high-energy electron beam-induced ionospheric modification experiments

8/6/2019 High-Energy Electron Beam-Induced Ionospheric Modification Experiments

1/26

June 30, 1996

. .PL- TR-96-22S6

HIGH-ENERGY ELECTRON BEAM-INDUCEDIONOSPHERIC MODIFICATION EXPERIMENTS

Brian GilchristTorsten NeubertLinda Habash Krause

University of MichiganSpace Physics Research LaboratoryDepartment of Atmospheric, Oceanic & Space Sciences2455 Hayward StreetAnn Arbor, MI 48109-2143

Final ReportMay 1993 - September 1996

Approved for public release; distribution unlimited.

PHILLIPS LABORATORY, 1 9 9 7 0 9 1 8 1 2 6Directorate of GeophysicsAIR FORCE MATERIEL COMMANDHANSCOM AFB, MA 01731-3010


2/26

. ,

"This technical report has been reviewed and is approved forpublication II

(/ /',-./)i ...7L / /fJ 'lO / A2W:~DAVID N. ANDERSONBranch ChiefKEpH M. GROVESContract ManagerCHARLES P. PIKEDivision Director

Qualified requestors may obtain additional copies from theDefense Technical Information Center (DTIC).If your address has changed, if you wish to be removed from themailing list, or if the addressee is no longer employed by yourorganization, please notify PL/IM, 29 Randolph Road, Hanscom AFB,MA 01731-3010. This will assist us in maintaining a currentmailing list.Do not return copies of this report unless contractualobligations or notices on a specific document require that it bereturned.


3/26

----------------------------------------,

REPORT DOCUMENTATION PAGE Form ApprovedOM8 No. 07040788P"JCIICtOOr't'lng ~urO@''''or ~~I\ C";',,!,,:,:,onot :ntC'n"'3tJO" I~~tll 'Mat!'O to av~aq@' 1 I"IOU(~r ."~r.\e' .. nC~I,,;OIl\qt~ t im! ' for " !' v1ewng Instructions, searcnlnq ~:':'5tln9 data sourceS,g,Jt!'ltnnq ,JndmalnUIr.:nq ~"e cara -~~. ~nd(CIT'OI!trnq.inc r~I~lnq ~"!oll~ctlon of Information )tnc ccmmtnn r~a,ralng thIsbyrd!" ~1.t lma(eor ~n' t '1 tht r a~ ot t""ScollKtlon ot Intormatlot', ,n('luOH'\'; \wg'i~tlon' 'or (!'CueIng tl"U\::)Ufo!n. =0Nasn,nqton I""ItadQuart!1'\ ServICe1.Jlr~oratt r?, Information O~ra[lon, and Re"OOrts.IZ1S jeHPflOn011\11,Highway. SUit@'12C4, ;"(hngtCn. IA 22202-4302 . .tnd t o (ne Offi(l~' Jf Manaqement dod BUdget. P'~perwor~ fteductJon J)rOJect (0704.0188), 'Nurunqton, ::::;C20503.1 . AGENCY U SE ONLY (Leave blank) 1 2 . R EPO RT D AT E 1 3 . REPORT TYPE AND OATES COVERED

30 June 1996 Final Report: May 1993-September 19964 . T ITLE A ND SU BT ITLE S . FUND ING NUM BERS

High-Energy Electron Beam-Induced Ionospheric PE 63402FM od if ic at io n E xp er im en ts PR 4643 TA GH WU AG

6 . A U TH O R(S ) Contract F19628-93-K-0003Brian E. Gilc_hristTorsten NeubertLinda Habash Krause7 . PERFO RM IN G O RG AN IZA TIO N N AM E(S ) A ND A DO RESS(ES ) 8. P ER FO RM IN G O RG AN IZ AT IO N

University o f M ic hig an R EPO RT N UM BE RSpace Physics R ese ar ch L ab or at or yDepartment of Atmospheric, Oceanic and Space Sciences2455 Hayward StreetAnn Arbor, MI 48109-2143

9 . S PO N SO R IN G I MON ITOR ING AGENCY NAME(S ) AND ADoRESS(ES ) 10 . SPONSOR ING I MONITOR INGA GENCY REPORT NU MBERPhillips Laboratory

29 Randolph Road PL-TR-96-22S6Hanscom AFB, MA 01731-3010Contract Manager: K ei th G ro ve s/ GP IA

11. SU PPLE ME NT AR Y N OT ES

12a. DISTR IBUT ION I AVA IL A B IL IT Y S TA T EM EN T 12b. D ISTR IBU TIO N CO DEApproved for public release;distribution unlimited

1 3. A B ST RA C T (MaXImum 200 words)The University of Michigan Space Physics Research Laboratory was given the task ofproviding scientific investigations of beam propagation physics associated with theinjection of a MeV electron beam into the earth's atmosphere. Motivation for thisstudy was driven by the reduction in weight and size of electron beam acceleratorsin this energy regime to the point which enables them to be flown on balloons,rockets, or spacecraft. Program goals included the modeling of dynamics of MeVel ec tr on b eam s injected from LEO spacecraft in to th e a tm os ph er e, analysis of theco ll is io na l an d b eam -p la sma interactions, and modeling of ionospheric modificationind.uced in the atmosphere, such as enhanced plasma densities, optical emissions andc on du ct iv it y c ha ng es . Results show that, for downward directed beams, radialdefocusing of be~ electrons due to scattering by the atmospheric neutrals issignificantly mitigated by the presence of the earth's magnetic field. In addition,substantial ionospheric modification occurs due to t he be am a tm os ph er e i nt era ct io n,resulting in plasma densities and c on du ct iv it ie s s ig ni fi ca nt ly ~a bo ve t he a mb ie ntvalues. Proposed future work includes the study of optical and bremsstrahlung-nissions to be used tor dia gn o sti.c ot beams injected rrom space, b ea m p ro pa ga ti ondynamics over l on g d is ta nc es , a nd m od if ic at io n o f th e a tm os ph er ic e lec tr ic p ot en ti al.

1. S UB JE CT T ER MS 15. NUM BER O F PAG ESRelativistic Electron Beam Propagation in Gas, Ionospheric 26Modification, Paraxial Envelope Equations. 16 . PR ICE C OO E

17. S EC UR IT Y C lA SS IF IC A nO N lB . S ECUR ITY C l ASS I FI CATION 1 9. S EC U RIT Y C LA S SIF IC A TI ON 20. L IM ITA TIO N O F A BST RA CTOF REPORT OF TH IS PAGE OF ABSTRACTUnclassified Unclassified Unclassified SAR-

..

Standard Form 298 (Rev 2a9)p'~rO


4/26

111

CONTENTS1. BACKGROUND...............................................................................................2. PROGRAM OBJECTIVES 13. RESULTS .....................................................................................3.1 BEAM-PLASMA INTERACTION 23.2. BEAM-ATMOSPHERE INTERACTION .................................................

3.2.1.MONTE CARLO SIMULATIONS ........................................................................3.2.2. ENVELOPE EQUATIONS 33.2.3. SIMULATION RESULTS 5

3.3. ATMOSPHERIC MODIFICATION 123.3.1. ELECTRON DENSITY 123.3.2. ELECTRIC CONDUCTIVITY 13

3.4. MIDDLE ATMOSPHERIC DYNAMICS 143.4.1. INTRODUCTION 143.4.2. ATMOSPHERIC ELECTRIC POTENTIAL 163.4.3. ATMOSPHERIC COMMUNITY INTERESTS 18

4. FUTURE STUDIES 185. PUBLICATIONS AND PRESENT ATIONS 196. REFERENCES 20


5/26

I V


6/26

1

1. BACKGROUNDElectron beam (linac) accelerators in the MeV range have been reduced in weight andsize to a point that enable them to be flown on balloons, rockets or spacecraft. This opensup the opportunity for a new generation of space-borne relativistic electron beamexperiments, as discussed in the AFGL study report by Banks et at. [1987]. Acting onthe results of the 1987 study, the Air Force solicited proposals under the SBIR programfor a sounding rocket experiment module including a linac. A contract was awardedSystem Planning Corporation (SPC) which is now heading a Phase 2 effort that wi11leadto a flight-ready sounding rocket module. The beam characteristics of the linac are: 1=100 r nA , E = 5.2 MeV, pulse length = 4 us, duty cycle = 0.1%.

On the background of these activities, the University of Michigan Space PhysicsResearch Laboratory was awarded the present contract by the Air Force to providescientific investigations of beam propagation physics and atmospheric interactions. Thecontract was in place at the University of Michigan in May 1993, and work was initiatedat the end of that month.

2. PROGRAM OBJECTIVESAt the outset, the following overall goals for the program were defined:

(1) to model the dynamics of MeV-energy beams injected from LEOspacecraft into the atmosphere, including beam-plasma interactions inthe ionosphere and collisional interactions in the atmosphere.

(2) to model the modification induced in the atmosphere such as enhancedplasma densities, optical emissions, and conductivity changes.

(3) to assess from such models changes induced in the atmosphericelectric structure and under what conditions triggering of upwardlightening is likely for beams injected over thunder clouds.

(4) to explore modifications to middle atmospheric dynamics induced byenergetic particle precipitation.

(5) to assess the interest in the aeronomy community for an activeexperiment of this nature.


7/26

2

Because of the transfer of Air Force technical oversight at a mid-point in the contractfrom ionospheric to space plasma disciplines, not all of the above areas received equalattention. In particular, emphasis was given to areas (1) and (2) with lesser efforts for theothers.

3. RESULTS3.1 BEAM-PLASMA INTERACTIONAs the beam exits the accelerator payload, it encounters the ambient ionospheric plasmaand is immediately subject to beam-plasma interactions (BPI). A quantitative model ofBPI as it manifests in active space experiments has yet to be developed, partly becausethe problem is a difficult one, and partly because experiments with adequate plasmadiagnostics have been relatively few. To develop a model for BPI (funded on othercontracts), collaboration was established with Dr. G. Khazanov of the Space PhysicsResearch Laboratory at the University of Michigan. Dr. Khazanov has written a kineticmodel based on the Boltzmann equation describing transport of suprathermal electrons inthe ionosphere-plasmasphere and energetic auroral particle precipitation. In our firsteffort, the characteristics of a 10 keV beam was studied under the assumption thatLangmuir turbulence elastically scatter beam electrons. It was clear from this study thatthe more difficult part is to parameterize properly the scattering of beam electrons byLangmuir turbulence. Thus, only elastic back-scattering was considered whileexperiments clearly show the importance of inelastic scattering. A paper describing theresults was published in Geophysical Research Letters [ Kh aza no v e t a i., 1993].3.2. BEAM-ATMOSPHERE INTERACTION3.2.1. Monte Carlo Simulations.Previous analytical efforts [Banks, 1987] were able to identify applicable models ofbeam atmospheric ionization, associated energy loss, and beam spreading withoutmagnetic field focusing. From this early effort, the importance of including magneticfield focusing effects was identified since beam spreading, due to elastic collisionprocesses, dramatically reduced the expected ionization density. It was, therefore,considered a top priority to investigate magnetic field effects on beam dynamics.


8/26

3

Efforts to model the collisional interaction of a beam with the earth's atmosphereincluding the earth's magnetic field has been studied in two ways. The first is theadaptation of already existing Monte Carlo algorithms, while the second is an analyticalapproach. We focused on predictions of energy deposition rates, the spreading of thebeam perpendicular to the ambient magnetic field, and the penetration depth of the beam.For this purpose a collaboration was begun with Professors W. Martin, Y. Y. Lau, andwith Dr. S. Wilderman, all of the Department of Nuclear Engineering at the University ofMichigan.

Two different Monte Carlo transport algorithms have been applied. One is the EGS4program, which has been used extensively and thoroughly benchmarked by both thehigh-energy physics and the medical physics communities. EGS4 uses the condensedrandom walk transport algorithm in which a particle's full transport path is broken intoroughly a hundred segments. Aggregate deflection and ionization through each segmentis assumed to be described by analytically derived distribution functions, and the fullparticle path is modeled by successive application of appropriate segment-wise scatteringdistribution data.

As a means of verification, a more computationally intensive algorithm, which treatsall elastic scattering events individually and employs best available differential scatteringdistribution data, was also applied to this problem. Both simulations produced similarresults, and we are satisfied that the algorithms can be used for our purpose.

Simulations were performed with the beam injected vertically downwards from 60km altitude (the altitude where beam-atmosphere interaction becomes significant at 5MeV). The atmosphere is modeled as eleven separate 2 km thick 'slabs' composed of 70%N2 and 30% 02, and with constant local density. Energy deposition is tallied for cubicregions of volume (1 km x 1 km x 100 m) for simulations without the earth's magneticfield and (100 m)3 for simulations which include the magnetic field. The energydeposition profiles in altitude and in lateral dimensions are determined in this way.Simulations of a balloon-borne beam accelerator were also performed with an upwarddirected beam injected at 40 km altitude.

3.2.2. Envelope EquationsWith the cooperation of Professor Y.Y. Lau, we investigated a class of closed formanalytical expressions to estimate spreading of relativistic electron beams, the so called"paraxial envelope equations" [Murphy et al., 1992; Lawson, 1988; Lee and Cooper,1976]. These equations may be adapted to the case of beam propagation over hundreds


9/26

( 4 )

of kilometers of distance while accounting for magnetic field focusing effects. Theenvelope equations utilize a paraxial approximation (i.e., vzvr) to simplify theequations of motion while still tracking the effects important to radial beam expansion.Using the envelope equations, it is possible to allow for expansion effects other thanelastic scattering, such as electrostatic self-expansion.The envelope equations are closed-form analytical expressions to estimate thespreading of relativistic electron beams [Humphries, 1990]:

(1)

d{(2

) = d{(2

) +__:q_j(2) drdz dz r 2 -1\ dzc

(2)

d{(2) = 1 6 nNZ (Z + 1)re2 In [ 2 0 4 ]dz r 2 f 3 4 ZO . 333c

(3)

The beam is propagating in the z-direction, and the radius of the beam, R, is a function ofz. These are the envelope equations per se, where we have only retained terms that are ofsignificance for the problem at hand. In 0), the first term on the right-hand siderepresents focusing processes arising from acceleration or deceleration of electrons withposition along the z-axis. The second term represents applied magnetic fields, in our casethe earth's magnetic field, while the third term describes defocusing processes fromscattering and deceleration. The final term represents the defocusing effect arising fromthe conservation of canonical angular momentum. In the equations, Ro is the beamradius at injection, 'Y is the relativistic factor, ~ is vzlc, c is the velocity of light, and weethe non-relativistic electron gyro frequency.

4


10/26

5

Equation (2) describes defocusing by emittance due to collisions (first term) anddeceleration (second term). Here (92) is the mean-squared divergence angle. The small-angle scattering formula is shown in (3), where Nand Z are properties of the medium, reis the Bohr radius, and yand B are functions of z. The equations are closed by the Betheformula (4), which gives the collisional stopping power for relativistic electrons.Itwas hoped that the equations could be simplified by the introduction of the so-called Nordsieck length [Lee and Cooper, 1976; Murphy et al., 1992]. However, theassumptions under which the Nordsieck length is derived (high-current beams) are notvalid for our case, and we are forced to solve the complete set of Equations (1)-(4).

3.2.3. Simulation Results

In the Monte Carlo simulations a beam is injected downwards from 60 km altitude orupwards from 40 km altitude. The upward injection simulates the case of a linac carriedon a balloon, and the downward injection the case of a sounding rocket experiment. Thealtitude of 60 km was chosen to limit the computational effort. The atmosphere abovethis altitude is relatively tenuous, and little scattering/ionization is experienced above 60km.

The atmosphere is the MSIS86 model for night-time, mid-latitude conditions[Hedin, 1987]. Figure 1 shows the fractional energy deposition of a 5 MeV beam injecteddownwards from 60 km altitude. The black curve represents the Monte Carlo results, andthe red curve represents the results using the Bethe formula. As can be seen, there is verygood agreement between the two. The Bethe formula predicts an altitude limit at 42 km;the Monte Carlo results peak here, but they predict that a small fraction of electrons willpenetrate to 40 km altitude. The good agreement was anticipated since, after all, theEGS4 code incorporates the Bethe formula.


11/26

50

10-6 10-4

Beam Down @ 60 kmE =5.0 MeV; B =0.4 G

60

55

45

Fractional Energy Deposition (m=-l )

Figure 1. Fractional Energy Deposition: The Bethe formula (red curve) andMonte Carlo Simulation (black curve).

Turning now to the problem of the radial expansion of the beam and the influenceof the magnetic field on this expansion, the case of a downward injected beam from 60km altitude and no magnetic field is shown in Figure 2. The fractional energy depositionas a function of altitude and horizontal distance is shown on a gray-scale together withthe analytical results obtained from the envelope equations. We notice an excellentagreement between the two approaches. It is also evident that without a magnetic field,the beam experiences substantial spreading in the horizontal direction, up to 8 kmat 42km altitude. As mentioned earlier, this spreading is caused by diffusion induced byelectron-neutral scattering.

6


12/26

5510-12

]''-"(!) 50t: IE 10-13. . . .. . . . .-:

4510-14

40-10 -5 0.0 5 10

Horizontal Distance (km)

Figure 2. Beam cross-section of fractional energy deposition as a function ofbeam radius and altitude. The Earth's magnetic field is not included. Thesimulations are shown on a greyscale and the solution of the envelopeequations with the red curve.

The question is now to what extent the magnetic field is able to contain the radialbeam spreading, and thereby focus and enhance the perturbation of the atmosphere. AMonte Carlo simulation was run with same beam-atmospheric parameters as those usedfor Figure 2 except that the earth's magnetic field was included. The result is shown inFigure 3. The beam is now well contained, and the beam radius is within 400 m, which iseven less than the 607 m gyro radius of a 5 MeV-beam injected perpendicular to themagnetic field (notice the change of horizontal scale). The analytical results are also inoverall agreement, with the exception that the envelope oscillates with altitude.

Fractional Energy Deposition (m"-3)E = 5.0 MeV; B = 0.4 G

60

7


13/26

55 -1010~] 50. -. . . . . 10-11-

high-energy electron beam-induced ionospheric modification experiments

Documents