P Massarotti Charged kaon lifetime Paolo Massarotti Charged

P. Massarotti Charged kaon lifetime Paolo Massarotti Charged Kaon Meeting 10 June 2005

Strategy Signal selection • Self triggering muon tag • K track on the signal side • Decay vertex Signal K extrapolated to the IP. correction applied along the path. K Kmn tag K Li m d. E/dx Li = step length Vertex efficiency and resolution functions needed wrt the proper time of the Kaon Paolo Massarotti Charged Kaon Meeting 10 June 2005 vtx

Why self-triggering tag? Tag self trg Tag eff Tag. Trg selfeff trg Paolo Massarotti Charged Kaon Meeting 10 June 2005

Definition of signal sample: • Cut at 100 Me. V on boost with 96% efficiency 70% signal 30% “bg” (not kaon bg less than 0. 2 %) Daughter P* with kaon mass hypothesis Paolo Massarotti Charged Kaon Meeting 10 June 2005

Reconstruction Efficiency: Neutral vertex technique Considering only kaon decays with a p 0 K X p 0 Ep, xp, tp X gg We look for the vertex asking p. K tm Ø clusters on time: (t - r/c)g 1 = (t – r/c)g 2 Kmn tag x. K l. K t 0 p K Ø p 0 invariant mass p 0 Eg, tg, xg Ø agreement between kaon flight time and clusters time Paolo Massarotti p± Charged Kaon Meeting 10 June 2005

Efficiency measurement e. G = e. Tr K e. Tr sec e. V kaon Paolo Massarotti Charged Kaon Meeting • charged daughter 10 June 2005

Efficency comparison Monte. Carlo kine vs Monte. Carlo reco fit window definition Fit window 15 : 30 ns MC reco MC kine a. G = (101. 7 0. 7) x 10 -2 b. G = (-. 59 3. 6) x 10 -3 T (ns) Paolo Massarotti Charged Kaon Meeting T (ns) 10 June 2005

Data and Monte. Carlo compared Data MC reco A 0 = (110. 7±. 012) A 1 = (2. 8±. 5)x 10 -3 Fit window 15 : 30 ns Paolo Massarotti Charged Kaon Meeting 10 June 2005

Resolution functions: MC (8: 10) ns (20: 22) ns t Reco – t true Convolution of the exponential decay function with bin by bin resolution function is needed Paolo Massarotti Charged Kaon Meeting 10 June 2005

Resolutions: Data vs MC To evaluate resolution on data we use the difference between T obtained with neutral vertex and T obtained with charged vertex: Data MCfit reco we Tneu – Tchar on MC and on Data and we obtain the smearing factors for MC resolutions Paolo Massarotti Charged Kaon Meeting 10 June 2005

Smearing factors MC reco Data We have fitted these distributions using 7 Gaussians: 3 for the core 4 for the peaks at ± 2. 7 and ± 5. 4 Smearing factors ~ 10% Paolo Massarotti Charged Kaon Meeting 10 June 2005

Measures of t data 15 : 30 t data ±. 06 Range (ns) 12. 420 16: 29 t data ±. 06 (ns) 12. 432 15 : 29. 12. 456 16: 28. 5 12. 425 15: 28. 5 12. 451 16: 28 12. 416 15: 27. 5 12. 489 15. 5: 28. 5 12. 447 15. 5: 29 12. 452 15. 5: 28. 12. 449 Range Paolo Massarotti Charged Kaon Meeting 10 June 2005

Fit stability : bin size t (ns) range = 16. : 26. ns Systematic effect due to bin size is negligible N bin Paolo Massarotti Charged Kaon Meeting 10 June 2005

Systematic effects: first estimate BP (ps) 10 DC Wall Fit stability Smearing factors (ps) range (ps) 25 25 Obtained making fit with and without B. P. or D. C. Wall and 10% uncertainty on wall thickness 30 Obtained making fit with and without smearing factors Systematic error = 47 ps Paolo Massarotti Charged Kaon Meeting 10 June 2005

Result t+ = (12. 456± 0. 052 stat ±. 047 syst) ns KLOE using 2 millions of events Paolo Massarotti Charged Kaon Meeting 10 June 2005

Conclusions The value obtained, t = (12. 456 ± 0. 052 stat ±. 047 syst)ns, has to be compared with Particle Data Group, t = (12. 385± 0. 024)ns. We have studied decay reconstruction efficiency and resolution functions on data Further systematic checks on signal selection and neutral vertex sample are on the way We have to complete the measure on negative charge and Work is in progress for the “time” measurement Paolo Massarotti Charged Kaon Meeting 10 June 2005