Ferric chloride hexahydrate (FeCl₃∙H₂O), potassium borohydride (KBH₄), tetraethyl orthosilicate (TEOS), ethanol (EtOH), polyethylene glycol (PEG 400) and Malachite green were procured from the Guangfu Fine Chemical Research Institute (Tianjin, China). All other chemicals were of analytical grade or the highest quality. Malachite green was selected for the decolorizing tests.

SiO₂-nZVI was artificially synthesized by using One-Step method in this study. Ferric iron was reduced by borohydride according to the following equation:

First, FeCl₃∙6H₂O (0.436 g) was diluted with the solution of ethanol/distilled water (7:3, 30 ml) and stirred in a serum bottle at room temperature. Then, 0.5 ml TEOS was added to the iron solution. The mixture was agitated in a magnetic stirrer at room temperature and this process required the protection of nitrogen gas. Black iron nanoparticles formed when 0.537 mol/l KBH₄ was added under an accurate control of flow rate (2 ml/min). The mixture was reacted at room temperature under a nitrogen flow for 15 min. The synthesized iron particles were separated from the solution by using magnets. The SiO₂-nZVI particles formed by this reaction were washed with deionized water to prevent immediate rusting.

Our study intends to set different preparation conditions to explore the similarities at different conditions of the formed layered structure of SiO₂-nZVI particles and examine antioxidant capability and reducing ability of SiO₂-nZVI coated particles prepared in different conditions.

Effects of ethanol/distilled water volume ratio: The SiO₂-nZVI prepared in various liquid systems containing different ethanol/distilled water volume ratio (1:2, 1:1, 2:1 and 3:1).

Effects of PEG: We explored whether PEG was added to the SiO₂-nZVI preparation system.

Effects of TEOS dosage: Various dosages (0.3 - 1.5 ml) of TEOS were added to the reaction mixture.

Effect of TEOS hydrolysis time: The reaction mixture was reacted at various TEOS hydrolysis time (0.5 - 4 h).

The SiO₂-nZVI nanoparticles which were prepared with different conditions were used in dye decolorization test to evaluate their reducing activity. The SiO₂-nZVI nanoparticles were incubated with Malachite green solution (100 mg/l). The samples were taken at different time interval and analyzed for dye degradation. The dye decolorization rate was estimated by measuring the absorbance at the λ_max (618 nm) of the dye in a UV-vis spectrophotometer (UV-2800, Unico Instruments Co., Ltd., Shanghai, China). The rate of decolorization was calculated from the difference between initial and final absorbance values.

TEM was used to visually identify the morphology of SiO₂-nZVI. TEM was performed at accelerating voltage of 200 kV. The sample was supported by a copper-mesh TEM grid and dried in air at room temperature. The elemental composition was performed by energy dispersive spectrometry (EDS). XRD was used to investigate the material structure of the SiO₂-nZVI nanoparticles. The crystallographic structures of these nanoparticles were analyzed by XRD at wavelength of 0.026˚, scan range from 10˚ to 80˚, voltage 40 kV, electric current 40 mA.

The effects of the ethanol/H₂O volume ratio used in the synthesis of the SiO₂-nZVI on the degradation of Malachite green were examined. Table 1 showed the decolorization rate of Malachite green by SiO₂-nZVI at various ethanol/distilled water volume ratio. The higher decolorazation rate was obtained at volume ratio 2:1 and 3:1. After 25 min incubation, the decolorization rate was above 99%. The lower decolorization rate was observed at volume ratio 1:2 and 1:1, about 92% Malachite green was decolorized within 25 min.

Our study indicated that a moderate ethanol/distilled water volume ratio improved the decolorization ability of SiO₂-nZVI. It was reported that SiO₂-nZVI nanoparticles tend to agglomerate rapidly and form larger aggregates in water, thereby losing their reactivity [13] [14] . In the ethanol/distilled water system, adsorbed water molecules on the surface of the generated nanoparticles are replaced by ethanol. Thus, the SiO₂-nZVI particles obtained lower surface tension and surface energy, so it can effectively reduce agglomeration force of nanoparticles.

PEG is a dispersing agent and acts as a surfactant by adsorption at solid-liquid interfaces; forming a layer of molecules in a membrane prevents particles from being in contact with each other, reduces the surface tension and capillary adsorption capacity, and suppresses crystal growth, resulting in a small size distribution of the nanoparticles [15] . Based on these, the effects of PEG used in the synthesis of the SiO₂-nZVI on the degradation of Malachite green were examined in this study, and the result was shown in Table 2. After reaction for 15 min, the removal percentage of Malachite green increased to 99.9% for SiO₂-nZVI synthesized in the presence of PEG and to 90.7% for

those synthesized in the absence of PEG. These results showed that the nanoparticles’ reducing ability was related to PEG.

Table 3 showed the effects of TEOS dosage in the synthesis procedure of SiO₂-nZVI on Malachite green decolorization. When the dosage of TEOS was 0.5 ml, the azo dye decolorization rate was the highest. The increase in TEOS dosage in SiO₂-nZVI synthesis system leads to a decrease in the decolorization ability of the nanoparticles. Just half of the dye was degraded by SiO₂-nZVI when the dosage of TEOS was 1.5 ml. With increasing TEOS content, a tight mesh structure was no longer formed, so oxygen would migrate to the particle surface, form an oxidation film, and decrease the availability of reactive nanoparticles, resulting in a reduction in the removal rate of dye water [12] .

As shown in Table 4, the influence of coating time in the synthesis procedure on Malachite green degradation were evaluated with increasing reaction time from 0.5 to 4 h. The results showed that the optimum TEOS hydrolysis time for the SiO₂-nZVI preparation was 4 h, at which 100% of Malachite green was degraded within 15 min.

It was reported that the reaction time was the main influence factor on the coating layer structure [16] . TEOS is hydrolyzed under alkaline conditions, after the first ethoxy hydrolysis, the second ethoxy hydrolysis rate would gradually accelerate [17] . If the reaction time is insufficient, TEOS may only complete the first step of the reaction, causing incomplete hydrolysis an imprecise reticular structure, and a pore size too large to decrease antioxidation and the reducing capacity. Therefore, in our study, a coating time of 4 h gave SiO₂-nZVI with the best reducing and antioxidation capacities.

Representative TEM images of SiO₂-nZVI were shown in Figure 1. It could be seen that the SiO₂-nZVI particles size was about 30 nm. Moreover, from the TEM images, we could clearly see the nanoparticles coated completely with a transparent film. The SiO₂-nZVI was spherical and formed linear chains in space, and the surfaces of the coated particles were smooth and uniform. In addition, SiO₂-nZVI did not form large aggregates. The chemical composition of SiO₂-nZVI was also determined by EDS. The elemental composition of SiO₂-nZVI particles were analyzed by EDS, and the spectra confirmed the presence of Fe and Si in the Core-shell SiO₂-coated iron nanoparticles.

Figure 2 shows the results of XRD analysed freshly synthesized nZVI, and freshly synthesized SiO₂-nZVI samples. The diffraction peaks of freshly prepared nZVI located at 2θ = 44˚, 64˚ and 83˚ respectively. The diffraction peak of 44˚ reflected that zero- valent iron existed in sample. The other two diffraction peaks indicated that the sample contained iron oxides. The diffraction peaks of freshly prepared SiO₂-nZVI were located at 2θ = 27˚ and 44˚. The diffraction peaks from Si (2θ = 27˚) were found in sample. These results indicated that the nZVI particles and SiO₂ formed composite by the optimizing synthesis methods.