Semileptonic Decays from Belle

1
Semileptonic Decays from Belle
Youngjoon Kwon Yonsei Univ. / Belle
2
Contents

• Introduction
• Analyses for Vub
• Inclusive B ? Xu l n with full-recon.
• Inclusive B ? Xu l n with end-point
• Exclusive B ? Xu l n with semileptonic tag.
  (prelim.)
• Analyses towards precise Vcb
• Inclusive B ? X e n with full-recon.
• B ? X l n hadronic mass moments (prelim.)
• B ? X l n lepton energy moments (prelim.)

PRL submitted, hep-ex/0505088
PLB accepted, hep-ex/0504046
PLB 614, 27 (2005)
3
Motivation

• Fundamental free parameters
• Majority of free parameters in the SM are related
  to quark (and lepton) flavors masses and
  mixings
• Quark flavor mixing through CKM matrix
• Unitarity triangle
• Direct test of CP violation mechanism in SM
• Vub still a weak link in CKM test
• Vcb as fundamental as Vub
• Super-high precision, experimentally
• A good testing ground of strong int. theory

4
Issues in Vub measurement

• Use B ? Xu l n (semileptonic)
• Cleaner, better understood than B ? hadronic
• More accessible than B ? leptonic
• Inclusive and Exclusive
• Experimentally challenging
• Tight selection criteria results in
• Limited phase space
• ? Extra theoretical uncertainties (from
  extrapolation)

5
Variables to suppress b ? c l n
(10 phase sp.)
(80)
(30)
6
Inclusive B ? Xu l n with End-point
7
Main features

• Measure DB(B ? Xu e n, pe gt pcut)
• Kinematic selection
• pe gt pcut
• pcut 1.9, 2.0, 2.1, 2.2, 2.3 and 2.4 GeV/c
• No requirement other than on the
• signal electron
• high statistics
• Large background from B ? Xc e n
• bkg. sys. uncertainty

OPE breaks down ? need shape ftn.!
8
b-quark Shape Function

• What is it?
• b-quark Fermi motion inside the B-meson
• How to get it?
• Universal at leading order for all b ? light
  quark transitions
• Use E(g) in B ? Xs g

9
Three models for SF
Neubert Paz, hep-ph/0504071
10
(No Transcript)
11
(No Transcript)
12
Event Selection

• Hadronic events
• Electrons in the barrel region
• eff. 94 fake rate 0.13
• J/y, y(2S), g, p0 veto
• Continuum suppression
• Fox-Wolfram moment
• Fisher disc. for energy-flow variables
• cos qth lt 0.75
• as q2 independent as possible
• Background estimated by
• b ? c MC and Off-resonance Data
• low-p region for check
• 1.5 lt pe lt pcut (GeV/c)

13
E(e) spectrum for B ? Xu e n
14
DB partial BF

• Two methods for DB ? Vub
• Standard method (DFN, JHEP 9906, 017)
• New method (BLNP, NP B699 (2004))

15
Vub from DB
DFN
BLNP
16
Inclusive B ? Xu l n with Full-Recon.
17
Full Reconstruction
4.2?105 evts./253 fb-1 effic. 0.33 purity 62
2.5?105 evts./253 fb-1 effic. 0.21 purity 59
18
Signal selection

• Lepton Selection
• p gt 1 GeV/c
• leptons in the barrel region
• J/y, g, p0 veto
• Correct charge for B candidate
• No other lepton

Nsl (9.15 ? 0.05) x 104

• For b ? u
• Total charge 0
• No reconstructed KS or K?

19
Kinematic selections
(New)
20
(1) MX lt 1.7 GeV

• Each bin content is calculated from Mbc fit
• 2-component fit to extract N(b?u)

21
(No Transcript)
22
(2) MX lt 1.7 GeV q2 gt 8 GeV2
23
(3) P lt 0.66 GeV

• First measurement using P

24
Signal Yield DB
25
Vub from DB

• New method
• Vub directly from DB
• R by Bosch, Lange, Neubert, Paz, NPB699 (2004)
• with an improved treatment of shape-function
  effects
• weak annihilation effects estimated

26
Exclusive B ? Xu l n with S.L. tag
27
Experimental Strategy
Unique opportunity at ee- B-factory!
B?p l n / r l n (140fb-1) ICHEP04
preliminary hep-ex/0408145
S/N
Full recon.
D() l n tag
B?w l n (78fb-1) PRL 93, 131803 (2004)
advanced n recon.
p l n / r l n (preliminary)
Lum.
trad. n recon.
Eff.
How well can we measure the q2 dist. (for p l n) ?
28
Analysis Method
Tag side reconstruction
4 decay modes
7 decay modes
Signal side reconstruction
or
Kinematics of double semileptonic decay
Back-to-back correlation of the two Bs
constrains their direction to the intersection of
the 2 cones.
To have intersection, need
2-fold ambiguity ? q2 calculated neglecting the B
motion in U(4S) q2 resolution 0.75 GeV2
29
Calibration with Bsig?Dln Decays

• Validity of the method for double semileptonic
  decay
• detection has been tested with

M(Kp-p-) on the signal side
xB2 distribution
Nobs147?12
( Nexpected 165)
Dots Data Histo MC
M(Kpp) GeV/c2
xB2

• The ratio Nobs/Nexpected 0.89 ? 0.08 is used to
  correct the MC efficiency for p l n and r l n
  detection.

The method works !
30
Signal Extraction

• q2 distribution is extracted by fitting the
  (mX,x2) distrib. for three q2 intervals.

mX (GeV/c2)
q28-16GeV2
q2gt16GeV2
q2lt8GeV2

• Fitting components p ln, r ln, other Xu ln, and
  BB
• PDFs are based on MC.
• Constraint for extracted Br
• Br(p ln) Br(r ln) Br(other Xu ln) Br(Xu ln)

82?13
65?20
31
Results
32
Vub from exclusive B ? Xu l n
140 fb-1
33
Analyses towards precise Vcb

• Inclusive B ? X e n (0th moment)
• B ? X l n hadronic mass moments (prelim.)
• B ? X l n lepton energy moments (prelim.)

PLB 614, 27 (2005)
34
PLB 614, 27 (2005)
35
B ? X l n moments

• Leptonic and hadronic moments
• Needed for global fit for precise Vcb
• Common features
• Preliminary with 140 fb-1
• Full-recon. in the tag side
• Boost from Btag
• QED correction by PHOTOS
• Mixing correction
• Unfolded by SVD method

36
Leptonic moments
37
Hadronic moments
38
Hadronic moments
39
Belle Vub Summary