Non-perturbative effects in the energy-energy correlation

TL;DR

The paper addresses non-perturbative power corrections to the energy-energy correlation in $e^+e^-$ annihilation and extends the fully resummed perturbative prediction by incorporating leading NP contributions via the dispersive method. It identifies leading NP effects from quark–gluon correlations, giving a $1/(Q\chi)$ dependence in the back-to-back region, and from quark–antiquark correlations, giving a $\ln Q^2/Q^2$ term, with non-integer power suppression due to PT–NP interplay. NP coefficients are linked to low-energy moments of the strong coupling, with the Milan factor entering the quark–gluon piece, enabling connections to hadronization observables. The framework provides default NP parameter values and a path for experimental tests of the NP-PT interplay in the back-to-back region.

Abstract

The fully resummed next-to-leading-order perturbative calculation of the energy-energy correlation in $e^+e^-$ annihilation is extended to include the leading non-perturbative power-behaved contributions computed using the ``dispersive method'' applied earlier to event shape variables. The correlation between a leading (anti)quark and a gluon produces a non-perturbative 1/Q contribution, while non-perturbative effects in the quark-antiquark correlation give rise to a smaller contribution $\ln Q^2/Q^2$. In the back-to-back region, the power-suppressed contributions actually decrease much more slowly, as small non-integer powers of 1/Q, as a result of the interplay with perturbative effects. The hypothesis of a universal low-energy form for the strong coupling relates the coefficients of these contributions to those measured for other observables.