Consider the Lagrangian \begin{equation} \mathcal{L}= -\frac{1}{4} F_{\mu \nu}F^{\mu \nu} - A_{\mu}J^{\mu} \ \ \ \ \text{ with } \ \ \ \ F_{\mu \nu}=\partial_\mu A_\nu - \partial_{\nu}A_{\mu}. \end{equation} The current $J^{\mu}$ is conserved and hence it satisfies $\partial_{\mu}J^{\mu}=0$.

I showed action $S$ is invariant under the gauge transformation \begin{equation} A_{\mu} \to A_{\mu} + \partial_{\mu} \alpha, \end{equation} where $\alpha$ is an arbitrary function that vanishes at infinity. Moreover I know that the equations of motions are $$\partial^{\mu}F_{\mu \nu} = J_{\nu}.$$

Question: Show that one can choose the function $\alpha$ so that after an appropriate gauge transformation one can set $A_0=f$, where $f$ is a given function.

I think I'm missing some fundamental stuff about gauge theory since e.g. I don't know why the coulomb gauge or the lorenz gauge are allowed.

My idea: Say $\alpha = \int (f- A_0) dt$, then we get under the gauge transformation . $$A_0 \to A_0' = A_0 + \partial_{0} \int (f- A_0) dt = A_0 +f - A_0 = f.$$ And then we redefine $A_0'$ as $A_0$ which is possible by gauge invariance and therefore obtain $A_0=f$.

Is what I'm doing correct, or am I doing something fundamentally wrong?
Moreover, what would the residual gauge transformation be?

  • $\begingroup$ Are you sure $A^0$ can be set to any function $f$? $\endgroup$ Commented Dec 9, 2020 at 5:50
  • $\begingroup$ well that's the question. Of course in physics we don't consider non-integrable functions e.g. everything is always smooth. or is that not the point $\endgroup$ Commented Dec 9, 2020 at 10:22

1 Answer 1


Suppose that the gauge field $A'_{\mu}$ does not satisfy the desired property, so let's say instead $A'_{0} = g$ for some function $g$. Choose $\alpha$ such that $$ \partial_0 \alpha = f-g, $$ which is guaranteed to have a solution. We apply a gauge transformation $A'_{\mu} \to A_{\mu} = A'_{\mu} + \partial_{\mu} \alpha $ and obtain that $$ A_0 = A'_0 + \partial_0 \alpha = g + f-g =f. $$ Therefore after an appropriate gauge transformation one can set $A_0=f$. Of course if we'd apply another gauge transformation such that $\partial_0 \alpha =0$ and we'd still have that $A_0=f$. So the residual gauge symmetry is $\partial_0\alpha =0$.


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.