COMPASS plans to measure GPDs

1
COMPASS plans to measure GPDs
Jean-Marc Le Goff DAPNIA/CEA-Saclay 24 Feb 2006,
Albuquerque workshop on Orbital Angular Momentum

• The COMPASS experiment
• GPDs at COMPASS
• now deep r production
• 2010 DVCS and HEMP

2
The COMPASS experiment
3
The COMPASS Collaboration(230 Physicists from 12
Countries)
Bielefeld, Bochum, Bonn (ISKP PI), Erlangen,
Freiburg, Heidelberg, Mainz, München (LMU TU)
Dubna (LPP and LNP), Moscow (INR, LPI, State
University), Protvino
CERN
Warsaw (SINS), Warsaw (TU)
Helsinki
Prag
CEA-Saclay
Nagoya
Lisboa
Torino(University,INFN), Trieste(University,INFN)
Burdwan, Calcutta
Tel Aviv
4
COMPASS fixed target experiment at
CERN muons
hadrons p/K, p Beam
2 . 108 µ/ spill (4.8s / 16.2s) 2 . 108
h/spill Beam polarisation 80 Beam momentum
160 GeV/c 150-270
GeV/c

COMPASS
SPS

LHC
COMPASS ? COmmon Muon and Proton Apparatus
for Structure and Spectroscopy
5
Physics goals

• hadron beams
• nucleon spectroscopy
• Primakoff-Reactions
• polarizability of ? and K
• Exotics glueballs and hybrids
• charmed mesons and baryons
• semi-leptonic decays
• double-charmed baryons

6
Spectrometer 2002 ? 2004
trigger-hodoscopes
DW45
straws
SM2 dipole
Muon-filter2,MW2
HCAL1
RICH_1
Gem_11
SM1 dipole
ECAL2,HCAL2
MWPC Gems Scifi
Polarised Target
Muon-filter1,MW1
straws,MWPC,Gems,SciFi
Veto
Gems,SciFi,DCs,straws
Silicon
Micromegas,DC,SciFi
SciFi
BMS
7
COMPASS and GPDs
8
measurement of GPDs
Collins et al.
Deeply Virtual Compton Scattering (DVCS)

?
?
Q2

hard
x ?
x - ?
soft


GPDs
Q2 large t ltlt Q2 ?
p
p
t ?2
Hard Exclusive Meson Production (HEMP)
meson
L
t ?2
Quark contribution
9
Complementarity of experiments
E190, 100GeV
At fixed xBj, study in Q2
Valence and sea quarks and Gluons
Valence quarks
0.0001lt xBj lt 0.01 Gluons
Hermes PRL87(2001) COMPASS plans
JLab PRL87(2001)
H1 and ZEUS PLB517(2001) PLB573(2003)
10
if Nµ ? 2 ? Q2 lt 11 GeV2
DVCS limited by luminosity now Nµ 2.108µ /SPS
spill Q2 lt 7.5 GeV2

?
At fixed xBj, study in Q2
11
m flux at COMPASS in 2010 ?
COMPASS
? sharing CNGS/FT ? new Linac 4 ? up to 10 times
more p improve m line
gtgtLINAC4
From Lau Gatignon
12
What can we measure now ?
13
DVCS, HEMP ?

• sDVCS small, bckg
• HEMP factorization gL
• vector meson decay
• ? RsT/sL
• ?0 largest s
• ?0 ? p p- , charged particules

Vanderhagen et al.
14
Diffractive r0 production

• Exp (NMC, E665, Zeus, H1, Hermes)
• lr lg S-channel helicity conservation SCHC
• exchanged object has natural parity P(-1)J NPE

15
Spin properties of amplitudes

• angular dist. of ? ? pp- spin density matrix
  elt
• they are bilinear combinations of
• helicity amplitudes
• ?? ? 1, 0 ?? ? 1, 0 ? 9 amplitudes
• if NPE ?
  5 amplitudes
• T00, T11 gtgt T01, T10 gtgt T-11
• SCHC 1 helicity flip 2 flips

16
r spin density matrix elt
04 00
0.01 lt Q² lt 0.05 lt Q² lt 0.3 lt Q² lt 0.6 lt
Q² lt 2.0 lt Q² lt 10 GeV2
Distribution
in terms of amplitudes
17
Determination of RsL/sT

• If SCHC holds

• sL is dominant at Q2gt2
• 2002 high stat
• in large Q2 range
• 2003 and 2004 data
• - much more stat
• - better high Q2 coverage

18
Conclusions on rho

• SCHC ? R
• stot R ? sL
• when Q2 gt 2 ? R gt 1 accurate sL
• we have transv. target spin asym ? E/H
• important for Ji sum rule (?EH)
• exploratory measurement
• (no exclusivity, nuclear target)

19
Towards a dedicated experiment for DVCS and HEMP
20
Additions to COMPASS setup
? ECal 1 or 2 ?? ? 12
Exclusivity at high energy dDM2
d(mpmp)2-mp2 gt 1 GeV2 need 0.25 GeV2 for
DM cut ? hermetic detector
21
A possible solution
30
ECAL0
12
4m

2004-2007
Funding by European Union (Bonn-Mainz-Warsaw-Sacl
ay) 45 sector recoil detector -
scintillating material studies (200ps ToF
Resolution over 4m) - fast triggering and
multi-hit ADC/TDC system
22
DVCS background

• not included
• Beam halo
• with hadronic contamination
• Beam pile-up
• Secondary interactions
• External Bremsstrahlung

• DVCS µp ? µp?
• with PYTHIA 6.1 simulate
• HEpP µp ? µpp
• ? ??
• Dissociation of the proton
• µp ? µNp
• ? Np
• DIS µp ?µpX
• with 1?, 1p, 2p,?
• Acceptance cuts
• qgmax 30
• Egmin 50 MeV
• qchargedmax 30

23
GPD measurement

• x?xB/(2-xB)
• and t are fixed
• loop over x

24
DVCS Bethe-heitler
BH calculable
Higher energy DVCSgtgtBH ? DVCS Cross section

• Smaller energy DVCSBH
• Interference term will provide
• the DVCS amplitude

25
DVCS BH with and .
Pµ-0.8 Pµ-0.8
t, ?xBj/2 fixed
26
at 100 GeV
BCA
Q24?0.5 GeV2
x 0.05 0.02
Model 1 H(x,0,t) q(x) F(t)
Model 2
H(x,0,t) q(x) e t ltb?2gt q(x) /
xat

• assuming
• L 1.3 1032 cm-2 s-1
• 150 days
• efficiency25
• 2 bins shown out of 18
• 3 bins in xBj 0.05, 0.1, 0.2
• 6 bins in Q2 from 2 to 7 GeV2

f
f
BCA
x 0.10 0.03

f
27
advantage of COMPASS kinematics
Model 1 H(x,0,t) q(x) F(t) Model 2 H(x,0,t)
q(x) e t ltb?2gt
q(x) / xat sensitive to different spatial
distributions at different x
28
HEMP filter of GPDs
Cross section
Vector meson production (?,?,?) ? H
E Pseudo-scalar production (p,? ) ? H
E


H?0 1/?2 (2/3 Hu 1/3 Hd 3/8 Hg) H? 1/?2
(2/3 Hu 1/3 Hd 1/8 Hg) H?
-1/3 Hs - 1/8 Hg
Transverse single spin asymmetry E/H

29
HEMP in 2010

• HEMP requires higher Q2 than DVCS
• with liq H target and same m flux as now
• r ? Q2 20 GeV2
• w, p, h, f ? Q2 7 GeV2
• if more m ? higher Q2

30
Roadmap for DVCS at COMPASS

• 2004-2009 r production sL and transverse spin
  asym
• 2005 "Expression of Interest" SPSC-E0I-005
• 2004-2007 recoil detector prototype
• 2008 ? proposal
• 2007-2009 construction of the recoil detector
• cryogenic target, ECal0
• 2010 dedicated GPD measurement

31
Measure OAM ?

• Ji sum rule relates total quark spin to GPDs H
  and E

• similar sum rule for gluons
• To do
• Measure H and E
• Perform flavor decomposition
• Extrapolate to t0
• Integrate over x at fixed x

• Sivers asymmetries are also measured at COMPASS

32
SPARE
33
Key role of calorimetry
ECAL2 from 0.4 to 2 mainly lead glass
GAMS ECAL1 from 2 to 12 good energy and
position resolution for
2-g separation in a high rate environment ECAL0
from 12 to 30 to be designed
for background rejection

Careful study of g and p0 production ?
COMPASS program with hadron beam
34
? angular distributions W(cos?, f, F)
depends on the Spin density matrix elements ?
23 (15) observables with polarized (unpolarized)
beam
f
This analysis only one-dimensional
angular distribution
We will use also ? f - F
35
Angular distributions
0.01 lt Q² lt 0.05 lt Q² lt 0.3 lt Q² lt 0.6 lt Q² lt
2.0 lt Q² lt 10 GeV2

f
Preliminary - Corrected for Acceptance,
smearing and efficiency (MCDIPSI gen) -
Background not subtracted

Statistical error only, limited by MC
36
Measurement of r and Im r
04 1-1
3 1-1
f
Distribution
beam polarisation
weak violation
Spin density matrices
If SCHC holds
37
COMPASS
6 angular distributions among 18
3 bins in xBj0.05, 0.1, 0.2
6 bins in Q2 from 2 to 7 GeV2
BCA in DVCS projections for 1 year
HERMES
one single bin
ltxgt0.11 ltQ2gt2.6