Скачать презентацию Study of Rare Kaon Decays at the CERN-SPS
1a70a7aaf2493c098abaf44fb89209bb.ppt
- Количество слайдов: 113
Study of Rare Kaon Decays at the CERN-SPS Villars, 27 September, 2004 A. Ceccucci for the B. NA 48 -Future Working Group September 27, 2004 A. Ceccucci/ CERN Villars 1
“CERN Director General Outlines Sevenpoint Strategy for European Laboratory” 18. 6. 2004 Official CERN Press Release Geneva 18 June 2004. At the 128 th session of CERN Council, held today under the chairmanship of Professor Enzo Iarocci, CERN Director General, Robert Aymar, outlined a sevenpoint scientific strategy for the Organization. Top of the list was completion of the Large Hadron Collider (LHC) project with start-up on schedule in 2007. This was followed by consolidation of existing infrastructure at CERN to guarantee reliable operation of the LHC, with the third priority being an examination of a possible future experimental programme apart from the LHC. ………. September 27, 2004 A. Ceccucci/ CERN Villars 2
NA 48 -Future Working Group CERN-SPSC-2004 -010 SPSC-EOI-002 • We identified the Rare Kaon Decays as the next logical step of the Kaon Programme at CERN • Short term (2004 -2010): NA 48/3 K+ → p+ nn **LETTER OF INTENT** • Longer term (>2010): – Assuming a new or refurbished SPS capable to deliver higher intensity/energy as the ultimate injector for LHC 0 ee ( NA 48/4 KL ) 0 NA 48/5 KL →p mm →p nn September 27, 2004 A. Ceccucci/ CERN Villars 3
CP-Violation Kaon Rare Decays and the SM NA 48/3 NA 48/5 (holy grail) |Vtd| NA 48/4 CP-Conservation Kaons provide quantitative tests of SM independent from B mesons September 27, 2004 The Physics case was reviewed by G. Isidori The state-of-the-art was reviewed by L. Littenberg A. Ceccucci/ CERN Villars 4
Letter of Intent to Measure the Rare Decay K+ → p+ n n at the CERN SPS Cambridge: D. Munday; CERN: N. Cabibbo, A. Ceccucci*, V. Falaleev, F. Formenti, B. Hallgren, A. Gonidec, P. Jarron, M. Losasso, A. Norton, P. Riedler G. Stefanini; Dubna: S. Balev, S. Bazylev, P. Frabetti, E. Goudzovski, D. Gurev, V. Kekelidze, D. Madigozhin, N. Molokanova, R. Pismennyy, Y. Potrebenikov, A. Zinchenko; Ferrara: W. Baldini, A. Cotta Ramusino, P. Dalpiaz, C. Damiani, M. Fiorini, A. Gianoli, M. Martini, F. Petrucci, M. Savrie’, M. Scarpa, H. Wahl; Firenze: E. Iacopini, M. Lenti, G. Ruggiero; Mainz: K. Kleinknecht, B. Renk, R. Wanke; UC Merced: R. Winston; Perugia: P. Cenci, M. Piccini; Pisa: A. Bigi, R. Casali, G. Collazuol, F. Costantini, L. Di Lella, N. Doble, R. Fantechi, S. Giudici, I. Mannelli, A. Michetti, G. M. Pierazzini, M. Sozzi; Saclay: B. Peyaud, J. Derre; Sofia: V. Kozhuharov, L. Litov, S. Stoynev; Torino: C. Biino, F. Marchetto *contact September 27, 2004 A. Ceccucci/ CERN Villars person 5
Main K+ decay modes competing with K+→p+ nn Decay m+ n p +p 0 p +p +p p+p 0 p 0 p 0 m + n p 0 e + n BR 63 % 21 % 6% 2% 3 % (called K+m 3) 5 % (called K+e 3) Suppression: m PID, kinematics g veto, kinematics CHV, kinematics g veto, m PID g veto, E/P BR(K+→p+ nn)~10 -10 !! September 27, 2004 A. Ceccucci/ CERN Villars 6
Framework • So far K+→p+ nn only studied with kaon decays at rest – This limits the statistics to a few events • We plan to collect ~100 events at the SPS by 2010 – We dubbed this initiative NA 48/3 – the name is not an issue at this early stage • Employ high energy kaons has the following advantages: – The larger cross section increases the kaon content in the beam – The rejection of backgrounds from K+→ p+p 0 is simplified • Tens of Ge. V of EM energy is deposited in the photon vetoes! – Accidental background are minimised • The use of unseparated beam becomes a possibility • 2/3 of the final state is invisible !! – The kaon and the pion must be redundantly measured to keep backgrounds under control – Muon and photon vetoes are essential September 27, 2004 A. Ceccucci/ CERN Villars 7
Kinematics Region I Re gio September 27, 2004 n. I I A. Ceccucci/ CERN Villars 8
Region I September 27, 2004 75 Ge. V/c Acceptance A. Ceccucci/ CERN Villars Region II 9
THE BEAM September 27, 2004 A. Ceccucci/ CERN Villars 10
Rationale p 0 = primary proton momentum pk = secondary beam momentum • Kaon production increases as p 02 – Use highest p 0 (that is 400 Ge. V/c protons from SPS) • For a fixed fiducial length the number of decays increases as pk. If p 0 is fixed the maximum is for: pk = 0. 23 p 0 • For unseparated beams the limitation comes from the detectors, not from the amount of protons September 27, 2004 A. Ceccucci/ CERN Villars 11
Choice of pk=75 Ge. V/c Pk (Ge. V/c 60 K+ flux @ production ( 3 1012 400 Ge. V/c protons) K+ survival over 102 m 90 120 1. 1 1. 3 (meas) 1. 5 1. 9 2. 4 2. 3 (meas) 0. 80 x 108 75 0. 83 0. 86 0. 89 K+ / Total beam rate x 10 -2 5. 2 6. 8 (meas) 5. 5 5. 6 5. 2 4. 7 (meas) K+ decays in 50 m x 106 8. 9 10. 7 11. 6 11. 4 75 Ge. V/c is about the maximum momentum for which a beam incorporating stages for large solid angle acceptance, momentum selection, K+ tagging, beam momentum measurement and tracking using standard beam elements can fit into the present length of 102 m September 27, 2004 A. Ceccucci/ CERN Villars 12
Layout of the Beam • SPS North Area High Intensity Facility – TCC 8 tunnel + ECN 3 cavern (where NA 48 is installed) • Modified K 12 beam line (102 m long) – Achromatic design (no net bend) – Triplet of radiation hard quadrupoles to capture large solid angle – Front-end achromat: • Four radiation hard dipoles • Narrow momentum bite Dpk/pk~1% RMS – Parallel section to house a He filled CEDAR – Final quadrupoles to make beam slightly converging – Second achromat to allow for redundant momentum measurement • Optics calculated using the TRANSPORT programme September 27, 2004 A. Ceccucci/ CERN Villars 13
NA 48/3: Beam Layout Dipoles Beam-line 102 m long September 27, 2004 about 17% K+ lost A. Ceccucci/ CERN Villars 14
Decay Tank • ~100 m long decay tank closed by thin Kevlar window • A 160 mm diametre beam pipe traverses the principal detectors to allow the beam to be transported in vacuum • Double magnetic spectrometer, achromatic design, with two large gap dipoles to provide horizontal deflections of -2. 1 and +3. 5 mrad for 75 Ge. V/c beam – Charged particles of all momenta are centered on the beam axis again downstream of the e. m. calorimeter • A magnetised hadron calorimeter provides a +18 mrad deflection, displacing the charged beam 160 mm 8 m further downstream to clear a 100 mm radius photon veto at the end of the hall September 27, 2004 A. Ceccucci/ CERN Villars 15
NA 48/3: Beam Across Tank September 27, 2004 A. Ceccucci/ CERN Villars 16
Already Available New high-intensity K+ beam for NA 48/3 Present K 12 (NA 48/2) SPS protons per pulse on T 10 Duty cycle (s. /s. ) Factor wrt 2004 1 x 1012 Beam: New HI K+ > 2006 3 x 1012 3. 0 4. 8 / 16. 8 Solid angle (msterad) 1. 0 0. 40 16 40 Av. K+momentum
1 HISTOGRAM NO 21 DISTRIBUTION OF P IN GEVC 0 INTERVAL 0 LESS THAN 72. 000 0 102. 000 M FROM THE TARGET SCALE FACTOR. . 100 X'S EQUAL 5637 ENTRIES TURTLE SIMULATION MOMENTUM DISTRIBUTION 72. 000 TO 72. 200 0 72. 200 TO 72. 400 0 72. 400 TO 72. 600 0 72. 600 TO 72. 800 0 72. 800 TO 73. 000 1 73. 000 TO 73. 200 20 73. 200 TO 73. 400 178 XXX 73. 400 TO 73. 600 576 XXXXX 73. 600 TO 73. 800 1210 XXXXXXXXXXX 73. 800 TO 74. 000 2068 XXXXXXXXXXXXXXXXXX 74. 000 TO 74. 200 2870 XXXXXXXXXXXXXXXXXXXXXXXXX 74. 200 TO 74. 400 3727 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 74. 400 TO 74. 600 4598 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 74. 600 TO 74. 800 5354 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 74. 800 TO 75. 000 5637 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 75. 000 TO 75. 200 5596 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 75. 200 TO 75. 400 5126 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 75. 400 TO 75. 600 4288 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 75. 600 TO 75. 800 3400 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 75. 800 TO 76. 000 2623 XXXXXXXXXXXXXXXXXXXXXXX 76. 000 TO 76. 200 1820 XXXXXXXXXXXXXXXX 76. 200 TO 76. 400 1068 XXXXXXXXX 76. 400 TO 76. 600 552 XXXXX 76. 600 TO 76. 800 250 XXXX 76. 800 TO 77. 000 61 X 77. 000 TO 77. 200 8 77. 200 TO 77. 400 0 77. 400 TO 77. 600 0 77. 600 TO 77. 800 0 77. 800 TO 78. 000 0 0 GREATER THAN 78. 000 0 0 TOTAL NUMBER OF ENTRIES = 51031 INCLUDING OVERFLOW AND UNDERFLOW 0 MEAN = 74. 981 RMS HALF WIDTH = 0. 710 0 HISTOGRAM NO 21 DISTRIBUTION OF P IN GEVC 102. 000 M FROM THE TARGET
= 74. 981 Ge. V/c, RMS = 0. 710 Ge. V/c September 27, 2004 A. Ceccucci/ CERN Villars 18
-60. 000 -20. 000 60. 000 TOTALS I**---**---**---**---**---**---**I--------60. 000 I I 0 -54. 000 I I 0 -48. 000 I I 0 -42. 000 I I 0 -36. 000 I I 0 -30. 000 I I 0 -24. 000 I 2 AQ$$$$$$OF 1 I 397 -24. 000 I 8 T$$$$$A I 932 -18. 000 I $$$$$$ I 4016 -18. 000 I A$$$$$$A I 5407 -12. 000 I 1$$$$$$4 I 8053 -12. 000 I Z$$$$$$Z I 8326 -6. 000 I 2$$$$$$2 I 9999 -6. 000 I $$$$$$$ I 9999 0. 000 I 3$$$$$$2 I 9999 0. 000 I $$$$$$$ I 9999 6. 000 I $$$$$$ I 8186 6. 000 I W$$$$$$W I 8536 12. 000 I $$$$$$2 I 4102 12. 000 I 4$$$$$$G I 5529 18. 000 I 2 GS$$$$$$$F 2 I 396 18. 000 I 3 Q$$$$T 2 I 787 24. 000 I I 0 30. 000 I I 0 36. 000 I I 0 42. 000 I I 0 48. 000 I I 0 -60. 000 -20. 000 60. 000 TOTALS 54. 000 I I 0 I**---**---**---**---**---**---**I-------I**---**---**---**I----60. 000 I I I I -54. 000 I I 0 I 2456666542 I I 134455554431 I -48. 000 I I 0 I 631337614046 I I 15428485392461 I -42. 000 I I 0 I 7435062357861 I I 59077738177357 I -36. 000 I I 0 TOTALS I 0000668218733483200000 I 9999 TOTALS I 0000785595737100770000 I 9999 -30. 000 I I 0 -24. 000 I 4 H$$$$K 5 I 603 TOTAL NUMBER OF ENTRIES = 51031 -18. 000 I 3$$$$$$ I 5010 -12. 000 I H$$$$$$K I 8414 -6. 000 I H$$$$$$H I 9999 0. 000 I L$$$$$$H I 9999 6. 000 I A$$$$$$E I 8483 12. 000 I $$$$$$6 I 5058 18. 000 I 4 L$$$$H 3 I 590 24. 000 I I 0 30. 000 I I 0 36. 000 I I 0 42. 000 I I 0 48. 000 I I 0 54. 000 I I 0 I**---**---**---**I-------I I I 134456555431 I I 213971951312 I I 61257420138277 I TOTALS I 0000899915863673340000 I 9999 BEAM TRANSVERE SIZE SPIBES 2 SPIBES 1 FTPC (KABES) Y [mm] X [mm] TOTAL NUMBER OF ENTRIES = September 27, 2004 A. Ceccucci/ CERN 51031 Villars 19
Spot at Wire Chamber 6: 1 HISTOGRAM NO 29 HORIZONTAL AXIS VERTICAL AXIS 0 X IN MM Y IN MM -60. 000 + -60. 000 -54. 000 -48. 000 -42. 000 -36. 000 -30. 000 -24. 000 -18. 000 -12. 000 -6. 000 0. 000 6. 000 12. 000 18. 000 24. 000 30. 000 36. 000 42. 000 48. 000 54. 000 Y [mm] X [mm] 0 0 -20. 000 FROM THE TARGET 60. 000 TOTALS I**---**---**---**I-------TO -54. 000 I I 0 TO -48. 000 I I 0 TO -42. 000 I I 0 TO -36. 000 I I 0 TO -30. 000 I 2321 11 I 10 TO -24. 000 I 159 XT$$$$SJE 73 I 334 TO -18. 000 I 8$$$$$$U 72 I 1947 TO -12. 000 I 1 E$$$$$$D I 4900 TO -6. 000 I G$$$$$$J 1 I 8143 TO 0. 000 I 1 L$$$$$$I 2 I 9999 TO 6. 000 I G$$$$$$82 I 9984 TO 12. 000 I G$$$$$$E 1 I 8252 TO 18. 000 I F$$$$$$E 1 I 4901 TO 24. 000 I AV$$$$$X 7 I 1929 TO 30. 000 I 2 ERYVWQSPFF 5 I 254 TO 36. 000 I 21 1 1 I 5 TO 42. 000 I I 0 TO 48. 000 I I 0 TO 54. 000 I I 0 TO 60. 000 I I 0 I**---**---**---**I-------I I I 1356776531 I I 15731887723651 I I 16166042762880 I TOTALS I 000000027056033304565390000000 I 9999 TOTAL NUMBER OF ENTRIES = 51031 INCLUDING UNDERFLOW AND OVERFLOW AS FOLLOWS UNDERFLOW OVERFLOW ACROSS 0 0 DOWN 0 0 0 HISTOGRAM NO 29 HORIZONTAL AXIS VERTICAL AXIS September 27, 2004 204. 858 M X IN MM Y IN MM A. Ceccucci/ CERN 204. 858 M Villars FROM THE TARGET 20
Muon Halo Calculation • The flux of “halo muons” crossing the WC has been calculated using the HALO program: • Single rate in the WC is dominated by muons from kaon decays • The total halo is about 7 MHz • Thank to the new beam design, the situation appears much better than in Na 48/2 September 27, 2004 A. Ceccucci/ CERN Villars 21
DETECTORS September 27, 2004 A. Ceccucci/ CERN Villars 22
NA 48/3 Detector Layout 10 MHz Kaon decays 800 MHz (p/K/p) Only the upstream detectors see the 800 MHz beam September 27, 2004 A. Ceccucci/ CERN Villars 23
Detectors • CEDAR – To tag positive kaon identification • GIGATRACKER – To Track secondary beam before it enters the decay region • ANTI – Photon vetoes surrounding the decay tank • WC – Wire chambers to track the kaon decay products • CHOD – Fast hodoscope to make a tight K-pi time coincidence • LKR – Forward photon veto and e. m. calorimeter • MAMUD – Hadron calorimeter, muon veto and sweeping magnet • SAC and CHV – Small angle photon and charged particle vetoes September 27, 2004 A. Ceccucci/ CERN Villars 24
2004 Test beam • It was of the utmost importance to test in 2004 the performance of the NA 48 detectors at intensities comparable to NA 48/3 (no SPS in 2005!) • This was a unique opportunity to collect data to validate our – simulated- understanding to quantify the necessary effort (technical and financial) to transform NA 48 into an experiment capable to address K+→p+ nn. • Thank to the extension granted by CERN we could test: – WC: raise intensity to about 30 times NA 48/2 – GIGATRACKER • • Tested a state-of the-art ALICE SPD assembly in our beam Use a thinner 25 micron MICROMEGAS amplification gap Read out KABES with 480 MHz FADC (former NA 48 tagger FADC) Read KABES at ~14 times the NA 48/2 rate – LKR: Complement the photon coverage with extra LKr electronics and a Small Angle Calorimeter SAC (CMS RCAL prototype) – CHOD test of prototypes • A few very preliminary results will be shown September 27, 2004 A. Ceccucci/ CERN Villars 25
Increase of beam Intensity • I 0 = Intensity of NA 48/2 K+ beam – Tune the K 12 beam to + 75 Ge. V/c – Open up the aperture of the P 42 line (X 3. 5) – Opening momentum bite DP/P from 5 to 20 % (X 4) – Turn on both K+ and K- polarities (X 1. 3) – Employ a shorter T 4 target (100 mm instead of 300 mm) (X 1. 6) • Tot ~ 29 I 0 • Accidentals in 29 I 0 beam NA 48/2 are dominated by pions rather than kaons – Expect cleaner situation with new beam September 27, 2004 A. Ceccucci/ CERN Villars 26
CEDAR • CErenkov Differential Counter with Achromatic Ring Focus • • He pressure adjusted to make it sensitive only to kaons Requires beam divergency < 0. 1 mrad Built at CERN in the 80’s (Bovet et al. ) for use in the SPS beam lines We will certainly need to upgrade the photon detectors and front-end electronicsto operate at the NA 48/3 rates (~60 MHz) Beam September 27, 2004 A. Ceccucci/ CERN Villars 27
CEDAR K/p Cedar-W Cedar-N September 27, 2004 A. Ceccucci/ CERN Villars 28
GIGATRACKER • Specifications: – – – Momentum resolution to ~ 0. 5 % Angular resolution ~ 10 mrad Time resolution ~ 100 ps Minimal material budget Perform all of the above in • 800 MHz hadron beam, 40 MHz / cm^2 • Hybrid Detector: – SPIBES (Fast Si micro-pixels) • Momentum measurement • Facilitate pattern recognition in subsequent FTPC • Time coincidence with CHOD – FTPC (NA 48/2 KABES technology with FADC r/o) • Track direction September 27, 2004 A. Ceccucci/ CERN Villars 29
FTPC SPIBES 2 SPIBES 1 GIGATRACKER ♠ momentum: use SP 1 and SP 2 to measure y = 40 mm displacement. Assuming σp~50µm from pixel and 350µm thick Si (0. 37% X 0) ü σ = (σp√ 2 ‡ σMS ) ⁄ 40 mm = 0. 25% ♠ direction: use SP 2 and FTPC. Assuming σp~100µm from pixel and similar from FTPC and no MS from FTPC (from SP 2 no influence) ü ∆Өх= σp√ 2 ⁄ 12. 4 m = 11µrad Tails in the beam? (Turtle simulation) 6. 25 12. 45 m September 27, 2004 ♠ time resolution: essential to obtain a low background due to accidental hits and to allow the pattern recognition (see result from test beam). For a pixel C≈ 100 f. F a risetime ~ 2 ns should be achievable for 130 nm technology and a good S/N. A. Ceccucci/ CERN Villars. Mara Martini 30
Proposal for SPIBES Beam square shape 5 x 5 cm 2 The cooling should be studied • September 27, 2004 Should avoid a substrate • 300 x 100 µm pixel cell An effort must be done to minimize the overall thickness to ≤ 350 µm of Si without loosing in yield. • 5 cm • The dimension of the pixel cell and of the chip must be optimized to fit the 2 n rule and to match the design requirements (PA, Discri, multiplexed TDC, power consumption, r/o bus) 80000 pixels in total to cover the beam A. Ceccucci/ CERN Villars. Mara Martini 31
SPIBES: Time scale & Resources • A Preliminary time scale has been established based on the experience gained in other silicon pixel detector projects • The baseline technologies are those already adopted in the most advanced current designs further extended (0. 13 micron technology for the pixel chip probably needed) • Novel solutions may have to be worked out for the hybrid and cooling • Preliminary work should start without delay • Front-end ASIC will require contribution of several expert designers for a period of two or three years. • Good definition of the DAQ environment is needed to develop a matching back-end readout and calibration system September 27, 2004 A. Ceccucci/ CERN Villars 32
Test of ALICE pixel in NA 48/2 beam 1 ALICE assembly 1 DAQ adapter card 30 m DAQ cables 30 m JTAG control cables LV and HV power supplies VME crate with r. o. module (Pilot) and JTAG controller JTAG multiplexer MXI interface to PC ALICE PTS software (Lab. View) PC remotely controlled from NA 48 control room September 27, 2004 A. Ceccucci/ CERN Villars 33
Single Chip Alice Assembly tested Assembly 7: • 150µm thick ALICE chip • 200µm thick sensor • 1. 1 % X 0 all together Mounted on a thin test-PCB Vfd=15 V Vop=50 V Sensor 8192 pixels Produced 2003, tested in ALICE p-TB 2003 September 27, 2004 A. Ceccucci/ CERN Chip Villars 34
MULTIPLICITY (200 nsec gate) I 0/4
MULTIPLICITY (200 nsec gate) for r/o window of 10 ns: 1 GHz x 10 ns x 1. 1 ~ 10 hits/ trig for σ = 100 ps we expect in a ± 2. 5σ: 0. 5 accident hits/trig September 27, 2004 A. Ceccucci/ CERN Villars 36
FTPC (KABES+FADC) • NA 48/2 – – KABES has achieved very good performance Position resolution ~ 70 micron Time resolution ~ 0. 6 ns Rate per micro-strip ~ 2 MHz • NA 48/3 – Intensity ~ 10 higher per unit area – 600 ns drift – The long drift (600 ns) makes a standalone pattern recognition very difficult or just impossible ( That’s why we plan to have SPIBES in front) – To reduce double pulse resolution and improve the time resolution one has to reduce the pulse duration and possibly read-out every micro-strip with 1 GHz FADC September 27, 2004 A. Ceccucci/ CERN Villars 37
KABES principle: TPC + micromegas Tdrift 2 Micromegas Gap 50 μm Operated @ Edrift=0. 83 k. V/cm Tdrift 1 + Tdrift 2 = 750 ns 48 strips with 0. 8 mm pitch September 27, 2004 A. Ceccucci/ CERN Very low discharge probability 38 Villars
NA 48/2 KABES IN NA 48/2 Position and time resolutions Time resolution: 0. 6 ns Using TOT to correct time slewing Space resolution from drift time measurement: 70 μm Tagged K track Time resolution 0. 6 ns (T 0)KABES- (T 0)DCH Spectrometer (ns) Tagging with reconstructed K± ± + September 27, 2004 A. Ceccucci/ CERN XStation 1 or 2 - XStation 3 (cm) Villars 39
TEST OF KABES IN 2004 LOW INTENSITY September 27, 2004 A. Ceccucci/ CERN Villars 40
KABES 25 micron amplification gap Recent lab test with 25 mm gap Width ~30 ns Width ~18 ns 50 mm gap 25 mm gap improvement of occupancy observed with 25 mm amplification gap September 27, 2004 A. Ceccucci/ CERN Villars 41
BEAM DATA: 50/25μ mesh 2003/2004 Time over Threshold (ns) September 27, 2004 A. Ceccucci/ CERN Villars 42
2004 KABES TEST HIGH INTENSITY FLUX PER UNIT AREA CLOSE TO NA 48/3 ! September 27, 2004 A. Ceccucci/ CERN Villars 43
TEST OF KABES with 480 MHz FADC September 27, 2004 A. Ceccucci/ CERN Villars 44
FADC Readout (1) • 8 bit FADC 960 MHz (2 interleaved 480 MHz) from NA 48 proton tagger • 9 FADC boards (18 channels “ 480 MHz mode”) • 18 FADC channels connected to KABES strips in station upstream UP and downstream K 1 K 1 23 24 25 26 K 2 K 2 23 24 25 26 FADC 16 14 12 10 FADC K 1 6 4 2 18 September 27, 2004 K 5 K 5 K 5 K 2 K 5 K 3 K 6 K 4 A. Ceccucci/ CERN Villars 23 24 25 26 27 FADC FADC 1 8 3 5 7 K 6 K 6 K 6 23 24 25 26 27 FADC FADC 9 11 13 15 17 45
Multiple pulses without Zero-suppression September 27, 2004 A. Ceccucci/ CERN Villars 46
FADC data (1) • Distribution of time over threshold – Low intensity run – Threshold set to 37 ADC counts (~27 m. V) – ± 2 samples over threshold ~20 ns time over threshold (KABES 25 µm mesh) September 27, 2004 A. Ceccucci/ CERN Villars 47
FADC data • Number of pulses per channel as a function of beam intensity x 1 x 3 x 7 x 12 – noisy channels 10, 12, 14, 16 all connected to K 1 September 27, 2004 A. Ceccucci/ CERN Villars 48
Multiple pulses with Zero-supp • Some data from RUN 16964 (14 x. I 0) • horizontal scale: FADC time units (2× 1. 04 ns) September 27, 2004 A. Ceccucci/ CERN Villars 49
Attempt to fit single pulses • fit of 500 FADC pulses from the low intensity RUN (16916) • 4 parameters function used: • uncertainty of 1. 7 counts per FADC value September 27, 2004 A. Ceccucci/ CERN Villars 50
Attempt to fit multiple pulses • “hand fit” of a six pulses event from run 16964 (x 14 I 0) September 27, 2004 A. Ceccucci/ CERN Villars 51
KABES FADC Conclusion • We read-out the micro-strips with FADC • Double pulse resolution capability is very good • To do list: – develop reconstruction from list of times in strips and measure resolution (space, time, angular) as a function of beam intensity September 27, 2004 A. Ceccucci/ CERN Villars 52
ANTI • Set of ring-shaped photon vetoes surrounding the decay tank • Specification: inefficiency to detect photons above 100 Me. V < 10 -4 • The NA 48 ANTI’s (AKL) need to be replaced • Extensive R&D Performed by American and Japanese groups • Claims that inefficiency as low as 10 -5 can be achieved • Baseline solution: Lead/ Plastic scintillator sandwich (1 -2 mm lead / 5 mm plastic scintillator) • Cost driver of NA 48/3 September 27, 2004 A. Ceccucci/ CERN Villars 53
Current NA 48 ANTI September 27, 2004 A. Ceccucci/ CERN Villars 54
WIRE CHAMBERS September 27, 2004 A. Ceccucci/ CERN Villars 55
NA 48 WC September 27, 2004 A. Ceccucci/ CERN Villars 56
Beam Test 2004: WC • With the exception of one sector in the Y view of DCH 3 all other channels could be operated at nominal HV at the highest beam intensity • Quite encouraging: plane efficiency decreases only by 1 -2% • Ability to operate the WC at intensity close to NA 48/3 cannot be overemphasised !! September 27, 2004 A. Ceccucci/ CERN Villars 57
WC @ 30 x I 0 September 27, 2004 A. Ceccucci/ CERN Villars 58
WC: Kinematical rejection Events accepted Region I Measured quantities September 27, 2004 A. Ceccucci/ CERN Villars 59
Double spectrometer layout Kevlar Mag 1 DCH 2 DCH 3 DCH 1 Mag 2 DCH 4 DCH 5 DCH 6 x He Vacuum Kevlar z NA 48 Wire chambers 0. 004 X 0 x Vacuum Straw tube 0. 0025 X 0 September 27, 2004 He z NA 48 Wire chambers 0. 004 X 0 A. Ceccucci/ CERN Villars 60
Double spectrometer Two independent measurement of Pp s( ) Single configuration Double configuration DCH 1 in vacuum Simulation with gaussian MS September 27, 2004 A. Ceccucci/ CERN Villars 61
WC (baseline solution) • NA 48/2 – Four larger drift chambers and one large-gap dipole magnet (MNP 33/1) – The Read out of the chambers is modern (2002) – Chambers enclosed in He by thin (0. 3 %) Kevlar membrane • NA 48/3 – Add two chambers and one copy of MNP/33 – Consider first tracking station operated in vacuum to improve the angular resolution (and missing mass) – For this we are evaluating the ATLAS/COMPASS Straw technology – Straws have been employed in high rate rare kaon decays experiment (AGS-E 871, KL→me) with rates as high as 750 KHz/wire September 27, 2004 A. Ceccucci/ CERN Villars 62
Can we do w/o beampipe? September 27, 2004 A. Ceccucci/ CERN Villars 63
WC (More elegant solution) • Replace all WC with straws and extend the vacuum down to the CHOD (original CKM idea) • This would allow to remove the beam-pipe, thus simplify the scheme to render hermetic the photon vetoes in the intermediate region between the LKR and SAC 3 at the end of the hall • This solution will likely reduce the number of e. m. showers generated in the beam-pipe. • These showers may lead to a non-negligible load on the chambers and read-out September 27, 2004 A. Ceccucci/ CERN Villars 64
CHOD • It is used to provide a fast trigger and (together with the gigatracker) to provide the time coincidence to associate the “right” kaon track to the pion track • Investigating three solutions: – Scintillator tiles – Fused silica to use cherenkov light – Multi-gap glass RPC (ALICE). • Excellent time resolution (< 50 ps) for rates up to 1 KHz/cm^2 • But in the hottest regions near the beam pipe of NA 48/3, rates up to 5 KHz/cm^ have been measured further studies needed September 27, 2004 A. Ceccucci/ CERN Villars 65
LKR • The NA 48 Liquid Krypton Calorimeter • Must achieve inefficiency < 10 -5 to detect photons above 1 Ge. V • Advantages: – – – It exists Homogeneous (not sampling) ionization calorimeter Very good granularity (~2 2 cm 2) Fast read-out (Initial current, FWHM~70 ns) Very good energy (~1%, time ~ 300 ps and position (~1 mm) resolution • Disadvantages – 0. 5 % X 0 of passive material in front of active LKR – The cryogenic control system needs to be updated September 27, 2004 A. Ceccucci/ CERN Villars 66
NA 48 Liquid Krypton Calorimeter 9 m 3 of Lkr (13212 cells) 1. 25 m deph (27 X 0) s(E)/E = 3. 2%/ E 9 % /E 0. 42% s(mgg)~1 Me. V/c 2 ; s(t) ~ 300 ps FAST: 70 ns FWHM (it can be reduced) EXCELLENT GRABULARITY: 2 2 cm 2 September 27, 2004 A. Ceccucci/ CERN Villars 67
2004 TEST OF LKR AS PHOTON VETO September 27, 2004 A. Ceccucci/ CERN Villars 68
LKR Hermeticity • According to simulation we should be able to address p 0 rejection inefficiency to 10 -5 -10 -6 complementing the NA 48/2 setup with: – A SAC and the end of the hole – Read out the corner of LKR which are usually not read out • We have prepared the above for the beam test in 2004 • Since we currently we have no beam sweeper, intensity of beam was reduced to 1/6 of NA 48/2 nominal: test of hermeticity at low intensity. September 27, 2004 A. Ceccucci/ CERN Villars 69
Small angle photon veto Pb. WO 4 crystals (CMS) ü Dimension of crystals 2 x 2 x 23 cm 3 ü 7 x 7 cm matrix ü ~ 25 X 0 ü Readout with light guides and PMT üInstalled for last week of data taking September 27, 2004 A. Ceccucci/ CERN Villars 70
Full LKr read-out YLKr (cm) Installed for last week of data taking New instrumented LKr regions work. XLKr (cm) September 27, 2004 A. Ceccucci/ CERN Villars 71
General definitions Region I p 0 Peak Region II September 27, 2004 A. Ceccucci/ CERN Villars 72
How to measure p+p 0 Rejection p 0 peak BEFORE APPLICATION OF EXTRA CLUSTERS IN LKR DATA MC AFTER APPLICATION OF EXTRA CLUSTERS IN LKR DATA MC 1486 MC 678 DATA 745862 MC 212569 DATA M 2(miss) (Ge. V/c 2)2 September 27, 2004 A. Ceccucci/ CERN M 2(miss) (Ge. V/c 2)2 Villars 73
LKR: Work in progress. . . • Detectors: – Read-out for LKr corners work – Must understand effect of e. m. showers initiated in ANTI and Beam pipe spilling into WC on tight one track trigger (underestimate the photon rejection) – SAC works; must improve reconstruction – Collected several millions of unbiased K+ →p+ p 0 • These data are invaluable to compare with our assumptions and to design the new experiment September 27, 2004 A. Ceccucci/ CERN Villars 74
MAMUD • To provide pion/muon separation and beam sweeping. – Iron is subdivided in 150 2 cm thick plates (260 cm 2 ) • Four coils magnetise the iron plates to provide a 1. 3 T dipole field in the beam region • Active detector: – Strips of extruded polystyrene scintillator (1 x 4 x 130 cm 3) – Light is collected by WLS fibres 1. 2 mm diameter • As shown by our Fermilab CKM friends, if backgrounds from K+→ m+ n remains a concern, a tracking RICH in place of the second magnetic spectrometer would completely remove it September 27, 2004 A. Ceccucci/ CERN Villars 75
MAMUD rough parameters (MARCELLO LOSASSO) Total weight ≈ 150 ton Overall Dimension 2. 6 m x 2. 8 m x 5. 25 m (Wx. Hx. L) Number of iron plates (2 x) 150 Current ≈ 4 KA Power ≈ 0. 8 MW Field integral on axis (from -5 m to +5 m) 6. 85 T m Magnetic field into a “good field region” (by 10 cm x 10 cm) ≈ 1. 3 T September 27, 2004 A. Ceccucci/ CERN Villars 76
Proposed Dipole configuration Pole gap is 11 cm V x 30 cm H Coils cross section 15 cm x 25 cm September 27, 2004 A. Ceccucci/ CERN Villars 77
Main results of simulation Field integral on axe Magnetic field at z=0 m September 27, 2004 A. Ceccucci/ CERN Villars 78
NA 48/3 Organisational Matters September 27, 2004 A. Ceccucci/ CERN Villars 79
Time Schedule • 2004 – – – Launch GIGATRACKER R&D Vacuum tests Evaluate straw tracker Start realistic cost estimation Complete analysis of beam-test data • 2005 – Complete of the above – Complete Specifications – Submit proposal to SPSC • 2006 -2008 – Costruction, Installation and beam-tests • 2009 -2010 – Data Taking September 27, 2004 A. Ceccucci/ CERN Villars 80
Tentative Indication of cost Element Cost (MCHF) Comments BEAM LINE 0. 5 Modified K 12 line CEDAR 0. 2 Photon Detectors GIGATRACKER 1. 4 Assuming 0. 13 mm VACUUM 0. 7 Upgrade of vacuum system ANTI 4. 2 Based on CKM estimate + elec. WC 3. 0 Two more chambers + R/O MNP 33/2 2. 5 Including prolongation of He tank CHOD 1. 0 2000 PMs, 500 CHF/channel LKR 2. 0 New supervion system and R/O A state-of the art calorimeter is 15 MCHF MAMUD 2. 0 SAC & CHV Cost of iron: ~500 KCHF 0. 5 TRIGGER & DAQ TOTAL September 27, 2004 19. 0 A. Ceccucci/ CERN Villars 81
Compatibility/Competition • NA 48/3 is fully compatible with COMPASS running – We need only 3 1012 per cycle (already available now on T 4!) • There are two approved competitors for beam time – LHC filling and CNGS • The extent of the project implies that there should not be in addition (at the stage of data taking) – A fixed target heavy ion programme – Any other experiment in ECN 3 competing for proton beam time • We understand that the SPS can deliver protons for fixed target physics even when LHC is being operated with ions September 27, 2004 A. Ceccucci/ CERN Villars 82
Comparing Apples with Apples NA 48/3 P 940 Accelerator CERN-SPS FNAL-MI Energy (Ge. V) 400 120 Duty cycle (s /s ) 4. 8 / 16. 8 = 0. 29 1. 0 / 3. 0 = 0. 33 1 HEP year (s) 107 Eff. Spill. Length (s / s) 3. 0 / 4. 8 = 0. 63 1. 0 / 1. 0 (? ) Total Rate (GHz) 0. 8 0. 23 Fraction of Kaons (%) 6 4 Flux of Kaons (MHz) 50 10 Decay fraction (%) 9 (50 m / 100 m) 17 (60 m / 60 m) Acceptance (%) 10 5 Events/y (BR~10 -10) 82 28 September 27, 2004 A. Ceccucci/ CERN Villars 83
Conclusions • We have found a fortunate combination where a compelling physics case can be addressed with an existing accelerator, employing the infrastructure (i. e. civil engineering, hardware, some sub-systems) of an existing experiment • We stress that this initiative in not a mere continuation of NA 48 • The working group has now becoming a proto-collaboration seeking the qualified participation of new collaborators September 27, 2004 A. Ceccucci/ CERN Villars 84
Kaons: Longer term (i. e. More Protons Needed!) • K 0 L→ p 0 e+e- and K 0 L→ p 0 m+m- (NA 48/4) • K 0 L → p 0 n n (NA 48/5) September 27, 2004 A. Ceccucci/ CERN Villars 85
K 0 L→p 0 ee(mm) : Perspectives • Detector s(gg) × 2 – Very ambitious, KTe. V/NA 48 already state of the art • KS-KL time dependent interference × 2 – Position experiment between 9 and 16 KS lifetimes (hep-ph/0107046) • KS-KL time independent interference × 3 – Assume constructive interference (theoretically preferred) • • Data Taking × 5 – Run in “factory mode”. After all E 799 -II run only for a few months to collect ~7 × 1011 KL decays Beam intensity × 4 – – • Need ~1012 protons/sec, slowly extracted, high energy (≤ 1 Te. V), DC explore the window of opportunity between current upper limit and SM Tot ~ × 240 → sens on BR ~ × 15 (on Im lt ~× 4 -15) Ideal Kaon Application for High Intensity/High Energy Machine September 27, 2004 A. Ceccucci/ CERN Villars 86
KL→p 0 nn @CERN? From KAMI proposal E 391 A J-PARC NA 48/5? SPS CERN may become competitive if the E 391 A technique works September 27, 2004 A. Ceccucci/ CERN Villars 87
Conclusions – A competitive programme can start pnow for charged kaons at the current SPS – For a very competitive neutral kaon decay experiment, ~ 1013 slowly extracted, high energy protons per second would be needed September 27, 2004 A. Ceccucci/ CERN Villars 88
SPARES September 27, 2004 A. Ceccucci/ CERN Villars 89
Optics drawing : September 27, 2004 A. Ceccucci/ CERN Villars 90
Detector Layout September 27, 2004 A. Ceccucci/ CERN Villars 91
Ratio of K+ production at 400 and 120 Ge. V/c p 0 vetoing is easier for high energy kaons, it pays off to go to highest proton energy!! The CERN-SPS is the best machine By LAU GATIGNON CERN/AB September 27, 2004 A. Ceccucci/ CERN Villars 92
ALICE PIXEL IN NA 48 BEAM ALICE pixel team members helping with the preparations and the test: Fadmar Osmic DOCT Petra Riedler September 27, 2004 Giorgio Stefanini Mike Burns A. Ceccucci/ CERN Peter Chochula Villars Alex Kluge Michel Morel 93
KABES IN NA 48/2 Reconstruction of K+ and K- beam spots Beam spot size from K ∼ 7 cm 2 Total Rate from π K ∼ 20 MHz September 27, 2004 A. Ceccucci/ CERN Villars 94
Photon Vetoes (KAMI) September 27, 2004 A. Ceccucci/ CERN Villars 95
Trigger • • • INPUT: 10 MHz DC; The single rate particle is dominated by muons from Kaon decays!! 2/3 of decays have 1 moun in filan state Important to reduce rate using simple cuts to avoid correlations MAMUD and ANTI used as a veto should reduce the rate to 2. 5 MHz LKR bidimensional pipelined trigger to count photons can be envisaged now (follow LHC experience) LKR will reduce rate to 1 MHz Further suppression based on multiplicity in DCH, SAC, CHV At this stage the rate will be sufficiently reduced to allow a processorbased system to analyse the WC tracking information applying mild cuts on the missing mass ( 100 KZH? ) High Level Trigger handled in PCFARM (cf. LHCB 1 MHz level 1…) ½ band-width reserved for down-scaled triggers Others triggers to collect less rare but still very interesting kaon decays September 27, 2004 A. Ceccucci/ CERN Villars 96
SPIBES • Si Sensor – Wafers of thickness < 150 micron can be obtained but bumpbonding yield needs to be investigated – Experience gained by LHC experiments very valuable – To achieve 100 ps time resolution 90% of the charge from the sensor has to be collected in ~1 ns • Pixel chip (Mixed-signal ASIC) – Very fast preamp and shaper – Very low time walk discriminator – High Resolution TDC • Front-end hybrid – Additional functionality is needed for clock, trigger distribution, data multiplexing and transmission, and controls – This requires one or more ASICs in the immediate vicinity of the pixel chip – For this design, to minimise thickness it is desirable to have power, data buses and the front-end hybrid functionality at the periphery of the detector plane September 27, 2004 A. Ceccucci/ CERN Villars 97
SPIBES • Silicon micro-pixel good candidate to achieve required time resolution – Capacitance of single pixel ~ 100 f. F – Rise time requirement < 2 ns – Must be demonstrated in R&D phase • The complexity of the analog/digital design with onchip TDC capability will most likely require the use of 0. 13 micron CMOS technology • Hybrid Pixel Technology – Pixel chip bump-bonded on Si sensor – State-of-the-art: ALICE SPD 150 micron chip on 200 micron sensor – Further reduction of material needs investigation September 27, 2004 A. Ceccucci/ CERN Villars 98
K 0 L→p 0 e+e- and K 0 L→p 0 m+m. Study Direct CP-Violation • Indirect CP-Violating Contribution has been measured (NA 48/1, see next slide) Direct CPV • Constructive Interference (theory) • CP-Conserving Contributions are negligible Indirect CPV CPC September 27, 2004 A. Ceccucci/ CERN Villars 0++, 2++ 99
0 K S →p 0 + ee and 0 K S →p KS →p 0 ee 0 + mm KS →p 0 mm NA 48/1 7 events, expected back. 0. 15 NA 48/1 6 events, expected back. 0. 22 BR(KS→p 0 ee) 10 -9 = 5. 8 +2. 8 -2. 3(stat) ± 0. 8(syst) BR(KS→p 0 mm) 10 -9 = 2. 9 +1. 4 -1. 2(stat) ± 0. 2(syst) |as|=1. 06+0. 26 -0. 21 (stat) ± 0. 07 (syst) PLB 576 (2003) |as|=1. 55+0. 38 -0. 32 (stat) ± 0. 05 (syst) September 27, 2004 Villars A. Ceccucci/ CERN La Thuile, Moriond 2004 100
K 0 L→p 0 ee (mm): Sensitivity to New Physics Isidori, Unterdorfer, Smith: Fleisher et al: Ratios of B → K modes could be explained by enhanced electroweak penguins and enhance the BR’s: September 27, 2004 A. Ceccucci/ CERN Villars * A. J. Buras, R. Fleischer, S. Recksiegel, F. Schwab, hep-ph/0402112 101
0 K L →p nn 0 • Purely theoretical error ~2%: SM 3 10 -11 • Purely CP-Violating (Littenberg, 1989) • Totally dominated from t-quark • Computed to NLO in QCD ( Buchalla, Buras, 1999) • No long distance contribution SM~3 × 10 -11 • Experimentally: 2/3 invisible final state !! • Best limit from KTe. V using p 0→eeg decay BR(K 0 → p 0 nn) < 5. 9 × 10 -7 90% CL Still far from the model independent limit: BR(K 0 → p 0 nn) < 4. 4 × BR(K+ → p+nn) ~ 1. 4 × 10 -9 Grossman & Nir, PL B 407 (1997) September 27, 2004 A. Ceccucci/ CERN Villars 102
Estimation of the Radiation Effects for Future NA 48 VERY PRELIMINARY! Fluence: Assuming 5000 spills/day, 50 MHz particles/cm 2/spill and a spill length of 4. 8 s: 1. 2 E 12 particles/cm 2 per day Assuming pion beam: feq= kf k…hardness factor (9 Ge. V pions: 3. 6, flat dist. ) feq=4. 32 E 12 (1 Me. V neutrons)/cm 2 per day Assuming 100 days of running: feq= 4. 32 E 14 (1 Me. V neutrons)/cm 2 per run >> Leakage current, depletion voltage, signal, operating conditions, …. Dose: 4. 32 E 14/(6. 24 E 9/(1. 66 Me. V g-1 cm 2))=114 k. Gy= 12 Mrad Need an accurate investigation ! September 27, 2004 A. Ceccucci/ CERN Villars. Mara Martini 103
Residual events after LKr rejection Events with no g on LKr and at least one g at high angle Events with at least one g on LKr Events with no g on LKr and at least one g at small angle September 27, 2004 A. Ceccucci/ CERN Villars 104
g Rejection with AKL Pocket 1 -7 |Thit-Ttrack+offset| < 10 ns September 27, 2004 A. Ceccucci/ CERN Villars 105
g rejection with SAC Events with a shower reconstructed in SAC rejected Fraction of surviving events: (2. 8 ± 0. 2) × 10 -3 September 27, 2004 A. Ceccucci/ CERN Villars 106
Residual events Look at NUT signals (SPY) September 27, 2004 A. Ceccucci/ CERN Villars 107
Comparison with TDC • Leading time distribution FADC TDC September 27, 2004 A. Ceccucci/ CERN Villars 108
FADC data (3) • Pulse height distribution – Threshold ~37 ADC counts September 27, 2004 A. Ceccucci/ CERN Villars 109
Data without Zero-supp (baseline) • Baseline: September 27, 2004 A. Ceccucci/ CERN – Mean: 20. 2 Villars – R. m. s. : 2. 4 110
Unexpected very big pulses (1) • ADC count 255 = 400 m. V • width of this pulse: ~450 ns September 27, 2004 A. Ceccucci/ CERN Villars 111
Unexpected very big pulses (2) occupies the whole FADC R/O time window! September 27, 2004 A. Ceccucci/ CERN Villars 112
Unexpected very big pulses (3) • Fraction of “big” pulses (>200 ADC counts) over “normal” pulses (<200 ADC counts) as a funcion of beam intensity September 27, 2004 A. Ceccucci/ CERN Villars 113