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: reflection and refraction



REPOSTING; IT WAS PROBABLY LOST SOMEWHERE

Recent philosophical comments are profound and interesting. Especially
those of Leigh and Bob Schiamanda. I am still meditating on what John M
wrote about the nature of "truth" in physics. To me a prediction is
truthful if it is confirmed in an experiment. Practical accuracy limits
are not the main issue in classical physics. Perhaps more people will
contribute to the "explaining explain" thread. We are not philosophers
but we are trying to explain things to people; the meaning of what it
means is worth debating.

But let us return to physics. We were trying to understand (explain to
ourselves) propagation of light through transparent materials. Then we
were distracted by the "what is understanding" question. It is an
experimental fact that the speed of light in glass is about 2/3 of its
speed in a vacuum. The Huygens principle ("it takes time for light to
be absorbed and reemitted by electrons") was used to explain this
behaviour. The value of v approaches c when the air pressure is decreased.

I have a problem with solids and liquids. Let me introduce it in a very
simple conext. Light travels through a fiber cable whose diameter does
not exceed one wavelength. (A technical term "monomode" is often used to
describe this kind of an optical wave guide). The cable is one meter long.
Suppose it is made from pure SiO2. The size of each molecule (3 nuclei
and 44 electrons) is about 3*10^-10 meters and molecules are closely
packed. We can thus say that the cable is composed of 3*10^9
"monomolecular layers".

At the speed od 2*10^8 m/s the time to pass through the cable is 5 ns,
or 2*10^-18 seconds per layer. Is this enough to abrsorb and reemit
a wave? Keep in mind that the wave frequency is of the order of
5*10^14 Hz (T=2*10^-15 s). How can a wave be absorbed and reemitted
during 1/1000 of its period?

Also keep in mind that the phase shift introduced by a plate of glass
of thickness lambda/4 (500 layers) is EXACTLY equal to 90 degrees. Does
it mean that 5 layers of glass will introduce a 0.9 degree shift
(without a wide spread of values around 0.9 degrees and without a large
not-shifted component)? And if my recollection is correct, 5 monoatomic
layers of a metal (evaporated on glass) are already very opaque.
Something does happen to light in a layer whose thickness is very small
in comparison with the lambda.

Why is the speed of light in water (three nuclei and 18 electrons per
molecule) higher than in SiO2? Why is speed of light in diamond (one
nucleus and 6 electrons per "molecule") lower than in SiO2? I am
referring to the radii of circles we draw when explaining (geometrically
deriving) the laws of reflection and refraction. The geometric reasoning
leads to truthful predictions about the relations between angles. Does it
mean we understand propagation of light in transparent materials? Or is it
a description without an explanation? I wish I had the Feynman's book
which was quoted here several days ago. Does he deal with the above
issues?
Ludwik Kowalski
P.S.
Who was that philosopher who said ...? something like --> "I do not
have to understand the chemistry of digestion to enjoy a good meal"?