Abstract

TiO2/SiO2/Fe3O4 (TSF) core-shell nanoparticles with good photocatalytic activity that are capable of fast magnetic separation have been successfully prepared by chemical coprecipitation and two-step sol-gel process. The as-prepared TSF nanoparticles were calcined at high temperature in order to transform the amorphous titanium dioxide into a photoactive crystalline phase. The calcined nanoparticles are composed of a Fe3O4 core with a strong response to external magnetic fields, a SiO2 intermediary layer, and a TiO2 outshell. Vibration sample magnetometer (VSM) analysis confirms the superparamagnetism of calcined nanoparticles, which can enhance the recoverable properties of the novel photocatalyst. When the TiO2/SiO2/Fe3O4 core-shell nanoparticles are added to the crude oily wastewater, they exhibit high photocatalytic activity in the degradation of crude oily wastewater. The oil concentration could be reduced to lower than 30 ppm within 20 minutes for the case of initial oil concentration less than 350 ppm. It has been found that the TSF nanoparticles could be easily separated from the wastewater and withdrawn by using an external magnetic field. The recovered TSF nanoparticles possess high efficiency in the degradation of crude oily wastewater even after three times successive reuse. The present results indicate that TSF core-shell nanoparticles possess great application perspectives in the degradation of crude oily wastewater.

1. Introduction

Photocatalytic technology offers a facile and cheap method for removing inorganic and organic pollutants from wastewater [14], since most pollutants could be degraded or mineralized by use of photocatalytic degradation technology [57]. Photocatalysts used in UV or near-UV light activated processes are semiconductor materials such as TiO2, ZnO, and CdS [8, 9]: among them, the nanosized titanium dioxide is one of the most widely used photocatalysts due to its high photocatalytic activity, low cost, good stability, and nontoxic nature [1012].

When TiO2 photocatalyst is employed in the degradation of wastewater by photocatalytic process, it could be suspended in wastewater directly or be supported on substrate materials firstly and then immersed in wastewater with substrates. Nanosized TiO2 photocatalyst can either be suspended in wastewater or be supported on substrate materials. Nanosized TiO2 photocatalyst immobilized on substrate materials (such as glass, zeolite, silica, and ceramic) would benefit its separation from the wastewater. However, the activity of TiO2 photocatalyst in the fixed state is reduced to a considerable extent because the effective surface area of TiO2 photocatalyst decreases dramatically after the immobilization of TiO2. In addition, TiO2 photocatalyst may easily fall off the substrate materials, which make their complete recovery from wastewater difficult [13, 14]. So developing recoverable TiO2 photocatalysts with high photocatalytic activity is meaningful and imperative. It has been reported that magnetic separation provides a suitable solution of this problem for removing TiO2 photocatalysts from wastewater and reusing by applying external magnetic field [1521]. The combination of TiO2 photocatalyst and magnetic oxide nanoparticles (Fe3O4, γ-Fe2O3, and α-Fe2O3) may enhance the separation and recoverable property of nanosized TiO2 photocatalyst. However, compared with nanosized TiO2, the magnetic oxide nanoparticles are much more unstable, especially under acidic conditions. Beydoun et al. [22, 23] reported a photo-dissolution phenomenon, which was found in the coating anatase TiO2 directly onto magnetite. Electronic interactions will occur between TiO2 and magnetite core, which not only deteriorates the photocatalytic activity of TiO2, but also changes the magnetic properties of magnetite core. Furthermore, the preparing process of TiO2 photocatalyst usually involves a high temperature annealing, magnetic oxide nanoparticles such as Fe3O4 or γ-Fe2O3, if treated concurrently, may transform to antiferromagnetic α-Fe2O3, which will reduce the property of magnetic response [24]. A suitable solution of this problem is the utilization of a passive interlayer SiO2 between the magnetic core and TiO2 shell. It has been found the SiO2 layer promotes the photocatalytic activity of the catalyst by decreasing the negative effect of magnetic core [25, 26]. However, in prior studies the photocatalytic activity of combined photocatalysts did not show much improvement when compared with anatase-form nanoparticles, probably because the size of TiO2 shell and magnetic core was not controlled reasonably.

In present study, the TSF core-shell magnetic nanoparticles, which constitute a Fe3O4 layer, a SiO2 intermediary layer, and a TiO2 outer shell, have been successfully prepared. The TSF core-shell magnetic photocatalysts exhibit high photocatalytic efficiency, which is the same as the well-known commercial photocatalyst P25, a mixture of 80% anatase and 20% rutile form of TiO2 produced by Degussa Chemical Company (Germany). As a result, the TSF core-shell magnetic photocatalysts can be efficiently recovered from the solution by using external magnetic field for many times without significant loss of photocatalysts and photocatalytic activity. The photocatalytic activity of the as-prepared TSF core-shell magnetic nanoparticles and the recovered ones has been studied by photocatalytic experiments in the degradation of crude oily wastewater.

2. Experiment

2.1. Synthesis of Fe3O4 Core

The Fe3O4 superparamagnetic cores were prepared by the chemical coprecipitation method. The Fe3O4 cores were synthesized with the mixed solution of ferric chloride hexahydrate (FeCl3·6H2O) and ferric chloride tetrahydrate (FeCl2·4H2O) at 80°C for 1 h, together with a suitable amount of ammonia (NH4OH). The solution was centrifuged and washed with distilled water, and the Fe3O4 cores were dried for 12 h. The average size of the prepared Fe3O4 cores is around 8–10 nm.

2.2. Synthesis of SiO2/Fe3O4 Nanoparticle

The SiO2 intermediary layer was prepared by using the sol-gel method. The above Fe3O4 cores were mixed with ethanol (50 mL) and ammonia under vigorous magnetic stirring. TEOS was added into the solution slowly and then aged for 5 h. After being washed with ethanol three times, SiO2/Fe3O4 (SF) nanoparticle was dispersed in ethanol for using.

2.3. Synthesis of TiO2/SiO2/Fe3O4 Photocatalyst

The outer layer TiO2 was prepared by using the sol-gel method. The SiO2/Fe3O4 (SF) nanoparticles were mixed with ethanol (50 mL) and ammonia under ultrasonic dispersion for 30 min. An appropriate amount of TBOT was added into the solution drop by drop. The final photocatalysts were washed with ethanol three times, dried at 60°C for 48 h, and finally were calcined at 450°C for 3 h.

2.4. Characterization

The crystalline structure of particles was examined by using a Rigaku /max-RB X-ray diffraction (XRD) spectrometry with Cu-Ka radiation. The size and microstructure of samples have been characterized with JEM 200CX transmission electron microscope (TEM) with 200 kV operating voltage and JEOL-2011 high resolution transmission electron microscopy (HRTEM). The magnetic property of samples was measured with Lakeshore 7307 vibration sample magnetometer (VSM). The porosity of samples was measured by the nitrogen adsorption-desorption isotherm and BJH methods on the micromeritics ASAP 2000 specific surface area instrument. A UV-2802PC ultraviolet-visible spectrometer was used to measure the UV/Vis absorption spectrum of the solutions to monitor the concentration of crude oil at different time intervals.

2.5. Photocatalytic Degradation of Crude Oily Wastewater

Photocatalytic activities of the nanoparticles have been evaluated by degradation experiments of modulated crude oily wastewater in a self-made photocatalytic reactor. An 8 W UV lamp was used as the ultraviolet light source and air-blowing apparatus has been used; 0.2 g/L TSF and 0.067 g/L commercial TiO2 were suspended in the modulated crude oily wastewater. To determine the change of crude oil concentration during photocatalysis process, a few milliliters of solution was taken from the mixture at different time. Then, the nanoparticles were separated from the solution with a magnetic bar. The solution was subsequently mixed with CHCl3, followed by centrifugation. Finally, CHCl3 solution was taken out and the oil concentration was measured with UV-Vis spectrometer (UV-2802PC Unico). According to the measurement, the oil concentration was calculated based on the concentration-absorbance curve obtained by a standard measurement.

3. Results and Discussion

3.1. Characterization of the Nanoparticles

The successful preparation of Fe3O4 magnetic nanoparticles and the TSF core-shell magnetic photocatalysts have been confirmed by X-ray diffraction analysis (Figure 1). Figure 1(a) shows X-ray diffraction patterns of Fe3O4 core. The diffraction peaks located at 30.0°, 35.4°, 43.0°, 53.6°, 57.2°, and 62.5° can be perfectly indexed to the crystal planes (220), (311), (400), (422), (511), and (440) of magnetite phase Fe3O4, respectively. The average crystal size of Fe3O4 was calculated by using Scherrer’s formula , whereβ is the width of the XRD peak at the half-peak height, is the X-ray wavelength in nanometers, andθ is the half diffraction angle of 2θ in degrees. The average crystal size of Fe3O4, determined by the data from , is around 11.7 nm, which is approximately consistent with the TEM observation mentioned later. After two-step sol-gel processes and calcination, the X-ray spectra confirm the transformation of amorphous TiO2 to anatase, as shown in Figure 1(b). The diffraction peaks indicated by in Figure 1(b) are indexed as anatase TiO2, while diffraction peaks indicated by in Figure 1(b) are indexed as the magnetite Fe3O4. The wave packet that appears in the range of ~ might result from the interlayer of amorphous SiO2. No clear glassy sharp peak has been observed due to the fact that the thickness of the coated SiO2 layer is very small and the diffraction peak of amorphous structures is very weak if compared with the crystalline diffraction peaks. It approves that the coating layer of Fe3O4 is amorphous SiO2. “” peak is the diffraction peak of TiO2 (25.38°, 48.04°, and 37.80°), and it is the coincidence of three diffraction peaks of anatase, so it approves that the coating of SF nanoparticles is anatase TiO2. According to Scherrer’s formula, the average TSF photocatalyst size was around 19.1 nm.

The structure characterization of the as-prepared nanoparticles has been examined with transmission electron microscope (TEM) and high resolution transmission electron microscope (HRTEM). Figure 2(a) shows the morphology of Fe3O4 nanoparticles, and the size of Fe3O4 nanoparticles is around 5–20 nm, which agreed with the result obtained from XRD analysis. The direct coating of silicon oxide onto the surface of the magnetite core was carried out by the hydrolysis of TEOS using sol-gel method, which can be seen from Figure 2(b). The SF nanoparticles had a core-shell structure in which magnetite was predominantly concentrated in the center of the SF nanoparticles. The direct coating of titanium oxide onto the surface of the SF nanoparticles was carried out by the hydrolysis of TBOT using sol-gel method, as can be seen from Figures 2(c) and 2(d). Before calcination, titanium oxide was amorphous and after calcination at 450°C for 3 h, the amorphous TiO2 transformed to the anatase phase, which is confirmed by the peaks in the XRD pattern (Figure 1(b)). By using Scherrer’s formula, the size of TSF photocatalyst is around 10–30 nm, which matches the results of XRD analysis.

The Fourier transform infrared (FTIR) spectroscopy spectrum of the calcined sample was investigated to confirm the structure of core-shell nanoparticles. As shown in Figure 3, TSF core-shell nanoparticles show more signals than P25. The band at ca. 1620 cm−1 can be assigned to the H-O-H stretching vibration, and the band at ca. 1100 cm−1 and 800 cm−1 corresponds to the asymmetric vibration and symmetric vibration of Si-O-Si. The presence of water is proved by the stretching mode at 3400 cm−1. This surface hydroxylation will benefit the photocatalytic degradation of organic contaminants. Notably, no absorption peaks corresponding to Fe3O4 core are revealed, suggesting that it was totally coated by outer shell.

The magnetic properties of the Fe3O4 nanoparticles and TSF core-shell magnetic nanoparticles have been measured by use of the vibration sample magnetometer. The magnetization curves of Fe3O4 nanoparticles and TSF core-shell nanoparticles are shown in Figures 4(a) and 4(b), respectively. According to Figure 4, the saturation magnetization and the residual magnetization of Fe3O4 nanoparticles are 67.7 emu/g and 1.6 emu/g, respectively, while the coercivity of Fe3O4 core is close to zero, indicating the existence of superparamagnetism characteristics. After two-step coating and calcination, the saturation magnetization and the residual magnetization of TSF photocatalyst are 16.7 emu/g and 0.74 emu/g, respectively, while the coercivity of TSF photocatalyst is 5.78 Gs, which is still close to zero, confirming the superparamagnetism nature of TSF photocatalyst. Due to the superparamagnetism nature, TSF photocatalyst can demagnetize easily. When additional magnetic field intensity decreases to zero, the residual magnetization of TSF photocatalyst also drops to zero quickly, which would benefit the removal and demagnetization of TSF photocatalyst. It would provide the feasibility of the recovery and reusing of TSF photocatalyst.

The surface area of TSF photocatalyst and P25 was investigated by using nitrogen adsorption-desorption isotherms. The calculated BET surface area of TSF photocatalyst and P25 is 40.5 m2/g and 50 m2/g, respectively. It is known that large specific surface area is beneficial to the photocatalytic activity.

3.2. Photocatalytic Activity

We further demonstrated the photocatalytic performance of TSF photocatalyst by the removal of crude oil from oily wastewater. For comparison, we also evaluated the photocatalytic performance of P25 anatase TiO2. Figure 5(a) shows the absorption spectra of oily wastewater exposed to UV light for different time intervals. The typical absorption peak at 245 nm gradually decreases as the time increases and completely disappears after 40 min, implying the complete degradation of crude oil by the photocatalysts. Figure 5(b) shows that the oil concentration of oily wastewater varies with the photocatalytic degrading time. It shows that with the increase of degrading time the oil concentration decreases rapidly. After 10 min photocatalytic degradation, the oil concentration is reduced from 176 to 20 ppm. It indicates that the TSF core-shell magnetic nanoparticles possess high photocatalytic activity and can purify the crude oily wastewater efficiently. While keeping the amount of TiO2 in TSF photocatalyst and P25 TiO2 the same, we found that TSF photocatalyst and P25 TiO2 show similar photocatalytic activity under identical conditions, as shown in Figure 5(b). The good photocatalytic activity of TSF photocatalyst may be caused by two reasons. One reason is the small size of anatase phase formed during the calcination process. Some previous papers pointed out the optimal size of anatase for photocatalysis is around 10 nm [2729]. The other reason is the formation of Ti-O-Si bond, which has been reported in many papers to enhance the photocatalytic efficiency [3032]. Meanwhile, SiO2 intermediary layer inhibits electrical contact and prevents photogenerated electrons from transferring into the lower lying conduction band of the iron oxide core, thus eliminating the possible photodissolution of iron oxide in the reaction process.

Figure 5(c) shows that the curves of oil concentration vary with photocatalytic degradation time for different initial oil concentration in the wastewater. It has been found that even the initial oil concentration is as high as 340 ppm, the crude oily wastewater could be purified quickly and the oil concentration can be reduced to less than 30 ppm within 20 minutes. It indicates that TSF nanoparticles exhibit high efficiency. It has been shown that the photocatalytic degradation of crude oil follows pseudo-first-order kinetics according to Langmuir-Hinshelwood model, and the photocatalytic reaction can be described simply by , where and are the actual and initial oil concentration and is the degradation rate parameter. The kinetic parameter and pseudo-first-order model fitting are summarized in Table 1 and Figure 5(d). The sample with initial oil concentration 100 ppm exhibits the highest photocatalytic efficiency; its kinetic constant is 0.0231 min−1, which is much higher than that with initial oil concentration of 178 ppm or 340 ppm. The increase of oil concentration leads to a decrease in photocatalytic activity.

3.3. Recovery Properties

After photocatalytic degradation experiments, the TSF core-shell magnetic nanoparticles within the wastewater in the container could be easily withdrawn with an external magnetic field (NdFeB magnet). When the magnet is located at the bottom outside of the container, all of the TSF core-shell magnetic nanoparticles could be separated from the wastewater and attracted to the bottom of the container within only 5 minutes. The photocatalytic activity of the recovered photocatalysts has not noticeably changed after three successive cycles under UV-Vis irradiation indicating that magnetically recoverable photocatalyst is stable and effective for the degradation of crude oil. The photocatalytic performance of TSF after three cycles under UV-Vis irradiation is illustrated in Figure 6. The slight decrease in photocatalytic activity might result from the absorbance of crude oil, but it could be recovered by a high temperature calcination of the contaminant. Present results indicate that the magnetically recoverable TSF core-shell nanoparticles are stable and effective for the degradation of crude oily wastewater.

4. Conclusions

In summary, TSF core-shell magnetic nanoparticles, which are constituted by a Fe3O4 core, a SiO2 protective intermediary layer, and a TiO2 outshell, have been prepared by use of chemical coprecipitation and sol-gel processes. The as-prepared TSF core-shell magnetic nanoparticles display superparamagnetic behavior with saturated magnetization of 16.7 eum/g, residue magnetization of 0.74 eum/g, and coercivity of 5 Gs. It has been found that the TSF core-shell magnetic nanoparticles possess high photocatalytic activity in the degradation of crude oily wastewater. With the addition of 0.2 g/L TSF core-shell magnetic nanoparticles, the oil concentration in the crude oily wastewater could be reduced to lower than 30 ppm within 20 minutes for the initial oil concentration less than 350 ppm. After photocatalytic experiments, the TSF core-shell magnetic nanoparticles could be easily separated from the wastewater and withdrawn back by use of external magnetic field (a NdFeB magnet). It has been found that recovered TSF core-shell magnetic nanoparticles also exhibit high efficiency in the degradation of crude oily wastewater even for three cycles. Present results indicate that magnetically recoverable TSF core-shell magnetic nanoparticles are promising nanomaterials for degrading the crude oily wastewater.

Conflict of Interests

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

Acknowledgments

This work is supported by the Jiangsu University Scientific Research Foundation for Advanced Talents (Grant no. 12JDG095) and the National Natural Science Foundation of China (Grant no. 51302112).