Precise Measurement of the +e+ Branching Ratio · Outline Physics Motivation Experimental Apparatus Data & Analysis Summary & Future Plans Precise Measurement of the π+ → e+ν

Outline Physics Motivation Experimental Apparatus Data & Analysis Summary & Future Plans

Precise Measurement ofthe π+ → e+ν Branching Ratio

E. Frlež (for the PEN Collaboration)

Department of PhysicsUniversity of Virginia

New Trends in High Energy PhysicsYalta, Crimea, Sept. 27 - Oct. 4, 2008


Outline

Physics MotivationPIBETA Experimental Programπ → eν: Theoretical Statusπ → eν: Experimental Status

Experimental ApparatusPEN DetectorCentral Detector RegionWaveform Digitizer

Data & AnalysisADC/TDC/DIG SpectraWaveform AnalysisANN Analysis

Summary & Future PlansPEN 2007-2009


PSI Experiment R-05-01 (PEN) collaboration members:

L. P. Alonzi,a V. A. Baranov,c W. Bertl,b M. Bychkov,a

Yu.M. Bystritsky,c E. Frlež,a N.V. Khomutov,c

A.S. Korenchenko,c S.M. Korenchenko,c M. Korolija,f

T. Kozlowski,d N.P. Kravchuk,c N.A. Kuchinsky,c D. Mekterović,f

D. Mzhavia,c,e A. Palladino,a D. Počanić,a∗ P. Robmann,g

O.A. Rondon-Aramayo,a A.M. Rozhdestvensky,c T.Sakhelashvili,g V.V. Sidorkin,c U. Straumann,g I. Supek,f

Z. Tsamalaidze,e A. van der Schaaf,g∗ E.P. Velicheva,c andV.V. Volnykhc

aDept of Physics, Univ of Virginia, Charlottesville, VA 22904-4714, USAbPaul Scherrer Institut, CH-5232 Villigen PSI, SwitzerlandcJoint Institute for Nuclear Research, RU-141980 Dubna, RussiadInstitute for Nuclear Studies, PL-05-400 Swierk, PolandeIHEP, Tbilisi, State University, GUS-380086 Tbilisi, GeorgiafRudjer Bošković Institute, HR-10000 Zagreb, CroatiagPhysik Institut der Universität Zürich, CH-8057 Zürich, Switzerland


PEN follows the PIBETA experiment

PIBETA program (precision checks of SM):

I π+ → π0e+νe—main goal◦ SM checks related to CKM unitarity

I π+ → e+νeγ(or e+e−)◦ FA/FV , π polarizability (χPT prediction)◦ tensor coupling besides V − A (?)

I µ+ → e+νeν̄µγ(or e+e−)◦ departures from V − A in Lweak

⇒ The PEN experiment:I π+ → e+νe◦ e-µ universality◦ pseudoscalar coupling besides V − A◦ ν-sector anomalies, Majoron searches, mh+, PS l-q’s, Vl-q’s, . . .


π → eν decay: SM predictions; measurements

Modern theoretical calculations:

Bcalc =Γ(π → eν̄(γ))Γ(π → µν̄(γ))calc

=

1.2352 (5)× 10−4 Marciano and Sirlin, [PRL 71 (1993) 3629]1.2356 (1)× 10−4 Decker and Finkemeier, [NP B 438 (1995) 17]1.2352 (1)× 10−4 Cirigliano and Rosell, [PRL 99, 231801 (2007)]

Experiment, world average [current PDG]:

Γ(π → eν̄(γ))Γ(π → µν̄(γ))exp

= (1.230± 0.004)× 10−4

PEN goal:δBB' 5× 10−4 .


π → eν Decay and the SM

Be/µ is given in SM to 10−4 accuracy; dominated by helicitysuppression (V − A). Deviations from this rate primarily causedby new pseudoscalar interactions (mass scale ΛeP):

∆Be/µ = 1−Bnewe/µBSMe/µ

∼√

2Gµ

m2πme(mu + md)

1Λ2eP

∼

(1TeVΛeP

2)× 103.

Thus (δB/B)exp = 10−4 probes ΛeP∼ 103 TeV! → Limits on:I charged Higgs in theories with richer Higgs sector than

SM,I R-parity violating SUSY; loop diagrams with certain SUSY

partner particles,I PS and V leptoquarks in various theories with dynamical

symmetry breaking,I non-zero mν and mixing; sterile ν’s; Majorans.





∼√

2Gµ

m2πme(mu + md)

1Λ2eP

∼

(1TeVΛeP

2)× 103.

Thus (δB/B)exp = 10−4 probes ΛeP∼ 103 TeV! → Limits on:

I charged Higgs in theories with richer Higgs sector thanSM,

I R-parity violating SUSY; loop diagrams with certain SUSYpartner particles,

I PS and V leptoquarks in various theories with dynamicalsymmetry breaking,

I non-zero mν and mixing; sterile ν’s; Majorans.





∼√

2Gµ

m2πme(mu + md)

1Λ2eP

∼

(1TeVΛeP

2)× 103.


SM,

I R-parity violating SUSY; loop diagrams with certain SUSYpartner particles,







∼√

2Gµ

m2πme(mu + md)

1Λ2eP

∼

(1TeVΛeP

2)× 103.



partner particles,







∼√

2Gµ

m2πme(mu + md)

1Λ2eP

∼

(1TeVΛeP

2)× 103.




symmetry breaking,






∼√

2Gµ

m2πme(mu + md)

1Λ2eP

∼

(1TeVΛeP

2)× 103.




symmetry breaking,I non-zero mν and mixing; sterile ν’s; Majorans.


Experimental Constraints on Lepton Non-UniversalityLepton coupling constants gl=e,µ,τ : gl = gl(1 + �l), ∆ij = �i − �j

Loinaz et al. PRD70 (2004).


PSI π → eν Experiment: Czapek et al. 1993

• PSI cyclotron, 78 MeV/c π+ beam• Segmented BGO calorimeter:

132 crystals, φ 55 mm, l=300 mm• Good energy (1.7 %) and timing rms (1 ns)• large solid angle (∼ 4π) coverage• π+ → e+ν tail simulated, not measured• PRL93: (1.235± 0.004± 0.004) · 10−4



• PSI cyclotron, 78 MeV/c π+ beam

• Segmented BGO calorimeter:132 crystals, φ 55 mm, l=300 mm

• Good energy (1.7 %) and timing rms (1 ns)• large solid angle (∼ 4π) coverage• π+ → e+ν tail simulated, not measured• PRL93: (1.235± 0.004± 0.004) · 10−4




132 crystals, φ 55 mm, l=300 mm

• Good energy (1.7 %) and timing rms (1 ns)• large solid angle (∼ 4π) coverage• π+ → e+ν tail simulated, not measured• PRL93: (1.235± 0.004± 0.004) · 10−4




132 crystals, φ 55 mm, l=300 mm• Good energy (1.7 %) and timing rms (1 ns)

• large solid angle (∼ 4π) coverage• π+ → e+ν tail simulated, not measured• PRL93: (1.235± 0.004± 0.004) · 10−4




132 crystals, φ 55 mm, l=300 mm• Good energy (1.7 %) and timing rms (1 ns)• large solid angle (∼ 4π) coverage

• π+ → e+ν tail simulated, not measured• PRL93: (1.235± 0.004± 0.004) · 10−4




132 crystals, φ 55 mm, l=300 mm• Good energy (1.7 %) and timing rms (1 ns)• large solid angle (∼ 4π) coverage• π+ → e+ν tail simulated, not measured

• PRL93: (1.235± 0.004± 0.004) · 10−4




132 crystals, φ 55 mm, l=300 mm• Good energy (1.7 %) and timing rms (1 ns)• large solid angle (∼ 4π) coverage• π+ → e+ν tail simulated, not measured• PRL93: (1.235± 0.004± 0.004) · 10−4


TRIUMF π → eν Program: D. Bryman et al. 1983 & 1993

• TRIUMF cyclotron, 83 MeV/c π+ beam• NaI(Tl) “TINA”: φ 460 mm, l=510 mm• Good energy and timing resolution• small (2.9 %) solid angle coverage• π+ → e+ν tail measured• PRD86: (1.218± 0.014) · 10−4• PLR93: (1.2265± 0.0034± 0.0044) · 10−4

• TRIUMF E1072 experiment proposed 2006◦ better tracking, larger solid angle (25 %)

Goal:δBB' 1× 10−3 .

◦ Engineering runs Aug.-Oct. 2008◦ Data taking 2009-10



• TRIUMF cyclotron, 83 MeV/c π+ beam

• NaI(Tl) “TINA”: φ 460 mm, l=510 mm• Good energy and timing resolution• small (2.9 %) solid angle coverage• π+ → e+ν tail measured• PRD86: (1.218± 0.014) · 10−4• PLR93: (1.2265± 0.0034± 0.0044) · 10−4


Goal:δBB' 1× 10−3 .




• TRIUMF cyclotron, 83 MeV/c π+ beam• NaI(Tl) “TINA”: φ 460 mm, l=510 mm

• Good energy and timing resolution• small (2.9 %) solid angle coverage• π+ → e+ν tail measured• PRD86: (1.218± 0.014) · 10−4• PLR93: (1.2265± 0.0034± 0.0044) · 10−4


Goal:δBB' 1× 10−3 .




• TRIUMF cyclotron, 83 MeV/c π+ beam• NaI(Tl) “TINA”: φ 460 mm, l=510 mm• Good energy and timing resolution

• small (2.9 %) solid angle coverage• π+ → e+ν tail measured• PRD86: (1.218± 0.014) · 10−4• PLR93: (1.2265± 0.0034± 0.0044) · 10−4


Goal:δBB' 1× 10−3 .




• TRIUMF cyclotron, 83 MeV/c π+ beam• NaI(Tl) “TINA”: φ 460 mm, l=510 mm• Good energy and timing resolution• small (2.9 %) solid angle coverage

• π+ → e+ν tail measured• PRD86: (1.218± 0.014) · 10−4• PLR93: (1.2265± 0.0034± 0.0044) · 10−4


Goal:δBB' 1× 10−3 .




• TRIUMF cyclotron, 83 MeV/c π+ beam• NaI(Tl) “TINA”: φ 460 mm, l=510 mm• Good energy and timing resolution• small (2.9 %) solid angle coverage• π+ → e+ν tail measured

• PRD86: (1.218± 0.014) · 10−4• PLR93: (1.2265± 0.0034± 0.0044) · 10−4


Goal:δBB' 1× 10−3 .




• TRIUMF cyclotron, 83 MeV/c π+ beam• NaI(Tl) “TINA”: φ 460 mm, l=510 mm• Good energy and timing resolution• small (2.9 %) solid angle coverage• π+ → e+ν tail measured• PRD86: (1.218± 0.014) · 10−4

• PLR93: (1.2265± 0.0034± 0.0044) · 10−4


Goal:δBB' 1× 10−3 .






Goal:δBB' 1× 10−3 .





• TRIUMF E1072 experiment proposed 2006

◦ better tracking, larger solid angle (25 %)

Goal:δBB' 1× 10−3 .






Goal:δBB' 1× 10−3 .






Goal:δBB' 1× 10−3 .

◦ Engineering runs Aug.-Oct. 2008

◦ Data taking 2009-10





Goal:δBB' 1× 10−3 .



The PIBETA/PEN Apparatus: 1998-2008


The PIBETA/PEN Apparatus: Basic Subsystems

◦ stopped π+ beam◦ active tracking degrader◦ active target counter◦ 240-det. CsI(p) calo.◦ central tracking◦ digitized PMT signals◦ stable temp./humidity

AT

MWPC1

MWPC2

PV

AD

AC1

AC2BC

CsIpure

π+beam

10 cm


Central detector region for the 2007/2008 run

ActiveDegrader

ActiveCollimator

Photomultiplier ModuleHamamatsu H2431-50

TargetCounter

12,5 mm

15,0 mm


4-wedge Tracking Active Degrader 2008

OverviewDeg2008Ver1.xxx

Universität Zürich

Allgemeintoleranz DIN 7168-f

Physik-Institut

Bearb.

der

POS.

Datum

ZEICH_Nr. ANZ. MATERIAL

MASSTAB:

FILE:

PROJEKT/TITEL:

BESCHREIBUNG

(±0.1)

Degrader 2008Overview

BLATT:

Robmann August 2002

PeterPro

x

y

x

z

y

x

y

z

π+Beam

Scintillation Properties BC-408

Light Output, %AnthraceneRise Time, nsDecay Time, nsPulse Width, FWHM, nsWavelength of Maximum Emission, nmBulk Light Attenuation Length, cm

640,92,1~2,5425380

Wrapping thickness: VM 2000 Foil+ Kapton

76 µm75 µm

top

bottom

left

right

5,00 mm1,50 mm

1,50 mm5,00 mm

1,50 mm5,00 mm

1,50 mm5,00 mm

ø 15,00 mm ø 12,00 mm

topHV cnannel S2

rightHV cnannel S1

bottomHV cnannel S3

leftHV cnannel S0

bottomHV cnannel S3

topHV cnannel S2

leftHV cnannel S0

rightHV cnannel S1

PEN Degrader 2008Version 1 Overview

Beam

◦ AD π+ + MWPC e+ Tracking =e+ pathlength in Target=suppresion of In-Flight Decays

◦ BC-408 Scintillator◦ 0.9 ns Rise Time◦ 2.1 ns Decay Time◦ 160 Phel/MeV◦ two x and two y Wedges◦ 3 cm Upstream of Target


Acqiris Digitizer for Target Counters

◦ High-Speed 10-bit PXI/CompactPCI◦ 1 ch=8 G/s, 2 ch=4 G/s, 4 ch=2 G/s◦ Acquisition memory: 256-1024 kp◦ Complete pre- and post-triggering◦ Low 350 ns dead time◦ 400 MB/s PCI bus transfers data◦ High-res. TTI for accurate timing◦ Device driver for Windows XP


PEN: Statistical Uncertainties

Relative statistical uncertainty for Ne2 = (1 + �)Np HT events:

∆Ne2Ne2

=

[1

Np+

(∆�)2

(1 + �)2

]1/2.

For LT prescaling factor f we find:

∆Ne2Ne2

[f + � + �2

fNp

]1/2.

Conservative choice of EHT = 46 MeV → tail fraction � = 0.02.Setting ∆Ne2/Ne2 = 2× 10−4, rπstop = 20, 000/sec, f = 1/16:

Np = 3.4× 107 andrtrig ∼ 75 (6 months net beam time).




∆Ne2Ne2

=

[1

Np+

(∆�)2

(1 + �)2

]1/2.


∆Ne2Ne2

[f + � + �2

fNp

]1/2.

Conservative choice of EHT = 46 MeV → tail fraction � = 0.02.

Setting ∆Ne2/Ne2 = 2× 10−4, rπstop = 20, 000/sec, f = 1/16:





∆Ne2Ne2

=

[1

Np+

(∆�)2

(1 + �)2

]1/2.


∆Ne2Ne2

[f + � + �2

fNp

]1/2.

Conservative choice of EHT = 46 MeV → tail fraction � = 0.02.Setting ∆Ne2/Ne2 = 2× 10−4, rπstop = 20, 000/sec, f = 1/16:



PEN: Systematics Uncertainties

I π/µ decay discrimination: low µ decay pileup, digitizedtarget waveforms

I π/µ decay normalization: Michel energy spectrum,absolute energy calibration (≤ 1× 10−4)

I Acceptance ratio for πe2 and Michel decays: RMD →PIBETA data & analysis (≤ 1× 10−4)

I Hadronic background in the detector: GEANT4 studies(≤ 1× 10−4)

I Time-zero definition and fπd/fsd ratio(≤ 2× 10−4 → ∆t0 = 5 ps)

I Total systematic uncertainty: ≤ 2− 3× 10−4)


π → eν Branching Ratio and π → eν Tail Fraction

Illustration from 2008 data: BR will be determined from HTdata, tail fraction deduced from prescaled LT runs.

Raw histograms show total target energy vs total CsIcalorimeter energy (left: HT, right: LT).


Timing Spectra

RMS 0.07891

DEG-TGT DSC Time (ns)-1 -0.8 -0.6 -0.4 -0.2 -0 0.2 0.4 0.6 0.8 1

Num

ber o

f Eve

nts

0

100

200

300

400

500

600

700

800

900 RMS 0.07891RMS 0.07891

DEG-CsI TDC (ns)0 50 100 150 200

Num

ber o

f Eve

nts

0

10000

20000

30000

40000

50000

HT Trigger+π68 MeV/c

DEG-CsI TDC (ns)0 20 40 60 80 100 120

Num

ber o

f Eve

nts

0

10000

20000

30000

40000

50000

HT Trigger+π68 MeV/c

DEG-CsI TDC (ns)0 20 40 60 80 100 120

Num

ber o

f Eve

nts

0

200

400

600

800

1000 HT Trigger+π68 MeV/c


Target Waveforms: π → eν Events

Digitizer Channel (0.5 ns/ch)

Deg

rade

r D

igiti

zer

Am

plitu

de

π → e ν

0

10000

20000

30000

40000

50000

60000

350 400 450 500 550 600 650 700

Digitizer channel (0.5 ns/ch)

Deg

rade

r D

igiti

zer

Am

plitu

de

π → e ν

0

10000

20000

30000

40000

50000

60000

400 450 500 550 600 650 700

Digitizer Channel (0.5 ns/ch)

Tar

get D

igiti

zer

Am

plitu

de

π → e ν

0

10000

20000

30000

40000

50000

60000

350 400 450 500 550 600 650 700


Tar

get D

igiti

zer

Am

plitu

de

π → e ν

0

10000

20000

30000

40000

50000

60000

400 450 500 550 600 650 700


Target Waveforms: Michel Events

Digitizer channel (0.5 x ns/ch)

Tar

get D

igiti

zer

Am

plitu

de

π → µ → e

0

10000

20000

30000

40000

50000

60000

100 200 300 400 500 600 700


Tar

get D

igiti

zer

Am

plitu

de

π → µ → e

0

10000

20000

30000

40000

50000

60000

350 400 450 500 550 600 650 700


Artificial Neural Network: Inputs and Output

I Multi Layer Percepton definition - linear ANNI Network Structure: ten input neurons, two hidden layers

(10 + 10) and one outputI Inputs: eight charge integrals, number of peaks

(TSpectrum), π+-Stop TimingI Output: 1 = π+ → e+ν, 0 = π+ → µ+ → e+

double p2e_mpl::value(int index,double in0,double in1, doublein2,double in3,double in4,double in5) {input0 = (in0 - 18.365)/2.02788; ...input5 = (in5 - 2.34097)/0.474034;return ((neuron0xa212580()*1)+0); }double p2e_mpl::neuron0xa212580() {double input = 0.0816982;input += synapse0xa212648(); ...return input } ...



I Multi Layer Percepton definition - linear ANN

I Network Structure: ten input neurons, two hidden layers(10 + 10) and one output

I Inputs: eight charge integrals, number of peaks(TSpectrum), π+-Stop Timing

I Output: 1 = π+ → e+ν, 0 = π+ → µ+ → e+





(10 + 10) and one output

I Inputs: eight charge integrals, number of peaks(TSpectrum), π+-Stop Timing

I Output: 1 = π+ → e+ν, 0 = π+ → µ+ → e+






(TSpectrum), π+-Stop Timing

I Output: 1 = π+ → e+ν, 0 = π+ → µ+ → e+






(TSpectrum), π+-Stop TimingI Output: 1 = π+ → e+ν, 0 = π+ → µ+ → e+



π → eν Discrimination with ANN: Monte Carlo

• ANN training on High Thr. MC data: Ee>48 MeV, tπG>10 ns

• Application on Low Thr. MC data, Ee>10 MeV, tπG>10 ns• Extract efficiency and false positive probability for π+ → e+ν

MLP Output-0.2 0 0.2 0.4 0.6 0.8 1 1.2

Nu

mb

er o

f E

ven

ts

1

10

210

310

410

5101A HT Trigger: Signal Background

0.00 percent±=99.98 2eπ∈ 0.01 percent±=0.08 Mm

>10 nsGπ

1E6 MC Events, t

=0.50on

MLP Output-0.2 0 0.2 0.4 0.6 0.8 1 1.2

Nu

mb

er o

f E

ven

ts

1

10

210

310

410

5101A LT Trigger: Signal Background


>10 nsGπ

1E6 MC Events, t

=0.95on



• ANN training on High Thr. MC data: Ee>48 MeV, tπG>10 ns• Application on Low Thr. MC data, Ee>10 MeV, tπG>10 ns

• Extract efficiency and false positive probability for π+ → e+ν

MLP Output-0.2 0 0.2 0.4 0.6 0.8 1 1.2

Nu

mb

er o

f E

ven

ts

1

10

210

310

410



>10 nsGπ

1E6 MC Events, t

=0.50on

MLP Output-0.2 0 0.2 0.4 0.6 0.8 1 1.2

Nu

mb

er o

f E

ven

ts

1

10

210

310

410



>10 nsGπ

1E6 MC Events, t

=0.95on



• ANN training on High Thr. MC data: Ee>48 MeV, tπG>10 ns• Application on Low Thr. MC data, Ee>10 MeV, tπG>10 ns• Extract efficiency and false positive probability for π+ → e+ν

MLP Output-0.2 0 0.2 0.4 0.6 0.8 1 1.2

Nu

mb

er o

f E

ven

ts

1

10

210

310

410



>10 nsGπ

1E6 MC Events, t

=0.50on

MLP Output-0.2 0 0.2 0.4 0.6 0.8 1 1.2

Nu

mb

er o

f E

ven

ts

1

10

210

310

410



>10 nsGπ

1E6 MC Events, t

=0.95on


Discrimination with Neural Networks: MC Simulation

0 10 20 30 40 50 60 70 80 90 1000

5000

10000

15000

20000

25000

1A HT threshold positron energy Entries 4416831A HT threshold positron energy

0 10 20 30 40 50 60 70 80 90 1000

5000

10000

15000

20000

25000

1A HT Michel positron energy Entries 3294841A HT Michel positron energy

-20 0 20 40 60 80 100 120 140 1600

500

1000

1500

2000

2500

3000

3500

4000

1A HT p2e positron timing Entries 112199

0.09 ns±=25.90 πτ

1A HT p2e positron timing

-20 0 20 40 60 80 100 120 140 1600

500

1000

1500

2000

2500

3000

1A HT Michel positron timing Entries 3294841A HT Michel positron timing


PEN Summary & Future Plans

I Developed clean π+ beam tunes with up to 30,000 stoppedπ+/sec at 85 MeV/c momentum.

I Digitized signals of beam detectors: forward (B0) beamcounter, active degrader (AD), and active target (AT).

I Two development runs, in 2007 and 2008, ramping upbeam stop and DAQ rates to design specifications.

I In 2008 run used position-sensitive four-wedge activedegrader, considering replaced it with mini-time-projectionchamber (mTPC) for 2009 data production run.

I Total pions stopped in 2007 and 2008 runs: > 8× 1010. Todate > 4.7× 106 π → eν decays recorded, correspondingto (δB/B)stat < 5× 10−4

I Detailed data analysis under way in preparation for a 2009run, planned to complete the required event statistics.

I PEN Web page: http://pen.phys.virginia.edu


PEN Summary & Future Plans

I Developed clean π+ beam tunes with up to 30,000 stoppedπ+/sec at 85 MeV/c momentum.

I Digitized signals of beam detectors: forward (B0) beamcounter, active degrader (AD), and active target (AT).

I Two development runs, in 2007 and 2008, ramping upbeam stop and DAQ rates to design specifications.

I In 2008 run used position-sensitive four-wedge activedegrader, considering replaced it with mini-time-projectionchamber (mTPC) for 2009 data production run.

I Total pions stopped in 2007 and 2008 runs: > 8× 1010. Todate > 4.7× 106 π → eν decays recorded, correspondingto (δB/B)stat < 5× 10−4

I Detailed data analysis under way in preparation for a 2009run, planned to complete the required event statistics.

I PEN Web page: http://pen.phys.virginia.edu

OutlinePhysics MotivationPIBETA Experimental Programe: Theoretical Statuse: Experimental Status

Experimental ApparatusPEN DetectorCentral Detector RegionWaveform Digitizer

Data & AnalysisADC/TDC/DIG SpectraWaveform AnalysisANN Analysis

Summary & Future PlansPEN 2007-2009

Documents

Precise Measurement of the +e+ Branching Ratio · Outline Physics Motivation Experimental Apparatus Data & Analysis Summary & Future Plans Precise Measurement of the π+ → e+ν