Double slit - properties of slit?

[Nugatory]

There is the screen housing the slits

You can save yourself just a ton of grief by calling that a "barrier"

 

[rude man]

You can save yourself just a ton of grief by calling that a "barrier"

10-4. 'Barrier' it will be from now on.

 

[akhmeteli]

I'm not sure what percentage of emitted electrons make it thru the slits but any that don't are absorbed in,or reflected by, the slit screen. They will not affect the interference pattern. Only electrons that make it thru the slits are recorded on the phosphorescent screen downstream.

I am trying to say that electrons passing through a slit at some distance from the edge of a slit are also affected by the slit screen (they interact with their image charges, which depend on the properties of the slit screen).

 

[PeroK]

I am trying to say that electrons passing through a slit at some distance from the edge of a slit are also affected by the slit screen (they interact with their image charges, which depend on the properties of the slit screen).

You're proposing different interrefence patterns depending on the composition of the barrier with the slits?

In fact, we can simplify the experiment to a single slit: you're proposing different single-slit diffraction based not only on the width of the slit but the composition of the barrier?

 

[akhmeteli]

You're proposing different interrefence patterns depending on the composition of the barrier with the slits?

In fact, we can simplify the experiment to a single slit: you're proposing different single-slit diffraction based not only on the width of the slit but the composition of the barrier?

That is correct. Please see the reference in my post #30 in this thread.

 

[PeroK]

That is correct. Please see the reference in my post #30 in this thread.

I guess what we have (for single-slit diffraction) is:

1) The heuristic explanation using the HUP for position and lateral momentum.

2) A better analysis in terms of the electron being in an infinite square well for a short time as it passes through the slit.

3) A more exact analysis in terms of how precisely (or imprecisely) the barrier provides an infinite square well potential - which may depend on the material of the barrier.

Is that about it?

 

[akhmeteli]

I guess what we have (for single-slit diffraction) is:

1) The heuristic explanation using the HUP for position and lateral momentum.

2) A better analysis in terms of the electron being in an infinite square well for a short time as it passes through the slit.

3) A more exact analysis in terms of how precisely (or imprecisely) the barrier provides an infinite square well potential - which may depend on the material of the barrier.

Is that about it?

I am not sure. There can be dynamic aspects as well, as an electron spends a finite time near the slit. Something about dispersion relation for the barrier material.

 

[hutchphd]

There is an entire science niche devoted to LEED (Low Energy Electron Diffraction) mostly to investigate the surfaces of solids including conductors. There is nothing of fundamental interest about wave-particle fru-fru in any these studies and papers. I think my name may be on one of them from long ago (I used to analyze atomic beams diffracting from surfaces but electrons were just a passing discussion).
This is interesting physics but no big deal. The fact that surfaces are approximately two dimensional is what is really interesting. Don't get me started.

 

[gentzen]

There is an entire science niche devoted to LEED (Low Energy Electron Diffraction) mostly to investigate the surfaces of solids including conductors. There is nothing of fundamental interest about wave-particle fru-fru in any these studies and papers.

What exactly are you trying to say here?

That you don't see wave-particle duality effects in the measurement data? That depends, the wave effects are often quite dominant (and "undesired") for transmission data. Those occur in practice when you have manufactured a wedge to measure the stopping power at different thicknesses in order to validate your models. Wave effects (or rather phase effects) also occur in backscatter data (this time "desired"), when used to measure dislocaltions, defects, or stress via channeling contrast.

Or do you mean that the models themselves don't include the wave effects? For this part the reply would be a bit more complicated. The model certainly contain a variety of direct local quantum effects, including parts where electrons are indistinguishable Fermions, and parts where the electrons are distinguishable for various reasons. On the other hand, the mixing of wave and particle effects as seen in the measurement results described above is still active research. But even there I don't understand which point you are trying to make with your statement.

 

[DrChinese]

What exactly are you trying to say here?

That you don't see wave-particle duality effects in the measurement data? That depends, the wave effects are often quite dominant (and "undesired") for transmission data. Those occur in practice when you have manufactured a wedge to measure the stopping power at different thicknesses in order to validate your models. Wave effects (or rather phase effects) also occur in backscatter data (this time "desired"), when used to measure dislocaltions, defects, or stress via channeling contrast.

...

Answering for hutchphd (who is certainly capable of addressing this without me), and stating the orthodox line:

The presence or absense of "wave-particle duality" is not dependent on the slit's composition. Several posters have implied the opposite, that such composition has not been sufficiently researched (with respect to duality).

On the other hand: it is well known that there are "some" edge effects, but again quantum interference itself is not dependent on those. Could edge effects be exaggerated to the point that interference completely disappears? Even if that were possible, it wouldn't change anything important we learn from the double slit experiment. It would simply be a completely different experiment.

 

[gentzen]

Answering for hutchphd (who is certainly capable of addressing this without me)

Thanks for your answer. It enabled me to learn that "LEED (Low Energy Electron Diffraction)" is actually a specific imaging technique where an energy filter is used to exclude those electrons that lost energy during inelastic scattering events (typically interactions with electrons from the valence or conduction band). So it makes sense that hutchphd called it "an entire science niche". I had misinterpreted LEED as the study of the actual scattering and diffraction of the low energy electrons (inside of the sample), and was a bit surprised why that should be a niche.

After learning this, I first worried whether those measurements I knew using wave effects in backscatter data to measure dislocations, defects, or stress were also done using an energy filter. Had I really always missed that? But after checking some more, my impression is that most of them didn't use an energy filter. So wave effects are still strongly visible even without energy filtering. But the relation between energy filtering and visibility of wave effects is cool: Of course, after an electron lost (a mostly random bigger amount of) energy, its wave length is changed, and therefore its contribution to a diffraction pattern becomes mostly noise.

And that relation between inelastic scattering and lost coherence also makes it clearer how the mixing of wave and particle effects can be included in the models. Only the elastic part of the model must directly take care of the coherence and the structure of the solid. This is great, because it is clear (at least in principle) how to do this, and existing work has focused on that part. And the inelastic part can then be used to estimate the loss of coherence for a given distance.

This is a nice step forward compared to a discussion I had some weeks ago, where someone suggested that the inelastic part would also need to take care of the coherence. My arguments why I thought that the elastic part was more important were only partly convincing. (It was clear that I had no idea what to do with the inelastic part, and that weakened my arguments significantly.)