Recent Results on Deeply Virtual Compton Scattering at HERMES H. Marukyan a ∗ (for the HERMES Collaboration) a Yerevan Physics Institute, Alikhanian Brothers str. 2, Yerevan, Armenia HERMES has measured azimuthal asymmetries in the hard electroproduction of real photons, the so-called Deeply Virtual Compton Scattering (DVCS) process, using 27.6 GeV HERA positron and electron beams and various polarized and unpolarized gaseous targets. New results have recently been obtained for the t, x and Q 2 dependencies of the transverse target-spin asymmetry (TTSA) from data accumulated in 2002-2004 with the positron beam and a transversely polarized hydrogen target. A model-dependent constraint on the total angular momentum carried by u- and d-quarks, Ju and J d , is obtained by comparing the extracted TTSA amplitude with theoretical predictions based on a certain GPD model. 1. Introduction An essential task in QCD is to understand the hadron structure at the level of quarks and glu- ons. Particularly, the contributions of these basic constituents to the nucleon’s spin both via their intrinsic spin and their orbital motion is of great interest. Recent interest in hard exclusive processes has resulted from the theoretical description of these processes in terms of the Generalized Parton Dis- tribution (GPD) formalism [1–3]. This formal- ism offers a consistent description of the nucleon structure. It incorporates as limiting cases the well-known nucleon form factors determined from elastic scattering as well as parton momentum distributions (PDFs) determined from measure- ments of inclusive and semi-inclusive deep inelas- tic lepton-nucleon scattering (DIS). Interest in the GPD framework has also been motivated by the fact that the unknown total angular momenta of quarks and gluons within the nucleon are en- coded in the GPDs [4]. The latter allows in prin- ciple access to the orbital angular momentum of quarks using the knowledge of their intrinsic spin contribution to the nucleon’s spin. One of the theoretically cleanest processes with which to probe GPDs is Deeply-Virtual Comp- ∗ We gratefully acknowledge the DESY managment for its support, the staff at DESY and the collaborating institu- tions for their significant effort, and our national funding agencies and the EU RII3-CT-2004-506078 programme for financial support. ton Scattering (DVCS), in which a highly virtual photon is absorbed by a parton within the target, which produces a single real photon in the final state, along with the recoiling nucleon being in its ground state. 2. DVCS at HERMES The DVCS process is experimentally indistin- guishable from the Bethe-Heitler (BH) process, in which a real photon is radiated by the incoming or outgoing lepton rather than the quark. The cross section for the leptoproduction of real pho- tons is therefore given by the coherent sum of the BH and DVCS amplitudes squared: dσ ∝|τ BH | 2 +|τ DV CS | 2 +τ ∗ BH τ DV CS +τ ∗ DV CS τ BH (1) At HERMES kinematics the BH process dom- inates over DVCS in the cross section. How- ever, the DVCS amplitude can be studied via the interference term by measuring various cross section asymmetries and their dependences on the azimuthal angle φ, defined as the angle be- tween the lepton scattering plane and the pho- ton production plane. The interference term τ ∗ BH τ DV CS + τ ∗ DV CS τ BH (I ) can be written in a series of Fourier moments in φ [5] I ∝±  c I 0 + 3  n=1 c I n cos(nφ)+ 3  n=1 s I n sin(nφ)  ,(2) where sign +(-) stands for the electron(positron) beam, c I 0 , c I n and s I n are given by a linear combina- Nuclear Physics B (Proc. Suppl.) 174 (2007) 19–22 0920-5632/$ – see front matter © 2007 Elsevier B.V. All rights reserved. www.elsevierphysics.com doi:10.1016/j.nuclphysbps.2007.08.078