Electrospinning Preparation and Photocatalytic Activity of Porous TiO 2 Nanofibers

1 Photocatalyst Group, Kanagawa Academy of Science and Technology, KSP East 412, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan 2 Institute of Environmental and Analytical Sciences, College of Chemistry and Chemical Engineering, Henan University, Kaifeng 475001, China 3 Research Institute for Science and Technology, Energy and Environment, Photocatalyst Research Division, Tokyo University of Science, 1-3 Kagurazaka, Shinjuku-ku, Tokyo 162-8601, Japan


Introduction
TiO 2 -based photocatalysts have been one of the most active areas in heterogeneous catalysis due to their great catalytic abilities for removing environmental pollutants relating to waste water, polluted air, and spilling water [1,2]. Over the past few years, it has been demonstrated that the photocatalytic activity strongly depends on the specific surface area of photoactive materials [3][4][5]. Many strategies have been developed to fabricate porous and hollow TiO 2 nanostructures to enhance their photocatalytic performance [6][7][8][9][10]. Among these methods for generating porous nanostructures, especially for porous nanofibers, electrospinning is a well-established top-down approach [11][12][13][14][15]. For instance, Li et al. synthesized porous TiO 2 -based nanofibers via a combined electrospinning and alkali dissolution method; they firstly prepared TiO 2 /SiO 2 composite nanofibers by electrospinning and sintering and then leached out silica with alkaline solution to produce porous TiO 2 microstructures [16]. Mesoporous TiO 2 nanofibers were prepared by electrospinning TiO 2 -nanoparticle-containing polymeric spinning solution [17]. Furthermore, in our previous work, TiO 2 fibers with interior hollow channels were fabricated via multifluidic electrospinning with the obvious improvement of photocatalytic activity [18]. In this paper, porous TiO 2 nanofibers were fabricated through an electrospinning method with carbon nanospheres as sacrificial template. Firstly, the spinning solution containing carbon nanospheres was prepared. Our aim was to retain the carbon nanospheres in the as-spun nanofibers, followed by calcination leading to the thermal decomposition of carbon nanospheres, leaving behind a porous TiO 2 nanofiber structure.  using glucose as reagents as previously reported [19]. In brief, 3.5 g glucose (Wako, Japan) was dissolved in 40 mL of water to form a clear solution, which was placed in a teflonsealed autoclave (50 mL) and maintained at 170 • C for 5 h. The black or puce products were isolated by centrifugation, washed several times under sonication with water and ethanol, then oven-dried at 60 • C for characterization and subsequent application.

Preparation of Porous TiO 2 Nanofibers.
In the typical electrospinning procedure, 0.3 g of carbon nanospheres was dissolved in 10 mL of ethanol to form a homogeneous solution, followed by adding 1.0 g of polyvinylpyrrolidone (PVP, Mw = 130,000, Aldrich) and 3 mL of acetic acid (Wako, Japan). After stirring for one hour, this solution was added to 1.0 mL of titanium tetraisopropoxide (Ti(OiPr) 4 , Aldrich) (Wako, Japan) and stirred for another hour. The mixed solution was fed through a nozzle at a feeding rate of 4 mL min −1 . The applied voltage was 22 kV and the working distance was 15 cm. After electrospinning process, the obtained fibers were calcined in air at 500 • C for 2 h at a heating rate of 5 • C min −1 to form porous TiO 2 nanofibers, which were denoted as P-TiO 2 nanofibers. For comparison, we have prepared TiO 2 nanofibers in the absence of carbon nanospheres, and the rest of the synthesis conditions were kept the same as those described above in the typical experiment.

Characterization.
The morphology of the nanofibers was examined using scanning electron microscopy (JSM-5400, JEOL, Japan). X-ray diffraction (XRD) patterns were recorded using a Rigaku RINT 1500 diffractometer with the use of Cu Kα radiation. Nitrogen adsorption isotherms were measured at −196 • C using a Micromeritics system (Gemini V, Shimadzu, Japan). UV-vis absorption was performed on UV-1700 (Shimadzu, Japan) spectrophotometer.

Results and Discussion
The crystalline structure and phase composition of the products were characterized by XRD. As shown in Figure 1  nanofibers, suggesting that the crystal phase is not changed during the process of generating pores.
Carbon nanospheres were investigated by SEM analyses. Figure 2 represents SEM image and the corresponding histogram of the diameters of carbon nanospheres. The nanospheres are monodispersed and uniform on a large step. Moreover, during the calcination carbon nanospheres only release nontoxic CO 2 and H 2 O but not toxic gases. Figure 3 shows SEM images of TiO 2 nanofibers and P-TiO 2 nanofibers before and after calcination. It was observed that the surface of nanofibers without carbon nanospheres was smooth (Figure 3(a)) while the surface of nanofibers with carbon nanospheres was lumpy at places where the nanospheres were present (Figure 3(c)). After calcination, the average diameters of the nanofibers decreased slightly (Figures 3(b) and 3(d)). This is because the organic components (PVP or carbon nanospheres) were removed by combustion and the inorganic component had crystallized. The carbon nanospheres were removed via calcination leaving behind pores in the nanofibers. The diameters of nonporous TiO 2 nanofibers and P-TiO 2 nanofibers were statistically determined to be 230 and 330 nm, respectively.
The porous structure within the P-TiO 2 nanofibers was further confirmed by the N 2 physisorption experiments. Despite its larger diameter, P-TiO 2 nanofibers have larger specific surface area (33 m 2 /g) than TiO 2 nanofibers (9 m 2 /g). As shown in Figure 4, the pores volume of P-TiO 2 nanofibers was larger than that of nonporous TiO 2 nanofibers. However, the P-TiO 2 nanofibers possess wide pore size distributions from several nanometers to tens of nanometers because of the complex interplay between the shrink of the nanofibers and the decomposition of organics during calcination.
More evidence on crystallinity of P-TiO 2 nanofibers was obtained from the TEM analysis. As shown in Figure 5, the distinct contrast derived from the difference of electron density in TEM image further confirms the porous structure of the nanofibers. The crystal lattice fringes indicates that the nanofibers are formed by geometrically random TiO 2 nanograins, which exhibit high crystallinity after calcination.
To determine the photocatalytic activity of P-TiO 2 nanofibers, the degradation experiments of MB against P-TiO 2 nanofibers, TiO 2 P25 (AEROXIDE), and TiO 2 nanofibers were carried out. In the dark (−200∼0 min), the relative concentration of MB in the solution is slightly decreased owing to the adsorption of MB on the catalysts; the concentration decrease is slightly greater for P-TiO 2 nanofibers than for TiO 2 nanofibers, which reflects that the surface area is greater for P-TiO 2 nanofibers. As shown in Figure 6(a), 90% of MB could be decomposed by P-TiO 2 nanofibers within 60 min and was degraded completely after 110 min. Although the most of MB was also decomposed in the presence of the P25 after the same period, the photocatalytic activity was lower than P-TiO 2 nanofibers. In contrast, the photocatalytic activity of TiO 2 nanofibers was lowest; about 50% MB remained after 60 min irradiation. The reaction rate constant of the MB degradation against P-TiO 2 nanofibers (0.035 min −1 ) is three times higher than that of TiO 2 nanofibers (0.011 min −1 ) (Figure 6(b)). The results indicate that P-TiO 2 nanofibers have greater photocatalytic activity than nonporous TiO 2 nanofibers, suggesting that the surface area is an important factor for adsorption and catalysis.

Conclusion
In conclusion, porous TiO 2 nanofibers were prepared via an efficient and simple electrospinning method by using carbon nanospheres as sacrificial template. Carbon nanospheres together with PVP could be removed at the same time Journal of Nanomaterials 5 during calcination. Microstructural analysis and properties evaluation revealed that porous TiO 2 nanofibers have higher surface area and enhanced photocatalytic activity compared to nonporous TiO 2 nanofibers. In addition, the method for synthesis of porous TiO 2 nanofibers presented here can be extended to synthesize other one-dimensional porous nanostructures.