Sb-doped TiO2 nanoparticle with varied dopant concentrations was synthesized using titanium tetrachloride (TiCl4) and antimony chloride (SbCl3) as the precursors. The properties of Sb-doped TiO2 nanoparticles were characterized by X-ray diffraction (XRD), scanning electron microscope (SEM), fluorescence spectrophotometer, and Uv-vis spectrophotometer. The absorption edge of TiO2 nanoparticles could be extended to visible region after doping with antimony, in contrast to the UV absorption of pure TiO2. The results showed that the photocatalytic activity of Sb-doped TiO2 nanoparticles was much more active than pure TiO2. The 0.1% Sb-doped TiO2 nanoparticles demonstrated the best photocatalytic activity which was better than that of the Degussa P25 under visible light irradiation using terephthalic acid as fluorescent probe. The effects of Sb dopant on the photocatalytic activity and the involved mechanism were extensively investigated in this work as well.

1. Introduction

In the past decade, has become a hot spot of research due to its outstanding photoelectric properties, high chemical stability, low cost, and nontoxicity and can be widely applied on photons device [1], photocatalysis [2], sensor [3], and so forth. Although has many excellent properties, the application is limited due to a relatively wide band gap (3.2 eV for the anatase) and the quick recombination of electron-hole pairs in taking place on a time scale of  s [4]. Therefore, several methods have been attempted to enhance its photocatalytic activity, such as modifying with metals [57], nonmetals [810], or other semiconductors [11, 12]. In addition, the wide band gap limits the absorption of catalysts to UV light only. It has been reported that the spectral response of photocatalysts can be extended to visible light region through the doping with other elements in order to make a donor or an acceptor level [13, 14] in the forbidden band of . The doping level can either capture excited electrons of from valence band or make electrons jump to the conduction band of . As a result, the long wave photons can be absorbed and the scope of absorption spectra region is extended to lower energy [15]. Previously, has been doped with a variety of elements, such as N [16], Fe [17], La [18], and so forth. The effect of Sb element on photocatalytic activity under visible light irradiation has also been studied by several research groups, showing high visible light absorption with UV-vis absorption spectrum measurements [19]. Although catalytic properties of Sb-doped have been found more effective, the photocatalytic mechanisms under visible light are not definitively clarified or elaborated.

In order to further study the catalytic properties of doped under visible light and the contribution of different free radicals in the catalytic reaction. In this work, pure and 0.1%5% Sb-doped photocatalysts were prepared by coprecipitation method from and . Moreover, the influence of Sb dopant on the structure of , the photoabsorption properties, and catalytic activity of these samples were studied systematically. Sb doping not only expands the absorption spectra of catalysts from UV light to visible light but also improves the catalytic activity of photocatalysts. The obtained results indicate that the catalytic activity of Sb-doped is better than Degussa P25 which absorbs the UV light only [20]. Furthermore, because hydroxyl radicals () and superoxide anion radical () can be generated on the surface of pure and Sb-doped under visible light irradiation (>420 nm), the terephthalic acid was employed as the fluorescent probe to evaluate the photocatalytic activity of catalyst [21]. For the understanding of the role of two kinds of free radicals in the catalytic process and the catalytic mechanism involved, scavengers of DMSO and p-benzoquinone are used to eliminate hydroxyl radicals () and superoxide anion radical (), respectively.

2. Experimental

2.1. Materials and Catalyst Preparation

All reagents used in the experiments were purchased from commercial sources as received, including titanium tetrachloride (, AR), dimethyl sulfoxide [, AR], n-butyl alcohol [, AR], and p-benzoquinone (, CP). The distilled water from Milli-Q system was used in the preparation of materials.

Pure and 0.1%5% Sb-doped nanoparticles were synthesized by coprecipitation method according to the literature [19], in which the dissolved in diluted HCl with deionized water and the obtained mixture solution was named Solution A. Dissolving in at a doping level ranging from 0.1% to 5% nominal atomic against , we obtained Solution B. Vigorously stirring Solution B until it is completely dissolved and then adding Solution B to Solution A under vigorous stirring at room temperature, titanium hydrous gels were precipitated upon neutralization with at pH8. The resulting precipitates were repeatedly washed to remove undesirable anions such as . The coprecipitated wet gels were treated in a butanol at 100°C for 2 h, followed by drying at 120°C for 10 h. The dried precipitates were calcined at 500°C in air to maintain the anatase phase of and milled with agate mortar in ethanol for 30 min.

2.2. Characterization of Catalyst

X-ray diffraction (XRD, DX-2700) patterns collected from 10° to 80° in 2 with 0.02° steps/s were used to identify average crystallite sizes and crystallinity of the nanoparticles using a powder X-ray diffractometer ( nm) with Cu K1 radiation from the monochromatized X-ray beam at 40 kV and 40 mA. The optical absorption spectra were measured using a UV-vis spectrophotometer (Shimadzu, UV-2500) equipped with an integrating sphere, a referenced against the compressed powders. Particle morphologies of samples were examined by scanning electron microscopy (SEM, JSM 6301F) at 15 kV. The Brunauer-Emmett-Teller (BET) surface area () of the samples was determined by nitrogen adsorption/desorption isotherm measurements at 77 K (ASAP 2010). The fluorescence emission spectrum was recorded at room temperature with excitation at 320 nm on a fluorescence spectrophotometer (Hitachi, F-4500).

2.3. Photocatalytic Activity Test

The fluorescence probe sensitive to the free radicals was applied to detect the photocatalytic activity of pure and 0.1%5% Sb-doped nanoparticles. Terephthalic acid as the fluorescent probe can readily react with hydroxyl radicals () and superoxide anion radical () to produce highly fluorescent products [22]. As shown in Scheme 1, the main products of terephthalic acid are 2-hydroxyterephthalic acid (2-HTA) or hydroxybenzoic acid (4-HBA), and only 2-HTA is highly fluorescent and thus can be easily determined by fluorescence spectroscopy [22, 23].


Photoreactivity experiments: 5 mg photocatalysts of P25 ; pure ; and 0.1%, 0.5%, 1%, and 5% Sb-doped were suspended in 20 mL aqueous solution containing 0.01 M NaOH (pH~11.5) and 500 M terephthalic acid, respectively. Before exposure to visible light, the suspension was stirred in the dark for 30 min. 3 mL of the solution was then taken out after irradiation for 20 min and centrifuged for fluorescence spectroscopy measurement. The emission intensity of the 2-HTA was monitored with the excitation at 320 nm. The photoreactivity in visible region was investigated with the irradiation of light source ( nm) obtained by the cut-off filter at 420 nm from a 500 W Xe lamp.

To study the reactivity of radicals formed from trapped holes solely, p-benzoquinone (0.05 mM) was used to quench radicals. Similarly, for the study of anion radicals, dimethyl sulfoxide (DMSO) (1 mM) was used to quench radicals formed on the Sb-doped surface. Scavenger of DMSO can react with to generate methyl sulfinic acid () and methyl radical () [24]. Detailed methods were performed as follows: 5 mg P25 ; pure ; and 0.1%, 0.5%, 1%, and 5% Sb-doped were suspended in 20 mL aqueous solution containing 0.01 M NaOH, 500 M terephthalic acid, and (1) without scavenger, (2) 1 mM DMSO, (3) 0.01 mM p-benzoquinone, respectively. Measurement method of fluorescence spectrum is the same as photoreactivity experiments except that the irradiation time is 10 min.

In this way, we were thus able to study separately the reactivity of two radicals and on the surface of Sb-doped by observing the fluorescent spectra of products.

3. Results and Discussion

For pure TiO2, the absorption edge is in UV region and the bandgap is around 3.2 eV. With the Sb doping, new dopant level was formed in the forbidden band of (Figure 1). When is exposed to light of energy greater than the bandgap energy, electrons are excited from its valence band into the conduction band to form spatially separated electron-hole () pairs by light absorption. The photogenerated charge carriers can be transferred to the surface of the catalyst and react with electron donors or acceptors adsorbed on the surface of the photocatalyst to generate reactive oxygen species such as and . As shown in Figure 1, in an aqueous environment, the photogenerated electrons are trapped by adsorbed oxygen to generate radicals (), and photogenerated holes are trapped by adsorbed water or hydroxyl bound superoxide anion () to generate bound hydroxyl radicals () [25].

Figure 2() shows the UV-vis absorption spectra of pure and 0.1%5% Sb-doped prepared by coprecipitation method. The pure had strong absorption only in the UV region corresponding to its band gap energy, while Sb-doped exhibited a new absorption band in the short wavelength region of visible light. The visible light absorption was enhanced by the dopant of pure and increased with Sb dopant. The color of the prepared catalysts changed from yellowish-white to yellow with the increasing dopant concentration of Sb.

Figure 2() shows X-ray diffraction patterns of with different Sb doping concentration; the anatase is the crystal phase of samples prepared by coprecipitation method. The average crystallite sizes of can be estimated from XRD spectra by using the Scherrer equation [26]: where is the crystallite size (nm), is the Scherrer constant (), is the wavelength of the X-ray radiation source (nm in this case), is the full width at half maximum intensity (radians), and is the angle between the incident and diffracted beams. The average crystallite sizes of are presented in Table 1. It is found that the average crystal sizes of with different Sb doping concentration are about 13.619.2 nm. The species are formed in the calcination step through oxidation of the incorporated (0.0760 nm) to [27]. The ionic radii of and are 0.0600 and 0.0605 nm, respectively [28], and the ionic radius of is similar to that of . It is suggested that ions are likely be substituted at sites within . ions partially replace until the solubility limit is reached.

Figures 2() and 2() show the SEM images of pure and 1% Sb-doped , respectively. They indicate that the surface morphologies of pure and 1% Sb-doped are almost undifferentiated. All samples are irregular in shape, and the size of particles is in the range of 2540 nm. The particle sizes of 1% Sb-doped are almost the same as that of pure due to the small amount of Sb [29]. There is a slight difference between the results of SEM and Scherrer formula, probably arising from aggregated particles. In addition, the specific surface area of the samples was obtained using the BET surface area measuring apparatus at the boiling point of liquid nitrogen. The specific surface area () of catalyst is among 40.60669.220 m2/g (as shown in Table 1).

Here, on the basis of the fact that both of these radicals and can react oxidatively with nonfluorescent probe terephthalic acid to generate product 2-HTA (highly fluorescent) or 4-HBA (Scheme 1), the catalytic activities can be monitored by fluorescence spectroscopy. Figure 3(a) shows fluorescence spectrum of terephthalic acid solution with P25, pure, and 0.1%5% Sb-doped , respectively. The results indicate that the samples of 0.1%1% Sb-doped have better photocatalytic activity compared with pure and Degussa P25 in visible light region, and 0.1% Sb-doped demonstrated the best photocatalytic activity. It could be suggested that Sb dopant acts as electron traps retarding electron-hole recombination and enhancing interfacial charge carriers transfer to the surface of the particles; however, the recombination rate will increase when the dopant concentration of Sb is too high [30]. Upon excitation with visible light, electrons are excited from oxygen atoms in the conduction band, probably to the Ti 3d orbitals, and are further trapped by adsorbed to produce radials. The photogenerated holes are trapped by adsorbed or to produce radicals bound to the surface [25]. These trapped electrons () or holes () react with the nearby adsorbed molecules, as shown in Figure 1.

It is known that the photocatalytic process will generate two kinds of free radical,   and , to drive the subsequent chemical reaction. Here, on the basis of that and h+/OH can be quenched by scavengers of p-benzoquinone and DMSO, respectively; it is expected to get more insight into the catalytic mechanism of pure and Sb-doped TiO2 by adding scavengers. As shown in Figure 3(b), DMSO inhibited the activity of catalysts obviously, because DMSO can react with and reduce the concentration of . The result indicates the existence of in solution and participation in the catalytic reaction of fluorescent probe. Similarly, the activity of catalysts had been also inhibited with scavenger by p-benzoquinone, showing the existence of and its involvement in the catalytic reaction at the same time. With the comparison of different Sb doping concentration, it was demonstrated that the scavengers can inhibit the photocatalytic activity of all the catalysts.

Doped demonstrates that doping Sb in can improve the photocatalytic activity of Sb-doped  . The XRD results indicate that the crystalline structure of the Sb-doped is anatase with a small mean diameter. It is accepted that the anatase phase and small mean diameter are very beneficial for photocatalytic reaction [31, 32]. The Sb atoms may be substituted at some of Ti sites in and forming a narrow Sb band under the conduction band, which was determined to be responsible for visible light sensitivity. The reactive oxygen species such as and on the surface of catalysts are considered beneficial for the photocatalytic process. On one hand, when the or species accept under light irradiation, they might be oxidized to form . On the other hand, the surface hydroxyl groups can also act as absorption centers for molecules. The e from the conduction band can be captured by molecules to produce . It is and , which possess high oxidizability, that oxidize terephthalic acid.

4. Conclusion

The pure and Sb-doped photocatalysts have been prepared by a coprecipitation method using and as precursor. The influence of Sb doping concentration on the catalytic property, using terephthalic acid as fluorescent probe, has been studied in this work. Sb ions can be isomorphously substituted into the lattice to generate a new doping energy level and change the electron transition way of from valence band to conduction band. In this case, it extends the scope of absorption spectrum to the visible light region and improves photocatalytic activity of catalysts. The obtained results show that the photocatalytic activity of 0.1%1% Sb-doped is higher than pure , and 0.1% Sb-doped presents the best photocatalytic activity, which is much better than Degussa P25. This superior catalytic activity mainly arises from the fact that the Sb-doped generates a new level in the forbidden band of and contributes to the capture of carriers as well as improving the separation of photogenerated electron hole. As for the catalytic mechanism of and in the fluorescence probe method, it can be attributed to the product fluorescence bursts from the oxidation of terephthalic acid to form 2-HTA by either one of these two radicals.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.


This work was partially supported by the National Science Foundation of China (21173068) and the Program for Innovative Research Team (Science and Technology) at the University of Henan Province (13IRTSTHN017).