Degradation of perchlorate using green method : Mechanism and kinetic process

This work analyzes the degradation of toxic perchlorate from wastewater and its conversion to nontoxic chloride ion through nano-scale zero valent iron (nZVI) coating. To prevent its agglomeration and to provide the required stabilization and more removal efficiency, it was coated with various coating agents which are co-friendly (green agents) and inexpensive. At first, nZVI was synthesized using green method. Thereafter, it was used for the removal of perchlorate. NZVI was characterized by X-ray diffraction (XRD), dynamic light scattering (DLS) and transmission electron microscopy (TEM) techniques. Data show that nZVI coated with starch had the size range of 20 to 60 nm and spherical morphology and thickness of about 60 nm. Analysis results of UV-Spectrophotometry and ion chromatography showed that perchlorate was removed with more efficiency (up over 90%) under optimal conditions, and for one week, it was coated with nZVI using starch. Also, the parameters of the removal efficiency include temperature, time reaction, pH and amount of nZVI. Activation energy (Ea) of 16.77 kJ mol -1 and constant rate (k) of 0.0242 min -1 were obtained from the removal of perchlorate under optimal conditions. The study shows that the obtained results improved more than previously.


INTRODUCTION
Perchlorate salts have been used widely in explosives, fireworks and propellant compositions (Sijimol et al., 2015;Hosseini et al., 2014Hosseini et al., , 2015)).The chlorine atom in perchlorate anion is a strong oxidizing agent, because of its higher oxidation state (+7) (Parker, 2009).Perchlorate is very resistant in the environment, and does not decompose for decades.Potential health effects are associated with perchlorate, because it can interfere with the ability of the thyroid gland to produce thyroid hormones (Srinivasan and Sorial, 2009;Motzer, 2001;McDougal et al., 2011).
For this reason, the US Environmental Protection Agency (EPA) has placed this anion on its Contaminant Candidate List (CCL) for drinking water, which requires the setting of MAC at 24.5 μg/dm 3 (Parker, 2009;Xiong et al., 2006).Thus, much attention is paid to perchlorate treatment before the discharge of wastewaters.
However, reduction of perchlorate from water has been *Corresponding author.E-mail: azamghavi@ymail.com.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License a challenge to researchers because perchlorate ions are non-volatile and highly soluble in water.Different treatment technologies including ion exchange (Xiong et al., 2006), biological treatment (Giblin, 2002), membrane filtration (Wang and Huang, 2008), electro dialysis (Malaisamy et al., 2011) and adsorption (Mahmudov and Huang, 2007) have been developed.These processes produce large volumes of concentrated waste residuals (Ahn et al., 2014).Zero-valent iron (ZVI) has been used as new, facile, simple and inexpensive tool for remediation of contaminated aquifers for 15 years.ZVI acts as a reducing agent of contaminants such as halogenated organics, heavy metals (Cd (II), As (II), Pb (II) and Cr (VI)), pesticides, nitro aromatic compounds, nitrates, and perchlorates (Grover, 2012;Zhou et al., 2010;Li, 2016).Nano zero valent iron (nZVI) is highly reactive due to its large surface area, which enables a more rapid degradation of contaminants, as well as the ability to treat otherwise recalcitrant contaminants (Comba et al., 2011).Prepared nZVI is agglomerated in aqueous media with its activity decreasing (Fan et al., 2006;Saleh et al., 2007).To prevent agglomeration and provide more efficiency of nZVI, it can be coated with hydrophobic agents such as carboxy methyl cellulose (CMC) (Kocur et al., 2013;Chen et al., 2012), carbon (Chen et al. 2016), poly acrylic acid (Mak and Chen, 2004), chitosan (Bing et al., 2009) and bentonite (Jin, 2011).A proper stabilizer must be co-friendly, inexpensive, should disperse metals well, and stimulate reaction on its surface.In this study, nZVI was synthesized via electrochemical reduction.Thereafter, the effect and activity of coated nZVI was compared with the degradation of perchlorate from solution samples at the desired time.Also, other effective parameters for removal including temperature, time reaction, pH, amount of nZVI and mZVI (micro ZVI), sustainability and efficiency of the nanoparticles, along with effect of oxygen in the reaction were researched (Oh, 2010).Finally, mechanism, rate constant (k) and activation energy (E a ) were studied.

Instrumentation
TEM images were obtained using electron microscopes (H-800, Hitachi, Japan).DLS analysis was performed with a Nicomp 380 Submicron Particle Sizer (PSS, Malvern, CA) at a measurement angle of 901° (internal He-Ne laser, wavelength 633 nm).DLS analysis indicated the dynamic particle size distribution of the nanoparticles in situ (aqueous solution).XRD analysis of the Fe 0 nanoparticles was carried out using a Model STOE diffractometer.Nickel filtered Cu-Kα radiation source was used to produce X-ray(λ =1.54178°A), and scattered radiation was measured with a proportional counter detector at a scan rate of 4° min -1 .The scanning angle was from 10 to 80°, operating at a voltage of 40 kV under potential current of 30 mA.A Hitachi model 3310 UV-Vis spectrophotometer with 1-cm quartz cells was used for recording absorbance spectra.

Preparation of coated ZVI nanoparticles
For the preparation of stabilized ZVI nanoparticles, 25 ml of 0.2 mol/L solution of FeSO4.7H2O was mixed with 5 ml of 1% starch solution (w/w), 1% CMC, and 1% nanoclay as coating.The mixture was purged under N2 gas for 30 min to remove dissolved oxygen (DO).Thereafter, Fe 0 nanoparticles were formed by adding dropwise 10 ml of 0.2 mol/L solution of sodium borohydride to the mixture.The color of mixture changed from reddish brown to light yellow and finally to black (Equation 1) (Fan et al., 2006;Zhao et al., 2007).Finally, nanoparticles were stored in the container until use. (1)

Degradation of the perchlorate by Fe 0 nanoparticles
Removal of perchlorate by nZVI was carried out using 50-ml glass vials.Desired amounts of ZVI nanoparticles suspension were added to 25 ml of perchlorate solution, with initial concentration of 2 mg L -1 (C0) after purging of N2 gas for 30 min.The pH and temperatures were adjusted.After adjustment of pH=7 and variable times of reaction, the remaining nZVI was separated from the solution using magnetic field and the remaining perchlorate (Ct) with chloride produced were measured.

Spectrophotometric determination of perchlorate
For monitoring of the perchlorate in the degradation process, a simple spectrophotometric method was used based on ion pairing formation with methylene blue reagent in acidic media, in this case, an aliquot of perchlorate solution.Its final concentrations of 1 ml of 1.5 mmol L -1 methylene blue solution and 1 ml of 2 mol L - 1 sulfuric acid solution were transferred into a 10-ml centrifuge tube and the contents were mixed well.The contents were diluted with deionized water and then 1 ml of chloroform, as an extraction solvent, was injected rapidly into a sample.The ion-pair product is extracted into chloroform.The mixture was centrifuged at 3800 rpm for 2 min; thereafter the supernatant aqueous phase was separated by a pipette.The absorption of the remaining organic phase was measured at λmax = 652 nm and finally, removal efficiency was calculated using the following equation: (2) where R is the ClO4 -reduction efficiency, C0 (mg L -1 ) is the initial concentration of ClO4 -, the solution and Ct (mg L -1 ) is the concentration of ClO4 -at every time.

RESULTS AND DISCUSSION
Characterization of coated Fe 0 nanoparticles Fe 0 nanoparticles were synthesized based on reduction of Fe(II) by sodium borohydride (NaBH 4 ) in the presence of starch solution, the best coating agent.The size of Fe 0 nanoparticles synthesized in starch solution was characterized by TEM, DLS and XRD.TEM images are demonstrated in Figure 1A; while Figure 1B revealed that the size distribution of the synthesized nZVI particles was widespread in the range of 20 to 60 nm (mean diameter of 40 nm).Also, DLS image of the coated nanoparticles is as shown in Figure 1C.The spectrum shows two peaks in which Peak 1 at 650 nm is related to 32.4% of the particles and Peak 2 at 103 nm is 67.6%.Comparison of TEM and DLS results confirm that Peak 1 correlates with the encapsulated nZVI particles and Peak 2 correlates with non-dissolved starch particles.Also, comparison of these images showed that thickness of the encapsulate is about 60 nm.XRD pattern of nZVI particles is as shown in Figure 2. The apparent peaks at 2θ of 44.64 and 65.16° indicated the presence of nZVI.The broad iron peak at 44.64° showed that the synthesized nZVI particles possess chemically disordered crystal structure (Oh, 2010).

Effect of the initial pH of solution
The effect of pH on the reduction of perchlorate was investigated at the pH range of 3.0 to 9.0, Fe 0 dosage of  , 2 ml of ZVI nanoparticle suspension, 2 mg L -1 of perchlorate initial concentration and temperature of 65°C.The reaction solution was stirred for about 30 min.Equations 3, 4 and 5 show the mechanism of perchlorate removal (Kocur et al., 2013;Chen et al., 2012).After completion of the process, their nanoparticles were separated by magnet and samples analyzed with the results as shown in Figure 3.As presented in this figure, the removal of perchlorate reached a maximum amount of 43.5% at pH=5.In the lower pHs, the removal efficiency reduced because of dissolution of nanoparticles.At higher pHs, reduction efficiency decreased rapidly due to masking of nZVI produced by Fe(OH) 3 .Thus, pH=5 was selected for future studies.

Effect of temperature
To study the effect of temperature on removal by nZVI, 25 ml of 2 mg L −1 of perchlorate solutions was transferred into the reactor and at temperatures of 25, 35, 45, 55 and 65°C.In this case, 0.3 g L -1 nZVI was used, and as previously explained, the initial pH was adjusted five times for 30 min.The results show that with increased temperature, the removal was enhanced up to 43.5% at a temperature of 65°C.Because increased temperature is not appropriate and not possible, temperature of 65°C is selected for future studies.

Effect of contact time
The effect of contact time on the removal of perchlorate was studied to reduce 25 ml of 2 mg L −1 of perchlorate solution at pH=5, Fe 0 dosage of 0.3 g L -1 and a temperature of 65°C.Here, the reduction of perchlorate was investigated at 20, 30, 40, 50 and 60 min, and the results are as shown in Figure 4.With increasing time, reduction of perchlorate increases.It can be seen that perchlorate removal at 60 min of the reaction is 3% more than 50 min, hence 50 min was selected as optimal time.

Effect of the amount of nZVI
To study the effect of nanoparticles concentration on removal of the perchlorate, pH = 5, perchlorate concentration of 2 mg L -1 , temperature of 65°C and contact time of 30 min were selected based on previous experiments and the amount of nZVI was changed from 0.15 to 0.6 g L -1 for 1 to 4 ml of ZVI nanoparticle , after 30 min, for better view of the effect of iron amount in less time, more than 73.8% of perchlorate was removed.However, when the Fe 0 concentration was 0.15 g L -1 , only 32.6% of perchlorate was removed.Thus, 0.6 g L -1 of Fe 0 concentration was selected for further works.

Optimum conditions
According to the results, the optimum conditions for removal of 25 ml of the perchlorate (2 mg L -1 ) (pH=5; temperature=65°C; reaction time=50 min and 0.6 g L -1 of iron nanoparticles) were chosen.Under these conditions, more than 90% of the perchlorate was reduced and was converted to non-toxicity chloride ion via electrochemical reduction mechanism by coating nZVI with emulsion starch of more than 90% (Figure 5).

Effect of various coating agents on degradation of perchlorate by nZVI
For dispersing reagents or modified surfaces, hydrophobic organic materials are commonly used.
Based on literature, nZVI was stabilized using hydrophobic organic materials, demonstrating enhanced reactivity along with good mixing and mobility (Chen et al., 2012).In current years, coatings done with green polyelectrolyte polymer have been tested with varying success; they are extremely inexpensive, non-toxic, soluble in natural water and biodegradable (Fan et al., 2013).In all the studied references, it was expressed that coated nanoparticles were used to remove contaminants immediately after synthesis.In this study, the effect of non-coated nZVI and some green coating agents such as nano clay, CMC and starch were investigated on nZVI performance for a week by measuring perchlorate removal percentage.Ion chloride is produced and their results are presented in Figures 6 and 7.It can be seen that the removal efficiency using modified nanoparticles with starch solution was maximum even after one week.So, starch solution was selected as stabilizer and coating agent.

Effect of oxygen on the reaction medium
In Equation 6, iron reacts with dissolved oxygen.In the nZVI/O 2 system, nZVI oxidation by dissolved oxygen produces Fe (II) and OH -via a two-electron transfer reaction from the Fe 0 surface for oxygen.At neutral pH, the oxidation of Fe (II) produces Fe (III) which can precipitate to form a layer of ferric oxide or hydroxide on Fe 0 surface further decreasing the rate of the process (Chen et al., 2010).
Degradation of 25 ml of the perchlorate solution (2 mg L -1 ) by nZVI was investigated under optimum conditions in the presence and absence of oxygen.The results showed that in the presence of oxygen, removal efficiency of the perchlorate was 10% and in the absence of oxygen, it was 90%.It is notable that to reduce DO the solutions were purged with purified N 2 for 30 min.

Perchlorate degradation by nano and micronized iron
The nano scale iron particles (1-100 nm) are characterized by high surface-area-to-volume ratios, high levels of steeped surface (Knighton and Geiger, 2010).NZVI has received much attention due to its smaller particle size, larger specific surface area, higher density of reactive surface sites and greater intrinsic reactivity of surface sites (Fan et al., 2013).This way, under optimum conditions, 0.6 g L -1 of 10 µm of iron particles was used rather than 0.6 g L -1 of iron nanoparticles.Thereafter, removal percentages of perchlorate were calculated with Equation 2. Removal percentage of perchlorate by micrometers of iron particles (mZVI) was 5% and by nanoparticles it was 90% (Figure 8).These values confirmed that iron micro particles are notable for removal of perchlorate.

Kinetic and activation energy of the perchlorate removal
The kinetic of the process was investigated at reaction times of 20, 30, 40 and 50 min and temperatures of 35, 45, 55 and 65°C, respectively.The results are shown in     9.According to the results, the degradation of perchlorate by nZVI can be interpreted using a pseudo-first-order rate model with correlation coefficient more than 0.9.The pseudo-first-order rate model is surveyed using Equations 7 and 8. ).The results show that pseudofirst-order constant rate improved twice when the temperature increased from 35 to 65°C, hence the reaction can be said to depend on temperature.When the constant rate versus reciprocal of the temperature was plotted logarithmically, a distinct linear relationship resulted, which allows computation of E a from the Arrhenius formula to be expressed as follows (Equation 9): (9) where E a (KJ.mol -1 ) is the Arrhenius activation energy or apparently the activation energy, A 0 is pre-exponential factor with the same dimension as k, and R is the universal gas constant (8.314J k -1 mol -1 ).The results here are presented in Figure 10.The apparent E a of the process is obtained as 16.77 kJ mol -1 which is much lower than activation energy using CMC-nZVI (52.5978.41kJ mol -1 ) and much more lower using nZVI alone (79.2 kJ.mol -1 ) (Zhao et al., 2007;Elliott et al., 2005).This result confirms the suitability of starch as a modifier of iron nanoparticles.

Conclusions
The perchlorate ion blocks the sodium iodide symporter (NIS) protein that normally acts as an iodide pump on the surface.Therefore, it is very toxic, hence its degradation from the aqueous solution and wastewater using various modified nano zero valent irons.In this method, perchlorate ion converts to non-toxicity chloride ion via electrochemical reduction mechanism and does not return to the environment.Also, this method is fast, simple, controllable, affordable and economical.Studies show that the stability of nanoparticles in their performance on perchlorate removal is important.Thus, they were coated with various coating agents which are co-friendly (green agents) and inexpensive and was tested as the performance of modified nZVI.The results show that the dimension which prepared iron nanoparticles in the presence of solution starch were 20 to 60 nm with a high efficiency.Under optimum conditions (temperature = 65°C, contact time = 50 min, pH = 5 and nZVI = 0.6 g L -1 ), along with the removal efficiency of perchlorate toxicity (2 mg L -1 )), nZVI coated with emulsion starch was more than 90% (Figure 10).The kinetic of the process was obtained as pseudo-firstorder with constant rate of 0.0242 min -1 . E a of the process was obtained as 16.77 KJ.mol -1 , which is much lower than the activation energy using CMC-nZVI (52.5978.41KJ.mol -1 ) and much more lower using non-coated nZVI (79.2 KJ.mol -1 ).

Figure 3 .
Figure 3.Effect of initial pH on removal of perchlorate (2 mg L -1 ) at temperature= 65˚C, time = 30 min and 0.3 g L -1 of coated nZVI.

Figure 4 .
Figure 4. Effect of contact time on removal of perchlorate (2 mg L -1 ) at initial pH = 5, temperature=65˚C and 0.3 g L -1 of coated nZVI.

Figure 7 .
Figure 7. Effect of various modifying agents on efficiency of nZVI in 5 days on the removal of perchlorate (2 mg L -1 ) at temperature=65˚C, initial pH = 5 and coated.

Figure 8 .
Figure 8.Effect of particle size (nano and micro iron) on the removal of perchlorate (2 mg L -1 ) at temperature=65°C, time= 60 min, initial pH = 5 and 0.6 g L -1 of ZVI.
t (min) and k is the observed pseudo-firstorder constant rate (min -1

Figure 10 .
Figure 10.Investigation of activation energy of the removal process in optimum conditions.

Table 1 .
Rate constants of perchlorate reduction by modified nZVI using starch at various temperatures under optimum conditions.