Application of pea-like yolk–shell structured Fe3O4@TiO2 nanosheets for photocatalytic and photo-Fenton oxidation of bisphenol-A

Uniform pea-like yolk–shell (PLYS) structured magnetic TiO2(PLYS-Fe3O4@TiO2) nanosheets have been prepared via a combined kinetics-controlled mechanical force-driven and hydrothermal etching assisted crystallization method and characterized. The resulting PLYS-Fe3O4@TiO2 nanosheets possess well defined yolk–shell structures with a large BET surface area (∼187.26 m2 g−1) and a strong magnetic susceptibility (∼17.4 emu g−1). The reaction rate constant was 24.2 × 10−2 min−1 as a result of oxidative decomposition of BPA using UV/PLYS-Fe3O4@TiO2/H2O2 system. This is 1.1 and 8.34 times faster than the BPA decomposition reaction rate constant in UV/TiO2/H2O2 and UV/Fe3O4/H2O2 systems, respectively. The synthesized catalyst also exhibited excellent recycle capability and excellent acid decomposition performance.


Introduction
Emerging organic contaminants (EOCs) are a burgeoning and extremely diverse class of contaminants that are not routinely monitored but have the great potential to enter the environment and may cause known or suspected adverse ecological and human health effects. EOCs of major concern include endocrine disrupting chemicals, pharmaceuticals and personal care products, surfactants, and various industrial additives as well as hormones. BPA, classied as an endocrine disruptor, is a contaminant of importance because it is extensively used in the production of polycarbonates, epoxy resins, and other plastics. 1 Due to the low concentration, high chemical stability and low biodegradability of BPA, the effectiveness of conventional treatment technologies such as adsorption, membrane ltration, and biological treatment are typically limited. 2 As an alternative, advanced oxidation processes (AOPs) have been demonstrated as a highly efficient technology for the destruction and mineralization of organic pollutants through powerful reactive oxidation species such as hydroxyl radicals (HOc). [3][4][5][6] Among various AOPs, homogeneous Fenton reaction based on ferrous ions and hydrogen peroxide is of great importance owing to the environmentally friendly characteristics. However, the narrow working pH range, difficulties to recover the dissolved metal ions and necessity for further treatment of ferric hydroxide sludge greatly hinder its wide application for practical water treatment. 7,8 To this end, heterogeneous Fenton process is developed as a valid approach to overcome these kinds of drawbacks.
Recently, Nemanja et al. 9 prepared an effective photo-Fenton catalyst of Fe/TiO 2 by deposition-precipitation method for degradation of thiacloprid. However, the catalyst separation issue seems to be a big challenge. Although Lejin et al. 10 synthesized Fe 3 O 4 magnetic particles via the co-precipitation of Fe 2+ and Fe 3+ method and observed the effect of different parameters, the degradation efficiency of 2,4-dichlorophenol is not high. Xiaoliang et al. 11 demonstrated the heterogeneous Fenton-like process for the treatment of methylene blue (MB), whereas the core-shell structured Fe 3 O 4 @C nanoparticles only worked well in acidic environment. Sheng-Tao et al. 12 showed the core-shell structured Fe 3 O 4 @SiO 2 nanoparticles as efficient Fenton-like catalyst in neutral environment for the degradation of MB but the mechanism still requires farther investigations. In addition, yolk-shell material with a distinctive cor-e@void@shell conguration, has stimulated considerable interest because of the void space between the core and the shell which can provide as a reactor. Dan et al. 13 reported that yolkshell structured Fe 3 O 4 @TiO 2 nanoparticles as a highperformance catalyst for the combination of photo-Fenton degradation of tetracycline. However, due to Fe 3 O 4 was fully covered by the TiO 2 shell, it can cause lots of disadvantages.
In this study, we design and prepare an advanced PLYS-Fe 3 -O 4 @TiO 2 nanosheets as a heterogeneous photocatalytic photo-Fenton catalyst. Instead of TiO 2 shell, TiO 2 nanosheets are coated which cannot only increase the surface area, but also shorter the penetration pathway of both H 2 O 2 and light. Most importantly, even aer 3 times cycle test, the catalyst still shows high activity for the degradation of BPA under a neutral condition.
2.2.2 Synthesis of Fe 3 O 4 @SiO 2 @TiO 2 nanoparticles. The core-shell Fe 3 O 4 @SiO 2 nanospheres were prepared according to a Stöber sol-gel method. 12,15 For a typical synthesis, an ethanol dispersion of the Fe 3 O 4 magnetite particles obtained above (6.0 mL, 0.05 g mL À1 ) was added to a three-neck round-bottom ask with ethanol (70 mL), deionized water (30 mL) and concentrated ammonia solution (2.0 mL, 28 wt%). The mixed solution was sonicated for 20 min. Then, 1.0 mL of TEOS was added dropwise in 5 min, and the reaction was allowed to proceed for 1 h under continuous mechanical stirring at room temperature. The resultant products (denoted as Fe 3 O 4 @SiO 2 ) were separated and collected with a magnet, followed by washing with deionized water and ethanol for 3 times, respectively.
The as-prepared Fe 3 O 4 @SiO 2 nanospheres were further coated with a TiO 2 shell through a kinetic-controlled Stöber method. Briey, the core-shell Fe 3 O 4 @SiO 2 nanospheres (0.1 g) were dispersed in ethanol (100 mL), and mixed with concentrated ammonia solution (0.80 mL, 28 wt%) under ultrasound for 15 min. Subsequently, 0.75 mL of TIPO was added dropwise in 5 min, and the reaction was allowed to proceed for 24 h at 45 C under continuous mechanical stirring. The resultant products (denoted as pea-like core-shell Fe 3 O 4 @SiO 2 @TiO 2 ) were separated with a magnet and washed with deionized water and ethanol for 3 times, respectively.
2.2.3 Synthesis of PLYS-Fe 3 O 4 @TiO 2 . The PLYS-Fe 3 O 4 @-TiO 2 nanoparticles were prepared through an alkaline hydrothermal etching assisted crystallization method. The above obtained Fe 3 O 4 @SiO 2 @TiO 2 nanoparticles (1.0 g) were mixed with NaOH solution (30 mL, 1.0 M), then transferred into a Teon-lined stainless-steel autoclave (100 mL in capacity). The autoclave was heated at 200 C for 24 h and then cooled to room temperature. The products were collected by a magnet and added in HCl solution (50 mL, 0.1 M) for 15 min, washed with deionized water until pH value was around 7 and subsequently dried at 60 C thoroughly in vacuum oven.

Materials characterization
X-ray diffraction (XRD) patterns were recorded on a Bruker D8X-ray diffractometer with Ni-ltered Cu Ka radiation (40 kV, 40 mA). Nitrogen sorption isotherms were measured at 77 K with a Micromeritics Tristar 3020 analyzer (USA). Prior to measurements, the samples were degassed in a vacuum at 180 C for 6 h. The Brunauer-Emmett-Teller (BET) method was utilized to calculate the specic surface areas (S BET ) using adsorption data in the relative pressure range P/P 0 ¼ 0.04-0.2. Using the Barrett-Joyner-Halenda (BJH) model, the pore size distributions were derived from the adsorption branches of the isotherms, and the total pore volumes (V) were estimated from the adsorbed amount at the relative pressure P/P 0 ¼ 0.995. Transmission electron microscopy (TEM) was carried out on a JEOL 2011 microscope (Japan) operated at 200 kV. For TEM measurements, the sample was suspended in ethanol and supported on a holey carbon lm on a Cu grid. Highresolution transmission electron microscopy (HRTEM) observations were performed on JEM-2100F transmission electron microscope with an accelerating voltage of 200 kV equipped with a post-column Gatan imaging lter (GIF-Tri-dium). Scanning electron microscopy (SEM) images were taken using a Zeiss ultra 55 ultrahigh resolutions thermal FEG with an in-lens electron optic operating at 3 kV. The magnetization was measured using a Vibrating Sample Magnetometer (EV9 including automatic sample rotation, Microsense, Japan) under a magnetic eld of 10 kOe and a temperature of 24 C.

Experimental procedure
All experiments were conducted in cylindrical batch reactors (F2.0 Â 10.0 cm, total volume: 50 mL) with a shaking water bath to mix and maintain the temperature (Fig. S1 †). A thermometer (Tecpel DTM-318) was used to measure the temperature. The light source ( Fig. S2 †) was HANSUNG G12T5 UV lamp (l ¼ 254 nm, 5 Â 10 W) and the light intensity in the centre of the BPA solution was 800 mW cm À2 . The reaction suspension was prepared by adding appropriate amounts of catalyst and H 2 O 2 into 30 mL BPA solution. The desired pH value was adjusted by HClO 4 or NaOH and measured by a pH meter (Orion 3 Star). Prior to addition of H 2 O 2 , the mixture was mixed in dark for 30 min to reach the adsorption/desorption equilibrium between the catalyst and pollutants. Aerwards, 1.0 mL of the suspension was removed using a 2 mL syringe at given time intervals and ltered via a membrane with a pore size of $0.45 mm. Furthermore, 10 mL 0.5 M n-butanol was added to the sample above to terminate the reaction and the BPA concentration in each sample was analysed on a high-performance liquid chromatography (Agilent 1260) with an Eclipse XDB C18 column (4.6 Â 250 mm, 5 mm) and a diode array UV detector (G4212B 1260 DAD, l ¼ 210 nm). For the reusability test, the catalysts were collected by magnetic separation, washed with deionized water several times, dried in vacuum and used it for the next reaction under similar experimental conditions. Experiments were carried out 3 times and all results were expressed as a mean value. The total organic carbon (TOC) was determined with a laboratory TOC analyzer (SIEVERS 5310C). Leached iron ions were detected by an Inductively Coupled Plasma Atomic Emission Spectrometer (ICP-AES; PerkinElmer 5300DV). Fig. 1 illustrates the diagrammatic sketch of the synthesis procedure which involves three steps. In step (1), the pre-synthesized magnetic Fe 3 O 4 nanoparticles were coated with silica layer via sol-gel method using TEOS as a precursor. In step (2), a further sol-gel process was used to coat TiO 2 shell on the silica layer using TIPO as the precursor. In step (3), aer a hydrothermal etching method followed by calcination process, PLYS-Fe 3 O 4 @TiO 2 nanosheets were formed.

Synthesis methods and characterizations
The uniform magnetite particles can be prepared through a facile solvothermal reaction based on a high temperature reduction of Fe(III) salts with ethylene glycol in the presence of trisodium citrate. SEM images clearly reveal that the obtained Fe 3 O 4 particles possess a uniform spherical shape with an average diameter of $130 nm ( Fig. 2(a)). The particles exhibit excellent dispersibility in polar solvents such as water and ethanol because of numerous citrate groups anchored on the surface, facilitating the subsequent coating with silica and titania. The PLYS-Fe 3 O 4 @TiO 2 nanospheres aer the rst sol-gel process show a relatively smooth surface with a diameter of $180 nm ( Fig. 2(b)). TEM images reveal that a silica layer with a thickness of $40 nm is uniformly coated onto the magnetic core, resulting in a well-dened core-shell structure.
The further sol-gel process leads to the formation of the pealike Fe 3 O 4 @SiO 2 @TiO 2 particles with an average thickness of TiO 2 layer is $120 nm (Fig. 2(c)). Aer the hydrothermal treatment, the PLYS-Fe 3 O 4 @TiO 2 nanoparticles with a uniform size of 900 nm are obtained (Fig. 2(d)). As represented in Fig. 2(e), PLYS-Fe 3 -O 4 @TiO 2 nanoparticles possess a unique yolk-shell structure that the diameters of the inner and of the outer layer are 200 and 320 nm. Fig. 2(f) clearly indicates that the lattice fringes of the nanosheet are 0.35 nm which can be analogous to the (101) planes of anatase TiO 2 . 16 The X-ray diffraction (XRD) pattern of PLYS-Fe 3 O 4 @TiO 2 ( Fig. 3(a)) shows six well resolved characteristic diffraction peaks of (320), (331), (400), (422), (511), (440), which are typical for Fe 3 O 4 crystalline phase. 17 In addition, new characteristic peaks can be clearly distinguished compared with XRD pattern of pure Fe 3 O 4 nanoparticles. N 2 sorption isothermal ( Fig. 3(b)) shows that the PLYS-Fe 3 O 4 @-TiO 2 nanoparticles have a nanoporous structure and the BET surface area is calculated to be 187.26 m 2 g À1 which is a little decrease compared with PLYS-Fe 3 O 4 @TiO 2 (208.38 m 2 g À1 ). The integrated energy dispersive X-ray spectroscopy (EDS) analysis of PLYS-Fe 3 O 4 @TiO 2 (Fig. S8 †) further conrms the presence of elements Ti, Fe and O. In addition, the Fe content is $13.17 wt%.
Meanwhile, the saturation magnetization value of pure Fe 3 O 4 and PLYS-Fe 3 O 4 @TiO 2 nanoparticle (Fig. 3(c)) were measured to be 51.7 and 17.4 emu g À1 respectively, which could be ascribed to the existence of TiO 2 nanosheets.
Furthermore, to observe the mechanism of the formation of PLYS-Fe 3 O 4 @TiO 2 double-shelled yolk-shell microspheres, a series of experiments were carried out. With the change of ammonia content (0.4, 0.5, 0.6 and 0.7 mL), the typical sandwich sphere structure was converted to a pea-like yolk-shell structure (Fig. S3 †). And using different diameter of mixing paddle (4, 5 and 6 cm), the length of pea-like particles was changed (Fig. S4 †) with different concentration of sodium hydroxide solution, the structure of PLYS-Fe 3 O 4 @TiO 2 spheres had a big change (Fig. S6 †), but there was no big difference for PLYS-Fe 3 O 4 @TiO 2 particles. However, the PLYS-Fe 3 O 4 @TiO 2 double-shelled structure will be collapsed under the high concentration of sodium hydroxide (Fig. S7A †). In addition, with higher concentration of hydrochloric acid, the pea-like yolk-shell structure would be changed (Fig. S7B †).
On the basis of the above observations, we propose a combination of kinetics-controlled mechanical force-driven growth (Fig. S5 †) and hydrothermal etching assisted crystallization method for the formation of the PLYS-Fe 3 O 4 @TiO 2 double-shelled yolk-shell nanoparticles. Because of the high initial ammonia content, the heterogeneous and homogenous nucleation processes simultaneously occur 18 and the pea-like core-shell Fe 3 O 4 @SiO 2 @TiO 2 nanoparticles can be formed under the mechanical force. 19,20 Subsequently, the amorphous silica layer is etched by NaOH solution rst, then the alkali solution can permeate and etch porous TiO 2 shell to form two layers titanate nanosheets. Furthermore, the reaction kinetic constants were evaluated through tting the experimental data with Langmuir-Hinshelwood model to better compare the catalytic performance. And the degradation kinetic curves can be assumed as pseudo rstorder kinetic eqn (1).
where C 0 is the initial concentration, C is the concentration at time t, k is the apparent rate constant, respectively. Fig. 4 shows the BPA decomposition reaction rate constants in different systems related to this study. Table S1 † Comparing reaction rate constants in different systems and conditions.
The results indicate that the relatively poor adsorption efficiency of BPA on the PLYS-Fe 3 O 4 @TiO 2 catalysts and the BPA oxidation capacity of H 2 O 2 directly is weaker than hydroxyl radical. It also suggests that the synergetic effect between photocatalytic process and photo-Fenton process, which not only inhibits the recombination between electrons and holes, but also accelerates the reaction speed of Fe 3+ to Fe 2+ to increase the kinetic constant. This is similar to the value given in the ref. 22 and 23.

Effect of H 2 O 2 dose.
Hydrogen peroxide plays the important role of BPA degradation in heterogeneous photo-Fenton reaction. The effect of different initial H 2 O 2 dose on the degradation of BPA in the heterogeneous photocatalytic photo-Fenton process was investigated (Fig. 5). When the H 2 O 2 dose was increased from 6.3 to 18.9 mM, the kinetic constant of BPA degradation increased from 1.7 Â 10 À3 to 19.0 Â 10 À3 min À1 correspondingly. However, the further increase of H 2 O 2 dose to 31.5 mM led to a decrease of kinetic constant to 9.6 Â 10 À3 min À1 . Theoretically, 72 mol of H 2 O 2 are needed to completely degrade 1 mol of the BPA (eqn (2)-(4)). From the results, it can be seen that the maximum degradation occurred with H 2 O 2 to BPA molar ratio which is 3 times as large as stoichiometric ratio.
Fe 2+ + H 2 O 2 / Fe 3+ + HOc + OH À The enhancement of degradation rate is because of the increase in HOc. 24,25 Moreover, further addition of H 2 O 2 dose did not improve the degradation efficiency, which may be explained by the scavenging effect of HOc at a higher H 2 O 2 dose eqn (5). 26,27 3.2.3 Effect of catalyst dose. The BPA concentration change with catalyst dose from 0.5 to 5.0 g L À1 and the pseudo rst order reaction rate constants are shown in Fig. 6. The maximum kinetic constant (24.2 Â 10 À3 min À1 ) was obtained with 1.5 g    L À1 , which is about 75.4% greater than the kinetic constant with 0.5 g L À1 (13.8 Â 10 À3 min À1 ). Furthermore, there is a decrease of kinetic constant which was 6.8 Â 10 À3 min À1 when the catalyst dose was 5.0 g L À1 .
The increase of the degradation rate might be attributed to a number of active sites on the surface of both Fe 3 O 4 and TiO 2 , which cannot only be occupied by H 2 O 2 , but also enhance the light utilization to generate more hydroxyl radicals. The decrease of kinetic constant aer 1.5 g L À1 might be due to three reasons, higher turbidity which can inhibit the further penetration of light into the reactor, the consuming of HOc by excess Fe 2+ and other radicals, such as HO 2 eqn (6) and (7) 28 and catalyst agglomeration. 29 Fe 2+ + HOc / Fe 3+ + OH À (6) 3.2.4 Effect of pH. It is well known that the pH value is an important parameter in the photocatalytic process. The effect of initial pH on BPA degradation in a range of 3 to 11 is shown in Fig. 7, which shows that the maximum kinetic constant at pH 3 is 33.4 Â 10 À3 min À1 . As the pH was increased from pH 5 to pH 11, the reaction rate constants were decreased to 25.4 Â 10 À3 , 14.5 Â 10 À3 and 9.9 Â 10 À3 min À1 , respectively. This may be attributed to that pH affects TiO 2 through the charge (eqn (8) and (9)) on the particle surface and aggregation size which can cause the change of specic surface area and light absorption. 30,31 On the other side, the increased oxidation efficiency at lower pH values can be due to the higher oxidation potential of hydroxyl radicals. Based on eqn (10) and (11) In addition, it should be noticed that catalytic activity of PLYS-Fe 3 O 4 @TiO 2 was slightly affected by pH values from 5 to 7. This phenomenon is important because one of the major drawbacks of Fenton reaction is the narrow range of pH.
3.2.5 Effect of initial concentration of BPA. The effect of the initial concentration of BPA on the photocatalytic photo-Fenton degradation was evaluated from 0.044 to 0.132 mM as shown in Fig. 8. The kinetic constant was increased from 4.5 Â 10 À3 to 24.2 Â 10 À3 min À1 with the increase of the initial concentration from 0.044 to 0.088 mM. When the initial concentration increased to 0.132 mM, the kinetic constant decreased to 2.5 Â 10 À3 min À1 .
This phenomenon may be related to the fact that the excessive dose of H 2 O 2 and catalyst for lower initial concentration of BPA which leading to the scavenging effect of HOc. 32 Furthermore, with the increasing of BPA concentration, more BPA molecules could be adsorbed on the surface of catalyst which can form the blocked active sites to decease the generation of HOc. 33,34

Mechanism investigation
According to the results of other researchers, n-butanol can react with HOc which generated from both surface and bulk solution. 35,36 The actual reactive species in the process was discriminated by determining the inuence of n-butanol as radical scavenger on the degradation of BPA. As shown in Fig. 9, 1 M n-butanol in solution can scavenge most of the HOc produced in the system and the kinetic constant decreased to 3.7 Â 10 À3 min À1 .
Based on all the information obtained above and previous studies by other researchers, 13,37-39 an assumed mechanism of BPA degradation by PLYS-Fe 3 O 4 @TiO 2 /H 2 O 2 system is illustrated in Fig. 10. The BPA is mainly removed by HOc which was generated from photo-Fenton and photocatalysis process as shown in eqn (3), (14), 40,41 (15) 42,43 and (16). 13 Beneting from the yolk-shell and nanosheet structure, BPA molecules can easily permeate into the surface of Fe 3 O 4 and degrade by hydroxyl radicals. Most importantly, the photo-induced electrons which generated from TiO 2 can not only promote the recovery of Fe 2+ from Fe 3+ , but also inhibit the recombination of electrons and holes.

Catalytic stability test
Recyclability is a crucial factor which can affect the catalyst application in economic perspective. [43][44][45] In order to observe the stability, the catalyst was collected by magnetic separation aer treatment, washed by deionized water and ethanol respectively, dried at 353 K and was evaluated by BPA degradation under the standard reaction conditions. As shown in Fig. 11(A), the degradation kinetic constant of reused catalyst is 24.2 Â 10 À3 , 21.1 Â 10 À3 and 20.3 Â 10 À3 min À1 for the rst, second and third run, respectively. In addition, the Fe leaching (Fig. S9 †) was less than 0.21 mg L À1 in the whole process. In addition, XRD analysis before and aer the reaction of PLYS-Fe 3 O 4 @TiO 2 catalyst in Fig. 11(B) showed that the characteristics of the catalyst did not change. The results demonstrated that PLYS-Fe 3 O 4 @TiO 2 may be used as a promising catalyst for BPA degradation because of good recyclability and stability.

Conclusion
In summary, pea-like yolk-shell structured PLYS-Fe 3 O 4 @TiO 2 has been successfully synthesized via a combination of kineticscontrolled mechanical force-driven growth and hydrothermal etching assisted crystallization method. And the catalyst was rst tested as a heterogeneous photocatalytic photo-Fenton catalyst for BPA degradation, catalyst and H 2 O 2 dose, initial pH and concentration of BPA are important variables on the degradation process. The rate constant of BPA degradation in PLYS-Fe 3 O 4 @TiO 2 / H 2 O 2 /UV system was 24.2 Â 10 À3 min À1 (pH ¼ 7). The BPA decomposition rate constant decreased with increasing pH. As a result of studying the rate of decomposition reaction according to the initial concentration of BPA, we found that there is an optimal BPA decomposition rate constant value at a certain concentration. Through the XRD analysis aer the cycling experiment and before and aer the reaction of the catalyst, the activity of the catalyst was still very stable, indicating that the catalyst had excellent stability and reusability. This study may provide useful information to further develop some effective heterogeneous photocatalytic photo-Fenton catalysts for degradation of organic pollutants.

Conflicts of interest
There are no conicts to declare.