Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/5/84; site mordor.UUCP
Path: utzoo!watmath!clyde!cbosgd!cbdkc1!desoto!packard!ihnp1!ihnp4!zehntel!dual!mordor!space@mit-mc
From: space@mit-mc
Newsgroups: net.space
Subject: Re: photon wavelengths and solar-sail
Message-ID: <791@mordor.UUCP>
Date: Fri, 1-Mar-85 17:44:57 EST
Article-I.D.: mordor.791
Posted: Fri Mar  1 17:44:57 1985
Date-Received: Sun, 3-Mar-85 04:25:00 EST
Sender: daemon@mordor.UUCP
Lines: 68

From: Rick McGeer (on an aaa-60-s) 

	Sorry, my message was quite incorrect.  As another correspondent
has pointed out, it is not the wavelength of the photon that is affected,
but rather the vector. Since momentum is a vector quantity, not scalar, the
total net momentum transfer to the sail is 2P, where P is the momentum of
the entire set of photons impacting the sail.  To see how this works,
remember that the *vector* momentum is conserved; to a first approximation,
we imagine the sail to be travelling in a linear direction x, with 0
momentum in this reference frame.  We then imagine a solid photon wavefront
with aggregate momentum P travelling in the positive x axis.  The photon
rebounds in a perfect elastic collision along the x axis...

Total Momentum before collision:

P (photons) + 0 (sail)

Total Momentum following collision:

-P (photons) + x (sail)

equating and solving, we get x = 2P.   (1)

	Of course, this is idealized.  The total momentum transfer to the
sail can be obtained by integrating the momenta of the photons before and
after the collision.  This is related to wavelength by the DeBroglie formula:

	mv = hf,

where h is Planck's constant and f = 1/lambda.

	The wavelength will have lengthed by Compton's formula:

	lambda' = lambda + h/mc (1 - cos theta)

where m is the mass of the particle that the photon collided with, and theta
is the angle of the deflection...

	In our idealized case, theta = pi, hence (assuming the entire mass
of the sail is involved in a collision), we get (note I'm assuming coherent
light):

	lambda' = lambda + 2 h / mc  (2)

	writing lambda as l, the total momentum of the photon wavefront
post-collision is:

	h/(l+2h/mc)

	hence the total momenta of the photon post-collision is:

	P[1-2h/(clm+2h)] (3)

	as a result, there's a net loss of momentum transfer from (1) due to
the (trivial) redshift noted in (2).  The total net momentum transfer to the
sail is then:

	2P[1-h/(clm+2h)]


	Bottom line: yes, as I said yesterday, there *is* a redshifting
effect.  However, its net effect on the momentum of the light sail is
*negative*, not positive, and it is trivial.  Sorry about that.  Hope this
clears everything up.  [NB -- the effect here is the *maximum* loss due to
the Compton effect...]


						Rick.