2
$\begingroup$

In the Lorenz Gauge, we can write Maxwell's equations as: $$\tag1 \Box A^\beta=\mu_0j^\beta.$$

We then go on to solve this by treating each component $A^\beta$ as an independent solution of the scalar wave equation with source, and summing Green functions. That is: $$\tag2 A^\beta=\mu_0\int \frac{1}{4\pi |r-r'|}j^\beta(ct-|r-r'|, r')\ \mathrm d^{3}r'.$$

My question is, how have we ensured that the Lorenz gauge condition is still met?

Sure to write $(1)$, we needed $\partial_\mu A^\mu = 0$, but how have we ensured our solution $(2)$ meets this condition?

$\endgroup$
2
  • $\begingroup$ Hint: what is $\partial_\mu j^\mu$? [also, its easier if you write $(2)$ in terms of $\mathrm dt'\mathrm d^3r'$] $\endgroup$ Commented May 16, 2017 at 18:36
  • $\begingroup$ I thought that you were still struggling so I decided to post a more detailed hint. I'm glad you could solve it on your own though :-) $\endgroup$ Commented May 18, 2017 at 21:31

1 Answer 1

1
$\begingroup$

Sketch of the argument:

  1. Write your equation in the form $$ A^\mu(x)=\int_{\mathbb R^4} G(x-x')j^\mu(x')\ \mathrm dx' $$ where $G$ is one of the propagators of the wave equation. For example, $G_\mathrm{ret}(x)\propto\delta(x^2)\Theta(x^0)$ which, after integration over $\mathrm dx^0$, leads to your second formula.

  2. Show that $$ \partial_\mu A^\mu(x)=\int_{\mathbb R^4} G(x-x')\partial'_\mu j^\mu(x')\,\mathrm dx' $$ where we have use the fact that $G$ depends only on the difference $x-x'$, and we have integrated by parts. Here one should argue that boundary terms vanish because of the kinematics of $G$.

  3. Argue by current conservation that the Lorenz gauge condition holds. One should note that, independently of the chosen gauge, $$ \partial_\mu F^{\mu\nu}=j^\nu $$ and therefore, due to the skew-symmetry of $F$, the current must be conserved, as a consistency condition for the above PDE to be well-posed.

$\endgroup$

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.