1 / 75

Exploring Rare Kaon Decays: Challenges and Prospects

Dive into the world of rare kaon decays at the Fermi School in Varenna, discussing unitarity, lepton flavor violation, and experimental challenges in detecting decays such as KL0. Learn about advanced techniques like the KOPIO method and recent experiment results.

yetta
Télécharger la présentation

Exploring Rare Kaon Decays: Challenges and Prospects

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Rare Kaon Decays - 3 Laurence Littenberg BNL E. Fermi School, Varenna - 26 July 2005 L. Littenberg – Varenna

  2. Organization • Introduction & general motivation • Lepton Flavor Violation, etc. • Brief review of Unitarity • K+ • KL0 • K • KLl+l- • KL0l+l- L. Littenberg – Varenna

  3. _ A(K++) U.T. & K • ReC/ImC ~ 0.0006 • Range of mC(mC) 1.25-1.35GeV gives 8.5% in charm amplitude • 2.5% in total amplitude, 5% in BR AT AC L. Littenberg – Varenna

  4. _ _ The Challenge of KL0 • B(KL0) ~ 310-11, need intense flux of K’s • rates inevitably rather high • hard to minimize both random vetoing & veto-blindness • Kinematic signature weak (2 particles undetectable) • Backgrounds with 0 up to 1010 times larger than signal • Veto inefficiency on extra particles, both charged particles and photons, must be 10-4 • Self-vetoing is a problem • shower spreading makes it hard to maximize both signal efficiency and veto power • Huge flux of neutrons in beam • can make 0 off residual gas – requires high vacuum • halo must be tiny • hermeticity requires photon veto in this beam • Need convincing measurement of background L. Littenberg – Varenna

  5. 1st dedicated KL0 experiment - E391a • KEK 12 GeV PS • 4° “pencil” beam • <pK> ~ 2 GeV/c • CsI calorimeter Pencil beam L. Littenberg – Varenna

  6. Pencil Beam 5 stages of collimators made of heavy metal (tungsten) 2 stages of sweeping magnets Thermal neutron absorber Pb/Be plug for control of /neutron flux Fine alignment using telescope GEANT M.C. agrees well with the measurements L. Littenberg – Varenna

  7. Covered by plastic scintillators (Charged Veto (CV)) Recycled 576 Un-doped CsI (70X70X300 mm3 (from E162) 50X50X500 mm3 (from KTeV)) CC03 (Tungsten + Scin.) E391a detector setup KL beam L. Littenberg – Varenna

  8. po produced at CC02 po produced at CV (?) KL popopo KL p+p-po KL  popo KL gg 2g analysis Data without tight veto M.C. for KL decays ( Without Normalization) PT(GeV/c) Reconstructed vertex (cm) L. Littenberg – Varenna

  9. Veto Optimization ~Main-barrel timing (low E sample)~ upstream KLgg pure sample gg B.G.sample ② ① early late downstream Backsplash should NOT veto! • Real photon hit  should veto. • Backsplash  should NOT veto. L. Littenberg – Varenna

  10. E391a Result from 10% of Run I • No events observed/expected background of 0.030.01 events (mainly K2) • 1.14109 KL decay, 0.0073 acceptance  s.e.s of 1.1710-7 • B(KL0)<2.86 10-7 @ 90% CL (c.f. 5.9 10-7 from KTeV) Z(cm) L. Littenberg – Varenna

  11. Better quality of data(online plots) Run-I Run-II PT(GeV/c) Reconstructed vertex (cm) Run-II analysis Run-III this fall ( Conditionally approved ) L. Littenberg – Varenna

  12. E391a status & prospects • First physics run Feb-June 2004 • 2.21012 12 GeV POT, 50% duty factor • 5 105 KL/pulse • Detector worked well • Nominal s.e.s. 410-10 • But acceptance ~ 15 lower than in proposal (0.0073) • first sight of the enemy • Halo neutrons, self-vetoing, etc. • Analysis of 10% of data  B(KL0)<2.86 10-7 • Run II, Feb-March 2005 • Many problems fixed, 60% of Run 1 • Run III, this fall, conditionally approved L. Littenberg – Varenna

  13. JPARC Phase I Beamlines L. Littenberg – Varenna

  14. 100 more KL Thicker photon vetoes Deeper, more granular crystals Faster electronics KEK-PS to J-PARC • 1014 interacting 30 GeV protons/cycle, 5srbeamline @ 16° • 22MHz KL @ 20m, <pK> = 2.1 GeV/c, 9%/5m decay • 4% acceptance • 23 events in 3 Snowmass years (competition from ) • S:B~1:1 L. Littenberg – Varenna

  15. Signal KL00 bckgnd 16° case 0 pT Z-decay Step by step at JPARC • “Step by step” approach, learning as they go • Different beam angles, lengths • Larger detector • Eventual goal – few 100 evts L. Littenberg – Varenna

  16. KL0 Experiment veto beam veto calor. prod. tgt L. Littenberg – Varenna

  17. KL0 Experiment veto prerad beam veto calor. prod. tgt L. Littenberg – Varenna

  18. In the KLCoM • Bckgnd mainly in discrete areas • Obvious for KL00 “even” • But even “odd” case not ubiquitous • K3 infests slightly different area • Even after all bckgrnds accounted for, still some clear space for signal • Can get factor 50-100 L. Littenberg – Varenna

  19. KOPIO Technique • High intensity micro-bunched beam from the AGS • Measure everything! (energy, position, angle, time) • Eliminate extra charged particles or photons • KOPIO: p0 inefficiency < 10-8 • Suppress backgrounds • Predict backgrounds from data:dual cuts • Use “blind” analysis techniques • Test predictions “outside the box” • Weight candidate events with S/N likelihood function L. Littenberg – Varenna

  20. 40 ns between microbunches AGS Provides • Proton Beam • 100TP/spill (upgraded from present 70TP) • ~5s spill, 2.3s interspill • Microbunching • Extract debunched beam resonantly between empty buckets • 25MHz frequency • 200ps bunch width • 10-3 interbunch extinction • Kaon Beam • 42.5o take-off angle • Soft momentum spectrum • 0.5-1.5 GeV/c • 3108 KL/spill • 8% decay • 10 GHz neutrons =200 ps L. Littenberg – Varenna

  21. Tests of Microbunching Tests of: Microbunch width Interbunch extinction Studied the RF extraction mechanism proposed for KOPIO & measured a microbunch rms width of 244 ps -- KOPIO spec is 200 ps rms Measured the inter-bunch extinction ratio (flux between bunches/within bunch). KOPIO requires ~ 10-3. 4.5 MHz 93 MHz L. Littenberg – Varenna

  22. KOPIO Concepts L. Littenberg – Varenna

  23. KOPIO Detector L. Littenberg – Varenna

  24. Preradiator – convert & measure  properties . . . . . .  e+ . . . 4m . e- . . . . . . Cathode strip drift chambers Extruded Scintillator & WLS fibers 64 Layers (4% X0/layer, 2.7 X0) 256 Chambers 288 Scintillator Plates (1200 m2) 150,000 Channels Readout L. Littenberg – Varenna

  25. KOPIO Prototype Measurements–BNL LEGS Tagged Photon Beams Preradiator Angular resolution: 25 mr at 250 MeV/c Simulations agree with measurements. L. Littenberg – Varenna

  26. Shashlyk Photon Calorimeter Shashlyk modules prototyped and tested in beams. Required specs have all been met APD L. Littenberg – Varenna

  27. Beam test of Calorimeter modules Simulation: Combined PR +CAL Energy Resolution L. Littenberg – Varenna

  28. Charged Particle Veto in vacuum L. Littenberg – Varenna

  29. Charged Particle Veto Performance Plastic Scintillator – backed up by  vetoes! MC 10-3 10-4 Data 10-5 10-6 • 290 L. Littenberg – Varenna

  30. Barrel Veto/Calorimeter • Cylindrical array of 840 modules with 2.5m ID • Both signal detection and vetoing functions • 1g in prerad + 1g in BV/C • Modified version of calorimeter shashlyk technology, pmt readout • Energy resolution calculated to be almost as good as calorimeter • Time resolution should be comparable • B V/C lined with thin, high-efficiency, charged particle veto scintillators US end of barrel sealed by wall of plate shower-counter vetoes L. Littenberg – Varenna

  31. D4 & downstream vetoes • Charged & g vetoes in D4 sweeping magnet • Field sweeps vertically • DS vetoes detect g’s emerging from the beam • Lead/scintillator plate sandwich counters • Hermeticity completed by catcher veto at the back 48D48 L. Littenberg – Varenna

  32. Photon Veto Efficiency KOPIO PV Estimates and Simulations based on improved E949 Measurements supplemented by FLUKA calculations 1 MeV Visible Energy Threshold L. Littenberg – Varenna

  33. E949 SPI Measurement K2 Decay + 2 1 L. Littenberg – Varenna

  34. Catcher: Hadron Blind Beam  Veto beam Aerogel Counter 420 modules ofPb-Aerogel counter L. Littenberg – Varenna

  35. Catcher R&D results Modules prototyped and tested in beams. L. Littenberg – Varenna

  36. KOPIO Detection Modes Primary detection mode: Secondary mode: 2 photons covert in preradiator 1 photon in preradiator, 1 in BV Reconstruct 1ste+e- in “Preradiator”, Point to K decay vertex in vacuum L. Littenberg – Varenna

  37. KL modes simulated for bkgnd studies Largest back-grounds L. Littenberg – Varenna

  38. Other Backgrounds • K+ contamination of beam: <0.001 of signal rate • KLK+e-, K-e+: ~ 0.001 of signal rate • nN p0N: negligible production from residual gas in decay volume if pressure<10-6 Torr. Requirements on reconstructed ZV(KL) suppress rate from US wall to <0.01 of signal rate • n: far smaller than neutron background • Hyperons: <10-5 of signal rate • Fake photons < 0.05 of signal rate assuming ~10-3 10-3 suppression from (vetoing)  (g/n discrimination) • Two KL giving single candidate: negligible due to vetoes • (KL pX)  (p p0e ): ~0.01 of signal rate • KS p0p0: ~4  10-4 of KL p0p0 background rate _ L. Littenberg – Varenna

  39. Kinematic Separation of Signal & Background Pion kinetic energy squared (T*2) vs Ln(Missing Energy) Signal Backgrounds L. Littenberg – Varenna

  40. Radiative Ke3 • Background from radiative Ke3 when • e+ annihilates before being vetoed & resulting  combines with radiated to make “0” • 2nd  from annihilation very soft, so missed • Then it’s critical to veto on - • But what if - is very slow? Either - • have long gate with big random loss • or let background creep in • Can fight this with kinematics - • If - really slow, p m • Then m2=(pK - p0 - m)2 should be  0: L. Littenberg – Varenna

  41. _ Optimized S/B vs. Signal for KOPIO(assuming SM B(KL0 )) L. Littenberg – Varenna

  42. 1st yr 2nd yr 3rd yr KOPIO: SM Precision L. Littenberg – Varenna

  43. Discovering New Physics - Early L. Littenberg – Varenna

  44. Constraining New Physics - Early L. Littenberg – Varenna

  45. 3rd yr 1st yr 2nd yr Discovering/Constraining New Physics L. Littenberg – Varenna

  46. KOPIO Summary & Outlook • Excellent discovery potential for non-SM physics • Unique connection with underlying parameters • Extremely rapid progress in first part of run • Experiment background rejection scoped for 10-11 • 5 discovery if BR 0.6< BRSM or >1.7 BRSM • If find Br ~ BrSM • Likelihood analysis using ~150 evts • Precision on BR: 14%; Im t: 7% • Rule out non-SM effects outside (1  0.24)  BRSM • Unique constraint on some BSM operators L. Littenberg – Varenna

  47. _ A New Challenge: K _ • Another clean short-distance dominated decay • Related to Ke4 the way K is related to Ke3 • Calc. by Geng et al., also by Valencia & LL. Latter obtain: • B(KL+-)  [1.8(1.4-)2 + 0.32]  10-13 • B (KL00)  (1.4-)2  10-13 • B (K++0)  [0.7(1.4-)2 + 0.72]  10-14 • Very interesting angular distributions in +- case •  and  terms correspond to different l states, therefore in principle can extract  and  separately • In practice -  term dominates almost completely • KL+- major experimental challenge: • Must reject KL+- 0 by >1012, K3 by similar factor. • Must reject KL+- by >108 • Best to get into KL c.m.(only at low momenta) •  veto & particle ID usually best at high momenta • KL00 seems even harder, but has certain advantages • Detector can focus on  detection & vetoing, KL00 suppressed • Experiments seeking KL0 can add this to their menu. • K++0 was the first to have been probed L. Littenberg – Varenna

  48. E391a limit on KL00 Data Signal MC signal box PT PT normalization box m (GeV/c2) m (GeV/c2) No events seen, normalized via KL00, BR<3.210-5 @ 90% CL (preliminary) One day run, ~100 times more data will be available, half with better quality L. Littenberg – Varenna

  49. E787 search for K++0 Target of opportunity in monitoring data for K++ Data Signal MC No events observed, 90% CL limit established: B (K++0)< 4.3 10-5 Far short of SM expectation, but comparable to K++-e+ Can also get limits on non-SM process K++0X0 L. Littenberg – Varenna

  50. Short distance part of B(KL+-) given by: KL+- where  is a kinematic factor  1, Y(xt) 1.02(mt/170)1.56, & YNL  310-4 = 1.7510-9A4Y2(xt)(0-)2 = (0.960.10) 10-9 (where 01.2) So could potentially measure  or be sensitive to BSM physics KL+- K3 Moreover there’s a very good measurement by AGS-871 (6000 events!) But a number of roadblocks need to be overcome: 1. B(KL+-) dominated by absorptive contribution from KL 2. Much larger than the dispersive part that contains BSD! 3. Long-distance dispersive part interferes with short-distance contribution L. Littenberg – Varenna

More Related