Threshold Criterion for a Space Simulation

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 87, NO. A3, PAGES 1565-1573, MARCH 1, 1982

Threshold Criterion for a Space Simulation Beam-Plasma Discharge E. P. SZUSZCZEWICZ, 1 K. PAPADOPOULOS, 2 W. BERNSTEIN 3 C S LIN 4 AND D. N. WALKER1 We have conductedan experimentaland theoretical studyof the thresholdcharacteristicsof a space simulationbeam-plasmadischargewith emphasison densityprofilesand a density-dependentignition criterion. The studyincludedvariousbeam-plasmaconditionscoveringbeam currentsfrom 8 to 85 ma, beam energies from 0.8 to 2.0 keV, and magnetic fields at 0.9 and 1.5 G. The study included experimenta!determinations of radial profilesof electrondensityfor eachof the selectedconditions extending from a low-density, pre-beam-plasma dischargestate to a strong beam-plasma discharge

condition. At beam-plasma discharge threshold it wasdetermined that(%,/rOc) = 5.4wasthedensitydependent ignition criterion. The experimental results are shown to agree with detailed model calculations,which considerthe beam-plasmadischargeto be producedby large-amplitudeelectron plasma waves resultingfrom the beam-plasmainteraction.

1.

INTRODUCTION

as

A cold electron beam propagatingthrough a weakly ionized plasma will, under proper conditions,producea modified beam-plasmastateknown as the beam-plasmadischarge (BPD). This dischargestate has received considerableattention in recent years as a result of increased interest in mechanisms for vehicle neutralization during spaceborne accelerator experiments [Bernstein et al., 1980; Carnbou et al., 1978; Galeev et al., 1976], enhancedbeam-plasmaionization processes [Bernstein et al., 1978], and, in general, single-particle or collective phenomena initiated by beams injected into neutral gas and charged-particleenvironments [Hess et al., 1971; Winckler et al., 1975; Hendrickson and Winckler, 1976; Carnbou et al., 1975; Monson and Kellogg, 1978; Szuszczewicz et al., 1979; Jost et al., 1980; Winckler, 1980]. The BPD appears at a critical energetic-electron-beam

current IBc, with the transitionfrom single-particlebehavior (IB < I• c, pre-BPD) to collective processes(I• > I• •, solid BPD) described as follows for conditions in which the plasma is created by the beam itself. 1. As an electron beam linearly interacts with a neutral gas, it collisionally produces a plasma with a density that varies directly with the magnitudeof the beam current for a fixed beam energy. 2. As the beam current is increasedto a critical value I• •, a two-stream instability sets in and the electric fields of the excited waves heat the electronsto energiescomparableto the ionization energy. The 'heated' electrons create an enhanced ionization process, which results in an avalanche breakdown, the BPD. Bernstein et al. [1979] have reported the dependence of this critical current IB• on various experimentalparameters

VB3/2

I•• orBø'7pL

(1)

where VB, B, P, and L are the beam energy (voltage), the superimposedmagnetic field, the ambient neutral pressure, and the beam length (gun-to-collectordistance), respectively. While the Ia c - Ia•(VB, B, P, L) relationship was establishedamongthe controllingsystemparameters (with

theIB• ocp-1 behaviorapplicable at P •< 10-5 torr),a clear dependenceon plasmadensity was expected. Early thoughts

suggestedthat top> to• satisfiedignitionthresholdcriteria [Bernstein et al., 1979; Getty and Smullin, 1963]. We have

since had the opportunity to test this idea under various beam-plasmaconditions.Our resultsinclude (1) time-delay data and associated analyses indicating that BPD ignition does indeed occur at a critical density, (2) direct measurements of plasma density near ignition threshold with deter-

minationsof •op/•o•, and (3) a theoreticalanalysisthat predicts the critical density criterion.

In subsequentsectionswe presentthe experimentaland theoretical details, which establishthe density-dependent

thresholdconditionsfor BPD at 3.5 •< •op/•oc •< 6.9. EXPERIMENT

CONFIGURATION

AND RESULTS

The experiment was conductedin a large vacuum chamber facility at the NASA JohnsonSpace Center with an experimental configurationsimilar to those employed in earlier investigations[Bernsteinet al., 1979]. The configurationis illustrated in Figure 1 with specificemphasison the pulsed

plasmaprobemeasurement andtheassociated prOCedUre for

%,/•%determinations. Not shownare the previouSly'de scribed diagnostics[Bernsteinet al., 1979;Jost et al., 1980]

including the 3914-• scanning photometer, the segmented currentcollector,the energeticelectronelectrostatic analyz-

I IonosphericDiagnosticsSection,E. O. Hulburt Center for Space Research, Space Science Division, Naval Research Laboratory, Washington, D.C. 20375.

er, andtheremotewavedetection antenna andspectrum analyzer system.The beam was generatedby a tungsten:

cathodePierce-type diodegun,mounted on a position-

controlledcart. In all casesthe beamwas injectedparallelto 2 ScienceApplications, Inc., McLean,Virginia22102. 3 Centerfor SpacePhysics,Departmentof SpacePhysicsand the magneticfield/• and terminatedon a 3 x 3 m target

Astronomy, Rice University, Houston, Texas 77001.

suspended about20 m abovethe gunaperture.A combina -

4 Bendix Field EngineeringCorporation,Columbia,Maryland tion of coil current and the earth's magneticfield established the B field at selected levels up to 1.5 G. In most casesthe beam was injected into a neutral gaswith This paperis not subjectto U.S. copyright.Publishedin 1982by

21O45.

the AmericanGeophysical. Union. Paper number 1A1703.

no pre-beamplasma;however,the experimental investig a1565

1566

SzuszczEwlCZ ET AL.' SPACE SIMULATION BEAM-PLASMA DISCHARGE

Equation (2) has the solution

Ne(t)= 7__ (1- e-vt)

(3)

v

where •/= lB •rNo/eAis the electron-ionpair productionrate, v is the loss rate, •r is the electron impact ionization cross section, No is the neutral density, and A is the beam cross section. At short times, the density will increase linearly with time (Ne = •/t), and at longer times the density will

approach thetimeindependent valueNe = •//v.If, indeed, there is a critical densityNec requiredfor BPD ignitionand if we define its relationshipto a critical beam currentby Nec = (•rNo/eAv)IBc, then the time delay td until Ne(t) = Nec can be obtained from (3) and written as

td= vln1-•-ff]

(4)

For lB • IBc, relatively long delay times shouldbe observed with a rapid decreasein td as lB is increased.This is exactly the behavior demonstratedin Figure 2, where the solidcurve has been normalizedto the data by selectingIBc = 20 ma and

Fig. 1. Experiment configuration.

tion included two cases in which the chamber was filled with

v = 400s-]. Thegoodagreement betweenthesimplemodel calculationsand the experimentalresults supportsthe concept of a critical density thresholdfor BPD ignition, but the normalizationprocedureprovideslittle measurefor an absolute value of Nec. Specific determinations of the critical density threshold criteria are detailed in the next two sec-

a plasma created by a Kauffman-type argon ion thruster. In these casesthe pre-beamplasmadensitywas lower than the critical density at BPD ignition. The investigation was conducted in two stages. First, pulsed-gunexperimentswere carried out to test the concept tions. of a density-dependentthresholdconditionagainstthe origiDIRECT MEASUREMENTS OF CRITICAL DENSITY nal ideas of Getty and Srnullin [1963]. After the concepts check, a seriesof direct density measurementswas conductTo establishquantitativelythe density dependentthreshed to quantitatively establish the density criteria. We de- old criterion for the ignition of a space simulationBPD, scribethe resultsin the order in which the experimentswere emphasiswas placed on the direct measurementof plasma conducted. density profiles over a range of beam-plasmaconditions covetingbeamcurrentsfrom 8 to 85 ma, beamenergiesfrom TIME DELAY ARGUMENTSFOR Ne c 0.8 to 2.0 keV, and magnetic fields at 0.9 and 1.5 G. The

In pulsed-beam experiments with lB >> IBc, Getty and Smullin [1963] identified three sequential phases in the temporal evolution into the BPD: (1) a quiescent stable period, (2) an intermediateperiod during which the gross featuresof the beamremainedunchangedbut waveswithf= fee were present, and (3) BPD ignition. They suggestedthat

phases1 and2 corresponded to thetimerequiredfor buildup of the ambient density to a critical value... a value at which the BPD ignited. To test this conceptin the spacesimulation beam-plasmaefforts at the JSC facility a series of pulsedbeam measurementswere performedto study the temporal evolution of the BPD in the configurationshownin Figure 1. All diagnostics sensitive to the transition to BPD demonstrated identical time delaysbetween beam current initiation

Legend= 0) Meosured Delaysß

C•iti••• =I• keV, • 1.56 g.uss --

andBPDignition.At lowpressure ( tOc -- I•

COS0i

(1+ (8.5/g)sin 20a) 1/2

Chen, F. F., Plasma Diagnostic Techniques,chap. 4, edited by R. H. Huddlestone and S. L. Leonard, Academic, New York, 1965. Galeev, A. A., E. V. Mishin, R. Z. Sagdeev,V. D. Shapiro,and V. I. Shevchenko,Dischargein the regionarounda rocket following the injection of electron beams in the ionosphere, $ov. Phys. Dokl., 21,641, 1976. Getty, W. D., and L. D. Smullin, Beam-plasmadischarge:Buildup of oscillations, J. Appl. Phys., 34, 3421, 1963. Hendrickson, R. A., and J. R. Winckler, Echo III: The study of electric and magnetic fields with conjugateechoesfrom artificial electron beams injected into the auroral zone ionosphere, Geophys. Res. Lett., 3, 409, 1976. Hess, W. N., M. C. Trichel, T. N. Davis, W. C. Beggs,G. E. Kraft, E. Strasinopoulos,and E. J. R. Maier, Artificial aurora experiment: Experiment and principal results, J. Geophys. Res., 76,

6067, 1971. (10) Holmes, J. C., and E. P. Szuszczewicz, A versatile plasma'probe,

We can see that the top/tOc ratio at threshold(with • • 1

Rev. Sci. Instrum., 46, 592, 1975.

Holmes, J. C., and E. P. Szuszczewicz,A plasmaprobe systemwith microperv) is rather insensitiveto the beam injectionangle automatic sweep adjustment,Rev. $ci. Instrum., 52, 377, 1981. for 0i •< 60ø. In Table 2 we presentthe valuesof top/tOc Jost, R. J., H. R. Anderson, and J. O. McGarity, Measured electron energy distributions during electron beam-plasmainteractions, computedon the basisof equation(10) by takingcos Oi/(1+ Geophys. Res. Lett., 7, 509, 1980. (8.5/•) sin2 0•)•/2• 1fornearparallelinjection. FromTable2 Linson, L. M., and K. Papadopoulos,Review of the statusof theory we findthe averagecomputedvalueto be top/tOc - 4.95.This and experimentfor injectionof energeticelectronbeamsin space, result, while subject to moderate uncertaintiesin ro/R, is Rep. LAPS-69/SAI-D23-459-LJ, Sci. Appl., Inc., LaJolla, Calif., taken to be in excellent agreementwith the experimentally April 1980. derived conditions (Table 1 and equation (4)) providing Monson, S. J., and P. J. Kellogg, Ground observationsof waves at 2.96 MHz generated by an 8- to 40-KEV electron beam in the complementaryargumentsthat confirm the original suggesionosphere, J. Geophys. Res., 83, 121, 1978. tion that the ignition of the BPD was coupled to a density- Papadopoulos,K., Theory of beam plasma discharge,in Proceeddependentthreshold criterion. We shouldmention that the ings of NATO Advanced Research Institute on Artificial Particle Beams in Space Plasma Physics, Plenum, New York, in press, above theory is consistentwith the observationalfacts that

the temperatureof the plasmaelectronsis much lower than the ionization potential [Szuszczewiczet al., 1979], and energetic tails are observed within the beam-plasma core [Szuszczewicz, 1982].

Acknowledgments. The various elements in this investigation were supportedin whole or in part by NASA grantNAGW-69, Rice University subcontractNAS 8-33777-1,andNASA/NOAA contract NA79RAE0039. Supplementaryfundingwas also providedby the Office of Naval Research under Program Element 61153N-33 in Task Area RR033-02.

We wish to thank J. C. Holmes for his critical

carein electronicsdesignof the p3 instrumentation, L. Kegleyfor technicalassistancein experimentexecution,and L. Linsonfor his review and comments on the manuscript. The Editor thanks R. J. Jost and P. J. Kelloggfor their assistance in evaluating this paper.

REFERENCES

Bernstein, W., H. Leinbach, P. Kellogg, S. Monson, T. Hallinan, O. K. Garriott, A. Konradi, J. McCoy, P. Daly, B. Baker, and H. R. Anderson, Electron beam injection experiments: The beam-plas-

1982.

Papadopoulos, K., and T. Coffey, Nonthermal features of the auroral plasma due to precipitating electrons, J. Geophys.Res., 79, 674, 1974a.

Papadopoulos, K., and T. Coffey, Anomalous resistivity in the auroral plasma, J. Geophys. Res., 79, 1558, 1974b. Papadopoulos,K., and H. L. Rowland, Collisionlesseffectson the spectrum of secondary auroral electrons at low altitudes, J. Geophys. Res., 83, 5768, 1978. Rowland, H. L., C. L. Chang, and K. Papadopoulos,Scalingof the beam-plasmadischarge,J. Geophys.Res., 86, 9215, 1981. Szuszczewicz, E. P., Direct measurementsof plasmacharacteristics in space-simulationbeam-plasmainteractions,paper presentedat 20th Aerospace Sciences Meeting, Am. Inst. of Aeronaut. and Astronaut, Orlando, Fla., Jan. 1982. Szuszczewicz, E. P., and J. C. Holmes, Surface contamination of

active electrodesin plasmas:Distortion of conventionalLangmuir probe measurements,J. Appl. Phys., 46, 5134, 1975. Szuszczewicz,E. P., and J. C. Holmes, Reentry plasmadiagnostics with a pulsed plasma probe, paper presented at 9th Fluid and PlasmaDynamics Conference,Am. Inst. of Aeronaut. and Astronaut., San Diego, Calif., July 1976. Szuszczewicz, E. P., and J. C. Holmes, Observations of electron temperaturegradientsin mid-latitudeEs layers, J. Geophys.Res., 82, 5073, 1977. Szuszczewicz, E. P., D. N. Walker, and H. Leinbach, Plasma

SZUSZCZEWlCZ

ET AL.:

SPACE SIMULATION

diffusion in a space simulationbeam-plasmadischarge,Geophys. Res. Lett., 6, 201, 1979.

Winckler, J. R., The application of artificial electron beams to magnetosphericresearch, Rev. Geophys. Space Phys., 18, 659, 1980.

Winckler, J. R., R. L. Arnoldy, and R. A. Hendrickson,Echo 2: A

BEAM-PLASMA

DISCHARGE

1573

study of electron beams injected into the high-latitudeionosphere from a large soundingrocket, J. Geophys.Res., 80, 2083, 1975. (Received June 29, 1981; revised October 2, 1981; accepted October 30, 1981.)