Measured data for $n(E)$ of air were fitted to equations for $n(E)$ and $k(E)$. The measured data for n(E) spanned 0.734 to 6.702 eV. To obtain fits to the equations for n(E) and k(E) in the absence of measured k(E) data, measured k(E) was taken to be zero. The calculation yielded negative k(E) over part of the fitted range, of the order of 10-9. The equations for n(E) and k.(E) were based on 1st order time dependent perturbation theory and were consistent with the principal of causality.
$\begingroup$
$\endgroup$
2
-
$\begingroup$ What was the error bar on your fit? At what confidence level can you reject the hypothesis that $k(E) \geq 0$? $\endgroup$– Michael SeifertCommented Mar 6, 2023 at 22:16
-
$\begingroup$ You should give a reference, preferably a link, where the equations you are talking about are defined. $\endgroup$– anna vCommented Mar 7, 2023 at 5:42
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
It is possible that taking $k(E)$ to be zero in some areas is incompatible with the data for $n(E$) because of the Kramers-Krönig theorem.
