Jump to content

Quick question about delayed choice interferometer exp


swift

Recommended Posts

In the wikipedia entry on wheeler's delayed choice experiment, a description is given of a simple interferometer set up.

 

The article says (if a second beam splitter is not present):

 

"Observing that photons show up in equal numbers at the two detectors, experimenters generally say that each photon has behaved as a particle from the time of its emission to the time of its detection, has traveled by either one path or the other, and further affirm that its wave nature has not been exhibited."

 

What about the experimental results rules out interpreting that the photon has travelled in a wave like way as opposed to particle like way?

Edited by swift
Link to comment
Share on other sites

This page, I assume: https://en.wikipedia.org/wiki/Wheeler%27s_delayed_choice_experiment

 

I think that paragraph is just badly written (*). I think all it really means is that the experiment can be described purely in terms of particles (because there is no interference pattern). In other words, there is no need to consider wave-like effects. But, of course, you can describe it terms of waves.

 

(*) It is badly written using pretentious terminology ("emitted into the entry port", "affirm"? who says that?) and hard to parse sentences.

Edited by Strange
Link to comment
Share on other sites

Yes, that page. Consider this, written further down:

 

"since in the first case the photon is said to "decide" to travel as a particle and in the second case it is said to "decide" to travel as a wave, Wheeler wanted to know whether, experimentally, a time could be determined at which the photon made its decision."

 

The article goes on to say some scientists believed in 'retro-causality' to explain the delayed choice experiment observations. They apparently felt that the method of travel that the photon takes is distinctly and unavoidably different in the two cases, inducing the need for such a peculiar theory as retro-causality.

 

But if it's only an option to consider the photon as having travelled as a particle and not a necessity, then why the need to explain how the photon sometimes travels as a particle, sometimes as a wave?

 

This suggests there is something about the first set up (the one without the second beam splitter) that excludes describing the method of travel as wave like but not in the second set up (the one with the second beam splitter).

 

I can't think what. I mean, in the first set up, assuming they slow the photon beam down so they receive one detection at a time, the fact that only one of the detectors records a result at a time and not both of them suggests a singular particle mode of being; a wave would hit both detectors at the same time. But then in the second set up, again, only one detector would record a result at any given time, suggesting a wave like description does not work here either. So why then the insistence that the photon travels like a particle in the first case but a wave in the second?

Edited by swift
Link to comment
Share on other sites

Well, one important point is that photons are quanta. So when you send a single photon through, it will EITHER go one way OR it will go the other. You wont find a single photon taking both paths.

 

Until you introduce interference, when it appears to go both ways. So one interpretation of this case is that it has started behaving like a wave and takes both paths. Note that this is an interpretation; i.e. a rationalization. It doesn't change what the underlying theory says happens. Other interpretations are available (I prefer to think of it as just a result of non-locality).

 

Edit: just read your post again, and that is pretty much the same as what you said. :)

Edited by Strange
Link to comment
Share on other sites

But even when you introduce the second beam splitter and witness interference, the photon does not behave completely as a wave. If it did, it would also be split at the second beam splitter and end up at both detectors at the same time. As we know, this doesn't happen, only one detector makes a detection.

 

I want to know why scientists considered the first set up as exhibiting particle like behaviour and second as wave like?

Edited by swift
Link to comment
Share on other sites

But even when you introduce the second beam splitter and witness interference, the photon does not behave completely as a wave.

 

Indeed. Photons are always a bit wavelike (frequency, interference, etc) and a bit particle-like (being quanta and localised in their effects). They are never one or the other.

 

I want to know why scientists considered the first set up as exhibiting particle like behaviour and second as wave like?

 

I don't think they did. That is just a kind of a "popular science" / journalese description.

Link to comment
Share on other sites

There is much fuss about such experiments. When an experimenter tells a journalist "test particle versus wave" and then this bizarre concept leads to even more bizarre thoughts, it makes a success paper.

 

But if you abandon the idea that a particle has one single location, these experiments become "almost" natural.

 

So it's mainly a problem of putting the wrong concept on an experiment in order to obtain puzzling misinterpretations.


 

Photons are [...] a bit particle-like (being [...] localised in their effects). [...]

 

This one time I disagree - maybe just a wording detail. I say instead "photons can localize their effects as much as needed and keep one single entity though", and this happens with zero delay whatever the distances.

 

For instance when an electron at a CCD detects the photon, the electron is delocalized before and after, and the photon doesn't need to localize more than the electron. Or an electron at an atom: a valence electron has the size of the molecule before and after absorbing the photon, and the photon localizes to the size of the molecule - not to a point. It's even the way interactions are computed, by an integral over the volume common to both particles.

 

If you describe multiple paths as the photon's shape and size being that of light: two beams after the splitter, again one after merging and possibly with fringes - and capable to shrink to the size of a detector pixel or delocalized electron, you get sensible interpretations of such experiments. What is necessary is that some properties (photon energy...) are kept in integer amounts, and that the reduction of the wavepacket is instantaneous.

 

An other aspect is that photons can choose other properties than their position, for instance their polarization or energy or momentum. Then, the interaction with the detector de-localizes the photon, which cannot be a point to interact. So it's more a matter of choice or adaptation than of position only.

 

Thinking only in terms of absorbed photons can lead to misinterpretations of QM. It is necessary to think at other processes too, especially at an atomic force microscope that pictures a molecule. There, it is clear that the interaction doesn't destroy a particle nor localize it to a point.

Link to comment
Share on other sites

But if you abandon the idea that a particle has one single location, these experiments become "almost" natural.

 

Yes. For me, considering the photon as a wave like 'potentia' (i.e. not real, in a state of nascent potential) until 'observation' presents itself as the most obvious interpretation. But I'm just trying to understand the line of scientists who resisted this idea of a 'potential wave' by asserting the photon is always real but travels as either a particle (as in set up 1, without second beam splitter) or wave (as in set up 2, with second beam splitter).

 

 

You both suggest scientists didn't really have this line, that it was just journalism and lay man misunderstanding? If that's the case, then the wikipedia article is wholly wrong as it presents a completely different story. It essentially suggests that set up 1 demonstrates particle behaviour only and set up 2 wave behaviour only and that scientists (or experimenters here) confirmed this. Quoted from the article:

 

"experimenters generally say that each photon has behaved as a particle from the time of its emission to the time of its detection" (on set up 1)

 

"it was conclusively shown that a photon could begin its life in an experimental configuration that would call for it to demonstrate its particle nature, end up in an experimental configuration that would call for it to demonstrate its wave nature" (when starting with set up 1 and changing it to set up 2 after the photon had already been emitted)

Edited by swift
Link to comment
Share on other sites

If that's the case, then the wikipedia article is wholly wrong as it presents a completely different story.

Or just over-simplified. <shrug> It's Wikipedia.

 

You could post a comment on the Talk page; maybe one of the regular editors will improve the wording.

Link to comment
Share on other sites

[...] If that's the case, then the wikipedia article is wholly wrong as it presents a completely different story. [...]

 

QM is already an old theory - though very healthy for its century - and still badly understood, including often by newspapers, teachers, books. One heavy reason is that in the past, it had many interpretations that have been disproven but are still available in human knowledge and are transmitted. I feel important to learn it in recent books and think at recent expriments, not just the double slit.

 

"Particle = local" is one such misconception, which persists from the time they were imagined as snooker balls.

 

For me, considering the photon as a wave like 'potential' (i.e. not real, in a state of nascent potential) until 'observation' presents itself as the most obvious interpretation. [...].

 

This interpretation - the abstract wave giving a probability to observe the concrete point particle localized by the detector, if I understand you properly - is wrong. It works at the double slit experiment, not at others. One website does presently damage by showing only this.

 

It is wrong, because a photon can choose at detection other properties than its position. Take a femtosecond pulse from a Ti:sapphire laser, its frequency and energy are poorly defined (very few light periods!) but its position and instant are very accurate. If an atom that has a narrow bandwidth absorbs it, then the photon's energy gets well defined, but the instant gets badly defined since this absorption takes time. And if a semiconductor superlattice absorbs it, the photon can become less localized than it was since the absorbing electron spans a big volume.

 

It is wrong, because an atomic force microscope observes all the time the same electron pair, over all locations at the molecular orbital. Searchwords : pentacene afm

http://sbkb.org/update/research/keep-your-eye-on-the-atom

obviously the microscope's tip didn't localize the electron pair.

 

The wave gets very real with such tools. Or as QM says, it's "observable" - and observed. Not as a statistics over many particles: at one particle pair, the same over the time.

Link to comment
Share on other sites

Ok, I've worked it out.

 

The wiki article wasn't wrong. Some experimenters did consider the photon to have travelled in a particle like way in set up 1 and wave like in set up 2.

The reason for this is as follows:

 

Set up 1 necessitates particle like behaviour because only one detector at a time makes a record. If the photon travelled as a wave, it would follow both paths and make an impression on both detectors.

That's what I thought but then my thinking was that set up 2 should also make an impression on both detectors if it is considered as wave like, which does not follow from the results. However, in set up 2, the wave would destructively interfere with itself on one path thus producing only one record at a given detector at a time, which is consistent the results. Hence, the results in set up 2 are coherent with a wave like propagation of light. The interference fringes exclude particle descriptions.

 

Of course, the idea that the photon 'decides' how to propogate based on the experimental set up and the need for retro causality to explain delayed choice results makes this theory less attractive than the 'wave potentia materialises into a particle upon observation' interpretation, in my view. But at least, I can see where these scientists were coming from now...

Edited by swift
Link to comment
Share on other sites

The wiki article wasn't wrong. Some experimenters did consider the photon to have travelled in a particle like way in set up 1 and wave like in set up 2.

 

So do you have a reference to who these experimenters were? If so, that could be added to the Wikipedia page...

Link to comment
Share on other sites

Ah no, I have no idea myself. I've just been trying to make sense of the wiki article itself. It definitely asserts that wheeler rejected the interpretation of 'many experimenters', that interpretation being wave/particle motion decision making and retro activity. I think I found a way of making sense of it, as per my last post, but I'm not sure if my explanation works.

 

Although, to be clear, perhaps I shouldn't have asserted that the article isn't wrong, rather that the story it tells can make sense.

 

There doesn't seem to be a decent resource for understanding the finer points of these experiments on the net, as far as I have found, unfortunately.


And, Enthalpy, wow, ok. That's kind of shaken things up for me. I'm getting to the more recent experiments but I'm trying to work my way through history to make understanding the more recent experiments easier.

 

I don't quite get the explanations you give but you're saying that it would be incorrect to consider the photon as being in a quasi state of potential that materialises (perhaps not in terms of it's location, but forms some properties at least) upon observation?

Edited by swift
Link to comment
Share on other sites

When observed with the atomic force microscope, the electron pair doesn't need to decide where it is. The microscope's tip observes the pair successively over its complete shape.

 

Better: if at some time, one or both electrons did chose a point where they are, subsequently the microscope wouldn't observe the same orbital shape. In addition, the observation would be noisy. And so on.

 

And anyway, the interaction is computed by integrating over both orbitals (observed orbital and microscope's tip atom) because the interacting particles are at all these positions simultaneously.

 

So to a question like "where are the electrons" my answer is "they are the orbitals". The probability density (when the wave function is expressed versus the position) gives chances to find the electrons in a smaller volume unit if the setup is a means to localize the target electron more accurately, say by using a high-energy electron, proton... If the sensing particle is big, the observed one keeps its full volume.

 

The wave function can also be expressed versus other variables, say as a linear combinations of eigenstates (in an atom, orbitals), a function of the momentum, and so on - and depending on the setup, the observed particle may reduce its choice of energies, momenta, polarization, not just the position. When the spread of momenta is reduced at an interaction, the spread of positions increases.

 

So a wavefunction doesn't serve just to determine a probability to localize a particle better. With the atomic force microscope, it's pretty much the wavefunction itself that we see. The founders of QM didn't have such nice tools, but we do, and this must help us get a clearer picture.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.