh ions with different oxidation
numbers would neighbor each other, judging from the charge
and distortion compensation by their cations. The Rh with higher
oxidation number can be easily reduced to Rh3+ by photo-
generated electrons at an early stage of the photocatalytic
reaction, indicating that the Rh with higher oxidation number
works as an electron acceptor. The highly efficient H2 evolution
Figure 3. H2 evolution from an aqueous methanol solution under
visible light irradiation over a Pt(0.1 wt %)/SrTiO3:Rh(1%) photocata-
lyst prepared by solid-state reaction at 1273 K for 10 h. Catalyst, 0.3
g; reactant solution, 150 mL of 10 vol % aqueous MeOH; light source,
300 W Xe lamp with cutoff filters (λ > 440 nm).
Figure 4. Diffuse reflectance spectra of (a) before and (b) after
photocatalytic reaction, and (c) action spectrum for H2 evolution from
an aqueous methanol solution over a SrTiO3:Rh(1%) photocatalyst. The
inset shows an expanded view of the spectra in the wavelength region
from 800 to 1300 nm. Catalyst, 0.3 g; cocatalyst, Pt (0.1 wt %); reactant
solution, 150 mL of 10 vol % aqueous methanol solution; light source,
300 W Xe lamp.
reduction. On the other hand, although the rate of H2 evolution
became slow after ca. 10 h, it was improved by evacuating a
gas phase and the induction time was not observed. The decrease
in activity with reaction time was probably due to the poisoning
by evolved H2 because H2 evolution was suppressed by the
purposeful addition of H2 into a gas phase. The maximum rate
of H2 evolution was 117 µmol h-1. The amount of the SrTiO3:
Rh photocatalyst used was 1.6 mmol, and it contained 16 µmol
of Rh. The number of electrons reacted was estimated to be
2.8 mmol judging from the amount of evolved H2. The electron
turnover number to amounts of doped Rh and surface Ti reached
175 and 300, respectively, after 27 h of irradiation. Therefore,
it can be concluded that H2 evolution photocatalytically
proceeded on Pt/SrTiO3:Rh.
Figure 4 shows diffuse reflectance spectra before and after
the photocatalytic reaction and an action spectrum for H2
evolution from an aqueous methanol solution over the optimized
Pt(0.1 wt %)/SrTiO3:Rh(1%) photocatalyst. An absorption peak
around 580 and 1000 nm observed before the reaction vanished
after the reaction, whereas a shoulder around 420 nm grew up.
The color of this photocatalyst changed from grayish purple to
yellow during the photocatalytic reaction in agreement with the
spectral change. The change in color and the observation of an
induction period indicate that some doped Rh species with a

Noble Metal Ion Doped SrTiO3 Photocatalysis
J. Phys. Chem. B, Vol. 108, No. 26, 2004 8995
Figure 5. Proposed band structure and visible light response of Rh-doped SrTiO3 photocatalyst.
from this fact that O2 evolution from water is a tougher reaction
on the SrTiO3:Rh photocatalyst under visible light irradiation
indicates that Rh3+ works as an electron donor. Therefore, the
absorption bands around 580 and 1000 nm, which have onsets
at 1.7 and 1.0 eV, are assigned to the transition from the valence
band and the donor level to the acceptor level, respectively.
On the other hand, the absorption band remaining after the
reduction by photogenerated electrons should be the transition
from the donor level formed with Rh3+ to the conduction band.
It is the key process for the photocatalytic H2 evolution on Pt/
SrTiO3:Rh under visible light irradiation. The electrons photo-
generated in the conduction band reduce water to form H2, while
holes formed in the electron donor level possessed thermody-
namic and kinetic potentials for oxidation of methanol. This
indicates that the doped Rh3+ formed the visible light absorption
center and the surface reaction center. The optimum doping
amount was 1%: it is not low in general. Although the donor
levels are basically discrete, they should partly overlap with O
2p orbital to form a hybrid orbital. Therefore, the migration of
holes is not so difficult. For example, in the case of ZnS doped
with 0.1% of Ni2+ ions, high activity has been realized despite
a quite small amount of doping.19 However, O2 evolution hardly
proceeded. The doping level in which holes are generated is
estimated to be +2.1 eV as shown in Figure 5. This potential
might thermodynamically be deep enough for oxidation of water
to form O2. However, dopants are not easily able to form a
reaction center for the O2 evolution because the reaction is
accompanied by four-electron oxidation. The kinetic limitation
of the lack of an active site is one of the reasons for no O2
evolution activity. On the other hand, a photocatalytic reaction
efficiently proceeds after Rh species with high oxidation
numbers are reduced. These species cannot be reduced in the
reaction in an aqueous silver nitrate solution because they work
as recombination centers in the absence of a hole scavenger.
This is another reason for no photocatalytic activity for O2
evolution. On the other hand, the methanol oxidation is easy
compared to O2 evolution from water from both thermodynamic
and kinetic standpoints. The standard redox potentials for
HCHO/CH3OH and H2CO3/CH3OH are 0.232 and 0.044 V.
These redox potentials are considerably more negative than E°-
(O2/H2O) (1.23 V). This indicates that O2 evolution is thermo-
dynamically difficult more than methanol oxidation. Moreover,
it is well-known that the overpotential for O2 evolution is quite
large (more than 0.5 V): it is a kinetic factor. In fact, O2
evolution from an aqueous methanol solution has never been
observed for any photocatalysts even if the photocatalyst
possesses a quite high activity for O2 evolution. It is obvious
than the oxidation of methanol.
4. Conclusion
SrTiO3:Rh(1%) is a novel oxide photocatalyst that is active
for H2 evolution from an aqueous methanol solution under
visible light irradiation. The visible light response was due to
the transition from the electron donor level formed by Rh ions
to the conduction band composed of Ti 3d orbitals of SrTiO3.
It was supported by diffuse reflectance spectra and the color
change that the reducible Rh species contributed to the formation
of visible light absorption band and the surface reaction sites.
Acknowledgment. This work was supported by Core
Research for Evolutional Science and Technology (CREST) of
Japan Science and Technology Agency (JST) and a Grant-in-
Aid for Scientific Research on Priority Areas (417) from the
Ministry of Education, Culture, Sports, Science and Technology
(MEXT) of the Japanese Government.
References and Notes
(1) Yamaguti, K.; Sato, S. J. Chem. Soc., Faraday Trans. 1 1985, 81,
1237.
(2) Kudo, A.; Domen, K.; Maruya, K.; Onishi, T. Chem. Phys. Lett.
1987, 133, 517.
(3) Sayama, K.; Arakawa, H. J. Photochem. Photobiol. A 1996, 94, 67.
(4) Domen, K.; Kudo, A.; Onishi, T.; Kosugi, N.; Kuroda, H. J. Phys.
Chem. 1986, 90, 292.
(5) Lehn, J. M.; Sauvage, J. P.; Ziessel, R.; Hilaire, L. Isr. J. Chem.
1982, 22, 168.
(6) Sakata, Y.; Yamamoto, T.; Okazaki, T.; Imamura, H.; Tsuchiya,
S. Chem. Lett. 1998, 1253.
(7) Kato, H.; Kudo, A. J. Phys. Chem. B 2002, 106, 5029.
(8) Ishii, T.; Kato, H.; Kudo, A. J. Photochem. Photobiol. A 2004,
163, 181.
(9) Matsumoto, Y.; Shimizu, T.; Sato, E. Electrochim. Acta 1982, 27, 419.
(10) Serpone, N.; Lawless, D. Langmuir 1994, 10, 643.
(11) Marusaka, H. P.; Ghosh, A. K. Sol. Energy Mater. 1979, 1, 237.
(12) Lam, R. U. E. t.; Haart, L. G. J. d.; Wiersma, A. W.; Blasse, G.;
Tinnemans, A. H. A.; Mackor, A. Stud. Inorg. Chem. 1983, 3, 255.
(13) Lam, R. U. E. t.; Haart, L. G. J. d.; Wiersma, A. W.; Blasse, G.;
Tinnemans, A. H. A.; Mackor, A. Mater. Res. Bull. 1981, 16, 1593.
(14) Matsumura, M.; Hiramoto, M.; Tsubomura, H. J. Electrochem. Soc.
1983, 130, 326.
(15) Watanabe, I.; Matsumoto, Y.; Sato, E. J. Electroanal. Chem. 1982,
133, 359.
(16) Hosogi, Y.; Tanabe, K.; Kato, H.; Kobayashi, H.; Kudo, A. Chem.
Lett. 2004, 28.
(17) Hara, M.; Hitoki, G.; Takata, T.; N. Kondo, J.; Kobayashi, H.;
Domen, K. Catal. Today 2003, 78, 555.
(18) Ishikawa, A.; Takata, T.; N. Kondo, J.; Hara, M.; Kobayashi, H.;
Domen, K. J. Am. Chem. Soc. 2002, 124, 13547.
(19) Kudo, A.; Sekizawa, M. Chem. Commun. 2000, 1371.