How to observe a photon without disturbing it?

[swift]

In the double slit experiment, it is said scientists wanted to observe which slit a singularly fired photon went through once they discovered even this produces an interference pattern.

 

On an atomic level of description, how exactly do they observe a singular photon? Ie what is used to extract information about the photons position at the slit without stopping the photon altogether?

And what physical effect does this have on the photon ie does it affect its momentum/energy?

 

To screen used to see the interference pattern actually absorbs the photon and essentially destroys it, so that can't be used to watch the photon pass through a given slit.

Same question for an electron.

[Strange]

The fact that a photon is destroyed if it is detected meant that this stayed as a thought experiment for a long time. Eventually someone worked out that you could use entangled photons to detect which slit the photon went through. In other words, you generate a pair of entangled photons (so there is a known relation between their polarization).

 

By putting a different polarizing filter on each slit and measuring the polarization of the other entangled photon (the one that doesn't go through the slit) you can determine which slit the photon went through. This destroys the interference pattern.

 

But then, if you erase the "which slit" information by putting a diagonal polariser in the path of the detector photon (i.e. the one that doesn't go through the slits) then you cannot determine which slit the other photon went through and the pattern re-appears.

 

Even more surprisingly, you can erase the information after the photon has hit the screen and it still allows the pattern to appear.

 

http://en.wikipedia.org/wiki/Quantum_eraser_experiment

http://en.wikipedia.org/wiki/Delayed_choice_quantum_eraser

The same thing works with electrons (using the spin of the electron).

[michel123456]

In the double slit experiment, it is said scientists wanted to observe which slit a singularly fired photon went through once they discovered even this produces an interference pattern.

 

On an atomic level of description, how exactly do they observe a singular photon? Ie what is used to extract information about the photons position at the slit without stopping the photon altogether?

And what physical effect does this have on the photon ie does it affect its momentum/energy?

 

To screen used to see the interference pattern actually absorbs the photon and essentially destroys it, so that can't be used to watch the photon pass through a given slit.

Same question for an electron.

 

This is the field of research of physics Nobel Prize Serge Haroche. it goes far over my head. Search this below

Quantum non-destructive observation of trapped photons, detection of field quantum jumps, reconstruction of non-classical field states, direct monitoring of decoherence and quantum feedback demonstrations (2006-2011)

Edited November 9, 2014 by michel123456

[swift]

"But then, if you erase the "which slit" information by putting a diagonal polariser in the path of the detector photon (i.e. the one that doesn't go through the slits) then you cannot determine which slit the other photon went through and the pattern re-appears."

 

I assume a diagonal polarising filter will stop any photon regardless of its polarisation. Where do they place this filter so as to ensure it is in the path of just the detector photon? If they place it over one slit, it is known the other photon has gone through the other slit. If they place it nearer the photon beam, how do they ensure capturing just one of the entangled photons?

Plus using entangled photons sounds like the kind of advancement that occurred long after they determined observation of the photon/electron causes wave collapse. What was their method of observation that caused collapse prior to this?

 

Also, how do two photons/electrons become entangled?

 

Thanks in advance.

[Strange]

I assume a diagonal polarising filter will stop any photon regardless of its polarisation.

 

It doesn't stop the photon, it just removes the polarisation information so that you can no longer determine which slit each photon went through.

 

Where do they place this filter so as to ensure it is in the path of just the detector photon?

 

There are diagrams on the Wikipedia page linked. It is POL1 in this diagram:

It is placed before Dp which measures the polarization of the entangled photon and so allows you to tell which slit the photon went through. The other detector, Ds, detects the interference pattern.

 

Plus using entangled photons sounds like the kind of advancement that occurred long after they determined observation of the photon/electron causes wave collapse. What was their method of observation that caused collapse prior to this?

 

Before someone worked out how to detect it using entangled photons, it was purely a thought experiment.

 

Also, how do two photons/electrons become entangled?

 

They are created by the cool-sounding "spontaneous parametric down conversion" (also explained on the Wikipedia page).

 

Thanks in advance

Edited November 15, 2014 by Strange

[swift]

Appreciate your teachings. I'm still unclear though.

 

What does it mean to 'remove polarisation information'? I thought a filter merely blocks components of the photon that are not aligned with it. Ie a vertical filter will absorb horizontal components.

 

As I currently understand it, if two photons are entangled they will already be polarised and have opposite polarisations. Passing it through a diagonal filter will reduce the amplitude of both a horizontally polarised photon and a vertically polarised one but the polarisation of the photon is still there- how would information regarding it be 'removed'?

 

Secondly, what thought exp gave them the revolutionary conclusion that observation causes wave collapse? It was such an unexpected result that it seems odd they'd reach that conclusion unless forced to by experimental observation.

[Strange]

As I currently understand it, if two photons are entangled they will already be polarised and have opposite polarisations. Passing it through a diagonal filter will reduce the amplitude of both a horizontally polarised photon and a vertically polarised one but the polarisation of the photon is still there- how would information regarding it be 'removed'?

 

It removes the unique circular polarization associated with each slit. But I'm afraid I don't understand any more details than is in the Wikipedia article.

 

 

Secondly, what thought exp gave them the revolutionary conclusion that observation causes wave collapse? It was such an unexpected result that it seems odd they'd reach that conclusion unless forced to by experimental observation.

 

It is the other way round: theory predicted what would happen; the thought experiment was a way of demonstrating it (in principle). It was predicted that if there was a way of detecting which slit a photon went through (without disturbing the it) then this would destroy the interference pattern. Then someone figured out how to do it using entanglement.

[swansont]

 

As I currently understand it, if two photons are entangled they will already be polarised and have opposite polarisations.

 

That's a specific application, but not a general requirement; as the SPDC wikipedia page says, there is also type I, which has identical polarizations. What's needed is that they have an indeterminate property and can't otherwise be identified. You might notice in the diagram of SPDC (4th one down) that the entanglement only takes place where the two emission cones overlap. For any other photons, you can tell which cone it was in, and that also tells you the polarization.

[swift]

If the detector, dp was removed, would there still be an interference pattern?

[Strange]

If the detector, dp was removed, would there still be an interference pattern?

 

If you remove the detector but leave the polariser? Interesting question. I don't know the answer to that.

[swift]

To collapse the wave function there is no polariser at pol1, just a detector, right? The entangled photon hits the detector, having not passed through a filter, it's polarisation detected and there is no interference pattern seen at ds. Have I got that right? If yes, then I'm asking what would happen if in the same situation you remove the detector dp, so there is no filter or detector. I'm pretty sure the answer is that there would be an interference pattern right?

[Strange]

In that case, yes. You are back to the standard double slit experiment, so you will get the interference pattern.

[swift]

Great. Ok, so here's my real question/ thought experiment. What if dp is placed a great distance (many light seconds) away such that an interference pattern is allowed to be displayed at ds before the first entangled photon is allowed to hit dp. Dp could be inserted or left out at random by a scientist next to it in fact. The photon going through the slits would have to display an interference effect but then after, it's entangled twin could be measured. What would happen here?

[Strange]

The time until the photon is detected at Dp is irrelevant. What you are describing is the "delayed choice quantum eraser" experiment (second link). Quantum effects are non-local in time as well as space.

[swift]

non- local as in the decision made by the scientist at dp affects the outcome at ds? so effect (@ds) occurs before the cause (@dp)?

That creates paradoxes. For instance, if electrons were used, which travel slower than light, once the result at ds is known, the scientist at ds could send a signal to the one at dp before the entangled electron has reached dp.

Think I'm very confused.

[Strange]

non- local as in the decision made by the scientist at dp affects the outcome at ds? so effect (@ds) occurs before the cause (@dp)?

That creates paradoxes. For instance, if electrons were used, which travel slower than light, once the result at ds is known, the scientist at ds could send a signal to the one at dp before the entangled electron has reached dp.

Think I'm very confused.

 

But the only result known at Ds is that an electron has been detected. It isn't even known, until a large number have been detected, whether an interference pattern has been formed or not. So I'm not sure what information could be sent from Ds to Dp that would create a paradox.

[swift]

Yes, well, theoretically, you could have dp sufficiently far that many electrons have reached ds before any of their entangled twins have reached dp. Hence, you can observe an interference pattern or not. At this point scientist at ds sends a light speed signal to scientist at dp, letting him know whether he observed an interference pattern. Let's say he did, so scientist at dp receives this message and deliberately measures the entangled twins to determine which slit they went through via their polarisation. Or if ds scientist doesn't see an interference pattern, he mesaages this and scientist at dp deliberately removes the detector dp, meaning the electron wave function collapsed despite no observation.

Edited November 17, 2014 by swift

[Strange]

But you can't tell anything about interference/non-interference from a single particle. It is a statistical thing. Where any given electron lands doesn't tell you anything; it could have landed in the "dark" area with low probability or the "light" area with high probability. It is only after seeing enough electrons that you see that most fall in the light area, creating the pattern. (Or not, as the case may be.)

[swift]

See edited post above..

[Strange]

So, it is actually a little subtler than that. Here is a good description that explains why it doesn't work like that.

http://van.physics.illinois.edu/qa/listing.php?id=25872

 

It all comes down to which photons/electrons you count as contributing to the interference pattern.

[swift]

Ah, so what I gather from that is its not the fact that you actually measure and find out the polarisation of the entangled pair that collapses the wave function, it's the fact that you could in principle. So due to the mere fact that you have placed a polarisation filter on each slit, the wave collapses and you will never see an interference pattern, whether you measure the entangled twin or not. So delaying measurement of the entangled twin is ineffectual. True.

Edited November 19, 2014 by swift

[Strange]

So due to the mere fact that you have placed a polarisation filter on each slit, the wave collapses and you will never see an interference pattern, whether you measure the entangled twin or not. So delaying measurement of the entangled twin is ineffectual. True.

 

It is not quite that simple. After all, placing a (delayed) diagonal filter in the path of the other photon means that the pattern appears, despite the presence of the circular polarisers.