Beyond the Standard Model
Searches for New Physics at the
B-Factories
K. Honscheid
Ohio State University
Aspen 2006
Where to look for New Physics?
K. Honscheid, Ohio State University, Aspen 2006
Experimental Strategies
b→sg penguins
b→sg penguins
g
W+, H+
s
b
b
W+, H+
d
B→tn
B,D →Xusll,Xusnn
s f
s
s 0
d K
Z, h0
s,u
l,n
l,n
Surprises
l, n
b,c
B,D →ll,nn
b,c
W+, H+
l, n
d
b
u
t
W+, H+
LFV
t→lg
n
t
~
n
~
c
l
g
From Hitlin, LHC Lavor Workshop
K. Honscheid, Ohio State University, Aspen 2006
Experimental Landscape (ca 2006)
1800
Integrated Luminosity
1600
Belle
550 fb-1
Million B
t Mesons
Leptons
Mesons
D
Million
1400
BaBar
1200
320 fb-1
1000
800
600
400
CLEO II.5
CLEO II
200
0
1990
1992
1994
1996
1998
2000
2002
2004
K. Honscheid, Ohio State University, Aspen 2006
Search for Lepton Flavor Violation in t Decays
• Lepton flavor is conserved in the Standard Model
– not protected by an underlying
conserved current symmetry
– Neutrino Masses
t
• Many SM extensions include LF violation
~n
l
~
c
g
• Observation of LF violation would be a clear signature for
new physics
K. Honscheid, Ohio State University, Aspen 2006
Search for t- → e-g
PRL 96, 041801 (2006)
n
ttag
Positrons
Electrons
tsignal
e
g
• 207106 e+e- → t+t- events
• Observe 1 event in a 2s2s signal box on 1.9±0.4 expected background
Br(t- → e-g) < 1.1  10-7 at 90% C.L.
Previous limits (all @ 90% CL):
Br(t- → m-g) < 0.68  10-7
Br(t- → m-g) < 3.1  10-7
Br(t- → e-g) < 3.8  10-7
(BaBar, PRL 95 041802)
(Belle, PRL 92 171802)
(Belle, PLB 613 22)
K. Honscheid, Ohio State University, Aspen 2006
Limits on New Physics – An Example
tlg and mSUGRA
Just an example with
tanb=55
Trilinear coupling A0=0
Inverted mn hierarchy
…
Br(teg)<1x10-8
-8
11x10-8 5x10-8 2x10
Current BaBar limit
O. Igonkina, SUSY 2005
K. Honscheid, Ohio State University, Aspen 2006
Search for New Physics in the Charm Sector
• D0 Mixing
Flavor eigenstates are not mass eigenstates
with M1,2 and G1,2
Using D0 K+p-, 400 fb-1
Wrong sign
• D0 Mixing is small in the SM
box diagram: x,y, < few x 10-5
long distance: x ~ 10-3, y ~ 10-2
• Strong phase in hadronic decays
x’ = x cosd + y sind
y’ = y cosd – x sind
• Proper Time Fit
hep-ex/0601029
K. Honscheid, Ohio State University, Aspen 2006
Experimental Limits on D0 mixing
hep-ex/0601029
World 95% CL in (x,y)
World 95% CL in y
95% CL Limits (assuming CP)
x’2 < 0.72 x 10-3
-9.9 x 10-3 < y’ < 6.8 x 10-3
Note: No-Mixing (0,0) has only 3.9% CL
G. Burdman and I. Shipsey, Ann. Rev. Nucl. Part. Sci. 53, 431 (2003)
[arXiv:hep-ph/030076]
(assuming d = 0, CPV = 0)
K. Honscheid, Ohio State University, Aspen 2006
Rare Charm Decays and CP Violation
Experiment
Decay mode
ACP (%)
BaBar
D+K-K+p+
1.41.0 0.8
BaBar
D+ f+p+
0.21.50.6
BaBar
D+ K*0 K+
0.91.70.7
CLEO II.V
D0 p+ p- p0
CDF
D0 K+K-
2.0 1.2  0.6
Direct CPV
CDF
D0 p+ p-
1.0 1.3  0.6
Direct CPV
FOCUS
D0 K+K-p+ p-
1.0 5.73.7
FOCUS
D+ KoK+p+p-
2.3 6.22.2
FOCUS
DS KoK+p+
-3.6 6.72.3
T violation
through triple
product
correlations
1 8
Notes
Res. Substr.
Of
D+K-K+p+
Dalitz plot analysis
from S. Stone
K. Honscheid, Ohio State University, Aspen 2006
Experimental Aside: Single B Meson Beams
Lots of interesting modes include one or more neutrinos.
“Beams” with a single, monochromatic B and without c, QED etc
would be very useful for : Btn, Bnn, BKnn,…
Fully reconstruct one of the Bs and study the remaining of the
event  closed kinematics, missing energy reconstruction
Tag types
Semileptonic D(*)l(np)
5K/fb-1
Hadronic
D(*) X
efficiency purity
3K/fb-1
X=np+mp0+pK+qKs
K. Honscheid, Ohio State University, Aspen 2006
Search for B tn
hep-ex/0507069
SM decay proceeds via W-annihilation diagram
Sensitive to new physics charged current
Analysis:
Undetected neutrinos result in large missing energy and few kinematic
constraints – high background.
Reduce the background by reconstructing the second B (“tag B”) in the
event in the copious decay mode B-→D*0l-nl
Reconstruct B+→t+ut with t+→l+nnbar or t+→h+n, where h = p, r, or a1
Require no additional charged tracks in the event
K. Honscheid, Ohio State University, Aspen 2006
Results for Btn
t+nte+ne
Expectations from U.T. fit
BaBar
SL tags
230M BB
MC Signal for Br=10-3
Energy in addition to signal candidate (GeV)
No signal in 230 M BB events
90% CL Upper Limit
(Combined with hadronic tags)
Br(B+→t+ut) < 2.6  10-4
Current BaBar
90% C.L.
Close to the discovery of this mode or NP
…even closer
hep-ex/0507034
New Belle analysis using 250 fb-1
BF ( B  tn )  1.8 10-4
K. Honscheid, Ohio State University, Aspen 2006
Higgs sensitivity
2HDM
Gambino, Misiak Nucl. Phys. B611 338
Hou Phys.Rev.D48:2342-2344,1993
mH(GeV)
Limits on 2-Higgs Models from
Btn
Current
limits
both
bsg
Btn
tanb
K. Honscheid, Ohio State University, Aspen 2006
Search for Bs m+m- at the Tevatron
hep-ex/0508058
Standard model prediction: Br(Bs m+m-)= (3.50.9) 10-9
No Signal found
CDF 360 pb-1
300pb-1
Bsm+m- < 2.010-7 CDF
Bsm+m- < 3.910-7 D0
Combined (90% CL)
Bsm+m- < 1.510-7
K. Honscheid, Ohio State University, Aspen 2006
EW Penguins: BKl+l-, BK*l+l-, and BXsl+l-
• With l+l- pair, can produce both pseudoscalar
and vector mesons
• SM: Br(B →Kl+l-) ~ 4  10-7 (± 30% theory)
~3 times that for K*
BK
B  K*
• Asymmetries (AFB,
CP)
• mm/ee ratio
Belle prelim. hep-ex/0410006, 0508009
BABAR hep-ex/0507005 (229M BB)
+ -
New Physics affects
Affects
• Rates
+ -
K. Honscheid, Ohio State University, Aspen 2006
BKl+l- and BK*l+l- : branching fractions
Theory errors
mainly due to
form factors.
B( B  K *
+ -
)WA  (1.18  0.17)  10-6
)WA  (0.45  0.05)  10-6
(rarest observed B decay)
B( B  K
+ -
(10-6 )
pole at low q2
q2
q2
K. Honscheid, Ohio State University, Aspen 2006
BK*l+l-: Lepton F-B Asymmetry
l
hep-ex/0508009
-
l
B
l
+
s
q
K
*
Lepton angular
distribution in
l+ l- rest frame
386 M BB
SM
AFB
NP scenarios
q2
constraints on Wilson coeffs describing
short-distance physics
K. Honscheid, Ohio State University, Aspen 2006
Getting Lost?
Little Higgs w
MFV UV fix
Generic Little Higgs
Extra dim w
SM on brane
Supersoft
SUSY breaking
Dirac gauginos
SM-like B physics
Generic extra dim w SM in bulk
SUSY GUTs
MSSM
MFV
small tanb
MSSM
MFV
large tanb
Effective SUSY
New Physics in B data
after G. Hiller
K. Honscheid, Ohio State University, Aspen 2006
Be ready for surprises
X(3872)
Y(3940)
B decays
Z(3930)
gg
B decays
BELLE
2003
2004
BaBar
2005
DsJ(2458)
DsJ(2317)
continuum
Y(4260)
ISR
Several new particles have been observed by the B Factories
Different production mechanism
Different JPC
For a nice review by Swanson see hep-ph/0601110
K. Honscheid, Ohio State University, Aspen 2006
CP Violation in the B System
ACP e i f
B0
q
 e -2i b
p
B
0
fCP
ACP e -i f
Present
CKM fit
Present
WA
BABAR 0.722±0.040±0.023
Belle 0.652±0.039±0.020
See talk by Aihara for more on CPV
K. Honscheid, Ohio State University, Aspen 2006
Bounds on new physics from UT fits
CKM Model confirmed by many measurements
Now look for New Physics as correction to CKM Model
The fit to the unitarity triangle constraints:
b
W
=
+ =
cNP
d
b
W
K. Honscheid, Ohio State University, Aspen 2006
Bounds on new physics from UT fits
Constraints: |Vub|, Dmd, SyK
First constraints on Qd, CBd
In terms of probed new physics mass
scale (L=m(cNP); gNP=coupling)
L(now)~5gNP TeV
add: a, g
Ligeti,
Silvestrini,
Agashe et al
K. Honscheid, Ohio State University, Aspen 2006
The next few years (2006 – 2008)
• LHCb is nearing completion
• Belle and BaBar
• 1 ab-1 (2006)
• 2 ab-1 (2008)
• Tevatron
• 2 fb-1 (2006)
• 8 fb-1 (2009)
1200
ICHEP08
1000
800
600
400
200
Jul-08
Jul-07
Jul-06
Jul-05
Jul-04
Jul-03
Jul-02
Jul-01
Jul-00
Jul-99
0
See talk by S. Stone on Thursday
K. Honscheid, Ohio State University, Aspen 2006
Super B-Factory Plans at KEK and Frascati
Design Luminosity ~ 1 x 1036 /cm2/sec
Synergy with ILC
Lots of R&D needed
New Beam pipe
More RF power
Interaction Region
Crab crossing
=30mrad.
by*=3mm
New QCS
Damping ring
Linac upgrade
KEK
L = 41035/cm2 /sec
Frascati
Next Super B Workshop in Frascati, March 16-18, 2006
K. Honscheid, Ohio State University, Aspen 2006
Have Ideas – Need Statistics
Mode
BRSM
Notes on New Physics
bsg
~3 10-4
BF, ACP, and AFB important
bsg
~10-5 each
BF not critical. Need events to
measure SCP
BXll
~10-6 each
BF, ACP, and AFB important
BXnn
~10-6 each
Up to 10-5 each
DXll
~10-6 each
Up to 10-5 each
Btn
~10-4
Experiments close to SM sensitivity
tlg
~10-40
Up to 10-8
Bll
<10-11
Up to 10-5
Dll
<10-9
Up to 10-6
K. Honscheid, Ohio State University, Aspen 2006
Additional Transparencies
K. Honscheid
Ohio State University
Aspen 2006
Let’s rule out at least one NP model
Observation of direct CP violation in B0K+p232x106 BB’s
B0K+pB0K-p+
BaBar 2004
ACP  -0.133  0.030  0.009
New Belle Result:
-0.113+/- 0.022+/- 0.008
 Superweak Model (Wolfenstein) is really out
K. Honscheid, Ohio State University, Aspen 2006
M0 [GeV]
Implications of Bs m+m-<1.510-7
mSUGRA prediction:
Dedes, Dreiner, Nierste PRL87:251804,2001
Excluded!
K. Honscheid, Ohio State University, Aspen 2006
Search for LFV in t- → lhh’
PRL 95, 191801 (2005)
t-→
e-K+K-
e-K+p-
e-p+K-
e-p+p-
e+K-K-
e+K-p-
e+p-p-
UL(10-7)
1.4
1.7
3.2
1.2
1.5
1.8
2.7
t-→
m-K+K-
m-K+p-
m-p+K-
m-p+p-
m+K-K-
m+K-p-
m+p-p-
UL(10-7)
2.5
3.2
2.6
2.9
4.8
2.2
0.7
hep-ex/0509016
Br(t- → e-Ks) < 5.6  10-8 at 90% C.L.
Br(t- → m-Ks) < 4.9  10-8 at 90% C.L.
K. Honscheid, Ohio State University, Aspen 2006
First Search for B0→t+t-
hep-ex/0501115
SM expectation ~ 10-7
(Compare: 10-11 for m+m-, 10-15 for e+e-)
Helicity (ml/mB)2 and Cabibbo (VtbVtd*) suppressed
Very clean theoretically
Small SM
Sensitive to new physics in the loops
t+
tt+
t-
Analysis
Reconstruct the tag B in decays to D(*) and up to
5 kaons and pions
Reconstruct the signal decay
B0→t+t-, with t→1-prong decays
Require no additional charged tracks in the event
t+
t-
Require little remaining energy: Eextra < 0.11 GeV
Use correlations between Eextra and the momenta
of the reconstructed t daughters in a neural net to
further reduce the background
K. Honscheid, Ohio State University, Aspen 2006
B0→t+t- Results
hep-ex/0501115
Using 232  106 Y(4S)→BB events we observe
263 ± 19 events with 281 ± 40 expected background.
First limit on this mode
Br(B0→t+t-) < 3.2  10-3 at 90% C.L.
2
2
mES  Ebeam
- pBtag
GeV
Yields a constraint on leptoquark parameter. For example
RH coupling to
generations i,j
Leptoquark doublet mass
K. Honscheid, Ohio State University, Aspen 2006
Measurement of B(BXsl+l-)
Analysis
Semi-inclusive technique
Xs is reconstructed from K+ or Ks +
0-4p (at most one p0 is allowed)
hep-ex/0503044
2/m 2 )
Decay width for bsll
(s=q
b
140 fb-1
used
d G b  s
dsˆ
+ -
  a
2
2 5
*
2
2
 GF mb VtsVtb
1 - sˆ 


3
4
p
48
p


2
2
2


 2
 1 + 2sˆ  C9eff + C10eff + 4 1 +  C7eff + 12 Re  C7eff C9eff *  
 sˆ 


em

MXs < 2.0 GeV
Electron or muon pair

Mll>0.2GeV
Charmonium veto
flavor
Sign Wrong
of C7 flipped
Theoretical prediction by Ali et al.
SM
MXs
q2
K. Honscheid, Ohio State University, Aspen 2006
Constraints on Ci from B(BXsl+l-)
P.Gambino, U.Haisch and M.Misiak PRL 94 061803 (2005)
Clean prediction for B(BXsll) with 1<q2<6GeV2 is available.
Combine Belle and Babar results
Sign of C7 flipped case with SM C9 and C10 values is unlikely.
BF
Belle
Babar
WA
SM
C7 = -C7SM
q2>(2mm)2
4.11±1.1
5.6±2.0
4.5±1.0
4.4±0.7
8.8±0.7
1<q2<6GeV2
1.5±0.6
1.8±0.9
1.60±0.5
1.57±0.16
3.30±0.25
Donut : 90% CL allowed region
C10NP
C7
SM
C10NP
C7 = -C7SM
C9NP
K. Honscheid, Ohio State University, Aspen 2006
2005: Sample used for AFB(BK*ll)(q2)
hep-ex/0503044
2001: Belle 1st observed
Treat q2, cos(θ) dependence of
bkgs.
Sample for BK* l l events 113±13
BK l l control sample 96±12
Consistent with flat
K. Honscheid, Ohio State University, Aspen 2006
New Measurement of AFB(q2) in K*ll
Forward-backward asymmetry is induced by interference btw
virtual photon and Z0 contributions.
Relative signs and magnitudes of C7 to C10 and C9 to C10 can be
determined from AFB(BK*ll)!!
We do not use Ci but Ai which is leading coefficients.
K. Honscheid, Ohio State University, Aspen 2006
Fit results for ratios of Wilson coefficients
Null test with K+ll
Projection to AFB
Best fit for negative A7 (SM like)
J/Y
Y’
SM
Best fit for positive A7 (non-SM like)
(This solution is also allowed)
K. Honscheid, Ohio State University, Aspen 2006
Confidence Level Contours
For any allowed A7, we obtain 95% CL
SM
SM
Sign of A9A10 is negative!!
Sign of A9A10 flipped case is excluded
The first and third quadrants
Sign of A7A10 is not determined yet.
Best fit
The second and fourth quadrants
A10/A7
ＳＭ
A9/A7
fit result
A7A10 sign flipped (to SM)
Both A7A10 and A9A10 signs flipped
A9A10 sign
flipped
K. Honscheid,
Ohio
State University, Aspen 2006
What is the ultimate possible UT precision?
Theoretical limits (continuum methods)
Many measurements will not be theory-limited for quite
some time
Ligeti:
K. Honscheid, Ohio State University, Aspen 2006