Nucleon Spin Structure at Very Low Q2 From EG4 Experiment

1
Nucleon Spin Structure at Very Low Q2 From EG4
Experiment

• Krishna Adhikari
• Physics Department
• Old Dominion University
• Dec. 11, 2007

2
Outline

• Introduction and Motivation
• Description of the Experiment
• My Work
• EC-timing calibration
• Raster correction
• Future work
• Summary

3
Motivation

• 1933 Discovery of nucleon anomalous magnetic
  moments
• Measurement (?p 2.79 ?N ?n - 1.91 ?N) with
  the Diracs
• Dirac prediction (?N 3.152510-14 MeV/T and 0)
  
• The first concrete signature of nucleon
  substructure.
• Interest in nucleon structure begins.
• Later experiments at powerful accelerators
  provided independent confirmations.

4
Scattering as a probe

• ? E E energy transfer
• Q2 -q2 (p2-?2) 4EEsin2(?/2) resolution
• W2 M22M?-Q2 (Final state invariant mass)2
• x Q2/(2M?) Bjorken scaling variable 0ltxlt1
  (elasticity momentum fraction carried by struck
  quark)

• Photon and lepton scattering - a very powerful
  the predominant method to probe the tiny
  composite systems such as the nucleons.

• Vast DIS data of last 40 years (all available
  target-beam types)
• Initial (DIS) SLAC data confirmed the
  quark-parton picture of the nucleon.
• Many later experiments - more precision and size
  of our knowledge of nucleon structure improving.
• Some surprising results too e.g., original
  EMC-Effect, violation of Gottfried sum rule,
  even some hints for quark substructure.

5
Nucleon Spin Thru the high Q2 microscope Spin
Crisis

• Spin-averaged quark structure studied and
  understood a lot .
• Spin-structure - not much known polarization
  techniques not available before.
• Naïve Parton Model (NPM) predicts Quarks give
  60 of the nucleon spin
• Last 3 decades - great advances in polarization
  technology.
• Many subsequent experiments extracted g1 and g2
  (functions of the quark-spin distribution)
• First experiments at SLAC (limited kinematics)
  seemed to confirm the NPM predictions.

• However, later, EMC experiment (at CERN
  published in 1988) (higher precision and wider
  kinematics) reported quarks share only 12
  (practically none) of the nucleon spin.

• This Spin Crisis sparked a large interest in the
  spin content measurement and related theoretical
  works.

6
Nucleon Spin in high Q2 .

• Subsequent theoretical advances in QCD clarified
  spin-picture more
• Bjorken sum rule is a precise test of QCD.
• Interpretation of existing DIS results
• verified the Bjorken sum rule with 10 accuracy
  
• only 30?10 of the nucleon spin is due to
  quarks.
• The rest is expected to be due to gluons and/or
  the orbital motions of the constituents, but has
  not been easy to meausre.

• Experiments to measure the gluon contribution
  are underway at RHIC (at BNL) and CERN.

7
Probing at the other end of the energy scale

• Low resolution information on long distance
  structure, static properties.
• spin crisis led Anselmino et al. to examine
  the previously unappreciated GDH sum rule
  (formulated in 1960s)
• The sum rule is connected to the DIS region as
  an analytic extension of Bjorken sum rule towards
  the real photon point.
• It implies a negative slope (w. r. t. Q2) of ?1
  at the photon point.
• Later, Burkert et al. - rapid transition of ?1
  between the real photon point and the DIS region
  is saturated by contributions from nucleon
  resonances.
• Then Ji et al. extended the GDH sum rule beyond
  the real photon point.
• This theoretical progress triggered a large
  interest in testing those predictions
• A large experimental spin-program underway at
  Jlab (includes EG4) and elsewhere.

8
GDH Sum rule

• Sum rules relations linking an integral over
  structure functions to quantities characterizing
  the target.
• Windows into the target structure and tools to
  test QCD.
• Some polarized sum rules Bjorken sum rule,
  Burkhardt-Cottingham sum rule, Ellis-Jaffe sum
  rule and Gerasimov, Drell and Hearn (GDH) sum
  rule.
• GDH Sum Rule (derived in 1966) connects static
  properties of the nucleon with the spin dependent
  absorption of real photons with total cross
  sections ?3/2 and ?1/2

?th pion production threshold
9
Test of GDH Sum rule and its Extension

• GDH-Collaboration at two accelerators ELSA and
  MAMI verified it for the first time.
• The "sum" on the left hand side of the GDH Sum
  Rule generalized to the case of virtual photons
  (i.e. Q 2gt 0)

where, K the flux factor of virtual photons ?
v 1?2 ?2 Q2/ ?2 and x0 Q2/ 2M ?th This
reduces to the GDH sum rule for Q20. In the DIS
limit the integral becomes
Where G1 is the first moment of g1. In the Q2 ?
0 limit also, the integral will take the same
form.
10
Studying the GDH Integral at various Q2

• See the transition from the high to low Q2
  regimes of QCD test predictions of ?PT, and
  phenomenological models
• Particularly interesting - the change of sign
  somewhere in (0 lt Q2 lt 1) GeV2
• Subject of several experiments (HERMES experiment
  at DESY for higher Q2 EG1a, EG1b and EG4 at JLab
  for the resonance region)

?PT
11
EG4 Experiment (E03-006)

• Data taken from February to May 2006.
• Goal Measurement of the extended GDH integral on
  the proton and deuteron at low Q2 (0.01 0.5
  GeV2).
• Our method - measure the helicity dependent
  cross-section difference

• How to measure the absolute cross-section
  difference?

N-, N ? the of events detected for the
opposite and same beam-target helicitites Ni, t,
f, ?? and PbPt ? the of incident electrons
(Faraday cup), target areal-density, the detector
acceptance, detector efficiency and the product
of beam-target polarizations respectively.

• How to extract g1?

• g2 is negligible at very low Q2 values, extract
  g2, then evaluate ?1, and the GDH sum .

12
EG4 Experimental Setup

• Target
• Standard cryogenic polarized NH3 and ND3 targets
  (Dynamic Nuclear Polarization Maintained at 1K
  and 5T longitudinal magnetic field.
• 12C and an empty cell targets for background
  data
• Positioned 1m upstream of the usual CLAS center
  to enhance low Q2 coverage.

Beam Longitudinally polarized (85 87) with
energies (3.0, 2.3, 2.0, 1.5, 1.3, 1.0) GeV
Detector CEBAF Large Acceptance
Spectrometer Used in Standard Configuration with
a few modifications.
13
CLAS Detector (CEBAF Large Acceptance
Spectrometer)
14
Gas Threshold Cherenkov Counter (CC)

• For triggering on electrons and separating
  electrons from pions .
• Single sector of CLAS used for inclusive trigger.
• Cross section measurement requires uniform
  detection efficiency at low Q2
• For EG4, a new CC in the 6th sector, designed and
  built by INFN Genova, Italy.
• Very high and uniform electron detection
  efficiency (99.9), a high pion rejection ratio
  (of the order of 10-3).

Old CC design
New CC
C4F10 perfluorobutane n 1.00153 Ppigt2.5 Gev/c
Overhead View of CLAS
15
Forward EM calorimeters (EC)

• Sampling, shower calorimeters (for energy and
  position of neutrals)
• Alternating (39) layers(10 mm) of scintillator
  (SC) lead sheets (2.2 mm) total thickness - 16
  radiation lengths.
• Each SC layer (36 parallel strips, strips rotated
  by 120? in each successive layer)
• 3 views U, V, W each of 13 layers for stereo
  information. (5 inner, 8 outer stacks, for
  improved hadron discrimination) (electron-pion
  rejection is 0.01.)
• Intrinsic energy resolution for showering
  particles 10/?E, 3 cm position resolution at 1
  GeV up to 60 detection efficiency for high
  momentum neutrons.
• Main functions
• Detection and primary triggering of electrons at
  energies above 0.5 GeV.
• Detection of photons at energies above 0.2 GeV.
• Allowing ?0 and ? reconstruction from the
  measurement of their 2? decays
• Detection of neutrons, with discrimination
  between photons and neutrons using TOF
  measurements.

16
My Work EC-timing Calibration

• What? Adjustment of EC time against SC time.
• Why? EC time is very critical in identifying
  neutral particles.
• What factors affect EC time?
• Systematic changes in the time response of EC and
  Sc over time due to hardware changes (eg.
  Replacement of cables, PMTs etc.) between
  different experiments.
• Also, calibrations of different detector systems
  are somewhat interconnected. (eg. in our case,
  the EC time is calibrated w.r.t. TOF Similarly
  TOF w.r.t.RF signals coming from the accelerator.
  Changes in other systems can affect the
  calibrations of the TOF and EC.

17
EC-timing

• How?
• Neutral particles do not give strong signals in
  Scintillation Counters (SC), therefore we must
  choose particles which give enough signal both in
  SC and EC to calibrate the time information. The
  charged particles are the best choice.
• Select a large sample of data (1-2 M events).
• Calibrate (EC time SC time) of electrons for
  each SC paddle by chi-squared minimization of the
  time difference (a five parameters model used).

Fig ECt SCt for all charged particles in
sector6, (run 51057)
Time in ns For electrons
?
The average timing resolutions or accuracies
obtained after a few calibration runs is 300
ps for electrons and 400-500 ps for the hadrons.
18
Examples of the use of EC-time
Fig Invariant mass of two photons as
reconstructed from the EC (The clear peak at the
?0 mass, indicates that there was an undetected
?0 in the reaction.)

• Fig The ? spectrum for neutrals EC. The
  peak at 1 is for the photons while the shoulder
  on the left is for the neutrons.

19
My Work Raster Correction

• Why Rastering?
• CEBAF generates High current beam (with a
  transverse dimension of 0.1 mm)
• Beam is rastered to prevent
• over-heating of the target
• local depolarization
• How?
• Two raster magnets ( one for x and the other for
  y direction) move the beam spirally giving it a
  larger transverse size.
• ADCs record the currents in the raster magnets.

20
Raster Correction

• Why Raster correction?
• To make corrections to the tracking (vertex x and
  y is assumed zero by the tracking code) which
  allows
• better rejection of events from up-beam and
  down-beam windows (especially for particles at
  small angles)
• reduction in accidental coincidences in
  multi-particle final states.
• To make correction to the ? angle (improves
  missing mass resolution for multi-particle final
  states)
• Target imaging (to look for mis-steered beam)

21
Raster Correction
Conversion of ADC counts to cm
Target imaging
X0, Y0, cx cy fit parameters.
Exclusive 50808
Z0 a fit parameter that defines the target
center.
Zc corrected vertex position - a function of ?,
?, sector- ? reconstructed py, px

• ? correction
• The track length for a particle in the 5T (50 kG)
  magnetic field of the target is different than
  the tracking code assumed which means that the ?
  rotation is incorrectly calculated.
• This is corrected using the equation ?c ?0
  50 q x /(100 33.356 pt)

22
Future Work

• Momentum Correction
• Background subtraction
• Final definition of all cuts
• Beam and target polarization determination.
• Acceptance and Efficiency of electron detection
  in CLAS (From simulations and comparison with
  known cross sections.)
• Radiative corrections
• Development of models
• Extraction of g1, integration for its moment(s)
• Neutron information extraction.

23
Summary

• As part of an attempt to understand the spin
  structure of nucleons, we plan to measure the
  spin structure function g1, its first moment ?1
  and the extended GDH integral at very low Q2
  (0.01 0.5 GeV2) values using EG4 data on proton
  and deuteron.
• The data is already available, and I worked on EC
  time calibration and Raster correction. EC timing
  is complete, Raster work is underway.
• I will start working on Momentum correction very
  soon.
• My thesis will be on the neutron spin-structure
  (extract g1, ?1 and the extended GDH integral for
  the neutron)

24
References

• (1) Wikipedia Standard Model. (http//en.wikipedi
  a.org/wiki/Standard_Model)
• (2) S. E. Kuhn, Nucleon Structure Functions
  Experiments and Models, HUGS 97.
• (3) S. E. Kuhn and G. E. Dodge, Private
  communications.
• (4) K. J. Slifer, Ph. D. thesis, Temple
  University.
• (5) R. MilnerHERMES physics, a historical
  perspective (A ppt presentation for HERA
  symposium June 30, 2007)
• (6) M. Ripani, Private communication.
• (7) K.G. Vipuli G. Dharmawardane, Ph.D. thesis,
  Old Dominion University
• (8) K.J. Slifer and A. Deur Private
  communications.
• (9) M.Battaglieri, et al. 2003 Jefferson Lab
  proposal E03-006
• (10) http//galileo.phys.virginia.edu/classes/sajc
  lub/gdh.html
• (11) M. Amarian et al., The CLAS forward
  electromagnetic calorimeter, Nucl. Instr. And
  Meth. 460 (2000) 239 265.
• (12) P. Bosted et al., Raster Corrections for
  EG1b, CLAS-NOTE-2003-008.
• (13) http//www.krl.caltech.edu/johna/thesis/node
  19.html
• (14) R. De Vita, Private communications.
• (15) A. Klimenko and S. Kuhn, Momentum
  corrections for E6, CLAS-NOTE-2003-005.
• (16) K. Park et al., Kinematics Corrections for
  CLAS, CLAS-NOTE-2003-012.
• (17) http//people.virginia.edu/xz5y/Research.htm
  l
• (18) M. Anghinolfi et al., The GDH Sum Rule with
  Nearly-Real Photons and the Proton g1 Structure
  Function at Low Momentum Transfer. Jlab PR
  03-006.
• (19) A. Deur, Experimental Studies of Spin
  Stucture in Light Nuclei, EINN07.

25

• CLAS Forward EM calorimeters (EC)
• Sampling, electromagnetic shower calorimeters
• To measure energies and positions of photons and
  neutrons.
• Alternating (39) layers(10 mm) of scintillator
  (SC) lead (Pb) sheets (2.2 mm) total thickness
  - 16 radiation lengths.
• Pb-SC sandwich given the shape of an equilateral
  triangle to match the hexagonal geometry of the
  CLAS. Each SC layer made of 36 parallel strips,
  with the strips rotated by 120? in each
  successive layer.
• Thus 3 orientations/ views (labeled U, V, W each
  of 13 layers) which provide stereo information.

• Each view further divided into inner (5 layers)
  outer (8 layers) stacks, to provide longitudinal
  sampling for improved hadron discrimination (
  electron-pion rejection is 0.01.).

• Each module 36 (strips) 3(views)2(stacks)
  216 PMTs. Altogether 1296 PMTs intrinsic energy
  resolution for showering particles 10/?E, 3 cm
  position resolution at 1 GeV up to 60
  efficiency for detecting high momentum neutrons.
• Has a projective geometry (successive
  layer-area larger) minimizes posi shower
  leakage at the edges of the active volume and
  minimizes the dispersion in arrival times of
  signals originating in different scintillator
  layers forms a truncated triangular pyramid with
  a projected vertex at the CLAS target point (5 m
  away) and base-area of 8 m2. The projective
  geometry to maximize position resolution for
  neutral particles.
• main functions Detection and primary triggering
  of electrons at energies above 0.5 GeV. particles
  or to select a particular range of scattered
  electron energy Detection of photons at energies
  above 0.2 GeV. Allowing ?0 and ? reconstruction
  from the measurement of their 2? decays
  Detection of neutrons, with discrimination
  between photons and neutrons using TOF
  measurements.

26
Examples of the use of EC-time
? of all the neutrals in sector -1 (50808) and
distance (in cm) EC hit from target. The two
bands represent the hits in the inner and outer
parts of the calorimeter.
50808
The ? spectrum for neutrals EC. The peak
at 1 is for the photons while the shoulder on
the left is for the neutrons.
?
The invariant mass of two photons as
reconstructed from the EC (The clear peak at the
?0 mass, indicates that there was an undetected
?0 in the reaction.)
27

• The projective geometry (truncated triangular
  pyramid - projected vertex at the CLAS center
  base 8 m2) to maximize position resolution for
  neutrals.

28
(No Transcript)
29
Introduction and Motivation

• Early 1920s - Investigation of spin begins with
  the Stern-Gerlach expt. and subsequent
  introduction of the concept of spin by Uhlenbeck
  and Goudsmit
• Spin also explains other puzzling findings such
  as the hyperfine splitting in atomic spectra
• Uhlenbeck and Goudsmit Spin - an intrinsic
  property just like mass and charge. (Point like
  electron would need to rotate infinitely fast if
  the classical tiny spinning orb picture were
  true. The spin appears in every way like an
  angular momentum, but it is unrelated to any
  spatial motion.)
• 1928 Diracs theory for spin-1/2 structureless
  particle ? q/2M Agreed with experimentally
  measured electron value. This establishes Diracs
  theory.

30
Introduction

• Initially, ? of nucleons remained unmeasured
  being comparatively too small (because of higher
  masses) and so they were believed to be
  structureless as well.
• 1933 - Stern (improved his apparatus sufficiently
  to measure the much smaller ?proton) and his
  collaborators measured value of ?proton
  disagreed with Dirac prediction by 150.
• The anomalous magnetic moments - the first clear
  indication of nucleons internal structure. (Of
  course, in the case of uncharged neutron, any
  magnetic moment at all would be anomalous.)
• Later experiments provided independent
  confirmations for the nucleons internal
  structure nucleon resonances scaling
  phenomena x lt 1.

31
Beam

• TJNAF electron beam.
• Energy upto 5.7 GeV (injector - 45 MeV 1500 MHz
  bunch structure)
• Hall B.