Remarkable Charge Separation and Photocatalytic Efficiency Enhancement through Interconnection of TiO2 Nanoparticles by Hydrothermal Treatment

Article abstract of DOI:10.1002/anie.201510000

Although tremendous effort has been directed to synthesizing advanced TiO₂, it remains difficult to obtain TiO₂ exhibiting a photocatalytic efficiency higher than that of P25, a benchmark photocatalyst. P25 is composed of anatase, rutile, and amorphous TiO₂ particles, and photoexcited electron transfer and subsequent charge separation at the anatase-rutile particle interfaces explain its high photocatalytic efficiency. Herein, we report on a facile and rational hydrothermal treatment of P25 to selectively convert the amorphous component into crystalline TiO₂, which is deposited between the original anatase and rutile particles to increase the particle interfaces and thus enhance charge separation. This process produces a new TiO₂ exhibiting a considerably enhanced photocatalytic efficiency. This method of synthesizing this TiO₂, inspired by a recently burgeoning zeolite design, promises to make TiO₂ applications more feasible and effective. Connections matter: Hydrothermal treatment of P25 TiO₂ selectively converts the amorphous component into crystalline TiO₂, which is deposited between the original anatase and rutile component particles to increase the particle inter-faces, and thus considerably enhances charge separation and photocatalytic efficiency.

Full text of DOI:10.1002/anie.201510000

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201510000
German Edition: DOI: 10.1002/ange.201510000
Photocatalysis
Remarkable Charge Separation and Photocatalytic Efficiency
Enhancement through Interconnection of TiO2 Nanoparticles by
Hydrothermal Treatment
Yusuke Ide,* Nozomu Inami, Hideya Hattori, Kanji Saito, Minoru Sohmiya, Nao Tsunoji,
Kenji Komaguchi, Tsuneji Sano, Yoshio Bando, Dmitri Golberg, and Yoshiyuki Sugahara*
Abstract: Although tremendous effort has been directed to
synthesizing advanced TiO2, it remains difficult to obtain TiO2
exhibiting a photocatalytic efficiency higher than that of P25,
a benchmark photocatalyst. P25 is composed of anatase, rutile,
and amorphous TiO2 particles, and photoexcited electron
transfer and subsequent charge separation at the anatase–rutile
particle interfaces explain its high photocatalytic efficiency.
Herein, we report on a facile and rational hydrothermal
treatment of P25 to selectively convert the amorphous compo-
nent into crystalline TiO2, which is deposited between the
original anatase and rutile particles to increase the particle
interfaces and thus enhance charge separation. This process
produces a new TiO2 exhibiting a considerably enhanced
photocatalytic efficiency. This method of synthesizing this
TiO2, inspired by a recently burgeoning zeolite design,
promises to make TiO2 applications more feasible and
effective.
TiO2 particles in consideration of crystallinity (density of
defects acting as recombination centers), size, porosity, and
morphology.[6–11] Recently, designed TiO2 particle interfaces
have been shown to enhance the charge separation through
inter-particle transfer of photoexcited electrons,[12] as well as
to enhance electron mobility within dye-sensitized solar cells
composed of TiO2 mesocrystals (oriented superstructures of
nanocrystal building blocks).[13] Despite this progress in
advanced TiO2 synthesis, however, it remains difficult to
obtain TiO2 that exhibits a photocatalytic efficiency higher
than that of a commercial TiO2 (P25), one of the best TiO2
photocatalysts,[14] which has often been used as a benchmark
photocatalyst.
P25 is composed of nanometer-sized particles of anatase
and rutile (the main crystallographic forms of TiO2) and
amorphous TiO2, and the anatase and rutile particles are
partially interconnected.[15,16] It was pointed out that a transfer
of photoexcited electrons at the rutile and anatase particle
interfaces enhanced charge separation and, hence, produced
a high photocatalytic efficiency.[17] Coupling of anatase and
rutile particles has consequently been widely investigated as
a means of enhancing the photocatalytic efficiency,[18,19] but it
has remained difficult to synthesize coupled TiO2 photo-
catalysts that are superior to P25. Herein, we report on
a facile hydrothermal treatment of P25 to selectively dissolve
the amorphous TiO2 component and deposit crystalline TiO2
between the partially interconnecting anatase and rutile
particles, increasing the particle interfaces and thus enhancing
charge separation. This treatment produces a new TiO2
material with considerably enhanced photocatalytic effi-
ciency. Our synthetic method was inspired by a recent
O
wing to its combination of semiconducting properties with
low cost, non-toxicity, and long-term stability, TiO2 stands out
as one of the most promising materials for many applications
designed to deal with environmental and energy issues,
including photocatalytic degradation of organic pollutants,
H2 production through photocatalytic water splitting, and
photovoltaics.[1–5] When TiO2 is used as a photocatalyst,
electron–hole pairs generated by its band gap excitation are
transferred to surface adsorbed substrates with competition
of charge recombination; the photocatalytic efficiency of
TiO2 therefore depends largely on its charge separation
efficiency, surface areas, and exposed reactive facets. Tremen-
dous effort has thus been directed to synthesizing advanced
[*] Prof. Y. Ide, Prof. Y. Bando, Prof. D. Golberg
International Center for Materials Nanoarchitectonics (MANA)
National Institute for Materials Science (NIMS)
1-1 Namiki, Tsukuba, Ibaraki 305-0044 (Japan)
E-mail: IDE.Yusuke@nims.go.jp
E-mail: ys6546@waseda.jp
Prof. Y. Sugahara
Department of Applied Chemistry
School of Advanced Science and Engineering
Waseda University
N. Inami, K. Saito, Dr. M. Sohmiya
Department of Earth Sciences, Waseda University
1-6-1 Nishiwaseda, Shinjuku-ku, Tokyo 169–8050 (Japan)
3-4-1 Ohkubo, Shinjuku-ku, Tokyo 169-8555 (Japan)
and
Kagami Memorial Research Institute for Materials Science and
Technology
Waseda University
2-8-26 Nishiwaseda, Shinjuku-ku, Tokyo 169-0051 (Japan)
H. Hattori, Prof. N. Tsunoji, Prof. K. Komaguchi, Prof. T. Sano
Graduate School of Engineering, Department of Applied Chemistry
Hiroshima University
1-4-1 Kagamiyama, Higashi-Hiroshima 739-8527 (Japan)
Supporting information for this article can be found under
http://dx.doi.org/10.1002/anie.201510000.
Prof. Y. Ide, Prof. Y. Sugahara
Graduate School of Creative Science and Engineering
Waseda University
1-6-1 Nishiwaseda, Shinjuku-ku, Tokyo 169–8050 (Japan)
3600
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3600 –3605

Angewandte
Communications
Chemie
material design leveraging the chemically selective weakness
of solids.[20] A good example of this approach is the synthesis
of zeolites, microporous solids used as industrial catalysts, for
which synthesis of specific architectures is a critical and
challenging issue for attaining better performance and new
applications.[21] Cejka and co-workers demonstrated that
ˇ
selective removal of Ge from Ge-containing zeolites com-
À
posed of silica layers connected through Ge O bonds
(regioselective weakness), and then reintroduction of more
À
Si O bonds between the layers had produced zeolites with
a range of precisely controlled pore sizes.[22] We successfully
extended this material design using microscopic (inter-
particle) regioselective weakness as a synthetic tool to
a design using macroscopic (particle-level) regioselective
weakness.
Recently,[23] we have investigated hydrothermal treat-
ments of TiO2-based compounds such as layered titanates
using tetrapropylammonium hydroxide (TPA) and NH4F as
an alkali source to dissolve starting materials and a mineral-
izer to recrystallize the dissolved species, respectively.[24,25]
This led to the discovery of hydrothermal conditions under
which anatase and rutile are stable, whereas less chemically
stable TiO2-based compounds dissolve to recrystallize. In the
present study, we treated P25 under identical conditions to
selectively dissolve the amorphous TiO2 component of P25.
We named the present hydrothermal product Hyd-P25. X-ray
photoelectron spectroscopy revealed that C, N, and F were
not introduced into Hyd-P25 (Supporting Information, Fig-
ure S1). Small amounts of surface-adsorbed FÀ were detected
in Hyd-P25. This is due to residual NH4F and thus it scarcely
affects the photocatalytic activity. Scanning electron micros-
copy (SEM) and transmission electron microscopy (TEM)
confirmed that the morphology and size of the primary
particles of Hyd-P25 were almost identical to those of P25
(Figure 1a–d and e upper). In contrast, the size of the
aggregated particles of Hyd-P25 (500–600 nm), determined
by dynamic light scattering (DLS) analysis (Figure 1e lower),
was considerably larger than that of P25 (200–300 nm). The
specific surface area of Hyd-P25 (37 m2 gÀ1), determined by
N2 adsorption/desorption isotherms (Figure 1 f), was lower
than that of P25 (50 m2 gÀ1). This supports our claim that Hyd-
P25 is composed of larger aggregates than P25. This reduction
in the surface area, which generally lowers the efficiency of
semiconductor photocatalyst particles, did not significantly
affect the photocatalytic activity of Hyd-P25 by enhancing
charge separation efficiency through interconnection of the
primary particles (see below).
Figure 1. Morphology and structure characterizations of P25 and Hyd-
P25 particles. a) SEM image of P25. b) SEM image of Hyd-P25. c) TEM
image of P25. d) TEM image of Hyd-P25. e) Particle size distribution
in P25 (gray lines) and Hyd-P25 (blue lines) determined by TEM
(upper) and DLS (lower). f) N2 adsorption (open)/desorption (filled)
isotherms of P25 (gray) and Hyd-P25 (blue). Values in the graph are
the BET surface areas of the samples.
presence of thin layers of crystalline TiO2 between primary
crystalline particles (Figure 2b). In contrast, such layers were
not observed for P25. X-ray diffraction (XRD) analysis
showed that Hyd-P25 contained only anatase and rutile
crystalline phases, and that the diffraction peaks relevant to
anatase and rutile for Hyd-P25 were more intense than those
for P25 (blue vs. gray peaks, Figure 2c). In light of the fact that
the crystallinity of anatase and rutile in P25 is probably high
enough[14,27] to prevent recrystallization on their particle
edges under the present relatively mild hydrothermal con-
ditions,[23] all of the above results indicate that the amorphous
TiO2 particles in P25 were selectively dissolved to deposit
anatase or rutile thin layers on the original anatase and rutile
particles, and that the newly developed crystallites acted as
binders to increase the particle interfaces by linking the
original anatase and rutile particles. The interconnecting
aggregates of anatase and rutile particles for P25[16] conse-
quently became larger aggregated forms after the hydro-
thermal treatment (Figure 1e lower). Because P25 includes
up to approximately 15 wt% of amorphous TiO2 particles,[26]
we cannot rule out the possibility that the amorphous
Careful TEM observations revealed that Hyd-P25 had
a larger number of particle interfaces (tight inter-particle
contacts) between primary particles than P25 (yellow arrows
in Figure 1c and d). The numbers of such particle interfaces
were 44 and 21 per 100 particles for Hyd-P25 and P25,
respectively
(Figures S2–S5).
High-resolution
TEM
(HRTEM) revealed that Hyd-P25 contained scarcely any
amorphous TiO2 particles, which are typically observed for
P25 (Figure 2a). Composition analyses (anatase:rutile:amor-
phous ratio in weight) of Hyd-P25 (84:16:0) and P25
(76:13:11) using a reported method[26] supported these
observations. The HRTEM image of Hyd-P25 showed the
Angew. Chem. Int. Ed. 2016, 55, 3600 –3605
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3601

Angewandte
Communications
Chemie
Figure 2. Characterization of amorphous TiO2 contained in P25 and its
conversion into anatase and rutile. a) HRTEM of P25. b) HRTEM of
Hyd-P25. c) XRD patterns of P25 (gray line) and Hyd-P25 (blue line).
The asterisk indicates a diffraction peak from Al2O3 used as a standard
for normalizing the XRD patterns.
component is converted into anatase and rutile particles
whose size is almost identical to those of the primary anatase
and rutile particles in P25, in addition to thin layers (Fig-
ure 2b).
Control experiments were conducted to shed light on the
formation mechanism of Hyd-P25. When P25 was hydro-
thermally treated with either TPA or NH4F, or without either
agent, the DLS particle sizes and XRD patterns of the
products were no different from those of P25. These results
imply that the amorphous TiO2 contained in P25 dissolves in
a TPA alkali solution to recrystallize into anatase and rutile
with the aid of NH4F mineralizer.
To obtain a deeper understanding of the formation
mechanism of Hyd-P25, additional hydrothermal experi-
ments were conducted using commercially available anatase,
rutile, and amorphous TiO2 particles under conditions iden-
tical to those used for the synthesis of Hyd-P25. First, we
hydrothermally treated only amorphous TiO2. XRD (data not
shown) and TEM (Figures S6 and S7) revealed that almost all
of the amorphous phase was converted into anatase and rutile
under the hydrothermal conditions (these particles were
different in morphology and size from thin layers of
crystalline TiO2 observed in Hyd-P25; Figure 2b). We then
hydrothermally treated a physical mixture of anatase, rutile,
and amorphous TiO2, the composition of which was identical
to that reported for P25 (anatase:rutile:amorphous =
73:14:13 in weight),[26] to investigate whether amorphous
TiO2 might be selectively converted into crystalline TiO2 to
bind anatase and rutile particles into larger aggregated forms.
Hydrothermal treatment of the physical mixture produced
a product (Hyd-AnRuAm) with a particle size larger than
that of the original anatase and rutile particles (each TiO2 is
aggregated), while the size of the untreated physical mixture
(AnRuAm) was similar to that of the main anatase compo-
Figure 3. Control hydrothermal reactions using commercial materials
of anatase, rutile, and amorphous TiO2. a) DLS of anatase (red), rutile
(green), Hyd-AnRuAm (purple), and AnRuAm (dashed purple line).
b) TEM of Hyd-AnRuAm. c) HRTEM of Hyd-AnRuAm. The inset shows
an expanded image of the area indicated by a yellow square.
nent, as seen in the DLS analysis (Figure 3a). TEM revealed
Hyd-AnRuAm to be composed of anatase and rutile particles
whose morphology and size were similar to those of the
original anatase and rutile primary particles (Figure S8), with
no amorphous TiO2 particle content (Figure 3b). Moreover,
HRTEM indicated that Hyd-AnRuAm had tightly contacted
anatase and rutile particles that were bound by smaller TiO2
crystallites (Figure 3c). These results support the hypotheses
that the amorphous TiO2 component of P25 also dissolves,
and that the recrystallized TiO2 thin layers bind the original
anatase and rutile particles.
We further investigated the charge separation efficiency
of Hyd-P25. For P25, the charges generated on rutile are
separated through electron transfer to anatase trapping sites,
or vice versa, between interconnecting rutile and anatase
particles, and this electron transfer explains the high photo-
catalytic efficiency of P25.[17] Here, we compared the charge
separation efficiency of Hyd-P25 and P25 by monitoring the
yields of radical species, such as trapped electrons, using
electron spin resonance (ESR) analysis. As shown in Fig-
ure 4a and b, when irradiated by UV light (l > 330 nm) in the
presence of molecular O2 as an electron acceptor, Hyd-P25
created considerably larger amounts of Ti3+, which were
formed by trapping the photoexcited electrons. Judging from
3602 www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3600 –3605

Angewandte
Communications
Chemie
catalytic activity of Hyd-P25 with
respect to the oxidative decompo-
sition of acetic acid in water to
evolve CO2, which is a well-known
reaction for testing the perfor-
mance of photocatalysts.[14] TiO2
samples suspended in aerated aque-
ous solutions of acetic acid were
irradiated by full-spectrum simu-
lated solar light (l > 300 nm, 1 Sun
(1000 WmÀ2) power), allowing band
gap (UV light) excitation of the
TiO2. As shown in Figure 4d, Hyd-
P25 exhibited a rate of CO2 evolu-
tion about five times higher than
that of P25, which is one of the best
TiO2 photocatalysts for the reac-
tion.[14] To our knowledge, only
a
small number of synthesized
TiO2 materials[6] have shown
higher activity toward the reaction
than P25, and the activity of Hyd-
P25 is by far the highest ever
reported. It should be noted here
that the photocatalytic activity of
Hyd-AnRuAm was nearly twice as
high as AnRuAm, suggesting the
versatility of the present hydrother-
mal reaction using chemically selec-
tive weak particles (even those
added later).
Figure 4. Photoproperties and photocatalytic performances of P25 and Hyd-P25. a) ESR spectra of
P25 (gray) and Hyd-P25 (blue). b) Ti3+ and O2À yields on P25 (gray) and Hyd-P25 (blue) determined
from ESR spectra. c) UV/Vis spectra of P25 (gray) and Hyd-P25 (blue). d) Time course of photo-
catalytic CO2 evolution from aqueous solutions of acetic acid containing P25 (gray) and Hyd-P25
(blue) under simulated solar light irradiation (l>300 nm, 1 Sun). e) Time course of photodegrada-
tion of methylene blue in aqueous solutions containing P25 (gray) and Hyd-P25 (blue) under
simulated solar light irradiation (l>300 nm, 1 Sun).
the g-values, two different Ti3+ types are formed in anatase for
Hyd-P25. One comprises the surface electron trapping sites
(g? = 1.980), and the other type comprises the lattice electron
trapping sites (g? = 1.990).[28] Because the structural regular-
ities of the interface TiO2 thin layers are similar to those of the
surface rather than the bulk, the notably larger amount of the
former suggests that a marked increase in the number of
particle interfaces plays an important role in trapping the
We also examined the photocatalytic degradation of an
organic dye, methylene blue, in water by Hyd-P25. Aqueous
solutions of methylene blue containing TiO2 photocatalysts
were irradiated by 1 Sun power-simulated solar light (l >
300 nm). As shown in Figure 4e, photodegradation was
almost complete after 40 min for Hyd-P25, whereas it took
nearly 2 hours for P25 under the same testing conditions,
indicating that Hyd-P25 demonstrated remarkably enhanced
activity compared to P25 for this reaction as well.
À
electrons in Hyd-P25. The yields of superoxide anions (O2 ),
on the other hand, which were formed for Hyd-P25 by
reduction of molecular O2 with photoexcited electrons on the
anatase surface,[29] were closely comparable to those for P25.
These results indicate preferential trapping of photogener-
ated electrons in newly developed anatase thin layers in Hyd-
P25, as well as a specific interfacial transfer of photogenerated
electrons from rutile to anatase. For these reasons, photo-
generated electrons and holes are separated more efficiently
in Hyd-P25 than in P25. It should be noted here that the UV/
Vis spectrum of Hyd-P25 was almost identical to that of P25
(Figure 4c), suggesting a close similarity in the anatase–rutile
ratios, which also affects the radical species yields (amorphous
TiO2 is known to be photocatalytically inactive[30]), and thus
the photocatalytic efficiency, between the two TiO2 forms.
The UV/Vis spectrum also confirmed that Hyd-P25 intro-
duced very few elements, such as N and F, which often affect
the photocatalysis of TiO2.
It has been reported that the composition of as-supplied
P25 in the same package was inhomogeneous; the amorphous
phase was sometimes not found, even if the sample powder
was collected from the same package.[26] In the present study,
we examined the reproducibility of Hyd-P25 to find that the
hydrothermal products synthesized from P25 powders col-
lected from the same position in a package with 11 wt% of
amorphous TiO2 showed similar results with respect to
syntheses, properties, and photocatalytic performance.
When using P25 powders collected from another position in
the same package with 6 wt% of amorphous TiO2, on the
other hand, we obtained a hydrothermal product showing
a photocatalytic activity lower (ca. 70%) than Hyd-P25.
These results are again indicative of the importance of
amorphous TiO2 to the interconnection of the primary
particles to enhance charge separation efficiency.
In summary, we successfully synthesized a TiO2 material,
Hyd-P25, which exhibits considerably enhanced charge
separation efficiency and photocatalytic efficiency by a hydro-
As expected, Hyd-P25 showed considerably higher photo-
catalytic efficiency than P25. We first examined the photo-
Angew. Chem. Int. Ed. 2016, 55, 3600 –3605
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3603

Angewandte
Communications
Chemie
irradiated by a solar simulator (San-ei Electric, XES-155S1) under
stirring. Acetic acid adsorption was negligible under these conditions.
The headspace CO2 was quantified by a Shimadzu GC-2010 plus gas
chromatograph equipped with a BID detector. Photocatalytic degra-
dation of methylene blue was performed as follows. A suspension
containing an aqueous solution of methylene blue (0.05 ppm, 40 mL)
and TiO2 (30 mg) placed in a Pyrex glass container (50 mL) was
irradiated with a solar simulator under stirring. The degree of
photodegradation was measured by monitoring the changes in
absorption of a methylene blue solution at 664 nm with an UV/Vis
spectrometer.
thermal treatment of the benchmark TiO2 photocatalyst P25.
The hydrothermal reaction involved selective dissolution of
the amorphous TiO2 component present in P25 to convert it
into anatase and rutile particles that bind with the partially
interconnected anatase and rutile particles to remarkably
increase the particle interfaces. This approach, using chemi-
cally selective weakness (amorphous TiO2 particles) in
a powder material (P25), was inspired by a recent design of
zeolites and metal–organic frameworks (MOFs), in which
chemically selective weaknesses in the individual particles in
a material are used to design new architectures and func-
tions.[20] Designing TiO2 nanoparticle interfaces, including
synthesis of TiO2 mesocrystals, has recently been a crucial
issue for enhancing the charge separation efficiency and
electron mobility (long-range electron connectivity) of photo-
catalysts and dye-sensitized solar cells, respectively.[7,12,13]
Although several methodologies have been reported for
such purposes, they often involve complicated procedures.
The present hydrothermal method is extremely simple and
offers the potential for large-scale production. Incidentally, it
was reported that heating of P25 powders in air at 973 K
produced anatase particles with a primary particle size and
photocatalytic activity almost identical to that of P25.[31] It is
thus difficult for P25 to form tight inter-particle contacts that
can enhance charge separation efficiency by simple sintering
at high temperatures. In addition to highly efficient photo-
catalysts and dye-sensitized solar cells, TiO2 materials exhib-
iting long-lived charge separation can be used as visible-light-
responsive photocatalysts, after surface modification, for the
efficient utilization of solar or indoor light.[32] Further
functionalization of Hyd-P25 and the extension of its
synthetic method to other TiO2 materials are currently
under investigation in our laboratory.
Acknowledgements
This work was partly supported by JSPS KAKENHI Grant
Number 26708027.
Keywords: charge separation · dye-sensitized solar cell ·
photocatalysis · titanium dioxide · zeolite
How to cite: Angew. Chem. Int. Ed. 2016, 55, 3600–3605
Angew. Chem. 2016, 128, 3664–3669
[1] A. Fujishima, X. T. Zhang, D. A. Tryk, Surf. Sci. Rep. 2008, 63,
515.
[2] A. Kudo, Y. Miseki, Chem. Soc. Rev. 2009, 38, 253.
[3] Y. Shiraishi, T. Hirai, J. Photochem. Photobiol. C 2008, 9, 157.
[4] H. Tada, T. Kiyonaga, S. Naya, Chem. Soc. Rev. 2009, 38, 1849.
[5] A. Hagfeldt, G. Boschloo, L. C. Sun, L. Kloo, H. Pettersson,
Chem. Rev. 2010, 110, 6595.
[6] H. Kominami, T. Matsuura, K. Iwai, B. Ohtani, S. Nishimoto, Y.
Kera, Chem. Lett. 1995, 693.
[7] Z. Bian, J. Zhu, J. Wen, F. Cao, Y. Huo, X. Qian, Y. Cao, M. Shen,
H. Li, Y. Lu, Angew. Chem. Int. Ed. 2011, 50, 1105; Angew.
Chem. 2011, 123, 1137.
[8] H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zou, G. Liu, S. C. Smith,
H. M. Cheng, G. Q. Lu, Nature 2008, 453, 638.
Experimental Section
[9] F. Amano, O. O. Prieto-Mahaney, Y. Terada, T. Yasumoto, T.
Shibayama, B. Ohtani, Chem. Mater. 2009, 21, 2601.
[10] E. Bae, N. Murakami, T. Ohno, J. Mol. Catal. A 2009, 300, 72.
[11] J. M. Macak, M. Zlamal, J. Krysa, P. Schmuki, Small 2007, 3, 300.
[12] Z. Bian, T. Tachikawa, T. Majima, J. Phys. Chem. Lett. 2012, 3,
1422.
[13] E. J. W. Crossland, N. Noel, V. Sivaram, T. Leijtens, J. A.
Alexander-Webber, H. J. Snaith, Nature 2013, 495, 215.
[14] O. O. Prieto-Mahaney, N. Murakami, R. Abe, B. Ohtani, Chem.
Lett. 2009, 38, 238.
[15] T. Ohno, K. Sarukawa, K. Tokieda, M. Matsumura, J. Catal.
2001, 203, 82.
[16] R. I. Bickley, T. Gonzalez-Carreno, J. S. Lees, L. Palmisano,
R. J. D. Tilley, J. Solid State Chem. 1991, 92, 178.
[17] D. C. Hurum, A. G. Agrios, K. A. Gray, T. Rajh, M. C. Thurna-
uer, J. Phys. Chem. B 2003, 107, 4545.
Hydrothermal reactions: P25 (supplied by Nippon Aerosil), TPA,
H2O, and NH4F were mixed at
a
molar ratio of
TiO2:TPA:H2O:NH4F = 1:0.8:5:0.2 in a Teflon-lined stainless-steel
autoclave, and the mixture was treated at 443 K for 7 days. The
product was washed repeatedly with water and dried. Control
experiments were conducted without TPA and/or NH4F. Amorphous
TiO2 (Wako Pure Chemical) or a physical mixture of amorphous
TiO2, anatase and rutile (JRC TIO-1 and JRC TIO-6, respectively,
supplied by the Catalysis Society of Japan) was also hydrothermally
treated under identical conditions.
ESR analysis: ESR measurements were conducted using a JEOL
JES RE-1X spectrometer (X-band). The magnetic field was cali-
brated and the radical yields were determined by ESR spectra
recorded with a Mn2+/MgO marker as an external standard reference.
TiO2 powder (10 mg) was placed in a Suprasil ESR tube (external
diameter 5 mm), which was evacuated at 423 K for 1 h and cooled to
room temperature. O2 (20 torr) was introduced into the tube and
retained for 10 min. The sample was photoirradiated at room
temperature using a 500 W Xe lamp at l > 330 nm for 5 min. The
sample was then evacuated for 10 min to remove excess O2 and
subjected to ESR measurements at 77 K.
[18] T. Kawahara, Y. Konishi, H. Tada, N. Tohge, J. Nishii, S. Ito,
Angew. Chem. Int. Ed. 2002, 41, 2811; Angew. Chem. 2002, 114,
2935.
[19] T. Ohno, K. Tokieda, S. Higashida, M. Matsumura, Appl. Catal.
A 2003, 244, 383.
ˇ
[20] R. E. Morris, J. Cejka, Nat. Chem. 2015, 7, 381.
Photocatalytic reactions: Photocatalytic acetic acid oxidation was
conducted as follows. An aqueous solution of acetic acid (5 vol%,
5 mL), saturated with molecular O2, was mixed with TiO2 (15 mg) in
a stainless-steel container (75 mL) equipped with a Pyrex glass
window by ultrasonication for 1 min, and the suspension was
[21] C. S. Cundy, P. A. Cox, Chem. Rev. 2003, 103, 663.
[22] W. J. Roth, P. Nachtigall, R. E. Morris, P. S. Wheatley, V. R.
ˇ
Seymour, S. E. Ashbrook, P. Chlubnµ, L. Grajciar, M. Polozij, A.
ˇ
Zukal, O. Shvets, J. Cejka, Nat. Chem. 2015, 7, 628.
[23] H. Hattori, Y. Ide, T. Sano, J. Mater. Chem. A 2014, 2, 16381.
3604 www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3600 –3605

Angewandte
Communications
Chemie
[24] S. Shibata, M. Itakura, Y. Ide, M. Sadakane, T. Sano, Micro-
[30] B. Ohtani, Y. Ogawa, S. Nishimoto, J. Phys. Chem. B 1997, 101,
3746.
porous Mesoporous Mater. 2011, 138, 32.
[25] K. Honda, M. Itakura, Y. Matsuura, A. Onda, Y. Ide, M.
Sadakane, T. Sano, J. Nanosci. Nanotechnol. 2013, 13, 3020.
[26] B. Ohtani, O. O. Prieto-Mahaney, D. Li, R. Abe, J. Photochem.
Photobiol. A 2010, 216, 179.
[31] T. Khoa Le, D. Flahaut, H. Martinez, H. K. Hung Nguyen, T. K.
Xuan Huynh, Appl. Catal. B 2015, 165, 260.
[32] Z. Bian, T. Tachikawa, P. Zhang, M. Fujitsuka, T. Majima, J. Am.
Chem. Soc. 2014, 136, 458.
[27] Y. Shiraishi, H. Hirakawa, Y. Togawa, Y. Sugano, S. Ichikawa, T.
Hirai, ACS Catal. 2013, 3, 2318.
[28] T. Rajh, A. E. Ostafin, O. I. Micic, D. M. Tiede, M. C. Thurna-
uer, J. Chem. Phys. 1996, 100, 4538.
[29] M. Chiesa, M. C. Paganini, S. Livraghi, E. Giamello, Phys. Chem.
Chem. Phys. 2013, 15, 9435.
Received: October 26, 2015
Revised: January 28, 2016
Published online: February 17, 2016
Angew. Chem. Int. Ed. 2016, 55, 3600 –3605
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3605