Titanium-defected undoped anatase TiO2 with p-type conductivity, room-temperature ferromagnetism, and remarkable photocatalytic performance

Article abstract of DOI:10.1021/ja512047k

Defects are critically important for metal oxides in chemical and physical applications. Compared with the often studied oxygen vacancies, engineering metal vacancies in n-type undoped metal oxides is still a great challenge, and the effect of metal vacancies on the physiochemical properties is seldom reported. Here, using anatase TiO₂, the most important and widely studied semiconductor, we demonstrate that metal vacancies (V_Ti) can be introduced in undoped oxides easily, and the presence of V_Ti results in many novel physiochemical properties. Anatase Ti_0.905O₂ was synthesized using solvothermal treatment of tetrabutyl titanate in an ethanol-glycerol mixture and then thermal calcination. Experimental measurements and DFT calculations on cell lattice parameters show the unstoichiometry is caused by the presence of V_Ti rather than oxygen interstitials. The presence of V_Ti changes the charge density and valence band edge of TiO₂, and an unreported strong EPR signal at g = 1.998 presents under room temperature. Contrary to normal n-type and nonferromagnetic TiO₂, Ti-defected TiO₂ shows inherent p-type conductivity with high charge mobility, and room-temperature ferromagnetism stronger than Co-doped TiO₂ nanocrystalline. Moreover, Ti-defected TiO₂ shows much better photocatalytic performance than normal TiO₂ in H₂ generation (4.4-fold) and organics degradation (7.0-fold for phenol), owing to the more efficient charge separation and transfer in bulk and at semiconductor/electrolyte interface. Metal-defected undoped oxides represent a unique material; this work demonstrates the possibility to fabricate such material in easy and reliable way and thus provides new opportunities for multifunctional materials in chemical and physical devices.

Full text of DOI:10.1021/ja512047k

Subscriber access provided by SELCUK UNIV
Article
Titanium-defected undoped anatase TiO2 with p-type conductivity, room-
temperature ferromagnetism and remarkable photocatalytic performance
Songbo Wang, Lun Pan, Jia-Jia Song, Wenbo Mi, Ji-Jun Zou, Li Wang, and Xiangwen Zhang
J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 06 Feb 2015
Downloaded from http://pubs.acs.org on February 7, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Titanium-defected undoped anatase TiO2 with p-type conduc-
tivity, room-temperature ferromagnetism and remarkable pho-
tocatalytic performance
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Songbo Wang,†,‡,ǁ Lun Pan,†,‡,ǁ Jia-Jia Song,†,‡ Wenbo Mi,§ Ji-Jun Zou*,†,‡, Li Wang,†,‡ and
Xiangwen Zhang†,‡
† Key Laboratory for Green Chemical Technology of the Ministry of Education, School of Chemical Engineering and
Technology, Tianjin University, Tianjin 300072, China.
‡ Collaborative Innovative Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, China.
§
Tianjin Key Laboratory of Low Dimensional Materials Physics and Preparation Technology, Faculty of Science,
Tianjin University, Tianjin 300072, China.
KEYWORDS: anatase TiO2, metal vacancy, p-type conductivity, ferromagnetism, photocatalysis.
ABSTRACT: Defects are critically important for metal oxides in chemical and physical applications. Compared with the
often studied oxygen vacancies, engineering metal vacancies in n-type undoped metal oxides is still a great challenge, and
the effect of metal vacancies on the physiochemical properties is seldom reported. Here, using anatase TiO2, the most
important and widely studied semiconductor, we demonstrate that metal vacancies (VTi) can be introduced in undoped
oxides easily, and the presence of VTi results in many novel physiochemical properties. Anatase Ti0.905O2 was synthesized
using solvothermal treatment of tetrabutyl titanate in ethanol-glycerol mixture and then thermal calcination. Experi-
mental measurements and DFT calculations on cell lattice parameters show the un-stoichiometry is caused by the pres-
ence of VTi rather than oxygen interstitials. The presence of VTi changes the charge density and valence band edge of TiO2,
and an un-reported strong EPR signal at g=1.998 presents under room temperature. Contrary to normal n-type and non-
ferromagnetic TiO2, Ti-defected TiO2 shows inherent p-type conductivity with high charge mobility, and room-
temperature ferromagnetism stronger than Co-doped TiO2 nanocrystalline. Moreover, Ti-defected TiO2 shows much bet-
ter photocatalytic performance than normal TiO2 in H2 generation (4.4-fold) and organics degradation (7.0-fold for phe-
nol), owing to the more efficient charge separation and transfer in bulk and at semiconductor/electrolyte interface. Metal-
defected undoped oxides represent a unique material; this work demonstrates the possibility to fabricate such material in
easy and reliable way and thus provides new opportunities for multifunctional materials in chemical and physical devices.
1. INTRODUCTION
ficult to engineer and manipulate metal defects in a stable
and reliable way.12
The physicochemical properties of metal oxides large-
ly depend on the concentration of intrinsic defects and
extrinsic impurities, and it has been demonstrated that
pure stoichiometric metal oxides are inactive in catalysis.1-
Since the discovery of photoelectrochemical (PEC) wa-
ter splitting in 1972,13 anatase TiO2 has attracted intensive
attention in photocatalytic hydrogen generation, dye sen-
sitized solar cell, pollutants removal and CO2 reduc-
tion.11,14-15 Many approaches have been adopted to improve
its photocatalytic performance such as tuning oxygen
defects,16 improving the crystallinity,17 utilizing slow pho-
ton effect,18-19 exposing active facet,20-21 and surface
flurination.22-23 Actually, the photocatalytic performance
of TiO2 greatly depends on the optical and electrical
properties that are mostly determined by the crystal
structure and defect.24 TiO2 inherently exhibits n-type
conductivity due to the existence of VO. When the con-
centration of acceptor dopants increases to a suitable lev-
el,8,25-26 the Fermi level of TiO2 will shift towards valence
band (VB) edge and its conductivity can be altered from
n- to p-type. For example, both Cr- and Fe-doped TiO2
3
Defect engineering is a key factor to tune the functional
properties of oxides, such as electronic structure, charge
transport, and catalytic performance.4-5 Specifically for
semiconductors, density functional theory (DFT) calcula-
tions predict that both oxygen vacancies (VO) and metal
interstitials are shallow donors responsible for n-type
conductivity, whereas metal vacancies (VM) are shallow
acceptors contributing to p-type conductivity.6 In the past
decades, VO related species are thought to be very im-
portant and techniques for VO engineering have been ex-
tensively studied,7-11 but only a few literatures focus on VM
control. Although some theoretical computations have
been conducted to study the role of VM in metal oxides,
experimental results are quite rare because it is quite dif-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
show p-type conductivity,4 and Cr-doped TiO2 photoelec-
us to explore the possibility of introducing VTi in TiO2
trodes exhibits photocathodic current.25 However, foreign
acceptors may have deteriorating effect on charge separa-
tion and transfer.27-28 Titanium vacancy (VTi, Scheme 1) is
acceptor type intrinsic defect, so increasing its concentra-
tion may lead to n→p transition and thus influence the
mobility of charge carriers. Besides, as the only negative
ionic defect in undoped TiO2, VTi has been predicted as
active site for water splitting.7,29 However, the study on
undoped anatase p-type TiO2 is quite rare, let alone its
performance in photocatalysis. Nowotny et al reported
that, annealing TiO2 at 1323 K for 3500 h can lead to Ti-
deficient oxide with p-type conductivity,27,30 but the ob-
tained TiO2 is not anatase phase any more. Bhattacharyya
and Basu et al reported undoped anatase p-type TiO2 thin
films deposited on p-Si substrate but neither checked the
structure characteristic nor explained the reason for the
p-type conductivity.31-33
that is the most important semiconductor oxide in both
chemical and physical applications. Herein, we report the
successful fabrication of undoped anatase TiO2 abundant
with VTi using simple solvothermal method, as well its
structure characteristics, electrical conductivity, ferro-
magnetism and photocatalytic performance, based on
experiments and density functional theory simulations.
The metal-defected TiO2 represents a new kind of semi-
conductor and provides new opportunities for TiO2-based
materials.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2. EXPERIMENTAL SECTION
2.1. Materials
Tetrabutyl titanate [Ti(OC4H9)4, TBT], ethanol,
glycerol, methanol, H2PtCl6·6H2O, rhodamine B (RhB),
and methyl orange (MO) were all purchased from Tianjin
Guangfu Fine Chemical Research Institute, China. Phenol
was obtained from J&K Chemical. Milli-Q ultra pure water
with a resistivity higher than 18.2 MΩ·cm was used in all
experiments. All the reagents were reagent grade and
used as received.
2.2. Synthesis of Ti-defected TiO2
2 g TBT was added into a solution of 20 mL glycerol
and 60 mL ethanol under magnetic stirring, then the so-
lution was transferred into a 100 mL Teflon-lined auto-
clave and heated at 180 °C for 24 h. The obtained powders
were collected, washed with absolute ethanol for several
times, dried at 60 °C for 12 h, and finally calcined in air at
470 °C for 1 h with a heating rate of 5 °C/min (referred as
defected TiO2). As reference, normal TiO2 was synthe-
sized with the same procedure but without glycerol in the
solvothermal treatment.
Scheme 1. Schematic representation of different defects in
undoped TiO2. VTi: Ti vacancy, VO: oxygen vacancy, Tiint: Ti
interstitial, Oint: oxygen interstitial, Ti3+: Ti4+ ion accepting
one electron, O−: O2− donating one electron.
Room-temperature ferromagnetic (RT-FM) oxides
have attracted great interests from magnetic fluids, bio-
medicine, magnetic resonance imaging, catalysis, and
environmental remediation.34 The discovery of carrier
mediated ferromagnetism at high temperature (~170 K)
inspires numerous attempts to construct diluted magnet-
ic semiconductors with RT-FM.35 Koinuma et al reported
cobalt-doped anatase thin film with Curie temperature
(TC) more than 400 K.36 But recently debate arises on
whether the ferromagnetism is really caused by the ho-
mogeneous Ti1-xCoxO2 or just by the dopants themselves.37
Meanwhile, the key of high TC is a large density of states
at the Fermi level, which is easier to achieve in p-type
semiconductors.35 Also, owing to the electronic structure
changed by defects, metal-defected undoped oxides may
show intrinsic high-TC ferromagnetism.38 Therefore, p-
type metal oxide with RT-FM are potential candidates of
multifunctional materials in both chemical and physical
devices.
2.3. Structure characterizations
XRD patterns were recorded using D/MAX-2500 X-ray
diffractometer equipped with Cu Kα radiation at 40 kV
and 140 mA at a scanning rate of 5 º/min. TEM analysis
was carried out using a Tecnai G2 F-20 transmission elec-
tron microscope with a field-emission gun operating at
200kV. Electron paramagnetic resonance (EPR) meas-
urement was performed with a Bruker EMX-6/1 spec-
trometer at 9.424 GHz (120 K) or 9.868 GHz (298 K) using
a 100 kHz field modulation and a 0.5 G standard modula-
tion width. X-ray photoelectron spectrum (XPS) analysis
was conducted with a PHI-1600 X-ray photoelectron spec-
troscope equipped with Al Kα radiation, and the binding
energy was calibrated by the C1s peak (284.8 eV) of con-
tamination carbon. Ti/O ratios in bulk and on the surface
were analyzed by titration method with deviation <0.5%
and XPS analysis, respectively (Details are listed in Sup-
porting Information (SI) and Figure S1). Specific surface
area (SBET
)
was calculated based on N2 adsorp-
Recently, we developed a facile method to fabricate
ZnO crystals abundant with Zn vacancies, and observed
p-type conductivity, RT-FM and excellent photocatalytic
activity.39 It demonstrates the possibility of reliable VM
engineering in oxides, and more importantly, suggests the
resulted materials may exhibit many interesting proper-
ties that have much potential in application. This inspires
tion/desorption isotherms recorded on a Micromeritics
TriStar 3000 instrument at 77 K, all samples were out-
gassed under vacuum at 300°C for 4 h. Steady-state pho-
toluminescence (PL) spectra were measured by a Horiba
JobinYvon Fluorolog3-21 with the excitation light at 325
nm. Positron annihilation lifetime was measured with a
fast/slow coincidence ORTEC system with a time resolu-
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
generalized gradient approximation was used for the
tion of ~201 ps full width at half-maximum. A 5×10-5-Bq
source of 22Na was sandwiched between two identical
samples (10×10 mm, 1 mm thickness).
exchange-correlation potential,43 and the plane-wave
cutoff was set to 400 eV together with performing a G-
centered 7×7×3 k mesh Brillouin-zone integration to ob-
tain accurate structure for all systems. The convergence
criterion for minimization of the Hellmann-Feynman
force was set to less than 0.01 eV/Å.
1
2
3
4
5
2.4. Measurements of electrical and magnetic proper-
ties
Electrochemical properties were measured using an
Autolab potentiostat/galvanostat (Model PGSTAT 302N)
in a three-electrode cell with a Pt sheet as the counter
electrode and an Ag/AgCl reference electrode. Na2SO4
(0.2 M) was used as electrolyte solution. The working
electrode was prepared by dip-coating TiO2 slurry on an
F-doped tin oxide (FTO) glass electrode (1 cm× 1 cm) and
heating at 40 °C overnight. All investigated working elec-
trodes were of similar thickness. Electrochemical imped-
ance spectroscopy (EIS) measurements were carried out
with a sinusoidal ac perturbation of 10 mV applied over
the frequency range of 0.01-105 Hz. Hall-effect measure-
ments were carried out in the van der Pauw configuration
(HL5550PC, Accent Optical Technologies Ltd., UK) at
room temperature. The contacts were applied on the four
corners of 0.5×0.5 cm2 pressed pellets that were annealed
at 300oC for 1 h in air. Magnetic properties were measured
using Qantum Design SQUID-VSM at a maximum field of
20 kOe in temperature range of 10-350 K.
6
7
8
9
A 2×2×1 anatase supercell was used for the calculation,
with one Ti atom removed to construct Ti-defected
crystal (Ti15O32, Ti/O=0.469) and one O atom inserted in
the interstitial position to construct O-interstitial crystal
(Ti16O33, Ti/O=0.485). The lattice parameters were
calculated with DFT-GGA method that provides accurate
lattice constants but underestimates the band gap. The
GGA+U method can improve the prediction of band
structure,44 so it was applied with U= 7.2 eV for Ti atom in
the static state and density of states calculation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3. RESULTS AND DISCUSSIONS
3.1. Formation of anatase TiO2
The intermediated powders formed by solvothermal
treatment of tetrabutyl titanate (TBT) in glycerol-ethanol
mixture show a sharp peak at 2θ = ca. 10° and several
weak peaks at higher 2θ degree (Figure S2, SI). Although
the pattern can not be indexed to any known structure in
database, some literatures refer it as titanium glycerolate
(TiGly).45-47 The solid state CP/MAS 13C NMR spectrum
(Figure S3, SI) shows three well resolved peaks at 73.8,
80.5 and 85.3 ppm, in accordance with the reported NMR
data of TiGly.46 From TG analysis (Figure 1a), the weight
50
2.5. Photocatalytic tests
Photodegradation of organic pollutants was
conducted in an opening quartz chamber (150 mL)
vertically irradiated by a 300 W high-pressure Xenon
lamp (PLS-SXE 300UV, Beijing Trusttech. Co. Ltd.)
located on the upper position. The light density was 35.2
mW/cm2 at 365 nm, and 33.5 mW/cm2 at 420 nm as
measured with a radiometer (Photoelectric Instrument
Factory, Beijin Normal University, Model UV-A). The
irradiation area was ca. 20 cm2. Reaction conditions:
(a)
(b)
100
90
80
70
60
50
40
(101)
Powders without
glycerol TG
40
30
20
10
0
Powders with
glycerol TG
Normal TiO
2
Defected TiO
2
temperature, 25± 0.2 C; C0(RhB)= 20 μmol·L–1; C0(MO)=
o
24.0 24.5 25.0 25.5 26.0 26.5 27.0
2θ
/degree
(200)
120 μmol·L–1; C0(Phenol)= 400 μmol·L–1; TiO2: 0.2 g·L–1; no
acid or alkaline reagents were added. For self-sensitized
degradation of RhB under visible light, an UV-420 cut was
equipped to eliminate UV light. Reaction was conducted
by magnetic stirring under atmosphere after stirring for
20 minutes in darkness to achieve adsorption equilibrium.
Samples were withdrawn, centrifuged and analyzed using
UV-vis spectrometer (U-3010, Hitachi Ltd.).
(004)
(204)
Powders with
glycerol DTA
(215)
(224)
20
30
40
50
60
70
80
90
0
100
200 300 400 500 600
Temperature/oC
2θ /degree
Photocatalytic hydrogen production was carried out in
a Pyrex top-irradiation reaction vessel connected to
closed glass gas system. 10 mg catalyst were dispersed in
120 mL aqueous solution containing methanol (30 vol.%).
1.0 wt.% Pt were introduced as co-catalyst by in-situ
photodeposition. The temperature of reaction solution
was maintained at 0 °C. The resultant hydrogen was
analyzed using an on-line gas chromatography (Bruker
450-GC, thermal conductive detector (TCD),
molecular sieve column, and N2 as carrier gas).
5 Å
Figure 1. (a) TG and DTA profiles of intermediated powders
produced by solvothermal treating TBT with and without
glycerol. (b) XRD patterns of defected and normal TiO2. In-
set: Enlarged XRD patterns of peak (101). (c), (d) TEM images
of defected and normal TiO2. Inset in (c): Fast-Fourier trans-
form pattern of defected TiO2.
2.6. Density functional theory calculations
The calculations were based on density functional
theory and the projector augmented wave method
implemented in the Vienna Ab initio Simulation
Package.40-44 The Perdew-Burke-Ernzerhof spin-polarized
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
80 (a)
80 (b)
loss of the powders in the range of 200-470 °C is 52.9%,
which is close to the theoretical weigh loss (52.1%) of TiG-
ly to TiO2.45
-0.5
-1.0
-1.5
-2.0
-2.5
1
2
3
4
60
40
20
0
60
40
20
0
g=1.990
3380 3400 3420 3440
Magnetic Field/Gauss
After calcination at 470 °C for 1 h, the final product
(referred as defected TiO2) can be undoubtedly identified
as anatase TiO2 (JCPDS no. 21-1272) without any other
phases or impurities (Figure 1b and Figure S4, SI). The
sole Raman-active modes (Figure S5, SI) at around 146,
399, 519 and 641 cm-1 confirm the formation of pure ana-
tase phase.48-49 TEM image (Figure 1c) shows the inter-
planar distances of (002) and (101) planes of anatase
TiO2,50 which means the exposed facet is (010) facet. The
fast Fourier transform pattern (inset in Figure 1d) also
supports this conclusion, with the angle of 68.3° identical
to the theoretical value of angle between (101) and (001)
facets.50-51 In addition, the reference sample synthesized
without glycerol (referred as normal TiO2) is also anatase
TiO2 exposed with (101) facet, see the XRD pattern (Figure
1b) and TEM image (Figure 1d).
5
g=1.998
ꢀ20
ꢀ20
g=1.998
6
ꢀ40
ꢀ60
ꢀ40
ꢀ60
7
8
9
Normal TiO2
Defected TiO2
Normal TiO2
Defected TiO2
ꢀ80
ꢀ80
ꢀ100
ꢀ100
3450
3340
3360
3380
3400
3500
3550
3600
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Magnetic Field/ Gauss
Magntic Field/ Gauss
O1s
Ti2p3/2
(c)
(d)
Defected TiO2
Normal TiO2
Defected TiO2
Normal TiO2
3.2. Ti defects in TiO2
Although the two TiO2 samples both exhibit anatase
phase, element analysis shows they have very different
composition. As shown in Table 1, the Ti/O molar ratio in
bulk determined by chemical titration is very close to the
surface ratio detected by XPS, indicating that the as-
prepared TiO2 samples have homogenous composition
from surface to inner bulk. Normal TiO2 produced with-
out glycerol shows nearly stoichiometric composition, but
the Ti/O ratio in defected TiO2 produced with glycerol is
only 0.452, much lower than the stoichiometric ratio. Ba-
sically, there are two possibilities responsible for the un-
stoichiometry of defected TiO2 (Scheme 1): Ti vacancies
and/or oxygen interstitials (Oint).
528
529
530
531
532
457.0 457.5 458.0 458.5 459.0 459.5 460.0 460.5
Binding Energy/eV
Binding Energy/eV
Figure 2. (a) Low temperature (120 K) and (b) room temper-
ature (298 K) EPR spectra of defected and normal TiO2. Inset
in (a): enlarged EPR spectrum of normal TiO2. (c) O 1s and
(d) Ti2p3/2 XPS spectra of defected and normal TiO2.
The optimized cell units containing VTi (Ti15O32,
Ti/O=0.469) and Oint (Ti16O33, Ti/O=0.485) are shown in
Figure 3, with the lattice constants summarized in Table 1.
Calculation predicts that the presence of VTi makes the c
axis to shrink heavily whereas the a and b axes to expand
slightly. On the contrary, the presence of Oint makes c axis
to expand. Actually in the XRD pattern (Inset in Figure
1b), the (101) peak of defected TiO2 is shifted towards
higher angle compared with that of normal TiO2, suggest-
ing the lattice constants are changed. As shown in Table 1,
To get more information about the defected TiO2, EPR
spectra were recorded. Under low temperature, many
defect-related species and vacancies in TiO2 can be de-
tected. The EPR signal of Ti3+ defects, O2 and oxygen va-
−
cancies (VO) locate at g=1.960~1.990, 2.020 and 2.003,10,48,52
respectively. As shown in the inset of Figure 2a, normal
TiO2 has very weak signal at g=1.990 at 120 K, which can
be assigned to the existence of trace Ti3+-oxygen vacancy
associates. Generally, the EPR signals of TiO2 are visible
only at sufficiently low temperature.14,49,52-53 Surprisingly,
defected TiO2 has a remarkably strong EPR signal at
g=1.998 under both room temperature (298 K) and 120 K
(Figure 2a, b). This signal can not be indexed to any spe-
cies reported in literatures, such as Ti3+, O2 , and VO, sug-
−
gesting the formation of a new species of defects. Moreo-
ver, the Ti2p XPS spectrum (Figure 2c) is well symmet-
rical, no shoulder band appeared at 456.9 eV that is asso-
ciated with Ti3+ defect,54 supporting that there is no Ti3+
and O2− species. In addition, the absence of peak at 457.3
eV in the Ti2p XPS spectrum roles out the existence of Ti-
H bonds,55 and the presence of other doped impurities
(mainly C atoms) is excluded by EDX and XPS analysis
(Figure S6, SI).
Figure 3. (a), (b), (c) Optimized cell structures of Ti-defected
(Ti15O32), O-interstitial (Ti16O33) and normal anatase (Ti16O32)
TiO2, respectively. (d), (e) Charge density difference of Ti-
defected (Ti15O32) and normal (Ti16O32) TiO2, respectively.
To determine the defect structure of defected TiO2,
DFT calculations on the cell structure were conducted.
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
Table 1. Lattice constants, molar ratio and electrical properties of normal and defected TiO2.
1
2
3
4
5
6
7
8
9
Resistivity×10-4 Mobility×10-4
Ti/O ratio
Hall coefficient
/(cm3/C)
Carrier
type
Samples
26-1272a
a/ Å
b/ Å
c/ Å
/(Ω cm)
/(cm2/V
s)
•
Bulk
Surface
0.500
0.498d
0.456d
0.500
3.785
3.787
3.790
3.804
3.811
3.785
3.787
3.790
3.804
3.811
9.514
9.513
9.487
9.481
9.372
9.511
0.500
0.496c
0.452c
0.500
0.469
0.485
-
-
-
-
n
p
-
Normal TiO2
1.497
0.223
-33.3
Defected TiO2
0.967
1.390
+134
b
Ti16O32
-
-
-
c
-
-
-
-
-
-
d
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
b
Ti15O32
0.469
0.485
-
b
Ti16O33
3.811
3.807
-
a
Standard JCPDS card of anatase TiO2
;
b cluster calculated using DFT; calculated from chemical titration; calculated from
XPS measurment.
the c axis of defected TiO2 is shortened whereas the a and
b axes are elongated compared with those of normal
TiO2.This experimental result is in agreement with the
computed VTi mode instead of the Oint mode. We further
calculated the charge density of VTi mode. The neighbor-
ing O and Ti atoms of VTi get less electrons compared
with those on normal sites (Figure 3d,e), which is con-
sistent with the increased binding energy in O 1s and Ti
2p XPS spectra of defected TiO2 (Figure 2c,d). Based on
the experiment and calculation, the un-stoichiometry of
defected TiO2 is attributed to the presence of Ti vacancies
in the lattice.
sibly due to the annihilation of orthopositronium atoms
formed in the large voids present in TiO2.16,56
In our recent work, Zn-defected ZnO was synthesized
by calcining zinc glycerolate (ZnGly),39 where the lamel-
leted structure of ZnGly plays crucial role in constructing
the metal defected crystals. Actually, TiGly has parallel
chains built up from edge-sharing Ti4+O6 octahedrals,47,61
similar to the structure of ZnGly, so the formation mech-
anism of Ti vacancies should be similar to that of Zn va-
cancies. During calcination, inter-chains organic groups
of TiGly is gradually removed, and the remaining Ti-O-Ti
parallel lattice chains simultaneously conjunct with each
other to form TiO2 crystals. It has been reported the de-
fects in metal oxides are sensitive to the circumstance (O-
rich or metal-rich) of synthesizing process.62-64 In particu-
lar, theoretical calculations predicted that metal vacancies
can be formed easily under O-rich environment.6,8 In the
present case, the removal of inter-chains glycerin groups
may result in O-rich environment between the Ti-O-Ti
parallel lattice chains, and thus metal defects are formed.
The defects in TiO2 were further characterized by pos-
itron annihilation. As shown in Figure S7 (SI) and Table 2,
four positron lifetime components are observed for nor-
mal and defected TiO2. The shorter lifetime component
(τ1) is caused by the free annihilation of positrons in de-
fect-free crystal.56 In a disordered system, small vacancies
or shallow positron traps can reduce the surrounding
electron density and thus increase the lifetime of τ1.56-59
The τ1 for normal and defected TiO2 is 264 ps and 287 ps,
respectively, longer than that of defect-free TiO2 crystal
(178 ps),56 indicating the existence of monovacancies or
point defects in TiO2.16, 56-59 The very low intensity (3.0%)
for normal TiO2 corresponds to the existence of small
amount of oxygen vacancies. Interestingly, the intensity
of I1 for defected TiO2 reaches 48.5%, which indicates
thecrystals have many monovacancies or point defects.
Hahn et al pointed out that metal vacancies in metal ox-
ides exist in form of point defect.60 In defected TiO2, met-
al vacancies are the major defects. Therefore, the lifetime
component of τ1 in defected TiO2 is attributed to Ti point
vacancies. Additionally, the longer lifetime components
of τ2 and τ3 are resulted from the positrons trapped by
larger size defects, and the longest component (τ4) is pos-
3.3. p-Type conductivity and room-temperature fer-
romagnetism
TiO2 is intrinsic n-type semiconductor without any
ferromagnetism. As mentioned above, VTi is acceptor-type
defects, so defected TiO2 may show p-type conductivity if
the concentration of VTi is high enough. As shown in Fig-
ure 4a, the valence band of defected TiO2 starts from 1.96
eV, which is closer to the Fermi level (0.0 eV in the va-
lence band spectrum) compared with that of normal TiO2
that starts from 2.12 eV, suggesting defected TiO2 possess-
es p-type characteristic.25,65
Hall Effect measurement was conducted to get direct
evidence for the type of electrical conductivity, see Table
1. The Hall coefficient is positive for defected TiO2, which
Table 2. Positron lifetime parameters (lifetime (τ), and intensity (I)) of normal and defected TiO2.
Samples
τ1/ps I1/% τ2/ps I2/% τ3/ps I3/% τ4/ns
I4/%
Normal TiO2
264
3.0
326
320
79.1
4.1
405
364
17.4
1.07
4.17
0.483
0.020
Defected TiO2
287 48.5
47.4
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
(a)
Defected TiO2
Normal TiO2
5 kHz
10 kHz
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
(b)
strength (M-H) at 300 K (Figure 5a). Well-defined hyste-
resis loop is observed, indicating the defected TiO2 crystal
1
2
3
4
5
6
7
8
is ferromagnetic at room temperature. From the M-T
curve in a field of 4000 Oe (Figure 5b), the Curie tempera-
ture is higher than 300 K. The spontaneous magnetic
moment deduced from the saturated magnetization value
is 0.025 emu·g-1 and the coercivity is about 18 Oe, which is
significantly higher than the reported magnetization of
Co-doped TiO2 nanocrystalline (Ti0.985Co0.15O2, 1.57×10-3
emu·g-1).71 In addition, according to DFT calculation (Ta-
ble S1, SI), the Oint mode does not have any ferromag-
netism.
1.96 eV
2.12 eV
9
ꢀ0.6
ꢀ0.4
ꢀ0.2
0.0
0.2
0
1
2
3
4
Binding Energy/eV
Potential / V vs. (Ag/AgCl)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
-1.0
Light off
(c)
-0.8
-0.6
-0.4
-0.2
0.0
0.03
0.02
0.0020
0.0015
0.0010
0.0005
0.0000
(a)
(b)
0.010
0.005
0.000
-0.005
-0.010
ꢀA/cm2
Light on
ꢀ1
0
20
40
60
80 100 120 140 160 180 200
0.01
Time/s
Defected TiO2 light on
Defected TiO2 light off
Normal TiO2 light on
Normal TiO2 light off
300 K
0.00
-300 -200 -100
0
100 200 300
Field/Oe
ꢀ0.01
ꢀ0.02
ꢀ0.03
4000 Oe
300 K
10000 20000
-0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1
0.0
Potential / V vs. (Ag/AgCl)
0
50 100 150 200 250 300
Temperature/K
ꢀ20000 ꢀ10000
0
Field/Oe
Figure 4. (a) XPS valence band spectra of defected and nor-
mal TiO2. (b) Mott-Schottky plots measured in a standard
three-electrode setup using defected TiO2 electrode as work-
ing electrode. (c) Current-voltage (I-V) curve of defected and
normal TiO2, inset in (c): transient photocurrent (chronoam-
perometric) curve of defected TiO2 at -0.45 V vs. Ag/AgCl.
Figure 5. (a) Magnetization (M-H) curves of defected TiO2
measured at 300 K. (b) M-T curve of defected TiO2 in the
field of 4000 Oe.
3.4. Photocatalytic performance
TiO2 is the most widely studied photocatalyst in ener-
gy and environment fields. Here we tested the photocata-
lytic activity of defected TiO2 in three types of reactions:
H2 generation under UV light, organic pollutant (RhB,
MO and phenol) degradation under UV light, and self-
sensitized RhB degradation under visible light. The pho-
toreaction under UV light involves the formation, separa-
tion and transfer of electron-hole pairs, surface reaction,
with charge recombination as the major side process,24
The sensitized reaction under visible light involves the
excitation of dye, injection of electron from dye to TiO2,
and then surface reaction, with the electron back-
injection as the side process.15 No matter what type of
photocatalysis is, the charge transfer in TiO2 crystal is one
key factor determining the efficiency.
confirms this material is truly p-type and conducts via
holes.66-68 As contrast, normal TiO2 is n-type with nega-
tive Hall coefficient. Furthermore, the presence of VTi
increases the charge mobility (1.39×104 cm2/V·s for defect-
ed TiO2vs. 2.23×103 cm2/V·s for normal TiO2). It was re-
ported that the mobility of Mn-doped p-TiO2 is lower
than that of pure TiO2.28 Unlike foreign p-type dopants,
VTi has no deteriorating effect on the charge transfer.
The electrochemical impedance measurements and
photoelectrochemical characterizations were also per-
formed to determine the conductivity of defected TiO2.
The evidence for p-type conductivity comes from the
negative slope of linear part of Mott-Schottky plots (Fig-
ure 4b).69 Correspondingly, defected TiO2 photocathode
shows obvious cathodic current under illumination (Fig-
ure 4c) with the photocurrent increasing progressively
with the increase of negative potential applied (due to
increased band bending), also see the transient current (I-
t) curves (Inset in Figure 4c) with a sharp cathodic cur-
rent peak appears upon the turn-on of light. On the con-
trary, normal TiO2 exhibits positive slope in the Mott-
Schottky plots (Figure S8, SI) and do not show any ca-
thodic current under similar measurement conditions
(Figure 4c). Besides, the photocurrent onset potential
(−0.40 V vs. Ag/AgCl) of defected TiO2 is more negative
than the VFB, which is also consistent with the feature of
p-type photoelectrode.70
The reaction rate shown in Figure 6a derived from
pseudo-first-order fitting of data in Figure S9 (SI) is used
to evaluate the degradation photoactivity. Obviously, de-
fected TiO2 is much more active than normal TiO2 under
both UV and visible illumination. The phenol degradation
rate of defected TiO2 is 7.0-fold of normal TiO2. Figure 6b
shows the time course of H2 evolution, defected TiO2 ex-
hibits stable H2 release rate of 29.8 mmol·g-1·h-1, whereas
normal TiO2 produces only 6.8 mmol·g-1·h-1 under the
same conditions.
Defected and normal TiO2 are both pure anatase
phase, and have similar morphology (Figure S10, SI),
comparable BET surface area (59 m2·g-1 and 72 m2·g-1 for
normal and defected TiO2, respectively) and identical
optical absorption (Figure S11, SI), so the influences of
DFT calculations indicate that the presence of VTi in
TiO2 causes ferromagnetism (Table S1, SI). The magnetic
response was measured as a function of magnetic field
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
2.0
storage under atmosphere (Figure S12, SI). Moreover, de-
Defected TiO2
Normal TiO2
(a)
5
4
3
2
1
0
1
2
3
4
5
6
7
8
fected TiO2 shows stable activity in 5 recycles of photo-
catalytic reactions (Figure S13, SI), and the surface defects
are well retained after the recycling as evidenced by XPS
analysis (Figure S12, SI). These results indicate that the
surface defects are stable under the reaction conditions
and long-term storage.
1.5
1.0
0.5
0.0
60
(a)
Normal TiO2
40
Total
9
20
0
Phenol-UV
Samples
RhB-vis
MO-UV
RhB-UV
10
11
12
13
14
15
16
17
18
19
20
21
22
-5
-4
-3
-2
-1
0
1
2
3
4
5
-20
120
(b)
-40
-60
29.86mmol·gꢀ1·hꢀ1
100
80
60
40
20
0
Defected TiO2
Normal TiO2
60
40
20
0
Defected TiO2
Total
-5
-4
-3
-2
-1
0
1
2
3
4
5
-20
-40
-60
6.81mmol·gꢀ1·hꢀ1
Energy/eV
2000
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Normal TiO2
Defected TiO2
(c)
(b)
Time/hour
1500
1000
500
0
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. (a) Reaction rate of photodegradation of organic
pollutants. (b) Time course of hydrogen generation.
Normal TiO2 light off
Normal TiO2 light on
Defected TiO2 light off
Defected TiO2 light on
morphology, surface area and optical property on the
photoactivity are excluded. The only difference is the
presence of metal defects in defected TiO2. Actually, it has
been reported that increasing the concentration of Ti va-
cancies can increase the electrical conductivity and the
mobility of electronic charge carriers.29 More specifically,
as shown in Figure 7a, the calculated density of states
(DOS) of defected TiO2 has a new band transition above
the top of original VB, which means the VB width is
raised. The VB width intrinsically controls the mobility of
holes, and wider VB results in higher mobility of holes
generated.72-73 Therefore, the charge transfer and separa-
tion in defected TiO2 is improved.
0
200
400
600
800
1000
350
400
450
500
550
600
650
700
Wavelength/nm
Z` /(ohm)
Figure 7. (a) Calculated total density of states of normal TiO2
(Ti16O32) and defected TiO2 (Ti15O32), (b) Steady state PL
spectra of defected and normal TiO2, (c) EIS Nyquist plot of
defected and normal TiO2.
4. CONCLUSIONS
We successfully synthesized undoped anatase TiO2
with large amount of (9.5 mol%) Ti vacancies using a fac-
ile solvothermal treatment method. The presence of
abundant VTi makes the c axis of anatase shrunk heavily
whereas a and b axes elongated slightly, produces a strong
EPR signal at g=1.998 even under room temperature, and
widens the valence band top edge. VTi switches TiO2 con-
ductivity from n-type to p-type with higher charge mobil-
ity, and endows room temperature ferromagnetism high-
er than Co-doped TiO2 nanocrystalline. Moreover, defect-
ed TiO2 exhibits remarkably higher photocatalytic activity
compared with normal TiO2 because of the more efficient
charge transfer in bulk and at semiconductor/electrolyte
interface. It indicates that the metal vacancies can change
the structure characteristics and physiochemical proper-
ties of undoped semiconductor and thus produce new
kind of functional materials for chemical and physical
devices.
The more efficient charge transfer and separation of
defected TiO2 is confirmed by experiments. Photolumi-
nescence (PL) emission signal caused by charge recombi-
nation can reflect the charge separation in bulk. As shown
in Figure 7b, defected TiO2 has smaller PL intensity than
normal TiO2, which indicates a more efficient transfer and
separation of photo-induced charge pairs. In electro-
chemical impedance spectroscopy, the radius of arc of
Nyquist plots can be used to evaluate the charge transfer
resistance at the semiconductor/electrolyte interface, and
smaller arc radius implies smaller charge transfer re-
sistance.74 Figure 7c shows that defected TiO2 exhibits
smaller radius, so the charge in defected TiO2 can be
quickly transferred to reactant through the solid/liquid
interface and consumed by chemical reaction. The quick
consumption of surface charges in turn retards the charge
recombination or back charge transfer side processes.
ASSOCIATED CONTENT
Supporting Information.
Details of composition analysis, XRD pattern of intermediat-
ed powders, NMR spectra of TiGly, XPS spectrum, positron
annihilation spectra, EDS spectra, Raman spectra, computed
ferromagnetism of defected TiO2, UV-vis DRS spectra, pho-
The stability of surface defects is important because
photocatalysis is very sensitive to them. XPS analysis re-
veals the surface Ti/O ratio, binding energies of Ti and
lattice O of defected TiO2 do not change after 4 months
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
(21) Liu, S.; Yu, J.; Jaroniec, M. J. Am. Chem. Soc. 2010, 132,
11914-11916.
todegradation data, stability test, and SEM images. This ma-
terial is available free of charge via the Internet at
1
(22) Ren, L.; Li, Y.; Hou, J.; Zhao, X.; Pan, C. ACS Appl. Mater.
Interfaces 2014, 6, 1608-1615.
(23) Xiang, Q.; Lv, K.; Yu, J. Appl. Catal. B 2010, 96, 557-564.
(24) Liu, G.; Yang, H. G.; Pan, J.; Yang, Y. Q.; Lu, G. Q.; Cheng,
H.-M. Chem. Rev. 2014, 114, 9559-9612.
2
3
4
5
6
7
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
(25) Cao, J.; Zhang, Y.; Liu, L.; Ye, J. Chem. Commun. 2013, 49,
3440-3442.
* jj_zou@tju.edu.cn (J.-J. Zou); Tel and fax: 86-22-27892340
8
9
Author Contributions
(26) Ruiz, A.; Cornet, A.; Sakai, G.; Shimanoe, K.; Morante, J.
R.; Yamazoe, N. Chem. Lett. 2002, 31, 892-893.
(27) Nowotny, M. K.; Bak, T.; Nowotny, J.; Sorrell, C. C. Phys.
Stat. Sol. (b) 2005, 242, R88-R90.
ǁ
These authors contributed equally.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Notes
(28) Li, X.; Wu, S.; Hu, P.; Xing, X.; Liu, Y.; Yu, Y.; Yang, M.;
Lu, J.; Li, S.; Liu, W. J. Appl. Phys. 2009, 106, 043913.
(29) Bak, T.; Nowotny, M. K.; Sheppard, L. R.; Nowotny, J. J.
Phys. Chem. C 2008, 112, 13248-13257.
(30) Nowotny, M. K.; Bogdanoff, P.; Dittrich, T.; Fiechter, S.;
Fujishima, A.; Tributsch, H. Mater. Lett. 2010, 64, 928-930.
(31) Bhowmik, B.; Dutta, K.; Hazra, A.; Bhattacharyya, P. Solid-
State Electron. 2014, 99, 84-92.
(32) Hazra, A.; Das, S.; Kanungo, J.; Sarkar, C. K.; Basu, S. Sens.
Actuators, B 2013, 183, 87-95.
(33) Hazra, A.; Das, S.; Kanungo, J.; Bontempi, E.; Sarkar, C. K.;
Bhattacharyya, P.; Basu, S. J. Mater. Sci.: Mater. Electron. 2013, 24,
1658-1663.
(34) Lu, A.-H.; Salabas, E. L.; Schüth, F. Angew. Chem. Int. Ed.
2007, 46, 1222-1244.
(35) Chambers, S. A. Sur. Sci. Rep. 2006, 61, 345-381.
(36) Matsumoto, Y.; Murakami, M.; Shono, T.; Hasegawa, T.;
Fukumura, T.; Kawasaki, M.; Ahmet, P.; Chikyow, T.; Koshihara,
S.; Koinuma, H. Science 2001, 291, 854-856.
(37) Sharma, P.; Gupta, A.; Rao, K. V.; Owens, F. J.; Sharma, R.;
Ahuja, R.; Guillen, J. M. O.; Johansson, B.; Gehring, G. A. Nat.
Mater. 2003, 2, 673-677.
(38) Janotti, A.; Van de Walle, C. G. Phys. Rev. B 2007, 76,
165202.
The authors declare no competing financial interest.
ACKNOWLEDGMENT
The authors appreciate the supports from the National Natu-
ral Science Foundation of China (21222607), the Tianjin Mu-
nicipal Natural Science Foundation, and the Program for
New Century Excellent Talents in Universities (NCET-09-
0594).
REFERENCES
(1) Gray, T. J.; McCain, C. C.; Masse, N. G. J. Phys. Chem. 1959,
63, 472-475.
(2) Henrich, V. E.; Dresselhaus, G.; Zeiger, H. J. Phys. Rev. Lett.
1976, 36, 1335-1339.
(3) Brown, G. E. Jr; Henrich, V. E.; Casey, W. H.; Clark, D. L.;
Eggleston, C.; Felmy, A.; Goodman, D. W.; Grätzel, M.; Maciel,
G.; McCarthy, M. I.; Nealson, K. H.; Sverjensky, D. A.; Toney, M.
F.; Zachara, J. M. Chem. Rev. 1999, 99, 77-174.
(4)Bally, A. R.; Korobeinikova, E. N.; Schmid, P. E.; Levy, F.;
Bussy, F. J. Phys. D: Appl. Phys. 1998, 31, 1149-1154.
(5) Nowotny, M. K.; Sheppard, L. R.; Bak, T.; Nowotny, J. J.
Phys. Chem. C 2008, 112, 5275-5300.
(6) Morgan, B. J.; Watson, G. W. Phys. Rev. B 2009, 80, 233102.
(7) Nowotny, J.; Bak, T.; Nowotny, M. K.; Sheppard, L. R. J.
Phys. Chem. B 2006, 110, 18492-18495.
(8) Na-Phattalung, S.; Smith, M. F.; Kim, K.; Du, M.-H.; Wei,
S.-H.; Zhang, S. B.; Limpijumnong, S. Phys. Rev. B 2006, 73,
125205.
(9) Hong, N. H.; Sakai, J.; Poirot, N.; Brize, V. Phys. Rev. B
2006, 73, 132404.
(10) Wang, W.; Ni, Y.; Lu, C.; Xu, Z. RSC Adv. 2012, 2, 8286-
8288.
(11) Lira, E.; Wendt, S.; Huo, P.; Hansen, J. O.; Streber, R.;
Porsgaard, S.; Wei, Y.; Bechstein, R.; Laegsgaard, E.;
Besenbacher, F. J. Am. Chem. Soc. 2011, 133, 6529-6532.
(12) Coey, J. M. D.; Venkatesan, M.; Fitzgerald, C. B. Nat.
Mater. 2005, 4, 173-179.
(13) Fujishima, A.; Honda, K. Nature 1972, 238, 37-38.
(14) Zuo, F.; Bozhilov, K.; Dillon, R. J.; Wang, L.; Smith, P.;
Zhao, X.; Bardeen, C.; Feng, P. Angew. Chem. Int. Ed. 2012, 51,
6223-6226.
(39) Pan, L.; Wang, S.; Mi, W.; Song, J.; Zou, J.-J.; Wang, L.;
Zhang, X. Nano Energy 2014, 9, 71-79.
(40) Kohn, W.; Sham, L. J. Phys. Rev. 1965, 140, A1133-A1138.
(41) Blöchl, P. E. Phys. Rev. B 1994, 50, 17953-17979.
(42) Kresse, G.; Furthmüller, J. Phys. Rev. B 1996, 54, 11169-
11186.
(43) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett.
1996, 77, 3865-3868.
(44) Dudarev, S. L.; Botton, G. A.; Savrasov, S. Y.; Humphreys,
C. J.; Sutton, A. P. Phys. Rev. B 1998, 57, 1505-1509.
(45) Zakharova, G. S.; Andreikov, E. I.; Osipova, V. A.; Yatluk,
Y. G.; Puzyrev, I. S. Inorg. Mater. 2013, 49, 1127-1132.
(46) Chen, Y.; Tian, G.; Ren, Z.; Tian, C.; Pan, K.; Zhou, W.; Fu,
H. Eur. J. Inorg. Chem. 2011, 2011, 754-760.
(47) Zhao, J.; Liu, Y.; Fan, M.; Yuan, L.; Zou, X. Inorg. Chem.
Front. 2015, DOI: 10.1039/C4QI00191E.
(48) Yu, X.; Kim, B.; Kim, Y. K. ACS Catal. 2013, 3, 2479-2486.
(49) Pan, L.; Zou, J.-J.; Wang, S.; Huang, Z.-F.; Yu, A.; Wang, L.;
Zhang, X. Chem. Commun. 2013, 49, 6593-6595.
(50) Zhao, J.; Zou, X. X.; Su, J.; Wang, P. P.; Zhou, L. J.; Li, G.
D. Dalton Trans. 2013, 42, 4365-4368.
(51) Pan, L.; Zou, J.-J.; Wang, S.; Liu, X.-Y.; Zhang, X.; Wang, L.
ACS Appl. Mater. Interfaces 2012, 4, 1650-1655.
(52) Liu, H.; Ma, H. T.; Li, X. Z.; Li, W. Z.; Wu, M.; Bao, X. H.
Chemosphere 2003, 50, 39-46.
(53) Naldoni, A.; Allieta, M.; Santangelo, S.; Marelli, M.;
Fabbri, F.; Cappelli, S.; Bianchi, C. L.; Psaro, R.; Dal Santo, V. J.
Am. Chem. Soc. 2012, 134, 7600-7603.
(54) Wang, R.; Sakai, N.; Fujishima, A.; Watanabe, T.;
Hashimoto, K. J. Phys. Chem. B 1999, 103, 2188-2194.
(15) Pan, L.; Zou, J.-J.; Zhang, X.; Wang, L. J. Am. Chem. Soc.
2011, 133, 10000-10002.
(16) Kong, M.; Li, Y.; Chen, X.; Tian, T.; Fang, P.; Zheng, F.;
Zhao, X. J. Am. Chem. Soc. 2011, 133, 16414-16417.
(17) Zhou, W.; Sun, F.; Pan, K.; Tian, G.; Jiang, B.; Ren, Z.; Tian,
C.; Fu, H. Adv. Funct. Mater. 2011, 21, 1922-1930.
(18) Li, Y.; Kunitake, T.; Fujikawa, S. J. Phys. Chem. B 2006, 110,
13000-13004.
(19) Wu, M.; Jin, J.; Liu, J.; Deng, Z.; Li, Y.; Depairs, O.; Su, B.-L.
J. Mater. Chem. A 2013, 1, 15491-15500.
(20) Yang, H. G.; Sun, C. H.; Qiao, S. Z.; Zou, J.; Liu, G.; Smith,
S. C.; Cheng, H. M.; Lu, G. Q. Nature 2008, 453, 638-641.
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
(55) Zheng, Z.; Huang, B.; Lu, J.; Wang, Z.; Qin, X.; Zhang, X.;
1
2
3
4
5
6
7
8
Dai, Y.; Whangbo, M.-H. Chem. Commun. 2012, 48, 5733-5735.
(56) Murakami, H.; Onizuka, N.; Sasaki, J.; Thonghai, N. J. Ma-
ter. Sci. 1998, 33, 5811-5814.
(57) Yan, J.; Wu, G.; Guan, N.; Li, L.; Li, Z.; Cao, X. Phys. Chem.
Chem. Phys. 2013, 15, 10978-10988.
(58) Jiang, X.; Zhang, Y.; Jiang, J.; Rong, Y.; Wang, Y.; Wu, Y.;
Pan, C. J. Phys. Chem. C 2012, 116, 22619-22624.
(59) Zheng, F.; Liu, Y.; Liu, Z.; Dai, Y. Q.; Fang, P. F.; Wang, S.
J. J. Cryst. Growth 2012, 353, 55-58.
9
(60) Hahn, B. P.; Long, J. W.; Rolison, D. R. Acc. Chem. Res.
2013, 46, 1181-1191.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(61) Wang, D.; Yu, R.; Kumada, N.; Kinomura, N. Chem. Mater.
1999, 11, 2008-2012.
(62) Zhang, S.; Ogale, S. B.; Yu, W.; Gao, X.; Liu, T.; Ghosh, S.;
Das, G. P.; Wee, A. T. S.; Greene, R. L.; Venkatesan, T. Adv. Ma-
ter. 2009, 21, 2282-2287.
(63) Wang, X. F.; Xu, J. B.; Zhang, B.; Yu, H. G.; Wang, J.;
Zhang, X.; Yu, J. G.; Li, Q. Adv. Mater. 2006, 18, 2476-2480.
(64) Xiang, B.; Wang, P.; Zhang, X.; Dayeh, S. A.; Aplin, D. P.
R.; Soci, C.; Yu, D.; Wang, D. Nano Lett. 2007, 7, 323-328.
(65) Li, X.; Guo, Z.; He, T. Phys. Chem. Chem. Phys. 2013, 15,
20037-20045.
(66) Kawazoe, H.; Yasukawa, M.; Hyodo, H.; Kurita, M.; Yan-
agi, H.; Hosono, H. Nature 1997, 389, 939-942.
(67) Look, D. C.; Reynolds, D. C.; Litton, C. W.; Jones, R. L.;
Eason, D. B.; Cantwell, G. Appl. Phys. Lett. 2002, 81, 1830-1832.
(68) Bian, J. M.; Li, X. M.; Gao, X. D.; Yu, W. D.; Chen, L. D.
Appl. Phys. Lett. 2004, 84, 541-543.
(69) Lin, Y.; Xu, Y.; Mayer, M. T.; Simpson, Z. I.; McMahon, G.;
Zhou, S.; Wang, D. J. Am. Chem. Soc. 2012, 134, 5508-5511.
(70) Scragg, J. J.; Dale, P. J.; Peter, L. M.; Zoppi, G.; Forbes, I.
Phys. Stat. Sol. (b) 2008, 245, 1772-1778.
(71) Bryan, J. D.; Heald, S. M.; Chambers, S. A.; Gamelin, D. R.
J. Am. Chem. Soc. 2004, 126, 11640-11647.
(72) Liu, G.; Niu, P.; Sun, C.; Smith, S. C.; Chen, Z.; Lu, G. Q.;
Cheng, H.-M. J. Am. Chem. Soc. 2010, 132, 11642-11648.
(73) Guan, M.; Xiao, C.; Zhang, J.; Fan, S.; An, R.; Cheng, Q.;
Xie, J.; Zhou, M.; Ye, B.; Xie, Y. J. Am. Chem. Soc. 2013, 135, 10411-
10417.
(74) Huang, Z.-F.; Song, J.; Pan, L.; Jia, X.; Li, Z.; Zou, J.-J.;
Zhang, X.; Wang, L. Nanoscale 2014, 6, 8865-8872.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
1
2
3
TOC
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment