Photon Self-Interference: Just a Misunderstanding?

[pmb_phy]

Has anyone ever heard of the double slit experiment being described as a photon interacting with itself? I've seen people mention this "explanation" but it doesn't sound very QM to me.

Does Feynman say this verbatim somewhere?

Thanks

Pmb

 

[Eye_in_the_Sky]

To say that the photon "interferes" with itself sounds perfectly correct to me. You have one photon and two slits. This one photon then has two amplitudes (one for each slit). These two amplitudes for this one photon then interfere.

But to say that the photon "interacts" with itself ... well, I think that that word choice is inappropriate.

 

[ZapperZ]

Has anyone ever heard of the double slit experiment being described as a photon interacting with itself? I've seen people mention this "explanation" but it doesn't sound very QM to me.

Does Feynman say this verbatim somewhere?

Thanks

Pmb

Read

L. Mandel, Rev. Mod. Phys., v.71, p.274 (1999).

There is a clear difference between single-photon interference (which is the explanation for the 2-slit observation) and 2-photon interference. Again, most people are confusing the "focus" of the issue here. It really isn't the photon that is the major cause here. It is the superposition if it's PATH! The photon is merely the carrier of this effect! It means that you can replace the photon with ANYTHING that exhibit the identical deBroglie wavelength and the same superposition. That's why you can get this with photons, electrons, protons, neutrons, buckyballs, etc. etc.

Zz.

 

[pmb_phy]

Read

L. Mandel, Rev. Mod. Phys., v.71, p.274 (1999).

There is a clear difference between single-photon interference (which is the explanation for the 2-slit observation) and 2-photon interference. Again, most people are confusing the "focus" of the issue here. It really isn't the photon that is the major cause here. It is the superposition if it's PATH! The photon is merely the carrier of this effect! It means that you can replace the photon with ANYTHING that exhibit the identical deBroglie wavelength and the same superposition. That's why you can get this with photons, electrons, protons, neutrons, buckyballs, etc. etc.

Zz.

To me the double slit simply determines the probability amplitude of a photon stiking the screen at a particular location. The "Photon interacts with itself" description simply gives me the willies!

Thanks folks

Pete

 

[ZapperZ]

To me the double slit simply determines the probability amplitude of a photon stiking the screen at a particular location. The "Photon interacts with itself" description simply gives me the willies!

Not if you get rid of that classical confinment of "particles".

Most people's inability (or refusal) to comprehend quantum phenomena is primarily due to the ingrained view of what the world should be based on what is familiar, which is the classical world. It is utterly arrogant for us to force the microscopic world to fit to this description. We are forcing onto it concepts that are simply vaguely applicable, and then we complain when the results make no "sense". What gives?

Zz.

 

[Nereid]

Not if you get rid of that classical confinment of "particles".

Most people's inability (or refusal) to comprehend quantum phenomena is primarily due to the ingrained view of what the world should be based on what is familiar, which is the classical world. It is utterly arrogant for us to force the microscopic world to fit to this description. We are forcing onto it concepts that are simply vaguely applicable, and then we complain when the results make no "sense". What gives?

Zz.

Oh you'd really enjoy some of the discussions we've been having in the Philosophy section Zapper!

 

[ZapperZ]

Oh you'd really enjoy some of the discussions we've been having in the Philosophy section Zapper!

I'll be right over as soon as I have a lobotomy! :)

Zz.

 

[pmb_phy]

Not if you get rid of that classical confinment of "particles".

Most people's inability (or refusal) to comprehend quantum phenomena is primarily due to the ingrained view of what the world should be based on what is familiar, which is the classical world. It is utterly arrogant for us to force the microscopic world to fit to this description. We are forcing onto it concepts that are simply vaguely applicable, and then we complain when the results make no "sense". What gives?

Zz.

Who says I am unable to or refuse to comprehend quantum phenomena? I understand it/comprehend if just fine thanks. I took two full courses of quantum in grad school. That doesn't mean I like a particular description/interpretation. Especially in this case. I don't see this as a photon interfering with itself. I see the slit as a quantum well and the well determines the probability of measuring the position a photon lands on the screen behind the double slit. There is no need for me to invoke images of a photon interfering with itself and that's the way I see it. That does not mean I don't comprehend QM.

Pmb

 

[ZapperZ]

Who says I am unable to or refuse to comprehend quantum phenomena? I understand it/comprehend if just fine thanks. I took two full courses of quantum in grad school. That doesn't mean I like a particular description/interpretation. Especially in this case. I don't see this as a photon interfering with itself. I see the slit as a quantum well and the well determines the probability of measuring the position a photon lands on the screen behind the double slit. There is no need for me to invoke images of a photon interfering with itself and that's the way I see it. That does not mean I don't comprehend QM.

Pmb

Then don't use a slit. Use a beam splitter, or a SQUID as done for Cooper pairs interference. The results are identical. If you think your slit is a quantum well, then you need to explain how your quantum well translates to all these setups that give the same phenomena.

After you have done that, then you should then address how an electron in an H2 molecule "interferes" with itself in two different but overlapping potential wells to create the bonding-antibonding states.

Zz.

 

[pmb_phy]

Then don't use a slit. Use a beam splitter, or a SQUID as done for Cooper pairs interference. The results are identical.

Yes. I'm well aware of that. Thank you.

If you think your slit is a quantum well, then you need to explain how your quantum well translates to all these setups that give the same phenomena.

The same way that the Hamiltonian of a given system determines the possible quantum states. Different systems may have the same Hamiltonian. There is a Hamiltonian associated with the slit. The possible wave functions are then determined by that Hamiltonian and the probability that a particle will be measured at a particular location will depend on the the wave function.

After you have done that, then you should then address how an electron in an H2 molecule "interferes" with itself in two different but overlapping potential wells to create the bonding-antibonding states.

Hamiltonian -> Probability distribution. The rest is all flowery tails. However if you can provide a measurement which indicates that what I've said, i.e. Hamiltonian - > wave function - > Probability density etc is wrong then I'm all ears.

What observation are you saying is made where a photon/particle interfers with itself and what are you saying that physically means?

Pete

 

[ZapperZ]

The same way that the Hamiltonian of a given system determines the possible quantum states. Different systems may have the same Hamiltonian. There is a Hamiltonian associated with the slit. The possible wave functions are then determined by that Hamiltonian and the probability that a particle will be measured at a particular location will depend on the the wave function.

Then show me the Hamiltonian for the slit and especially wher you incorporate the "potential well". You might want to submit this for publication, since Marcello has already derived this using photon interference.[1] And then justify how the similar Hamiltonian can be used in SQUIDs interference pattern. I'd like to see where the "potential well" is for that one.

What observation are you saying is made where a photon/particle interfers with itself and what are you saying that physically means?

When a photon passes through only one slit, and we know it passes through only that slit, you have no interference pattern. When it is allowed to pass through both slits, we have interference pattern. Same thing with the SQUID. If I detect the cooper pair going through one arm of the SQUID loop, I see no Fraunhoffer current pattern. But if I leave it alone, I get that pattern. More same thing, this time with H2 molecule. If I confine the electron only to one of the H atom, I detect no change in the ground state. But if I let them overlap, then the electron will occupy both atoms simultaneously to produce the "interference" effect that results in the bonding-antibonding state.

And I have no idea what you want by "physically means". "Meaning" of things tend to be a matter of tastes. We might as well argue about favorite colors.

Zz.

[1] T.V. Marcella Eur. J. Phys. v.23 p.615 (2002).

 

[pmb_phy]

Then show me the Hamiltonian for the slit and especially where you incorporate the "potential well".

Since the idea is the same with an electron I'll refer to electrons instead. The Hamiltonian for such a potential as a slit is not easy to express. Its an infinite potential at the wall of the slit and zero potential where the slit is open.

I'm not certain if you understand the term "potential well" in the way I've intended it here. For an example of a potential well see diagram at
http://www.geocities.com/physics_world/qm/harmonic_oscillator.htm

A potential well for a slit would have infinite walls and zero potential inside the slit.

When it is allowed to pass through both slits, we have interference pattern.

I agree.

 

[ZapperZ]

Since the idea is the same with an electron I'll refer to electrons instead. The Hamiltonian for such a potential as a slit is not easy to express. Its an infinite potential at the wall of the slit and zero potential where the slit is open.

I'm not certain if you understand the term "potential well" in the way I've intended it here. For an example of a potential well see diagram at
http://www.geocities.com/physics_world/qm/harmonic_oscillator.htm

A potential well for a slit would have infinite walls and zero potential inside the slit.

Then your potential well is the physical confinment within the slit. This then is inconsistent with, for example, the SQUID experiment. Each branch of the SQUID loop can be made as thick as one care to. It doesn't affect the nature of the interference pattern, so the pattern has nothing to do with physical confinemnt. In fact, we KNOW how quantum well/quantum wire states behave since we can already fabricate those nowadays. These things do not show the interference pattern of.

Zz.

 

[Pieter Kuiper]

Has anyone ever heard of the double slit experiment being described as a photon interacting with itself? I've seen people mention this "explanation" but it doesn't sound very QM to me.

Does Feynman say this verbatim somewhere?

The quote is due to Paul Dirac (The Principles of Quantum Mechanics, 1930):

"Each photon then interferes only with itself. Interference between two different photons never occurs."

But different radio transmitters and lasers do interfere...

 

[danim]

The quote is due to Paul Dirac (The Principles of Quantum Mechanics, 1930):
"Each photon then interferes only with itself. Interference between two different photons never occurs."

Maybe I'm wrong, but I think you can deduce from this statement the main idea behind the the bose-einstein and fermi-dirac statistics. That's shortly why you don't have the exclusion principle for bosons.

But different radio transmitters and lasers do interfere...

Please don't mix the photon with the superposition principle acting in the electromagntic waves. I think to use this picture you have to use the quantum field theory too.

I think the overall discussion comes from trying to picture the quantum mechanics properties in a language herited from the classical world. It's easier to use math. (maybe I'm a fool, but really can't picture the wave function unless as an integral of the action of the infinite possible paths, which I really can't visualize) Try to use the "is like a" expression and apologize for not giving the right picture and you don't need this kind of fighting. Otherwise we can have in mind the others difficulties to picture their ideas.
tank you

 

[Pieter Kuiper]

Maybe I'm wrong, but I think you can deduce from this statement the main idea behind the the bose-einstein and fermi-dirac statistics. That's shortly why you don't have the exclusion principle for bosons.

??
The double slit interference pattern is the same for photons and electrons.

 

[pmb_phy]

??
The double slit interference pattern is the same for photons and electrons.

The physics is the same, yes. I.e. there is an interference pattern in both. See Feynman lectures V-III if you have them.

Pete

 

[reilly]

What exactly does it mean for a photon to interfere with itself? As far as I can figure, it can happen only as a result of a nonlinear light-light interaction, and that's easy enough to rule out as a possible explanation. However, the self interference of waves is quite another matter, and it provides the basic classical explanation for diffraction and interference -- the path length inerference discussed by Zapper, above.

Is there not a dilemma? The quantum and classical fields in diffraction and interference phenomena are, in practice identical, -- same momentum, angular momentum, same energy and momentum density. The mathematial struture of the interference is identical for both quantum and classical fields. Classically the energy density at the detector shows the standard interference effect, for example, for two slits. The actual photon density is energy density/(hbar omega), and for free fields this classical = quantum is rigorous.

The best thing to say is; the photon is governed by a probabiliy density , which is governed by interference phenomena. Photons will interfere with themselves only in extreme high energy circumstances in which they undergo a nontrivial self interaction, and very likely photons at such extremes no longer exist as such.

Under normal circumstances and with normal the meaning of the following words, photons do not, in fact cannot interfere with themselves. The description of photons, however, does show the familiar classical interference effects.

There's always a danger of confusing the THING and its NAME. Could not be more true here: the description involves interference. But, there's no physical way for the photon to interfere with itself. Fields interfere, photons do not. Same is true for electrons.

Regards,
Reilly Atkinson

 

[member 11137]

Ok Reily, but electron has a spin and spin can interact with the field or the way around. Isn't it a possible way to begin a description of a self-interaction of the electron with itself via the field existing around it? Thank you for explanations concerning this subject; I am actually trying to understand this Papapetrou forces which seem to be an illustration of my question.

 

[pmb_phy]

Thanks Reilly. That's exactly what I had in mind.

Pete

 

[marlon]

What exactly does it mean for a photon to interfere with itself?

Basically we don't know. What happens between the foton-source and the two slits is unkown and it even does not matter. Indeed it is known that one photon-wavefunction will have to interfere with itself in order to explain the correct observed interference-pattern at the screen.

So basically, reilly, i completely agree with you on this one. Yet i would not call this "self-interference" non linear in nature. Trust me, i can see your point when you stated this in your above post, but i just want to point out that this self-interference is NOT the same as the photons self-energy. This self-energy only arises (well, QED...) when interactions between photons or some surrounding lattice occur. Or when we adopt the virtual vacuum and its fluctuations in QFT. Yet these interactions are certainly NOT non-linear since photons do NOT interact with each other, unlike gluons for example...


regards
marlon

 

[bruce2g]

I don't think you can say that "the photon interferes with itself" because the photon does not actually exist until until it strikes the screen.

Bohr's principle of complementarity states that the photon is not a particle while wave-like behavior is occurring. So it's not a particle when it goes through the slits, it's a wave. The wave is just the probability amplitude that a particle will be detected at a certain place at a certain time, so it's not possible to even know for certain whether or not the photon exists at a certain time.

Since you do not even know whether the photon exists until it hits the screen, then I believe it is fair to say that the photon does not exist until it hits the screen.

At least, that's how I see it.

Bruce

 

[member 11137]

..."I don't think you can say that "the photon interferes with itself" because the photon does not actually exist until until it strikes the screen"...(bruce2g)

I do not reject basic principles of the QM, certainly not; believe me. But in despite of this I have a problem with this argumentation which is remembering me a discussion with my father during a beautiful summer night as I was a young boy and telling him: you know pa, there are so many stars in the sky that we cannot see... and he answered me: as long as we don't see these objects they do not exist for the science... ok ok ok because we need rationalism to do science; as said I do not reject this method. But in despite of our momentary incapacity to measure or to observe something, it doesnot mean that this thing definitively do not exist, or...? This basic principle of the QM is "just" explaining us that we need a scientific method and a scientific theory of the measurement in general. And it's good.

It doesn't avoid that the wave associated with an invisible (potential future visible) photon could at each moment interfere with waves due to the rest of the world (the geometric field f.e.) and the way around; am I wrong in my representation? Thanks for help to correct.

 

[marlon]

I don't think you can say that "the photon interferes with itself" because the photon does not actually exist until until it strikes the screen.

I get your point, yet what you are saying here is untrue. The photon does exist, only we know nothing about it prior to any measurement of any kind. Basically, in this state, the photon wavefunction is the superposition of all possible photon-states that can be a&cquired after any measurement...

Bohr's principle of complementarity states that the photon is not a particle while wave-like behavior is occurring. So it's not a particle when it goes through the slits, it's a wave.

This is partially true. The particle-wave-stuff is a DUALITY. This means that you can replace any vision with the other at any given time. So you can replace the word particle by the word wave at any time. You cannot, however, use them both simultaneously...

The wave is just the probability amplitude that a particle will be detected at a certain place at a certain time, so it's not possible to even know for certain whether or not the photon exists at a certain time.

The photon exist at any time, we only do NOT know what state it is in...

Since you do not even know whether the photon exists until it hits the screen, then I believe it is fair to say that the photon does not exist until it hits the screen.

At least, that's how I see it.

Bruce

Given my reasoning, i state that this is a wrong conclusion.

regards
marlon

 

[danim]

??
The double slit interference pattern is the same for photons and electrons.

Certainly you can see the interference pattern once the superposition principles holds for the wave function of these particles. So you have the crossed term when making psi^2.
It's not what we are discussing. We are discussing if two photons can have the same quantum state. Two or more particles with half spin, three half, etc. cannot stay in the same quantum state. It's not the case for integer spin particles. The former are called fermions because they obbey the fermi-dirac statisctics ressembling for the exclusion principle. The latter do not obbey this principle. I think the first askin was related to this.

 

[prj45]

Re: Photons interfering with themselves.

Considering they move through space at light speed, this means they don't move through time right? So, surely, they must interfere with themselves, as from our point of view, they'll be everywhere at once (until we observe them and bring them into our timeframe right).

Sorry that sounds like B Movie stuff, and yes, I'm just making it up as I go along, but just trying to stretch my brain!

 

[J C Oelund]

Reality versus Understanding

Ultimately I suggest that what Physicists are doing is building models (generally mathematical) that explain and encapsulate behaviors of reality. These models have consequences, often making predictions of behavior. When we test our models against reality (experiment), it will either validate our model or not.

Understanding? This is a sense of intuition that our models are correct, but such a sense can only be based on experience, and observation. Well, observaton of things moving at the speed of light, or on unimaginably small scales is not part of our common experience.

I suggest that in the end, all we have is building and testing such models. Questions that lie outside of this realm really have no meaning, except as they form the basis for construction of such models (basic assumptions and postulates). I suggest they really have no meaning unless they produce something testable in the real world.

But yes, asking such questions is a vital way to test and generate or extend such models, so not wise to discourage such imaginings. Perhaps there is another way to look at the world, that will make the underpinnings clear - and bring yet another, "but of course, how could it be any other way?".

 

[Mentz114]

Dirac's statement is not to be taken lightly. The point he was making is that we must distinguish 2 possibilities when a lot of identical things go through the slits -
1. The wave functions of different items interfere.
or
2. They don't, but each things split wavefunctions interfere.
This is an important distinction.