Speaker
Description
Generalized parton distributions (GPDs) are off-forward matrix elements of quark and gluon operators that enclose information on the total angular momentum of partons, and so on the spin of hadrons (cf. EMC measurements and spin puzzle). In addition, GPDs enable tomography of the nucleon allowing to study spatial distribution of partons as a function of their momentum. To access GPDs one needs to consider exclusive processes, such as deeply virtual and timelike Compton scattering (DVCS and TCS). At LO these processes are mainly sensitive to GPDs in a restricted kinematic domain, namely $x = \xi$, where $x$ represents the average fraction of longitudinal momentum carried by an active parton, while $\xi$ is the so-called skewness variable. The process that avoids this constraint is double deeply virtual Compton scattering (DDVCS) for which an electron scatters off a nucleon and produces a lepton pair. The extra virtuality with respect to DVCS and TCS allows for LO access to GPDs at $x\neq\xi$.
Due to the lepton-pair production, DDVCS cross-section is smaller than that of DVCS by roughly a factor $\alpha_{\rm em}$. Consequently, measurements of DDVCS with the relevant statistics for GPDs extraction require experiments with large luminosity. One of them will be the Electron-Ion Collider (EIC). In this talk, the importance of DDVCS for GPD physics, being one of the pillars of the EIC programme, will be highlighted. Elements of impact studies for EIC and JLab have been worked out by means of PARTONS software and EpIC Monte Carlo event generator. The phenomenological study lays on a new formulation of DDVCS based on the methods developed by R. Kleiss and W. J. Stirling in the 1980s, that will be shown as well.
