Semi-Inclusive Deep Inelastic Scattering processes from small to large P_T

TL;DR

The paper addresses describing hadron production in unpolarized SIDIS across the full transverse-momentum range by combining two regimes: low-P_T contributions from intrinsic transverse motion via unintegrated (TMD) distributions and high-P_T contributions from collinear pQCD with higher-order corrections. It demonstrates that LO intrinsic-motion terms reproduce data for P_T ≲ 1 GeV/c, while α_s^1 pQCD processes explain large-P_T data, with a smooth transition near P_T ≈ 1 GeV/c and K-factor adjustments to approximate NLO effects. The authors fit ⟨k_perp^2⟩ and ⟨p_perp^2⟩ using MRST01 PDFs and Kretzer FFs, achieving good agreement with EMC and ZEUS measurements for dσ/dP_T, dσ/dφ_h, and ⟨cos φ_h⟩, and they provide detailed predictions for upcoming HERMES, COMPASS, and JLab measurements of cross sections and azimuthal moments in the low-P_T domain. The work highlights the role of TMDs in SIDIS and offers experimentally testable predictions to probe intrinsic motion and its impact on azimuthal observables.

Abstract

We consider the azimuthal and $P_T$ dependence of hadrons produced in unpolarized Semi-Inclusive Deep Inelastic Scattering (SIDIS) processes, within the factorized QCD parton model. It is shown that at small $P_T$ values, $P_T \lsim 1$ GeV/c, lowest order contributions, coupled to unintegrated (Transverse Momentum Dependent) quark distribution and fragmentation functions, describe all data. At larger $P_T$ values, $P_T \gsim 1$ GeV/c, the usual pQCD higher order collinear contributions dominate. Having explained the full $P_T$ range of available data, we give new detailed predictions concerning the azimuthal and $P_T$ dependence of hadrons which could be measured in ongoing or planned experiments by HERMES, COMPASS and JLab collaborations.

Figures (17)

• Figure 1: Three dimensional kinematics of the SIDIS process.
• Figure 2:
• Figure 3: The normalized cross section $d\sigma/dP_T^2$: the dashed line reproduces the $\mathcal{O}(\alpha_s^0)$ contributions, computed by taking into account the partonic transverse intrinsic motion at all orders in $(k_\perp/Q)$, Eq. (\ref{['sidis-Xsec-final']}). The solid line corresponds to collinear and pQCD contributions, computed at LO, with a $K$ factor ($K=6$) to account for NLO effects, Eqs. (\ref{['LOxs']})-(\ref{['pert-exp-K']}). The data are from EMC collaboration measurements EMCpt. $\langle k_\perp\rangle$ and $\langle p_\perp\rangle$ are fixed as in Eq. (\ref{['parameters']}).
• Figure 4: The normalized cross section $d\sigma/dP_T$: the dashed line reproduces the $\mathcal{O}(\alpha_s^0)$ contribution, computed by taking into account the partonic transverse intrinsic motion at all orders in the $(k_\perp/Q)$ expansion, Eq. (\ref{['sidis-Xsec-final']}); the solid line corresponds to the SIDIS cross section as given by LO contributions and a $K$ factor ($K=1.5$) to account for NLO effects, Eqs. (\ref{['LOxs']})--(\ref{['pert-exp-K']}). The data are from ZEUS collaboration measurements Derrick96. $\langle k_\perp\rangle$ and $\langle p_\perp\rangle$ are fixed as in Eq. (\ref{['parameters']}).
• Figure 5: The cross section $d\sigma/d\phi_h$: the solid line is obtained by including all orders in $(k_\perp/Q)$, the LO corrections and a $K=6$ factor to account for NLO effects. The data are from EMC measurements EMC2. $\langle k_\perp\rangle$ and $\langle p_\perp\rangle$ are fixed as in Eq. (\ref{['parameters']}).