Preparation of Magnetic-ZnO Nanocomposite by High Energy Milling Method for Methyl Orange Degradation

A magnetic Fe3O4/ZnO nanocomposite (NCs) was prepared by a high energy milling (HEM) method. In the present study, the ZnO catalyst was prepared through two ways. The ZnO was synthesized by coprecipitation method (ZnO (S)), and ZnO directly used a commercial product (ZnO (Ald)). The prepared NCs were characterized using X-ray diffraction (XRD), vibrating sample magnetometer (VSM), Fourier transform infrared (FTIR), transmission electron microscope (TEM), and UV-Vis spectrophotometer. The XRD refinement indicates that Fe3O4 nanoparticle (NP) is a single phase and well indexed to cubic spinal structured magnetite. The Fe3O4/ZnO (S) and Fe3O4/ZnO (Ald) NCs are consisted of Fe3O4 and ZnO phases. The VSM result show that Fe3O4 NP, Fe3O4/ZnO (S), and Fe3O4/ZnO (Ald) NCs possess super-paramagnetic properties with saturation magnetization (Ms) is 102 emu.g-1, 28 emu.g-1 and 26 emu. g-1, respectively. The TEM observation shows that the average diameter of Fe3O4 is approximately 15 nm, while the thickness both of ZnO shell is ranging 20 nm - 50 nm. The average diameter of TiO2 P25 particle as catalyst was observed about 20 nm. The photocatalytic activity of catalysts were evaluated based on the degradation of methyl orange (MO) dye solution. The result shows that at pH = 7, the Fe3O4/ZnO (Ald) NC can degrade the pollutant in MO dye solution to 99 %, where as at pH = 3, the catalyst TiO2 P25 degrade only 96%.


INTRODUCTION
Metal oxide semiconductor photocatalysts have been extensively studied in the fields of environmental purification. Among the various oxide semiconductor photo-catalysts zinc oxide (ZnO), and titanium dioxide (TiO 2 ) from the commercial Degussa product P25 are widely used as materials for photo-catalytic processes, because of its high photosensitivity, and environmentally friendly nature. ZnO and TiO 2 P25 have proven to be the most suitable materials for removing pollutants from the water with their high surface-to-volume ratio, high photosensitivity, good quantum efficiency, and non-toxic to the nature (Abdollahi et al. 2012;Álvarez et al. 2010). However, a main problem in industrial application of ZnO and TiO 2 P25 nanoparticles in slurry reactor system is the difficulties encountered in recollection of the nanoparticles from the treated waters. The separation step such as filtration is not enough to prevent the possible large scale loss and the potential secondary pollution caused by the loss ZnO and TiO 2 P25 nanoparticles (Abdollahi et al. 2012;Ahadpour Shal and Jafari 2014;Xu and Li 2014).
Recently, magnetically separable photocatalysts of the Fe 3 O 4 /ZnO NCs have attracted attention because of their scientific and technological importance in the environmental purification (Feng et al. 2014;Nikazar et al. 2014). In this case, both the Fe 3 O 4 magnetic core and the ZnO shell are of interest. The incorporation of Fe 3 O 4 magnetic core into the semiconductor ZnO possesses the property not only of high surface area to volume ratio but also the magnetic photo-catalysts have dual functions. Fe 3 O 4 magnetic component can be easily and effectively collected the catalysts with the help of a magnetic field, and the photocatalytic properties of ZnO shell can eliminate the organic pollutants in the wastewater.
Several methods have been already established for the preparation of magnetic photo-catalyst and their photo-catalytic properties. Nikazar et al. (2014)  In this work is focused that the magnetic Fe 3 O 4 /ZnO NCs were produced by a mechanochemical activation using high energy milling (HEM) process for methyl orange (MO) degradation. It is well known that the mechanochemical or the activation of chemical reactions by mechanical energy can lead many applications, it is from waste processing to the production of advanced materials and enhanced mechanical properties (Giwa et al. 2012;Iwasaki et al. 2010).
The ZnO NP as a catalyst is obtained in two ways. Firstly, ZnO is made from ZnCl 2 .6H 2 O precursor using coprecipitation method. Secondly, the ZnO was taken from a commercial product (Sigma-Aldrich) without further purification. The weight percent ratio between ZnO and Fe 3 O 4 is arranged of 1 : 4. The obtained Fe 3 O 4 /ZnO NCs were annealed at 550ºC for 2 hours, and then characterized using X-rays diffractometer, vibrating sample magnetometer, transmission electron microscope, and UV-Vis spectrophotometer. The photo-catalytic efficiency of Fe 3 O 4 /ZnO NCs at different pH solution of MO dye photodegradation was reported. These results were compared with the TiO 2 catalyst from commercial Degussa product P25 (Germany). In the frame of this research TiO 2 P25 was used as comparative material.

Methods
The Fe 3 O 4 NPs were synthesized with coprecipitation method using iron (III) chloride, FeCl 3 .6H 2 O and iron (II) chloride, FeCl 2 .4H 2 O with molar ratio of 2 : 1 according to the method used by Winatapura et al. (2014) with modification. The synthesis was performed at 70 o C using the 1-M NaOH and NH 4 OH (25%) solution as precipitate agents. The chemical reaction of this process is written as follows (Ahadpour Shal and Jafari 2014): The ZnO NPs was synthesis using coprecipitation method from ZnCl 2 .6H 2 O raw materials. In this work, 1-M of NaOH solution was added drop wise to ZnCl 2 .6H 2 O at 80 o C until pH= 12. The reaction can be written as: The white precipitate was collected, purified with water and dried at 100 ºC.
The preparation of Fe 3 O 4 /ZnO NCs was conducted by HEM (Certi-Prep 8000 M machine). The Fe 3 O 4 NPs and ZnO powders (with 1 : 4 weights ratio) were mixed in a Tungsten Carbide vial and milling for 10h. The weight ratio of Fe 3 O 4 /ZnO NCs and agate balls was arranged of 1 : 5. The Fe 3 O 4 /ZnO (S) NCs was annealed at 550 º C for 2h. This sample was labeled by Fe 3 O 4 /ZnO (S). The same procedure was then repeated for preparation of Fe 3 O 4 and ZnO commercial product (Sigma Aldrich), and labeled with Fe 3 O 4 /ZnO (Ald). X-ray diffraction (XRD) patterns of the NCs were collected on a Pan-Analytical diffractometer with CuK" radiation (# = 0.154 nm). The XRD data was analyzed using general structure analysis system (GSAS) (Larson and Von Dreele 2004). The magnetization of the sample was measured using a VSM (Oxford 1,2T). TEM images of NCs were obtained with a transmission electron microscope (JEOL JEM-1400) at an accelerating voltage of 120kV.
Photo-catalytic degradation of MO was performed in a slurry batch reactor which consisted of cylindrical beaker glass, magnetic stirrer and an UV-pen Ray 100 watt, # = 356 nm (Model: UVP Pen Ray 90-0016-01), located at the center of the reactor. The distance between an UV-pen lamp and MO dye solution was arranged of about 5 cm. In all experiments, 100 mL MO dye in water solution of 30 ppm concentration was adjusted to pH 3 and pH 7. Then 0.1 g of each Fe 3 O 4 /ZnO (S) and Fe 3 O 4 /ZnO (Ald) NCs catalyst is added, and the mixture was stirred magnetically to obtain homogeneous suspension. Before irradiation, the reaction mixture was put in darkness for 3 h to achieve maximum adsorption of the MO onto the catalyst surface. After 3 h, a sample was taken and photocatalyst particles were separated using a magnetic bar. The 455 nm of MO dye concentration remaining was determinated by UV-Vis spectrophotometer (Perkin-Elmer, Lambda 25) at a wavelength. These results are compared with TiO 2 P25 from the comemercial Degussa product (Sigma Aldrich).

RESULTS AND DISCUSSION
The refinement result of XRD pattern of the prepared Fe 3 O 4 NP, synthesized zinc oxide, ZnO (S) and zinc oxide commercial product (Sigma Aldrich), ZnO (Ald) are shown in   Figure 1(b) and 1(c) that both ZnO preparation results and the commercial product showed a single phase. A preferred growth orientation lies along the (101) crystallographic direction at 2" ! 36.35#. Figure 2 shows the refinement result of XRD Fe 3 O 4 /ZnO (S) and Fe 3 O 4 /ZnO (Ald) NCs produced very good quality of fitting with R factor is very small, and goodness of fit value !2 (chisquared). It is seem that all the diffraction peaks can be readily indexed to Fe 3 O 4 and ZnO phases. The diffraction peaks of Fe 3 O 4 phase are indexed by " ", while peaks intensity of the ZnO phase that marked by "+". Some peaks of intensity enhancement are detected due to peak overlapping, as seen in Fig. 2. This is in a good agreement with work done by Feng et al. (2014). Fig. 3 presents magnetization M curve of prepared Fe 3 O 4 NPs, before and after ZnO coating, versus applied field H between $10 and +10kOe measured by VSM. The magnetization saturation (M s ) value of Fe 3 O 4 NPs measured at room temperature exhibits an excellent magnetic saturation (M s ) is around 102emu.g -1 (Fig. 3(a)). The magnetic property Fe 3 O 4 NPs is superparamagnetic behaviour with no coercivity due to their small size. The M s Value was observed around 93 and 65emu.g -1 for Fe 3 O 4 bulk and NPs, respectively (Wei et al. 2012  The TEM micrograph of prepared Fe 3 O 4 NPs, Fe 3 O 4 /ZnO (S) NCs, Fe 3 O 4 /ZnO (Ald) NCs, and TiO 2 P25 Nps are presented in Fig. 4. The most of aggregated Fe 3 O 4 particles are spherical; with the average size of the most particles is nearly 15nm, as seen in Fig. 4a. Fig.  4b and 4c show TEM image of Fe 3 O 4 /ZnO (S) and Fe 3 O 4 /ZnO (Ald) NCs. It can be seen clearly that the whole Fe 3 O 4 NPs coated by the ZnO shell. The morphology of these Fe 3 O 4 /ZnO NCs was generally the same which shows a coreshell like structures. The dark part (black colour) of the Fe 3 O 4 /ZnO NCs is the Fe 3 O 4 core and the ZnO shell is the light parts (grey colour), as it is seen in Fig. 4(b) and 4(c). The thickness of the ZnO shell in both of Fe 3 O 4 /ZnO (Ald) and Fe 3 O 4 /ZnO (S) is between 20-50nm. This result is in a good agreement with Choi et al. (2011), which the magnetic composite consist of nanoscale grains and has a super-paramagnetic behaviour. TEM image of TiO 2 nanoparticles from commercial Degussa product P25 are shown in Fig. 4 (d). It is clear spherical and non homogenous structure can be seen in Fig. 4 (d) having diameter $ 20nm.
The photo-catalytic activity was expressed by means of the degradation efficiencies of methyl orange (MO) dye for different catalysts and condition (pH = 3 and 7) are presented in Fig. 5 and 6. This is due to the pH value of dye solution is an important parameter in photo-catalytic degradation reactions that are taking place on the surfaces of semiconductors. It determines the surface charge properties of the photocatalyst and the adsorption behaviour of pollutants.
Based on Fig 5 (a)       It has been reported by Álvarez et al. (2010), Apopei et al. (2014), and Xue et al. (2013) that TiO 2 P25 has an excellent photoactivity for removal pollutants in pharmaceutical and personal care products, 4-Chlorophenol and pollutants in methyl orange from an aqueous solution. The high photo-catalytic activity of TiO 2 P25 is probably due to finer crystallite size (20-25nm), and BET specific surface area $ 50m 2 .g -1 (Salazar 2010). Further, TiO 2 P25 has unique crystal structure, which is composed of anatase (80%) and rutile (20%). In spite of TiO 2 P25 remains the most widely used photo-catalyst in water treatment, the difficulty in the recovery of the catalyst because of its small size to search alternative materials. The decreased of photocatalytic activity of the Fe 3 O 4 /ZnO in acidic condition is most likely due to the photocorrosion of ZnO. It has been reported that the zinc hydroxide surface (Zn-OH) can become charged by reacting with H + (acidic environment) or OH -(basic environment) ions due to surface amphoteric reactions (Abdollahi et al. 2012), Feng et al. 2014 (Abdollahi et al. 2012;Wei et al. 2012). Figure 6. shows the relationship between irradiation time and degradation rate of MO dye solution at pH = 7 treated by Fe 3 O 4 /ZnO (S), Fe 3 O 4 /ZnO (Ald), TiO 2 P25, and non-catalysts under UV light. Based on the Table 1., the degradation rate of MO dye solution is obtained of 99% for Fe 3 O 4 /ZnO (Ald) NCs, 96% for Fe 3 O 4 /ZnO (S) NCs with 0.1g/100ml (1g/L) respectively, and 88 % of TiO 2 P25 NPs. While the decolouration rate of the MO dyes without any catalysts is about 3 %. It is clear that at neutral condition (pH = 7) the photo-catalytic activities of the Fe 3 O 4 /ZnO (Ald), and Fe 3 O 4 /ZnO (S) NCs show a better performance than that of TiO 2 P25 after 180 minutes UV-irradiation.
In fact, the weight fraction of ZnO catalyst in the Fe 3 O 4 /ZnO (Ald) and Fe 3 O 4 /ZnO (S) NCs with TiO 2 Degussa P25 NPs is the same. It is suggested that mechanical energy can induce growth of nanocrystals in ZnO powder and change their surface properties so that affect significantly the photocatalytic activities of Fe 3 O 4 /ZnO NCs for methyl orange dye degradation in neutral condition. Beside that, the presence of magnetic Fe 3 O 4 NPs with high M s Value (102emu.g -1 ) as the core in the ZnO system shell provides additional surface active site for adsorption of organic pollutants in MO dyes solution. This is consistent with the results conducted by Nikazar et al. (2014). They used precipitation method for Fe 3 O 4 /ZnO preparation, followed with calcinations temperature of 550 °C during 2h for phenol solution degradation after 5h UV-irradiation and produce high photocatalytic activity.
The high photo-catalytic performance Fe 3 O 4 /ZnO NCs can be attributed to the different band gaps between ZnO (E g = 3.37 eV) and Fe 3 O 4 (E g = 0.1 eV) and work function of ZnO and Fe 3 O 4 which promote interfacial electron hole separation in the photo-catalytic process.  (Ahadpour Shal and Jafari 2014). It has been reported that Fe 3+ ions in Fe 3 O 4 can act as photo-excited electron-trapping site to prevent the fast recombination photo-induced charge carriers, and also as a trigger for the enhanced photo-catalytic activity observed in Fe 3 O 4 -TiO 2 in visible light (Feng et al. 2014;Xu and Li 2014 The apparent first-order kinetic equation used to fit the experimental data was determined by equation : Where K app, is apparent rate constant, C 0 and C t represent the initial concentration and concentration at particular time, t of MO dye solution. The half life (t ! ) is determined by equation (A. Fisli 2014): The corresponding linear transforms in ln (C t /C o ) as a function of irradiation time are given in Fig. 7 and Fig. 8. At acidic environment (pH = 3), the TiO 2 P25 NPs decomposed the MO dye solution rapidly and completely in 180 minutes (Fig. 7). This phenomenon shows the highest photo-catalytic activity with the degradation rate constant (K app ) obtained of 0.8732 hour -1 and the half life (t ! ) is about 0.7938 hour, as seen in  However, at pH = 7 the reaction rate of Fe 3 O 4 /ZnO (S) and Fe 3 O 4 /ZnO (Ald) NCs can degrade the MO dye solution completely up to 96% (Fig. 8b) and 99% (Fig. 8c) Based on the Figure 8a and 8b, we can obtain the same half life (t ! ) for degradation process by TiO 2 P25 catalist (Fig. 8a) and the Fe 3 O 4 /ZnO (S) NC (Fig. 8b). The values are t ! = 0.6700 hour and t ! = 0.6705 h ($ 47 min) for the degradation of the MO dye solution by TiO 2 P25 catalyst and the Fe 3 O 4 /ZnO (S) NC, respectively under UV light, see Table 1. The results show that their reaction rates are almost the same under UV-irradiation. Evidently, the degradation of MO dye solution decreases rapidly with increasing the reaction time which is indicated of the MO dye reduction from coloured aqueous to colourless. appear to the most effective for removal of MO dye solution at neutral environment (pH = 7). In this condition, the mechanical energy can induce the growth of nanocrystals in zinc oxide powder, and change their surface properties. As a result, they can remove almost of the pollutants in the methyl orange (MO) dye solution. Furthermore, the Fe 3 O 4 /ZnO NCs can be easily collected and separated from the solution after the treatment process using a magnetic bar.