1. What is the dielectric constant of a star, especially its corona? Is it of order 1 or is it quite large?

  2. How much do "impurities" (elements other than H and He) affect the dielectric constant?

  • $\begingroup$ Thanks for the edit, I've adjusted the wording a bit to better fit the site. Just fyi the answer will be different at radio frequencies and optical frequencies (visible light). $\endgroup$ – uhoh Mar 18 '20 at 1:19
  • 1
    $\begingroup$ Trying for the world's biggest capacitor here? :-) . Now, as pointed out, plasmas conduct, but depending on the frequency band you're interested in, you may find some ranges where the plasma does hold energy rather than conduct it. $\endgroup$ – Carl Witthoft Mar 18 '20 at 19:28
  • 2
    $\begingroup$ @JamesK there's nothing wrong with talking about the dielectric constant of conductors or other lossy materials; they are simply complex valued rather than real: en.m.wikipedia.org/wiki/Permittivity#Complex_permittivity $\endgroup$ – uhoh Mar 19 '20 at 9:24
  • 1
    $\begingroup$ @CarlWitthoft ditto. $\endgroup$ – uhoh Mar 19 '20 at 9:25
  • 1
    $\begingroup$ @JamesK I found Determination of the Dielectric Constant of a Plasma which gives a formula and might be helpful for the corona - I have to think of it. Do you think we could estimate all the parameters? $\endgroup$ – B--rian Apr 15 at 8:14


This not (yet) a full answer, since I only focus on approaching part 1 of the question, namely the calculation of the dielectric constant of the corona.

Dielectric constant of the corona

The Sun's corona

... extends millions of kilometres into outer space and is most easily seen during a total solar eclipse [...] measurements indicate strong ionization in the corona and a plasma temperature in excess of $1\,000\,000 {\rm K}$ much hotter than the surface of the Sun.

So let's dig a bit into plasma physics.

According to I. M. Podgornyi's article Determination of the Dielectric Constant of a Plasma, in the absence of a magnetic field, the dielectric constant $\varepsilon$ of a plasma can be written

$$ \varepsilon(\omega) = 1 - {{4\pi {e^2}n} \over {{m_e}{\omega ^2}}} \cdot {1 \over {1 - i{{{v_{col}}} \over \omega }}}$$

where $\omega$ is the angular frequency of the electromagnetic wave, $e = 1.60217662 \cdot 10^{-19} {\rm C}$ is the charge of an electron, $m_e = 511 {\rm keV} = 9.10938356 \cdot 10^{-31} {\rm kg}$ is the mass of an electron, $i$ the imaginary unit.

The frequency of collisions $ν_{col}$ with charged particles is given by

$${v_{col}} = {{{{2.10}^{ - 5}}{n_i}Z} \over {T_e^{3/2}\left( {\rm eV} \right)}}.$$

Here, we have $T_e \approx 80\ldots90 {\rm eV} = 10^6 {\rm K}$ the temperature of a plasma given in electron volt, and (average?) $Z$ the charge number.

The next step towards an answer would be to figure out all parameters in this equation, in particular the average composition of the plasma (in terms $n$ and $Z$).

I did not yet manage that (yet), see also my question Physical properties of the stellar corona?


The following publications might be helpful for determining $n_i$ and $Z$ for some or all consitutents of the coronal plasma:


  • 1
    $\begingroup$ The solar corona is practically a collisionless plasma, so you can safely neglect the collision term here (even in the photosphere it should be negligible unless $\omega$ corresponds to radio waves). $\endgroup$ – Thomas Apr 20 at 21:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.