In supersymmetric gauge theories, the vector potential is a part of a vector supermultiplet which is represented by a real superfield $V$. Expanded out in components, the Lagrangian for such a field is given by

$$\mathcal{L}=-\frac{1}{4g^2}F_{\mu\nu}^a F^{\mu\nu a}+\cdots,$$

where the extra terms are in terms of the extra superpartners in this multiplet.

Upon gauge fixing, one introduces Faddeev-Popov ghosts to counter the overcounting in the path integral. The gauge-fixed Lagrangian in, say, the Lorenz gauge takes the form

$$\mathcal{L}_{\text{g.f.}}=\text{Tr}\left[-\frac{1}{4g^2}F_{\mu\nu} F^{\mu\nu}+\overline{c}D_{\mu}\partial^{\mu}c+\cdots\right].$$

Thus, since the vector potential $A$ transforms non-trivially under SUSY transformations, but the ghosts do not, it is clear that the introduction of ghost fields has broken supersymmetry at ghost level $n_g\neq 0$.

Is there a gauge-fixing method which circumvents this and allows for intrinsically supersymmetric perturbative calculations (for instance introducing a super partner for the ghosts and absorbing the broken supersymmetry into a $\mathbb{Z}_2$-graded version of extended BRST symmetry)?

  • 1
    $\begingroup$ Can you give a reference where I can read how the overcounting problem occurs? $\endgroup$
    – my2cts
    Dec 18, 2018 at 13:06
  • $\begingroup$ This paper might be useful. $\endgroup$
    – Nogueira
    Dec 18, 2018 at 13:20
  • $\begingroup$ @my2cts Any QFT book will discuss this in the quantization of non-abelian gauge theories in the path integral formalism (cf section 16.2 of Peskin and Schroeder). $\endgroup$ Dec 18, 2018 at 14:13
  • $\begingroup$ Why have you concluded that susy is unbroken at ghost number zero? Wouldn’t that imply that observables are in fact unaffected? $\endgroup$ Jul 11, 2022 at 6:56

1 Answer 1


In SUSY Yang-Mills the gauge parameter is not just a Lorentz-scalar, it is a full chiral superfield. So when we gauge fix, we don't just use a Lorentz-scalar ghost, we use a chiral superfield ghost. Likewise, the constraint isn't just a scalar, it needs to be some kind of supersymmetric object.


Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.