On non-singlet physical evolution kernels and large-x coefficient functions in perturbative QCD

TL;DR

This paper investigates the large-x behavior of non-singlet physical evolution kernels in QCD, revealing a universal single-logarithmic enhancement at all orders and providing predictions for higher-order coefficient functions.

Contribution

It establishes the all-order universality of large-x behavior of non-singlet kernels and predicts the structure of higher-order contributions in perturbative QCD.

Findings

Universal single-logarithmic large-x enhancement at all orders.

Predictions for highest ln^n(1-x) contributions to coefficient functions.

Exponentiation form in Mellin-N space with non-negligible 1/N corrections.

Abstract

We study the large-x behaviour of the physical evolution kernels for flavour non-singlet observables in deep-inelastic scattering, where x is the Bjorken variable, semi-inclusive e^+ e^- annihilation and Drell-Yan lepton-pair production. Unlike the corresponding MSbar-scheme coefficient functions, all these kernels show a single-logarithmic large-x enhancement at all orders in 1-x. We conjecture that this universal behaviour, established by Feynman-diagram calculations up to the fourth order, holds at all orders in the strong coupling constant alpha_s. The resulting predictions are presented for the highest ln^n(1-x) contributions to the higher-order coefficient functions. In Mellin-N space these predictions take the form of an exponentiation which, however, appears to be less powerful than the well-known soft-gluon exponentiation of the leading (1-x)^(-1) ln^n(1-x) terms. In particular in deep-inelastic scattering the 1/N corrections are non-negligible for all practically relevant N.