Muon Production Using a High Power Cyclotron L

Muon Production Using a High Power Cyclotron L. Calabretta, INFN-LNS, Catania on behalf of DAEd. ALUS collaboration Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 1

DAEd. ALUS, a Decay-At-rest Experiment for d. CP studies At a Laboratory for Underground Science, provides a new approach to search for CP violation in the neutrino sector. The design utilizes high-power proton beam to produce neutrino flux with energy up to 52 Me. V from pion and muon decay-at-rest. The experiment searches nm ne for at short baselines corresponding to the atmospheric Dm 2 region. The ne is best detected in a large (>100 kton) water Cerenkov detector, preferable Gd-doped , via inverse beta decay. Experiment proposed by J. Conrad (MIT) and M. Shaevitz (Columbia Univ. ) Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 2

Lay-out of DAEd. ALUS experiment. Three proton sources are used to deliver neutrino at a >100 kton water Cerenkov detector placed at 1. 5 km underground The duty factor is flexible, but beams must be off for 40% of time to measure background 25% DF 10% DF Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 3

Cyclotrons Are a Viable Technology PSI is current world power leader in this energy range ~ 1. 3 MW average, 590 Me. V protons Higher power face with two major problems: 1. Capture of more beam current… space-charge at injection 2. Clean and safe extraction… beam losses <200 W (~10 -4) Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 4

Proposed Solution: Acceleration of H 2+ ions • Two protons for every ion (1 em. A = 2 pm. A) • Perveance of 5 em. A H 2+ at 35 ke. V/amu is the same as 2 em. A of 30 ke. V protons. Axial injection of 2 em. A protons at 30 ke. V is within state of the art. • The electromagnetic field to dissociate H 2+ is higher than for H-, magnetic field as high as 5÷ 6 Tesla are permitted; • Advantage of extraction by stripping foil: – Clean turn separation at extraction is not necessary; – multiple beams can be extracted simultaneously! Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 5

Main Downtime Causes - electrostatic elements - controls problems - cooling/site power - RF not prominent! Performance 2009 Reliability: 89. 5% Beam trips: 25. . 50 d-1 PSI-HIPA operational data 2009, courtesy of M. Seidel PSI Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 6

The base cyclotron module for DAEDALUS will deliver 10 m. A proton beam @ 800 Me. V, duty cycle 25%, average power <2 MW> The beam dynamics and the related technical problems for the two accelerators have been investigated: Peak current 5 m. A of H 2+ < 1. 25 m. A> H 2+ 60 Me. V/n <150 k. W>/600 k. W peak Space Charge effects and Electrostatic Deflectors Injector cyclotron, compact & resistive Superconducting Coils, Losses due to residual gas < 1 m. A> H 2+ 800 Me. V/n 25%<2 MW> 8 MW peak Stripping extraction Main cyclotron, separated sectors superconducting Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 7

Extraction lines Stripper foil Multiple beam extractions Stripper foil Last equilibrium orbit Stripper foil e k I li it RF cav S P y 1 st equilibrium orbit Double gap cavity, Vmax=250 k. V Outer diameter 14. 5 m Injection line Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 8

Superconducting Ring Cyclotron Parameters Einj <Rinj> 60 Me. V/n 1. 94 m Emax <Rext> 800 Me. V/n 4. 85 m <B> at Rinj Pole Gap Hill width Outer radius 1. 07 T 70 mm <29° 7. 3 m <B> at Rext Bmax Sector height Sector weight 1. 88 T 4. 72 T <6 m 830 Tons Flutter 4 Cavities type 1. 4 1. 97 Pill Box (PSI) Sectors N. 2 Cavities type 6 Double gap RF 49. 2 MHz Harmonic 6 th V-peak DR at Rinj 1000 k. V > 18 mm DE/turn DR at Rext 2. 5 4. 8 Me. V 2. 3 mm Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 9

Stripper extraction by carbon foil: - 2 MW beam @ 800 Me. V crossing a stripper foil 1 mg/cm 2 thick release 20 W due to nuclear interaction! - The electrons removed by the strippers have a power of ≈ 550 W Electrons are the main source of stripper damage, but we can remove it! If B=0. 2 T Re=9 mm electrons catcher H 2 + beam Peak current 20 m. A proton, <I>=4 m. A Safe limit < 2500 K Stripper foil emerging protons Courtesy I. Okuno, Riken Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 10

Beam losses due to stripping with the residual gases along the acceleration are affordable: H 2+, I=2. 5 m. A, 4 MW losses <90 W <90 W !!! High energy gain/turn is useful to reduce beam losses present simulation <3 Me. V/turn> Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 11

RIKEN RSC Courtesy Iroki Okuno, RIKEN Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 12

Control Dewar Side Shield Superconducting (Open for mainte. ) Bending Magnet Upper Shield Upper Yoke Side Yoke Lower Yoke SC Main Coil rf-Cavity SC Trim Coil Lower Shield Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 13

RIKEN Sector Magnet - - Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 14

Preliminary study of superconducting magnet and cryostat made by J. Minervini Group, @ MIT-PSFC ar. Xiv. org > physics > ar. Xiv: 1209. 4886 Sector of DAEd. ALUS SRC Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 15

Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 Courtesy F. Meot & Malek Haj Tahar, Brookhaven 16

Vertical beam envelope B 2 without gradient Radial beam envelope B 2 with 0. 68 k. Gauss/cm gradient Injection Main Parameters: Large beam emittance at injection (60 AMe. V) 37 p mm. mrad (13 times the measured emittance @ ion source). Low magnetic gradient of 0. 68 k. Gauss/cm, only on B 2 Courtesy F. Meot & Malek Haj Tahar, Brookhaven Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 17

DAEd. ALUS Superconducting Ring Cyclotron Space charge effects are negligible during acceleration in the ring cyclotron Vertical beam size along the acceleration in the radial range from 4 to 4. 9 m, snapshot at 0° azimuth. The left figure has no charge space effects, 0 m. A. The right figure is evaluated with a 5 m. A beam H 2+ current (10 m. A proton) Simulation made by J. Yang and A. Adelman (PSI), using OPAL code Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 18

0. 67 nsec 2. 3 mm 4 mm Histogram of 5 m. A H 2+ beam at the stripper foil position, simulation include space charge effects (OPAL code) Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 19

Request for future Meson Factories (Cywinski et Al. , doi: 10. 1016/j. physb. 2008. 11. 203 Physica. B(2009), doi: 10. 1016/j. physb. 2008. 11. 203) (i) A pure pulsed mode (ideally at 25 k. Hz, T=40 microsec. ), with power >500 k. W (ii) An electrostatically tailored pulse mode (e. g. 5 ns, 25 k. Hz), with power >50 k. W (iii) A quasi-CW mode, with power > 2 MW (i) If macro pulse is 4 msec long and 36 msec off (25 k. Hz), duty cycle is 0. 10, Peak current=10 m. A average power 0. 8 MW! (ii) The SRC, here presented, deliver a train of 1 nsec width pulses , with a period of 20. 3 nsec (49. 2 MHz), so selection of single bunch is feasible. 1 pulse in a 20 nsec period with repetition rate of 25 k. Hz duty cycle 0. 0005, Ipeak=10 m. A 4 k. W, power limit is due to current limit of H 2+ sources. Can be achieved simultaneously at the previous mode on independent lines! (iii) This request can be satisfied quite easily by the described SRC operated in cw with average power up to 5 MW in a single beam line or 4 MW at two beam lines Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 20

The Injector Cyclotron • 60 Me. V/amu, peak current 5 m. A H 2+ • Normal conducting coils ~ 4. 4 meter coil diameter • Axial injection (spiral inflector) 4 RF Cavities, with Voltage in the range 70 -250 k. V, 49. 2 MHz • 2 Electrostatic Deflectors Vmax=55 k. V ar. Xiv: 1207. 4895 Proposed to drive Iso. DAR experiment to search for sterile neutrino at Kamioka Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 21

1 Me. V/n 61. 7 Me. V/n Vertical beam size vs. turns number for different beam current Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 22

Deflector septum 0. 5 mm thick Extraction efficiency 99. 98%, if beam power is 600 k. W on the septum 120 W! Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 23

Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 24

A Test Stand was installed at Best Cyclotron System Laboratory to investigate the injection problems of high intensity H 2+ beam VIS Solenoid Cyclotron Quads. Injection Voltage 60 ke. V; H 2+ current injected 7 m. A; 4 s en< 1 p mm. mrad; E. S. Inflector worked at 10 k. V. Emittaance meter Buncher DCCT Beam Stop - Steerer Injection efficiency through the spiral inflector >95% with H 2+ current of 7 m. A The spiral inflector was tested to operate up to voltage higher than ± 13. 5 k. V Solenoid Requested RF voltage 70 k. V. achieved < 60 k. V! Vacuum Pump Experiment supported also by INFN, CSN 5 Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 25

A new Ion Source and an injection beam line for H 2+, including a RF buncher will be build at MIT to deliver I>40 m. A of H 2+ Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 26

H 2+ current 5 m. A Current delivered by the H 2+ ion source: 50 m. A acceptance efficiency 10%, without buncher With buncher the current requested from the source could be reduced at 30 -40 m. A 3 rd 5 -8 m. A at post 1. 7 - 2. 8 k. W @ 350 ke. V Liner Serious problem at the injection due to the space charge effect? But for sure we have thermal problem Liner Dee 85% beam lost @ injection 7 -12 m. A at 2 nd post 1. 4 -2. 4 k. W @ 200 ke. V Dee Beam from SI Backintegrated orbit It is mandatory to replace copper with other materials! Liner Tungsten, tungsten carbide or GLIDCOP? 17 -27 m. A at 1 st post 1. 7 -2. 7 k. W @ 100 ke. V Liner Dee Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 27

Source Ion Source Vacuum pump Focusing lens Spiral inflector Median Plane Vacuum pump RFQ 90° Dipole magnet Ion Source Injection into Cyclotron by a RFQ, first proposed by R. Hamm (Jacow, Cycl. Conf. 1981, pag. 359), reduce greatly the beam power lost in central region. Cyclotron yoke RFQ is able to bunch up to 90% of the injected beam into a bunch length of ≈20° RF. Build a RFQ with working frequency of 49. 2 MHZ is feasible. Ion source able to deliver H 2+ current up to 10 -15 m. A are available. Beam current in excess of 8 m. A of H 2+ could be injected inside the acceptance phase of the cyclotron. The power lost in the central region Huddersfield could be reduced at 2015 values < 500 W! Workshop on "Future Muon Sources", 12 -13 January

Summary • H 2+ ions can be a key to high-power cyclotrons for many applications • Compactness and relative lower cost of cyclotrons (<130 M€ for the whole system) could be a real option for Future Muon Sources • Exciting times ahead! Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 29

Thanks for your attention! Workshop on "Future Muon Sources", Huddersfield 12 -13 January 2015 30