F Decays Working Group Activities

1. FDecays Working Group Activities GENERALMEETING systematics studies concerning theshad measurement first results in h  p+ p- p0 Dalitz plot parameters h  g g g: wg background and a new upper limit Federico Nguyen - October 10th 2002

2. GENERALMEETING Trackmass, Acceptance and Filfo studies for the shad measurement

3. some sourcesof background can be seen in the Mtrk - distribution: -p+ p-p0:at high values of Mtrk which is Mpp2 dependent -m+ m- g:at Mtrk 104 MeV Trackmass Systematics (I) p+ p- p0 Mtrack (MeV) p+ p- g mp m+ m- g M2pp(GeV2)

4. Trackmass Systematics (II) 2 MC samples differing only for Fp values eFZK(Mpp)  eLNF(Mpp) atthe level 2% Switched OFF PPGTAG and PPFILT

5. Geometrical Cuts Systematics (I) we studied acceptance cuts dependence on Fp parameters by generating stand alone ppg MC samples (NTOT=106 events) with different Mr or Gr values covering the whole phase space our acceptance selection: 400 < +,-<1400 and (pt>160 MeV or |pz| >90 MeV) 00 <  < 150 or 1650 <  < 1800 E > 10 MeV there is no significant dependence on Fp parameters in acceptance studies

6. geom [(772.6)-(770)]/(772.6) Mpp2 (GeV2) [(772.6)-(780)]/(772.6) Mpp2 (GeV2) Mpp2 (GeV2) Geometrical Cuts Systematics (II) dependence onMr - we fixed the other parameters

7. [(143.7)-(140)]/(143.7) Q2 [(143.7)-(150)]/(143.7) Q2 Geometrical Cuts Systematics (III) dependence onGr - we fixed the other parameters geom Mpp2 (GeV2) Mpp2 (GeV2) Mpp2 (GeV2)

8. Filfo Systematics (I) FILFO/par=5: ~ 2000 refiltered files 6 pb-1 from different periods of 2001 DATA taking • Filfo efficiency is evaluated from DATA by reprocessing RAW • without using Filfo as a filter • to overcome the long • reconstruction time, a prefilter, based on CALORIMETER and PATTERN RECOGNITION, is used, then TRACK, VERTEX modules are called... Mpp2 (GeV2) errors ~0.5-1% per bin Mpp2 (GeV2) Overall efficiency: DATAFILFO, SMA= (94.82±0.07) %

9. Filfo Systematics (II) • signal events vetoed by FILFO/par=5 • by the following: • Bhabha-filter (83.5%) • Cosmic-filter (13.9%) • Machine Background filter (4.3%) 4734 events BHABHA[3955events] COS[659 events] MB[205 events] Mpp2 (GeV2)

10. Filfo Systematics (III) FILFO/par=5: 2  105 stand alone MC events • we wanted to know the difference in applying Filfo at the beginning or at the end of our analysis chain • resulting systematic correction: •  = (-0.71 ± 0.07)% Overall efficiency (Filfo at first): MCFILFO, SMA= (96.66±0.07) % Mpp2 (GeV2) systematic Filfo checks have been performed we plan to study Filfo changes in different periods of DATA taking Mpp2 (GeV2) Overall efficiency (Filfo at the end): MCFILFO, SMA= (97.37±0.07) %

11. GENERALMEETING f  h g, h  p+ p- p0: resolution, fit procedure and results

12. Dalitz Plot Parameters: Definition Ti kinetic energy of the pi Q mh - 2mp+ - mp0 usual decay rate in terms of slope parameters: G(h  p+ p- p0 )  |M(X,Y)|2 |M(X,Y)|2  1 + aY + bY2 + cX2

13. Dalitz Plot Parameters: Resolution (I) Y resolution has been obtained by using 1.3105 MC events it can be estimated either from p or from gg (fromp0) score = 0.034 score = 0.042

14. Dalitz Plot Parameters: Resolution (II) score = 0.028

15. as a 1st approximation b  c  0, , that has to be fitted on: Fit Procedure and Results (I) where: N(X,Y) = # of events in the Dalitz plot e(X,Y) = efficiency (from an independent MC sample) PS(Y) = phase space as a function of Y bins

16. Fit Procedure and Results (II) 16 pb-1from 2000 DATA have beenstudied (the same selection for the mixing angle analysis,Kloe Memo 266) dN/dY ~ 1 + aY Y

17. Fit Procedure and Results (III) resolution studies have been optimized, 1st glance at DATA shows sensitivity to the quadratic slope parameter dN/dY ~ 1 + aY + bY2 Y

18. GENERALMEETING f  h g, h  g g g: selection cuts, possible background and preliminary limit

19. after kin. fit kinematic fit has been studied on MC (Ntot=3104 events) before kin. fit Eg (MeV) Eg (MeV) Analysis Requirements Pre-selection: 4 prompt clusters Angle between two clusters > 15° Etot(clusters) > 900 MeV; Ptot < 110 MeV • Final Selection: • Pkin-fit(2>20) > 0.01 • |M - 140| > 30 MeV • E vs. cosplane cut • |Mggg– 550|< 45 MeV

20. e+ e- w g1 4 p0 g2 Background Problem w peak  Mrecoil of g1 (MeV) inconsistent with MC estimated background 59events perhaps a missing background: wg can be rejected by p0 cut, still we must know the tails g1 non p0 least energetic g Mgg (MeV)

21. Preliminary Results on 2001 DATA from MC: sGauss = 13 MeV PAW raw fit: a+bx+cx2+Nsignal*Gauss Mggg (MeV) Mggg (MeV) Nsignal = 9 26, corresponding to 616 events out of 142 pb-1 Nupper (90% C.L.) = Nsignal + 1.28*sNsignal = 42, eTOT = 21% BR(hggg) < Nupper/(L*sF*BR(f hg)*eTOT) = 3.5 10-5

22. Summary and Outlook shad: NO significant Fp parameters dependence both in acceptance and trackmass, filfo efficiency has been studied, we started evaluating trackmass efficiency from DATA. h  p+ p- p0 : resolution and fit procedure have been checked, work ahead: genuine 2D fit, 2001-2002 DATA analysis. h  g g g: wg background has been studied, but a MC code is needed, reachable limit (~ 500 pb-1): BR(h  g g g) < 2 10-5 .