Surface charge in dielectric barrier discharge plasma actuators

D. F. Opaits, M. N. Shneider, Richard B. Miles, A. V. Likhanskii, and S. O. Macheret

Citation: Physics of Plasmas (1994-present) 15, 073505 (2008); doi: 10.1063/1.2955767

View online: http://dx.doi.org/10.1063/1.2955767
View Table of Contents: http://scitation.aip.org/content/aip/journal/pop/15/7?ver=pdfcov
Published by the AIP Publishing

Articles you may be interested in

Characterization of nanosecond pulse driven dielectric barrier discharge plasma actuators for aerodynamic flow
control
J. Appl. Phys. 113, 103302 (2013); 10.1063/1.4794507

Spatiotemporal structure of a millimetric annular dielectric barrier discharge plasma actuator

Phys. Fluids 25, 017103 (2013); 10.1063/1.4774334

Airflow influence on the discharge performance of dielectric barrier discharge plasma actuators
Phys. Plasmas 19, 073509 (2012); 10.1063/1.4736995

Forcing mechanisms of dielectric barrier discharge plasma actuators at carrier frequency of 625 Hz
J. Appl. Phys. 110, 113301 (2011); 10.1063/1.3664695

Power consumption, discharge capacitance and light emission as measures for thrust production of dielectric
barrier discharge plasma actuators
J. Appl. Phys. 110, 013305 (2011); 10.1063/1.3603030

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.239.20.174 On: Tue, 25 Nov 2014 03:58:24
 PHYSICS OF PLASMAS 15, 073505 共2008兲

Surface charge in dielectric barrier discharge plasma actuators

D. F. Opaits,1,a兲 M. N. Shneider,1 Richard B. Miles,1 A. V. Likhanskii,2 and S. O. Macheret3
1
Department of Mechanical and Aerospace Engineering, Princeton University, Princeton,
New Jersey 08544, USA
2
The Pennsylvania State University, University Park, Pennsylvania 16802, USA
3
Lockheed Martin Aeronautics Company, Palmdale, California 93599, USA
共Received 10 April 2008; accepted 17 June 2008; published online 21 July 2008兲

Direct measurements of the dielectric surface potential and its dynamics in asymmetric dielectric
barrier discharge 共DBD兲 plasma actuators show that the charge builds up at the dielectric surface
and extends far downstream of the plasma. The surface charge persists for a long time 共tens of
minutes兲 after the driving voltage has been turned off. For a sinusoidal voltage waveform, the
dielectric surface charges positively. With the voltage waveform consisting of nanosecond pulses
superimposed on a dc bias, the sign of the dielectric surface charge is the same as the sign 共polarity兲
of the bias voltage. The surface charging significantly affects DBD plasma actuator performance.
© 2008 American Institute of Physics. 关DOI: 10.1063/1.2955767兴

I. INTRODUCTION II. SURFACE POTENTIAL MEASUREMENTS

A. Experimental setup
Asymmetric dielectric barrier discharge 共DBD兲 plasma
actuators have been reported in a number of works which For nonintrusive surface potential measurements we
have addressed their potential use for aerodynamic used the Trek Model 247-3 Electrostatic Voltmeter with Trek
control.1–22,24–29 This application relies on a DBD-induced Model 6000B-13C Electrostatic Voltmeter Probe. It has high
resolution 共around 1 mm兲, fast response time 共less than 3 ms
wall jet with typical velocities of several meters per second.
for a 1 kV step兲, and an operating range from 0 to ⫾3 kV dc
It is understood that electric charge deposition on the dielec-
or peak ac.
tric surface limits the current and thus allows operation of The description of the asymmetric DBD plasma actuator
nonequilibrium discharges at atmospheric pressure. The im- 共Fig. 1兲 can be found in our earlier paper.24 The electrodes
portant role of the surface charge and its finite relaxation were made of copper foil. The width of the electrodes was
time 共the “memory voltage”兲 in the discharge physics was equal to 25 mm and their spanwise dimension was 75 mm. A
pointed out by Massines et al.30 and Gadri31 for the parallel- 100 micron thick kapton tape was used as the dielectric. The
plate DBD configuration, and by Roth et al.32 for One Atmo- discharge was ignited by applying high voltage to one of the
sphere Uniform Glow Discharge Plasma 共OAUGDP™兲. electrodes. The other electrode was grounded.
Since the surface charge plays an important role in DBD
physics, it warrants a special study. In the work of Font
et al.,21 the surface potential in a DBD driven by sinusoidal B. Sinusoidal voltage experiments
voltage was measured using a V-dot probe technique, and the First, experiments were conducted for the 3 kHz 10 kV
voltage across the plasma was shown to be less than half of peak-to-peak sinusoidal voltage profile. Since the sinusoidal
that which is applied across the electrodes. Recently the profile is widely used in DBD experiments,1–6,13–15,17–20 it is
same group22 presented the time evolution and the spatial interesting to compare its surface potential distribution with
extent of the surface potential. In this paper, we describe the one obtained in pulsed DBD.16,24–29 The actuator was run
electrostatic voltmeter measurements of surface charge in for 15 s and then the surface potential was measured. The
DBD plasma actuators driven by various voltage profiles, result is presented in Fig. 2, where the distance is measured
from the edge of the exposed electrode. The plasma extends
most notably by pulses plus dc bias but also by sinusoidal
only a few millimeters downstream of the edge of the ex-
voltages. The work reported here extends our previously re-
posed electrode, whereas the dielectric charge extends centi-
ported work which also used the electrostatic probe.24 The meters away from it, far downstream of the plasma. The
results for the sinusoidal profile recently reported by Enloe maximum surface potential 共1.5 kV兲 is comparable to the
et al.22 are consistent with the work reported here and with applied peak voltage 共5 kV兲. This means that the electric
our previous results. For sinusoidal voltages the two ap- field of the surface charge can significantly distort the elec-
proaches are mutually complementary since Enloe et al. tric field produced by the voltage applied to the electrodes.
measure the surface potential using a different method. Also, even though the applied voltage is symmetrical 共sinu-
soidal with no bias voltage兲 the time-averaged surface charge
has a positive polarity. This is possibly due to the much
a兲
Electronic mail: [email protected]. higher mobility of electrons compared to positive ions which

1070-664X/2008/15共7兲/073505/5/$23.00 15, 073505-1 © 2008 American Institute of Physics

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.239.20.174 On: Tue, 25 Nov 2014 03:58:24
 073505-2 Opaits et al. Phys. Plasmas 15, 073505 共2008兲

FIG. 1. 共Color online兲 Asymmetric dielectric barrier discharge plasma

actuator.

leads to a much higher electron current into the exposed

electrode during the positive half-cycle of the sinusoidal
voltage compared to the positive ion current into the exposed
electrode during the negative half cycle. With this reasoning FIG. 2. Surface potential distribution. Applied voltage profile: Sinusoidal
3 kHz, 10 kV peak-to-peak. Electrodes edge at 0.
one might conclude that there is a net negative current into
the exposed electrode and positive current flow from the
DBD to infinity. This picture is consistent with results from
creates a noticeable surface charge. As the number of pulses
other two experiments described in our earlier work.24 Simi-
increases, the surface charge builds up and moves farther
lar behavior is observed in asymmetric rf discharges, where
away from the plasma. After thousands of pulses, the surface
there is a positive current flux to a virtual electrode at
charge is close to its steady-state value. We have also found
infinity.23
that 2 kV bias voltage itself builds up some surface charge at
The persistence of surface charge far downstream of the
the dielectric. The charge is smaller then the one built by a
plasma can be due either to surface mobility of the charges
single pulse but is still significant. We have also found that
or to deposition of the charges onto the surface from the gas
positive pulses added to positive bias are significantly more
jet downstream of the plasma. Estimates show that while the
efficient in depositing the charge on the surface than other
dielectric surface charge mobility is too low, the deposition
combinations of the pulse and bias polarities. This effect is
of ions from the DBD-generated tangential gas jet is likely to
illustrated in Fig. 5.
persist downstream of the plasma. Indeed, in the DBD gen-
The purpose of the next set of experiments was to study
erated jet downstream of the visible plasma, plasma elec-
the evolution of the surface charge with time after the dis-
trons, no longer regenerated by ionization, rapidly 共in tens of
charge has been turned off. The results for positive polarity
nanoseconds兲 recombine with positive ions and attach to
of the bias voltage are presented in Fig. 6. 共In the case of
oxygen molecules, which results in an ion-ion plasma jet
negative polarity the potential distribution looks very
with an estimated ion-ion recombination time on the order of
similar.24兲 As seen in Fig. 6, the charge persists for a very
several milliseconds. At a jet velocity of several m/s, this
long time. In 30 min it depletes by only 15%–20%. There is
translates into several centimeters of ion-ion plasma and thus
uniformly slow charge depletion from the surface in the re-
into several centimeters of dielectric surface charge.
gion of low gradients of the charge concentration and a rela-
tively fast depletion close to the exposed electrode in the
C. Pulses plus dc bias experiments
region of high gradients. The fact that the depletion occurs at
In this set of experiments the voltage profile consisted of the same rate for both positive and negative charge is con-
3 kV 8-ns 共FWHM兲 ionization pulses superimposed on 2 kV sistent with the charged particles at the dielectric surface
dc bias voltage. The pulse repetition rate was 10 kHz and the being positive and negative ions from the gas rather than
plasma was run for 10 s. The results of surface potential electrons.
measurements are presented in Fig. 3. As in the case of a An experiment was conducted with a plasma actuator in
sinusoidal voltage, the charge builds up far downstream the which a conventional Kapton tape was substituted by a low-
plasma region. As can be seen from Fig. 3, the magnitude of static Kapton tape of the same thickness.24 The results show
the surface potential is quite close to that of the bias voltage, that the depletion rate is faster for the low-static Kapton tape,
while the magnitude of the ionization pulse voltage has little but the characteristic time is still of the order of tens of
effect. minutes.
We next examined how quickly the surface charge builds Note that Roth et al.32 also observed the slow depletion
up. At a selected dc bias voltage, a predetermined number of of the residual surface charge on the dielectric after the driv-
pulses were sent to the actuator, and then the surface poten- ing voltage was turned off. The typical time periods were up
tial was measured. A typical result is shown at Fig. 4. Even to 30 s, which is much shorter than the times we observed.
though for a small number of pulses the accuracy of mea- This difference may be associated with the difference in di-
surements is low, it can be seen that even a single pulse electric materials and configuration of the electrodes.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.239.20.174 On: Tue, 25 Nov 2014 03:58:24
 073505-3 Surface charge in dielectric barrier discharge plasma actuators Phys. Plasmas 15, 073505 共2008兲

FIG. 3. 共Color online兲 Surface potential distribution. Applied voltage profile: 2 kV dc bias, 3 kV pulses at 10 kHz PRR.

III. CALCULATION OF THE SURFACE CHARGE lated 2D electric potential distribution for the case of 3 kV
DENSITY negative pulses, 2 kV positive dc bias, right after the applied
voltage was turned off. Figure 8 shows the calculated surface
In order to interpret the experimental data and to infer
charge for the case of 3 kV negative pulses, 2 kV positive
the surface charge density from the measured surface poten-
bias. As seen in Fig. 8, the surface charge is mainly located
tial, we used a numerical model based on the two-
⬃5 – 20 mm downstream of the exposed electrode. The peak
dimensional Poisson solver for the electric potential. We con-
surface charge density corresponds to ⬃1 ion per 5
sidered the DBD plasma actuator consisting of 2 electrodes
⫻ 5 共nm兲 共Ref. 2兲 site, which is a coverage of a few percent.
of length 2.5 cm, separated by the 100 micron Kapton tape,
i.e., the same configuration as in the experiments. The elec-
IV. THREE-ELECTRODE CONFIGURATION
trodes were assumed to be infinitely thin. The size of the OF DBD PLASMA ACTUATORS
computational domain was 25 mm by 1 mm. The computa-
tional domain included the entire lower 共encapsulated兲 elec- One of the results of our numerical simulations7,28 of
trode and 5 mm of the upper 共exposed兲 electrode. The reason DBD actuators driven by repetitive nanosecond pulses super-
for neglecting the remaining part of the upper electrode is its imposed on a dc bias was the prediction of a significant
insignificant contribution to the electric potential on the sur- effect of the dc bias voltage on the plasma produced tangen-
face. Since the electric potential distribution was measured tial gas velocity and thrust. However, the experiments26 did
experimentally after the DBD was turned off, and the recom- not show such an effect. The reason was found to be the
bination time of the plasma is of the order of microseconds, surface charge which builds up in tens or hundreds of pulses
the gas above the dielectric was considered noncharged. and which shields the applied bias field. Indeed, the DBD
In the computations, the distribution of charge on the induced tangential flow velocity and thrust depend quite
dielectric surface that would match the experimentally mea- strongly on the voltage amplitude.4–7,15,24–28 Thus, a reduc-
sured profile was obtained. Thus, having the experimentally tion in the effective voltage on the plasma caused by the
measured potential, we reproduced the 2D electric potential surface charge results in a significant decrease in the actuator
distribution and the surface charge for each temporal mea- performance. The numerical model7,27 could not include
surement as the charge decayed. Figure 7 shows the calcu- large numbers of pulses and intervals between those due to

FIG. 4. 共Color online兲 Surface potential for a fixed number of pulses. Ap-
plied voltage profile: Positive 2 kV dc bias, negative 3 kV pulses at 10 kHz FIG. 5. 共Color online兲 Surface potential after five ionizing pulses for differ-
PRR. ent pulses and bias polarities.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.239.20.174 On: Tue, 25 Nov 2014 03:58:24
 073505-4 Opaits et al. Phys. Plasmas 15, 073505 共2008兲

FIG. 6. 共Color online兲 Surface potential vs time for positive surface charge.
共Negative 3 kV pulses, positive 2 kV dc voltage.兲

FIG. 8. 共Color online兲 Calculated charge density on the dielectric surface at

different moments of time after the DBD was turned off. 共Negative 3 kV
numerical difficulties; the total number of pulses in the com- pulses, positive 2 kV dc voltage.兲
putations never exceeded 4–5. Since several pulses deposit
only little charge on the surface, the modeling did not cap-
ture the potential shielding and the performance decrease plasma can be expected to result in enhanced induced gas
that occur after hundreds and thousands of pulses. velocity and thrust.
As a solution to the surface charge build up problem, it Note that if a relatively thick dielectric layer is used,
was proposed26 to switch the bias periodically. This im- then simply cutting the dielectric layer off over a part of the
proved the DBD actuator performance.26 However, in this lower electrode would produce a “step” which can be detri-
paper we found that the surface charge builds up far down- mental to aerodynamic drag. A simple modification of the
stream from the plasma. It would therefore be desirable to idea would then be to put a third electrode on the surface and
remove the surface charge as it builds up downstream of the to electrically connect this electrode with the encapsulated
exposed electrode. We therefore proposed a modification to electrode, rather than cutting the dielectric off.
the actuator configuration. In the new configuration, the di- The three-electrode DBD configuration resembles that of
electric covers only part of the lower electrode, leaving the the sliding discharge.9–11 However, in sliding discharges the
rest of the electrode exposed, as shown in Fig. 9. In such a plasma actually reaches both exposed electrodes. In contrast,
case the surface charge cannot build up far downstream and in the three-electrode DBD discharge proposed here, the
the dc bias voltage is not shielded. Some preliminary results electrons would not reach the auxiliary electrode whose pur-
of thrust measurements can be found in Ref. 24. The surface pose is simply to reduce or eliminate the charge accumula-
potential profile for the configuration of Fig. 9 after 10 s of tion on the dielectric surface.
running the plasma is shown in Fig. 10. A comparison of
Figs. 3 and 10 shows that the surface charge has indeed been V. CONCLUSIONS
substantially reduced by exposing the electrode downstream
of the plasma. Although gas dynamic measurements are out- In this work, surface charge dynamics in DBD plasma
side the scope of this paper, the surface charge reduction actuators were studied. Direct dielectric surface potential
accomplished by exposing the electrode downstream of the measurements showed that the electric field of the surface
charge can dramatically distort the field produced by the
voltage applied to the electrodes and thus reduce the actuator

FIG. 7. 共Color online兲 Calculated 2D electric potential distribution based on

the experimental data for the electric potential on the dielectric surface right
after the voltage applied to DBD was turned off. 共Negative 3 kV pulses,
positive 2 kV dc voltage.兲 FIG. 9. 共Color online兲 Modified DBD plasma actuator.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.239.20.174 On: Tue, 25 Nov 2014 03:58:24
 073505-5 Surface charge in dielectric barrier discharge plasma actuators Phys. Plasmas 15, 073505 共2008兲

10
C. Louste, E. Moreau, and G. Touchard, Proceedings of the ESA-IAS-IEJ
Joint Symposium, Berkeley 共Laplacian Press, Morgan Hills, CA, 2006兲.
11
B. Arad, Y. Gazit, and A. Ludmirsky, J. Phys. D 20, 360 共1987兲.
12
D. Orlov, G. Font, and D. Edelstein, Proceedings of the 46th AIAA Aero-
space Sciences Meeting and Exhibit, Reno, NV, 2008 共AIAA, Washington,
D.C., 2008兲, Paper No. AIAA-2008-1409.
13
C. A. Borghi, A. Cristofolini, M. Carraro, and G. Neretti, Proceedings of
the 37th AIAA Plasma Dynamics and Lasers Conference, San Francisco,
CA, 2006 共AIAA, Washington, D.C., 2006兲, Paper No. AIAA-2006-3380.
14
Y. V. Shcherbakov, N. S. Isanov, N. D. Baryshev, V. S. Frolovskij, and V.
S. Syssoev, Proceedings of the 31st AIAA Plasma Dynamics and Lasers
Conference, Denver, CO, 2000 共AIAA, Washington, D.C., 2000兲, Paper
No. AIAA-2000-2670.
15
M. L. Post and T. C. Corke, Proceedings of the 43th AIAA Aerospace
Sciences Meeting and Exhibit, Reno, NV, 2005 共AIAA, Washington, D.C.,
2005兲, Paper No. AIAA-2005-563.
16
D. F. Opaits, D. V. Roupassov, A. Yu. Starikovskii, I. N. Zavialov, and S.
G. Saddoughi, Proceedings of the 43th AIAA Aerospace Sciences Meeting
and Exhibit, Reno, NV, 2005 共AIAA, Washington, D.C., 2005兲, Paper No.
AIAA-2005-1180.
17
FIG. 10. 共Color online兲 Surface potential distribution at modified DBD L. S. Hultgren and D. E. Ashpis, Proceedings of the 41th AIAA Aerospace
plasma actuator. Applied voltage profile: 2 kV dc bias, 3 kV pulses at Sciences Meeting and Exhibit, Reno, NV, 2003 共AIAA, Washington, D.C.,
10 kHz PRR. 2003兲, Paper No. AIAA-2003-1025.
18
J. R. Roth, Phys. Plasmas 10, 2117 共2003兲.
19
S. Grundmann and C. Tropea, Proceedings of the 46th AIAA Aerospace
Sciences Meeting and Exhibit, Reno, NV, 2008 共AIAA, Washington, D.C.,
performance. For a sinusoidal voltage profile, significant 20
2008兲, Paper No. AIAA-2008-1369.
positive charge builds up at the dielectric surface in the re- A. Hoskinson and N. Hershkowitz, Proceedings of the 46th AIAA Aero-
space Sciences Meeting and Exhibit, Reno, NV, 2008 共AIAA, Washington,
gion far downstream of the plasma. For the “pulses plus dc D.C., 2008兲, Paper No. AIAA-2008-1370.
bias” voltage waveform the dielectric surface potential de- 21
G. I. Font, C. L. Enloe, T. E. McLaughlin, and D. Orlov, Proceedings of
pends primarily on the bias voltage. In particular, the sign of the 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, NV, 2007
the surface potential coincides with the bias polarity. The 共AIAA, Washington, D.C., 2007兲, Paper No. AIAA-2007-188.
22
C. L. Enloe, T. E. McLaughlin, J. W. Gregory, R. A. Medina, and W. S.
surface charge builds up relatively quickly 共in a few pulses兲 Miller, Proceedings of the 46th AIAA Aerospace Sciences Meeting and
and depletes slowly 共in tens of minutes兲. Based on the results Exhibit, Reno, NV, 2008 共AIAA, Washington, D.C., 2008兲, Paper No.
of the surface potential measurements, a three-electrode con- AIAA-2008-1103.
23
Yu. P. Raizer, M. N. Shneider, and N. A. Yatsenko, Radio-Frequency
figuration of DBD plasma actuators was proposed.
Capacitive Discharges 共CRC, Boca Raton, 2006兲.
24
D. Opaits, A. Likhanskii, M. Shneider, G. Neretti, S. Zaidi, S. Macheret,
ACKNOWLEDGMENTS and R. Miles, Proceedings of the 46th AIAA Aerospace Sciences Meeting
and Exhibit, Reno, NV, 2008 共AIAA, Washington, D.C., 2008兲, Paper No.
This work was supported in part by NASA Glenn Re- AIAA-2008-1372.
25
search Center 共Dr. David Ashpis兲, and in part by the Air A. Likhanskii, V. Semak, D. Opaits, M. Shneider, R. Miles, and S. Mach-
eret, Proceedings of the 46th AIAA Aerospace Sciences Meeting and Ex-
Force Office of Scientific Research 共Dr. John Schmisseur兲. hibit, Reno, NV, 2008 共AIAA, Washington, D.C., 2008兲, Paper No. AIAA-
2008-1380.
1
E. Moreau, J. Phys. D 40, 605 共2007兲. 26
D. Opaits, A. Likhanskii, M. Shneider, G. Neretti, S. Zaidi, S. Macheret,
2
J. R. Roth, D. M. Sherman, and S. P. Wilkinson, Proceedings of the 46th and R. Miles, Proceedings of the 38th Plasma Dynamics and Lasers Con-
AIAA Aerospace Sciences Meeting and Exhibit, Reno, NV, 1998 共AIAA, ference, Miami, FL, 2007 共AIAA, Washington, D.C., 2007兲, Paper No.
Washington, D.C., 1998兲, Paper No. AIAA-1998-328. AIAA-2007-4532.
3
J. Pons, E. Moreau, and G. Touchard, J. Phys. D 38, 3635 共2005兲. 27
A. Likhanskii, M. Shneider, D. Opaits, S. Macheret, and R. Miles, Pro-
4
D. Orlov, T. Corke, and M. Patel, Proceedings of the 46th AIAA Aerospace ceedings of the 38th Plasma Dynamics and Lasers Conference, Miami,
Sciences Meeting and Exhibit, Reno, NV, 2006 共AIAA, Washington, D.C., FL, 2007 共AIAA, Washington, D.C., 2007兲, Paper No. AIAA-2007-4533.
28
2006兲, Paper No. AIAA-2006-1206. A. Likhanskii, M. Shneider, S. Macheret, and R. Miles, Proceedings of the
5
C. L. Enloe, T. E. McLaughlin, R. D. VanDyken, K. D. Kachner, E. J. 44th AIAA Aerospace Sciences Meeting and Exhibit, Reno, NV, 2006
Jumper, and T. C. Corke, AIAA J. 42, 589 共2004兲. 共AIAA, Washington, D.C., 2006兲, Paper No. AIAA-2006-1204.
6 29
C. L. Enloe, T. E. McLaughlin, R. D. VanDyken, K. D. Kachner, A. Likhanskii, M. Shneider, S. Macheret, and R. Miles, Proceedings of the
E. J. Jumper, T. C. Corke, M. Post, and O. Haddad, AIAA J. 42, 595 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, NV, 2007
共2004兲. 共AIAA, Washington, D.C., 2007兲, Paper No. AIAA-2007-633.
7 30
A. V. Likhanskii, M. N. Shneider, S. O. Macheret, and R. B. Miles, J. F. Massines, R. B. Gadri, A. Rabehi, Ph. Decomps, P. Segur, and Ch.
Appl. Phys. 103, 053305 共2008兲. Mayoux, J. Appl. Phys. 83, 2950 共1998兲.
8
J. P. Boeuf and L. C. Pitchford, J. Appl. Phys. 97, 103307 共2005兲. 31
R. Ben Gadri, IEEE Trans. Plasma Sci. 27, 36 共1999兲.
9 32
C. Louste, G. Artana, E. Moreau, and G. Touchard, J. Electrost. 63, 615 J. R. Roth, J. Rahel, X. Dai, and D. M. Sherman, J. Phys. D 38, 555
共2005兲. 共2005兲.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.239.20.174 On: Tue, 25 Nov 2014 03:58:24