Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Potentials and curl

  1. Oct 23, 2008 #1
    I'm trying to get a picture of magnetic potential, as in how to relate it to spin precession.
    (So have I got the right picture so far?)

    Classically, B is measured as the derivative of curl (which is a circulation integral) "around" a conducting current (that is, perpendicular to the direction of current I).
    Gauss' Law says the field is divergence-free, which means it circulates linearly, and the integral is just BL, for any circular integral along a circumference L, of a potential around say, a 1-d current in a wire.

    Einstein and de Haas tried to measure precessional forces to explain magnets, by envisaging electrons as being in orbits and precessing, as a gyroscope does in a gravitational potential. It didn't quite explain magnetism because it didn't allow for Pauli's exclusion principle that 'admits' two electrons, which share an orbital; magnetic forces are in fact related to electron's spin angular momentum, not their orbital momentum. Or electrons rotate "in place", rather than orbiting like gyroscopes.

    Aharonov-Bohm phase precession is due to a magnetic potential in a curl-free region, which rotates the spin precession angles in a divided beam of electrons.
    It looks like a bump for 1/2 the beam, and like a dip for the other 1/2. One spin phase shifts up or 'squeezes' over the dip, the other shifts down or 'expands' over the bump in potential. The phase-difference is preserved, as momentum carries the matter-wave to a detector.

    So it's about how to relate the geometry of a magnetic field (potential) to the geometry of spin in electrons.
    Is potential just equal to the precession angle, since a change in this angle, will change the local potential gradient?

    It's a question (I think), of seeing the geometry and topology involved. The algebra is the really tricky bit with complex quantum spaces with a dimension of [tex] \mathbb C^{2n} [/tex]. Anyways, there's this somewhat dated article in a '81 SciAm about the A-B geometry and neutron spin-precession. Quite an interesting geometric picture; they relate geodesics (over an inclined conic surface) to parallel transport, then what a "path-lifting" does, and so on.

    You can relate the observed patterns of interference to the topology and geometry of spin precession. Spin is a 'component' of the Schrodinger wavefunction(?), and charge is too(?), whereas position and momentum are another component, or IOW the wavefunction is a "mixed wave" of spin, charge, and mass wave components? There's a major difference between the A-B experiment and neutron interferometry, wrt to the magnetic field applied, since the neutrons cross classical field-lines, or do not travel through a curl-free region, but see a field that rotates the precessional angle, over time, whereas the electrons see just a potential and preserve a single phase shift?

    Have I left something out of this picture?
  2. jcsd
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Can you offer guidance or do you also need help?
Draft saved Draft deleted