948 IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 4, APRIL 2013

Development of Large-Area Switchable PlasmaDevice for X-Band Applications

Lee W. Cross, Member, IEEE, Mohammad J. Almalkawi, Member, IEEE,and Vijay K. Devabhaktuni, Senior Member, IEEE

Abstract— This paper demonstrates a large-area, lightweight,conformal plasma device that interacts with propagatingX-band microwave energy. The active elements are ruggedplasma-shells— hollow ceramic shells encapsulating a controlled-pressure gas that can be ionized to controlled plasma para-meters. Plasma-shells are electrically excited by frequencyselective surfaces that are transparent to the frequency bandof interest. The result is equivalent to large-area free-spaceplasma confined in a discrete plasma slab. A novel structureis designed with the aid of full-wave simulation and fabri-cated as a 76.2-mm square array, and transmission perfor-mance is tested across different drive voltages and angles ofincidence. Switchable attenuation of 7 dB is measured acrossthe passband when driven with 1400 Vpp at 1 MHz. Plasmaparameters are estimated from theory and full-wave simula-tion, with electron density estimated to be 3.6 × 1012 cm−3.The proposed structure has potential for use on mobile platforms.

Index Terms— Bandpass filter, frequency selective surface(FSS), gas plasma, microwave, plasma absorber.

I. INTRODUCTION

LARGE-SCALE plasma can significantly attenuate radiofrequency (RF) communication, as seen during space-

craft vehicle reentry where communication blackout lasts forseveral minutes because of the surrounding plasma sheath[1]–[3]. Plasma attenuates low-power RF energy used forcommunication and radar tracking purposes, as well ashigh-power electromagnetic pulse and high-power microwaveenergy [4]–[6].

The ability to rapidly switch large-scale plasma volumes ishighly desirable for the creation of large-area electromagnetic(EM) devices with tunable operating frequency or transmis-sion, absorption, and reflection properties.

Although the concept of using plasma as a microwaveabsorber or reflector has existed for decades [7], [8], very fewdevices were demonstrated in literature. Such devices primar-ily consist of long, fragile plasma tubes. Anderson et al. [9]presented devices using cylindrical mercury lamps as switch-able plasma volumes, sharing all the problems of mercury

Manuscript received January 18, 2013; revised February 21, 2013; acceptedFebruary 21, 2013. Date of current version April 6, 2013. This work wassupported in part by the National Science Foundation EDI under Award1000744.

The authors are with the Electrical Engineering and ComputerScience Department, University of Toledo, Toledo, OH 43606 USA(e-mail: leecross98@yahoo.com; mohammad.almalkawi@utoledo.edu;Vijay.Devabhaktuni@utoledo.edu).

Color versions of one or more of the figures in this paper are availableonline at http://ieeexplore.ieee.org.

Digital Object Identifier 10.1109/TPS.2013.2251369

(a) (b)

Fig. 1. (a) Plasma-shell cutaway showing internal plasma. (b) Proposedswitchable device made of patterned conductors around plasma-shell array.

lamps: fragility, limited life, modulation speed limited to kHz,and toxic mercury vapor. In [10], Vardaxoglou demonstrateda solid-state switchable plasma device by illuminating apatterned semiconductor wafer, but is likewise fragile, sizelimited, and very restricted in available material propertiesand thicknesses. Murphy et al. [11] introduced a large-area plasma sheet reflector that operated by a pulsed elec-tron beam and exhibited low levels of RF interaction. Lari-galdie and Caillault showed significant X-band sheet reflectionbut only in pulsed mode with magnetic confinement [12].Scharer et al. [13], [14] demonstrated large-volume plasma andRF reflection from ultraviolet photoionization of an organicseed gas [e.g., tetrakis-(dimethylamino)-ethylene] using apulsed laser.

There are several key challenges in creating a large-scaleplasma device. Initially, plasma must be encapsulated in a waythat maintains controlled gas composition, pressure, and purity.Next, power must be delivered to the plasma volume in away that provides controlled EM properties across a frequencyrange of interest.

None of the devices demonstrated in literature are suitablefor large-area, rugged, conformal, rapidly switchable plasmadevices. This paper proposes a concept that overcomes theselimitations by using an encapsulated plasma component calledthe plasma-shell shown in Fig. 1(a), and driving it directly withhigh-voltage alternating current (ac) energy through conduc-tive layers, referred to as plasma excitation surfaces, that aretransparent to RF energy in a frequency range of interest asshown in Fig. 1(b).

Plasma-shells are hollow ceramic shells that hermeticallyencapsulate any gas mixture at controlled pressure. The shell

0093-3813/$31.00 © 2013 IEEE

CROSS et al.: DEVELOPMENT OF LARGE-AREA SWITCHABLE PLASMA DEVICE 949

is a strong refractory oxide layer capable of direct contact withplasma without damage or gas contamination. Plasma-shellscan be manufactured in any shape and size from 0.5 to 10 mmand are lightweight and rugged.

Plasma excitation surfaces must have large conductor areain contact with plasma-shells yet be transparent to one ormore microwave bands, and some types of frequency selectivesurface (FSS) elements have such properties. FSSs are struc-tures that interact with EM energy propagating through freespace in the microwave frequency range, loosely defined as0.3–30 GHz [15], [16]. FSSs are used in radar, communica-tion, instrumentation, and power transfer applications and areused as hybrid radomes, spatial bandpass or bandstop filters,dichroic reflectors and subreflectors, absorbers, and polarizers.

This paper is organized as follows: initially, a novel plasmadevice is proposed in Section II that emulates the physicalphenomenon of a large-scale plasma sheet in free space acrossa frequency range of interest and the design of the elec-tromagnetic structure is presented using full-wave simulationtools. In Section III, the proposed plasma device is fabricatedand tested in the off- and on-states, and key parametersare measured including passband frequency response in theoff- and on-states, effect of angle of incidence, and effectof drive voltage. Plasma parameters are estimated and waysto optimize the device are discussed. Finally, results of thispaper are summarized in Section IV along with descriptionsof potential applications of this technology.

II. DESIGN AND FABRICATION

A. Proposed Concept

The proposed device consists of an array of unit cells. Eachplasma excitation surface unit cell is an FSS element com-posed of a conductive sheet with etched slots. The Jerusalemcross pattern is employed in this paper, but any shape can beused that meets the following criteria:

1) bandpass characteristic;2) free-standing (no unconnected floating elements); and3) large conductor area coverage.

The FSS element is repeated on the conductive surface ona closely spaced regular grid. Plasma excitation surfaces arereadily fabricated as single-sided printed circuit boards (PCB).

Both plasma excitation surfaces are laminated onto an arrayof plasma-shells slightly smaller than FSS elements. Thiscomposite material can be driven with a high-voltage ac powersource to create a volume of plasma that directly interacts withpropagating EM energy through the plasma excitation layersthat are effectively transparent to X-band energy. Switchabletransmission and reflection performance through the plasmavolume is accomplished in this way.

B. Analysis

A candidate structure was optimized using full-wavesimulation, and the resulting Floquet port model shownin Fig. 2 is the unit cell of a symmetric second-orderFSS with spacing of 6.35 mm. Floquet port simulation inANSYS high frequency structure simulator (HFSS) models

Fig. 2. HFSS Floquet port model with material parameters and dimensions.The unit cell is composed of single plasma-shell in center sandwiched by twosingle-layer PCBs, with outer dielectric slabs for port tuning.

an infinite FSS from a single unit cell with good speed.This model simulates one response from 5 to 18 GHz. Simpli-fications were made to the model to improve simulation timeincluding flat conductors, merging plasma-shell electrodeswith plasma excitation surfaces, and omitting shell side walls.

FSS element spacing of 0.21 λ at 10 GHz gives stronginterelement coupling for center frequency and bandwidthstability across scan angle. Element geometry provides severaldegrees of freedom for tuning center frequency and bandwidthand was adjusted so that an entire plasma-shell could fit acrossthe center + slot, to minimize performance deviation causedby placement error.

Both outer dielectric slabs increase port coupling strengthfor this structure to achieve sufficient return loss; the responsewas undercoupled without them. The dielectric was real-ized with commonly available unclad Rogers 5880 substratematerial [17].

The simulated wideband response in Fig. 3 shows centerfrequency ( fc) of 10.1 GHz and 3 dB bandwidth of 2.8 GHz.Insertion loss for the infinite array is 1.0 dB, and return loss is12.5 dB. The response of one single-sided Rogers 4003C PCBis shown for verification with measured results, and higher fcof 14.0 GHz is evident with the absence of electroded plasma-shells that decrease element resonant frequency [18].

C. Implementation

Fabricated PCBs in Fig. 4 were made using standard photo-etching and plating processes. The top layer includes a finite12 × 12 FSS array (i.e., there were 160 shells available for thisexperiment) and the bottom layer is bare except for a perimetercopper flange filled with vias for continuous electrical contactwith the mounting plate. One PCB was trimmed to 114 mmoverall size for use as the top PCB.

Plasma-shell size is 4.23 mm × 4.23 mm × 2.24 mm, with176-μm shell thickness. Fill gas is 0.1% argon with balanceneon, a common Penning mixture, at 175 Torr [19].

The plasma-shell electroding and mounting processes areshown in Fig. 5 where each shell is printed with electrodes on

950 IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 4, APRIL 2013

-20

-15

-10

-5

0

-30

-20

-10

0

5 6 7 8 9 10 11 12 13 14 15 16 17 18

| S11

| (dB

)

|S21

| (dB

)

Frequency (GHz)

Off

Single PCB

Fig. 3. Simulation of proposed device shows good second-order off-stateresponse, with return loss included for reference. One single PCB response(without plasma-shells) is simulated for comparison with measured results.

(a) (b)

Fig. 4. Fabricated PCB layers. (a) Top plasma excitation surface with center12 × 12 array. (b) Blank bottom with perimeter flange for ground continuitywith mounting plate.

(a) (b) (c)

Fig. 5. Plasma-shell assembly method. (a) Conductive pads with + slotare printed on top and bottom of each shell. (b) Conductive epoxy dotsare stenciled onto PCB. (c) Shell is placed onto substrate, achieving optimalcontact (fabrication trial seen through glass slide pictured above).

the top and bottom, with a + shaped slot of width of 0.35 mm.To mechanically and electrically attach shells to the PCB, firstan array of silver conductive epoxy dots are stenciled to thePCB, then shells are placed using a pick and place machine.To mount the second PCB to the array, silver epoxy dots arestenciled to the top PCB and placed on the bottom PCB/shellassembly. Alignment is maintained with pins that are removedafter the entire assembly is cured.

The assembled device is shown in Fig. 6(a) without dielec-tric slabs. To mount the device to a large ground plane fortesting, the bottom PCB (seen as the copper border) is screwedto the mounting plate and the seam is covered with aluminum

(a) (b)

Fig. 6. Fully assembled device (a) without dielectric slabs and (b) withdielectric slabs taped in place, showing weak plasma glow during initial turn-on test.

VNA

DUT

Receive Antenna

TransmitAntenna

Sustainer

610 mm Plate

76.2 mmAperture

Fully AnechoicChamber

RotationStage

775 mm 825 mm

Trigger

Fig. 7. Test setup for transmission response measurement from 5 to 18 GHzwith angle of incidence sweep.

tape. The mounting plate is connected to one sustainer outputand the top PCB (isolated from the ground plane) to the otherwith a wire. The device is shown in Fig. 6(b) with dielectricslabs taped in place to the top and bottom PCBs, and an orangeglow is seen while the device is sustained at low voltage.

III. EXPERIMENTAL RESULTS AND DISCUSSION

A. Test Setup

Device frequency response is measured with the test setupin Fig. 7 where a Rohde & Schwarz ZVB20 vector networkanalyzer (VNA) measures transmission scatter (S)-parameter(S21) for different plasma states and incidence angles. Thedevice under test (DUT) is mounted to a 610-mm (24 in.)square plate that can be rotated, and the plate is substitutedwith reference plates for calibration. Two 2–18-GHz quad-ridge horns are placed in the far field for uniform illumina-tion. Plate edge effects and small sample size limited usefulmeasurement range to ∼7–18 GHz. Return loss (S11) mea-surements were attempted but the setup lacked the dynamicrange and placement repeatability that are required for thesmall sample.

Fig. 8(a) shows the actual measurement setup wherethe 610-mm plate is seen in the center, shown coveredwith absorber for a calibration measurement. The sustainer(not seen) is mounted on the far side of the Styrofoam columnout of antenna line of sight, and the column and sustainer arerotated to sweep angle of incidence. The DUT is mounted on

CROSS et al.: DEVELOPMENT OF LARGE-AREA SWITCHABLE PLASMA DEVICE 951

(a)

(b)

Fig. 8. (a) Fully anechoic chamber test setup with two horns and610-mm plate on rotation stage in center. (b) High-voltage device testingwithout dielectric slabs, showing intense plasma-shell light emission seenthrough 1.524-mm-thick Rogers 4003C PCB substrate.

the rear side of the plate in Fig. 8(b), and high-intensity orangelight is emitted during operation.

The sustainer used in this experiment is a high-voltagesquare wave generator capable of bridged output drive of1500 Vpp at 1 MHz, using a bipolar H-bridge schemecommonly used in plasma display panels [20], with energyrecovery circuits [21], [22] to recover capacitive displacementcurrent for increased drive efficiency. The sustainer is notdesigned for continuous operation so all test data were mea-sured with 10-ms drive bursts at a repetition rate of 4 Hz,and the VNA was triggered to capture data within this largeon-state window. This does not affect measurements becauseplasma attains steady-state operation on extremely short timescales. The test device is capable of continuous operation,limited only by the thermal limitations of the attached PCBs.

Measurement data were time gated with a 2-ns gatingwindow to eliminate multipath and referenced to measured76.2-mm aperture plate through-calibration. The small samplesize manifests some measurement challenges including surfacewaves, impedance discontinuities at array edges, and apertureresonances, none of which were accounted for in postprocess-ing. However, the data are sufficient for conceptual validation.To accommodate such challenges, fabrication of a largerdevice is advised for precise measured results.

-30

-20

-10

0

5 6 7 8 9 10 11 12 13 14 15 16 17 18

|S21

| (dB

)

Frequency (GHz)

Off

On

Single PCB

Fig. 9. Measured wideband transmission response of device in off- andon-states driven at 1200 Vpp, 1 MHz, normal incidence. Off-state and singlePCB responses agree with HFSS model. Measured data may only be usefulabove nulls at 7 and 8.5 GHz that do not appear in simulation.

B. Wideband Results

The test device was measured in the off- and on-states, andwith a single PCB (without plasma-shells) for verification withthe HFSS model. Measured data in Fig. 9 show off-state centerfrequency of 11.0 GHz, bandwidth of 4.2 GHz, and minimuminsertion loss of 5.1 dB. On-state center frequency increasesto 12.6 GHz, bandwidth reduces to 3.0 GHz, and minimuminsertion loss increases to 9.2 dB. Absolute insertion loss isnot likely to be accurate because of finite array effects andcalibration error; however the measured center frequency isexpected to be reliable.

The on-state response has two important characteristicsversus off-state: 1) higher center frequency, and 2) higherinsertion loss. The frequency shift is confirmed by the on-stateresponse being higher than the off-state response >15 GHz,and these characteristics will be examined more closely duringplasma modeling.

Measured results compare well with simulation, wheremeasured passband is higher by 1.0 GHz for the off-state andsingle PCB, and measured off-state bandwidth is 4.2 GHz.Nulls at the lower passband are not present in simulation; thissuggests that the measured nulls are artifacts of the finite array.

C. Angle of Incidence Sweep

The device was measured in the off- and on-states asangle of incidence was increased from normal (0°) to 60° inincrements of 15°, shown in Fig. 10. Bandwidth and centerfrequency appear relatively stable up to 45°, but either theresponse or the test setup falls apart at 60° with insertion lossreaching an unrealistically low value and nulls moving nearpassband center.

Key design features contribute to scan angle stability: elec-trically small and closely spaced FSS elements, and low-dielectric-constant slabs on the outer surfaces. With tallershells it is possible to design a device that meets the theoreticalcriterion for an optimally scan-independent FSS.

952 IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 4, APRIL 2013

-30

-20

-10

0

8 9 10 11 12 13Frequency (GHz)

|S21

| (dB

)

0° Off

0° On-30

-20

-10

0

8 9 10 11 12 13

Frequency (GHz)

|S21

| (dB

)

15° Off

15° On

-30

-20

-10

0

8 9 10 11 12 13Frequency (GHz)

|S21

| (dB

)

30° Off

30° On-30

-20

-10

0

8 9 10 11 12 13

Frequency (GHz)

|S21

| (dB

)

45° Off

45° On

-30

-20

-10

0

8 9 10 11 12 13

Frequency (GHz)

|S21

| (dB

)

60° Off

60° On

(a) (b)

(c) (d)

(e)

Fig. 10. Measured transmission response in off- and on-states versus angleof incidence, driven at 1200 Vpp, 1 MHz. Center frequency and bandwidthare stable at: (a) 0°; (b) 15°; (c) 30°; and (d) 45°. The device or test setupfalls apart at (e) 60°. Results show good scan independence, and switchableattenuation is significant across passband at all angles.

D. Drive Voltage Sweep

Increasing drive voltage has a significant effect on pass-band attenuation. Fig. 11 shows that each voltage increaseof 200 V increases average on-state passband attenuation by1.4 dB. Following the observed trend, an improved sustainerthat outputs 2400 Vpp would more than double switchableattenuation from 5.9 dB at 1200 Vpp to 14.2 dB. It is believedthat increasing drive voltage directly increases electron densityand gas conductivity, thereby increasing attenuation.

E. Plasma Medium Model

Measured results verify that the test device provides sig-nificant switchable attenuation. To better understand the inter-action between the device structure and plasma medium, theplasma volume can be modeled as a homogenous conductivedielectric medium. Complex permittivity (ε) is defined asε = ε0(ε

′r + jε′′

r ), consisting of real and imaginary relativepermittivity (ε′

r and ε′′r , respectively), and both frequency-

dependent components for a cold, collisional, and weakly

-20

-15

-10

-5

0

8 9 10 11 12 13

Frequency (GHz)

|S21

| (dB

)

0

1

2

3

4

5

6

7

8

800 1000 1200 1400Voltage (V)

Avg

. Ins

erti

on L

oss

(dB

)

Off800 V1000 V1200 V1400 VAvg. IL

Fig. 11. Measured transmission response in off- and on-states with sweptdrive voltage, at 1 MHz and normal incidence. Increasing drive voltageincreases attenuation by 1.4 dB/200 V, with maximum average passbandattenuation of 7.0 dB at 1400 Vpp.

ionized plasma are

ε′r = 1 − ω2

p

ω2 + ν2 (1)

ε′′r = ω2

pν/ω

ω2 + ν2 (2)

as functions of plasma frequency (ωp , in rad/s), microwavedrive frequency (ω, in rad/s), and electron collision frequency(ν, in rad/s) [2]. Material dielectric loss tangent, tan(δ), is

tan (δ) = ωε0ε′′r + σ

ωε0ε′r

(3)

with frequency-independent gas conductivity (σ , in S/m) [23].Plasma frequency is a function of electron density (ne, incm−3) given as

ωp =√

nee2

ε0me(4)

with elementary charge (e = 1.6 × 1019 C) and electron mass(me = 9.1 × 10−31 kg), simplifying to 56 500

√ne.

Gas number density (N , in cm−3) is calculated from theideal-gas equation of state and is 5.6 × 1018 cm−3 for shellsused in this experiment at 300 K. Electron collision frequencyis a function of gas pressure for noble gas plasmas andis available for electron temperature (Te) of 300 K in [24,Table I], and for higher Te in [25]. For shells used in thisexperiment, ν = 4.3 × 109 rad/s.

To calculate ε, first ne must be estimated which is possiblewith the HFSS model by simulating the on-state frequencyshift proportional to ne. The resulting value is 3.6 × 1012 cm−3

producing a frequency shift of 1.5 GHz, matching measuredresults. Calculated plasma frequency is 17.1 GHz, and calcu-lated values for ε′

r and tan(δ) at 10 GHz are −1.88 and 0.105,respectively. Conductivity shifts the on-state response down,

CROSS et al.: DEVELOPMENT OF LARGE-AREA SWITCHABLE PLASMA DEVICE 953

TABLE I

MOMENTUM-TRANSFER COLLISION RATE FOR NOBLE GASES

Gas ν/N (cm3/s)

He 7.6 × 10−9

Ne 7.6 × 10−10

Ar 2.2 × 10−9

Kr 1.8 × 10−9

Xe 5.3 × 10−8

-20

-15

-10

-5

0

-30

-20

-10

0

5 6 7 8 9 10 11 12 13 14 15 16 17 18

|S11

| (dB

)

|S21

| (dB

)

Frequency (GHz)

Off

On

Fig. 12. HFSS simulation of device in off- and on-states, with plasmaproperties of ne = 3.6 × 1012 cm−3, ν = 4.3 × 109 rad/s, and σ = 0.45 S/mfor normal incidence. Simulated and measured transmission responses agree.

and a simulated value of 0.45 S/m matches the measured off-state to on-state increase in minimum insertion loss of 4.2 dB.

The transmission response of the HFSS model withplasma modeled as a frequency-dependent complex permit-tivity medium is shown in Fig. 12. The shape of the simulatedon-state transmission response agrees with measured resultsin Fig. 9, particularly with increased passband frequency andattenuation in the on-state, validating the complex permittivityplasma model.

F. Power Usage

Device power usage is measured to be ∼150 W (powerdensity of 25 kW/m2), by measuring sustainer input powerwith and without the device connected, operating at 1200 Vppand 1 MHz. This value does not measure sustainer conversionefficiency, which is assumed to be high. Plasma power dissi-pation per unit volume can be estimated for noble gas plasmasdominated by two-body recombination by [24]

P

V= kn2

e�i e,

(W

cm3

)(5)

where k is the two-body dissociative rate constant and �i isthe energy required to ionize a noble gas using an electronbeam from Table II. For plasma-shells used in this experimentwith gas volume of 0.0284 cm3 and ne estimated previously,predicted device power usage is 55 W (power density of9.3 kW/m2).

Drive power can be dramatically reduced by using heliumgas with a much lower rate constant because of order of

TABLE II

DISSOCIATIVE RECOMBINATION RATE CONSTANT AND ELECTRON BEAM

IMPACT IONIZATION POTENTIAL FOR NOBLE GASES

Gas k (cm3/s) �i (eV)

He 1.0 × 10−8 41.5

Ne 1.8 × 10−7 36.2

Ar 9.1 × 10−7 26.2

Kr 1.6 × 10−6 24.3

Xe 2.7 × 10−6 21.9

magnitude longer plasma lifetime. Predicted power densitydrops by a factor of 16 (i.e., to 0.6 kW/m2). The mostsignificant driver of power usage is electron density, whichshould not be higher than needed. Further power reductionmay be achieved by reducing total plasma volume. An exampleof this is locating smaller plasma-shells in the center of eachFSS element where there is strongest EM interaction, notwasting power in areas of low-RF coupling.

Large-area plasma devices clearly require significant, butnot impractically large, power sources. The delivery of acpower in the ∼1-kV range at MHz frequencies is practical withclass E amplifiers, that can operate at tens of MHz at >90%efficiency, high power density, and can scale up to arbitrarilyhigh power levels. This is a practical power solution forplatforms, versus alternative approaches using lasers, electronbeams, and pulsed high-voltage supplies.

Switchable plasma devices can be very lightweight; thestructure is composed primarily of plasma-shells weighing71 mg each and dielectric material. FSS materials oftenrequire low dielectric constants, so foams and dielectrics withengineered voids are lightweight.

IV. CONCLUSION

This paper demonstrated a novel concept for RF–plasmainteraction in a scalable plasma-shell device. The proposedX-band switched plasma device was fabricated and measured,showing good scan angle independence and significant atten-uation in the passband averaging 7 dB. There are manyother avenues for increasing attenuation including optimizingthe plasma-shell gas composition and pressure, shell size,drive voltage, plasma excitation surface geometry, and usingadditional layers.

This concept has the potential to be scaled for use inmobile systems. Power dissipation can be reduced over anorder of magnitude with gas optimization and structuralchanges. Power sources can produce high-voltage, high-frequency waveforms with maximum efficiency and powerdensity. The proposed structure can be very lightweight,composed primarily of hollow shells and dielectric foam.Conformal structures are possible.

The concept can be adapted for operation in absorptivemode by altering the structure. Additional layers can be addedto create an electrically large graded plasma absorber, or asingle thick plasma layer can use tuned electron density fortotal absorption.

954 IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 41, NO. 4, APRIL 2013

Potential applications of large-area switchable plasma-shelldevices include: 1) active radomes; 2) switchable EM aper-tures; and 3) plasma metamaterials. This experiment was animportant initial step for practical large-scale plasma devicesto overcome traditional limitations concerning encapsulatingand driving large plasma volumes.

ACKNOWLEDGMENT

The authors would like to thank Dr. C. Bloebaum and Dr.P. Collopy of the NSF ESD Program. The authors also wouldlike to thank Dr. A. Weisshaar for his kind encouragement andsupport.

REFERENCES

[1] M. P. Bachynski, “Electromagnetic wave penetration of reentry plasmasheaths,” J. Res. Nat. Bureau Standards Sect. D, Radio Sci., vol. 69D,no. 2, pp. 147–154, Feb. 1965.

[2] J. P. Rybak, “Causes, effects and diagnostic measurements of the reentryplasma sheath,” Dept. Electr. Eng., Colorado State Univ., Fort Collins,CO, USA, Tech. Rep., Dec. 1970.

[3] R. A. Hartunian, “Causes and mitigation of radio frequency RF blackoutduring reentry of reusable launch vehicles,” Aerospace Corporation, ElSegundo, CA, USA, Tech. Rep. ATR-2007(5309)-1, Jan. 2007.

[4] J. Benford, A. S. John, and E. Schamiloglu, High Power Microwaves.New York, USA: Taylor & Francis, 2007.

[5] U. Jordan, G. D. Anderson, M. Backstrom, A. V. Kim, V. Arkady,M. Lisak, and O. Lundén, “Microwave breakdown in slots,” IEEE Trans.Plasma Sci., vol. 32, no. 6, pp. 2250–2262, Dec. 2004.

[6] R. S. Smith, T. E. Gary, D. C. Coleman, and J. T. Pappalardo, “Gas-breakdown transmit-receive tube turn-on times,” IEEE Trans. PlasmaSci., vol. 14, no. 1, pp. 63–66, Feb. 1986.

[7] H. Shapiro, “Electromagnetic scattering properties,” Ph.D. dissertation,Dept. Eng. Appl. Sci., California Inst. Tech., Pasadena, CA, USA, 1957.

[8] H. M. Musal, “On the theory of the radar-plasma absorption effect,” GMDefense Research Laboratories, Santa Barbara, CA, USA, Jul. 1963.

[9] T. Anderson, I. Alexeff, E. J. Raynolds, E. Farshi, M. S. Parameswaran,P. E. Pradeep, and J. Hulloli, “Plasma frequency selective surfaces,”IEEE Plasma Sci., vol. 35, no. 2, pp. 407–415, Apr. 2007.

[10] J. C. Vardaxoglou, “Optical switching of frequency selective sur-face bandpass response,” IEEE Electron. Lett., vol. 32, no. 25,pp. 2345–2346, Dec. 1996.

[11] D. P. Murphy, R. F. Fernsler, and R. A. Meger, “X-band microwave prop-erties of a rectangular plasma sheet,” Naval Research Lab, Washington,DC, USA, Tech. Rep. NRL/MR/6754–99-8368, May 1999.

[12] S. Larigaldie and L. Caillault, “Dynamics of a helium plasma sheetcreated by a hollow-cathode electron beam,” J. Appl. Phys., vol. 33,no. 24, pp. 3190–3197, Jul. 2000.

[13] Y. S. Zhang and J. E. Scharer, “Plasma generation in an organicmolecular gas by an ultraviolet laser pulse,” J. Appl. Phys., vol. 73,no. 10, pp. 4779–4784, Jan. 1993.

[14] K. L. Kelly, B. White, and S. Tusk, “Laser ionization and radio frequencysustainment of high-pressure seeded plasmas,” J. Appl. Phys., vol. 92,no. 2, pp. 700–709, Jul. 2002.

[15] B. A. Munk, Frequency Selective Surfaces. New York, USA: Wiley,2000.

[16] B. A. Munk, Finite Antenna Arrays and FSS. New York, USA: Wiley,2003.

[17] Rogers Corp. (2013, Jan. 8). RT/Duroid 5870/5880 HighFrequency Laminates, Rogers, CT, USA [Online]. Available:http://www.rogerscorp.com/documents/606/acm/RT-duroid-5870-5880-Data-Sheet.pdf

[18] Rogers Corp. (2013, Jan. 8). RO4000 Series High FrequencyCircuit Materials, Rogers, CT, USA [Online]. Available:http://www.rogerscorp.com/documents/726/acm/RO4000-Laminates—Data-sheet.pdf

[19] F. M. Penning and C. C. Addink, “The starting potential of the glowdischarge in neon-argon mixtures between large parallel plates,” Physica,vol. 1, nos. 7–12, pp. 1007–1027, May 1934.

[20] E. F. Peters, “Control apparatus for supplying operating potentials,” U.S.Patent 3 959 669, May 25, 1976.

[21] B. K. Velayudhan, W. G. Jeffrey, and A. W. Carol, “Energy recovery inplasma display panel,” U.S. Patent 6 917 351, Jul. 12, 2005.

[22] L. F. Weber, W. W. Kevin, and B. W. Mark, “Power efficient sustaindrivers and address drivers for plasma panel,” U.S. Patent 4 866 349,Sep. 12, 1989.

[23] C. A. Balanis, Advanced Engineering Electromagnetics. New York,USA: Wiley, 2012.

[24] R. J. Vidmar, “On the use of atmospheric pressure plasmas as electro-magnetic reflectors and absorbers,” IEEE Trans. Plasma Sci., vol. 18,no. 4, pp. 733–741, Aug. 1990.

[25] P. Baille, J. S. Chang, A. Claude, R. M. Hobson, G. L. Ogram, andA. W. Yau, “Effective collision frequency of electrons in noble gases,”J. Phys. B, Atomic Molecular Phys., vol. 14, no. 9, no. pp. 1485–1495,May 1981.

Lee W. Cross (M’12) received the B.S. and M.S.degrees in electrical engineering from The Univer-sity of Toledo, Toledo, OH, USA, in 2002 and2007, respectively, where he is currently pursuingthe Ph.D. degree with research in microwave filterdesign.

He was with Imaging Systems Technology as aDirector of RF Research and is the Principal Inves-tigator of research projects developing technologyutilizing high-power RF and plasma devices.

Mohammad J. Almalkawi (M’12) received theB.S. degree in telecommunications engineering (withHons.) from Yarmouk University, Irbid, Jordan, theM.S. degree in electrical engineering from the SouthDakota School of Mines & Technology, Rapid City,SD, USA, and the Ph.D. degree in engineering fromthe University of Toledo, Toledo, OH, USA, in 2007,2009, and 2011, respectively.

He was an Intern-RF Modeling Engineer withFreescale Semiconductor, Inc., Tempe, AZ, USA,in 2011, working on modeling RF/microwave com-

ponents for the development of product-level models. In 2012, he joinedthe EECS Department, University of Toledo, as a Post-Doctoral ResearchAssociate. He is currently a Research Assistant Professor. His current researchinterests include RF/microwave circuit design and modeling, UWB antennas,computational electromagnetics, and electromagnetic compatibility.

Dr. Almalkawi is currently a member of the Editorial Board of theInternational Journal of RF and Microwave Computer-Aided Engineering.

Vijay K. Devabhaktuni (SM’09) received theB.Eng. degree in electrical and electronics engi-neering and the M.S. degree in physics from BirlaInstitute of Technology and Science, Pilani, India,in 1996, and the Ph.D. degree in electronics fromCarleton University, Ottawa, ON, Canada, in 2003.

He has been an Associate Professor with theEECS Department, University of Toledo, Toledo,OH, USA, since 2008. He was a Professor withPenn State Erie Behrend in 2005. His R&D interestsinclude applied electromagnetics, biomedical appli-

cations of wireless sensor networks, computer aided design, device modeling,image processing, infrastructure monitoring, neural networks, optimizationmethods, RF/microwave devices, and virtual reality applications. He has co-authored more than 150 peer-reviewed papers, and is advising 11 M.S. andPh.D. students and two research faculty.

Dr. Devabhaktuni currently serves as the Associate Editor of the Interna-tional Journal of RF and Microwave Computer-Aided Engineering and theIEEE Microwave Magazine. He is a registered member of the Associationof Professional Engineers, Geologists, and Geophysicists of Alberta. He isa member of TC 36 on New Technologies and Applications for the IEEEInternational Microwave Symposium 2013. He was the recipient of theCarleton University’s Senate Medal at the Ph.D. level, and several teachingexcellence awards in Canada and USA. He held the Natural Sciences andEngineering Research Council of Canada postdoctoral fellowship and spentthe tenure researching at the University of Calgary during 2003–2004. He heldthe prestigious Canada Research Chair in Computer-Aided High-FrequencyModeling and Design at Concordia University, Montreal, during 2005–2008.