A Precision Measurement of the Neutral Pion Lifetime via the Primakoff Effect

1
A Precision Measurement of the Neutral Pion
Lifetime via the Primakoff Effect

• Eric R. I. Clinton
• University of Massachusetts Amherst
• On behalf of PrimEx Collaboration
• (Jefferson Lab Experiment E99-014)
• Duke University Seminar
• May 31, 2007

2
OutlineWhere Im going, and how I intend to get
there.

• Who we are and a brief history of PrimEx
• Physics
• Theory and Motivation
• Some of World Data
• Photo-nuclear and other physics measured by
  PrimEx
• Compton, pair production as calibration reactions
• Experimental Set-up
• Im an experimentalist. Go figure.
• Data Analysis
• Calibration reaction results
• How other analysis groups looked at the pion data
• My analysis
• Simulation
• Acceptance correction, resolution effects
• Results
• Radiative width
• Error Evaluation
• Summary of collaboration p0 width measurements

3
The PrimEx Collaboration
4
History of the PrimEx collaboration

• January of 1999, the TJNAF PAC approved E99-014
  with an A- rating. E99-014 is more commonly
  known as PrimEx.
• In June of 2002, the PAC upgraded E99-014 from
  and A- to A rating.
• PrimEx's main goal
• The measurement of the neutral pion decay width.
• Data was collected in Fall 2004 on 208Pb and 12C
  targets
• 12C results will be shown.
• Measure Primakoff cross section to 1.5 or better.

5
The neutral pionas opposed to the partisan pion

• p0 discovered by Bjorklund, Crandall, Moyer and
  York at Berkeley Cyclotron in 1950.
• Pre-QCD, Gp0?gg 0
• Primakoff Effect
• Predicted Henry Primakoff (Phys. Rev 81, 899,
  (1951))

The Primakoff Effect Photo-pion production in
the Coulomb field of a high Z nucleus
6
Physics Motivation

• Adler, Bell, Jackiw and Bardeen discover triangle
  diagrams that alter PCAC predictions for p0 decay
• Post QCD, O(p4) anomalous Lagrangian is
  constructed by Wess, Zumino and Witten.
• This permits transitions between even and odd
  numbers of pseudo-scalars
• The (zero order, m u,d 0) decay width of the ?o
  ? ??.?
• Real world quark masses are not -0- MeV
• Mass of u,d quarks on order of 5-7 MeV.

7
Corrections Zero-eth order width 7.725 eV

• Adler and Bardeen -- Phys. Rev. 182, 1517-036
  (1969)
• Non-renormalization theorem gives quark mass
  correction
• Bijens Prades, Z.Phys. C64 (1994), 475
• ??and ? mixing -- 2-3 increase in width
• Recent theoretical calculations
• Goity, Bernstein, Holstein -- Phys. Rev. D66,
  076014, 1-10 (2002)
• NLO Theory Calculation -- 8.1 0.081 eV.
• Ananthanaravan Moussallam -- JHEP 0205 (2002)
  052
• NLO Theory Calculation -- 8.06 0.08 eV.
• For comparision, PDG width 7.84 eV.
• Thus, ?(?o ? ??) is the most accurate prediction
  in QCD, depending only on the number of colors.

8
World Data
9
Photo-nuclear processes in PrimEx

• Total Cross Section --

• Primakoff
• Coherent
• Incoherent
• Interference
• See last term in Total CS

--Fe.m.(Q) pion electromagnetic form factor,
FN(Q) is the nuclear matter dist. form factor
(both corrected for FSI). Csin2qp is the
isospin and spin dependent part of p0
photoproduction in a single nucleon, 1-G(Q)
reduces cross section at small momentum transfer
(Pauli principle), and dsH/dW is p0
photoproduction on a single nucleon.
10
Compton and Pair Production

• Atomic Compton cross section
• Compton scattering off atomic electrons
• Klein-Nishina, radiative corrections, all in hand
• Calibration reaction
• Use HyCal to detect and measure photons and
  electrons
• Significant result in unto itself
• New 5 GeV Compton measurement
• Pair production
• Measure pair production during physics runs
• Pair production rate is a monitor of relative
  photon flux
• Another Calibration reaction
• Use HyCal to detect e- and e co-incidences

11
The Jefferson Lab AcceleratorIn Newport News, VA
12
The Jefferson Lab Accelerator
13
Conceptual PrimEx set-up

• Electron beam enters the tagger
• Bremms. off a thin gold radiator
• Electron tagged, photon energy known to 0.1
• Photon beam interacts with the target
• Produces p0s, pairs, Compton, signals
• Sweep magnet can select signal going to HyCal
• Magnet fully onp0
• Magnet off Compton
• Partially on pairs
• Physics signal travels thru helium bag
• Detection in HyCal of signal
• Charged particle veto can help sort out some of
  data

14
Photon TaggerPost bremmstrahlung electron
momentum analyzer

• 384 E- counters provide energy information
• Accessing only the highest energy photons
  (5.6-4.9 GeV)
• 0.1 energy resolution for PrimEx tagged photon
  energy range.
• 61 T-counters, provide timing information
• Only the highest energy 11 T-counters and 56
  E-counters active.
• Sub nanosecond timing resolution
• 1 or better uncertainty photon flux
  (luminousity)
• Flux 107 photons/second

15
The BullseyeWhat are we shooting all these
photons at?

• 12Carbon Target
• Highly Order Vacuum Deposited Pyrolytic Carbon
• Mouthful for VERY homogeneous, pure 12Carbon
• At UMass, PrimEx targets were studied
• Carbon (1in x 1in x .38 in)
• Micrometer and water displacement density
  measured
• Outsourced elemental analysis
• Result --
• Pyrolytic Carbon is VERY homogenous, pure
  12Carbon
• Error on atoms/cm2 0.04

The Carbon targets
Electron micrograph of carbon target
16
Pair Spectrometer

• Pair Spectrometer constructed for PrimEx and Hall
  B
• The PS must have a flat acceptance curve over
  PrimEx energy range
• No acceptance gaps
• Dipole magnet will sweep pairs from the physics
  targets into the PS
• Pair production rates will be used to calibrate
  relative photon flux.
• Total Absorption Counter for absolute flux
  measurement--calibration

17
The Hybrid Calorimeter

• Highly segmented array of lead tungstate and lead
  glass crystals
• 7.3 meters downstream of the targets.
• The interior array of crystals
• 1152 lead tungstate modules
• 2.05 x 2.05 x 18 cm3, 20 Xo, and 2.0 cm Moliere
  radius.
• The outer array
• 576 lead glass modules
• 3.84 x 3.84 x 45 cm3 and 17 Xo and a Moliere
  radius of 3.6 cm
• Small detector size
• Fine angular resolution ( 0.02o)
• Energy resolution
• 4.5 for lead glass
• 1.2 for lead tungstate.
• Identify multi-photon backgrounds
• Charge Particle Veto Counters
• Offline removal of charged events

18
Data AnalysisWe've had data for 2 years now...
And all I'm getting is this lousy Ph.D

• Lets define a few key terms...
• Photon Flux
• Number of Photons on physics target
• Tagger Energy
• Energy of photon inferred from bremm'ed electron
• Cluster Energy
• Energy deposited in/measured by HyCal in event.
• Invariant Mass
• gg Invariant mass from reconstructed 4-vectors of
  decay photons
• Elasticity
• ratio a cluster pair energy sum and tagger energy
• Hybrid Mass
• Projection of 2-D elasticity and invariant mass
  onto new axis

19
Photon fluxSince its the largest bit of our
error budget
20
ComptonSignal and radiative effects
COMPTON SIGNAL
RADIATIVE EFFECTS IN THE DATA
21
ComptonPreliminary results with radiative
corrections

• Summary of Compton analysis
• Agreement with theory at the level of few
• Work in progress to reduce the systematic errors
  to 1 2 level

Radiative corrections I. virtual loops, and soft
double Compton scattering, Brown and Feynman II.
hard double Compton scattering Mork, and Mandl
and Skyrme
22
Pair productionSignal and differential cross
section
Theory by Alexandr Korchin

• Summary of pair analysis
• Agreement with theory 3.8
• Work in progress to reduce the systematic errors
  to 1 2 level

23
I. Larins Approach--Kinematic fitting to the
condition Elasticity 1
24
D. McNultys approach --Elastic and Inelasltic
yield extraction
To get elastic yield, project data below onto
Invariant Mass axis a function of p0 production
angle
Then extract the Inelastic yield by
fitting slices in elasticity and plotting
extracted yield
25
My Method

• Project events onto axis perpendicular to the
  kinematic correlation between Elasticity and Mgg
  for elastic events
• Fit peak in mass distribution

26
Sample "Hybrid Mass" fits
Interference Region
Primakoff Region
Coherent Region
Interference Region
27
Preliminary p0 Cross SectionsEfficiency/Acceptanc
e Uncorrected
PRELIMINARY
28
Monte CarloGambling time

• MC used efficiency calculations
• Theoretical lineshapes thrown at simulated HyCal
• Primakoff, nuclear coherent and incoherent, and
  interference
• MC data "conditioned" to look like physical data
• Energy smearing, electronic noise, resolutions
  built in
• Then, the data got the "full treatment"
• Run thru event selection, same cuts...
• Geometric and Reconstruction Efficiencies
• As function of photo-nuclear process and p0 angle

29
Efficiency Corrected Differential Cross
SectionsMine, not the other guys'
30
Larin and McNultys results

• Larins Results
  

• D. McNultys Results

31
Ta-da!!p0 radiative width 7.88 eV 2.05
(Stat) 2.00 (Sys)
PRELIMINARY
McNulty 7.929 eV 1.6 (Stat) 2.4 (Sys)
Larin 8.00 eV 2.1 (Stat) 2.5 (Sys)
32
Estimated Systematic Errors (preliminary)
Combined from two analysis groups a Experimental
b Analysis c Theoretical input

33
Summary
?(?0???) 7.93eV ? 2.1 (stat) ? 2.0
(syst) Lifetime (8.200.24)x10-17 sec PDB
average (8.40.6)x10-17 sec
34
Future worktowards a final result and publication

• Adding HyCal Lead Glass detectors into analysis
• Incoherent process
• Discussion regarding shape of incoherent
• Evolve cross section to single photon energy
• Data collected over a range of photon energies
• Photon flux is known as a function of energy to
  0.1
• Determine final result
• Finish writing and defend dissertation

35
The 400 lb. gorilla would like to extendThanks
to

• Jennifer Clinton
• Obligatory thanks to my wife for putting up with
  a graduate student salary
• George and Susan Clinton
• My parents
• Rory Miskimen
• Advisor
• David Lawrence
• Mentor
• Mike Wood (formerly UNC-TUNL)
• Mentor

36
Support provided

• by the DOE and Jefferson Lab
• in part by NSF MRI grant PHY-0079840
• in part by RFBR Grant 04-02-17466

37
Extra slides
38
Adding HyCal Lead glass
39
Geometric acceptance
PRELIMINARY