Plasma Chem Plasma Process (2012) 32:141–152 DOI 10.1007/s11090-011-9332-1 ORIGINAL PAPER

Oxidization of SO2 by Reactive Oxygen Species for Flue Gas Desulfurization and H2SO4 Production Mindi Bai • Jie Hu

Received: 17 August 2011 / Accepted: 8 November 2011 / Published online: 22 November 2011 Ó Springer Science+Business Media, LLC 2011

Abstract A strong ionization dielectric barrier discharge was used to produce a high concentration of reactive oxygen species that were then injected into a simulated flue gas in a duct to remove SO2 by oxidation. Sulfuric acid (H2SO4) was produced through the following two reactions: (1) O3 oxidation of SO2–SO3, which then reacted with H2O to produce H2SO4; and (2) reaction of O2? with H2O to produce OH radicals, which then rapidly and non-selectively oxidized SO2–H2SO4. When the molar ratio of reactive oxygen species to SO2 was 4:1, the SO2 removal efficiency was 94.6%, the energy consumption per cubic meter of flue gas was 13.3 Wh/m3, the concentration of recovered H2SO4 was 4.53 g/l, and the H2SO4 recovery efficiency was 28.8%. The H2O volume fraction in the simulated flue gas affected the SO2 removal efficiency, whereas the O2 and CO2 volume fractions did not. These results prove that oxidation by reactive oxygen species is a feasible method for flue gas desulfurization. Keywords Reactive oxygen species  OH radicals  Removal SO2  SO2 removal efficiency  H2SO4 Introduction Wet limestone-gypsum flue gas desulfurization (FGD) is widely used to treat flue gas in coal-fired power plants in the United States, Germany, Japan and other developed countries [1, 2]. However, this technology is too expensive for developing countries, requires a large area, and produces large volumes of wastewater and low quality by-products [3, 4]. As an alternative to this treatment method, many studies have demonstrated that nonthermal plasma can be used for efficient SO2 removal from polluted flue gas [5–12]. Saavedra [12] highlighted that nonthermal plasma can be used for simultaneous removal of several pollutants at atmospheric pressure. Furthermore, the initial investment and operation costs for nonthermal plasma are relatively low. M. Bai (&)  J. Hu Dalian Maritime University Environmental Engineering Institute, Dalian 116026, China e-mail: [email protected]

123

142

Plasma Chem Plasma Process (2012) 32:141–152

Nonthermal plasmas for flue gas desulfurization are usually created by electron beam irradiation or pulsed corona discharge [13]. FGD using electron beam irradiation has been demonstrated at 200 MW power plants in Chengdu, China [14]. Techniques have been developed for recovery of ammonium salt that is generated as a byproduct during this process, when NH3 absorbs SO2 from the flue gas. However, there are other problems with this technique, such as the large size of the plasma source and vacuum system, and X-ray radiation hazards. Bernie and Penetrante [15] suggested that the high capital cost and X-ray hazards of the electron beam method have discouraged its use in many pollution control applications. To resolve the presented disadvantages in electron beam irradiation, many scholars have studied the treatment of flue gas using the pulse corona discharge. Encouraging results have been obtained for the FGD, but the energy efficiency obtained by electron beam irradiation is twice that obtained by the pulsed corona discharge [13]. In the pulsed corona discharge zone, a large number of electrons lose considerable energy through vibrational excitation, which does little to enhance the desired reaction of FGD. High desulfurization efficiency is with injected ammonia under the pulsed corona discharge. Sadakat [16] pointed out that the pulse corona method is a thermal chemical reaction, and that the main products [(NH4)2SO3] are likely to decomposed into SO2 and NH3 again above 30°C. To solve this problem, in this paper a strong ionization dielectric barrier discharge (DBD) method was used to ionize and dissociate O2 into a high concentration of reactive oxygen species, which were then injected into a duct containing a simulated flue gas to produce OH radicals. These radicals rapidly and non-selectively oxidized SO2 to produce sulfuric acid (H2SO4). The whole plasma chemistry reaction takes place in a duct, without the need for other conventional treatment methods. The basic principles and the plasma chemistry reactions for SO2 removal, the plasma source, and the effect of H2O, O2, and CO2 volume fractions on SO2 removal efficiency are discussed.

Experimental Reactive Oxygen Species Generation and Plasma Chemistry Reactions for SO2 Removal The basic principles for flue gas desulfurization by reactive oxygen species injected into the duct are shown in Fig. 1. High concentrations of reactive oxygen species, such as O2?, O, O(1D), O2-, O2(a1Dg), and O3 [12], are produced by a strong ionization dielectric barrier discharge in a reactive oxygen species generator (1 in Fig. 1). However, among these species, only O2?and O3 are injected into the duct (3 in Fig. 1) to react with SO2 and produce H2SO4. This is because O, O(1D) O2(a1Dg) are short-lived. The main reaction is of O2? with H2O to produce OH radicals, which then rapidly and non-selectively oxidize SO2 to H2SO4 mist. This occurs because the rate coefficient and oxidization potential of OH radicals are about 10-12 cm3/s and 2.80 V, respectively. The secondary reaction involves O3 oxidation of SO2–SO3, and subsequent reaction of SO3 with H2O to produce H2SO4. The H2SO4 mist passes through a high-voltage direct current electric field and is captured and collected as liquid H2SO4 [17] in an electric acid mist remover (4 in Fig. 1). The purified experimental gas is discharged from the duct by an induced-draft fan (6 in Fig. 1). The rectangular (L 9 W 9 H, 280 9 50 9 220 mm) reactive oxygen species generator (Fig. 2a, b) contains discharge electrodes, ground electrodes, spacers, and dielectric

123

Plasma Chem Plasma Process (2012) 32:141–152

143

2

Purified flue gas

O2 1

Simulated flue gas

O2+

5

O3 4 6

3

H2SO4 Fig. 1 Generation of reactive oxygen species for removal of SO2 from simulated flue gas in a duct with the following components: 1 reactive oxygen species generator, 2 high-voltage high-frequency power supply, 3 duct, 4 electric acid mist remover, 5 high-voltage DC power supply, 6 induced-draft fan

layers. A high-voltage, high-frequency discharge output from the high-voltage, high-frequency power supply (2 in Fig. 1) is applied to the discharge electrodes, and the peak voltage (6 kV), current (100 mA) and waveform (11.5 kHz) are measured by a oscilloscope (TDS3032, Tektronix, Beaverton, OR) (Fig. 3). The field strength E, which was calculated as the peak voltage (shown in Fig. 3) divided by the electrode gap spacing (0.2 mm), was divided by the total gas density n to acquire the reduced electric field strength E/n. This is discussed in detail in elsewhere [18]. The mean electron energy depended primarily on the reduced field strength, E/n, which is discussed in detail elsewhere [14]. When the electric field strength reached [97.2 kV/cm in the discharge channels, the mean electron energy increased to [8 eV, which is around 6 eV higher than the mean electron energy for the pulsed corona discharge and about 3–4 eV higher than that for the conventional dielectric barrier discharge. An image of the strong ionization discharge formed in the discharge channels is shown in Fig. 2c. The energy of electrons ranges from ionization energy (8.4 eV) to dissociation energy (12.5 eV). These high energy electrons are deposited on O2 to ionize or dissociate it into a high concentration of reactive oxygen species, such as O2?, O, O(1D), O-, O2-, O2(a1Dg), and O3, in the discharge channel. The remaining low energy electrons do not contribute to the desired reaction Because the reactive oxygen species O, O-, O(1D), O2-, and O2(a1Dg) are shortlived (approximately 1 9 10-8 s), only O2-, O2? and O3 are injected into the duct (3) for oxidation of SO2. The plasma chemistry reactions and their rate coefficients are presented in Table 1. The rate coefficients of 1–6 are taken into account with values of E/n = 150 Td in the discharge channel. Reaction

Rate coefficient

O- ? O2(a1Dg) ? O3 ? e

Reference

k7 = 3.0 9 10-10 cm3/s

(7)

[19]

O( D) ? O2 ? O ? O2(a Dg)

k8 = 3.4 9 10-11 cm3/s

(8)

[19]

O(1D) ? O2 ? O ? O2

k9 = 6.3 9 10-12 cm3/s

(9)

[19]

k10 = 1.0 9 10-14 cm3/s

(10)

[19]

1

O ? O2 ? O3

1

When O2?, O2- and O3 are injected into the duct, the electron energy of O2- produced through electron deposition on O2 is so low that it cannot oxidize SO2. Only O2? and O3 can oxidize SO2 to produce H2SO4 through the following two major reactions. The first of these is reaction of O2? with H2O to produce OH radicals, and subsequent oxidation of

123

144

Plasma Chem Plasma Process (2012) 32:141–152

Enclosur

High-frequency high-voltage Power

(b) Ground electrode Cooling water

O2 HO2 inlet Spacer Dielectric layer

Discharge electrode

The exposure time: 1/120s

(a)

(c)

Fig. 2 Structure of the reactive oxygen species generator and discharge image in the discharge channel. a Structure of reactive oxygen species. b Photo of the reactive oxygen species generator. c Photo of discharge image

Fig. 3 Voltage, current and waveform output from highvoltage high-frequency power supply

I2

V2

V:1kV/div, I:50mA/div, t:25µs/div

Table 1 Plasma chemistry reactions involved in reactive oxygen species generation Electron impact reaction

Rate coefficient 150 Td

O2 ? e ? O2? ? 2e

k1 = 8.0 9 10-10 cm3/s -10

3

Reference (1)

[19]

O2 ? e ? O ? O ? e

k2 = 7.9 9 10

cm /s

(2)

[19]

O2 ? e ? O- ? O

k3 = 3.2 9 10-11 cm3/s

(3)

[19]

O2 ? e ? O2-

k4 = 5.6 9 10-12 cm3/s

(4)

[19]

1

-9

3

O2 ? e ? O ? O( D) ? e

k5 = 2.3 9 10

O2 ? e ? O2(a1Dg) ? e

k6 = 7.2 9 10-10 cm3/s

cm /s

(5)

[19]

(6)

[19]

SO2–H2SO4 mist. The second reaction involves O3 oxidation of SO2–SO3, and subsequent reaction of SO3 with H2O to produce H2SO4. The plasma reactions proceed to completion in the duct (Fig. 4, length 1 m) between where the reactive oxygen species are injected and the entrance of the electric acid mist remover. The plasma reactions are described below.

123

Plasma Chem Plasma Process (2012) 32:141–152

145

Entrance of electric

Hole of injecting the reactive oxygen species

acid mist remover

1m (The entire process from removal SO 2 to production of H2SO4 mist) Fig. 4 Photograph of the plasma reaction duct

First O2? reacts with H2O to form water cluster ions, O2? H2O, and these then dissociate to form OH radicals. O2? ? H2O ? M ? O2? H2O ? M ?

?

O2 H2O ? H2O ? H3O ? OH ? O2

k11 = 2.5 9 10-28 cm6/s

(11)

[20]

k12 = 1.2 9 10-9 cm3/s

(12)

[20]

Because the rate coefficient and oxidization potential of OH radicals are about 10-12 cm3/s and 2.80 V, respectively, the OH radicals can rapidly and non-selectively oxidize SO2–H2SO4 mist. SO2 ? OH ? HSO3

k13 = 7.5 9 10-12 cm3/s

(13)

[19]

HSO3 ? OH ? H2SO4

k14 = 1.0 9 10-12 cm3/s

(14)

[19]

H2SO4 is also produced by O3 oxidization of SO2–SO3, which then reacts with H2O to produce H2SO4 mist. SO2 ? O3 ? SO3 ? O2

k15 = 3.0 9 10-12 cm3/s

(15)

[21]

SO3 ? H2O ? H2SO4

k16 = 6.0 9 10-15 cm3/s

(16)

[19]

This method for the SO2 removal is very different to conventional gas ionization discharge, in which the total flue gas passes through the plasma source [5]. Compared with conventional gas ionization discharge, the present method requires a smaller volume of plasma, has a simplified procedure, lower investment cost and energy consumption, and does not require additional catalysts, reductants, oxidants or the use of other technologies. This method also produces liquid H2SO4, which is an important raw material in the chemistry industry. System for Simulated Flue Gas Desulfurization The experimental setup for simulated flue gas desulfurization is shown in Fig. 5. The experimental gas contained SO2, O2, CO2, and N2 standard gases, which were from

123

146

Plasma Chem Plasma Process (2012) 32:141–152

O2

O2

N2

1

2

5

7

RH,T

Purified experimental gas

3 8

9

Air

L

10 H2O SO2

CO2

4 6 11

11 12

13

H2SO4 liquid Fig. 5 Experimental setup for simulated flue gas desulfurization. 1 Reactive oxygen species generator; 2 high-voltage high-frequency power supply; 3 duct; 4 electric acid mist remover; 5 high-voltage DC power supply; 6 induced-draft fan; 7 reactive oxygen species detector; 8 gas-mixing chamber; 9 temperature and humidity regulator; 10 mass flow meter; 11 flue gas analyzer; 12 ion chromatograph; 13 acid storage tank

compressed gas cylinders, and dry air. The flow rate was adjusted using mass flow meters. The experiment gases were mixed in the gas-mixing chamber (8 in Fig. 5) before adjusting the temperature and humidity (9 in Fig. 5) so that they were similar to those in flue gas from a coal-fired power plant. The simulated gas was introduced into the duct (3 in Fig. 5), which was a stainless tube with an inner diameter of 20 mm. The reactive oxygen species (O2?,and O3) with concentrations of 200–300 mg/l produced by DBD in the reactive oxygen species generator (1 in Fig. 5) were injected into the center of the duct. These species were transferred through a PTFE tube with an inner diameter 6 mm. In the duct, O2? and O3 oxidized SO2 to H2SO4 mist. This mist passed through the high-voltage direct current electric field and was captured and collected as liquid H2SO4 in an electric acid mist remover (4 in Fig. 5). The electric acid mist remover was a grounded, titanium steel cylinder with the innner diameter of 200 mm, thickness of 1.5 mm and length of 1,200 mm that contained a star-shaped electrode wire with the average diameter of 3 mm. The electric field strength in the middle of the acid mist remover was 14 kV/cm. Residual reactive oxygen species in the experimental gas were removed by heat treatment, and the final purified experimental gas was discharged from the duct using an induced-draft fan (6 in Fig. 5). A flue gas analyzer (11 in Fig. 5) (Photon ? PGD-100, Madur, Vienna, Austria) was used for online monitoring of the gas (SO2, N2, O2, and CO2) volume fractions, temperature, and pressure. An ion chromatograph (12 in Fig. 5) (ICS-1500, Dionex, Sunnyvale, CA) was used for quantitative analyses of the composition of the recovered acid. The reactive oxygen species detector (7 in Fig. 5) included an O2? density tester and detector (BMT964, BMT MESSTECHNIK GMBH, Stahnsdorf, Germany). The BMT964 detector was used to measure the O3 concentration using a microprocessor-based dual beam photometer (UV 254 nm). We developed a ball (ø 6 mm) ion detector for detection of the O2? density in the high velocity duct. The density was measured at output ranges from 104 to 1010 cm-3 and air velocity ranges from 0.2 to 50 m/s, as discussed in detail by Bai and Yang [22]. The electron charges collected using a ball probe were transferred into a single microcurrent, and the O2?density could be calculated using the following

123

Plasma Chem Plasma Process (2012) 32:141–152

ni ¼

147

I eVo pd2

where ni is the O2? density (cm-3), I is the micro current, Vo is the gas flow, e is the electron charge, d is the diameter of ball probe.

Results and Discussion Influence of Experimental Parameters on the SO2 Removal Efficiency The influences of the molar ratio of reactive oxygen species (O3, O2?) to SO2, and H2O, O2, and CO2 volume fractions on the SO2 removal efficiency were investigated. The SO2 removal efficiency is defined as follows: gSO2 ¼ ðCSO2 :in  CSO2 :out Þ= CSO2:in  100% where gSO2 is the SO2 removal efficiency, CSO2.in is the input concentration of SO2 (ppm), and CSO2.out is the output concentration of SO2 (ppm). The input and output concentrations were measured using the flue gas analyzer (11 in Fig. 5). As an important index of flue gas desulfurization technology, the molar ratio of reactive oxygen species (O3, O2?) to SO2 is usually used to determine the addition of reactive oxygen species. The results for the influence of the molar ratio of reactive oxygen species (O3, O2?) to SO2 are shown in Fig. 6. When n increased from 0 to 1, the SO2 concentration decreased from 800 ppm to 247 ppm, and the SO2 removal efficiency was 69.1%. When n was 2, the SO2 concentration decreased to 116 ppm, and the SO2 removal efficiency was 85.5%. The highest SO2 removal efficiency (94.6%) was achieved with n = 4, and the SO2 concentration decreased to 43 ppm. The energy consumption for SO2 removal per cubic meter of flue gas was 13.3 Wh/m3. This value was calculated by dividing the 16 W output from the high-voltage high-frequency power supply, which produced reactive oxygen species at n = 4, by the total flue gas flow rate of 20 l min-1, which did not include O2 input to the reactive oxygen species generator (1 in Fig. 5). These results show the SO2 removal efficiency depends on the molar ratio of reactive oxygen species (O3, O2?) to SO2, and the optimum molar ratio is n = 3–4. 100

SO2 removal efficiency / %

Fig. 6 Effect of the molar ratio n between the reactive oxygen species and SO2 on SO2 removal efficiency Conditions: 800 ppm SO2, 12% O2, 10% H2O, 10% CO2 in N2 at 60°C, flow rate = 20 l min-1

80

60

40

20

0 0

1

2

3

4

Molar ratio n

123

Fig. 7 Effect of H2O concentration on SO2 removal efficiency. Conditions: n = 4, 800 ppm SO2, 12% O2, 10% CO2 in N2 at 80°C, flow rate = 20 l min-1

Plasma Chem Plasma Process (2012) 32:141–152 80

SO2 removal efficiency / %

148

60

40

20

0

2

4

6

8

10

H2O concentration (v/v)

The results for the influence of the H2O volume fraction are presented in Fig. 7. When the H2O volume fraction increased from 0 to 3.4%, the SO2 removal efficiency increased by 47.2% (from 20 to 67.2%). A further 8.6% increase in the SO2 removal efficiency (to 75.8%) was observed when the H2O volume fraction was increased to 6.8%. The highest SO2 removal efficiency (77.8%) was achieved with a H2O volume of 9.3%, and this removal efficiency was 2% higher than that with 6.8% H2O. These results show that the H2O volume fraction affects the SO2 removal process. This is largely because the production of OH radicals relies on the reaction of H2O with O2?, and the H2O volume fraction is important in this reaction. The maximum SO2 removal efficiency was achieved with a H2O volume fraction of [9%. The typical H2O content in flue gas from a coal-fired power plant is [9%, and will be sufficient for flue gas desulfurization by reactive oxygen species without the addition of water vapor. The results for the influence of the O2 volume fraction are presented in Fig. 8. At 60°C, when the O2 volume fraction was increased from 4 to 17%, the SO2 removal efficiency remained at around 94%. By comparison, at 80°C, when the O2 volume fraction was increased from 4 to 17%, the SO2 removal efficiency remained at around 82%. These results show that the O2 volume fraction has almost no influence on the SO2 removal 100

SO2 removal efficiency / %

Fig. 8 Effect of O2 concentration on SO2 removal efficiency. Conditions: n = 4, 800 ppm SO2, 10% H2O, 12% CO2 in N2 at 60°C or 80°C, flow rate = 20 l min-1

80

60

40

60 80

20 4

6

8

10

12

14

O2 concentration (v/v)

123

16

Plasma Chem Plasma Process (2012) 32:141–152 100

SO2 ,NOremoval efficiency / %

Fig. 9 Effect of CO2 concentration on SO2, NO removal efficiency. Conditions: 800 ppm NO, 800 ppm SO2, 10% H2O, 12% O2 in N2 at 80°C or 45°C, flow rate = 20 l min-1

149

80

60 NO: 60

, n=2

SO2: 60

, n=4

SO2: 80

, n=4

40 3

6

9

12

CO2 concentration (v/v)

efficiency. Because the reactive oxygen species (O3 and O2?) are injected from outside the duct, the O2 content in the mixed experimental gas does not affect the amount of reactive oxygen species and OH radicals produced. Therefore, there is no need to regulate the O2 volume fraction in flue gas desulfurization by reactive oxygen species. The results for the influence of the CO2 volume fraction on SO2 removal efficiency are presented in Fig. 9. When the CO2 volume fraction was increased from 3 to 12% at n = 4, the SO2 removal efficiency remained at around 94% in the 60°C experiment and 82% in the 80°C experiment. These results show that the CO2 volume fraction has little influence on the SO2 removal efficiency. This is because CO2 is a stable gas, and these results indicate that the volume fraction of CO2 does not need to be controlled in flue gas desulfurization by reactive oxygen species. The results for the influence of the CO2 volume fraction on NO removal efficiency are presented in Fig. 9. When the CO2 volume fraction was increased from 3 to 12% at n = 2, the NO removal efficiency remained at around 97% in the 60°C experiment. The energy consumption of NO removal per cubic meter of flue gas is 8.3 Wh/m3. Two major conclusions can be drawn from these results. The first of these is that the CO2 volume fraction has little influence on the NO removal efficiency because CO2 is a stable gas. The second is that the reactive oxygen species remove NOx with a lower energy consumption (8.3 Wh/m3) than that for SO2 removal (13.3 Wh/m3), but with almost the same removal efficiency. In future studies we will investigate the basic principles for flue gas denitrification by the reactive oxygen species injected into the duct. Figures 8 and 9 show that the input flue gas temperature affects the efficiency of desulfurization. High gas temperatures are unfavourable for SO2 removal because they increase the kinetic energy of air molecules, leading to increasingly thermal motion and the loss of reactive oxygen species. Low desulfurization temperatures may cause problems such as equipment corrosion, the need for a large heat exchanger, and the difficulties with flue gas discharge into atmosphere. The optimum temperature range is 60–80°C. Analysis of the Recovered Acid The acid was collected in an acid storage tank (13 in Fig. 5). Constant volume samples were removed from the tank, diluted by a factor of 1,000, and then injected into the

123

150

Plasma Chem Plasma Process (2012) 32:141–152

Fig. 10 Chromatogram for SO42-. Conditions: n = 4, 800 ppm SO2, 12% O2, 10% H2O, 10% CO2 in N2 at 60°C, flow rate = 20 l min-1. a Standard chromatogram for SO42-. b Chromatogram for SO42- in the recovered acid liquid

–

SO42 10mg/L

0

5

10

15

20

25

30

20

25

30

t / min

(a)

–

SO42 4.53mg/L

0

5

10

15

t / min

(b)

ion chromatograph (12 in Fig. 5). The type of acid can be determined based on the retention time of the peak in the chromatogram, and acid concentration can be calculated from the peak area using an external standard method. Ion chromatography was performed for the dilute acid sample from the acid storage tank (Fig. 10b) and the diluted standard sample ([SO42-] = 10 mg/l) for comparison (Fig. 10a). The SO42concentration calculated for this sample by dividing the peak area in Fig. 10b by that in Fig. 10a was 4.53 mg/l.

Influence of the Molar Ratio of the Reactive Oxygen Species to SO2 on the SO42Recovery Efficiency The influence of n on the SO42- recovery efficiency are presented in Fig. 11. The SO42recovery efficiency was calculated as follows: g = experimental mass of SO42- recovered per mole of SO2/theoretical mass of SO42- recovered per mole of SO2. When n increased from 1.1 to 3, the SO42- recovery efficiency increased in an approximately linear manner from 4.4 to 26.8%, which is an increase of 11.8% for every unit increase in the molar ratio. When n was 3.7, the SO42- recovery efficiency was 28.8%, and only 2% more than that for n = 3, which is an increase of 2.85% for every unit increase in the molar ratio. This indicates the optimal n for the SO42- recovery efficiency is between 3 and 4. These results agree with the results for the influence of the molar ratio on the SO2 removal efficiency.

123

Plasma Chem Plasma Process (2012) 32:141–152 30

SO42- recovery efficiency / %

Fig. 11 Effect of molar ratio n between the reactive oxygen species and SO2 on SO42recovery efficiency. Conditions: n = 4, 800 ppm SO2, 12% O2, 10% H2O, 10% CO2 in N2 at 60°C, flow rate = 20 l min-1

151

20

10

0 0

1

2

3

4

Molar ratio n

Conclusion 1 A strong ionization discharge method was used to dissociate and ionize O2 into reactive oxygen species, which were then reacted with H2O to generate OH radicals. This is a new method for production of a large number of OH radicals. 2 The influence of the molar ratio (n) of reactive oxygen species to SO2 on the SO2 removal efficiency was determined, and the optimal value of n was 3–4. Doubling the volume fraction of CO2 or O2 had little influence on the SO2 removal efficiency, whereas a H2O volume fraction of[9% was required for SO2 removal. With n = 4, the SO2 removal efficiency was 94.6%, and the energy consumption per cubic meter was 13.3 Wh/m3. The concentration of H2SO4 in the recovered liquid was 4.53 g/l. 3 The experimental results show that injection of reactive oxygen species into flue gas ducting is a feasible method for SO2 removal and liquid H2SO4 production. The plasma chemistry reaction can proceed to completion within the duct (length 1 m) without the need for other conventional treatment methods. This is a new green method for flue gas desulfurization. Acknowledgments This work was supported by the National High-Tech Research & Development Plan of China (Grant No. 2008AA06Z317) and the National Natural Science Foundation of China (Grant No. 50778028).

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

Ortiz FJG, Vidal F, Ollero P et al (2006) Ind Eng Chem Res 45:1466–1477 Zhong Y, Gao X, Huo W (2008) Fuel Process Technol 89:1025–1032 He ZQ, Tian YB, Shen JJ et al (2008) Power Syst Eng 4 Jia SH (2005) J Electr Power 1 Zhang ZT, Bai MD, Bai XY et al (2006) J Air Waste Manag Assoc 56:810–815 Callen MS, Cruz MT, Marinov S (2007) Fuel Process Technol 88:251–258 Ma HB, Chen P, Zhang ML et al (2002) Plasma Chem Plasma Process 22:239–254 Zhu Y, Chae JO, Kim KY et al (2002) Plasma Chem Plasma Process 22:187–195 Wu Y, Jie L, Wang NH et al (2003) J Electrostat 57:233–241 Lazaroiu G, Zissulescu E, Sanduc M (2007) Energy 32:2412–2419

123

152

Plasma Chem Plasma Process (2012) 32:141–152

11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

Bai MD, Zhang ZT, Yi CW et al (2006) Plasma Chem Plasma Process 26:177–186 Saavedra HM, Pacheco MP, Pacheco-Sotelo JO (2007) IEEE Trans Plasma Sci 35:1533–1539 Obradovic BM, Sretenovic GB, Kuraica MM (2011) J Hazard Mater 185:1280–1286 Hao JM, Wang SX, Lu YQ (2005) Chemical Industry Publishing House, Beijing Penetrante BM, Bardsley JN, Hsiao MC (1997) Jpn J Appl Phys 36:5007–5017 Sadakat M (1995) Jpn Proc Inst Electrostat 19:252–281 Bai MD, Wang SL, Chen ZG et al (2010) IEEE Trans Plasma Sci 38:127–132 Eliasson B, Kogelschatz U (1986) Jpn Phys B 19:1241 Jowke JJ, Morrow R (1995) IEEE Trans Plasma Sci 23:661–671 Person J C, Ham DO (1988) Radiat Phys Chem 31:1–8 Demore WB, Sander SP, Golden DM et al (1997) Evaluation number 12, vol 97. JPL Publication, p 4 Bai MD, Yang B, Xue XH et al (2009) IEEE Trans Plasma Sci 37:2257–2260

123