Md. Naimuddin (on behalf of CDF and D0 collaboration) University of Delhi Recontres de Moriond

Bs Mixing at the Tevatron

Md. Naimuddin(on behalf of CDF and D0 collaboration)

University of DelhiRecontres de Moriond

19th March, 2006

03/19/2006 Md. Naimuddin 2

– Introduction to Mixing

– Why study Mixing

– Fermilab Tevatron

– Data Taking

– CDF and D0 Detectors

– Analysis strategy

– Bs mixing at CDF

– Bs mixing at D0

– Results and Conclusions

OUTLINE


Mixing: The transition of neutral particle into it's anti-particle, and vice-versa

Introduction

First observed in the K meson system.

In the B meson system, first observed in an admixture of B0 and Bs

0 by UA1 and then in B0 mesons by ARGUS in 1987.

These oscillations are possible because of the flavor-changing term of the Standard Model Lagrangian,

..,,2

chW

b

s

d

Vtcug

L

L

CKML


1)1(2

11)1(

)(2

11

23

2242

32

AiA

AiA

iA

VVV

VVV

VVV

tbtstd

cbcscd

ubusud

Where VCKM is the CKM matrix;

Wolfenstein parametrisation- expansion in ~ 0.22

0*** tbtdcbcdubud VVVVVV

CP violation in the Standard Model originates due to a complex phase in the CKM matrix (quark mixing).

iubub eVV || i

tdtd eVV ||1

*

*

*

*

cbcd

tbtd

cbcd

ubud

VV

VV

VV

VV

Introduction

(0,0) (0,1)

(,)

*ubud

*tbtd

VV

VVarg

*cbcd

*tbtd

VV

VV

*tbtd

*cbcd

VV

VVarg

*cbcd

*ubud

VV

VVarg

*cbcd

*ubud

VV

VV

(0,0) (0,1)

(,)

*ubud

*tbtd

VV

VVarg

*cbcd

*tbtd

VV

VV

*tbtd

*cbcd

VV

VVarg

*cbcd

*ubud

VV

VVarg

*cbcd

*ubud

VV

VV


Mixing in B mesons

Light and Heavy B mesons mass eigenstates differ from flavor eigenstates:

B|qB|pB| L

B|qB|pB| H

122 qp

2LH MM

m

2

LH

LH MMm HL

Time evolution of

B0 and states

mtte

BBPt

cos2

cosh2

)(

mtte

BBPt

cos2

cosh2

)(

In case 0,

0B

mttP temu cos1

1, 2


2*2

22222

6

/tbtddBdBdB

WttFd VVBfm

π

)mηF(mmG=Δm

standard model Expectation:Δms= 14 - 28ps-1 Δmd=

0.5ps-1

measurement of Δms/Δmd

Vts/Vtd

constrain unitary triangle

f2BdBBd = (228+30+10 MeV)2

|Vtd| from md limited by ~15%

consider ratio

2

22

2

2

2

2

td

ts

B

B

td

ts

BB

BB

B

B

d

s

V

Vξ

m

m=

V

V

Bf

Bf

m

m=

Δm

Δm

d

s

dd

ss

d

S

Determine |Vts|/|Vtd| ~ 5% precision

02.021.12 035.0014.0

Why Bs mixing is important

K mixing direct & indirect CPVBd mixing heavy top mass mixing neutrino mass 0

Bs mixing ????


Fermilab Tevatron

Highest Luminosity achieved: 1.7x1032 cm/s2

Expected: 2 – 3 x 1032 cm/s2


Data TakingExcellent performance by the Tevatron and we are doubling our data set every year.

Data taking efficiency is ~ 85%.

We already have collected about 10 times of run 1 data.

0

200

400

600

800

1000

Luminosity used in the currentanalysis

CDF

CDF

D0

Pb-1


CDF Detector

• Solenoid 1.4T• Silicon Tracker SVX

• up to ||<2.0• SVX fast r-

readout for trigger• Drift Chamber

• 96 layers in ||<1• particle ID with

dE/dx• r- readout for

trigger• Time of Flight

• →particle ID


D0 Detector

• 2T solenoid• Fiber Tracker

• 8 double layers• Silicon Detector

• up to ||~3• forward Muon +

Central Muon detectors• excellent coverage

||<2• Robust Muon

triggers.


We need to:

Reconstruct the Bs signal. Know the flavor of the meson at it’s production time (Flavor tagging) and get D2

Proper decay length resolution

Analysis strategy

(t)N(t)N

(t)N(t)N(t)A

mixunmix

mixunmixmix

N

NNε WR

Measuring Bs oscillations is much more difficult than Bd oscillations because of the fast mixing frequency.In order to measure,

N

NND WR

2

22

2

tsm

eysensitivit DNBSS

Reconst-ruction

Flavor tagging

Propertime

tmΔcos*D s


2

22

2

tsm

eysensitivit DNBSS

Semileptonic modes collected by the two track trigger

Reconst-ruction

Flavor tagging

Proper time


More than 2300 Bs signal candidates in 765 pb-1

Reconst-ruction

Flavor tagging

Proper time

2

22

2

tsm

eysensitivit DNBSS

)(0 KDB BsDs-()+


Currently using only semileptonic decay of the Bs

Bs Ds X (Ds ) ( K+K-)

Reconst-ruction

Flavor tagging

Proper time

2

22

2

tsm

eysensitivit DNBSS

26,710±560

7,422±281

5,601±102

1,519±96

Untagged Tagged


Soft Lepton Tagging ( or e):• Charge of the leptons provides the flavor of b.Jet Charge Tagging:• Sign of the weighted average charge of opposite B jet provides the flavor of b.Same Side(Kaon) Tagging (New at CDF)• B meson is likely to be accompanied by a close K/ • particle Id helps

b-hadron

Trigger leptonFragmentationpion

B

PV D

neutrino

LxySoft lepton

Jet charge

0 b<jetQ

0 b>jetQ

Opposite side Same side

e or

Reconst-ruction

Flavor tagging

Proper time

2

22

2

tsm

eysensitivit DNBSS


Reconst-ruction

Flavor tagging

Proper time

2

22

2

tsm

eysensitivit DNBSS

Opposite side Tagging: )%014.0020.055.1(2 D

Same side Kaon Tagging:

%3.28 2.32.4

D %0.4 9.0

2.12

D


Reconst-ruction

Flavor tagging

Proper time )%21.048.2(

06.0

08.02

D

2

22

2

tsm

eysensitivit DNBSS

Using only Opposite side e, , secondary vertex or jet charge for tagging

Tagger tuned using Bd mixing measurement

md = 0.506±0.020±0.014 ps-1


Easier to calculate for Hadronic decays

Proper decay time given as :

TB

xy P

mLct Reconst-

ruction

Flavor tagging

Proper time

2

22

2

tsm

eysensitivit DNBSS

KP

mLct

TB

xy *

TB

TlD

P

PK s

Distance from PV to SV in transverse plane

Boost meson back to its rest frame

Due to missing neutrino, it becomes little tricky in semileptonic sampleMissing momentum increased ct error ( ct)

Proper decay time given as:

where

from MC


2

22

2

tsm

eysensitivit DNBSS

Reconst-ruction

Flavor tagging

Proper time

Best estimate of errors used in track fitting, any additional scaling determined from data

Scale factor = 1.17 +/- 0.02

(largest systematic)

Pull of J/ vertex from the primary vertex for prompt J/ s


Fitting procedure

Amplitude Fitting approach: Introduce amplitude

Fit for A at different ms

Traditionally used for Bs mixing search Easy to combine experiments

tmADtP s cos1~)(

Likelihood approach:

Search for min, which minimizes the –ln(L),

where min is the estimator of the true frequency.

Difficult to combine the results from different experiments.

)(ln)(ln)( minmin LLL


Amplitude fit method

• Fit to data - free parameter

• Obtain as a function of

• Measurement of gives and otherwise

• At 95% CL• sensitivity • excluded

• Systematics is given by

A

A

AsysA AA

1

t)]m(A±[Γe=p sΓt cos1

2

1

sm

smA 0

A

1 1.645 =A1 1.645 <+A A

A 1


Bs mixing limit

Limit: ms > 8.6 ps-1 @ 95% confidence

Sensitivity: 13.0 ps-1 @ 95% confidence


Bs mixing limit

Dominant sources of systematic uncertainty:Scale factor, K factors and Sample composition


Expected Limit: 14.1 ps-1 @ 95% confidence

A clear deviation of 2.5 from 0 can be seen at

ms~19 ps-1


Likelihood scan

ms < 21 ps-1 @ 90% CLMost probable value of ms = 19 ps-1

Resolution K-factor variation by 2% BR (BsDsX) ccbar contamination BR (BsDsDs)

Syetematics


Probability for this measurement to lie

in the range

16 < ms < 22ps-1

Given a true value:

ms > 22 ps-1: 5.0%

ms = 19 ps-1: 15%

Simulate ms = infinity by randomizing flavor

tagging variable

Consistent results from parameterized MC studies

Significance Test


CKM fit without new D0 result


Disclaimer:

Produced only yesterday by CKM group so I don’t have much information about the inputs.

CKM fit with new D0 result


CONCLUSIONS Excellent performance of the Tevatron.

Development of Same Side Kaon Tagger which will ease the total tagging power by about 2.5 (CDF)

D0 has excellent opposite side tagger performance.

CDF preliminary:

D0 Final (hep-ex/0603029, submitted to PRL):

We are on the edge of Bs mixing measurement and entering into the precision era.


Sensitivity: 13.0 ps-1 @ 95% confidence

ms < 21 ps-1 @ 90% CLMost probable value of ms = 19 ps-1

Likelihood scan


Expected Limit: 14.1 ps-1 @ 95% confidence

Amplitude Scan


FUTURE IMPROVEMENTS

CDF:

Improved selection in hadronic modes using Neural Networks

Use semileptonic events from other triggers.

Improve vertex resolution.

Use same side Kaon tagger.

Several major improvements are expected in the future:

More data: On track for 4-8 fb-1 of data by 2009.

D0:

More decay modes (DsK*K, 3, Ks, Bse Ds X )

Use of same side tagging.

Use of hadronic modes.

Additional Layer of silicon.

Proposal to increase the bandwidth.


• Back-up slides


Backup-1

Time evolution follows from a simple perturbative solution to the schrondinger’s equation

(t)B|

B(t)|iΓMiΓM

iΓMiΓM

(t)B|

B(t)|

22222121

12121111

dt

di

Eigenvalues are, LHLHLH

im ,,, 2

B|B|

q/p1q/p1B|B|

2

qpB| L

Bεq/p1

q/p1

Measure of the amount by which |BL> and |BH> differ from CP eigenstates

In case 0,


In the limit of no CP violation in mixing (q/p =1), unmixed and mixed decay probabilities become

mttP temu

cos11

,

These oscillations can be used to measure the fundamental parameters of the standard Model and have other far reaching effects such as breaking the matter anti-matter symmetry of the Universe.Constraining the CKM matrix redundantly using different measurements of the angles/sides is a sensitive probe of New Physics

Backup-2


Backup-3


Backup-4


Backup-5


Backup-6


Backup-7

WA+CDF+D0


Comparison with other experiments at ms=15 ps-1

Backup-8

Documents

Md. Naimuddin (on behalf of CDF and D0 collaboration) University of Delhi Recontres de Moriond