Journal of Modern Physics, 2012, 3, 1692-1696
http://dx.doi.org/10.4236/jmp.2012.330207 Published Online October 2012 (http://www.SciRP.org/journal/jmp)
Effect of Waveform Parameters on Pulsed Glow
Discharge in Air
Fengbo Tao1, Zhicheng Zhou1, Yong Ma1, Qiaogen Zhang2
1Jiangsu Electric Power Research Institute, Nanjing, China
2School of Electrical Engineering, Xi’an Jiaotong University, Xi’an, China
Email: hvtaofb@gmail.com
Received August 25, 2012; revised September 26, 2012; accepted October 4, 2012
ABSTRACT
The nanosecond single pulse was employed here to generate a large volume glow discharge between the wire-plane
electrodes in air. In order to find requirements on pulse waveform for generation of a large volume discharge at atmos-
pheric pressure, the effect of pulse risetime, pulse width, and amplitude on glow discharge has been widely investigated
in this paper. Results reveal that a large volume glow discharge can be generated in an inhomogeneous electric field
under the single pulse with the faster risetime, the lower peak amplitude. The pulse width has almost no influence on
the density of glow discharge, but which has a great influence on the transition from glow discharge to streamer dis-
charge. A model of inter-shielding-effect has been proposed to explain the influence of waveform parameters on pulsed
glow discharge.
Keywords: Pulsed Discharge; Large Volume Glo w Discharge; Inte r-Shielding-E f fect
1. Introduction
Gas plasma generated by pulse discharge is widely used
in ozone generation [1-3], detoxification of gaseous pol-
lution [4-6], material surface treatment [7,8], synthesis of
nanostructured material [9,10], etc. In order to improve
the efficiency of the applications mentioned above, the
homogeneous, large volume discharge is required. From
the previous works, it can be seen that nanosecond pulse
discharge in non-uniform electric field constructed by
needles-plane or wire-plane electrodes is an efficient way
to obtain large volume discharge. For instance, Ryo Ono
and Tetsuji Oda have used the pulsed discharge between
needles-plane electrodes to generate ozone and measured
the ozone distribution in the discharge gap. R. A. Roush
and R. K. Hutcherson focused on the exhaust gas detoxi-
fication efficiency by large volume discharge under dif-
ferent pulse risetime and pulse width [11-13]. Although
the large volume discharge is wid ely used in the industry
applications, the effect of waveform parameters on the
formation of large volume discharge is no t very clear yet.
In order to find requirements on pulse waveform for ge-
neration of a large volume discharge at atmospheric pres-
sure and understand more clearly the mechanism of pul-
sed discharge, a single pulse voltage with adjustable rise-
time, pulse width, amplitude etc is employed here to in-
vestigate the formation of the large volume discharge
between the wire-plane electrodes with non-uniform
electric field.
2. Experimental Set-Up
Figure 1 shows the schematic experimental circuit em-
ployed in this research. A single pulse with the pulse
risetime from 10 ns to 500 ns, pulse width from 100 ns to
2 µs, peak amplitude from 10 kV to 50 kV was applied
on the wire-plane electrodes. A rogowski brass plane of
60 mm in diameter was used as the cathode, and a brass
wire of 0.1 mm in diameter was used as the anode. Also,
the wire of 100 mm in length, longer than th e diameter of
the cathode, was used to prevent the edge effect of the
two electrodes. The gap distance was fixed at 8mm. Both
electrodes were sealed in a polymethyl methacrylate con-
tainer. The air pressure in the chamber can be adjusted in
the range of 2 kPa to 0.2 MPa.
The pulse voltage was measured with the voltage di-
vider consisting of (R1 & C1) and (R2 & C2), which has
the response time less than 5 ns. The discharge current
was measured with a current transducer (rogowski coil),
which has the response time less than 5 ns. The voltage
and current signals were recorded by a digitizin g oscillo-
scope (Tektronix DPO4054) with a bandwidth of 500
MHz. Considering that the discharge images are in cor-
respondence with one-off discharge process caused by
the nanosecond single pulse, there will be no superposi-
tion of repetitious discharges, so the exposure time is
C
opyright © 2012 SciRes. JMP
F. B. TAO ET AL. 1693
Figure 1. Schematic diagram of the electrode system and
the experimental set-up.
(a1)
(a2)
(b1)
(b2)
Figure 2. Pulsed discharge images with different pulse rise-
times at different air pressures ((a1) 20 ns/6 kPa; (a2) 200
ns/6 kPa; (b1) 20 ns/15 kPa; (b2) 200 ns/15 kPa).
determined by the pulsed discharge time. Furthermore,
due to the high resolution and the high sensitivity of the
ordinary camera, detail of the discharge channel can be
observed more clearly. Therefore, the images of the dis-
charges were captured by a camera (FUJIFILM FinePix
S6500) with a resolution of 2848 × 2136 pixels in a sin-
gle shot with the exposure time of (1/3) s.
3. Experimental Results and Discussions
3.1. Effect of Pulse Risetime
Figure 2 shows the images of pulsed glow discharge in
wire-plane air gap with the pulse risetime of 20 ns and
Figure 3. Schematic diagram of inter-shielding-effect be-
tween avalanches.
200 ns respectively. The pulse width and peak amplitude
are fixed at 200 ns and 20 kV respectively. Experiments
were carried out at the air pressure of 6 kPa and 15 kPa.
For all images in this paper, the upper is wire anode elec-
trode and the opposite is plane gr ound electrode.
From Figure 2, it can be seen that at the same gas
pressure, in the case of fast pulse with risetime of 20 ns,
the white spots called spot glow by Anatoly Nikolaevich
Maltsev near anode [14], are much more in number than
those in the case of the slow pulse with risetime of 200
ns. This can be explained by the inter-shielding-effect.
Because of the different delay times in the formation of
the primary electrons arou nd the anode, some avalanches
grow ahead and distort the electric field in the gap, which
leads to the suppressing of the adjacent avalanches. The
schematic diagram of inter-shielding-effect is illustrated
in Figure 3, in which E0 is the applied electric field, A is
the preceding avalanche, and B is the posterior avalanche,
ES is the space charge electric field formed by A around
the head of the posterior avalanche B. When ES is large
enough to cause the total electric field unsatisfied for the
development of B, the growth of the posterior avalanche
B will stop.
Assuming that there is the delay time (
0) of the pri-
mary electron formation. Avalanche A grows ahead of
avalanche B at the time of
0 early. The growth of ava-
lanche A results in accumulation of the space charges in
the head of avalanche A as well as the decrease of elec-
tric field ES around the head of the posterior avalanche B,
which leads to the suppressing of avalanche B and the
decrease of white spots around the anode. The following
expression can also clarify the effect of pulse risetime on
glow discharge [15] .

12
0
0
1,40
2cc c
E
ntrrnQ
n

 

 (1)
where, t0 is the pulse risetime when applied E/n has
reached (E/n)0, rc is critical radius of avalanche head, μ is
the electron mobility, ξc is critical length of avalanche, Q
Copyright © 2012 SciRes. JMP
F. B. TAO ET AL.
1694
is the averaged momentum transfer cross section. From
expression (1), we can see that with the decrease of pulse
risetime, expression (1) can be quite easily satisfied, i.e.
the overlapping condition between electron avalanches
can be fulfilled. Furthermore, with the increase of air
pressure, the risetime of pulse is required to decrease
further to fulfill expression (1).
3.2. Effect of Pulse Amplitude
Figure 4 shows the images of the pulsed glow discharge
under the pulse amplitude of 20 kV and 35 kV respec-
tively. The experiments were carried out at different gas
pressures: 20 kPa, 40 kP a and 60 kPa. The pulse risetime
and pulse width are fixed at 20 ns and 120 ns respec-
tively.
In Figure 4, the number of discrete channels in pulsed
glow discharge decreases with the increase of peak am-
plitude. Moreover, the discrete channels can easily come
(a1)
(a2)
(b1)
(b2)
(c1)
(c2)
Figure 4. Pulse discharge images with different pulse peak
amplitude at different air pressure ((a1) 20 kPa/20 kV; (a2)
20 kPa/35 kV; (b1) 40 kPa/20 kV; (b2) 40 kPa/35 kV; (c1)
60 kPa/20 kV; (c2) 60 kPa/35 kV).
into streamer discharge (the white channels illustrated in
Figures 4(a2)-(c2) in the case of higher pulse amplitude
due to the current heat effect. As mentioned above, the
early or late growth of avalanches will form the uneven
distribution of the current in each discrete channel. With
the increase of the pulse amplitude, the higher overvolt-
age promotes the growth of avalanche A as well as the
formation of space charge electric field (ES), which re-
sults in the further increase of inter-shielding-effect on
avalanche B. Therefore, the increase of pulse amplitude
will cause the decrease of the discrete channel number as
well as the increase of current in each channel. Due to
the uneven distribution and the increase of the current in
discrete channels, the early propagated ch annel will tran-
sit into streamer or spark discharge, which results in the
current and the luminescence decrease in other discrete
channels, as shown in Figures 4(a2)-(c2).
The discharge current waveforms with different pulse
amplitude at the air pressure of 20 kPa are illustrated in
Figures 5(a) and (b). It can be found that there are some
fluctuations on the current waveforms, which are related
to the formation of avalanches during the discharge
process. Due to the greater inhibitio n between av alanch es
at high pulse amplitude, the number of fluctuations on
the current waveform under 35 kV pulse (Figure 5(b)) is
(a)
(b)
Figure 5. Current waveforms of pulse discharge under dif-
ferent amplitude in 20 kPa. (a) Up = 20 kV; (b) Up = 35 kV.
Copyright © 2012 SciRes. JMP
F. B. TAO ET AL. 1695
less than that under 20 kV pulse (Figure 5(a)).
From Figure 5, It can also be found that due to the
streamer discharge, the peak discharge current under 35
kV pulse is larger than that under 20 kV pulse, the volt-
age collapse time under 35 kV pulse is less than that un-
der 20 kV pulse. A slight oscillation on the tail of the
current waveform under the pulse amplitude of 35 kV
can also be seen, which is caused by the streamer dis-
charge illustrated in Figure 4.
3.3. Effect of Pulse Width
Effect of the pulse width on the pulsed glow discharge is
investigated by varying the pulse width from 120 ns to
1600 ns at the air pressure of 40 kPa. Figure 6 illustrates
the images of large volume discharges under different
pulse width with the fixed pulse risetime of 20 ns and
peak amplitude of 20 kV.
From Figure 6, it can be clarified that the pulse width
in the range from 120 ns to 1600 ns has little influence
on the number and luminescence of the discrete channels
in the pulsed glow discharge, but the increase of the
pulse width will cause the transition from glow discharge
to streamer or spark discharge easily. As discussed in
part A, the discharge mode in the gap applied by pulsed
voltage is dominated by the pulse risetime. Therefore, fo r
a given pulse risetime, the number of the discrete chan-
nels is basically invariable and is not affected by the
pulse width as shown in Figure 6. However, when in-
creasing the pulse width, the higher energy will be in-
jected into the discharge channel, which results in the
(a)
(b)
(c)
(d)
Figure 6. Images of pulse discharge with different pulse
width (a) 120 ns; (b) 200 ns; (c) 500 ns; (d) 1600 ns.
transition from glow discharge to streamer or spark dis-
charge in the early propagated channel due to the uneven
distribution of th e current. In Figure 6, the luminescence
of the other discrete channels is nearly not influenced by
the pulse risetime, which can be considered that the in-
creased pulse energy is almost injected into the early
propagated channel with the increase of pulse width.
The discharge current waveforms under different pulse
width are illustrated in Figure 7, the part in the range of
0 - 150 ns is zoomed at the top right corner. From Figure
7, it can be seen that at the first 60 ns, the pulse discharge
current waveforms for different pulse width are almost
the same, which is controlled by the growth of the pulsed
glow discharge channels as shown in Figure 6, i.e. the
number of the discrete channels as well as its distribution
is determined by the pulse risetime. After the pulse rea-
ches the peak value, the discharge development is deter-
mined by the pulse width. With the increase of the pulse
width, the increased energy is injected into the early pro-
pagated channel, resulting in the increasing of the dis-
charge current including the increase of the luminescence
in the early propagated channel.
4. Conclusion
In this paper, pulsed glow discharge is generated in air
between wire and plane electrodes with non-uniform
electric field. The effect of pulse risetime, pulse width
and amplitude on glow discharge has been widely inves-
tigated. Results reveal that a large volume glow dis-
charge can be generated in an inhomogeneous electric
field more easily under the pulse with faster risetime.
The pulse width has almost no influence on the mode of
glow discharge, but has a great influence on the transi-
tion from glow discharge to streamer discharge. A model
of inter-shielding-effect has b een proposed to explain the
Figure 7. Pulse discharge current under different pulse
width.
Copyright © 2012 SciRes. JMP
F. B. TAO ET AL.
Copyright © 2012 SciRes. JMP
1696
influence of waveform parameters on the pulsed glow
discharge.
REFERENCES
[1] W. Samaranayake, Y. Miyahara, T. Namihira, et al.,
“Ozone Generation in Dry Air Using Pulsed Discharges
with and without a Solid Dielectric Layer,” IEEE Trans-
actions on Dielectrics and Electrical Insulation, Vol. 8,
No. 4, 2001, pp. 687-697. doi:10.1109/94.946724
[2] R. Ono and T. Oda, “Ozone Production Process in Pulsed
Positive Dielectric Barrier Discharge,” Journal of Physics
D: Applied Physics, Vol. 40, No. 1, 2007, pp. 176-182.
doi:10.1088/0022-3727/40/1/011
[3] M. Simek and M. Clupek, “Efficiency of Ozone Produc-
tion by Pulsed Positive Corona Discharge in Synthetic
Air,” Journal of Physics D: Applied Physics, Vol. 35, No.
11, 2002, pp. 1171-1175.
doi:10.1088/0022-3727/35/11/311
[4] Y. Yankelevich and A. Pokryvailo, “High-Power Short-
Pulsed Corona: Investigation of Electrical Performance,
SO2 Removal, and Ozone Generation,” IEEE Transactions
on Plasma Science, Vol. 30, No. 5, 2002, pp. 1975-1981.
doi:10.1109/TPS.2002.805320
[5] F. Fresnet, G. Baravian, L. Magne, et al., “Influence of
Water on NO Removal by Pulsed Discharge in N2/H2O/
NO Mixtures,” Plasma Source Science and Technology,
Vol. 11, No. 2, 2002, pp. 152-160.
doi:10.1088/0963-0252/11/2/305
[6] T. Namihira, S. Tsukamoto, D. Wang, et al., “Influence of
Gas Flow Rate and Reactor Length on NO Removal Us-
ing Pulsed Power,” IEEE Transactions on Plasma Sci-
ence, Vol. 29, No. 4, 2001, pp. 592-598.
doi:10.1109/27.940952
[7] T. Yuji and Y. Sung, “Surface Treatment of TiO2 Films
by Pulse Plasma for Dye-Sensitized Solar Cells Applica-
tion,” IEEE Transactions on Plasma Science, Vol. 35, No.
4, 2007, pp. 1010-1013. doi:10.1109/TPS.2007.896918
[8] N. Takeuchi, K. Yasuoka and S. Ishii, “Surface Modifica-
tion of Thin Rods by Theta-Pinching Metallic Plasmas,”
IEEE Transactions on Plasma Science, Vol. 34, No. 4,
2006, pp. 1112-1115. doi:10.1109/TPS.2006.876484
[9] W. D. Yang, P. N. Wang, F. M. Li and K. W. Cheah,
“Synthesis of Oxygen-Free Nanosized InN by Pulse Dis-
charge,” Nanotechnology, Vol. 13, No. 1, 2002, pp. 65-68.
doi:10.1088/0957-4484/13/1/315
[10] H. Yang, W. Nam and D. Park, “Production of Nanosized
Carbon Black from Hydrocarbon by a Thermal Plasma,”
Journal of Nanoscience and Nanotechnology, Vol. 7, No.
11, 2007, pp. 3744-3749. doi:10.1166/jnn.2007.003
[11] R. Ono and T. Oda, “Dynamics of Ozone and OH Radi-
cals Generated by Pulsed Corona Discharge in Humid-Air
Flow Reactor Measured by Laser Spectroscopy,” Journal
of Applied Physics, Vol. 93, No. 10, 2003, pp. 5876-5882.
doi:10.1063/1.1567796
[12] R. Roush, R. Hutcherson, M. Ingram, et al., “Effects of
Pulse Risetime and Pulse Width on the Destruction of
Toluene and NOX in a Coaxial Pulsed Corona Reactor,”
Proceeding of 22nd IEEE International Power Modulator
Symposium, Boca Raton, 25-27 June 1996, pp. 79-84.
[13] D. Wang, M. Jikuya, S. Yoshida, et al., “Positive- and Ne-
gative-Pulsed Streamer Discharges Generated by a 100-ns
Pulsed-Power in Atmospheric Air,” IEEE Transacitions
on Plasma Science, Vol. 35, No. 4, 2007, pp. 1098-1103.
doi:10.1109/TPS.2007.902132
[14] A. N. Maltsev, “Dense Gas Discharge with Runaway Elec-
trons as a New Plasma Source for Surface Modification
and Treatment,” IEEE Transactions on Plasma Science,
Vol. 34, No. 4, 2006, pp. 1166-1174.
doi:10.1109/TPS.2006.878994
[15] J. I. Levatter and S.-C. Lin, “Necessary Conditions for the
Homogeneous Formation of Pulsed Avalanche Discharge
at High Gas Pressure,” Journal of Applied Physics, Vol.
51, No. 1, 1980, pp. 210-222. doi:10.1063/1.327412