1

Bd Mixing Measurements with the BaBar Detector

Cecilia VoenaUniversità “La Sapienza”

& INFN Roma

ICHEP2002

on behalf of theBABAR Collaboration

2

B0B0 Mixing-Proceeds through box diagram

22 2 2 2 2

02( / ) | |

6 d d d

Fd w B t W B td B B

Gm m e S m m m V B f

2 2(210 40MeV)d dB BB f

(D. Becirevic and L. Lellouch, at this conference)

- Extraction of Vtd limited by theoretical hadronic uncertainties

B0

- md sensitive to Vtd CKM matrix element

if q/p 1 => CP violationin mixing (assuming CPT invariance )

B0 B0

b

b

d

d

W W

t

t

V*td

H H11H12

H21H11

q

pH21

H12

1

2

md 2Re H12H21 1

2

23533+00-23 MeV

Mass difference betweenHamiltonian eigenstates

md constraint in(,) plane

F.Parodi

3

B0B0 Mixing at Asymmetric B Factories

23 x 106 BB (B0D*-l+ and dilepton)32 x 106 BB (hadronic)

-Mixing and lifetime with B0D*-l+l decays-Mixing with hadronic decays (PRL 88:221802[2002])-Mixing with dilepton events (PRL 88:221803[2002])

new result

Before B factories:

At BaBar, asymmetric B factory, we measure:

-can perform time dependent measurements-high statistics sample of fully reconstructed Bd,no Bs

md = 0.472 ± 0.017 ps-1

4

B0B0 Mixing with Fully Reconstructed B0

(4s)

= 0.55

Tag B

z ~ 160 m Reco Bz ~ 80 m

l+z|z|> 250 m

K

D0

-K+

-

K+

+

-

1. Reconstruct one B meson in the flavor eigenstates D*-l+l or hadronic (e.g.D*-+ )2. Reconstruct the vertex position

3. Reconstruct inclusively the vertex of the “other” B meson (BTAG)4. Determine the flavor of BTAG

5. Compute the proper time difference t z/c 6. Fit the t spectra of mixed (B0B0 or B0B0) and unmixed (B0B0)events

e- e+

5

t Distribution of Mixed andUnmixed Events

Decay Time Difference (reco-tag) (ps)

UnMixedMixed

0

10

20

30

40

50

60

-8 -6 -4 -2 0 2 4 6 8

perfect flavor tagging & time resolution

Decay Time Difference (reco-tag) (ps)

UnMixedMixed

0

10

20

30

40

50

60

-8 -6 -4 -2 0 2 4 6 8

realistic mis-tagging & finite time

resolution

Unmix

xMi

f (Δ t) 1 1 2 cos( ) ResolutionFunction4

Bd

d

d

| Δt |/τ

B

e tτ

mw Δ Δ

Decay time distribution for signal events

oscillation frequencyB0 lifetime Mistag probability

B0B0

B0B0 ,B0B0

-+

6

B0D*-l+ Selection

• Cannot reconstruct B0 mass and energy because of the missing => use m(D*)-m(D0) distribution

• Reconstruct candidate D* using full decay tree

D*+ D0+ , D0 K-+,K-+0,K-+-+,Ks+-

• Combine with lepton candidate (p* > 1.2 GeV)

• Require consistency with D*-l kinematics (angles, missing )

signal

background

7

Sample Composition

signal sample

Peaking background: real D*non from semileptonicdecays-uncorrelated lepton from other B-fake lepton-continuum events

Combinatoric: fake D*m(D*)-m(D0) (MeV)

• Select several control samples to characterize (fraction,t shape) the main sources of backgrounds

8

Flavor Tagging

For electrons, muons and Kaons use the charge correlation

b c

d d

l-

B0 D, D*

W-

0

0

l

l

B

B

Lepton Tag

b

d

B0

W- W+c s

d

0

0

0

0

kaons

kaons

Q

Q

B

B

Kaon Tag

Hierarchical, Mutually Exclusive, Tagging CategoriesHierarchical, Mutually Exclusive, Tagging Categories

Neural network exploits information carried by non-identified leptons and kaons, soft pions from D* decays

K* 0 K- +

Tagging Category

Efficiency(%)

Mistag Fractionw(%)

Q=(1-2w)2

(%)

Lepton 11.1 8.6 7.6

Kaon 34.7 18.1 14.1

Neural Net

54.2 37.7 3.3

9

B0 D*-l+ Mixing Likelihood Fit Results

• Simultaneous fit to all signal and background samplesfloat m, B0 , mistag probabilities, signal andbackground t resolution parameters, fraction of charged B

md=0.492±0.018±0.013ps-1

Mixing asymmetry

( ) ( )( )

( ) ( )

(1 2 )cos( )

unmixed mixedmix

unmixed mixed

N t N tA t

N t N t

w m t

B0 =1.523+0.024-0.023±0.022ps

Preliminary

+

-

Floating parameters:Innermost m, B0

m, B0, mistag m, B0, fB+

m, B0, fB+,mistag All signal t par Default fit

errorellipse

correlation coefficient (m, B0) = -0.22

6B0-6B0

10

Mixing with Hadronic Events

• Fully reconstructed B0D(*)-+/+/a1+,J/K*0

md=0.516±0.016±0.010 ps-1

mixA

PRL 88[2002]

• All t and mistag parameters simultaneously extracted from data

used for sin2 measurement

• lifetime measurements with hadronic events

0 = 1.546 0.032 0.022 ps = 1.673 0.032 0.023 ps /0 = 1.082 0.026 0.012

PRL 87 [2001]1-2w

/2md

6347 B0

(purity ~86%)• Largest syst. is B0 lifetime

11

Mixing with Inclusive Dilepton Events

Very precise mixing measurement• Select events with 2 high

momentum leptons (sample contains ~50% B+)

• Largest syst. is B0 lifetime and resolution function parameterization

md= 0.493±0.012±0.009 ps-1

PRL 88[2002]

with the same sample:T and CP violation in mixing

|q/p|=0.998±0.006±0.007

PRL 88[2002]

99010 dileptons(B0B0 purity ~40%)

12

B0 Lifetime with Partially Reconstructed B0D*-l+

Kinematic constraints at the (4S):

-pD* = p+

-M2= (pB0- pD* - pl)20

B0 = 1.529 ±0.012 ±0.029 psPRL 89[2002]

reconstructlepton and only

in a small cone around D*direction, and from simulation

missing neutrino

13

Summary• New simultaneous measurement

of md and B0 with the exclusive

B0D*-l+l sample

md= 0.492±0.018±0.013 ps-1

B0 = 1.523+0.024-0.023±0.022 ps

• Combined BaBar result for

Bd mixing frequency:

Preliminary

md= 0.500±0.008±0.006 ps-1

2% error

14

Backup slideSignal t Model

15

Backup slide

Background t Model

16

Backup slideSystematic error

17

Backup slideSystematic error

18

Backup slide

Tagging Category

Efficiency(%)

Mistag Fractionw(%)

B0/B0 diff.w(%)

Q=(1-2w)2

(%)

Lepton 11.1 0.2 8.6 0.9 0.6 1.5 7.6 0.4

Kaon 34.7 0.4 18.1 0.7 -0.9 1.1 14.1 0.6

NT1 7.7 0.2 22.0 1.5 1.4 2.3 2.4 0.3

NT2 14.0 0.3 37.3 1.3 -4.7 1.9 0.9 0.2

All 67.5 0.5 25.1 0.8

Mistag fraction determined from simultaneous fit to Bflav sample

19

Backup slide

Signal Sample

Combinatoric Enriched : Sideband

Peaking Bkgs “l not from Brec” Enriched : D* with

flipped l

Continuum D* Enriched : Off-resonance Data

Fake l Enriched :l candidate fails e, ID

20

Backup slide