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

'Detection' at a slit

  1. Sep 29, 2010 #1
    (1) In the double slit experiment, if I fire individual electrons at the slits, is there a chance that an electron could interact with one of the atoms making up a slit in such a way as to alter the energy state of one of its shell electrons. i.e. I could imagine there is a probability of ionization? If this change of state occurs at a slit this would presumably mean that the slit that the electron went through could in principle be determined even though I choose not to bother doing this.

    (2) Is this a factor when doing the experiment for real? e.g if I set the initial energy such that this interaction has a high probability of occurring, then do I lose the interference pattern?

    (3) ...Or... is it in fact essential to transfer this information from the quantum-microscopic world to the classical-macroscopic world in order to lose the interference pattern?

    (4) Extending this further, does this mean that any interaction (i.e. transfer of energy) between two quantum objects in free space is an 'observation' and would cause a collapse of their wavefunctions?

    (5) Finally, on a similar note, what, in fact, causes the incident wavefunction to diffract at the slit? is it simply the potential well formed by the atoms of the slit? If so, do we assume that most of the time there is no change of state to the atoms forming the slit, i.e. that the potential well is a 'static and unchanging' environment that our incident wavefunction moves through.


  2. jcsd
  3. Sep 30, 2010 #2
    Isn't it so that you are implicitly assuming that there is an "electron" traveling along some path, while it has been repeatedly said that one should not think this way because it leads to contradictions - unless you move to some alternative version of the theory, like the one by Bohm and de Broglie.
  4. Sep 30, 2010 #3
    @arkajad I think you missed his point.

    What I understood the question to be was:

    "Is it possible for an electron during a double slit experiment to interact with the damn wall."

    Which is a very good question, IMO. In these kinds of thought-experiments, professors and textbooks greatly simplify the problem so the student can focus on the main issue.

    Once we think about how to put together a real experiment with real sources of error, the question becomes important -- what constitutes a detector?

    My own personal guess is that with the appropriate setup, the amplitude to interact with the wall is small enough to ignore. Surely, if the numbers experiment aren't right, there are also amplitudes for the electron bouncing off the wall and the electron getting lodged within the wall and amplitudes for the electron going through the parts of the wall with no holes. In a perfect world, we could account for all these.

    But the basic rule is that in a given quantum event, the more interaction that occurs, the smaller the amplitude. Each event involving an interaction with the wall (including being detected by it) individually have a much smaller amplitude. As long as the sum of all the interaction events is still negligible, we can ignore it in our analysis.

    I don't think the wall's ability to detect would destroy the interference completely. It's just as it is for the case when using an unreliable detector at one of the holes: the interference pattern is eroded, but not gone.
  5. Sep 30, 2010 #4
    Of course. It can be easily calculated and modeled with a computer. All depends on the strength and the details of the coupling. When the one slit detects - you have a reduction of the wave packet. But when the coupling is weak - it will be a rare event. Moreover the details of such reductions depend on the discriminative power of the "slit detector" and on what will it detect: position, momentum, energy, some other quantity? The devil is in the details.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook