Degradation of Aniline Wastewater Using Dielectric Barrier

0 downloads 0 Views 2MB Size Report
Aniline degradation increased with increasing discharge power. Under the same conditions, .... sel to reduce aniline oxidation in the air. A stainless steel plate ...
Plasma Science and Technology, Vol.17, No.3, Mar. 2015

Degradation of Aniline Wastewater Using Dielectric Barrier Discharges at Atmospheric Pressure∗ WU Haixia (武海霞)1 , FANG Zhi (方志)2 , XU Yanhua (徐炎华)1 1

College of Environment, Nanjing Tech. University, Nanjing 210009, China School of Automation and Electrical Engineering, Nanjing Tech. University, Nanjing 210009, China

2

Abstract

Aniline is a toxic water pollutant detected in drinking water and surface water, and this chemical is harmful to both human and aquatic life. A dielectric barrier discharge (DBD) reactor was designed in this study to investigate the treatment of aniline in aqueous solution. Discharge characteristics were assessed by measuring voltage and current waveforms, capturing light emission images, and obtaining optical emission spectra. The effects of several parameters were analyzed, including treatment distance, discharge power, DBD treatment time, initial pH of aniline solutions, and addition of sodium carbonate and hydrogen peroxide to the treatment. Aniline degradation increased with increasing discharge power. Under the same conditions, higher degradation was obtained at a treatment distance of 0 mm than at other treatment distances. At a discharge power of 21.5 W, 84.32% of aniline was removed after 10 min of DBD treatment. Initial pH significantly influenced aniline degradation. Adding a certain dosage of sodium carbonate and hydrogen peroxide to the wastewater can accelerate the degradation rate of aniline. Possible degradation pathways of aniline by DBD plasmas were proposed based on the analytical data of GC/MS and TOC.

Keywords: non-thermal plasma (NTP), dielectric barrier discharge (DBD), aniline, degradation, wastewater treatment

PACS: 52.77.−j, 52.80.Wq DOI: 10.1088/1009-0630/17/3/10 (Some figures may appear in colour only in the online journal)

1

Introduction

O3 exhibit a strong oxidizing ability, which enables them to participate in the contaminant decomposition processes. NTP processes can rapidly and efficiently degrade various organic compounds, including 2,4-dinitrophenol [9] , phenol and its derivatives [10,11] , acetophenone [12] , ethylenediaminetetraacetic acid [13] , and organic dyes [14,15] . DBD is considered an effective discharge source for generating NTP because of its advantages of easy operation, stable and micro-discharge, as well as large treatment area. Using DBD technology to degrade toxic compounds in water is a new approach for wastewater treatment that has attracted significant interest in recent years. Manoj Kumar et al. used a parallel-plane coaxial DBD reactor to degrade crystal violet in water [15] . Young et al. reported a gas phase DBD reactor that was submerged in water; and the treated water in their experiment was used as one of the electrodes and was also used to degrade Orange II [16] . Biljana et al. designed a coaxial DBD reactor to treat azo dyes [17] . Lu et al. designed a granular activated carbon packed bed DBD reactor to remove pentachlorophenol from wastewater [18] . These studies

Aniline is frequently used in drug production, dye synthesis, agriculture, as well as polymer and rubber industries [1] . Even at very low concentrations, aniline is known to be carcinogenic and toxic to aquatic life. Aniline also reacts easily in the blood, converting hemoglobin into methahemoglobin and thus preventing oxygen uptake [2] . Therefore, industrial wastewater containing aniline should be treated before being discharged into the receiving water. At present, various technologies, such as physical methods, chemical oxidation, and biodegradation, have been applied for aniline removal from wastewater [3−6] . However, applications of these methods are restricted by poor efficiency, secondary contamination, and high cost [7] . In recent years, researchers have attempted to convert or decompose organic pollutants by using certain non-thermal plasma (NTP) processes, which generate a substantial amount of activated species as well as ultraviolet radiation via ionization, excitation and dissociation [8] . The species ·OH, ·H, ·O2 H and

∗ supported by National Natural Science Foundation of China (No. 51377075), the Natural Science Foundation of Jiangsu Province of China (No. BK20131412), the Environmental Protection Scientific Foundation of Jiangsu Province of China (No. 201004)

228

WU Haixia et al.: Degradation of Aniline Wastewater Using DBD at Atmospheric Pressure demonstrated that different DBD reactors could satisfactorily oxidize organic compounds in wastewater, and most focus was placed on plate-plate or coaxial DBD reactors. However, limited literature is available on using a needle-plate reactor for aniline degradation in aqueous solution. Moreover, the reported studies mainly focused on comparing the treatment effects before and after the treatment with fixed treatment parameter, and few studies concerning the degradation behavior and the mechanism of aniline by DBD have been reported. In this study, a new needle-plate DBD reactor with four rods used as high voltage electrodes (in water or at water–gas interface) was designed to investigate the feasibility of the degradation of aniline in water. The influence of various parameters, including applied voltage, treatment distance, pH value, as well as the addition of sodium carbonate and hydrogen peroxide, on the degradation efficiency was studied when the designed DBD reactor was used.

2

voltage power supply was used to power the reactor; the amplitude ranged from 0 kV to 30 kV and the frequency ranged from 1 kHz to 25 kHz. The voltage applied to the electrodes was measured by using a voltage probe (Tektronix P6015A). The discharge current and transported charges were measured by placing a 50 Ω non-inductance resistor (R1) between the bottom electrodes and the ground. The voltage and current waveforms were recorded by a digital oscilloscope, Tek TDS2014 (100 MHz, 1 GS/s). The light emission images from the discharge gap were recorded by a CCD digital camera (Canon PowerShot G6) mounted for capturing side pictures of the discharges. The images were taken with exposure time of 1 s. The activated species generated during the DBD treatment were measured by using an Ocean Optic S4000 spectrometer in the 200– 900 nm range with a resolution of 0.7 nm FWHM. The light emitted was collected by an optical fiber located at 1 cm from the glass vessel outer wall of the reactor, which was placed in a dark room to avoid external light interferences.

Experimental setup

The experimental system mainly consisted of an alternating current high-voltage power supply and a DBD reactor, as shown in Fig. 1. The DBD reactor consisted of a transparent quartz glass vessel, a high-voltage electrode and a grounded electrode. Four cylindrical stainless steel rods connected in parallel were used as highvoltage electrodes. One end of the steel rod was mechanically made to be hemispherical in shape to ensure that the breakdown occurred at a relatively low voltage. Compared with plate-shaped high-voltage electrodes, hemispherical-shaped high-voltage electrodes possess a stronger electric field around the top of the rod, and can thus trigger the occurrence of a breakdown at lower voltages across the gap between the high-voltage and the grounded electrodes. The length and diameter of the rods were 50 mm and 20 mm, respectively, and each rod was covered by a ceramic tube serving as dielectric barrier. The dimensions of the ceramic tube were as follows: inner diameter 20 mm; outer diameter 26 mm; length 50 mm; and thickness 3 mm. The distance between the two rods was 25 mm. The treated water was held in a transparent quartz glass vessel, which had an inner diameter of 200 mm, a wall thickness of 1.6 mm and a height of 50 mm. A cylindrical organic glass with four holes (the four electrodes go through the four holes) was used to cover the vessel to reduce aniline oxidation in the air. A stainless steel plate with a diameter of 30 cm was tightly attached to the outside wall of the quartz glass vessel as a grounded electrode. In this study, the gap between the top of the high-voltage rod electrode and the 250 mL treated water surface is defined as the treatment distance, which can be adjusted from 0 mm to 8 mm, as shown in Fig. 1(b). During the experiments, discharges occurred in the air–water interface (3 mm and 8 mm) and the water surface (0 mm). An adjustable high-

(a)

(b) Fig.1 Schematic diagram of the experimental system (a) and treatment distance (b)

Fig. 2(a) and (b) show the typical discharge image as well as the voltage and current waveforms that are measured during the treatment with peak applied voltage of 23 kV, frequency of 17 kHz, and treatment distance of 0 mm. A strong light emission around the top of the high-voltage rod electrodes at the interface of the rod electrodes with the water surface can be observed in Fig. 2(a). The discharge extends to a certain distance along the rod electrode surface, demonstrating a typical gas–liquid discharge. As shown in Fig. 2(b), many fine- and narrow-pulsed current spikes corresponding to microdischarges are present in each half-cycle of the applied voltage, indicating the occurrence of a filamentary DBD. The peak current is approximately 30 mA. The discharge power P can be calculated from the measured voltage and the current data by using the following formula: Z t0 + f1 u (t) i (t) dt, (1) P =f t0

229

Plasma Science and Technology, Vol.17, No.3, Mar. 2015 where f is the repetition frequency of applied voltage pulse, and u(t) and i(t) are the instantaneous discharge voltage and discharge current measured, respectively. With the instantaneous discharge voltage u(t) and current i(t) data shown in Fig. 2(b), Eq. (1) can be used to calculate the value of discharge power P , which is 21.5 W. In this study, the discharge power was varied by changing the amplitude of the applied voltage with fixed repetition frequency of the applied voltage of 17 kHz.

and conductivity of the solution were measured to be 6.2 and 17 µs/cm, respectively. The temperature of treated solutions rose during DBD treatment. The initial temperature of the solutions was 24 ◦ C, which increased to 34 ◦ C within 2 min of treatment. The temperature rose further for longer treatment times. To maintain consistent measurement conditions for long treatment times, solutions were rapidly cooled down to 34 ± 1 ◦ C with ice water before any measurement was taken. The concentration of aniline was measured through spectrophotometric quantification [19] . The pH value was measured by using a pH meter (model pHS3C, Shanghai Precision & Scientific Instrument Co., Ltd.). Total organic carbon (TOC) was measured by using a TOC-VCPH (Shimadzu, Japan) analyzer, which works through combustion method. The intermediates were detected by GC/MS (GC-TOF, Micromass).

Fig.3

Molecular structure of aniline

The degradation efficiency (percentage) was calculated by using the following formula: degradation efficiency (%) =

C0 − Ct × 100%, C0

(2)

where C0 and Ct are the initial and the final concentrations of the aniline solution, respectively.

3 3.1

Results and discussion Effect of treatment distance and discharge power

Treatment distance and discharge are two parameters that influence the degradation efficiency. To determine the effects of treatment distance on aniline degradation, three distances (i.e., 0 mm, 3 mm, and 8 mm) from the high-voltage electrode top to the water surface were considered while the discharge power was fixed at 21.5 W. To maintain the discharge power at 21.5 W for the three treatment distances, the applied peak voltages were adjusted accordingly; that is, 23 kV, 20 kV, and 18 kV for 0 mm, 3 mm, and 8 mm, respectively. Fig. 4 shows the experimental results of the degradation efficiency as a function of treatment time for three treatment distances. At each treatment distance, the conversion of aniline gradually increased with increasing treatment time. At the 6 min treatment time, the degradation efficiency reached 64.25% when the electrodes just touched the water surface (treatment distance of 0 mm), whereas the degradation efficiency was only 49.11% when the electrodes were 8 mm under the solution surface. The discharge power was adjusted by changing the applied peak voltage. The voltage

Fig.2 Light emission images at treatment distances of 0 mm, 23 kV (a), 3 mm, 20 kV (c), and 8 mm, 18 kV (d), and voltage and current waveforms at treatment distance of 0 mm, 23 kV (b)

All of the chemicals were of analytical grade and were used without further purification. Aniline (C6 H7 N, formula weight: 93.12) was purchased from Sinopharm Chemical Reagent Co., Ltd., and its molecular structure is given in Fig. 3. The target aqueous solutions were prepared by dissolving liquid aniline in deionized water. A 250 mL test solution (with a height of 8 mm in the vessel) with a 100 mg/L initial concentration of aniline was used for each DBD treatment. The pH value 230

WU Haixia et al.: Degradation of Aniline Wastewater Using DBD at Atmospheric Pressure

3.2

values of 15 kV, 17 kV, 20 kV and 23 kV were chosen, and the discharge powers were calculated to be 14.25 W, 17.96 W, 20.4 W, and 21.5 W, respectively. The treatment distance was fixed at 0 mm. Fig. 5 shows the aniline decomposition efficiencies as a function of treatment time for different discharge powers. At a certain treatment time, the degradation efficiency increased with an increase in discharge power. At the 10 min DBD treatment, the removal rate at 21.5 W was 84.32%, whereas the rate was only 18.31% at 14.25 W. Discharge was weak, and it occurred only on the interface surface of rod electrodes with water. Figs. 2(c) and 2(d) show that no discharge was found under the water when the high-voltage electrodes were under the water surface at treatment distances of 3 mm and 8 mm, whereas Fig. 2(a) shows that the discharge was strong and that the discharge area increased when the high electrode tops just touched the water surface at a treatment distance of 0 mm. In this case, the discharge also extended to the interface of rod electrodes with water surface. Thus, more activated species and UV intensity were produced with the increase of discharge area at a treatment distance of 0 mm, and aniline was decomposed effectively, as shown in Fig. 4. At a higher discharge power with a fixed treatment distance of 0 mm, more reactivated species may be generated in the DBD discharge regime. Consequently, more species can react with aniline molecules, resulting in an increased degradation rate, as shown in Fig. 5.

Effect of pH

The level of pH is an important factor that influences the removal efficiency. NaOH (0.1 mol/L) or H2 SO4 (0.05 mol/L) solutions were used to adjust the initial pH values to 3, 5.1, 7, 8.5, and 10. The used dosage of sulfuric acid is low. Therefore, the effects of the reaction between sulfuric acid and aniline can be disregarded. The conductivity of the solution affects the plasma treatment of pollutants [20] . After NaOH or H2 SO4 was added, the conductivity of the solutions changed accordingly. Therefore, Na2 SO4 was used to adjust the initial conductivity of the solutions to 800 µS/cm prior to the treatment to avoid the effects of conductivity on the treatment results. Fig. 6 shows the aniline decomposition efficiency as a function of treatment time for five pH values when the discharge power and treatment distance were fixed at 18.2 W and 0 mm, respectively.

Fig.6 Aniline decomposition efficiency as a function of treatment time for different pH values

As shown in Fig. 6, the initial pH values influence the removal rate of aniline. The decomposition of organic compound by DBD was lower for experiments conducted in acidic and neutral solutions than that in weak acid and alkaline solutions. A large number of H+ in the acidic solution may reduce the effective role of high-energy electrons. A few activated species such as HO2 and O2 were formed, and the formation of ·OH, ·H, ·O2 H and O3 was also reduced. Therefore, the decomposition of aniline in alkaline solutions was higher than that in acidic ones. In alkaline solutions, the selfdecomposition of ozone accelerated and the production of ·OH by ozone increased [21] . In other words, the existence of OH− species enhances the decomposition of O3 into the hydroxyl radical.

Fig.4 Effect of treatment distance on the variation of degradation efficiency versus treatment time of aniline

O3 + OH− → HO2 + O− 2

(3)

+ O− 2 + O3 + H → HO3 + O2

(4)

HO3 → ·OH + O2

(5)

The hydroxyl radical can decompose aniline into inorganic carbon, which exists in the form of HCO− 3 or CO2− (radical scavenger) in the solutions. At the be3 ginning, much more ·OH and aniline decomposition into inorganic carbon was obtained at a pH value of 10 than

Fig.5 Aniline decomposition efficiency as a function of treatment time for different input powers

231

Plasma Science and Technology, Vol.17, No.3, Mar. 2015 that at pH value of 8.5. With an increase in the abovementioned radical scavenger, ·OH and aniline decomposition was decreased in the solution with pH value of 10. This may be the reason why the degradation rate is higher at a pH value of 8.5 than that at pH value of 10.

3.3

to Acero et al., carbonate and bicarbonate are sometimes accelerators of hydroxyl radical in the simultaneous presence of hydrogen peroxide and ozone [23] . In this study, carbonate addition showed a synergistic effect on the degradation of organic compound with an increase of hydrogen peroxide and ozone in the system. Therefore, in longer treatment time, the degradations of aniline with 50 mg/L and 100 mg/L sodium carbonate addition were higher than that with 0 mg/L addition.

Effect of sodium carbonate

Industrial wastewater usually contains different inorganic ions, such as chloride ion, sulfate, and carbonate. These inorganic ions affect the efficiency of wastewater treatment by DBD plasma. To analyze the relationship between concentrations of sodium carbonate and degradation of aniline, an experiment was designed, in which different concentrations of sodium carbonate were added to the aniline solutions. Sodium carbonate solution concentrations were 0 mg/L, 50 mg/L, and 100 mg/L (calculated from the concentration of Na+ ). According to the abovementioned sodium carbonate concentration, the pH values of the solutions were 6.2, 6.9, and 7.2, and the conductivities were 17 µs/cm, 150 µs/cm, and 260 µs/cm. Fig. 7 shows the changes of degradation rate with treatment time for different concentrations of sodium carbonate at 20.96 W discharge power, and 0 mm treatment distance.

3.4

Effect of hydrogen peroxide

To determine the influence of H2 O2 addition on degradation efficiency, different concentrations of H2 O2 (0.1%, 0.2%, 0.4%, and 0.8% volume concentrations) were added to the aniline solutions just before the DBD treatment. Fig. 8 shows the variation of degradation rate with treatment time for different concentrations of H2 O2 at 20.96 W discharge power and 0 mm treatment distance. Degradation of aniline could be inhibited or accelerated in the presence of different dosages of H2 O2 . The degradation rate of aniline was about 100% with addition of the 0.8% volume concentration of H2 O2 at different treatment times. For low H2 O2 concentrations (0.1% and 0.2%), the removal efficiencies of aniline were higher than that without addition of hydrogen peroxide for treatment time less than 4 min, and became lower for treatment time longer than 4 min. Higher aniline degradation rates with 0.8% dosage of H2 O2 may be ascribed to more efficient generation of ·OH according to the following formulae [15,24] : − H2 O2 + ·O− 2 → ·OH + OH + O2

(6)

H2 O2 + hν → 2 · OH

(7)

Fig.7 Variation of the degradation rate as a function of treatment time for different concentrations of sodium carbonate

The degradation of aniline decreased with increasing amounts of sodium carbonate addition before the 6 min DBD treatment. However, the treatment time beyond 6 min changed the influence of sodium carbonate on aniline removal rate. The degradation rate after the 10 min treatment was 93.38% when 50 mg/L sodium carbonate was added to the aniline solution. 0 mg/L and 100 mg/L of sodium carbonate correspond to 78.32% and 88.19%, respectively. The present results are different from the experimental results reported by Huang et al [22] . The reason may be the fact that the additions of sodium carbonate increases the liquid conductivities, leading to lower rates of activated species formation [20] . The reaction may explain why the degradation rate in 6 min was lower in 100 mg/L of sodium carbonate addition than that in 0 mg/L and 50 mg/L. According

Fig.8 Variation of the degradation rate as a function of treatment time for different concentrations of hydrogen peroxide

The increase of active ·OH concentration thus accelerates the degradation rate. When small amounts of H2 O2 (0.1%, 0.2% and 0.4%) were added, ·OH concentration increased first in a short treatment time (less than 4 min), whereas OH may react with H2 O2 when the treatment time went beyond 4 min, which could be expressed by the following formulae [24,25] : ·OH + H2 O2 → H2 O + HO2 · 232

(8)

WU Haixia et al.: Degradation of Aniline Wastewater Using DBD at Atmospheric Pressure HO2 · + · OH → H2 O + O2

(9)

of the toxic organic compounds (RH) in water, and degrades hydrogen from the compounds. R· is generated and oxidized by O2 in water into ROO·, which is cracked into smaller organic molecules in a series of reactions [27,28] . The reactions by which toxic organic molecules are degraded are as follows:

When higher concentrations of ·OH and H2 O2 are both present simultaneously for longer treatment times, these reactions become more significant, which reduces the effective level of both ·OH and H2 O2 in the solution. As a result, the degradation efficiency is decreased [25] . For 0.8% H2 O2 dosage, the aniline concentration decreased to 75.32% after 2 min without DBD treatment. By contrast, there is enough ·OH to treat the organic compound completely, and the two abovementioned reactions have no significant effect on the treatment process. Therefore, the degradation efficiency was almost 100% with 0.8% H2 O2 dosage because both chemical and plasma-induced reactions took effect at the same time.

3.5

H2 O → ·OH + ·H + H2 O2 + H3 O + H2

(10)

O3 → O2 + O

(11)

O + H+ → ·OH

(12)

OH + RH → R · +H2 O

(13)

R · +O2 → ROO·

(14)

TOC is the amount of carbon bound in an organic compound, and is often used as a nonspecific indicator of water quality [15] . Fig. 10 shows changes of TOC as a function of treatment time at 21.5 W discharge power and 0 mm treatment distance. The TOC of the solution treated for 10 min decreased by 15.57% from 81.05 ppm to 68.43 ppm. The reduction in TOC was lower than the reduction of aniline concentration. This result indicates that most of the aniline is oxidized into other organic byproducts instead of carbon dioxide and water.

Byproducts and reaction mechanism

Various oxidizing species, which contribute to the decomposition of aniline, were formed in DBD discharges. Fig. 9 shows the time-averaged emission spectra measured during the DBD treatment (21.5 W, 0 mm). A collection of species was produced in the discharge space, and the concentration of these species is a reflection of the quantity of all the species generated in gas phase. To test the concentrations of the active state species (such as OH+ and N2 O+ ) is difficult because they have very short lifetimes (only about 10−8 s). These activated species in gas are transferred into the solution near the high-voltage electrodes, and dissociate contaminant molecules. In addition to the activated species, electrical breakdown in water produces UV radiation. The production efficiency of ·OH is relatively high when UV light is present [26] .

Fig.10 Variations of TOC with DBD treatment time at 21.5 W discharge power and 0 mm treatment distance

Fig.9 ment

In order to identify the byproducts, GC/MS analysis was carried out. The main products produced after 10 min DBD treatment at 20.4 W discharge power and 0 mm treatment distance were 2-nitrophenol, phenol, benzoquinone, alkanes, and other unidentified trace products, as shown in Fig. 11. According to the abovementioned data and by referring to the work of Brillas et al. [29] , Meguru et al. [30] and Chen et al. [31] , aniline was degraded to other products after DBD treatment, and the possible reaction pathways are shown in Fig. 12.

Emission spectrum measured during the DBD treat-

High-energy electrons and free radicals generated from DBD provide sufficient energy to dissociate contaminant molecules. The interaction among these free radicals initiates chain reactions with each other or with water contaminants and results in aniline decomposition. Hydroxyl radical, a highly reactive intermediate, is responsible for strong oxidizing character of the discharge. It attacks the high density of electron cloud

Fig.12 aniline

233

Proposed reaction pathways for degradation of

Plasma Science and Technology, Vol.17, No.3, Mar. 2015

Fig.11

4

GC/MS chromatograph of aniline intermediates after DBD treatment

Conclusions

9

A DBD reactor was designed and used to degrade aniline in aqueous solutions. Four cylindrical stainless steel rods connected in parallel were used as highvoltage electrodes, and one end of the steel rod was made to be hemispherical in shape to reduce the breakdown voltage. Experimental results show that the aniline degradation using DBD treatment is feasible and effective. The discharge process was analyzed by emission spectra. The species generated were mainly O3 , OH+ , N2 O+ , O, NO, and H2 O+ . These radical species were produced in gas phase, and transferred into the wastewater to react with the aniline. The degradation rates of the aniline solution were influenced by the treatment distance. A maximum degradation ratio was obtained when the high-voltage electrodes just touched the water surface (0 mm treatment distance). TOC removal was lower than aniline removal for incomplete oxidation of organic compound during DBD treatment. The degradation efficiency increased with the increase of the discharge power. A decreasing trend of degradation efficiency was observed in weak acid, alkaline and acidic solutions. Adding a certain concentration of sodium carbonate and hydrogen peroxide in the solutions could accelerate the degradation rate of aniline. On the basis of TOC and GC/MS data, the possible mechanism of aniline degradation by DBD was elucidated.

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

References 1 2 3 4 5 6 7 8

26 27

Abram F S H, Sims I R. 1982, Water Res., 16: 1309 Khan M F, Boor P J, Kaphalia B S, et al. 1995, Fund. Appl. Toxicol., 25: 224 Devulapalli R, Jones F. 1999, J. Hazard. Mater., 70: 157 Brillas E, Casado J. 2002, Chemosphere, 47: 241 Jadhav S R, Verma N, Sharma A, et al. 2001, Sep. Purif. Technol., 24: 541 Pereira R, Pereira L, Madalena Alves M, et al. 2011, Water Res., 45: 191 Kang Y. 2010, Modern Agrochemicals, 9: 15 (in Chinese) Bai Y H, Chen J R, Li X Y, et al. 2009, Rev. Environ. Contam. Toxicol., 201: 117

28 29 30 31

234

Zhang J B, Zheng Z, Zhang Y N, et al. 2008, J. Hazard. Mater., 154: 506 Chen Y S, Zhang X S, Dai Y C, et al. 2004, Sep. Purif. Technol., 34: 5 Bubnov A G, Burova E Y, Grinevich V I, et al. 2006, Plasma Chem. Plasma Proc., 26: 19 Wen Y Z, Jiang X Z, 2001, Plasma Chem. Plasma P., 21: 348 Kim Y K, Kim S A, Lee S B, et al. 2005, Plasma Proc. Polym., 2: 252 Abdelmalek F, Ghezzar M R, Belhadj M, et al. 2006, Ind. Eng. Chem. Res., 45: 23 Manoj Kumar Reddy P Subrahmanyam Ch. 2012, Ind. Eng. Chem. Res., 51: 11097 Young S M, Jin O J, Jin C W. 2008, J. Chem Eng., 142: 56 Biljana P D, Goran M R, Bratislav M O, et al. 2011, J Hazard. Mater., 192: 763 Lu N, Li J, Wang X. 2012, Plasma Chem. Plasma Process., 32: 109 Daniel J W. 1961, Analyst, 86: 640 Grymonpre D R, Sharma A K, Finney W C, et al. 2001, Chem. Eng. J., 82: 189 Roth J A, Sullivan D E. 1981, Ind. Eng. Chem. Fundamen., 20: 137 Huang F M, Li C, Wang H L, et al. 2010, Chem. Eng. J., 162: 250 Acero J L, Konrad S, Gunten UV. 2000, Ozone Sci. Eng., 22: 305 Malik M A, Ghaffar A, Malik S A. 2001, Plasma Sources Sci. Technol., 10: 82 Biljana P D, Goran M R, Bratislav M O, et al. 2011, J. Hazard. Mater., 192: 763 Laroussi M. 2002, IEEE Trans. Plasma Sci., 30: 1409 Nickelsen M G, Cooper W J, Kurucz C N, et al. 1992, Environ. Sci. Technol., 26: 144 Sugiarto A T, Masayuki S. 2001, Thin Solid Films, 386: 295 Brillas E, Mur E, Sauleda R, et al. 1998, Applied Catalysis B: Environ., 16: 31 Meguru T, Masakazu I. 2001, Thin Solid Films, 386: 204 Chen X Y, Xue Z Y, Wu D, et al. 2009, Water Treatment, 35: 16

(Manuscript received 4 March 2014) (Manuscript accepted 21 August 2014) E-mail address of corresponding author FANG Zhi: [email protected]