🚀 go-pugleaf

In article <1iamerINNkuq@flash.pax.tpa.com.au> anon.281@pax.tpa.com.au writes:
%
%
%Here is a repost of a recent posting I made.  The other posting was cut
%off after a series of dashes I had in the posting due to the anonymous
%posting software. 
%=================
%
%I have a question regarding the Quantum Eraser experiment described in
%Scientific American July 1992 (see below) performed by Raymond Y. Chiao of the
%University of California at Berkeley.  First the relevent passages from the
%article below (I think copyright laws allow me to do this for the purposes of
%discussion) then my question about the experiment.

%Here's a diagram of the apparatus:
%
%                                     D
%                 ____M____       |_______|
%                    / \             /
%          S ==> \\/     \       \\/
%           __   /\\       \     /\\ <== F
%Incident  |  \/             \ /
%=======>==| C >           ======== <== B 
%Photon    |__/\             / \
%                \         /     \// <== F
%                  \     /       //\
%                 ___\_/___        __\___
%                     M           |      |
%                                     D
%
%C = Down Converter
%S = Polarization Shifter
%M = Mirrors
%B = Beam Splitter
%F = Polarizing Filters
%D = Detectors
%
%
%I assume that you can know the polarization of the incident photon on the
%down-conversion crystal (otherwise you could not know which path each photon
%took with the polarization shifter there anyway).  After one photon
%has passed through the polarization shifter, we then have two photons polarized
%at 90 degrees to each other.  Let's define the axis parallel to the direction
%of polarization of one photon the x-axis, and the axis parallel to the direction
%of polarization of the other photon the y-axis.  Now, my initial problem was
%with understanding how the polarizing filters "erased" the information -
%initially I had visualised the polarization of each filter being parallel to
%each of the photons (which wouldn't work).  However, I think it *would* work in
%erasing the information if the filters were each placed at 45 degrees to the x
%and y axes, as then each filter would have a 50% chance of passing each
%photon.  The orientation of the filters isn't described anywhere in the
%article.  So is my understanding correct?

Your understanding is correct.  The two photons are produced horizontally
polarized, and the waveplate (when positioned at 45 degrees to the hor-
izontal) rotates one of them to vertical polarization.  We showed that
as you change the waveplate angle from 0 to 45, the visibility drops from
close to 100% down to 0.  

In order to "erase" the information, the polarizers must be placed at
45 degrees to the horizontal, as you conclude.  We also presented data
showing how as you turned them towards 45, the visibility increased again.

Actually, you can also place one at 45 and the other at -45; in this
case, you get an "antifringe."  Instead of a dip, you observe a peak.

All of these phenomena can also be understood as a manifestation of the
effective spin-singlet state produced by the interferometer, which is
the same kind used in studies of Bell Inequalities.  Individually, the
photons are unpolarized, but whenever you detect a photon at one port
after a polarizer, the other photon immediately becomes orthogonally
polarized.  Once you change the path lengths to move off the dip, however,
this changes, and you simply have two non-interfering photons, one
horizontal and one vertical.  In that case, if you put both polarizers
at 0 degrees, you get no counts at all even away from the dip.  On the
other hand, if you put one at 0 and the other at 90 so you get a lot of
counts far from the "dip," you also get a lot of counts at the "dip."
-- 
Aephraim M. Steinberg                    | "WHY must I treat the measuring
UCB Physics                              | device classically??  What will
aephraim@physics.berkeley.edu            | happen to me if I don't??"
                                         |       -- Eugene Wigner