home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!bonnie.concordia.ca!cerberus.ulaval.ca!cornu.phy.ulaval.ca!yergeau
- From: yergeau@cornu.phy.ulaval.ca (Francois Yergeau)
- Newsgroups: sci.physics
- Subject: Re: Blue Sky
- Message-ID: <1992Jul28.153403.10122@cerberus.ulaval.ca>
- Date: 28 Jul 92 15:34:03 GMT
- References: <Jul.27.14.46.15.1992.6472@ruhets.rutgers.edu> <1992Jul27.224915.15755@das.harvard.edu>
- Sender: news@cerberus.ulaval.ca
- Organization: Universite Laval, Quebec
- Lines: 43
-
- In article <1992Jul27.224915.15755@das.harvard.edu> love@geophysics.harvard.edu (Jeff Love) writes:
- >Wow. Okay, maybe I didn't read Jackson's book before I posted my question.
- >I guess what I was hoping for was a heuristic arguement as to why blue light
- >is scattered whilst red light is not (at least not as much), or maybe a simple
- >dimensional arguement could yield the k^4 dependence; I admit that I haven't
- >tried to put one together.
-
- In his book "Optics", K.D. Moller offers just such an argument, but it
- requires that you admit a couple of things that only further study will
- prove (one may try Jackson). Here goes:
-
- We have a polarizable body of index n2, imbedded in a medium of index
- n1. A wave impinges on the body, inducing forced dipole oscillations
- in it. Said dipole oscillations will in turn re-radiate an outgoing
- wave, whose amplitude will be inversely proportional to the distance
- from the body. Let's try to build an expression for the outgoing
- intensity, using dimensional analysis:
-
- I
- - = f(r,V,lambda,n1,n2,theta,phi)
- I0
-
- where r is the distance from the scattering body, theta and phi denote
- the direction, V is the volume of the scattering body and lambda the
- wavelength. n1, n2, theta and phi fall out of the analysis, being
- dimensionless. Now if we assume that the _amplitude_ of the dipole
- radiation is proportional to the volume of the body and inversely
- proportional to the distance, the _intensity_ will be proportional to
- V^2/r^2, or length^4; we must then have a lambda^-4 dependance to
- balance the units.
-
- QED, though not that convincing.
-
- BTW, the often quoted requirement that the body must be much smaller
- than lambda is implicit in the assumption that the radiated amplitude
- is proportional to V. If the body is too big, dipole radiation from
- different parts of it will be out of phase and cancel out, and the
- total amplitude will not grow as V anymore.
- --
- Francois Yergeau (yergeau@phy.ulaval.ca) | Quand le doigt montre la lune,
- Centre d'Optique, Photonique et Laser | l'imbecile regarde le doigt.
- Departement de Physique | -proverbe chinois
- Universite Laval, Ste-Foy, QC, Canada |
-