201 MHz NC RF Cavity R&D Derun Li Center for Beam Physics Lawrence Berkeley National Laboratory WG3 at NuFact 2004 July 28, 2004

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 2 Collaborators R. MacGill, J. Staples, S. Virostek, M. Zisman Lawrence Berkeley National Laboratory R. Rimmer, L. Philips, G. Wu Jefferson National Laboratory D. Summers University of Mississippi W. Lau, S. Yang Oxford University, UK

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 3 Outline

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 4 Introduction Muon cooling channels call for normal conducting RF cavities with highest possible accelerating gradients –Muon beam: secondary, unstable and has short decay time (~ 2  s at rest) Created with LARGE 6-D phase space → High gradient with large beam aperture (iris) Strong external magnetic field needed to confine muon beams → Normal conducting Muon beam decays, manipulation must be done quickly, including cooling → High accelerating gradient –Goal: design and engineering for RF cavity with high shunt impedance, large beam aperture and withstand high peak RF field → Higher gradient for the same input RF power (less RF power) → No surface breakdown

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 5 Introduction (cont’d) Shunt impedance of an RF cavity High peak RF field: Kilpatrick number Required gradient at 201 MHz: ~ 16 MV/m – Kilpatrick: 15 MV/m Required gradient at 805 MHz: ~ 30 MV/m – Kilpatrick: 26 MV/m

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 6 Choice of NC RF Cavity Conventional approach: NC “Ω” cavity with open iris –“High” shunt impedance for open iris cavities –high peak surface field (high E pk /E acc ~ 2) –Shunt impedance reduces with the increase of iris → Very difficult (if not possible) to achieve the gradient required for a muon cooling channel Taking advantage of muon beam’s penetration property, we choose a RF cavity with irises terminated by windows or grids –Pillbox like cavity –Lower peak surface field (low E pk /E acc ~ 1) –Independent phase control, higher transit factor –High shunt impedance, but with large windows Be window

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page MHz Cavity R&D Prototype of 201 MHz cavity with curved Be windows –Cavity design Body profile Be windows, grids Ports Coupler RF (ceramic) windows Tuners –Cavity fabrication going reasonably well Cu sheets + spinning techniques E-beam welding Ports extruding Cleaning, … –Progress and status of cavity fabrication –Placed purchase order of 21-cm radius of curved Be windows

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page MHz Cavity Concept Spinning of half shells and e-beam welding Water cooling channels Cavity design accommodates different windows At NuFact 2003 !

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page MHz Cavity Design Spinning of half shells using thin Cu sheets and e-beam welding to join the shells. Four ports across the e-beam joint at equator. Cavity design uses pre-curved Be windows, but also accommodates different windows or grids.

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 10 Cavity Body Profile De-mountable pre-curved Be windows pointing in the same direction to terminate RF fields at the iris 2 o tilt angle Spherical section at the equator to facilitate fabrication of ports (± ~ 6 o ) Elliptical-like nose shape to further reduce peak surface field 6-mm Cu sheet permits spinning technique and mechanical tuners similar to SCRF ones Stiffener ring Bolted Be window Spinning E-Beam welding Port extruding Curved Be windows

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 11 The cavity parameters The cavity design parameters (~1.2 m diameter, 0.43 m long) –Frequency: MHz – β = 0.87 –Shunt impedance (V acc 2 /P w ): ~ 22 M  /m –Quality factor (Q 0 ): ~ 53,000 –Curved Be window radius and thickness: 21-cm and 0.38-mm (better performance with significant savings, compared to pre- tensioned flat Be windows) Nominal parameters for a cooling channel in neutrino factory –16 ~ 17 MV/m accelerating field –Peak input RF power ~ 4.6 MW per cavity (85% of Q 0, 3τ filling) –Average power dissipation per cavity ~ 8.4 kW –Average power dissipation per Be window ~ 100 watts

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 12 Spun half shells + RF & CMM measurements CMM scans, RF frequency and Q measurements of half shells; Cu tape for better RF contacts. 3 CMM scans per half shell conducted at 0 o, 45 o, 90 o, respectively. Measured frequency: MHz (simulated frequency: MHz)

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 13 E-Beam welding at JLab Preparation for e-beam welding of the stiffener ring (left); after the e-beam Welding (above) Stiffener ring

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 14 Recent progress for the welding f measured = MHz

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 15 Recent Progress Port extruding We have successfully developed techniques to extrude ports across e-beam welded joints.

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 16 What’s Next? Cavity has been cleaned and ready for nose welding and ports annealing in next two weeks

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 17 Status Cavity cleaning at J-Lab 2~3 Months delay for using NASA e-beam welder –Extruding ports –Cosmetic welding Continue engineering designs –Loop coupler: conceptual → engineering design –Supporting structure (vacuum): developed –RF windows (SNS type 4” coaxial window) –Tuner: mechanical Cavity test at MTA this year

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 18 Window for muon RF cavity Performance for an ideal window –Transparent to muon beams  Low-Z material –Perfect electric boundary to RF field  Good electrical conductivity –Mechanical strength and stability  No detuning of cavity frequency under RF heating Beryllium is a good material for windows –High electrical & thermal conductivity with strong mechanical strength and low-Z Engineering solutions being explored so far –Thin, flat Be foils (pre-tensioned) –Curved Be foils –Grids (Ph.D thesis of M. Alshaor’a at IIT)

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 19 Curved Be window R&D –Designed, fabricated and tested pre-tensioned flat Be windows They work, but expensive; balance between thickness & RF gradient –Progress on FEA modeling and engineering design of all approaches –Fabricated pre-curved windows of S.S. and Be for 805 MHz cavity –Cu frames + curved Be foils: better performance with BIG savings ANSYS simulations: mechanical vibration modes Fabricated pre-curved Be window: 16-cm in diameter and mm thick

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 20 More on Be Windows Model validation –Preliminary measurements on mechanical vibration frequency of curved Be windows for 805 MHz cavity agree with FEA modeling Ti-N coatings at LBNL recently –Curved Be windows –Pre-tensioned Be windows Placed purchase order of 3 curved Be windows for 201 MHz cavity (21-cm radius, 0.38-mm thick) Baseline design for MICE cavities

WG3 at NuFact 2004 July 28, MHz NC RF Cavity R&D Derun Li Osaka University, Osaka, Japan Page 21 Summary