This document is made available through the declassification efforts and research of John Greenewald, Jr., creator of: The Black Vault The Black Vault is the largest online Freedom of Information Act (FOIA) document clearinghouse in the world. The research efforts here are responsible for the declassification of hundreds of thousands of pages released by the U.S. Government & Military. Discover the Truth at: http://www.theblackvault.com DEPARTMENT OF THE AIR FORCE IIEADQL'ARTERS 37ifll AIR BASE WING (An i C) KlRTLA~D AIR FORCE BASE, ~EW MEXICO Col Lance K. Kawane 377 ABW/CV 2000 Wyoming Blvd SE Kirtland AFB, NM 87117 Mr. John Greenewald Dear Mr. Greenewald In accordance with 5 U.S.C. § 552a, The Privacy Act, 5 U.S.C § 552, The Freedom of Information Act (FOIA), DoD Policy Memorandum 01-CORR-101, entitled "Withholding of Personally Identifying Information Under the Freedom ofI nformation Act," and DoD 5400.7-R, your 25 July 2014 request for the document titled Directed Energy Weapons Test Facility has been partially exempted from release under 5 U.S.C § 552(b)(3) and (b)(6). The requested document would include the release of information exempted from release under 5 U.S.C § 552(b)(3). This statute allows withholding information under the FOIA that is exempted from disclosure by other federal statutes. In this case, 10 U.S.C. § 130 authorizes the Secretary of Defense to withhold "technical data with military or space application in the possession of, or under the control of, the Department of Defense, if such data may not be exported lawfully outside the United States without an approval, authorization, or license under the Arms Export Control Act." The redacted portions contain technical information that is not releasable outside the Department of Defense. Furthermore, the redacted portions are controlled by export restrictions which cover sensitivities relating to current military data not authorized for release under the Arms Export Control Act. Private information of non-public figures, which is exempt under 5 U.S.C § 552(b)(6), is also redacted pursuant to the Department of Defense (DoD) Policy Memorandum, dated 9 November 2001 (requiring the withholding of names of DoD personnel currently assigned to a "particular component, unit, organization, or office with the [DoD]," unless they are senior officials or their names are known by the public). Accordingly, the requested document is released to you, with the exception of information which is exempt under the FOIA. If you wish to appeal this decision, you will need to write to the Secretary ofthe Air Force. Your letter must be received within 60 (sixty) calendar days from the date of this letter. Please include your reasons for reconsideration and attach a copy of this letter. Mail your appeal to the following address: Secretary of the Air Force Thru: 377 MSG/SCOK (FOIA Manager) 2051 Wyoming Blvd SE Kirtland AFB. NM 87117 Sincerely /~-~ LANCE K. KAWANE. Colonel. USAF Vice Commander UNCLASSIFIED AD NUMBER AD339671 CLASSIFICATION CHANGES TO unclassified FROM confidential AUTHORITY 31 Aug 1975, DoDD 5200.10 THIS PAGE IS UNCLASSIFIED UNCLASSIFIED AD NUMBER AD339671 CLASSIFICATION CHANGES TO confidential FROM secret AUTHORITY 31 Aug 1966, DoDD 5200.10 THIS PAGE IS UNCLASSIFIED ADS39671 DEFENSE DOCUMENTATION CENTER FOR SCIENTIFIC AND TECHNICAl INFORMATION CAMERON STATION. ALEXANDRIA. VIRGINIA NOTICE: When government or other drawings, speci fications or other data are used for any purpose other than in connection with a definitely related government procurement operation, the u. s. Government thereby incurs no responsibility, nor any obligation whatsoever; and the fact that the Govern ment may have fon:aulated, furnished, or in any way supplied the said drawings, specifications, or other data is not to be regarded by implication or other wise as in any manner licensing the holder or any other person or corporation, or conveying any rights or permission to manufacture, use or sell any patented invention that may in any way be related thereto. NOTICE: THIS DOCUMENI' CONTAINS INFORMATION AFFECTING THE NATIONAL DEFENSE OF THE UNITED STATES WITHIN THE MEAN- lNG OF THE ESPIONAGE LAWS TITLE 18, 1 U.S.C., SECTIONS 793 and 794. THE TRANSMISSION OR THE REVELATION OF ITS CONTENI'S IN ANY MANNER TO AN UNAUTHORIZED PERSON IS PROinBITED BY LAW. ASD-TDR-63-29 S 7 i Copy No. of 182 Copies ·t ) I , b . ... ·C~ e o7;;;~f;dnEnergy Weapons Test Facility :...L..! (-:.-~:: ,...-.., -- .! ASD Ttd•nical Docu•e•tary Report No. ASD·TDR-63·29 AUGUST 1963 • AFSC Projed No. 3805 DIRECTORATE OF ARMAMENT DEVELOPMENT Det 4, AERONAUTICAL SYSTEMS DIVISION All fOICI IYIUMS COMMAND • UNITID SfA Til All fOICI IGLIN AIR FORCE aASE, fLORIDA (Pr.pCHed .. nder ~ontroct No. AF 011(635).2795 by lon Phyai~oca~rot!!._"- BYrlington, Mou, o..thor A. S. ~ • -• GROUP-4 Downgraded at 3 year intervals .. · Declassified after 12 years. I .. j Qualified requesters may obtain copies from DDC. Orders will be ex pedited if placed through the librarian or other person designated to re quest documents from DDC, This document contains information affecting the national defense of the United States within the meaning of the Espionage Laws (Title 18, U. S. C. , 1ection1 793 and 794), Transmiuion or revelation in any manner to an unauthorized person is prohibited by law, When US Government drawings, specifications, or other data are used !or any purpose other than a definitely related government procurement operation, the government thereby incurs no reeponsibility nor any obli gation whatsoever; and the fact that the government may have fo:rmulated, !urnished, or in any way supplied the said drawings, specifications, or other data is not to be rega:rded by implication or othe:rwise, ae in any manner licensing the holder or any other person or corporation, or con veying any rights or permission to manufacture, use, or sell any patented invention that may in any way be related thereto. DDC release to OTS not authonzed. DETN::HMENT 4 Directorate of Armament Development AERONAUTICAL SYSTEMS DIVISION Air Force Systems Command UNITED STATES AIR FORCE Eglin Air Force Base, Florida REPLY TO ATI'N OF: ASQWR-67-3352 SUBJECT: Technical Information Concerning the Air Force Atomic Particle Beam Space Weapons Research Program TO: Recipients of ASD-TDR-63-29 1. This technical report is one of the documents published under the subject program (AFSC Project 38o5). This report is the result of theoretical and experimental investigation concerning the basic feasibility of particle beam accelerators operated as space weapons. This report is intended to provide you with the latest information con cerning particle beam accelerator weapons concepts and related topics. Subsequent reports in the subject program will also be sent to your office , if there is sufficient interest within your organization concerning advanced radiation space weapon systems concepts. 2. We would appreciate your bringing this report to the • attention of any persons in your organization who might have an interest in the subject of radiation space weapons. If your office has no immedi ate use for this report please it to your technical library or fon~ard return it to this Detachment. We would also appreciate your reviewing the distribution list of this report and informing this office of any suggested corrections or additions to the list in order to provide proper distribution of futu.re reports. In addition, lve invite any technical or editorial comments concerning this report', or requests for more in formation concerning the Atomic Particle Beam Space Weapons Research Pro gram from members of your organization. FOR THE DIRECTOR Assistant Chief, Weapons Laboratory \ NOTICE Thil Final Report was prepared by Ion Physics Corporation under Air Force Contract AF08(635)-Z795, "Directed Energy Weapons Test Facility". The work was administered under the direction of Weapons Laboratory (ASQWR) Detachment 4, ASD. The studies began on May 1, 196Z and ended on December 31, 196Z. Dr. A. S. Denholm was overall program manager with R. Britton acting as project engineer on that part of the studies concerned with the ultra high voltage facilities for the study of vacuum insulation, The •tudies required many talents. The major contributors were: K. Arnold A. :r. Gale R. Britton I. Kohlberg R. Cheever S. V. Nablo P. DeBeurs P. Wiederhold A. S. Denholm :1. Weisman This report concludes work under the contract and is the only report. It i• cla. .i fied SECRET because of the data it contains related to directed energy weapons technology. '· ii TABLE OF CONTENTS NOTICE CONTENTS ILLUSTRATIONS TABLES LIST OF SYMBOLS SUMMARY I INTRODUCTION 1 1. 1 GENERAL 1 z 1. Z PURPOSE OF THE PROGRAM Z FACILITY FOR THE STUDY OF DIRECTED ENERGY WEAPONS CONCEPTS 2.1 CHOSEN WEAPON CONCEPT AND ITS JUSTIFICATION 4 2. 2 FACILITIES AND MAJOR PROBLEM AREAS 9 z. 3 POWER SUPPLY AND ENERGY STORAGE 17 Z. 3. 1 General 17 z. 3, Z Facility Power Supply and Eneray Store II z. 3, 3 · Application of Faciliti. . to the Development of Ener1y Storaae 28 for Weapon• Z. 4 PARTICLE ACCELERATOR SYSTEMS 33 z. 4. 1 Injector Con•ideration• for the Facility 33 z. 4. 2 Beam Focu•inl Durin1 Acceleration 38 • • 1 : - • - • - • - • - • • - • iii TABLE OF CONTENTS (continued) • • - • .II.: • • - • - • .•. • .. - • • 3 FACILITY FOR HIGH VOLTAGE VACUUM INSULATION AND 106 POWER PRODUCTION STUDIES 3. 1 INTRODUCTION TO THE PROBLEM 106 3. 2 PROGRAM--GOALS AND APPROACH llf 3. 2.1 Goals 114 , 3. 2. 2 Philoeopby of Anroach 115 3. 3 DEVELOPMENT AND DESCRIPTION OF EFFORT 119 3. 3. 1 Sleteme for Devel2!iDI Hilh Voltale in Vacuum 119 3. 3. 2 Cboeen Method 130 3. 3. 3 Review of ExietiDI and Pro22•ect Facilitiee 132 3. 3. .. Allocation of Stucliee to ExietiDI aDd Propoeed Facilitiee lff 3. 4 GROWTH PLAN FOR FACILITIES TO SOLVE THE PROBLEMS 150 OF HIGH VOLTA CiE INSULATION IN SPACE REFERENCES lU APPENDIX JI--Cba:ra• Flow DoWD aa Accelerato:r Tube aad the JmpUcatl~ to PoteDtial Gradial ' iv ILLUSTRATIONS Fla. Page 1 Z Mev Directed Energy Facility 14 z Circuit Ou•line-Simple Potential Drop Accelerator 15 3 Neutral Particle Injection Acc~lerator 16 4 a) Lile-Stre. . Relationehip for Typical Capacitor Material Zl b) Enerty Deneity-Streu Relationship for Typical Capacitor 21 Material 5 Accelerator Coupled for Maximum Beam Stiffneu 24 6 a) Driven Terminal Concept for Maintenance of Beam Energy 26 During Dt.charge b) Voltage Relationehipe with Ruonant Pulaina of Terminal 26 (C2>> Cl) 7 Maintenance of Beam Enerty ueing an Aaxiliary Beam and 27 2 Terminal• 8 SuperconductinJlnductive EnerJy Stora1e Syetem 31 9 CouplinJ of Inductive Enerty Storaae Syetem to Accelerator 31 Tube - •10 Occluded Ciae So1uce: Axial Extraction (30) •36 • I • ( • • = • • • • • • • • - • • ' v ILLUSTRATIONS ( eonti.Dued) Ft1• Paae I I ' 2S 1 Million Volt Hl&h Vaeuum Tut Facility 108 26 1/2 Million Volt Pre. .u re to Vaeuum Feedthrou&h Buehinl 109 2.7 HIVE Faeility 110 28 Volt&Je-Oap Rel&tioneblpe for Volume and Surface 118 IDeul&tion in Vaeuum 29 Inverted HiJh Volt&Je Oenerator 120 30 Conventional Volt&Je Source, Internally Loeated Vacuum 128 Ineulation Reeearch Apparatue by Leo Jedynak 31 0. 4 Million Volt~ Hilh .Y_acuum-Hi&h Fl!.ld Facility 135 32 Cut-a-way View - MiV Facility 136 33 Z Million Volt Facility- with Propo. .d Modification• 137 34 4 Million Volt Vacuum Breakdown Facility 138 , 35 Type D BuehiDI 139 36 Volt&Je which can be Supported with CyliDdrical Oeometry 140 Uein1 SF6· R ie the number of coaxial Shell• 37 Z. 5 MY Vacuum IDeulation Facility 148 38 4 MY Vacuum Ineulation Facility 149 39 OroWth Plan--Rich Voltaa• ID•ulatiOil FaclUti. . • Stwlle• 151 ---. , TABLES 1 Outline Paramet•tera: Propoaed Hilh Pulae Current Acceleratou 11 2 Outline Parameter•: Preaent Hlth Pulae Current Accelerator• 13 3 A Review of Cia• Oiacbarte Proton lnjectore •34 I b • t • ) • I - • I • I vii ., i LIST OF MAJOR SYMBOLS Symbol Definition Dimenaiona -3 N ambient plaama denaity em p -3 N' total plasma density em p -3 N neutral density em n projection (beam) velocity em/sec length of beam travel em z e initial energy flux ergs/em /sec 0 z e measured energy flux ergs/em /sec m m positive ion mass g p b distance of closest approach for coulomb em interactions mass of incident particle g mass of target particle g m reduced mass: m.= ml mz/(ml + mz) g 0 DeBroglie Wavelength em -10 e electronic charge 4. 8 x 10 em -9 a. Bohr Radius 5. Z9 x 10 em • v general notation for velocity em/aec Dirac '• conatant 1. 05 X 10•ll g cmZ/aee atomic acreening diatance (a. • =a. z-1 1 3 ) em • p impact parameter em .J • I viii i ·'. Symbol Definition l)imenl ion• ' ratio of dbtance of clo•e•t approach b, e compared to the ICreeninJ lentth, CL , : b/CL • I = 2 8 v "Bohr velocity" ; v e /'h 2. 3 x 10 cm/1ec • • kinetic energy of beam particle er11 w colli•ion probability I. drift chamber lensth em c -3 em ambient pla1ma temperature h Debye length em . I 3 p ma•• den1ity gram em •pecific heat cal/gram K conductivity cal/•ec•K em T temperature •Kelvin X di8tance z D diffu1ivity em /•ec z crPr pre••ure ks/cm di1placement em c •peed of lisht cm/•ec c • speed of 1ound cm/•ec G •hear modul\18 k.s/cmz 2 E Youns modulu• kt/cm propqation velocity of lonsitudinal cm/•ec di•placement• propqation velocity of tran•verse cm/•ec di•placement• ex thermal expan•ion coefficient 1x Symbol Definition Dimenaiona cal/ mechanical heat equivalent 4. 17 joule a = -6 microsecond 1 x 10 sec average distance of the incoming particle from nucleus proton maaa kg· m electron mau kg e At delivery time aec z atomic number v f3 c B radiation converaion length or radiation em length .... linear absorption coefficient w electron energy, where(d dW x >. = (dW) Mev c dx 10ft rad wavelength •• permittivity of vacuum 8. 85 x 10-l~ F/m -6 permeability of vacuum 1. ~5 x 10 H/m • X SUMMARY The objective of the proJr&m de•cribed in thi• rt:port wa• to evolve de •iJn plan• and an evolutionary pro1ram for a blah-vacuum te•t facility •uitable for u•e in the Directed EnerJy Weapon• Development Proaram. Hiah vacuum i•, o! cour•e, the environment of •pace. Facilitie• for the development and te•tina of luer type weapon• are dif ferent from tho•e for particle beam weapon• (ion•, electrona), and thb, toJether with the .fact that a larJe laaer teatinl facility 1• beinl developed elaewhere. led to a concentration on facilitie• for the particle beam weapon. A major problem in the acceleration o! particle• to biJh eneraie•, par• ticularly ~a•inJ a •in&le potential drop device, i• the •upport of larJe potential• in wa• the vacuum environment. It hoped that a facility •uitable for the development and te•tinl o£ directed enerJy weapon• component• would be •uitable abo for the •tudy of technique• for the aupport of very biJh voltaa•• in •pace. Such a d~aal p~arpo•e facility wa• found to be conceivable, but not advi•able. Con•equently, the 1tudy proceeded alon1 path• which are interrelated b~at aeparate. Pro1reu alonJ one path determined the facilitie• for the development of directed enerJY ••eapona,and alona the other, !acilitie• for •olvinl the hiah volta1e in•~alation problem. A •tudy of facility requirement• wa• impouible without a clo•e examifta tion o! the •everal directed eneriY weapon• concepu. Thi• •bowed the need for information on taraet dan.aJe and beam/environment iftteraction•. Con•eq~aently, the1• areal were 1tudied theoretically at 1ome lenJth, which confirmed the belief that adequte experimental data on tar1et &Del environmental effect• were needed to determine the utility of particle bt·am weapoD• • • The development of a potential drop particle accelerator with adequate •to red eneriY and with reveraible polarity ia the faateet and m.oet economical ap proach to beam• of ione or e1ectrone of aufficient power denaity to produce the required taraet and environmental interaction data. Jt appean that a 'drift' tube ,. ZS meteu lona would beain to aupply uaeful information on environmental effecta. Thia drift tube would be extended aa experiment• pro1reu and the utility of the particle beam weapon ia confirmed. The realisation of hith power potential drop accelerator• require• reaearch and development in aeveral areaa which are diaeuaaed in the report. Theae areaa include power aupply, eneray etoraae. ion and electron injection, .and beam handlina. A 1rowth plan for faciUtiea to aolve beam production and acceleration problema ia preeentecl aa part of an overall arowth plan for particle beam weapon• (FiJ. Zl). A major part of a facility for the atudy of very biJh voltaae eflecta in the apac.e environment ia the vacuum chamber inaide which the atudiee are made. The voltaae• which are required ineide thia chamber dictate ita timenaioaa. The atandard method for obtaininl hilh voltaJe inaide a vacu11m ayatem ia to aenerate outaide and uae a feeclthrou1h buahina. It appeara pouible to extend thie tecb niqlle, which hae been developed to above 1 MV at IPC, to abo11t 5 MV. However, • a more fruitf11l approach ia to develop 1eneratora which can operate inaide the chamber 11ain1 vacoum for their external ineulatlon. Generatore of thie form are almoat directly applicable to the apace borne accelerator. The report concllldea with a frowtb plan for hi1h vac1111m facilitiee which would permit the atudy of potential• 11p to 8 MV &Del develop the biJh voltaae tecbnoloay required for the operation of acceleratora in tpace. . I 1. INTRODUCTION 1.1 GENERAL Thb atudy i8 related to Directed Energy Weapon• concepti and con aequently it is worthwhile diacuuina the preaent dgnificance of the nomen clature. The title "Directed Energy" preaumably waa coined when the poaai bility of uain& electromagnetic radiation directly for taraet deetruction waa being conaidered. The main adYanta.ae of a weapons acheme baaed on a beam of radiation ia in the fa at taraet interception which il poa aible becauae of the velocity of propagation, which il the ultimate. Obrioualy, laaer weapon• fall directly in thia category. Ion and electron beam approach•• to a weapon are not atrictly "Directed Energy" unleaa one aaaumea energy to include kinetic energy, in which caae such mundane meana of deatruction aa the gun and bullet could be termed a directed energy weapon 1yatem. It ia neceuary then to define the term "Directed Enersy" aa applied to a weapon ayatem, and it il aucaeated that the nomenclature be applied to a weapon ayatem where the velocity of propasation la a alsniflcant fraction of the apeed of Uaht. It il a peculiarity of the aeveral concepti that electrical forcea are required in the penultimate ata1e of the weapon. The varioua directed eneriY concepti will be diacua1ed in 1ome detail later, but it il worth introducblg 1ome of the ideaa at thla point. Poaalble approache1 to a directed energy weapon include laaer beam•. pla1ma projection and electron or ion beama. Ion beama may be neutraU•ed before projection to aive an uncharaed atomic or molecular beam. Laaer beam weapoM are attrac tive for focu1inc on a taraet but auffer from a very poor power efflc:iency and the fact that·damage effecta are auper£icial and are likely to cauae only aublima tion of material. Pla1ma projeetlOil concepti have ao far been ellmlaated due to the cllfftculty ln contalnilll a denae jllaema. ewer uae.ful projectloa diata.ncea. Ion or electron beam weapon• appear the moat feaaible; but with the preaent atate of the art are far removed from practicality. Proarama auch aa the preaent one are aimed at cloainJ thia gap. For obvioua reaaona, directed enerJy weapon• are more feaaible • for operation in a very low preaaure environment, and the altitude required for the operation of a directed energy concept ia of more than pa. .i na inter eat. In conaiderinJ the utility of any concept the moat lo1ical place to atart would be in determining if it could alanlficantly damaae a target. Conaequentl'Y, a conaiderable eection of thia report il devoted to target damage effectl. Of comparable importance, and related to minimum altitude for operation, i1 the interaction of the directed eneray with the environment. Thil hal allo been treated at aome lenJth, and the need for thi1 will become clear from the pro aram philoaophy outlined below. 1. Z PURPOSE OF THE PROGRAM The purpoae of thiA program wa1 to conceive teat facilitiea for the development of directed energy weapona componentl, and hopefully, for the ultimate teating of a "prototype" weapon. To be realiatic the ultimate facility for a complete weapon W011ld not be built without the prior building and operation of amaller facilitiea for the proving and development of each of the component part• of a weapon. It waa the aim of the program to produce a acheme for auch a planned growth towarda the ultimate facility, with indication• of the timinJ and funding which would be involved. Obviou1ly the facilitie1 could not be conceived without a cloae examina tion of the varioua approachea to a weapon and their weakne. .u . The extent of the weaknea ••• then determine whether or not a particular approach il worthy of te1t facility conaideratioDI. A further requirement of the facilitie 1 wa1 that they be auitable for examinin1 the electrical ioaulatiOD atreqth of hiJh vacuum at very hiah volt&Jea. Thia waa primarUy to aupport a cODc:ept of elec:troatatic ener11 atoraa• uaiq the apace environment as a dielectric which was proposed under another con tract (AF08(635) -1636), but such 1tudiea of vacuum in1ulation are aho desir able in support of high energy accelerators in space, even where the space • environment is not the energy storage medium. It was posaible to conceive facilities which could serve both the com ponents testing and the vacuum insulation tests but this was not considered a sound approach. The design of the dual purpose facility !or reasonable dimen sions requires the making of as sumptiona on vacuum voltage insulation at the higher potentials which may not be valid. If these asaumptiona proved to be incorrect, the facility could be o£ minor utility and, since the concept o£ a directed energy weapon is not necesaarily tied to the support of megavolt po tentials across vacuum gaps, the ultra high voltage vacuum insulation facility should be separate from the facility for examining other directed energy prob lem areas. In support o£ this two facility approach it should be noted that all present methods of accelerating charged particles to energies above about 1 MV use graded accelerator tubes. ' 2. FACILITY FOR THE STUDY OF DIRECTED ENERGY WEAPONS CONCEPTS 2. 1 CHOSEN WEAPON CONCEPT AND ITS JUSTIFICATION The preliminary deai1n of a facility uaeful for the atudy of the di rected ener&y ayatem muat be baaed upon aome initial, neceuarlly reatrictive, auumptiona concernina the nature of the projector ltaelf. In order that thb atudy could accompllah uaeful concluaiona concernin1 not only the type of fa cillty required, but evolve, u well, the phUoaopby of the experimental program ·., which could be conducted with it, an early decialon waa made on the weapon aya- tem type that ahould be conaidered in thla proJram. The conaiderationa upon which thia deciaion waa baaed and the flexibility it permit• wlll therefore be outlined. . . 1 Continuin1 atudiea at Ion Phyaica Corporation under AF08(635) -2166 2 3 4 and earlier effort• ehewhere ' ' have provided conaiderable inliJht into the fundamental limitation• of the varioua approachea to energy projection in the extraterreatrial environment. Of thoae technique• conaidered, the charJed particle or electroatatic accelerator, the plaama projector and recently devel oped aourcea of coherent radiation (laaera) have received the 1reateat attention due to their relatively promiainJ characterlatlca for the application. The tactical utllity of each of theae may be evaluated on the baaia of five primary criteria , · namely: diverJence of the projected beam, eneriY efficiency of the weapon aya tem, enerp deneity at the plane of the pro;ector, velocity of the directed beam and flnally, mode of interaction at the tarJet itaelf. Since the forementioned re...,rta have treated theee criteria, for all but the laaer, to varylnJ dear•••· only a brief qualitative review of the llJniflcance of each to total ayatem conaid eratlone will be diacuued ln turn below. The neutral particle ayatem will be cODaidered aa ayn•ymwe with a charaed particle projector &I it 11 eubject (at lealt wlthln the projector ltaell), to the aame llmltatlOAI impoeed by electroetatlc acceleratlOD techDlq1ae1. The diveraence ansle ex: of the projected beam ln the charaed particle systems arises from repulaive coulomb forcea within the beam itself as well aa from the thermal motion or temperature of the accelerated particlea. In addition, the effect of the ambient environment muat alao be considered in evalu atina "diversence." Thus, the effecta of acatteri.ng, char1e-exchange and ex citation durin& the drift phaae muat be determined in addition to the interaction of the beam with the ambient electric and magnetic fielda. The former phe- S nomena (coulomb, thermal) are much better underatood at the present time than are the latter, particularly for the relativistic regime of drift velocitie1. In view of this unbalance, conaiderable ef!ort haa been devoted to analysia of theae latter effects with the charaed particle 1ystem, and the results are pre a ent ed in Section Z. 6 under Drift Tube Considerat.lOh;... A useful review of our knowledge of the divergence problem for the plasma ayatem ia pruented in Ref. 6 and, along with the graphical reaulta of Ref. Z, provide a uaeful intr\ duction to the complexity of the divergence - range conaiderationa 10 funda mental to weapon• evaluation. The eneray efficiency fl of the aystem ia broadly defined aa the ratio of the eneray contained within the (uaeful cone and pulae lenath of the) projected beam to the total eneray expended in generation of the "ahot" pulae aa well aa in ita focuaina. Thia flaure of merit ia of prime importance in ap- 7 proacbin& the power aupply problema implicit in the directed eneriY concept due to the extreme energy atoraae requirement• of even a hi&h efficiency pro jector. All typea of eneray lou mechanism• muat be conaidered, incluclinc thoae auociated with the atoraae ayatem itself. At the preaent time we can only make uaeful efficiency estimate• for the projector per ae while excludinc the primary aupply efficiency itaelf: i. e. , for the eneray atoraae ayatem, elec trical power aource and pulae enercy converter. Tbia parameter is unqueation ably the beat defined, at preaent, of the five conaidered here in view of a rea aonably exact knowledce of the atate of the art of the varioue "eourcee" them- 1 7 ..1 vea coupled with t~oae data relatlnc to aupply ayatema outUned elaewhere. ' The eneray denllty ( ol the beam at the plane of the projector u ..u provide• a mea1ure of the capabllity of the 1y1tem provided that kill mechanbm• and diveraence are adequately under1tood !or the device. II one .. could neslect the "po•t-acceleration" effect• on beam diveraence and the re lultlna decrement in energy den1ity at the taraet, it would be u~eful to char acterise the beam itael! in term• of iu directivity. For a beam of energy t denlity and divergence hal!-an&le a; the beam directivity il expreued a• z z (/11' ex . (Ampere•/cm /steradian where cx:h the beam hal!-an&l• at the crou- over or minimum beam radiu1.) If one can further define thil beam radiua 11r" 2 2 near the projector plane, then the directivity becomu (J/11'r ) (l/1!'<X ) or 2 2 J /11' • where J il the beam ener1y and • le the beam emittance; 1. e. , rex:. Thi• parameter can then include both ( and ex: in a re1trictive 1en1e. The velocity v of the directed be·am b & •. critical parameter u it conatitute1 the rea1on "why" for theae directed energy atudie1. Tactical con aideration• of the effect• of projection velocity are pre1ented in Ref. 1 and 2 and indicate that for the intended application (apace vehicle• 1n Keplerian earth orbit1) projector velocitie• of > .. 1 c are required for prel8nt tracking errore (0. 5 milliradian1). The lead an11e required varie• inver1ely u the projection velocity and 11, of coune, 1mo.H for relativhtic velocitlu for thele "near - earth" trajectoriu. In addition, the beam velocity for char1ed particle 1y1tem1 11 intimately connected to the divergence or expan1ion con1ideration• durin1 pro jection. In particular, for a given power den•ity within the beam, the char1e denlity and hence the radial electric field (•• derived !rom a 1olution of Poiuon'• equation) will vary lnvenely •• the velocity v. The •elf-foculinl effect• of relatlvhtlc beam• are of intere1t here, particularly for electron 1tream1, and the ltrenath ol thl1 force produced by the asimuthal field a• well a• the radiu• of curvature ol the particle in the ambient field, bo~ vary directly a• the par tic:le velocity. . The mode ol interactioo ol the beam at the tar1et cletermbaea the ~ . l kUl mechwam a1acl hence the tactical udllty ol the endre ayatem. Implicit 1D 6 conelderatione of the interaction mode are technique• of weapon counter- 2 6 meaeuree. Recent work conducted un.der thie ' (Directed EneriY) and re- 8 lated programe (ORION) hae provld.ed only •cant information relating par ticularly to the effect• of high energy, hiah den1ity plaama impact on variou• materlab, and hae yielded a poor proanoeil for the ueefulneu of the plaema 9 eyatem a1 a tactical weapon. There ie a eimllar paucity of data for the high denaity photon beam and the nature of it• interaction at a eolid eurface, althouah 10 aeveral aovernment eupported procrame are now directed to t.hi• area. Rather more data are available relatlnJ to the interaction of hiah current denaity cbaraed particle beam• from hiah lntenaity accelerator etudlee, particularly with ener- 11 getic beam a. Experiment• deeigned epecifically for the inve•tication of the metallursy of damaae mechaniem• are etill wantin1 ln the ener1y re1lon of in terelt to thi1 program, and 1ome deftnltive data in thll area for "heavy" charged particle• will hopefully be available ln the near future. In eummary then, it il eeen that of the five critical parameter• con lidered here in the evaluation of the eeveral weapon concepti, the leaet data are available for elucidation of the "kill" mechanlem. However, with the in formation available, the hiah lntenllty cbarced particle beam would appear to be the moet effective, ba1ed larcely upon cou.ntermea1ure con1lderation1 for the varioue ey1tem1. The main aclvantaae of the blah intenelty photon 1ource or op tical ma1er obviouely llee ln lte hiJh dbe..:tivity; however thb advantage. ie much leu obvioue (ae compared to the particle concept) for hi1h power laeen a1 12 Sace ba• pointed out. The~e llmltatione are yet to be fully demonetrated on lara• aperture la1eu at u1eful power levele. Sy1tem1 are under etudy elee- 10 where for evaluation of the1e capabUltie1, and although partially communlca- tiooe oriented, wUl be readily ueeful tor the purpo1e1 of thle proaram. The facUlty etudy outlined ln tbl1 report therefore punuee cooeldera tionl relating primarily to a charced particle linear acceleration eyetem. A1- 1umm, that our pueent kDowledce ol th.e, .,.._ ....n. o.,l ldll eneray nqulremefttl le acle4auate. there woulcl appeu to M ao • barriel'l to the cleYelopmeot of a low ene-.ry (l Mev) prototype ay11tem which conatltutea the early baail of the facility. The critical area a of beam propagation and tars et interaction may then be atudied in detail with thb facUlty in order that data may be acquired at ener1y deneltiea of lntereat to the pro&ram. The Mark I (2 Mev) facility there fore repreaenta the firat prototype terreatrial directed eneriY eyatem which will be capable of providing the experimental data required for the evaluation and extrapolation of thoae parametera outlined above and, in particular, for the atudy of beam propaJation in a aimulated environment. The analytical con aiderationa outlined in Section• 2. 5 and Z. 6 of thie report provide a foundation for the experimental pro1ram propoaed for thh (Mk I) facility. A aummary of order of masnitude data relatin& to the criteria ueed in ayatem evaluation ia preaented below and ie intended to be indicative of the mid -1962 state of the art in thia area for instrument• that miaht realietically be considered for the directed eneriY application. Reference• are included for each ayuem. The "prognoeie" for target interaction effecta 1n each caae 11 a beet eetimate baaed upon conalderatlona of pouible kill mechanlama at the propagation velocitlea and power denaitlea achieved to date. Any comment• concerning vulnerability to countermeaaurea for each ayatem would be prema ture in view of our limited knowledge of theae interaction mechaniama. Projector Type Dlveraence Power Power Vel. Pulae Interaction and Referencea (~) Eff. ('1) Denalty (I) (v) Lenath Effect -Z 5 2 lon Beam < 10 r >50,. 10 w/cm .2c 10 -l. &ood- (2, 4, 7) excellent -3 7 z 6 Electron Beam < 10 r > 95,. 10 w/cm ""'C 10- • excellent (11, 13) -1, -•o z Optical Maaer -5 r 10 6 w/cm c \0-6. aood (13, 14) -Z 3 P1aama > 10 r <40,. Low .OOOSc 10- • poor l (2, 6, 15) I • l • The ener1y denaity of the beam at the beam crouover near the plane of the projector b a commonly uaed.~dterion. For our purpoau, the product of the ener1y denaity and beam velocity conetitute a more uaeful parameter, t 2 namely the eneriY flux or power denalty of the beam, atated in watta/cm . 1Z Since the concept of beam directivity aa uaed in optica, h pertinent to thla problem, it h uaeful to conaider the equivalent beam power directivity 6 ex- 2 preued aa t/w ex: • where CX:la the half an1le of the beam meaaured at the mini- z mum beam diameter or crouover point. The unite of 6 are then watta/ em I steradian and are lar1ely of uae in injector-accalerator evaluation but i1nore the critical problema of eubaequent power lou durin1 the drift phase. Since tar1et interaction effecte are Tate dependent, it 11 felt that the uee of power denelty ahould be adopted rather than the ueual ener&Yiunit area aa the pulle period illmpllcit in ~. The importance of pulee period T iel&rJely under etood on the baele of ener1y diaaipation conaiderationa. It haa be~n 1•nerally concluded that T muat be short(preferably leaa than 1 maec) in order that bulk ther z. mal conductivity conaiderationa no lonaer plan an important role (Section 2. S. 3). 2. 2 FAClLlTIES AND MAJOR PROBLEM AREAS The pre'rioue section has indicated that the moat promlein1 approach to a directed energy weapon lies in the acceleration of ion or electron beama. The actual ener1y and eneriY denaity at a tar1•t wblch would be required will depel)d upon the reaulta of experiment• on the interaction of beama with the apace environ ment and on tar1et dama1e wblch will be made in the early eta1•• of the pro posed facWtiea. ' The information on Table I ia for direct potential drop machine• which appear to be the belt approach to hi&h poaitive or neaative beam cur· rent• for pulae duration• up to 1 millleecond. Several forma of hi&h volta&• aenerator are available for the accelerator, and thoae which are worth con eidedn& are the Dynamatron, Coekcroft Walton, Van de Graaff and lnaula~ed Core Traneformer (ICT). The Dynamatron, Cockcroft Walton and ICT are limited to potential• of 3 - 4 Mev at the preaent time, but can produce current• 10 to 100 timea treater than the Van de Graaff, which hal a maximum output of about 1 milliampere. Both the Dynamatron, Cockcroft Walton and the ICT could ~· made for hiaher voltaaee, but laraely becauee of the lncreaain1 dimenalone the power t•neration become• le. . eUlcient. Becauee the requirement for a pubed accelerator ia etored eneriY rather than hich ccmtinuoua current, the Van de Graaff ia the beet form of senerator provided lt hae eufflcient current capacity to aupply any lealtaae auociated with the etored eneriY. Vande Gruff deeian• exiat for potential• up to IS Mev and euch accelerator• are at preeent under conetructlon. The current capacity of 1 ma ehould be quite adequate to eupply the eneriY etoraae eyetem (eee Table I). The acceleration of an electron beam 1• a more tractable problem than the acceleration of an ion beam, and unleee otherwi1e elated the more dil licult problem: 1. e., the ion beam accelerator, la belna pureued. Two ap .._tac•• proachee to the accelerator are ehOWD oo l't,e. 1, 2 &Ddl. The of .i the ch&ra• exchaqe machlae (na. l) lie lA the l• eource bellla at arowul po telltial rather than ln the termiaal becauee the operation of a hip curr•t ewrce . 1 10 1 l Table 1 Outline Parameters: Propo1ed High Pulse Current Accelerator• See Note Number Terminal potential (MV) z 10 3 3 Maximum pulse duration (uc} 10- 10- 1 Maximum terminal drop (kv) 100 100 -3 Beam current for 10 sec pulse (amp} 1 1 3 -5 Beam current for 10 sec pulse (amp} 100 100 4 3 3 Maximum charge flow (coulomb} 10- 10- 3 4 Maximum energy in pulse (joule• "'Z.I0 "'10 7 8 Energy density on 1 cmz (j/mz) Z.10 10 5 4 4 Terminal capacitance needed(~td) 10 10 6 4 5 Stored energy in system (joules} 2.10 5. 10 7 Time to charse from zero potential (aec) zo 100 8 Current drain due to leakage ba.a) z 10 9 3 Volume of capacltora (m ) o.zz 5.5 10 I 11 Not. . to Table 1 1. Thh ia ample duration from both coneideratione of tarset damas• and the holdinJ of a typical hilh velocity tar1et without continuoue tracldnJ durint firinJ. z. The operation of pulled accelerator~ at HVEC IUII••t• that a 100 kv drop with a Z Mev machine b quite acceptable. Information on the allowable drop will be obtained from the firet eta1e pro1ram. It ie likely that greater thaD 100 kv drop• will be allowable with the 10 Mev accelerator, and if eo, thil can be ueed to increaee the beam current x pube duration where d ..i red. The allowable drop becom. . much lee• where matnetic deflection follow• acceleration. Approach•• to reducing terminal drop are dhcueeed ln S.ction 2. 3. 2. 3. For poeitive ion operation. 4. ror electron operation. s. 2 Focueein1 to epot eiaee le11 than 1 em wlll be pouible. 6. The terminal capacitance of Van de Graaff machine• h ueually 100 - .a. 200 .... 7. The eneriY etored in Van de Graaff type machine h ueually: 2 Mev -zso joule•. 10 Mev - 6ZSO joulee. 8. Baeed on 1 ma charpnJ current belna available. Somewhat le11 may be available for the 2 Mev machine, but thie wW not be eiJftificant. .. 9. Ba. .d on me1 ohm x .,._( prodllct of 10 • 3 3 10. Baeed on ener11 deneitiee of 1. S j/in • Deneitl. . ae hiJh ae 2 j/in are available. For an e ..e ntlally DC appllcatloo hi1her deneltlee may be poe elble elnee capacitor llfe and volta1• etre. . h eloeely related to volta1e reveraale. However, lzdtlal outllae deelp uelq 200 kv commercially available wdtl ae module• 1lve lar1er volume• than Table 1 but the dlmeneiODI are etlll practical. Smaller moclule volta1•• wUl JlVe better eaariY deneltiee. • 12 Table z Outline Parameten: Preaent Hi1h Pulae Current Accelerator• Potential Peak Pulae Pulae Tube :;, Location Accelerator EneriY Current Duration Ener1y Len1th Renaaealaer Linac 77 Mev 800 ma 4.51'• zso j S.Z M P.I. (N.Y.) (electron) Yale Linac 60 Mev 700 ma 4.51'• 190 j 7.5 M (electron) Livermore Pulae Trana- 1. 7 Mev 150 a o.z5 .... 64 j (Aatron Project) former (electron injector) • Livermore Injector 100 kev Za 1ma zoo j (proton) Berkeley C. W. Injector 370 kev lZO ma 1 ma 45 J (proton) • Source proridea pulae duration to Z5 ma., but duration• abcwe 1 m•. not con•idered of intere•t here. Beam cllameter wa• 4" and beam cllver1ence about 10°. 14 • ... II~~ ·~· i .f-1 - ..--: "" ... ai j_.l j_ l_ 1 laea. Pulaecl lon or Electron Sclurce CWt 0 I 0 0 I Q. 0 I 0 0 - t "' F'IQ. 2 Circuit Outline - Simple Potential Drop Accelerator CAB Energy S10rage Capacitance c 8 Beam Stiffening Cclpacitance Rp ProtectiVe Resistors 16 in a terminal may lead to thermal problema and large pumping apeed require + menta dow the accelerator tube. Starting with a 1. 3 ampere aource of H 3 (Fig. 3) which ia believed attainable within the next two yean. a one ampere + beam output thould be pouible. In the neutraliaing canal the H diuoclatea 3 to 3 of H. each of which are injected aa neutral particle• with a 50~ efficiency ao that the neutral 'current' to the terminal ia about 2 ampere1. In the ter minal. converaion to poaitive ions il about 50~ efficient. 10 that a 1 ampere beam il then accelerated to ground through the terminal potential. However, the ayatem of Figa. 1 and 2 ia obvloualy a aimpler con cept and ia preferred at the preaent time. The fact that only a pulaed beam il required mean• that the source can be operated on a ga• pulae ba1il, po11ibly with pumping in the terminal, to alleviate the problem of pumping down the tube. Thermal problema in the terminal are alao reduced becau1e the aource baa to operate only in a pulled mode. The major problem area• 1n the accelerator concept are: 1) The ion aource. Thb il treated together with the problem of injection into the accelerator tube 1n Section 2. 4. 1. 2) Power aupply, energy atorage and beam coupling (tube prob lema). Thia ia treated in Section 2. 3 .. 3) Beam handling. Thla il treated 1n Section 2. 4. Z. 4) Drilt tube. Thil component il needed to investigate the interaction of the beam with the apace environment. Section Z. 6 ahowa that lianificant data on thl1 can be ob tained with a ground baled ayatem of reaaonable dimenaiona. Z.3 POWER SUPPLY AND ENERGY STORAGE 2. 3. 1 General In thia aection conaideration will be given to the power 1uppllea and eneray atorea, both for the high powered accelerator• which are propo1ed for target damage and environmental interaction atudie• and for the ultimate directed energy weapon which il a11umed to be a potential drop machine. The power aupply and pouible energy 1tore for the ultimate weapon require con1icleratlon Iince that 1y1tem will require a teat facUlty aa the weapoa i1 clevelopecl. 17 2.3.2 · FacUlty Power Supp!y and Enerv Store The parameter• of the accelerator~ which are outlined in Table 1 were determined both by the need for a certain ran1• of ener1y denaitiea on a tarJet over a uaeful area and by a reaaonable extenalon of preaent teclmolOJY· The potential drop accelerator can operate either with a ateady applied vo1ta1• or with an impulle voltaJ•· The former baa been choeen bec:oouee lt b almpler, 1lvea better beam control a11d there b much Jreater famlliarity with the tech nique. However, l11 the ultimate weapon aft lmpulee vo1ta1• machUle uain1, for example the Marx circuit if adequate eneriY ca11 be etored capacitively, mi1ht be uaed aince the euperior total volt..Je inaulatlon 1tre111th obtained with lmpul. . rather than direct voltaJ• la attractive. The moat powerful hlJh voltase lmpulae generaton which have been made are probably 7. 5 MV, 180 KJ at General Electric Company, PittaU.eld and 8. Z MV 420 KJ at the Khar'kov 17 Electrotec:hnlcal Inatltute (USSR). In the teet facility the power aupply problem il e11entlally one of voltale convereio11 to the hilh potential required by the accelerator. Apart from the ability to produce hilh potential a, the volta1e convenlon device baa to be capable of eupplyin1 eufflcient current to atore the required efterJy l11 a reaaonable time &ftd to aupply the leaka1• a11oclated primarily with the eneriY etore. In decidin1 on the power converalon device thoae approach•• capable of developln1 •llniflcantly 1reater than 1 .MV DC were conaldered. Machin•• of intere at were the Va11 de Ora.afi (belt machine), variable capacitance 1e11era tor, variable reluctance 1enerator, lnaulated core tranafonner, Cockroft Walton multiplier and the Dynamatron. All of the. . machine• except thelaat are produced or are bein1 uamlned experimentally in the HVEC aroup of com pule• (aee Refs. 18, 19). The Dynamatron, which ia an r.f. coupled DC 20 eupply baa been produced up to 3 Mev. The belt charJlDC macblDe wae choaea for the follow!Dt reuODe: . ~- • It auppllea adequate current ~1 ma). • It ia the aimpleat and cheapeat approach. • It haa developed potentiala approximately three tlmea higher than any of the other machlnea, and ia capable of further extenalon. • It haa been amply proven. • Polarity can be reveraed by aimple awitching . • It ia adaptable to apace operation. The accelerator ahown on Fig. 1 uaea a Van de Gra&ff aupply and haa dlmen aiona correapondlng to a 3 Mev machine which can accommodate a lar1e beam tube. None of the voltage auppllea mentioned above could po. .l bly aupply the inatantaneoua power delivered in the accelerator beam, thua an energy atore ia required. Thia atore ia armed over a relatively long period compared with the delivery time of the beam and conaequently the power from the aupply can be fairly amall. The lnatantaneoua beam power with the Z Mev accelerator ia ZOO Mw in the electron pulae, whereaa the power a vall able in the Van de Gra&ff la only Z kw. The atored energy needed by the accelerator la determined by the allowable terminal droop during the delivery of the beam and the char1e in the puhe (AQ • C 6 V). The terminal drop in potential muat be limited becauae it influence• beam containment and focuaaing. Any poat acceleration beam bend ing ualng ma1netlc flelda would be very aenaltive to terminal droop becauae of the variation in particle momentum, and it la becauae of thla that a horisontal machine la propoaed with both target chamber and drift chamber in line. The allowable droop will be the aubject of early experiment in the facUlty, but ex perience with puhed accelerator. at HVEC lndlcatea the drop in potential ahould be lltnlted to 100 kv in a Z Mev machine. A lar1er drop would probably be allow able at 10 Mev, but 100 kv haa alao been ~aed in that deal1n outline becauae lt la d. .l rad to lean • marpn for lar1er baUD c:harl••· 19 3 The drop of 100 kv and the beam charae of 10 • C indlcate1 that a terminal capac:!tance of 10,000 ...,.._, ia required. It ia nece11ary to en1ure that the dlmen1ion• of thb capacitance bank are not exceuive and that the leakage current II tolerable. The dlmen•ion1 of capacitor• for varioua application• can be determined from allowable eneriY den1itiea which are related to the particular application. The eneray den1ity of a given capacitor ia proportional to the aquare of the voltage at which it il operated, and thil ye»ltage hal a maxi mum value related to the lifetime de•ired, the number of voltage reveraall and their aeverity. Fiprea 4a and 4b 1how typical relation1hip1. The bank in thia ca1e wUl not experience volta&• rever•al• except in the ca1e of breakdown. 3 EnariY den1iti., a• hi&h a• Z j/inch are obtainable in an optimum package (e.g., optimum capacitance and volta&• value). At the hi&h voltage ratin&• and low capacitance ratin&• de1ired here the eneray den•ity ia rather poor. However, the dlmen1ion1 of a 10,000 ...,.._, Z Mev bank ba1ed on a SO kv, 0. 1 td capacitor ia indlcated on Fia. 1 where it can be 1een that the bank i1 rea•onably 3 compact. The eneray den•ity of thi• 0. 1 ..,t unit I• 0. S j/lnch . It ll intere•tln& to compare thi1 with the maximum ener1y den1ity obtainable in a hi1h dlelectric 3 1tren1th material auch a• mylar, which ia about SO j/inch • The leaka1e current In the bank il determined by the me1 ohm x..,t product, .. which II a mea1ure of the quality of a capacitor. A value of me1 ohm x..,t • 10 1• typical of a 1ood unit, which pv•• alaaka1• current of Z - ampere• for the Z MY 1tora1• bank, and 10 ..., ampar., for the 10 MV bank. The voltal• acro11 the capacitor~ in the bank hal to be controlled by re•btance 1radin1, and thll wUl pve an added leaka1• currant ol parhapl 40 I' ampere• and ZOO ..., am par•• re1pactively. Other lo•••• 1uch a• that due to corona wl11 have to be limited by 1ood hllh volta1• de•liD ln1lde the pra11ure tank. .. The enerpe1ln the Z MY and 10 MV 1tora are re1pectively Z x 10 5 and 5 x 10 joule1, which can be compared with enual•• of Z50 joul. . and 6Z50 joule• in typical accelerator•. Obvlou1ly •om• thouaht hal to be alvan to Cha poa.u.ulty o1 bnalrdown Ia the ayatem &Dd the affect of thaea lara• aaaral•• zo Zl bein1 dumped. Breakdown could take place either in the hi1h preuure 1a• environment or throu1h the vacuum of the accelerator tube. Even thou1h the atored ener1y b of the order needed for the electrohydraull.c formin1 of metab, dbchar1• in the hi1h preuure 1a• h not expected to cauae aerioua c:llfficultiea, althou1h thl.a ahould be confirmed by experiment. With re1ard to breakdown l.n the accelerator tube, which l.a more ~ly, it l.a pouible that the dbchar1• Zl curre"lt would be limited by apace char1e effecta. Experiment• at HVEC where hi1h pulaed current• from a vacuum arc aource ("" 4 ampere• of elec tron•) were accelerated to about Z Mev ahowed that the maximum current waa z limited to about 1 ampere per em of tube area per MV/ m tube 1radl.ent. Thia zz confll.cta with experiment• by Braach and Lan1• who obtained 1000 ampere• throu1h an accelerator tube ual.n1 a Z. 4 MV impulae 1enerator, apparently aomewhat to the detriment of the accelerator tube. In theae laat experiment•, z which were with a rather unuaual tube, the current waa 5 amperea per em per MV/m.. A,al.n, the effect of atored ener1y on accelerator tube breakdown ahould be the aubject of experiment. The current limit mentioned above ml1ht apply 84lU&lly well to the controlled dl.achar1• of the beam. and for 100 ampere• of electron• the inalde diameter of the tube ahould be at leaat 4 inchea. In conalderin1 protection a1al.nat dumpln1 exceulve ener1y (current) at breakdown the ftrat atep l.a to l.ncreaae the output impedance of the bank by addl.n1 a realatance between capacitor and terminal. Thia haa two effecta. It limit• the diachar1e current and increaaea the dl.achar1• time. which allow• the dumplac of moat of the bank eneriY via another path, for example a crow bar circuit. Unfortunately thla realatance 1a limited by the allowable potential drop acrou it when the accelerator operatea, and in the ca. . of the 100 ampere pulae a maximum value would be about 500 0. Thla pv•• a bank dl.achar1e time conatant of 5 .,. tee; which would require a faat actf.ns crowbar to divert the dl.achar1• (lf tbl1 were found to be neceuary). Spark 1ap crowbar circuit• Z3 have beea Mnloped tor de&dlll h'&DIIIIittlq tubea up to 350 kY, withia 2.,.. Allll tMH tubea Ia aome Hapecta al'e quite almllal' to acceln&tol' abea. Thea• crowbar• operate at atmo1pherlc pre11ure, and a cr~wbar to operate wltbln the accelerator tank at hiah pre11ure could be a much falter device 1inee breakdown at hiah preuure develop• much more rapidly than at atmo1pherlc pre11ure. The uncertalntie1 a• aoclated with the operation of an accelerator with lar1• amount• of atored energy •u11• 1t1 that the terminal capacitance ahould be inereaaed in ataaea to 10,000 l'o.-1 during the experimental ataae, perhapa in atepa of 1000 .,..,.t. Another aapect of eneray etoraae and particle acceleration il that of beam atiffening. The term 1atiffenina• refer• to the maintenance of accel erator tube gradient in apite of interaction• of the beam with the tube. A loll of tube gradient can be by two eauaea. The flnt ia due to the movement of beam eharae down the tube which cauaea induced eharae flow in the accelerator column and lo•• of aradi ent, particularly at the aource end of the tube. Theae induced charae• flow in the atray capacitance• between the electrode• of the accelerator tube• a1 well aa in the atoraae bank while the front of the beam pane• down the tube, but once the charge in the tube (beam) reachea equilibrium, there la no fur ther net flow of charge in theae atray capacitance• by induction. The atray capacitance between aectlon• on a 3 Mev Van de Graaff wa1 meaaured and found to be approximately 250 .,..,.t, and with thi1 value the lou of gradient due to induced charae flow waa found to be unimportant (••• Appendix II). Th• aecond cau1e of lou of tube aracUent la actual interception of part of the beam by the electrode 1y1tem, and thl• lou lncrea1e1 a• lona a• the pulae la1t1 and beam il intercepted. The areate1t 1tlffneu i1 obtained by coupllq the eneriY 1toraae bank to each electrode a• 1hown on Fil• 5, which would alve 0. 4 .,.t between electrode•. However, thil con1tdct1 de1i1n and complicate• protection, and the belt approach il to increa1e the inter electrode capacitance to an adequate value while retalnin1 molt of the 1tored eneriY Ia a eeparate buk. Typlc:ally the 1radlent oa the accelerator tube crowbar aeam Fig. 5 Accelerator Coupled for Maximum Stiffness 2t i' f , would be SO kv per electrode aection, and if 1~ of the 100 ampere electron beam waa intercepted by the firat electrode, an interelectrode capacitance of 1000 IJ.td would Jive a potential drop of 10 kv in the firat electrode aectlon by the end of the pulae. Beam interception will have to be a aubject for experi ment, with initial experiment• uain1 tubea over-deaigned with re1ard to atiff neaa. In concluaion some commenta will be made on the maintenance of beam potential, atill uaing terminal capacitance but with more aophiaticated technique a. In some inatances, liners have been provided in accelerator tanka to reduce the variation in beam potential during delivery. ~4 The liner is pulsed during the beam firing and the potential variation b impressed on the terminal of the accelerator through the stray capacitance to compensate the droop. Thia approach can be uaed uaing the energy atorage bank for coupling aa ahown in Fig. 6a. The low voltage bank ia charged to reverse potential through a high reaistance before beam firing. When the beam pulae atarta, awitch a ia closed (spark gap) and the linear fall in potential of the H. V. bank ia largely compensated by the awing in voltage of the low voltage bank. Figure 6b shows the voltage variationa. Thia technique could be applied eventually to the proposed accelerator ayatems either to increaae their cur rent capability or to decrease the variation in beam potential for a Ji9Jen current. An intereating approach to the exact maintenance of beam potential 25 which waa propoaed by R. J. Van de Graaff ia ahown on a Z Mev beam ma- chine on Fig. 7. The charge which flowa from the Z Mev terminal in the pri mary beam ia exactly compenaated by a flow of charge in a beam between the Z and 4 Mev terminal a. In other worda, the charge on the Z Mev terminal re maina conatant ao that there il no chanJe in the 2 Mev beam potential during the pulae. An electron beam ia eaaier to produce and handle than an ion beam, ao that if the primary beam conaiata of electrODa, a compenaatiq electrOG 25 \! f +2MV C (Energy Store) 1 L 200Mn ~L-----~~--c-~-----s~~~~ Fig ea Driven Terminal Concept for Maintenance of Beam Energy During Discharge + 4V t-- t• 6t Fig. 8b Voltage Relationships with Resonant Pulsing of Ttrlllnal (~>C1) Z6 21 + .. f ~ .5 j I + N ~ _______ - · .. .-.............. ..._ ___ ...... . . ~ beam llowa from the • WV to the Z WV terminal, and lf the primary beam conalata of lema, a compenaatiDc electrOJl beam llowa from the 2 MY to the • MY terminal. Simple analyaia ahowa that without any dearadation ln beam potential the followlna relationahip bolda: maximwn u ..f ul beam enerp delivered eneriY atored in hlaher volta&• terminal where k la the ratio of the potential of the upper terminal to the lower terminal. l'or example, in the •· 2 MV concept, about half of the eneray atored at • MV could be delivered aa uaeful beam without any dearadation of potential. Thla concept may be important to the directed eneriY weapon• protram becaw.e a lara• part of the atored eneriY can be ueed in a mono eneraetic beam. z. 3. 3 Application of J'acUltiu to the Development of Eneray Storaae for Weapona General It b vbuallaed that the D. E. Teat Fac:ility wUl be udll&ed for in· vead.aattona on larae enerty atoraae •yatema and their couplinJ to acceleratora. Hlth voltaae vacuum breakdown 1tuc:llea are belna conducted with the objective of boldlna multl-m11Uoa volt potend.ah acrou vacuum aap1. U volta&•• in the 1 8 order of 10 - 10 volta can be inaulateclin vacuum, eneray 1tora1• can be ob tained in autflclently blJh denaitl. . by mean• of electric field• in vacuum. Thil ~. ll a major objective of the atudle1 and tut facWtlea deacribeclin Sectloa 3. • Baaed oa praeently available data OD vacuum breakdown, a •olld ). c:Uelectric capacitor bank would offn a better, but aWl far from ideal eolud.on. Uainc the beat available cll.elactric material• ucl auumlna that they wU1 aup 6 port 10 volte/cm for extended period• o1 tlma lA a apace ellvlrfAIDaent, auch a 3 8 ayatam would have a volume of about ZOOO m for etoraa• of 10 joulea. Jn- vead.iatlOAI ol c:Uelectric breakdown and aoUd dielectric capacitor• in the vac uum (apece) eariroament could be performed in the hi&h volta&• vacuum break dowa ladlitr (aecti• J). Howwer, thl1 c•ceptla aot very attractive aad 11 !. aot P"•••tly "'-1 ,.... .. ._ A better, '-'at certalaly mora caatpllcata4 method, t ll tM etorac• ol eaero la btcJa ~~~a~aedc 4el4a. ,· t Inductive Ener1y Stora1e Large quantitiea of ener1y can be atored induc tively in ma1net coUa generating a hl1h ma1netic field. The ener1y denaity b given by: B~ 3 W •- joulea/m ~IJ. with Bin Webera/m~ -7 1J. = 411' X 10 A field of 15. 7 Webera/m ~ = 157 kUogauu would repreaent an 8 3 energy denaity of 10 joule/m • Becauae of the high ohmic louea in con- ventional magnet coUa, the powe;f required to keep the energy in atorage would be prohibitive. Therefore, thia concept would only be feaaible if auperconducting coU• are u•ed. But even taking into account the volume of the coil and the required helium liquefier with auociated cryogenic equip ment, energy den•itie• two to three order• of magnitude higher than in capacitor bank• may be pouible. The highe•t field• that have been generated in auperconducting coU• to date are about 70 kUogauu. Material• with critical field• in exceu ~6 of 100 kUogauu have already been reported although theae material• are ~6 ~7 pre•ently not •uitable for large ma1net coil•. Recent development• ' indicate that •uperconducting coila 1enerating more than 100 kUogau•• can . probably be buUt in the not too di•tant future • One concept of an inductive ener1y atorage ayatem with which it b pouible to develop relatively hilh volta1••• ia •hown achematically in s s Fi1. 8. With awitchea SZ and open and cloaed, the atora1• coil ia 3 1 charged up to a certain c.arrent level. The char1in1 time depend• on the DC generator and the time conatant of circuit 1. When the coil ia char1ed, s Sz ia clo. .d and opened. The eneriY ia now kept in atora1e by the per 1 alatent current in circuit ~ (Sz la a auperconcluctin1 awitch). Since there 1• s no reaiatance in circuit Z, there are no ener1y lo••••· can now be 3 s clo•ed and ener1y diachar1e take• place throu1h the load when 1• opened. 2 29 . .. ·---··- ······..- It b quite po. .l ble that the beat approach to the DC aenerator ln thia circuit would be all electro:maanetic machine - aho auper conductina. Many problema remain to be aolved before the fealibUity and appli cability. of thil concept to D. E. Weapon• h demonatrated. Some of the .. prob lema are: • Dealan of very larae auperconductint colla a•neratina blah maanetic flelda at high current denlitiea. • Development of a hiah voltaae. faat-actinc auperconductlng . awitch (S 1n Fig. 8) . 2 ,v ) • Deeian of •upport etructurea to contain the lar1• mechanical forcu 1enerated in the coil. tim•• • Obtainin1 aufficlently faat dhchar1• while maintainlna th• atoraae coil in the aupe rconductina atate . • lnYeatiaation of matchlna output lmpedancea and couplinJ of the atorage ayatem to, for example, an accelerator tube. • Protective circuitry to protect the equipment and diapoae of the etored energy in ca1e the coil ao•• normal in an uncon trollable manner. . . • CoU and ay1tem protection aa&lnat hi1h volta&•• t•nerated )' durln1 dlachar1e . • DeYelopment of cryogenic ayatema and a nelium liquefyinJ plant to operate the •yatem unattended in a •pace en.vironment. The•• problema are of conaiderable maanitude and aome of them are currently under lnveatiaation aa part of Contract AF08(635) -2166. Thia proaram le preeently 1n an early •tate of development, but •om• comment• with ret)Mct to the D. E. Teat Facility can be made. Utility of the Propo. .d Fac:Ultiea to Eneru Storaae Studlee The moat promlllnt approach to a D. E. Weapon lnYolvee the ac:c:eleratioli of c:haraed particle• to h11h potential•. Howenr, hlp vo1t•1• limitation• of the coil &ad ewitch ln the coacept of l"il·· 8 would prnent the Ule OJl a linlle C:oll . eyetem ADd nec:e. .l tate a bank etruc:ture eimllar to a capac:ltor bank. Ill ita i l' eoa elem•ary form. the accelerator wbe with ladlacd'ft ...1 '11 etorac• J'ta. ..qbt M •• elaowa 1a 9. The laducdYe etoraa• eyetam 11 a curreat 30 l ,--------/,:: ~ ~~ ... D.C. Load Generator Impedance L----------1 Fig. a Stlperconducting Inductive Energy Storage System _ ~Accelerator / TUbe Fig. 9 Coupling of Inductive Energy Storage System to Accelerator Tube device aa oppoted to a capacitor bank which le a voltaae device. Conae quently, the dhcharae voltaae 11 highly unaltive to the load impedance and operation of the dhcharge awitchea muat be properly aynchronised. Parallel capacitor• may be needed aero. . the dbcbarge awitchu to provide a auffi ciently long rbe time of the dhcharge voltage to prevent the coil from goina normal during dhcharge and also to provide the dulred pulse width. After preliminary atudiea on amaller ayatema, it la vbualized that theae experiments can be continued by aubatltuting a properly d.esianed inductive ayatem for the capacitor bank of the D. E. Component• Teat Facility. The colla and diacharge switchea require a liquid helium environment and are mounted in a auitable dewar system filled with liquid bellum. For these ex perimenta, the atorage system need not be poaitioned in a vacuum chamber. Breakdown atudiea and the development of dewara, helium liquefying plant and cryogenic recirculating syatema for unattended operation in space would benefit from the availability of the high voltage vacuum breakdown test facility. The aize of the cryogenic eneray atorage ayatem, ita cooling require menta and the time required to develop thh ayatem la difficult to predict and depend• on the aolution of the problema outlined above. Preaent effort• are on atorage coila made with Nb Zr wire with a critical magnetic field of 70 kilogauaa. Future availability and aubaequent uae of more aophiaticated ma teriah with higher critical field• h anticipated but depend• on pro1reu made in .nateriala development. Development of the cryogenic awitch h likely to be one of the moat difficult and time conaumina problema, but preliminary experiment• can probably be performed ualng a conventional awitch by elimina tina the penhtent current mode of operation ln Fia. 8 and awitching from charae to dhcharge operation directly. A ..u m1n1 a aulficient effor·t put into the inductive energy a tora ge inveatiaationa and aatlafactory proarea • toward• the aolution of the problema outlined above, the feaaibillty atudiea and experi ment• may be completed ln 1 to 1 1/Z yean (1964). Completion of dealan· and development of a prototype eneray atoraae ayatem for operation in conjunction t with. the D. E. Component• Teet Facility may require another two (2) yeara (1966). Parallel with thie l;ut effort, a program should be conducte~ to atudy and develop a prototype apace cryogenic ayatem including helium liquefier, dewar, transfer and recirculating ayatema to operate the storage device. Poaaibly prototype hardware for facility teats resulting from auch a program could also be available at that time (1966). As pointed out before, due to the very early state of development of this energy storage concept, the time esti mates made can be no more than a rough guide. Acceleration of the program by increased efforts may be poaaible while delays could occur as a result of presenUy unforeseen difficulties. Z.4 PARTICLE ACCELERATOR SYSTEMS z. 4. 1 Injector Conaiderations for the Facility The baaic requirement of the positive ion or electron source for the facility is that it provides a monoenergetic "parallel" beam of charged particles for injection into the accelerator tube. System a of cylindrical a ym- metry will be considered here and apace requirements in the high voltage ter minal ahall not generally be conddered a limitation. Since the system will be operated largely in the pulaed mode, aource duty cycle and gaa efficiency must alao be considered for terminal applications in the propoaed accelerator. Since projection analyaes have thusfar been restricted to low Z ionic species, largely on the bada of total voltage -velocity requirements as well as ~rget penetration. proton injectors will be treated here and the problema o! utilisation of the sys tern with a negative ion injector will be excluded. Several exbtlng high energy acceleratora, such as the proton syn chrotron and linac, accelerate short pulses of charged particles to high energy so that the duty cycle may be quite small. The in.rectora used with these sys tems are well su.ited for our purposes, in which high peak currents for pulse periods of one millisecond or leu are required at low duty cycle; 1. e. leaa than one pulae per second due to the charting and atoraa• llmlcatlons of the generator. Table 3 presents a review of thoae lon sources now ln exbtence ' p.., ''""''~'· ,•. ' Table 3 A Renew of Gas Discharge Proton Injectors v Total X Current Aperture Current Hl: Hz S<Nrce Type Reference amperes kv Pulse Length Density 2 Occluded Ga. 30 0.625 10 - lOJlS 40:0. 6 0. 74 A/em (dlecbarge) .. 2 31 0.400 20 400JlS 9: 1 1. 15 A/em 2 .. Hot Cathode 32 ""2.0 100 25ms 9: 1 2. 1 A/em "" Gas Discharge 2 J)Qoplasmatron 33 -1.0 low 1001'S 2~1 6.5 A/em 2 J)Qoplasmatron 34 0.120 70 lms 7: 1 56 A/em 2 J)Qoplasmatron 35 0.275 45 d. c. 4:1 7 A/em ·-----~· which are capable of currents approachin1 the ampere ranse. Since only source. of the ga. discharge type, {generally referred to as the masnetically confined arc), are capable of the high emiuion current dendties required for ,, ' the injector application, a survey of the other source types (r. f., P.I.G., spark, etc.) would be academic. For modest pulse duration, the gas discharge source can be opera ted in the arc pulsed mode, Coupled with a pulsed gas valve, this injector can lead to relatively high gas efficiencies which are compatible with the pumping capabilities of the tube-terminal auembly of such a machine par ticularly where very low duty cycles are of interest. The high gas efficiency in pulled operation is the obvious advantage of the first injector considered, the occluded gas source of Crawford et al. In this geometry, the molecular {and/ or atomic) hydrogen is supplied from hydrogenated titanium discs from which the gas is extracted subsequent to triggering by a suitably matched pulse forming network. Gas is injected into the arc only during the pulse period and hence the gas efficiency can exceed that of the externally fed source due to conductance and valving considerations. Since the occluded gas capacity of titanium is very high {400 c. c. at STP/gm of metal), adequate storage poses 5 . no problem. Crawford has reported 10 pulses at low repetition rates at high 30 ·output levels with no degradation in performance. {see Fig. 10) It is now necessary to examine the other qualifications of the sources enumerated for the proposed application. The average current density at the emission aperture of the source is a meaningless figure for optical design un less some indication of beam divergence is included. A review of injector acceleration tube matching considerations is presented in the next section and suggests that considerable effort is required in this area for the optimisation of a high current electrostatic machine. Perhaps the injector most representative of the state-of-the-art of source -tube matching is the 100 ma proton injector described in Ref. 34. With the 0. OZO inch diameter aperture used in this •our~•· the beam current 35 Kovar Metal Stainless Steel Anode-Extraction Glass wanr Titanium Insulator SandWICh Fig. 10 Occluded Gas Source: Axial Extraction (30) 36 deneity at an extraction potential of 70 kv 1• of the order of ten• of ~mpere•/ z 36 em • Conventional Einsel lenaea are inadequate at thla level and a wedae tank analoa waa uaed for len• deai1n. The lena shape waa modeled in the tank and the axial potential obtained. Thb information wae tbell used for calculation of the equipotential dia1ram with an aaaumed beam radius and diver1ence. From the particle orbit and equipotential data, lene aeometries were achieved to adequately handle the beam. In matchln1 the column to the eource, a four electrode lent ayatem waa used which would provide an injection ener1y of 60 kev to a linear-high gradient column of 538 kv/m. Injection beam requirement• were derived by ray traclna backward• through the column and the lena 1yatem waa then de aisned to match the extracted output to the tube input. Tbia matching ie uau ally referred to aa tource emittance - tube acceptanc~ mating and aimply ttated, require• that the tube can adequately accept and focua the current aup lled to it within the pbaae apace area characterised by the tource emittance. + The above mentioned injector provided an 85 ma H beam with an emittance of 100 mrad-cm in a beam diameter of 1 em to an accelerator column of cal culated acceptance of 179 mrad-cm. The contiderations below indicate that adequate computational technique• are now available to permit the deaign of a matched injector-accelerator tyatem capable of providins proton beam en 3 ergy fluxea in the lOZ - 10 joulee/cm?./m1ec, (pulaed power den1itie1 of z 6 10 w/cm ), with a two mlllioo volt machine. Since the faclUty will al1o be capable of ute a• an electron beam accelerator for blah lntenaity pulaed beam propaaation and bombardment 1tudie1, aome comment• concernina electron pn optic:• 1hould be made here. ?.9,38 In tbb caae, the tec:hnlquea of optical dealp are reaeonably well ct.velopad. and pulaed electron beam a Clta•) of Z Mev eneray have already been realiaed in the ampere ranae at thb facility. A ZO ampere, 100 kev electron pn or in jector caa be de..,_. for the fadllty wltll exlatlac te~•• while u ex I taw• to 100 unpe&-ee total cuneat 11 feaaible either with the uea of exletiq bllh CUI"rant denelty matrix or lleld eml. .l • cathode• for low duty cycle puleed ,. operadOD. t. f ,, This study has served to point up the inadequacy of preaent beam diagnostic techniques for injector evaluation with low divergence, high cur- 39 40 rent density beams. Emittance measurements are usually accomplished ' through the insertion of variable position apertures directly into the beam and scanning of the beam at a plane downstream for determination of the beam di vergence characterizing that portion of the beam. Any of these systems suffer from the perturbations induced due to physical interruption of the beam, the associated secondary particle effects and beam potential distor.tion. These effects are not considered serious at low beam current densities while their application to high intensity pulsed beams becomes difficult and the results highly suspect. Other indirect techniques must be developed for pulsed beam study if reliable emittance measurements in the 11rad-cm region of interest for prototype charged particle projector studies. 2. 4.Z . Beam Focusing During Acx:eleration In a fixed-voltage particle accelerating system, the power delivered to a target is directly proportional to beam cuiTent, which is chiefly limited by the mutual repulsion among the beam particles, the so-called space charge force. The essential problem in accelerating high currents is the design of electrodes with focusing properties which offset the space charge force in such a way that the beam leaves the accelerating system with suitable diameter and divergence. In order to define the problem more precisely, it is necesaary first to discuss the practical constraints within which the problem must be set. In present Van de Graaff machines, since the high voltage terminal and charging belt are insulated from the containing vessel by gas, under pres sure, it is necessary to provide a vacuum path from the terminal to ground for beam· acceleration. The vacuum tube is usually constructed of a series of con ducting discs separated by glass ring insulators and having holes in their cen ters through which _the beam may pass. The discs are tapped into a resistor bank between the terminal and around in such a manner that the potential b aradecl uniformly alona the ·tub~. The constant aradient insures maximum 38 ineulation etrength for the tube. In addition to the potential grading, the diece &leo eerve to prevent charge accumulation along the tube and to ehield the beam from extraneoue infiuencee. The vacuum tube conetitutee a leak- age path from the terminal to ground. The volt-ampere characterhtic of thil path typically exhibit• current run-away when the voltage exceed• a limit which depend• on the geometry and length of the tube. The reaeone for thil phenomenon are not yet completely underetood and none of the vari oue pouible explanation• will be detailed here. Suffice it to eay that current run-away lead• to overload of the high voltage eupply and muet be avoided. Thue, for a tube of fixed geometry, there exht1 a minimum length for. each voltage. The etandard tube geometry ueed by HVEC requires aomewhat over one meter to atand off Z MV without run-away. The large number (over 40) of discs required for such a tube make it expedient that they all be identical and of aa simple geometry aa pouible. Thia, together with the adviaability of linear potential variation, leavea virtually no room for deaigning focuaing propertiea into the tube. It haa thua been natural to think of accompliahing beam forming at low energy 1n the immediate vicinity of the particle aource while performing moat of the beam acceleration in the uniform field tube, which haa virtually no focualng effect at all. The beam handling problem la then convenienUy divided into two parta. The flrat of theae Ia concerned with beam flow from the particle aource up to a potential of S0-100 kv. In thi1 region, uaually of the order of centi meter• in length, apace charge ia of firat order importuace becauae the epace charge denaity and force are quite large near the particle aource where the beam move a very alowly. The particle aource and the electrode• to form the beam in thia initial etage of acceleration will be called the injector. The aec ond part of the problem concern• the fiow from the injector to the final energy of Z MV 1n the (uniform field) acceleration tube. Becauae the average now epeed 1n the tube ia ao very large, the epace charge deneity b rather emall ao that the force due to it ia aecODd order in compariaOD with the applied force. 39 ________ _ .,"' Since 1pace char1• effect• are of primary importance ln the injector and only ..c ondary importance in the acceleration tube, the analytical techniquu which have been developed for treating each of the .. etagu are quite different. The problem with the acceleration tube il not one of deelgn but rather one of determining what beam input conditione lead to acceptable out put characteriltice in the etandard tube duign. Thil problem can be eolved ueing ray-tracing technique• with a coaree space charge approximation. For tunately, suitable computer programs are already in exletence at HVEC ~1• 4z They have 10 far not been applied to beams of the high currents deelred in the preeent application, but there are no conceptual difficulties standing in the way. The flret etep in high current accelerator design would then be the application of theee program• to determine the acceptable range of acceleration tube input conditione. The problem of dedgnlng an injector stage to produce a beam having characterbtice within thla acceptable range is far more difficult. Suitable ana .lytical tool• and the a11ociated computer programs have been under development at IPC for eeveral years. A brief outline of this work as of about one year ago ilgiven in Ref. 43. Achievement• Iince that time have not yet been formally diaclosed. However, they enable, in theory, the analytical design of injector etas•• to any deeired desree of accuracy. However, all of the neceeeary com puter prosrame have not yet been written and there are, in fact, many question• yet remainin1 about how beet to apply the theory from the etandpoint of economy of computer uease. Neverthele11, theee queetlons do not prohibit the analytical deaisn of injectors but only stand in the way of improvin1 the technique• pre• - ently in uae. The overall acceleration eyetem il current limited by epace charse effects in the injector etase. Voltas• breakdown limits the etrensth of the focuains flelde attainable in the injector and hence the capacity for compenaa tion of epace chars• defocuelns. Injector• for other application• have already been clealcnecl (thouch not teeted) which ahoulcl be capable of cleliTerina beam 40 currenta of up to 0. 5 amp of protona or ZO amp of electron• at 100 kv. With the improved analytical toola which can be made available in the near future, it may prove pol 1ible to double or triple the•e currenta. It 1a reaaonable to predict that currents of this magnitude can be brought through the 1tandard Z MV acceleration tube 10 &I to emerge in a well collimated beam with a croaa z aectlonal area of the order of 1 em • 41 • - - •• - - - IIIJ!tv) (I) F=;. =lltl ~2 =.-,..~ ...:.-.r~~ - - . ...,_ •. - ...._ . - - r ( ~:. T ~ - T 1 K &T ) L&T: \ .l ,._ ---.-.- l lit i - lliX J.IJ •• ·•· -A•x- • _.q • =I &t 8x - ............. . ... --------··~···-··-·· .. i ' ,. - - - - - Material .. • - • - - ~· • - • - - • - -• - - - - - •I' - - ·.. - • - - - Ill • • •• t. -····-~- - 1 ' .I • J l 1 .......... ., .,. ....... . . - .Z •=J. .. It • ¥ I t-z • •. ~ -c l 8 • ~.~ • ... s_ I - ••• • -- - • - '·II - • - ··- • ... - · - ·. .. (14) - ,. I • - . - - - - -- - - - -- - ... -"·"'--.'f.-: ~·<;. ·~---'·•;,,- .,.~~- :i f TableS I' Some Characterimca of Structural Material• i Tenaile Break Atomic Expanaion z z Melting O...dty Weiaht Coefficient E q/cm G q/cm Fmax Point AH L. Material A ax lo-s X 10S X lOS kg/cmZ x 103 ~ kcal/mol z.o z.o Z.7 Al Z7 7.0 Z.3 940 68 . 9.0 Cu 64 1.4 1Z.O 4.0 Z.8 1350 73 a.z No 96 0.49 30.0 lS.O Z8 Z900 Ut 0 17.7 w 184 0.4Z 36.0 14.0 4Z 36SO 15.4 Ta 181 0.69 zo.o 9. 3 3Z70 Fuaecl Z.9 0.05 1900 Quarts 7.8 Steel 56 1. 8 zo.o 7.0 7.8 1700 95 • 61feC • I .... l;-l • ••-L • rx- E 1=-• (19) (- .. . .,1! . . } I ~· 't ' '· .. Mean Diatanoe • Density .. E Density! ~· -~ ~ at 2.7 AI f 8.0 cu 0 11.4 Pb at ~ L___-- L__ --- --- .,I I I I I I I I I I I I I I I I « Kl I I I I I I I I I I I I I I I I I I I I J I ~ 00 ~ Energy (Mev) Fig. 12 Mean Range of Protons in Metals • i • j ..... ~- .. • - - - - - • - -· I • I I - : 1·-· - - ·--·--···· •• -· ' t • (30) . " . . . " .. . - - . . 2 2 . ,.,. .. . • • .. ~ 0 • - ..... - .... .... - ..~.... ~-- Material • - • • • • • . • - - - · - ~) I I · on Clx B t. radiation 0 • ...~ ~~·-n ...,. ~. ... .-.~- ----... -.....- .-......... ... _ ..-~R. I - I v r.• • - • - -.111111111111, • I ·-~- •• I ,. i ;:'t ·- I I I • ·- .. --·~·-- - ' .. ' l i .. - · - .. II t I 1.- I = -. - (38) - - · •• - - I , - - . .I I 63 C~•o) 'UHO. UOII""W ..... ••IUM ~ Q !:::: 4( ~ 2 , 0 a~ s 0 8 ~ ~ f -'() G) 'i c 0 il 0 co i 0 If N g ..... .5 0 ~ ·s () z ~ e ! ~--~n-----~----~~----~-----8 '"----------~-- ..... - • Sl • •=·c~ {40) • • I - .-=-T II ·-··· .. I .. - - ... ... • • ~·- ~·- ..II - •.•.•.0 1.0 .··.-. - ~ - - - .. ' ' • - - , • • •• • ..• .. ·- - ... ,... ", . ... .,~-·- .. - ..... ,., .... ' . ~~ k · - ·· , ·, ·, . • I j I . I\ l {. ' \ I \ I •\ ' ---~ H 71 .. - - -·· - - - - -- -··· ...( , ' \ ~ j. :. ! f i: l.. - - - --·- ~·-""· ... .. I I v - · - - · - . . ..... . ··~~.. - T=T~-z·z·z·) • - - . ··~' - }o(llf; - ho ,. . . .it 1 --3/Z • t • -- ~~~~~~~ f ' ~ • . I _ _ ( _._... .. blljSL ., •• · - · - - • - · -· - - .... • • • = ...... ·~~.., --··' -.......- - - -- --·····-- _ ... .. -1/3* z 0.• =o. • •• • • - '=--··e-···)1 t- ··-~--- I ··---·-- ··-. ... .. ~· > l • - .. l"l•llPalF • crl~-...,.c. e & . : •• • •- · I .. •o• ·--·I .. ..... Ill-· ~~ - - - E. -dv .. ll - · + ..... - ..... . _. • - .. L • .....z t - z ...., . • • • 14;11 I .....l :. ...... ... -.- ........ .......I t - .. (b) (3) (A) • ~• -.· (b) (3) (A) •• -·-except - ...,..--.1 • WE,)IIZ • - (b) (3) (A) - • 1{-•112 I - 1 111111f:7 -I} -. > I • I I .... 11·-'!~-I\-·· --· ...- --... -·--·z-. ... - · · ~·- - - -· - . ) - - ---------· - (b) (3) (A) --·· .. • ••• --·1 .... --·- ............................ • __, . . · - -..• ,.Ax-· - - - - ~~~--··13- t"' · - - i··~~ (b) (3) (A) - •= -··-·· --- ~-·'••vr•• .1(~..,·· (77) ...1 .--26 .. - (b) (3) (A) -----··--···-· .... ·:'!!'~~--·------ (b) (3) (A) 1! ... , . • I • • • - --· (b) (3) (A) - - - - - - - - ,....._ __ _ -··-7· --· -·----· ·-.. __b ________ - - - - - - •• -------------· .. • ------= -~~·-~_. w~ I '111111~*' - •. --. ...· =· ~ ' !I' ~·-·· t+ ' '··~'IIJool"f>'~">~.~>llll'.,..•tt.i'. . f .. l«••-·-·"""·'· ..... -...... ., (b) (3) (A) • -··--~- (b) (3) (A) - (b) (3) (A) •• • wflllllllll•n9· - - - · I - .... ··-··--~~·- ·-1 .l_ __ - · · ..- ... 0. . - - - Ill~-)· ~-·1 ......, - ·-· •)'•- ____ ................... .· ~--------·-·--·--- '" (b) (3) (A) - • - - - - - - - - - - - - · following - - - t ,, f I (b) (3) (A) - - · · - - - - - -------~·--··-- ..... ~­ ,,.~., • (b) (3) (A) - • (b) (3) (A) -· (b) (3) (A) • T (b) (3) (A) • - • • I I I • "'''"\'~------ ··-··------·-·· (b) (3) (A) (b) (3) (A) ' ~.;. ' ' ( b ) ( 3 ) ( A ) • I • • • .... lf'~f~' ~~.,. '' ' ' ' 'r. -}"".'),. ~ ~-~-· .~. .. ·~·-·:. .·-··-.>-_.~---·..IT""::;• ~.:-~f~~---~ Months I 0 2 4 8 8 10 12 14 18 18 Pu ton a.ne DevelopMent (Packaged) I uno !"V'· ~ f I IOOCI !I!P«• f ulsecl ! E....,. Gun DevelopMnt (PaokCIOid) MDIIInltor 1\lbe Protection Studies £MIW atDrage Design a Fabrication a MD 111 ator - ModltloGtlon Fabrication : 2 M1V Storclge plus Tube Testa .0 -1 .._ Mindlina a Analyalng Equipment r , Multlllll 'nlroet Chcllllbera Drift ...... (25 IHtera) ' i vault /'frwM)h Construction CGIII!IIIte AooelerGtor Proving Ia - 1 . ... aou c r - o t dnleln a Acoelen: ~<t. L;.: ..... a a ~fac ~re latnr' ~ :•. ~.J! ,_ ,......,_ , Fig. 21 Program Outline -2 Mev. 2 KJ Pulse Accelerator (Facility A) (b) (3) (A) e I I rr-------t I 1 n---....... r ........ . I~~- n---...... I aet 1 11----tl I~~------ 18 -·- 28 27 - • (b) (3) (A) • ~.~ ··~ ~·-~·~,-. ~-- .. - - -------- - 0 e ,, 1 I • --- ... 0 0 - w . ---- ZO MeV Van de Graaff Accelerator HIGH VOLTAGE ENGINEERING CORPORATION Fig. 23 Emperor Tandem Accelerator 10 MV Terminal I ' ' ' (b) (3) (A) ' ~~· ,~,, x~>- ·~{~ - ......................... ............................. • · - · · · -.. ·--~ -~- • ~.· (b) (3) (A) IOZ (b) (3) (A) - - (b) (3) (A) ..................... ·-· - · - - - - - - - - - .. .. ....................... . - ' ft 1 ' (b) (3) (A) • < ,lf.' ~- (b) (3) (A) (b) (3) (A) 3. FACILITY FOR HIGH VOLTAGE VACUUM INSULATION AND POWER PRODUCTION STUDIES 3. 1 INTRODUCTION TO THE PROBLEM Earlier sections have shown that the most fruitful approach to a directed energy weapon for operation in space is by the acceleration of charged particles. These particles can be accelerated using techniques whereby their travel is synchronized with an accelerating gradient produced by a high frequency voltage at several regions along the length of an acceleration tube - the linear accelerator principle, or by a single potential drop device. The former is not only less flexible but more complicated and the latter is considered the better approach. It becomes particularly attractive if adequate energy can be stored in a reasonable volume at the potentials required for the device. Consequently, the ability to generate and support very high voltages in space is highly desirable. 7 It is expected that potential differences of 10 volts or more will be required, 8 with a stored energy of the order of 1 o joules. However, the generation and support of high voltages is not ealiy in the la!:.oratory, and is still less so in space. High voltage insulation problems of a conventional nature are often solved by the use of pressurized gases, and this leads, in turn, to pressure tanks of massive proportions: for example, the total weight of a 5. 5 MV Van de Graaff accelerator manufactured by High Voltage Engineering Corporation is 65, 600 lbs. of which over 50" is accounted for by the weight of the tank necessary to contain the pressurized gas which insulates tt.e generator from ground. It is therefore an obvious step to examine the use oi vacuum as an insulator when high voltages are to be generated and utilb.ed in :>p.ilce. Further, if sufficiently high voltages can be insulated with realizable dimensions in vacuum it would be possible to store the energy required by the accelerator uainJ the space environment as dielectric. For example, it has been SUJJested by another contractor under AF08(635)-1636 that adequate enerJy can be stored in a system of concentric spheres, the larger havinJ a diameter of 80 meters and the smaller a diameter of 40 meters. DA-ta which is available on the insulating properties of vacuum is fairly profuse, but unfortunately there has been little consistency in the experimental techniques used: furthermore, until recently experiments had only b~en performed up to 700 k/ ~ which is far short of required potential differences. Even at low voltages, the mechanisms of electrical breakdown in vacuum are not understood, . h b ff d 76, 77, 78 although many conJectures ave een o ere • For these reasons, it is not possible to design a Directed Energy Weapon to operate at tens of mega volts by extrapolation from existing data, and to obtain better data a research program was initiated under Contract AF08(635 )-Z 166. The contract required the study of vacuum breakdown up to 1 MV. In this, the study has been success ful, and the voltage range covered to date has been 250 s V s 1700 kv over the -4 -8 pressure regime 10 < p < 10 torr. A brief resume of the progress under Contract AF08(635)-Zl66 will be given here since it is most pertinent to the dis cussions which follow. The facility which has been built has a vacuum chamber which is cylindrical, 4 ft. long and Z! ft. in diameter; it contains ports for windows, gauge plates, and the pumping system, which consitlts of a liquid nitrogen trap, refrigerated baffle, mercury diffusion pump and fore-pump. At both extremities of the chamber there is a pressure tank housing a 1. 3 million volt Van ue Graaff generator, while a special pressure-to-vacuum high voltage feedthrough bushing communicates the potential of eacr. generator to electrodes in vacuum. The gap between these electrodes can be adjusted in the range 0-60 em under vacuum conditions. Figure ;5 shows the facility, while FiM.lC shows the basic high vol ... age bushing. 108 ...... ...... ..... § > 0 ...... 109 "' ... Ql) ....... N :::E ·~ ..... ..... ·~ 0 c > -0 ..... - 0 > Ill w ::l ::l ):: ~ "c.'. Iii ..... ...'c":' g 01) ...c: ~ ::l rJl ...c: c 01) 110 \ j 1,(< ' "J r .. ... .._· _ . - \ The original requirement was a facility rated at 1 million volts: with generators of opposite polarity, this requires that each bushing supports 500 kv. However, the state of the art at the commencement of this Contract was such that only large ceramic bushings some 3 ft •. in length could support 500 kv in vacuum, so that for compactness and reliability, the first step in the Contract had to be the design and manufacture of two h. v. bushings of smaller dimensions and of superior performance to the ceramic type. As Fig.Z6 shows, the final bushing consists of two aluminum-glass stacks bonded to a central flange: both stacks are insulators, one operating in vacuum and one in the pressurized envi ronment of the generator. A central shaft electrically connects the two extremi ties of this bushing, which are at high potential, while the center flange is at ground potential, and bolts to the end-plates of the vacuum chamber. Both stacks are voltage-graded internally'by means of a cylindrical urethane resistor which carries a current between the high potential ends of the bushing and ground. This resistor is connecte'd at various positions along its length to the aluminum rings in the insulating stacks, so that the potential of these rings is fixed and a uniform electric field exists along the exterior of the insulating stacks. Considerable effort has gone into the development and modification of this bushing, and of its urethane resistors, and such a bushing has held 1. 2 -4 -7 million volts in vacuum at p ,.. l 0 torr, and 800 kv at p ,.. 10 torr, voltage levels which are well over the contractual obligations. This can be considered to be a minor breakthrough in the state-of-the-art, in particular when it is noted that the dimensions of this bushing are less than one-third those of a ceramic bushing necessary to support hal£ the voltage. It is felt that this success shows that the development of similar bushings to support 5 million volts is perfectly .feasible. Althou1h a ~ajor breakthrou1h in the ability to 1upport very hilh volta1es at high field strengths across vacuum insulated 1aps has not been achieved, there are indications that thi1 may not be altolether out of the ques tion. It has, Cor example, proved possible to support about 1 mUUoa volts 111 across a 1 em gap between a 1/8" diameter positive sphere and a plane, where both electrodes were of 304 stainless steel and buffed to a good polish, and the -4 ambient pressure was p,.. 10 torr. Under these circumstances, the macro- 6 scopic electric field at the surface of the sphere was -7 x 10 V/cm. Other experiments have examined vacuum breakdown voltages V as a function of inter electrode gap d for different materials, for similar materials with different sur face finishes, for different organic and non-organic coatings on cathodes, for diffe ..: ent electrode geometries, etc. Still further experiments have investigated bushing modifications and improvements with the aim of evolving design criteria for high voltage equipment in vacuum. Besides the nominally 1 million-volt facility, a second facility at present rated at 400 kv, has recently become operational. This is termed the HIVE system because it will be used to examine breakdown phenomena in high vacuum at high electric field strengths. The system is shown in Figure 27. This facility will be used to determine the effect of various electrode materials, residual gases, etc. Data acquisition should be much faster with this smaller system. It is felt that the achievements to date under Contract 2166 indicate that continued research on vacuum breakdown phenomena is justified, and is likely to contribute substantially to the Directed Energy Weapon Program. In addition to the support of high voltages at high field strengths between vacuum insulated electrodes, one of the most important aspects of this program is the generation of high voltages in space. There are several likely DOD applications of accelerators in space where the technology for externally vacuum insulating would be of great utility for packaging for minimum mass. In this context the 78 79 inverted Van de Graaff generator • is relevant. This is a Van de Graaff generator built inside a voltage-graded accelerator column in a pressurized gaas, while the outside of the column is in vacuum. The investigation and devel opment of this type of generator is a logical step, based on the tllcl,nolog~· already developed under the present contract, and is discussed in more detail later. Other topics pertinent to the Program include energy storage at high volt ages, and the improvement of acceleration tubes beyond the present rating of 10 0. 5 million volts per foot. However, such research is also important to other projects, one of which is the use of electrostatic shielding to protect personnel and equipment 81 from harmful radiation in space. This concept has recently been examined for a pair of concentric spheres (the inner sphere being the shielded volume) in the environment of the inner Van Allen belt, where the radiation hazard is due principally to relatively low energy protons. U a large potential difference exists between the spheres, the inner sphere being positive, a large fraction of the incident proton flux may be prevented from reaching the inner sphere because of their interaction with the applied electric field. It is cJear that both present and future research on vacuum breakdown phenomena, and the generation of high voltages in space, is highly relevant to this program. This reaearch is also currently of importance in the design and development of high energy particle separators. In this apparatus, an incident beam of charged particles passes between and parallel to a pair of electrodes which may be as large as 3 ft. wide and 30 ft. long. The electric force on the particles is counteracted by the application of a magnetic field at right angles to the beam and the electric field, in such a way that selected particles are undevi ated, while the remainder are removed from the beam. In conclusion, it is apparent that a continuation of this research is directly relevant to the Directed Energy Weapon Program. Progress under the present Contract 2166 is such that it is likely that continued research at present and higher total voltages will greatly improve the state-of-the-art of the weapon, as well as being pertinent to other projects which include electrostatic shielding for space vehicles, and the selection of high energy nuclear particles. z 3. PROGRAM - CiOALS AND APPROACH z. 3. 1 Cioals The characteristics of the facilities for research on high voltage phenomena and the development of high voltage components for the space environ ment are obviously determined by the required progr•ms. The prime aims of these high voltage vacuum research and development programs may be expressed in the following goals;. the first group containing scientific and engineering ground work and the second group, application of this groundwork to the invention and development of new de\ices for use in space. I. Fundamental Investigations a) Theory - Explanations for the complete vacuum breakdown mechanism, both through the volume of vacuum and along the interface (surface) between vacuum and a solid insulator. b) Materials - Conductors for support of intense electric fields in vacuum. Conductors for support of large total voltages across single gaps in vacuum. Conductive anode and cathode materials for use in the above situations when polarities are fixed. Insulators for support of high voltages or intense fields and which, at the same time, may serve as high strength structural members in vacuum. Insulators having high tensile strength and ductility combined with low outgassing rates in vacuum. c) Formulae and techniques - Required for the design of higt. voltage vacuum insulated apparatus and for the optimization of such devices as regards power-to-mass and power-to volume ratios. IL Application to Space-Borne Mechanisms a) Particle accelerator tubes for the generation of high current beams with very large total energy. b) Electric aenerators or converters for producing high voltages from mechanical or low voltage electrical power. c) Enerp storaae devices capable of rapid discharae at hilh voltal•• A perfect understanding of electrical breakdown phenomena in vacuum would open the way for rapid development of new electrode and insulator materials as well as the creation of engineering design formulae. Therefore, it is felt that the early part of this program must lean heavily towa i fundamental research which can unveil this theoretical understanding. Later, when carefully controlled experiments ha"e been performed at the multi-million volt level, and results have become reliable and repeatable, a materials study program should be initiated to produce engineering data on the electrode or insulator characteristics of a large variety of structurally useful materials. With the experience gained from testing a wide range of materials, new formulae may be established for design of either space-borne or terrestrial vacuum insulated equipment. The applications of high voltage-vacuum research would, of course, be the final goals. Briefly, the technology is required for the construction of particle accelerator tubes in space capable of conducting high current beams and linearly extrapolative in length to any total voltage. In parallel with this, a high voltage generator design also capable of linear extrapolation to any total voltage is desired. For pulsed beam operation, an energy storage device is required - also in parallel with the accelerator tube and designed for extension to high volt- ages. 3. 2. 2 Philosophy of Approach In planning the future direction of effort on vacuum insulation at the million-volt level, one might review the progress which has been n'l&de at IPC to date. A rundown of notable findings is as follows: 1) Granberg's Relationship for plane parallel-uJliforrn = field eleitrodes (VE = C from which V kd"i may be derived) has been fow1d to hold to 1. 7 MV. Previous experiments stopped at 0. 7 MV. = = = *In the expressions, V voltage, E electric field, d electrode separations, C and k are constants. .l Z) k (in V = kd.a) is an inverse function of the area of either electrode. Between two ZO centimeter diameter 5 electrodes, Kmax is typically 3 x 10 V-cm·O. 5. If one electrode is changed to an 0. 3 centimeter sphere, k may be as high as 106 V-cm·O. 5. max a) As determined by sphere-to-plane experiments. b) Strongly indicated by biased grid experiments in which 15o/o area grid --properly biased -- raised k by 50 to 80%. 3) k is not noticeably affected by electrode finish if V > o. 5 MV. = 4) Granberg's Law (VE C) does not hold for asymmetric field configurations (sphere-to-plane). 5) Voltage of the total system is strongly affected by residual pressure, e. g. -4 V at 5 x 10 torr ,.,. 50 to 1 OO% higher than -7 V at 2 x 10 torr. The location of this effect -- in the vacuum gap (volume insulation) or along the bushings (surface insulation) has not been determined. (The phenomenon was first noticed at CERN, then also at Livermore, California.) 6) Vacuum gap insulation is virtually destroyed by exposure to a copious source of charged particles (such as those produced by an ion vacuum gauge). 7) Gross electron field emission does not occur until macro scopic fields exceed approximately 1 o6 volts per em. a) Determined by computation of fields in small sphere-to-plane geometries. b) Also determined by experiments with a series of disc-shaped field intensification electrodes. 116 In Fig. 28, one may see the importance of the Granberg Relationship wherein, for any gap and electrode configuration, voltage obtainable is directly proportional to k value. In the list of experimental finding• we have two strong handles on k; namely, area effects and pressure effects. It therefore seems in order to concentrate on these phenomena until they are understood. Area effects should be studied by further and more exacting experi ments with grids and electrical biasing. It should not be too difficult to establish a direct relationship between k and electrode area, when all other system param eters are maintained constant. Pressure effects should probably be approached from the "particle" point of .riew. In the vacuum gap, one would expect to find many types of par ticles: for instance, neutrals of many different species, and ions --also in vari ous atomic numbers, of both polarities, and at different energy levels. Experi ments must definitely be established to study these particles. The effects of external circuitry (inductive - capacitive - resistive - rectifier networks) on vacuum insulation should also be examined, with special attention being given to changes in electrode conditioning. Examination of the results from this work would almost surely result in improved theoretical under standing of the problem. The appro~ch to this fundamental program containing rather many experiments is best solved by having several complete and sepamte facilities differing mainly in the voltage range covered. The range at present is from 4 6 7 about 1 0 to 1. 7 x 1 0 volts. Extension of this to 10 volts would be extrem~ly useful,but, unfortunately, technical problems as well as expense will more than 6 o likely limit the voltage of the next facility step to 4 x 1 volts. An advantage of several complete systems, each coverins a different voltage range, is that experiments performed in one chamber may be checked in another chamber, thereby changing a sreat.many wall and circuitry effects which supposedly do not affect tl. • data. U chanses in data are noted, their orisin may 117 TypioGI Volume Insulation by Cronberg Rekltionship V • Kd+O.S 1 ,of> _,. ... o"' .a. \<' f» "'' 'f) ~'f. ;- .J1! TyPiCal SUrface Insulation if Voltage Groding is Used V•Kzd+I.O to Pbint where Volume Breakdown overrules. ( e's are Best Volues) Gap-d (em) FIQ. 28 Voltage-Gop Relationships for Volume and Surface Insulation in Vacuum be traced down and eliminated. Another advantage of several systems with differ ent voltage ranges is that a small chamber is more suited to repetitive testing (such as materials testing) where setup time becomes important, while a high voltage and, therefore, large system is more suited to the heavily instrumented, tedious experiment (such as studying charge-to-mass ratio of particles in the gap) in which internal mechanisms may be left in place for long periods. The applications section of the program may best be handled accord ing to the needs of the device. Power source ·development, for example, may be carried on progressively from the smallest chamber to the largest, starting with an 0. 5 MV design to fit the HIVE (Fig. Z9) and moving to the larger systems as voltages demand. Development of accelerator tubes, energy storage devices, and so forth may be carried on in a similar fashion. The two facilities in opera tion at present, as well as those planned for the future, are equipped with ports, flanges, and vacuum systems of sufficient size, number, and pumping capacity to accommodate the testing of these high voltage devices, with the limit that the design voltage of the device tested may be no greater than the rated voltage of the facility. 3. 3 DEVELOPMENT AND DESCRIPTION OF EFFORT Hip 3. 3. 1 Systems for Developing Voltage in Vacuum Apparatus for the study of phenomena at the million-volt level in vacuum has, until recently, been nonexistent except for the special case of the voltaae-graded accelerator tube. The closest approach prior to 196Z was proba bly the experiment by J. Ci. Trump and R. J. Van de Ciraaff who, in 1947, reached 0. 7 MV in a vacuum chamber of useful dimensions. Because of the dif ficulties which still frustrate research in this field, an ov•all review bas been made. I 119 Resistor Voltaoe Oractlno Control Unit - N 0 High vacuum Prototype I to be tested in HIVE Fig. 29 INVERTED HIGH VOLTAGE GENERATOR The possible methods for producing high voltages in vacuum may be classified as follows: I. Conventional voltage sources A. Externally located: High voltage is generated outside of the vacuum chamber by conventional means (e. g. cascaded transformer-rectifier, RF pumped capacitor rectifier, charge-carrying drum or belt) and conducted into the vacuum with special feedthrough bushings. B. I~ternally located: High voltage is produced inside the vacuum chamber by special generators designed to fit and operate within the insulated columns which sup port the terminals. The generating mechanism cou,ld operate by any of the principles presently used for external generation of high voltage as listed in A. II. Direct particle charging A. External source: Beams of high energy charged par ticles (positive or negative) are generated by conven tional means (e. g. cyclic accelerators, direct poten tial drop accelerators), conducted through the wall of the vacuum chamber, and collected in terminals sup ported on special standoff insulator columns. B. Internal source: Alternatively, on the insulated ter minal one may place a radioactive isotope which, through natural decay, creates high energy charged particles. These particles emanate from the ter minal and collect on the vacuum chamber wall, thereby constituting a charging current. Maximum terminal volta1e in either II. A or II. B can approach, but not exceed, the maximum particle ener1y in Mev. Surprisingly, all four of the.e methods have been tried and each has been rather successful. A detailed discussion of each follows. lZl Conventional Voltage Sources, externally located. This method of obtaining high voltage in vacuum has been the most popular, and is probably the method most likely to succeed if the voltage required is not beyond the capabili ties of the bushings. The state of the art, as of January, 1963, for conventional high voltage generators and for appropriate vacuum chambers is, at present, some four times better than the art of bushings. High voltage generators operating on the belt charging principle are commercially available which will yield 5. 5 million volts either positive or nega tive. The high voltage components of these generators are insulated with com pressed gas. Therefore, two of these generators with opposing polarity will yield a potential difference of 11 million volts, but located in a compressed gas environ ment. It must be further noted that the voltage limit of such generators is nor mally set by the permissible electric field in the self-contained vacuum-insulated particle accelerator tube. These tubes, until the recent development of the inclined field principle, have been limited to a maximum gradient of about 0~ 5 million volts per foot. If the accelerator tube is allowed to fill with insulating gas, or else removed entirely, the typical Van de Graaff generator will attain a somewhat higher voltage. With a higher insulating gas pressure and other refine ments, one can attain 1. 0 MV. per foot of column length. With conservative opera tion, however, one should be able to realize a SO% voltage increase to 8. 25 mil lion volts each. This would indicate a maximum potential difference of 16. 5 mil. lion volts attainable in the gas environment. Aside from their high voltage-producing ability, belt-charged aen erators are rather current-limited. At present, the highest current model pro duces approximately Z milliamperes. Other conventional generators are com mercially available which can produce currents of 10 to 30 milliamperes in the ranae of 1. 0 to 4. 0 MV. The Cockcroft-Walton, which operates on the cascaded (series pumped) voltaae doubler-rectifier principle, can reach 4. 0 MV and ZO ma. The Dynamitron, which operates on the parallel capacitively pumped capacitor rectifier principle caa produce 3. 0 MV at 10 to ZO rna. The llleulatinl Core 1Z2 :' I Transformer (ICT), which employs the parallel masnetically pumped transformer rectifier principle, can yield 3. 0 MV at 30 ma with present desisns. Therefore, it may be seen that several methods are available for reachins the ·multi-million volt level with currents of around 20 ma. A vacuum system to contain million-volt potentials must have mini mum interior dimensions in feet of approximately the intended megavolts squared. (See Section 3. 3. 3. ) This allows reasonable values between terminals and from terminals to walls. The attainable vacuum in the chamber must be better than -4 2 x 10 torr to prevent the occurrence of Townsend type (glow) discharges across high voltage gaps (beginning of the discharge range defined by Paschen's Law). Vacuum systems have been built with spherical chambers 38. 5 feet 9 in diameter and capable of pressures as low as 5 x 10- torr. The pressure level is therefore better than the required minimum and t!1e diameter is suitable for a rating of 6. 3 million volts. This voltage rating is for an open gap double-terminal configuration such as is sketched in ~ig. 34 *, and assumes a k = 3 x 1 o5 V -em -O. 5 in the working gap and k = 2. 4 x 1 o 5 V -em -O. 5 maxima in other areas of the chamber. By addition of potential dividing surfaces such as is shown in Fig. 34, the total voltage may be raised by the square root of two, or to about 8. 8 million volts. However, this voltage limit would orily be attainable when electrodes were 5 -0. 5 . such as to allow a k of 4. 3 x 10 V -em 1n the working gap. In addition to a power supply and a vacuum system, the third essen tial to the vacuum insulation facility is a means for carrying the high voltage into the vacuum chamber. This implies the u~e of a !eedthrough bushing. The origi nal effort by IPC (Contract AF08(635)-ZI66) for obtaining one million volts in high vacuum was, of course, mainly oriented around the development of bushings, since other elements in the system could be readily enpneered to the specifica tion. A prototype bushing (Type I) was desiped and following approximately six months and several hundred hours of developmental work, it could attain 0. 8 MV in..!!!J!! -5 -4 vacuum (p < 10 torr) and l.Z MV in optimum vacp~ (p ... 5a10 torr). l *see ra1e I 118 Ul \ The data obtained from tests of the Type I bushing (Fig. Z6 ) were used in an improved design (Type II), which is now ready for electrical tests. Photographs of the completed Type I and Type II bushings are shown in Figs. Z6 and 35.* Although the Type II bushing has not yet been voltage-tested, the construction features are such that its performance should be slightly better than the Type I. Therefore, a predicted performance in the present vacuum chamber is 0. 8 mil lion volts in high yacuum and 1. Z million volts at optimum vacuum. These values are expected with either polarity, which was not possible with the Type I. It is evident that this general design for a high voltage-vacuum research facility (two gas insulated power sources of opposite polarity, bushings to feed the high potentials through from pressurized gas to vacuum, and a central vacuum chamber for containing the experiments) depends entirely upon the continued development of bushings. Since the present bushings operate at only 0. 8 MV in high vacuum, reaching the :1: S MV level implies a factor of six improvement in performance. A linear extrapolation of the vacuum insulation column of the type I and II bushings indicates that 1. Z MV per foot surface insulation should be pos sible over a length of three or four feet in optimum vacuum (1. Z MV over one foot has been attained). If additional effort is put on research, it is felt that approximately 1. 0 MV per foot could be obtained in high vacuum and extrapolated to a length of several feet (0. 8 MV over one foot has been obtained). These extrapolations are made along a curve of slope 1 as shown in Fig. Z8. Since o. the typical vacuum gap breakdown curve has a slope of S, it is necessary that the volume and surface insulation curves intersect at some point as shown in s . 5 -0. the same fiaure. For instance, at 5 MV, a Cranbera k o f 4 x 10 V -em u required to have volume breakdown equal voltaae araded surface breakdown. Therefore, one must conclude that around 5 MV or less, a chanae in slope will occur in the surface insulation curve and above that point, insulator length would have to be increased with the square of the increase in voltaae. Alternatively, *See paae f 139 124 potential dividina sur£ace1 could be uud at intervals of Z or 3 million volts alona • the insulator column. Potential dividers are shown in Fia. 33. Unfortunately, the use of potential dividing surfaces is detrimental to vacuum performance because of the large increase in surface area and associated outgassing. Further more, the larae surface area of the dividers must be electrically conditioned just as any other electrode in vacuum. The electrical effect of this would be a reduc tion in the workinJ value of k for every increase in area. The internal electrical design of a. S MV bushing where the high volt age conductor passes through the around plane appears to present a more difficult problem than the insulator column design. The electrical insulation which is required to take the high radial stress in this repon may be either a solid, a liquid, or a compressed gas. Solid or gaseous insulations are preferred because they are more compatible with maintenance of~ vacuum systems. Gaseous insulations are also preferred over solid because of the self-healina properties of a gas following a breakdown. Furthermore, electrodes in a compressed gas will undergo "positive conditioning" whereas electrodes separated by a solid insulator generally exhibit a "negative conditioning" or loss of voltage-holding ability after each breakdown. The total voltage at the feedthrough may be supported across a sinale radi.a l gap. Alternatively, the gap may be split up by the use of potential dividing surfaces, thereby forcing the field distribution. This technique is advantageous ___ where breakdown is strongly~ dependent (the usual case in compressed 1as insulation) or where breakdown is stron1ly total volt&...J;;.e.. _d ependent (exhibited typically in high vacuum). Where breakdown is initiated at a maximum field E , the maximum m voltage (V ) which can be supported using interpotential shields at optimum radii m can be shown to be 116ee pa1e I 13 7 125 where Rz i8 the outside diameter and k is the number of shells (including the outside). There is little point in using more than two interpotential shields = (k 3) as V is then 95% of the maximum pouible. There is insufficient infor m mation in the megavolt range on solid breakdown and also very little on high pres- 84 sure gas breakdown. Data supplied by Philp (private communication) has been * used to calculate the curves on Fig. 36; there it can be seen that a suitable coaxial system with a lZ" diameter and pressurized at 300 psi of SF can with 6 stand 5 MV. It then remains to develop a glass system which can withstand such pressures at that diameter. A two-phase material approach (glass fibre-glass) looks promising, but may require considerable experimentation to develop a surface-flashover strength equal to that presently attained with vitreous Pyrex glass. Potential dividing surfaces may also be advantageous where the voltage which can be supported is a function of gap to a power & , and a is less than one as follows: v = kd 6 Ifa.>l Potential dividing reduces total V. max a.= 1 Potential dividing has no effect. a.< 1 Potential dividing increases total V. This relationship is not true in all cases since electrode area effects must also be considered. For instance, adding a single potential divider shield doubles the active electrode area and this area increase may reduce k such that the total voltage attainable is lessened. The optimum design of the bushing feedthrough is therefore a rather difficult one, and any design for voltages greater than the present I. Z MV will have to be followed closely by experimental development. However, our present technical "know-how" seems quite sufficient for the development of bushings to z. at least the S or 3. 0 MV level. * See page I 1~0 ,- The use of conventional voltage sources, externally located, for a very high voltage vacuum facility design therefore hinges entirely on bushing development. Further development is practical up to 3 MV or so. Beyond that point, feedthrough problems become severe, and especially when one considers that the total fat.~ity stored capacitive energy can be released across the short radial gap in this bushing if the insulation fails at that point. Stored energy is, of course, proportional to the system capacitance times the square of the system voltage. Because of the voltage limit which can be foreseen, this approach to high voltage vacuum facilities does not present the possibility of extensive advance ment in voltage capability. One further point on this type of facility design is that where very high currents must be obtained in vacuum, this approach is necessary. The ext~rnal supply can, of course, be increased to almost any required size and power without overloading the bushing current limitation. In our present appli cation, however, such high currents are not needed. Conventional Voltage Sources, internally located. The inverted power generator, located inside the vacuum chamber, achieves a large reduction in overall space required by the facility as may be seen in Fig. 34 for a typical design. The generators in this facility are located almost entirely within the bounds of the 16 foot diameter spherical vacuum chamber. 1f external generators were used, they would add some 8 feet each or 16 feet total to the height of the facility in order to achieve the same total system voltage. The idea for inverting the Van de Ciraaff generator by putting the belt inside a pressurized column and having the dome and outside of the column in vacuum is not new. However, it has never been developed at the multi-million volt level. A recent desip (Fig. 30) by Leo Jedynak at MIT has generated 0. 55 million volta at 30 microampa. The column is 5j-" diameter by 12" long and fits in a ZO" dia. x Z8" tall vacuum chamber. These overall dimensions of a one-half million volt aenerator indicate the capabilities of an inverted machine. 127 Electrode ACUuetment Upper Electrode ( grounclecl to chamber thru current metera) Lower Electrode (o o.sa Viewing Part to MV positive MJp.A max. ourrent) Sepnted Insulating Alulllinum Cha11ber Column c20•dta. by 2a•ta11> (5f.dla. by t2•tong) VGOUUIII Pwnp SF -lneuiatlng Gal (&Opel) .I 30 Conventional Voltage Source, Internally Locoted F~g. Vac..- tnsulotion Reaearch Apparatus by Leo Jedynak ' ! 1Z8 Further atrong reaaona for wbhing to develop the inverted generator are connec~ed with problema of feeding high voltage into vacuum U a atandard generator ia employed. Aa already noted, the high voltage preuure-to-vacuum feedthrough buahing concept can be developed to around 3 million volta with reasonable effort, but the same effort could make rapid advances in the inverted generator desian. The main baais for choice between atandard generators plua buahings and inverted generators is in power or current required. The present 9 o3 Type I-II million-volt bushing is capable of conducting aome 10 watts (1 amps 6 at 10 volts), while an inverted power generator of the same column size could 3 -3 6 develop 10 watts (10 amps at 10 volta). The choice therefore seems to lie at the 1 o3 watt level for a 1 million-volt machine. The current which will be required for high voltage-vacuum research is expected to be small --probably less than 100 microamperes for vacuum break down studies at the 4 Mv level.(Experiments at the I. 5 Mv level generally require only I 0 microampere-.) Capability of the inverted power generators is expected to be 1000 microamperes at 4 Mv or at least one magnitude greater than predicted needs. The belt generator is only one of many methods for producing high v:>ltage which may also be inverted. Other possibilities include the gas insulated Cockcroft-Walton, Dynamitron, and the variable capacitance electrostatic gen erator. The Insulating Core Transformer would not be ao suitable because it ia less amenable to packaging in a small diameter column. The electrostatic sen erator may also be vacuum insulated, thereby offerinslighter weight conatruc tion. Vacuum insulation is a desired characteristic, in any case, for a syatem which must operate in space. Direct Particle Charging, external source. Electron beam charsms of an insulated terminal in vacuum has been experimented with at Arsonne National Laboratory by De Oeeter. Potentials up to o. 3 MV were obtained in a small vacuum chamber with maximum voltase limited by breakdown to the walls of the chamber. 129 At 0. Z MV, electron current• of 100 microampere. could be collected with 80 to 90tfo efficiency. Further experiment• in electron beam charging are planned at Argonne in connection with the deaign for a croased field velocity selector. Direct Particle Chargina, internal aource. Charaing of an insulated 90 90 terminal in a vacuum system by means of a radioactive beta source (Sr -Y mixture) has been investigated by J. W. Kennedy, et al, at Washington University, St. Louie. Potentials of approximately 0. 3 MV were reached, at which point prebreakdown currents were equal to the charging current (a one millicurie beta source emits a current of S. 92 micromicroamperes). Experiments with direct particle or beam charging have certainly proved the high voltage capability of such techniques, especially when one con siders that breakdown or prebreakdown currents to the chamber walls limited the peak voltage. Our intent, however, is to obtain a facility in which experi ments will not be adversely affected by any stray particles other than the mainly neutral ones which are outgassing from the walls and being steadily removed by the vacuum pump. The particle charging system, therefore, must be discounted since a very small percentage of ions being added to the chamber residual gas z. could completely upset most high voltage experiments (see Section 3. Z, Item 6). 3. 3. z Chosen Method The facility concept which appears to offer the highest prospects for extrapolation to very high total voltages is the conventional voltage source, internally located design. The only problems which are foreseen with this approach are that the vacuum chamber must be unreaaonably large for rather low total workina voltaae• (4 to 10 MV). However, the vacuum chamber size is determined atrictly by ~~voltage and ia unaffected by the method for pro ducina or introducina thla voltaae. Therefore, the chamber size problem would be preaent no matter which concept ia uaed. 130 The main advantage of internally located power generators is that increases in voltage can probably be made by linearly increasing the length of the generator (assuming it is a belt charged machine). This increase can con tinue until volume voltage breakdown occurs from the generator dome to the chamber wall or to the opposing terminal. The approach is also very attractive because it is that required for the development of accelerators in space. making the optimum use of the natural environment. Investigation of designs for an inverted Van de Graaff by IPC indicates the feasibility of one million volts per foot of column length and 1 milliampere of current in a 9" outside diameter -- exclusive of vacuum insulation to the walls. Proflotype designs of an inverted supply could be developed, for example based on existing IPC facilities, in the following order: 1) A working mockup -- air insulated. Z) An o. 5 million-volt generator to fit the HIVE, gas insulated. 3) U 1) is succesdul, a second generator of the same size but with incorporation of further ideas. "' 4) A 1. 5 million-volt generator to fit MiV, using results from Z) combined with Type II bushing column. z. 5) A 5 nlillion volt generator to be tested in Mi V -Z (7' diameter chamber) but for use in the 4 Mv system if successful. 6) Ditto of 4) for opposite polarity sourc~. 7) Design of 5 Mv generator for test in 4 Mv system and use in 8 Mv system. "' Note: Step 4) may be omitted if the 71 diameter MiV-Z chamber is completed by that date. A sketch of an o. 5 Mv - o. Z rna prototype inverted Van de Graaff generator is shown in Fig. Z9. This machine would be designed to fit the HIVE as listed in Z) above. It is expected that all mechanical design difficulties of the inverted generator could be ironed out at the 0. 5 Mv level. 131 Deaip problema foreseen at the moment are mainly those connected with containment of insulatins aas at ZOO to 350 pound• per square inch to insu late the internal works of the column to a level of 1. 0 million volts per foot of lensth. A mechanical tension rod which is a sood electzoical insulator is there fore needed to hold the column in axial compression. Presently available insu lating materials are either brittle or their insulatins qualities are poor. Tanaen~ tial stresses in the column rings must be removed from the slass. This can probably be accomplished by increasing the number and strensth of metal rins•• and by adjusting their modulus of elasticity such that they carry the major por tion of the radial bursting pressure. Other problems in the inverted design will be mainly those of pack asing. Their solution, however, is expected to become evident during construe~ tion of the initial mockup and 0. 5 Mv machine. The chosen approach is, therefore, to pursue the development of o. z. inverted high voltase power senerators from the 5 to the 5 Mv level, at z. which point a suitable vacuum chamber would be constructed to house two 5 Mv supplies and thereby yield 4 to 5 Mv total. Further facility growth will be covered in Section 3. 4. 3. 3. 3 Review of Existina and Proposed Facilities Vacuum breakdown research at the present under AF08(635)~Zl66 is carried out in two facilities: HIVE: a Z!' diameter spheroidal chamber with capability o. of approximately 4 million volta. (Fia. 27) MiV: a Z!' diameter by 4' lenJth cylindrical chamber with voltaae capability of approximately 1. 5 million. (Fig. 25) These facilities, in particular the MiV, are, by a considerable mea sure, the moat advanced vacuum insulation teat facilities for the very hiJh volt age range of which we are aware. It is natural, then, that advances to hiJher potential• ahoulcl be baaed 011 theae facilitiea, &lUi that ia the philoaophy which haa bea adopted. 132 ',. . ' . ' Studies under AF08(63S)-Z795 have developed concepta for extenaiona in potential to 4 and 8 million volta in two additional facllitiea of 16 and 64 foot diameters, n. hereafter designated aa Phase I and Phaae In addition, it ia propo1ed to under• take certain improvements and modifications to the two exlating ayatema to bring their experimental performance up to the level which has been calculated for n. Phases I and A basic criterion for experimental value of a vacuum breakdown facility has been determined empirically through experiments in the 1 million-volt system. This criterion is that the minimum chamber dimensions for testing of typical elec- z trode materials up to the full system voltage must be D • (MV) feet. For larger chambers, the optimum geometry ia a sphere and permits experiments with elec trodes of diameter D/Z at maximum separations of D/Z at which condition system design voltages could be attained with a Cranberg value of k S; 3 x 10 5 volt-em .o. 5 in the gap. Application of this criterion to this same 1 MV facility reveals that . the Z!• diameter x 41 long vacuum chamber places a aerioua limit on maximum chamber voltage. By replacing the vacuum chamber with a 7. 01 diameter aphere 1 2 2 (1.3 MV V.d.G. supply maximum x Z) feet • (2.6) • 7.0 z. it should be possible to conduct "xperimenta up to the 6 MV level with this facility. A similar calculation for the HIVE facility with 1. 881 minimum dimen sion indicates that ita capability should be I. 37 MV. Since the preaent bushings for this system are not capable of 1. 37/Z MV. small inverted 0. 5 MV power z. supplies could be applied to this chamber to yield a factor of 5 improvement in working voltaae. The vacuum breakdown facilities would then stand as listed in the. following table. 133 Original Maximum Nominal Chamber Ultimate Operational Facility Potential Rating Min. Dim. Vacuum Date -8 HIVE 1. 37 MV 1.0 MV 1. aa• lxlO torr January 163 MiV-2 2. 6 2. 0 7. 0 8xlO.a March '62 (MiV ·1) 9 Phase I 4. 0 4. 0 16.0 5xlo· October '64 proposed 9 Phase II a. o a. o 64.0 lxlo· March '65 proposed HIVE Facility HIVE, standing for High Vacuum, High Field, is a small, versatile research facility of advanced design (Fig. 31 ). It incorporates many features essential to high voltage as well as high vacuum which were discovered or proven on the 2 million-volt research apparatus, such as: high polish on interior sur faces, large radii on edges of windows and ports which are reached by electric fields, 304 stainless steel construction.(~ good electrode material), freedom from organics by mercury diffusion pumping, ceramic bushings, and gold metal gaskets. Electrodes up to 8" diameter may be changed by removing a window, while larger electrodes or multiple samples on a turntable may be set up by open ing the 30" flange. A fast pumping system permits one to reach 5 x 1 0 -S torr (sufficient for starting voltage tests) from atmosphere within 10 minutes. An -8 ultimate pressure of 1 x 1 0 torr is expected. Bu•hings on the HIVE are ~ IPC design. They operate from atmos phere to high vacuum and initial tests have shown good performance to 0. 21 MV in high vacuum. The bushing i8 of brazed ceramic construction, and is insulated internally by sulphur hexafluoride at 2 atmospheres. High voltage for the ~VE is obtained from two air-insulated Van de o. Graaffs which are capable of ZS MV and 100 microamperes each, and which include polarity revereal switches. 134 HV Source 0.25 Million Volta± Provision for Turntable for Multiple Tests Electrodes · GGps to I' 2 Windows 120° Apart -e" Mercury Diffusion Pllnp -17cfm Fore Pump HV Source + 0.25 Million Volt construction : Stainless st•l ThroughoUt Gold Galketl - - - Fig.31 Q4 MILLION VOLT HIVE HIGH VACWM -HIGH FIELD FACILITY 135 1,36 .. '" "~' ....... ~ . ' .".t a.: ~ t . ~ ·\ \\ ~ . . /I \ i \ \ \ I / ~ Q ~ - c3 I ~ I I >. > I I 2: .~ ~ \ ~ ~\ .. \ ~ · .. ·:·· ' I ~ ' t " •.·. .. · I ..... .... . ··:'. " ..... '... "... ~· "'": ~J''.~7'i.~''F"'~-n;""·'· ~-: ."'"'~-:~ '?~I!II!~Lifl 5 a VOCWIII Cl'lclllller (Stainllss SIMI lntlmcllly Palilllld) Potlntiol Dividing SUrtace 8.5' DiO Sphere I All Joints Gald ~) Hilll \laltOII IMtlinOs TWPe II DniCJI _..... Troclc for (')~ ., ·,;. :• --Clolinl t: ~ PU111P Down Tillie - Afllcleplllnt to 1cr• in 30 llin. . \ F~ 33 2 MILLION VOLT FACILITY -WITH PROPOSED MODIFICATIONS \ ~ II &a. .ri :; •I f :::; ! ~ i I It I :::) ~ ~,! ID ~ !II I ~ w u~;: :J 1- >- IH cnc )JI II :> ~~~ '·I !I I ,.J ·~ a h~ .5 I § ~ 139 ~ ..... ~ ""' II' , f:- ;:. c .... Q. ..... ..... ~ rJ ...c: ~ 0 ) ( t f f 8.0 f 5.0 > 4.0 z . 7 • • 10 II 12 OUtllde Diameter ( R XI) -lnotlea 1 , ..... Vottacle WNah - ... 8uppor1lcl With CyHndrtoal GeoNtry Uling IF • R il the NuMblr of Coo*~ SMIIa. (Field Dota E,., 1 Taken froM Philp.) 140 Because of the rapid set-up time pouible in this system, its first purpose in the vacuum breakdown program will be to study special electrode materials. Also, because of its organic free vacuum design (compared with vinyl acetate bonded bushings in the Z million-volt chamber) it will be used to study the effects of different gases (e. g. H , He, NZ.' Oz.• COZ.' SF ) at pres- 2 6 -S +3 sures between 10 and 10 torr. Mi V -Z Facility • Modifications proposed for the present Z million-volt facility (Mi V -1) would consist, mainly, of a new spherical vacuum chamber of 7 feet diameter. This large chamber would reduce radically the number of and effects of charged particles crossing the gap be.tween electrodes and the chamber walls. Electrodes of diameters to 3' could be accommodated in lieu of the 1 foot maximum in the present z.. 51 diameter system. (See Figs. 33 and 37) Should this vacuum chamber be obtained and used as an environment for the Type II bushings, their performance should be markedly improved. The insulator column of these voltage-araded bushings, on the vacuum side, may be considered as a series of electrodes -- each affected by other electrodes on either side of it and each affected by the proximity of the large area ground plane electrode --better known as the "tank!' It is the sum total of these electrode effects, paralleled by the vacuum-solid interface surface flashover effects, which act to limit bushing voltage. lt is these electrode surfaces, also,· which must undergo electrical conditioning each time that voltage is applied, as indi cated by a gradual increase in peak voltaae as a function of time. ln the present million-volt system, the chamber diameter is 76 centimeters (Z. 5 feet) and the bushina terminal diameter 30 centimeters. This Jives a minimum electrode aap o. of 2.3 centimeters and a nominal maximum of 94 million volts across this aap -o 5 5 if k • Z x 1 0 V -em • • Since k values much above this are unusual, it becomes quite clear that bushins terminal voltaae must be limited primarily by wall clear ance in this present system. Therefore, if further aains in bushina voltaae and, therefore total sy~tem voltal•• are to be macle, they will be piae4 moat rapidly ad laupeaalvely by the lAitallatlGD.ol. an ealai'IH vacuum chambu. * <••• n 1• 3z) 141 Other improvements to the desian would include a lar1e surface area condenser (liquid nitro1en cooled) inside the chamber, potential dividin1 surfaces to improve bushing performance, quick opening port for chan1e of electrodes, large rouJhinl pump for 1 S-30 minute pumpdown from atmosphere, "reduced weld" chamber design to avoid leaks, and a track for aliiJlment of chamber halves when openin1 and closing the system. Instrumentation improvements planned would include a new optical system for scanning and inspecting electrodes during operation, an RF telemeter ing system for monitoring current magnitudes and waveforms to the electrodes and for controlling bias voltages to grids with respect to electrode potential, and recording equipment for obtaining better data and partially processing it as it is • received (e. a. x-y recorder, analog computer, plots on loa-loa paper of V •kd • 4 MV Facility Results from a lona and rather detailed study of the vacuum break down program indicate that the most senaible increase in system energy per step is by factor• of two. The basic reason for this is that dimensions required for vacuum inaulation increase typically with the square of the voltaae. Therefore, this facility is proposed as a 16 foot diameter sphere with voltaae capability of twice the present system with 4 foot length tank. A spherical chamber design has been chosen from investigations of electric field plots made with varioue configurations euch as: long cylinder-flat ends, long cylinder-hemispherical ends, and sphere. A sphere is the only geometry which permits reaeonably large electrode areas with amall wall effects. (See Figs. 34 and 38) Special features of thie system have not been clearly established, except that they will include those prcaposed for Mi V -Z and will emphasize the ability for rapid changeover of experiments. In keeping with this, lar1e pumps are planned so that pumpdown time may be held to 30 minutes or less. Power sources will be of the inverted desip and will require a modeat z. development program to attain the required 5 million volt r&tinl for both polari ties. 142 ., ... ~ ' 8 MV Facility The general design of an 8 to 1 0 MV .facility would be identical with the predeacribed 4 MV .facility. However, most dimension• would be increased by a factor of four to account for the non-linearity of voltage-gap relationships. It is expected that inverted high voltage power generators could be extended to two or, at most, three times the length of those required for the 4 MV facility (5 to 7. 5 .feet for 8 MV facility). These generators would fit in the top and bot r tom of a sphere approximately 64 feet in diameter {8 MV)l • 64 feet • ] The construction of a vacuum system oJ-this size several years in the future {c. 1965) is not expected to create any great difficulty, in view of the • fact that a 38. 5 foot diameter chamber was completed during the summer of 1962 whidJ has given excellent vacuum performance. The main problem with this large vacuum system may be connected with obtaining large inorganic fluid vapor diffusion pumps. Most large di.ffuaion pumps used today are oil diffusion which will eventually contaminate the aystem with oil. High voltage syatems give poor performance when contaminated with organics, aa displayed by a loss in maxi mum voltage and an increase in x-radiation levels. Therefore, moat high voltage-vacuum insulated device• employ mercury diffusion pumping, with the preeent size limit being 24" diameter compared with 48" or more for oil diffu eion pumpe. We may therefore say that, aside from the expense of a large vacuum eyatem, and the possibility of having to make one'• own vacuum pumps, there appears to be no insurmountable problema involved in the conetruction of an 8 to 10 MV vacuum reeearch facility. • Coutructecl by F. J. Stolte• Corpora&iOG at Kial of Prua• alae PeDuylvaDla for ue by Oeaeral Electric Compaay Space TechaoloiJ Cater. 143 3. 3. 4 Allocation of Studies to Existing and Proposed Facilities It is the purpose of this section to examine the allocation of studies to the existing and proposed facilities for research on vacuum breakdown phenomena. These facilities are: Present After Voltage Rating Modification (a) HIVE System -high vacuum and 0.4 MV up to 1. 0 MV high electric field • (b) MiV System -million-volt system 1. 7 MV up to 2. 5 MV and the proposed facilities: (c) 4 MiV System - 4 million-volt system 4.0 MV (d) 8 Mi V System - 8 million-volt system 8.0 MV (a) HIVE System (see Fig. 27) This system, which has recently become operational, is intended to complement the existing million-volt system. By virtue of its design and voltage rating, HIVE is well suited to a comprehensive study of electrode materials, electrode coatings, and states of finish of electrode surfaces. This is because interelectrode gaps at 400 kv will be about 1 em: thus plane parallel electrodes about 10 em in diameter can be used to produce uniform electric fields in the gap. It is then much more economical to carry out a series of tests with electrodes of this size, rather than those of 25 em diameter normally used in the million-volt facility. Furthermore, the time which elapses whenever electrodes are changed, and the conditioning time for bushings and electrodes, will be shorter than for the other facility. It is intended that promising electrode materials, coatings and fin ishes which arise from research in the HIVE system will be examined at hlaher voltaae levela in the million-volt facUlty. However, atudiea in the HIVE ayatem can be of a fundamental as well as applied nature, since the system is instru mented to examine pre-breakdown currents between the electrodes and Fowler Nordheim plots can be made: these may be of extreme significance in understand- 82. ing the phenomena of breakdown. It is also planned to investigate the effect of ambient pressure and the nature of the ambient gas on the breakdown voltage, and to examine the influence of external circuih y on this voltage and on the waveform of interelectrode discharges. The high voltage feed-through bushings on the HIVE system can be adapted to allow the introduction of liquid nitrogen to a volume adjacent to the electrodes. These may be cooled by conduction to a low temperature, so that it will be possible to observe the effects of reduced electrode temperature on breakdown. Finally, it is hoped to use this facility to measure the ratio of charge to-mass of ions which cross the electrode gap. Such particles would pass through an aperture in an electrode and out into a separately pumped chamber, where they would be deflected and analyzed by a magnetic field as in a mass spectrom eter. This experiment may resolve the nature of Cranberg-type clumps which have a postulated role in breakdown. (b) MiV System (see Fig. 32.) This system has been operational for nearly a year, and during that period it has been used for experiments of two types. In the first place, the design of the pressure-to-vacuum feed-through bushings was new, so that experi ments were made to explore the behavior of the bushings and their voltage sup port capability: this was done by using a variety of field rings, terminations and urethane resistors with the bushings, to determine an optimum geometry. Th<: second type of experiment has been measurement of the breakdown voltage V as a function of electrode separation d at high voltages, for plane parallel elec trodes of various materials, for coated electrodes and for electrodes having sphere-to-plane aeometry. 145 It appears that for larse electrode separations, the proximity of the vacuum chamber wall8 may be limitina the voltages which the bushings can sup- port. Replacement of the existing vacuum chamber with a spherical chamber has been discussed in a previous section, but the anticipated experimental pro aram will be the same, whether this modification is accepted or not. With particular emphasis on the Directed Energy Weapon Program, breakdown tests to determine V as a function of d in hard vacuum will be con tinued for essentially plane parallel, polished· and coated electrodes of ZO em 83 diameter or larger. It has been suaaested by Van de Ciraaff that voltage- dividina shields surrounding the interelectrode gap would help to provide uniform field conditions between the electrodes, and would render the interelectrode gap less sensitive to bushing inconsistencies: these suagestions will be adopted and results compared with those for electrodes without benefit of shields. Electrode parameters will be determined principally from information gained in HIVE experiments. In addition, this facility is equipped for tests with heated elec trodes (in contrast to the HIVE system). for the investigation of breakdown between electrodes having back-biased grids over their surfaces, and for the testing of an initial model of an inverted Van de Ciraaff generator. However, it is probable that research into the fundamentals of vacuum breakdown will be as important to the Directed Energy Weapon Proaram, and the million-volt system is suitable for this. Such an investisation includes experi ments with uniform, shielded saps described above, with instrumentation to determine pre-breakdown currents and the current density distribution over the electrode surface, and the use of phosphor-covered electrodes for visual obser vation. The contribution of photoelectric, thermionic and field emission to breakdown must be considered, and can be examined experimentally in the million-volt system. It is also planned to observe the effect of introducina radio-active isotopes to the proximity of electrodes, since this technique allows the bombardment of electrode surfaces by controlled amounts of ions, electron• or hip eaeray photou. Finally, those fu.~dam ental studies which are made with the HIVE system -- investigation of th ~ambient preuure effect, etc. •• can be performed in the million-volt system at higher voltage• and larger electrode separations. (c) 4 Million Volt System (see Fig. 37 ) It i• envisaged th.at fabrication of this system, and development of the power supplies for it, wiLl take lZ months. On completion of this phase, it is anticipated that an examination of breakdown voltages as a function of inter electrode distance ,.,.ill be rna de for promising electrode materials and coatings, with the object of ascertainin& whether an immediate breakthrough in the support of high voltages in vacuum is possible. Experimental and theoretical studies in cc. 1ection with the HIVE and MiV systems will be used in determining the nature of these experiments, but it is expected that they will follow closely the type of experiments already performed in these chambers at lower total voltages. (d) 8 Million Volt System (see Fig. 38 ) The initial experi.r.2lental program to be carried out in this system will be the same as that forth~ 4 MiV system outlined above. It is felt that a study at these voltage levels, -,.hich emphasizes both the fundamental and applied aspects of the problem, is the best approach to the high voltage insulation and generation problema posed by -the Directed Energy Weapon Program. Full inter change of experimental result~ between the systems is planned, although it is expected that information will .,rimarily follow the directions shown below. MiV SYSTEM -- £ . _jJ ··,- :;~:- A.. . • . .; .,_..-: g_ _ \\..---. \ ~ e . ·, ' '" Z.S MV VACUUM INSULlTION FACILITY- FIGUf!E ~1 ION ftMVSICS CORPORATION SUtlLI" .. C:TON, MASS. _f:F~ II.' AIC Y-r1.6._: ;;· •-t.IL.- 1119 I ' I " I -. • I •..• •. • .I . • •·. ~ ... ~ ... Ill 'C lll "'t ."...'. It a.. = • t .. " tit ~ G ... 3. 4 GROWTH PLAN FOR FACILITIES TO SOLVE THE PROBLEMS OF HIGH VOLTAGE INSULATION IN SPACE A proper treatment of the high voltage vacuum insulation problem would create physical equations that could be used for solution of any voltage-cur rent-gap-geometry situation. Such equations could conceivably allow calculation of material performance (Cranberg k values) from data such as atomic constitu ents, crystalline structure, electrical and thermal conductivity, specific heat, yield strength vs. temperature curves, ductility, work function, etc. The back ground for such formulae must therefore contain data from a wide voltage and gap range and also a variety of geometrical configurations. The basic idea of a par ticulate beam directed energy weapon implies high energy beams and therefore high total voltage acceleration devices. It is therefore necessary that one study electrical breakdown in vacuum at very high voltages and at voltages at least equal to the maximum voltage steps which will be designed into the final particle accelerator. With this high voltage goal in mind, a growth plan has been developed for attainment of 8 to 10 million volts across single gaps in vacuum. The figure of 10 million, however, is based on extrapolation from presently available data which only reaches 1. 7 million volts. Therefore, it must be und.erstood that the maximum voltage of each proposed facility may only be determined experimentally. The proposed growth plan is outlined on Fig; 39 which also has inset the present and projected potential~ available in vacuum according to the plan. The inset also includes the approximate dimensions of the vacuum chambers required to attain these potentials according to extrapolation of the beat available data. The one to two million volt facility (FiJ. ZS) was the first system developed by IPC. Experimentally, the power sources have proven capable of Z. 7 million volta potential difference. The combination of both power sources, Type I buahinss and present vacuum chamber have achieved only 1. 7 million volta potential difference. As previously explained, it is now realized that voltase• z. much higher than 0 million will not be obtained in this chamber, and yet only because of ite email else. The firat step in facility srowth ehoWd therefore be the inetallation of a larser vacuum chamber on thi• facility, thereby rai•lal it• ~­ !!!&.. z. volta1e to the 0 MV re1ion and it• maximum to approximately 2. 6 milUOD • • • -- I .It! I! f ~ :! il I I 1 > i t 9 ~ ~ I ! ~- !I ! ! !I ! ! ., 0> I ~---:- -......- -=='====== ~ ' ===--==.,==·--=-=--=:.-:-.:-=-~·-==;-~i=liB"===;;===I 0 I I ii: I I s I I • • • • • •• •• • ••••••• 1 ,• •••••••••••••••••••• 1 I :X: ,: . .................................., I •••••••••1 111 .SP ~ I I ~ : f I I I J• •• .._• • .s: I 1 . A .. 8 ., W ... a ... . J .. O .. j . . ' . O .. J . U .. I . 4 I iJ I JQJDJ-=V ':Htl\ I .S! I I += I i 0 c N~ f~ :e ., ., d5 - 0 ~ :e If) ., (I) ~---------------------~· volta. (Fig. 39, line S). It aeems advisable, in ~iew of the small expenae and the large gains to be reaped by this step, that it be undertaken in the immediate future. The 0. 4 MV HIVE facility shown in Fig. 27 was the second vacuum breakdown system to be developed at IPC. Individually, the bushings and power supplies of this chamber in the initial test stages have reached a potential differ ence of 0. 38 MV. There is every indication, however, that this chamber would be suitable for approximately 1. 0 MV total if suitable power were available. The power source which seems most applicable is the inverted 1enerator, and possibly the inverted Van de Graaff concept. Such a power source could be. developed in a small package to fit the HIVE, with a design voltage of around 0. S MV. If it proves satisfactory, fabrication of a second unit would give a total of 1. 0 MV for the system. Therefore, the second step in facility growth is envisaged as develop ment of prototype inverted power generators and the incorporation of these gen erators in a vacuum breakdown facility. (Fig. 39, lines 2. and 3) Further advantages from this plan are that working experience with the inverted supply could be com pared with past experiences with external supplies plus bushings in the two million volt facility. The third stage planned in facility growth is the development of power sources for a 4 million volt facility. (Fil• 39, line 7) Each source should have z. a nominal rating of approximately 5 MV positive or ne1ative, Thiil volta1e is within the limits of operation for the enlar1ed chamber proposed for the Z MV facility. Therefore, by proper flan1e and port desip, the 2.. 5 MV inverted power 1enerator may be built and tested before construction is be1un on the total 4 MV facility with 16 foot diameter chamber. This 1enerator, externally insulated by vacuum, is almost immediately applicable as the supply for a vacuum insulated (space borne) accelerator, and this sub pro1ram mates with the accelerator pro lrama as shown on Fils• 22. and 39. Followin1 aucceuful development of a 2. 5 MV power 1enerator, the fifth etep iD facility arowth would be actual fabrication and a ..e mbly of the 4 NV facility iDclucflDa .fabrication of a· •ecoacl 2. 5 NV aenuator. uz T It is felt that considerable experimental reuarch time and effort mi1ht be oriented around the 4 MV machine and that its abilities be pushed to the utmost before much thought is put toward higher ener1y levels. It is conceivable that high voltage-vacuum studies will be more or less completed at the 4 to S· million volt level. The attainment of higher energies across single gaps could then prove to be unnecessary. However, if higher energy levels are required, development of inverted power sources would be extended to the 5 MV level, using the 16 foot diameter facility for test purposes. (Fig. 39, line 10) Fabrication of a second 5 MV supply would follow, simultaneously with construction of a suitable vacuum chamber, radiation shielding, and instrumentation for performance of experiments at the 10 MV level. The estimated funding requirements to carry out this program are included on Fig. Z4 a, b and c. us REFERENCES 1. "Power Source• for Directed Energy Weapon•," AF08(635)-Zl66, Second Quarterly Report, January, 196Z, Ion Phy1ic1 Corporation, Burlington, Mall., (Secret). z. "Investigation of Jet Pla1ma Weapons, 11 AF08(635)-470, Final Report, November, 1960, Litton Indultries, Beverly Hilll, Calif., (Confidential). 3. "Investigation of Ion Beam Projecton," AF08(635)-307, Final Report, Augu1t, 1960, Pla1madyne Corporation, Santa Ana, Calif., (Confidential). 4. "Investigation of Ion Beam Projecton," AF08(635)-1636, Fifth Quarterly Report, July, 196Z, G. E. Tempo, Santa Barbara, Calif., (Secret). S. Cheever, R. N. and Bloch, G. E., "Optical Design Methods for High Cur rent Density Ion Engines," ARS Paper No. Z431-6Z, March, 196Z. 6. "Investigation of Pulsed-Train Plasmoid Weapon1," AF08(635)-1671, Final Report, March, 196Z, Aerojet-General Corporation, Downey, Calif. , (Secret R. D.). 7. "Power Sources for Directed Energy Weapona," AF08(b35)-Zl66, Second and Third Quarterly Reporta, January, 196Z and July, 196Z respectively, Ion Phy1ics Corporation, Burlington, Mall., (Secret). 8. Private communication, V. Jo1eph1on, Aero1pace Corporation, Loa Angelea, Calif. 9. Rosenthal, J. E., "Phy1ical Optica of Coherent Radiation Syatema," .!• Applied OptiCI No. Z, 169, March, 196Z. 10. Private communication, J. D. McClelland, Aero1pace Corporation, Lo1 Angelea, Calif. 11. "Study of Nuclear Weapona X-Ray Effecta," TR61-377, Tech. Opa, Burlington, Maa1., Auguat 1, 1961. (Secret) U. Sage, S. J., "High Efficiency Optical Maaer Parameter•," Applied l• Optic• No. 2, 17 3, 1962. 13. Opitz, W • , Proc. of Sec. Sympo1ium on Electron Beam Procee1ea, Cambridge, Ma. .. , 1960. Svelto, 0. • "Pumping Power Conaiderationa in an Optical Maaer," l• Applied Optic• No. 6, 745, 1962. 14. Tingye, Li and Sim, S. D., "A Calorimeter for Energy Measurement• l· of Optical Maser•, Applied Optic•, No. 3, 3Z5, 1962. II 15. Gloersen, P., "Pulsed Plasma Accelerators," ARS Paper No. 2129·61, October, 1961. 16. von Ardenne, M., "Tables of Electron and Ion Physics," Vol. 1, 135, Deutscher Verlag, Berlin, 1956. 17. Vorobyev, A. et al., "High Voltage Te1ting Equipment and Measure ments," Gosenergoizdat, Mo1cow (1960). Available in translation from Technical Documents Liaison Office, MCLTD, Wright-Patter1on AFB, Ohio as SOV/4809. z, 18. Denholm, A. S. et al., Aatronautic1, No. 6, 46, 1962. 19. "Industrial Proceuing Application• of Electron-Beam Radiation," The Engineer, January 8, 15 and 22, 1960. 20. Cleland, M. R., "Voltage Multiplication Apparatus," U.S. Patent 2, 875,394, February 24, 1959. 21. Private communication, C. Goldie, High Voltage Engineering Corpora tion, Burlington, Ma11. ..!!!• 22. Brasch, A. and Lange, F., Z. Phy•ik, 10, 1931. 23. Grotz, G., "A High-Voltage Multigap Crowbar," preunted at the Hydro gen Thyratron and Modulator Sympo•ium, Fort Monmouth, New Jersey, May, 1962. l 24. Roger•, E. J., ElectroniCI, 58, March 30, 1962. 25. Van de Graaff, R. J., "ApparatUI for Voltage Stabilization of Con•tant• u.s. Potential High-Voltaae Generators," Patent 2,847,611, Au,uat, 1958. f • j l5) f ; 26. Kunsler, J.E., Rev. of Modern Phyaica _ll, 4, 1961. 27. Hulm, J. K., Chandraaekhar, B.S. and Riemerama, H., Sci. Paper 62-108-278-P3, Cryogenic• Conference, Loa Angelea, Calif., Auguat, 1962. 28. Handbuch der Phyeik, "Nuclear Inatrumentation I, 11 44, 64-98, Springer- Verlag, Berlin, 1959. - 29. Livingaton, M.S. and Blewett, J.P., Particle Acceleratore, Chapter• 3 and 4, McGraw Hill Book Co., Inc., New York, 1962. 30. Crawford, R. B. et al., "An Occluded Gas Ion Source," UCRL-3103, August, 1955. 31. Ehlera, K. W. et al., "Development of an Occluded Gas Ion Source," Rev. Sci. Inat. 29, 614, 1958. 32. Lamb, W. A. S. and Lofgren, E. J., 11High Current Ion Injector," Rev. Sci. l'lst. 27, 907, 1956. 33. Frohlich, H., " A High Current Ion and Electron Bf!am Generator," .!• Nukleonik 183, 1959. 34. Allison, R. W. et al., "Development of a 100 ma Proton Source and Lena Syatem," UCRL-9743, June, 1961; Rev. Sci. Inat. 32, 1331, 1961. 35. Demirchanov, R. A. et al., "A Proton Injector of High Precieion," J. T. P. (in pre11) ••• manuscript obtained via private communication with J. Kistemaker. 36. Galejs, A. D., "Ion Optical Study of an Aaymmetric Einsel Lena," HVEC E-Report No. 307, July 2, 1962; "Particle Trajectoriea in Acceleration Tubes," HVEC E-Report No. 267, February, 1961. 37. Cheever, R.N., Bloch, G. E. and Peck, L. G. , "Optical Design Method for High Current Denaity Ion Engines," Ion Physica Corporation; Burlington, Mase., 22TR-014, February, 1962. 38. Zworykin, V.K. et al., "Electron Optic• and the Electron Microacope," Wiley and Sona, New York, 1945. 39. Wittkower, A. B. et al., "Space Charge Expanaion of a Poaitive Hydro ..!!• sen Ion Beam Leavins a Duoplaamatron Ion Source," Rev. Sci. Inat• No. 5, 196Z. 15£. ;~· ' 40. Salvat, M. et al., "A Mea1urement of the Emittance of the Van de Graaff Beam," CEN.S.PA 7678, Aprilll, 1958. 41. Galej1, A. and Rote, P. H. , "Particle Trajectoriel in Accelerating Tube•," HVEC E-Report Z67, February, 1961. 42. Galej1, A. and Rou, P. H., "Calculation of the Effect of Space Charge on Ion Beam• in an Electro1tatic Field," HVEC E-Report No. Z76, April 19, 1961. High Voltage Engineering Corporation, Burlington, Mall. 43. "Ion Sourcea/Ion Beam Neutralization," Section 3. 4, a propoaal aub mitted to APGC, Eglin AFB, 14 May 1962, by Ion Phy1ic1 Corporation, (fori'J)~.J'ly po)odrich-High Voltage Aatronautic1, Inc., Burlington, Mau. ). {C onfldential 44. Carllaw, H.S. and Jaeger, J. C., Conduction of Heat in Solidi, Clarendon Preaa, Oxford, England, 1959. 45. Leighton, R. B. , Principle• of Modern Phy1ica, 495, McGraw Hill Book Co., New York, 1959. 46. Wright, K. A. and Trump, J. G., "Back•cattering of Megavolt Electron• From Thick Target•," Jour. Applied Phya. !_!, 687,, 1962. 47. Richtmeyer, F.K. and Kennard, E. H., Introduction to Modern Phy1ic1, oO, McGraw Hill Book Co., New York,, 1947. 48. Livingaton, M.S. and Blewett, J.P., Particle Accelerator., Ch. 14, 508 McGraw Hill Book Co. , New York, 196Z. 49. National Bureau of Standard•, Handbook 55, Feb. Z6. 1954. SO. "Energy Di11ipation by Fa•t Electron•," National Bureau of Standard•, Monograph 1, Sept. 10, 1959. 51. Price, B. T., Horton, C. C. and Spinney, K. T., Radiation Shielclins, Pergamon Pre••• New York, 1957. 52. Panof•ky, W.K.H. and Philip•, M., Cla••ical Electricity andMaaneti1m, Addilon-Wuley, R.eaclin1, M&-1., 1962. 53. "Pul•e Heatina of Material• by Hiahly Concentrated Electron Beam• •. " AFCRL -62-167, Contract AF19(604)-8810, Proj. 4619, Final Report, Eitel-McCullough, Inc., San Carlo•, Calif., January 196Z. 15'7 54. "Structural R.eaponae to lntenae Electromaanetic Radiation," Space Science• Lab. , G. E. , Philadelphia, Pa. , .AFOSR.-2483, Contract AF49(638)1030, Proj. 9782, July, 1961. .!• 55. McCoy, T. M. , Space Aeronautic a (R. and D Handbook, 60-61) B-5, 1961. 56. Koelle, H. H. , Handbook of Aatronautical En1ineerint McGraw Hill Book Co., New York, 1961. .!!• 57. Bohr, N., Kgl. Danake Videnakab Selakab. Mat-fysk. Medd• 8, 1948. 58. Bethe, H. and Ashkin, J. , Experimental Nuclear Phyaica, Vol. I, E. Segre, Editor, Chap. Z, John Wiley and Sona, Inc., New York, 1953. .!• 59. Allison, S. K. and Warshaw, S.D., Revs. Modern Phys • 264, 1937. .!!.• 60. Williams, E. J., Revs. Modern Phya • 217, 1945. 61. Roui, B. B. and Greisen, K.I., Revs. Modern Phys. £• 240, 1941. 6Z. Fermi, E., Phys. Rev. ~· 485, 1940. 63. Williams, E. J., Phys. Rev. _!!, 29Z, 1940. 64. Bethe, H., Phys. Rev., 89, 1256, 1953. 65. Bohm, D. and Grou, E., Phya. Rev. ~· 1851, 1949. 66. Bohm, D. and Groaa, E., Phya. R.ev. 75, 1864, 1949. 67. Sturrock, P.A., Phya. Rev. !!.!• 1426, 1960. !.!!.• 68. Sturrock, P. A., Phya. Rev. 1488, 1958. 69. Finkelstein, D and Sturrock, P. A., Plaama Pbyaica, J. Drummond, Editor, McGraw Hill Book Co., New York, (Chap. 8) 1961. ..!• 70. Fainberg, Y. A., Pluma Pbya. (J. Nucl. Ener1y, Part C) 203, 1962. 71. Bohm, D. and Pinea, D., Phya. Rev.,~· 348, 1952. 71.. Gould, R. W., Plaama Dynamica, Chap. 4, F. H. Clauser, Editor, .Adcliaon-Wealey, Readina, Maaa., 1960. !!• 73. Looney, D. H. and BI'OWD, S.C., Pbya. Rev. 965, 1954. 74. Von aooa, 0., J. Malia. Ph-p. !• 10'7, 1960. 158 !!• 75. Trump, J. 0. and Vande Oraaff, R.. J., J. App. Phy1. 327, 1947. 76. Denholm, A. S. , Can. J. Phy1. 36, 476, 1958. 77. Hawley, R.., Vacuum~, 310, 1960. 78. Power Source1 for Directed Energy Weapon1, Amendment No. Z, POT ZM 61-104, Eglin Air Force Bale, Florida. 79. Jedynak, L., Invutigation of Vacuum Breakdown, MIT, Dept. of Elec• trical Engineering, D. Sc. The1i1, 196Z. 80. Van de Oraaff, R.. J., R.ose, P. H. and Wittkower, A. B., Nature 195 , un. 196Z. 81. Vogler, F. H., 11Electro1tatic Shielding for Space Vehiclea," I. A. S. Meeting, N.Y. C. January l!, 1963. 8Z. Lyman, E., Univeraity of Olinoi1, Dept. of Phy1ic1 (private communica tion). 83. Van de Oraaff, R. J. , High Voltage Engineering Corporation, Burlington, Mall. (private communication). 84. Philp, S., MIT, Dept. of Electrical Engineering, (private communication). 159 (b) (3) (A) - · • I 11' ' ' f (b) (3) (A) ; ' (b) (3) (A) I & I • .r r • - - · 'r r. - - ...= = • 1 1 . ._ .. .. ................. . a' - ao - ~ ............................ (b) (3) (A) -··· • .. •• II-., r.--------• •- -~···- ··---- - - - - - · - - - · · - __ . .... .,. ...... - ~.~~- - · - - - - · - - - - - - (b) (3) (A) ,._ - "..) 1 X - (b) (3) (A) - I - a - J~ •. I - I I -·· I. 5 (b) (3) (A) • - - · · - (b) .. •. (3) t (~A)r.iliiililliiir. ~ ~ j~ 6(v·jj(6 ilj • .. - ... - ~ /J/f +&1 (a-x)/t IP I -··<Mw>i'z" Ill' I ~z. - - I - ··-·- - .. ___ _. ......... ....... . ~ APPENDIX II CHARGE FLOW DOWN AN ACCELERATOR TUBE AND THE IMPLICATION TO POTENTIAL GRADING To allow a high current beam to pass effectively down an accelerator tube it is essential that the paaaage of the beam should not aiplificantly influence the accelerating gradient down the column. The desian of accelerator tubes and associated parts to maintain gradient with high currents is designated "beam stiffening. 11 This loas of gradient can either be by influence charge movements or by direct interception of a fraction of the beam. A. THE EFFECT OF INFLUENCE CHARGE FLOW To simplify the treatment, certain assumptions are made concerning the equivalent circuit of the generator during the pulse discharge, and also con cerning the charge denaity of the beam. The circuit which is aaaumed is shown ln Fig. (U- 1 ). Capacitance CA G ls large 1illce it contains several times the stored energy to be extracted by the beam. In the specific case of interest C AG esc' ls 1 0, 000 t-J+A.f. The interelectrode capacitances CSB' etc. are assumed I equal, which is reasonably correct even for an accelerator without special capacitances added for beam stiffening, as can be seen from the following table, which shows measurement• made on a 3 MV accele:r:ator. Capacitances Between Adjacent Potential Rings on a 3 Mev Van de Graff Column Ring• Number 1/Z 10/11 Z4/ZS 34/35 46/47 6Z/63 Capacitance (t'f.ll) Z61 Z39 Z35 Z54 Z31 Z40 U·l I I c Cso I D I E I F G G Fig. :1-1 Equivalent Circuit-Accelerator Tube System Theae meaaurementa were made with the tank removed, but bearing in mind that the capacitance of the relatively large terminal to ground (tank) ia perhapa 150J.&t'f, theae value• would not be appreciably changed by the preaence of the tank. Conaider a amall packet of charge~ at the front of the beam. As it moves from region A to region B there il a correaponding charge movement in the external circuit. Part of the charge takes the path A to B, diacharpng CSB' and the remainder path AGB, discharging C AG and charging CBQ" The flow through the two paths ia shared directly &I the capacitance of the two routea. The voltage drop Av auociated with the flow ~ is Av • *where (1) -1 CAQ CS (n-1) C = CSB + -l and CSB = CSC = CSD •••• = CS; there are CAQ+ CS (n-1) n sections in the tube. CAQ CS i.e. c • c s + (n -- -1 = ) ~ - C : A - O - + CS (Z) (3) In moving from B to C this charge ~ cause• the 1ame drop in voltage Av acroas CSC and 10 on. The increaae in voltage acroaa, for example, CSB when "gap BC is being traversed ia Av1 where, CA Av 1 0 Av• = Av x (5) ~= and the decrease in voltage across C AO when, for example, gap BC is being traversed, is Av" where, 1 (6) CAO = (n-1) C AO + c 5 For the complete traverse of the accelerator tube, the total drop in voltage acros a C due to q is AG p - ..2 nAv11 = (7) nCAO + CS CAO c The net change in voltage (Avn) across any for a complete traverse of 5 the accelerator tube by ~ is = Av Av+(n-l)Av1 n - _i = • (8) nCAG + CS nCAO It follows from these considerations that the first interelectrode gap suffers the greatest drop in gradient due to influence charge movements, and that after the tube is full of charge there ie no further drop in the gradient of that gap, except ineofar as the potential of C AG falle. The maximum drop in potential of CSB ie euentially when the tube ie just full of charge. Each element of charae in the tube will have eubtracted from CSB a net amount of charge depending on the position of that element of char1• in the tube. Tbie ia W11atrated in Fi. . ll-Z where the char1• lo. . in the capadtance aero•• the llrat la&enlectrocle pp ( 0 .. x ) la plotted a1alnat the 1 ca., poeltlon of char1• procluclal that loiS. D-4 .--11 II II. • I~ I I Cse Csc Cso CsE +80 c 8V 8 Fig.I-2 Charge Displacement on First lnterelectrode Capacitance for 4 Gap Tube. V Position of Charge qp Causing Displacement (cn-n C~a+cs] Cs~V • qp Cs(nC~Ca) D·5 The linear char1e density distribution ia obviously si!Mficant to the determination of the total AQ (AQT) due to all the char1e in the tube. Let the = linear char1e density diatribution be N f (x). It is required to determine the 1 chan1e of potential across the first interelectrode gap due to all the charge in the beam. From Fig. (li-Z) it can be seen that AQ, due to any char1e between the first interelectrode 1ap, is a function of the position of that charge, such that • (9) and in the remainder of the tube, nx -x = 1 AQ (10) Z x (n-1) 1 Considerin1 the element of length 6x and the linear char1e density (N ) then from expres a ion (4) 1 AQ =X- 6x ((n-1) CAG + Cs] (11) 1 x 1 nCAG + CS nx -x and AQ • 1 (lZ) Z x (n-1) 1 The total loss of char1e of the first interelectrode capacitance is then pven by: AQ • T (U) u-6 Conatant Velocity Beam: Conaider the caae of a beam movlnJ with conatant velocity (v) throuJh the tube auch that ~ • I. Thla la approximately the caae for an eneraetlc electron beam (v • c). Then N l • ~ la conatant. dx] (nx -x) AQ = 1 T n-1 (14) For the caae in mind, C ACi = 1 0, 000 ......,.£, c = Z50 1-LtU• n • 40, I • 100 ampere a, 5 = 8 z. -Z -7 v c • 3 x 10 M/S, x • 5 x 10 M and AQT • 1. 63 x 10 C. The drop 1 AQT ""CS' • ln potential acro11 the firat sap ia 650 volta. Compared with cap poten- tial of 50 kv, this can be nealected. Constant Ma11 Beam, Variable Velocity: Consider the caae of a beam beinJ accelerated by a constant 1raclient (Et) but nealect relativity effect• (low eneray ion beam). With an injection eneray V, the velocity of the particle• in the beam {v) is aiven by: v ;\) ~ ZeEtx (15) m m v I and N • l_ • Ze (V + E x) (16) I v m t X U-1 For the ca. . in mind, C AG = 10, 000 ~. c = Z50 !J..,_{, n = 40, I = 1 ampere, 5 = z. -Z -19 V • 100 kv (. .y ), Et ZO kv/cm, x • 5 x 10 M, e • 1. 6 x 10 C, 1 -Z7 m • 1. 67 x 10 kl (proton). -7 Subetltutlnlin (17) thi• live• AQT • 1. 6 x 10 C and the drop in poten- cAQ;T- • tial acrou_ the fir•t l&p 650 volt•. Compared with the 1ap potential, thb drop is not •ipdficant, but '• within one order of bein1 so. A lower inter electrode capacitance, a lower acceleratinl 1radient, a hi1her current or a.1reater particle mau than that a••umed could make the fall in 1radient due to influence charge effect• aipficant. In conclu•ion, it should be noted that the inte1rated chan1e• of char1e of all the interelectrode capacitance• of the tube due to the pre•ence of the beam ie zero. Thi• can be confirmed by referrin1 to Fil• ( U- 3 ) which •howe, for a 4-gap tube, the effect of an element of char1e ~ on the char1e di•plac,ement in each interelectrode capacitance plotted aaain•t the po8ition of ~ in the tube. For example, ~ at ordinate x is auociated with the followin1 di•placement char1••• capacitance CSl!S- -Yz • -3yl " esc- +yl " CSD- +yl " CSE- +yl It can be •••n that the total di•placement char1• i• zero and thl• ie the ca•e for all the char1•• in the tube irre•pectiv• of linear char1e di•tribution. U·l · · ~I 1~1 I~ +~Q X -~Q Electrode 0 2 4 • Fig.I-3 Charge Displacement on Each Capacitance V Position of Charge qp Causing Displacement ' . [Cn-1) C~a+Cs) Cs ~VI • qp C (nCAG+Cs) 8 Curve Cas- effect on oapaoltanoe C• curve csc- effect an capaoltanoe Cac curve c 80 - effect on oapacltanoe c 10 cww Car- etreot on OIPIDIIInOI Ca u-• B. DIRECT INTERCEPTION The effect of direct char1e interception by the electrode• on the vo1ta1e distribution ia cumulative over the duration of the pulse. The fraction of the be~ which will be intercepted by any liven electrode ia related to beam optica, scat terinl• etc., and will not be diacuued here beyond auumina a percenta1e inter ception. It b quite poaaible that leaka1e 'conduction' in the tube would be more important in losinl charae than beam interception. -3 c. The char1e in the beam pulse ie 10 With an inter electrode capaci- .u 1.,. tance of ZSO .... and interception of 0. of the beam at the firat electrode there would be a potential drop of 4 kv acroaa the first sap, or a chan1e of 1radient of 8.,., which ia alpficant. It would seem adviaable then to increase the interelec trode capacitance to a value which should be determined by experiment, atartinJ at perhaps S, 000 14...,f and decreaeina the value until deleterioua effect• are noted. U-10 ,.- ASD-TDR-63-29 INITIAL DISTRmUTION 2 Def Dir of R and E 1 AEDC (AER) 1 Joint Advance Study Group 1 AFWL (WLL) 5 Dir Advanced Rsch Proj Agency 1 AFWL ( WLL-3) 1 Wpns Sys Eva1 Gp 2 AFSWC ( SloJoiL) 1 Hq USAF ( AFC IN) 1 AFMDC ( RRRT) 1 Hq USAF ( AFOAP) 1 AFMTC (MTBAT) 1 Hq USAF (AFORQ-OT) 1 Hq AMD(DCS Rand D) 1 Hq USAF ( AFRDDG) 1 OAR (RROS) 1 Hq USAF ( AFRDPC) 1 AU ( AUL-9764) 1 Hq USAF (AFRDR-NU-3) 1 AF Tech Applications Cen/TD 1 Hq USAF ( AFRSTC) 1 USAFSS (ORC-R) 1 AFSC ( SCFRA) 1 AFSP Comn Cen (-SCP) 1 AFSC ( SCGB-2) 1 SAC (DI) 1 AFSC ( SCLDS) 1 SAC (DPLBC) 1 AFSC ( SCSAS) 1 SAC (DOPL) 1 AFSC ( SCTAE) 1 SAC (OASCI) 1 AFSC. ( SCTD) 1 1 Strat Aerospace Div 1 AFSC (MSFAL) 1 AAC (ACS/Intell) 2 RTD (RTNW) 1 DASA (Doc Lib Br) 3 ASD (ASNXR) 1 Atomic Wpn Tng Cmd (DASA) 1 ASD (ASNCS) 1 Opnav ( OP-75) 1 ASD (ASNL) 1 Opnav (OP-76) 1 ASD (ASl\CP) 1 Buweps (RT) 1 ASD (ASRMS-12) 1 ONR (Physics Div) 1 ASD (ASRNE) 1 Naval Rsch Lab (Code 624o) 1 ASD (ASRNG) 1 USA Research Office - (CRD-AA-1P) 1 ASD (ASRNGW) 1 US Army Nuclear Def Lab 1 ASD (ASRNRS) 1 Picatinny Arsenal (Tech Info Sec) 1 ASD (ASRSMD) 1 Picatinny Arsenal (ORDBB-VC2) 1 ASD (ASZ-5) 1 Picatinny Arsenal (SMUPA-VE1) 1 ASD (IMSE) 1 USA Missile Comd(AMSI-RFE) 1 FTD (TDATA) 1 USA Electronic Camd (Tech Lib) 1 FTD (TDCA) 1 USA Test and Eval Comd (Tech Lib) 1 FTD (TDCE) 1 USA Canbat Development Comd 1 FTD (TDEW) 15 DOC 1 FTD (TDFS) 1 CIA( OCR/Li h/ILS) 3 SSD (SSTRE) 1 Dir USAF Proj RAND 1 SSD (SSTRS) 3 Scientific and Tech Info Fac 1 SSD (SSTSW) (S-AK/DL) 1 BSD (BSRA) 1 Lewis Research Center (NASA) 1 Dep IG for Insp (AFIPA) 1 USAEC (Tech Info Ex) 2 ESD (ESAT) 1 USAEC (Div of Tech Info) 1 ESD (ESL) 1 USAEC (Div of Rsch) 1 ESD (ESRD) 1 Los Al.as Scientific Lab (Tech 1 AFCRL (CRRDA) Lib) 2 RADC (RAALD) 1 Argonne National Lab (Tech Lib) 1 ~ (RAOPA) 1 Brookhaven National Lab (Tech 1 RADC ( RASG) Lib) INITIAL DISTRIBUTION (Continued) 4 Univ of California 1 Grad Rsch Cen of the Southwest 1 Inst for Def Analyses 1 Elec Engineering Dept, NY Univ 1 Analytic Services, Inc 1 General Electric TEMPO Div 1 General Electric Co, Gen Engr Lab 25 Ion Physics Corp 1 Electro-Optical Systems 1 Aerojet-General Nucleonics 1 North Americnn Aviation 1 Raytheon Co 1 Stanford Research Inst APGC PGF 1 PGAPI 2 S.ACPO 1 T.ACIJJ 1 ASQWR 15 Jl'lll Jflfl' J'~ J'~ 1 !II tl ~t1 till''' 1 JJI If. ~ 1 ( 11, ~ ' 1 iJhl ltfil rli IIJ n, ll ;I Jl, 'tifa I ' ';'11 fl, il•~. II :: 1.,! "•r .. o if 1ilri f •t, 'lr I0 t~~ 1. il[f ii ,,(1 I rl I r ji ,J ~ f 1 ; ' f .. lf:o~ ~FI"I'")"Pt' :S:c:J FP:--r~e-'~ u,i rrfnUp'fl i HJ,r jff If :!;.ffHt•t I II ~ l J ""· '! i a !a:ou_if. .~,~~ i .. §"' ! Ii ! i i i }Jill' }I'll' ~~~ ~~~ ~1 11 i1 Ul ~~~~~~~ Ul rllr' d r ~~~ f J 'tf tl jthl il~ ilj f. rli ila ,,, ,& f flfti . l ';'11 u~ ';'II u~ i. 'I .,., i. ' .,., 1 htit 0 •1, I0 I.~~~ J~f 1 l,r,.. ( fl[l, rl ' 1 1 \ 1',1' f 1 ; ' f P:o;; ~fll"nl"t' p,.~ Pflrof')"l"t' \ \~ mm:,rrpl mm;,'rt1 • 1 ,.. I I 1 1 JI'JII; ,~. • ~ t J i ~ l I I . ~ ,~1~···1 I I I t~tf'W;I l1 l"lfl JJ.~hJ.~ fa~ JJ.~hl. ~~~ e:.:c e:.:c ..~o~.u..~=EI ..lol~ .. ..~tis . ' ' I l :r~.~ I I :rj~~ UH ~~ !1,1 ) tJltt 0 0 !·~ ti~H !i~ d~H t4d ~~~ ~~~ =aJti ·t 1 ~~~ a f~~ 'J(l ~~~ Jl}ll ~~~ ,J!ltil 1 1 ~~~ ii l!f li i!i(it Jii ljit .!~ ilhll ilhll1 i•l iii f ~ I ~ e -~-- -~ftLEI I I -~!U.e~ ifJfijUfi. I I Jl~ftjHji. l! I u h .!I.Ui.s ...;.-~.-"=a ==-=~ ..loi~.J..Idej e!O:ri 1 I • ~ ~ l I l !rjl~ ~~ J ~ H ]j t l 1 IP 1 o· -1 ·If' 0 !1 i J., n~H Jt• Jl, !·! I 1 ·•ttl ,u t•ltl ilf ~~~ ja "I 1 1 f ti•t! II Jlfl1 Jlt IJIJ i!l'i' lfi lri i.fft ltdJt JIJIJt IU IU