Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] refraction question



On 04/23/2010 06:43 AM, Edmiston, Mike wrote:
In response the this question, daryl@darylscience.com gave an
explanation based upon the light being continuously absorbed and
reemitted. I've seen this explanation in several physical science
textbooks aimed at both the high school level and also the conceptual
physics level in college.

It's also in Feynman ... not exactly the introductory
or conceptual-only level.

I don't like this explanation very well. Do the rest of you accept
this?

There is a good physics argument to be made.

OTOH as the proverb says, no matter what they're doing,
they can always do it wrong.

In this case that means encrusting the physics with
layers of bogus assumptions and "interpretation".

This explanation seems to promote the idea that the
interaction between the E&M radiation and the atoms in the medium
involve complete absorption that excites an electron to a higher
quantum state, then reemission.

The right way to say it is in terms of scattering.

Interpreting the scattering in terms of "complete"
absorption and discrete "quantum states" is unhelpful
and has no basis in physics.

This is the description we use for
spectroscopic line absorption and line emission.

That's also unhelpful. It is spectacularly unhelpful
in the wings of the line, far from the center of the
resonance.

There are lots of things involving light and/or atoms
that don't involve "complete" transitions between
discrete "quantum states". The idea that "everything"
must be fully discrete is one of those ideas, like
the Fitzgerald-Lorentz contraction, that experts in
the field stopped using 100 years ago, but somehow
still lingers in Parade magazine and in intro-level
textbooks.

Example: Pulsed NMR: Suppose you have a single
two-state system. You can apply a π pulse that
flips the spin from one quantum state to "the
other" quantum state. But what if you apply a
π/2 pulse???? It's not hard to do.

Example, perhaps more on-topic: In some sense,
the low-frequency limit of the index of refraction
is the _dielectric constant_. I assume we all
agree that a box of argon gas has a nontrivial
dielectric constant, and that it does not involve
"complete" transitions to excited atomic energy
levels. OTOH the magnitude of the dielectric
constant can be computed _in terms of_ the energy
levels via perturbation theory..........

The assumption that "everything" is completely
quantized is high on the list of widespread and
pernicious misconceptions.