Aerosol-spraying synthesis of SiO2/TiO2 nanocomposites and conversion to porous TiO2 and single-crystalline TiOF2

Article abstract of DOI:10.1039/b909692b

Single-crystalline TiOF₂ was prepared by HF-etching the SiO ₂/TiO₂ nanocomposite obtained via aerosol-spraying, which exhibited high activity and durability in visible-light driven photocatalysis. The Royal Society of Chem

Full text of DOI:10.1039/b909692b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Aerosol-spraying synthesis of SiO2/TiO2 nanocomposites and conversion
to porous TiO2 and single-crystalline TiOF2w
Jian Zhu,a Dieqing Zhang,a Zhenfeng Bian,a Guisheng Li,a Yuning Huo,a Yunfeng Lu*b
and Hexing Li*a
Received (in Cambridge, UK) 15th May 2009, Accepted 8th July 2009
First published as an Advance Article on the web 31st July 2009
DOI: 10.1039/b909692b
Single-crystalline TiOF2 was prepared by HF-etching the SiO2/
TiO2 nanocomposite obtained via aerosol-spraying, which exhibited
high activity and durability in visible-light driven photocatalysis.
Photocatalysis has led to intensive attention both in dealing with
the environmental pollution by degrading organic substances
and in producing new energy by splitting water.1 TiO2
Fig. 1 Fabrication of porous TiO2 and single-crystalline TiOF2.
photocatalysts have been the focus of numerous investigations
Fig. 1 illustrates the principal steps involved in the preparation
of single-crystalline TiOF2. First, a mixture containing TiCl4,
glycerine, ethanol and tetraethoxysilane (TEOS) with molar ratio
of 1.0 : 45 : 37 : 0.25 was stirred vigorously at 25 1C for 24 h,
followed by diluting with 50 ml anhydrous ethanol and atomizing
in N2 flow through a high-temperature zone at 450 1C. The
as-received SiO2/TiO2 was calcined at 450 1C for 2 h, followed by
stirring at 30 1C in 100 ml solution containing HF, ethanol and
water (volume ratio = 1 : 1 : 2) for 40 h, and finally evaporating
the solvents at 80 1C. For comparison, pure TiO2 particles were
prepared by aerosol-spraying in the absence of TEOS. The
F-doped TiO2 sample was prepared following a procedure
reported by Yu et al.7 and calcined at 450 1C for 2 h. P25 was
commercially available and used without further purification.
The scanning electronic micrograph (SEM, JEOL JSM-
6380LV) and transmission electronic micrograph (TEM,
JEM-2100) images (see Fig. S1, ESIw) demonstrated that the
undoped TiO2 sample obtained via aerosol-spraying was present
as hollow spherical particles and remained almost unchanged
after being calcined at 450 1C for 2 h. No significant reaction
occurred by treating the TiO2 powder in HF solution at
room temperature since the reaction was thermodynamically
forbidden.8 As shown in Fig. 2, the TiO2/SiO2 sample after being
calcined at 450 1C for 2 h was comprised of solid particles with a
dimpled surface morphology. The related selected area electronic
diffraction (SAED, JEM-2100) image revealed that the TiO2/SiO2
sample showed poor crystallization degree, which was further
confirmed by its X-ray diffraction (XRD, D/MAX-2000) pattern.
The FTIR spectra (see Fig. S2, ESIw) collected on a Nicolet
Magna 550 spectrometer demonstrated that, besides the O–H and
O–Ti vibrations at 3450, 1630 and 607 cmÀ1 observed in pure
TiO2,9 the TiO2/SiO2 sample displayed two new peaks around
1030 and 920 cmÀ1, corresponding to the Si–O–Si and Si–O–Ti
bonds.10 This confirmed that the TiO2 and SiO2 were not present
in separated phases, but in a composite oxide. When the TiO2/SiO2
sample was treated with 20% NaOH solution, energy
dispersive X-ray spectroscopy (EDX, JEM-2100) analysis
demonstrated that the Si species was completely removed and
the molar ratio of Ti and O was 1 : 2, suggesting the formation of
owing to their low cost, non-toxicity, high stability, and easy
3
preparation.2 To date, the development of doped TiO2 and
non-titania4 visible photocatalysts is still the major research
topic both in academia or practical applications since visible
light (wavelength 4400 nm) accounts for 43% of the solar
energy. Meanwhile, the improvement of quantum efficiency is
also an attractive and challenging research subject. It has been
proved that the high crystallization degree may enhance quantum
efficiency due to the reduced recombination rate of photo-
electrons and holes.5 More recently, we reported that the core–
shell TiO2 microspheres exhibited enhanced quantum efficiency
owing to multiple light reflections within the chamber.6
It has been confirmed that the F-doped TiO2 exhibits a spectral
response in the visible region owing to the building up of inter-
mediate energy levels.7 However, this catalyst exhibited poor
activity and durability due to the low content of F species
incorporated in the TiO2 lattice and the leaching of F-dopants.
Titanium oxydifluoride (TiOF2) is expected to be more active and
stable since a large amount of F species are covalently bonded
with Ti. However, no studies on the photocatalytic performance
of TiOF2 have been reported so far since the reaction
between TiO2 and HF is thermodynamically forbidden
(DG = 267.4 kJ molÀ1).8 The TiOF2 prepared by either the
reaction of TiF4 (g) and SiO2 (s) or the hydrolysis of TiF3Cl9 is
impure and poorly crystallized, which disfavor the photocatalytic
activity. We present here a facile approach to prepare novel single-
crystalline TiOF2 by HF-etching the SiO2/TiO2 nanocomposite
obtained through aerosol-spraying. As expected, this TiOF2 is
much more active and durable than the F-doped TiO2 during
photodegradation of 4-chlorophenol under visible light irradiation.
a Department of Chemistry, Shanghai Normal University, Shanghai,
China. E-mail: HeXing-Li@shnu.edu.cn; Fax: 86-21-64322272;
Tel: 86-21-64322272
b Chemical & Biomolecular Engineering Department,
University of California, Los Angeles, CA 90095, USA
w Electronic supplementary information (ESI) available: Experimental
details and characterizations, including EDX, TEM and SEM images,
N2 adsorption–desorption isotherms, XRD patterns, XPS spectra,
FTIR spectra, 4-chlorophenol degradation curve under visible light
irradiation without adding photocatalyst. See DOI: 10.1039/b909692b
ꢀc
This journal is The Royal Society of Chemistry 2009
5394 | Chem. Commun., 2009, 5394–5396

View Article Online
only À92 kJ molÀ1, which could not compensate the positive
Gibbs energy of the TiO2–HF reaction to TiOF2. The EDX,
HRTEM and SAED images (Fig. S6, ESIw) revealed that only
TiOF2 polycrystals rather than single crystals were obtained
when the TiO2/SiO2 sample without calcination was treated in
HF solution, which further confirmed the essential role of the
TiO2/SiO2 nanocomposite in the formation of single-crystalline
TiOF2 since calcination at high temperature could strengthen the
interaction between Si, O and Ti. The XRD patterns (Fig. S7,
ESIw) further demonstrated that the important role of HF
amount in generating single-crystalline TiOF2. With the increase
of HF amount, the TiO2/SiO2 nanocomposite first transformed
to pure anatase TiO2 with poor crystallization, then to the
mixture of TiO2 and TiOF2, and finally to pure TiOF2 single
crystals. These results demonstrated the reaction between
TiO2/SiO2 and HF proceeded stepwise from TiO2 to TiOF2
nanocrystals, followed by crystallographically oriented alignment
into the TiOF2 single crystal. The reaction between TiO2/SiO2
and HF was essential for driving the reaction between TiO2 and
HF to form TiOF2 since no reaction occurred if the pure TiO2
was added into HF solution. According to N2 adsorption–
desorption isotherms, the SBET and VP of the pure TiO2 obtained
Fig. 2 SEM, TEM and SAED (insets) images of (a) SiO2/TiO2, (b) TiO2,
and (c) TiOF2 samples, together with the corresponding XRD patterns (d).
TiO2. The XRD and SAED image in Fig. 2 confirmed that the
as-prepared TiO2 was present as poorly crystallized anatase while
the SEM and TEM images showed irregular TiO2 particles with
porous structure. The N2 adsorption–desorption isotherms
(Fig. S3, ESIw) measured on a Quantachrome NOVA 4000e
analyzer at 77 K further confirmed the transfer from nonporous
TiO2/SiO2 particles to porous TiO2 particles during NaOH
treatment, with an increase of pore volume (VP) from 0.030 to
0.23 cm3 gÀ1 and the enhancement of surface area (SBET) from 14
by HF etching were determined as 67 m2 gÀ1 and 0.20 cm3 gÀ1
,
almost the same as determined for the TiO2 obtained by NaOH
etching. The close packing led to the low SBET (5.6 m2 gÀ1) and
VP (0.015 cm3 gÀ1) of the single-crystalline TiOF2.
The UV-vis diffuse reflectance spectra (DRS, MC-2530)
spectra in Fig. 3A demonstrated that the pure TiO2 was inactive
in visible area due to its large energy band gap (3.2 eV). The
F-doped TiO2 showed spectral response in visible area owing to
the incorporation of F-dopants into the TiO2 lattice, generating
intermediate energy levels.7 Both the polycrystalline and single-
crystalline TiOF2 samples obtained by HF etching TiO2/SiO2
nanocomposite before and after calcination at 450 1C displayed
similar patterns, suggesting they comprised similar energy
levels. The TiOF2 exhibited much stronger absorbance for
visible light than the F-doped TiO2 due to the higher F content
(F/Ti = 2). PL emission spectra (PLS, a Varian Cary-Eclipse
500) have been widely used to investigate the efficiency of
charge carrier trapping, migration, and transfer in order to
understand the fate of electron–hole pairs in semiconductor
particles since PL emission results from the recombination of
free charge carriers.12 As shown in Fig. 3B, the PL intensity
using a 280 nm laser as excitation source increased in the order
of single-crystalline TiOF2, polycrystalline TiOF2, F-doped
to 75 m2
g
À1. The porous structure could be attributed to
aggregates of TiO2 nanoparticles (see Fig. 1). When the
TiO2/SiO2 sample was treated in HF solution, regular cubic
particles were obtained. EDX analysis (Fig. S4, ESIw) demon-
strated the complete removal of the Si species and a Ti : O : F
molar ratio of 1 : 1 : 2, suggesting the formation of pure TiOF2.
The XRD pattern and the corresponding HRTEM and SAED
images confirmed the typical Pm3m structure, corresponding to
ordered lattice fringes with an inter-planar space of 0.38 nm
between (100) planes. The X-ray photoelectronic spectrum (XPS,
PHI 5000 VersaProbe ESCA, Fig. S5, ESIw) demonstrated that
the F species in the TiOF2 displayed only one peak of binding
energy ca. 685.3 eV and the Ti : O : F molar ratio was determined
as 1 : 1 : 2, which further confirmed the formation of pure TiOF2.10
The previously reported F-doped TiO2 displayed two peaks
around 684.3 and 688.4 eV for F 1s, which could be assigned to
the F species adsorbed on the TiO2 surface and those incorporated
into the TiO2 lattice.11 The F/Ti molar ratio was calculated as only
6.0% in the F-doped TiO2 sample. The lower binding energy of
the F in TiOF2 than that of the F incorporated into the TiO2
lattice in F-doped TiO2 implied stronger F–Ti bonding strength in
TiOF2, taking into account that the F accepted partial electron
density from Ti due to its higher electronegativity.
The TiO2/SiO2 nanocomposite played a crucial role in the
formation of single-crystalline TiOF2. When the mechanical
mixture of SiO2 and TiO2 was treated by HF solution at
room temperature, the SiO2 was removed via forming SiF4.
However, no reaction between TiO2 and HF occurred since
the Gibbs free energy of the SiO2–HF reaction to SiF4 was
Fig. 3 UV-Vis spectra (A) and PLS spectra (B) of (a) TiOF2 single
crystals obtained by HF etching the TiO2/SiO2 composite calcined at
450 1C, (b) TiOF2 polycrystals obtained by HF etching the TiO2/SiO2
composite without calcination, (c) F-doped TiO2, and (d) TiO2.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 5394–5396 | 5395

View Article Online
Table 1 Structural parameters and photocatalytic performancesa
Catalyst
F/Ti SBET/m2 gÀ1 VP/cm3 gÀ1 Yield (%)
P25
0
0
45
81
0.20
0.34
0.21
0.11
0.015
6
7
36
54
62
Porous TiO2
F-doped TiO2
Polycrystalline TiOF2
0.060 86
2.0 17
5.6
Single-crystalline TiOF2 2.0
a
Reaction conditions: 0.050 g catalyst, 50.0 mL 5.0 Â 10À5
M
W Xenon lamps (l 4 420 nm),
4-chlorophenol, three 500
T = 30 1C, stirring rate = 800 rpm, reaction time = 6 h.
TiO2, and undoped TiO2, indicating the increase of recombination
rate between photoelectrons and holes.12 The single-crystalline
TiOF2 displayed a lower recombination rate than the poly-
crystalline TiOF2 owing to its higher crystallization degree
which facilitated the transfer of photo-induced electrons from
the bulk to the surface and thus, could effectively inhibit their
recombination with the photo-induced holes.13 Both the
polycrystalline TiOF2 and F-doped TiO2 exhibited lower
recombination rate than the undoped TiO2. A possible reason
was the increase of oxygen vacancies due to the replacement of
O by F, taking into account that the oxidation numbers of O
and F are À2 and À1, respectively. These oxygen vacancies
could capture photoelectrons and thus, inhibit their
recombination with photo-induced holes. This could also account
for the lower recombination rate of the polycrystalline TiOF2
than that of the F-doped TiO2 since the former contained much
higher F-content (F/Ti = 2) than the latter (F/Ti = 6%).
The photocatalytic performance was examined by using photo-
degradation of 4-chlorophenol under visible light irradiation (light
intensity = 600 mW cmÀ2, wavelength 4420 nm). Detailed
experimental procedures are described in ESI.w Table 1 lists
structural parameters and catalytic activities of different catalysts.
The undoped TiO2 and P25 displayed very low activity
(4-chlorophenol degradation yield o 10%) since they could not
be activated by visible light. The single-crystalline TiOF2 exhibited
higher activity than the F-doped TiO2, despite its lower SBET and
VP. This could be attributed to either stronger absorbance of
visible light or higher crystallization degree which effectively
inhibited the photoelectron–hole recombination.13 Despite the
lower SBET and VP, similar F-content and absorbance ability
for visible light (see Fig. 3A), the single-crystalline TiOF2 still
exhibited higher activity than the polycrystalline TiOF2, possibly
owing to the lower photoelectron–hole recombination resulting
from the enhanced crystallization degree and oxygen vacancies
(see Fig. 3B). This could be further confirmed by the photo-
catalytic degradation of 4-chlorophenol under UV-light
irradiation (Fig. S8, ESIw).
Fig. 4 Recycling test of TiOF2 and F-doped TiO2 photocatalysts.
Reaction conditions are given in Table 1.
In summary, this work developed a facile approach to synthe-
size single-crystalline TiOF2 by HF etching the TiO2/SiO2 nano-
composite prepared by aerosol-spraying. The TiOF2 was more
active and durable than the F-doped TiO2 in photocatalytic
degradation of 4-chlorophenol under visible irradiation, which
could be attributed to the intense spectral response in the visible
region, the high crystallization degree, the enhanced F–Ti inter-
action, and the strong hydrothermal stability. The well defined
TiOF2 single crystals also showed powerful optical, photo-
electrical, and biological properties, which might offer new
opportunities of applications in these areas.
This work was supported by the National Natural Science
Foundation of China (20825724), Shanghai Government
(2007CG59, S30406, 07dz22303, DZL807) and by Key
Laboratory of Resource Chemistry of Ministry of Education.
Notes and references
1 (a) M. R. Hoffmann, S. T. Martin, W. D. Choi and
W. Bahnemann, Chem. Rev., 1995, 95, 69; (b) A. L. Linsebigler,
G. Lu and J. T. Yates, Chem. Rev., 1995, 95, 735.
2 (a) M. A. FoxandM. T. Dulay,Chem. Rev., 1993, 93, 341; (b)I. Justicia,
´
P. Ordejon, G. Canto, J. L. Mozos, J. Fraxedas, G. A. Battiston,
R. Gerbasi and A. Figueras, Adv. Mater., 2002, 14, 1399.
3 (a) H. Yamashita, M. Harada, J. Misaka, M. Takeuchi, K. Ikeue and
M. Anpo, J. Photochem. Photobiol., A, 2002, 148, 257; (b) J. C. Yu,
G. S. Li, X. C. Wang, X. L. Hu, C. W. Leung and Z. D. Zhang,
Chem. Commun., 2006, 2717; (c) S. Sakthivel and H. Kisch, Angew.
Chem., Int. Ed., 2003, 42, 4908; (d) R. Asahi, T. Morikawa,
T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269.
4 (a) K. Maeda, K. Teramura, D. Lu, T. Takata, N. Saito, Y. Inoue
and K. Domen, Nature, 2006, 440, 295.
5 (a) S. Park, J. H. Lim, S. W. Chung and C. A. Mirkin, Science, 2004,
303, 348; (b) H. G. Yang and H. C. Zeng, Angew. Chem., Int. Ed.,
2004, 43, 5930.
6 H. X. Li, Z. F. Bian, J. Zhu, D. Q. Zhang, G. S. Li, Y. N. Huo,
H. Li and Y. F. Lu, J. Am. Chem. Soc., 2007, 129, 8406.
7 J. C. Yu, J. G. Yu, W. K. Ho, Z. T. Jiang and L. Z. Zhang, Chem.
Mater., 2002, 14, 3808.
8 S. Kanno, S. Tamata, H. Yamashita and S. Azuhata, Electr. Eng. Jpn.,
2000, 132, 157.
As shown in Fig. 4, the single-crystalline TiOF2 could be used
repetitively eight times without significant decrease in activity,
while the F-doped TiO2 displayed a 32% decrease in activity
upon use five times. The EDX analysis demonstrated that the
F-content in the single-crystalline TiOF2 remained unchanged
while 25% F-dopants in the F-doped TiO2 were leached
off within five recycles. The inhibition of F-leaching in the
single-crystalline TiOF2 could account for its high durability.
Meanwhile, the single-crystalline TiOF2 displayed strong hydro-
thermal stability, leading to the preservation of well defined cubic
single crystals after repetitive use (Fig. S9, ESIw).
9 (a) Y. Kuroda, T. Mori, K. Yagi, N. Makihata, Y. Kawahara,
M. Nagao and S. Kittaka, Langmuir, 2005, 21, 8026; (b) C. P. Sibu,
S. R. Kumar and P. Mukundan, Chem. Mater., 2002, 14, 2876;
(c) Y. Tanaka and M. Suganuma, J. Sol–Gel Sci. Technol., 2001, 22, 83.
10 D. Li, H. Haneda, S. Hishita and N. Ohashi, Chem. Mater., 2005,
17, 2588.
11 (a) H. Park and W. Choi, J. Phys. Chem. B, 2007, 108, 4086;
(b) Y. Q. Wang and P. M. A. Sherwood, Chem. Mater., 2004, 16, 5427.
12 L. Q. Jing, Y. C. Qu, B. Q. Wang, S. D. Li, L. B. Yang, W. Fu,
H. G. Fu and J. Z. Sun, Sol. Energy Mater. Sol. Cells, 2006, 90, 1773.
13 B. Ohtani, Y. Ogawa and S. I. Nishimoto, J. Phys. Chem. B, 1997,
101, 3746.
ꢀc
This journal is The Royal Society of Chemistry 2009
5396 | Chem. Commun., 2009, 5394–5396