G Model

AEUE-51564; No. of Pages 5 ARTICLE IN PRESS

Int. J. Electron. Commun. (AEÜ) xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Electronics and

Communications (AEÜ)
journal homepage: www.elsevier.com/locate/aeue

Experimental study on the surface wave driven plasma antenna

Fatemeh Sadeghikia ∗ , Mahmoud Talafi Noghani, Mohammad Reza Simard
Aerospace Research Institute, Shahrak Qods, P.O. Box 1465774111, Tehran, Iran

a r t i c l e i n f o a b s t r a c t

Article history: This experimental study examines the characteristics of a plasma antenna. For these experiments, we
Received 7 September 2015 use a surface wave excitation at 450 MHz with RF power levels up to 40 W and gas pressure between
Accepted 31 January 2016 0.25 mb and 0.6 mb. Measurements of visible light from the plasma column confirm that the length of
the antenna is linearly proportional to the square root of the excitation power and the proportionality
Keywords: factor is a function of gas pressure. Observations of the VSWR of the antenna show that decreasing the
Plasma antenna
excitation power and increasing the pressure lead to the reduction of the resonant length of the antenna.
Surface wave driven excitation
The gain of the plasma antenna was measured between 110 and 300 MHz. The results show that the gain
Antenna measurement
Radiation characteristics
of a plasma antenna is considerably less than a conventional monopole antenna.
VSWR © 2016 Elsevier GmbH. All rights reserved.

1. Introduction powers, considering constant excitation frequency of 450 MHz to

The difficulty in changing the antenna element in different
application with various frequencies is a concern for communicat-
ing. During the last few years, research on the plasma antennas
has greatly expanded [1–3]. One reason is that the researchers and
engineers now realize that relative to the conventional metallic
antenna, plasma antennas, in particular the surface wave driven
(SWD) ones, can generally be more flexible in operating condi-
tions for special technical applications. Plasma antennas find its
applications in variety of fields such as military applications, faster
internet, public safety networks, radio and television broadcasting
and space communications.
The properties of the SWD depend on the amount of power
absorbed per unit length of the plasma column and on the discharge
conditions, namely the composition and the pressure of the gas, the
dimensions of the plasma tube, the wall material and the excita-
tion wave mode and frequency [4]. Although the specific design
of the wave launcher does not affect the plasma parameters, the
launcher determines the efficiency of the power transfer from the
power generator to plasma and, to some extent, it imposes wave
propagation mode [4].
Recent progresses with numerical simulations of SWDs in [5]
lead us to the experimental study to better understanding of the
physical process and proving the previous investigations. Experi-
ments were undertaken over the ranges of pressures and excitation
∗ Corresponding author. Tel.: +98 2188366039.
E-mail addresses: sadeghi kia@ari.ac.ir (F. Sadeghikia), mtnoghani@ari.ac.ir
(M. Talafi Noghani), Mohammadreza.simard@gmail.com (M.R. Simard).
determine the major characteristics of a SWD plasma antenna
including the antenna height, the resonant length, VSWR and gain
values.
This paper is organized as follows. Section 2 constrains a short
review of the principal theory in a SWD plasma antenna, with
the corresponding equations. In Section 3, we concentrate on the
experimental setup. Section 4 is devoted to the description of
experimental results including measurement of the antenna height
(Section 4.1). We then characterize the VSWR and the resonant
length of the antenna in Section 4.2 and the radiation pattern and
gain of the antenna will be discussed in Section 4.3. This investiga-
tion will be summarized in Section 5.
2. Theory
In a cylindrical configuration, the possible modes of propaga-
tion of surface waves along the plasma column are defined by their
field intensity dependence upon the azimuthal angle ϕ. Accord-
ing to [4,6], the mode selection relies essentially on the value of
the product f · R, where f is the excitation frequency and R is the
radius of the antenna. When this value is less than approximately
2 GHz cm, the plasma column can be sustained only with the m = 0
surface wave. This product value is independent of the gas nature
and pressure. When f · R is increased starting from 2 GHz cm, one
can achieve the discharge with dipolar mode (m = 1). Increasing f · R
repeats the preceding situation with the larger modes.
A condensed analysis of the plasma radiation and power launch-
ing constraints on excitation of a surface wave on a plasma column
were presented in [7,8]. The case of a plasma column excited
below ωp , the angular plasma resonance frequency, leads to a

http://dx.doi.org/10.1016/j.aeue.2016.01.024
1434-8411/© 2016 Elsevier GmbH. All rights reserved.

Please cite this article in press as: Sadeghikia F, et al. Experimental study on the surface wave driven plasma antenna. Int J Electron
Commun (AEÜ) (2016), http://dx.doi.org/10.1016/j.aeue.2016.01.024
G Model
AEUE-51564; No. of Pages 5 ARTICLE IN PRESS
2 F. Sadeghikia et al. / Int. J. Electron. Commun. (AEÜ) xxx (2016) xxx–xxx

Fig. 1. Experimental setup of a SWD plasma antenna.

negative permittivity and permits the propagation of a surface speed of 25 m3 /h and ultimate total pressure of 2 × 10−3 mbar.
wave along the column. Let ω be the angular excitation frequency, A pirani pressure gauge with the pressure range of 10−3 to
when (ωp /ω)  1, the plasma occludes the wave as if it were as a 760 torr and a gas handling system that admitted Argon at different
perfect conductor. pressures are also connected to the tube in order to determine an
Following the analysis of [9], for a given pressure, the height of optimum pressure for the antenna.
the antenna should increase as the square root of the applied RF The surface wave launcher consisted of a copper collar of length
power and may be written as: 30 mm mounted less than 4 mm below a circular hole cut in the top
 of a grounded box. RF power up to 40 W at 450 MHz was applied via
h ≈ B(p) P0 (1) a directional coupler and a double stub tuner. The reflected power
where theoretical slope B(p) derived from global model as: was measured using a power meter connected to the coupled port
 of the directional coupler. By controlling the stub length of the dou-
2 ble stub tuner, it was possible to decrease the reflected power to
B(p) = (2)
CK(p)vm (p) a minimum value. So, in measurements, the reflected power was
less than 0.01 of the applied power.
C is a constant with a value of C ≈ 5 × 109 m−4 S [10]. vm (p) is the The gas can be ionized using the intense electric field developed
electron-neutral collision frequency for momentum transfer and in the gap between the collar and the box and igniting the plasma
K(p) is a function of pressure for a given geometry. Eq. (1) shows along the column.
that for a given pressure, controlling the resonant length of the A second coupling collar (for transmission or receiving) was
antenna is possible by excitation power. mounted 40 mm below the pump collar which was used to apply
Ref. [9] shows that the plasma electron density n0 at the base of communication signals to the antenna via a band stop filter with a
the antenna is: center frequency at 450 MHz. The filter attenuated the 450 MHz RF

n0 ≈ A(p) P0 (3) power signal up to 70 dB.
Tests were conducted using the plasma column as a receiving
where, or transmitting antenna over the range from 80 to 300 MHz. Exper-
 iments were undertaken for different RF power levels and plasma
2C vm (p)
A(p) = (4) pressures to determine the antenna characteristics.
K(p)

The electron density steadily decreases away from the launcher
and the plasma column is axially non-uniform. The minimum den-
sity nD below which the wave no longer propagates is [10]:
nD = 1.2 × 104 (1 + εg )f 2 (cm−3 )
where f, the applied excitation frequency is expressed in MHz and
εg is the tube relative permittivity.
3. Experimental setup
Surface wave driven plasma monopole antenna was employed
in this investigation using the arrangement shown in Fig. 1. This
arrangement consisted of a 1 m Pyrex tube with the thickness of
1 mm and permittivity of 4.82, and the diameter of 25 mm con-
nected to a rotary vacuum pump of VARIAN DS602 with pumping
4. Experimental results
4.1. Antenna height
(5)
The height of the plasma antenna as a function of the applied
RF power was determined by observing the visible light from the
plasma. Fig. 2 plots the height of the antenna (h) as the square root
of the applied power at the pressure of 0.4 mb. The used origin of
this measurement is the top part of the launcher. The plot shows
that the height of the antenna increases linearly proportional to the
square root of the applied power and the slope of the plot, B(p), in
this case is around 6.77 ± 0.04 cm/W1/2 , which was predicted by the
global model as 6.4 cm/W1/2 [9,10]. The comparison shows that the
results with argon column at 0.4 mb agree, approximately, with the
global model. It is noticeable to mention that the resulting slope in

Please cite this article in press as: Sadeghikia F, et al. Experimental study on the surface wave driven plasma antenna. Int J Electron
Commun (AEÜ) (2016), http://dx.doi.org/10.1016/j.aeue.2016.01.024
 G Model
AEUE-51564; No. of Pages 5 ARTICLE IN PRESS
F. Sadeghikia et al. / Int. J. Electron. Commun. (AEÜ) xxx (2016) xxx–xxx 3

0.4

0.35
Antenna Height (m)

0.3

0.25

0.2

0.15

0.1 Analytical
Measurement @ p=0.4
0.05
1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 Fig. 4. Slope of height B(p) for plasma antenna as a function of pressure; solid line
1/2 is a theoretical value derived from the global model [9] and the dotted line is the
The Square Root of Excitation Power (W )
measurement results (error bar plot).
Fig. 2. Height of the SWD plasma antenna (error bar plots) as a function of the square
root of excitation power. 17
x 10
10
0.5 n03
Measurement @ p=0.25 mb
0.45 8
Measurement @ p=0.3 mb n02

Plasma Density (m )
-3
30 W
Measurement @ p=0.4 mb
0.4
Measurement @ p=0.5 mb
Antenna Height (m)

6
0.35 20 W
n01
0.3
4
0.25 10 W

0.2 2
0.15
nd
0.1 0
0 0.05 0.1 0.15 0.2 h 0.25 0.3 0.35 h3 0.4
1 h2
0.05
1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 Antenna Height (m)
1/2
The Square Root of Excitation Power (W )
Fig. 5. The plasma density along the antenna (error bar plots) as a function of the
antenna height.
Fig. 3. The height of the antenna (error bar plots) as a function of excitation power
at different pressures.

the excitation frequency of 450 MHz in this experiment, using (5),

[9] was measured as 9.5 ∓ 0.4 cm/W1/2 with a COTS1 tube, which
was higher than the predicted slope. This difference may be due to
the presence of mercury vapour in COTS tubes.
The measurements were repeated for several different
pressures between 0.25 and 0.5 mb. Variations of the height of the
plasma antenna as a function of the square root of the excitation
power at different pressures are shown in Fig. 3. Lower pressures
are consistent with higher height of the antenna possibly due to
the less electron-neutral collision frequency and consequently an
increment in B(p). Fig. 4 plots the resulting slope of the height as a
function of pressure (dotted line) which is compared with the the-
oretical slope derived from the global model in (2) (the solid line).
The measured slope within the error bars is synchronous with the
global model but is somewhat higher than the predicted especially
for lower pressures. The discrepancy between the results may be
attributed to some factors such as the error in measurement pro-
cess, leakage of gas into the system during testing or the thickness
of the Pyrex tube. Other possibility is the limitations of the global
model.
Based on the antenna height measurement established
in this investigation, estimation of K ≈ (17 ± 2) × 10−17 m2 W−1
is acceptable for the pressure of 0.4 mb which will give
A(p) = (0.16 ± 0.01) × 1018 m−3 W1/2 . Although it was not possible
to measure the plasma density along the column, it is possible to
infer its value at the base of the antenna (n0 ) using (3). Moreover, for
1
Commercial off the shelf.
the minimum density is calculated as bD = 1.41 × 1016 m−3 , which
is equivalent to the plasma frequency of 1.08 GHz.
The linear decrease of the plasma density along the antenna was
obtained analytically as well as experimentally in [9,10]. Because
of the essentially linear variations in number density with length
along the plasma column, a continuous range of plasma density
exists between n0 and nD in the column. Consider the curve of
the axial distribution of the cross section average electron density
within the error bars sketched in Fig. 5 at the pressure of 0.4 mb
and the increase of the excitation power from 10 to 30 W. The
plasma column initially sustained at 10 W and remains unchanged
when the power increased and becomes the tail of the higher
power plasma column. For example, for an RF power of 10 W, the
maximum density is n = n01 ≈ (0.53 ± 0.03) × 1018 m−3 at the exci-
tation point and the column ends at n = nD , where the antenna
height is h1 . With increasing the excitation power up to 20 or 30 W,
the maximum density varies to n = n02 ≈ (0.75 ± 0.04) × 1018 m−3 or
n = n03 ≈ (0.92 ± 0.05) × 1018 m−3 respectively, with the equivalent
antenna height as h2 and h3 .
4.2. VSWR measurement
Since VSWR is a critical parameter in an antenna, the VSWR
value of the antenna has been measured from 20 MHz to 350 MHz
over the pressure ranges between 0.3 mb and 0.5 mb at different
excitation powers. According to [11], the finite thickness of the
monopole antenna makes the antenna electrically longer than its
physical length. So, the physical length must be slightly shortened

Please cite this article in press as: Sadeghikia F, et al. Experimental study on the surface wave driven plasma antenna. Int J Electron
Commun (AEÜ) (2016), http://dx.doi.org/10.1016/j.aeue.2016.01.024
G Model
AEUE-51564; No. of Pages 5 ARTICLE IN PRESS
4 F. Sadeghikia et al. / Int. J. Electron. Commun. (AEÜ) xxx (2016) xxx–xxx

2.5 0.038

0.036

0.034
2
VSWR (dB)

0.032

)
-1/2
T (mW
0.03

0.028
1.5
0.026
p = 0.30 mb
p = 0.40 mb
0.024
p = 0.50 mb
1 0.022
0.05 0.1 0.15 0.2 0.25 0.3 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
Antenna Height (Wavelength) Pressure (mb)

Fig. 6. The VSWR of the plasma antenna as a function of the antenna height for the Fig. 8. Slope of the resonant length as a function of the pressure for plasma antenna.
excitation power of 31.62 W at different gas pressures.

possible to produce a correct resonant length by controlling the

in order to create a resonant length antenna and the shortening applied power.
factor varies depending on the thickness of the antenna.
Experimental measurements of the voltage standing wave ratio
4.3. Gain measurement
(VSWR) at the signal collar of the plasma antenna terminal were
performed using a network analyzer over the frequency range
The radiation pattern of the plasma antenna was measured by
of signal. Fig. 6 shows the results for the excitation RF power of
moving the antenna in an arc with the radius more than 200 m
31.62 W at different pressures. Based on these results, the resonant
over the probe antenna in the elevation and azimuth plane. The
length of the antenna is essentially a function of the gas pressure
measurements were done in the open air and there were no
and increasing the gas pressure decreases the resonant length of
obstructions in the vicinity. Since monopole plasma antenna are
the antenna. The resonant length of the antenna is considered here
concerned, in all cases, it is important to be able to obtain an
as the middle point of the antenna height (in wavelength) in the
azimuthally pattern similar to a conventional metallic antenna.
region which VSWR is less than 2.5:1. Fig. 7 plots the resonant
However, the measurement result, as shown in Fig. 9, reveals
length at different RF powers and pressures. Increasing the exci-
that the azimuth pattern is not completely circular at the reso-
tation power increases the resonant length of the antenna in a
nant frequency. At the first glance, one may conclude that there
constant pressure which is due to the higher average plasma den-
is azimuthally asymmetric dipolar mode of surface wave (m = 1)
sity in the antenna and the antenna performance becomes closer to
propagating along the column and sustaining the plasma. How-
the metallic antenna. It can be concluded that, in plasma antenna,
ever, in this experiment, f · R = 0.5625 GHz which is definitely less
the shortening factor of the resonant length is a function of the
than 2 GHz and the condition m = 1 is not satisfied. Therefore, the
plasma frequency in the column (excitation power) and the pres-
assumption of dipolar mode of surface wave along the plasma col-
sure of the gas as well as the thickness of the antenna.
umn seems to be invalid. As another hypothesis, the non-circular
The resonant length of the antenna increases approximately as
the square root of the excitation power with a slope T(p), which is
nearly constant in a certain pressure. Fig. 8 plots the resulting slope
as a function of pressure. This implies that:

H0 = T (p) P0 (6)

which H0 is the resonant length of the plasma antenna in wave-

length and T(p) is a constant for a given pressure. Eq. (6) therefore
shows that, for a given transmitting frequency, it should be

0.18
Resonant Length (Wavelength)

0.16

0.14

0.12

0.1

p = 0.30 mb
0.08
p = 0.40 mb
p = 0.50 mb
0.06
10 15 20 25 30 35
Excitation Power (W)

Fig. 7. Variations of the resonant length of the plasma antenna as a function of

excitation power at different gas pressures. Fig. 9. The measured azimuthal pattern of the plasma antenna.

Please cite this article in press as: Sadeghikia F, et al. Experimental study on the surface wave driven plasma antenna. Int J Electron
Commun (AEÜ) (2016), http://dx.doi.org/10.1016/j.aeue.2016.01.024
G Model
AEUE-51564; No. of Pages 5 ARTICLE IN PRESS
F. Sadeghikia et al. / Int. J. Electron. Commun. (AEÜ) xxx (2016) xxx–xxx
-5
-10
-15
Gain (dB)
-20
-25
-30
-35
-40
-45
100 120 140 160 180 200 220 240
Frequency (MHz)
Fig. 10. The measured gain of the plasma antenna.
azimuth pattern may be due to the configuration of the wave
launcher.
Since gain is a significant factor of the antenna, the gain of the
plasma antenna was measured between 110 and 300 MHz. Fig. 10
plots the resulting gain of the antenna in for the antenna excited by
16.59 W, which is equivalent to 25 ± 0.7 cm visible light at 0.4 mb.
The results show that the gain increases in the resonant region.
However, the gain of the plasma antenna is considerably lower
than the gain of a conventional metallic antenna. The small average
plasma frequency in the column seems to be the main reason of this
shortage, as was predicted numerically previous in [12]. Moreover,
since the tube is not completely filled with plasma, the power flux
of the wave decreases toward the end of the column and the sur-
face wave power is gradually expended in sustaining the discharge.
So, the surface wave cannot deposit more energy into the top part
of the tube and thus leads a non-uniform distribution of the con-
ductivity along the column. In this case, the plasma column has a
linearly decreasing density from 6.333 × 1017 m−3 at the excitation
point to nD at the end of the column. The axial non-uniformity of
the plasma density is a shortcoming in the antenna application of a
plasma column and some remedies are necessary for this problem.
Another possibility of the reduced gain may be due to the way in
which the signals are coupled to the antenna via the collar. Since
there is no physical metallic link between the transmitting signal
and plasma, a capacitive coupling is required because of the Pyrex
envelope which may affect the antenna gain.
Moreover, the advantages of plasma antenna, including nearly
zero RCS, when it is not energized, and also controllable length or
fast transmission, provide inevitably limited gain.
5. Conclusion
This experimental study has established some major character-
istics of a surface wave driven plasma monopole antenna. It has
Please cite this article in press as: Sadeghikia F, et al. Experimental study on the surface wave driven plasma antenna. Int J Electron
Commun (AEÜ) (2016), http://dx.doi.org/10.1016/j.aeue.2016.01.024
5
been confirmed that the length of the antenna increases as the
square root of the excitation power. The slope variation of this curve
has been presented as a function of pressure which is consistent
with a simple global model within a precision of 5%. Estimations of
the resonant length of the plasma antenna, based on the measured
VSWR, have proved that the resonant length of a plasma antenna
varies dynamically with excitation power and gas pressures. The
results have shown that decreasing the gas pressure or increas-
ing the excitation power, increases the resonant length of the
antenna.
Observations of the radiation pattern of the antenna have shown
that the azimuth pattern is not completely circular at the resonant
frequency maybe due to the configuration of the wave launcher.
260 280 300
We found that the tapered density profile of the plasma along the
column may reduce the gain of the plasma antenna. Another possi-
bility of the reduced gain may be due to the capacitive coupling
required to couple the signals to the antenna. Although plasma
antennas have advantages including nearly zero RCS, when not
energized, controllable length and fast transmission, they have
inevitably limited gain. Whether or not the gain is a critical fac-
tor in an antenna design and some remedies are necessary for this
problem. We have undertaken some studies to increase the gain of
the SWD plasma antenna and currently, more investigation is being
carried out.
References
[1] Anderson T. Theory, measurements and prototypes of plasma antennas. In:
52nd IEEE International Symposium on Antennas and Propagation (APSURSI).
2014. p. 567–8.
[2] Anderson T. Plasma antennas VSWR and co-site and parasitic interference
reduction or elimination. In: 52nd IEEE International Symposium on Antennas
and Propagation (APSURSI). 2014. p. 366–7.
[3] Anderson T. Plasma antennas for lowering co-site interference among closely
spaced antennas. In: 40th IEEE International Conference on Plasma Science.
2013. p. 1.
[4] Popov OA. High Density Plasma Sources: Design, Physics and Performance. New
Jersey: Noyes Publication; 1995.
[5] Sadeghikia F, Hodjat-Kashani F, Rashed-Mohassel J, Ghayoomeh-Bozorgi J.
Characterization of a surface wave driven plasma monopole antenna. J Elec-
tromagn Waves Appl 2012;26(2–3):239–50.
[6] Margot-Chaker J, Moisan M, Chaker M, Glaude VMM, Lauque P, Paraszczak J,
et al. Tube diameter and wave frequency limitations when using the electro-
magnetic surface wave in the m=1 (dipolar) mode to sustain a plasma column.
J Appl Phys 1989;66:4134–48.
[7] Morrow IL, James JR. Fundamental limitations on excitation of a surface wave
on a plasma column. In: 40th IEEE International Symposium on Antennas and
Propagation (APSURSI), vol. 4. 2002. p. 272–5.
[8] Ye HQ, Gao M, Tang CJ. Radiation theory of the plasma antenna. IEEE Trans
Antennas Propag 2011;59(5):1497–502.
[9] Rayner JP, Wichello AP, Cheetham AD. Physical characteristics of plasma anten-
nas. IEEE Trans Plasma Sci 2004;32(1):269–81.
[10] Moisan M, Zakrzweski Z. Plasma sources based on the propagation of electro-
magnetic surface waves. J Phys D: Appl Phys 1991;24(7).
[11] Balanis A. Antenna Theory, Analysis and Design. New Jersey: John Wiley & Sons
Inc.; 2005.
[12] Sadeghikia F, Hodjat-Kashani F. A two element plasma antenna array. Eng
Technol Appl Sci Res 2013;3(5):516–21.