Let $\mathcal{L}$ be the Lagrangian for usual QED with scalar, charged particles (with photons and electrons as well):

$$\mathcal{L} = -\frac{1}{4}F_{\mu\nu}F^{\mu\nu}+\bar{\psi}\left(i\gamma^{\mu}\partial_{\mu}-m \right)\psi - \partial_{\mu}\phi^*\partial^{\mu}\phi-M^2\phi^*\phi$$

I was trying to show that, because of symmetries, that the Lagrangian above could be written in the same way if $\partial_{\mu} \to D_{\mu}$. However, I have to find $D_{\mu}$. The symmetry I am considering is such that

$$\psi \to \psi e^{i\Lambda(x)}$$ $$\phi \to \phi e^{i\Lambda(x)}$$ $$A_{\mu} \to A_{\mu}-\frac{1}{e}\partial_{\mu}\Lambda(x)$$

My attempt

Since $\psi$ and $\phi$ have no crossed terms between them, I figured that we should add a $\mathcal{L}_{int}$ (interactions) such that it would cancel the term


that results from the fermionic part. Hence, $\mathcal{L}_{int} = -e\bar{\psi}\gamma^{\mu}A_{\mu}\psi$ seems to do the trick. However, the scalar particle bit does not seem to trivially simplify, as there are some remaining terms:


This would give $D_{\mu} = \partial_{\mu}+ieA_{\mu}$.

Is my procedure correct? Is there a more intuitive way of solving this?

Edit: Corrected misuse of concepts as pointed in the comments.

  • $\begingroup$ How can the fields not be coupled if both have charges? Because A has spin 1? $\endgroup$
    – user303670
    Jun 7, 2021 at 17:29
  • $\begingroup$ @Duepietri My bad, wrong use of concepts. I should've said that there are no crossed terms between them. I apologise. $\endgroup$
    – miniplanck
    Jun 7, 2021 at 17:32
  • $\begingroup$ I would just comment for mu! $\endgroup$
    – user303670
    Jun 7, 2021 at 17:47
  • 1
    $\begingroup$ @Duepietri Already took care of that :) $\endgroup$
    – miniplanck
    Jun 7, 2021 at 17:53
  • 1
    $\begingroup$ Look how the Dirac Lagrangian is written here: quantummechanics.ucsd.edu/ph130a/130_notes/node508.html $\endgroup$
    – user303670
    Jun 7, 2021 at 19:41

1 Answer 1


Thanks to @Duepietri, I've come to a conclusion. Taking into account that $D_{\mu}$ needs to be gauge invariant, the minimal coupling

$$D_{\mu} = \partial_{\mu} + ieA_{\mu}$$

would give some remaining terms, of which we could compose an interaction:

$$\mathcal{L}_{int} = -e\bar{\psi}\gamma^{\mu}A_{\mu}\psi -ieA_{\mu}\left[\left(\partial^{\mu}\phi^*\right)\phi-\phi^*\partial^{\mu}\phi\right]-e^2A_{\mu}A^{\mu}\phi^*\phi$$

since we can identify $\mathcal{L}_{free}$ with the same expression as in the original post, with $\partial_{\mu}$ instead of $D_{\mu}$.

For those who wish to read a bit more about this, I would recommend the online note by Matthew Schwartz (2012) on Scalar QED or his book, Quantum Field Theory and The Standard Model by Matthew D. Schwartz.


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.