1
$\begingroup$

The problem is of a conducting medium in a uniform field $E_0$. A spherical cavity of radius $a$ is formed inside said medium.

I already found the potential inside and outside the cavity

$\Phi_{\text{int}}=- \frac{3}{2}E_0r\cos(\theta)$

$\Phi_{\text{ext}}= -E_0r\cos(\theta)-\frac{1}{2}E_0a^3\frac{\cos(\theta)}{r^2}$

But I can't find a way to find the surface charge density on the cavity.

It occurs to me to take the value of the potential $\Phi_{\text{int}}$ and calculate the electric field using $E=-\nabla \Phi_{\text{int}}$ and then use Gauss's law to find $\sigma$ but it doesn't work out the result. The book tells me that the result for this density should be

$\sigma=-\frac{3}{2}KE_0\cos(\theta)$

But I don't know who that "$K$" is. I guess it's the dielectric constant which is defined as $K=\epsilon/\epsilon_0$. If it is the dielectric constant, I don't know where to get it from. Can you give me any suggestion?

$\endgroup$
4
  • $\begingroup$ Yes, you should use Gauss's law, as you suggest. This is a common/typical homework problem so you should be able to find many examples by googling for things like "surface charge density" and "gauss's law." $\endgroup$
    – hft
    Commented May 27, 2022 at 18:08
  • $\begingroup$ In particular, integrate the maxwell equation $\nabla \cdot E = \rho/\epsilon_0$ over a pill box that has one side in the cavity and one side in the material and you will see how the surface charge is related to the field difference. $\endgroup$
    – hft
    Commented May 27, 2022 at 18:09
  • $\begingroup$ Actually, now I am a little confused, since both your fields seem to have zero radial component at the boundary of the cavity (i.e., $E_r=0$ at $r=a$). $\endgroup$
    – hft
    Commented May 27, 2022 at 18:24
  • $\begingroup$ @hft Sorry, the inner field was missing an $r$. I already modified it $\endgroup$
    – Kale_1729
    Commented May 27, 2022 at 18:27

1 Answer 1

0
$\begingroup$

I already found the potential inside and outside the cavity

$\Phi_{\text{int}}=- \frac{3}{2}E_0r\cos(\theta)$

$\Phi_{\text{ext}}= -E_0r\cos(\theta)-\frac{1}{2}E_0a^3\frac{\cos(\theta)}{r^2}$

Integrating $\vec \nabla \cdot \vec E = \rho/\epsilon_0$ over a pillbox (small cylinder) oriented with its long direction along $\hat r$, we find: $$ \int_{V_{pillbox}}dV \vec \nabla \cdot \vec E = Q_{enc}/\epsilon_0 = \int_{boundary}\vec {dS}\cdot \vec E = \delta S(E_{ext}^r(r=a) - E_{int}^r(r=a)) $$ So, that: $$ \sigma/\epsilon_0 = -E_{int}^r(r=a)\;, $$ since $E_{ext}^r(r=a)=0$.

Or: $$ \sigma = \epsilon_0 \frac{\partial \Phi_{int}}{\partial r} = - \frac{3}{2}\epsilon_0E_0\cos(\theta) $$


Another way to approach this problem is to open up your copy of Jackson's "Classical Electrodynamics" (3rd Edition) to page 18 and see that: $$ (\vec D_2 - \vec D_1)\cdot \hat n = \sigma\;, $$ but in our case, we have $\vec D_2\cdot\hat n = 0$ at the boundary (per your equations). And we have $-\vec D_1\cdot\hat n = -\epsilon_0E_{int}^r$, since the cavity is empty.

So, again we find: $$ \sigma = -\epsilon_0E_{int}^r $$

$\endgroup$
3
  • $\begingroup$ Exactly, that's what I find. But in the book, the answer it gives me doesn't have $\epsilon_0$, it has $K$ and that's what has me confused $\endgroup$
    – Kale_1729
    Commented May 27, 2022 at 19:36
  • $\begingroup$ Don't know what to tell you. What book are you using? $\endgroup$
    – hft
    Commented May 27, 2022 at 20:19
  • 1
    $\begingroup$ I use Reitz Milford's Fundamentals of Electromagnetic Theory. But I'll leave it as you put it. I'm not going to cling to the answer the book gives. It may be wrong. Thank you very much $\endgroup$
    – Kale_1729
    Commented May 27, 2022 at 23:06

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.