Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 1 Charged kaons at KLOE Roberto Versaci on behalf of the KLOE collaboration

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 2 I get it! СПАСИБО!!!

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 3 Outline ● Vus with charged kaons TAG mechanism K +    (  ) Semileptonic decays Charged kaon lifetime ● Conclusions

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 4 V us form charged kaons From semileptonic decays From leptonic decay (Marciano, Phys.Rev.Lett.93:231803,2004) Form factorRadiative corrections Phase space ( +, 0 ) Lattice QCD

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 5 Kaon pair production The  decays at rest producing a kaon pair: K L K S or K + K - The detection of a K guarantees the presence of the K with known momentum  Tag mechanism Normalization to the number of tags allows a precise measurement of absolute BRs

Roberto Versaci e + e - collisions from  to  – Novosibirsk / Ev/0.5MeV P* [MeV]  data Tag mechanism K ± events tagged using two body decays (about 85%): K ±   ±  ±  0  1.5 x 10 6 K + K - ev/pb -1 Two-body decays identified as peaks in the momentum spectrum of secondary tracks in the kaon rest frame P*(m  ) To minimize the impact of the trigger efficiency tags must provide themselves the Emc trigger of the event: self-triggering tags N self-trg Tag  2 x 10 5 pb -1

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 7 Measurement of the absolute branching ratio Published on Phys.Lett.B 632:76-80,2006

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 8 Overview Normalization N TAG given by 175 pb -1 from self-triggering Counting events in the distribution of secondary track momentum in the kaon rest frame p* Background subtraction Efficiency related to DC reconstruction only (tracking plus vertexing), evaluated on data

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 9  region Entries MC includes radiative process P * [MeV] ● Signal given by K + decay in the DC FV (40 cm <  < 150 cm) Using ~60 pb -1 ● Background given by events with  0 in the final state: Signal

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 10  DC evaluation ● Efficiency has been evaluated on a second uncorrelated sample of ~115 pb -1 using only calorimeter information ● Double K  events have a typical signature in the EMC i.e. 2 isolated clusters with energy in the range 80 < E CLU < 320 MeV ● A correction O(10 -4 ) to the efficiency has been evaluate from MC:

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 11 Result Total number of events: Total accuracy: 0.27%

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 12 Using the updated result from MILC: we obtain: Result

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 13 V us - V ud plane unitarity Inputs: V us = ± (K l3 KLOE) V ud = ± (Marciano) V us /V ud = ± (K  2 KLOE) Fit results: V us = ± V us = ± V ud = ± V ud = ± Fit results assuming unitarity: V us = ± V us = ± P(  2 ) = 0.43

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 14 Measurement of the K ± semileptonic decays absolute branching ratios absolute branching ratios&

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 15 Semileptonic overview 4 independent normalization samples (2 tag x 2 charges) 410 pb -1 self-triggering tags from 2001 and 2002 data Fit of the charged secondary square mass spectrum m 2 lept K    and K      rejected cutting on p*(m  ) Efficiency evaluated from MC and corrected for Data/MC ratio

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 16    e+e+e+e+   Two tracks vertex in the FV: 40 cm <  < 150 cm Track of charged secondary extrapolated to EMC Two body decays cut: p*(m  ) < 195 MeV/c  0 reconstruction: 2 neutral clusters in EMC with TOF matching the kaon decay vertex Mass of charged secondary from TOF measurement Signal selection

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 17 K  0  0 K  0 K nucl.int. K3K3 Ke 3         with a   undergoing a Dalitz decay, or with a wrong cluster associated to    give a m l 2 under the Ke3 peak  cut requiring (E miss – P miss ) < 90 MeV       with early    , give m l 2 under the K  3 peak  rejected using the missing momentum of the secondary track in the pion rest frame (P* sec < 90 MeV) Background (I)

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 18 Background (II)       K nucl.int. The cuts reject ≈ 96% of the background events The efficiency on the signal is ≈ 50% for both K e3 and K  3 The residual background is ≈ 1.5% of the selected K ± l3 sample. It has m lept 2 ≈ m  2

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 19 Event counting  Fractional accuracy: 0.9% for K e3 and 1.2% for K  3  Systematic error studies to be completed  Dominated by the knowledge of selection efficiency Fit m 2 lept spectrum with linear combination of Ke3, K  3 shapes, and bck contribution. Average of the four data samples. Preliminary

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 20 V us from semileptonic decays ns

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 21 Measurement of the charged kaon lifetime

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 22 Two different methods to measure  : #1: using K decay length #2: using K decay time Allow cross check of systematics. Method #1: using K decay length K    self-triggering tag Signal K decay vertex (using DC only) Signal K track extrapolated backwards to the IP de/dx taken into account  2mm step Efficiency evaluated directly on data

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 23 00 E ,t ,x  ±± E ,x , t  K  tag tt pKpK pKpK t0t0 lKlK xKxK E,t,xE,t,x  trk+vtx evaluation Efficiency has been evaluated directly on data Look for a charged vertex on a sample selected requiring a neutral vertex Neutral vertex from timing of the neutral clusters fired by the  s from the   0 decay

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 24    ns Proper time fit Preliminary The proper time is fitted together with the efficiency and taking into account resolution effects too Fit between 16 and 30 ns

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 25 Two different methods to measure  : #1: using K decay length #2: using K decay time Allow cross check of systematics. Method #2: using K decay time K    self-triggering tag Tag K track extrapolated backwards to the IP Second kaon helix extrapolated forwards Step along the helix looking for a  0 decay vertex For each photon:

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 26 K ± at KLOE - summary Absolute BR(      ) with 0.27% accuracy Phys.Lett.B 632:76-80,2006 K    l  l absolute branching ratios and lifetime: preliminary results BR(K     0 ) finalizing Using 2 fb -1 collected KLOE will be able to measure: K    l  l form factors, BR(K      l  l ) and the ratio BR(K  e )/BR( K  ) for e-  universality About 5 x 10 4 Ke2 events produced with 2fb -1

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 27 Spare slides

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 28 Beam pipe (spherical, 10 cm , 0.5 mm thick) + instrumented permanent magnet quadrupoles (32 PMT’s) Drift chamber (4 m   3.75 m, CF frame)  Gas mixture: 90% He + 10% C 4 H 10  stereo–stereo sense wires  almost squared cells Electromagnetic calorimeter  lead/scintillating fibers (1 mm  ), 15 X 0  4880 PMT’s  98% solid angle coverage Superconducting coil (B = 0.52 T) The KLOE design was driven by the measurement of direct CP parameter  '  KLOE detector

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 29  E /E  5.7% /  E(GeV)  t  54 ps /  E(GeV)  50 ps  vtx (  ) ~ 1.5 cm (   from K L        )  p /p  0.4 % (tracks with  > 45 ° )  x hit  150  m (xy), 2 mm (z)  x vertex ~1 mm KLOE detector

Roberto Versaci e + e - collisions from  to  – Novosibirsk / Ev/0.5MeV P* (MeV)  data Tag mechanism (I) K ± events tagged using two body decays (about 85%): K ±   ±  ±  0  1.5 x 10 6 K + K - ev/pb -1 Two-body decays identified as peaks in the momentum spectrum of secondary tracks in the kaon rest frame P*(m  )

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 31 Tag mechanism (II) To minimize the impact of the trigger efficiency on the signal side we restrict our normalization sample N TAG to 2-body decays which provide themselves the Emc trigger of the event: self-triggering tags Emc trigger: 2 trigger sectors over threshold ~50 MeV The  fires two sectors:  Trigger ~ 35% The photons from the   fire two sectors  Trigger ~ 75%

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 32 Tag bias C TB = BR MC (with tag) / BR MC (without tag) Measuring the BRs we must take into account a correction due to the bias on the signal sample induced by the tag selection Tag bias The correction C TB is evaluated from MC and is given by: Tag     

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 33  region Tag bias estimated from MC: C TB = / Entries MC includes radiative process P * [MeV] ● Signal given by K + decay in the DC FV (40 cm <  < 150 cm) Using ~60 pb -1 ● Background given by events with  0 in the final state:

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 34  DC evaluation Efficiency evaluated on ~115 pb -1 sample Double K  events selected using Emc 1) Self-triggering K -  - tag 2) Ask for: 1 cluster with 80 < E CLU < 320 MeV no cluster with 20 < E CLU < 80 MeV 3) No requirements for E CLU < 20 MeV Low E radiative 

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 35 Normalization N TAG given by 175 pb -1 from 2002's data selftriggering Counting events in the distribution of secondary track momentum in the kaon rest frame p* Fit together signal and backgrounds Km2 and 3-bodies Efficiency related to DC reconstruction only (tracking plus vertexing), evaluated on data

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 36 p*(MeV) ev/MeV Efficency evaluated on a sample selected requiring a neutral vertex p* fit cut at 180 MeV/c fitting the p* distribution with  peak from EMC sample   peak requiring the   3-body decays from MC N  |FIT from MC 175 pb -1 (’02 data) N   1383 N   2309 Status: finalizing

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 37 Averages accounting for correlations:  0.040BR(K  3)  BR(Ke3) Tag K -  2Tag K +  2Tag K -  2Tag K +  2 Tag K -  2Tag K +  2Tag K -  2Tag K +  2  2 /dof for the 4 measurements: error accounts for The error accounts for the data and Monte Carlo statistics used in the fit, the MC statistics for the efficiency evaluation, the Data/MC efficiency corrections, and the systematics on the tag selection. It is dominated by the error on Data/MC efficiency correction. Still to be evaluated the systematics due to the signal selection efficiency, to the nuclear interaction, and to the momentum dependency of the tracking efficiency Kl3 preliminary results

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 38 V us from semileptonic decays Plot by F. Mescia

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 39 V us from semileptonic decays * Vusxf+(0) = (5) chi2/dof = 1.9/4 calcolato con Tau_L=50.84(23)ns (media KLOE) BR(Kl3) i 5 di KLOE, compresi i preliminary K+/-l3 fattori di forma, parametrizzazione quadratica, mediando su KLOE, KTeV,ISTRA+,NA48: lambda+' = (80) lambda+'' = (35) lambda0 = (95) * Unitarieta`: Vusxf_+(0) = (22) calcolato da f+(0)=0.961(8) (Leutwyler,Ross) Vud = (27) (Marciano) * Vusxf+(0) = (4) calcolato con Tau_L=50.99(20)ns (media KLOE-PDG) Tutti i BR Kl3 disponibili, compresi KLOE prel. K+/- Fattori di forma, c.s. KLOE  L = 50.84(23) ns Five KLOE BR(Kl3) Form factors quad. param. KLOE + PDG  L = 50.99(20) ns All available BRs Form factors quad. param. Imposing unitarity f+(0) = 0.961(8) (Leutwyler, Roos) Vud = (27) (Marciano)

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 40 00 E ,t ,x  ±± E ,x , t  K  tag tt pKpK pKpK t0t0 lKlK xKxK E,t,xE,t,x  trk+vtx evaluation The K track on the tagging side is extrapolated backwards to the signal hemisphere Step along the extrapolated kaon looking for the best neutral vertex Using timing of the neutral clusters fired by the  s from the   0 decay FV  40 cm ≤  ≤ 150 cm

Roberto Versaci e + e - collisions from  to  – Novosibirsk /26 41 Extremely well known within SM: R K SM = (  0.001)x10 -5 Probe  -e universality: non-universal terms from LFV sources in SUSY extensions At KLOE the measurement is extremely challenging, At KLOE the measurement is extremely challenging, especially the PID due to huge K  2 background O(4x10 4 ) Produced about 5 x 10 4 events with 2fb -1 Produced about 5 x 10 4 events with 2fb -1