Preparation of Particle Reusable Heterogeneous Catalyst Fe 3 O 4 / ATP for Methylene Blue Decolorization

New simple wet-ultrasonic method was employed to synthesize akind of particle heterogeneous catalyst Fe3O4/attapulgite (Fe3O4/ATP), which used attapulgite particles as catalyst carrier and Fe3O4 as active component loaded on the carrier. The catalyst is low cost and easy to prepare, and the effects of various preparing factors on its catalytic performance were investigated and discussed. The heterogeneous catalyst was characterized using SEM, FT-IR and XRD for its structure and catalysis properties. Fe3O4/ATP was used as the catalyst to decolorize high concentration (100mg/L) methylene blue (MB) in the heterogeneous Fenton system. Box-Behnken design (BBD) method was used for experimental design, data analysis and optimization. The influences of catalyst dosage, H2O2 initial concentration and pH value on MB decolorization ratios in Fe3O4/ATP/H2O2 heterogeneous Fenton system were studied. The heterogeneous catalyst Fe3O4/ATP has excellent catalytic performance and more than 99%MB decolorization were achieved under the optimum conditions. Higher decolorization ratio was obtained for recycled use of Fe3O4/ATP. Comparison experiments showed MB decolorization in Fe3O4/ATP/H2O2 system is really a catalytic process. Furthermore, Fe3O4/ATP can be easily separated from the solution and reused due to its particle form.


I. INTRODUCTION
The types and quantity of manufactured dyes are increasing with the rapid development of dye industry in China. This leads to increasing discharged dye wastewater into nature water systems annually. China has the most dye production in the world, but disposal rate of dye wastewater was less than 30% of the total, and most of that was not properly disposed. Various methods can be used for dye wastewater decolorization and degradation, including biological treatment [1], flocculation precipitation [2], adsorption [3], chemical oxidation [4], and catalytic oxidation [5], etc. In recent years, more and more researchers focused on advanced oxidation Fenton method and Fenton-like method to treat dye wastewater [6]- [9], especially the heterogeneous Fenton method. For examples, Valero -Luna et al. [10] prepared BaFe 12  was obtained with catalyst dosage of 0.75 g/L and hydrogen peroxide dosage of 12 mmol. Jiang et al. [11] synthesized novel Fe 2 O 3 nanosheets with double-shell hollow morphology by replica molding from diatomite framework, which was applied to dispose malachite green dye wastewater. Palas et al. [12] tested the catalytic performance of LaCuO 3 in photo Fenton-like oxidation in treating the food dye, Tartrazine under visible and UV light irradiation. Ma et al. [13] prepared the Fe-g-C 3 N 4 /GMC catalyst to treat acid red dye wastewater, which showed excellent catalytic properties in the range of pH 4 ~ 10.
Clay minerals have been used as the catalyst carrier, including montmorillonite, kaolin, bentonite, etc. Among those, attapulgite has gained a lot of research attention [14]- [16] due to its peculiar structural characteristics suitable for adsorbent and catalyst carrier [17], [18]. Zhang and Nan. [19], [20] prepared Fe 2 O 3 /attapulgite heterogeneous Fenton catalyst, and used it to degrade or decolorize surfactant SDBS and dyes MB and CR. More than 99% degradation or decolorization ratios can be obtained under the optimal conditions. Some composites containing Fe 3 O 4 and attapulgite were synthesized, such as attapulgite /Fe 3 O 4 /Au nanocomposites [21], attapulgite /Fe 3 O 4 magnetic nanoparticles [22] and attapulgite /Fe 3 O 4 /polyaniline nanocomposites [23], [24], for water treatment or water analysis. As can be seen, most Fe 3 O 4 /ATP composites were made to nanoparticle form in order to obtain high surface area. But nanocomposites have an obvious defect: it is difficult for nanocomposites to separate and reuse when they were used as adsorbents and catalysts. Although some nanocomposites can be separated by magnetic properties, the process is not as easy as that of particle composites.
To obtain a high-efficient, low-cost and easy-to-separate catalyst, in this study, attapulgite clay was used as a raw material to prepare a kind of particle heterogeneous catalyst Fe 3 O 4 /ATP, and its preparation conditions, reacting conditions and recycling properties of MB decolorization were studied.

B. Preparation and Characterization of Heterogeneous
Catalyst Fe 3 O 4 /ATP 20 g of ATP clay were weighed and mixed with water evenly and then aged for 24 h at room temperature. The aged ATP was granulated to small particle (D=1mm) and dried for 2 h at 105 ℃, then roasted for 2 h at 600 ℃. 5mmol Fe(acac) 3 was added into a 250 mL conical flask contained 80 mL oleylamine (or ethanol) and 20 mL n-hexaneinto, fully mixing. 5 gram spherical ATP particles were immersed in the mixture solution for a certain time in an ultrasonic water bath. The ATP particles were then separated from solution before being dried at a vacuum drying oven for 2 h under 0.07 Mpa and 40 ℃. After that, the particles were placed into muffle furnace and roasted without air for 2 h at 190 ℃. Thus, the particles heterogeneous catalyst Fe 3 O 4 /ATP was obtained. The process of preparing Fe 3 O 4 /ATP catalyst is shown in Fig.  1. Scanning electron microscope (SEM), transmission electron microscope (TEM), Fourier transform infrared spectroscopy (FT-IR) and X-ray diffraction (XRD) were used to characterize Fe 3 O 4 /ATP catalyst sample. A field-emission scanning electron microscope was utilized to determine the crystal morphology. Measurements were made on a JSM-5500 SEM instrument using a digital imaging process in room temperature at an acceleration voltage of 20 kV. To avoid charge accumulations, the samples were prepared electrically conductive by sputter coating with a thin layer of gold in vacuum conditions. And all samples were dispersed in ethanol before tests. A transmission electron microscopy was utilized to investigate the micromorphology and microstructure. Measurements were made on a JEM-1200EX TEM instrument. FT-IR spectroscope was used to confirm the chemical structure, any changes in the compositional or functional group during preparation of ATP-base catalyst. Measurements were made on a Nicolet AVTAR 360 FT-IR spectrometer after samples were mixed with 300mg of spectroscopic grade KBr and ground in an agate mortarin the range of 4000-400 cm −1 at room temperature. To investigate the crystalline structure and stability of ATP and Fe 3 O 4 /ATP composites, powder X-ray diffraction patterns wererecorded by X-ray diffract meter (Panalytical X Pert PRO XRD) using Cu K α radiation (λ=0.1542nm) at a rate of 0.02°/s in the range of 5° to 80°with an operating voltage of 40 kV and electric current of 150 mA.

C. Experimental Methods
Methylene blue (MB) was chosen as a target pollutant as a common cationic dye and its molecular weight of 373.90. A stock solution containing MB (1000mg/L), which can be diluted to the required concentrations for later degradation experiments, was prepared using distilled water. To obtain 100mg/L MB dye solution, 10mL MB solution (1000mg/L) was added in a conical flask and diluted with 90mL distilled water. The pH of the solution was adjusted by adding sodium hydroxide solution or sulfur acid. ATP-base catalyst samples with required dosage of hydrogen peroxide (30%w/w) were then added into the conical flask contained 100mL, 100mg/L MB solution .After being sealed the conical flasks were placed in a constant-temperature water bath. For H 2 O 2 solution, solid catalyst and MB solution, metric units of concentration used in this study were mmol/L, g/L and mg/L, respectively. The MB concentration was measured using a INESA 752N type ultraviolet-visible spectrophotometer by methylene blue spectrophotometry at λ of 664nm. The degradation ratio, i.e. the removal degree of MB can be calculated using the equation of η= where C 0 is the initial concentration of MB wastewater and C t is the concentration at contact time t.
Box-Behnken design (BBD) is a three-level incomplete factorial design extensively used in experimental design. BBD is performed to optimize the initial reaction rate andthe derivative of the initial portion of the kinetic curve, and it can be used to arrange in order the importance of the various influencing factors.
In order to explore the possibility of recycling and the stability of Fe 3 O 4 /ATP as a catalyst, at least 10 times repeated decolorization experiments were performed. To guarantee the same experimental conditions, for each run, the initial concentration of MB and H 2 O 2 were 100mg/L and 9.8mmol/L separately, and reused 0.5 g catalyst which was just washed by distilled water after reaction. The analytical methods for each run were also the same as mentioned above.

A. Optimization of Fe 3 O 4 /ATP Preparation
Catalysis effect of Fe 3 O 4 /ATP was influenced by many factors during the process of synthesis, such as solvent used, ultrasonic time, ultrasonic frequency, ultrasonic temperature and precursor concentration.
Solvents are very important during the preparation of Fe 3 O 4 /ATP. They might influence the dispersing condition of active components thereby they will influence the loading capacity of active components on the catalyst. We compared the MB decolorization ratios when using two different mixed solvents (alcohol and n-hexane, oleylamine and n-hexane) during the catalyst preparation. It can be seen from Fig. 2 that it achieves higher MB removal ratio (over 80%) as using alcohol and n-hexane solvents than that (70%) while using oleylamine and n-hexane. Furthermore, oleylamine has high viscosity, which is difficult to be mixed evenly with other solvent. The truth that it also generates a lot of waste indissolvable liquid, can't meet the low carbon demand. As a result, using ethanol + n-hexane as solvents is more efficient and friendlier to environment.
Precursor (Fe(acac) 3 ) solution concentration also had a great influence on the catalytic performance of Fe 3 O 4 /ATP. Fig. 3 shown that when precursor Fe(acac) 3 solution concentrations changed from 10 to 60 mmol/L during catalyst preparation, the decolorization ratios of MB firstly increased then went down. And Fe(acac) 3 solution concentrations of 30 mmol/L showed the highest catalytic performance in this heterogenous Fenton system. High precursor concentration enhances the number of Fe species loaded on ATP particles, but too much Fe species loaded would decrease the specific surface area of ATP particles which is tightly associated with catalytic and degrading performance.  In order to introduce Fe species into ATP and gain enough Fe 3 O 4 nano-particles on ATP surface, ATP particles should be immersed in precursor solution for almost 12 hours. While using ultrasonic can significantly reduce the immersing time. Ultrasonic frequency, ultrasonic time and ultrasonic temperature significantly affected the performance of catalyst, as can be seen in Fig. 4. From Fig. 4(a) and 4(b), the MB removal efficiency increased with increasing of the ultrasonic frequency and ultrasonic time. It is believed that higher frequency and longer time can cause the fast and efficient mass exchange, leading to high-load capacity. But the ultrasonic frequency and the ultrasonic time should better not be more than 100 kHz and 40min separately, due to safety considerations. As can be seen from Fig. 4(c), high ultrasonic temperature does not result in high-load capacity and high MB decolorization. It has the highest MB decolorization at 25℃ because too high temperature is not good for adsorption and too low temperature is not good for thermal motion of molecule.  (Fig. 5a). After the introduction of iron species into ATP, nano-Fe 3 O 4 particles were coated evenly onto the surface of ATP (Fig. 5b).
The FT-IR results of ATP and Fe 3 O 4 /ATP can be seen in Fig. 6. The strong wide absorption bands appeared at 3420cm -1 associated with the surface hydroxyl groups. The weak sharp absorption bands at 1630cm -1 is flexural vibrations of water H-O-H. The strong wide absorption bands at 1040cm -1 is dissymmetry stretching vibrations of Si-O-Si, and the absorption bands at 777cm -1 and 476cm -1 are symmetry stretching and flexural vibrations of Si-O. Some peaks were observed at lower wavenumbers (480cm -1 and 565cm -1 ) corresponding to vibration modes of Fe-O bonds of Fe 3 O 4 nanoparticles on the surface of ATP.

C. Process Optimization of MB Decolorization by Fe 3 O 4 /ATP/H 2 O 2 System
BBD method is used for the experimental design, data analysis and optimization. Reaction conditions of heterogeneous catalysis system such as temperature, pH value, H 2 O 2 dosage and catalyst dosage have great effect on MB decolorization ratio. It is known that there is a positive correlation between temperature and MB decolorization. It takes 24 h to obtain 99% MB decolorization ratio at 20℃, while it takes only 1 h to obtain the same ratio at 60 ℃when using heterogeneous Fenton system to decolorize MB . Thus all the experiments were conducted at 60℃. Parameters chosen for process optimization are pH (2~12), Fe 2 O 3 /ATP dosage (2g/L~10g/L) and H 2 O 2 dosage (19.8 mmol/L~98 mmol/L). The data of 17 experimental runs of BBD method were analyzed by Design Expert 8.0.6.1 software. The optimized parameters for maximizing the initial rate were also obtained using the established equation. According to the analysis by Design Expert 8.0.6.1 software, the analysis of variance (ANOVA) results of the quadratic model of the decolorization ratio was shown in Table I. The calculated F value of 59.78, much larger than the critical value of 3.68 for F 0.05 (9,7), implying that the derived quadratic polynomial model is significant. The model correlation coefficient of R 2 =0.9872, suggests that the there is good agreement between the experimental and predicted values of the degradation efficiency of MB. The function presented below was employed as the prediction model: where Y is the decolorization ratio of MB (%), X 1 is pH value, X 2 is H 2 O 2 concentration, X 3 is Fe 3 O 4 /ATP dosage. The derived model is adequate to perform the process variables optimization of Fe 3 O 4 /ATP/H 2 O 2 system for the decolorization of MB, significant at 95% confidence level (p=0.0273<0.05). Fig. 8 shows the 3D response surface modeling representing the effects of catalyst dosage, initial H 2 O 2 concentration and the initial pH on the decolorization of MB by heterogeneous Fenton Fe 3 O 4 /ATP/H 2 O 2 system. It can be seen from Fig. 8(a) and Fig. 8(b)

D. Repeated Decolorization Experiments
10 times repeated decolorization experiments were performed to test the recycling ability of Fe 3 O 4 /ATP as a catalyst, as shown in Fig. 9. At the first run, the decolorization efficiency of MB is only 64.6%, in subsequent runs, the decolorization efficiency increases. At the fifth, sixth and seventh runs, the ratios are almost 100%, and at next several runs, the efficiency decreases a little, but still is over 90%. This phenomenon could be explained that the products of MB degradation in the first several runs, such as acetic acid, ethyl alcohol, etc., gradually activate the Fe 3 O 4 /ATP and make it more efficient than before. The decreasing of decolorization ratio in the last several runs may be explained by that adsorbing too more MB or its degradation products on catalyst hinder the surface catalytic reactions between Fe 3 O 4 and H 2 O 2 .

IV. CONCLUSIONS
In this study attapulgite was used as the carrier to prepare heterogeneous catalyst Fe 3 O 4 /ATP successfully which demonstrated higher efficiency in MB degradation. Systematic experiments under various factors during the preparation of the catalyst were performed to optimize the catalyst performance and preparation methods. Ethanol solvent is superior to oil amine solvent for leading better catalytic performance and less pollution to the environment when used in Fe 3 O 4 /ATP catalyst preparing. Low temperature and moderate concentration of the predecessor solution will benefit the catalytic performance. Ultrasonic treatment can effectively shorten the dipping time of the catalyst in predecessor solution and improve the load capacity of the active components on catalyst supporter at the same time. As a result, the optimal preparing conditions of Fe 3 O 4 /ATP catalyst are: ethanol is as solvent, n-hexane is as dispersant, predecess or solution temperature is 25 ℃, predecessor solution concentration is 30 mmol/L, ultrasonic time is 30 min, and ultrasonic frequency is 100 KHz. Catalyst dosage, pH value and hydrogen peroxide dosage are all the important factors influencing the removal rate of MB, and BBD coupled with RSM was used to optimize the important parameters of heterogeneous Fenton-like reactions on Fe 3 O 4 /ATP for MB degradation. The results showed that the order of influence is pH>Fe 3 O 4 /ATP dosage>H 2 O 2 dosage. pH value has crucial effect on MB decolorization ratio in Fe 3 O 4 /ATP/H 2 O 2 system. MB removal ratios are high at acid and alkaline solution but low at neutral solution. The model predicted that the optimal conditions were pH 2, Fe 3 O 4 /ATP 10g/L and H 2 O 2 98mmol/L. Through the recycle experiments of the catalyst, it can be seen that the catalyst has good repeatable performance. The catalysis capacity of the recycled catalyst can be even better than that of initial ones. According to comparison experiments, Fe 3 O 4 /ATP/H 2 O 2 system has strong catalysis effect on MB degradation, rather than adsorption or direct oxidation.

CONFLICT OF INTEREST
The authors declare no conflicts of interest to this work.

AUTHOR CONTRIBUTIONS
Chunyuan Qian and Lingyu Dong conducted the research; Ting Zhang and Chunyuan Qian analyzed the data; Ting Zhang wrote the paper; all authors had approved the final version.