Re: dielectric constant

[Ludwik Kowalski <kowalskiL@MAIL.MONTCLAIR.EDU>]

"John S. Denker" wrote:

> > Only in gasses is K very close to unity at w->0.
>
> So why not conclude that figure 31-5 (on page 31-8)
> represents a dilute gas?  Further evidence in support
> of this conclusion comes from the fact that n peaks
> at around n=2 or n=3;  in a dense substance it could
> be a million or so.  Atomic resonances have high Q.

Aren't gas resonances atomic resonances? Intuitively
I would expect higher Q for gasses because atoms
do not interact. Therefore "peaking" for gasses would
by higher than for solids and liquids.

> > Somehow I feel that nobody can
> > calculate n(w) quantitatively to produce agreement
> > with experimental data for real substances,
>
> What's the evidence for this inability?  Spectroscopy
> is a pretty well-established discipline.  Spectroscopists
> have been calculating lineshapes to high accuracy for a
> long time.  Really high accuracy.

I have no evidence. My feeling is justified by the absence
of clear statements in textbooks. For example, something
like this: " the wavelength dependence of n for glass or
water can be very accurately calculated in QED."

The line shape, I suppose, depends on parameters of two
levels. How can this be translated into the frequency
dependence of n or K?
Ludwik Kowalski