3
$\begingroup$

Let $D$ be a compact region in $\mathbb{R}^3$ with a smooth boundary $S$. Assume $0 \in \text{Int}(D)$. If an electric charge of magnitude $q$ is placed at $0$, the resulting force field is $q\vec{r}/r^3$, where $\vec{r}(x)$ is the vector to a point $x$ from $0$ and $r(x)$ is its magnitude. Show that the amount of charge $q$ can be determined from the force on the boundary by proving Gauss's law:$$\int_S \vec{F} \cdot \vec{n}\,dA = 4\pi q.$$

I am familiar with the approach in basic textbooks, but I would be interested in seeing a derivation/proof using the language of differential topology.

$\endgroup$
6
  • $\begingroup$ Provided that "differential topology" is not a standard mathematical term as far as I know, you should take a look at the Stokes' theorem for differential forms. Once you understand it, the proof is straightforward starting from the $\nabla\cdot E=\varrho$ of Maxwell's equations. $\endgroup$
    – yuggib
    Commented May 12, 2015 at 5:04
  • $\begingroup$ @yuggib it is in fact a standard term (see e.g, Milnor's book Topology from the differentiable viewpoint) and yes, Stokes theorem is the right way to look at it if one wants to be fancy :) $\endgroup$
    – Danu
    Commented May 12, 2015 at 7:49
  • $\begingroup$ The Wikipedia page on Stokes theorem has a short explanation under the section "special cases" $\endgroup$
    – Danu
    Commented May 12, 2015 at 8:00
  • 1
    $\begingroup$ @Danu Ok, I see...it is a utilized terminology as the topology of functions on differentiable manifolds. Nevertheless it is not the best semantical choice (in my opinion), for it seems that you "differentiate topologies" in some sense. Just a matter of personal taste anyways ;-) As for the mathematical fanciness, it seems exactly what the OP asked... $\endgroup$
    – yuggib
    Commented May 12, 2015 at 8:27
  • $\begingroup$ Related: physics.stackexchange.com/q/104693/2451 $\endgroup$
    – Qmechanic
    Commented May 12, 2015 at 12:21

1 Answer 1

1
$\begingroup$

First, we will compute $\text{div}\,F$. The partial derivatives are given by$${{\partial F}\over{\partial x_i}} = q {\partial\over{\partial x_i}}\left({{x_i}\over{r^3}}\right) = q\left({1\over{r^3}} - {{3r^2 {{x_i}\over{r}} x_i}\over{r^6}}\right) = 0.$$Thus, $\text{div}\,F = 0$ away from the origin. Consider now a ball $B$ of radius $r$ centered at the origin contained entirely in the interior of $D$. Then$$\int_B F \cdot \mathbf{N}\,dA = q \int_b {r\over{r^3}}\,dA = {q\over{r^2}} \int_B dA = {{4\pi r^2 q}\over{r^2}} = 4\pi q.$$Finally, consider the manifold $M$ consisting of the space between $B$ and $D$ with those as boundary. Then, by the Divergence Theorem,$$\int_B F \cdot \mathbf{N}\,dA - \int_D F \cdot \mathbf{N}\,dA = \int_M \text{div}\,F\,dV = 0 \implies \int_D F \cdot \mathbf{N}\,dA = 4\pi q.$$

$\endgroup$

Your Answer

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