GPD a GPD at t COMP COMPASS ASS at C t CERN ERN 1- DVCS CS - PowerPoint PPT Presentation

GPD a GPD at t COMP COMPASS ASS at C t CERN ERN 1- DVCS CS 2- HEMP HEMP Nicole d’Hose – CEA – Université Paris-Saclay 1

Deeply virtual Compton scattering (DVCS) D. Mueller et al , Fortsch. Phys. 42 (1994) ℓ’ X.D. Ji , PRL 78 (1997), PRD 55 (1997) A. V. Radyushkin , PLB 385 (1996), PRD 56 (1997) ℓ Q² large , x B   * DVCS: ℓ p  ℓ ’ p’  the golden channel hard because it interferes with GPDs soft the Bethe-Heitler process Generalized Parton also meson production Distributions ℓ p  ℓ ’ p’  ,  ,  or  or J/  ... small The GPDs depend on the following variables: The variables measured in the experiment: E ℓ , Q 2 , x B  2  /(1+  ), t (or   *  ) and  ( ℓℓ’ plane/  * plane ) 2

Deeply virtual Compton scattering (DVCS) ℓ’ From Goeke, Polyakov, Vanderhaeghen, PPNP47 (2001) ℓ Q² large , x B   * q(x) DGLAP hard GPDs soft ERBL Generalized Parton Distributions small The amplitude DVCS at LT & LO in  S (GPD H ) : Real part Imaginary part H(𝑦,  ,𝑢) H(𝑦,  ,𝑢) +1 𝑒𝑦 +1 𝑒𝑦 − 𝑗 𝜌 H(𝑦 ±  , 𝑦, 𝑢) H = ׬ 𝑦−  + 𝑗ε = P ׬ 𝑦−  −1 −1 𝑢 ,  fixed ReH (  , 𝑢) = න 𝑒𝑦 ImH (𝑦, 𝑢) In an experiment we measure + 𝐸(𝑢) 𝑦 −  Compton Form Factor H 3

COM COMPAS ASS: S: Versatile facility with hadron (   , K  , p …) & lepton (polarized   ) beams of high energy  200 GeV LHC COMP COMPASS ASS SPS 4

The DVCS experiment at COMPASS DVCS : μ p  μ ’ p  μ μ’  p SM2 Muon Wall SM1 ECAL2/ HCAL2 ECAL1/ RICH HCAL1 NIM A 577 (2007) 455 and NIM A 779 (2015) 69 Two stage magnetic spectrometer for large angular & momentum acceptance Particle identification with: - Ring Imaging Cerenkov Counter - Electromagnetic calorimeters ( ECAL1 and ECAL2 ) - Hadronic calorimeters 5 - Hadron absorbers

The DVCS experiment at COMPASS DVCS : μ p  μ ’ p  μ μ’  p New equipements:  2.5m LH2 target  4m ToF Barrel CAMERA 24 inner & outer scintillators separated by  1m 1 GHz SADC readout, 330ps ToF resolution  ECAL0 : 2 × 2 m2 ECAL0 Shashlyk modules + MAPD readout one module is made of 9 cells (4  4 cm 2 ) = 194 modules or 1746 cells 6 CAMERA

ECAL2 ECAL1 ECAL0 CAMERA recoil proton detector surrounding the 2.5m long LH2 target 2012: 1 month pilot run |  2016 -17: 2 x 6 months   data taking + SIDIS on unpolarized protons 7

COMPASS 2012 Selection of exclusive evts with recoil detection x Bj > 0.03 10<  <32GeV Comparison between the observables given by the spectro or by CAMERA with  0 contamination DVCS : μ p  μ ’ p  2) For the 1) For the proton proton 1) azimuthal momentum angle 2) applied applied cut cut 3) 1) 4) 3) For the 4) proton track applied applied cut cut 8

 0 background estimation COMPASS 2012  0 are one of the main background sources for excl. photon events. Two possible case: Visible (both  detected  subtracted)  the DVCS photon after all exclusivity cuts is combined with all detected photons below the DVCS threshold: 4,5,10 GeV in ECAL0, 1, 2 Visible leaking  0 in the data Invisible (one  lost  estimated by MC)   Semi-inclusive LEPTO 6.1  Exclusive HEPGEN  0 (Goloskokov-Kroll model) Comparing the two components to the data allows the determination of their relative normalisation. The sum of the 2 components is normalized to the visible  0 contamination in the M  peak 9

DVCS cross section at E  =160 GeV COMPASS 2012 BH | 2 + In d   |T |T BH Interference Term + |T DVCS | 2 Bethe-Heitler (BH) DVCS   * μ ’ μ p  Φ (rad) φ φ

DVCS cross section at E  =160 GeV COMPASS 2012 BH | 2 + In d   |T |T BH Interference Term + |T DVCS | 2 Bethe-Heitler (BH) DVCS   * μ ’ μ p  DVCS dominates - Study of d  DVCS /dt DVCS ampl. via interference BH dominates Reference yield Jlab, HERMES, Only for H1, ZEUS, COMPASS H1, COMPASS Φ (rad) φ φ

DVCS cross section at E  =160 GeV COMPASS 2012 when BH is not dominant At COMPASS using polarized positive and negative muon beams: S CS,U  = ] All the other terms are cancelled in the integration over  calculable can be subtracted Note the symmetric acceptance Flux for transverse 12 virtual photons

COMPASS 2012 Transverse extention of partons in the sea quark range d  DVCS /dt= e - B’| t| = hep-ex/1802.02739, subm. to PLB COMPASS ‘ 4 weeks in 2012 10 times more stat in 2016-17 ‘ 13

COMPASS 2012 Transverse extention of partons in the sea quark range 2 (x B ) >  2B’ (x B ) d  DVCS /dt= e - B’| t| = <r  hep-ex/1802.02739, subm. to PLB At COMPASS: < x Bj >=0.056; < Q 2 >=1.8 GeV 2 ; COMPASS t varies from 0.08 to 0.64 GeV 2 4 weeks in 2012 At small x Bj and small t: : ‘ c 0 DVCS 10 times more stat in 2016-17 Dominance of Im H (with respect of Re H and other CFF F ) ‘

COMPASS 2012 Transverse extention of partons in the sea quark range 2 (x B ) >  2B’ (x B ) d  DVCS /dt= e - B’| t| = <r  hep-ex/1802.02739, subm. to PLB At COMPASS: < x Bj >=0.056; < Q 2 >=1.8 GeV 2 ; t varies from 0.08 to 0.64 GeV 2 ‘ At small x Bj and small t: : c 0 DVCS 10 times more stat in 2016-17 Dominance of Im H (with respect of Re H and other CFF F ) ‘

COMPASS 2016-17 First insight Exclusivity variables Comparison between the observables given by the spectro or by CAMERA DVCS : μ p  μ ’ p  1) 2) 3) 4)

First insight Distributions in  COMPASS 2016-17   * μ ’ Only 1.3% of 2016-17 data μ  p Bethe-Heitler (BH) DVCS 0.005 < x Bj < 0.01 0.01 < x Bj < 0.03 x Bj > 0.03 No ‘’invisible’’  0 still to be removed BH expected a significant to contribute DVCS contribution only will allow to study BH MC is d  DVCS /dt = e -B ’|t | normalized to this bin c 0 DVCS = This research is part of the Blue Waters sustained-petascale computing project, which is supported by the National Science Foundation (awards OCI-0725070 and ACI-1238993) and the state of Illinois. Blue Waters is a joint effort of the University of Illinois at Urbana-Champaign and its National Center for Supercomputing Applications. This work is also part of the "Mapping Proton Quark Structure using Petabytes of COMPASS Data" PRAC allocation supported by the National Science Foundation (award number OCI 1713684 ).

Beam Charge and Spin Diff. @ COMPASS Re Re H > 0 at H1 D CS,U  < 0 at HERMES Value of x Bj for the node? I = Re F 1 H c 1 Predictions with VGG KM10 The knowledge of Re F 1 H and Im F 1 H is essentiel to play with the dispersion relation to extract the D-term JLab HERMES 1 < Q 2 < 8 GeV 2 E  = 160 GeV COMPASS 2 years of data 18

GPDs and Hard Exclusive Meson Production Quark contribution 4 chiral-even GPDs: helicity of parton unchanged H q ( x ,  , t) E q ( x ,  , t) qq meson For Vector Meson Q 2 γ * L   H q ( x ,  , t) E q ( x ,  , t) For Pseudo-Scalar Meson x - ξ x + ξ + 4 chiral-odd or transversity GPDs: helicity of parton changed (not possible in DVCS) GPDs p p’ H q ( x ,  , t) E q ( x ,  , t)  T T   H q E q = 2 + E q H q ( x ,  , t) E q ( x ,  , t) T T T Gluon contribution at the same order in  S T T Factorisation proven only for  L Vector meson qq γ * L  T is asymptotically suppressed by 1/Q 2 but large contribution observed L model of  T with transversity GPDs - divergencies regularized by k T of q and q and Sudakov suppression factor meson + qq Q 2 γ * + T + M 0 - , + + sensitive to H q x - ξ - x + ξ + T The meson wave function and to a twist-3 meson wave function GPDs Is an additional non-perturbative term - + p p’

Exclusive  0 production on unpolarized proton COMPASS 2012 e p  e  0 p Leading twist should be dominant but  only a few % of The other contributions arise from coupling between chiral-odd (quark helicity flip) GPDs to the twist-3 pion amplitude A large impact of E T should be clearly visible in  TT and in the dip at small |t| of  T 20

Exclusive  0 production on unpolarized proton COMPASS 2012 e p  e  0 p A dip at small t would indicate a large impact of E T  TT large (impact of E T ) ?  LT smaller but significantly positive 21 hep-ex/1903.12030, subm. to PLB

Exclusive  production on unpolarized proton COMPASS 2012 22

Conclusions From 2016-17 data sum and difference of DVCS x-sections with polarized  + and  -  transverse extension of partons as a function of x Bj  Im H ( ξ,t ) and Re H ( ξ,t ) for D-term and pressure distribution HEMP  0 ,  ,  ,  , J/   universality of GPDs - transverse GPDs - flavor decomposition COMPASS++/AMBER starting in 2022 Letter of Intent Draft 1.0: https://arXiv.org/abs/1808.00848 New collaborators are welcome: https://nqf-m2.web.cern.ch 23