In the $q\bar{q}\to ZZ$ process, the following Feynman diagram in LO appears:
This means for each vertex, the Feynman amplitude contains a term proportional to $(g_V-g_A\gamma^5)$, which makes $D$-dimensional calculations of traces very difficult. In the paper where the figure above is from (doi:10.1016/0550-3213(91)90475-d), the author mentions the following reasoning in order to deal with $\gamma^5$ matrices:
There are many other papers (like doi.org/10.1103/PhysRevD.43.3626), where they similarly discard the terms with traces containing $\gamma^5$ in them, however, I do not understand the reasoning behind it. The quoted text above mentions something about these terms having to equal to zero,...
- ... since we are dealing with two identical vector bosons and thus making it symmetric...
- ... and the term being proportional to the antisymmetric tensor contracted with independent momenta at the same time. I would be really thankful, if someone could explain these conditions in a little bit more detail, and why exactly the discarding of such terms with traces containing $\gamma^5$ matrices justifiable.
An additional, but a little unrelated question: Can this strategy somehow replicated using the HMBV scheme? (i.e. calculating the corresponding traces in $D$ dimensions, and then discarding the terms proportional to the Levi-Civita tensor contracted with the $4-D$ dimensional parts of the momentum)
