home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.physics
- Path: sparky!uunet!gatech!destroyer!wsu-cs!igor.physics.wayne.edu!atems
- From: atems@igor.physics.wayne.edu (Dale Atems)
- Subject: Re: The instantaneous transfer of information in QM calculations
- Message-ID: <1993Jan11.031132.1521@cs.wayne.edu>
- Sender: usenet@cs.wayne.edu (Usenet News)
- Organization: Wayne State University, Detroit, MI
- References: <481@mtnmath.UUCP> <1993Jan10.164016.16419@cs.wayne.edu> <484@mtnmath.UUCP>
- Date: Mon, 11 Jan 1993 03:11:32 GMT
- Lines: 75
-
- In article <484@mtnmath.UUCP> paul@mtnmath.UUCP (Paul Budnik) writes:
- >In article <1993Jan10.164016.16419@cs.wayne.edu>, atems@igor.physics.wayne.edu (Dale Atems) writes:
- >> This is certainly not true of the spin-1/2 analog. I can do that
- >> calculation using only the "linear" part of QM, and the prediction is
- >> valid as interaction with Stern-Gerlach magnets does not affect the
- >> spin part of the wave function. The predicted correlations depend only
- >> on the singlet state wave function before detection; what happens to the
- >> wave function as a result of detection is not relevant.
- >
- >Then you will not be able to prove Bell's inequality is violated. It is
- >important to understand that tests of Bell's inequality involve more than
- >just two detections. They also involve a knob or device for affecting those
- >detections. In the photon case it is a polarizer. In the case your describe
- >it would be some apparatus that would force the spin to be aligned along
- >a certain axis if a detection is made. You will only be able to prove
- >the expected correlations violate Bell's inequality if the measurement
- >or alignment of the spin of one particle influences the probability of
- >detection of the other. Since these two events, the alignment of the spin
- >of one particle and the detection of the other are space-like separated,
- >the only law of physics that would predict such an influence is some
- >version of quantum collapse.
-
- |
- |\
- | \ |
- | \ |
- | \ --- --- |
- | -----------------[ ]---------------------- |
- | --- --- \ |
- | Magnet 1 Source Magnet 2 \ |
- | \ |
- | \|
- | |
- Detector 1 Detector 2
- Site 1 Site 2
-
-
- Consider an experimental setup designed to test the version of Bell's
- inequality that applies to spin-half particles in the singlet state. The
- particles are charged and are deflected by Stern-Gerlach magnets whose
- orientation determines which component of spin we are measuring. Let's
- say one deflector is oriented to determine the spin along unit vector a,
- the other along unit vector b where the directions of a and b are
- arbitrary. The place where the particle lands depends on which way it
- was deflected and thus determines whether the measured component of
- spin was + or -.
-
- Bell's inequality for this situation takes the form of a relationship
- between correlation functions < (S(1,a) S(2,b) > where S(i,a) is the
- component of particle i's spin along a, and similarly for b. To determine
- such a correlation function experimentally, *in principle* all you need
- to do is repeat the experiment a sufficient number of times, measuring
- the spins of both particles and forming the product S(1,a) S(2,b) for
- each pair, sum the products and divide by the number of pairs detected.
- I don't see where you need to assume anything about what happens to the
- wave function of one particle after the other is detected. One simply
- measures spin components and computes a statistical average.
-
- Of course, a finding that the observed correlations violate Bell's
- inequality says nothing about locality if they could have been produced
- by an exchange of information. As I understand it, the basic
- premise behind Bell's proof is that all components of each particle's
- spin are fixed when the singlet state is prepared and do not change
- afterward. I believe that the point to making the relevant intervals
- space-like is to ensure that the spins cannot be prepared according
- to the settings of the magnets by signals sent from the magnets to
- the source, and cannot be changed during flight by signals sent
- from one particle to the other, or from a magnet or detector to
- the distant particle.
-
- ------
- Dale Atems
- Wayne State University, Detroit, MI
- Department of Physics and Astronomy
- atems@igor.physics.wayne.edu
-