Recent + future MEG results (and LFV implications)

in the “SuperB” era

Giovanni Signorelli INFN Sezione di Pisa

XVII SuperB Workshop and Kick-off Meeting

La Biodola (Isola d’Elba) 1 June 2011

3

Lepton flavor violation• LFV decays in the SM is radiatively induced by neutrino masses and mixings at a negligible

level

• All SM extensions enhance the rate through mixing in the high energy sector of the theory (other particles in the loop...)

• Clear evidence for physics beyond the SM

- background-free • Restrict parameter space of SM extensions

! e!! !e

W

"

=3α

32π

�∆m2

23s13c13s23

M2W

�2

relative probability ~ 10–54

10-14

10-13

10-12

10-11

10-10

10-9

10-8

B.R.(!"

"e)

1.21.11.00.90.80.70.60.5The unified third generation Yukawa coupling (MG)

BR

(µ→

eγ)

• LFV is related to a “new” lepton-lepton coupling

• A wide field of research

- LFV decays

- Anomalous magnetic moment for the µ, τ- Muon-to-electron conversion

- (LFV in B-meson decays)

Many processes

4

yij �̄iFµν�jσµν

γ

µ e

γ

τ µ,e

γ

µ µ

Ze

µ e

µ→ eγ τ → µγτ → eγ

(g − 2)µ µ−N → e−Nµ→ eee

µ,τ

b

γ

d l

l

e e

eNPNP NP NP NP

NP

B → ��̄�

B → ��̄�Xs

τ!"’# $%&'

(-$!)*&%+

• LFV is related to a “new” lepton-lepton coupling

• A wide field of research

- LFV decays

- Anomalous magnetic moment for the µ, τ- Muon-to-electron conversion

- (LFV in B-meson decays)

Many processes

5

yij �̄iFµν�jσµν

γ

µ e

γ

τ µ,e

γ

µ µ

Ze

µ e

µ→ eγ τ → µγτ → eγ

(g − 2)µ µ−N → e−Nµ→ eee

µ,τ

b

γ

d l

l

e e

eNPNP NP NP NP

NP

B → ��̄�

B → ��̄�Xs

τ!"’# $%&'

(-$!)*&%+

Processes are correlatedModel-dependent correlations

6

Barbieri et al,, Nucl. Phys B445 (1995) 225Hisano et al., Phys. Lett. B391 (1997) 341Masiero et al., Nucl. Phys. B649 (2003) 189Calibbi et al., Phys. Rev. D74 (2006) 116002Isidori et al., Phys. Rev. D75 (2007) 115019...

The LFV wheel

7

µ→ eγ

µ→eee

µ−N → e−N τ →µγ

τ →eγ

�Zα

π

� �mτ

mµ

�2÷4

(g − 2)µ

LFV couplings

Bµeγ

10−12∝

�∆aµ

10−9

�2(αe.m.)

× tan2 β

≡ O(1)∝∝

∝

,&''&- .&/01#

Present limits

8

µ→ eγ

µ→eee

µ−N → e−N τ →µγ

τ →eγ

(g − 2)µ

× tan2 β

1.2× 10−11

3.3÷ 4.5× 10−8

aexpµ − aSM

µ =(296± 81)× 10−11

B-factories

BNL E821

SINDRUMII

1× 10−12

B(µTi→ eTi) < 4.3× 10−12

B(µAu→ eAu) < 7× 10−13

SINDRUM

MEGA@BNL

5626

5667

2333

2344

566<

! 89-* $&% :;?

Future prospects

9

µ→ eγ

µ→eee

µ−N → e−N τ →µγ

τ →eγ

(g − 2)µ

× tan2 β

10−16 → 10−18

few × 10−13 2× 10−9

∼ 10−15÷16 ∆aµ = (XXX ± 34)× 10−11

3.6σ → 8σ

562?→

562?→

562?

%"--9-=

→562>

562<→

562?→

mu2e COMET

MEG

HeidelbergGm2 FNAL

SuperB1÷

The MEG collaboration

10

KEK

Tokyo U.Waseda U.

KEK

INFN & U PisaINFN & U Roma

INFN & U GenovaINFN & U Pavia

INFN & U Lecce

PSI UCIrvine JINR DubnaBINP Novosibirsk

The MEG collaboration

11

Tokyo U.Waseda U.

KEK

INFN & U PisaINFN & U Roma

INFN & U GenovaINFN & U Pavia

INFN & U Lecce

PSI UCIrvine JINR DubnaBINP Novosibirsk

J. AdamM. Hildebrandt P.-R. Kettle O. KiselevA. Papa S. Ritt

X. Bai E. Baracchini T. DokeY. Fujii T. Haruyama T. Iwamoto A. Maki S. Mihara T. Mori H. Natori H. Nishiguchi Y. Nishimura W. Ootani R. Sawada Y. Uchiyama A. Yamamoto

A. Baldini C. Bemporad G. Boca P. W. Cattaneo G. Cavoto F. Cei C. Cerri A. De Bari M. De Gerone S. DussoniK. Fratini L. Galli F. Gatti M. GrassiD. Nicolò M. PanareoR. Pazzi† G. Piredda F. Renga M. Rossella

B. Golden G. LymmW. Molzon

F. Sergiampietri G. Signorelli F. TenchiniC. VoenaD. Zanello

D. N. Grigoriev F. Ignatov B. I. Khazin A. Korenchenko N. Kravchuk D. Mzavia†

A. Popov Yu. V. Yudin

Time scale

• A experiment at the Paul Scherrer Institut (PSI)• The decay• The detector

• Overview of sub-detectors• Calibration methods

• Analysis of 2009 run• Status

• Run 2010 • 2011 and Next year(s)

12

µ→ eγµ→ eγ

Engineering

first limit (<2.8 x 10 -11)

This presentation

data under analysis

starting soon!

2012

Signal and Background

13

“Signal” “RMD” “Accidental”

The accidental background is dominant and it is determined by the experimental resolutions

~~~~~µ+

!e+

"µ

"e µ+ e+

"µ

"e

~~~~~µ+

! e+

"µ

"e

~~~~~ µ+

!

e+

µ+

Ee = Eγ = 52.8 MeV

θeγ = 180ºteγ ~ 0

14

MEG experimental method

Easy signal selection with μ+ at rest

• µ: stopped beam of 3 x 107 μ /sec in a 205 μm polyethylene target

- PSI !E5 beam line

• e+ detection

magnetic spectrometer composed by solenoidal magnet and drift chambers for momentum

plastic counters for timing

• γ detection

Liquid Xenon detector based on the scintillation light

- fast: 4 / 22 / 45 ns- high LY: ~ 0.8 * NaI- short X0: 2.77 cm

~~~~~ µ+

!

e+

µ+

Some detector pictures

15

LXe detectorDC system

Beam Line

16

The photon detector• γ Energy, position, timing

• Homogeneous 0.8 m3 volume of liquid Xe

• 10 % solid angle

• 65 < r < 112 cm

• |cosθ| < 0.35 |ϕ| < 60o

• Only scintillation light

• Read by 848 PMT

• 2’’ photo-multiplier tubes

• Maximum coverage FF (6.2 cm cell)

• Immersed in liquid Xe

• Low temperature (165 K)

• Quartz window (178 nm)

• Thin entrance wall

• Singularly applied HV

• Waveform digitizing @2 GHz

• Pileup rejection

Liq. Xe

H.V.

Vacuum

for thermal insulation

Al Honeycomb

window

PMT

Refrigerator

Cooling pipe

Signals

fillerPlastic

0 100 cm50

17

Xe properties• Liquid Xenon was chosen because of its unique properties among radiation

detection active media

• Z=54, ρ=2.95 g/cm3 (X0=2.7 cm), RM=4.1 cm

• High light yield (similar to NaI)

• 40.000 phe/MeV

• Fast response of the scintillation decay time

•τsinglet= 4.2 ns

•τtriplet= 22 ns

•τrecomb= 45 ns

• Particle ID is possible

• α ~ singlet+triplet, γ ~ recombination

• Large refractive index n = 1.65

• No self-absorption (λAbs=")

α-particle

electron

Xe Xe

Xe Xe

e e

e

Calibrations

27

LEDPMT Gain

Higher V with

light att.

Alpha on wires

PMT QE & Att. L

Cold GXe

LXe

Laser

Laser

relative

timing calib.

Nickel ! Generator

9 MeV Nickel _-line

NaI

quelle

onoff

Illuminate Xe from

the back

Source (Cf)

transferred by

comp air ! on/off

Proton Accelerator Li(p,!)Be

LiF target at

COBRA center

17.6MeV !

~daily calib.

also for initial

setup

KBi

Tl

F

Li(p, !0) at 17.6 MeV

Li(p, !1) at 14.6 MeV

µ radiative decay

"0! !!"- + p ! "0 + n

"0 ! !! (55MeV, 83MeV)

"- + p ! ! + n (129MeV)

LH2 target

!

e+

e-

ee!!

##µµ

##Lower beam intensity < 107

Is necessary to reduce pile-

ups

A few days ~ 1 week to get

enough statistics

Xenon

Calibration

18

[MeV]!E0 5 10 15 20

Num

ber

of e

vent

s

0

200

400

600

800

1000

1200

!"#$%&#'(!! )*+*,%-./01)232 9

!"#$%&"'(

454

!' !65"

77

89

!':

• The precise knowledge of the calorimeter energy scale is crucial for the experiment

• constant check of Xe light yield and purity

- trigger threshold

- systematic error on energy scale

• Different calibrations have different time-scales

19

γ-energy scale calibration

Process Energy Frequency

Charge exchange 55, 83, 129 MeV year - month

Proton accelerator 14.8, 17.6 MeV week

Nuclear reaction 9 MeV daily

Radioactive source 1.1 -4.4 MeV daily

!!

p ! !0n

!0! ""

7Li(p, !17.6)8Be

58Ni(n, !9)59Ni En

ergy

Freq

uenc

y

2009: efficient physics run

20

- 2008 run BR<2.8 x 10-11

Nucl. Phys. B834, 1–12 (Apr. 2010)

January - October - detector dismantling - improvement (after run 2008)

-DCH-Electronic

- re – installation - LXe purification - CW calibration - another experiment in the area had

“exciting results” (µp)

October - !º calibration

November – December - MEG run

Running conditionsMEG run period

– Live time ∼84% of total time– Total time ~ 7 weeks– μ stop rate: 3x107 μ/s– Trigger rate 6.5 ev/s ;– Total data taken: 93 TB

Programmed beam shutdowns

Beam line tests and Maintenance

Analysis principle• A µ→eγ event is described by 5 kinematical variables

• Ee, Eγ, (Δϑ, Δφ), teγ

• Likelihood function is built in terms of Signal, radiative Michel decay RMD and

background BG number of events and their probability density function PDFs

• Extended unbinned likelihood fit

- fit (Nsig, NRMD, NBG) in a wide region

• PDFs taken from

- data

- MC tuned on data

21

• 48 ≤ Eγ ≤ 58 MeV• 50 ≤ Ee ≤ 56 MeV• | Teγ | ≤ 0.7 ns• | φeγ |, | θeγ | ≤ 50 mrad

22

(nsec)e t-2 0 2

(MeV

)E

44464850525456586062

Right Sideband

Left Sideband

E Sideband

Blind box

RMD

• We adopt a blind-box likelihood analysis

strategy

• The blinding variables are Eγ and teγ

- Hidden until analysis is fixed

• Three independent analyses

- different pdf implementation

- Fit or input NRMD, NBG

- Different statistical treatment (Freq. or Bayes)

• Use of the sidebands

- our main background comes from accidental

coincidences

- RMD can be studied in the low Eγ sideband

Analysis principle

Pdfs and resolutions

23

µ→eγ runs

• Average upper tail for deep conversions• σR = (2.1 ± 0.15) %

• Systematic uncertainty on energy scale < 0.6%

• Resolution functions of core and tail components• core = 390 keV (0.74%)

• Positron angle resolution measured using multi-loop tracks• σ(φ) = 7.1 mrad (core)• σ(ϑ) = 11.2 mrad

• Overall angular resolution combining • XEC+DCH+target

• σ(φ) = 12.7 mrad (core)• σ(ϑ) = 14.7 mrad

Eγ Ee+ teγ

(nsec)e t-1 0 1

Num

ber

of e

vent

s /(0

.080

nse

c)

0200400600800

10001200140016001800200022002400

(MeV)eE50 51 52 53 54 55 56

Num

ber

of e

vent

s (0.

10/ M

eV)

0

1000

2000

3000

4000

5000

0.048 0.05 0.052 0.054 0.056 0.058

Num

ber

of e

vent

s (0.

5 / M

eV)

0

200

400

600

800

1000

1200

1400

(MeV)E50 52 54 56 58 60

Num

ber

of e

vent

s / (0

.64

MeV

)

0

100

200

300

400

500

600

700

48 50 52 54 56 58

• 40 MeV < Eγ < 48 MeV• σt is corrected for a

small energy-dependence• (142 ± 15) ps• stable within 15 ps

along the run• MEGA had on RMD

• 700 ps resolution

!º

γ bck

RMD

Michel

!"#$%&'()(*+%,

!"#$%&'()(*+%,

!"#$%&'()(*+%,

Normalization• The normalization factor is obtained from the number of observed Michel

positrons taken simultaneously (pre-scaled) with the µ→eγ trigger• Cancel at first order

- Absolute e+ efficiency and DCH instability

- Instantaneous beam rate variations

24B.R. = Nsig x (1.01 ± 0.08) # 10-12

theory

resolution

acceptance

O(1)

~18k107

Likelihood fit result• Nsig < 14.5 @ 90% C.L., Nsig best–fit value = 3.0• Nsig = 0 is in 90% confidence region

- C.L @0: 40÷60% depending on the statistical approach

-

25

(sec)eT-0.5 0 0.5

-910!

Even

ts /

( 5.6

e-11

sec

)

0

5

10

15

20

25

30

(sec)eT-0.5 0 0.5

-910!

Even

ts /

( 5.6

e-11

sec

)

0

5

10

15

20

25

30

(GeV)eE0.05 0.051 0.052 0.053 0.054 0.055

Even

ts /

( 0.0

0024

GeV

)

0

5

10

15

20

25

30

35

40

45

(GeV)eE0.05 0.051 0.052 0.053 0.054 0.055

Even

ts /

( 0.0

0024

GeV

)

0

5

10

15

20

25

30

35

40

45

(GeV)E0.048 0.05 0.052 0.054 0.056 0.058

Even

ts /

( 0.0

004

GeV

)

0

10

20

30

40

50

60

70

(GeV)E0.048 0.05 0.052 0.054 0.056 0.058

Even

ts /

( 0.0

004

GeV

)

0

10

20

30

40

50

60

70

(rad)-0.04 -0.02 0 0.02 0.04

Even

ts /

( 0.0

04 )

0

5

10

15

20

25

(rad)-0.04 -0.02 0 0.02 0.04

Even

ts /

( 0.0

04 )

0

5

10

15

20

25

(rad)-0.04 -0.02 0 0.02 0.04

Even

ts /

( 0.0

04 )

0

5

10

15

20

25

(rad)-0.04 -0.02 0 0.02 0.04

Even

ts /

( 0.0

04 )

0

5

10

15

20

25 Accidental BGRMDSignalTotal

Dashed lines : 90% C.L. UL of Nsig

Fitting was done by three groups with different parametrization, analysis window and statistical approaches, and confirmed to be consistent (Nsig best fit = 3.0-4.5, UL = 1.2-1.5×10-11)

!"#$%&'()(*+%,!"#$%&'()(*+%, !"#$%&'()(*+%,

!"#$%&'()(*+%, !"#$%&'()(*+%,

Upper limit• From the analysis of the 2009 data our limit on the BR is the following:

- cfr. MEGA limit BR < 1.2 x 10–11 @ 90% C.L.

• Sensitivity: • 6.1 x 10-12 average 90% upper limit on null-signal toy experiments • BR < (4 ÷ 6) x 10-12 from the SideBands

• On going activity• better understanding of the spectrometer• reduction of systematics on back-to-back alignment• better usage of sideband information in the likelihood

• We plan to present a combined 2009/2010 analysis this summer26

!"#$%&'()(*+%,

27

(MeV)eE50 51 52 53 54 55 56

(MeV

)E

48

49

5051

52

53

54

5556

5758

(MeV)eE50 51 52 53 54 55 56

(MeV

)E

48

49

5051

52

53

54

5556

5758

ecos-1 -0.9995 -0.999 -0.9985

(nse

c)et

-3.5

-3

-2.5

-2

-1.5

-1

-0.5

0

ecos-1 -0.9995 -0.999 -0.9985

(nse

c)et

0

0.5

1

1.5

2

2.5

3

3.5

(MeV)eE50 51 52 53 54 55 56

(MeV

)E

48

49

5051

52

53

54

5556

5758

8152 3

ecos-1 -0.9995 -0.999 -0.9985

(nse

c)et

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

12 3

56 7

8 9

10

!"#$%&'()(*+%,

!"#$%&'()(*+%, !"#$%&'()(*+%,

!"#$%&'()(*+%,

!"#$%&'()(*+%,

!"#$%&'()(*+%,

Blue lines are 1(39.3 % included inside the region w.r.t. analysis window), 1.64(74.2%) and 2(86.5%) sigma regions.For each plot, cut on other variables for roughly 90% window is applied.

Event distibution

ecos-1 -0.9995 -0.999 -0.9985

(nse

c)et

0

0.5

1

1.5

2

2.5

3

3.5

!"#$%&'()(*+%,

Signal region

RightSB

LeftSB

Event display• Events in the signal region were checked carefully• An event in the signal region

28

-80 -60 -40 -20 0 20 40

-40

-20

0

20

40

xy

-80 -70 -60 -50 -40 -30 -20 -10 0

-50

-40

-30

-20

-10

0

-100 0 100 200 300-100

-80

-60

-40

-20

0

20

40

60

80

100

1

3

12

42

148

517

1806

6304

21999

z

0

20

40

60

80

100

What’s next?• Data taking was restarted from Aug. 5 to Nov. 6 2010

- !º calibration from 23/8 to 9/9

- accident to the beam transport solenoid on Nov. 6

- ~ 2 x 2009 statistics

• An accident on Nov. 6 put a premature end to the 2010 run• Analysis ongoing

- 2009 & 2010 data together

• Run 2011 soon starting

- physics data taking from June to December

29

2009 run

2010

!º

µ on target

Sensitivity prospect• Data from the two months of stable data taking of the MEG experiment in 2009 give

a result competitive with the previous limit

• Plans to reach its design sensitivity (few x 10–13) within 2013

30

Back to the wheel...

31

µ→ eγ

µ→eee

µ−N → e−N τ →µγ

τ →eγ

(g − 2)µ

× tan2 β

10−16 → 10−18

few × 10−13 2× 10−9

∼ 10−15÷16 ∆aµ = (XXX ± 34)× 10−11

3.6σ → 8σ

562?→

562?→

562?

%"--9-=

→562>

5627→

562?→

mu2e COMETMEG

HeidelbergGm2 FNAL

SuperB

Thank you

32

Back-up slides