JO U R N A L OF EN V I RO N M EN T A L S CI E NC ES 2 6 ( 20 1 4 ) 2 2 4 9–2 2 5 6
Available online at www.sciencedirect.com
ScienceDirect www.journals.elsevier.com/journal-of-environmental-sciences
Experimental study of NO2 reduction in N2/Ar and O2/Ar mixtures by pulsed corona discharge Xinbo Zhu1 , Chenghang Zheng1 , Xiang Gao1,⁎, Xu Shen1,2 , Zhihua Wang1 , Zhongyang Luo1 , Kefa Cen1 1. State Key Laboratory of Clean Energy Utilization, Zhejiang University, Hangzhou 310027, China. E-mail: [email protected] 2. Zhejiang Electric Power Design Institute, Hangzhou 310012, China
AR TIC LE I N FO
ABS TR ACT
Article history:
Non-thermal plasma technology has been regarded as a promising alternative technology
Received 20 December 2013
for NOx removal. The understanding of NO2 reduction characteristics is extremely
Revised 16 February 2014
important since NO2 reduction could lower the total NO oxidation rate in the plasma
Accepted 6 March 2014
atmosphere. In this study, NO2 reduction was experimentally investigated using a
Available online 29 September 2014
non-thermal plasma reactor driven by a pulsed power supply for different simulated gas
Keywords:
the specific energy density (SED), and the highest conversion rates were 33.7%, 42.1% and
Non-thermal plasma
25.7% for Ar, N2/Ar and O2/Ar, respectively. For a given SED, the NO2 conversion rate had the
Pulsed corona discharge
order N2/Ar > Ar > O2/Ar. The highest energy yield of 3.31 g/kWh was obtained in N2/Ar
NO
plasma and decreased with increasing SED; the same trends were also found in the other
compositions and operating parameters. The NO2 reduction was promoted by increasing
NO2
two gas compositions. The conversion rate decreased with increasing initial NO2 concentration. Furthermore, the presence of N2 or O2 led to different reaction pathways for NO2 conversion due to the formation of different dominating reactive radicals. © 2014 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.
Introduction Nitrogen oxides (NOx) constitute one of the most hazardous air pollutants from coal-fired power plants and gasoline engine exhaust. Regarded as a precursor of ozone, acid rain and photo-chemical smog, NOx emission is becoming a crucial issue throughout the world, and NOx treatment is an urgent task due to increasingly stringent environmental regulations. Up to now, many technologies have been developed for NOx treatment such as selective catalytic reduction (SCR), selective non-catalytic reduction (SNCR), activated carbon adsorption and wet scrubbing (Skalska et al., 2010). In the past two decades, non-thermal plasma has attracted considerable attention due to its capability for simultaneous removal of multiple pollutants
under ambient conditions with relative low energy cost (Van Durme et al., 2008; Yu et al., 2010a,b). In non-thermal plasma, energy deposited to the reactor generates highly energetic electrons. These electrons collide with background gas molecules, resulting in the formation of reactive species like ions, radicals and excited molecules, which are able to initiate gas phase reactions and convert pollutants to final harmless products (Gallon et al., 2011; Tu and Whitehead, 2012). NO conversion processes in plasma have been experimentally investigated by many researchers in terms of various operating parameters including reactor geometry (Mizuno et al., 1995; Wu et al., 2005), power supply (Matsumoto et al., 2010; Mok et al., 1998), polarity (Masuda and Nakao, 1990), gas composition (Arai et al., 2004; Sun and Yin, 2009), temperature (Eichwald et al., 2002) and additives (Cha et al., 2007; Lin et al., 2010). Kinetic models
⁎ Corresponding author. E-mail: [email protected] (Xiang Gao).
http://dx.doi.org/10.1016/j.jes.2014.09.010 1001-0742/© 2014 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.
2250
J O U R NA L OF EN V I RO N M EN T A L S CI EN C ES 2 6 ( 20 1 4 ) 2 2 49–2 2 5 6
were also introduced to understand the mechanism of plasmainduced NOx conversion (Mok and Ham, 1997; Orlandini and Riedel, 2000). More recently, a few researchers reported high conversion rates using non-thermal plasma-based technologies for NOx treatment in industrial applications (Lee et al., 2003; Yoshida et al., 2009b). It is generally recognized that in industrial application of a plasma-combined deNOx process, plasma is employed as a gas phase oxidizer for NO to NO2 conversion. NO2 can be easily converted to nitric acid or adsorbed by wet scrubbing (Obradovic et al., 2011). However, some researchers found that the oxidation of NO to NO2 in the plasma environment was not straightforward (Penetrante et al., 1997). More NO molecules would be oxidized to NO2 by plasma at high energy density, however when NO2 concentration became higher, these NO2 molecules could react with N or O radicals via reverse reactions: NO2 + O → NO + O2 or NO2 + N → 2NO, producing NO molecules and suppressing the total NO conversion efficiency. More importantly, it is difficult to deal with NO in the aforementioned wet flue gas desulfurization (WFGD) process. As mentioned above, the suppression of NO2 reduction to NO should be investigated to prevent the decrease of NOx conversion rate and optimize the working conditions of plasma. In this study, a reactor driven by a pulsed corona power supply was employed to investigate the conversion characteristics of NO2 to NO in various gas compositions. The discharge characteristics under different simulated gas compositions were evaluated. The influences of energy density and initial concentration on NO2 conversion were also studied. Finally, the effects of N2 and O2 content on NO2 conversion were investigated and the reaction pathways were compared for both N2 and O2 addition cases. In order to facilitate the understanding of plasma-induced chemical reactions, especially electron–molecule impact reactions, argon was used as a background gas for all tested conditions.
Furthermore, as an inert gas, no argon-containing products were formed in the reactions.
1. Materials and methods 1.1. Experimental setup Fig. 1 shows a schematic diagram of a pulsed corona discharge system designed and built for this study. The system consists of four parts: a pulsed corona discharge reactor, a pulsed power supply, a gas control system and gas analysis units. A detailed drawing of the pulsed corona discharge reactor is illustrated in Fig. 2. The reactor was designed in a wire-to-plate configuration. The reactor body was a quartz glass box with the geometry of 600 mm length, 90 mm width and 60 mm height. Two stainless steel plates were placed at the inner sides of the reactor as the grounded electrodes. Eight stainless steel discharging wires with an interval of 50 mm were mounted on the center line of the reactor top side as the high voltage electrodes. The diameter of the high voltage electrodes was 2 mm with an effective discharge length of 40 mm. The distance between the high voltage electrodes and the grounded electrodes was 40 mm. The reactor was powered by a pulsed power supply. A model PPCP-1-300-400 pulsed power supply (Zhongwuhaitong, Sichuan, China) was designed to generate high voltage pulses with a rise time of 100–200 nsec, a duration of 600–800 nsec and an adjustable repetition pulse frequency of 10–300 Hz. The maximum amplitude of the pulse voltage was 60 kV and the maximum output power was 500 W. The pulsed corona discharge voltage was measured at the discharging cable by a 1000:1 P6015A voltage probe (Tektronix Inc., USA), while the current waveform was measured at the grounded electrode by a Model 6585 current probe (Pearson Electronics Inc., USA). The voltage and current waveforms
Oscilloscope Gas analyzer
Pulse power supply
Ozone analyzer
Current prober
Mixing chamber
Plasma reactor
HV probe
Grounded N2
O2
Fig. 1 – Schematic of the experimental set-up.
NO
NO2
Ar
CO2
2251
JO U R N A L OF EN V I RO N M EN T A L S CI E NC ES 2 6 ( 20 1 4 ) 2 2 4 9–2 2 5 6
Discharging wires
Inlet
Outlet
Grounded electrodes
Fig. 2 – Structure of the pulsed corona discharge reactor.
where, CNO2-intlet and CNO2-outlet are the concentrations of NO2 at the inlet and outlet of the reactor, respectively. The concentration of ozone was monitored by a Model 106M UV-absorption ozone analyzer (2B Technologies, USA) at the reactor outlet.
were monitored and recorded with a DPO4034B oscilloscope (Tektronix, Inc., USA). The discharge power was calculated by multiplying voltage and current waveforms, while the total energy consumption was obtained by integrating discharge power over the discharge periods. In this paper, all the results are presented in terms of specific energy density (SED, J/L), which is calculated as follows:
2. Results and discussion
ð1Þ
2.1. Discharge characteristics
where, U (kV) and I (mA) are the voltage and current respectively, while Q denotes the gas flow rate (L/min).
As a key parameter of non-thermal plasma-based air pollution control technology, energy density is affected by both discharge voltage and pulse frequency. Fig. 3a shows the typical voltage and current waveforms at the peak voltage of 38 kV. As mentioned in Section 1, the plasma power was calculated by multiplying voltage and current waveforms while the energy consumption in one pulse period was obtained by integrating the plasma power over one pulse period. As shown in Fig. 3b, for a single pulse period, all energy is deposited to the reactor in the first 700 nsec and the total deposited energy is about 34.65 mJ. In this study, the energy density was adjusted by changing the pulse frequency, while the discharge peak voltage was fixed at 38 kV. Fig. 4 shows the relationship between SED and pulse frequency for the aforementioned three gas compositions. As the pulse frequency increases from 10 to 300 Hz, the SED increases monotonically from 1.16 to 38.03 J/L, 2.39 to 49.03 J/L, and 2.46 to 82.56 J/L, for Ar, N2 (2%)/Ar and O2 (2%)/Ar, respectively. For a certain pulse frequency, the SED values were generally in the order: O2 (2%)/Ar > N2 (2%)/ Ar > Ar. In a plasma environment, electrons are generated via electron-induced ionization reactions, i.e., e + Ar → 2e + Ar+, e + N2 → 2e + N+2 and e + O2 → 2e + O+2 for Ar, N2 and O2,
1.2. Reagents and methods The experiments were carried out at room temperature and atmospheric pressure using the gas mixtures of NO2/Ar, NO2/N2/Ar and NO2/O2/Ar. All the feed gases used in this study were of high-purity grade (> 99.99%). Gas streams from compressed gas cylinders were regulated by a series of calibrated DB-07B mass controllers (Sevenstars, Beijing, China). The total gas flow rate was fixed at 5 L/min and the initial NOx (NO + NO2) concentration was 210 ppm unless otherwise mentioned.
1.3. Chemical analysis The concentrations of NO and NO2 at reactor inlet and outlet were measured online by a Testo 350XL gas analyzer (Testo, Germany), and the conversion rate of NO2 was defined as follows: cNO2 −inlet −cNO2 −outlet 100% cNO2 −inlet
a
Voltage
Voltage (kV)
30
350 10
Current
20 5 10 0
0
Power (kW)
40
ð2Þ
Current (A)
ηNO2 ¼
b
35
300
30
250
25
200
20
150
15
100
Power Energy
50 -10 200 400 600 800 1000 1200 1400 1600
Time (nsec)
-5
5 0
0 0
10
Energy (mJ)
SED ¼ 60UI=Q
0
500
1000
1500
Time (nsec)
Fig. 3 – Discharge characteristics of the pulsed corona reactor. (a) Typical voltage and current waveforms across the reactor; (b) power and energy deposited in the corona discharge reactor (peak voltage 38 kV, one pulse discharge in pure argon).
2252
J O U R NA L OF EN V I RO N M EN T A L S CI EN C ES 2 6 ( 20 1 4 ) 2 2 49–2 2 5 6
90 Ar N2/Ar O2/Ar
Specific energy density (J/L)
80 70 60 50 40 30 20 10 0 0
50
100
150
200
250
300
Frequency (Hz) Fig. 4 – Effect of pulse frequency on specific energy density.
respectively. The decrease in the SED can be attributed to two main reasons: (a) larger ionization potentials of Ar, and (b) the absence of vibrational excitations and dissociation. Compared with the ionization potential of N2 (15.6 eV) and O2 (12.6 eV), fewer electrons can be generated at a given voltage in pure Ar (15.8 eV) conditions, which decreases the discharge current. Nevertheless, vibrational excitations and dissociations of gas molecules, which consume energetic electrons, only occur in the presence of N2 and O2. At this point, the SED in pure Ar plasma was lower compared with the other two gas compositions. For O2/Ar plasma, electrons were consumed via electron attachment to O2 due to its electronegativity. For the same reason, the SED of O2/Ar is higher than the N2/Ar case. Similar observations were obtained by Mok et al. (2000).
rate increases from 2.8% to 33.7% in the SED range of 1.2– 38.3 J/L. In the case of N2/Ar, the conversion rate increases from 5.1% to 42.1% for the SED range of 2.5–49.0 J/L, while for O2/Ar the conversion rate increases from 4.3% to 25.9% in the SED range of 2.4–52.4 J/L. Previous studies have shown that increasing plasma energy effectively enhances the electric field, electron density and gas temperature in the discharge, all of which may contribute in different ways to improve the conversion of NO2 (Tu et al., 2011; Tu and Whitehead, 2012). Fig. 5 also shows that when the SED exceeds 31.8 J/L for pure argon plasma, similar results are observed for N2/Ar and O2/Ar when the SED exceeds 43.8 and 50 J/L respectively. In pure Ar plasma, due to the low initial concentration of NO2 molecules compared with background Ar, the probability of NO2 direct dissociation reactions induced by high energy electrons was negligible. Metastable Ar(3P2) atoms and Ar+ ions as active species are generated via Ar + e → Ar(3P2) + e and Ar + e → Ar+ + 2e in the plasma zone. With lower ionization potential of NO2 compared to that of Ar, Penning ionization (Ar(3P2) + e → Ar + NO+ + e) and charge transfer mechanisms (Ar+ + NO → Ar + NO+ + e) played key roles in the conversion of NO2. NO+ finally recombined with a low energy electron via NO+ + e → NO or collided with a high energy electron, via NO+ + e → NO + O. These reactions were responsible for the NO2 removal in pure Ar in this study (Hu et al., 2004; Kutasi, 2011; Tsuji et al., 2002). In the presence of N2 or O2, active species including N and O radicals and metastable nitrogen were efficiently formed via electron impact reactions: k1a ¼ f ðeÞ
e þ N2 →e þ N þ N e þ N þ N→e þ N2 ðAÞ
2.2. Effect of energy density
NO2 conversion rate (%)
45 Ar N2/Ar O2/Ar
40 35
k2 ¼ f ðeÞ:
e þ O2 →e þ O þ O
The effect of SED on the conversion rate of NO2 for different gas compositions was investigated. The experimental data on the variation of NO2 conversion rate at the reactor outlet are plotted in Fig. 5 in terms of SED. It indicates that the NO2 conversion rate increases with increasing SED for all three gas compositions, which means that a part of the NO2 was reduced in the reactor. In pure Ar plasma, the NO2 conversion
k1b ¼ f ðeÞ
k3 ¼ 6:01 10−34 ½M
O þ O2 þ M→O3 þ M
ðR4Þ
NO2 þ N→N2 O þ O
k5 ¼ 1:21 10−11
ðR5Þ
k6 ¼ 3:01 10−12
10
NO þ O3 →NO2 þ O2
k7 ¼ 1:3 10−11
k8 ¼ 2:93 10−11
NO þ O þ M→NO2 þ M
15
ðR3Þ
k4 ¼ 1:03 10−11
NO þ N→N2 þ O
20
ðR2Þ
NO2 þ O→NO þ O2
NO2 þ N2 ðAÞ→N2 þ NO þ O
25
ðR1bÞ
Once N and O radicals formed, the most accepted mechanisms of NO2 conversion to NO in plasma in the reactor are listed as follows (Arai et al., 2004; Herron, 1999; Tochikubo et al., 2009; Yan et al., 1999):
NO2 þ N→NO
30
ðR1aÞ
k9 ¼ 9:02 10−32 ½M
k10 ¼ 1:73 10−14
ðR6Þ ðR7Þ ðR8Þ ðR9Þ ðR10Þ
5 0
0
10
20
30
40
50
Specific energy density (J/L) Fig. 5 – Effect of specific energy density on NO2 conversion rate.
N2 O þ O→N2 þ O2
k11 ¼ 5:00 10−3
ðR11Þ
where, M denotes background gas molecules including N2, O2 or Ar. The unit of two-body reactions is cm3/(molecule·sec), while the unit is cm6/(molecule2·sec) for three-body reactions.
2253
JO U R N A L OF EN V I RO N M EN T A L S CI E NC ES 2 6 ( 20 1 4 ) 2 2 4 9–2 2 5 6
The conversion rate had an order of N2 (2%)/Ar > Ar > O2 (2%)/Ar for a given SED. For example, the conversion rate was 35.03%, 30.13% and 23.22% for all three gas compositions at 30 J/L, respectively. In N2 (2%)/Ar plasma, the main reaction pathways of NO2 conversion were R5, R6 and R7 due to the absence of oxygen. The generation of N radicals was an important step of NO2 removal in N2/Ar plasma. Some researchers (Ferreira et al., 2003; Tatarova et al., 2007) reported that the dissociation degree of N2 increased in high Ar fraction gas streams. Thus the amounts of N+2 ions increased by direct electron impact and fast charge transfer. These N+2 ions then went through a dissociative recombination reaction: e + N+2 → N + N. The generated N radicals were beneficial for NO2 conversion (Hu et al., 2002). In addition, the quenching of N2(A) as a long-lifetime metastable species also played a significant role in the conversion process (Aerts et al., 2012). In O2 (2%)/Ar, NO2 was mainly reduced by O radicals via NO2 + O → NO + O2 (R4). Due to the electronegativity of oxygen, the amount of high-energy electrons generated in the plasma was reduced in the presence of oxygen. These electrons were responsible for the formation of O radicals, thus the conversion rate decreased compared with the pure argon case at the same SED. Energy yield, as an important parameter to evaluate the ability of NO2 conversion in different gas compositions, was also investigated. As shown in Fig. 6, the energy yield of NO2 conversion was in reverse ratio with SED. The highest energy yield of 3.31 g/kWh was observed in N2 (2%)/Ar at 2.39 J/L, while the highest energy yields for Ar and O2 (2%)/Ar were 2.77 and 2.57 g/kWh, respectively. In addition, the energy yield decreased with increasing SED for all three gas compositions. Energy deposited to the reactor was used not only for NO2 conversion, but also for background gas heating, excitation and dissociation. With increasing SED, more high energy electrons and reactive species were generated in the reactor, thus the probability of collision between NO2 and the reactive species decreased at higher SED. The highest energy yield observed in this study was in the N2 (2%)/Ar case. Nevertheless, for a given SED, the order of energy yield was N2 (2%)/ Ar > Ar > O2 (2%)/Ar. The results indicated that the addition of nitrogen improved the energy efficiency because of the formation of N radicals and metastable nitrogen, while the
introduction of oxygen suppressed the energy efficiency due to its electronegativity.
2.3. Effect of initial concentration The effect of initial NO2 concentration on NO2 conversion rate in pure argon is shown in Fig. 7. The experiments were carried out with an initial NOx concentration in the range of 63.5 to 271.2 ppm at a fixed SED. As shown in the figure, with the initial NO2 concentration ranging from 63.5 to 271.2 ppm, NO2 conversion rate decreases from 39.7% to 26.1%. The converted NO2 concentration increases almost linearly from 25.2 to 70.8 ppm. As mentioned before, high energy electrons and reactive species generated by plasma were capable of NO2 conversion. The number of high energy electrons and reactive species produced by the pulsed corona discharge remained the same under a given SED. As the initial NO2 concentration increased, the concentration of electrons and reactive species became relatively low compared to the NO2 concentration. Thus the probability of the reactions between single NO2 molecules and high energy electrons or reactive species became lower, which led to an insufficient conversion of NO2 and accordingly, the conversion efficiency decreased. However, the increased initial NO2 concentration increased the probability of collisions between NO2 molecules and the high energy electrons or reactive species, which resulted in better utilization of the high energy electrons or reactive species. At this point, the converted NO2 concentration increased with increasing initial NO2 concentration.
2.4. Effect of gas composition 2.4.1. NO2 conversion in N2/Ar Fig. 8a shows the effect of N2 content on NO2 conversion in N2/ Ar plasma. The initial concentration of NO2 was 207.2 ppm; the content of N2 was 0% to 8% with Ar balance. The SED was kept at 31.8 J/L. The results show that in the absence of N2, as an extreme case, the concentrations of NO and NO2 are 77 and 127.1 ppm, respectively. As mentioned before, in pure Ar plasma, NO2 can be converted via charge transfer mechanisms. With varied nitrogen content from 0% to 2%, the NO2 conversion rate increases slightly from 38.65% to 39.81%.
80
3.6
2.4 2.0 1.6 1.2 0.8
NO2 Conversion Rate Converted NO2
38
70
36
60
34 50 32 40
30 28
Converted NO2 (ppm)
2.8
NO2 conversion rate (%)
Ar N2/Ar O2/Ar
3.2
NO2 energy yield (g/kWh)
40
30
0.4 26
0.0
0
5
10
15
20
25
30
35
40
45
50
55
Specific energy density (J/L) Fig. 6 – Effect of specific energy density on NO2 energy yield.
50
100
150
200
250
20 300
Initial NO2 concentration (ppm) Fig. 7 – Effect of initial concentration on NO2 conversion rate.
2254
J O U R NA L OF EN V I RO N M EN T A L S CI EN C ES 2 6 ( 20 1 4 ) 2 2 49–2 2 5 6
b
NO NO2 O2
80
0.20
60 0.15
40
NO2 conversion (%)
50 100
20 0
200
48
190 NOx (NO+NO2) NO2 conversion
46
180
44 170
42
160
40 0
1
2
3
4
5
6
7
8
0.10
210
38
0
1
2
3
N2 content (%)
4
5
6
7
8
NOx concentration (ppm)
52
O2 content (%)
NOx concentration (ppm)
0.25
a
120
150
N2 content (%)
Fig. 8 – Effect of N2 content on NOx conversion. (a) NO, NO2 and O2 concentrations; (b) NO2 conversion and total converted NOx concentration.
NO NO2 O3
120 100
1.0
80
0.5
60 40 0
Fig. 9 shows the variation of NO, NO2 and O3 with O2 content in O2/Ar plasma. The gas composition was NO2 210 ppm, and the content of O2 was 0% to 8% with argon balance. The SED was the same as the experiments in N2/Ar. The results show that increasing the oxygen content gradually increased the concentration of NO2 from 127.1 to 170.3 ppm, while the NO concentration decreased accordingly from 77 to 29 ppm (Fig. 9a). These results indicated that the addition of oxygen suppressed the conversion of NO2 to NO with increasing oxygen content. In O2/Ar atmosphere, the oxidation pathways, i.e., ozoneand O radical-containing reactions, dominated the process of NO2 conversion (Yoshida et al., 2009a). O radicals can be generated via both electron-induced oxygen dissociation (R2) and quenching of metastable Ar by O2 molecules. The dissociation of oxygen and quenching of metastable Ar produced ground state O (3P) and excited O (1D), which
1.5
a
140
20
2.4.2. NO2 conversion in O2/Ar
1
2
3
4
5
O2 content (%)
6
7
8
0.0
45
b
220
40
NO2 conversion (%)
160
O3 concentration (ppm)
NOx concentration (ppm)
180
decreases with increasing N2 content. NOx can be reduced to N2O or N2 via R5 or R8, as proven by other researchers (Okumura et al., 1994). The presence of slight amounts of oxygen may be attributed to R4 or recombination of O radicals generated from NO reduction in R5, which was confirmed by Penetrante et al. (1995) and Zhu et al. (2005).
200 180
35
160
NO2 conversion NOx
30
140 120
25
100 80
20 15
NOx concentration (ppm)
When the nitrogen content was between 2% and 3%, the conversion rate rises remarkably from 39.81% to 49.71%. When the nitrogen content exceeds 3%, the conversion rate increases slightly and reaches 50.33% at 8% nitrogen content. In N2/Ar atmosphere, N radicals and metastable nitrogen were generated during the discharge via electron-induced reactions (R1a, R1b) and charge transfer reactions. These two kinds of active species may induce conversion reactions upon collision with pollutants (Aerts et al., 2013). Due to the existence of metastable Ar(3P2), the excitation of ground state N2 molecules by Ar(3P2) + N2 → Ar + N2(A) cannot be negligible (Masoud et al., 2005). Thus NO2 conversion was dominated by conventional reduction reactions with N radicals and metastable N2(A) via R5, R6 and R7 (Jõgi et al., 2009). With increasing nitrogen content, electron excitation and dissociation of nitrogen became more and more important, which led to a boost of N radical generation. These radicals accelerated the aforementioned NO2 reduction reactions and finally reached a higher conversion rate. When N2 exceeded 3%, the NO2 conversion rate was saturated. Since N2 possessed a larger cross section than Ar did, the generated N radicals could be consumed via recombination reaction N + N + M → N2 + M (M_N2, Ar). This reaction suppressed further NO2 conversion in N2/Ar atmosphere at higher N2 content. In addition, the total concentration of NOx (Fig. 8b)
60 0
1
2
3
4
5
6
7
8
O2 content (%)
Fig. 9 – Effect of O2 content on NOx conversion. (a) NO, NO2 and O3 concentrations; (b) NO2 conversion and total NOx concentration.
JO U R N A L OF EN V I RO N M EN T A L S CI E NC ES 2 6 ( 20 1 4 ) 2 2 4 9–2 2 5 6
oxidized NO to NO2. These O radicals can also react with O2 via three-body reaction and form O3 via R3. These O radicals and ozone can both contribute to NO oxidation, inhibiting further NO2 conversion. As mentioned before, O can also reduce NO2 to NO, however the reaction rate constant of NO oxidation with O radicals in an oxygen-rich environment is 3.00 × 10−11 cm−3/(mol·s), which is much larger than that of reduction reaction R4. Thus with increasing oxygen content, the net converted NO2 decreased, resulting in NO2 conversion rate decrease, as shown in Fig. 9b. Since no N2O and N2 were formed in the oxygen-rich plasma, the change in total NOx conversion can be regarded as negligible. In the presence of NO2, the detected ozone concentration was below 1 ppm. In order to prove the role of ozone in the NO2 removal process, the ozone concentration was monitored at the reactor outlet, as shown in Fig. 10. The variation of oxygen content from 0% to 8% resulted in ozone formation from 0 to 31.3 ppm. Compared to the concentration of NOx in the NO2/O2/Ar case, the ozone concentration was one order of magnitude lower, which suggests that ozone was depleted via NO oxidation and NO2 reduction reactions in the presence of NOx (Yan et al., 1999).
2255
O2 inhibited the conversion process. The highest energy yield of 3.31 g/kWh was observed in N2/Ar and decreased with higher SED; the same trends were observed in other two gas compositions. (3) The presence of N2 and O2 led to different reaction pathways for NO2 conversion due to the formation of different reactive radicals respectively. In the case of the N2/Ar gas mixture, N radicals and metastable N2(A) played crucial roles in the NO2 conversion. The conversion rate increased dramatically in the N2 content range of 2%–3% and it reached equilibrium when N2 content exceeded 3%. In the O2/Ar case, O radicals and O3 played an important role via oxidation pathways, and the conversion rate decreased monotonically over tested O2 concentrations. For higher conversion efficiency under the given conditions, a wet scrubbing unit should be employed for NO2 absorption.
Acknowledgments This work was supported by the National Science Fund for Distinguished Young Scholars (No. 51125025) and the National Natural Science Foundation of China (Nos. 51076140, 51206143).
3. Conclusions The discharge characteristics and NO2 conversion performance within a pulsed corona discharge reactor in simulated gas compositions of Ar, N2/Ar and O2/Ar were investigated experimentally. The results were summarized as follows. (1) The specific energy density (SED) increased from 1.16 to 38.03 J/L, from 2.39 to 49.03 J/L and from 2.46 to 82.56 J/L for N2/Ar, Ar and O2/Ar, respectively, with increasing pulse frequency from 10 to 300 Hz. For a given pulse frequency, SED had the order O2/Ar > Ar > N2/Ar. (2) Higher SED led to higher NO2 conversion rate: the conversion efficiency increased from 2.8% to 33.7%, 5.1% to 42.1%, and 4.27% to 25.94% for N2/Ar, Ar and O2/Ar. The conversion rate was also affected by gas composition: the addition of N2 enhanced NO2 conversion while
35 O3 concentration
O3 concentration (ppm)
30 25 20 15 10 5 0 0
1
2
3
4
5
6
7
8
O2 content (%) Fig. 10 – Ozone formation without initial NO2 in O2/Ar mixture.
REFERENCES Aerts, R., Tu, X., De Bie, C., Whitehead, J.C., Bogaerts, A., 2012. An investigation into the dominant reactions for ethylene destruction in non-thermal atmospheric plasmas. Plasma Process. Polym. 9 (10), 994–1000. Aerts, R., Tu, X., Van Gaens, W., Whitehead, J.C., Bogaerts, A., 2013. Gas purification by nonthermal plasma: a case study of ethylene. Environ. Sci. Technol. 47 (12), 6478–6485. Arai, M., Saito, M., Yoshinaga, S., 2004. Effect of oxygen on NOx removal in corona discharge field: NOx behavior without a reducing agent. Combust. Sci. Technol. 176 (10), 1653–1665. Cha, M.S., Song, Y.H., Lee, J.O., Kim, S.J., 2007. NOx and soot reduction using dielectric barrier discharge and NH3 selective catalytic reduction in diesel exhaust. Int. J. Plasma Environ. Sci. Technol. 1 (1), 28–33. Eichwald, O., Guntoro, N.A., Yousfi, M., Benhenni, M., 2002. Chemical kinetics with electrical and gas dynamics modelization for NOx removal in an air corona discharge. J. Phys. D. Appl. Phys. 35 (5), 439–450. Ferreira, C.M., Tatarova, E., Guerra, V., Gordiets, B.F., Henriques, J., Dias, F.M., et al., 2003. Modeling of wave driven molecular (H2, N2, N2–Ar) discharges as atomic sources. IEEE Trans. Plasma Sci. 31 (4), 645–658. Gallon, H.J., Tu, X., Twigg, M.V., Whitehead, J.C., 2011. Plasma-assisted methane reduction of a NiO catalyst—low temperature activation of methane and formation of carbon nanofibres. Appl. Catal. B Environ. 106 (3–4), 616–620. Herron, J.T., 1999. Evaluated chemical kinetics data for reactions of N(2D), N(2P), and N2(A 3Σu +) in the gas phase. J. Phys. Chem. Ref. Data 28 (5), 1453–1483. Hu, X., Nicholas, J., Zhang, J., Temi, M.L., Filippis, P.D., Agarwal, P.K., 2002. The destruction of N2O in a pulsed corona discharge reactor. Fuel 81, 1259–1268. Hu, X., Zhao, G.B., Zhang, J.J., Wang, L., Radosz, M., 2004. Nonthermal-plasma reactions of dilute nitrogen oxide mixtures: NOx-in-argon and NOx + CO-in-argon. Ind. Eng. Chem. Res. 43 (23), 7456–7464.
2256
J O U R NA L OF EN V I RO N M EN T A L S CI EN C ES 2 6 ( 20 1 4 ) 2 2 49–2 2 5 6
Jõgi, I., Bichevin, V., Laan, M., Haljaste, A., Käämbre, H., 2009. NO conversion by dielectric barrier discharge and TiO2 catalyst: effect of oxygen. Plasma Chem. Plasma Process. 29 (3), 205–215. Kutasi, K., 2011. Modelling of NO destruction in a low-pressure reactor by an Ar plasma jet: species abundances in the reactor. J. Phys. D. Appl. Phys. 44 (10), 105202. Lee, Y.H., Jung, W.S., Choi, Y.R., Oh, J.S., Jang, S.D., Son, Y.G., et al., 2003. Application of pulsed corona induced plasma chemical process to an industrial incinerator. Environ. Sci. Technol. 37 (11), 2563–2567. Lin, H., Guan, B., Cheng, Q., Huang, Z., 2010. An investigation on the principal paths to plasma oxidation of propylene and no. Energy Fuel 24 (10), 5418–5425. Masoud, N., Martus, K., Becker, K., 2005. VUV emission from a cylindrical dielectric barrier discharge in Ar and in Ar/N2 and Ar/air mixtures. J. Phys. D. Appl. Phys. 38 (11), 1674–1683. Masuda, S., Nakao, H., 1990. Control of NOx by positive and negative pulsed corona discharges. IEEE Trans. Ind. Appl. 26 (2), 374–383. Matsumoto, T., Wang, D.Y., Namihira, T., Akiyama, H., 2010. Energy efficiency improvement of nitric oxide treatment using nanosecond pulsed discharge. IEEE Trans. Plasma Sci. 38 (10), 2639–2643. Mizuno, A., Shimizu, R., Chakrabarti, A., Dascalescu, L., Furuta, S., 1995. NOx removal process using pulsed discharge plasma. IEEE Trans. Ind. Appl. 31 (5), 957–962. Mok, Y.S., Ham, S.W., 1997. Conversion of NO to NO2 in air by a pulsed corona discharge process. Chem. Eng. Sci. 53 (9), 1667–1678. Mok, Y.S., Ham, S.W., Nam, I.S., 1998. Evaluation of energy utilization efficiencies for SO2 and NO removal by pulsed corona discharge process. Plasma Chem. Plasma Process. 18 (4), 535–550. Mok, Y.S., Kim, J.H., Nam, I.S., Ham, S.W., 2000. Removal of NO and formation of byproducts in a positive-pulsed corona discharge reactor. Ind. Eng. Chem. Res. 39, 3938–3944. Obradovic, B.M., Sretenovic, G.B., Kuraica, M.M., 2011. A dual-use of DBD plasma for simultaneous NOx and SO2 removal from coal-combustion flue gas. J. Hazard. Mater. 185 (2–3), 1280–1286. Okumura, T., Matsuda, S., Satoh, Y., Sakai, Y., Tagashira, H., 1994. Decomposition of NO2 in a glow discharge and electron impact ionization and attachment coefficients in NO2/air mixtures. J. Phys. D. Appl. Phys. 27 (4), 801–806. Orlandini, I., Riedel, U., 2000. Chemical kinetics of NO removal by pulsed corona discharges. J. Phys. D. Appl. Phys. 33, 2467–2474. Penetrante, B.M., Hsiao, M.C., Merritt, B.T., Vogtlin, G.E., Wallman, P.H., 1995. Comparison of electrical discharge techniques for nonthermal plasma processing of NO in N2. IEEE Trans. Plasma Sci. 23 (4), 679–687. Penetrante, B.M., Hsiao, M.C., Merritt, B.T., Vogtlin, G.E., 1997. Fundamental limits on NOx reduction by plasma. SAE Technical Paper Series, 971715, p. 971715.
Skalska, K., Miller, J.S., Ledakowicz, S., 2010. Trends in NOx abatement: a review. Sci. Total Environ. 408 (19), 3976–3989. Sun, B., Yin, S., 2009. The characteristics of NO reduction in the reactor with dielectric barrier discharge. ASME 2009 3rd International Conference of Energy Sustainability. San Francisco, USA, pp. 19–23 (July). Tatarova, E., Guerra, V., Henriques, J., Ferreira, C.M., 2007. Nitrogen dissociation in low-pressure microwave plasmas. J. Phys. Conf. Ser. 71, 012010. Tochikubo, F., Uchida, S., Yasui, H., Sato, K., 2009. Numerical simulation of NO oxidation in dielectric barrier discharge with microdischarge formation. Jpn. J. Appl. Phys. 48 (7), 076507. Tsuji, M., Nakano, K., Kumagae, J., Tsuji, T., 2002. Decomposition of NO by microwave discharge of NO/He or NO/Ar mixtures. Bull. Chem. Soc. Jpn. 75, 607–614. Tu, X., Whitehead, J.C., 2012. Plasma-catalytic dry reforming of methane in an atmospheric dielectric barrier discharge: understanding the synergistic effect at low temperature. Appl. Catal. B Environ. 125, 439–448. Tu, X., Gallon, H.J., Whitehead, J.C., 2011. Electrical and spectroscopic diagnostics of a single-stage plasma-catalysis system: effect of packing with TiO2. J. Phys. D. Appl. Phys. 44 (48), 482003. Van Durme, J., Dewulf, J., Leys, C., Van Langenhove, H., 2008. Combining non-thermal plasma with heterogeneous catalysis in waste gas treatment: a review. Appl. Catal. B Environ. 78 (3–4), 324–333. Wu, Z.L., Gao, X., Luo, Z.Y., Wei, E.Z., Zhang, Y.S., Zhang, J.Z., et al., 2005. NOx treatment by DC corona radical shower with different geometric nozzle electrodes. Energy Fuel 19 (6), 2279–2286. Yan, K., Kanazawa, S., Ohkubo, T., Nomoto, Y., 1999. Oxidation and reduction processes during NOx removal with corona-induced nonthermal plasma. Plasma Chem. Plasma Process. 19 (3), 421–443. Yoshida, K., Rajanikanth, B.S., Okubo, M., 2009a. NOx reduction and desorption studies under electric discharge plasma using a simulated gas mixture: a case study on the effect of corona electrodes. Plasma Sci. Technol. 11 (3), 327–333. Yoshida, K., Yamamoto, T., Kuroki, T., Okubo, M., 2009b. Pilot-scale experiment for simultaneous dioxin and NOx removal from garbage incinerator emissions using the pulse corona induced plasma chemical process. Plasma Chem. Plasma Process. 29 (5), 373–386. Yu, L., Li, X., Tu, X., Wang, Y., Lu, S., Yan, J., 2010a. Decomposition of naphthalene by dc gliding arc gas discharge. J. Phys. Chem. A 114 (1), 360–368. Yu, L., Tu, X., Li, X., Wang, Y., Chi, Y., Yan, J., 2010b. Destruction of acenaphthene, fluorene, anthracene and pyrene by a DC gliding arc plasma reactor. J. Hazard. Mater. 180 (1–3), 449–455. Zhu, A.M., Sun, Q., Niu, J.H., Xu, Y., Song, Z.M., 2005. Conversion of NO in NO/N2, NO/O2/N2, NO/C2H4/N2 and NO/C2H4/O2/N2 systems by dielectric barrier discharge plasmas. Plasma Chem. Plasma Process. 25 (4), 371–386.
Available online at www.sciencedirect.com
ScienceDirect www.journals.elsevier.com/journal-of-environmental-sciences
Experimental study of NO2 reduction in N2/Ar and O2/Ar mixtures by pulsed corona discharge Xinbo Zhu1 , Chenghang Zheng1 , Xiang Gao1,⁎, Xu Shen1,2 , Zhihua Wang1 , Zhongyang Luo1 , Kefa Cen1 1. State Key Laboratory of Clean Energy Utilization, Zhejiang University, Hangzhou 310027, China. E-mail: [email protected] 2. Zhejiang Electric Power Design Institute, Hangzhou 310012, China
AR TIC LE I N FO
ABS TR ACT
Article history:
Non-thermal plasma technology has been regarded as a promising alternative technology
Received 20 December 2013
for NOx removal. The understanding of NO2 reduction characteristics is extremely
Revised 16 February 2014
important since NO2 reduction could lower the total NO oxidation rate in the plasma
Accepted 6 March 2014
atmosphere. In this study, NO2 reduction was experimentally investigated using a
Available online 29 September 2014
non-thermal plasma reactor driven by a pulsed power supply for different simulated gas
Keywords:
the specific energy density (SED), and the highest conversion rates were 33.7%, 42.1% and
Non-thermal plasma
25.7% for Ar, N2/Ar and O2/Ar, respectively. For a given SED, the NO2 conversion rate had the
Pulsed corona discharge
order N2/Ar > Ar > O2/Ar. The highest energy yield of 3.31 g/kWh was obtained in N2/Ar
NO
plasma and decreased with increasing SED; the same trends were also found in the other
compositions and operating parameters. The NO2 reduction was promoted by increasing
NO2
two gas compositions. The conversion rate decreased with increasing initial NO2 concentration. Furthermore, the presence of N2 or O2 led to different reaction pathways for NO2 conversion due to the formation of different dominating reactive radicals. © 2014 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.
Introduction Nitrogen oxides (NOx) constitute one of the most hazardous air pollutants from coal-fired power plants and gasoline engine exhaust. Regarded as a precursor of ozone, acid rain and photo-chemical smog, NOx emission is becoming a crucial issue throughout the world, and NOx treatment is an urgent task due to increasingly stringent environmental regulations. Up to now, many technologies have been developed for NOx treatment such as selective catalytic reduction (SCR), selective non-catalytic reduction (SNCR), activated carbon adsorption and wet scrubbing (Skalska et al., 2010). In the past two decades, non-thermal plasma has attracted considerable attention due to its capability for simultaneous removal of multiple pollutants
under ambient conditions with relative low energy cost (Van Durme et al., 2008; Yu et al., 2010a,b). In non-thermal plasma, energy deposited to the reactor generates highly energetic electrons. These electrons collide with background gas molecules, resulting in the formation of reactive species like ions, radicals and excited molecules, which are able to initiate gas phase reactions and convert pollutants to final harmless products (Gallon et al., 2011; Tu and Whitehead, 2012). NO conversion processes in plasma have been experimentally investigated by many researchers in terms of various operating parameters including reactor geometry (Mizuno et al., 1995; Wu et al., 2005), power supply (Matsumoto et al., 2010; Mok et al., 1998), polarity (Masuda and Nakao, 1990), gas composition (Arai et al., 2004; Sun and Yin, 2009), temperature (Eichwald et al., 2002) and additives (Cha et al., 2007; Lin et al., 2010). Kinetic models
⁎ Corresponding author. E-mail: [email protected] (Xiang Gao).
http://dx.doi.org/10.1016/j.jes.2014.09.010 1001-0742/© 2014 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.
2250
J O U R NA L OF EN V I RO N M EN T A L S CI EN C ES 2 6 ( 20 1 4 ) 2 2 49–2 2 5 6
were also introduced to understand the mechanism of plasmainduced NOx conversion (Mok and Ham, 1997; Orlandini and Riedel, 2000). More recently, a few researchers reported high conversion rates using non-thermal plasma-based technologies for NOx treatment in industrial applications (Lee et al., 2003; Yoshida et al., 2009b). It is generally recognized that in industrial application of a plasma-combined deNOx process, plasma is employed as a gas phase oxidizer for NO to NO2 conversion. NO2 can be easily converted to nitric acid or adsorbed by wet scrubbing (Obradovic et al., 2011). However, some researchers found that the oxidation of NO to NO2 in the plasma environment was not straightforward (Penetrante et al., 1997). More NO molecules would be oxidized to NO2 by plasma at high energy density, however when NO2 concentration became higher, these NO2 molecules could react with N or O radicals via reverse reactions: NO2 + O → NO + O2 or NO2 + N → 2NO, producing NO molecules and suppressing the total NO conversion efficiency. More importantly, it is difficult to deal with NO in the aforementioned wet flue gas desulfurization (WFGD) process. As mentioned above, the suppression of NO2 reduction to NO should be investigated to prevent the decrease of NOx conversion rate and optimize the working conditions of plasma. In this study, a reactor driven by a pulsed corona power supply was employed to investigate the conversion characteristics of NO2 to NO in various gas compositions. The discharge characteristics under different simulated gas compositions were evaluated. The influences of energy density and initial concentration on NO2 conversion were also studied. Finally, the effects of N2 and O2 content on NO2 conversion were investigated and the reaction pathways were compared for both N2 and O2 addition cases. In order to facilitate the understanding of plasma-induced chemical reactions, especially electron–molecule impact reactions, argon was used as a background gas for all tested conditions.
Furthermore, as an inert gas, no argon-containing products were formed in the reactions.
1. Materials and methods 1.1. Experimental setup Fig. 1 shows a schematic diagram of a pulsed corona discharge system designed and built for this study. The system consists of four parts: a pulsed corona discharge reactor, a pulsed power supply, a gas control system and gas analysis units. A detailed drawing of the pulsed corona discharge reactor is illustrated in Fig. 2. The reactor was designed in a wire-to-plate configuration. The reactor body was a quartz glass box with the geometry of 600 mm length, 90 mm width and 60 mm height. Two stainless steel plates were placed at the inner sides of the reactor as the grounded electrodes. Eight stainless steel discharging wires with an interval of 50 mm were mounted on the center line of the reactor top side as the high voltage electrodes. The diameter of the high voltage electrodes was 2 mm with an effective discharge length of 40 mm. The distance between the high voltage electrodes and the grounded electrodes was 40 mm. The reactor was powered by a pulsed power supply. A model PPCP-1-300-400 pulsed power supply (Zhongwuhaitong, Sichuan, China) was designed to generate high voltage pulses with a rise time of 100–200 nsec, a duration of 600–800 nsec and an adjustable repetition pulse frequency of 10–300 Hz. The maximum amplitude of the pulse voltage was 60 kV and the maximum output power was 500 W. The pulsed corona discharge voltage was measured at the discharging cable by a 1000:1 P6015A voltage probe (Tektronix Inc., USA), while the current waveform was measured at the grounded electrode by a Model 6585 current probe (Pearson Electronics Inc., USA). The voltage and current waveforms
Oscilloscope Gas analyzer
Pulse power supply
Ozone analyzer
Current prober
Mixing chamber
Plasma reactor
HV probe
Grounded N2
O2
Fig. 1 – Schematic of the experimental set-up.
NO
NO2
Ar
CO2
2251
JO U R N A L OF EN V I RO N M EN T A L S CI E NC ES 2 6 ( 20 1 4 ) 2 2 4 9–2 2 5 6
Discharging wires
Inlet
Outlet
Grounded electrodes
Fig. 2 – Structure of the pulsed corona discharge reactor.
where, CNO2-intlet and CNO2-outlet are the concentrations of NO2 at the inlet and outlet of the reactor, respectively. The concentration of ozone was monitored by a Model 106M UV-absorption ozone analyzer (2B Technologies, USA) at the reactor outlet.
were monitored and recorded with a DPO4034B oscilloscope (Tektronix, Inc., USA). The discharge power was calculated by multiplying voltage and current waveforms, while the total energy consumption was obtained by integrating discharge power over the discharge periods. In this paper, all the results are presented in terms of specific energy density (SED, J/L), which is calculated as follows:
2. Results and discussion
ð1Þ
2.1. Discharge characteristics
where, U (kV) and I (mA) are the voltage and current respectively, while Q denotes the gas flow rate (L/min).
As a key parameter of non-thermal plasma-based air pollution control technology, energy density is affected by both discharge voltage and pulse frequency. Fig. 3a shows the typical voltage and current waveforms at the peak voltage of 38 kV. As mentioned in Section 1, the plasma power was calculated by multiplying voltage and current waveforms while the energy consumption in one pulse period was obtained by integrating the plasma power over one pulse period. As shown in Fig. 3b, for a single pulse period, all energy is deposited to the reactor in the first 700 nsec and the total deposited energy is about 34.65 mJ. In this study, the energy density was adjusted by changing the pulse frequency, while the discharge peak voltage was fixed at 38 kV. Fig. 4 shows the relationship between SED and pulse frequency for the aforementioned three gas compositions. As the pulse frequency increases from 10 to 300 Hz, the SED increases monotonically from 1.16 to 38.03 J/L, 2.39 to 49.03 J/L, and 2.46 to 82.56 J/L, for Ar, N2 (2%)/Ar and O2 (2%)/Ar, respectively. For a certain pulse frequency, the SED values were generally in the order: O2 (2%)/Ar > N2 (2%)/ Ar > Ar. In a plasma environment, electrons are generated via electron-induced ionization reactions, i.e., e + Ar → 2e + Ar+, e + N2 → 2e + N+2 and e + O2 → 2e + O+2 for Ar, N2 and O2,
1.2. Reagents and methods The experiments were carried out at room temperature and atmospheric pressure using the gas mixtures of NO2/Ar, NO2/N2/Ar and NO2/O2/Ar. All the feed gases used in this study were of high-purity grade (> 99.99%). Gas streams from compressed gas cylinders were regulated by a series of calibrated DB-07B mass controllers (Sevenstars, Beijing, China). The total gas flow rate was fixed at 5 L/min and the initial NOx (NO + NO2) concentration was 210 ppm unless otherwise mentioned.
1.3. Chemical analysis The concentrations of NO and NO2 at reactor inlet and outlet were measured online by a Testo 350XL gas analyzer (Testo, Germany), and the conversion rate of NO2 was defined as follows: cNO2 −inlet −cNO2 −outlet 100% cNO2 −inlet
a
Voltage
Voltage (kV)
30
350 10
Current
20 5 10 0
0
Power (kW)
40
ð2Þ
Current (A)
ηNO2 ¼
b
35
300
30
250
25
200
20
150
15
100
Power Energy
50 -10 200 400 600 800 1000 1200 1400 1600
Time (nsec)
-5
5 0
0 0
10
Energy (mJ)
SED ¼ 60UI=Q
0
500
1000
1500
Time (nsec)
Fig. 3 – Discharge characteristics of the pulsed corona reactor. (a) Typical voltage and current waveforms across the reactor; (b) power and energy deposited in the corona discharge reactor (peak voltage 38 kV, one pulse discharge in pure argon).
2252
J O U R NA L OF EN V I RO N M EN T A L S CI EN C ES 2 6 ( 20 1 4 ) 2 2 49–2 2 5 6
90 Ar N2/Ar O2/Ar
Specific energy density (J/L)
80 70 60 50 40 30 20 10 0 0
50
100
150
200
250
300
Frequency (Hz) Fig. 4 – Effect of pulse frequency on specific energy density.
respectively. The decrease in the SED can be attributed to two main reasons: (a) larger ionization potentials of Ar, and (b) the absence of vibrational excitations and dissociation. Compared with the ionization potential of N2 (15.6 eV) and O2 (12.6 eV), fewer electrons can be generated at a given voltage in pure Ar (15.8 eV) conditions, which decreases the discharge current. Nevertheless, vibrational excitations and dissociations of gas molecules, which consume energetic electrons, only occur in the presence of N2 and O2. At this point, the SED in pure Ar plasma was lower compared with the other two gas compositions. For O2/Ar plasma, electrons were consumed via electron attachment to O2 due to its electronegativity. For the same reason, the SED of O2/Ar is higher than the N2/Ar case. Similar observations were obtained by Mok et al. (2000).
rate increases from 2.8% to 33.7% in the SED range of 1.2– 38.3 J/L. In the case of N2/Ar, the conversion rate increases from 5.1% to 42.1% for the SED range of 2.5–49.0 J/L, while for O2/Ar the conversion rate increases from 4.3% to 25.9% in the SED range of 2.4–52.4 J/L. Previous studies have shown that increasing plasma energy effectively enhances the electric field, electron density and gas temperature in the discharge, all of which may contribute in different ways to improve the conversion of NO2 (Tu et al., 2011; Tu and Whitehead, 2012). Fig. 5 also shows that when the SED exceeds 31.8 J/L for pure argon plasma, similar results are observed for N2/Ar and O2/Ar when the SED exceeds 43.8 and 50 J/L respectively. In pure Ar plasma, due to the low initial concentration of NO2 molecules compared with background Ar, the probability of NO2 direct dissociation reactions induced by high energy electrons was negligible. Metastable Ar(3P2) atoms and Ar+ ions as active species are generated via Ar + e → Ar(3P2) + e and Ar + e → Ar+ + 2e in the plasma zone. With lower ionization potential of NO2 compared to that of Ar, Penning ionization (Ar(3P2) + e → Ar + NO+ + e) and charge transfer mechanisms (Ar+ + NO → Ar + NO+ + e) played key roles in the conversion of NO2. NO+ finally recombined with a low energy electron via NO+ + e → NO or collided with a high energy electron, via NO+ + e → NO + O. These reactions were responsible for the NO2 removal in pure Ar in this study (Hu et al., 2004; Kutasi, 2011; Tsuji et al., 2002). In the presence of N2 or O2, active species including N and O radicals and metastable nitrogen were efficiently formed via electron impact reactions: k1a ¼ f ðeÞ
e þ N2 →e þ N þ N e þ N þ N→e þ N2 ðAÞ
2.2. Effect of energy density
NO2 conversion rate (%)
45 Ar N2/Ar O2/Ar
40 35
k2 ¼ f ðeÞ:
e þ O2 →e þ O þ O
The effect of SED on the conversion rate of NO2 for different gas compositions was investigated. The experimental data on the variation of NO2 conversion rate at the reactor outlet are plotted in Fig. 5 in terms of SED. It indicates that the NO2 conversion rate increases with increasing SED for all three gas compositions, which means that a part of the NO2 was reduced in the reactor. In pure Ar plasma, the NO2 conversion
k1b ¼ f ðeÞ
k3 ¼ 6:01 10−34 ½M
O þ O2 þ M→O3 þ M
ðR4Þ
NO2 þ N→N2 O þ O
k5 ¼ 1:21 10−11
ðR5Þ
k6 ¼ 3:01 10−12
10
NO þ O3 →NO2 þ O2
k7 ¼ 1:3 10−11
k8 ¼ 2:93 10−11
NO þ O þ M→NO2 þ M
15
ðR3Þ
k4 ¼ 1:03 10−11
NO þ N→N2 þ O
20
ðR2Þ
NO2 þ O→NO þ O2
NO2 þ N2 ðAÞ→N2 þ NO þ O
25
ðR1bÞ
Once N and O radicals formed, the most accepted mechanisms of NO2 conversion to NO in plasma in the reactor are listed as follows (Arai et al., 2004; Herron, 1999; Tochikubo et al., 2009; Yan et al., 1999):
NO2 þ N→NO
30
ðR1aÞ
k9 ¼ 9:02 10−32 ½M
k10 ¼ 1:73 10−14
ðR6Þ ðR7Þ ðR8Þ ðR9Þ ðR10Þ
5 0
0
10
20
30
40
50
Specific energy density (J/L) Fig. 5 – Effect of specific energy density on NO2 conversion rate.
N2 O þ O→N2 þ O2
k11 ¼ 5:00 10−3
ðR11Þ
where, M denotes background gas molecules including N2, O2 or Ar. The unit of two-body reactions is cm3/(molecule·sec), while the unit is cm6/(molecule2·sec) for three-body reactions.
2253
JO U R N A L OF EN V I RO N M EN T A L S CI E NC ES 2 6 ( 20 1 4 ) 2 2 4 9–2 2 5 6
The conversion rate had an order of N2 (2%)/Ar > Ar > O2 (2%)/Ar for a given SED. For example, the conversion rate was 35.03%, 30.13% and 23.22% for all three gas compositions at 30 J/L, respectively. In N2 (2%)/Ar plasma, the main reaction pathways of NO2 conversion were R5, R6 and R7 due to the absence of oxygen. The generation of N radicals was an important step of NO2 removal in N2/Ar plasma. Some researchers (Ferreira et al., 2003; Tatarova et al., 2007) reported that the dissociation degree of N2 increased in high Ar fraction gas streams. Thus the amounts of N+2 ions increased by direct electron impact and fast charge transfer. These N+2 ions then went through a dissociative recombination reaction: e + N+2 → N + N. The generated N radicals were beneficial for NO2 conversion (Hu et al., 2002). In addition, the quenching of N2(A) as a long-lifetime metastable species also played a significant role in the conversion process (Aerts et al., 2012). In O2 (2%)/Ar, NO2 was mainly reduced by O radicals via NO2 + O → NO + O2 (R4). Due to the electronegativity of oxygen, the amount of high-energy electrons generated in the plasma was reduced in the presence of oxygen. These electrons were responsible for the formation of O radicals, thus the conversion rate decreased compared with the pure argon case at the same SED. Energy yield, as an important parameter to evaluate the ability of NO2 conversion in different gas compositions, was also investigated. As shown in Fig. 6, the energy yield of NO2 conversion was in reverse ratio with SED. The highest energy yield of 3.31 g/kWh was observed in N2 (2%)/Ar at 2.39 J/L, while the highest energy yields for Ar and O2 (2%)/Ar were 2.77 and 2.57 g/kWh, respectively. In addition, the energy yield decreased with increasing SED for all three gas compositions. Energy deposited to the reactor was used not only for NO2 conversion, but also for background gas heating, excitation and dissociation. With increasing SED, more high energy electrons and reactive species were generated in the reactor, thus the probability of collision between NO2 and the reactive species decreased at higher SED. The highest energy yield observed in this study was in the N2 (2%)/Ar case. Nevertheless, for a given SED, the order of energy yield was N2 (2%)/ Ar > Ar > O2 (2%)/Ar. The results indicated that the addition of nitrogen improved the energy efficiency because of the formation of N radicals and metastable nitrogen, while the
introduction of oxygen suppressed the energy efficiency due to its electronegativity.
2.3. Effect of initial concentration The effect of initial NO2 concentration on NO2 conversion rate in pure argon is shown in Fig. 7. The experiments were carried out with an initial NOx concentration in the range of 63.5 to 271.2 ppm at a fixed SED. As shown in the figure, with the initial NO2 concentration ranging from 63.5 to 271.2 ppm, NO2 conversion rate decreases from 39.7% to 26.1%. The converted NO2 concentration increases almost linearly from 25.2 to 70.8 ppm. As mentioned before, high energy electrons and reactive species generated by plasma were capable of NO2 conversion. The number of high energy electrons and reactive species produced by the pulsed corona discharge remained the same under a given SED. As the initial NO2 concentration increased, the concentration of electrons and reactive species became relatively low compared to the NO2 concentration. Thus the probability of the reactions between single NO2 molecules and high energy electrons or reactive species became lower, which led to an insufficient conversion of NO2 and accordingly, the conversion efficiency decreased. However, the increased initial NO2 concentration increased the probability of collisions between NO2 molecules and the high energy electrons or reactive species, which resulted in better utilization of the high energy electrons or reactive species. At this point, the converted NO2 concentration increased with increasing initial NO2 concentration.
2.4. Effect of gas composition 2.4.1. NO2 conversion in N2/Ar Fig. 8a shows the effect of N2 content on NO2 conversion in N2/ Ar plasma. The initial concentration of NO2 was 207.2 ppm; the content of N2 was 0% to 8% with Ar balance. The SED was kept at 31.8 J/L. The results show that in the absence of N2, as an extreme case, the concentrations of NO and NO2 are 77 and 127.1 ppm, respectively. As mentioned before, in pure Ar plasma, NO2 can be converted via charge transfer mechanisms. With varied nitrogen content from 0% to 2%, the NO2 conversion rate increases slightly from 38.65% to 39.81%.
80
3.6
2.4 2.0 1.6 1.2 0.8
NO2 Conversion Rate Converted NO2
38
70
36
60
34 50 32 40
30 28
Converted NO2 (ppm)
2.8
NO2 conversion rate (%)
Ar N2/Ar O2/Ar
3.2
NO2 energy yield (g/kWh)
40
30
0.4 26
0.0
0
5
10
15
20
25
30
35
40
45
50
55
Specific energy density (J/L) Fig. 6 – Effect of specific energy density on NO2 energy yield.
50
100
150
200
250
20 300
Initial NO2 concentration (ppm) Fig. 7 – Effect of initial concentration on NO2 conversion rate.
2254
J O U R NA L OF EN V I RO N M EN T A L S CI EN C ES 2 6 ( 20 1 4 ) 2 2 49–2 2 5 6
b
NO NO2 O2
80
0.20
60 0.15
40
NO2 conversion (%)
50 100
20 0
200
48
190 NOx (NO+NO2) NO2 conversion
46
180
44 170
42
160
40 0
1
2
3
4
5
6
7
8
0.10
210
38
0
1
2
3
N2 content (%)
4
5
6
7
8
NOx concentration (ppm)
52
O2 content (%)
NOx concentration (ppm)
0.25
a
120
150
N2 content (%)
Fig. 8 – Effect of N2 content on NOx conversion. (a) NO, NO2 and O2 concentrations; (b) NO2 conversion and total converted NOx concentration.
NO NO2 O3
120 100
1.0
80
0.5
60 40 0
Fig. 9 shows the variation of NO, NO2 and O3 with O2 content in O2/Ar plasma. The gas composition was NO2 210 ppm, and the content of O2 was 0% to 8% with argon balance. The SED was the same as the experiments in N2/Ar. The results show that increasing the oxygen content gradually increased the concentration of NO2 from 127.1 to 170.3 ppm, while the NO concentration decreased accordingly from 77 to 29 ppm (Fig. 9a). These results indicated that the addition of oxygen suppressed the conversion of NO2 to NO with increasing oxygen content. In O2/Ar atmosphere, the oxidation pathways, i.e., ozoneand O radical-containing reactions, dominated the process of NO2 conversion (Yoshida et al., 2009a). O radicals can be generated via both electron-induced oxygen dissociation (R2) and quenching of metastable Ar by O2 molecules. The dissociation of oxygen and quenching of metastable Ar produced ground state O (3P) and excited O (1D), which
1.5
a
140
20
2.4.2. NO2 conversion in O2/Ar
1
2
3
4
5
O2 content (%)
6
7
8
0.0
45
b
220
40
NO2 conversion (%)
160
O3 concentration (ppm)
NOx concentration (ppm)
180
decreases with increasing N2 content. NOx can be reduced to N2O or N2 via R5 or R8, as proven by other researchers (Okumura et al., 1994). The presence of slight amounts of oxygen may be attributed to R4 or recombination of O radicals generated from NO reduction in R5, which was confirmed by Penetrante et al. (1995) and Zhu et al. (2005).
200 180
35
160
NO2 conversion NOx
30
140 120
25
100 80
20 15
NOx concentration (ppm)
When the nitrogen content was between 2% and 3%, the conversion rate rises remarkably from 39.81% to 49.71%. When the nitrogen content exceeds 3%, the conversion rate increases slightly and reaches 50.33% at 8% nitrogen content. In N2/Ar atmosphere, N radicals and metastable nitrogen were generated during the discharge via electron-induced reactions (R1a, R1b) and charge transfer reactions. These two kinds of active species may induce conversion reactions upon collision with pollutants (Aerts et al., 2013). Due to the existence of metastable Ar(3P2), the excitation of ground state N2 molecules by Ar(3P2) + N2 → Ar + N2(A) cannot be negligible (Masoud et al., 2005). Thus NO2 conversion was dominated by conventional reduction reactions with N radicals and metastable N2(A) via R5, R6 and R7 (Jõgi et al., 2009). With increasing nitrogen content, electron excitation and dissociation of nitrogen became more and more important, which led to a boost of N radical generation. These radicals accelerated the aforementioned NO2 reduction reactions and finally reached a higher conversion rate. When N2 exceeded 3%, the NO2 conversion rate was saturated. Since N2 possessed a larger cross section than Ar did, the generated N radicals could be consumed via recombination reaction N + N + M → N2 + M (M_N2, Ar). This reaction suppressed further NO2 conversion in N2/Ar atmosphere at higher N2 content. In addition, the total concentration of NOx (Fig. 8b)
60 0
1
2
3
4
5
6
7
8
O2 content (%)
Fig. 9 – Effect of O2 content on NOx conversion. (a) NO, NO2 and O3 concentrations; (b) NO2 conversion and total NOx concentration.
JO U R N A L OF EN V I RO N M EN T A L S CI E NC ES 2 6 ( 20 1 4 ) 2 2 4 9–2 2 5 6
oxidized NO to NO2. These O radicals can also react with O2 via three-body reaction and form O3 via R3. These O radicals and ozone can both contribute to NO oxidation, inhibiting further NO2 conversion. As mentioned before, O can also reduce NO2 to NO, however the reaction rate constant of NO oxidation with O radicals in an oxygen-rich environment is 3.00 × 10−11 cm−3/(mol·s), which is much larger than that of reduction reaction R4. Thus with increasing oxygen content, the net converted NO2 decreased, resulting in NO2 conversion rate decrease, as shown in Fig. 9b. Since no N2O and N2 were formed in the oxygen-rich plasma, the change in total NOx conversion can be regarded as negligible. In the presence of NO2, the detected ozone concentration was below 1 ppm. In order to prove the role of ozone in the NO2 removal process, the ozone concentration was monitored at the reactor outlet, as shown in Fig. 10. The variation of oxygen content from 0% to 8% resulted in ozone formation from 0 to 31.3 ppm. Compared to the concentration of NOx in the NO2/O2/Ar case, the ozone concentration was one order of magnitude lower, which suggests that ozone was depleted via NO oxidation and NO2 reduction reactions in the presence of NOx (Yan et al., 1999).
2255
O2 inhibited the conversion process. The highest energy yield of 3.31 g/kWh was observed in N2/Ar and decreased with higher SED; the same trends were observed in other two gas compositions. (3) The presence of N2 and O2 led to different reaction pathways for NO2 conversion due to the formation of different reactive radicals respectively. In the case of the N2/Ar gas mixture, N radicals and metastable N2(A) played crucial roles in the NO2 conversion. The conversion rate increased dramatically in the N2 content range of 2%–3% and it reached equilibrium when N2 content exceeded 3%. In the O2/Ar case, O radicals and O3 played an important role via oxidation pathways, and the conversion rate decreased monotonically over tested O2 concentrations. For higher conversion efficiency under the given conditions, a wet scrubbing unit should be employed for NO2 absorption.
Acknowledgments This work was supported by the National Science Fund for Distinguished Young Scholars (No. 51125025) and the National Natural Science Foundation of China (Nos. 51076140, 51206143).
3. Conclusions The discharge characteristics and NO2 conversion performance within a pulsed corona discharge reactor in simulated gas compositions of Ar, N2/Ar and O2/Ar were investigated experimentally. The results were summarized as follows. (1) The specific energy density (SED) increased from 1.16 to 38.03 J/L, from 2.39 to 49.03 J/L and from 2.46 to 82.56 J/L for N2/Ar, Ar and O2/Ar, respectively, with increasing pulse frequency from 10 to 300 Hz. For a given pulse frequency, SED had the order O2/Ar > Ar > N2/Ar. (2) Higher SED led to higher NO2 conversion rate: the conversion efficiency increased from 2.8% to 33.7%, 5.1% to 42.1%, and 4.27% to 25.94% for N2/Ar, Ar and O2/Ar. The conversion rate was also affected by gas composition: the addition of N2 enhanced NO2 conversion while
35 O3 concentration
O3 concentration (ppm)
30 25 20 15 10 5 0 0
1
2
3
4
5
6
7
8
O2 content (%) Fig. 10 – Ozone formation without initial NO2 in O2/Ar mixture.
REFERENCES Aerts, R., Tu, X., De Bie, C., Whitehead, J.C., Bogaerts, A., 2012. An investigation into the dominant reactions for ethylene destruction in non-thermal atmospheric plasmas. Plasma Process. Polym. 9 (10), 994–1000. Aerts, R., Tu, X., Van Gaens, W., Whitehead, J.C., Bogaerts, A., 2013. Gas purification by nonthermal plasma: a case study of ethylene. Environ. Sci. Technol. 47 (12), 6478–6485. Arai, M., Saito, M., Yoshinaga, S., 2004. Effect of oxygen on NOx removal in corona discharge field: NOx behavior without a reducing agent. Combust. Sci. Technol. 176 (10), 1653–1665. Cha, M.S., Song, Y.H., Lee, J.O., Kim, S.J., 2007. NOx and soot reduction using dielectric barrier discharge and NH3 selective catalytic reduction in diesel exhaust. Int. J. Plasma Environ. Sci. Technol. 1 (1), 28–33. Eichwald, O., Guntoro, N.A., Yousfi, M., Benhenni, M., 2002. Chemical kinetics with electrical and gas dynamics modelization for NOx removal in an air corona discharge. J. Phys. D. Appl. Phys. 35 (5), 439–450. Ferreira, C.M., Tatarova, E., Guerra, V., Gordiets, B.F., Henriques, J., Dias, F.M., et al., 2003. Modeling of wave driven molecular (H2, N2, N2–Ar) discharges as atomic sources. IEEE Trans. Plasma Sci. 31 (4), 645–658. Gallon, H.J., Tu, X., Twigg, M.V., Whitehead, J.C., 2011. Plasma-assisted methane reduction of a NiO catalyst—low temperature activation of methane and formation of carbon nanofibres. Appl. Catal. B Environ. 106 (3–4), 616–620. Herron, J.T., 1999. Evaluated chemical kinetics data for reactions of N(2D), N(2P), and N2(A 3Σu +) in the gas phase. J. Phys. Chem. Ref. Data 28 (5), 1453–1483. Hu, X., Nicholas, J., Zhang, J., Temi, M.L., Filippis, P.D., Agarwal, P.K., 2002. The destruction of N2O in a pulsed corona discharge reactor. Fuel 81, 1259–1268. Hu, X., Zhao, G.B., Zhang, J.J., Wang, L., Radosz, M., 2004. Nonthermal-plasma reactions of dilute nitrogen oxide mixtures: NOx-in-argon and NOx + CO-in-argon. Ind. Eng. Chem. Res. 43 (23), 7456–7464.
2256
J O U R NA L OF EN V I RO N M EN T A L S CI EN C ES 2 6 ( 20 1 4 ) 2 2 49–2 2 5 6
Jõgi, I., Bichevin, V., Laan, M., Haljaste, A., Käämbre, H., 2009. NO conversion by dielectric barrier discharge and TiO2 catalyst: effect of oxygen. Plasma Chem. Plasma Process. 29 (3), 205–215. Kutasi, K., 2011. Modelling of NO destruction in a low-pressure reactor by an Ar plasma jet: species abundances in the reactor. J. Phys. D. Appl. Phys. 44 (10), 105202. Lee, Y.H., Jung, W.S., Choi, Y.R., Oh, J.S., Jang, S.D., Son, Y.G., et al., 2003. Application of pulsed corona induced plasma chemical process to an industrial incinerator. Environ. Sci. Technol. 37 (11), 2563–2567. Lin, H., Guan, B., Cheng, Q., Huang, Z., 2010. An investigation on the principal paths to plasma oxidation of propylene and no. Energy Fuel 24 (10), 5418–5425. Masoud, N., Martus, K., Becker, K., 2005. VUV emission from a cylindrical dielectric barrier discharge in Ar and in Ar/N2 and Ar/air mixtures. J. Phys. D. Appl. Phys. 38 (11), 1674–1683. Masuda, S., Nakao, H., 1990. Control of NOx by positive and negative pulsed corona discharges. IEEE Trans. Ind. Appl. 26 (2), 374–383. Matsumoto, T., Wang, D.Y., Namihira, T., Akiyama, H., 2010. Energy efficiency improvement of nitric oxide treatment using nanosecond pulsed discharge. IEEE Trans. Plasma Sci. 38 (10), 2639–2643. Mizuno, A., Shimizu, R., Chakrabarti, A., Dascalescu, L., Furuta, S., 1995. NOx removal process using pulsed discharge plasma. IEEE Trans. Ind. Appl. 31 (5), 957–962. Mok, Y.S., Ham, S.W., 1997. Conversion of NO to NO2 in air by a pulsed corona discharge process. Chem. Eng. Sci. 53 (9), 1667–1678. Mok, Y.S., Ham, S.W., Nam, I.S., 1998. Evaluation of energy utilization efficiencies for SO2 and NO removal by pulsed corona discharge process. Plasma Chem. Plasma Process. 18 (4), 535–550. Mok, Y.S., Kim, J.H., Nam, I.S., Ham, S.W., 2000. Removal of NO and formation of byproducts in a positive-pulsed corona discharge reactor. Ind. Eng. Chem. Res. 39, 3938–3944. Obradovic, B.M., Sretenovic, G.B., Kuraica, M.M., 2011. A dual-use of DBD plasma for simultaneous NOx and SO2 removal from coal-combustion flue gas. J. Hazard. Mater. 185 (2–3), 1280–1286. Okumura, T., Matsuda, S., Satoh, Y., Sakai, Y., Tagashira, H., 1994. Decomposition of NO2 in a glow discharge and electron impact ionization and attachment coefficients in NO2/air mixtures. J. Phys. D. Appl. Phys. 27 (4), 801–806. Orlandini, I., Riedel, U., 2000. Chemical kinetics of NO removal by pulsed corona discharges. J. Phys. D. Appl. Phys. 33, 2467–2474. Penetrante, B.M., Hsiao, M.C., Merritt, B.T., Vogtlin, G.E., Wallman, P.H., 1995. Comparison of electrical discharge techniques for nonthermal plasma processing of NO in N2. IEEE Trans. Plasma Sci. 23 (4), 679–687. Penetrante, B.M., Hsiao, M.C., Merritt, B.T., Vogtlin, G.E., 1997. Fundamental limits on NOx reduction by plasma. SAE Technical Paper Series, 971715, p. 971715.
Skalska, K., Miller, J.S., Ledakowicz, S., 2010. Trends in NOx abatement: a review. Sci. Total Environ. 408 (19), 3976–3989. Sun, B., Yin, S., 2009. The characteristics of NO reduction in the reactor with dielectric barrier discharge. ASME 2009 3rd International Conference of Energy Sustainability. San Francisco, USA, pp. 19–23 (July). Tatarova, E., Guerra, V., Henriques, J., Ferreira, C.M., 2007. Nitrogen dissociation in low-pressure microwave plasmas. J. Phys. Conf. Ser. 71, 012010. Tochikubo, F., Uchida, S., Yasui, H., Sato, K., 2009. Numerical simulation of NO oxidation in dielectric barrier discharge with microdischarge formation. Jpn. J. Appl. Phys. 48 (7), 076507. Tsuji, M., Nakano, K., Kumagae, J., Tsuji, T., 2002. Decomposition of NO by microwave discharge of NO/He or NO/Ar mixtures. Bull. Chem. Soc. Jpn. 75, 607–614. Tu, X., Whitehead, J.C., 2012. Plasma-catalytic dry reforming of methane in an atmospheric dielectric barrier discharge: understanding the synergistic effect at low temperature. Appl. Catal. B Environ. 125, 439–448. Tu, X., Gallon, H.J., Whitehead, J.C., 2011. Electrical and spectroscopic diagnostics of a single-stage plasma-catalysis system: effect of packing with TiO2. J. Phys. D. Appl. Phys. 44 (48), 482003. Van Durme, J., Dewulf, J., Leys, C., Van Langenhove, H., 2008. Combining non-thermal plasma with heterogeneous catalysis in waste gas treatment: a review. Appl. Catal. B Environ. 78 (3–4), 324–333. Wu, Z.L., Gao, X., Luo, Z.Y., Wei, E.Z., Zhang, Y.S., Zhang, J.Z., et al., 2005. NOx treatment by DC corona radical shower with different geometric nozzle electrodes. Energy Fuel 19 (6), 2279–2286. Yan, K., Kanazawa, S., Ohkubo, T., Nomoto, Y., 1999. Oxidation and reduction processes during NOx removal with corona-induced nonthermal plasma. Plasma Chem. Plasma Process. 19 (3), 421–443. Yoshida, K., Rajanikanth, B.S., Okubo, M., 2009a. NOx reduction and desorption studies under electric discharge plasma using a simulated gas mixture: a case study on the effect of corona electrodes. Plasma Sci. Technol. 11 (3), 327–333. Yoshida, K., Yamamoto, T., Kuroki, T., Okubo, M., 2009b. Pilot-scale experiment for simultaneous dioxin and NOx removal from garbage incinerator emissions using the pulse corona induced plasma chemical process. Plasma Chem. Plasma Process. 29 (5), 373–386. Yu, L., Li, X., Tu, X., Wang, Y., Lu, S., Yan, J., 2010a. Decomposition of naphthalene by dc gliding arc gas discharge. J. Phys. Chem. A 114 (1), 360–368. Yu, L., Tu, X., Li, X., Wang, Y., Chi, Y., Yan, J., 2010b. Destruction of acenaphthene, fluorene, anthracene and pyrene by a DC gliding arc plasma reactor. J. Hazard. Mater. 180 (1–3), 449–455. Zhu, A.M., Sun, Q., Niu, J.H., Xu, Y., Song, Z.M., 2005. Conversion of NO in NO/N2, NO/O2/N2, NO/C2H4/N2 and NO/C2H4/O2/N2 systems by dielectric barrier discharge plasmas. Plasma Chem. Plasma Process. 25 (4), 371–386.
