A Novel Final Focus Design for High Energy Linear Colliders

A Novel Final Focus Design for High Energy Linear Colliders
of 3
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  SLAC-PUB-9722April 2003 A Novel Final Focus Design for High Energy Linear Colliders * Pantaleo Raimondi and Andrei SeryiStanford Linear Accelerator Center; Stanford University, Stanford, California 94309 USAAbstractThe length, complexity and cost of the presentFinal Fo-cusdesignsor linearcollidersgrowsvery quickly with the beamenergy. In this letter, a novel final focus system spresented nd comparedwith the one proposed or NLC[ 11. This new design s simpler,shorterand cheaper, ithcomparable andwidth, olerances nd tunability. More-over, he engthscales lower han inearly with energyal-lowing for a more lexible designwhich is applicable vera much argerenergy ange. 1 INTRODUCTION The designof a final focus system or linear colliders sdrivenprimarily by the necessityo compensatehromatic- ity of the final doublet (FD). This chromaticity scalesasL*/@‘, whereL* (typically 2-4 m) is thedistance rom theinteraction oint (IP) to the FD and ?* (about -O.lmm) isthe betatron unction at the IP. As an example, he “tradi-tional” designof the NLC Final Focus l] with L* = 2 m,& = 10 mm and 3; = 0.12 mm s shown n Fig.1.Figure 1: Opticsof the raditionalFinal Focus or theNLC. The major disadvantage f the “traditional” final focussystem s that the chromaticity of the FD is not locallycompensated. s a direct consequencehereare ntrinsiclimitations on the bandwidthof the systemdue o the un-avoidable reakdown f theproperphaseelationsbetween the sextupolesand the FD for different energies. Thisprecludeshe perfectcancellation f the chromaticaberra- tions. Moreover, he system s very sensitive o any distur-banceof the beamenergy n between he sources f chro-maticity, whetherdue to longitudinalwake-fieldsor syn- chrotron adiation. In particular, he bends n the systemhave o be long and weak to minimize the additionalen-ergy spread enerated. n addition, he phase lippageofthe off-momentum articlesdrastically imits the dynamicaperture f the system.Therefore ery ong andproblem-atic collimation sections re equired n order o eliminate * Work supported y the U.S. Department f Energy,ContactNumberDE-AC03-76SF00515. these articles hat would otherwise it the FD and/orgen-eratebackgroundn the detector.The collimation section optics tself alsobecomes source f aberrations inceverylargebetaand dispersion unctionsare required. As a re- sult of all these imitations, he engthof the beamdeliverysystembecomes significant raction of the lengthof theentireaccelerator, ndscaling o higherenergiess difficult. 2 “IDEAL” FINAL FOCUS SYSTEM Taking nto account he disadvantages f the traditionalapproach, ne can formulate he requirementsor a more“ideal” final focus: 1)Thechromaticityshouldbecorrected as ocally as possible. 2) The numberof bendsshouldbe minimized. 3) The dynamicaperture r, equivalently, hepreservation f the inearoptics shouldbe as argeaspossi- ble. 4) The systemshouldbe as simpleaspossible.5) The system houldbe optimized or flat beams 3].Figure2: Optical ayout of the new inal focus.It is straightforward, tarting rom the IP, to build sucha system:1) A Final Doublet s required o provide ocus-ing. 2) The FD generates hromaticity,so two sextupolesinterleaved ith thesequadrupoles nda bendupstreamo generate ispersionacross he FD will locally cancel hechromaticity. 3) The sexupoles enerate eometricaber-rations, so two more sextupolesn phasewith them andupstream f the bendare equired.4) In general our morequadrupoles re needed pstream o match the incomingbeta unction (see he schematicn Fig.2).The second rderaberrations re cancelledwhen he xandy-pairsof sextupoles reseparated y matrices: whereall nonzero arameters rearbitrary. The sextupoleintegrated trengths S aredetermined y the equations: KSFS = -F3&~l ;KSD~ = -D3Ksol Kz andZ’ are he beam oordinates t the IP, &i is the hor-izontal chromaticityof the systemupstream f the bend,&s is the chromaticitydownstream, , is the vertical chro-maticity. RF, RD are he ransfermatrices efined n Fig.2.The angular ispersion t the IP, Q’, s necessarily onzeroPresented t the 7th European articleAcceleratorConferenceEPAC 2000), Vienna,Austria,June26-30,200O  in the new design,but can be small enough hat it doesnot significantly ncreasehe beamdivergence. art of thehorizontalchromaticitymustbe generated pstream f the bend n order o cancel he second rderdispersion.The third order geometricaberrations enerated y thesextupoles-are: 1222 = K~DK~F&R$&~~ U3444 = K~DKs&,,R:,,~C~I~ VI244 = u3224 = -KsDKsF[~I&,&~~ +Y~R~,,,R$,, - 4~34R~lzRo34R~12R~34]/2 where $12 and $34 are elementsof transfer matrix be-tween SF1 and sol. The term us444 s small if the lastquadrupoles defocusing. CT1222s negligible or typicalparameters f flat beams. CT1244 and us224 an be madeto vanish.Similar constraints old for third orderchromo-geometric berrations. ll these onstraints anbe satisfiedwith the simplesystem escribed bove.A systemwith the samedemagnification s he NLC FF andcomparable p-tical performance anbe built in a lengthof about300 m. 1~I III I IIFigGe 3: O$cs of ;h”,new EC Fin? Focu?&%&. 3 PERFORMANCE OF THE NEW FF The new FF systemhas potentially much better perfor-mance han the traditionaldesign. The “minimaI” opticsconcept an be further mproved y addingmoreelementsto minimize residualaberrations.An additionalbend up- streamof the second extupole air decreaseshromatic-ity through he system. An additionalsextupole pstreamand n phasewith the last one further reduces berrationsin x-plane. Sincesuchsystem s “ideal” up to third order,additionaldecapoles an give further mprovements.Thenew optics is shown n Fig.3. The flat beamparametersaregiven n Table1. The new systemhasan L* = 4.3 m, which is twice the srcinal value. This allows the use oflargebore superconductinguadrupoles ndsimplifies he designof the detector.Although he chromaticity s dou-bled due o the argerL*, the performance f the system sstill better han or the srcinal NLC FF design.Table1: BeamparametersBeamenergy,GeV500Normalizedemittances ee IYQ, @m>4 IO.06Beta-functions& / pa,at IP (mm)9.5 JO.12Beamsizes6, / a, at IP (nm)19712.7Beamdivergence 2: 8, at IP brad)21123Energyspread ~ ( 10w3)3Dispersion’11;at IP (10m3)5.4Figs.4,5,6 omparehebandwidth f the NLC FF and henewdesign.Figs.45 show he P beamsize sa functionof energy nd he uminosity eduction sa functionof energyspread.Beamsize and uminosity n Fig.5 were mprovedwith two additionaldecapoles. ig.6 shows he bandwidthin terms of the beamsize t the final doublet. The band-width is derived rom the variationof the beta unctionandthe beam sizes as they actually contribute o luminosity,which s determined y tracking.Thebeamsizebandwidth is narrower han the beta function bandwidthbecause fhigherordercross-plane hromaticaberrations.While the IP bandwidth or these wo systemss comparable,he FDbandwidth s muchwider for the new FF.Fig.7 shows he halo particle distribution at the face ofthe final doublet for the traditional FF and for the newFF. The beam s very distorted n the traditionalFF whilethe nonlinear ermsarestill negligible or the new FF. Thenonzero ispersion cross he FD in the new system as it-tle affect on the dynamicaperture. n addition, he designaperture f the NLC final doublet s about ra = 10 mmwhile for the new FF with twice longer L” this aperturecanbe as argeas r. =40 mm. Therefore he collimationrequirementsor the new FF may be relaxedby a factorof at leastone hundred n the IP phase, ndby a factor ofat least 3 for the FD phase nd energywithout increasingparticle osses t the FD.Due o the shorter engthof the system, herewould alsobe ess egenerationf thebeam alo n the inal focus tselffrom beam-gas cattering, educing n additionalsource fbackground. - (P, fPJ' I2(P, fPJ' I2-----(fj/pJ"----(fj/pJ".ax 'a"x 'a"cll0QY =oY =on- UL,UL,1.1-0OooOOOOOC1.0D*P----------0.9t 8---__ ___%\0.60.5 ,-0.010 -0.005 O.&O OiO5 O.dlO Figure4: IP bandwidth f the raditionalNLC Final Focus.Normalizedbetatron unctionsandnormalized uminosity equivalent eam izeversus nergy ffset A El E, andnor-malized uminosity versus ms energy pread ~. 1.6-o1.5- 0-(i-j, ,$'2c----<p,/p~"z000.6 AJ3E.5 , , -* ***Bc1-0.010 -0.005 0.000 0.005 0.010 Figure5: IP bandwidthof the New NLC Final Focus. I ’  - Traditional FF. x--o-- TraditionalFF.Y- NewFp.X---- NewFF,Y Figure6: FD bandwidth f the Traditional ndNew NLCFinal Focus. Normalized etatron unctionsat the finaldoublet ersus nergy ffsetA El E. lOO-0 0SO-D-100, , I . I I I I . I . I I t-1M) -80 -60 40 -20 0 20 40 60 80 100 Figure7: Beamat the entrancef the inal doublet or the traditionalNLC FF and or thenewFF.Particles f the n- comingbeam replaced n a surface f an ellipsoidwith dimensionsN,(z, z’, y, y’, E) = (800,8,4000,40,20)times arger han henominal eam izes. 4 SCALING WITH E AND ENERGY To maintain ptimalperformancef the systemwith larger incomingbeamemittances,he bend ield must ncrease like Bu cx 4. The ncreasedield is necessaryo holdconstanthe contribution f high orderaberrationso the IP beam ize,aswell as he contribution f the P angulardispersion,$ to thebeam ivergence.The scaling o higherenergiess mucheasierwith the new design. For a wide range f parameters,he IP spotsizedilution s dominated y theenergy pread reated ysynchrotronadiationn thebends. his scalesike ab,; gK y&3j2gll L2I pJ3 (5)"" gwhere & is the angular ispersion roduced y the bendswhere hebend ength s assumedo beproportionalo the total lengthof the system . The erms n the parenthe-sis areconstantf the P angular ispersions proportional to the beamdivergence nd f we conservativelyssumethat the normalized mittance ill be the same t higherenergies.n this case he engthof the system caleswith energy s L c( y7/ro. If, however,he achievableormal-izedemittance cales pproximatelynversely ith energy,as s assumedn [2], then he scaling ecomes o( y2i5.In this case,with thenewdesign,heFF or a 3 TeVcenter of mass nergy ollidercouldbeonly about 00m ong. 1.0“1TeV” parameters“5Tev” parameters - OhOn’y\ + cvE+synch.rad0.72505001000Beam energy. GeV \ \*- 2500 Figure8: Luminosity s beam nergyor thenewFF,bend field optimizedat eachenergy, eamwith energy pread. With andwithout synchrotronadiation.The “1TeV” pa- rametersorrespondo Table. , he STeV”set orresponds to [2] with YE = 50/l . 10vs m, p* = 9.5/0.14 mm,nE = 0.2%andc*(at 2.5TeV/beam)=1/0.54nm. The beam also emits synchrotron adiation in thequadrupoleshich becomes oreof a problemat higher energies. This can be reducedn the new designbe-cause he argerbandwidth llows the FD quadrupolesobe engthenedo minimize he synchrotronadiation en-eratedn them.For hepresentedptics, hedependencef the uminos- ity onbeam nergys shownn Fig.& Clearlya ixed ength final focushasa wide range f energies heret couldop- erate. f thebeam arametersrom 2] areassumed,his FF canoperate lmost p o 5TeV n thecenter f mass. 5 CONCLUSION We have eveloped newFinal Focus ystemhat hasbet- ter propertieshan he systems o ar consideredndbuilt. It is muchshorter, rovidinga significant ost reductionfor thecollider.Thesystem as imilarbandwidth nd ev- eral orders f magnitudesargerdynamicaperture.This reduceshebackgroundsnd elaxeshedesign f thecol- limation section. t is alsocompatiblewith an L' whichis twice as ong as hat n the raditionalNLC FF design,which simplifiesengineeringf the nteraction oint area.Finally, ts favorable calingwith beam nergymakest at- tractive or multi-TeV olliders.We believe hat further mprovements f the perfor-mance f the system repossible.We would ike to thank N. Phinney, . Raubenheimerand ? Tenenbaumor veryusefuldiscussions. 6 REFERENCES [l] NLC ZDR Design Group, A Zeroth-Order esign Reportfor the Next LinearCollider”, SLAC Report-4741996).[2] J.P Delahaye,G. Guignard,J. Irwin, T.O. Raubenheimer,R.D. Ruth, . Wilson,P.B.Wilson, “A 30 GHz 5-TeVLinear Collider”, PAC 1997Proceedings,1998).p.482. [3] P. Raimondi,A. Seryi A novelFinal Focus esign or futureLinear Colliders”, SLAC-PUB-8460,May 2000, submittedto Phys.Rev.Lett.
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks