CLIC and Other Options for MultiTe V Lepton

CLIC and Other Options for Multi-Te. V Lepton Physics Tor Raubenheimer Accelerator Research Division Head, SLAC P 5 Meeting Fermilab February 1 st, 2008 Tor Raubenheimer

Introduction • Outline – – CLIC concept (X-band Two-Beam Accelerator) Technology status Outstanding issues LC roadmap and other options • Assumptions – Believe that the motivation for Te. V-scale LC remains the same but timescale is slower, motivating a broad look at LC technology • Caveats – Evaluation of outstanding issues for CLIC design is my opinion – Suggestions for ‘other options’ is also my opinion • These views are not endorsed by SLAC, the GDE, or … • I (and SLAC) are committed to developing the ILC as the near-term solution for a 500 Ge. V LC February 1 st, 2008 Page 2 Tor Raubenheimer

What is CLIC? • CLIC = Compact LInear Collider – Developed by CERN originally as a 30 GHz and 150 MV/m that is based on a two-beam accelerator concept • Two-beam concept is an efficient way to transform rf frequency from long-pulse low-frequency short-pulse high-frequency and thereby drive high gradients • Concept is elegant but still waiting for demonstrations and detailed costs illustrating the benefits – Developed parameters from 500 Ge. V 3 Te. V • Recently changed parameters to 12 GHz and 100 MV/m to reduce cost and better utilize GLC/NLC R&D – Development program to demonstrate ~100 MV/m by 2010 – CTF 3 test facility should demonstrate TBA concept on a similar timescale February 1 st, 2008 Page 3 Tor Raubenheimer

Two-Beam Accelerator Concept (from R. Corsini; 2006 parameters) February 1 st, 2008 Page 4 Tor Raubenheimer

Ma in CLIC RF Module ~ 2 meters Be am Dr iv e. B ea m ~1 Accelerating structure, A +100 MV/m, 64 MW, 229 mm 10 0 A rf distribution Power Extraction Structures: -6. 5 MV/m, 136 MW, 210 mm February 1 st, 2008 Page 5 Tor Raubenheimer

CLIC Schematic (2007 Parameters for 3 Te. V) Similar number of klystrons as 500 Ge. V ILC Drive beam complex efficiently generates high power beam Main linacs have deccelerator structures adjacent to accelerator structures in single tunnel – all LLRF and complicated electronics are elsewhere February 1 st, 2008 Page 6 Injector systems similar to other LC concepts Tor Raubenheimer

CLIC Linear Collider Parameters February 1 st, 2008 Page 7 Tor Raubenheimer

Possible CLIC Siting Option CERN site Prevessin IP under CERN Prevessin site Phase 1: 1 TEV extension 19. 5 km Phase 2: 3 Te. V extension 48. 5 km st February 1 , 2008 Page 8 Detectors and Interaction Point Tor Raubenheimer

Proposed Timescale (from JPD presentation to CERN SPC) February 1 st, 2008 Page 9 Tor Raubenheimer

Cost for TBA versus Conventional LC • Major study needed as part of CLIC CDR but characteristics can be understood. The TBA has a large central infrastructure that generates drive beam – Cost per Ge. V of TBA is likely cheaper than that of a conventional klystron-based linear collider – Initial cost of the TBA is higher than that of a klystron-based collider – Location of cross-over and slopes is unknown for present technologies February 1 st, 2008 Page 10 From 1998 comparison of 1996 NLC versus X-band TBA costs by G. Loew Cms Ge. V Tor Raubenheimer

GLC/NLC >50 MV/m Operation Breakdown Rate at 60 Hz (#/hr) with 400 ns Pulses Single Structures Eight Structure Average NLC/GLC Rate Limit Breakdown performance continued to improve with time BDR ~ exp(- t / 400 hrs) over the 2000 hrs operation February 1 st, 2008 Unloaded Gradient (MV/m) Page 11 Tor Raubenheimer

100 MV/m Structure testing at NLCTA (Structures from GLC/NLC program in early 2000’s) • Run slotted, a/l = 0. 18, 75 cm NLC structure (H 75 vg 4 S 18) with 150 ns pulses - at 102 MV/m, breakdown rate = 6 10 -6 • Run early NLC, non-slotted, 53 cm, smaller aperture (a/l = 0. 13) structure (T 53 vg 3 MC) at short pulses – unloaded gradient at a 10 -6 breakdown rate with 100 ns pulses is 105 MV/m and more recently it achieved similar gradient with 200 ns ramped pulse. • Building CERN-designed structures for future tests at SLAC and KEK February 1 st, 2008 Page 12 Tor Raubenheimer

Single Cell Accelerator Structure Testing (Understand Fundamental Breakdown Limits) Goals • Study rf breakdown in practical accelerating structures: dependence on circuit parameters, materials, cell shapes and surface processing techniques Difficulties • Full scale structures are complex and expensive Solution • Single cell Traveling wave (TW) and single cell standing wave (SW) structures with properties close to that of full scale structures This program, now, has a strong participation from both KEK and CERN. Time of flat pulse after filling time Variety of Single Cell Accelerator Structures Manufactured KEK February 1 stat , 2008 SW accelerator structure test with a/l~0. 21. In this type of structures loaded and. Tor unloaded gradients are the same Page 13 Raubenheimer

CTF 3 – CLIC Test Facility • Large-scale LC test facility to demonstrate TBA concept 2004 2005 Thermionic gun Linac DL CR 2007 Photo injector / laser tests from 2008 30 GHz production (PETS line) and test stand CLEX 2007 -2009 building in 2006/7 TL 2 2007 -2008 Beam up to here 150 Me. V 30 Major milestones in 2007: Combiner Ring (CR) installed CLEX building started Februaryfinished, 1 st, 2008 equipment Pageinstallation 14 Tor Raubenheimer A - 140 ns

RF Unit Demonstrations (What is necessary before construction? ) • The ‘RF Unit’ is the acceleration element that is replicated through the main linacs – Usually thought of as the minimal element that needs demonstration before construction -- CLIC is different – In GLC/NLC: two 75 -MW klystrons, SLED-II rf pulse compression system and 4. 8 meters of accelerator structure operating at 50 MV/m loaded ~250 Me. V per rf unit • Pieces demonstrated in 2004; System demo canceled – In ILC: a modulator and klystron, an rf distribution system, and 3 cryomodules with 26 1 -meter rf cavities operating at 31. 5 MV/m ~1 Ge. V per rf unit • Pieces to be demonstrated in 2010; System demo in ~2012 – In CLIC: a 2. 5 Ge. V 100 Amp drive beam is fed into ~600 meters of decellerator structures that accelerate the primary by ~60 Ge. V • Pieces in 15 ~2012 in CTF 3 no RF Unit demo February 1 st, demonstrated 2008 Page Torbut Raubenheimer

Outstanding Issues for CLIC • Program to develop high-gradient accelerator structures by 2010 – May not achieve 100 MV/m at desired breakdown rate but, given present results, will probably be close • Systematic cost estimate needed – Working with GDE to develop costs using same methodology as applied to ILC – aiming for 2010 -timescale • Tighter alignment and jitter tolerances – Aiming to demonstrate stabilization techniques by 2010 • Program to demonstrated TBA-concept in CTF 3 by 2012 and accelerate beams to ~1 Ge. V – Concept demonstrated but drive beam parameters quite different from CLIC and will not demonstrate an ‘RF Unit’ • Not clear what is necessary to launch construction and the collaboration is discussing options Tor Raubenheimer February 1 st, 2008 Page 16

Understanding the Gradient Choice G = A sqrt(P * Rs) P = rf power / meter Rs = shunt imp. / m Relative TPC • Cost optimum is a balance between costs proportional to length, i. e. tunnel & structures and costs GLC/NLC X-band proportional to the rf At low gradient, cost power sources increases due to larger • Have to reduce rf power cost per MW by 2 x or double shunt imped. to increase G by 40% February 1 st, 2008 Page 17 length costs At high gradient, cost increases due to higher rf power costs Unloaded Gradient (MV/m) Tor Raubenheimer

CLIC Gradient Optimization • CERN developed a detailed cost estimate using the TESLA estimate and the US Technical Options Study (2003) costing – Not entirely clear what is included and what drives the frequency scaling but the basic form makes sense – Believe that there is an assumption that above 10 GHz, the gradient is independent of frequency – Main point: very high gradients don’t make cost sense February 1 st, 2008 Page 18 Cost Previous New Optimum Tor Raubenheimer

Approaches to a Linear Collider (Four Options) • Superconducting rf (1. 3 GHz) – Strong international support through ILC collaboration – Gradients of 30 MV/m in cavities yielding 20 MV/m average – Technology well advanced (1 Ge. V test facilities under construction at Fermilab and KEK 2011 or 2012) – Can be stretched to ~1 Te. V energy • Normal conducting rf (11 ~ 12 GHz) – Strong international support through CLIC collaboration • CLIC recently adopted 12 GHz down from 30 GHz – Gradients of 100 MV/m yielding 80 MV/m average – Technology fairly well advanced (test facility at SLAC demonstrated 300 Me. V at 50 MV/m in 2004 and CTF 3 at CERN aiming for 1 Ge. V at 100 MV/m in 2012 - 2014) – Certainly reach 1 Te. V and maybe multi-Te. V energies February 1 st, 2008 Page 19 Tor Raubenheimer

Approaches to a Linear Collider (2) (Four Options) • Normal Conducting rf (cont. ) – Two NC rf source concepts have been considered: • Klystron-based linacs with klystrons along accelerator • Two-Beam accelerator with drive beam powering linac • Possible to consider a staged implementation using first klystron-based and then TBA-based rf power to reduce risk • Advanced concepts (laser and plasma) – Small lab and university-based collaborations – Gradients of many Ge. V per meter have been demonstrated – Technology has many challenges – working to develop roadmap illustrating development of acceleration concept and beam quality concepts – Some concepts (PWFA) use conventional rf linacs as drivers or injectors February 1 st, 2008 Page 20 Tor Raubenheimer

A Roadmap for Multi-Te. V Lepton Colliders Normal conducting - Two-Beam-based Multi-Te. V LC Normal conducting – Klystron-based 350 Ge. V LC Plasma Acc 4 th Generation SR Sources Superconducting RF 5 th Generation SR Sources? 500 Ge. V LC The LC roadmap illustrates Neutrino source options and connections between them. Selecting a path requires additional information such as LHC results and technology status Timescale (personal guess) February 1 st, 2008 Multi-Te. V LC 2010 2020 Page 21 2030 Neutrino ring Muon collider (few Te. V) 2040 2050 Tor Raubenheimer

One Possible Path to Multi-Te. V Lepton Physics Normal conducting - Two-Beam-based Multi-Te. V LC Normal conducting – Klystron-based 350 Ge. V LC Plasma Acc 4 th Generation SR Sources Superconducting RF 5 th Generation SR Sources? Multi-Te. V LC 500 Ge. V LC Neutrino source Neutrino ring Muon collider (few Te. V) Timescale (personal guess) February 1 st, 2008 2010 2020 Page 22 2030 2040 2050 Tor Raubenheimer

RF Power Source R&D • Developing rf power sources for ILC: – Marx solid state modulator – broad applicability of technology – Sheet beam klystron – broad applicability of SBK concept • Developed rf power source for GLC/NLC: – SLED-II system delivered >500 MW – Two-Pac modulator fabricated but never tested – halted in 2004 – X-band klystrons operated at 75 MW and 1. 5 us but limited by breakdowns → Consider new output structures or reduced power levels using knowledge from high gradient studies • Future program to complete X-band rf source program – Could provide a more conservative option to CLIC design – Power sources for compact radiation sources and other compact installations February 1 st, 2008 Page 23 Tor Raubenheimer

GLC/NLC RF Power Sources • Good success with modulator, pulse compression and rf distribution development. Klystrons achieved peak power and pulse length specs but BDR was too high Output Power (Gain = 3. 1, Goal = 3. 25) Combined Klystron Power February 1 st, 2008 Page 24 Tor Raubenheimer

Staged Approach to TBA • Should re-optimize the NC rf source but as a start: • Use the (nearly developed) GLC/NLC power source to power the CLIC accelerator structures at a loaded gradient of ~60 MV/m – Need to solve klystron BDR problem but assuming success • Increase gradient by ~20% for same cost per meter • Easy to perform systems demonstration of an rf unit • Simple improvements in pulse compression could increase power per meter 10% cost reduction • Build lowest reasonable energy LC with klystrons – Commission X-band main linac, BDS, sources and detectors – Use infrastructure to start testing TBA drive beam dynamics while operating klystron-based collider and then move to TBA. February 1 st, 2008 Page 25 Tor Raubenheimer

Another Possible Path to Multi-Te. V Lepton Physics Normal conducting - Two-Beam-based Multi-Te. V LC Normal conducting – Klystron-based 350 Ge. V LC Plasma Acc 4 th Generation SR Sources Superconducting RF 5 th Generation SR Sources? Multi-Te. V LC 500 Ge. V LC Neutrino source Neutrino ring Muon collider (few Te. V) Timescale (personal guess) February 1 st, 2008 2010 2020 Page 26 2030 2040 2050 Tor Raubenheimer

Comment on Spin-off Applications Both NC and SC rf technology have many additional applications Normal conducting RF • • Compact high gain FELs Storage ring injectors Medical linacs Industrial radiation sources Superconducting RF • High gain FELs • Recirculating linacs and CW applications • Industrial accelerators (no present applications) • To date, NC technology has been simpler and cheaper to implement (at least for small-scale applications) • SC technology is better suited for CW applications and NC is better suited to short high-current beam pulses • Both technologies can have comparable efficiencies and deliver comparable beam power February 1 st, 2008 Page 27 Tor Raubenheimer

Applications Example: High Gain FELs Roughly equal number of normal conducting and superconducting– based FEL sources February 1 st, 2008 Many FELs use higher harmonics for bunch compressions; SLAC was asked to build 12 GHz klystrons for Trieste, Page 28 Frascati and. Tor Raubenheimer PSI

Yet Another Possible Path to Multi-Te. V Lepton Physics Normal conducting - Two-Beam-based Multi-Te. V LC Normal conducting – Klystron-based 350 Ge. V LC Plasma Acc 4 th Generation SR Sources Superconducting RF 500 Ge. V LC PWFA accelerator could likely work with either SC or NC driver linacs – SC option illustrated here. Timescale (personal guess) February 1 st, 2008 5 th Generation SR Sources? Multi-Te. V LC Neutrino source Neutrino ring Muon collider (few Te. V) 2010 2020 Page 29 2030 2040 2050 Tor Raubenheimer

Example: Plasma Wakefield Acceleration (PWFA) • Acceleration gradients of ~50 GV/m (3000 x SLAC) – Doubled energy of 45 Ge. V beam in 1 meter plasma • Major questions remain – Beam acceleration – Emittance preservation – New facilities being developed February 1 st, 2008 Page 30 Tor Raubenheimer

Future PWFA Opportunities A Te. V Plasma Wakefield Accelerator based Linear Collider Single stage afterburner… … or optimized design using low energy bunch train to accelerate single high energy bunch Other applications: • Apply MT/m focusing gradients in plasma ion column to radiation production (Ion Channel Laser) • New phenomena may sources February(trapped 1 st, 2008 electrons) Page 31 offer high brightness Tor Raubenheimer

X-band R&D Funding Requirements • X-band R&D was cut from ~20 M$ / year to ~3 M$ per year after 2004 ITRP decision – 3 M$ / year funds US High Gradient Collaboration pursuing fundamental R&D on structure gradient limitations – US and KEK are working with CERN testing high-gradient structure prototypes. Need additional funds to support this. • Also urge funding for X-band power source R&D in US – Complete GLC/NLC rf power source development to facilitate a staged approach to CLIC while pursuing fundamental R&D on alternate rf power sources – Infrastructure is already in place relatively inexpensive to use; however it will be difficult to maintain capability without a program • Complete R&D program would ramp to ~10 M$ / year – Roughly 20% of projected FY 10 US SCRF and ILC programs February 1 st, 2008 Page 32 Tor Raubenheimer

Summary • Critical time for linear collider R&D program – Science case for a Te. V-scale collider remains strong • Need to consider what we as a community need to do to maintain options for energy frontier lepton probes – Options exist with different reaches, timescales, risks and costs • ILC is the most developed but X-band options also exist • Don’t really know the costs and risks of the different paths – Should have much more information in 2010 ~ 2012 • Develop multiple linear collider technologies: need R&D on SC, NC and advanced acceleration concepts – Great potential & many applications of the technology across science – Strong collaborations with ILC GDE as well as CERN and KEK – Extensive infrastructure exists to support X-band plasma R&D February 1 st, 2008 Page 33 Tor Raubenheimer