1
$\begingroup$

I am confused on some identities in Mark Srednicki's QFT book, section76. Let $j^{\mu}=\bar{\psi}\gamma^{\mu}\psi$ and $j^{\nu}_A=\bar{\psi}\gamma^{\nu}\gamma_5\psi$ be $U(1)$ and $U(1)_A$ current, respectively. Then we have three identities (see eqs. (76.17)-(76.19) in the book)

$$ \begin{aligned} &\partial_{\mu} \langle j^{\nu}(x)j^{\nu}(y)j^{\rho}_A(z) \rangle=0 \cr &\partial_{\nu} \langle j^{\nu}(x)j^{\nu}(y)j^{\rho}_A(z) \rangle=0 \cr &\partial_{\rho} \langle j^{\nu}(x)j^{\nu}(y)j^{\rho}_A(z) \rangle=0 \end{aligned} $$

In the next, it says

Note that there are no contact terms in eqs. (76.17)-(76.19), because both $j^{\mu}$ and $j^{\mu}_A$ are invariant under both $U(1)$ transformation.

This is my confusion: why does such invariance give vanishing contact terms?

$\endgroup$
1
  • 1
    $\begingroup$ The same question was asked here. Although it doesn't have an answer, there is a helpful comment by Anonjohn, saying "The right hand side of the ward identity is the contact terms. But each contact term is the variation of the operator under the U(1) gauge(axial) symmetry times a delta function. Since the currents themselves are invariants under the symmetry, the contact terms vanish" $\endgroup$ Commented Jun 17, 2021 at 14:01

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.