CP violation at B-factoris

• Introduction• 1/• 2/• 3/• Vub, D mixing, Bs• Conclusion

Pavel KrokovnyKEK, Tsukuba, Japan

KEK

Unitarity triangle

1)1(

2/1

)(2/1

23

22

32

AiA

A

iA

VVV

VVV

VVV

V

tbtstd

cbcscd

ubusud

* *ub ud tb td* *cb cd c

* * *ub ud cb cd tb td

b cd

V V V V

V V V V

V V V V V Vi k :=1, = + + =0

+ 1

3

+ = 0

Using unitarity requirement:

We need to measure all the angles to test SM.

CPV in Mixing

B0 fCP

B 0

Af

Af

2ie

2

B0

fCPB 0Af

Af

2ie

2

02

0

( )

( )CP

i CPf

CP

A B fe

A B f

0 02

0 0

2

2

2Im ( ) ( ) 1 | |( )

( ) ( )1 | |

[ sin cos ] 1 | |

CP

CPCP

CP

C

CP C

CP

P

PCP

f

CP CPff

CP CP

fCP

f

f

f

f f

B f B fA t

B f B f

m t m t C

S

CS

Difference in decayrate for B0 and B0

CP Violation

Direct CPV in charged B Decays

wk sti iδ2A e e

B f

wk stΓ(B f) Γ(B f) for 0 and δ 0

2

1 2( ) wk sti iB f A A e e st

B fwk

wk

B f2

1 2( ) wk sti iB f A A e e

One rate asymmetry is not sufficient to extract physical parameters:

• measure A and A, but need A1, A2, wk, st

CP violation through interference of decay amplitudes

1A

1A

2A

2A

1 22 2

1 2 1 2

2 sin( )sin( )( ) ( )

( ) ( ) 2 cos( )cos( )wk

CPt

w

s

stk

A Af fA

f f A A A A

dj

j d

G - G= =

G +G + +

KEKB and Belle

KEKB Collider

3.5 GeV e+ & 8 GeV e– beams

3 km circ, 11 mrad crossing angle

L= 1.65 x 1034 cm–2s–1

world record

L dt = 740 fb–1 @ Υ(4S)+off(~10%)

Event shape*2 *2

ES beam Bm E p * *B beamE E E

(mES) 3 MeV (E) : mode dependent

qq events(q=u,d,s,c)

BB events

(4S) rest frame

Event shape variables examples:

Kinematics and event shape

1 from the “golden” bccs mode-

Measuring cos2

B → D(*)0 h0 with D0 → KS(same weak phase

as B0 → J/K0)

cos2 consistently measured to be > 0!

sin21=0.75 ±0.44±0.22 cos21=1.79 ± 0.53 ± 0.33cos21>0 @ 98.3% CL

cos>0 solutions excluded:

β = (21.3±1.0)°

CPV in B0 → D+D-

sin2 from Penguin Decays

• no weak phase in b→(qq)s penguin decays– expect to measure S = sin(2) [just as in B → ψ KS]– contributions from suppressed diagrams expected to be

small (sin(2) = sin(2eff) sin(2) ~ 0.01-0.1)• if new physics introduces weak phase in decay, we could

measure something different than sin(2)

newphysicsb

sssd

dg

t

SK

0B,

W

Standard Model New Physics

b

d0B

sssd SK

,

NP weak phase no weak phase from decay

sin2 in bsqq Penguins

No significant deviations from the WA from the value from the bccs modes: sin2 = 0.68±0.03

sin2

sin2eff 1

eff

HFAG advises extreme caution when interpreting this average (assumption of Gaussian errors not justified for all measurements)

Determination of 2()Time-dependent CP asymmetry: )cos( )sin( )( tmCtmStA dd

Tree:

00-0 A A2/1 A

Penguin:

)sin(2 )sin(21 0 eff2 CSC

Without penguin: Including penguin:

Use isospin relations to estimate the penguincontribution:

-00 A A Neglecting EWP, h+ h0 (I=2)=pure tree

Gronau-London, PRL, 65, 3381 (1990)Lipkin et al., PRD 44, 1454 (1991)

0 C )2sin( S

2() Measurement using B

.)(04.0.)(10.061.0

.)(05.0.)(08.055.0

syststatS

syststatA

t (ps)

PRL 78, 211801 (2007)

Sorting out penguin pollution

Triangle analysis6 observables6 unknowns

WA values

°±°

B isospin analysis

WA values

°< 2<°

2

21

22

21

2

21

2

coscossinsin14

1

4

9

coscos

1 LL ff

dd

d

B0+– has 3 polarization states with different CP eigenvalues

Fortunately, longitudinal polarization dominates pure CP-even stateNo significant 00 signal small penguin contribution

B CP fit results

PRD 76, 011104 (2007)

B →

0 3.6 63.9

0 6

0 0 0 6

+ +0.026L 0.013

(25.5 2.1 ) 10 [arXiv: 0705.2157v2]

(16.0 2.2 2.3) 10 [PRL 97, 261801 (2006)]

(1.07 0.33 0.19) 10 [PRL 98,111801 (2007)]

f 0.992 ± 0.024 [arXi

Br B

Br B

Br B

+0 +0.023L 0.02700L

0

0.05-0.06

v:0705.2157] f 0.905 ± 0.042 [PRL 97, 261801 (2006)]f 0.87 ± 0.13 0.05 [PRL 98,111801 (2007)]

0.12 0.13 0.10 [PRL 97, 261801 (2006)]

0.17 0.20 [arXiv:0705.

CPA

S

2157]0.01 0.11 0.06 [arXiv:0705.2157]C

Small penguins !

Dominantly longitudinally polarized !

Large branching fraction !

arXiv: 0705.2157v2

B0 tag

B0 tag

BABAR

Time-dependent CP asymmetries in B0 →

0B BaBar: 383 MBB

729 60 signal events

with B →

| |16.5 (90% C.L.)

effo

With more data information onS00 and C00 will allow to resolve

discrete ambiguities

00

000.5 0.9 0.2 0.4 0.9 0.2

SC

0 0 085 28 17 (3.6 σ significant, incl. syst.)

B

0 0 0 6

00L

(0.84 0.29 0.17) 10f 0.70 ± 0.14 0.05

Br B

arXiv:0708.1630

B0()0 Dalitz Analysis

B0

B0–

Various (24) Patterns of Interferences Information on Relative Phases

Interference by B0B0 oscillation + Interference Between , ,

Dalitzt

Pentagon analysis

°< 2<°

PRL 98, 221602 (2007)

2 results:

B

BB

~90o±10o

Determination of

Relative phase: (B– DK–), (B+ DK+)

includes weak (γ/φ3) and strong (δ) phase.

3*1 ~~ AVV uscbA i

csub ei(ρ~AλV~V ~)3*2A

If D0 and D0 decay into the same final state, 000~ DreDD i

B– D0K–: B– D0K– :

Amplitude ratio:

Possible D0 / D 0 final states:CP eigenstates, KK)Flavor eigenstates (K)Three-body decays (KS)

Gronau & London, PLB 253, 483 (1991) Gronau & Wyler, PLB 265, 172 (1991)Atwood, Dunietz, & Soni, PRL 78, 3257 (1997),Atwood, Dunietz, & Soni, PRD 63, 036005 (2001)Giri, Grossman, Soffer, & Zupan, PRD 68, 054018 (2003)Bondar, PRD 70, 072003 (2004)

0 0| ( ) | / | ( ) |~ 0.1r A B D K A B D K

Parameters are obtained from the fit to Dalitz distributions of D Ksπ+π– from B±DK± decays

Dalitz analysis method

Using 3-body final state, identical for D0 and D0: Ksπ+π-.Dalitz distribution density:

22222 ||),(

ssss KKKK

dmdmmmd A

222 |),(| ss KKmmA

(assuming СР-conservation in D0 decays)

000~ DreDD i

),( 22 ss KK

mmf is determined from D*– D0π–, D0 Ksπ+π– decay model uncertainty of the result

),,( 3r

A.Giri, Yu. Grossman, A. Soffer, J. Zupan, PRD 68, 054018 (2003)A.Bondar, Proceedings of the Belle Workshop, September (2002)

Dalitz analysis: sensitivity to the phase

Belle/Babar results on γFit parameters are x= r cos(φ3+δ) and y= r sin(φ3+δ)

(better behaved statistically than r, φ3 and δ)Belle/Babar measurements in good agreement

Note: σγ depends significantly on the value of rB

Contours do not include Dalitz model errors

Contours do not include Dalitz model errors

rB

2γ

Dalitz analysis (Belle)

BDK

BD*K

BDK*

Combined for 3 modes: φ3=53°+15 3° (syst)9° (model)

8°<φ3<111° (2σ interval)

rDK =0.159+0.054 0.012(syst)0.049(model)

CPV significance: 78% rD*K=0.175+0.108 0.013(syst)0.049(model)

rDK*=0.564+0.216 0.041(syst)0.084(model)

-0.099

-0.050

-0.155

-18

φ3=66 -20 °(stat) φ3=86 -93°(stat) φ3=11 -57°(stat)+37 +23+19

from BD(*)0K, D0KS (Babar)

B- DK- B+ DK+

Dalitz distribution of selected B± candidates

2

1

Stat Syst Model

rB 0.142 rB 0.198 0.016 rB

* 0.206 rB* 0.282

=(92±41±11±12)°

Constraints of the Unitarity Triangle

Estimated average with GLW, ADS, Dalitz and sin(2φ1+φ3)

~80o±30o

Model-independent approachA.Bondar, A.Poluektov hep-ph/0510246

50 ab-1 at SuperB factoryshould be enough for model-independent γ/φ3 Measurement with accuracy below 2°

~10 fb-1 at ψ(3770) needed to accompany this measurement.

A.Giri, Yu. Grossman, A. Soffer, J. Zupan, PRD 68, 054018 (2003)

rB=0.2

Summary in pictures

CPV angles

CP-conserving quantities

Summary in pictures

Summary

φ1 is measured with 1º accuracy. Still there is room for improvement.

Remarkable progress in φ2 measurement: accuracy of O(10º) is achieved using and Still limited by statistic - improvements in the future.

The φ3 remains the most difficult angle of the Unitarity

Triangle to measure. Good perspectives with higher statistics since the theoretical uncertainties are very low. Model error can be eliminated using charm data: CLEOc/BES.

No indication for a New Physics is found so far...Much more data is necessary LHCb, Super B factory.

PEP-II and BaBar

e-

e+

1.5T magnet

ElectromagneticCalorimeter (EMC)

Detector of Internally Reflected Cherenkov Light (DIRC)

Drift Chamber (DCH)

Silicon Vertex Tracker (SVT)

Instrumented Flux Return (IFR)

3.1 GeV e+ & 9 GeV e– beams

L= 1.12 x 1034 cm–2s–1 (June 28, 2006)

L dt = 380 fb–1 @ Υ(4S)+off (~10%)

Future plans of KEKB• Next milestone

– Accumulation of 1000 fb-1 (740 fb-1 at present)

• Near term plans– L = 2 x 1034cm-2s-1

• Benefits of crab crossing• Higher beam current (HER) 1400 -> 1500mA with several reinforcements of vacuum components• Search for better machine parameters

– Beam emittance, x*, x etc.• Long-term future plan

– SuperKEKB• Major upgrade plan of KEKB• Design luminosity: ~ 1036cm-2s-1 • Not yet approved (Construction hope to start in FY 2009)

Bρρ

2

21

22

21

2

21

2

coscossinsin14

1

4

9

coscos

1 LL ff

dd

d

B0+– has 3 polarization states with different CP eigenvalues

021.0029.0014.0978.0

Lf

030.0 9410 03400400

..L .f

BelleFortunately, longitudinal polarization dominates, therefore, pure CP-even state

BaBar (PRL 95, 041805, (2005)):

Belle (hep-ex/0601024): cos

# of

Eve

nts

No significant 00 signal small penguin contribution

BaBar (PRL, 98, 111801 (2007)):

000=(1.07 ±0.33 ±0.19) 10-6)( BB

ADS methodD. Atwood, I. Dunietz and A. Soni, PRL 78, 3357 (1997); PRD 63, 036005 (2001)

Enhancement of СР-violation due to use of Cabibbo-suppressed D decays

B–D0K– - color allowed D0K+π– - doubly Cabibbo-suppressedB–D0K– - color suppressed D0K+π– - Cabibbo-allowed

Interfering amplitudesare comparable

coscos2)(

)(3

22DBDBDK rrrr

KDBBr

KDBBr

fav

suppR

ADS method

Belle results (357 fb-1)

Suppressed channel not visible yet:

34.89.7 10)0.10.0(

DKR

Using rD=0.060±0.003, for maximum mixing (φ3=0, δ=180°):rB<0.18 (90% CL)

3119 1013 DKR

3106][*

1020

KDD

R

31813][* 1011 KDD R

rB<0.23 (90% CL) for BDKrB<0.16 for BD*K

BaBar results (211 fb-1)

GLW method

M. Gronau and D. London, PLB 253, 483 (1991); M. Gronau and D. Wyler, PLB 265, 172 (1991)СР eigenstate of D-meson is used (DCP).

CP-even : D1 K+K–, π+ π –

CP-odd : D2 KS π0, KS ω, KS φ, KSη…

)()(

)()(

2,12,1

2,12,121

KDBBrKDBBr

KDBBrKDBBr,Α

32

3

coscos21

sinsin2

rr

r

for D1

for D2

32

002,12,1

2,1 coscos21)(/)(

)(/)(

rrDBBrKDBBr

DBBrKDBBrR

4 equations (3 independent: ), 3 unknowns ),,( 3r

Additional constraint:

СР-asymmetry:

A1,2 of different signs

2211 RARA

GLW averages

DC

P K

*D

* CP K

DC

P K

GLW method

0 0

0 0

2

0 0

0 0

2

( ) ( )

[ ( ) ( )]/ 2

1 2 cos

( ) ( )

( ) ( )

2 sin

1 2

cos

sin

coscos

CP CPCP

CP CPCP

CP CP

BR B D K BR B D KR

BR B D K BR B D K

r r

BR B D K BR B D KA

BR B D K BR B D K

r

r r

0.81 0.10 0.05

1.07 0.10 0.04

0.19 0.12 0.02

0.35 0.09 0.05

CP

CP

CP

CP

R

R

A

A

First evidence (3.4) for direct CP violation in B → DK decays

0CPB D K 0

CPB D K

85 14 signal events177 20

signal events

Dalitz analysis (Belle)

Belle result (357 fb-1) PRD73,112009

331±17 events 81±8 events 54±8 events

BDK

BD*KBDK*

B- B+ B- B+ B- B+

Dalitz analysis: D0 Ksπ+π– decay

Statistical sensitivity of the method depends on the propertiesof the 3-body decay involved

(For |M|2=Const there is no sensitivity to the phase θ)Large variations of D0 decay strong phase are essential

Use the model-dependent fit to experimental data from flavor-tagged D* D0π sample ( >2x105 events)

Model is described by the set of two-body amplitudes + flat nonresonant term

As a result, model uncertainty in the

γ/φ3 measurement

ρ-ω interference

Doubly Cabibbosuppressed K*

)(GeV22sK

M

)(GeV22sK

M

)(GeV22sK

M )(GeV22sK

M

)(GeV 22

M )(GeV 22

M

M

(

GeV

2 )

Ksπ

–

2

D0 Ksπ+π– decay model

Dalitz analysis (Belle)Fit parameters are x= r cos(φ3+δ) and y= r sin(φ3+δ)(better behaved statistically than ) are obtained from frequentist statistical treatment based on PDFs from toy MC simulation.

3,, r3,, r

BDK

BD*K BDK*

easier to combine results

x–= 0.025 -0.080

y–= 0.170 -0.117

x+= –0.135 -0.070

y+= –0.085 -0.086

x-= –0.128 -0.146

y-= –0.339 -0.158

x+= 0.032 -0.116

y+= 0.008 -0.136

x–= –0.784 -0.295

y–= –0.281 -0.335

x+= –0.105 -0.167

y+= –0.004 -0.156

+0.072

+0.093

+0.069

+0.090

+0.167

+0.172

+0.120

+0.137

+0.249

+0.440

+0.177

+0.164

sin(2φ1+φ3) from B0 D*π decay

)sin()cos(18

1)( /||(*)0 tmStmCeDBP Bt

B

)sin()cos(18

1)( /||(*)0 tmStmCeDBP Bt

B

),2sin()1(1

2312

L

R

RS

11

12

2

R

RC

where

Use B flavor tag, measure time-dependent decay rates:

Decay : (*)00 )( DBB

+

B D-+

Btag B0

Btag B0

CP violation

02.0R

sin(2φ1+φ3) result (211 fb-1) hep-ex/0601018result (357 fb-1) hep-ex/0604013

*

0

DB

BB

rec

tag

*

0

DB

BB

rec

tag

*

0

DB

BB

rec

tag

*

0

DB

BB

rec

tag

*

0

DB

BB

rec

tag

*

0

DB

BB

rec

tag

*

0

DB

BB

rec

tag

*

0

DB

BB

rec

tag

Lepton tags, D* final state

New

New

sin(2φ1+φ3)

90% CL

68% CL

|sin(2+)| > 0.64 @ 68 % C.L.|sin(2+)| > 0.42 @ 90 % C.L.

Combine partial and fully reco results and use the R parameters from SU(3) symmetry

Frequentistconfidence level

|2+| = 90o 43o

30% theoretical error on rf

DBB )( 00 *00 )( DBB

|sin(2φ1+φ3)|> 0.44 |sin(2φ1+φ3)|> 0.52

R~0.02 from

New

New

Br(BDS(*)π)/Br(BD(*)π)

D0-D0 mixing

D0-D0 mixing

D0-D0 mixing