IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 10, OCTOBER 2013
3069
A Comparative Study of Water Electrodes Versus Metal Electrodes for Excitation of Nanosecond-Pulse Homogeneous Dielectric Barrier Discharge in Open Air Tao Shao, Senior Member, IEEE, Cheng Zhang, Member, IEEE, Zhi Fang, Member, IEEE, Yang Yu, Dongdong Zhang, Ping Yan, Member, IEEE, Yuanxiang Zhou, Member, IEEE, and Edl Schamiloglu, Fellow, IEEE Abstract— Atmospheric pressure low-temperature plasmas produced by dielectric barrier discharge (DBD) provide a promising approach for civilian application of pulsed power technology. In this paper, repetitive nanosecond pulses were generated using a magnetic compression solid-state pulsed power generator, and the rise time and pulse duration of the nanosecond pulse are ∼ 30 and 70 ns, respectively. The DBD in open air is created using two kinds of electrodes, i.e., water and metal electrodes. The electrical, luminous, and optical characteristics of the DBDs under these two electrodes are studied and compared. The experimental results show that no filaments are observed and the discharge is homogeneous when water electrodes are used. The DBD still behaves in a filamentary mode when the discharge gap is extended to 4 cm in the case of metal electrodes. The results are validated by fast images taken by an intensified charge-coupled device camera. In addition, some discussion about the experimental results is presented. Improvement of discharge uniformity is due to the effect of resistive stabilization using water electrodes. Index Terms— Dielectric barrier discharge (DBD), discharge emission, filamentary discharge, gas discharge, homogeneous discharge, metal electrode, nanosecond pulse, nonthermal plasma, open air, pulsed discharge, pulsed power, water electrode. Manuscript received April 7, 2013; revised July 10, 2013; accepted August 5, 2013. Date of publication September 11, 2013; date of current version October 7, 2013. This work was supported in part by the National Natural Science Foundation of China under Contract 51222701, Contract 51207154, and Contract 11076026, in part by the National Basic Research Program of China under Contract 2014CB239505, and in part by the Opening Project of State Key Laboratory of Electrical Insulation and Power Equipment at Xi’an Jiaotong University under Contract EIPE12204. T. Shao, C. Zhang, Y. Yu, and D. Zhang are with the Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, China, and also with the Key Laboratory of Power Electronics and Electric Drive, Chinese Academy of Sciences, Beijing 100190, China (e-mail:
[email protected];
[email protected]). Z. Fang is with the School of Automation and Electrical Engineering, Nanjing University of Technology, Nanjing 210009, China (e-mail:
[email protected]). P. Yan is with the Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, China, and also with the State Key Laboratory of Electrical Insulation and Power Equipment, Xi’an Jiaotong University, Xi’an 710049, China (e-mail:
[email protected]). Y. Zhou is with the State Key Laboratory of Power Systems, Electrical Engineering Department, Tsinghua University, Beijing 100084, China (e-mail:
[email protected]). E. Schamiloglu is with the Department of Electrical & Computer Engineering, University of New Mexico, Albuquerque, NM 87131 USA (e-mail:
[email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TPS.2013.2279254
I. I NTRODUCTION
N
OWADAYS, pulsed power technology has been applied to many kinds of civilian applications, and it is no longer confined to defense applications and basic physics research [1]. Applications of pulsed power in material surface modification, treatment of pollutant gases and water, and biomedical sterilization have been extensively reported [2]. Among these various applications, pulsed power is always used for an excitation of low-temperature plasmas. Pulsed power is characterized by the sharp release in a short time interval after an energy accumulation over a relatively long period of time and can produce high peak levels of power into a given load [1]. Therefore, some applications can benefit from very fast excitation of low-temperature plasmas using pulsed power. For instance, heat loss may be decreased, plasma chemical reaction can be improved, and then, the applicable efficiency can be enhanced, which has been confirmed experimentally in some comparative studies of gas discharge plasmas [3]–[7]. At atmospheric pressure, the dielectric barrier discharge (DBD) is an important method for producing low temperature plasmas [8]. Compared with corona discharges, arc discharges and other forms of atmospheric-pressure discharges, the configuration of DBD avoids the transition of the discharge to spark or arc mode and can generate moderate plasmas [8]–[10]. For certain applications, homogeneous discharges at atmospheric air are an important aim for different investigations. Therefore, many researches are carried out in different gases, barrier dielectrics, electrode geometries, driving power sources, and some additional methods, including preionization. The driving power source is one of the key factors for producing atmospheric-pressure homogeneous discharges. Generally, DBDs are generated by ac power sources of 50 Hz, 10 s of kilohertz or radio frequency of periodic sine or square waves [10]. In these cases, the DBD has a relatively low energy input because only spikelike currents imposed on the ac current waveform can enter into the reacting plasma field [3]. From the experimental researches of a comparison of highvoltage ac and pulsed discharge, it is concluded that more powerful plasmas may be generated using pulsed high voltage, and it is much more efficient in the case of pulsed power [4]–[7], [11]–[17], such as the production of plasma/reactive species.
0093-3813 © 2013 IEEE
3070
IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 10, OCTOBER 2013
Pulsed power generators, however, vary widely in performance according to load requirements; different results may be obtained from using various pulsed power generators and kinds of discharge configurations [18]. It is known that many DBDs being used in industry are still in a filamentary mode [8], so the homogeneity of DBDs can improve the industrial applications, such as for surface modification, the homogeneous DBD can treat the polymer surface much uniformly, and avoid partial ablation. In any event, largevolume homogeneous discharges are desirable for various kinds of plasma applications. The so-called homogeneous or glowlike discharge can be achieved under some conditions, which are determined by gas parameters such as the type of gas mixture, pressure and flow rate, as well as power-source parameters [19]–[22]. The transition mechanisms between a homogeneous mode and a filamentary mode are, however, not fully understood. Experimental and theoretical works are, therefore, still being conducted and reported in this field. This paper reported herein continues the cycle of research on a nanosecond-pulse DBD in atmospheric air [20], [23]–[27]. This paper described herein pertains to the use of a magnetic compression solid-state pulse generator to produce a stable homogeneous DBD in atmospheric air via two kinds of electrode configurations. The goal of the experimental work described in this paper is to present a comparative study of water versus metal electrodes for excitation of a nanosecond-pulse homogeneous DBDs in open air. II. E XPERIMENTAL A RRANGEMENT For industrial applications, compact all-solid-state switched generators are convenient, and a pulsed power generator based on magnetic pulse compression system is a good choice [28]. In this paper, a home-made all-solid-state repetitive nanosecond-pulse generator with single stage magnetic pulse compression is used [18], [27]. This generator can provide repetitive high-voltage pulses of up to 30 kV with a rise time of 40 ns and a full-width-at-half-maximum pulse of 70 ns. A high-power resistor of low inductance was connected with the discharge system in parallel and used to obtain a relatively stable output voltage. The output pulse voltage can be changed by the ac input voltage. The pulse repetition frequency (PRF) of the output pulses can be varied by changing the frequency of the trigger modulator, and the range of the PRF can be chosen from single pulse to up to 1 kHz. A schematic diagram of the experimental setup can be seen in Fig. 1. Two kinds of electrode configurations were used in the experiment. One discharge type used water electrodes, and the DBD was created between two cylindrical glass containers with the same diameter of 35 mm. Each container is filled with a 16.7% saline soultion (< 100 ). Two parallel glass plates with a thickness of 1 mm serve as dielectric layers. The other type employed tranditional metal electrodes, and the DBD was generated between two circular planeparallel aluminum electrodes of 50 mm in diameter. Both electrodes were covered by 100-mm × 100-mm area, 1-mmthick glass plates. To keep a good contact between the glass plates and aluminum electordes, one layer of aluminum foil
Fig. 1.
Schematic diagram of the experimental setup.
(the same diameter as the aluminum electrodes) was coated on the glass plate in advance. All these experiments were performed in open air and at room temperature. Applied voltages were measured by a high-voltage probe (Tektronix, P6015, 1000:1) and discharge currents were measured by a current monitor (Pearson, Model 4100, 1 A/V). All signal measurements were recorded using a digital oscilloscope (Tektronix, DPO 2024, 1 GS/s, 200 MHz). Two kinds of image recordings were used for the discharge photographs. One was a Canon EOS500D digital camera with a Tamron lens (model A001), which took time-integrated images of the DBD on a time scale of the order of hundreds of milliseconds or longer. The other was an intensified charge-coupled device (ICCD) camera (Princeton Instruments, PIMAX2), which can be used at a short exposure time of 2 ns. Synchronization between the discharge and ICCD gating was achieved by a step voltage setup triggered by the trigger modulator of the generator [29]. For the configuration of water electrodes, both side- and front-view photographs were taken to study the discharge uniformity. Only side-view images were taken when metal electrodes were used. The optical emission spectra of the plasmas were measured by a spectrometer (Avantes, AvaSpec-2048, spectral range: 200–1000 nm, resolution: 0.12 nm) [23], [27]. The optical fiber was placed vertically near the discharge gap while the distance was constant at 20 mm. III. E XPERIENTIAL R ESULTS A. Electrical Characteristics of DBD For the case of water electrodes, Fig. 2 shows typical DBD voltage and current waveforms, where the experimental conditions are that the gap spacing was 2 mm and the applied PRF was 1 kHz. The left waveforms in Fig. 2 are the measured consecutive pulse train waveforms of 10 pulses, and the right waveforms give an expanded view of a single pulse. It can be seen that the repetitive discharges occurred for each voltage pulse and the corresponding discharge current was also repetitive, indicating the stability of the discharge. With an
SHAO et al.: COMPARATIVE STUDY OF WATER ELECTRODES VERSUS METAL ELECTRODES
3071
Fig. 4. Typical waveform of DBD applied voltage and current. Air gap spacing was 2 mm, and the applied PRF was 1 kHz and metal electrodes. Fig. 2. DBD waveforms of applied voltage and discharge current in a (a) and (c) repetitive mode and (b) and (d) single mode, respectively. Air gap spacing was 2 mm, and applied PRF was 1 kHz and water electrodes.
Fig. 3. Applied voltage, discharge current, and displace current waveforms. Air gap spacing was 3 mm, applied voltage was 25 kV, and water electrodes.
applied pulse of 25 kV, the positive peaks of the discharge current of DBD are ∼17 A, which is significantly different from the discharge current in an ac DBD with an amplitude of 10–100 s mA [3]–[7], [20], [21]. For the case of fast pulse excitation, the effect of displacement current (because of the charging of capacitive load) should not be neglected. In this case, when the air gap was extended to 3 mm and only a single voltage pulse was applied, no discharge was observed and the measured current was similar to the displacement current, as shown in the red thin line of Fig. 3. Although repetitive pulse voltages of 1 kHz were applied, a stable DBD occurred and a discharge current was presented in the blue dotted line of Fig. 3. Therefore, the current waveforms in Fig. 3 show a clear distinction between the discharge and displacement currents. For metal electrodes, typical waveforms for applied voltage and current are shown in Fig. 4. The experimental parameters are: 1) 1-mm glass in thickness and 2) 2-mm air gap spacing. According to Fig. 4, the rise time and pulsewidth at the half peak of the applied voltage are ∼40 and 80 ns, respectively,
whereas the waveform of the discharge current was much sharper and possessed a pulsewidth of ∼20 ns. At a peak applied voltage of ∼23 kV, the discharge current was ∼60 A, which was much higher than that in Figs. 2 and 3. The bipolar pulse characteristic of the current was reported in pulsed DBDs previously [4]–[7], [13]–[15]. In the stage of primary discharge, the discharge current increases with the applied voltage. The discharge current pulse rises sharply and extinguishes quickly within 20 ns. In the initial stage of the primary discharge, a slow rising current part corresponds to a displacement current characteristic. When the applied voltage drops, some charges are accumulated on the dielectric barrier and lead to a secondary discharge. It can be found that the intensity of the secondary discharge is significantly decreased. According to an equivalent electrical model [4], [10], the discharge voltage and current across the air gap, voltage across the dielectric layer and displacement current, instantaneous input power, power across the air gap, and dielectric layer can be calculated. To compare the electrical parameters, air gaps of 2 mm were fixed under the conditions of water and metal electrodes. According to the measured applied voltage and total current shown in Figs. 2 and 4, real-time data for voltage, current, and instantaneous power are calculated and shown in Fig. 5. The peak values of the main electrical parameters of the discharge are also given in Table 1. For metal electrodes, the calculated discharge voltage and current across the air gap are shown in Fig. 5(a) and (b). It can be seen that the calculated voltage across the air gap appears bipolar, but the voltage across the dielectric layer still remains unipolar and its peak value is beyond the pulse amplitude of the voltage across the air gap. Furthermore, the instantaneous total input power, instantaneous power consumption in the air gap, and power consumption in the dielectric layer are shown in Fig. 5(c). All these power consumption curves have two pulses, but the secondary pulses are much lower than the first pulse. It can be explained from the current waveforms in Fig. 5(b). The peak value of the secondary discharge pulse is much lower than that of the primary discharge. This can be attributed to power dissipation and
3072
IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 10, OCTOBER 2013
TABLE I P EAK VALUE OF E LECTRICAL PARAMETERS OF THE DBD S
Fig. 6. Front-view discharge images with different exposure times (applied voltage: 27 kV, gap spacing: 1 mm, PRF: 1000 Hz, and water electrodes).
only about one third of the amplitude of metal electrodes. The decrease of electrical parameters using water electrodes can be attributed to that the saline water limits the discharge intensity and behaves as ballast [19]. Some measured deviation results from impedance mismatch in the DBD with water electrodes. The total power consumption of the discharge in the air gap can reach a peak value of 1 MW and 300 kW under the conditions of metal and water electrodes, respectively. Provided that the current density is calculated by a division of the discharge current across the air gap and the effective discharge area for the homogeneous discharge (the calculation method can be found in [10]), the estimated electron density is ∼1011–1012 cm−3 under these experimental conditions. According to the estimation of the average electron density under the experimental conditions of Fig. 5, the electron densities are estimated to be ∼4.2 × 1011 cm−3 and 2 × 1011 cm−3 under the conditions of metal and water electrodes, respectively. B. Discharge Images
Fig. 5. Typical waveform of DBD applied voltage and current. Air gap spacing was 2 mm (a)–(c) with metal electrodes and (d)–(f) with water electrodes.
charge accumulation in the primary discharge [9], [10]. It is known that the secondary discharge is induced by the charges accumulated on the surface of the dielectric during the primary discharge. The total power provides the excitation energy only during the primary discharge, much of which supports the DBD plasma in the air gap and the other residual is stored or consumed by the dielectric [10]. Therefore, the secondary pulse curve of the air gap power is positive, and the secondary pulse curves of both total and dielectric powers are negative. In contrast with data for metal electrodes, the calculated data for water electrodes can be seen in Fig. 5(d) and (e). It can be found that the calculated voltage peak across air gap in the case of water electrodes is nearly the same as that of metal electrodes, but the discharge current peak is
With regard to the difference between the homogeneous and filamentary modes, discharge images are used to distinguish whether or not the mode is homogeneous [20]. When metal electrodes are used, the side-view photographs of the electrode can be used. In the case of water electrodes, the photographs from the side or front views are used to judge uniformity. In this paper, both images of long-time and shorttime exposures are used to investigate this discharge. Some characteristics of these discharge images are reported in [23]– [25]. Here, only some typical images are shown and used to compare the experimental results between metal and water eletrodes. In the case of water electrodes, Fig. 6 shows the frontview discharge images at different exposure times. It can be seen that no filaments were observed and the discharge was homogeneous and glowlike in the entire discharge region as well as a variation of luminosity. For all these cases, the discharge always exhibits a diffuse and homogeneous mode under the conditions of 1–3-mm air gaps [20]. For metal electrodes, some typical discharge images can be seen in Fig. 7, where the gap spacing was changed from 1–4 mm and the thickness of the glass dielectric was fixed at
SHAO et al.: COMPARATIVE STUDY OF WATER ELECTRODES VERSUS METAL ELECTRODES
3073
Fig. 7. Side-view discharge images at different air gaps (applied voltage was 25 kV, PRF: 500 Hz, and metal electrodes).
Fig. 9. Side-view discharge images at different air gaps (applied voltage was 25 kV, PRF: 50 Hz, and metal electrodes).
Fig. 8. Discharge images at different PRFs (applied voltage was 25 kV, PRF: 500 Hz, and metal electrodes).
1 mm. When the gap distance was < 3 mm and the applied PRF was < 500 Hz, the discharge was always homogeneous besides a variation of luminosity. In these discharge images, the applied voltage was 25 kV, PRF was 500 Hz, and exposure times of these images were set at 0.1 s. In these experiments, it was found that the luminosity was enhanced with the increase of PRF. When the gap was as large as 4 mm, the discharge mode was changed from diffuse to filamentary with the increase of PRF. As shown in Fig. 8, the discharge still appeared to be diffuse with a PRF of 250 Hz, whereas with the increase of PRF, more filaments were observed from these images. It can be seen that some characteristics are the same with results ever reported [10], where the applied PRF is the important parameters influencing the transition of DBD from homogeneous to filamentary modes. Figs. 6–8 were taken by a digital camera with a long exposure times. Furthermore, the homogeneous characteristic is proved by a high-speed ICCD camera with an exposure time of 2 ns, and the temporal behavior of discharge can be found in [24] and [26]. It can be found that DBD using water electrodes can generate homogeneous atmospheric-pressure plasmas in air. In the case of metal electrodes, the discharge at a 1-mm air gap is diffuse and homogeneous, and the discharge at a 4-mm air gap will transform to a filamentary mode. Fig. 9 shows some ICCD images at different air gaps and the discharge images were taken when the discharge current was close to the maximum of the current pulse. It can be observed
that the discharge was in a homogeneous mode when the air gap was 1 mm. With the increase of air gap, the discharge was transformed to a filamentary mode. Some constricted filaments were observed in the discharge image at an air gap of 4 mm. In the case of water electrodes, the front-view discharge images can be obtained because of liquid electrodes. According to the 2-D images presented in [24], 3-D images can also be obtained using an attached function of the ICCD camera in conjunction with data processing software [29]. Fig. 10 shows time-dependent discharge images in a nanosecond time scale. The ICCD camera only can capture one image during one pulse, but the time-dependent images can be detected owing to the good reproducibility of the discharge. An adjustable time delay was added to track the evolution of the discharge. The ICCD camera was used with a short exposure time of 2 ns, and the time delay between adjacent images of the discharge was also set at 2 ns after the time interval between the triggered pulses is reduced. The frame schematic of the ICCD trigger gates was shown in the discharge current waveform of Fig. 10. The results further validate that the discharge behaved in a homogeneous mode and was without any filaments. These characteristics are also consistent with those discharge images taken at a long exposure times in Figs. 6–8. The 3-D images with an exposure time of 2 ns indicate that the discharge does not develop in a radially uniform fashion, but starts from the center and spreads outward from the discharge electrode step-by-step [29], [30]. C. Discharge Light Emission In the experiments, optical emission spectroscopy is also used for plasma diagnostics. It can be applied to identify the various chemical species presented in the discharge gap [5], [22], [23], [27]. Under the same experimental conditions in Fig. 4, typical emission spectra of the DBD plasma are shown in Fig. 11(a). It clearly shows that for the air plasma, there was strong emission from nitrogen species < 400 nm, notably the nitrogen second positive system owing to the radiative deexcitation between N2 (C3 u -B3 g ) bands [23]. The N2 (1-0), N2 (0-0), N2 (0-1), N2 (0-2), and N2 (0-3) bands are the dominant ones for emission. The emission from other species
3074
IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 10, OCTOBER 2013
Fig. 10. ICCD images of the nanosecond-pulse DBD at an air gap of 2 mm in open air. (Applied voltage: 25 kV, air gap: 2 mm, PRF: 50 Hz, and water electrodes, these 2-D images are from [24]). (b)–(h) are the 2-ns images at the time scales shown in the current waveform in (a).
(such as the first negative bands of N+ 2 at 391.6 nm) was not clear or their emission was very low compared with that from the nitrogen second positive system. Under different experimental conditions, it can be observed that the spectra exhibited spectral signatures of the same group of plasma species. Furthermore, when water electrodes were used, the same spectral signatures can be seen and only the intensity of luminosity was reduced. According to Fig. 11(a), the emission spectra in the range 368–383 nm of the second positive system of nitrogen are used to determine the rotational and vibrational temperatures. A software code can be used to determine the rotational and vibrational temperatures of the discharges by comparing the experimental and model spectra of nitrogen second positive system [5], [23]. The rotational and vibrational temperatures are 312 and 2790 K, respectively, as shown in Fig. 11(b). The nanosecond-pulse plasma is characterized with a low rotational temperature and high vibrational temperature, and, therefore, is under strongly nonequilibrium conditions.
In our case, the effect of applied voltage and PRF on the rotational and vibrational temperatures was studied [27]. For metal electrodes, the dependence of rotational and vibrational temperatures on the applied voltage and PRF is shown in Figs. 12 and 13, respectively. In these experiments, air gap was fixed, and glass barriers of 1- and 2-mm thickness were used. Fig. 12 shows the variations of the rotational an vibrational temperatures with the applied voltage when the PRF was fixed at 1 kHz. The rotational temperature increases slightly with the applied voltage, which may be resulted from the increased discharge intensity when the applied voltage increases. The vibrational temperature practically remains unchanged. In addition, the rotational temperature of the excited plasma in the case of 2-mm barrier dielectric is ∼20 K lower than that of 1-mm barrier dielectric. Fig. 13 shows the variations of the rotational and vibrational temperatures with the PRF when the applied voltage was fixed at 25 kV. Similar to the tendency of rotational and vibrational temperatures with the applied voltage, it is obvious that the
SHAO et al.: COMPARATIVE STUDY OF WATER ELECTRODES VERSUS METAL ELECTRODES
1.0
337.0
(a)
0.8
Intensity (a.u.)
3075
357.6
0.6 315.8
0.4 0.2
380.5 375.6
353.6
313.5
371.0
0.0
399.9405.9 394.3
310 320 330 340 350 360 370 380 390 400 410 (b)
Simulation Experiment
Intensity (a.u.)
1.0 0.8
Trot=312;Tvib=2790
0.6 0.4 0.2 0.0 368
370
372
374
376
378
380
Wavelength (nm)
Vibrational temperature (K) Rotational temperature (K)
Fig. 11. (a) Typical optical emission spectrogram of the plasma in open air. (b) Experimental and simulated spectrogram of the DBD plasma.
Fig. 13. Rotational and vibrational temperatures of the DBD plasma under different PRFs (air gap spacing: 2 mm and metal electrodes). (a) Rotational and (b) vibrational temperatures of the DBD plasma under different PRFs.
380 360
1mmDielectric 2mmDielectric
340 320 300
(a)
280
3000 2800 2600 2400 2200 2000 17
(b) 18
19
20
21
22
Voltage (kV)
23
24
25
26
Fig. 12. Rotational and vibrational temperatures of the DBD plasma under different applied voltages (air gap: 2 mm, PRF: 1 kHz, and metal electrodes). (a) Rotational and (b) vibrational temperatures of the DBD plasma under different applied voltages.
rotational temperature increases slightly with the PRF, but the variation of vibrational temperature is weak. The rotational temperature of the excited plasmas in the case of 2-mm barrier dielectric is ∼30 K lower than that of 1-mm barrier dielectric. In our case, the data show that the emission intensity was sensitive to the applied voltage and PRF. The rotational
temperature indicates insensitivity to the applied voltage and PRF as shown in Figs. 12 and 13, but it may be affected by the discharge configuration, such as the change of barrier dielectric thickness. It is due to the fact that the variation of the dielectric will change the discharge intensity and affect light emission. Under the same experimental conditions, a comparison of metal and water electrodes for the effect on the rotational and vibrational temperatures of the excited plasmas is shown in Fig. 14. It can be seen that, for the metal and water electrodes, the vibrational temperatures are nearly the same and in the range 2500–3000 K. Both the applied voltage and PRF have a slight influence on the vibrational temperature. The rotational temperatures are in the range 300–350 K. This suggests that increasing the applied voltage or the PRF cannot clearly change the gas temperature in our case (the gas temperature can nearly be identified by the rotational temperature). Though the similar spectra are presented using metal or water electrodes, the rotational temperatures of the excited plasma using metal electrodes is ∼30 K higher than that using water electrodes. This is consistent with the experimental results of electrical parameters shown in Fig. 5 and Table 1. IV. D ISCUSSION Compared with filamentary microdischarges with sinusoidal excitation, stable and homogeneous DBDs driven by nanosecond pulses can be applied to generate low temperature plasmas in open air [20]. From our experimental results, it is shown that the fast rise time and short pulse duration of the voltage in pulsed excitation is beneficial to achieving discharge uniformity. It is known that the formation of a filamentary path
3076
Fig. 14. Rotational and vibrational temperatures of the DBD plasma. (a) Effect of applied voltage. (b) Effect of PRF (air gap: 2 mm, dielectric: 1 mm, PRF: 1 kHz, metal and water electrodes).
is a streamer development. Generally, a primary avalanche converges toward a single streamer filament at expense of simultaneous developments of other avalanches, and then, a filamentary channel of streamer forms in the gap. In the case of a short gap, if the initial electron density is high enough for sustaining simultaneous developments of some initial avalanches, and then, a homogeneous discharge develops in the gap [26], [31]. In addition, for a nanosecond-pulse discharge, the enhancement of electric field near the cathode is different from traditional discharge because of different drift velocity and movement direction of electrons and ions [19], [32]. Therefore, for a relatively short gap, initial avalanches are developed simultaneously, and heads of developing avalanches overlap before the streamer formation. The discharge may behave in a diffuse and homogeneous mode as shown in Fig. 10. As the gap spacing is increased, the initial density of seed electrons is, however, decreased and the spacecharge field no longer becomes uniform in response to the weak charge accumulation. Therefore, the discharge develops some filaments. Uniformity of the DBD plasma can be improved by shortening the applied voltage rise time and duration. In addition, the electrodes have an influence on discharge uniformity. In the case of metal electrodes, the experimental study shows that the discharge was homogeneous at 1-mm gap spacing, whereas it was filamentary when the gap spacing was increased to 4 mm. Similar results are obtained
IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 10, OCTOBER 2013
except that the discharge was always uniform with water electrodes. The configuration of the water electrodes can be considered as a resistance connecting in series with the discharge circuit, and it plays a role such as a distributed ballast that limits the discharge current and, therefore, avoids the glow-to-filamentary transition [33]. In addition, the discharge current with water electrodes is much lower than that with metal electrodes, which is also useful for the discharge uniformity [34]. Therefore, the suppression of the instabilities using water electrode is desirable for improving discharge uniformity [19], [20], [34], [35]. From the 2- or 3-D evolution shown in Fig. 10, it can be found that the discharge developed along the radial direction of the electrode [24], [26], which is similar to the discharge images of dielectric barrier glowlike discharge taken in an alternating-voltage DBD in atmospheric-pressure helium also show this characteristic [29], [30]. In our case, a radial evolution is also found in a nanosecond-pulse DBD in open air regardless of gap spacing. The radial evolution of discharge is associated with the axial electrical field and residual charges [26], [29], [30]. The axial electric field is maximal and density of residual charges is higher at the center of the electrode than at the edge. Much ionization takes place at the center because the ionization coefficient strongly depends on the electric field. Therefore, the discharge develops much rapidly at the center because of the charge accumulation, which establishes stepby-step the discharge breakdown as indicated by the evolution of the light emission. The emission spectrum indicates that the plasma is under a highly nonequilibrium condition [5], [23], [27] and the low rotational and high vibrational temperatures. There are the same plasma groups from the spectrum measurement, and the plasmas are with almost the same vibrational temperatures under the two electrode conditions, respectively. The rotational temperature in the case of water electrodes is, however, lower than that with metal electrodes. This can be attributed that resistive stabilization using water electrodes decreases the discharge current and reduces the heat power. V. C ONCLUSION In general, the DBDs in atmospheric air driven by ac power source always appear in the form of filamentary discharges. From our experimental results, the DBDs driven by nanosecond pulses can be used to improve the discharge uniformity, which can enrich the plasma generation technology. For a comparative study, two types of electrodes (water and metal electrodes) are used. The electrical characteristic, optical emission spectrum, and discharge images are also obtained. It can be found that a uniform DBD was produced in open air using a compact all-solid state nanosecond-pulse generator when the gap spacing was < 3 mm. It also can be found that the discharge current and emission intensity are smaller in the case of water electrodes, which also resulted in a lower rotational temperature when compared with that of metal electrodes. Furthermore, the discharge uniformity will benefit from the use of water electrodes, which is due to the resistive stabilization. The use of water electrodes can improve the required
SHAO et al.: COMPARATIVE STUDY OF WATER ELECTRODES VERSUS METAL ELECTRODES
conditions of obtaining a homogeneous discharge. In addition, it is found that the discharge starts from the center and expands radially outward the discharge and covered the entire electrode surface step-by-step. An intense optical emission from the nitrogen second positive system < 400 nm indicates that the plasma is with low rotational and high vibrational temperatures. ACKNOWLEDGMENT The authors would like to thank Dr. H. Luo (Tsinghua University, Beijing, China) for his help on the ICCD camera and Prof. X. Lu for providing the spectra fitting software. R EFERENCES [1] J. O. Rossi, E. Schamiloglu, and M. Ueda, “Advances in high-voltage modulators for applications in pulsed power and plasma-based ion implantation,” IEEE Trans. Plasma Sci., vol. 39, no. 11, pp. 3033–3044, Nov. 2011. [2] H. Akiyama, T. Sakugawa, T. Namihira, K. Takaki, Y. Minamitani, and N. Shimomura, “Industrial applications of pulsed power technology,” IEEE Trans. Dielectr. Electr. Insul., vol. 14, no. 5, pp. 1051–1064, Oct. 2007. [3] K. Okazaki and T. Nozaki, “Ultrashort pulsed barrier discharges and applications,” Pure Appl. Chem., vol. 74, no. 3, pp. 447–452, 2002. [4] J. M. Williamson, D. D. Trump, P. Bletzinger, and B. N. Ganguly, “Comparison of high-voltage AC and pulsed operation of a surface dielectric barrier discharge,” J. Phys. D, Appl. Phys., vol. 39, no. 20, pp. 4400–4406, 2006. [5] Q. Xiong, X. Lu, K. Ostrikov, Z. Xiong, Y. Xian, C. Zou, and Y. Pan, “Pulsed dc- and sine-wave-excited cold atmospheric plasma plumes: A comparative analysis,” Phys. Plasma, vol. 17, no. 4, pp. 043506-1–043506-8, 2010. [6] T. Martens, A. Bogaerts, and J. van Dijk, “Pulse shape influence on the atmospheric barrier discharge,” Appl. Phys. Lett., vol. 96, no. 13, pp. 131503-1–131503-3, 2010. [7] J. Hirschberg, T. Omairi, N. Mertens, A. Helmke, S. Emmert, and W. Viol, “Influence of excitation pulse duration of dielectric barrier discharges on biomedical applications,” J. Phys. D, Appl. Phys., vol. 46, no. 16, p. 165201, 2013. [8] U. Kogelschatz, “Dielectric-barrier discharges: Their history, discharge physics, and industrial applications,” Plasma Chem. Plasma Process., vol. 23, no. 1, pp. 1–46, 2003. [9] M. Laroussi and X. Lu, “Power consideration in the pulsed dielectric barrier discharge at atmospheric pressure,” J. Appl. Phys., vol. 96, no. 5, pp. 3028–3030, 2004. [10] T. Shao, K. Long, C. Zhang, P. Yan, S. Zhang, and R. Pan, “Experimental study on repetitive unipolar nanosecond-pulse dielectric barrier discharge in air at atmospheric pressure,” J. Phys. D, Appl. Phys., vol. 41, no. 21, pp. 215203-1–215203-8, 2008. [11] H. Ayan, G. Fridman, A. F. Gutsol, V. N. Vasilets, A. Fridman, and G. Friedman, “Nanosecond-pulsed uniform dielectric-barrier discharge,” IEEE Trans. Plasma Sci., vol. 36, no. 2, pp. 504–508, Apr. 2008. [12] F. Iza, J. L. Walsh, and M. G. Kong, “From submicrosecond-to nanosecond-pulsed atmospheric-pressure plasmas,” IEEE Trans. Plasma Sci., vol. 37, no. 7, pp. 1289–1296, Jul. 2009. [13] X. Lu, Q. Xiong, Z. Xiong, Y. Xian, F. Zhou, J. Hu, W. Gong, C. Zhou, Z. Tang, Z. Jiang, and Y. Pan, “Effect of nano- to millisecond pulse on dielectric barrier discharges,” IEEE Trans. Plasma Sci., vol. 37, no. 5, pp. 647–652, May 2009. [14] E. Panousis, N. Merbahi, F. Clement, M. Yousfi, J. F. Loiseau, O. Eichwald, and B. Held, “Analysis of dielectric barrier discharges under unipolar and bipolar pulsed excitation,” IEEE Trans. Dielectr. Electr. Insul., vol. 16, no. 3, pp. 734–741, Jun. 2009. [15] J. L. Walsh, D. X. Liu, F. Iza, M. Z. Rong, and M. G. Kong, “Contrasting characteristics of sub-microsecond pulsed atmospheric air and atmospheric pressure helium-oxygen glow discharges,” J. Phys. D, Appl. Phys., vol. 43, no. 3, p. 032001, 2010. [16] D. Yang, W. Wang, L. Jia, D. Nie, and H. Shi, “Production of atmospheric pressure diffuse nanosecond pulsed dielectric barrier discharge using the array needles-plate electrode in air,” J. Appl. Phys., vol. 109, no. 7, pp. 073308-1–073308-5, 2011.
3077
[17] D. Yang, W. Wang, S. Zhang, K. Tang, Z. Liu, and S. Wang, “Multiple current peaks in room-temperature atmospheric pressure homogenous dielectric barrier discharge plasma excited by high-voltage tunable nanosecond pulse in air,” Appl. Phys. Lett., vol. 102, no. 19, pp. 194102-1–194102-5, 2013. [18] T. Shao, D. Zhang, Y. Yu, C. Zhang, J. Wang, P. Yang, and Y. Zhou, “A compact repetitive unipolar nanosecond-pulse generator for dielectric barrier discharge application,” IEEE Trans. Plasma Sci., vol. 38, no. 7, pp. 1651–1655, Jul. 2010. [19] Z. Fang, T. Shao, S. Ji, J. Pan, and C. Zhang, “Generation of homogeneous atmospheric-pressure dielectric barrier discharge in a large-gap argon gas,” IEEE Trans. Plasma Sci., vol. 40, no. 7, pp. 1884–1890, Jul. 2012. [20] T. Shao, Y. Yu, C. Zhang, D. Zhang, Z. Niu, J. Wang, P. Yan, and Y. Zhou, “Excitation of atmospheric pressure uniform dielectric barrier discharge using repetitive unipolar nanosecond-pulse generator,” IEEE Trans. Dielectr. Electr. Insul., vol. 17, no. 6, pp. 1830–1837, Dec. 2010. [21] D. Yang, Y. Yang, S. Li, D. Nie, S. Zhang, and W. Wang, “A homogeneous dielectric barrier discharge plasma excited by a bipolar nanosecond pulse in nitrogen and air,” Plasma Sour. Sci. Technol., vol. 21, no. 3, p. 035004, 2012. [22] S. Zhang, W. Wang, L. Jia, Z. Liu, Y. Yang, and L. Dai, “Rotational, vibrational, and excitation temperatures in bipolar nanosecondpulsed diffuse dielectric-barrier-discharge plasma at atmospheric pressure,” IEEE Trans. Plasma Sci., vol. 41, no. 2, pp. 350–354, Feb. 2013. [23] T. Shao, Y. Yu, C. Zhang, H. Jiang, P. Yan, and Y. Zhou, “Generation of atmospheric pressure plasma by repetitive nanosecond pulses in air using water electrodes,” Plasma Sci. Technol., vol. 13, no. 6, pp. 735–739, 2011. [24] T. Shao, Z. Niu, C. Zhang, Y. Yu, H. Jiang, W. Li, P. Yan, and Y. Zhou, “ICCD observation of homogeneous DBD excitated by unipolar nanosecond pulses in open air,” IEEE Trans. Plasma Sci., vol. 39, no. 11, pp. 2062–2063, Nov. 2011. [25] T. Shao, C. Zhang, Y. Yu, Z. Niu, H. Jiang, J. Xu, W. Li, P. Yan, and Y. Zhou, “Discharge characteristic of nanosecond-pulse DBD in atmospheric air using magnetic compression pulsed power generator,” Vacuum, vol. 86, no. 7, pp. 876–880, 2012. [26] T. Shao, C. Zhang, Y. Yu, Z. Fang, and P. Yan, “Temporal evolution of nanosecond-pulse dielectric barrier discharges in open air,” Europhys. Lett., vol. 97, no. 5, p. 55005, 2012. [27] T. Shao, C. Zhang, Y. Yu, P. Yan, and E. Schamiloglu, “Experimental study of nanosecond-pulse dielectric barrier discharge in open air,” in Proc. IEEE Int. Power Modulator High Voltage Conf., Jun. 2012, pp. 435–438. [28] D. Zhang, Y. Zhou, J. Wang, and P. Yan, “A compact, high repetition-rate, nanosecond pulse generator based on magnetic pulse compression system,” IEEE Trans. Dielectr. Electr. Insul., vol. 18, no. 4, pp. 1151–1157, Aug. 2011. [29] H. Luo, Z. Liang, B. Lv, X. Wang, Z. Guan, and L. Wang, “Radial evolution of dielectric barrier glowlike discharge in helium at atmospheric pressure,” Appl. Phys. Lett., vol. 91, no. 23, pp. 2315041–231504-3, 2007. [30] L. Mangolini, K. Orlov, U. Kortshagen, J. Heberlein, and U. Kogelschatz, “Radial structure of a low-frequency atmosphericpressure glow discharge in helium,” Appl. Phys. Lett., vol. 80, no. 10, pp. 1722–1724, 2002. [31] J. I. Levatter and S. Lin, “Necessary conditions for the homogeneous formation of pulsed avalanche discharges at high gas pressures,” J. Appl. Phys., vol. 51, no. 1, pp. 210–222, 1980. [32] T. Ito, T. Kanazawa, and S. Hamaguchi, “Rapid breakdown mechanisms of open air nanosecond dielectric barrier discharges,” Phys. Rev. Lett., vol. 107, no. 6, pp. 065002-1–065002-4, 2011. [33] M. Laroussi and I. Alexeff, “The resistive barrier discharge (RBD),” in Proc. IEEE Int. Conf. Plasma Sci., Las Vegas, NV, USA, Jun. 2001, p. 169. [34] E. E. Kunhardt, “Generation of large-volume, atmospheric-pressure, nonequilibrium plasmas,” IEEE Trans. Plasma Sci., vol. 28, no. 1, pp. 189–200, Feb. 2000. [35] J. Rahel and D. M. Sherman, “The transition from a filamentary dielectric barrier discharge to a diffuse barrier discharge in air at atmospheric pressure,” J. Phys. D, Appl. Phys., vol. 38, no. 4, pp. 547–554, 2005.
3078
IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 10, OCTOBER 2013
Tao Shao (M’10–SM’12) was born in Hubei, China, in 1977. He received the B.Sc. degree from the Wuhan University of Hydraulic and Electrical Engineering, Wuhan, China, in 2000, the M.Sc. degree in electrical engineering from Wuhan University, Wuhan, in 2003, and the Ph.D. degree in electrical engineering from the Graduate University, Chinese Academy of Sciences, Beijing, China, in 2006. He is currently with the Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing. From 2011 to 2012, he was a Visiting Scholar with the Department of Electrical and Computer Engineering, University of New Mexico, Albuquerque, NM, USA. Since 2012, he has been a Visiting Scholar with the State Key Laboratory of Power Transmission Equipment and System Security and New Technology, Chongqing University, Chongqing, China. His current research interests include high-voltage insulation, gas discharge, plasma application, and measurement. Dr. Shao is a Senior Member of the Chinese Society of Electrical Engineering. He is a recipient of the 2012 Lu Jiaxi Young Talent Award from CAS K. C. Wong Education Foundation.
Cheng Zhang (M’13) was born in Jiangsu, China, in 1982. He received the Ph.D. degree in electrical engineering from the Graduate University, Chinese Academy of Sciences, Beijing, China, in 2011. He is currently with the Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing. His current research interests include gas discharge and application. Dr. Zhang is a member of the Dielectrics and Electrical Insulation Society of the IEEE.
Zhi Fang (M’10) was born in Heilongjiang, China. He received the B.S., M.S., and Ph.D. degrees in electrical engineering from Xi’an Jiaotong University, Xi’an, China, in 1999, 2002, and 2005, respectively. He is currently an Associate Professor with the Nanjing University of Technology, Nanjing, China. His current research interests include atmospheric pressure gas discharge plasmas, and the development of atmospheric pressure plasma sources for materials surface processing applications.
Yang Yu was born in Shandong, China, in 1985. He received the B.Sc. degree in electrical engineering from the Shandong University of Technology, Zibo, China, in 2008, and the M.Sc. degree in electrical engineering from the Graduate University, Chinese Academy of Sciences, Beijing, China, in 2011. His current research interests include nanosecond pulse dielectric barrier discharge.
Dongdong Zhang was born in Shanxi, China, in 1980. He received the B.Sc. and M.Sc. degrees in electrical engineering from the Dalian University of Technology, Dalian, China, in 2002, and the Ph.D. degree in electrical engineering from the Graduate University, Chinese Academy of Sciences, Beijing, China, in 2008. He is currently with the Institute of Electrical Engineering, Chinese Academy of Sciences. His current research interests include pulsed power source.
Ping Yan (M’04) was born in Beijing, China, in 1965. She received the B.Sc. degree from the Department of Electrical Engineering, Tsinghua University, Beijing, in 1988, and the Ph.D. degree from Akita University, Akita-ken, Japan, in 2000. She is currently with the Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing. Her current research interests include high-voltage insulation and pulsed power.
Yuanxiang Zhou (M’04) was born in Fujian, China, in 1966. He received the B.E. degree from Tsinghua University, Beijing, China, in 1988, the M.E. degree from the Electrical Power Research Institute, China, in 1991, and the Ph.D. degree from Akita University, Akita-ken, Japan, in 1999. He engaged in insulation status monitoring and diagnosis of electrical equipment on site as he was serving for the China Electrical Power Research Institute, China, from 1991 to 1995. From 1999 to 2000, he was with the National Institute for Resources and Environment, National Institute of Advanced Industrial Science and Technology, Japan, as a New Energy and Industrial Technology Development Organization Fellow. He worked at the Massachusetts Institute of Technology, Cambridge, MA, USA as a visiting scientist in 2008. He is currently a Professor with Tsinghua University. His current research interests include organic and inorganic dielectrics, high-voltage technology and environmental protection, electrical equipment, and on-site detection and diagnosis. Dr. Zhou is the Deputy Secretary-General of the China Electrotechnical Society.
Edl Schamiloglu (M’90–SM’95–F’02) was born in The Bronx, NY, USA, in 1959. He received the B.S. degree from Columbia University, Columbia, NY, USA, in 1979, the M.S. degree in plasma physics from Columbia University in 1981, and the Ph.D. degree in engineering (minor in mathematics) from Cornell University, Ithaca, NY, USA, in 1988. He joined the University of New Mexico, Albuquerque, NM, USA, as an Assistant Professor, in 1988, and he is currently a Gardner-Zemke Professor of electrical and computer engineering and directs the Pulsed Power, Beams, and Microwaves Laboratory. He lectured at the U.S. Particle Accelerator School, Harvard University, Cambridge, MA, USA, in 1990, and at MIT in 1997. He has co-edited Advances in High Power Microwave Sources and Technologies (Piscataway, 2001) (with R. J. Barker) and he has co-authored High Power Microwaves (Taylor & Francis, 2007) (with J. Benford and J. Swegle). He has co-authored over 90 refereed journal papers, 170 reviewed conference papers, and four patents. His publications have been cited over 3000 times. He has been PI on over $18M of contracts and grants expenditured at UNM. Dr. Schamiloglu has received the Sandia National Laboratories Research Excellence Award as part of the Delphi/Minerva Team in 1991, the UNM School of Engineering Research Excellence Award twice (Junior Faculty in 1992 and Senior Faculty in 2001), the titles of UNM Regents’ Lecturer in 1996 and Gardner-Zemke Professor in 2000, and the Lawton-Ellis Award in 2004. He is a fellow of EMP (sponsored by the Summa Foundation), and an Associate Editor of the Journal of Electromagnetic Waves and Applications (JEMWA). He was the General Chair of the IEEE PPPS-2007 Conference, Albuquerque, NM, 2007. He was selected as the Outstanding Engineering Educator by the IEEE Albuquerque Section in 2008. He is a recipient of a 2011 and a 2012 UNMSTC Creativity Award.
