High Current Transport Experiment

Transcription

1 HlFAN 1052 Design of a Proof of Principle High Current Transport Experiment S.M. Lurid*, R.O. Bangerter', J.J. Barnard*, C.M. Celata', A. Faltens', A. Friedman*, J.W. Kwan', E.P. Lee', and P.A. Seidl' * Lawrence Livermore National Laboratory (LLNL) 'Lawrence Berkeley National Laboratory (LBNL) 1999 Division of Plasma Physics Meeting Seattle, Washington USA November, 1999

2 Abstract Preliminary designs of an intense heavy-ion beam transport experiment to test issues for Heavy on Fusion (HF) are presented. This transport channel will represent a single high current density beam at full driver scale and will evaluate practical issues such as aperture filling factors, electrons, halo, imperfect vacuum, etc., that cannot be fully tested using scaled experiments. Various machine configurations are evaluated in the context of the range of physics and technology issues that can be explored in a manner relevant to a full scale driver. t is anticipated that results from this experiment will allow confident construction of next generation ntegrated Research Experiments leading to a full scale driver for energy production.

3 An ntegrate Research Experiment (RE) s eing Planned to Test Most Remaining ssues for Heavy-ion Fusion (HF) - K+ ions at 200 MeV, Etot 30 kj, ntensity goal? 3~10 ~ W/cm2 & match focusing 1.6 MeV accel (60m) (r-2m) to 9.4 MeV final focus mag nets ferromagnetic core inner m router m 32-beam / quadrupole magnet array - beam m8 cm bore > 3.5 cm half-lattice period m

4 Simulation Experiment - 3L = 56x10-3pC/m, K = 6. 3 ~0-4, 1 E, = mm-mrad measured strong relative space charge (driver like) - s2/(2kcrx>) Fllennmo fur w w 01 ra10ro~1i:o,lawrence But absolute strength weak relative to a driver BERKELEY N A T O N A L LABORATORY

5 Transport Understood n deal Theory and in Simulations WARP transverse sliceyy simulations [D.P. Grote] of a single-beam of an unoptimized RE design with acceleration, electric and magnetic fringe fields, lattice structures, etc. Various quadrupole alignment and field errors simulated Each color represents an overlay of five simulations with different sds of errors (random seeds ) Normalized RMS Edge Emittance F 2.5 n El m g 2.0 i E (m) RMS emittance growth due to expected machine errors is consistent with maintaining beam focusability on target W 1.o

6 A High Current Transport xperiment (HCX) s Needed Next-step RE/Driver machines will be expensive. For confident design: Non-scalable issues in transport need to be explored experimentally at driver scale line-charge density Dirty physics issues must be better understood Technological components need to be engineered, prototyped, and tested This program is being carried out with present group resources in the Virtual National Laboratory (VNL, an LBNL LLNL collaboration) by: Developing and prototyping technology for multi-beam transport, eg. arrays of quadrupole magnets for beam focusing, etc. Verify transport and needed machine aperture clearances at full scale in beam self-fields with a single-beam transport experiment A single-beam High Current Transport Experiment is being designed to verify transport at full driver scale line-charge density and pulse duration. This experiment will explore physics and technology issues associated with high current transport and will enable confident design of next-step RE/ Driver experiments.

7 HCX Philoso Test transport issues not addressed by scaled experiments and idealized simulations Purpose not to validate present thinking, but to find where it may be wrong Understanding will help select a more optimal beam size and thus downselect from the large range of beam number (4 to thousands) presently being considered Test transport limits to failure for confident RWdriver design Find and improve limits at full driver scale space-charge ( h pc) and pulse duration (AT 2 0 ps and less) Maximize average beam current density J and aperture fill factor - Push Technology to Limit 0 Transport limits are inextricably linked to technology Experiment should be assembled from hardware designed for the purpose

8 Hex: Physics issues Transport Limits at High Aperture Fill Factors mperfect alignment and focusing fields mage charges from beam proximity to conducting pipe Electrons 0 Production from lost halo particles and beam proximity to structuredapertures Self field potential - 6 kv on-axis and a long pulse ps can allow electron trapping Magnetic quadrupole focusing lacks intrinsic electron sweeping Streaming instabilities can rapidly degrade beam quality Beam Hato Halo particles can be lost resulting in electron production and liberation of desorbed gases Possible degradation of transport limits Matching Bunch Head, Transverse/Longitudinal Equilibration nstability, Space-Charge Waves, etc.

9 Electric Quadrupoles nhibit Electron Trapping Whereas Magnetic Quadrupoles Total electrostatic potential at the point of max beam envelope excursion in an x-focusing quadrupole \ Electric Focusing Magnetic Focusing Potential Contours 75.9 n E 0 v > Hyperbolic - / Electrodes (cm) Clear Aperture rp = 2.05 clpl x kv\;, Edge Radius ofbeam Envelope Grounded Beam Pipe = 2.47cm rp

10 HCX: Technology ssues (1) Superconducting Quadrupole Focusing Magnets Develop compact, high-field quadrupoles Wire radius rw cm, Field B, Tesla Employ mass production techniques for cheap construction Winding geometries for acceptable field quality Minimize axial and radial dead space in cryostats Employ features of array magnets to the extent economical 0 Gain experience with Cryo-systems Superconducting magnet power supplies Quench protection Diagnostics Must measure detailed distribution accurate time dependent measures of mismatch, misalignments, electrons, 'halo, etc, needed to understand transport limits Must be compact for eventual use in RE/driver

11 ~~~ ~ HCX: Technology ssues (2) Alignment, Matching, and Steering Systems (possibly automatic) needed for beam Matching Steering Electron sweeping Trimlprevent halos Vacuum Driver-like high vacuum systems Use liquid He in quadrupole cryostats for combined function cryo-pumping

12 Alternating Gradient Focusing Lattice F-Quad k -- ' TL -- * rll dl d2 Drift D-Quad Drift 1 1 de X 3 Electric Quadrupole... A V Axial Coord., s

13 Transport Limits: Transverse RMS Envelope Equations Neglecting image charges, centroid offsets, and nonlinear self-fields (emittance constant) to calculate the beam envelope radii in a lattice period ~ 2 2 d rx statistical beam envelope radii EX 2K - + Kxrx ds rx + r 2 d r -Y ds K = EX = rx E X 2 2K ~ r3 x Y rx + r Y T Y... 27~~~y r snpcs [(x )(X ) 7-Y 2 1/2 - (xxl) ] ( &xn = YbPbE, normalized )... P- = 2 ( Y ) Dimensionless Perveance measures space-charge strength RMS Edge Emittance - measures x-x' phase-space area (beam size)sqrt(thermal temperature) The matched beam solution together with parametric constraints from engineering, higher-order theory, and simulations are used to design the lattice

14 Transport Lirnits: ~ ntroduce 0 and o0the phase advance of particle oscillations per lattice ( 2 ~ in ) the presence and absence ( = 0 ) of space-charge 2 2 cosoo =: 1 - (WqL 2 1 [(l -;+o-;p4 1 For maximum current transport ( 0 -> o ),the matched envelope,quatioi.; can be approximately solved to obtain [E.P. Lee, Particle Accel. 52,115 (1996).] Scaling of Solutions average force balance - rx 2L r rx = J A d s 2L 0 envelope excursion

15 Transport Limits (2): Lattice parameters must be chosen consistent with stability restrictions, materials limits, practical engineering tolerances and rules of thumb that account for detailed effects, dirty physics, etc. For example: o0 < 80 to account for envelope and space-charge instabilities Field limits in magnetic materials, superconductors, voltage breakdown limits, materials strengths, source technology, etc. Aspect ratios and geometries to limit undesirable nonlinear applied focusing fields Machine alignments tolerances 1 An important choice is the machine aperture or pipe radius. Practical experience with scaled experiments and simulations shows: For d o o<< 1 r = 1.25 ~ M a x [X ]r + A fmage For charge forces t For Centroid Oscillations TypicalA = 1 cm to 0. 5 cm Depending on details of lattice, steering, etc.

16 Transport Limits (3): A systems code incorporating numerous constraints was written to design lattices that minimize the aperture needed to transport a given current B P = Quadrupole Field at rp B W = B P ( rw / y P ) = Potassium ( K + ) with 0 - F i e l d at r w = inner v4 = Rod-to-Ground Voltage of Quadrupole Rod w i r e r a d i u s of magnet c o i l 1A ( h pc/m) most attractive: Consistent with driver-like injector technology Practical and economical electric and magnetic focusing quadrupoles Allows more viable acceleration and pulse compression experiments in a continued series of experiments

17 HCX: Lattice Options with C S ~ = 80 Potassium * Baseline Case (38.96 amu, singly ionized) = 810 ma, K = , E, E = 2. 0 MeV, Case * Sync, Mag L cm 17 Sync, Elec 18 FODO, Mag 21 FODO, Elec 20 i $ lvg[r,l cm Max[r,l cm rp cm Sync,Mag Sync,Elec L cm Bw Tesla ~ Cesium (132.9 amu, singly ionized) E = 2. 0 MeV, = ma, K = , Case rw cm = mm-mr E,-- = mm-mr Avg[r,l Max[r,l rp rw Bw cm cm Tesla cm cm a V kif - df cm OGO ~ ~~~ ~ mr

18 Matched Envelope --- 2MeV, 810 ma, K* Beam. S, Axial Coordinate (cm) - - rx - ry = Max[rX] = Max[ry] = Min[ry]. = cm = cm cm

19 Lattice Schematic --- 2MeV, 81 0 ma, K+ Beam R-Z Cross-Section (2 quadrupole cryostat) X-Y Cross-Section (in magnet) 6 n E v u -z -4 E, 2 n' , Cryostat Coils Free Space (Diagnostics, Pump, etc) Bellows Beam Pipe nsulation Coils Co a r/yo ke Pipe

20 3x3 array, cos(28), circular cross section eff = m A Superconducting Quadrupole Magnet Array for a Heavy on Fusion Driver, S. Caspi et al. (1998). Vector Potential and Stored Energy of a Quadrupole Magnet Array, S. Caspi (PAC99). L A W R E N C E BERKELEY noeiourewo.o-bd.p NATONAL LABORATORY

21 h,> 3x3 array, pancake design [ Caspi, Hinkins, and Martovetskv(LLNL)1 W Quad &ray r r rrrr r G=47 T/m = 6295 A harmonics, forces z l o d =QO =CO mefe 26 strand Rutherford cable 23 turns. The LLNL Quadrupole Array, S. Caspi, R. Hinkins, N. Martovetsky, SC-MAG-676 (1999). DC

22 Advanced Magnet Labs (AML) is Constructing Prototype Superconducting Magnets for HCX.AML winds magnet coils with robots, using round, flexible cables -Two prototype magnets near z L Y completion: 1: rp = 2.3 cm, L,= B, = 3-4 Tesla 15 cm, 2: r,, = 6.0 cm, Leff= 50 cm, B, = 4 Tesla 37-strand cable, d-2rnrn.lbnusc program is developing and supplying the cable (H. Higley), and will test the finished quadrupoles OOErovlewO 8 rd.ript L A W R E N C E B E R K E L E Y N A T O N A L LABORATCJRY

23 Electrostatic Quadrupoles 4-beam quadrupole arrays already developed induction linacs can be readily adapted to meet HCX needs i i 1 i i! 1 1li P /i4 E 15i l p d w,&,,* Total Axial Length, Ltot = 2 8 cm Clear Aperture Radius, rp = 2. 3 cm 2 5 cm Breakdown Voltage, vs,m a x kv EffectiveAxialLength,~,,, (Rod-to-Ground) -

24 HCX njector -- Near Term An existing injector system is based on a diode with a surface-ionization source, a biased column of electrostatic quadrupoles (for acceleration with focusing), and a Marx-based pulser [S.S. YU et al, Fusion Eng Design, 32,p 309 (1996)].K+ Alumino-silicate sources up to 1 7 c m diameter ( amu) andcs" ( amu) sourcestested Marx presently 2 p~ upgrading to 4 ps (possibly -7 ps) Beam uniformity emerging from diode poor (optics and/or vaccum?) but is being fixed for HCX = ma K+ : E = 2 MeV E, > 0. 6 mm-mr = 430 ma cs+: E = 2 MeV E, > 0.3 mm-mr t-dode (73 Cn) Sdurce --ELECTROSTATC -*a QUADRUPOLES E x t r a c l h electrode c This injector is adequate for early stages of the experiment and improved K+ and Cs+ injector systems with long duration pulsers ( 2 0 ps ) are under development

25 ~ ~- HCX njector -- Planned mproved K+ single-beam injector systems are under development with: E = 2 MeV > 800 ma Current, < 1 mm-mu Normalized RMS Edge Emittance - Tp > 20 ps Kinetic Energy Pulse Duration Will allow intense beam experiments over full driver scale pulse durations ssues: Beam uniformity must be addressed Source Alumino-Silicate Vapor onizer Beam Uniformity Less More Neutral Flux Cs' High K+ too large? acceptable

26 Bevatron Building at LBNL a Possible Site for HCX 100+ meters of straight axial length available -20 meters needed initially -- should not interfere with present site uses f needed: Sufficient for extended experiments on focusing, compression, and chamber transport Possible to use for early stages of RE or (with extensions) a full RE Facilities present and conveniently located to AFRD group Possible Beamline Layout

27 Possible Phase of the High Current Experiment Design should be flexible to allow progress in the event of any RE delays 0 Capable of reconfiguration to exploring a wide range of issues for RE/Driver 0 Choose site to allow a wide range of experiments to be carried out ssues addressable in an expanded program: Beam acceleration to MeV (short pulse for economy) Bunch compression (short pulse) Longitudinal space-charge waves Final focus systems Unneutralized, partially neutralized, and channel transport from final focus lenses to the focal spot

28 Conclusions We are designing a High Current Transport Experiment (HCX) to test fullscale intense heavy-ion beam transport and enable high confidence nextstep ntegrated Research Experiments (RE) and Driver designs for Heavy on Fusion (HF) Test non-scalable issues -- Electron effects with magnetic focusing and strong self-fields -- Halo and lost particles -- mperfect vacuum, materials, etc. Develop technology needed for next-step experiments -- Superconducting quadrupole magnets: Small aperture (rp = 2-4 cm), high-field (BP = 3-6 Tesla) -- njectors: High current (-1 A), long-pulse (-20 ps), uniform charge ( p ) -- Diagnostics: Compact flexible for beam position and phase space Status Preliminary design work has begun -- Options in machine parameters are being analyzed, simulated -- Component development and prototype -- Teams of engineers and scientists are being assembled Timescales consistent with programmatic needs -- Early assembly work is expected by mid-year Main program 2-3years -- f needed: Machine upgrades for acceleration, compression, etc.