KARLSRUHE INSTITUTE OF TECHNOLOGY (KIT)

BELLE

Search for the decay B0 → τ+τ− at BelleGK-Workshop 2015, Freudenstadt

Michael Ziegler | September 28, 2015

KIT – University of the State of Baden-Wuerttemberg and

National Laboratory of the Helmholtz Association

www.kit.edu

How to detect New Physics?

High Energy Frontier

New resonances at high masses

. . .

or

High Intensity Frontier

Mixing in neutral meson systems,e.g. K0 − K0, B0 − B0

Rare B decays

. . .

SuperKEKB

KEKB

LHC

Tevatron

(gNP/g)2

MNP[TeV

]

10-1

1

10

10 2

1 10 102

Michael Ziegler – B0 → τ+τ− 2/16

Rare B Decays

Small branching ratios in the SM⇒ effects from NP could be seen

B0 → `+`−

b

d

W

u, c, t

u, c, t

Z0

l+

l−

b

d

W+

W−

ν

l+

l−

u, c, t

B(B0 → e+e−)SM = (2.48± 0.21)× 10−15

B(B0 → µ+µ−)SM = (1.06± 0.09)× 10−10

B(B0 → τ+τ−)SM = (2.22± 0.19)× 10−8

Values from: Bobeth, C. et al., Bs,d → `+`− in the Standard Model withReduced Theoretical Uncertainty. Phys. Rev. Lett. 112, 101801 (2014)

Michael Ziegler – B0 → τ+τ− 3/16

Search for B0 → τ+τ− at Belle

Experimental Setup

KEKB: asymmetric energye+e− collider√s ≈M(Υ(4S))

Final data sample:772× 106 BB pairs

Michael Ziegler – B0 → τ+τ− 4/16

Reconstruction

Challenge of B0 → τ+τ−

τ leptons decay in the detector→ at least 2 neutrinos in the final state

But at Belle: only two B mesons in an event

Initial state well known

Reconstruct one of them in a hadronic final state - called BTag

In the rest of the event, search for the signal side - called BSig

Michael Ziegler – B0 → τ+τ− 5/16

Tag Side - The hadronic full reconstruction

BTag reconstructed in a large number of hadronic decay channels

Multivariate selection at each stage

Final probability for reconstructed BTag candidate being a B meson

Remaining tracks and energy deposition in the detector must comefrom second B meson

Michael Ziegler – B0 → τ+τ− 6/16

Results of the Full Reconstruction

Beam constraint mass

M tagbc =

√E2

beam − ~p2tag

]2[GeV / cbcM5.24 5.25 5.26 5.27 5.28

2Ev

ents

per

0.5

MeV

/ c

0

50

100

150

200

250

300

350

310×

classicalnewBackgroundData

Efficiency of Full reconstruction: O(10−3)

Michael Ziegler – B0 → τ+τ− 7/16

Signal Side

Reconstruct τ in different channels . . .

Decay mode B in %

τ− → e−νeντ 17, 83τ− → µ−νµντ 17, 41τ− → π−ντ 10, 83

. . . and combine two oppositely charged τ candidates to BSig

Selection

No charged track and π0 in rest of event

More cuts on kinematic variables like missing mass, . . .

AfterwardsDivide sample according to the final state

Michael Ziegler – B0 → τ+τ− 8/16

What do we expect so far?Final state: one τ decays to e the other to π ( + neutrinos)

0.0 0.2 0.4 0.6 0.8 1.0 1.2EEcl in GeV

0

100

200

300

400

500

600

700

800

Even

ts p

er 5

0 M

eV

Belle MC preliminary

expected backgroundsignal (BSM×105 )

What is EEcl?

Remaing energy in the calorimeter that is not used in the reconstruc-tion

Michael Ziegler – B0 → τ+τ− 9/16

Background Composition

0.0 0.2 0.4 0.6 0.8 1.0 1.2EEcl in GeV

0

100

200

300

400

500

600

700

800

Even

ts p

er 5

0 M

eV

Belle MC preliminary

e+ e− →uu,dd,sse+ e− →ccmisrec. B+

misrec. B0

Background from light quarks suppressed

Main background: misreconstucted B0. e.g.:

B0 → D−(→ KLπ−)e+νe

Michael Ziegler – B0 → τ+τ− 10/16

Background SuppressionNeuroBayes networks are used

Input variables

Missing mass

Invariant mass of thecharged tracks

Mbc of BTag

candidate

. . .

FOM =ε

σ/2 +√

NB

0.0 0.2 0.4 0.6 0.8 1.0NeuroBayes discriminator

0

2

4

6

8

10

12

14

16

18

Nor

med

eve

nts

per 0

.01

Belle MC preliminary

backgroundsignal

0.0 0.2 0.4 0.6 0.8 1.0Cut on nout

0.000

0.001

0.002

0.003

0.004

0.005

0.006

0.007Fi

gure

of M

erit

Belle MC preliminary

Michael Ziegler – B0 → τ+τ− 11/16

EEcl after the cut

0.0 0.2 0.4 0.6 0.8 1.0 1.2EEcl in GeV

0

20

40

60

80

100

120

140

160

Even

ts p

er 5

0 M

eV

Belle MC preliminary

e+ e− →uu,dd,sse+ e− →ccmisrec. B+

misrec. B0

signal (BSM×105 )

Signal extraction

Counting experiment in the signal window EEcl < 0.2 GeV

Michael Ziegler – B0 → τ+τ− 12/16

Expected Sensitivity

A scan over different branching ratios for B0 → τ+τ− is performed toestimate the expected sensitivity

0.2 0.4 0.6 0.8 1.0B(B0 →τ+ τ− ) 1e 3

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.001−p

0

Belle MC preliminary

95% CL

90% CL

B(B0 → τ+τ−)expected < 7.5× 10−4

Michael Ziegler – B0 → τ+τ− 13/16

Conclusion

NP can be seen in rare B decays

Knowledge of the initial state makes reconstruction of decays withmany neutrinos in the final state possible

Expected limit (@ 95% CL):

B(B0 → τ+τ−)expected < 7.5× 10−4

Just the beginning of the search for B0 → τ+τ−

Michael Ziegler – B0 → τ+τ− 14/16

Results for B0 → µ+µ−

Year1985 1990 1995 2000 2005 2010 2015

Lim

it (9

0% C

L) o

r B

F m

easu

rem

ent

10−10

9−10

8−10

7−10

6−10

5−10

4−10

−µ+µ → 0sSM: B

−µ+µ → 0SM: BD0L3CDFUA1ARGUSCLEO

CMS+LHCbATLASCMSLHCbBaBarBelle

2012 2013 2014

10−10

9−10

8−10

Source: LHCb & CMS Collaborations, Nature, 522 (2015), p. 68-72

Michael Ziegler – B0 → τ+τ− 15/16

2005 2010 2015 2020 2025 2030 2035Year

10-9

10-8

10-7

10-6

10-5

10-4

10-3

10-2

Lim

it on

B0→τ

+τ−

SM: B0 →τ+ τ−

BaBarBelle (expected)

Michael Ziegler – B0 → τ+τ− 16/16