ALPI and PIAVE Status and perspectives. Anna Maria Porcellato Lab. Naz. INFN Legnaro (PD) Italy

Transcription

1 ALPI and PIAVE Status and perspectives Anna Maria Porcellato Lab. Naz. INFN Legnaro (PD) Italy

2 Istituto Nazionale di Fisica Nucleare Laboratori Nazionali di Legnaro ALPI Complex CN accelerator Tandem XTU High Energy Physics Building

3 Contents Evolution of ALPI - The resonators: Pb/Cu, full Nb, Nb/Cu QWRs Status of PIAVE, the ALPI Positive Ion Injector - The resonators: full Nb QWRs, RFQs ALPI performance and upgrading

4 Tandem+ALPI and the PIAVE-Injector Upgraded Under Commissioning SC Booster ALPI Positive Ion Injector PIAVE ECRIS Alice 350kV platform EX. Hall 3 Reliable XTU-Tandem EX. Halls 1 and 2

5 ALPI Evolution ALPI: Linear Accelerator for Heavy Ions Funded in 1988 (97 Pb/Cu cavities, operating at 3 MV/m) Medium β section completed in 94 (44 QWR, Pb/Cu, 160 MHz) 1 st high β cryostat installed at the end of 1995 (Nb/Cu, 160 MHz); 2 nd and last in low β cryostats on line between 1996 and 1999 (12 QWR, Nb, 80 MHz) First beam on target in May 1994; Since then ALPI is operating according to the PAC schedule.

6 Tandem+ALPI Complex XTU-tandem: 15 MV routinely available, 30 8 MeV/u for H 127 I; 70% beam-time AVAILABLE ALPI: Linear superconducting QWR accelerator; ~ 40 MV equivalent voltage for 12 C 127 I; 30% b-time DATA Accelerated beams in ALPI (2) 8 p-na on target with A=100 masses available 700 ps (FWHM) bunch length on the exp. target 12 C 240 MeV (20 MeV/u), 82 Se 630 MeV (7.7 MeV/u) TO BE NOTED Completed: Energy Upgrade Programme In progress: Programme for efficient cooling of the first 3 cryostats

7 XTU-ALPI: Beam on Target I : :5500 time (years) Beam on Target (hours)

8 Accelerated beams in ALPI (1) Date Ion Cavities β in β out Ε/q/cavity Output Energy Output Energy [MeV/Ch-u/cav.] [MeV] [MeV/u] 21-Feb S 9, Feb S Mar Ni Mar Ca Nov S 9, Nov Ni 6, Nov Se 11, Feb Ni Feb Ni Feb Se 11, Mar Se 11, Jun S Jun Se 11, Jul S e6, Jul Cu 6, Nov Fe Nov C

9 Accelerated beams in ALPI (2) Date Ion Cavities β in β out Ε/q/cavity Output Energy [MeV/Ch-u/cav.] [MeV] 20-Feb Feb-02 2-May Mar May-02 1-Jun Jun Oct-02 8-Nov Nov Nov-02 1-Dec-02 5-Dec Dec Dec Dec-02 8-Jun-03 4-Jul-03 3-Nov Nov C O Ca Ca Ni 11, O 7, S 9, S 9, Se 11, C Ca Ca O Ru 12, S C Ni Ca O Se 11,

10 Accelerated beams in ALPI (3) Date Ion Cavities β in β out Ε/q/cavity Output Energy [MeV/Ch-u/cav.] [MeV] 19-Feb-04 5-Mar Mar Mar Mar-04 1-Apr Apr Apr Cr Ca +10, Cr Ni S Ni Se Zr

11 ALPI-layout CR12 CR13 CR14 CR15CR16 CR17 CR18 CR19 CR20 B4 E X. H A L L B3 COLD BOX 3 CR10 CR9 CR8 CR7 CR6 CR5 CR4 CR3 B2 β=0.11, 160 MHz,Nb/Cu (Pb/Cu) β=0.056, 80 MHz, full Nb. P I A V E β= MHz Nb/Cu. TANDEM EXPERIMENTAL HALLS 1, 2

12 2004 ALPI currents on target Ion [pna] on targetlinac transmissioni target/i source 36S Ni Se Zn Ni

13 ALPI medium β Pb/Cu resonators 50 installed resonators from 1992 to 1994 Pb/Cu, 160 MHz Average =2,68 MV/m Four cavities still installed in the bunching cryostats

14 9 Ea [MV/m] ALPI Mar 1995 Pb/Cu Nb Nb/Cu CR4 low beta CR5 low beta CR6 low beta CR7 CR8 CR9 CR10 CR12 CR13 CR14 CR15 CR16 CR17 CR18 CR19 CR20 (high beta) Cryostat Performance of ALPI accelerating cavities in 1995; four other cavities are installed in the bunching cryostats

15 QWR Nb sputtering at Legnaro 1988: a research project on QWR Nb sputtering at LNL starts (Palmieri) 1991: first sputtered prototype; 1993: three prototypes reach 6 7 W dissipated power 1995: Four high β resonators installed in ALPI (4 7W); process improvements allows to reach 7 7W 1998: Four new cavities, still on line, operate at 6 7W 1998: The substitution of Pb with Nb allows to reach 4 MV/m in a medium β ALPI resonator 2003 Upgrading of ALPI medium β resonators completed.

16 ALPI status in 1999 Medium and high β sections operational; Low β section installed but not usable; Improvement in medium β cavity performance obtained by replacement of Pb film with Nb; Leaks in the cryogenic and beam line valves. Why not associate cryostat maintenance with cavity upgrading? ALPI medium β upgrading possibility at: - Low cost, without necessity of devoted resources - No interference with ALPI operation Better ALPI reliability Use of sputtering technology developed at LNL

17 Cryostat maintenance Change of the leaking cryogenic valve (external actuator) Change of the leaking Viton sealing in the cryostat beam line valves, from resonator integral conditioning (changed + shielded through stainless steel rings) Gaskets in the He circuits (changed + fixed by silver plated screws) RF lines (Mechanical adjustments + Cu re-plating) Beam damage around beam ports: in the mylar wrapping of the thermal shield (shielding of aluminated Mylar by tantalum protection) Upgrading of the resonators

18 The increase in performance by Nb/Cu sputtering Ea at 7 W [MV/m] Out of frequency Installed Nb/Cu resonators To be repaired Removed Pb/Cu resonators Sep03 Apr 99 Jan 01 Dec 00 Jan 01 Jan 03 May 02 Jul 03 Oct 00 Mar 02 Oct 01 Mar 01 Rf line damaged Feb CR7 CR8 CR9 CR10 CR12 CR13 CR14 CR15 CR16 CR17 CR18 CR19 (high cryostat Beta) CR20 (high Beta) The accelerating fields reached by Nb/Cu resonators at 7 W (blue bars) are compared with the ones obtained by the same resonators when still Pb plated (black bars). Two cavities are not in operation. Black arrows indicate a decrease in performance (June 2003) with respect to previous results

19 Pb Nb Results (operational) E a (@7 W): from 2.4 (Pb) to 4.4 MV/m (Nb) Despite some limitations in the copper base (not best Cu quality, sharp edges & Sharp holes in high-i regions, hidden volumes in brazed joints) No interference with ALPI operation Same cryostat, control system, rf hardware and software 4.4 MV/m 790 kev/u (@ β=β op ;Φ=0)

20 Nb/Cu resonator advantages Relatively high accelerating fields High mechanical stability Frequency not affected by changes in the He bath pressure (<O.01 Hz/mbar!) Locked up to 7 MV/m without necessity of fast or soft tuners Very reliable Easy to put into operation No degradation with time after installation

21 High β section: ALPI sputtered Nb/Cu resonators Q 1.00E E+09 1W 3W 7W 1.00E+08 CR20-1 CR20-2 CR E+07 CR beam port Ea [MV/m] pick-up coupler ALPI β=0.13, 160 MHz resonator views; The cavity has no brazed joint; beam ports are jointed to resonator by indium gaskets

22 Low β section Ea [MV/m] Bulk Nb resonator perforamance 0 CR04-1 CR04-3 CR05-1 CR05-3 CR06-1 CR06-3 Cavity Ea av May 2003: 6.75 MV/m Nb, double wall, 80 MHz Excellent performance, no degradation along years Installed in between Their cooling down was difficult, upgrading of cryogenic line necessary and scheduled (spring 2004) Sensitivity to He bath pressure drift: 1Hz/mbar

23 B3 CR12 CR13 CR14 CR15CR16 CR17 CR18 CR19 CR20 COLD BOX B4 E X. H A L L 3 CR10 CR9 CR8 CR7 CR6 CR5 CR4 CR3 B2 β=0.056, 80 MHz, full Nb β=0.110, 160 MHz,Pb/Cu β=0.11, 160 MHz,Nb/Cu β= MHz Nb/Cu TANDEM ECR EXPERIMENTAL HALLS 1, 2

24 He transfer to the lower β branch 1. More total P for cooling (30 70 mbar), refrigerator and cryostats valves changed (done in 07/03) Liquid Gas 2. More He Flux 2 4 compressors (160 to 205 g/s) (received 09/03) Main Cold Box 4.5 K 7 K 3. Equalization of hydraulic routes to various cryostats (scheduled in 04-05/04): P ~ K on each cryostat 3. Equalization of hydraulic routes to cryostats (similarly for thermal shield gaseous He)

25 Some work on the mechanical tuners needed : Installation of the 3 lower-β cryostats 1998: the CR06 cavities, locked on line at up to an energy of ~ 6 MV/m ( 58 Ni shift) - fast f 0 changes: no problem - slow f 0 changes: (He P < 10 mbar/min kept for days) compensated by the slow tuner in f 0 feedback mode (poor reproducibility of tuner mechanics noted). The low-β cavity commissioning was interrupted (cryogenics problems) May 2003: the linac operation team can test all 12 cavities at the same time: a) cavity performancen is very good; b) fast f 0 changes OK; c) mechanical tuners could not compensate for P variations of mbar/min (mechanical faults) November 2003 January 2004: systematic tests on all the resonators (chaired by the resonator designer), to fix the slow tuning problems April 2004: Systematic test on PIAVE QWR with the old tuner: the He pressure fluctuations are lower than in ALPI. The cavities can be locked adjusting continuously the frequency by the old tuners while the He pressure changes up to 50 mbar/minute May 2004: Four cavity equipped with new design tuners, backlash free, are installed in ALPI.

26 1. Fast f 0 changes (acoustic vibrations, Hz): the dampers keep the frequency oscillations within a few Hz (no need of fast tuners) - checked in March ALPI cryogenic plant has drifts in He bath pressure: - it was designed for Pb/Cu resonators whose sensitivity to change in bath pressure is < 0.01 Hz/mbar - it has to feed both LHe circuit and the thermal shield circuit operating at 7 bar, thus making regulation difficult 0.23bar Pressure 28/05 drifts dalle in ALPI alle cryostats Frequency locking: keep the QWRs at 80 MHz ± 2 3 Hz 3. Slow f changes (He-P variations 0 seconds minutes): a slow tuner following the pressure fluctuations is required, the present one worked in PIAVE, where f are lower, but is not enough reliable in ALPI; a new type, based on leverage, is under test in ALPI B2 CR4 CR5 CR6 CR7 CR8 secondi Some mechanical improvement needed

27 ALPI output energy Tandem injector ALPI output [MeV/u] ALPI Dec 1998 ALPI 2003 (low beta not included) ALPI 2003 (low beta included) 12 C Si Ca Cu 17+ A 0 16 O S Ni Br Zr I Tandem 15 MV; F,F 1998 : 11 Pb/Cu cryostats 2003: 13 Nb/Cu cryostats ALPI output Energy Gain: 45% for 6 C;69% for 127 I Linac Energy Gain: 91% for 6 C;132% for 127 I (low β not included)

28 TTFn ALPI output energy once added 6 more Cryostats Zr 13,17+ Tandem at 15 MV TTFn ALPI output energy [MeV/u] ALPI output energy [MeVu] Cavity # TTFn C 6+ Tandem at 15 MV TTFn ALPI output energy [MeV/u] ALPI output energy [MeVu] Cavity #

29 PIAVE Status

30 Why to switch from a Tandem to a q+ injector? E/A [MeV/u] ALPI Output with the two Injectors Tandem Tandem+ALPI (G-F) (1-10 pna) PIAVE+ALPI ( pna) New Injector A MORE CURRENT HEAVIER MASSES MORE BEAM TIME AVAILABLE (TWO INJECTORS)

31 PIAVE design values for RFQs Reference beam: 238 U 28+ (pre-bunched) from an ECR on a 315 kv platform SRFQ1 SRF2 In Out In out Energy KeV/u MeV Beta Voltage kv Length cm N of cells m a cm R cm φ s deg E p,s MV/m U J Acceptance (norm) 0.8 mm mrad Output long. emittance 0.7 ns kev/u

32 PIAVE Injector Layout Two SRFQs ECR + LEBT Eight QWRs to ALPI

33 q/m 28/238 Energy Range kev/m β range Current 1 eµa T. Emittance 0.5 mmmrad (norm) L. Emittance < 0.7 nskev/u PIAVE the new injector Typical Beams ECRIS Element q I (eµa) 16 O Ar Kr Xe Cu Ag SRFQs QWRs In Feb 2003 a small fire due to a short circuit forced to stop ECR commissioning to SC booster New start-up of the ECR Ion Source in March 2004 in the previous conditions

34 Beam Commissioning from ECRIS to SRFQs Transmission: 89% Buncher t < 500 ps, no transverse steering induced Matching: obtained (as computed) at the SRFQ1 entrance

35 Two superconducting RFQs in a cryostat SRFQ1 SRFQ2 beam line 1.38 m 0.74 m 37 kev/u (β = ) 585 kev/u (β = 0.035)

36 SRFQ2 structure Nb tank, full Nb, RRR 250, 3 mm thick 0.74m Trapezoidal-shaped stems, the vertical ones displaced with respect to horizontal ones Modulated electrodes, rod-type, carved inside Empty, baking cylinders 0.8m Ti Stiffening ribs Ti /Nb end flange

37 Milestones on SRFQs SRFQ1 SRFQ2 Frequency MHz Length m Diameter m Weight Kg V_interelectrode kv Modulated cells E s,p MV/m E s,p /E a B s,p T Stored Energy J P diss (set) W Q 1x10 8 2x10 8 Performance Q 0 vs E p : both above specs Both installed in the line cryostat Leak check of cryogenic circuit OK Locking conditions Tested both in the test cryostat and on the beam line for relativily short times (hours): Slow f 0 changes and their control Fast f 0 changes and their control

38 SRFQ1 - SRFQ2: performance achieved SCRFQ1 SCRFQ2 1.E W Q 1.E W 1.E+07 SRFQ2 SRFQ Peak Surface Field [MV/m]

39 Frequency shift induced by He bath pressurization F = 39 Hz/mbar for RFQ2 F = 48 Hz/mbar for RFQ1-12 khz -8 y = x frequency shift [khz] P [mbar]

40 How do we cope with slow drifts? By deforming both tuning plates in frequency feedback mode One for + f, the other for f (no mechanical backlash) Tuners performance Frequency range: khz (huge) Sensitivity: 0.5 Hz (good) Speed: ~ 2 3 Hz/s Cryogenic-Plant Specs P=1.2 ±0.05 bar; P/ t < 2 mbar/min 1.33 Hz/s In the test cryostat: Margin: factor 2 f < 1 khz/day, at < 1 2 Hz/s IMPORTANT FOR HE IN REFRIGERATION CYCLE NO PROBLEM!

41 On line test April 2004 Conditioning: faster than in the test cryostat. The couplers located on the cavity bottom probably helped rf antennas: OK both for Q and movement Cavity Q similar to the ones measured in laboratory for both RFQs (if the dissipative load coming from the fast tuners is taken into account) SRFQ2: locking condition (20 Hz bandwidth) maintained for up to 60 minutes, lost for He pressure changes out specification or sometimes for not optimal tuning operation Fast tuners: increase the locking window from from 20 Hz to 200 Hz (SRFQ2) and 70 Hz (SRFQ1, to be increased). Microphonics at negligible level: unlocking at 80 MHz (master) due only to He pressure drifts. Slow tuners will work together with the fast tuners in the next test.

42 SRFQ2 locking to the master oscillator (80 MHz) P*10 [mbar], Ti [K], He level*10 [%] % E acc,n Tuner 2 not properly working 84 % E acc,n 21.4 mb/min 3.6 mb/min Time [min] (84 % E acc,n A/q = Pb 29+ e.g.)

43 Low-β QWR s for PIAVE PIAVE 80 MHz, β=0.047 QWR 80 MHz, low- β QWR cryostat 1.00E+10 PIAVE bulk Nb QWR's Q at 7W Mechanical damper High gradient No fast tuner required Built and tested Installed and operational 1.00E+09 Q 1.00E E+07 new specifications old specifications Ea (MV/m)

44 2 ALPI low-β section: Cavities locking to the linac clock to be improved To be cooled efficiently 4. CRYOGENIC SYSTEM UPGRADE 3. LOW BETA CAVITIES UPGRADE PIAVE 1. ECRIS OPERATION 1

45 On line tests of PIAVE low-β QWR s Good performance for CrQ1-PIAVE resonators:locked up to 5 MV/m making use of the old type slow tuner controlled by an improved version of the soft tuner program Q values lower than the ones measured in laboratory for CrQ1-PIAVE resonators; two cavities of the cryostat out of frequency; microphonic noise more relevant than in CrQ1- PIAVE resonators. Not negligible power consumption at 4 K due to the rf dissipation in the RF input line in strong overcoupling condition used to make locking possible.cryostat removal from the beam line to fix the problems foreseen soon. The resonators should recover the performance by HPWR

46 Coming next PIAVE beam test foreseen in October 2004 Final test of all the automatic procedures on the refrigerator, operating with the whole cryogenic load, scheduled in September 2004 Beam injection of a PIAVE beam in ALPI foreseen by the end of the year Improvement of ALPI cryogenic system to allow operation of low β resonators completed by July 2004 Substitution of old tuners with new designed ones in low β resonators in progress.

47 Present ALPI output energy Tandem injector ALPI output [MeV/u] ALPI Dec 1998 ALPI 2003 (low beta not included) ALPI 2003 (low beta included) 12 C Si Ca Cu 17+ A 0 16 O S Ni Br Zr I Tandem 15 MV; F,F 1998 : 11 Pb/Cu cryostats 2003: 13 Nb/Cu cryostats ALPI

48 ALPI injected by PIAVE Performance of different ECR sources compared with LNL negative source beam intensity PIAVE-ALPI output energy when fed by typical ECR beams Present ALPI authorization limit Constraints ALPI Completion Performance of ALPI after completion Reachable ALPI Energy stripping the ALPI beam

49 ECRIS currents, compared to LNL negative source Ion A state of charge [LNL ALICE I [Atomki]1 I [S.nanogan] 2 IKF SERSE I [Argonne ECRIS] I after Tandem Stripping e micro A e micro A e micro A e micro A e micro A e micro A e micro A C N N O O O F F Ca Ca Ar Ar Ar Ar Fe Fe Fe Ni Ni Ni Ni Ni Cu Cu Ge No isotope separation M. Cavenago, private communication

50 ECRIS currents, cont Ion A state of charge [LNL ALICE I [Atomki]1 I [S.nanogan] 2 IKF SERSE I [Argonne ECRIS] e micro A e micro A e micro A e micro A e micro A e micro A Kr Kr Kr Kr Kr Kr Zr Ag Sn Sn Xe Xe Xe Xe Xe Xe Ta Pb Au Au U U U No isotope separation M. Cavenago, private communication

51 Energy output PIAVE injection, 2004 ALPI MeV/u C O N Ar Ni Kr Sn Xe Ta Au Pb U High current High energy C O N Ar Ni Kr Sn Xe Ta Au Pb U A PIAVE: - 2 SRFQs at the design accelerating fields -8 low β resonators at 5 MV/m ALPI - 12 low β resonators operating at 4.4 MV/m - 44 medium β resonators operating at 4.4 MV/m - 8 high β resonators operating at 5.5 MV/m

52 Present authorization limits For ions from Si to Pb: E < 20 MeV/u I < -30 pna on target For ions from C to Al E< -26 MeV/u I<2 pna on target

53 Constraints ALPI optics was designed for B*ρ=3.3 Tesla*m An accurate beam transport analysis in ALPI is required The RFQs were designed for 5µA; the spatial charge effect must be evaluated for larger current (negligible at 5µA) Maximum possible A/Q for the PIAVE RFQs is 8.5 RFQ transmission is 45%. At high current the beam losses can produce a not negligible thermal load (0.2W for 1 µa of 238 U).

54 ALPI completion The original ALPI layout foresaw further 6 high β cryostats. The cryogenic lines, magnets, diagnostic box are installed. Only the cryostats and their equipment (cavities, vacuum, rf controller and amplifier ) are missing. The insertion of a low β cryostat is also possible in the low energy branch (useful in case of very heavy ions). A cryostat is in this case available. It is necessary to verify that the cryogenic system can provide the design power (1.3 K and K) It is necessary to study in detail the beam optics for the different beams

55 Upgraded ALPI Energy output PIAVE injection MeV/u C O N Ar Ni Kr Sn Xe Ta Au Pb U C O 6+ High current High energy High energy (Bro>3.5) 14 N Ar Ni Kr Sn Xe Ta Au Pb U A Upgraded ALPI: ALPI high β resonator operating at 5.5MV/m

56 Increasing the energy by beam stripping The final section of ALPI magnetic lattice can not transport the heavier beams at maximum reachable energy. To decrease the beam rigidity and increase the reachable beam energy it is possible to strip the beam in ALPI. Two options are possible: stripping in the diagnostic box before ALPI U band or stripping in the box after it In the first case you can select better the beam accelerated in the second linac branch, making radioprotection issues less demanding In the latter case it is possible to have instead multicharge acceleration in the second linac branch, increasing in such way ALPI output current The output energy is similar in the two cases so only the first option is analyzed

57 ALPI-layout CR12 CR13 CR14 CR15CR16 CR17 CR18 CR19 CR20 B4 E X. B3 Stripper position COLD BOX H A L L 3 CR10 CR9 CR8 CR7 CR6 CR5 CR4 CR3 B2 β=0.11, 160 MHz,Nb/Cu (Pb/Cu) β=0.056, 80 MHz, full Nb. P I A V E β= MHz Nb/Cu. TANDEM EXPERIMENTAL HALLS 1, 2

58 ALPI output energy Beam stripped before the U band MeV/u ALPI 2004 Upgraded ALPI Ca 10, Ni 14, Kr 18, Sn 16, Ta 24, Pb 25, Ar 8, Fe 13, Ge 13, Zr 17, Xe 17, Au 29, U 28, A Upgraded ALPI: ALPI high β resonator operating at 5.5MV/m