Transcript Document

KEK, August 13, 2003
CP Asymmetries:
New Results from Belle
•Weak interaction: CP, CKM matrix
•B(eauty) mesons & CP asymmetry
•B production: e+e– -> (4S) at KEKB
•Belle experiment
Highlights for 2003
Selected CP & other results (preliminary)
•Plans
Kay Kinoshita
University of Cincinnati
Belle Collaboration
CP Violation
Why is it of interest?
• matter-antimatter asymmetry in universe
requires CP-violating interactions (Sakharov 1967)
Weak interaction - the only known force that
• violates CP symmetry (a little)
.. but to 1998, only in KL (1963)
• violates P symmetry (maximally)
… but preserves CP symmetry (mostly)
right-handed particles, left-handed antiparticles.
no coupling to LH particles, RH antiparticles.
• allows particle to change identity (flavor)
What is source of CP asymmetry in KL? in universe?
… step back, look at weak coupling strengths…
KEK, August 13, 2003
2
Weak coupling strengths
Standard Model = 12 fermion flavors (+antifermion)
+ strong, EM, weak forces,
unification of EM+weak
3* generationsx2 typesx2 ea,
stable, but for weak interaction
*generations distinguished only by mass
Couplings:
• no generation x-ing, except
charged-current quark
• leptons: ~same
quarks: all different, approx
generation-conserving
±
W "charged current"
Generation
type
Q/|e|
1
2
3
lepton
–1
0
+2/3
–1/3
e
e
µ
µ

(no strong)
quark
(strong)
up
down
0
Z "neutral current"

charm truth
strange beauty
u
c
t
seen
suppressed
d
s
b
not seen
Inelegant!!
KEK, August 13, 2003
3
Elegance restored: GIM mechanism
Picture
• strong doublets, generations“degenerate,” perturbed by weak force:
c
t
new doublets ud'
s'
b'
no generation x-ing, universal W-coupling (=gF, seen in leptons)
d', s', b' are linear combinations of d, s, b:
d'
s' =
b'
M
d
s
b
Cabibbo-Kobayashi-Maskawa (CKM) matrix
complex
preserves metric
“ orthogonality
}
= unitary
Explains (Glashow-Iliopoulos-Maiani)
• suppression of flavor-changing neutral currents
• multiplicity of charged current couplings
• AND …….
KEK, August 13, 2003
4
… for >2 generations, e.g. 3, {9R+9I} dof constrained by unitarity:
4 free parameters, incl. 1 irreducible imaginary part
>>>> CP Violation >>>> (Kobayashi-Maskawa 1973)
Makoto
Kobayashi
First 3rdgeneration particle ()
seen 1975
Toshihide
Maskawa
KEK, August 13, 2003
5
3-generation Matrix
d
{1/gF x
couplings} =
u
c
t
Vud
Vcd
Vtd
s
Vus
Vcs
Vts
b
Vub
1-l2/2
l l3A(r-ih)
Vcb unitarity
-l
1-l2/2 l2A
Vtb
l3A(1-r-ih) -l2A
1
from decay rates,
Vji*Vjk=ik
{i=1,k=3}: Vub*Vud+Vcb*Vcd+Vtb*Vtd=0
=> Vub*Vud + 1 + Vtb*Vtd = 0
Vcb*Vcd
Vcb*Vcd
-(rih)
explicit parametrization(Wolfenstein):
l = 0.220 ± 0.002
A = 0.81 ± 0.08
|r-ih| = 0.36 ± 0.09
|1-r-ih| = 0.79 ± 0.19
-(1-r-ih)
(r, h): "unitarity triangle"
Self-consistent if CKM is correct
KEK, August 13, 2003
6
CKM CP phenomenology
CP asymmetry - due to ≥3 generations
-> need process w. all 3 (occurs with many B decays),
interference in  2 processes (but not too many)
-> probe different angles w different decays
(r,h)
1.0
B0->ππ
B0->rπ

VudVub
*
VcdVcb*
B0-> D*π
B0->D*r
B->DCPK


0.0
Program: are all asymmetries
consistent w single
KM phase?
Vtd Vtb*
VcdVcb*
first result
B0->J/Ks
B0->Ks
B0->D(*) D(*)
1
1.0
KEK, August 13, 2003
7
e.g. B -> J/ Ks for sin 21 (Sanda/Bigi/Carter)
tree (real Vij)
+
mixing+tree (Vtd*2)
(no cc of hadronic phase under CP)
CP asymmetry from x-term(s) - no theoretical uncertainty:  arg(Vtd2) = 21
This is only the cleanest, simplest - "golden mode"
KEK, August 13, 2003
8
CP phenomenology: variations
e.g. B -> J/π0 2 paths, different phases, + mixing
Tree
aVcb*Vcd
Penguin
aVtb*Vtd
(relative amplitudes, strong phase not known)
mixing+ "
aVtb*2Vtd2VcbVcd*
mixing+ "
aVtb*2Vtd2VtbVtd*
Bottom line: "direct" CP asymmetry possible
"direct" asym
KEK, August 13, 2003
9
B production:

}B
(4S)
t=0
e+ B2
e-

CP=–1,
conserved
+e-
-> (4S)
What else is needed?
• >107 B's just to get started - KEKB
• hadron (K/π) ID dE/dx, Aerogel, TOF
• lepton ID - CsI, multilayer µ
• <<200µm vertexing –
BB threshold
double-sided silicon strip
2nd decay
at t=t
z≈tbgc
~200 µm
B1
first B decay
(t=0), break CP
CP mode @ t=t
BR < 10–3
flavor tag @t=0: e, µ, K±, ...
effective eff ~30%
KEK, August 13, 2003
10
KEKB
bg = 0.425
Lmax = 1.06 X 1034 cm–2s–1 (world record)
Data (6/1999–6/2003)
Ldt = 158 fb–1@{(4S)+off(~10%)}
(1.52x108 B events)
s(E*beam) = 2.6 MeV
IP size = 77µm(x) x 2.0µm(y) x 4.0mm(z)
KEK, August 13, 2003
11
Belle detector
Charged tracking/vertexing
- SVD: 3-layer DSSD Si µstrip
– CDC: 50 layers (He-ethane)
Hadron identification
– CDC: dE/dx
– TOF: time-of-flight
– ACC: Threshold Cerenkov (aerogel)
Electron/photon
– ECL: CsI calorimeter
Muon/KL
– KLM: Resistive plate counter/iron
KEK, August 13, 2003
12
…the people
274 authors, 45 institutions
many nations
KEK, August 13, 2003
13
Belle physics results
58 papers published or in press (1st in 3/2001)
54 abstracts submitted to XXI Lepton-Photon (Fermilab)
(4S)
18 - CP asymmetry in B decay
25 - B decay non-CP
8 - charm hadrons
1 - QCD
2 - tau
Physics topics overlap in many
analyses, e.g., discovery of new
charmonium states in B decays.
KEK, August 13, 2003
14
Recent highlights
CP and related
• time-dependent CP measurements
update of J/Ks (1)
with J/π0(~1), D*+π–(21+3), Ks(1)
• evidence/observation
B-> K*l+l-, π0π0, D+D-, π0r0
• new method for 3: Dalitz plot analysis
D0K+ {D0->Ksπ+π–}
Charm
• new processes/ particles
new excited Ds states
new charmonium
first radiative: D0->g
• difference of CP lifetimes in D (yCP)
KEK, August 13, 2003
15
B decay reconstruction at
~3% background
s~10MeV
Kinematics for final selection:
exploit
• exclusive pair production of B
• narrow resolution of collision energy
E
Signal region
E E*cand–E*beam 0 (E*beam s 1/2/2)
s~10-50 MeV, depending on mode
Mbc (Beam-constrained mass)
Mbc (E*beam2-p*cand2 )1/2
s~3MeV/c2
Mbc
KEK, August 13, 2003
16
Time dependent measurements:
sin21: {charmonium}+Ks tag (CP=–1)
Fully reconstructed
BELLE-CONF-0353
2911 used
in fit
KEK, August 13, 2003
17
sin21: {charmonium}+KL tag (CP=+1)
J/ KL
1399±67 signal
pB* (cms)
[2332 events, purity ~0.60]
KEK, August 13, 2003
18
Flavor tagging: all remaining particles
–
high-p
lepton
(p*>1.1
GeV):
b->l
or π
–
l-
l+
b
c
s
D*+ + D0
π
net K charge b->K
medium-p lepton, b->c-> l+
*+->D0π+}
–
soft
π
b->c{D
K
hard π b->{c}π–X
* multidimensional likelihood, >99%
• wrong-tag fraction w
classify events based on
expected w (MC) - 6 bins.
(actual w -B0 mixing amplitude in data)>
• effective efficiency
= (1-2w): net (28.7±0.5)%
Flavor-specific B0
(B->D*+ l )
mixing amplitude <-> w
KEK, August 13, 2003
19
z: vertex reconstruction
Constrained to measured IP in r-
• st~1.43 ps (rms)
• Overall eff. = 87%
Validate resolution via lifetime
B0 = 1.551±0.018 ps
(PDG02: 1.542±0.016)
z
m
m-
K-
z
K-
KEK, August 13, 2003
20
Fitting t distribution
use data wherever possible to validate
unbinned maximum likelihood fit, includes
- signal root distribution (analytic)
- wrong tag fraction (const)
flavor-specific tags -> w, mixing oscillation (md)
- background: right & wrong tag (MC, parametrized)
- detector & tagging z resolution
parametrized,evt-by-evt - params from lifetime fits,
validate by measuring B, D, md
accounts for detector resolution, poorly measured tracks,
physics (e.g. charm), approximation of t=z/bgc
Checks- separate opp CP tags, verify null signal for
flavor-specific tags
KEK, August 13, 2003
21
Measurement of sin21 with 142 fb–1
5417 evts
raw
sin21= 0.733±0.057±0.028
Poor tags
Raw asymmetry
consistent with no
direct CPV.
Good tags
BELLE-CONF-0353
KEK, August 13, 2003
22
Separate CP odd/even
Raw asymmetry (all r-bins)
CP = -1 sample
sin21 = 0.73±0.06
CP = +1 sample
(B0->J/ KL)
sin21 = 0.80±0.13
BELLE-CONF-0353
KEK, August 13, 2003
23
Latest average from {cc}K
sin21 (Belle 2003, 140 fb1) =0.733±0.057±0.028
sin21 (BaBar 2002, 81 fb1) =0.741±0.067±0.033
sin21 (NEW World Av.)
=0.736±0.049
KEK, August 13, 2003
24
_
b->{c c d} decays: B->J/π0 (CP=+1)
tree
+
penguin
(relative amplitudes, strong phase not known)
B->J/π0
88.6±10.0
S=–0.72±0.42 (stat. only)
(expect -sin21 if penguin is small)
BELLE-CONF-0342
KEK, August 13, 2003
25
_
b->{c c d}: observation of D+ DW
0
B
b
d
-
d
c
c
d
W-
D()0
()
D
b
B
u,c,t
g
d
d
c
D()-
c
d
D()
Yield : 24.3 ± 6.0
Significance : 5.0
Efficiency : 7.95 %
B= (2.46 ± 0.61 ± 0.42) x 10–4
May be used in the future for time-dependent
CP asymmetry measurements
KEK, August 13, 2003
26
More time-dependent sin21 – or new physics
modes dominated by b->sqq penguins
In the absence of New Physics, S = sin21
KEK, August 13, 2003
27
Time-dependence in b->s penguins:
(Ks excluded) poor tags
good tags
+0.18
S = 0.51 ± 0.26 ± 0.05 -0.00
3rd error: due to uncertainty in CP content.
(A=-0.01±0.16±0.04)
S = 0.43±0.27±0.05
BELLE-CONF-0344
[sin21(world average) = 0.736±0.049]
KEK, August 13, 2003
28
b->s penguins: search for new physics
Theoretically cleanest
example:
68±11
Belle (78 fb–1):
S=-0.73±0.64±0.22
In the absence of New Physics, S=sin21eff = sin21
KEK, August 13, 2003
29
CP Asymmetry in B->KS
BELLE-CONF-0344
Control Sample
S=–0.09±0.26
(A=-0.15±0.29±0.07)
sin21eff = –0.96 ±0.50+0.09
–0.11
differs by
3.5s
[Current sin21(world average) = 0.736±0.049]
KEK, August 13, 2003
30
Probability & statistics
SM
sin21eff
Feldman-Cousins
treatment
including
systematic from
CP in the
background finds
this value ruled
out at 99.95% CL
or 3.5s
KEK, August 13, 2003
31
Systematic uncertainties
S
A
–
backgnd K+K–Ks, f0(980)Ks
+0.06
–0.00
+0.00
–0.08
background fraction-other
0.05
±0.04
Vertex algorithm
0.02
0.05
Total
+0.09
–0.11
±0.07
Fitting near |S|=1
±0.04
Correlation between A and S ? A=-0.15 ±0.29±0.07
If A is fixed to zero, S=-0.99±0.50
KEK, August 13, 2003
32
Observation of B0->π0π0
Gronau, London PRL 65, 3381 (1990): isospin analysis
of B->ππ to obtain 2 without time-dependence
π0π0 is most difficult step.
B=[1.7±0.6±0.2]x10–6
KEK, August 13, 2003
36
Charm: observation of new charmonium state
in B±->K±π+π–J/
bins of π+π–J/ mass
M=3871.8±0.7±0.4 MeV/c2
BELLE-CONF-0352
KEK, August 13, 2003
37
Observation of first radiative D decay
D0->g (->K+K–)
• large bg from D0->π0
(also first observation)
distinguish by ->KK
helicity angle
distribution
B=[2.60
]x10–5
+0.70 +0.15
–0.61 –0.17
BELLE-CONF-0346
KEK, August 13, 2003
38
Summary
Belle in August 2003:
• KEKB L= 1.06x1034cm–2s–1 <- @design! (1x1034cm–2s–1)
• 152M B pairs: sin21 has become a "precision" measurement
• first results on alternative probes of sin21 (or new physics!)
B->J/π0 - penguin may be small (need more data)
surprise deviation in B->Ks - 3.5s - hints of new physics?
consistency with SM in other b->sss
• developing sensitivity to 2, 3
• observations/hints in many modes, possibly CP in future
• large charm sample - new states, modes, searches for NP
Next
• 500 fb–1 by 2005
• Luminosity >@ design
• the CP challenge: just getting started - stay tuned!
KEK, August 13, 2003
39