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

I Stern-Gerlach using magnets

  1. Aug 11, 2017 #1
    Is there any study involving real magnets behaviour shot through Stern-Gerlach gates?
    I've seen something using rather large magnets[1], but still the pattern showed two bulges. I if suppose using very small magnets the split-up would be much clearer. After all these magnets are not simply classical objects, but they consist of particles with quantum behaviour.

    [1]
     
  2. jcsd
  3. Aug 11, 2017 #2

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Why do you think silver atoms are not "real magnets"? How do you define "real magnets"?
     
  4. Aug 11, 2017 #3
    Oh, in fact my intention was to delete the word real. There should be an advantage of using small regular magnets (neodymium for example), that would consist of a number of atoms like from 1 micron total size down to even 10 atoms (at this size I suppose we don't need magnetic materials anymore) or less. The advantage is you could literally watch them as the pass through. Silver atoms could do but I'm not sure if you can see their orientation.

    If we use for example 1um neodymium magnets the number of atoms oriented in the N-S direction of the magnet is not 100% but is much lower whereas for a single atom the entire field is produced by atom itself.
     
    Last edited: Aug 11, 2017
  5. Aug 11, 2017 #4

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    A 1 micron iron ball will have tens of billions of atoms. You aren't going to see QM effects that way.

    A 10 atom iron ball is not more visible than a silver atom. I don't see the point.
     
  6. Aug 13, 2017 #5
    There are ways of manipulating individual atoms but I'm not sure if you can make them keep their magnetic moment lined up to their lattice. But I was only interested in some experiments tracking normal magnets.
     
  7. Aug 13, 2017 #6

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    I still don't see the point. (See #4)
     
  8. Aug 16, 2017 at 5:18 PM #7
    In the experiment with real magnets it seems that unlike iron balls, the magnets would separate into two groups. The smaller the magnets, I suppose, the better they would separate. However, electrons can remember states from previous measurements. I don't expect these little magnets to do that.
     
  9. Aug 16, 2017 at 6:38 PM #8

    PeterDonis

    User Avatar
    2016 Award

    Staff: Mentor

    What do you mean by this?
     
  10. Aug 17, 2017 at 5:44 AM #9
    I mean you can prepare some particles by a measurement for example for up/down axis, then all particles that were found in up state, will be found in an up state on the following measurement that needs to be also for Up/down. "Remember" is not the right word, though.
     
  11. Aug 17, 2017 at 8:13 AM #10

    Nugatory

    User Avatar

    Staff: Mentor

    This is about as mysterious as the way that the books on my bookshelf remember that I put them there yesterday so they're still there today and will be there tomorrow if someone doesn't take them out. Why would you expect the magnets to behave differently?

    And a not completely unrelated question: what is the classical prediction for the behavior of the magnets in that video you posted at the start of the thread?
     
  12. Aug 17, 2017 at 3:40 PM #11
    The classical prediction results are similar to the experimental results. But I've seen many presentations (not conclusions from scientific research) that say the magnets will do roughly an even distribution between the two poles.
    http://www.m-hikari.com/astp/astp2011/astp13-16-2011/alrabehASTP13-16-2011.pdf

    For the first question, supposing we pass an electron beam through a SG gate oriented up/down (we ignore the Lorentz force). As half of the beam is deflected up, half down, they will separate into two groups. Up to this point the magnets should do the same (the Gaussian distributions will look similar but electrons will be more localised). If we place a second SG gate oriented the same direction and pass the upper beam through , I don't see any reason the magnets will all be deflected upwards as electrons will, because there is no difference between the magnets from the upper beam and lower beam unless you use a SG apparatus that has two identical poles. This way the upper beam will orient N-S and the lower, S-N or vice-versa, depending on what poles we use. But in a experiment the magnets may act like electrons. I suppose it is possible to create a quantum mechanical model (instead of the classical one) that can show what will happen.
     
  13. Aug 18, 2017 at 2:49 AM #12

    vanhees71

    User Avatar
    Science Advisor
    2016 Award

    The point is that the classical prediction leads to a continuous spread of the silver atoms due to the thermal initial conditions of the ensemble (in the original Frankfurt setup of Stern and Gerlach's experiment of 1922). The great surprise was that rather they found what was called "direction quantization", predicted by the wrong Bohr-Sommerfeld model, where they were lucky that they didn't predict three lines rather than the observed two without knowing the possibility of half-integer spins yet and the correct amount of the deflection because they didn't know about the correct gyro-factor of 2 for spin-1/2 particles but calculating effectively with spin 1.

    A very amusing historical account on the SG experiment, including the importance of heavily smoking cheap cigars in the lab, can be found here:

    http://physicstoday.scitation.org/doi/10.1063/1.1650229
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted