Measurement of sin2  at B A B AR Douglas Wright Lawrence Livermore National Laboratory B A B AR For the B A B AR Collaboration 31st International Conference in High Energy Physics Amsterdam, 24 July 2002 Update B 0  (cc)K 0 Update B 0  D *+ D *- NewB 0  K 0 S NewB 0  J/   0

July 24, 2002Doug Wright, ICHEP Amsterdam2 11 22 33 Weak interaction of quarks in the Standard Model CP violation will arise from complex component of V ub, V td Unitarity Triangle CKM Matrix

July 24, 2002Doug Wright, ICHEP Amsterdam3 B 0 -B 0 mixing introduces time-dependant CP violation f CP b d d b u,c,t top quark box introduces : V tb V td *  m d = m H - m L WW Direct CP violation if multiple amplitudes with different phases Sensitive to overall phase of f even if no Direct CP Violation

July 24, 2002Doug Wright, ICHEP Amsterdam4 Theoretically clean: Tree level dominates and CP only from B 0 -B 0 mixing Relatively large branching fractions Clear experimental signatures CP asymmetry in B 0  (cc)K 0 f =  f e -i2   f =±1 B0B0 K0K0

July 24, 2002Doug Wright, ICHEP Amsterdam5  Boost:  = 0.55 Start the Clock Coherent BB pair B0B0 B0B0 Experimental technique at the  (4S) resonance  4S  e + e -   (4S)  B B Exclusive B meson and vertex reconstruction   KSKS B tag B rec  K-K- Flavor tag and vertex reconstruction 

July 24, 2002Doug Wright, ICHEP Amsterdam6 PEP-II top luminosity: 4.60 x cm -2 s -1 (exceeded design goal 3.0 x ) PEP-II delivered 99 fb -1 BaBar recorded94 fb -1 In this analysis On peak81 fb -1 88M BB pairs Off peak10 fb -1 PEP-II top luminosity: 4.60 x cm -2 s -1 (exceeded design goal 3.0 x ) PEP-II delivered 99 fb -1 BaBar recorded94 fb -1 In this analysis On peak81 fb -1 88M BB pairs Off peak10 fb -1 SLAC B Factory performance 9 GeV e - on 3.1 GeV e + Boost  = 0.55 IP beam size 147  m x 5  m

July 24, 2002Doug Wright, ICHEP Amsterdam7 B A B AR Detector SVT: 97% efficiency, 15  m z hit resolution (inner layers, perp. tracks) SVT+DCH:  (p T )/p T = 0.13 %  p T % DIRC: K-  separation GeV/c  GeV/c EMC:  E /E = 2.3 %  E -1/4  1.9 %

July 24, 2002Doug Wright, ICHEP Amsterdam8 Vertex and  t Reconstruction Reconstruct B rec vertex from charged B rec daughters Determine B Tag vertex from All charged tracks not in B rec Constrain with B rec vertex, beam spot, and  (4S) momentum Remove high  2 tracks (to reject charm decays) High efficiency: 95% Average  z resolution ~ 180  m (dominated by B Tag ) ( ~ 260  m)  t resolution function measured from data Beam spot Interaction Point B REC Vertex B REC daughters B REC direction B TAG direction TAG Vertex TAG tracks, V 0 s z

July 24, 2002Doug Wright, ICHEP Amsterdam9 B Flavor tagging method Using tracks with or without particle identification, and kinematic variables, a multilevel neural network assigns each event to one of five mutually-exclusive categories: Lepton tag: primary leptons from semileptonic decay Kaon1 tag: high quality kaons, correlated K - and  + s (from D * ) Kaon2 tag: lower quality kaons,  s from D * Inclusive tag: unidentified leptons, low quality K,  leptons No tag: event is not used for CP analysis b sc K-K- New and improved tagging method

July 24, 2002Doug Wright, ICHEP Amsterdam10 perfect tagging & time resolution Tagging errors and finite  t resolution dilute the CP asymmetry typical mistagging & finite time resolution Need to know mistag fraction w and  t resolution function R in order to measure CP asymmetry. Can extract these from data with B 0 -B 0 mixing events.

July 24, 2002Doug Wright, ICHEP Amsterdam11 Use self-tagged B flav sample to measure w and R Fully reconstruct self-tagged modes: B flav sample is x10 size of CP sample Apply B tag to other side, fit for B 0 -B 0 mixing B 0  D (*)-      a 1  N tagged =23618 Purity=84% B 0  J/  K *0 (K     N tagged =1757 Purity=96%

July 24, 2002Doug Wright, ICHEP Amsterdam12 Tagging performance from B flav sample CategoryEfficiency (  ) Mistag Fr. (w) Q=  (1-2w) 2 Lepton 9.1    0.3 Kaon    0.4 Kaon    0.4 Inclusive 20.0    0.3 Total 65.6   0.7 This new tagging method increases Q by 7% compared to the method used in our previous result: PRL87 (Aug 01). ~1-2w

July 24, 2002Doug Wright, ICHEP Amsterdam13 Recent addition: B 0  c K s where  c K + K -  0 or K + K s  - sin2  golden sample: (cc)K S (  f = -1) SampleN tagged Purity J/  K s (  +  - ) 97497% J/  K s (  0  0 ) 17089%  (2S) K s 15097%  c1 K s 8095%  c K s 13273% Total150692% Energy-substituted mass J/K s ( +  - ) J/K s ( 0  0 ) (2S)K s cKscKs  c1 K s

July 24, 2002Doug Wright, ICHEP Amsterdam14 sin2  samples: J/  K L and J/  K *0 J/ K L J/K *0 (K s  0 ) Signal J/  Bkg Fake J/  Bkg Data  Use m B constraint to determine p KL  J/  background shape estimated from Monte Carlo  Fake J/  background shape estimated from data sidebands  Vector-Vector mode: mixture of CP+ and CP-  Use angular analysis to determine CP- fraction  Treat CP- component as dilution  effective  f  f = +0.65±0.07 N tagged = 147 Purity = 81%  f = +1 N tagged = 988 Purity = 55%

July 24, 2002Doug Wright, ICHEP Amsterdam15 sin2  likelihood fit Simultaneous unbinned maximum likelihood fit of CP and mixing samples Fit Parameters34 total sin2  1B CP Mistag fractions w for B 0 and B 0 tags 8B flav Signal  t resolution function R 8B flav Background properties 17B CP +B flav (mostly from m ES sidebands in data) B lifetime fixed (PDG 2002)  B = ps Mixing frequency fixed (PDG 2002)  m d = ps -1

July 24, 2002Doug Wright, ICHEP Amsterdam16 sin2  =  sin2  =  Fit results  f =-1 f =+1 sin2  =  (stat)  (sys)

July 24, 2002Doug Wright, ICHEP Amsterdam17 sin2  fit results by decay mode Consistency of CP Channels: P(  2 ) = 57%

July 24, 2002Doug Wright, ICHEP Amsterdam18 (cc)K S with lepton tag N tagged = 220 Purity = 98% Mistag fraction 3.3%    t 20% better than other tag categories sin2 = 0.79  0.11 Dramatic effect in golden modes with lepton tag

July 24, 2002Doug Wright, ICHEP Amsterdam19 Cross-check on data control samples Observed no asymmetry as expected

July 24, 2002Doug Wright, ICHEP Amsterdam20 Sources of Systematic Error  sin2  Description of background events0.017 CP content of background components Background shape uncertainties Composition and content of J/  K L background0.015  t resolution and detector effects0.017 Silicon detector alignment uncertainty  t resolution model Mistag differences between B CP and B flav samples0.012 Fit bias correction0.010 Fixed lifetime and oscillation frequency0.005 TOTAL0.033 Steadily reducing systematic error:July 2002 = July 2001 = 0.05

July 24, 2002Doug Wright, ICHEP Amsterdam21 Standard Model comparison One solution for  is in excellent agreement with measurements of unitarity triangle apex Method as in Höcker et al, Eur.Phys.J.C21: ,2001  =  (1- 2 /2)  =  (1- 2 /2)

July 24, 2002Doug Wright, ICHEP Amsterdam22 Consistent with the Standard Model expectation of | f |=1 and nominal fit sin2  =  for (cc)K s modes alone. If another amplitude (new physics) contributes a different phase, then In the Standard Model | f | = 1 (which we assume in the nominal sin2  fit) S f  -  f sin 2  C f  0 Fit | f  | and S f using the clean (cc)K s modes (  f =-1, N tagged = 1506, Purity = 92%): Search for non-Standard Model effects in (cc)K S | f | =  (stat)  (syst) S f =  (stat)  (syst)

July 24, 2002Doug Wright, ICHEP Amsterdam23 Cabbibo-suppressed mode with tree level weak phase same as b  ccs Penguin contribution uncertain, expected to be small < 0.1 Tree Not a CP eigenstate, mixture of CP even (L=0,2) and CP odd (L=1) Resolve using angular analysis (in transversity basis) (b  ccd) mode B 0  D *+ D *- N tagged = 102 Purity = 82% Reconstruct D *+  D 0  + or D +  0, but not both to  0 mode B0B0 D * - D * + B0B0 D * - D * + V cd

July 24, 2002Doug Wright, ICHEP Amsterdam24 CP composition of B 0  D *+ D *- We measure CP odd fraction (corrected for acceptance) to be small: R  = 0.07  0.06 (stat)  0.03 (syst)

July 24, 2002Doug Wright, ICHEP Amsterdam25 CP asymmetry fit B 0  D *+ D *- | + | = 0.98  0.25 (stat)  0.09 (syst) Im( + ) = 0.31  0.43 (stat)  0.10 (syst) If penguins are negligible, then Im( + ) = - sin2  Im( + ) measurement ~2.7  from BaBar sin2  in charmonium, assuming no penguins. Improved fitting strategy since winter conferences: Parameterize in terms of CP even ( + ) and odd (  ) components, include angular information from partial-wave analysis Fix CP odd component to  =1, Im(  ) =

July 24, 2002Doug Wright, ICHEP Amsterdam26 B0B0 B0B0 D*+D*+ D _ D*+D*+ D+D+ D* _D* _ D+D+ D* _D* _ CP conjugation strong phase (b  ccd) mode B 0  D *+ D - D * D 56 fb -1 N tagged = 85 Purity = 52% S +- =  1.41  0.20 C +- = 0.53  0.74  0.13 S -+ = 0.38  0.88  0.05 C -+ = 0.30  0.50  0.08 Update to full data set in progress D*D not CP eigenstate Possible strong phase contribution (and still have penguins) Different (but related) decay time distributions for B 0  D *+ D - B 0  D *- D +

July 24, 2002Doug Wright, ICHEP Amsterdam27 (b  ccd) mode B 0  J/   0 Cabibbo and color-suppressed mode with comparable tree and penguin contributions Ntagged= 49 Purity = 59%  f = + 1 V cd Tree: ~V cb V cd * ~ O( 3 ) same weak phase as b  ccs Penguin:~V cb V cd * + V ub V ud * ~ O( 3 ) adds additional weak phase

July 24, 2002Doug Wright, ICHEP Amsterdam28 CP asymmetry fit for B 0  J/   0 In absence of penguins C  =0, S  = - sin2   f = + 1

July 24, 2002Doug Wright, ICHEP Amsterdam29 Charmless decay dominated by (b  sss) gluonic penguins Weak phase same as b  ccs, but sensitive to new physics in loops sin2  from penguin mode B 0   K S N tagged = 66 Purity = 50%  f = - 1 Small branching fraction O(10 -5 ) Significant background from qq continuum Using only   K + K -

July 24, 2002Doug Wright, ICHEP Amsterdam30 CP asymmetry fit for B 0   K S Fix |  K | = 1, fit: S  K = (stat)  0.09 (syst) Cross check on B +   K + S  K = 0.26  0.27 If no new physics, S  K = sin2 Analysis of B 0   ’ K S in progress  t (ps)

July 24, 2002Doug Wright, ICHEP Amsterdam31 Begun to probe the same CP-violating phase and possibly new physics via penguin modes: B 0  K 0 S B 0  J/   0 and open charm modes: B 0  D *+ D *- B 0  D *+ D - Conclusion New measurement of sin2  from charmonium modes (88 x10 6 BB) Submitted to PRL July 17, 2002 (hep-ex/ ) Results have been improving by more than just luminosity gain The Standard Model remains unscathed, but the high statistics future of BaBar will provide further opportunities to challenge the theory. (First BaBar results) Million BB pairs July 00 Feb 01 July 01 Mar 02 July 02 sin2  = ± (stat) ± (syst)

July 24, 2002Doug Wright, ICHEP Amsterdam32

July 24, 2002Doug Wright, ICHEP Amsterdam33 sin2  in subsamples

July 24, 2002Doug Wright, ICHEP Amsterdam34 Tagging performance Category Efficiency (  )Mistag Fr. (  )  MistagQ=  (1-2  ) 2 Lepton 9.1     0.3 KPiorK 16.7     0.4 KorPi 19.8     0.4 Inclusive 20.0     0.3 Total 65.6   0.7

July 24, 2002Doug Wright, ICHEP Amsterdam35 Monte Carlo correction We evaluated the size of any potential bias on sin2  by fitting the full MC in two ways: Fitting data-sized signal MC samples with mistag fractions and  t resolution fixed to the MC truth values (see plot). Average bias =  Same as above except mistag fractions and  t resolution from Breco MC. Average bias =  One possible source of bias comes from neglecting the known correlation between the mistag fractions (or dilutions) and sigma  t. Estimates from toy and full MC inticate a bias at the level of We correct the fitted sin2  by subtracting and assign a systematic error of to this correction.