Abstract

Hydrangea-like F-doped TiO2 microspheres have been synthesized on a large scale by a simple hydrothermal process using potassium titanium oxalate as the titanium source, ammonium fluoride and hydrogen peroxide as the etchant. The photocatalytic activities were evaluated using carbamazepine as the target organic molecule under UV and visible light irradiation. Structural characterization indicates that the hydrangea-like TiO2 microspheres, with an average diameter of 2.80 μm, are composed of numerous anatase TiO2 petals. Moreover, it is found that both the NH4F and H2O2 dosages have important effects on the formation of the hydrangea-like structures. In addition, photocatalytic experiments show that the hydrangea-like TiO2 microspheres calcined at 500°C exhibit high photocatalytic efficiency under both UV and visible light irradiation. The enhanced photocatalytic activity can be attributed to the successful fluorine doping, good crystallinity, and the unique nanostructures.

1. Introduction

Heterogeneous photocatalysis using nanosized TiO2 catalyst under ultraviolet light illumination is an efficient method for the purification of wastewater [13]. However, a major barrier to the widespread use of TiO2 as photocatalyst is its relatively large optical band gap (𝐸𝑔 = 3.2 eV), which limits its photoresponse to visible light [4, 5]. To overcome this limitation, approaches such as mental (or anion) doping, nonmental (or cations) doping, and compositing with other semiconductors have been developed [6, 7]. Among them, it has been found that fluorine doping is the most effective process as it not only can promote the activity of TiO2 by slowing down the recombination of photogenerated electrons and holes [810] but also can induce a visible-light-driven photocatalysis by the creation of oxygen vacancies [11, 12].

On the other hand, in a practical photocatalytic process, it is very difficult to separate and recover these finely powdered TiO2 from a solution. Recently, three-dimensional (3D) hierarchical structures closely packed with different building bricks have attracted considerable attention not only because of their high surface-to-volume ratios for the improvement of photocatalytic activity but also because of the large size of the complete structures for enhancing of separation efficiency [13, 14]. Therefore, it is still scientific importance to explore novel 3D TiO2 microspheres with visible light activity and high separation efficiency [15]. In this paper, we report a hydrothermal approach for the synthesis of hydrangea-like F-doped TiO2 microspheres. The as-prepared products were then characterized by XRD, SEM, TEM, and XPS techniques. The photocatalytic activities of the products before and after calcination at 500°C were evaluated by photocatalytic degradation of carbamazepine under UV and light-emitting diode (LED) light irradiation.

2. Experimental Procedures

Typically, 0.7 g of potassium titanium oxalate was dissolved in 15 mL of distilled water, then 15 mL of 30% H2O2, 0.4 mL of 37% HCl, and 400 mg of NH4F is added to the solution. After five minutes of stirring, the final mixture was directly transferred into a 50 mL Teflon-lined stainless autoclave. The autoclave was maintained at 140°C for 24 h and afterwards allowed to cool to room temperature naturally. Finally, the white precipitate was collected, washed with distilled water and ethanol three times, respectively, then dried at 80°C for 12 h, and calcined in air at 500°C for 2 h.

The crystalline structure of the sample was characterized by X-ray diffraction (XRD) with Cu Kα radiation (λ = 1.5406 Å). The size and morphology of the sample were analyzed using a scanning electron microscopy (SEM, Hitachi S-4800) and a transmission electron microscopy (TEM, Phillips Tecnai 10) with an accelerating voltage of 100 kV. The BET surface area and pore size distribution of the product were measured by N2 adsorption-desorption test (Quantachrome, ASIC-2 measuring instrument). The UV-visible absorption diffuse reflectance spectra were measured on a TU-1901 spectrophotometer equipped with a labsphere diffuse reflectance accessory. The X-ray photoelectron spectroscopy (XPS) measurements were carried out using a VG ESCA Lab Mark II system with Mg Kα excitation.

Photocatalytic degradation of carbamazepine under UV and blue LED light irradiation was carried out in a Pyrex cylindrical batch photoreactor (containing 400 mL reaction slurry, as shown in previous work [2]) and a 100 mL beaker (containing 80 mL reaction slurry, as shown previously [16]), respectively. Agitation was provided by magnetic stirrer. The aqueous slurry, prepared with a given amount of catalyst 1.0 g/L and carbamazepine in concentration of 0.1 × 10−5 mol/L, was stirred in the dark for 30 min to ensure that the carbamazepine was adsorbed to saturation on the catalysts. A 10 W UV lamp (254 nm, GPH212T5L/4, Germany) and a 3 W blue LED lamp (~470 nm, Exploit 220024, China) were used as UV and visible light source, respectively.

The concentration of carbamazepine was determined by HPLC (Agilent 1200, USA) provided with a UV-Vis detector. A 4.6 mm × 250 mm (5 μm) XDB-C18 column was used. The analysis was carried out isocratically with an 60/40 (v/v) methanol/water mobile phase, and the flow rate was set at 1.0 mL/min.

3. Results and Discussion

The XRD patterns of the hydrangea-like F-doped TiO2  microspheres before and after calcinations at 500°C for 2 h are shown in Figure 1. Before calcination, the strong and sharp diffraction peaks indicate that the uncalcined products are well crystalline. All diffraction peaks can be perfectly indexed to the anatase phase of TiO2 (JCPDS 21-1272). No characteristic peaks of other impurities are detected in the XRD pattern, indicating that the F-doped TiO2 microspheres with high purity could be obtained under current facile synthetic conditions. Calcination at 500°C for 2 h does not change the phase and composition of the hydrangea-like F-doped TiO2 microspheres.

The morphologies and microstructures of the hydrangea-like F-doped TiO2  microspheres were characterized by scanning electron microscopy (SEM). Figure 2(a) shows a low-magnification SEM image of the as-prepared sample, which performs that the products consist of large-scale microspheres. The average external diameter of the microspheres is 2.85 μm, as observed by measuring 100 microspheres. A high-magnification SEM image (Figure 2(b)) reveals that the as-prepared products are hydrangea-like with many randomly attached petal-like structures. Moreover, the surface of the petal is not smooth but is composed of numerous TiO2 nanoparticles (as shown in Figure 2(c)). The morphology and the size of the hydrangea-like TiO2 microspheres remain unchanged, while it was calcined at 500°C for 2 h (see Figures 2(d)2(f)). The typical TEM images of the hydrangea-like microspheres before and after calcination at 500°C for 2 h are shown in Figure 3, and it is clear that the microsphere is solid with an external diameter of about 2.87 μm, which is in agreement with the SEM observation.

It has been found that both NH4F and H2O2 play important roles in the formation of the hydrangea-like structures. Without any addition of NH4F and H2O2, no precipitation can be obtained. Increasing H2O2 dosage to 15 mg without adding any NH4F, only aggregated nanoparticles can be observed. These two experiments indicate that the formation of hydrangea-like microspheres can be attributed to the HF generated during the hydrothermal process [17]. In addition, it has been found that microspheres closely packed with needle-like nanostructures could still be obtained when NH4F was replaced by NaF, while no microspheres could be produced when NH4Cl was used. This further confirms the crucial role of HF for the formation of hydrangea-like microspheres (see Figures 4(a) and 4(b)). Furthermore, the hydrangea-like microspheres could only be obtained when suitable amount of H2O2 was added in the reaction system (as shown in Figures 4(c) and 4(d)).

Figure 5(a) shows the nitrogen adsorption-desorption isotherms of the hydrangea-like F-doped TiO2 microspheres before and after calcinations at 500°C. Before calcination, the isotherms are of the typical type IV pattern with distinct H2 and H3 hysteretic loops in the range of 0.4–0.9 P/P0 and 0.9–1.0 P/P0, respectively, indicating the existence of ink-bottle and slit-shaped pores according to the IUPAC classification [18]. It has been pointed out that the bimodal pores are beneficial to the enhancement of photocatalytic performance due to faster diffusion of various reactants and byproducts [19]. After calcinations at 500°C, the hysteresis loops shifted to a higher relative pressure (P/P0) range and the area of the hysteresis loops gradually became smaller. This indicates the increase of average pore size and the decrease of pore volume [20]. The corresponding pore size distribution of the hydrangea-like F-doped TiO2 microspheres (see Figure 5(b)) was determined using the Barrett-Joyner-Halenda (BJH) method from the desorption branch of the isotherm. The BET surface area and average pore diameter of the hydrangea-like F-doped TiO2 microspheres before and after calcination are 2.74 m2/g and 8.0 nm, 1.03 m2/g, and 20.5 nm, respectively.

The optical band gaps of hydrangea-like F-doped TiO2 microspheres before and after calcination were studied by the UV-vis optical absorbance spectrum. The relationship between the absorption coefficient (α) and the photon energy (𝜈) can be written as shown in (1) [21]:(𝑎𝑣)𝑛=𝐵𝐸𝐸𝑔,(1) where 𝐵 is the constant related to the effective masses associated with the valence and conduction bands, 𝐸=𝜈 is the photon energy, 𝐸𝑔 is the band gap energy, and 𝑛=1/2 or 2, depending on whether the transition is indirect or direct. Figure 6 shows the absorption spectra of the hydrangea-like F-doped TiO2 microspheres before and after calcination. The inset shows the plots of (𝛼𝜈)1/2  versus the (𝜈). The band gap energy (𝐸𝑔) for the samples can be calculated by extrapolating the linear portion of (𝛼𝜈)1/2  versus the (𝜈) plot to α = 0. On the basis of these results, the optical band gaps for the products before and after calcinations are 2.92 and 2.97, respectively, which are lower than that of Degussa P25 TiO2  (𝐸𝑔 = 3.18–3.28 eV [22, 23]). The slight redshift of the optical band gap might be the result of Fanions successfully doped in the hydrangea-like TiO2 microspheres [8, 24].

X-ray photoelectron spectroscopy (XPS) measurements were carried out to confirm the chemical compositions and the existence forms of the elements in the F-doped TiO2 microspheres. The results are shown in Figure 7. In the spectrum, elements of Ti, F, O, and C can be observed. The C 1s at 284.6 eV is due to the adventitious hydrocarbon originated from the instrument itself. Figure 7(b) gives the high resolution XPS spectra of F 1s regions from the surfaces of the sample. It can be seen that there are two kinds of F states observed in the F 1s XPS spectrum. The low binding energy of around 683.8 eV could be ascribed to the Fanions physically adsorbed on the surface of TiO2, while the high binding energy located at 691.2 eV could be ascribed to the F atoms substituting for the O atoms, forming the Ti-F bonds [8, 25].

Figure 8(a) shows the degradation curves of carbamazepine as a function of reaction time in the presence of different photocatalysts under UV light irradiation. The removal rate dramatically increases with the addition of calcinated F-doped TiO2 microspheres, which is comparable to that of well-known commercial photocatalyst Degussa P25. The good photocatalytic activity of the products may be caused by the following three factors. First, the physically adsorbed Fanions on the surface of TiO2 can increase the photocatalytic degradation efficiency since the ·OH radicals generated on F-TiO2 surface are more mobile than those generated on pure TiO2 under UV irradiation [8, 26, 27]. Second, the enhancement of anatase crystallization can accelerate the transmission rate of photogenerated electrons and holes, and thus the photocatalytic activity of TiO2 is improved [28]. Third, the unique nanostructures are beneficial to the enhancement of photocatalytic performance due to faster diffusion of the reactant and byproducts [14].

The degradation curves of carbamazepine in the presence of hydrangea-like F-doped TiO2 microspheres before and after calcination under blue LED light (~470 nm) irradiation are shown in Figure 8(b). It has been found that LED light irradiation has no effect on the carbamazepine removal, no more than 0.5% of carbamazepine could be decomposed even after 24 h irradiation. It can also be seen that under identical conditions, the calcinated F-doped TiO2 microspheres possesses the highest removal efficiency. Though the removal rate of carbamazepine is only 46.7% at the reaction time of 24 h, the use of LED as the light source will make the application of heterogeneous photocatalysis practical since LED is now widely used in daily life.

Microspheres have taken an advantage over powder catalysts for separating the catalyst from solution by filtration or sedimentation. In our experimental, the TiO2 hollow microspheres can be separated from an aqueous suspension in less than 4 h by sedimentation (as shown in Figure 9), while the aqueous suspension of Degussa P25 is still turbid even after several days.

4. Conclusions

In summary, we report the hydrothermal synthesis of hydrangea-like F-doped TiO2 microspheres without using any templates or surfactants. It has been found that both of the NH4F and H2O2  dosages have important effects on the formation of the hydrangea-like structures. The hydrangea-like F-doped TiO2 microspheres calcined at 500°C exhibits higher photocatalytic activity than that of Degussa P25 under both UV and visible light irradiation. The enhanced photocatalytic activity can be attributed to the successful fluorine doping, good crystallinity, and the unique nanostructures. In addition, the use of LED as the light source will make the application of heterogeneous photocatalysis in practice possible.

Acknowledgments

This work was financially supported by the National Natural Science Foundation of China (Grant no. 51108406) and the Fundamental Research Funds for the Central Universities (2011FZA4022), to the authors we are grateful.