Recent Belle and Babar Results on CP violation in B 0 → f K s

1. Recent Belle and Babar Results on CP violation in B0→ f Ks Slides extracted from : T. Browder’s talk at LP2003 Y.Sakai’s talk at CERN on 13 October 2003 + my personal comments Yuehong Xie 31 October 2003

2. Today’s focus: New physics or not? b→s penguin LP2003: Belle B0fKs “sin2f1” = -0.96  0.50 3.5 σ from SM +0.09 -0.11 LP2003: Babar B0fKs “sin2b” = +0.45  0.43  0.07 2.1 σ discrepancy WA Average B0fKs “sin2b” = -0.14  0.33 2.6 σ from SM b→ccs WA NEW 2003 WA: b  ccs “sin2b” = +0.736  0.049

3. Kobayasi-Maskawa Scheme CPV: due to a complex phase in the quark mixing matrix CKM matrix Important element of SM Unitarity triangle * VudVub Wolfenstein representation 2 (a) * Vtd Vtb 3 (g) 1(b) Vcd Vcb *

4. G ® - G ® 0 0 ( B f ) ( B f ) º d CP d CP A CP G ® + G ® 0 0 f ( B ) ( B f ) CP d CP d J/y Vcb B0 Vcs * KS  mixing J/y V* td Vtb 2(a) B0 B0 1(b) 3(g) KS V* Vtb td - Mixing Induced CPV (bccs) B0 ｆｃｐ ｆｃｐ = B0 mixing B0 B0 = -xfsin2f1 sin(DmDt)

5. - q p 2Iml 1+ |l|2 |l|2 -1 |l|2 +1 A A S = A = l = xf SM: bs Penguin phase = (cc) K0 - CP Violation in b sqq ACP = S sin(DmDt) + A cos (DmDt) C(Babar)=-A(Belle) Mixing induced CPV Direct CPV f + + New Physics with New Phase Sbs¹ Sbc , A can¹ 0 Sbs=sin2f1, A=0 “b  ccs: sin2f1” (SM reference)deviation

6. Flavor Tagging l+ u l- p+ ー d n ー n W+ ー W- ｃ ー ー ｓ ー K+ L ｂ B0 ｄ D* p- slow High-p (primary), low-p (secondary) leptons Strangeness (b  c  s) Fast p, slow p Figure of merit(Q) =ε(1-2 w)2a.k.a effective tagging efficiency Babar: Neural Net based approach, 28.1±0.7 % Belle: Likelihood based approach, 28.7±0.5 %

7. Flavor Tagging Performance _ B0  D*ln B0 –B0 mixing (OF-SF)/(OF+SF) ~(1-2 w)cos(Δm t) 12 r-bins, 6 divisions in r. B0 and B0 tags treated separately. _ • Efficiency > 99.5% • eeffective = 28.7  0.5% determined by data

8. Belle 2003 sin2f1: CP samples - 140 fb-1 , 152 x 106 BB pairs 81% CP=+1 2911 events are used in the fit (CP=-1)

9. sin2f1 Fit Result Preliminary 5417 events @ 140 fb-1 (78 fb-1:0.719±0.074±0.035) consistent with no direct CPV

10. Compare CP odd and CP even sin2f1 0.730.06 CP=-1 opposite! CP=+1 0.800.13 (statistical errors only) (all r-bins)

11. Systematic Uncertainty Small uncertainty in analysis procedure stat err. = 0.057

12. Measurement of sin2f1 (BaBar 2002) 81 fb-1 CP Eigenstate Sample hep-ex/0207042, PRL 89, 201802 (2002)

13. Current Resultsfor sin2f1 sin2f1 (Belle 2003,140 fb-1) =0.733±0.057±0.028 sin2f1 (BaBar 2002, 81 fb-1)=0.741±0.067±0.033 (no updated yet) sin2f1 (New 2003 World Av.) = 0.736±0.049 [PDG2003: 0.731 ±0.056] Good SM reference

14. b  s Penguin Decays mode Br(K0: x10-6)* B0fK0 7.8  1.1 theoretically clean B0h’K0 65.2  6.0 unexpected large Br B0 K+K-K0 25.2  2.7 CP=+1: 1.03  0.15  0.05 cf) B0 J/yK0 850 50 much smaller than Golden mode f + e.g.) squark penguin [* HFAG LP03 averages]

15. Hunting for phases from new physics 110 fb-1 70±9 Theoretically cleanest example: 140 fb-1 68±11 Belle In the SM, sin(2φ1)eff = sin(2φ1) (Bψ KS )

16. BaBar 2003:CPV in Bφ KS BaBar 2003: 110 fb-1 (A=0.38±0.37±0.12) BaBar 2003: sin2φ1eff (φ KS) = +0.45±0.43±0.07

17. BaBar 2003:Bφ KS Systematic Issues sin2φ1eff (φ KS) = -0.18±0.51±0.09 81 fb-1: (A=0.80±0.38±0.12) sin2φ1eff (φ KS) = +0.45±0.43±0.07 110 fb-1: (A=0.38±0.37±0.12) Data size increased and was reprocessed. Extensive checks with data and Toy MC. The large change is attributed to a 1σ statistical fluctuation.

18. BaBar 2003:Bφ KS Systematic Issues Systematics are small and well understood from bc cbar s studies

19. My comment: Direct CP violation in B±φ K± Acpdir(B±f K±) =+0.04  0.09  0.01 (hep-ex/0309025) Compared with Acpdir(B0f Ks) = +0.38  0.37  0.12 Model independent parameterization (R. Fleischer, hep-ph/0103121 ) [Acpdir(B±f K±) + Acpdir(B0f Ks)]/2 = o(1) + o(l2) [Acpdir(B±f K±) - Acpdir(B0f Ks)]/2= o(l) + o(l2) NPI=0 SM NPI=1 SM Higher precision of A measurement is needed to distinguish new physics from SM contribution

20. p+p- B0→h’KS p+p-h,rg gg p+p- B0→f KS K+K- p+p- B0→K+K-KS (K+K-¹ f) Penguin: B0h’KS ,fKS ,K+K-KS Nsig(fKS)=68(64%) 106 ev Nsig(K+K-KS)= 199(55%) 361 ev CP=even 103  16% Nsig(h’KS)=244(58%) 421 ev ∫L dt = 140 fb-1

21. Fit Results Fit sin2f1 B0 fKS B0K+K-KS B0h’KS

22. Comparison of 78/140fb-1

23. Systematic Errors

24. fKs K+K- mass Bφ KS : Systematic Effect of Backgrounds with CPV CP in the background: (7.2±1.7)% K+K- KS (measured) +1.9 : (1.6 )% f0 KS (CP=+1) -1.5 Effects: included in the systematic error S: +0.001/-0.084 Correlation between A and S ? A = -0.15 ±0.29±0.07 ±10MeV If A is fixed to zero, -xfS = -0.99±0.50 • [cf) K+K-KS: +0.54 ± 0.24, hKS: +0.43 ± 0.27]

25. fKs : Dt distribution Poor tags Good tags

26. Belle 2003: CP Asymmetry in Bφ KS 140 fb-1 Poor flavor tags Good flavor tags

27. Belle 2003: CP Asymmetry in Bφ KS 140 fb-1 Belle (A=-0.15±0.29±0.07) +0.09 Belle: sin2φ1eff = -0.96 ±0.50 -0.11 3.5σ off Current WA: sin(2φ1)=0.731±0.056

28.  fKs : sideband asymmetry

29. B±φ K± Control Sample -xfS No sin-like asymmetry. [ cf) B+ hK+: -xfS = +0.10 ± 0.14 ]

30. Lifetime Check consistent with PDG values t = 1.537 0.015 B0 t = 1.671 0.018 B

31. Toy MC Check : fKs [other modes are also reasonable]

32. Significance • B0K+K-KS, h´KS • Consistent with sin2f1. sin2f1 • B0  f KS • Likelihood curve • 3.5s deviation • from 0.731 (PDG2003) Belle claims “hint for new physics”

33. My comment on significance What is the uncertainty of the 3.5 s deviation from SM? 1.Theory uncertainty in SM is in the order of o(l2) (see page 38) 3.5 → 3.5 ± ? 2. Statistical fluctuation of “3.5” itself should be reported. From p31, s(d(S)) ≈ 0.1 + 0.9 -0.6 Number of s from SM:3.5 +0.9 -0.6 ± ? ) sdeviation from SM Final report: ( 3.5

34. “sin2f1” 2002 Status

35. “sin2f1” 2003 Status “SM” Av = -0.14±0.33 (2.6s away)

36. _ Summary of CPV in b(c c d) decays: Old Belle value used sin(2φ1eff ) Errors are large for these modes so that it is difficult to verify whether there is large penguin pollution. There is a 2.5 σ “hint” for penguin pollution in Babar’s result for BD* D*

37. Sensitivity to new physics phases The “Killer App” for the super B Factory

38. My comment: Precision requirement Acpmix (B0f Ks) - Acpmix (B0J/y Ks) = o(1) + o(l) + o(l2) NPI=1 NPI=0 SM (R. Fleischer, hep-ph/0103121 ) An experimental precision of O(l2) for measurement of mixing induced CP asymmetry is needed to distinguish between SM and new physics contribution in B0f Ks ! Can the B Factories reach this precision? Hopefully no! Can the super B factory do the work? Probably yes, if there will be a B factory one day which will be really super. How much better will LHCb do the work? To be demonstrated …

39. My comment: credibility • Belle: • Detailed report and many cross checks • Consistent result when using bigger data sample • Lower background with lower efficiency • the 3.5 s deviation includes neither SM theory uncertainty, which is about a few percent, nor fluctuation of error on S • Babar: • No detailed report • More than 1s change toward SM in 2003; • Higher background and higher efficiency • Helicity angle of f is accounted for in the likelihood to distinguish between f KS and combinatorial background, including K+K-KS.

40. Personal feeling • Statistics for b →s penguin modes are very limited • Purity for b →s penguin is ~ 60%; it is on average >90% for ccs channels. • There is indeed a trend to have a smaller effective sin(2b) in B decays with dominating or big penguin contribution. • So far it is still an experimental problem! Discrepancy between Babar, Belle and SM can still be explained as fluctuation. • Need a huge increase of statistics and more careful study of background, to establish a discovery of new physics or converge to SM!