The byproduct generation analysis of the NOx conversion process in dielectric barrier discharge plasma

Yajie Zhanga, Xiaolong Tang*ab, Honghong Yiab, Qingjun Yua, Jiangen Wanga, Fengyu Gaoa, Yueming Gaoa, Dianze Lia and Yumeng Caoa
aCollege of Civil and Environmental Engineering, University of Science and Technology Beijing, Beijing, 100083, China. E-mail: txiaolong@126.com; Fax: +86 010 62332747; Tel: +86 010 62332747
bBeijing Key Laboratory of Resource-Oriented Treatment of Industrial Pollutants, Beijing, China

Received 2nd April 2016 , Accepted 24th June 2016

First published on 28th June 2016


Abatement of NOx through non-thermal plasma (NTP) processes has been developed over the past several years. Since discharge plasma contains a large amount of highly active species during the reaction process, NOx is removed during our desired conversion reaction, as well as production of other byproducts, such as O3 and N2O. The effect of reaction conditions, such as oxygen content, discharge power, operation time, initial NO concentration and gas residence time, on generation characteristics of O3 and N2O was investigated. Results showed that, with increasing oxygen concentration and discharge power, the production of N2O and O3 increases. Additionally, O3 concentration decreases with increasing operation time; the higher input power, the higher the temperature increases, causing a greater reduction rate of O3, which also leads to a reduction of NO2 production. NO concentration and gas residence time also exert effects on the generation of byproducts.


1. Introduction

The control of nitrogen oxide (NOx, x = 1, 2) emissions, mainly from stationary combustion and mobile sources, has become an increasingly important issue since it is considered to be one of the major air pollutants due to its serious damage to the environment and human health.1 In China, the amount of NOx in the form of NO emitted from stationary sources accounts for more than 90% of the total NOx emissions.2 In recent years, non-thermal plasma (NTP) technology in the treatment of gaseous pollutants, especially in the research and application of NOx removal, has been widely investigated.3–5 Numerous studies have been focused on seeking new methods to improve NOx removal efficiency and solve the problem of high energy consumption at the same time, such as the plasma-assisted selective catalytic reduction of NO,6–8 plasma derived catalytic oxidation process,9 NO decomposition with NTP10 and adsorption-discharge plasma catalytic method.11

The collisions of high-energy free electrons produced by electrical discharge with gas molecules produce chemically active species (such as radicals, ions and neutrals), which contribute to NOx formation and conversion.12 Our intention of using non-thermal plasma is to selectively transfer input electrical energy to the electrons which generates free radicals, and promote the desired chemical changes.13 However, due to the presence of large amounts of highly active species in the plasma, NOx abatement occurs in addition to the desired products during the conversion process, as well as the production of other undesired byproducts, such as N2O and O3, which influences the conversion rate of NOx to some extent. In addition, partial input electrical energy can also expend in heating the gas stream during reaction process at the same time, which also influences the removal process and generation of products.

With the increasingly stringent emission standards, it is necessary to study the formation of byproducts during the treatment process of NOx. Most of the time, the outlet concentration of NO has gained more concern than byproducts because, as reported,14,15 “NOx removal is generally a problem of NO removal”, “the generation of byproduct N2O may be negligible”. Thus, it is easy to ignore the impact of formation and conversion of byproducts during the NOx removal process. Actually, however, A. Gal et al.16,17 had proposed the byproduct-suppression effect a dozen years ago. They also claimed that it is urgently necessary to find a method for suppressing byproduct generation. The generation of N2O influences the removal efficiency of NOx, as founded by G. B. Zhao et al.18 They also stated that the formation and conversion of byproducts were rarely reported and explained, especially for N2O.19 As a strong oxidizer, the ozone causes formation of NOx (such as NO2, NO3, etc.),20 which also influences the removal of NOx. As presented by Mok et al.,21,22 O3 was found to be the most important one in the oxidation of NO. Jõgi I. and Li et al.23,24 also investigated the role of the ozone during NOx abatement process. However, the majority of studies have been focused on using a computational model and reaction kinetics of the reaction process to analyze the role of the ozone in the NOx conversion process.25,26 Therefore, for further understanding about the conversion mechanism of NO in discharge plasma, intensive investigations are required on the formation and conversion characteristics of N2O and O3.

In the present experiment, the amount of O3 and N2O, and temperature of the inner electrode/outer wall (reaction zone) were measured. Furthermore, the impact on generation of byproducts by reaction conditions (discharge power, oxygen content, discharge time, initial NO concentration and gas residence time) was also investigated using DBD plasma alone in the NO/N2, NO/O2/N2 system.

2. Experimental

The experimental setup is shown in Fig. 1. Experimental gas was prepared by mixing gas of high purity N2, O2 and NO/N2. Non-thermal plasma was obtained in a conventional coaxial cylindrical reactor consisting of quartz tube (inner diameter 21 mm, outer diameter 24 mm and length 370 mm) as a dielectric barrier between the inner high voltage electrode (stainless-steel-screw rod, 4.5 mm) and a grounded electrode (copper wire mesh) on the outer wall. The DBD reactor’s power source is a CTP-2000P instrument made by Nanjing Suman Electronics co., Ltd, China. Discharge power was measured by the V-Q Lissajous program. Ozone concentration in the gas stream was measured with an ozone analyzer based on the UV absorption method (2B Technologies, Model 106-M), with the uncertainty of ±0.01 ppm (10 ppb). The concentration of NO and NO2 were detected by a fuel gas analyzer (Kane, KM9106), with the uncertainty of ±1 ppm. The N2O concentration was analyzed online using a mass spectrometer (Extrel CMS, LLC, MAX300-LG), with an uncertainty of ±0.01 ppm. The temperature was measured by an IR thermometer (Fluke Corp., 566), with an uncertainty of ±0.1 °C. All measurements were repeated three times. The NOx conversion was defined as the fraction of NO converted to N2, expressed as follows:
image file: c6ra08488e-t1.tif
where Cin,NO is the concentration of NO in the inlet of the reactor and Cout,NO, Cout,NO2, Cout,N2O are the concentration of NO, NO2 and N2O in the outlet of the reactor, respectively.

image file: c6ra08488e-f1.tif
Fig. 1 Flow chart of the experimental system.

3. Results and discussion

3.1 Effects of discharge power and oxygen concentration

Fig. 2 shows the conversion rate of NOx, evolution of N2O, O3, and NO2 as functions of discharge power and oxygen concentration in the NO/N2 and NO/O2/N2 system, respectively. The mixture contained 530 ppm NO. The oxygen concentration was 0, 1.5%, 3%, 5% and 10% respectively. The gas flow rate was set as 1 L min−1. The major reactions and corresponding rate constants are listed in Table 1. As shown in Fig. 2(a), in a mixture absent of oxygen, the conversion rate of NOx initially increases quickly with the enhancement of the discharge power, and then increases slowly until reaching at 97% or so eventually. As shown in Fig. 2(b), with the augment of the discharge power, the N2O concentration increases to a maximum, and then reduces to a constant value. In Fig. 2(c) and (d), it is observed that the ozone concentration also increases with increasing discharge power, while only little NO2 was detected. In the gas stream of NO/N2, the formation of ozone is due to most of the NO decomposing into O2 and O, then the ozone is generated by reaction (1) (Table 1).
image file: c6ra08488e-f2.tif
Fig. 2 (a) Effect of discharge power on the conversion of NOx; (b) effect of discharge power on the concentration of N2O; (c) effect of discharge power on the concentration of O3; (d) effect of discharge power on the concentration of NO2 (where a% represents a mixture containing a% volume of oxygen).
Table 1 Reactions and corresponding rate constants
Reactions Rate constants/(cm3 s−1) Ref.
O + O2 → O3 (R1) 5.6 × 10−34 (300/T)2.23 31
O + O3 → 2O2 (R2) 8.0 × 10−12[thin space (1/6-em)]exp(−2060/T) 31
e* + N2 → N* + N + e (R3)   28
N + O2 → NO + O (R4)   19
NO + O3 → NO2 + O2 (R5) 2.3 × 10−12[thin space (1/6-em)]exp(−1450/T) 32
NO + O + M → NO2 + M (R6) 5.0 × 10−33[thin space (1/6-em)]exp(9000/T) [M] 32
NO2 + O → NO + O2 (R7)   32


In the mixture containing oxygen, the conversion and formation of products becomes rather complicated. When O2 concentration is low (for example ≤3%), the NOx conversion rate still rises with increasing discharge power, but more slowly than that of a mixture without oxygen. With higher oxygen concentration, the conversion rate of NOx becomes much lower at a given discharge condition, and it even decreases to a negative value with the enhancement of input energy. The negative removal rate of NOx constantly decreases with the growing oxygen concentration and input power. As shown in Fig. 2(c) and (d), the concentration of N2O and O3 continuously increases with the augment of discharge power. Under a given discharge condition, the yield of N2O and O3 increases with increasing oxygen concentration. As shown in Fig. 2(d), when the O2 content is ≤5%, the NO2 concentration increases with the improvement of discharge power, and then decreases to a constant value. When O2 concentration reaches 10%, NO2 production continuously increases by enhancing the input power.

The reason why the NOx conversion rate reduces and more byproducts are generated when oxygen is present will be discussed as follows. Firstly, with the augment of the discharge power, discharge intensity increases as well, which contributes to the occurrence of reactions (3) and (4) (Table 1). Consequently, the formation of NOx is exacerbated. Secondly, since the dissociation energy of O2 (5.2 eV per molecule) is lower than that of N2 (9.8 eV per molecule),27 thus it is more readily available to interact with O2 for electrons to produce strongly oxidizing species (O and O3) at given discharge condition. These oxidizing species promote the oxidation reaction of NO to NO2.20 Additionally, oxygen is electronegative and thus reduces the discharge current caused by the electron attachment process.27 Then the energy input to the reactor is reduced by the reduction of current. As a result, the generation rate of N radicals is lowered by the presence of oxygen at a given discharge condition, which causes a decrease in NO removal efficiency.19

In the present experiment, the yield of O3 in the outlet was detected at a level of several hundreds ppb. This may due to the violent electronic collision reaction causing large numbers of O3 reduced in a short time. Dorai R. et al.28 computationally investigated the DBD processing of simulated diesel exhaust. Results showed that, a large number of O3 was produced within several μs time after the injection of discharge energy, but the amount of O3 quickly decreased by reaction (5) in several ms. G. B. Zhao et al.19 calculated the O3 conversion by the rate constant of reaction (4), they found O3 conversion, as a function of gas residence time, can be completely converted in 0.1 s of residence time. Their simulation results showed that the ozone concentration is <1 ppm and NO3 concentration is <0.001 ppm for all experiments. Since N2O5 is easily decomposed to NO2, NO or N2 in discharge plasma, we didn’t detect N2O5 (or other high oxidizing states) in our experiment.

3.2 Effect of discharge time

Fig. 3 shows the production of O3/NO2/N2O influenced by discharge time. The first mixture contained 530 ppm of NO, and the oxygen concentration was 0, 1.5%, 3%, 5% and 10%, respectively (Fig. 3(a) and (b)). The second mixture contained 530 ppm of NO and 3% oxygen (Fig. 3(c)). As we can see in Fig. 3(a) and (c), they both have a rapid increase of ozone production within several minutes, and then decrease drastically with increasing discharge time, forming a peak value. The reduction rate of ozone yield is enhanced by the growing O2 concentration (Fig. 3(a)) as well as the enhancement of discharge power (Fig. 3(c)). The amount of ozone is influenced by the discharge intensity, discharge time and oxygen concentration at the same time. Fig. 3(b) shows that the NO2 concentration increases to a maximum with the increasing time, and then decreases down to a constant value. With the increase of oxygen concentration, the NO2 concentration increases under the same discharge condition. In addition, when oxygen concentration is high enough (such as 10%), there is no longer a maximum of the amount of NO2. That is, NO2 concentration almost rises to a constant value with the increase of operation time. In the first reaction system (Fig. 3(a) and (b)), with increasing operation time, the amount of ozone and NO2 had similar trend. As shown in Fig. 3(d), the mean temperature of the inner electrode/outer wall (reaction zone) rises with the increasing operation time and input power. The temperature increase of the inner electrode is slightly higher than that of the outer wall.
image file: c6ra08488e-f3.tif
Fig. 3 (a) Effect of discharge time on the production of O3; (b) effect of discharge time on the production of NO2; (c) effect of discharge time on the production of O3; (d) effect of discharge time on the mean temperature of the inner/outer electrode (where a% represents a mixture containing a% volume of oxygen, solid shape (■) represents inner electrode and the hollow shape (□) represents the outer wall of the reaction zone).

The current flow induced by the migration of charged species (electrons and ions) induces a temperature increase in the reactor through joule heating, dielectric loss, and gas heating in the plasma channel.29 The oxidizing species, O3 and O, formed by the violent collision of energetic electrons, promote the generation of NO2 (reactions (5) and (6)). According to Mok et al.,21,30–32 the ozone is believed to dominantly affect the oxidation of NO. That’s because the lifetime of the O radical is very short and the rate constant of the reaction (6) is small. Stamate E. et al.33 also found that after ozone injection, the concentration of NO immediately decreases and the concentration of NO2 increases by reaction (5) accordingly. However, the resonance structure of the ozone determines that it is very unstable and easily decreases with a temperature rise during the reaction process.14,34 In the reaction zone, the greater the power input is, the more heat will be produced, resulting in a reduction of the ozone with increasing operation time.

With regards to the mixture with a low O2 concentration, the reduction rate of the ozone concentration is much greater than that of the mixture with high O2 concentration with increasing discharge time. When O2 concentration increases, more O2 can be excited to generate more O3 and O radicals, providing a greater oxidizing source for the reaction system, allowing the oxidation atmosphere (formed by O3 and O) to be maintained for longer. It is observed that both the amounts of the ozone and NO2 have a maximum, and then their values both decreases with the increasing operation time. As Jõgi I. investigated,23 the production of NO2 is characterized by a linear relation with the inlet ozone concentration. With increasing temperature, the ozone concentration reduces, which limits the generation rate of NO2 according to reaction (5).

When the oxygen concentration is the same in a different gas stream, the production of the ozone in a different reaction system (Fig. 3(c)) is mainly influenced by discharge intensity and the temperature increase in the reactor. With the violent electronic collision reaction in the reactor, a great number of highly active free radicals are formed in the reaction zone. The higher the discharge power, the more highly active free radicals will be generated, and the more chance for the side reactions to be excited. Since the temperature increase depends primarily on the energy input in the reactor, the temperature rises by the improvement of the discharge power.35,36 Consequently, the reduction rate of the ozone increases with increasing input power.

Before sharply reduced by the temperature increase, the ozone has generated a great amount in the reaction system, thus forming a highly oxidizing atmosphere at the reaction zone in the initial stage. During this period, the reaction zone can provide a large amount of oxidizing species (O and O3), which contributes to the formation of NO2. Therefore, making full use of the highly oxidizing atmosphere at the initial stage during the reaction process can effectively contribute to the oxidation from NO to NO2.

3.3 Effect of the initial NO concentration

The generation of O3 and N2O influenced by the initial NO concentration is plotted in Fig. 4. As shown in Fig. 4(a), in the absence of oxygen, the NOx conversion rate decreases with increasing NO concentration in the gas stream. The conversion rate has a rapid decrease when the concentration of NO is 850 ppm. The yield of N2O rises to a maximum and then decreases to a constant value with the growing discharge power as shown in Fig. 4(b). Meanwhile, the O3 concentration follows a similar trend (Fig. 4(c)). At a given discharge power, production of O3 increases with the growing NO concentration. When oxygen is present (3%), in the mixed gas with a low initial NO concentration, the conversion rate of NOx is slightly higher than that of the mixed gas containing a high NO concentration, but the difference isn’t notable. At a given discharge condition, the N2O concentration increases with the increase of NO concentration as shown in Fig. 4(b). However, the ozone yield follows the opposite trend (Fig. 4(c)).
image file: c6ra08488e-f4.tif
Fig. 4 (a) Effect of NO concentration on the conversion of NOx; (b) effect of NO concentration on the evolution of N2O; (c) effect of NO concentration on the evolution of O3; where ■ (solid) and □ (hollow) respectively represent the mixture without/with oxygen.

When injecting more NO to the gas stream in the system of NO/N2, increasing NO concentration increases the amount of O generated by NO molecules. Correspondingly, the amount of ozone has risen according to reaction (1). As a result, it promotes the generation of NxOy (x = 1, 2; y = 1, 2) as we discussed in Section 3.1. Likewise, when the amount of NO injected in the NO/O2/N2 gas stream increases, the production of N2O and NOx removal rate also has risen and decreased respectively. However, the O3 yield decreases with growing NO concentration (under higher discharge power (>18 W)). As shown in Fig. 4(c), the yield of O3 in the NO/O2/N2 system initially has little change with the increase of the NO concentration, while the change of O3 yield in the NO/N2 system is more obvious: it rises with the increasing NO concentration. One reason for this may be related to the uncertainties in our measurements. When O2 is present, additionally, there exists a competition between NOx (NO, NO2) and O2 for O by reactions (6) and (7).37 At a higher discharge power (>18 W), the O3 yield of the NO/O2/N2 system decreases with the increase of NO concentration as shown. In the gas stream of NO/N2, almost all NO was decomposed. When O2 is present, as discussed above, it decreases the NO reduction rate. Furthermore, more NOx will be produced. As a result, the competition between NOx (NO, NO2) and O2 for O in the gas stream of NO/O2/N2 is more notable than that of the NO/N2 system which provides fewer NOx for it. In addition, the consumption on O3 by reaction (5) is another reason for the reduction of O3.

3.4 Effect of gas residence time

Fig. 5 shows the influence of gas residence time on the yield of O3, NO2 and N2O in the NO/O2/N2 system under the same discharge condition. The mixture gas contained 3% by volume of oxygen and 530 ppm of NO. Since we had discussed the dependence of ozone generation with the discharge intensity and operation time above, to avoid the thermal effect resulting from the discharge process on products, we investigated ozone production only in the initial stage of the reaction. As shown in Fig. 5(a), with the increase of gas residence time, both O3 and NO2 have a tendency to rise. Meanwhile the concentration of N2O has changed little with the increasing gas residence time as shown in Fig. 5(b). Under same reaction condition, the gas exposure to plasma increases by increasing the residence time. Then the probability of the electron impact reactions and secondary reactions for emission reduction increases.38 Moreover, the electrons will obtain more energy through a larger acceleration time in the reaction zone.39,40 The increase of residence time of the gas in the plasma region was found to be beneficial for removing pollutants. However, more O is generated to promote higher ozone production, and also causes a rise of NO2 concentration by reactions (5) and (6).
image file: c6ra08488e-f5.tif
Fig. 5 (a) Effect of gas residence time on the evolution of O3, NO2; (b) effect of gas residence time on the evolution on N2O.

4. Conclusions

In the present work, the influence on the conversion rate of NOx and generation characteristics of products by reaction conditions (discharge power, oxygen content, discharge time, initial NO concentration and gas residence time) were investigated. Results indicate that: (a) with higher oxygen concentration and discharge power injected into the reaction zone, the production of N2O and O3 increases, owing to the short residence time, while a large amount of O3 was not detected; (b) with the increase of operation time, a temperature increase in the reaction zone induced by the migration of charged species reduces the yield of O3, and causes a reduction of NO2 as well; a higher input power causes a higher temperature increase within the same operation time; (c) when injecting more NO to the gas stream in the system of NO/N2, the increasing amount of O decomposed by NO promotes generation of O3; when injecting more NO in the NO/O2/N2 mixture, the yield of O3 is reduced mainly by the competition for O between NOx (NO/NO2) and O2 and the consumption of O3 by the reaction NO + O3 → NO2 + O2; (d) the chance for gas stream exposure to plasma increases, and more violent reactions induced by longer gas residence time in the reaction zone benefits NOx removal, while the production of byproducts also rises.

Acknowledgements

This work was financially supported by National Natural Science Foundation of China (20907018, 21177051, 21507004), Program for New Century Excellent Talents in University (NECT-13-0667), the Fundamental Research Funds for the Central Universities (FRF-TP-14-007C1, FRF-TP-15-046A1), the special project on air pollution control of Beijing Municipal Science & Technology Commission (Z141100001014006), and China Postdoctoral Science Foundation (2015M580996).

References

  1. S. X. Wang and J. M. Hao, J. Environ. Sci., 2012, 24, 2 CrossRef CAS.
  2. J. M. Hao, H. Z. Tian and Y. Q. Lu, Environ. Sci. Technol., 2002, 36, 552 CrossRef CAS PubMed.
  3. H. H. Kim and A. Ogata, Eur. Phys. J.: Appl. Phys., 2011, 55, 13806 CrossRef.
  4. H. T. Q. An, T. P. Huu, T. Le Van, J. M. Cormier and A. Khacef, Catal. Today, 2011, 176, 474 CrossRef.
  5. A. M. Vandenbroucke, R. Morent, N. De Geyter and C. Leys, J. Hazard. Mater., 2011, 195, 30 CrossRef CAS PubMed.
  6. B. K. Cho, J. H. Lee, C. C. Crellin, K. L. Olson, D. L. Hilden, M. K. Kim, P. S. Kim, I. Heo, S. H. Oh and I. S. Nam, Catal. Today, 2012, 191, 20 CrossRef CAS.
  7. C. E. Stere, W. Adress, R. Burch, S. Chansai, A. Goguet, W. G. Graham, F. D. Rosa, V. Plama and C. Hardacre, ACS Catal., 2014, 4, 666 CrossRef CAS.
  8. S. Bröer and T. Hammer, Appl. Catal., B, 2000, 28, 101 CrossRef.
  9. J. P. Sousa, M. F. Pereira and J. L. Figueiredo, Catal. Today, 2011, 176, 383 CrossRef CAS.
  10. Y. Takahara, A. Ikeda, M. Nagata and Y. Sekine, Catal. Today, 2013, 211, 44 CrossRef CAS.
  11. Q. Q. Yu, H. Wang, T. Liu, L. P. Xiao, X. Y. Jiang and X. M. Zheng, Environ. Sci. Technol., 2012, 46, 2337 CrossRef CAS PubMed.
  12. D. Z. Xiao, C. Cheng, J. Shen, Y. Lan, H. B. Xie, X. S. Shu, Y. D. Meng, J. G. Li and P. K. Chu, J. Appl. Phys., 2014, 115, 033303 CrossRef.
  13. P. Talebizadeh, M. Babaie, R. Brown, H. Rahimzadeh, Z. Ristovski and M. Arai, Renewable Sustainable Energy Rev., 2014, 40, 886 CrossRef CAS.
  14. A. M. Zhu, Q. Sun, J. H. Niu, Y. Xu and Z. M. Song, Plasma Chem. Plasma Process., 2005, 25, 371 CrossRef CAS.
  15. M. Schmidt, R. Basner and R. Brandenburg, Plasma Chem. Plasma Process., 2013, 33, 323 CrossRef CAS.
  16. A. Gal, M. Kurahashi and M. Kuzumoto, J. Phys. D: Appl. Phys., 1999, 32, 1163 CrossRef CAS.
  17. J. C. Kramlich and W. P. Linak, Prog. Energy Combust. Sci., 1994, 20, 149 CrossRef CAS.
  18. G. B. Zhao, X. D. Hu, M. C. Yeung, O. A. Plumb and M. Radosz, Ind. Eng. Chem. Res., 2004, 43, 2315 CrossRef CAS.
  19. G. B. Zhao, S. V. B. Garikipati, X. D. Hu, M. D. Argyle and M. Radosz, AIChE J., 2005, 51, 1800 CrossRef CAS.
  20. E. Stamate, C. Irimiea and M. Salewski, Jpn. J. Appl. Phys., 2013, 52, 05EE03 CrossRef.
  21. Y. S. Mok and I. S. Nam, Chem. Eng. Technol., 1999, 22, 527 CrossRef CAS.
  22. A. Gal, A. Ogata, S. Futamura and K. Mizuno, J. Phys. Chem. A, 2004, 108, 7003 CrossRef CAS.
  23. I. Jõgi, E. Stamate, C. Irimiea, M. Schmidt, R. Brandenburg, M. Hołub, M. Bonislawski, T. Jakubowski, M. L. Kääriäinen and D. C. Cameron, Fuel, 2015, 144, 137 CrossRef.
  24. B. Li, J. Zhao and J. Lu, Int. J. Environ. Res. Public Health, 2015, 12, 1595 CrossRef CAS PubMed.
  25. H. L. Chen, H. M. Lee, S. H. Chen, T. C. Wei and M. B. Chang, Plasma Sources Sci. Technol., 2010, 19, 055009 CrossRef.
  26. K. Leistner, A. Nicolle and P. D. Costa, J. Phys. Chem. C, 2012, 116, 4642 CAS.
  27. Y. S. Mok, J. H. Kim, I. S. Nam and S. W. Ham, Ind. Eng. Chem. Res., 2000, 39, 3938 CrossRef CAS.
  28. R. Dorai and M. J. Kushner, J. Appl. Phys., 2000, 88, 3739 CrossRef CAS.
  29. H. H. Kim, Y. Teramoto, A. Ogata, H. Takagi and T. Nanba, Plasma Chem. Plasma Process., 2016, 36, 45 CrossRef CAS.
  30. Y. S. Mok, V. Ravi, H. C. Kang and B. S. Rajanikanth, IEEE Trans. Plasma Sci., 2003, 31, 157 CrossRef CAS.
  31. J. H. Li, W. H. Goh, X. Yang and R. T. Yang, Appl. Catal., B, 2009, 90, 360 CrossRef CAS.
  32. B. Eliasson and U. Kogelschatz, IEEE Trans. Plasma Sci., 1991, 19, 309 CrossRef.
  33. E. Stamate, W. Chen, L. Jørgensen, T. K. Jensen, A. Fateev and P. K. Michelsen, Fuel, 2010, 89, 978 CrossRef CAS.
  34. M. A. Malik and K. H. Schoenbach, Plasma Chem. Plasma Process., 2014, 34, 93 CrossRef CAS.
  35. H. H. Kim and A. Ogata, International Journal of Plasma Environmental Science & Technology, 2012, 6, 43 Search PubMed.
  36. S. Li, W. Zheng, Z. C. Tang and F. Gu, Heat Transfer Eng., 2012, 33, 609 CrossRef CAS.
  37. J. Z. Sun, T. Stirner and D. Z. Wang, Plasma Sci. Technol., 2002, 4, 1227 CrossRef CAS.
  38. P. Talebizadeh, H. Rahimzadeh, M. Babaie, S. J. Anaghizi, H. Ghomi, G. Ahmadi and R. Brown, PLoS One, 2015, 10, e0140897 Search PubMed.
  39. X. L. Tang, F. Y. Gao, J. G. Wang, H. H. Yi and S. Z. Zhao, Ind. Eng. Chem. Res., 2014, 53, 6197 CrossRef CAS.
  40. S. Yu, Q. Z. Chen, J. H. Liu, K. L. Wang, Z. Jiang, Z. L. Sun, J. Zhang and J. Fang, Appl. Phys. Lett., 2015, 106, 244101 CrossRef.

This journal is © The Royal Society of Chemistry 2016