An Updated High Precision Measurement of
the Neutral Pion Lifetime via the Primakoff
Effect
A. Gasparian
NC A&T State University, Greensboro, NC
for the PrimEx
Collaboration
Outline
 Physics Motivation
 Different methods of lifetime measurements
 The PrimEx experiment and our first results
 Control of systematic errors
 Summary
0 decay width
 0→ decay proceeds primarily via the chiral anomaly in QCD.
 The chiral anomaly prediction is exact for massless quarks:
 2 N c2 m3
    
 7.725 eV
576 3 F2
0
 Corrections to the chiral anomaly prediction:
(u-d quark masses and mass differences)
Calculations in NLO ChPT:
(J. Goity, at al. Phys. Rev. D66:076014, 2002)
~4% higher than LO, uncertainty: less than 1%
 Recent calculations in QCD sum rule:
(B.L. Ioffe, et al. Phys. Lett. B647, p. 389, 2007)
0→
Γ(0) = 8.10eV ± 1.0%
 Γ() is only input parameter
 0- mixing included
Γ(0) = 7.93eV ± 1.5%
 Precision measurements of (0→) at the percent level will provide
a stringent test of a fundamental prediction of QCD.
A. Gasparian
PAC33, January 15, 2008
2
Decay Length Measurements (Direct Method)
 Measure 0 decay length
1x10-16 sec too small to measure
solution: Create energetic 0 ‘s,
L = vE/m
But, for E= 1000 GeV, Lmean  100 μm
very challenging experiment
 Major limitations of method
 unknown P0 spectrum
0→
1984 CERN experiment:
P=450 GeV proton beam
Two variable separation (5-250m) foils
Result:
(0) = 7.34eV3.1% (total)
 needs higher energies for improvement
A. Gasparian
PAC33, January 15, 2008
3
e+e- Collider Experiment
 DORIS II @ DESY
 e+e-e+e-**e+e-0e+e-
 e+, e- scattered at small angles (not detected)
Results:
Γ(0) = 7.7 ± 0.5 ± 0.5 eV ( ± 10.0%)
 Not included in PDG average
0→
 only  detected
 Major limitations of method
 knowledge of luminosity
 unknown q2 for **
A. Gasparian
PAC33, January 15, 2008
4
Primakoff Method
12C
ρ,ω
target
Primakoff
d 3 Pr
8Z 2  3 E 4
2
2
 
F
(
Q
)
sin

e
.
m
.
3
4
d
m Q
Nucl. Coherent
Interference
Nucl. Incoh.
Challenge: Extract the Primakoff amplitude
A. Gasparian
PAC33, January 15, 2008
5
Previous Primakoff Experiments
 DESY (1970)
 bremsstrahlung  beam,
E=1.5 and 2.5 GeV
Targets C, Zn, Al, Pb
 Result: (0)=(11.71.2) eV
10.%
 Cornell (1974)
 bremsstrahlung  beam
E=4 and 6 GeV
 targets: Be, Al, Cu, Ag, U
Result: (0)=(7.920.42) eV
5.3%
 All previous experiments used:
 Untagged bremsstrahlung  beam
 Conventional Pb-glass calorimetry
A. Gasparian
PAC33, January 15, 2008
6
PrimEx Experiment
 Requirements of Setup:
 high angular resolution (~0.5 mrad)
 high resolutions in calorimeter
 small beam spot size (‹1mm)
 Background:
 tagging system needed
 Particle ID for (-charged part.)
 veto detectors needed
 JLab Hall B high resolution, high
intensity photon tagging facility
 New pair spectrometer for
photon flux control at high
intensities
 New high resolution hybrid multi-channel calorimeter (HYCAL)
A. Gasparian
PAC33, January 15, 2008
7
PrimEx Milestones
 Proposal approved in 1999 by PAC15,
re-approved by PAC22 (E02-103) in 2002 with A rating.
 In 2000, NSF awarded a collaborative MRI grant of $1 M
to develop the experimental setup.
 Full support of JLab (Engineering group, machine-shop,
installation, etc.).
 In 4 years the experimental setup, including procurement of all
hardware, was designed, constructed and tested.
 Commissioning and data taking was performed in Sept-November
2004 run.
 Preliminary results released at the April, 2007 APS meeting
with AIP press conference.
 First publication is expected in Spring, 2008.
A. Gasparian
PAC33, January 15, 2008
8
Luminosity Control: Pair Spectrometer
Measured in experiment:
 absolute tagging ratios:
 TAC measurements at low intensities
 relative tagging ratios:
 pair spectrometer at low and high
intensities
Scint. Det.
 Uncertainty in photon flux at
the level of 1% has been reached
 Verified by known cross
sections of EM processes
 Compton scattering
 e+e- pair production
A. Gasparian
PAC33, January 15, 2008
9
Electromagnetic Calorimeter: HYCAL
 Energy resolution
 Position resolution
 Good photon detection efficiency
@ 0.1 – 5 GeV;
 Large geometrical acceptance
PbWO4 crystals
Pb-glass
HYCAL
only
A. Gasparian
resolution
budget
Kinematical
constraint
PAC33, January 15, 2008
10
0 Event selection
We measure:
 incident photon energy: E and time
 energies of decay photons:
E1, E2 and time
 X,Y positions of decay photons
Kinematical constraints:
 Conservation of energy;
 Conservation of momentum;
 m invariant mass
Three groups analyzed
the data independently
A. Gasparian
PrimEx Online Event Display
PAC33, January 15, 2008
11
Differential Cross section
Experimental Yield
per 
A. Gasparian
GEANT:
 acceptances;
 efficiencies;
 resolutions;
PAC33, January 15, 2008
Diff. cross section
12
Fit to Extracted 0 Decay Width
 Theoretical angular
distributions smeared with
experimental resolutions
are fit to the data
 Combined average from
three groups:
Γ(0)  7.93 eV  2.1%(stat.)  2.0% (syst)
A. Gasparian
PAC33, January 15, 2008
13
Fit to Extracted 0 Decay Width:
208Pb Target
 Theoretical angular
distributions smeared with
experimental resolutions
are fit to the data
A. Gasparian
PAC33, January 15, 2008
14
0 Decay width (eV)
PrimEx Current Result
A. Gasparian
() = 7.93eV2.1%2.0%
±1.%
PAC33, January 15, 2008
15
Estimated Systematic Errors
Type of Errors
Errors in current data
Errors in this proposal
Photon flux
1.0%
1.0%
Target number
<0.1%
<0.1%
Background subtraction
1.0%
0.4%
Event selection
0.5%
0.35%
HYCAL response function
0.5%
0.2%
Beam parameters
0.4%
0.4%
Acceptance
0.3%
0.3%
Model errors (theory)
1.0%
0.25%
Physics background
0.25%
0.25%
Branching ratio
0.03%
0.03%
Total
2.0%
1.3%
A. Gasparian
PAC33, January 15, 2008
16
0 Forward Photoproduction off Complex
Nuclei (theoretical models)

Coherent Production A→0A
Leading order processes:
(with absorption)

Primakoff
Next-to-leading
order:

(with absorption)
0 rescattering
A. Gasparian
Nuclear coherent
PAC33, January 15, 2008
Photon shadowing
17
Γ(0) Model Sensitivity
 Incoherent Production A→0A´
 Two independent approaches:
 Glauber theory
 Cascade Model (Monte Carlo)
 Photon shadowing effect
Deviation in Γ(0) less than 0.2%
 Overall model error in Γ(0) extraction is controlled at 0.25%
A. Gasparian
PAC33, January 15, 2008
18
Luminosity Control:
Pair Production Cross Section
Theoretical Inputs
to Calculation:
 Bethe-Heitler
(modified by nuclear form
factor)
 Virtual Compton scattering
 Radiative effects
Atomic screening
Electron field pair production
A. Gasparian
Experiment/Theory = 1.0004
PAC33, January 15, 2008
19
  e     e
Δσ/ΔΩ (mb/6.9 msrad)
Verification of Overall Systematics:
Compton Cross Section
Compton Forward Cross Section
0.085
Klein-Nishina
Primex Compton Data
Uncertainties:
Statistical
Systematic
0.080
P R E L I M I N A R Y
0.075
0.070
0.065
Data with radiative corrections
0.060
0.055
4.9
5.0
5.1
5.2
5.3
5.4
5.5
Energy (GeV)
10
Uncertainties:
No Deviation
Experiment / Theory
Statistical
P R E L I M I N A R Y
5
Deviation (%)
 Average stat. error:
0.6%
 Average syst. error:
1.2%
Experiment To Theory Comparison
0
-5
 Total: 1.3%
-10
4.9
5.0
5.1
5.2
5.3
5.4
5.5
Energy (GeV)
A. Gasparian
PAC33, January 15, 2008
20
Beam Time Request
12C
target
208Pb
7 days
target
6 days
Compton and pair prod.
4 days
Empty target
2 days
Calibrations/checkout
6 days
HYCAL config. change
2 days
Tagging efficiency
1 days
Total
28 days
 Will provide 0.44% Primakoff statistics on each target
 Response to TAC comment on statistical error:
2% statistical error on current result
Primakoff stat. (1.46%)
+ fit error
 0.44% error from proposal requires (1.46/0.44)2 = 11 times more events:
 Increase DAQ rate from 1 kHz to 5 kHz
 Implement more stringent trigger, (factor of 2-3)
A. Gasparian
PAC33, January 15, 2008
21
Summary
 The 0 lifetime is one of the few precision predictions of QCD.
 Percent level measurement is a stringent test of QCD at these
energies.
 High resolution, high intensity tagging facility together with recent
developments in calorimetry make the Primakoff method the
best way to reach the required accuracy in 0 decay width.
 A high resolution experimental setup including an EM calorimeter and pair
spectrometer has been designed, developed, constructed and commissioned
with first physics run in Fall, 2004.
 Compton and pair-production cross section measurements demonstrate that
the systematic errors are controlled at the 1.3% level.
 The experimental setup is capable of percent level cross section
measurements. Collaboration has developed required expertise.
 Control of model error in 0 lifetime at 0.25% level has been reached.
 Our first result: Γ(0)  7.93 eV  2.1% (stat.)  2.0% (syst.)
 Requesting 28 days of beam time to reach the goal of 1.4% on 0 life time.
A. Gasparian
PAC33, January 15, 2008
22
 “… There is nothing new to be discovered in physics now. All that
remains is more and more precise measurement.”
William Thomson (Lord Kelvin), 1900
A. Gasparian
PAC33, January 15, 2008
23
The End
A. Gasparian
PAC33, January 15, 2008
24
0 Event selection (cont.)
Three groups analyzed
the data independently
A. Gasparian
PAC33, January 15, 2008
25
Model dependence of Γ(0) Extraction
Model error in Γ(0) Extraction can be controlled at < 0.25%
A. Gasparian
PAC33, January 15, 2008
26
Some results on Coherent Production A→0A
• Electromagnetic form factors
12C
E=5.2 GeV
208Pb
• Strong form factors
208Pb E=5.2 GeV
Without shadowing
With shadowing
A. Gasparian
PAC33, January 15, 2008
27
Estimated Systematic Errors on Compton
(preliminary)
Photon flux
1.0%
Target thickness (+impurity)
0.05%
Coincidence timing
0.03%
Coplanarity
0.065%
Radiative tail cut
0.098%
Geometric cuts stability
0.65%
Background subtraction
0.40%
Yield fit stability
0.063%
Total
A. Gasparian
1.27%
PAC33, January 15, 2008
28
PrimEx Collaboration
North Carolina A&T State University
Idaho State University
Jefferson Lab
Catholic University of America
CIAE Beijing, China
Beijing University, China
ITEP Moscow, Russia
Duke University
Northwestern University
University of Sao Paulo, Brazil
RIKEN, Japan
USTC, China
George Washington University
A. Gasparian
University of Massachusetts
University of North Carolina Wilmington
MIT
Arizona State University
Norfolk State University
Lanzhou University, China
IHEP Protvino, Russia
Kharkov Inst. of Physics and Tech. Ukraine
IHEP, China
Yerevan Physics Institute, Armenia
JINR Dubna, Russia
Hampton University
PAC33, January 15, 2008
29
Compton as Stability Control
(maybe to question section)
Compton Cross Section Time Stability
0.32
σ (mb)
Uncertainties:
0.31
Theory
PrimEx Compton Data
Statistical
P R E L I M I N A R Y
0.30
0.29
0.28
0.27
2% Band
0.26
0.25
E = 5.220 GeV
A. Gasparian
PAC33, January 15, 2008
19
N
O
V
14
N
O
V
09
N
O
V
N
O
V
02
0.24
30
Primakoff Method
ρ, ω
d 3 Pr
8Z 2  3 E 4
2
2
 
F
(
Q
)
sin

e
.
m
.
3
4
d
m Q
Challenge: Extract the Primakoff amplitude
A. Gasparian
PAC33, January 15, 2008
31
Compton Cross section
Compton Total Cross Section
0.32
0.31
0.30
Theory
PrimEx Compton Data
Uncertainties:
Statistical
Systematic
P R E L I M I N A R Y
0.29
0.28
0.27
0.26
0.25
0.24
0.23
4.9
5.0
5.1
5.2
5.3
5.4
5.5
Energy (GeV)
A. Gasparian
PAC33, January 15, 2008
32
Compton Cross section: Theory
Pure QED process: Calculable at the percent level
 Leading Order: Klein-Nishina
 Corrections to LO:
 Rad. correction (virtual/soft)
 Double Compton (hard emiss.)
 Klein-Nishina + full rad. Corr.
(Monte Carlo Method)
 Klein-Nishina + full rad. Corr.
(Numerical Integration Method)
A. Gasparian
PAC33, January 15, 2008
33
Trigger Improvement
A. Gasparian
PAC33, January 15, 2008
34
Impact of Giant Excitation of Nucleus on 0
Primakoff production
•
With nuclear collective excitation, the longitudinal momentum transfer
in 0 photo-production is Δin= Δ+Eav, where the average excitation
energy Eav for 12C is ~20-25 MeV.
•
The ratio of the cross section of the 0 photo-production in the
Coulomb field with nuclear excitation to “elastic” electromagnetic
production can be estimated as:
d in
(q 2  2 ) 2
7
12
d  1.4 N

10
at
Primakoff
peak
(q


)
for
C
2
d el 2m p ZAEav (q 2   in )
d
•
Nuclear Giant Excitation effect for lead is small as well.
A. Gasparian
PAC33, January 15, 2008
35