ckm studies and new physics searches with charm two...
TRANSCRIPT
Beijing 8/05 Charm L1 Ian Shipsey 1
CKM Studies and New Physics Searches with Charm
K-
π-
e+
K+
ν
0
0
0
0
(3770)
,
D
D
D
D K eK π
ψ
ν+ − − +→ →
→Ian Shipsey, Purdue University
Two themes:1) Why Charm Physics allowsB physics to reach its full Potential2) Charm physics as a probe of New physics Beyond the Standard Model
Topical Seminar on the Frontiers of Particle Physics (Beijing)Aug. 13 – 17, 2005
Beijing 8/05 Charm L1 Ian Shipsey 2
Outline of the Lectures
Charm as a Probe of New Physics:MixingCP Violation & Rare Decays
Summary & Outlook
Overview: How Charm Physics Helps B PhysicsPrecision Quark Flavor Physics
Experiments That Contribute To Charm Physics
Precision CKM Physics: LifetimesHadronic DecaysLeptonic Decays and Decay constantsSemileptonic Decays and CKM matrix elementsTests of Unitarity
Lecture1
Lecture2 Charm is a vast
topic I will not cover all of it
Beijing 8/05 Charm L1 Ian Shipsey 3
What we knew before November ‘74:**Search for Charm
M.K. Gaillard, B.W. Lee & J.L. RosnerRev. of Mod. Phys. 47, 277 (1975)
(written before the Nov. Revolution)
What we knew 30 years later:A Cicerone for the Physics of Charm
S Bianco, F.L. Fabbri, D. Benson & I. Bigi Nuovo Cimento 26, 1 (2003) hep-ex/0309021
** Nov. 1974 marked the discovery of the J/ψ ccbar state. This “November Revolution” helped to solidify the Standard Model.
Reviews to develop a Perspective
Beijing 8/05 Charm L1 Ian Shipsey 4
Big Questions in Flavor PhysicsDynamics of flavor? Why generations?
Why a hierarchy of masses& mixings?
Origin of Baryogenesis?Sakharov’s criteria: Baryon number violationCP violation Non-equilibrium3 examples: Universe, kaons, beauty but Standard Model CP violation too small, need additional sources of CP violation
Connection between flavor physics & electroweak symmetry breaking?
Extensions of the Standard Model (ex: SUSY) contain flavor & CP violating couplings that should show up at some level in flavor physics, but precision measurements and precision theoryare required to detect the new physics
Beijing 8/05 Charm L1 Ian Shipsey 5
Testable predictions are now being made:
M(Bc)
Charm decay constant fD
Semileptonic D/B formfactors
Harder- first test July 2005
Hardest- First test July 2005
Precision theory? In 2003 a revolution in Lattice QCDBEFORE AFTER
After 30 years of struggle Lattice QCD demonstrated that it can reproducea wide range of massdifferences and decay constants for the firsttime. These were postdictions.
Easier, the 1st prediction Nov. 2004 - a success.Theory/expt Theory/expt
Beijing 8/05 Charm L1 Ian Shipsey 6
Lattice QCD Prediction Mass of the Bc
lattice agrees with CDF to 2 parts in 1,000
bc
Beijing 8/05 Charm L1 Ian Shipsey 7
Charm Physics: The ContextFlavor physics is in the “sin 2β era’ akin to precision Z. Over constrain CKM matrix with precision measurementsDiscovery potential is limited by systematic errors from non-pert. QCD
LHC may uncover strongly coupled sectors in the physicsBeyond the Standard Model. The LC will study them. Strongly coupled field theories an outstanding challengeto theory. Critical need: reliable theoretical techniques& detailed data to calibrate them
Complete definition of pert. and non-pert. QCD Goal: Calculate B, D, Y, ψ to 5% in a few years, and a few % longer term.
Charm can provide the data to test and calibrate non-pert. QCD techniques (especially true at charm threshold)
This Decade
The Future
The Lattice
Beijing 8/05 Charm L1 Ian Shipsey 8
2005
The discovery potential of B physics is limited by systematic errors from QCD:
Form factors in semileptonic (β) decay ,ub cbV V
Decay constants in B mixing ,td tsV V
D system- CKM matrix elements are tightly constrained to <1% by unitarityWork back from measurements of absolute rates for leptonic and semileptonic
D decays yielding decay constants and form factors to test QCD techniques.thatcan then be applied to the B system. + determinations of Vcs Vcd
Precision Quark Flavor Physics: charm’s role
Bd Bd
In addition as Br(B D)~100% absolute D branching ratios normalize B physics.
ρ
η
|f(q2)|2|VCKM|2
CLEO-c BESIIIprogram
Probe of new physics: Dmxing, DCPV D rare (& strong phases for CKM φ3/γ)
J/ψ probe of glueballs and exotics test of QCD Not covered here
Beijing 8/05 Charm L1 Ian Shipsey 9
Theoretical errorsdominatewidth ofbands
2004
precision QCD calculationstested with precision charmdata
theory errors of afew % on B system decay constants & semileptonicform factors
500 fb-1 @ BABAR/Belle
Precision theory + charm = large impact
+
Beijing 8/05 Charm L1 Ian Shipsey 10
Theoretical errorsdominatewidth ofbands
Now
precision QCD calculationstested with precision charmdata at threshold
theory errors of afew % on B system decay constants & semileptonicform factors
500 fb-1 @ BABAR/Belle
Precision theory + charm = large impact
+
Plot usesVub Vcbromexclusiveecaysonly
Beijing 8/05 Charm L1 Ian Shipsey 11
Lattice predicts fB/fD with a small errrorIf a precision measurement of fD existed (it does not)
Precision Lattice estimate of fB precision determination of VtdSimilarly fD/fDs checks fB/fBs precise once Bs mixing seen
Importance of measuring absolute charm leptonic branching ratios: fD & fDs Vtd & Vts
2 2 2( .) Bd td tbV Vra n fte co st ⎡ ⎤= ⎣ ⎦1.0%(expt)Winter 2005
td tbV Vif was known to 3%
would be known to ~5%Bdf
tdV1
ubV
~15% (LQCD)hep-lat/0409040
~ 100%
PDG04
c
c
D
D
ff
δ
Bd Bd
22( ) / ( .)D cdD
B D const f Vµν τ +++ → =
|fD|2
νλ
|VCKM|2
s inaccessible accessible
B d B
D Ds
f ff f+
td tsV / V
known from unitarity to 1%cdV
ρ
η ~ 16%
D+
Beijing 8/05 Charm L1 Ian Shipsey 12
|f(q2)|2
|VCKM|2
HQS
1) Measure D→π form factor in D→πlν. Tests LQCD D→π form factor calculation.2) BaBar/Belle can extract Vub using tested LQCD calc. of B→π form factor.3) Needs precise absolute Br(D →πlν) & high quality dΓ (D →πlν)/dEπ neither exist.
2 D 2 2cd2 |V | |f (q )|
qdd
π→+
Γ∝
b u l νπB
c d l νπD
Importance of Absolute Charm SemileptonicDecay Rates
When Vub is determined fromexclusive semileptonic (β) decay
β
Vub0.87 30.51(3.76 0.16 )10ubV + −
−= ±form factor
Theory Error 18%Expt. Error 4%
known from unitarity to 1%cdV
2 B 2 2ub2 |V | |f (q )|
qdd
π→+
Γ∝
~ 45%
PDG04
BB
δ
Charm semileptonic decaystest form factor predictions
( )8% precision/ /
Br B lBABAR Belle CLEO
π ν→ (World Average Summer 2005)
Beijing 8/05 Charm L1 Ian Shipsey 13
3exp(41.3 1.0 1.8 ) 10cb theoV −= ± ± ×
Importance of precision absolute charm hadronic branching ratios I
As B Factory data sets grow, & calculation of F improve dB(D Kπ)/dB(D Kπ)
dVcb/Vcb=1.2%
Vcb Zero recoil in B → D*l+ν & B → Dl+ν2* 2 2
2 ( ) ( ) cbd B D F q Vdq
νΓ→ ∝l
2 2max( ) 0.91 0.04F q q= = ±
Lattice & sum rule
ALEPH, DELPHI,L3,OPAL.BABAR/BELLE,ARGUS/CLEO/CDF
(World Average Summer 2005)
Now: several key charm branchingratios have errors between 7-26%
needs B(D Kπ)
becomessignificant
Vub/Vcb from at hadron machines requires:( )( )
-b
-b c
p ν
ν
Γ Λ →
Γ Λ →Λ
l
l
B(/\c→pKπ) poorly known: 9.7% > B >3.0% at 90% C.L
Beijing 8/05 Charm L1 Ian Shipsey 14
( )( )
*+o
o +B
DB1
D
h
h
−
−
Γ →=
Γ →
The importance of precisionabsolute Charm BRs II
HQET spin symmetry test
Test factorization with B → DDs
since D*+→π+Do is most useful mode, this compares Do/D+ absolute rates
Compare Bo→D(*)+h- and B+→D(*)oh- rates to extract color suppressed amplitudes
Need Abs BrDS
To test models of B decay charm branching ratios are important:
Beijing 8/05 Charm L1 Ian Shipsey 15
The importance of precisionabsolute Charm BRs III
Test of the Standard Model. Precision: Z →bb and Z →cc (Rb & Rc)is systematically limited by knowledge ofabsolute charm branching ratios
To understand the Higgs at LHC/LCB(H → bb) B(H → cc)precision will depend on absolute charm branching ratios
Beijing 8/05 Charm L1 Ian Shipsey 16
The Unity of Quark Flavor Physics
CharmPhysics
(Semileptonic& hadronic)
B decays
)( ulvXbBr →)( clvXbBr →
Radiative decays (future)Oscillations
dm∆
sm∆
Form Factors,F(1), duality…
,,, 2πµbc mm
ξ,, BB Bf
Moment analysis...)( γsb →
cb
ub
VV
ts
td
VV
γ
α
β
η
ρ
,..., ρπππ→B
CharmPhysics
(Leptonic)
)()(
πππKBBr
BBr→→
DKB →
Theory
),/( 0KJaCP Ψ+ other charmonium
Kaon Physics
)1(~
)(
ρηε
πυυ
−
−
K
KBr
KB(future)
Plot inspiredBy A. Stocchi
Beijing 8/05 Charm L1 Ian Shipsey 17
Charm physics:1974
Charm discovery SLAC and BNL November 1974
e+ e- multihadron enhancementJ/Ψ width << 2 MeV (beam width)
SPEAR in 1974
Goldhaber, Perl, Richter 1974
hadrons
e e+ −
µ µ+ −
Beijing 8/05 Charm L1 Ian Shipsey 18
Charm Physics: 1974
p Be e e+ −+ →
e e+ −
Broad band probe, clean final state
Beijing 8/05 Charm L1 Ian Shipsey 19
g
ge−
e+
*γ c
c
/J Ψ
glueballs?CLEO-C
γ
's are narrow, insufficient energy to decay toopen charm, (i.e. DD) or (cu)(cu)Ψ
C=-1 easy to produce with virtual photon, but decay into three gluons suppressed
+
Despite this, 88% of / are hadrronic(only the ~12% to e / ) used in sin2 measurements
J decayse µ µ β−
Ψ
Radiative / Br~6% are very useful for glueball searchesJ Ψ
T h e / a s a fro n tie rJ Ψg
g
ge−
e+
*γ
1C = −
c
c
/J Ψ
1C = −
Beijing 8/05 Charm L1 Ian Shipsey 20
The Charmonium Spectrum
L=0 L=1
OpenCharm
Beijing 8/05 Charm L1 Ian Shipsey 21
Open Charm Production at Threshold
D meson discovered 1976 Goldhaber/Trilling (LBL)At Ψ’’ = Ψ(3S) = Ψ(3770)
1L′′Ψ → =0D
0D
( 1)L+ −
0D0D
Mark IIIEvents whereboth D’sare reconstructedcircamid-1980’s
0 0'' ,D D D D+ −Ψ →
The role of the Ψ(3770) in charm physicsis analogous to the role of the Y(4S)in B physics
0
0
,D K
D K
π
π
− +
+ −
→
→
Beam constrained mass
Beijing 8/05 Charm L1 Ian Shipsey 22
Absolute Branching Ratio Measurements at the Y(4S) and ψ(3770)
#X Observed( )efficiency x #D's produced
Br D X→ =
In B decay absolute branching ratios are measured at the Y(4S)
( (4 )) (4 )Y S Y SLdt Nσ =∫ g
(4 ) ( (4 ) ) 2Y S BN Br Y S BB N→ =g
With sufficient statistics, it is possible to eliminate the Y(4S) Br assumption,and complications from the fraction of B+ B+ and B0 B0 at the Y(4S)by tagging (fully reconstructing one B in the event)
( (4 ) ) 1Br Y S BB→ =
( ) #Br B X X→ =
#X Observed( )efficiency x #Btags
Br B X→ =
Full B reconstruction has a low efficiency ε(tag) <0.7%, But is becoming a stapleas B Factory data sets grow. Similarly, charm branching ratios can be measured at the Ψ(3770)
The number of B’s produced iswell known
To measure an absolute DBr must know #D’s produced
Beijing 8/05 Charm L1 Ian Shipsey 23
Absolute Charm Branching Ratios at Threshold
e+ e−
0D
0D
K+
π−
e+
e-
#X Observed( )efficiency x #D's produced
Br D X→ =
Where the # of D’s producedIs the # of tags
tag
X
ψ(3770) → DD
The tag efficiency at theψ(3770) is about 20% as theD has large branching ratios to 2-body final states
ψ(3770) ~6 .6nb ( ~x6 Y(4S))
Meson Factory tagfigure of merit:
-1
-1
(BB) tag Ldt=500fbB Factory ~ 1Charm factory (DD) tag Ldt=3fb
σ ε
σ ε= ∫
∫
Beijing 8/05 Charm L1 Ian Shipsey 24
Open Charm Production
experiments at Ψ(3770) # D’s produced Mark III 9.6 pb-1 2 x 10 4
BES II ~30pb pb-1 6 x 104
CLEO-c (now running, goal) 3 fb-1 4 x 107
BESIII (under construction) 15 fb-1 2 x 108
1 9 8 4 1 9 8 8 2 0 0 0 2 0 0 5 2 0 1 0Y e a r
CLEO-c phys. run
BESII BESIII Construction
BESIIIEngineer & phys. run
MARKIII
0.001
0.01
0.1
1
10
100
1000
10000
J/psi psi(2S) psi(3770) Ds Pairs(4100)
Psi Family
Num
ber o
f Eve
nt (M
illio
n)
MARKIIIBESI/IICLEOCBESIII
As we will see, the Ψ(3770)is by far the best place to determineabsolute charm branching ratiosBut until BESII/CLEO-c recentlyoperated there, nobody had done so since 1984.
Beijing 8/05 Charm L1 Ian Shipsey 25
Charm Hadrons
light lightj L S= +rrrQS
r
lightSr
Lρ
light QJ j S= +rr r
HQET
( )cq
QCD
*
*
Corrections go like /
( ) ~ 142( ) ~ 46
Qm
m m D D MeVm m B B MeV
Λ
∆ = −
∆ = −
Heavy quark (Q) hadrons: Q spin decouples ∝1/mQ. Spin of Q andtotal spin j of light quark are separately conserved quantum numbers.
1/2 (degenerate doublet)lightJ j⇒ = ±
3/2lightj =
1/2lightj =
1/2lightj =
cu cd spectroscopy
The same description works forheavy baryons
L=0 L=1
Similar for cs.. Excitement in 2003 Discovery: DsJ(2317) & DsJ(2460)(Spectroscopy not covered here)
Beijing 8/05 Charm L1 Ian Shipsey 26
Fixed Target + −e e pp E791 FOCUS LEP CLEO BaBar/Belle CDF
Beam Hadron Photon 0+ − →e e Z + −e e pp K-π+ ~ 2 × 104 ~ 2 × 105 ~ 104 /expt. ~ 2 × 105 ~ few 106 ~ 106
σt ~ 40 fs ~ 40 fs ~ 100 fs ~ 140 fs ~ 160 fs ~ 50 fs
Many Experiments Produce Charm
Results used in these lectures have been obtained by the following Collaborations:
In 2003-2005:
BESII CLEO-c Beam (3770)e e+ − → ψ K-π+ ~ 2.7 ×103 ~ 5 x104
σt Not
applicable Not
applicable Note:K-π+ is # reconstructed in published analyses, not total collected.
The B Factories and CDF now have the largest charm samples.
Exceptionally low background charm samples were obtained at BESII & CLEO-c ideal formeasuring absolute charm branching ratios.
udsu
cu
W+
Κ-
π+
D
Common normalizingmode:
Beijing 8/05 Charm L1 Ian Shipsey 27
Charm Production near/at the Y(4S)
D’s move ~150 microns, but, energy of D’s is a prioriunknown and the charm, and anti-charm hadron typesare not strongly correlated
Non-resonant production to all hadrons u,d,s,c σcc≈ 1 nb OFF ON
√s→
↑σ
+ σ(B c) ≈ 1 nb 1+1 =2 nbD, Ds, Λc,Ξc…
e−
e+
*γ c
cglue
anti-charmedhadron
charmedhadron
CLEO
0
0
Reconstructing a D does not mean the charm hadron in the event is a , , , , , or any other ground state or excited charm hadron is also a possibility
S c c cD a D D+ Λ Ξ Ω
absolute charm Br’sdifficult
3 experiments,CLEO #c’s=34x106 produced BABAR,BELLE #c’=x20 CLEO and growing
Beijing 8/05 Charm L1 Ian Shipsey 28
Charm Production at the Z0
Br(Z0 cc) ~ 11%3 x 106 c’s /LEP exptD, Ds, Λc,Ξc…
<p> ~ 40 GeVExcellent reconstruction of
Charm vertices
Main LEP contributions to charm physics: (my opinion)- electro-weak measurements: Rc =(Z0 cc)/ (Z0 had)
AFBc (charge asymmetry in e+ e- cc)
See: http://lepewwg.web.cern.ch/LEPEWWG/- c-quark fragmentation function- Ds decay constant
As at 10GeV, the flavors of charm anti-charm pairs produced in Z0
decays are not correlated absolute charm Br’s are difficult
Will notdiscussIn lecture
Hep-ex9909032
D*
While LEP is no more:Giga Z (ILC at the Z pole)or parasitic has some support
Beijing 8/05 Charm L1 Ian Shipsey 29
Charm at CDF
Great potential if charm stays within the trigger bandwidth: expect107 D0→Κπ reconstructedin 2fb-1
CDF can address D mixing,DCPV and rare decays (the formeris hard and no result has yet appeared)CDF cannot measure absolute branching ratiosas charm mesons are not produced correlated
Summer 2002
Cabbibosuppressed
CDF now ~1fb-1
Charm analysis so faruses <130 pb-1
Online trigger on detached vertices in silicon: designed for beauty (also collects charm)
Beijing 8/05 Charm L1 Ian Shipsey 30
B Factories: ~120 x 106 ccbar pairs / 100 fb−1
Substantial continuum rate (B ⇒ DX rate ~equal, but softer)Hard fragmentation: higher momentum, lower combinatoricsGood with neutral daughters
Charm Factories: ~640 x 103 ccbar pairs / 100 pb−1
Modest Rates; low momenta: silicon not useful hereConstrained Kinematics: very cleanGood with neutral daughters & especially neutrinos
Hadron colliders:Very high rateDetached vertex & lepton triggersLimited modes
Fixed Target (photo production & hadro production)High rateMore modes accessible than collidersLarge boosts good for lifetimes
Summary of Experimental Issues
Beijing 8/05 Charm L1 Ian Shipsey 31
Lifetimes
Naïve spectator model for charm
Muon decay:3
52
192πµmGF
o =Γ
23
52
192 cscF
o VmGπ
=Γ0)32( Γ+=Γc
Scaling from the muon:sc
1365
107102.25.1
105.051 −− ×=×⎟
⎠⎞
⎜⎝⎛=τ
τ(D+) ~1,000 fs τ (D0) ~400 fs. Not too bad. Including baryons lifetimes vary between ~100 and 1000 fs, non-spectator processes and higher order corrections
(700 fs)
µµv
eev
,e µ x 3 colorsud
sc
, , 3e uµ ×, , 3ev v dµ ×
Beijing 8/05 Charm L1 Ian Shipsey 32
Charm Hadron Lifetimes
2 5
3192FG mµ
µ πΓ =
(2 3)charm µΓ = + Γ2 5
23 700
192F c
charm cs charmG m V fsτ
πΓ = ⇒ =
Lifetime needed to compare Br(expt) to Γ (theory)Interpreted within O.P.E.
2 4, ,( ) (1/ ) ( ) (1/ )c spect c PI WAWS c cH O m H O mΓ =Γ + +Γ +
Spectator effects (PI.WA,WS) are O(1/mc3) but phase space enhanced
Brτ
= Γ
D+
Ds+
baryons
Muon decay:
µµv
eev
Naïve spectator model:
τ(D+) ~1,000 fs τ (D0) ~400 fs.
,e udµ
Gross features of lifetime hierarchy can be explained
Beijing 8/05 Charm L1 Ian Shipsey 33
Charm Lifetimes
Charm quarks more influenced by hadronicenvironment than beauty quarks.
Errors on lifetimes are not a limiting factor in the measurement of absolute rates.
D+ 7 ‰, D0 4 ‰, Ds 8 ‰, Λc3%, Ξ 0 10%, Ξ+c 6 %, Ω c 17%
some lifetimes known as precisely as kaon lifetimes.
x10x1.3
PDG2004DominatedBy FOCUS2002 results
SELEX, FOCUS, CLEO E791 E687
PDG2004
)( c+Ξτ
1040 7 fs±
)( 0Dτ
)( +Dτ
)( sDτ
( )cτ Λ
)( 0cΞτ
)( cΩτ
501 6 fs±
410.3 1.5 fs±
442 26 fs±
200 6 fs±
1310112 fs+
−
69 12 fs±
Charm beauty
0
( ) 2.5( )DD
ττ
+
≈ 0
( ) 1.1( )BB
ττ
+
≈ PDG2004
( )psτ ( )psτLifetimes are PDG2004 except Dswhich is a PDG2004 + FOCUS average.
Beijing 8/05 Charm L1 Ian Shipsey 34
# events
Charm meson lifetime measurements are still being improved. Most recently…
Charm baryon lifetimes, less well known new measurements from SELEX τΩc= 69.3 ±14.4±8.6 fs Nucl. Phys. A755 180 (2005) (comparable to World best (FOCUS))B factories well-suited to improving these measurements
New average
Beijing 8/05 Charm L1 Ian Shipsey 35
D Nonleptonic Decays
Nonleptonic decays dominate the total rate
fsucD
fsdcD
5.15.410:)(
9.67.1042:)(
00 ±=
±=++
τ
τ 5.2/ 0 ≈+ ττ
Quarks or hadrons? ….in betweenCompare to kaons and B-mesons:
psdsK
psusK
16.07.178:)(
2012390:)(
00 ±=
±=++
τ
τ 70/ 0 ≈+ ττ
0/ 1.08 0.008τ τ+ = ±
Hadrons
quarks00
( ) : 1643 10
( ) : 1528 9
B bu fs
B bd fs
τ
τ
++ = ±
= ±
Beijing 8/05 Charm L1 Ian Shipsey 36
The lifetime hierarchy (quark diagram level)
1
%83.3~
0
≡∝
→ +−
τ
π
BRA
BRKD
74.0%11.2~
000
=
→
ABR
KD π
For the Do (D+) the two states are distinct (identical) 100% Destructive interference predicts : A = 1 – 0.74 = 0.26Measure: A = 0.54 Difference due to hadronic final state interactions
++ → π0KD
u uc s
ud
u u
c sdu
d dc s
ud
d d
c sdu
++ → π0KD
B decays : small BRs to 2- body final states (phase space)2-body decays dominate D decays (multi-body decays found to be quasi 2 body) Is the D0 D+ lifetime hierarchy understandable in terms of 2 body hadronic decays?
0D K π+ +→
0D K π− +→
0 0 0D K π→
Beijing 8/05 Charm L1 Ian Shipsey 37
…..and at the hadronic level
2/32/10
31
32)( AAKDA +=→ +−π
)32
31)( 2/32/1
000 AAKDA +−=→ π
03 / 2( ) 3A D K Aπ+ +→ =
( ) οο 79003.037.0/ 022/12/3 ±=−=±= δδδAAMany similar cases. Substantial modification of hadronic 2-body BR’s due to FSI.
Rosner hep-ph/9903543
The presence of strong phases between amplitudesis an important ingredient in mixing studies and in CP violation
δδiII eAA =
Simple factorization picture describes 2 body hadronic decays established for B’s. For charm sizeable final state interactions are the norm.
( )2 2 21/2 3/2 1/2 3/2 1/2 3/2 3/2 1/22 cosA A A A A A δ δ+ = + + −
22 2 20 0 0 01/ 2 3/ 2( ) ( )A D K A D K A Aπ π− +→ + → = +measure:
Isospindecomposition
2 203/ 2( ) 3A D K Aπ+ +→ =
:extractmeasure: :extract
find:
0D K π− +→
0 0 0D K π→0D K π+ +→
is onlyschematic
Beijing 8/05 Charm L1 Ian Shipsey 38
New measurement of Λc+ Mass
hep-ex/0507009 232 fb—1
Sub. to PRD
New Result:2286.46 ± 0.14 MeVusing 4627+264 events
Shift from PDG:2284.9 ± 0.6 MeVNot measured since 1991!Best previous CLEOII, 1134 events (6 modes)
Currently charm meson masses known to 0.5 MeV (dM/M= 0.25%) Measure modes w/ large daughter mass:
Λc+ ⇒ Λ KSK+ & Σ0 KSK+
Low Q: 178 & 101 MeV Λ KSK+
Σ0 KSK+
---- = PDG
Inv. Mass
(Minimizes bias in track momentum measurement)
Beijing 8/05 Charm L1 Ian Shipsey 39
Status of Absolute Charm Branching RatiosBrτ
= ΓMeasured very precisely
Poorly knownMode PDG04 (%) Error (%)
D+ µν 100
D0 45Ds
+ µν 0.60 ±0.14 24
2.46.52526
1.7
Do K-π+ 3.80±0.09D+ K-π+ π+ 9.2±0.6Ds
+ φπ+ 3.6±0.9Λc pK-π+ 5.0±1.3
J/ψ µ+µ− 5.88 ±0.10
Key hadronic charm decay modes used to normalizeB physics
#X Observed( )efficiency x #D's produced
Br D X→ = #D’s producedis not well known.
0.170.050.08+
−
eπ ν− + 0.230.110.39 .04+
− ±
Charm produced at B Factories/Tevatron or at dedicated FT experiments allows relative rate measurements but absolute rate measurements are hard because backgrounds are sizeable & because # D’s produced is not well known.
Backgrounds are large.
|fD|2
νλ
|VCKM|2
|f(q2)|2|VCKM|2
(In early 2004)
Beijing 8/05 Charm L1 Ian Shipsey 40
(PDG) Measurement of B(Do→ K-π+)(Dominated by Y(4S) & Z0)
• Method: • Detect D*+→π+Do, Do →K-π+
compare to:D*+→π+Do, Do → unobserved
• Problem: Systematic error due to background extrapolation
B (%) Error(%) Source3.82±0.07±0.1 3.6 CLEO3.82±0.09±0.1 3.8 ALEPH3.83±0.09 2.3 PDG
α is Ρbetweenthrust axis& slow π+
(4 of 8 intervals shown)
(CLEO & ALEPHUse same technique)
(Thrust is a measureof the direction of theprimary quark pair in the event)
* 0 ~ 6D D Q MeVπ+ +→
Beijing 8/05 Charm L1 Ian Shipsey 41
(PDG) B(D+→Κ-π+π+) Y(4S) &ψ(3770)
• Method (CLEO): Measure:
B (%) Error(%) Source9.3±0.6±0.8 10.8 CLEO9.1±1.3±0.4 14.9 MKIII
9.1±0.7 7.7 PDGFrom 9.6 pb-1 (1984)
0 0*
* 0
( )( ( ))
( )B D BDB D D
DD K
KB π π
π ππ
+
+ − + +
−
+ +
++→→ →
→
(this bootstrap method can never yield a measurement of B(D+ →K-π+ π+) more accurate than B(Do →K-π+)•Method (MKIII): ψ′′→D+D- full reconstruction, limited by size of data sample
Assume this ratio is of Strong decays is given by isospin symmetry
The determination of B(D+→φπ+), which had a 25 % error also bootstraps on B(Do →K-π+).
Beijing 8/05 Charm L1 Ian Shipsey 42
Recall: the D hadronic scale sets the B hadronic scalebecause B D ~100%, All D hadronic BRs based on D Kpi a himeasurement. This is potentially a “house of cards”
Can we do better?
Beijing 8/05 Charm L1 Ian Shipsey 43
0, ,, S
K KK K
π π π
π π π π π
+ − + −
+ − + − + −
0B D∗
0D
SD∗
( )SD
π −
γ
• Ds+ from Ds
*+ → Ds+γ is not reconstructed
• Pair D*- (→ D0 π-) & γ , assume from B0→ Ds*+D∗−
This result independent of B(Ds
+→ φπ+) :
B(B0→ Ds*+D*-) = (1.88 ± 0.09(stat) ± 0.17(syst) )%
Data sample:124 million B pairs
Signal: 7488 ± 342
* *2 2( ) ( )miss beam BD D
m E E E p p pγ γ= − − − + +
New Measurement of B(Ds+→ φ π+)
• Ds+→ φ (→K+K−) π+ fully reconstructed
B(Ds+→ φ π+) = (4.81 ± 0.52(stat) ±0.38(syst))%
B (B0 → Ds*+ D*−) x B (Ds
+→ φ π+) = (8.81± 0.86(stat)) x10-4
Divide by (A)
Β0 → Ds*+ D*− : partial reconstruction1:
2:
(A)
B(Ds+→ φ π+) = (3.6 ± 0.9)% (PDG)
*2 2
*( )s
ES beam D Dm E p p= − +
uuur uuur
Signal 247 ±19
Β0 → Ds*+ D*− : full reconstruction
Recoil mass
CLEO Similar Partial recons. Β0 → Ds*+ D*−
0( ) / ( )sD D Kφπ π+ − +Γ → Γ →
HepHep--ex/0502041ex/0502041PRD 71 091104 (2005)PRD 71 091104 (2005)
13% total error (7.5%) syst
(25%)
BIG improvement!
Beijing 8/05 Charm L1 Ian Shipsey 44
e+ e−
0D
0D
K+
π−
e+
e-
#X Observed( )efficiency x #D's produced
Br D X→ =
Where the # of D’s producedis the # of tags
tag
X
ψ(3770) → DD
Can we do better? Recall:Absolute Charm Branching
Ratios at Threshold
Beijing 8/05 Charm L1 Ian Shipsey 45
D & Do +
D+s
c+
Unique Opportunities at Charm Thresholds
• Unique event properties– Only DD not DDx produced– Can get DoDo, D+D-, DsDs ,
ΛcΛc– Probably other charmed
baryons as well (not yet measured)
• Large cross sections
σ(Ds Ds) ~ 0.5 nb
R (units of σ(µ+µ−))-+
σ(µ+µ−)= 5.4 nb at 4 GeV
ψ(3770) → DD √s ~4140 → DsDs
Beijing 8/05 Charm L1 Ian Shipsey 46
31
-1
CESR CESR-c: 12 wigglers (damping at low E) 9/03-3/04 6 wiggler Pilot Run L=4.6 10 (as expected)Continue pilot run 9/03-4/05 x4 ended April '05281 pb (3770) ( 30 MarkIII, 15 BESII) 3 year CLE
at ψ
→
×
× ×
-1
-1s s
O-c program started April 2005
Year 1 goal: 3 fb @ (3770) ( ) ( 10increase)
Year 2 goal: 3 fb @ ~ 4140 MeV D D threshold ( 130 BES)Year 3 goal: 1 billion J/ ( 20 BES)
DDψ
ψ
×
××
Minor modifications:replaced silicon with 6 layerlow mass inner drift chamber summer ’03. + B 1.5T→1.0T
CLEO III Detector CLEO-c Detector
CESR (10 GeV)CESR-c (3-4GeV)
CESR-c/CLEO-c Runplan & Status
Beijing 8/05 Charm L1 Ian Shipsey 47
BEPCII/BESIII Project Design• Two ring machine• 93 bunches each• Luminosity
1033 cm-2 s-1 @1.89GeV 6× 1032 cm-2 s-1 @1.55GeV 6× 1032 cm-2 s-1 @ 2.1GeV
• New BESIII
Status and Schedule• Most contracts signed• Linac installed 2004• Ring installed 2005• BESIII in place 2006• Commissioning
BEPCII/BESIIIbeginning of 2007
X5 CESR-c designX15 CESR-c currentperformance
Beijing 8/05 Charm L1 Ian Shipsey 48
ψ(3770) Analysis Technique
• Common to all analyses, fully reconstruct 1D “the tag” thenanalyze decay of 2nd D to extract exclusive or inclusive properties
Unique to charm: high tagging efficiency:~22% of all D’s produced are reconstructed.
Compared to <1% of B’s at the Y(4S)
CLEO-c DATA
281 pb-1 ~ 1.8 million DD pairs
e+
Dsig
e−
D tagπ −
K +
π −
π +
π +
K −
Tagging creates a single D beam of known 4-momentum
(3770),
DD K
DD K
ψ
ππ ππ
+
+ − + +
−
− + − −→
→
→
Pure DD, no additional particles (ED = Ebeam).Low multiplicity ~ 5-6 charged particles/eventGood coverage: ν reconstruction Pure JPC = 1- - (mixing, CP, strong phase)
e+e- ψ(3770) DD
Beijing 8/05 Charm L1 Ian Shipsey 49
Absolute Charm Branching Ratios at Threshold
# ( )Observed in tagged events( )detection efficiency for ( ) #D tags
KB D KK
π ππ ππ π
+ − −− + − −
+ − −→ =•
Independent ofIndependent ofL and cross L and cross sectionsection
Single tags
D candidate mass (GeV)
Double tags
D K π π+ − + +→,
DD K
K π
π
π
π+
− +
− +
−
+
−→
→
2 2| |BC beam DM E p= −
D candidate mass (GeV)
Dbeam EEE −=∆Kinematics analogous to Υ(4S) BB: identify D with
σ(MBC) ~ 1.3 MeV, x2 with π0
σ(∆E) ~ 7—10 MeV, x2 with π0
:D beamE E⇒
15120±180377±20
: 10 /D beam bc bcE E M Mδ⇒ × ↓
Beijing 8/05 Charm L1 Ian Shipsey 50
(log scale)!
2484±51(combined)
1650±42(combined)
6 D+ Modes6 D+ Modes3 D0 Modes3 D0 Modes
Single tags Double tags
Signal shape: ψ(3770) line shape, ISR, beam energy spread & momentum resolution, Bgkd: ARGUS
Global fit pioneered by Mark IIINDD & Bi’s extracted from single and double tag yields with χ2 minimization technique.
D0D0
D+D+
i i iDDN N B ε= ijjiDDij BBNN ε=
ij ji
j ij
i j ijDD
ij i j
NB
N
N NN
N
εε
εε ε
=
=
Beijing 8/05 Charm L1 Ian Shipsey 51
Stat. errors:~2.0% neutral, ~2.5% chargedσ(systematic)~ σ(statistical).
ε syst. dominatesMany systematics evaluated using data, so will shrink as √L
0.0140.0060 0
( ) 0.776 0.024( )e e D De e D D
σσ
+ − + −+−+ −
→= ±
→0.170.08( DD) =(6.39 0.10 )nbσ +
−±
D0 ModesD+ Modes
Normalized to PDG
six modes more precise than PDG.
To bepublishedin PRLhep-ex/0504003
55.8/pbwill updateby 11/05
Beijing 8/05 Charm L1 Ian Shipsey 52
Comparison with PDG 2004B (%) Error(%) Source
3.82±0.07±0.1 3.6 CLEO3.82±0.09±0.1 3.8 ALEPH3.80 ±0.09 2.4 PDG
3.91±0.08 ±0.09 3.1 CLEO-c
CLEO & ALEPHD*+→π+Do, Do →K-π+
compare to:D*+→π+Do, Do → unobserved(Q~6MeV)
THEN:
NOW:
Do →K-π+
CLEO-c as precise as anyprevious measurement
π+
thrust α
CLEO-c
Beijing 8/05 Charm L1 Ian Shipsey 53
1
0
Decay / (%)2004 1
2.4 3.1 0.5( )(1.3)7.7 3.9 0.6( )(1.5)
12.5% ( ) 3.2S
B BPDG CLEO fb
D K stat sysD K stat sysD BABAR
δ
π
π π
φπ
−
− +
+ − + +
+
→
→
→ − −
CLEO-c goal is 3/fb
CLEO-c & BES III set absolute scale for all heavy quark measurements
Submitted to PRL
B (%) Error(%) Source9.3±0.6±0.8 10.8 CLEO9.1±1.3±0.4 14.9 MKIII9.1±0.7 7.7 PDG9.52 ±0.25±0.27 3.9 CLEO-c
0 0*
* 0
( )( ( ))
( )B D BDB D D
DD K
KB π π
π ππ
+
+ − + +
−
+ +
++→→ →
→
Method (CLEO) Bootstrap:Measure:
THEN:
Assume isospin
NOW:
B(D+→Κ-π+π+)
Conclusion: the charm hadronicscale we have been using for last 10 years is approximately correct
-1
-1
(BB) tag Ldt=500fb#B tags @B Factory ~ 1#D tags @Charm actory (DD) tag Ldt=3fbF
σ ε
σ ε= ∫
∫
Mostprecise
See Ecklund Radcorr, UCSD 3/05
Beijing 8/05 Charm L1 Ian Shipsey 54
Leptonic Decays Decay Constants
M
+++ BDD s ,,
Q
q
QqV
*W
+λ
λυ
2 2 2Wq m M= =
Fixed
(PseudoscalarMeson)
the meson decay constant ƒMmeasures the probability for the Q and q to have zero separation the annihilation probability is ∝ to wave function overlap
50 ( , ) (1 ) ( , )2F
QqGM V J M u k v p sµ
µ σ γ γ= −
50 ( ) Mq Q P p if pµ µγ γ =
22 12 (0)Mf ∝ Ψ
22 12 (0)Mf M = Ψ(For a meson with two heavy quarks) (Rosner)
22 22 2 22( ) 1
8F
Qq qQ MG mM V f Mm
Mν
π− ⎛ ⎞
Γ → = −⎜ ⎟⎝ ⎠
lll
This possibly repeats Jeff Richman’s lecture
Beijing 8/05 Charm L1 Ian Shipsey 55
b c
u u
u
dW −
fπ
b
d b
dt
tW WBf Bf
τ −
u
,d sW −
Mf
vτ
γMf
q
q
M e+
e−
Decay constants are important in many processes
M
+++ BDD s ,,
Q
q
QqV
*W
+λ
λυ
This possibly repeats Jeff Richman’s lecture
Beijing 8/05 Charm L1 Ian Shipsey 56
22 22 2 22( ) 1
8F
Qq qQ MG mM V f Mm
Mν
π− ⎛ ⎞
Γ → = −⎜ ⎟⎝ ⎠
lll
Decay is forbidden as ml 0 Helicity suppression
M+λ λυ
0=J
M+λ λυ
0=J
Γ(π + → e+νe )Γ(π + → µ+νµ)
≈10−4
5
( ) : ( ) : ( )
10 :1:10s e s sD e D Dµ τν µ ν τ ν+ + + + + +
−
Γ → Γ → Γ →
≈
Γ(D+ → +ν ) ∝ |Vcd |2 ≈ (0.22)2
Γ(Ds+ → +ν ) ∝ |Vcs |2 ≈ (0.97)2
Γ(B+ → +ν ) ∝ |Vub |2 ≈ (0.003)2 This possibly repeats Jeff Richman’s lecture
Beijing 8/05 Charm L1 Ian Shipsey 57
B(eν) B(µν) B(τν) D+ 8.2 x 10-9 4.2 x 10-4 1.1 x 10-3 Ds
t 7.5 x 10-8 5.7 x 10-3 5.5 x 10-2 B+ 7.5 x 10-12 3.2 x 10-7 7.1 x 10-5
π+ 1.2 x 10-4 99.99% K+ 1.6 x 10-5 63.5%
Page 58 Estimate of the leptonic Br’s using fBs = fBs=200 MeVfDs = 260 MeV, fD = 220 MeV
At first sight it is remarkable that : B(Ds+ → µ+νµ) = ( )B K µµ ν+ +→
While (ƒM2 M ) → constant
Γ(total) ∝ Μ5 so leptonic branching ratio becomes smaller as M↑If we compare ratesinstead of branching ratios:The leptonic rate is higher for the DS than for the K+
Γ(K + → µ+νµ ) = 5.13×107 s−1
Γ(Ds+ → µ+νµ ) = 6.9 ×109 s−1
Γ(Ds+ → τ +νµ) = 6.6 ×1010s−1
This possibly repeats Jeff Richman’s lecture
Beijing 8/05 Charm L1 Ian Shipsey 58
B(D+→lν)/ τD+ : fD+2|Vcd|2
B(DS →lν)/ τDs : fDs2 |Vcs|2
* Charm meson lifetimes known 0.3-2%* 3 generation unitarityVcs, (Vcd) known to 0.1% (1.1%) fD+ fDs
D meson Decay Constants
222
222
81 ||)1()( cqD
DDFq Vf
MmmMGD
q+
+
+ −=→Γ +λ
λλ πυ
|fD|2
νλ
|VCKM|2
M+++ BDD s ,,
Q
q
QqV
*W
+λ
λυ
Both the D+ and Ds leptonic decays are targetsat charm threshold facilities
Beijing 8/05 Charm L1 Ian Shipsey 59
fD+from Absolute Br(D+ → µ+ν)
1 additional track (consistent with a muon)Compute missing mass2: peaks at 0 for signal
Tag D fully reconstructed
Mark III PRL 60, 1375 (1988)
~9pb-1 2390 tags
4
11.1 12953 119
( ) 10 MeVMkIII 7.2 290BESII 12.2 0.11 371 25
DB D fµν+ −
+− −
→ ×< <
± ±
~33pb-1
5321 tags
S=3 B=0.33
BES II hep-ex/0410050
pµ
MKIII
BESII2 2 2( ) ( )beam DtagMM E E P Pµ µ+= − − − −
uuuuur uur
MM2
Beijing 8/05 Charm L1 Ian Shipsey 60
fD+from Absolute Br(D+ → µ+ν)
2 2 2( ) ( )beam DtagMM E E P Pµ µ+= − − − −
uuuuur uur 1 track µ consistent no showersD+ → µ+νD-Hadronictag
CLEO-cProposal:
2 2MM (GeV )
0D K π+ +→
PRD 70, 112004PRD 70, 112004
MC1fb-1
X20 DATA
Beijing 8/05 Charm L1 Ian Shipsey 61
fD+from Absolute Br(D+ → µ+ν)
2 2 2( ) ( )beam DtagMM E E P Pµ µ+= − − − −
uuuuur uur 1 track µ consistent no showers
D µ ν+ +→
~57 pb-1
D+ → µ+νD-Hadronictag
8 signal candidates
0D K π+ +→
2 2MM (GeV )
CLEO-cProposal:
MC1fb-1
X20 DATA
2 2MM (GeV )
28,651 tags
PRD 70, 112004PRD 70, 112004
¼ the data collected so far
Beijing 8/05 Charm L1 Ian Shipsey 62
K+π− π− K+π− π− π0
Ksπ−
Ksπ−π0
Ksπ+π− π−
K+K− π−
15
8K
fully recon
structed D
-
From 2
81
pb-1at Ψ
(37
70
)
Preliminary
Fully reconstructed D- tagAnd now with a larger data sample
All the data collected so far
Beijing 8/05 Charm L1 Ian Shipsey 63
• MC Expectations from 1.7 fb-1, 6 x data
0 0.25 0.50
200
400
600
MM (GeV )2 2
µ ν signal+
π π + ο
τ ν, τ π ν + +
sum
peak fromK π
ο +
100
50
0
50 signal events
D+ → µ+ν from CLEO-c Data281 pb-1 at ψ(3770)
D+→π+K0
D+→µ+ν
MM (GeV )
Num
ber
of E
vent
s/0.
01 G
eV2
2 2
-0.05 0 0.05
5
10
15
0 0.25 0.50
20
40
60
80
100
120
D+→π+K0
D+→µ+ν
Preliminary
Beijing 8/05 Charm L1 Ian Shipsey 64
Expt/LQCD consistent at 45% CLNow: LQCD error ~8%CLEO-c error 8%CLEO <5% within a year
Need latest LQCD predictions to few %by summer 2006 f D+ & 2007 f Ds
-1with 3fb : to 2.3%
to 1.9% @ s ~ 4140D
Ds
f
f MeV+
Comparison to the lattice
BES III may make definitive measurements
Beijing 8/05 Charm L1 Ian Shipsey 65
Extraction of the decay constant
0.225 0.0023(KTeV)
( ) 1.040 0.007 ps (PDG)
cd usV V
Dτ +
= = ±
= ±
Beijing 8/05 Charm L1 Ian Shipsey 66
COMPARISON
0
5
10
15
20
25
30
Err
or
(%)
DfsDf
)KD(Br ππ→+
)D(Br S φπ→
)KD(Br π→0
PDG04
Compare B factories & Charm FactoriesCompare B factories & Charm Factories
CFactory3 fb-1
Statistics limited
BFactory 400 fb-1
Systematics & Background limited
%Error
New BABARMeasurement
Beijing 8/05 Charm L1 Ian Shipsey 67
2004
Precision theory + charm decay constants = large impact
Beijing 8/05 Charm L1 Ian Shipsey 68
Semileptonic Decays: formalism
Q W-
q
q Hadron
νi
Qf
lVQiQf
( )22 2 2 2D X D X X Dq p p m m E mµ µ= − = + −
2 2( ) ( ) ( )( ) ( )( )X D D X D XP P J D P f q P P f q P Pµ µ µ+ −= + + −
Nuclear β decay Vud, K→πeν VusD →Kev Vcs b →clν Vcb b →ulν Vub
•Weak current is understood. The work is in the hadronic matrix element expressed in terms of form-factors. for D→Pseudoscalar λ+ ν
•For λ = e, contribution of f−(q2)→0,
Traditional way to determine magnitudes of CKM elements
•Kinematics:
=X
The form factor measures the probability that the final state mesonwill be produced
Beijing 8/05 Charm L1 Ian Shipsey 69
The hadronic current
B
b
q
c
q
−W−l
λυ
(*)D
D
c
q
c
q
+W+l
λυ
(*)K
Beijing 8/05 Charm L1 Ian Shipsey 70
Two Key Kinematic Configurations
b
q
c
q−l v2max
2 qq =
c
q 2min
2 qq =
v
−l
c
q
s
q+l υ
2max
2 qq =
s
q 2min
2 qq =
υ
−l
υ
Form factormaximum
Form factorminimum
Beijing 8/05 Charm L1 Ian Shipsey 71
Goals for a threshold Semileptonic Decay
( )2
3
23
22 24
( )cq PV Pf
D Pdq
qd ν
π +
Γ →=
l
Contraction with leptonic current
Quark models, HQET, Lattice & other methods have all been invoked to calculate form factor absolute normalizations. These calculations have been done
mostly at q2 =0 or q2 =q2max. (i..e w=1, just like F in Vcb in B →D* lν)
For PS PS decays data is mostly at q2 =0. Need to interpolate. The q2 dependenceis intrinsically non-perturbative. Models make phenomenological guesses about the shapes of the form factors as a function of q2..
To find Vcs & Vcd need f from theory at one fixed q2 point.
The lattice calculates rates as a function of q2. Charm semileptonic decays provide a powerful test of the lattice predictions. Once validated, the lattice can be used with confidence in the extraction of CKM matrix elements in other D (& also B) semileptonic decays .
0 0.5 1 1.5 2 2.5 3
3P
2q
cleanest theory
highest rate
D lπ ν→
22 2
pole
(0)( )1 /
ff qq m
++ =
−
Beijing 8/05 Charm L1 Ian Shipsey 72
K-
π-
e+
K+
ν
Semileptonic Decays
Tagging creates a single D beam of known 4-momentum
Semileptonic decays are reconstructed with nokinematic ambiguity
Hadronic Tags: 32K D+ 60K D0
0miss missU E pº - =rss
ν+−→ eKD0
Eve
nts
/ (
10 M
eV )
(~1300 events)
U = Emiss– |Pmiss| (GeV)
D+ results are new, D0 update ICHEP
Accepted for Publication in PRLAugust 12 2005Hepex/0506053 &0506052
CLEO-c 1/5 data
0
0
0
0
(3770)
,
D
D
D
D K eK π
ψ
ν+ − − +→ →
→
Beijing 8/05 Charm L1 Ian Shipsey 73
More Cabibbo allowed modes
0 *
* 0 D K e
K Knp
- +
- -
®®
U = Emiss– |Pmiss| (GeV)
Eve
nts
/ (
10 M
eV )
Eve
nts
/ (
10 M
eV )
Eve
nts
/ (
10 M
eV )
Eve
nts
/ (
10 M
eV )
ν++ → eKD 0
(~550 events)
U = Emiss– |Pmiss| (GeV)
U = Emiss– |Pmiss| (GeV)
c s Cabibbo Favored 57 pb-1 Data
(~90 events)
*0
*0
D K e
K K
n
p
+ +
- +
®
®
(~420 events)
Historically Cabibbo allowed modes: provide a significantbackground to Cabibbosuppressed modes, makingthe latter particularlychallenging…..
Beijing 8/05 Charm L1 Ian Shipsey 74
Cabibbo suppressed modes0D eπ ν− +→
0D eπ ν− +→
U = Emiss– |Pmiss| (GeV)
Eve
nts
/ (
10 M
eV )
(~110 events)
Eve
nts
/ (
10 M
eV ) νπ ++ → eD 0
(~65 events)
U = Emiss– |Pmiss| (GeV)
57 pb-1 Data
Compare to:state of the art measurementat 10 GeV (CLEO III)PRL 94, 11802
Note:kinematicseparation.
∆m
S/N ~40/1
S/N ~1/3
* 0
0
( ) (:
)
s
s
Tag with
obs
D D
D
m merva
mble
π
π ν
π π π
+
+ −
→
→
∆ = −
l
l l
Beijing 8/05 Charm L1 Ian Shipsey 75
*0D K e n+ +®0
*0
( )( )
D eD K e
r nn
+ +
+ +
G ®G ®
2(GeV/c )mpp
U = Emiss– |Pmiss| (GeV)
0D er n+ +®
(~30 events)
U = Emiss– |Pmiss| (GeV)
0 *D K e n- +®
(~30 events)
D ew n+ +®(8 events)
(5σ)
57 pb-1 Data
1st Observation.1st Observation.
E791 PLB 397325(1997)Relative rate:
S/N ~1/2S/N ~15/1
0D er n- +®
More Cabibbo supressed modesOnly measurementuntl now
*
( ) ( )/( ) ( )B e D eB K D Ke
r n r nn
+ +
+
G ® G ®G ® G ®l l
Useful for Grinstein’sDouble ratio Vub2/ Vcb2
Beijing 8/05 Charm L1 Ian Shipsey 76
0102030405060708090
100
1 2 3 4 5 6 7 8 9 10 11
Decay modes
Err
or
(%)
PDG
CLEO-c3fb-1%
BB
δNormalized to PDG
CLEO-c already all modes more precise than PDG.
Full CLEO-c data set (later BESIII) will make significant improvements in the precision with which each absolute charm semileptonic branching ratio is known
0
0*
0
5 :
6 :7 :
D K e
D K eD e
n
np n
+ +
+ +
+ +
®
®®
0
0
*0
8 :9 :
10 :
11 :
s
s
s
D eD K e
D K e
D e
r nn
n
f n
+ +
+
+
+
®®
®
®
0
0 *
0
0
1 :2 :3 :4 :
D K eD K eD eD e
nn
p nr n
- +
- +
- +
- +
®®®®
Preliminary Results Similar analysis by BES II
55.8/pbwill updateby 11/05
Accepted for Publication in PRLAugust 12 2005Hepex/0506053 &0506052
Beijing 8/05 Charm L1 Ian Shipsey 77
From the semileptonic measurements…
0.1)()(
0
0
=→Γ
→Γ++
+−
νν
eKDeKD
2.04.1)()(
0
0
±=→Γ
→Γ++
+−
νν
eKDeKD
0
0
0
0
( ) 1.00 0.05( ) 0.04( ) CLEO-c( )( ) 1.08 0.22( ) 0.07( ) BES-II( )
D K e stat sysD K eD K e stat sysD K e
νννν
− +
+ +
− +
+ +
Γ →= ± ±
Γ →
Γ →= ± ±
Γ →
0
( ) (15.1 0.5 0.5)%
( ) (6.1 0.2 0.2)%xcl
exclB D Xe
B D Xeν
ν
+
→ = ±
±
±
→ = ±
∑∑
Long standing puzzle in D semileptonic decays
Isospin requires
PDG gives
CLEO-c & BES IIsolve the problem
Is anything missing?sum up all the individualdecay modes
0
PDG (11%)PDG (( ) (6.87 0.28)%
( ) (17.2 1.94
)%
%.1 )
B DB D
Xe Xe+
+
+
→ =
→
±
= ±Compare to inclusive rate form PDG (not very precise)room for additional modes
Beijing 8/05 Charm L1 Ian Shipsey 78
Question: what could the additional modes be?
Beijing 8/05 Charm L1 Ian Shipsey 79
2q
2qdd
Γ
LQCD : shape correct:
Lattice comparison – the shape
D0→π-e+ν
22
2
2
2
~ 0.4GeV B Factoryq
~ 0.1 GeV FOCUS~ 0.025 GeV CLEO-c
qδ
Improved precision soon:The threshold advantage 1) Low background crucial forπ final state2) neutrino direction known
CLEO-c 1/5 data FOCUS all data
S/N ~40/1 S/N ~1/2.5
S/N ~6/1S/N >300/1
Beijing 8/05 Charm L1 Ian Shipsey 80
Lattice comparison: the normalization If we assume that the lattice shape is OK ⇒we can use branching fraction measurements to validate the normalization
LQCD : normalization agrees with data!
Total lattice Br agrees with experiment for PDG: Vcs, Vcd
Beijing 8/05 Charm L1 Ian Shipsey 81
Vcs and Vcd
• Assuming that shape and normalization of the form factors are OK:
Using isospin averaged widths
i.e. combining D0 and D+
Vcs =0.953±0.017(exp) ±0.067(th)Vcd =0.214±0.009(exp) ±0.016(th)
LQCD errors dominate expt.errors
Need 1fb-1 at ψ(3770) (soon) and LQCD to few % within 1-2 years
Artuso LP03not official CLEO-c
Expt. errors Vcs ~2%Vcd~4%
Result is theory limited
Beijing 8/05 Charm L1 Ian Shipsey 82
U = Emiss - Pmiss
CLEO-c: (& BES III) PS → PS & PS → V absolute form factor magnitudes & slopes to a few%. Stringent test of theory! (Unique)
D0 →πlν
D0 →Klν πdpdΓ
(GeV/c)pπ
CLEO-cMC
Lattice QCD
πdpdΓ
D0→ πlνD0 →πlν
CLEO-cMC
(GeV/c)pπ
1fb-1
Testing the Lattice with (semi)leptonic Charm Decays
1fb-1
This is what the data will look like soon (x5 existing data)
Beijing 8/05 Charm L1 Ian Shipsey 83
With 1fb-1 @ ψ(3770) Rlskexp ~3% uncertainty
10% uncertainty
With 1fb-1 @ 4140 Γ(Ds→lν) /Γ(Ds→ηlν) independent of Vcs Rlsk
exp ~ 3% uncertainty
δVcs /Vcs = 1.6% (now ~7%*) δVcd /Vcd = 1.7% (now: 5.4%)D eπ υ+→D Ke υ+→
Then Tested lattice to calc. B→πlv is available for precise exclusive Vub* 3 flavor unquenched LQCD + D Kev (last sldie) (note W decays at LEP in hadronic to leptonic 1.3%)
LatticeExperiment
Artuso LP03
Beijing 8/05 Charm L1 Ian Shipsey 84
Unitarity Tests Using Charm
2nd row: |Vcd|2 + |Vcs|2 + |Vcb|2 = 1 ??CLEO –c: improves precision (if theory D →K/πlν good to few %)& 1st column: |Vud|2 + |Vcd|2 + |Vtd|2 = 1 ?? with similarprecision to 1st row
⎟⎟⎟
⎠
⎞
⎜⎜⎜
⎝
⎛=
⎟⎟⎟
⎠
⎞
⎜⎜⎜
⎝
⎛
=⎟⎟⎟
⎠
⎞
⎜⎜⎜
⎝
⎛
bsd
VVV
VVVVVV
bsd
tbtstd
cbcscd
ubusud
'''
Compare ratio of long sides to 1.3%
|VubVcb*||VudVcd*|
|VusVcs*|
uc*=0
uc*
(3fb-1)
Beijing 8/05 Charm L1 Ian Shipsey 85
Inclusive D0/D+ Absolute Semileptonic Branching Fractions (CLEO-c)
sc
, , 3e uµ ×, , 3ev v dµ ×
Naïve spectator model:
DD00/D/D++ lifetime difference due to hadronic width (Pauli int. DD++))
0 0( ) / ( ) ( ) / ( )D D B D e X B D e Xτ τ+ + + += → →( ) ~ 20%B D e X+→
0PDG ( ) (6.87 0.28)%PDG ( ) (17.2 1.9)%
B D e XB D e X
+
+ +
→ = ±
→ = ±
Now: precision measurements of Γ(D→Xlν) and Γ(Ds→Xlν) needed to constrain background to Vub in B inclusive semileptonic decay
But gluon emission enhances hadronic rate( ) ~ 7 %B D e X+→
Historically: B(D→Xlν) important to interpret charm lifetime hierachy
Isospin symmetry requires Γ(D+→Xlν)= Γ(D0→Xlν)
DD00 DD++
dB/B=11%dB/B=11%
Beijing 8/05 Charm L1 Ian Shipsey 86
D+ → K-π+ π+
D0→K-π+
3σ 3σ3σ 3σ
Β(D→Xlν) Analysis Technique 1: Tag Modes
:identify D with
E & M
D beam
bc
E E⇒
∆
Dbeam EEE −=∆
Control regionsevaluate smallbackgrounds in Dtagging
Only golden modes+
D0→K-π+ &
D+ → K-π+ π+
Kinematics analogous to Υ(4S) BB:
2 2| |b c b ea m DM E p= −
74K D0 tags47K D+ tags
Details (probably will not discuss)
Beijing 8/05 Charm L1 Ian Shipsey 87
( )(2 )e ta g eD DY B D e X N ε ε+= →
( ) ( 2 )D ta gD DN ta g N ε=
Measure e spectrum in tagged events:
Correct for ε : Independent of L and NDD/( )( )
e e
D
YB D e XN tag
ε+→ =
Measure raw spectra / / in tagged events:16, 50 MeV bins (0.2<p<1.0 GeV) Right sign: ( / / )Wrong sign: ( ( / / )an unfolding procedure gives true populations
e K
K D D e K XK D D e K X
π
π π π
π π π
± ± ±
+ − − − + + + +
+ − − − + − − −
← →
← →
( ) ( ) ( )
e e eK
e K
Ke K
Y p A p N pe f f e
f fK f f K
π
π π π
π
επ ε π
ε
=
⎛ ⎞ ⎛ ⎞⎛ ⎞⎜ ⎟ ⎜ ⎟⎜ ⎟=⎜ ⎟ ⎜ ⎟⎜ ⎟⎜ ⎟ ⎜ ⎟⎜ ⎟⎝ ⎠ ⎝ ⎠⎝ ⎠
is probabilty a is identified as an e are efficiencies including PIDef π π
ε
+
0S
& from data:
ee for e, D for K &K for low mom.
ijf
K
ε
γ π π π
ππ π
− + +→
→
Β(D→Xlν) Analysis Technique 2: Electron spectrum
Subtract charge symmetric background (wrong sign) & D background (∆E sidebands) Extrapolate to p=0 GeV
Tally number of tagsOverview
MoreDetail
Measured True
Details (probably will not discuss)
Beijing 8/05 Charm L1 Ian Shipsey 88
RS
D+ →Κ- π+π+
D-→ e-X ν
D+ →Κ- π+π+
D-→ e+X ν
WS
Electron right-sign Electron Wrong-signTag Sideband
Raw measured spectra for D+
Measures charge symmetric backgroundmostly π0 Dalitz decays & γ conversions
Measures false tag background
Similar distributionsfor D0
Right sign Wrongt sign
RS 8.3K
WS 228
RS 127
WS 10
Details (probably will not discuss)
Beijing 8/05 Charm L1 Ian Shipsey 89
Corrected Spectra & Results
D+
D0
Momentum (GeV)
DΓ(D
e+ X
ν)/DDp
(p
s-1
GeV
-1 )
0
0.1
0.2
0.3
0.2 0.4 0.6 0.8 1
0
( ) (1 5 .1 0 .5 0 .5)%
( ) (6 .1 0 .2 0 .2 )%xc l
exc lB D X e
B D X eν
ν
+
→ = ±
±
±
→ = ±
∑∑
Unfolded background subtracted efficiency corrected lab spectrum – no FSR correction
0
(2.5%)(( ) (6.45 0.17 0
) (16.19 0.2(3.5%
0 0.36)1 )%
)%. 5
B DB D Xe
Xeνν
+
→ = ± ±
→ = ± ±
Sys errors EID 2%, Hadron ID 1% FSR 1% p 0 1%
Compareto CLEO-cexcl.
Incl & excl. consistent, some room for additionalexclusive modes
-1
0
0 1
PDG (11%)PDG (4.1%)
( ) (17.2 1.9)%
(( ) (6.87 0.28)%
( )) (0.1557 0.0
(0.1572 0.0041 0.0019 0.0035)p
37)pss
0
B D e X
D eB D e X
D e XX
+ +
+ +
+
+ −
→ = ±
Γ → = ±
→ = ±
± ±
±
Γ → =NOW
THEN
Fit low p data to polynomial to extrapolate p=0,( 8% has p<200 MeV) dB/B
0
0
C L E O -c ( ) / ( ) 2 .5 1 0 .0 4P D G ( ) / ( ) 2 .5 3 0 .0 2
B D B DD Dτ τ
+
+
= ±
= ±Excellentagreement!
preliminary
’79 DELCO
600 evts
10,100evts
Beijing 8/05 Charm L1 Ian Shipsey 90
Additional Slides
Beijing 8/05 Charm L1 Ian Shipsey 91
CLEO-c INPUTS TO CKM ANGLE φ3 /γ
See talks by Asner, Fleischer, Grossman, Petrov CKM2005
1. Gronau-London-Wyler Method– B- →DCPK-
– Statistics Limited so far– D mixing parameters can alter γ
2. Atwood-Dunietz-Soni Method– B- →D K-
– Requires very large Ldt – CLEO-c measure rD & cos δD
3. Dalitz plot Method– B- → DK-, D→ Ksπ+π-, π+π-π0, KsK+K-
– limited by uncertainty due to Dalitz plot model– CLEO-c quantum correlated D’s reduce model dependence
Beijing 8/05 Charm L1 Ian Shipsey 92
0 0D DK
mixingπ+ −
→
→
•BFac/Tevatron: large DCSD interferes with signal unknown phase δ
(CLEO-c measures δ same for ADS)
D CP violation, mixing & rare decays sensitive to new physics at high mass scales through intermediate particles entering loops.
udsu
cu
W+
Κ-
π+ usdu
cu
W+
D0 π-
Κ+CF DCSD
CLEO-c Probes of New Physics (+ input to φ3 /γ)
Μixing: ψ(3770)→DD(C = -1) Coherence simplifies study DCSDUnmixed: D0→ K-π+ D0→ K+π-
mixing: D0→ K-π+ D0→ D0 →K-π+ (+Klv Klv)Sensitivity: 10-4 current limit: 10-3
y’= ycosδ-xsinδx’= xcosδ+ysinδ
D0
Beijing 8/05 Charm L1 Ian Shipsey 93
ψ(3770)L=1 C=-1 CP anti-correlated
–cos δD ~ ±10%
CLEO-c Measurement of Strong Phase
Dr
Beijing 8/05 Charm L1 Ian Shipsey 94
•CP violating asymmetries•Unique:L=1,C=-1 CP tag one side, opposite side same CP• CP=±1 ← ψ(3770) → CP=±1 = CPV Sensitivity (two body final states) Acp < 0.01•CP tagging allows many searches for CPV in Dalitz plots sensitive to amplitudes
Rare charm decays. Sensitivity: 10-6
CLEO-c Probes of New Physics (+ input to φ3 /γ)
2. Dalitz plot Method– B- → DK-, D→ Ksπ+π-
– limited by uncertainty due to Dalitz plot model currently 10-110
– CLEO-c CP tagged D’s reduce Dalitz plot model dependence, compare CP+ DD Ksπ+π-,CP- DD Ksπ+π-, & untagged D Ksπ+π-,Extrapolation of Belle study CLEO-c 285 pb-1/6fb-1
(3770 + DsDs) reduce model error to ~70, ~10
CLEO-c
D Ksπ+π-
Beijing 8/05 Charm L1 Ian Shipsey 95
Theoretical errorsdominatewidth ofbands
2004
Today: the potential of quark flavor physics to discover new physics is limited by theory errors
precision QCD calculationstested with precision charmdata from CLEO-c
theory errors of afew % on B system decay constants & semileptonicform factors
+500 fb-1 @ BABAR/Belle
Beijing 8/05 Charm L1 Ian Shipsey 96
2004
precision QCD calculationstested with precision charmdata from CLEO-c
theory errors of afew % on B system decay constants & semileptonicform factors
500 fb-1 @ BABAR/Belle
Precision theory + charm = dramatic increase in the potential of quark flavor physics to discover new physics
+
Theoretical errorsdominatewidth ofbands
Plot usesVub, Vcbfrom exclusivedecays only
Beijing 8/05 Charm L1 Ian Shipsey 97
Charm at CDF
The hadronic B trigger a major milestone~150 VME boards find and fit tracks in Silicon, offline accuracy in a 15µs pipeline
Secondary vertex level 2 trigger|D|> 100 µm (2 body)
Summer 2002
Beijing 8/05 Charm L1 Ian Shipsey 98
Beijing 8/05 Charm L1 Ian Shipsey 99
Differential Semileptonic Rates P Pev
D0→K-e+ν D0→π-e+ν
2 2 2cs,d2 |V | |f (q )|
qdd +
Γ∝
Raw q2 No efficiency correction. Results by summer hoped for
2qdd
Γ
2q
CLEO PRL 94/, 11802
(PRL 94, 011601 (2005)
STATUS2005 (CLEO IIIFOCUS BES II)
LQCD : shape & rate correct: precision~10%
22
2
2
~ 0.4GeV CLEO q
~ 0.025 GeV CLEO-c
qδ
(1) No kinematic ambiguity (2) rest frame of D q2 resolution x 16 better
πeν
Keν
CLEO-c
2
ddq
Γ
shaperate
First unquenched LQCD for D→π/K e ν
Beijing 8/05 Charm L1 Ian Shipsey 100
Beijing 8/05 Charm L1 Ian Shipsey 101
Lattice
There is remarkably little difference between the shapes of pole, modfiied pole and lattice
Beijing 8/05 Charm L1 Ian Shipsey 102
Pole mass comparison, good consistency
*
1.93 0.05 0.031.89 0.05 0.0351.91 0.04
( ) 2.12s
Expt pole massFOCUSCLEO IIIWorld Avgm D
± ±± ±± 5.1 σ below the spectroscopic mass
Beijing 8/05 Charm L1 Ian Shipsey 103
Differential Semileptonic Rates P Vev
CLEO-c*0
*0
D K e
K K
n
p
+ +
- +
®
®
Preliminary, first lookat the data, also D ρeνno results at this time
The J=1 vector particle results in more complicated decay angular distributions
Beijing 8/05 Charm L1 Ian Shipsey 104
Theory & CharmSpectator model,
All lifetimes equal Factor of ~15x spread in lifetimes (0.3 for beauty)Large final-state interactions in charm
OPEs, lifetimes qualitative
HQET: Useful for spectroscopy, but larger (1/m)N corrections,
& no heavy-to-heavy decays like b ⇒ c l ν
Lattice QCD:Useful for charm & bottom
(but charmonia more relativistic…)Good opportunities to test in charm sector ( c.f. CLEO-c BESIII program )Only approach that is systematically improvable & testable to a few %
B vs. D decay & Penguin example:B decay: integrate out top ⇒ local 4-quark operator D decay: dominant strange light ⇒ not short-distance!
Will concentrate on the lattice in this lecture
Beijing 8/05 Charm L1 Ian Shipsey 105
SummaryCLEO-c 1st data (6 wigglers) summer 2004, updated + new results today.Detector performing well, data is excellent & is well understood.fD+ is now known to 22% (was 100%) D0→π-e+ν now 14% (was 45%) 1st observations: D0 →ρ e+ν, D+ →ω e+νMost precise measurements of D+ →K-π+ π+
x4 data by April. Three year program starts 4/05several % precision in all key charm quantities, + probe for new physics & glueballs
Lattice goal to calculate in D,B,Y,ψ to 5% in a few years and a few % longer term CLEO-c about to provide few % tests of lattice calculations (& other QCD techniques)in D system & in onia, quantifying the accuracy for application of LQCD to the B system
BABAR/Belle/CDF/D0 (later LHC-b/ATLAS/CMS SuperKEKb) + theorycan reach few % precision for Vtd, Vts, γ and exclusive Vub,Vcb.
CLEO-c maximizes the sensitivity of the worldwide heavy quark flavor physics program to new physics this decade and paves the way for understanding beyond the Standard Model physics at the LHC/LC in the next decade(s).
Beijing 8/05 Charm L1 Ian Shipsey 106
The importance of precisionabsolute Charm BRs IV
b →bb
-b c +cud+ccsν→ l
QCD prediction:(Neubert & Sachrajda, Nucl. Phys. B483,339 1997)
(Plot from Parodi at HF9)
areanti-correlatedSL cBR n
Theory can accommodate present values butexperimental errors are large test becomesmore incisive if abs charm BR’s were known precisely
As a function of the effective charm quark mass & scaleat which QCD corrections are evaluated
( ), is low compared to theory A possible explanation is that the c quark effective mass is low large decay rate for b ( )
( )is negatively correlated to The simultaneous mea
SL
c c c SL
BR B b cccs d
n n n Br
ν= →
→ →→ = +
l
csurement of and n can clarify the theoretical pictureSLBr
To measure nc need absolute charm Br’s
Beijing 8/05 Charm L1 Ian Shipsey 107
Belle predict excellent q2 resolution and low bkgd result expected soon
Beijing 8/05 Charm L1 Ian Shipsey 108
Beijing 8/05 Charm L1 Ian Shipsey 109
Charm vs. Beautyλ = sinθC ~ 0.22 Cabibbo quark-mixing angle
Decay rate: O(1) O(λ4)“Cabibbo”-suppression of b decay compensates Γ ~ m5 behaviorLong B life inspired Wolfenstein’s famous CKM parameterization…Advent of silicon revolutionizes b physics, as well as c
Mixing Ampl. : O(λ2) O(λ6) x f( mt )Charm decays too fast to mixB mixing enhanced by top mass ( actually, Bs enhanced too much! )
Fully Reconstruct: 20% < 0.1%At factories producing meson pairs: Can use full-recon. “Tag”:Study rest of event after full reconstruction of one hadronic decay-- get direction of other meson-- lower background (less combinatorics, reject continuum)
Similar numbers of single tags for charm & beauty, but:-- useful # of Double Tags (reconstruct both D’s) for charm !
since NDouble ~ (εB)2 : MUCH smaller for B’s than D’s
Very different but intimately linked
Beijing 8/05 Charm L1 Ian Shipsey 110
Christine Davies EPS2005