NNLO QCD corrections to unpolarized and polarized SIDIS

TL;DR

This paper computes NNLO QCD corrections to SIDIS processes, enhancing the precision of theoretical predictions for unpolarized and polarized scattering, which is vital for understanding hadron structure and future collider experiments.

Contribution

It provides the first detailed NNLO calculations for SIDIS coefficient functions, including all partonic channels and their phenomenological implications.

Findings

NNLO corrections significantly improve prediction stability.

Corrections reduce scale dependence in SIDIS cross sections.

Results are relevant for future Electron-Ion Collider analyses.

Abstract

The semi-inclusive deep-inelastic scattering (SIDIS) process requires the presence of an identified hadron H$'$ in the final state, which arises from the scattering of a lepton with an initial hadron P. By employing factorization in quantum chromodynamics (QCD), SIDIS provides essential knowledge on the hadron structure, enabling the exploration of parton distribution functions (PDFs) and fragmentation functions (FFs). The coefficient functions for SIDIS can be calculated in perturbative QCD and are currently known to the next-to-next-to-leading order (NNLO) for the cases, where the incoming lepton and the hadron P are either both polarized or unpolarized. We present a detailed description of these NNLO computations, including a thorough discussion of all the partonic channels, the calculation of the amplitudes and master integrals for the phase-space integration as well as the renormalization of ultraviolet divergences and mass factorization of infrared divergences in dimensional regularization through NNLO. We provide an extensive phenomenological analysis of the effects of NNLO corrections on SIDIS cross sections for different PDFs and FFs and various kinematics, including those of the future Electron-Ion Collider (EIC). We find that these corrections are not only significant but also crucial for reducing the dependence on the renormalization and factorization scales $\mu_R$ and $\mu_F$ to obtain stable predictions.