Color-Kinematics Relation from the Feynman Diagram Perspective

TL;DR

This paper investigates the color-kinematics relation in gluon tree amplitudes using Feynman diagrams, finding it holds for four gluons but not for five or six, unless non-local gauge transformations are applied.

Contribution

It provides explicit calculations of color-kinematics combinations for multiple gluon cases and explores the role of local gauge transformations in satisfying the relation.

Findings

elta=0 for N=4 gluons

elta ot=0 for N=5 and 6 gluons

Local gauge transformations can restore elta=0 for N=5, but not for N=6

Abstract

Feynman diagrams for gluon tree amplitudes are studied in the Feynman gauge and in any number of spacetime dimensions. The color-kinematics combinations $\Delta=n_s-n_t-n_u$ of numerators are explicitly calculated for $N=4,5,6$ gluons to see whether the color-kinematics relation $\Delta=0$ is satisfied. This is a tedious task because of the presence of four-gluon vertices, and the large number of Feynman diagrams, numerators, and $\Delta$ combinations involved, especially when $N=6$. For on-shell amplitudes, it is found that $\Delta=0$ for $N=4$, but $\Delta\not=0$ for $N=5$ and $N=6$ owing to the presence of the four-gluon vertex. However, a {\it local} generalized gauge transformation can bring about $\Delta=0$ for $N=5$, but not for $N=6$. This raises the question whether gluon amplitudes satisfying the color-kinematics relation contain non-local interactions.