Has FTL communication really never been tested in this way?

[Andreas Most]

scerir wrote:
> Andreas Most:
>> The "spooky action at a distance" is about correlations
>> of measurements and not about single measurements.
>> People tend to forget this sometimes.[/color]
>
> There is a general agreement that it is incorrect to say
> that the experimenter=92s arbitrary choice of measuring position
> or momentum of the idler photon on the right side of the apparatus
> somehow CAUSED a specific collapse of the signal photon wavepacket
> on the left side.
>
> Only nonlocal, instantaneous, UNCAUSED correlations-at-a-distance
> are predicted by quantum theory. Clearly, the collapse phenomenon
> is nonlocal and NONCAUSAL in nature.[/color]

"Collapse" is actually nothing physical. It is only a notion we have
introduced to explain when we know "what" about the physical state.
E.g., you cannot assign a time stamp to the collapse. By choosing
an appropriate reference frame one could come to the conclusion that
the collapse took place right after the creation of the entangled pair.

> However whether or not fringes, in coincidence detection, show up
> on the left side of the apparatus, DEPENDS ON THE ARBITRARY
> CHOICE by the experimenter of measuring - even in the future
> (delayed choice) - on the right side of the apparatus, position
> or momentum of the idler photon.
>
> Now, what (imo) Cramer is trying to do is to *study* that apparent
> contradiction between 'UNCAUSED/NONCAUSAL' and 'DEPENDS ON THE
> ARBITRARY CHOICE ...'.
>
> For this purpose he needs a clean source of entangled photons,
> to remove the coincidence detection unit, to perform the *same*
> measurement on (say) 1000 idler photons, and see what happens
> on the other side.
>
> Is that correct?[/color]

You cannot remove the coincidence unit. This has nothing to do
about whether you have a "clean source" of entangled photons.
The measurement on the idler so-to-say selects the valid pairs.
There is no way to settle this in advance.
(Someone once came up, that it is a type of book keeping, that
the idler photon determines which photons on the screen are
taken into account. I don't like this type of picture, because
it silently implies that the outcome of a measurement is fixed
since the creation of the pair, which is disproved by the violation
of Bell's inequality. But it is maybe not so bad as a guideline...)

The only way Cramer could be right would be quantum mechanics being
wrong in this respect. From my point of view, however, this would
violate observations in quantum statistics that are well described
by quantum mechanics.

Andreas.

 

[PostReplies]

On Mon, 5 Nov 2007 19:04:01 +0000 (UTC), Andreas Most
<Andreas.Most@nospam.de> wrote:

>Definitely, he will see no interference pattern whatever he does on the
>sender side. Only if he correlates his actions on the sender side with
>the measurements on the receiver side he is able to extract an
>interference pattern.
>[/color]

I thought I understood the classical two slit experiment but I must
not (I'm not a physicist). In the classical experiment if you rig the
experiment such that only photons that went through the slits can
possibly hit the back wall of the apparatus you should see a definite
interference pattern on the back wall. If you then put a detector on
the slits to record their location information the pattern should
disappear. The detector is causing a collapse of the wave function and
therefore each photon no longer has wave characteristics so cannot
interfere with itself. You can even slow the experiment down to firing
one photon a second if you want to convince yourself that different
photons aren't interferring with each other causing the pattern.

I thought that the EPR paradox was a thought experiment in which if
you take that experiment but add a second beam of entangled photons
then when you collapse the wave function for a photon in beam A it
would in theory collapse the wave function for the entangled photon in
beam B. I thought according to QM that happens instantaneously no
matter how far apart the two photons are. That is "spooky action at a
distance" so Einstein considered it a paradox.

I know that in theory FTL communication cannot happen. The "why" part
confuses me. If the photons in the entangled beam actually do take on
particle-like properties then wouldn't that beam no longer be able to
produce an interference pattern?

 

[Ben Rudiak-Gould]

Gerry Quinn wrote:
> I'm not convinced your 'simplified' version of the experiment is
> actually the same experiment at all![/color]

It's not intended to be the same, just to illustrate the importance of
the coincidence counter. It's pretty similar though, aside from being
discrete and classical and omitting the slits.

> The primary function of the coincidence counter, as described in the
> linked URL, is to separate valid pairs of entangled photons.[/color]

That's wrong; it's a misunderstanding by the guy who wrote that page,
and it's presumably the cause of all his other misunderstandings. The
two detectors have very narrow detection cross sections, and
deliberately miss most of the photons that pass them by. If the
coincidence counting had the purpose you suggest, it would make sense to
replace the fixed detector by one with a much wider cross section, since
this would give you a much larger data set. In reality this would
destroy the signal: the better the detector at D2, the less difference
there will be between the two graphs labeled "Measurement at D1". These
graphs do not show measurements at D1. They show a slice through the
parameter space of the nonseparable function f(x1,x2) that relates
detector position to coincidence count. With a wider detector at D2
you'd instead get integral f(x1,x2) dx2, which would look completely
different.

-- Ben

 

[Andreas Most]

PostReplies wrote:
> On Mon, 5 Nov 2007 19:04:01 +0000 (UTC), Andreas Most
> <Andreas.Most@nospam.de> wrote:
>
>> Definitely, he will see no interference pattern whatever he does on the
>> sender side. Only if he correlates his actions on the sender side with
>> the measurements on the receiver side he is able to extract an
>> interference pattern.
>>[/color]
>
> I thought I understood the classical two slit experiment but I must
> not (I'm not a physicist). In the classical experiment if you rig the
> experiment such that only photons that went through the slits can
> possibly hit the back wall of the apparatus you should see a definite
> interference pattern on the back wall. If you then put a detector on
> the slits to record their location information the pattern should
> disappear. The detector is causing a collapse of the wave function and
> therefore each photon no longer has wave characteristics so cannot
> interfere with itself. You can even slow the experiment down to firing
> one photon a second if you want to convince yourself that different
> photons aren't interferring with each other causing the pattern.
>
> I thought that the EPR paradox was a thought experiment in which if
> you take that experiment but add a second beam of entangled photons
> then when you collapse the wave function for a photon in beam A it
> would in theory collapse the wave function for the entangled photon in
> beam B. I thought according to QM that happens instantaneously no
> matter how far apart the two photons are. That is "spooky action at a
> distance" so Einstein considered it a paradox.[/color]

The old Kopenhagen interpretation considered the so called collapse
of the wave function as a physical process. This implies a lot of
paradoxes not to mention that nobody has ever been able to prove that
there is actually a collapse happening.
Nowadays most physicists avoid the term collapse. Quantum mechanics
does not actually describe a collapse but it describes the probabilities
of possible measurement outcomes. This is called the minimal
interpretation.
An entangled pair cannot be considered as two separate objects
like you would do in classical physics. It is a superposition
of states or loosely spoken a superposition of possible measurement
results.

It is actually worth noting about what you said about EPR, that the term
instantaneously immediately raises the question:
In which reference frame? In a certain frame the measurement on particle
B might have been the first. So, which measurement actually collapses
the wavefunction of the other particle?
Going a bit beyond the minimal interpretation you could consider the
wave function being the information content about a quantum system.
This goes back to Heisenberg, I think already in the 1930s.

> I know that in theory FTL communication cannot happen. The "why" part
> confuses me. If the photons in the entangled beam actually do take on
> particle-like properties then wouldn't that beam no longer be able to
> produce an interference pattern?[/color]

In an quantum eraser experiment the signaling photons do not make
up an interference pattern. Only if you apply a certain measurement
on the idler photon and set up the appropriate coincidence between
the two photons (thereby selecting a subset) you will be able to
see an interference pattern.
The point is that you cannot control the measurement outcome of your
idler photon. That is, you need to know the result in order to select
or deselect the corresponding signaling photon for the interference
pattern. By doing so you need to communicate classical information
about the measurement on the idler which is not FTL.

By no means Cramer will be able to circumvent the necessity of a
coincidence unit.

Andreas.

 

[scerir]

Andreas Most:
> You cannot remove the coincidence unit.
> This has nothing to do about whether you have
> a "clean source" of entangled photons.
> The measurement on the idler so-to-say
> selects the valid pairs. There is no way
> to settle this in advance.[/color]

If you want to measure the two-photon interference,
you have to observe, and register, the behaviour
of both entangled photons, the idler photon at its
wing, the signal photon at its wing.

This has much to do with Wheeler's (and Bohr's) saying:
'No elementary phenomenon is a phenomenon until it is
a registered phenomenon.'

Of course, because we are performing a two-photon
interference experiment, we must use an appropriate
source of entangled photons (this source must have
a 'large' size).

Now we can also ask something like this.

Imagine we want to observe what happens at the signal
photon wing only. That is to say: we want to see if there
is some visible interference pattern, made by signal
photons, on the two-slit interferometer screen,
without looking at the other side (where somebody
else is measuring the idler photons).

(We do not use the coincidence detection unit here,
because we are interested in whether there is some
visible interference pattern, made by signal photons,
on the two-slit interferometer screen, without looking
at the other side, where idler photons are measured).

And imagine we cannot see any interference pattern
on that screen. (Apparently this is very strange
because we have a two-slit interferometer and a beam
of photons, and we do not see any interference
pattern).

What is the reason?

The possible reason seems (to me) this one. Signal
photons cannot cause their interference pattern on
the screen because their momentum uncertainty is large.
And their momentum uncertainty is large because the
source of entangled photons we (must) use to perform
a two-photon interference experiments has a 'large'
size (divergence of the beam).

 

[Andreas Most]

scerir wrote:

> And imagine we cannot see any interference pattern
> on that screen. (Apparently this is very strange
> because we have a two-slit interferometer and a beam
> of photons, and we do not see any interference
> pattern).
>
> What is the reason?
>
> The possible reason seems (to me) this one. Signal
> photons cannot cause their interference pattern on
> the screen because their momentum uncertainty is large.
> And their momentum uncertainty is large because the
> source of entangled photons we (must) use to perform
> a two-photon interference experiments has a 'large'
> size (divergence of the beam).[/color]

This has nothing to do with our poor set up of the experiment.
Imagine a quantum eraser experiment with two slits where the
left slit is covered with a horizontal polarization filter and
the right slit is covered with a vertical one. If you shoot linearly
polarized photons (at an angle of 45°) through it there will be no
interference pattern on the screen because left and right linearly
polarized electromagnetic waves do not interfere.
Now, If you chose to measure circular polarization on the idler photon
and select those events where the idler photon has e.g. right circular
polarization then you would see the interference pattern for the
selected signaling photons on the screen.
(By choosing the left circularly polarized idler photons you would
see a shifted interference pattern)

You could argue now that it is possible to use circularly polarized
photons from the very beginning in which case you would see an
interference pattern on the screen without the need of any
coincidence unit. The question is then, what is the use of the idler
photon if not to decide on the type of measurement. And, does the type
of measurement actually change the interference pattern?
The answer by QM and experiment is definitely: No.

Andreas.

 

[scerir]

Andreas wote:

> > Signal photons cannot cause their interference pattern
> > on the screen because their momentum uncertainty is large.
> > And their momentum uncertainty is large because the
> > source of entangled photons we (must) use to perform
> > a two-photon interference experiments has a 'large'
> > size (divergence of the beam).[/color][/color]

> This has nothing to do with our poor set up of the
> experiment.[/color]

Since we were talking about the possibility of 'signaling',
at a distance, using a two-photon interference set-up,
having removed the coincidence detection unit, I was only
pointing out that the usual set-up, i.e. the usual SPDC
source itself, might not allow any single-photon interference
pattern, at the signal wing, for essential reasons (divergence
of the beam). More below.

> Imagine a quantum eraser experiment with two slits where the
> left slit is covered with a horizontal polarization filter and
> the right slit is covered with a vertical one. If you shoot linearly
> polarized photons (at an angle of 45°) through it there will be no
> interference pattern on the screen because left and right linearly
> polarized electromagnetic waves do not interfere.
> Now, If you chose to measure circular polarization on the idler photon
> and select those events where the idler photon has e.g. right circular
> polarization then you would see the interference pattern for the
> selected signaling photons on the screen.
> (By choosing the left circularly polarized idler photons you would
> see a shifted interference pattern).[/color]

Yes, I know these interesting experiments.
http://www.arxiv.org/abs/quant-ph/0106078
http://icpr.snu.ac.kr/resource/wop.pdf/J01/1998/033/R04/J011998033R040383.pd
f

> You could argue now that it is possible to use circularly polarized
> photons from the very beginning in which case you would see an
> interference pattern on the screen without the need of any
> coincidence unit. The question is then, what is the use of the idler
> photon if not to decide on the type of measurement. And, does the type
> of measurement actually change the interference pattern?
> The answer by QM and experiment is definitely: No.[/color]

Again, an important distinction is to be made.
Two-photon interference and one-photon interference
are obviously different phenomena. In the first case
you need a coincidence detection unit of some sort
(two clocks at least). In the second you do not need
any coincidence device.

It seems to me (I may be wrong of course) that
these position/momentum correlated photons
'signaling' machines are based on a sort of ... fusion :-)
of the one-photon and the two-photon interference
phenomena (you perform a specific measurement on the idler
photons and, at a distance, without checking the coincidences,
an interference pattern would appear, or disappear,
at the signal wing).

Now it is known, since long time, there is a weird
'complementarity' principle between the one-photon
and the two-photon interference. In the sense that
the more you can see the first interference, the less
you can see the second interference, and viceversa.

See, i.e., these papers:

M.A.Horne, A.Shimony, A.Zeilinger, 'Two-Particle Interferometry',
Phys.Rev.Lett. 62, 2209 (1989).

M.A.Horne, A.Shimony, A.Zeilinger,
'Two-Particle Interferometry', Nature, 347, 429 (1990).

D.M. Greenberger, M.A. Horne and A. Zeilinger,
'Multiparticle Interferometry and the Superposition Principle',
Physics Today 46 8, (1993).

and these specific experiments ...
http://www.arxiv.org/abs/quant-ph/0112065
http://josab.osa.org/abstract.cfm?id=35389

Since the 'complementarity' principles, in general,
presuppose a 'smooth' transition from the visibility
of a phenomenon to the visibility of the other,
here we can also expect (imo) a smooth transition
from the visibility of a single-photon interference
to the visibility of a two-photon interference,
and viceversa. If there is an intermediate situation
in which both interferences are (badly) visible,
and if - in this intermediate situation - it is possible
to perform 'signaling' (in principle) experiments,
I cannot say.

Summing up. I think WE CAN AGREE that in the two-photon
interference we need a coincidence detection unit,
and in the usual single-photon interference we do not
need such a device. I am pointing out that it is not
just about the use of the coincidence unit, or the use
of specific detectors. There is much more physics beyond,
there are many essential principles involved here.

Regards,
s.

 

[PostReplies]

On Wed, 14 Nov 2007 19:14:10 +0000 (UTC), "scerir" <scerir@libero.it>
wrote:
>
>The possible reason seems (to me) this one. Signal
>photons cannot cause their interference pattern on
>the screen because their momentum uncertainty is large.
>And their momentum uncertainty is large because the
>source of entangled photons we (must) use to perform
>a two-photon interference experiments has a 'large'
>size (divergence of the beam).
>[/color]

Reading all the posts I still don't see why there wouldn't be an
interference pattern. What causes the pattern is each photon
interferring with itself. Since each photon that registers on the back
of the apparatus had to go through a slit in order to even register,
and since we don't know which slit, then according to QM it seemingly
goes through both slits and interferes with itself. Isn't that the
crux of the dual slit experiment? That the source of the photons just
so happened to generate another entangled beam of photons I don't see
how that affects the dual slit experiment.

As far as I can see, polarization doesn't matter in Cramer's
experiment since he's only using the particle vs wave
complimentariness in the experiment--will a particle-like pattern show
up or an interference patttern.

Also, I don't think his experiment has anything to do with quantum
erasers. If you make a which-slit detection and then send the photon
through another 'which path did it take?' choice then you lose the
intereference pattern--not too surprising given what is known about
how QM works. But I don't see what that has to do with Cramer's
experiment. Neither do I see what any kind of delayed choice
experiment has to do with it since the experiment can be performed
without that.

I expect that the sender can at will turn on and off the interference
pattern on his side in the same way as the classical dual slit
experiment. Apparently, that will have no effect on the receiver side
since that would violate Relativity. Perhaps once the results of the
exxperiment are published it will be more understandable to me about
why it failed.

 

[Andreas Most]

PostReplies wrote:

> Reading all the posts I still don't see why there wouldn't be an
> interference pattern. What causes the pattern is each photon
> interferring with itself. Since each photon that registers on the back
> of the apparatus had to go through a slit in order to even register,
> and since we don't know which slit, then according to QM it seemingly
> goes through both slits and interferes with itself.[/color]

That is not quite true. It doesn't matter whether we know or we do not
know which slit the photon has taken. If we could possibly know which
path the photon has taken there will be no interference pattern.
That is a subtle but important difference and it is actually the reason
why the coincidence unit in the quantum eraser experiment is mandatory.

> Isn't that the
> crux of the dual slit experiment? That the source of the photons just
> so happened to generate another entangled beam of photons I don't see
> how that affects the dual slit experiment.
>
> As far as I can see, polarization doesn't matter in Cramer's
> experiment since he's only using the particle vs wave
> complimentariness in the experiment--will a particle-like pattern show
> up or an interference patttern.[/color]

"Polarization" is here a tool to tag the photons path. There are other
ways of doing that but polarization is probably the easiest to accomplish.

> Also, I don't think his experiment has anything to do with quantum
> erasers. If you make a which-slit detection and then send the photon
> through another 'which path did it take?' choice then you lose the
> intereference pattern--not too surprising given what is known about
> how QM works. But I don't see what that has to do with Cramer's
> experiment. Neither do I see what any kind of delayed choice
> experiment has to do with it since the experiment can be performed
> without that.[/color]

IMHO the quantum eraser experiment exhibits all the features to
understand what is going on in a delayed choice experiment and why
Cramers experiment will fail. Actually, what Cramer sets up is a
quantum eraser without a coincidence unit. Since the outcome is
already known it is clear that Cramer will fail.

> I expect that the sender can at will turn on and off the interference
> pattern on his side in the same way as the classical dual slit
> experiment. Apparently, that will have no effect on the receiver side
> since that would violate Relativity. Perhaps once the results of the
> exxperiment are published it will be more understandable to me about
> why it failed.[/color]

Andreas.

 

[Andrew Ray]

In post 23, Ben Rudiak-Gould said

... if you replace the detectors with photographic plates, there will
be no interference pattern on either plate.

Putting aside the question of FTL communication and Cramer's experiment, this appears to be an important part of the argument that coincidence detection is an essential part of Dopfer's experiment (or, at least, a show-stopping consequence of not having it).

Ben, by "either plate", do you mean 'a plate at either position of D1' (i.e. positions f and 2f beyond the Heisenberg lens), or are you saying that furthermore, there will be no fringes recorded at D2 (i.e. on the far side of the slits), regardless of where the film on the lens leg is? If the former, then do you predict there will always be fringes on the plate at D2?