Ionization Cooling – neutrinos, colliders and beta-beams

  • View

  • Download

Embed Size (px)


Ionization Cooling neutrinos, colliders and beta-beams. David Neuffer July 2009. Outline. Front End and Cooling IDS neutrino factory Study 2A ISS baseline example Target-capture, Buncher, Rotator. Cooler Shorter bunch train example(s) - PowerPoint PPT Presentation


  • Ionization Cooling neutrinos, colliders and beta-beamsDavid Neuffer

    July 2009

  • OutlineFront End and Cooling IDS neutrino factoryStudy 2A ISS baseline exampleTarget-capture, Buncher, Rotator. Cooler Shorter bunch train example(s)nB= 10, Better for Collider; as good for -FactoryVariation 88 MHz Rf cavities in solenoids major constraint? up to 15MV/m, ~2T AlternativesUse lower fields (B, V), use magnetic insulation ASOL lattice, use gas-filled rf cavitiesLarge Emittance Muon Collider optionLow-Energy Cooling discussionERIT resultsIon cooling for Beta-beams

  • Official IDS layout

  • Neutrino Factory-IDSFor IDS need baseline for engineering

  • ISS Study 2B baselineBase lattice has B=1.75T throughout buncher and rotatorrf cavities are pillboxgrouped in same-frequency clusters7 to 10 MV/m Buncher; 12.5 Rotatorwith 200 to 395 Be windows,750 windows in Rotator

    Cooling Lattice is alternating-solenoid with 0.75 half-period0.5m pillbox rf cavity1cm LiH absorbers15.25MV/m cavities

  • IDS - Shorter VersionReduce drift, buncher, rotator to get shorter bunch train:217m 125m 57m drift, 31m buncher, 36m rotatorRf voltages up to 15MV/m (2/3)Obtains ~0.26 /p24 in ref. acceptanceSimilar or better than Study 2B baselineBetter for Muon Collider80+ m bunchtrain reduced to < 50mn: 18 -> 10 -3040m500MeV/c

  • Shorter Buncher-Rotator settings Buncher and Rotator have rf within ~2T fieldsrf cavity/drift spacing same throughout (0.5m, 0.25)rf gradient goes from 0 to 15 MV/m in buncher cavities Cooling same as baselineASOL lattice 1 cm LiH slabs (3.6MeV/cell)~15MV/m cavitiesalso considered H2 cooling

    Simulated in G4Beamlineoptimized to reduce # of frequenciesHas 20% higher gradient ASOL lattice

  • Rf in magnetic fields?Baseline has up to 12 MV/m in B=1.75T (in 0.75m cells)short version has up to 15MV/m in B=2.0TExperiments have shown reduced gradient with magnetic field

    Results show close to needed ? 14MV/m at 0.75T on cavity wallhalf-full or half-empty ?Future experiments will explore these limitswill not have 200 MHz in constant magnetic field until summer 2010Open cell cavities in solenoids?did not show V /B limitation

  • Solutions to possible rf cavity limitationsFor IDS, we need an rf cavity + lattice that can workPotential strategies:Use lower fields (V, B) Use Open-cell cavities?

    Use non-B = constant latticesalternating solenoid

    Magnetically insulated cavitiesIs it really better ???Alternating solenoid is similar to magnetically insulated latticeShielded rf latticeslow B-field throughout rf -RogersUse gas-filled rf cavitiesbut electron effects?

  • Lower-field (?) VariantUse B=const for drift + buncherLow-gradient rf ( < 6 MV/m) B= 1.5 to 2.0 T ?Use ASOL for rotator + Cooler (and/or H2 cavities)12 MV/m rf Rotator15 MV/m cooler0.75 half-cells

    Simulation: fairly good acceptanceLose some low energy musbunch train shortened~0.25 /24p after 60m H2 cooling~0.19 /24p after 60m LiH cooling

  • Change cavity material-PalmerBe windows do not show damage at MTAno breakdown?Model: Energy deposition by electrons crossing the rf cavity causes reemission on the other sideless energy deposition in Behigher rf gradient threshold ~2 gradient possible with Be cavities ??calculated in model extrapolation to 200MHz ?

  • Variant: 88 MHz Front end Drift ~90mBuncher ~60m166100 MHz, 06MV/m Rotator ~58.5m10086 MHz, 10.5 MV/m Cooler ~100m85.8MHz, 10 MV/m1.4cm LiH/cell ASOL

  • 88 MHz examplePerformance seems very good~0.2 /p24smaller number of bunches > ~80% in best 10 bunches

    Gradients used are not huge, but probably a bit larger than practicalup to ~10 MV/m~2T magnetic fieldsWith 10 MV/m (0.75m cells) probably not free of breakdown problems

    redo with realistic gradients6MV/m ?

  • Plan for IDSNeed one design likely to work for Vrf/B-fieldrf studies are likely to be inconclusive Hold review to endorse a potential design for IDS likely to be acceptable (Vrf/B-field)April 2010 ?Use reviewed design as basis for IDS engineering study

  • Cooling for first muon colliderImportant physics may be obtained at small initial luminosity +- Collider

    + + - -> Z* , HS L > 1030 cm-2s-1

    Start with muons fron neutrino factory front end:3 1013 protons/bunch1.5 1011 /bunch~12 bunches both signs!t,rms, normalized 0.003m L,rms, normalized 0.034mAccelerate and store for collisionsUpgrade to high luminosity

  • Proton Source: X -> -Factory/-Collider Project X based proton driver8 GeV SRF linac , 15 Hz1.21014/cycleH- inject full linac pulse into new Accumulatorsmall dp/pLarge N6 =120 mm-mradBunch in harmonic 4 adiabatic OK !! (2kV)Transfer into new Buncher100kV h=41250 turns (2ms)short ~1 m bunches !!31013/bunchBF = 0.005 = 0.4

  • Large Emittance Muon ColliderProton Linac 8 GeVAccumulator,BuncherHg targetLinacRLAsCollider RingDrift, Bunch, Cool200mDetectorUse only initial front-end coolingAccelerate front-end bunch train; collide in ring

  • Must be upgradeable to high-luminosityMEMC Upgradesreduce t to 0.001minitial part of HCC1300MHz rfcombine 12 -> 1bunchL -> 3 1032High luminosity Cool to 0.000025

    ParameterSymbolHEMC MEMC LEMC ValueProton Beam PowerPp2.4 MW4MW4MWBunch frequencyFp60 Hz60Hz15HzProtons per bunchNp310135101341013Proton beam energyEp8 GeV8 GeV50 GeVNumber of bunches nB1211+/-/ bunchN10111.5101221012Transverse emittancet,N0.003m0.001m0.000025Collision **0.06m0.040.01Beam size at collisionx,y0.013cm0.0063cm0.0005cmBeam size (arcs)x,y0.55cm0.32cm0.05cmBeam size IR quadmax5.4cm3.2cm0.87cmCollision EnergyE+,E_1 TeV (2TeV total)1 TeV 1 TeV Luminosity turns nt100010001000Luminosity cm-2s-1L0410302.710321.51034

  • Other cooling uses- not just high-energy muons!. Stopping beam (for 2e, etc.)C. Ankenbrandt, C. Yoshikawa et al., Muons, Inc.

    For BCNT neutron source Y. Mori - KURRI

    For beta-beam sourceC. Rubbia et al

    (dE/ds)/E= gL(dp/ds)/p


    P( (MeV/c)




  • Virtual detectorr = 3 mend of NF/MC drift region & from 100k POT MERIT-like targetryRevisit Use of NF/MC Front End to Stop Muons with Momentum-dependent HCCmatching (not done)100k Mu-s w/ Bent Sol Spread at start of HCC.Mu-s midway to end of HCC (20,836/100,000)Mu-s at end of HCC. Displayed is 5398/100k, but stopping rate is 3519/100k.17025P(MeV/c)s stoppedPotential to enhance yield via P vs. y correlation in bent solenoid.C Yoshikawa

  • FFAG-ERIT neutron source (Mori, KURRI)Ionization cooling of protons/ ions is unattractive because nuclear reaction rate energy-loss cooling rate

    But can work if the goal is beam storage to obtain nuclear reactionsAbsorber is beam target, add rfERIT-P-storage ring to obtain neutron beam (Mori-Okabe, FFAG05)10 MeV protons ( = v/c =0.145)10Be target for neutrons5 Be absorber, wedge (possible)Ep=~36 keV/turnIonization cooling effects increase beam lifetime to ~ 1000 turns not actually cooling

  • Observations of Cooling-PAC09ERIT ring has been operatedBeam lifetime longer than without energy-recover rfagrees with ICOOL simulationBeam blowup is in agreement with simulationmultiple scattering heating in agreement with ICOOL

  • -beam Scenario (Rubbia et al.)-beam another e sourceProduce accelerate, and store unstable nuclei for -decayExample: 8B8Be + e++ or 8Li8Be + e-+ *

    Source production can use ionization coolingProduce Li and inject at 25 MeVnuclear interaction at gas jet target produces 8Li or 8B7Li + 2H 8Li + n6Li + 3He 8B + pMultiturn storage with ionization cooling maximizes ion production8Li or 8B is ion source for -beam acceleratorC. Rubbia, A. Ferrari, Y. Kadi, V. Vlachoudis, Nucl. Inst. and Meth. A 568, 475 (2006). D. Neuffer, NIM A 583, p.109 (2008)e

  • -beams example: 6Li + 3He 8B + n Beam: 25MeV 6Li+++ PLi =529.9 MeV/c B = 0.59 T-m; v/c=0.094 Jz,0=-1.6Absorber:3He -gas jet ?dE/ds = 110.6 MeV/cm , If gx,y,z = 0.13 (g = 0.4), =0.3m at absorberMust mix both x and y with zN,eq= ~ 0.000046 m-rad, x,rms= ~2 cm at =1mE,eq is ~ 0.4 MeV

    Could use 3He as beam 6Li target ( foil or liquid)

  • -beams alternate: 6Li+3He 8B + n Beam: 12.5MeV 3He++ PLi =264 MeV/c B = 0.44 T-m; v/c=0.094Absorber: 6Li - foil or liquid jetdE/ds = 170 MeV/cm, LR=155cm at (Li-6= 0.46 gm/cm3)Space charge 2 smaller If gx = 0.123 (g = 0.37), =0.3m at absorberN,eq= ~ 0.000133m-radx,rms= 2.0 cm at =0.3m, x,rms= 5.3 cm at =2.0mE,eq is ~ 0.3 MeVln[ ]=5.34

  • Cooling Ring for Beta-BeamsAssume He-3 beam B=0.44T-m, =0.094Cooling ring parametersC =12m (?)Absorber0.01 cm Li wedget = ~0.3m, = ~0.3mrf needed2 MV rfInjection charge strip He+ to He++ (?)Extraction kicker after wedgeNuFACT09miniworkshop: July27-29Solenoid1.38T-mCooling wedge=0.3m, =0.3m rf

  • SummaryRf in magnetic field problem must be addressed Need rf configuration that can work with high confidenceNeed to establish scenario Use as basis for engineering study

    Further meetings/studiesNuFACT 2009miniworkshop at Fermilab (July 27-28)front end and beta-beam cooling9-11am WH3NE1:30-4PMFront End ReviewApril 2010?

  • Future Funding ??