Searching for New Physics at a Bfactory: Why, Where and HowMaurizio Pierini 1 SLAC experimental Seminar 2/14/06 Searching for New Physics at a Bfactory: Why, Where and How Maurizio

1Maurizio Pierini

SLAC experimental Seminar 2/14/06

Searching for New Physics

at a Bfactory: Why, Where and How

Maurizio PieriniUniversity of Wisconsin,

Madison

2Maurizio Pierini


Outline

Why: We have several reasons to consider

the Standard Model as an effective theory. But switching on new physics we generate the flavor problem

Where: all FCNC processes are loop induced in

the SM. New Physics can contribute as virtual states. We can learn a lot on where this is possible fromthe overconstraint of the CKM matrix. UTfit in and beyond the Standard Model

How: two possible scenarios from present bounds:

Minimal Flavor Violation: interplay between K and B physics at high luminosity facilitiesLarge Flavor Violation in 23 sector:(time dependent and time integrated) studies of bs decays

natural in many flavour models, given the strong

breaking of family SU(3)

3Maurizio Pierini


Why New Physics at all?

Hierarchy problem: Renormalization of Higgs mass gives contributions ~Plank, but we want it to be few hundreds GeV. We don't want finetuning, so we need something (a symmetry?) that protects mH

Dark Matter: Standard Model does not provide a satisfactory candidate for dark matter ('s are not enough)

Grand Unification: coupling constants of weak, strong and electromagnetic interactions do not cross in a single point if we assume the Standard Model. They do in NPscenarios (SUSY)

4Maurizio Pierini


Going Beyond: Flavor Problem

EW scale

NP contribution to EW precision, FCNC processes, CPV, etc.

The SM works beautifully up to a few hundredGeV's, but it must be an effective theory validup to a scale Mplanck

L(MW)=2H†H+(H†H)2+Lgauge+LYukawa+L5/+L6/2SM SM

The new contributions in general introduce new sources of CP violation and flavor mixing. The consistency of the Standard Model becomes a puzzle in this framework.We should see some discrepancy!!!!

NP contribution to g2, bs, etc

5Maurizio Pierini


CP Violation in the SM

V CKM=Vud Vus VubVcd Vcs VcbVtd Vts Vtb ≃ 1−

2

2 A3 − i

− 1−2

2A2

A31− − i − A2 1

The mass eigenstates are not the eigenstates of weak

interaction. This feature of the SM Hamiltonian produces the

(unitary) mixing matrix VCKM

Since there are three families of quark, there is onephase that cannot be reabsorbed into the definition of the quark fields. This phase allows CP violation in the SM.

u

d

c t

s b

5

6Maurizio Pierini


Testing SM: Unitarity Triangle

The unitarity of the matrix can be visualized as a triangle in a complex plane

normalized:

+i

several experimentalobservables depend

on ρ and η:

normalized:

1 +i

7Maurizio Pierini


|Vub|

|Vcb|

Unitarity Triangle Analysis

Adding the “Classic” constraints...

K

∆md εK

∆md

∆ms

http://www.utfit.org/

M.Bona, M.Ciuchini, E.Franco, V.Lubicz,

G.Martinelli, F.Parodi, M.P.,

P.Roudeau, C.Schiavi,

L.Silvestrini, A.Stocchi

hepph/0501199

8Maurizio Pierini


sin2β

...to the new bounds from B factories

γ α

cos2β

β from D0π0

2β+γ

sin2β

Unitarity Triangle Analysis

http://www.utfit.org/

9Maurizio Pierini


Unitarity Triangle fit

Vub/Vcb + ∆md + ∆md/∆ms + εK + cos2β + β + + + 2β+γ + sin2β

η = 0.342 ± 0.022 [0.300, 0.385] @ 95% Prob.

ρ = 0.216 ± 0.036 [0.143, 0.288] @ 95% Prob.

sides+Kaon physics+

angles ρ

η

10Maurizio Pierini


Tension in the fit? inclusive from HFAG Vub=(4.38±0.19±0.27)103

exclusive: BRs from HFAG;form factor from quenched LQCD

Vub=(3.80±0.27±0.47)103

incl.+excl.

Vub = (4.22 ± 0.20) 10-3

from all the other inputs:

Vub = (3.48 ± 0.20) 10-3

11Maurizio Pierini


sin2 =0.7910.034from indirect determination

sin2=0.6870.0320.013From direct measurement

If we take it seriously,we want to go beyond the SM

The discrepancy remainseven including the model independentestimation of

theoretical erroron sin2

(Ciuchini,M.P,Silvestrini

hepph/0507290)

Tension in the fit

12Maurizio Pierini


ρ = ± 0.18 ± 0.11η = ± 0.41 ± 0.05

Assuming no NP at tree level the effect of the D0D0

mixing to is negligible wrt the present error

semileptonic decays are clean

We have a NP free

determination ofρ andη

First Step: NPindependent fit

referencestarting point for NP model

building

M.Bona, M.Ciuchini, E.Franco, V.Lubicz, G.Martinelli, F.Parodi,

M.P., P.Roudeau, C.Schiavi, L.Silvestrini, A.Stocchi, V.Vagnoni hepph/0509219

13Maurizio Pierini


J. M. Soares and L. Wolfenstein, Phys. Rev. D 47 (1993) 1021; N. G. Deshpande, B. Dutta and S. Oh, Phys. Rev. Lett. 77 (1996) 4499 [arXiv:hepph/9608231] J. P. Silva and L. Wolfenstein, Phys. Rev. D 55 (1997) 5331 [arXiv:hepph/9610208] A. G. Cohen et al., Phys. Rev. Lett. 78 (1997) 2300 [arXiv:hepph/9610252] Y. Grossman, Y. Nir and M. P. Worah, Phys. Rev. Lett. B 407 (1997) 307 [arXiv:hepph/9704287]

|εK|EXP = Cε⋅|εK|SM

∆mdEXP = Cd⋅∆md

SM

model independent assumptions

∆msEXP = Cs⋅∆ms

SM

ACP(J/ψK0) = sin (2β + 2φBd)

not yet available

5 unknowns

αEXP = αSM - φBd

6 available constraints

Second Step: Fit to NP

14Maurizio Pierini


The UTfit beyond the SM

large NP witharbitrary phase(~4.3% Prob.)

SM or small NP witharbitrary phase

or large NP with SM phase (~95.7% Prob.)

“SM” η = 0.379 ± 0.039 ”NP” [0.398, 0.381] @ 95% Prob.

“SM” ρ = 0.246 ± 0.053 “NP” [0.230, 0.212] @ 95% Prob.

15Maurizio Pierini


Bounds on NP parameters

CBd = 1.27 ± 0.44 φBd = 4.7 ± 2.3oPresent determination

Impact of B factories for 2008 on NP bound

now

2008

16Maurizio Pierini


New sources of CPV in b→s transitions are – much less (un)constrained by the UT fit– natural in many flavour models, given the strong breaking of family SU(3)

Pomarol, Tommasini; Barbieri, Dvali, Hall; Barbieri, Hall; Barbieri, Hall, Romanino; Berezhiani, Rossi; Masiero et al; …

Where do we go from here?

Completing the physics program for'08 BaBar can testboth the scenarios

MFVmodels

LargeFV

models

New sources of CPV in s→d and/or b→d transitions are

• strongly constrained by the UT fit • “unnecessary”, given the great success and consistency of the fit

17Maurizio Pierini


First Case: New Physicsis Minimal

Flavor Violating in thebd sector. No

additional CP violation

18Maurizio Pierini


We can determine and in a universal way for (MFV) NP and SM. εK and ∆md are not used.

ρ = 0.258 ± 0.066 UUTfit

Buras et al. hepph/0007085

η = 0.319 ± 0.039 UUTfit

The Starting Point

18

19Maurizio Pierini


From UT to Rare decays

Dimension 4 operators: FCNC effective Z vertex C=CSM+∆C (constrained by BR(B → Xsl

+l)

and BR(K++))

Dimension 5 operators: (chromo)magnetic penguin

C7eff=(C7

eff)SM+∆C7eff (constrained by BR(B→Xsγ))

Dimension 6 operators: penguins, boxes subleading NP contributions to rare decays

Rare decaysSM functions + ∆C, ∆C7eff

C.Bobeth, M.Bona, A.J.Buras, T.Ewerth, M.P., L.Silvestrini, A.Weiler, hepph/0505110

20Maurizio Pierini


SM like solution

Oppositesign of C

Oppositesign of C7

eff

Constraint on NP contributions

20

21Maurizio Pierini


In MFV models (at low/moderate tanβ) rare decays can be only slightly enhanced w.r.t the SM. Strong suppressions still possible at present.

The Lesson from MFV

If this is the case, we need very high statistics

22Maurizio Pierini


Second Case: New Physicsbrings additional CP violation in the

bs sector We will restrict to SUSY in what follows

23Maurizio Pierini


In general, SUSY generates additional processesthat can contribute to bs transitions

In this case thetransition is generatedby a strong interactionWe gain a factor gs/gW

that can compensate the suppression of the propagator

~1/mq~2

SUSY & bs decays

24Maurizio Pierini


m11

m12

m13

m21

m22

m23

m31

m32

m33 md

00

0ms

0

00

m b

m112

m122

m132

m 212

m222

m 232

m312

m322

m332

ABm d

2

00

0ms

2

0

00

m b2

AB

m q2 0

12AB

13AB

12∗BA

023

AB

13∗BA

23∗BA

0 quark rotation( generating CKM matrix)

Effective interaction(mass insertion)

between second and third family

chirality of incoming and outgoing squarks (LL,LR,RL,RR)

average squark mass

Mass Insertions

24

25Maurizio Pierini


gluinos contribute to rare decays only through(chromo)magnetic penguins; (electro) penguin operators are suppressed

Strong bounds from the combination of bsll and bsγWe can impose the experimental results to bound thesize of the δ's

Bertolini et al., NPB353; Gabbiani et al., NPB477; Buras, Romanino, Silvestrini, NPB520

Ciuchini et al, PRD67; Hiller, PRD69; Gambino,Haisch,Misiak, PRL94

Effect of SM operators

The new amplitudes have arbitrary phases.We are going to look for them in CPViolating processes

26Maurizio Pierini


Re(d23)LLvsIm(d23)LL

No bsllWith bsll

Re(d23)RRvsIm(d23)RR

Re(d23)LRvsIm(d23)LRRe(d23)RLvsIm(d23)RL

Constraints on δ's

26

tanβ = 10 mg = mq = µ =

350 GeV, ~ ~

27Maurizio Pierini


The Effect of δRL on S

SKvs Im(d23)RLSkvs Im(d23)RL

Skvs Im(d23)RLSk'vs Im(d23)RL

Using BR's and direct CP asymmetry(as for the SM fit)Switching on onlyone of the four

chiralities

27

28Maurizio Pierini


Accessing the δ's with Exp.

mixing

decay

decay

B0

B0

fCP

f CP=

qp⋅

A f CP

A f CP

=∣ f CP∣⋅e− 2iCP

mixing

decay

AfCP=

Bphys0 tfCP− Bphys

0 tfCP

Bphys0 tfCPBphys

0 tfCP=CfCP

cosmdtSfCPsinmdt

C f CP=

1− ∣ f CP∣2

1∣ f CP∣2

S f CP=−

2CP ℑ f CP

1∣ f CP∣2

With only one CKM term in the decay (A = A)

C=0 ; S=sin2

Standard Model predictions

29Maurizio Pierini


sidebands

signal region

MeV

GeV/c2

The beamenergy substituted mass

The energy difference(halfCM energy)

BoostLab frame CM frame

two largely independent analysis variables

dominated by beam energy spread

dominated by energy resolution

B Candidate Reconstruction

mES=Ebeam∗2 − pB

∗2

E=E beam∗ − s /2

Ebeam∗ = s /2

P=E ,p P∗=E∗ ,p∗

mES≈2.6 MeV /c2

E≈10÷40 MeV /c2

30Maurizio Pierini


e+e

(βγ)Υ(4s) = 0.56

B mesons pair oscillating in a coherent state

l +

K+

π+

φΚ+

Κ−

K0S

π−

π+

z

t≈z

c∣ z∣~200m

Breco

BtagY(4S)

Vertex Reconstruction

31Maurizio Pierini


log scaleMixedevents

∆t p.d.f.(from data)

∆t = trecttag

log scaleUnmixedevents

We can fit directly on data the parameterization of the Vertex resolution function, using a sample of fully

reconstructed B events

32Maurizio Pierini


Separately determine D for each tag category.

τΒ=1.6 ps

Overall taggingperformance

Mistag (ω) measurement (from data)

Amix=NoMix t − Mix t NoMix t Mix t

f UnmizedMixed

={e− ∣t∣/B

4B

[1 1− 2cos md t ]} R t

∆md = 0.501ps1 (fixed to PDG'04)

D=(12ω)<1 due to mistagsT=2π/∆md

ε=(74.90.2)%

Q = ε(12ω)2

= (300.4)%)

33Maurizio Pierini


The BC Vertexing (I)

y

xπ0

π+ π

Ks

~4 µm

γ

~200 µm

γ

~30 µm

Several of these channels do not have charged tracks from the B vertex. But we can extrapolate back the KS: Using the constraint of the beam

spot on the transverse plane Requiring the KS to decay in the

inner part of the SVT

Beam spot constraint on transverse plane

e+e

γ

Ks

π+

π yz

Btag

Breco

γ

Beam

34Maurizio Pierini


KS xy decay length

# insufficienthits in the SVT

signal MC DistributionBC vertexing resolutionposition of SVT layers

∆t resolution

We split our sample in 4 KS classes

Class I: 1 z & 1 φ hit in SVT layers 13 for both π± Class II: not Class I events with 1 z & 1φ SVT hit for both π± Class III: 1 SVT hit on both π±

Class IV: no SVT hit

Class I&II determine S

All classes determine C

The BC Vertexing (II)

34

35Maurizio Pierini


z resolution vs. Ks xy decay

length

∆t resolution vs. cos θ

Validation on J/ψ Ks(standard vtx. vs BC vtx).

Comparison data vs. MC to parameterize ∆t: ~1% difference in the outliers fraction (quoted systematic effect for BC vtx)

The BC Vertexing (III)

36Maurizio Pierini


Signal

Bkg

qq Background rejection Main source of background from e+e →qq. We can use the different topology to reject them

Several possible topology variables (|cos(θSPH)| , L2/L0, Fisher F(p,θ), NN)

Cut in the selection (8090% eff. on signal) Residual discriminating power

in the likelihood fit

|cos(θSPH)|

F(p,θ)

L2/L0

Signal

Bkg

B produced ~ at rest isotropic topology

qq events jetlike

36

37Maurizio Pierini


K*00 K00

|cos(HEL)| for

mφ bkg

Final states with same multiplicity but different resonances ( vs. f0)

We use helicity and invariant mass

Final states with higher multiplicity Strongly suppressed by lower cut on E

m(KK)

∆E

BB Background rejection

Irreducible and combinatoric background is taken into account in the likelihood

38Maurizio Pierini


Structure of the Likelihood

Shape ofKinematicVariables

=e− N SNqq N BB∏j

N TOT

NS⋅PSj mES ⋅ PS

j E ⋅ PSj f L2, L0 ,... ⋅ PS

j m , H ⋅ PSj t

N qq⋅Pqqj mES ⋅Pqq

j E ⋅ Pqqj f L2, L0 , ...⋅ Pqq

j m , H ⋅ Pqqj t

N BB⋅PBBj mES ⋅PBB

j E ⋅ PBBj f L2, L0 , ...⋅ PBB

j m , H ⋅ P BBj t

Shape ofTopologicalVariables

(Fisher, NN, ...)

Additional Variables(mass, helicity, ...)for BB bkg rejection

t information(to measures S and C)

L

Signal

qq Background

BB Background

Poisson factor for Extended

Maximum Likelihood Fits

39Maurizio Pierini


Fit of B0φK0 events

S=0.50±0.25− 0.040.07

C=0.00±0.23±0.05

KS

11412 events

KL

9818 events

KL

KSACP

ACP

40Maurizio Pierini


Fit of B0KSπ0 events

bkgsig t

S=0.35− 0.330.30 ±0.04

C=0.06±0.18±0.03

30023 events

Phys.Rev. D71 (2005) 111102

ACP

41Maurizio Pierini


Fit of B0K0SK

0SK

0S events

all KS→π+π−

88±10 events

~100% w/ good vertex

*sPlot

all events

one KS→π0π0

45±7 events

one KS with both pions ≥4 SVT hits~97% w/ good vertex

S=0.63− 0.320.28 ±0.04

C=− 0.10±0.25±0.05

Phys.Rev.Lett. 95 (2005)011801

hepex/0507052

hepex/0507052

t

ACP

42Maurizio Pierini


Fit of B0K*0γ events

tPhys.Rev. D72 (2005) 051103

S=− 0.21±0.40±0.05C=− 0.40±0.23±0.03

30023events

SK S

0 ≈− 2ms

mb

sin 2 ≈0

CK

S0 ≈0

in SM we expect

because ofhelicity suppression42

ACP

43Maurizio Pierini


Where We Are Now...

44Maurizio Pierini


And What It Means...

adding S measurements

extrapolationto 2008

Re(d23)RLvsIm(d23)RL

No bs time dep. With bs time dep.

BaBar abd Belle will reduce the bulk of parameter space

45Maurizio Pierini


we can do betterwith K* decays

Other smoking guns?

Puzzling situationin K BR's In SM we expect R~Rc~Rn~1

Still 1.5 includingradiative corrections

E.Baracchini, M.Ciuchini, G.Isidori, M.P, L.Silvestrini

in preparation

CP violation in K+ has also implications on ACP of other decays, depending on the assumed hadronic model.

In principle, we should get a bound on to compare to other determinationsMain Limitation: the presence of penguins imply hadronic uncertanties.

46Maurizio Pierini


Using Dalitz: a new CKM bound

A(B0 → Κ∗+ π−) = Vts Vtb* × P1 - Vus Vub

* × {E1-P1GIM}

A(B+ → Κ∗0 π+) = - Vts Vtb* × P

1 + Vus Vub

* × {A1-P

1GIM}

A(B0 → Κ∗0 π0) = - Vts Vtb* × P1 - Vus Vub

* × {E2+P1GIM}

Vus Vub*~ λ4Charming Penguin ~ λ2

Formally equivalent to BK decays The hadronic parameters are

numerically different than for K But we can access Abs and Arg of

the amplitudes through the interference in Dalitz Plots

K*+(0) K*0()0

B0()

K+(0)0

A(B+ → Κ∗+ π0) = Vts Vtb* × P

1 - Vus Vub

* × {E1+E

2+A

1-P

1GIM}

2⋅2⋅

46

M.Ciuchini, M.P. & L.Silvestrini, hepph/0601233

47Maurizio Pierini


It is possible to experimentally access I=3/2 amplitude

A0=AK∗− 2 AK 00 =− V ub∗ V usE1E2

A=AK∗0 2 AK 0=− V ub∗ V us E1E2

from K+0 Dalitz plot

from K0+0 Dalitz plot

Assuming (for the moment) no EW Penguins, the ratio of these quantities and their CP conjugatedmeasures

R0=A0

A0 =V ub V us

∗

V ub∗ V us

= e− 2i =A−

A =R∓Same argumentapplies to higherK* resonances

Cancellation of Penguins

48Maurizio Pierini


Inclusion of EW Penguins EW penguins are suppressed ~em respect to strong penguins but they are enhanced by 2 respect to I=3/2 amplitude They provide an O(1) correction. We use

to write

R becomes

where

Q9=32Q1

suu− Q1scc3Q1

scc−12

Q3s ; Q10=

32Q2

suu− Q2scc3Q2

scc−12

Q4s

H eff ∝V ub∗ V usC1−

32

V tb∗ V ts C9 Q1

suu− Q1scc

V ub∗ V usC2−

32

V tb∗ V ts C10Q2

suu− Q2scc − V tb

∗ V ts H peng I =1/2

R0 = R∓ = e− 2i Arg 1EW

EW=−32

C9C10

C1C2

V tb∗ V ts

V ub∗ V us

=32

C9C10

C1C211− 2/2 2

2 i

Q7 and Q8 neglected

|C7,8|<<|C9,10|

48

49Maurizio Pierini


Bound on CKM from Arg(R0)

A measurement of B0K+0 Dalitz exists, which determines both K* and K*(1430) and provides Arg(R0)with an error of 18o. We do not have K0 Dalitz plot yet, so the relative phase of B and B cannot be fixedWe can assume a perfect agreement to SM and ...

Possible improvements fitting directly for R

(cancellation of systematics) using all BaBar & Belle current data adding charged modes

The precision is already comparable to from DK

crossing point

0=−32

C9C10

C1C2C9C10

1− 2 /22

2...20o of error

...40o

of error

BaBar, hepex/0408073

49

Same technique gives with the Bs(M.Ciuchini, M.P.,

L.Silvestrini,in preparation)

50Maurizio Pierini


What Next: A Linear SuperB?Using the present technology and the R&D for the ILC, we recently proposed a new kind of machine to build

A small version of the ILC that runs at the Y(4S), with a reduced boost (~0.20.3) and a better vertex resolution (thanks to smaller beamspot and limited multiple scattering). Possible luminosity as large as ~35 1036 (now it is ~1034) Worse knowledge of the CM energy

50

J.Albert et al., physics/0512235

51Maurizio Pierini


even in this situation same∆z resoultion as now

Furthermore

Main Differences w.r.t. BaBar

More Energy spreadless B Less

Boost

BaBar

SuperB

51

52Maurizio Pierini

SLAC experimental Seminar 2/14/06 52

Benefits of a better vtx detector

Still Under Investigation

53Maurizio Pierini


How Things Will Look Like

S(K) vs Im(23)RL

S('K) vs Im(23)RL

assuming 50ab1

54Maurizio Pierini


Conclusions The combination of current constraints on and suggests Consistency of the SM Stringent constraint on New Physics in bd (MFV) No implication on bs (Large FV still possible)In MFV: NP can suppress, but not enhance, the ratesIn Large FV: We expect effects in bs decays Several channels studied, thanks to new techniques Current data show some pattern (S<sin2) but errors are still largeAdditional information can come from rates and amplitudes on Dalitz K puzzle New Bound on CKM from K*We need more data and interplay with LHC and ILCa new linear super B factory can say the last word on flavor physics

55Maurizio Pierini


Backup Slides

56Maurizio Pierini


DIRC (PID)144 quartz bars

11000 PMs

1.5T solenoid

EMC6580 CsI(Tl) crystals

Drift Chamber40 layers

Instrumented Flux ReturnIron / Resistive Plate Chambers or Limited Streamer Tubes (muon /

neutral hadrons)

Silicon Vertex Tracker5 layers, double sided strips

e+ (3.1GeV)

e− (9GeV)

Asymmetric geometry (in order to optimize performances for the

boost of the restframe respect to LAB)

57Maurizio Pierini


The OPE and decay amplitudesSince mb~4GeV and mW~80GeV, weak interaction canbe replaced by an effective local theory, contracting the W propagator to a point (similarthing to be done with t quark)

1p2mW

2 O( )1p2

mW2

This operation breaks the ultraviolet behaviour of the theory.

To remove the after integrating out the heavy degrees of freedomwe need to renormalize the theory, which introduces new operators and effective couplings, as a function of an unphysical regularization scale (µ).

∫ d 4 p

p6≈ ∫ dp

p3 0 ∫ d 4 p

p4≈ ∫ dp

p∞

2 fermions 2 bosonsin the loop 2 fermions

1 boson inthe loop

57

58Maurizio Pierini


The effective Hamiltonian

After the renormalization of the effective theory we get

Penguinoperators

EW Penguinoperators

(cromo)magnetic operators

Tree level operators

59Maurizio Pierini


Contractions of the Heff

Contracting (by Wick theorem) the effective Hamiltonian on certain initial and final states

AB0 K− =K − ∣H eff∣B0= ∑i=1,10

Ci K− ∣Qi∣B0

B π

K

π

K

B

All the perturbative physics (scale > µ) in the Wilson coeff.Ci(µ). All the nonperturbative physics (scale < µ) in thematrix elements. The unphysical dependence on µ has to cancelout. Every operator can produce several diagram topologies whencontracted on the initial and final states. For example, the tree level operators can be contracted into tree levelcontractions <Q>DE(µ) and <Q>CE(µ)

DE CE

59

60Maurizio Pierini


The RGI combinations One can rearrange the contractions of various operators into Renormalization Group Invariant combinations, that represent the physical quantities defining the decay amplitude (Buras &Silvestrini, hepph/9812392). For example the tree level contributions T and C correspond to the RGI's E1 and E2

E1=C1<Q1>DE+C2<Q2>CE E2=C1<Q1>CE+C2<Q2>DE

Penguins are more complicated Every RGI correponds to a contraction of the JµJµ interaction term of the Standard Model (i.e. RGIs are the physical quantities)

61Maurizio Pierini


B

B πc (cu)

K,π

K, π

π

P1(GIM)

A2

Disconnected Annihilation

Charming and GIM penguins(cu)

K, π

π

B Connected Annihilation

62Maurizio Pierini


Preliminary

A comment about radiative effects

What we calculate is the weak decay amplitudeThe experimental measurement includes radiative effect

|A(Bf)|2 = |AW(Bf)|2 G()experimentalobservable

the calculated decay amplitude the QED

correction(dependent on the energy cutoff)

Baracchini & Isidorihepph/0508071

We have to correct each result according to the energy cutoff

63Maurizio Pierini


Predictions for Rare K and B Decays

63

64Maurizio Pierini


In SM the photon is almost fully polarized: AR ~ ms/mb AL

New Physics effects can enhance AR

CP Asimmetry from a final state with mixed CP content. In SM C~0, S~ 2ms/mb •sin(2β)

bR sLbL

L

~mb

sLbL sR

R

~ms

helicity suppression

B0K*0γ In The SM

65Maurizio Pierini


Integrating

Taking the maximum

Both the methods have assumptions. At least in one case they are clear

(flat prior on P). And this is not a 2D Gaussian....

Documents

Searching for New Physics at a Bfactory: Why, Where and HowMaurizio Pierini 1 SLAC experimental Seminar 2/14/06 Searching for New Physics at a Bfactory: Why, Where and How Maurizio