Q. Xiao, L. Gao / Journal of Alloys and Compounds 551 (2013) 286–292
molar ratio of W to Ti is controlled at 10%. The as-prepared samples are labeled as C,
287
W-codoped TiO2. After stirring for 0.5 h, the solution is poured into a Teflon-lined
stainless autoclave with 200 ml capacity. The autoclave is sealed and heated, and
then kept at a certain temperature in a range from 150 °C for 12 h. The resulting
product is separated by centrifugation, and washed with distilled water and alcohol
for several times, respectively, and finally, dried at 80 °C for 6 h. For comparison
purposes, the samples of pure TiO2, C-doped TiO2, and 10% W-doped TiO2 are also
prepared by similar procedures, respectively. The as-prepared samples are labeled
as undoped TiO2, C-doped TiO2, and W-doped TiO2, respectively.
2.2. Characterization methods
The X-ray diffraction (XRD) patterns of the synthesized samples are obtained by
a Brucker D8-advance X-ray powder diffractometer (XRD) with Cu K
a radiation
(k = 0.15418 nm). Raman spectrum analysis is conducted on a Labram HR800 Laser
Raman Spectroscopy made by Jobin Yvon, France, using the 632.8 nm He–Ne ion la-
ser as an excitation source. The laser power on the sample is 10 mW. Transmission
electron microscopy (TEM) images are obtained using a JEM-2100F transmission
electron microscope. X-ray photoelectron spectra (XPS) measurements are per-
formed in a VG Scientific ESCALAB Mark II spectrometer equipped with two ultra-
high vacuum (UHV) chambers using Al Ka radiation (1486.6 eV) to investigate the
surface properties. The binding energy of the XPS spectra is calibrated with the ref-
erence to the C 1s peak (284.8 eV) arising from adventitious carbon. Ar+ sputtering
is applied to clean the surface of the samples. Nitrogen adsorption and desorption
isotherms are collected at 77 K on a Micromeritic ASAP 2010 instrument. The spe-
cific surface areas are calculated using the Brunnauer–Emmett–Teller (BET)
equation and the pore size distributions are calculated by applying the Barrett–
Joyner–Halenda (BJH) method using the desorption branch of the isotherms. The
UV–vis absorption spectra are measured under the diffuse reflection mode using
the integrating sphere attached to a Shimadzu 2450 UV–vis spectrometer. The
powders are pressed to form a pellet and BaSO4 is used as a reference.
Fig. 1. The XRD patterns of the as-prepared samples. (a) undoped TiO2; (b) C-doped
TiO2; (c) C, W-codoped TiO2; (d) W-doped TiO2.
titania or the tungsten oxide species exist as a highly dispersed
polymeric form over the titania surface, which could not be de-
tected by XRD. The averaged crystallite sizes D is determined
according to the Scherrer equation D = kk/bcosh [33], where k is a
constant (shape factor, about 0.9), k is the X-ray wavelength
(0.15418 nm), b is the full width at half maximum (FWHM) of
the diffraction line, and h is the diffraction angle. The values of b
and h of anatase are taken from anatase (101) diffraction line.
The calculated averaged crystallite sizes of the as-prepared sam-
ples are shown in Table 1. It is found that the crystallite sizes of
the as-prepared samples are about 12–14 nm.
2.3. Photocatalytic activity measurements
The photocatalytic activities of the samples are evaluated by the degradation of
terephthalic acid (TA) in an aqueous solution under visible light irradiation. The
analysis of ÅOH radical’s formation on the sample surface under visible light irradi-
ation is performed by fluorescence technique using terephthalic acid, which readily
reacted with ÅOH radicals to produce highly fluorescent product, 2-hydroxytereph-
thalic acid [32]. The intensity of the peak attributed to 2-hydroxyterephtalic acid is
Å
Fig. 2 shows the Raman spectra of the as-prepared samples. It is
found that the typical Raman peaks of anatase are detected in each
sample, which reports that the characteristic Raman peaks of ana-
tase locate at 144 cmÀ1 (Eg), 399 cmÀ1 (B1g), 515 cmÀ1 (A1g) and
639 cmÀ1 (Eg) [34]. The characteristic Raman peaks for WO3 were
observed by Daniel et al. at 807, 715, 324, 293, and 270 cmÀ1
[35]. Interestingly, the Raman spectra in the present work did
not show any trace of WO3, implying that WO3 does not exist as
a separate crystalline oxide phase. There is also a significant de-
crease in the intensity of the Raman peaks with the increase of
the W doping concentration. On the basis of these Raman spectro-
scopic observations, it can be inferred that the W ion is doped in
the titania lattice, which is in agreement with the XRD results. In
addition, the main peak intensity (146 cmÀ1) of C-doped TiO2 with
no change in the peak position decreases compared with that of
undoped TiO2, which can be inferred that carbon is doped in the
TiO2 lattice [36].
known to be proportional to the amount of OH radicals formed [32]. The selected
concentration of terephthalic acid solution is 5 Â 10À4 M in a diluted NaOH aqueous
solution with a concentration of 2 Â 10À3 M. It has been proved that at these exper-
imental conditions (low concentration of terephthalic acid, less than 10À3 M, room
temperature), the hydroxylation reaction of terephthalic acid proceeds mainly by
ÅOH radicals [32]. Five hundred milligrams of the prepared samples is added to
100 mL of the 5 Â 10À4 M terephthalic acid solution in 2 Â 10À3 M NaOH under
ultrasonic vibration for 10 min. Prior to light irradiation, the reactor is left in the
dark for at least 30 min until an adsorption–desorption equilibrium is finally estab-
lished. A 100 W tungsten lamp fixed at a distance of 150 mm above the surface
solution is used as visible light source, and a UV cut-off filter is used to completely
remove any radiation below 420 nm to ensure illumination by visible light source
only. The average irradiation intensity of 100 W tungsten lamp was about
0.7 mW cmÀ2. The radiant flux was measured with a power meter from Institute
of Electric Light Source (Beijing). Sampling is performed in every 15 min. Solution
after filtration through 0.45 lm membrane filter is analyzed on a Hitachi F-4500
fluorescence spectrophotometer. The product of terephthalic acid hydroxylation,
2-hydroxyterephthalic acid, gave a peak at the wavelength of about 425 nm by
the excitation with the wavelength of 315 nm.
3. Results and discussion
The structure and morphology of the sample are further
examined by transmission electron microscopy (TEM) and high-
resolution transmission electron microscopy (HRTEM). The TEM
and corresponding HRTEM images of the C, W-codoped TiO2 sam-
ple are presented in Fig. 3. It can be seen that the primary crystal-
lite size is about 13 1 nm (Fig. 3a), which is in agreement with the
value of the crystallite size determined by XRD (as shown in
Table 1). Fig. 3b shows clear lattice fringes of primary crystallite
(0.35 nm corresponding to) of the same sample, The HRTEM image
(Fig. 3b) shows that the fringe spacing is 0.35 nm, which corre-
sponds to the (101) crystallographic plane of anatase, and the
HRTEM picture and its corresponding FFT (inset in Fig. 3b) further
demonstrates that the particle is single crystalline in nature, and
there are no indications of secondary phases or impurities visible
in the HRTEM pictures, suggesting that all dopant atoms are homo-
geneously incorporated into the TiO2 nanocrystallines.
3.1. XRD, Raman spectroscopy and TEM analysis
Fig. 1 shows the XRD patterns of the as-prepared samples. The
diffraction peaks of each sample can be indexed to anatase phase
with lattice parameters a = b = 0.37852 nm, c = 0.95139 nm, and
space group I41/amd (141) (JCPDS No. 21-1272). No traces of impu-
rity peaks other than TiO2 are observed. In the case of XRD of C-
doped or C, W-codoped TiO2, no characteristic peak from carbon
was detected in XRD, indicating that C ion has been substituted
into the crystal lattice sites of the titania or amorphous carbon ex-
ists over the titania surface. It is worth noting that no WO3 phase
can be observed in all the XRD patterns of C, W-doped TiO2 and
W-doped TiO2. On the basis of this, it can be inferred that either
the W ion has been substituted into the crystal lattice sites of the