Synthesis of anatase TiO2 rods with dominant reactive {010} facets for the photoreduction of CO2 to CH4 and use in dye-sensitized solar cells

Article abstract of DOI:10.1039/c1cc13034j

Single crystalline anatase TiO₂ rods with dominant reactive {010} facets are directly synthesized by hydrothermally treating Cs _0.68Ti_1.83O₄/H_0.68Ti _1.83O₄ particles. The nanosized rods show a comparable conversion efficiency in dye-sensitized solar cells (DSSCs), and a superior photocatalytic conversion of CO₂ into methane to the benchmark P25 TiO₂ nanocrystals.

Full text of DOI:10.1039/c1cc13034j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 8361–8363
www.rsc.org/chemcomm
COMMUNICATION
Synthesis of anatase TiO rods with dominant reactive {010} facets for
2
the photoreduction of CO to CH and use in dye-sensitized solar cellsw
2
4
a
b
b
a
b
Jian Pan, Xia Wu, Lianzhou Wang, Gang Liu,* Gao Qing (Max) Lu and
a
Hui-Ming Cheng
Received 24th May 2011, Accepted 13th June 2011
DOI: 10.1039/c1cc13034j
Single crystalline anatase TiO2 rods with dominant reactive
010} facets are directly synthesized by hydrothermally treating
particles. The nanosized rods show
a comparable conversion efficiency in dye-sensitized solar cells
titanate particles. Due to the unique surface atomic/electronic
structure, the nanosized rods show a superior activity in
converting CO into CH and a comparable energy conversion
efficiency as a working electrode in DSSCs to the benchmark
P25 TiO nanocrystals.
Lepidocrocite-type bulk titanate Cs0.68Ti1.83
prepared according to the previously reported solid state
{
Cs0.68Ti1.83
O
4
/H0.68Ti1.83
O
4
2
4
(
DSSCs), and a superior photocatalytic conversion of CO
2
into
2
methane to the benchmark P25 TiO nanocrystals.
2
4
O powder was
1
8
Photocatalysis is a promising process to convert solar energy.
A high percentage of reactive facets in photocatalysts by
crystal facet engineering has been actively pursued due to
reactions. Its protonated form H0.68Ti1.83
by ion-exchange in 1 M chloric acid solution. The morphology
of both Cs0.68Ti1.83 particles with a size of
several hundreds of nanometres is irregular (see Fig. S1, ESIw).
By hydrothermally treating Cs0.68Ti1.83 powder in an
4
O was obtained
O and H0.68Ti1.83O
4 4
the competitive advantages in optimizing photocatalytic
2
reactivity and/or selectivity. Faceting an anatase TiO
photo-
O
4
2
catalyst has attracted increasing interest since a large percentage
3
of high-energy {001} facets was realized. Very recently, it has
aqueous solution at 180 1C for 24 h, all irregular particles
disappeared and uniform rods with tetragonal sides and
slightly truncated pyramid ends are formed as shown in
Fig. 1a. X-Ray diffraction patterns in Fig. S2 (ESIw) confirm
been reported that {010} with both a favorable surface atomic
structure and surface electronic structure is most reactive
4
among low index facets {001}, {010} and {101}. Although
that these micrometre-sized rods are anatase TiO
group: I4 /amd, JCPDS No. 21-1272). With H0.68Ti1.83
a precursor and Cs CO as a pH mediator agent, nanosized
2
(space
À2
the surface energy of {010} (0.53 J m ) is theoretically
À2
1
O
4
as
determined to be slightly higher than that (0.44 J m ) of
À2
2
3
{
101} and much lower than that (0.90 J m ) of {001}, it is
surprising that no {010} appears in the equilibrium shape of
5a
anatase. On the other hand, a couple of experimental studies
have validated the possibility of growing {010} dominant
anatase by using a correct organic or inorganic morphology
4,5b
controlling agent.
An alternative route to prepare {010}-rich
anatase was to hydrothermally treat pre-synthesized sodium
titanate nanotubes without any morphology controlling agent
6
in a basic solution environment. The origin of stabilization of
{
010} facets is that O-terminated {010} has a lower surface
energy than O-terminated {101} and {001} in the basic
environment, which provides a powerful mechanism for growing
7
010} dominant anatase. In this work, we showed that {010}
{
dominant anatase rods can directly grow from lepidocrocite-type
a
Shenyang National Laboratory for Materials Science,
Institute of Metal Research, Chinese Academy of Sciences,
72 Wenhua Road, Shenyang 110016, China.
E-mail: gangliu@imr.ac.cn; Fax: +86 24 23903126;
Tel: +86 24 83978238
b
ARC Centre of Excellence for Functional Nanomaterials,
School of Engineering and Australian Institute of Bioengineering and
Nanotechnology, The University of Queensland, Qld. 4072, Australia
w Electronic supplementary information (ESI) available: Sample
preparation, XRD, XPS, Raman data, SEM images, performance
tests. See DOI: 10.1039/c1cc13034j
Fig. 1 SEM and HRTEM images of (a) micron-sized and (b) nano-
sized TiO rods. The insets in (a) and (b) show a schematic shape of an
anatase rod with {010}, {101}, {001} and a high-magnification SEM
image of nanosized TiO rods. The insets in (c) and (d) are the
2
2
low-magnification TEM images of the micron- and nanosized rods.
This journal is c The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 8361–8363 8361

View Article Online
anatase rods with a similar shape to that from Cs0.68Ti1.83
O
4
Fig. 2a, in contrast to the micro-sized rods, the intrinsic
absorption edge of the nano-sized rods has a blue-shift of
4 nm. The derived bandgap from the plots of the transformed
Kubelka–Munk function vs. the energy of light is 3.27 and
3.32 eV for the micro-sized and nano-sized rods, respectively.
The marginally larger bandgap of the nano-sized rods is
apparently not caused by the quantum size effect because the
are formed (see Fig. 1b). To identify the exposed facets of the
rods, high resolution (HR) transmission electron microscopy
(
TEM) was used to reveal surface atomic structures. The
HRTEM images in both Fig. 1c and d, recorded along the
001] direction of micro- and nanosized rods, give three sets of
[
˚
lattice fringes with spacings of 4.8, 3.5 and 3.5 A, which are
%
assigned to {002}, {101} and {101} facets, respectively.
critical size of TiO
,10
2
is extremely small (5 nm). In the previous
4
Clearly, the surface of these well faceted rods can be identified
as lateral {010}, {101} and top {001} (see the inset in Fig. 1a),
according to the SEM and TEM images together with the
crystallographic symmetries of anatase.
work,
it has been verified that {001} has a smaller bandgap
than both {101} and {010} due to substantially different
atomic configurations on each surface, and {010} has a
bandgap close to {101}. Therefore, the rational reason for
the bandgap difference between the two kinds of rods may be
inferred to be a slightly higher percentage of {001} facets in the
micro-sized rods. According to the valence band (VB) spectra
in Fig. 2b, the VB maxima of both rods are at 1.96 eV, which is
consistent with the reported value (1.93 eV) of {010} dominant
The compositions of the rods and their states were investigated
with X-ray photoelectron spectroscopy (XPS) and Raman
+
spectroscopy. A number of Cs ions were present in the rods
as seen by the obvious XPS signal of Cs 3d in Fig. S3 (ESIw).
These ions can be easily removed by a simple ion-exchange
process with protons and subsequent calcination in air. Both
the crystal phase and morphology of the rods show no
detectable change after the above treatment. Prior to the
4
anatase single crystals. The unchanged VB maximum
therefore indicates that the conduction band (CB) minimum
of the nano-sized rods is raised by 0.05 eV with respect to the
micro-sized rods (by ca. 0.12 eV compared to conventional
anatase with a bandgap of 3.2 eV). The determined bandgap,
CB minimum and VB maximum are given in Fig. 2c.
+
removal of Cs , the binding energy of Ti 2p and O 1s is,
however, shifted to a higher energy. An additional peak at
ca. 532 eV is also seen in the O 1s XPS spectrum as a result of the
presence of surface terminated Cs–O bonds. After the treatment,
the binding energy of both Ti 2p3/2 and O 1s is shifted to 458.3
and 529. 5 eV, comparable to the previously reported values
Anatase TiO
2
is the most widely investigated semiconductor
serving as an electron collector to support a molecular
1
sensitizer in dye-sensitized solar cells (DSSCs), while TiO2
1
9
+
12
in clean anatase TiO2. The possible influence of the Cs
crystals with well-defined facets are seldom investigated.
adsorption and surface treatment process on surface atomic
structures was estimated by Raman spectroscopy. Different
from the substantial role of surface terminated Ti–F bonds in
Here, we explore the possible advantages of {010} facets in
DSSC applications as indicated in Fig. 3a and b. Due to the
2
À1
small specific surface area (4 m g ) of the micro-sized rods
and thus a limited dye adsorption ability, the micro-sized rod
working electrode gives a quite low short-current (Jsc) of
1
0
affecting B1g and A1g modes, the surface Cs–O bonds do not
exert an obvious influence on the Raman active modes of
anatase as shown in Fig. S4 (ESIw), suggesting that the surface
atomic structure of anatase is well retained.
À2
4.2 mA cm and solar energy conversion efficiency (Z) of
1.72%. However, it is interesting to find that the nano-sized
À2
The combination of UV-visible absorption spectra with
XPS valence band spectra of caesium-free anatase rods was
used to determine electronic band alignments. As shown in
rods can substantially improve Jsc and Z up to 16.5 mA cm
and 7.73%, respectively. These values are comparable to those
Fig. 3 (a) Photocurrent density–voltage curves and (b) action spectra
of the dye-sensitized solar cells with (i) the micro-sized, (ii) nano-sized
anatase rods with dominant {010} and (iii) P25 TiO
working anodes; (c) time-dependence of photocatalytic conversion
of CO into CH with (i) the nano-sized rods and (ii) P25 TiO
nanocrystals loaded with 1 wt% Pt; (d) average conversion rate of
CO into CH for the initial 2 h with the nano-sized rods loaded with
different percents of Pt cocatalyst.
2
nanocrystals as
Fig. 2 (a) UV-visible absorption spectra; (b) XPS valence band
spectra; (c) schematic of electronic band alignments of caesium-free
micro-sized (i) and nano-sized (ii) anatase rods. The inset is plots of the
transformed Kubelka–Munk function vs. the energy of light.
2
4
2
2
4
8
362 Chem. Commun., 2011, 47, 8361–8363
This journal is c The Royal Society of Chemistry 2011

View Article Online
(
16.9 and 7.69%) of the reference electrode fabricated with the
benchmark P25 TiO nanocrystals, whose specific surface area
is nearly two times higher than our nano-sized rods (48 vs.
4 2
(ii) a coexistence stage (5–18 h) of Cs0.68Ti1.83O and TiO ;
2
(iii) a possible ripening and recrystallization stage after 18 h.
In stages (i) and (ii), the gradual release of titanium species and
concomitant pH value increase of the solution from an initial
8.4 to a final 12.1 play a central role in controlling
2
À1
2
5 m g ). A reasonable explanation for the good performance
of the {010} dominant anatase rods can be attributed to the
unique surface atomic structure featuring both 100% five-
coordinated Ti (Ti5c) atoms and very flat bond configurations
the nucleation and growth of TiO , and obtaining a high
2
percentage of {010}. Regarding the formation processes of
(
see Fig. 4a in ref. 4), which may contribute to the effective
adsorption of dye molecules and promote electron transfer
from excited molecules to TiO . In addition, the electrodes of
the nano-sized rods from H0.68Ti1.83
from the above include
4
O , the obvious differences
a
faster dissolution rate of
2
4 4
H0.68Ti1.83O than Cs0.68Ti1.83O and a higher initial pH value
the {010} dominant rods have a marginally larger open-circuit
voltage Voc than that of P25 as a result of their higher CB
minimum (739 vs. 727 mV).
(10.8) mediated by additional Cs
pH value (11.5) during the reaction. These features result in a
much higher density of TiO nucleation and consequently
smaller rods. The pH value mediator Cs CO can be replaced
2 3
CO and slightly increased
2
Photocatalytic conversion of CO into chemical fuels such
2
2
3
as CH and CH OH is a highly important yet very challenging
4
by other weak basic agents such as Na CO and K CO to
2 3 2 3
3
1
3
research topic. This is largely due to the multi-carrier
transfer processes required not only by the photooxidation
prepare similar nano-sized rods (see Fig. S7, ESIw).
The authors thank NSFC (Nos. 50921004, 51002160,
of H
CO
2
O with holes from VB but also the photoreduction of
À
21090343), Solar Energy Initiative of CAS for financial
2
+
with electrons from CB, for instance, CO
2
+ 8e
+
support. GL thanks the IMR SYNL-T.S. Keˆ Research
0
8
H
- CH
pH = 7]. We estimated the photocatalytic conversion activity
of CO into CH of the Pt-loaded nano-sized {010} dominant
anatase rods in the presence of H O vapor. The rods give a
superior activity to P25 TiO nanocrystals in generating CH
4
+ 2H
2
O [E (CO
2
/CH
4
) = À0.24 V vs. NHE,
Fellowship.
2
4
Notes and references
2
1
(a) M. R. Hoffmann, S. T. Martin, W. Choi and
D. W. Bahnemann, Chem. Rev., 1995, 95, 69; (b) H. Tada,
T. Kiyonaga and S. Naya, Chem. Soc. Rev., 2009, 38, 1849;
(c) A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253;
2
4
throughout the whole reaction duration of 10 h, as shown in
Fig. 3c. The deviation of the CO2 conversion rate from
the linear relationship may be caused by the continuously
increased concentration of CH4 in the reaction chamber
during reactions. Systematic studies on the activity dependence
on the amount of Pt cocatalysts suggest that 1 wt% loading is
optimal.
(
(
d) X. L. Hu, G. S. Li and J. C. Yu, Langmuir, 2010, 26, 3031;
e) J. H. Pan, H. Q. Dou, Z. G. Xiong, C. Xu, J. Z. Ma and
X. S. Zhao, J. Mater. Chem., 2010, 20, 4512; (f) Z. G. Zou,
J. H. Ye, K. Sayama and H. Arakawa, Nature, 2001, 414, 625;
(
g) K. Maeda, K. Teramura, D. L. Lu, T. Takata, N. Saito,
Y. Inoue and K. Domen, Nature, 2006, 440, 7082;
h) H. J. Yang, J. H. Yang, G. J. Ma, G. P. Wu, X. Zong,
Z. B. Lei, J. Y. Shi and C. Li, J. Catal., 2009, 266, 165.
2 (a) W. Q. Fang, X. Q. Gong and H. G. Yang, J. Phys. Chem. Lett.,
011, 2, 725; (b) G. Liu, J. C. Yu, G. Q. Lu and H. M. Cheng,
(
2 4
The excellent performance in converting CO into CH of
the {010} dominant rods can be understood as the synergistic
effects of both the unique surface atomic structure and higher
2
Chem. Commun., 2011, 47, 6763.
H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zou, G. Liu, S. C. Smith,
H. M. Cheng and G. Q. Lu, Nature, 2008, 453, 638.
2
CB minimum of {010}. The adsorption of reactant H O and
3
CO molecules on the photocatalyst surface is a prerequisite
2
for the subsequent electron transfer and conversion reactions.
4 J. Pan, G. Liu, G. Q. Lu and H. M. Cheng, Angew. Chem., Int. Ed.,
011, 50, 2133.
2
Theoretically, H O molecules at low coverage can be
2
5
(a) M. Lazzeri, A. Vittadini and A. Selloni, Phys. Rev. B, 2002,
65, 119901; (b) C. T. Dinh, T. D. Nguyen, F. Kleitz and T. O. Do,
ACS Nano, 2009, 3, 3737.
dissociatively adsorbed on the (010) surface with 100% Ti5c
atoms, while the molecules can only be molecularly adsorbed
14a
on (101). Furthermore, the interaction of CO
6
7
8
J. M. Li and D. S. Xu, Chem. Commun., 2010, 46, 2301.
A. S. Barnard and L. A. Curtiss, Nano Lett., 2005, 5, 1261.
T. Sasaki, M. Watanabe, H. Hashizume, H. Yamada and
H. Nakazawa, J. Am. Chem. Soc., 1996, 118, 8329.
2
on the (010)
is predicted to be stronger than that on both (101) and
1
4b
(
001). All these features favor the adsorption of CO
O on (010). Equally important, it has been experimentally
verified that the photoexcited electrons in a more negative CB
2
and
H
2
9 G. Liu, H. G. Yang, X. W. Wang, L. N. Cheng, J. Pan, G. Q. Lu
and H. M. Cheng, J. Am. Chem. Soc., 2009, 131, 12868.
0 G. Liu, C. H. Sun, H. G. Yang, S. C. Smith, L. Z. Wang, G. Q. Lu
and H. M. Cheng, Chem. Commun., 2010, 46, 755.
1
1
3a
have a greater ability to reduce CO
2
.
In the current case, the
electrons from a more negative CB of {010} dominant rods
11 B. Oregan and M. Gra
1
1
¨
tzel, Nature, 1991, 353, 737.
2 J. G. Yu, J. J. Fan and K. L. Lv, Nanoscale, 2010, 2, 2144.
3 (a) T. Inoue, A. Fujishima, S. Konishi and K. Honda, Nature,
979, 277, 637; (b) V. P. Indrakanti, J. D. Kubick and
can effectively reduce CO as indicated in Fig. 3d.
2
Finally, it is useful to discuss the possible growth processes
of the {010} dominant TiO2 rods from lepidocrocite-type
1
H. H. Schobert, Energy Environ. Sci., 2009, 2, 745; (c) S. C. Roy,
O. K. Varghese, M. Paulose and C. A. Grimes, ACS Nano, 2010,
4 4
Cs0.68Ti1.83O /H0.68Ti1.83O . By monitoring the crystalline
4
, 1259; (d) Q. Liu, Y. Zhou, J. Kou, X. Chen, Z. Tian, J. Gao,
S. Yan and Z. G. Zou, J. Am. Chem. Soc., 2010, 132, 14385;
e) S. Yan, S. Ouyang, J. Gao, M. Yang, J. Feng, X. Fan, L. Wan,
structure and morphology evolution of intermediate solid
products with reaction times of 0–24 h (see Fig. S5 and S6,
ESIw), three typical stages can be identified for the formation
(
Z. Li, J. H. Ye, Y. Zhou and Z. G. Zou, Angew. Chem., Int. Ed.,
2010, 49, 6400; (f) N. Zhang, S. Ouyang, P. Li, Y. Zhang, G. Xi,
T. Kako and J. H. Ye, Chem. Commun., 2011, 47, 2041.
4 (a) A. Ignatchenko, D. G. Nealon, R. Dushane and K. Humphries,
J. Mol. Catal. A: Chem., 2006, 256, 57; (b) V. P. Indrakanti,
J. D. Kubicki and H. H. Schobert, Energy Fuels, 2008, 22, 2611.
of the micro-sized TiO rods from Cs0.68Ti1.83
2
4
O : (i) dissolution
(
0–4 h) of Cs0.68Ti1.83O to increase both the concentration of
4
1
soluble titanium species from nil to the critical point for
+
TiO nucleation and the aqueous pH value by releasing Cs ;
2
This journal is c The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 8361–8363 8363