Relative Phase between Strong and EM Decays at BESIII and CLEOc Marco Destefanis Università degli Studi di Torino 1 on behalf of the BESIII Collaboration.

Relative Phase between Relative Phase between

Strong and EM DecaysStrong and EM Decays

at BESIII and CLEOcat BESIII and CLEOcMarco Destefanis Università degli Studi di Torino

1

on behalf of the BESIII Collaboration

Quarkonium 2013

The 9th International Workshop on Heavy Quarkonium

IHEP, Beijing, April 22-26, 2013

OverviewOverview

• BESIII experiment

• Motivation

CLEOc and SND results

• Investigated processes

• Summary

2

3

Charmonium Physics D-Physics Light Hadron Spectroscopy -Physics ...

Physics program

The BESIII Experiment @ IHEPThe BESIII Experiment @ IHEP

BEijing Spectrometer III

e+e- collisions

S tuned depending on energy

D.M. Asner et al, Physics at BES-III, arXiv:0809.1869v1 [hep-ex] (2008)

4

Beam energy:1.0-2.3 GeV

Design Luminosity:1×1033 cm-2s-1

Achieved Luminosity: 6.5×1032 cm-2s-1 @ ψ(3770) Optimum energy:

1.89 GeV Energy spread:

5.16 ×10-4

No. of bunches:93

Bunch length:1.5 cm

Total current:0.91 A

Circumference:237m

e-e+

BEPCII Storage RingsBEPCII Storage Rings

Beijing Electron-Positron Collider II

5

MDC: small cell & He gasxy= 130 mp/p = 0.5% @1GeVdE/dx = 6%

EMC: CsI crystals, 28 cm E/E = 2.5% @1 GeV z = 0.6 cm/E

TOF: T = 80 ps Barrel 110 ps Endcap

Magnet: 1T Superconducting

Muon: 9 layer RPC

Trigger: Tracks & ShowersPipelined; Latency = 2.4 msData Acquisition: Event rate = 3 kHz Thruput ~ 50 MB/s

Zero Degree Detector (ISR)

BESIII DetectorBESIII Detector

J/J/ψψ Strong and Electromagnetic Decay Strong and Electromagnetic Decay AmplitudesAmplitudes

Resonant contributionsΓJ/ψ ~ 93KeV → pQCD

pQCD: all amplitudes almost real [1,2]

QCD -> Фp ~ 10° [1]

Non-resonant continuumpQCD regime

AEM

6[1] J. Bolz and P. Kroll, WU B 95-35.[2] S.J. Brodsky, G.P. Lepage, S.F. Tuan, Phys. Rev. Lett. 59, 621 (1987).

Strong → A3g

Electromagnetic → Aγ

Non-resonant Continuum → AEM

hadrons

hadrons

hadrons

J/J/ψψ Strong and Electromagnetic Decay Strong and Electromagnetic Decay AmplitudesAmplitudes

• If both real, they must interfere (Фp ~ 0°/180°)

• On the contrary Фp ~ 90° → No interference

J/ψ → NN (½+½-) Фp = 89° ± 15° [1]; 89° ± 9°[2]

J/ψ → VP (1-0-) Фp = 106° ± 10° [3]

J/ψ → PP (0-0-) Фp = 89.6° ± 9.9° [4]

J/ψ → VV (1-1-) Фp = 138° ± 37° [4]

• Results are model dependent

• Model independent test:interference with the non resonant continuum

[1] R. Baldini, C. Bini, E. Luppi, Phys. Lett. B404, 362 (1997); R. Baldini et al., Phys. Lett. B444, 111 (1998)[2] M. Ablikim et al., Phys. Rev. D 86, 032014 (2012).[3] L. Kopke and N. Wermes, Phys. Rep. 174, 67 (1989); J. Jousset et al., Phys. Rev. D41,1389 (1990).[4] M. Suzuki et al., Phys. Rev. D60, 051501 (1999).

7

8

J/J/ψψ Strong and Electromagnetic Decay AmplitudesStrong and Electromagnetic Decay Amplitudes

J/J/ψψ →→ NN NN

Favoured channel 3g match 3qq pairs

Without EM contribution p = n, due to isospin

EM contribution amplitudes have opposite sign, like magnetic moments

BRnn expected ~ ½ BRpp

But the BR are almost equal according to BESIII[1]:

BR(J/ψ → pp) = (2.112 ± 0.004 ± 0.027)•10-3

BR(J/ψ → nn) = (2.07 ± 0.01 ± 0.14)•10-3

Suggests 90° phase

[1] M. Ablikim et al., Phys. Rev. D 86, 032014 (2012).

A3g,A R<<1A3g A R 1

2

3

3

)/(

)/(p

g

ng

AA

AA

ppJBr

nnJBrR

9

Cross section for eCross section for e++ee---> -> ππ++ππ--ππ00

Interference of φ(1020) amplitudes @ SND experiment[1]

Shape indicates full interference path

phase ~ 180°

φ decay in agreement with PQCD

A3g and AEM are both real

[1] M.N. Achasov et al., PRD 63, 072002 (2001).

10

Phase Reconstruction @ CLEOcPhase Reconstruction @ CLEOc

φ

S. Dobbs et al., Phys. Rev. D 74, 011105 (2006).

11

Phase Reconstruction @ CLEOcPhase Reconstruction @ CLEOc

Evis = event energy

Charged π FF

S. Dobbs et al., Phys. Rev. D 74, 011105 (2006).

Was an Interference Already Seen?Was an Interference Already Seen?

e+e- → hadrons

e+e- → µ+µ-

e+e- → e+e-J.Z. Bai et al., Phys. Lett. B 355,

374-380 (1995)

Yes

without the strong contribution

12

Investigated ProcessesInvestigated Processes

Inclusive scenario: does not see anything

The phase is there, but the mean goes to 0

Interference |3| * fgf

Sum over all the final states ||3 ffg

Closure approximation 1|| ff

But orthogonal states0|3 g

If we sum over all the channels, the interference ≈ 013

Investigated ProcessesInvestigated Processes

Exclusive scenario: could see interference effects

• e+e+ -> J/ψ -> pp, nn NN

BR ~ 2.17x10-3 σcont~ 11 pb

• e+e- -> J/ψ -> ρπ VP

BR ~ 1.69% σcont~ 20 pb

• e+e- -> J/ψ -> 2(π+π-)π0

BR ~ 5.5% σcont~ 500 pb

14

15

Investigated ProcessesInvestigated Processes

Exclusive scenario: could see interference effects also on

• e+e- -> J/ψ -> π+π-

• e+e- -> J/ψ -> K+K-

• e+e- -> J/ψ -> K0K0

proposed and under study [1]

All the other channels for free

Even number of π: strong decay forbidden

-> interference must be

seen[1] H. Czyz, and J. Kühn, Phys. Rev. D80: 034035 (2009)

σcont~ 11 pb

21FF

S

10

1

W

Continuum Cross SectionContinuum Cross Section

16

σcont~ 500 pb

0

1

W

σcont~ 20 pb

6

1

W

pppppppp

ρπρπρπρπ

55ππ55ππ

V. Druzhinin et al., Rev. Mod. Phys. 83, 1545 (2011) ; B. Aubert et al., Phys. Rev. D73,012005 (2006)

Phase GeneratorPhase Generator

• Event generator

• Monte-Carlo method (100000 iterations)

• Cross section evaluation at each point

• Beam spread gaussian (0.93 MeV)

• Radiative correction (simple model to be optimized)

• Max radiation 300 MeV (~20% ECM)

• Cross section:2

3217

2

2/1012][

i

risris

i

outin eCiWW

eCC

W

cBBnb

17

Simulated Yields for eSimulated Yields for e++ee---> -> pppp

continuum referenceσ ~ 11 pb

beam energy spread + radiative corrections

(to be optimized)

no corrections beam energy spread(0.93 MeV)

Δφ = 0°

Δφ = 90°

Δφ = 180°

18

Simulated Yields for pp -> Simulated Yields for pp -> µµ++µµ--

continuum referenceσ ~ 18 pb

no corrections beam energy spread

Δφ = 0°

Δφ = 90°

Δφ = 180°

19

Phase SignPhase Sign

* red: Δφ = -90°

blue: Δφ = +90°Maximum differences at the 1% level

20

pppppppp

Energy Points ChoiceEnergy Points Choice

ρπρπ

55ππ

pppp

21

Maximum interference: 0°

Depends on the process

Energy Points ChoiceEnergy Points Choice

ρπρπ

55ππ

pppp

22

2 pts at low Wfix the continuumfix the slope

Maximum interference: 0°

Depends on the process

Energy Points ChoiceEnergy Points Choice

ρπρπ

55ππ

pppp

23

2 pts at low Wfix the continuumfix the slope

2 pts at deep positions

Maximum interference: 0°

Depends on the process

Energy Points ChoiceEnergy Points Choice

ρπρπ

55ππ

pppp

24

2 pts at low Wfix the continuumfix the slope

2 pts at deep positions

Maximum interference: 0°

Depends on the process

Energy Points ChoiceEnergy Points Choice

ρπρπ

55ππ

pppp

25

2 pts at low Wfix the continuumfix the slope

2 pts at deep positions 1 pt Beginning of the BW

Maximum interference: 0°

Depends on the process

Energy Points ChoiceEnergy Points Choice

What happens at 90°

Gradient calculation

The deep corresponds roughly to the maximum gradient

(σ90-σi)/σ90

i = 70i = 70

i = 100i = 100

i = 80i = 80

i = 110i = 110

26

i = i = 7070

i = 100i = 100

i = i = 8080

i = 110i = 110

pppppppp ρπρπρπρπ

Energy Points ChoiceEnergy Points Choice

3050 MeV

3060 MeV

3083 MeV

3090 MeV

3093 MeV

27

Luminosity HypothesisLuminosity Hypothesis

• 5 values of Luminosity: 8.6•1031, 1032, 2•1032,

5•1032, 1033 [cm-2s-1]

• Time: 1 day = 86400 s

• Injection efficiency = 0.8

• Reconstruction efficiencypp = 0.67ρπ = 0.385π = 0.20

• Rate = L•T•εinj•εrec • σ

Integrated Luminosity

Lint/day = L • T • εinj

6•1036, 6.9•1036,

1.4•1037, 3.5•1037,

6.9•1037 [cm-2]

28

Precision of the FitPrecision of the Fit

29

10°

90°

170°

Statistical error for:

pp circle

ρπ triangle

170°

• Lower sensitivity

(No 0°-90° and 90°-180° symmetry)

2 parameters:

φ and σcont

30

5 days Lint = 1.4x1037 [cm-2]

points: 3050,3060, 3083,3090, 3093

MeV

ℓ1 : ℓ1 : ℓ2 : ℓ2 : ℓ1

Fit resultsFit results

10°

90°

170°

Statistical error:

pp circle

ρπ triangle

Open points:

1:1:0.5:0.5:2

Very low sensitivity to Luminosity ratiosBest and simplest choice: 1:1:1:1:1

31

J/J/ψψ Scan Scan3 parameters:

φ, σcont and Bout

Points Par Inj. eff. Δφ [°] Δσ [pb] ΔBout

5 3 0.7 29.3 1.3 0.7•10-3

5 3 0.8 26.7 1.3 0.7•10-3

6 3 0.8 6.1 0.9 0.4•10-5

12 3 0.7 6.3 0.9 0.7•10-4

12 3 0.8 5.9 0.9 0.7•10-4

σcont = 11 pb Bout = 2.17•10-3

3 parameters: 3096.9 needed (1 point more with high

statistics)

Δφ = +90°pppppppp

32

J/J/ψψ Phase Phase

Energyrequested

[MeV]

Energycollected

[MeV]LLintint [pb [pb-1-1]]

3050 3046 14.0

3060 3056 14.0

3083 3086 16.5

3090 3085 14.0

3093 3088 14.0

3097 3097 79.6

33

J/J/ψψ Phase – Real Data Phase – Real Data

Ecm(GeV) 𝓛 (pb-1)3.0500 14.895±0.029

3.0600 15.056±0.030

3.0830 4.759±0.017

3.0856 17.507±0.032

3.0900 15.552±0.030

3.0930 15.249±0.030

3.0943 2.145±0.011

3.0952 1.819±0.010

Ecm(GeV) 𝓛 (pb-1)3.0958 2.161±0.011

3.0969 2.097±0.011

3.0982 2.210±0.011

3.0990 0.759±0.007

3.1015 1.164±0.010

3.1055 2.106±0.011

3.1120 1.719±0.010

3.1200 1.261±0.009

3.0969 79.6

B.X. Zhang, Luminosity measurement for J/psi phase and lineshape study.

34

ee++ee---> -> μμ++μμ-- Phase Phase ReconstructionReconstruction2 good charged tracks:

|Rxy|<1cm, |Rz|<10cm;|cos|<0.8.

No good neutral tracks in EMC:

0<T<14 (x50 ns)

E>25MeV (|cos|<0.8), E>50 MeV (0.86<|cos|<0.92)

,charged<10o.Vertex fit to impove the momentum resolution:

2vertex<100.

Veto e+e:

Each charged track has an energy deposit in EMC;

E/p<0.25.

Veto cosmic rays:

=|Tof(+)-Tof()|<0.5

Momentum window cut:

•|p±-pthe|<3

Leptonic decay

Contributions from Aγ and AEM

35

ee++ee---> 2(-> 2(ππ++ππ--) Phase ) Phase ReconstructionReconstruction

4 good charged tracks:

|Rxy|<1cm, |Rz|<10cm.

Vertex fit to improve the momentum resolution.

Veto bkg from -conversion

(2(e+e)):

All angles between and , 10o<+<170o.

Veto events which have multi-tracks:

Minimum angle between () pairs: ()>170o.

G-Parity

Contributions from Aγ and AEM

36

ee++ee---> 2(-> 2(ππ++ππ--))ππ00 Phase Phase ReconstructionReconstruction

4 good charged tracks:

|Rxy|<1cm, |Rz|<10cm.

At least 2 good neutral tracks in EMC:

0<T<14 (x50 ns);

E>25MeV (|cos|<0.8), E>50 MeV (0.86<|cos|<0.92)

,charged<10o.

PID for each charged track:

prob()>prob(K)

Vertex fit:

2vertex<100.

3-C kinematic fit:

Loop all photons, choose the combination with the minimum 2

3C(<200).

0 selection:

|M()-0.135|<0.02 GeV/c2

9.0|E-E|

|cos|0

21decay

0

p

）（

Multi-combination from intermediate processes

Contributions from Aγ and AEM

37

2 good charged tracks:

• |Rxy| < 1 cm, |Rz| < 10 cm;

• back-to-back tracks: 178° < θ < 180°;

• p < 2 GeV/c;

• |cos| < 0.92

Analysis in Barrel + End Cap.

ppbar Events Reconstructionppbar Events Reconstruction

M. Ablikim et al., Phys. Rev. D 86, 032014 (2012).

38

ISR Radiative CorrectionsISR Radiative Corrections

Comparison of different generators

•KKMC Phase Space

•KKMC 1 + cos2θ

•ConExc

•Babayaga

ISR on/off

Check at each energy point

Reconstruction Efficiency and Systematic Errors

SummarySummary

• J/ψ decay amplitude phase: 0° (theory) but 90°

(data)

• Energy points collected: 3046, 3056, 3086, 3085,

3088

• Statistical significance enough to discriminate between different theoretical predictions

• Precision of fit → Luminosity dependence

• Analysis is ongoing

Next StepsNext Steps

• Complete the presented analysis

• Analyze more final states

• More refined ISR evaluation

39