Radiative and Rare B-Decays at BaBar
Carsten Hast, SLAC
for the BaBar Collaboration
Pittsburgh, October 17
All results are preliminary unless journal ref. is given
limit values are 90% CL unless otherwise specified
Overview
Since this talk is nicely framed between a theoretical introduction
and a Belle summary and outlook I will focus on newer results and
analysis techniques from BaBar
•
FCNC
•
Purely Leptonic Decays
b  Knn
b  tn
b  mn
•
•
b  Kll and b  K*ll
B  Xsll
•
Radiative Decays
B0  K2*0(1430) g and B+  K2*+(1430) g
Beauty 2003, October 17th
Carsten Hast, SLAC
2
The BaBar Detector
EMC
6580 CsI(Tl) crystals
1.5 T solenoid
DIRC (PID)
144 quartz bars
11000 PMs
e- (9
e+ (3.1 GeV)
GeV)
Drift Chamber
40 stereo layers
Instrumented
Flux Return
iron / RPCs (muon
/ neutral hadrons)
Silicon Vertex Tracker
5 layers, double sided
strips
SVT+DCH: (pT)/pT = 0.13 %  pT + 0.45 %, good dE/dx
DIRC:
K- separation 4.2  @ 3.0 GeV/c  2.5  @ 4.0 GeV/c
EMC:
Very good electron identification and 0 reconstruction
IFR:
Decent muon identification
Beauty 2003, October 17th
Carsten Hast, SLAC
3
PEP- II Lumi Performance and Recorded BaBar Lumi
124.1106 BB
Best Peformance
PEPII peak Luminosity:
6.582x1033 cm-2 sec-1
Integrated Luminosity
Shift: 135.2 pb-1 in 24
hours: 391.2 pb-1
89.7106 BB
82 fb-1
on-peak data
for analysis
Beauty 2003, October 17th
On(off)-peak
Run 1+2
82 (10)fb-1
Run 3
31 (4) fb-1
Run 4
6 (0)fb-1
Carsten Hast, SLAC
4
Standard Variables in U(4S) Frame
e+e- U(4S)  BB
B produced almost at rest in U(4S) frame
For B decay with no missing particles use beam energy to constrain
mass and energy of the reconstructed B
*
E  EB*  Ebeam
0 for signal
mES 
(E
2
beam
 PB2

mB for signal
*
*
E  Esignal

E
B
beam
background
background
E and mES provide uncorrelated measurement of energy and mass
Beauty 2003, October 17th
Carsten Hast, SLAC
5
B+K+nn Search
A rare flavour changing neutral current
(FCNC) bsnn decay
SM prediction: Br (BKnn)  410–6
Summed over all neutrino species
• The best upper limits (@90% CL):
– CLEO: Br (BKnn)  2.410–4
– BABAR: Br (BKnn)  9.410–5
PRL 86 2950 (2001)
hep-ex/0207069
Semi-Leptonic B-tags: BDlnX (X  g, 0 or nothing) 50.7 fb-1
• New search (80 fb-1) with hadronic B-tags
• B  D0(*) () (K) (K0s) (0)
(D*  D0 and D0  K , K0, K3, Ks)
• B constraining kinematic variables: E and mES
Beauty 2003, October 17th
Carsten Hast, SLAC
6
B+K+nn Search II
• Identify tag B in signal E and mES area
• All remaining tracks and neutrals
belong to the recoiling
signal-side B
• One well identified Kaon
• no 0
• Missing momentum not
in the beam pipe
• “extra” Energy < 300 MeV
Data
B+B MC
Beauty 2003, October 17th
BKnn MC
B0B0 MC
Carsten Hast, SLAC
cc MC
u,d,s MC
7
B+K+nn Upper Limit
• Selected events: 3
• Expected background events: 2.70.8
Br ( B Knn ) < 1.05  10  4 @ 90% CL
hep-ex/0304020
• Combining this result with the previous, (Br (BKnn)  9.410–5)
statistically independent BABAR result, yields a new limit:
Br ( B Knn ) < 7.0  10  5 @ 90% CL
Beauty 2003, October 17th
Carsten Hast, SLAC
Going to PRL
8
Btn Search
2 analyses of the 1999-2002 data sample: ~81 fb1
Reconstruct one B meson as
SemiLeptonic B-tag
B  Dln X
Hadronic B-tag
B  DXhad
The remaining neutrals and tracks are signal candidates
• t  (e, m) n(e, m) nt
• t(e, m) n(e, m) nt
• t(, 0, ) nt
very clean but low efficiency (~ 0.25 – 0.30 %)
The two samples of tag B mesons are statistically independent
Beauty 2003, October 17th
Carsten Hast, SLAC
9
t Signal Selection
SemiLeptonic B-Tag
• t+t event veto
•
Hadronic B-Tag
Only one charged track having low
impact parameter
Track is not identified as a kaon
Track is an identified lepton
Residual neutral energy is
used to model PDFs to extract signal
and background contributions
•
•
•
Background
Events/0.07 GeV
Signal
•
•
•
•
•
Eleft (GeV)
•
•
Three preselections
• 1 track, 0 0
• 1 track, 1  0
• 3 tracks, 0  0
Track is not identified as a kaon
Track is identified as lepton or pion
Additional cuts on
• Missing momentum
• Residual neutral energy
• Track momentum
• Invariant masses
Background mainly from Vcb events
Signal efficiency: 11.3%
Eleft (GeV)
Signal efficiency: 22.5%
Beauty 2003, October 17th
Carsten Hast, SLAC
10
Btn Results
Hadronic B-Tag
SemiLeptonic B-Tag
40.9  3.8
Expected Bkg from MC
39.9  2.8
Observed events onpeak data
47.7  7.0
Events/0.07 GeV
Expected Bkg in onpeak data
(sideband extrapolation)
Selection
Total Bkg
Data
candidates
te ne nt
tm nm nt
t nt
t0 nt
t nt
6.7 2.0 0.6
5.0 1.7 0.4
11.2 2.5 0.5
10.4 2.6 1.2
4.3 1.4 0.3
10
8
6
7
4
All
37.6 4.7 1.5
35
Br ( B  t  n ) < 7.7  10  4
hep-ex/0304030
Combined limit
Br ( B  t  n ) < 4.9
Eleft (GeV)
 10  4
hep-ex/0303034
Beauty 2003, October 17th
Br ( B  tn ) < 4.110 4 @ 90% CL
SM prediction: < 9.2105
Carsten Hast, SLAC
hep-ex/0305034
11
Bmn Search
Data sample 1999–2002: ~81fb–1
Signal Monte Carlo
Two Body decay  p(m) ~ mB/2
B rest frame
•
One well identified muon
•
All the rest is from the companion B
CM frame
– do particle identification
– No additional leptons allowed
•
After the companion B was found,
p(m) is re-reconstructed in the
signal B rest frame
Beauty 2003, October 17th
Carsten Hast, SLAC
12
Bmn Search
All other tracks and neutrals make the Companion-B
Signal Monte Carlo
Beauty 2003, October 17th
Carsten Hast, SLAC
13
Upper Limit for Bmn
Data
• Signal Selection efficiency: 2.09%
• Selected events: 11
+1.8
• Expected background: 5.0
1.4
• Bg fluctuation prob. to the signal: 4%
Bmn MC
Data
Sideband
Signal box
B ( B mn ) < 6.6  10  6
@ 90% CL
BABAR-CONF-03-007
Beauty 2003, October 17th
Carsten Hast, SLAC
14
b K(*) l l
penguin diagram b s ll
box diagram b s ll
Updated result on 113 fb-1 accepted by PRL
Reconstruct K+, Ks   +-,
K*0  K+ -,
K*+  Ks +
Lepton identification for
e+e- with p(e) > 0.5 GeV/c,
m+m- with p(m) > 1.0 GeV/c
Beauty 2003, October 17th
Carsten Hast, SLAC
15
b K(*) l l Peaking Background
Events with the same final state:
B  J/y K(*), y’ K(*)
y’ veto
E [GeV]
E [GeV]
J/y veto
Mll [Gev/c2]
Mll [Gev/c2]
Signal is scattered in this above area
500 fb-1
B  J/y K(*), y’ K(*) Monte Carlo
little to no contribution
outside these veto bands
Beauty 2003, October 17th
Carsten Hast, SLAC
16
b K(*) l l Peaking Background II
Background from lepton miss-identification
B  h+h- K(*)
h+h- K(*) events in data convolved with rates for h to fake e, m
0.33 events
# Events
# Events
0.01 events
mES [Gev/c2]
mES [Gev/c2]
Included as part of the fit to data
Beauty 2003, October 17th
Carsten Hast, SLAC
17
b K l l Results
2-D unbinned maximum LH fit to mES and E
on all 4 decays K+ee/mm and Ks0ee/mm
8.4 excess over background
Br(bK ll) = (6.5 +1.4-1.3 +/- 0.4 ) 10-7
Beauty 2003, October 17th
accepted by PRL
Carsten Hast, SLAC
18
b K* l l Results
3-D unbinned maximum LH fit to mES, E and mK
3.3  excess over background
Br(bK* ll) = (8.8 +3.3-2.9 +/- 1.0 ) 10-7
Beauty 2003, October 17th
Constrain: K*ee = 1.33 X K*mm
Gtot(B0) = 1.08 X Gtot(B+)
accepted by PRL
Carsten Hast, SLAC
19
b K(*) l l Results
Comparison of mll in data with simulation,
normalized to the measured Br
Fits well
Beauty 2003, October 17th
Carsten Hast, SLAC
20
Semi-Inclusive B Xs l l
Sum of exclusive modes approach:
reconstruct Xs with 1 K± or K0s  +-, # ( + 0)  2, # 0  1
10 modes, ~50% of BR(B  Xs l+l-) (75% if assumed KS=KL)
Similar analysis as b K(*) ll but with higher combinatoric background
B Xs
e+e-
B Xs m+m-
mES [Gev/c2]
Beauty 2003, October 17th
mES [Gev/c2]
Carsten Hast, SLAC
21
Semi-Inclusive B Xs l l Results
Mode
Nsig
( stat)
Significance
(stat)
Efficiency
B
Xs e+e
29.0 
8.3
4.0
2.5%
B
Xs m+m
12.4 
6.2
2.2
1.2%
Sum
41.4 
10.3
4.6
+
Data
Monte Carlo
Br(B  Xs e+e) = (6.6  1.9  1.91.6) 106
Br(B  Xs m+m) = (5.7  2.8  1.71.4) 106
Br(B  Xs l+l) = (6.3  1.6  1.81.5) 106
Beauty 2003, October 17th
Carsten Hast, SLAC
22
B0  K2*0(1430) g and B+  K2*+(1430) g
BaBar Preliminary
BaBar Preliminary
Br (B0  K2*0(1430) g)
Br (B+  K2*+(1430) g )
BaBar (12.2 +/- 2.5 +/- 1.1)10-6
(14.4 +/- 4.0 +/- 1.3)10-6
Belle ( 13 +/- 5 +/- 1) 10-6
Cleo
(16.6 +/- 5.9 +/- 1.3) 10-6
Beauty 2003, October 17th
Carsten Hast, SLAC
23
Summary (1)
•
•
FCNC
b  Knn3
bg events expected 3 seen
Purely Leptonic Decays
b  tndifficult
b  mnlooks
•
quite interesting
Radiative Decays
B0  K2*0(1430) g and B+  K2*+(1430) g
Beauty 2003, October 17th
BaBar is in the game
Carsten Hast, SLAC
24
B s l l Summary
Br(bs ll) in good agreement within experiments and prediction
Next step: measure angular and kinematic distributions
…and have fun with the next talk covering all the other interesting rare results measured by Belle
Beauty 2003, October 17th
Carsten Hast, SLAC
25