Title: Study of spin structure of nucleon in COMPASS - measurement of ?G/G
1Study of spin structure of nucleon in COMPASS -
measurement of ?G/G
- Kaori Kondo
- University of Bochum
- On behalf of the COMPASS collaboration
2Structure of Nucleon
quark
gluon
orbital angular momentum
A small fraction of nucleon spin is carried by
quarks!! Where does the rest of the nucleon spin
come from?
naïve parton model ?S1 CERN, SLAC, DESY,
JLAB ?S 0.30
large ?G ??G(x) ?
g(x) Probability to detect spin (spin-) gluon
in spin proton as a function of Bjorken x.
3PGF probes gluons
Direct measurement of ?G is provided by Photon
Gluon Fusion (PGF) in polarized DIS
- Double spin asymmetry
- Experiment measure
Pb beam polarization
Pt target polarization
f dilution factor
4COMPASS collaboration
COmmon Muon and Proton Apparatus for Structure
and Spectroscopy
Bielefeld, Bochum, Bonn (ISKP), Bonn (PI),
Burdwan and Calcutta, CERN, Chubu, Dubna (LPP
and LNP), Erlangen, Freiburg, (Heidelberg),
(Helsinki), KEK, Lisbon, Mainz, Miyazaki, Moscow
(INR), Moscow (LPI), München (LMU), München
(Technische Universität), (Nagoya), Prague,
Protvino, Saclay, Tel Aviv, Torino (University
and INFN), Trieste (University and INFN),
Warsaw (SINS), Warsaw (TU), Yamagata
To study hadron structure and hadron spectroscopy
- 240 physicists from over 28 Institutes
- start taking data in 2002
DG/G, DS, Transversity, Pentaquark,
Exclusive ?0, ? physics
5COMPASS_at_CERN
COMPASS
6COMPASS Spectrometer
- 2 stages spectrometer
- with 2 magnets (SM1, SM2)
- Tracking (SciFi, Silicon, MicroMega, GEM, MWPC,
DC, Straw) - PID (RICH, ECAL, HCAL, muon filter)
Trigger Hodoscopes
m filter
MWPC
ECalHCal1/2
Straws
SM2
Drift chamber
RICH
Target
SM1
SciFi, Silicon
m beam
MicroMegas
7(No Transcript)
8Polarized beam and target
simultaneous measurement
- 160 GeV µ beam
- 2 108 µ/spill (4.8 s / 16.2 s)
- Pb 76 longitudinally naturally polarised
- luminosity 5 1032 cm-2 s-1
Spin orientation is reversed to cancel out the
systematic error from detector acceptance
difference between the cells
- by magnetic field rotation (3 times /day)
- by DNP microwave frequency change
- deuteron target (6LiD)
- Pt 50 longitudinally polarised by DNP
- cell size f30mm x 600mm long
- 2 or 3 cells oppositely polarized at the same
time - spin orientation can be reversed
target
µ beam
target
µ beam
9Accessing ?G/G
There are 2 ways to tag PGF process
q c OPEN CHARM charmed meson production
event ? no background asymmetry, less MC
dependent. ? small statistics q u,d,s HIGH
pT HADRON PAIRSevents with 2 hadrons with high
pT ? Large statistics ? physical background
model (MC) dependent,
Leading order analysis in the moment..
10Open Charm
PGF is the main reaction for charmed pair
production in DIS, Most frequently, D mesons are
produced in the independent fragmentation of c
pair.
(BR 4)
- Kaon identification with RICH
- up to 50 GeV/c
- Kinematical cut to suppress combinatorial
background - z(D0)gt0.25 zED/n
- cosqKlt0.5
(BR 68)
- 30 D0 comes from D decays
- combinatorial background highly suppressed
11D tagging
12Reconstructed D mesons
Effective signal
13Extracting ?G/G
- f dilution factor 0.4
- Pb beam polarization 0.8
- Pt target polarization 0.5
- S/(SB) determined from fit
- Parameterization to calculate aLL from kinematics
with one reconstructed meson - Neural network trained with AROMA Monte Carlo
- Strong correlation between NN aLL and AROMA aLL
correlation factor 82
14?G/G result from open charm
D0 D result 2002 2004
(prelim.)
systematic error under investigation no clear
systematic effect seen
NLO analysis, resolved photon contribution study
planed.
15High pT Hadron Pairs
PGF in the other method..
- High Q2 analysis
- Q2gt1 (GeV/c)2
- low statistics (10)
- LEPTO-MC
- Low Q2 analysis
- Q2lt1 (GeV/c)2
- additional resolved photon background events
- high statistics
- PYTHIA-MC
16High pT Hadron Pairs Q2 gt1(GeV/c)2
PGF
QCD compton
LO DIS
17High pT Hadron Pairs Q2 gt1(GeV/c)2
- pT1, pT2 gt 0.7 GeV/c, SpT2 gt 2.5 (GeV/c)2
- 0.1 lt y lt 0.9
- small x region A1d is small, LODIS and QCDC
neglected - Monte Carlo generator LEPTO
2002-2003 data result
(prelim.)
systematic error false asymmetry mainly
contributes
18High pT Hadron Pairs Q2 lt1(GeV/c)2
Resolved photon
Direct
neglect
neglect
MC event generator PYTHIA is used for low Q²
analysis
19High pT Hadron Pairs Q2 lt1(GeV/c)2
Contribution from resolved photons unknown
polarised PDFs of the photon known unpolarised
PDFs of the photon ? use unpolarised PDFs to
constrain polarised PDFs
- Adds a limited uncertainty
- to the estimation of (?G/G)(xg)
- ?G/G
- 0.010 (minimum), 0.023 (maximum)
20High pT Hadron Pairs Q2 lt1(GeV/c)2
- Systematic error mostly comes from PYTHIA tuning
2002-2003 data result
publishedPLB 633 (2006) 25-32
2002-2004 data result
(prelim.)
systematic error exp. syst. (0.014), MC syst.
(0.052), photon contribution (0.013)
21?G/G Summary
GRSV200 Glück et al.,Phys. Rev. D63(2001)
094005
22- P.Liebing, RBRC
- Spin 2006, Kyoto
23Summary
- COMPASS presents new results of ?G/G using 2002
2004 data in three different methods - open charm channel
- High pT hadron pairs Q2gt1 (GeV/c)2
- High pT hadron pairs Q2lt1 (GeV/c)2
- Small ?G is preferred