Peter Kri

1
Rare decays at B factories

• Peter Križan
• University of Ljubljana and J. Stefan Institute

2
Fundamental Questions in Flavor Physics
Are there new sources of CP violation?
Experiments b?s CPV, compare CPV angles from
tree and loops Are there new operators with
quarks enhanced by New Physics ? Experiments
AFB(B?Kl l), B?K ? rates and asymmetries Are
there right-handed currents ? Experiments b?s ?
CPV, B?V V triple-product asymmetries Are there
new flavor changing neutral currents
? Experiments b?s ??bar, D-Dbar mixingCPVrare,
t???
Data sample of 50ab-1_at_Y(4S) needed to address
these questions
3
Contents

• FCNC b?s decays
• b?sg inclusive rate, CP violation
• b?d penguins B ? rg, wg decays
• Measurement of Afb vs q2 in B ? K l l- decays
• Decays with gt1 neutrino
• Purely leptonic decays B- ? t- nt , B0 ? t t-
• B- ? K- n n
• Semileptonic decay B- ? D() t- nt
• ... Only a limited
  selection of topics.

4
Why FCNC decays?
Flavour changing neutral current (FCNC) processes
(like b?s, b?d) are fobidden at the tree level in
the Standard Model. Proceed only at low rate via
higher-order loop diagrams. Ideal place to
search for new physics.
5
Radiative Decays

• Inclusive Br(b?sg) C7, SF for Vub
• B?Kg isospin asymmetry (D-) sign of C7
• Mixing induced CPV
• Direct CPV in B?Xsg
• B?Xd g
• 
  
• red discussed in this talk

6
Inclusive Br(b?sg)

• Motivation measure C7 and shape function
  parameters (from Eg spectrum for Vub
  extraction)
• Inclusive and semi-inclusive ?

38 modes, 55 rate
7
Inclusive Br(b?sg)
BF(B ? Xs g, Eggt1.6 GeV )(3.550.24 0.09 0.03)
x 10-4
-0.10

• consistent with SM expectations (3.570.30) x
  10-4
• However theory error is expected to get reduced
  by a factor of 2 (NNLL calculations are under
  way)
• ?interesting for charged Higgs mass limit
• ?need a better measurement of inclusive rate
• Also important to fix the value of Wilson coeff.
  C7 and to determine the spectrum shape.
• At 5 ab-1 Egmin down to 1.5 GeV and stat. error
  5

8
B?Xsg CP Asymmetry

• Sensitive to NP right handed currents
• Theoretically clean.
• Standard Model Zero.
• g is polarized, and the final state is almost
  flavor specific.
• Helicity flip of g suppressed by ms/mb?S 0.06
  sin2f1
• Time dependent CPV requires vertex reconstruction
  with KS?pp-

g
Vertex recon. eff. at Belle 51 (SVD2), 40 (SVD1)
Possible at ee- B-factory
9
B0?KSp0g time dependent CPV
Atwood, Gershon, Hazumi, Soni, PRD71, 076003
(2005)

• M(KSp0) lt 1.8GeV/c2
• NP effect is independent of the resonance
  structure.
• Example Belle 386MBB
• Two M(KSp0) regions(MR10.8-1.0GeV/c2/MR2
  lt1.8GeV/c2)
• 70-11 (45-11) events in MR1(2).

S 0.08 0.41 0.10 A 0.120.270.10
Good tag (0.5ltrlt1.0)
10
B0?KSp0g time dependent CPV

• Results
• Belle hep-ex/0507059
• S(B?Kg, K?KSp0)-0.01 0.52 0.11
• S(B? KSp0g)0.08 0.41 0.10
• BaBar PRD 71 (2005) 0501103
• S(B?Kg, K?KSp0)-0.210.40 0.05
• Prospects
• Add more modes B?KSfg (with angular analyisis),
  higher K resonances, B?KShg (recent observation
  by BaBar),...

50ab-1
5ab-1
Present Belle (stat./syst.)
0.04
0.14
0.41 / 0.10
Acpmix(B?Kg, K?KSp0)
0.005
0.011
0.051 / 0.038
Acpdir(B?Xsg)
11
Acp(B?Xsg) vs SUSY models
50ab-1
5ab-1
Mixing CPV
Direct CPV
mSUGRA tanb30
U(2) tanb30
U(2) tanb30
mSUGRA tanb30
Acpdir
Acpmix
SU(5)nR tanb30 non-degenerate
SU(5)nR tanb30 degenerate
SU(5)nR tanb30 non-degenerate
SU(5)nR tanb30 degenerate
T. Goto, Y.Okada, Y.Shimizu,T.Shindou,
M.Tanaka hep-ph/0306093, also in SuperKEKB LoI
12
b?d penguins

• Supressed by Vtd/Vts 2 vs b ? sg

• Interesting
• Measurement of Vtd/Vts
• Addresses the same physics issue as Bs- Bs mixing
  (from a different perspective box vs loop)
• CP violation could be sizeable in SM (order 10)
• New physics could be quite different from b ? sg

Difference in dynamics
Form factor ratio
13
Belle First observation of B ? rg, wg
Ref PRL 96, 221601 (2006)
Reconstructing rare B meson decays at Y(4s) use
two variables, energy diference DE and beam
constrained mass Mbc
_
14
Vtd/Vts from B ? rg, wg
The measured branching fraction BF(B ? r/wg)
1.32 0.34 (exp.) 0.10 (theo.) Translates to
Vtd/Vts 0.199 0.026 (exp.) 0.018
(theo.) which is compatible with SM constraints
based on fits of other CKM parameters.
Ref PRL 96, 221601 (2006)
-0.31
-0.09
-0.025
0.015
Together with the evidence of B?K0 K decays
modes, this demonstrates the existence of a new
quark level transition b?d
15
Implications of Belles observation of b?d ?

• Comparison with the recent observation of Bs
  mixing at Tevatron
• green Belle measurement of Vtd/Vts in b?d ?
• yellow CDF measurement of Vtd/Vts from Bs
  mixing

The width of the Bs mixing contour is limited by
theory while B?d ? needs much more data.
16
b?d ? future prospects

• With 1-2 orders of magnitude more statistics (5
  ab-1, 50 ab-1)
• Direct CP violation and time-dependent CPV with
  B??0 ? and B ? ??
• Measurements of inclusive b ? d ?

17
B ? K l l-

• b ? s ll- was first measured in B ? K ll- by
  Belle (2001).
• Important for further searches for the physics
  beyond SM
• Ci Wilson coefficients

18
Particularly sensitive forward-backward
asymmetry in K ll
Forward event
Backward event
? and Z contributions in B?K l l interfere
and give rise to forward-backward asymmetries
c.f. ee- ?? ?-
19
Sample used for AFB(B?Kll)(q2)
Ref PRL 96, 251801 (2006)
Treat q2, cos(?) dependence of bkgs.
Sample for B?K l l 11313 events
Unbinned fit to the variables q2 (di-lepton
invariant mass) and cos(?) for the B?K l l
data. Fit parameters A9/A7 and A10/A7 (Ai
leading term in Ci)
20
Control sample B?Kll
B?K l l control sample 9612 events
Consistent with flat
Integrated asymmetry
21
Constraints on Wilson coefficients from AFB(B?K
l l)(q2)
Projections of the full fit to q2, cos(?)
Integrated FB asymmetry
J/y y
Observed integrated AFB rules out some radical
New Physics Models with incorrect
signs/magnitudes of C9 and C10 (red and pink
curves)
22
Results of the unbinned fit to q2 and cos(?)
distributions for ratios of Wilson coefficients.
SM
A10/A7
Best fit
A9/A7
A7 constrained from b?s ? to be close to SM
Ref PRL 96, 251801 (2006)
at 95 C.L.
23
AFB(B?K l l)(q2), BaBar
BaBar 229 M BB PRD 73 (2006)
 092001 
Integrated FB asymmetry AFB gt0.55 (_at_ 95 CL)
First bin excludes SM (blue) at 2s level?
24
AFB(B?K l l-)q2 at Super B Factory
q02
Precision with 5ab-1 dC9 11 dC10 14 d
q02/q02 11
4AFB zero-crossing q02 will be determined
with 5 error with 50ab-1
25
Purely leptonic decay B?t n

• Proceed via W annihilation in the SM.
• Branching fraction
• Provide information of fBVub
• Vub from B?Xu l n fB cf) Lattice
• Br(B?tn)/Dmd Vub / Vtd
• Expected branching fraction
• Vub (4.39 0.33)10-3 (HFAG)
• fB (216 22) MeV (lattice)

BF(B ? t nt) (1.59 0.40)10-4
26
Charged Higgs contribution to B?t n

e0 SUSY corrections to b Yukawa coupling
Br(SM) 1.59 x 10-4
27
Full Reconstruction Method

• Fully reconstruct one of the Bs to
• Tag B flavor/charge
• Determine B momentum
• Exclude decay products of one B from further
  analysis

Decays of interest B?Xu l n, B?K n n B?Dtn,
tn
B
e- (8GeV)
e(3.5GeV)
?(4S)
p
B
full reconstruction B?Dp etc. (0.10.3)
Offline B meson beam!
Powerful tool for B decays with neutrinos
28
Fully reconstructed sample

• Belle (447M BB) ? 4.12x105 B0B0
• 6.80x105 BB-

29
Event candidate B- ? t- nt
30
B?t n (Belle)
hep-ex/0604018 - preliminary

• NBB (produced) 447M
• NBB- (full recon.)
• 6.80 x 105 (purity 0.57)
• t decay modes
• Cover 81 of t decays
• Event selection
• Main discriminant residual ECL energy

Obtained Eresidual
Fit to Eresidual ? 21.26.7 signal events. ?4.2s
significance including systematics
-5.7
31
Impact of B- ? t- nt

• From BF(B ? t nt) ?Product of B meson decay
  constant fB and CKM matrix element Vub?use
  Vub from HFAG ? fB
• Use BF(B ? t nt) with Dmd ? constraint in the
  (r,h) plane

The common uncertainty from fB cancels in this
ratio.
32
Charged Higgs limits from B- ? t- nt
If the theoretical prediction is taken for fB ?
limit on charged Higgs mass vs. tanb
33
B?t n prospects

• Expected precision
• at Super-B
• 13 at 5 ab-1
• 7 at 50 ab-1
• Search with D() l n tag will help.
• ? BaBar 232M BB PRD 73 (2006) 057101
• Tag eff 1.75 x 10-3
• Signal selection eff. 31
• Similar S/N to Belle (full recon. sample)

34
Future Prospects B???
95.5C.L. exclusion boundaries
DfB(LQCD) 5
(for BFobs BFSM)
Extrapolations (T.Iijima)
Lum. DB(B?tn) exp DVub
414 fb-1 36 7.5
5 ab-1 10 5.8
50 ab-1 3 4.4
50ab -1
rH
35
B ? tt-

• BaBar (232M BB) PRL 96 (2006)
  241802
• Challenging measurement 2-4 neutrinos per event!
• Fully reconstruct one B (?D()X, Xcombination of
  up to 5 pions and kaons), 280k events
• t decay modes lnn, p-n, r-n (51 of tt decays)
• Reject events with KL, KS, K and employ neural
  network (kinematics of charged track momenta and
  ECL residual energy)
• ?26319 events (expect 28140 from sidebands, MC)
  
• BF(B ? tt-) lt 4.1 x 10-3 (90CL) SM
  prediction 0.12 x 10-6
• ? First ever limit on this channel
• Constrains leptoquark couplings and tanb
  enhancements

36
B- ? K- n n

• B?K()?? is a particularly interesting and
  challenging mode (with B ? t? as a small
  background), theoretically clean
• Experimental signature B ? K nothing
• The nothing can also be light dark matter with
  mass of order 1 GeV. Direct dark-matter searches
  cannot see the Mlt10 GeV region.
• SM prediction (3.81.2) x 10-6
• B ? t? analysis is a proof that such a one prong
  decay can be studied at a B factory
• Present limits
• BaBar (89M BB) BF(B?K??) lt 52 x 10-6 PRL 94
  (2005)101801
• Belle (275M BB) BF(B?K??) lt 36 x 10-6
  hep-ex/0507034

-0.6
37
B- ? K- n n prospects
MC extrapolation to 50 ab-1
5s
Observation of B ? K n n
SM prediction G.Buchalla,
G.Hiller, G.Isidori (PRD 63 014015)
Extra EM calorimeter energy
Fig. From SuperKEKB LoI
38
Charged Higgs search in B?D t n

• Charged Higgs contribution

c
b
H/W
t
nt
Decay amplitude
Br(SM) 8 x 10-3
Tauonic decay is the most sensitive

• Analysis reject events with p, K, reject Dtn
  contamination, no remaining charged or p0 tracks,
  cut on the ECL residual energy, angle between two
  ns and missing mass.

39
B?D t n (MC studies)

• Signal selection efficiency

10.2 2.6
26.1 13.3

• Expectation at 5 / 50 ab-1 for B decay

5ab-1 5ab-1 5ab-1 5ab-1 50ab-1 50ab-1 50ab-1 50ab-1
Mode Nsig Nbkg S dB/B Nsig Nbkg S dB/B
280 550 12.7 7.9 2800 5500 40.3 2.5
620 3600 12.7 7.9 6200 36000 40.3 2.5
5s observation possible at 1ab-1
40
B?D t n constraint on charged Higgs

• Once branching fraction is measured, we can
  determine R.

Reach in R
M.Tanaka, Z.Phys. C67 (1995) 321
BF error
at 5ab-1
? Reach in R 11
41
B?D t n ? charged Higgs
Constraint From b?sg
D(Form-factor) 5
D(Form-factor) 15
42
Super-B and LHCb complementary

• Clean environment ? measurements that no other
  experiment can perform. Examples CPV in B g fK0,
  B g hK0 for new phases, B g KSp0g for
  right-handed currents.
• B-meson beam technique ? access to new
  decay modes proof B gtn Example discover B
  g Knn.

• Measure new types of asymmetries. Example
  forward-backward asymmetry in b g smm, see
• Rich, broad physics program including B, t and
  charm physics. Examples searches
  for t g mg and D-D mixing with unprecedented
  sensitivity.

?LHCb
?LHCb
?LHCb
43
Summary

• Radiative, electroweak and tauonic B decays are
  of great importance to probe new physics.
• We are starting to measure B?tn, Dtn, AFB(Kll),
  ACP(Kp0g) etc. at the current B factories.
• ?Hot topics in the coming years !
• ?Watch out for updates (including this
  week)...
• For precise measurements, we need a Super-B
  factory!
• ?Observe K() nn, zero crossing in AFB, D()tn
• ?Expected precision (5ab-1?50ab-1)
• Br(tn) 13?7
• Br(D()tn) 7.9?2.5
• q02 of AFB(Kll) 11?5
• ACP(Kp0g) tCPV 0.14?0.04

44
Additional slides
45
Charged Higgs from Br(b?sg)
46
Radiative decays prospects
Summary by M. Nakao 1st Super-B workshop at
Hawaii (2004)
47
B- ? e- ne, m- nm

• Helicity supressed with respect to B ? t?
• B- ? m-nm
  SM prediction 0.4 x 10-6
• (Possibly better for charged Higgs limits than tn
  at high stat)
• Present limits
• Belle (152M BB) BF(B- ? m-nm) lt 2 x 10-6
  hep-ex/0408132
• BaBar (89M BB) BF(B- ? m-nm) lt 6.6 x 10-6 PRL
  92 (2005)221803
• B- ? e- ne
  SM prediction 0.00001 x 10-6
• Present limit
• Belle (65M BB) BF(B- ? e- ne) lt 5.4 x 10-6
  BELLE-CONF-0247

48
B- ? ee-, mm-

• No new results from B factories...
• BaBar (120M BB) BF(B ? ee-) lt 0.083 x 10-6
  BF(B ? mm-) lt 0.061 x 10-6
  PRL94(2005)221803
• Belle (85M BB) BF(B ? ee-) lt 0.19 x 10-6
  BF(B ? mm-) lt 0.16 x 10-6 PRD 68
  (2003)111101
• .... With present statistics we could be
  competitive with Tevatron
• Limits from Tevatron hep-ex/0508058
• BF(Bd ? mm-) lt 0.032 x 10-6 SM prediction
  0.0001 x 10-6
• BF(Bs ? mm-) lt 0.12 x 10-6 SM prediction
  0.0035 x 10-6

49
B?D t n (MC studies)

• Use fully reconstructed samples.
• t decay modes
• Analysis cuts
• Reject events with p, KL
• Reject D t n contamination
• No remaining charged or p0 tracks
• ECL residual energy
• Angle between two ns
• Missing mass

Signal
BG
50
B ? t nt
hep-ex/0604018 - preliminary
BF(B ? t nt)(1.06 0.34 (stat.) 0.18 (syst.)) x
10-4
-0.28
-0.16
? Product of B meson decay constant fB and CKM
matrix element Vub fB Vub (7.73 1.24
(stat.) 0.66 (syst.)) 10-4 GeV Using Vub
(4.39 0.33)10-3 from HFAG fB (176 28
(stat.) 20 (syst.)) MeV First measurement of fB!
fB (216 22) MeV (an unquenched lattice
calc.) HPQCD, Phys. Rev. Lett. 95, 212001 (2005)

-1.02
-0.58
-23
-18