AD-A2 5 8 009

, 1UL-Il-'-92- 4

September 1992

Preliminary Investigation into the Development ofHardening Devices Using Vacuum MicroelectronicsTechnology

by Chance M. Glenn

DTICS ELECTENOV19 199211

U.S. Army Research LaboratoryHarry Diamond Laboratories

Adelphi, MD 20783-1197

Approved for public release; distribution unlimited.

The findings In this report ae not to be construed as an offiial Dlparentof the Army position unless so designated by other authorized documents.

Citation of manufacturer's or trade names does not constitute an officialendorsement or approval of the use thereof.

D*toy this report when it is no longer needed. Do not return It to theoriginator.

PAGES'ARE

MISSINGIN

ORIGINALDOCUMENT

REPORT DOCUMENTATION PAGE T Foz A~q,'~oIBB4 8PUNIK "e00Ill b~Jd4 k(t WO W I Io of ZofoNWda WAesmmd o e I hom, per 'ineoewe, rdk9r le W¶N for twlfeAN WWilcU0AS meewir eilrg dma ,ourmgmhlr iW "aid rn'Wf vie dm ta eded ai 0WT'V'eZmi rer:edi O dm01m vi0W~v Said coiWftelb NOgW" Vi b~fe4 ati WN ciw WWi Of~ Nof0040@0W of ir~d,4.hdJlgIe weeti i rhotef.10oWai4t =ed.Sem Rm.. Okeass lOw IrdomIme Oowommt -i e~t.41 Jeff~ooOtiVe "S.4 Suito 1206. All cVoA 202=43M. aid to Vie Oftie of aid ReuPawm Fred (0704441 W= IV,000

1. AGENCY USE ONLY (aAM WNW) 2. REPORT DATE &. REPORT TYK AND "UTs COI4RED

ISeptember 1992 Progress, from March to September 1992

7.Z P AFOMND O&NIZTIO$4 4AEI N aAS(S. PFUNDRING IRMSAS~TO

REPORT NJMSIR

Harry Diamond Laboratories2800 Powder Mill Road HDL-PR-92-4Adelphi, MD 20783.1197

9. SPONSORINGIMONITORINO AG&NCY KAME(S) AND ADORESS(ES) Ilk SPO.ESORW01OMOITORNIOAGENCY REPORT HUM0SM

U.S. Army Research Laboratory2800 Powder Mill RoadAdelphi, MD 2078.1

11. SUPPLEMENTARY NOTES

AMS code: 612120.14011HDL PR: 2FE25

12a. WSThUUTIONIAVAILASLITY STATEMENT IM. 0SThISUION coca

Approved for public release; distribution unlimited.

I.AGSITRACT (liaMan" 200WwdmI)

Recently there has been a resurgence of interest in the vacuum valve. Manufacturing capabilities in the area ofsemiconductor microelectronic devices have paved the way for the emergence of vacuum microelectronic (VpAE)technology. The possibility of these new devices becoming a robust, high-speed alternative to solid-stateelectronics has caught the interest of the system hardening community. The most fundamental question asked iswhether or not these field emission devices can be incorporated into electronic systems as limiters-t4he mostbasic hardening device for in-band front-door hardening. This investigation shows that not only is it possible forthese devices to be developed into an alternative to conventional semiconductor limiters, but there may also bemarked operational advantages.

14. SU O JE C T TER M S 1 .N M M O A I

Microelectronics, semiconductors, vacuum, field emission, hardening, work 18function, Fowler-Nordheim equations, switching, limiters IL. PA"E CODE

17. SECURITY CLASSIFICAnOW IA. SECURITY CLALSW1CATION IT. SIECURITY CLASSIFICA11011 XL LIMTATION OP AMTRACTOF REPORT OF TWOS PAGE WP AWREACT

Unclassified Uncl assified Unclassified U LNON 7040-01-2W50M SWWW I"m2W RT~v 24UO

P"*0%"bW t0M1 S3W1tUIIIMI

Contents

Page

1. Introduction ............................................................................................................................ 5

2. Literature Search Results ............................................................................................... 6

2.1 Technology Overview ................................................................................................ 62.2 Fabrication ........................................................................................................................ 8

3. Experim entation ............................................................................................................ 11

4. H ardening D evice Configurations .............................................................................. 11

4.1 Conditional Short to Ground ................................................................................... 114.2 Lim iting Attenuator .................................................................................................. 12

5. Future W ork ........................................................................................................................ 13

Literature Cited ........................................................................................................................ 13

Bibliography ............................................................................................................................. 14

D istribution .............................................................................................................................. 17

Figures

1. Spike leakage of a typical PIN diode lim iter .................................................................. 62. Schem atic energy diagram for field em ission ............................................................... 73. Plane-to-plane and point-to-plane field emitter geometries ........................................ 84. Schem atic of a typical field em ission gated diode ...................................................... 95. Progression diagram of a UV lithographic fabrication technique ........................... 106. Progression diagram of a eutectic wet-etching fabrication technique ..................... 107. Possible configuration for a dual VILE element limiter ............................................... 128. I-V characteristics of a typical Si FEA ........................................................................... 129. Possible configuration for a limiting attenuator circuit using a VtLE element ........ 13

Accesion For

NTIS CRA&IDTIC TABUnannounced L]

-M C C_ -• Justification .............................

B y ................................................

Distribution/

Availability Codes

Avail and I orDist Special

3

1. IntroductionThe importance of this investigation hinges on the question of whetheror not VR.E technology can be exploited to design a better limitingdevice. Most limiter designs are nothing more than a conditional shortcicut to ground. As they stand, conventional limiters ate semicon-ductor p-n junction devices. There is much concern over the speed androbustness of conventional limiters in ever-increasing hostile environ-ments.

The speed of a semiconductor device such as a metal semiconductorfield effect transistor (MESFET) is ultimately limited by the time ittakes for an electron to travel from the source to the drain. Impurityand phonon collisions within the lattice of the solid lead to electronvelocity saturation at near the speed of sound [1]. Vacuum valvesoperate by electrons passing from cathode to anode uninterrupted bymolecular collisions. This is called ballistic transport. Typical voltages(100 V) and dimensions (1 Izm) yield transit times less than 1 ps [2].Complete switching times are impaired by parasitic capacitances,which are characteristic of small devices, as well as packaging capaci-tance. Still, VgE provides the technology for extremely fast switchingand, ultimately, extremely fast limiting.

Most semiconductor devices rely on low-voltage transport in a low-density electron gas. Ionizing radiation can cause an excitation ofcarriers, changing the density of the electron gas, thus leading tosignificant shifts in bias voltage. The result may be transient upset, orpermanent damage, which is evident from previous experimentation[3]. Vacuum valves have greater immunity to such perturbations sincethe source of electrons is either a metal or highly doped siliconcathode. Vacuum valves also work at much higher voltages thansemiconductors, again making them far less sensitive to large voltagepulses.

In any protection device that operates by short-circuiting delicatecomponents, this very low impedance alternative path must be initi-ated before the transient pulse causes damage. Spike leakage is energythat "leaks" through before the device is completely turned on (see fig.1). VtiE technology provides the hope of reducing this energy whilemaintaining a sufficiently robust device.

5

Figure 1. Spike leakageof a typical PIN diodelimiter.

Espike =f P( t) dt

tot

2. Literature Search Results

The first step in this investigation was a literature search. The goal wasto gain a basic knowledge of the technology as well as become awareof the latest developments in the field. This was necessary in order tomake rea5onable judgments as to whether this technology could lenditself to the development of hardening devices comparable to existingsemiconductor models.

2.1 Technology Overview

The emission of electrons from the cathode, modulation of the electronbeam by a grid, and collection at the anode is the basis of operation ofmost vacuum electronic devices. The energy required to strip anelectron free from its parent nucleus is the work function, which isobviously different for different materials. Traditional vacuum tubesexploit the process of thermionic emission, where electrons are "boiledoff" from the cathode by a heater. The electron stream is then modu-lated by a grid and collected at the anode. Microelectronic fabricationtechnology has provided the ability to create devices with this samefundamental operation, yet on a microscopic scale. With VAE devices,electron emission is induced by the high electric fields produced dueto specific geometry. This is field emission.

6

Field emission provided one of the first successes of quantum me-chanics when Fowler and Nordheim were able to explain it in termsof tunnelling through a triangular energy barrier [11 (see fig. 2). TheFowler-Nordheim equation, derived from this work in a basic form,describes the current density I in terms of the work function 0 and theelectric field E. It is, approximately,

1= aE 2-l0)exp , (1)

where a and P are constants. The experimental study of field emissionthen became very active in the 1930's with attempts to confirm theFowler-Nordheim theory using experimental techniques. Since then,work has been done to further understand and refine this process inorder to fabricate devices operating under field emission principles.

In field emission, it is the high electric field at the cathode surfacewhich allows quantum-mechanical tunnelling through the approxi-mately triangular energy barrier. Two particular field emitter geom-etries emerged: point-to-plane and plane-to-plane (see fig. 3). Thehigh concentration of the electric field at very sharp tips makes point-to-plane geometry more ideal. Present technology can manufacturefield emission tips whose radii of curvature are in angstroms. Plane-to-plane geometry requires well-defined, ultra-flat electrode surfaces.Electrical breakdown between flat electrical surfaces is not yet wellenough understood to make predictions and exploit these predictionsin the fabrication of a stable device. What is known is that microscopic,whisker-like protrusions on the plane are factors in initiating electricalbreakdown.

Figure 2. Schematic Image potentialenergy diagram forfield emission.

Work4functionFer •lev High electnc field

* *000(E > 109 //m)

O0000

ofenergy band

7

Figure 3. (a) Plane-to- (a) Anod (b) nodeplane and (b) point- *,_,__to-plane field emitfter 1geometries.

tpa fl Caetodt c:i:orr

The merger of point-to-plane and plane-to-plane geometries led to theconcept of field emitter arrays (FEA's). Several thousand field emittertips are spaced closely together to form an FEA. One advantage of theFEA structure is that large arrays can be built, thus allowing highcurrents to flow. The maximum current density ..,, (A/cm2 ) of theFEA is on the order of 4 x 104 and is independent of the applied voltage[4]. Using this maximum current density and a typical value of 100,A / tip, we can see that a typical FEA has about 4 x 108 tips/cm 2.Thesevalues are being pushed to higher and higher limits through continu-ous experimentation with new materials and the application of betterfabrication techniques.

2.2 Fabrication

If we consider the distance (d) as the length between the cathode tipand the anode surface (see fig. 4), we must have some criteria todetermine whether an electron travelling along that path is indeedtravelling within a vacuum. When (d) is much smaller than theelectron mean flee path (Ae) for collision with residual gas, we considerthe medium a vacuum [5]. We define X, as

T [cm] (2)273 p1',(V)

where P,(V) is the probability of collision, T is the absolute tempera-ture in kelvins, and p is the pressure in torr. The practical limits onfabrication technology allow for d - 0.5 gm. Now, given that theprobability of collision in a particular distance x is given by

p(x) = 1 - e-x/ = x/4 for x/Ae << 1 , (3)

8

Figure 4. Schematic ofa typical field emissiongated diode. A node

(Vacuum) ¶' ;' ', 2 Otats

+100V di I' S102 ,, ,film

F Metal or silicon base.__cathodei

thend s 10-2,, (4)

if we assume that the device can tolerate one in every one hundredelectron collisions, or p(x) = 1/100. For this calculation we let thetemperature T = 300 K and the pressure p a 760 tort.

Fabrication techniques developed from integrated circuit technologyhave been directly adapted to the fabrication of FEA's. Some of thesetechniques are as follows: anisotropic etching of metals and semicon-ductors, etching of a unidirectional solidified eutectic, and the use ofdeposition techniques, in conjunction with ultraviolet (LUV) or elec-tron beam lithography. Figure 5 is a progression diagram illustratinga UV lithography fabrication process developed by the Naval Re-search Laboratory, uirag silicon (Si) and silicon-dioxide (SiO) [1].Figure 6 shows a chemically selective wet-etching technique devel-oped by Science Applications International Corporation (SAIC), us-ing a semiconductor-metal eutectic composite material Si-TaSi2. Inboth techniques there is an attempt to control the tip height whileminimizing the tip radius.

Field emission is very sensitive to surface work function, electric field,and the field-enhancing tip shape. Although this is indeed true,cathodes have still shown very long lifetimes [1,6]. Surface cleanliness,atomic contamination, and lack of tip uniformity have been shown toplay a large role in FEA performance over time.

9

Figure S. Progressiondiagram ofa UV S1 SiOlithographicfabrication technique.

Mask- Resist

HF etch

Anisotropic etching of Si

Metal

Oxide depositionand metallization Si02

UIV

Uthographic definition Resistof anode and grid

Metal

Si0 2Grid Cathode

Completed device

(a) (b) (c) (d)

Figure 6. Progression diagram of a eutectic wet-etching fabrication technique.

10

3. Experimentation

At the present time, SAIC has development underway for severalprototype vacuum diode limiters for the Army Research Laboratory(ARL). Their designs must be compatible for microwave circuitry.After the devices are designed and fabricated, they will undergotesting here at ARL, Adelphi. Later reports will present the results ofUtis experimentation.

Contacts through the Naval Research Laboratory (NRL) have pro.duced the possibility of another manufacturer. The MicroelectronicsCenter of North Carolina (MCNC) at Research Triangle Park has beenworking with NRL for several years in their research and fabricationtechniques [7,8]. It may be possible to tap into the resources they havedeveloped in order to obtain workable devices for microwave re-search.

4. Hardening Device Configurations

The primary hardening device under consideration is the limiter.Conventional limiters operate by providing a conditional short toground. The conditions are determined by the switching device (i.e.,PIN diode). As stated earlier, we are convinced that VgE devices havethe potential to present advantages over PIN diodes in turn-on speedand power handling capability. If this is the case, then limiter circuitscan be designed that do not allow as much spike leakage as PIN diodeswith thin I-regions, and can handle more power than PIN diodes withthick I-regions. More power handling capability compared to PINdiodes is desirable since the Vg.E device can then handle electromag-netic pulse (EMP) signals. Limiter designs come in two classes at thepresent time: the first is the conditional short to ground and the secondis the limiting attenuator.

4.1 Conditional Short to Ground

Figure 7 shows a typical two-element limiter. Here Vp.E devices aresubstituted for PIN diodes. One of the immediate disadvantages ofthis device can be visualized from the I-V characteristics of a typical SiFEA, shown in figure 8. The turn-on voltage of the device is about 120to 130 V compared to 0.7 V for Si PIN diodes. When the goal is toprotect the front door of communications systems, the power neededto turn on the limiter may be well above the system's threshold.However, research is ongoing to try to bring down the turn-onvoltage.

11

Figure 7. Possibleconfiguration for a I, - " -- - _dual VpE element 0 0limiter.

-. r

Figure 8 I-Vcharacteristics of atypical Si FEA.

7

S3

2

1

0 20 40 60 80 100 120 140 160 180 20010 30 50 70 90 110 130 150 170 190

VC (volts)

4.2 Limiting AttenuatorAnother limiter configuration is a design that provides limiting byabsorption. The goal is to reduce the reflected power. This may beadvantageous for military systems. Figure 9 is an example of apossible limiting attenuator. The nonlinear property of the VýAE deviceswitches in a x-network attenuator when the input power exceeds athreshold. Another advantage of a circuit similar to this is that it maybe possible to design in the threshold level. Work is continuing at ARLto develop better circuit designs for this type of liatitet [9]. It may bepossible to incorporate a VpE device into the design.

12

Figure 9. Possible R,configuration for alimiting attenuator0circuit Using a VIAEelement.

R2 R3

VI&Eelement

0 , 0

5. Future Work

Several things are yet to be done to complete this work. First of all,efforts must continue to gain further knowledge of the VWE technol-ogy, as well as staying abreast of the latest developments in the field.This will be a result of continued literature search and study as well asobservation and experimentation. The fact that this is a relatively newfield means that progress will be made quickly from level to level. Itis our hope to successfully apply this technology to hardening devices,which calls for a deeper understanding of the entire design/fabrica-tion process. Secondly, we must be able to develop a microwavelimiter design into a tangible realistic circuit. Although the basictechnology alludes to a feasible device, we must find out whetherpackaging restrictions will deny us the advantages we seek. Thirdly,complete microwave characterizations of manufactured devices mustbe made if indeed we are able to overcome some of the design barriers.Not only will we attempt to verify other experimentation being doneat other facilities, but we will also attempt to generate new character-izations of VgE devices in the microwave arena. Finally, we willattempt to bring ARL into the network of research and informationregarding this technology.

Literature Cited

[1] N. A. Cade and R. A. Lee, Vacuum Microelectronics, GEC Journal ofResearch, 7, No. 3 (1990), pp.129-138.

[2) H. F. Gray, The Physics of Vacuum Microelectronic Devices, IEEE Trans-actions on Electron Devices, 36, No. 11, pt.1 ( November 1989).

13

[3] A. L. Ward, R. J. Tan, and R. Kaul, Calculated and Measured Silicon PINLimiter Short-Pulse Damage Thresholds, IEEE-MTr-S Symposium Di-gest, 2 (June 1989), 761.

[41 T. Utsumi, Vacuum Microelectronics: What's New and Exciting, IEEETransactions on Electron Devices, 38, No. 10 (October 1991).

[5] I. Brodie, Physical Considerations in Vacuum Microelectronics Devices,IEEE Transactions on Electron Devices, 36, No. 11 (November 1989).

[6] G. W. Jones, C. T. Sune, and H. F. Gray, Fabrication of Silicon Point,Wedge and Trench FEAs, Technical Digest of IVMC 91, Nagahama(1991).

17] H. F. GrayJ. L. Shaw, G. W.Jones, and C. T. Sune, High Current DensitySilicon FEAs, Technical Digest of IVMC 91, Nagahama (1991).

[8] G. W. Jones, C. T. Sune, S. K. Jones, and H. F. Gray, MicrostructuralGated Field Emission Sources for Electron Beam Application, Naval Re-search Laboratory, Washington D.C. (1991).

[9] Chance M. Glenn, Sr., Roger Kaul, and David A. Sumner, Design of aLimiting Attenuator, Sixth National Conference on High Power Micro-wave Technology (August 1992).

Bibliography

Brodie, I., Pathways to Vacuum Microelectronics, IEEE Transactions on Elec-tron Devices, 36, No. 11, November 1989.

Campisi, G. J., and Gray, H. F., Microfabrication of Field Emission Devices forVacuum Integrated Circuits Using Orientation Dependent Etching, NavalResearch Laboratory, Washington, D.C., 1987.

Ganguly, A. K., Phillips, P. M., and Gray H. F., Linear Theory of a Field-Emitter-Array DistributedAmplifier, Naval Research Laboratory, Wash-ington, D.C., February 1990.

Gray, H. F., Campisi, G. J., and Greene, R. F., A Vacuum Field Effect TransistorUsing Silicon Field EmitterArrays, International Electron Devices Meet-ing 1986.

He, J., Cutler, P. H., and Miskovsky, N. M., Generalization of Fowler-NordheimField Emission Theory for Nonplanar Metal Emitters, Tri-Service/NASACathode Worlkhop, March 1992.

14

Kirkpatrick, D. A., Bergeron, G. L., Czarnaski, M. A., Hickman, J. J., Levinson,M., Nguyen, Q. V., and Ditchek, B. M., Vacuum Field Emission from aSi. TaSi2 Semiconductor-Metal Eutectic Composite, Applied Physics Let-ters, 59, No. 17, October 1991.

Kirkpatrick, D. A., Mankofsky, A., and Tsang, K. T.,Analysis of Field Emissionfrom Three-dimensional Structures, Applied Physics Letters, 60, No. 17,April 1992.

Kirkpatrick, D. A., and Schoen, P. E., Measurements of Vacuum Field EmissionFrom Bio-Molecular and Semiconductor-Metal Eutectic Composite Micro-structure, IEEE Transactions on Plasma Science, 19, No. 5, October1991.

Lally, P. M., Goren, Y., and Nettesheim, E. A, j n X-Band TunedAmplifier Witha Field-Emission Cathode, IEEE Transactions on Electron Devices, 36,No. 11, November 1989.

Shaw, J. L., and Gray, H. F., A Sub-Micron Resolution UHV Probe Station forCharacterization of Field Emitter Arrays, Technical Digest of IVMC 91,Nagahama (1991).

15

Distribution

Administrator DirectorDefense Technical Information Center US Army Ballistic Research LaboratoryAttn: DTIC-DDA (2 copies) Attn: SLCBR-VL-A, E. M. VogelCameron Station, Building 5 Aberdeen Proving Ground, MD 21005-5066Alexandria, VA 22304-6 145 US Army Combined Arms Center

OSD Attn: CACDA; MID; DE DIV, J. DeBrouxDDDRER & AT FT Leavenworth, KS 66027Attn: J. Whittier US Army Electronics Technology & DevicesThe Pentagon, 3D1089 LaboratoryWashington, DC 20301-3080 LhrtrAttn: SI. CET-IB-H, E. BaidyDefense Advanced Research Project Agency Attn: SLCET-M, H. HeislmairAttn: ESTO, E. Cohen Atn: SLCET-M, V. GelnovatchAttn: ESTO, E. Sobolewski Attn: SLCET-MH, T. Burke3701 N. Fairfax Dr FT Monmouth, NJ 07703Arlington, VA 22203-1714 US Army Foreign Science Technology Center

HO Attn: AIFRTA, T. CaldwellDefense Nuclear Agency 220 7th Street NEAttn: RAEE, G. Baker Charlottesville, VA 22901-53966801 Telegraph Rd CommanderAlexandria, VA 22310-3398Cmade US Army Laboratory Command, VulnerabilityCommander Assessment LabBelvoir RDE Center Attn: SLCVA-TAC, L. GomezAttn: STRBE-JMC, J. Owen Attn: SLCVA-TAC, D. AlvarezAttn: STRBE-JMC, C. J. Wanner White Sands Missile Range, NM 88002FT Belvoir, VA 22060-5605 US Army Materiel Command

Missile Research, Development, and Attn: AMC-DE-AR-E, J. KreckEngineering Center 5001 Eisenhower Avenue

Attn: AMSMI-RD-DE-UB, R. Conrad Alexandria, VA 22333-0001Attn: AMSMI-RD-DE-UB, B. Jennings Deputy CommanderRedstone Arsenal, Al 35898-5000 US Army Strategic Defense Command

US Army Armament Resc~arch, Development, Attn: CSSD-H-LS, N. Belland Engineering Center PO Box 1500

Attn: SMCAR-AE, G. Taylor Huntsville, AL 35807-3801Attn: SMCAR-AEB, P. L. Marinkas DirectorPicatinny Arsenal, NJ 07806-5000 Vulnerability Lethality Assessment

US Army Aviation Systems Command Management OfficeAttn: AMSLV-ES, J. Snyder Attn: AMSTE-TC-C, D. Bassett4300 Goodfellow Blvd Attn: R. ReitzSt. Louis, MO 63120-1798 Aberdeen Proving Ground, MD 21005-5001

17

Distribution (cont'd)

Walter Reed Army Institute of Research USAISCAttn: SGRD-UWI-D, H. Bassen Arn: AMSLC-IM-VA, Admin Ser. Br.Washington, DC 2037-5100 Attn: AMSLC-IM-VP, Tech. Pub. Br.

Commander Harry Diamond LaboratoriesNaval Air Systems Command Attn: Laboratory DirectorsAttn: AIR 5161G CG-1, Room 940, Atn: SLCHD-SD-TL, Library

D. Fellin Attn: SLCHD-SD-TL, Library (Woodbridge)Amn: C. Caposel Atn: SLCHD-CS, Chief ScientistWashington, DC 20361 Atn: SLCHD-DD, J. Scully

Naval Research Laboratory Ann: SLCHD-NW, Deputy Director

Attn: Code 4000, W. Wilis Ann: SLCHD-NW-EH, Chief

Attn: Code 4650, T. Wieting Atn: SLCHD-NW-EP, Chief

Attn: Code 6800, G. Borsuk Ann: SLCHD-NW-ES, Chief

Attn: Code 6835, W. Anderson An: SLCHD-NW-HP, A.ief

Washington, DC 20375-5000 Attn: SLCHD-NW-HPM, A. KehsAttn: SLCHD-NW-HPM, F. J. Agee

Space and Naval Warfare Systems Command Attn: SLCHD-NW-HPM, H. BriskerAtn: PMW-145, CPT J. Fontana Attn: SLCHD-NW-P, J. GwaltneyWashington, DC 20363--5101 Attn: SLCHD-NW-RF, C. Fazi

Attn: SLCHD-NW-RF, ChiefAir Force Electronic Systems Division Atn: SLCHD-NW-RF, G. TranAttn: ESD/ICC, CPT K. E. Sears, JR Ann: SLCHD-NW-RF, J. TatumHanscom AFB, MA 01731 Attn: SLCHD-NW-RF, L. Jasper

Department of the Air Force Atn: SLCHD-NW-RF, M.. BerryAir Force Phillips Laboratory Attn: SLCHD-NW-RF, R. KaulAttn: PL,.PW, W. L. Baker Attn: SLCHD-NW-RF, R. TanAttn: PLwXPW, P. Vail Attn: SLCHD-NW-RF, S. KaplanKirtland AFB, NM 87117-6008 Atn: SLCHD-NW-RF, S. Saddow

Antn: SLCHD-NW-RF, T. TurnerAir Force Rome Air Development Center Ann: SLCHD-NW-RP, Chief

Attn:RADC 1. ooksAttn: SLCHD-NW-RP, ChiefAttn: RADC, J1. Rooks Attn: SLCHD-NW-RS, A. BromborskyAntn: RADC/RBCT, T. PestaAn:SCDN RChe

Griffiss AFB, NY 13441 Attn: SLCHD-NW-RS, L. Libelo

Commander Attn: SLCHD-NW-TN, ChiefWright Research Development CTR Attn: SLCHD-NW-TS, ChiefAttn: WRDC/ELM, T. Kemerley AtMn: SLCHD-SLCTO, N. BergBldg. 620, Area B Attn: SLCHD-ST-MW, W. WiebachWright Patterson AFB. OH 45433-6543 Attn: SLCHD-TA-ES, R. Wardell

US Army Research Laboratory Attn: SLCHD-NW-RF, C. Glenn (25 copies)

Attn: AMSLC-TP-PA, T. WhiteAttn: SLCSM-C3, J. Sczepanski

18