Is Wave-Particle Duality Really Real? An Analysis of the Double Slit Experiment

[Frame Dragger]

I agree with what you are saying. dBB gets by because the context is non-local. Another interesting set of interpretations is the Time Symmetric group (including Relational BlockWorld RBW), in which a future context is allowed to influence the present. These have the "benefit" of being local, contextual (non-realistic) and time symmetric. Of course you swap one assumption for another, so whether the result is net better is a matter of preference.

This is true, and possibly experimentally verifiiable. If phonons can be detected emerging from a sonic event horizon it would be a step towards showing that a process like Hawking Radiation might exist... and that hinges on virtual particle annihilation (or lack thereof)... which is also a kind of single particle along a worldline view, but not in one temporal "direction". I suppose we'll have to wait and see.

 

[eaglelake]

In the book he makes an example of the double slit experiment we all did in high school and says that if we fire one photon at the slitted screen, we'll get an interference pattern on the rear screen.

If you fire one photon through the slits you will get one dot on the second screen. A single photon is always detected as a single particle (a dot), never as a wave. In fact, the single dot doesn’t reveal any wave properties. It is only when you detect many photons does the distribution of scattered photons begin to look like an interference pattern. There are sites on the web that show how the interference pattern is built up one photon at a time.

I am bothered that an author would make such a statement, but apparently it is all too common. No experiment has ever shown a particle behaving like a wave. The point is this – a quantum particle is not a wave.

 

[Frame Dragger]

If you fire one photon through the slits you will get one dot on the second screen. A single photon is always detected as a single particle (a dot), never as a wave. In fact, the single dot doesn’t reveal any wave properties. It is only when you detect many photons does the distribution of scattered photons begin to look like an interference pattern. There are sites on the web that show how the interference pattern is built up one photon at a time.

I am bothered that an author would make such a statement, but apparently it is all too common. No experiment has ever shown a particle behaving like a wave. The point is this – a quantum particle is not a wave.

Annnnd... you're wrong. In fact, even when passing C60 or Rubidium atoms (one at a time) there is interference consistant with passage through both apertres of the experiment. There are sites on the web that show how 8th dimension lizards run the country... that doesn't make them accurate. Interference patterns of the type you describe are not consistant with a purely particle-theory of light. Light's wave-like properties are as well established as its particle-like properties.

As for experiment with photons... you'rre wrong again? I don't know what else to say... you can in fact set up simple controls for these experiements which eliminate interence by more than specifically polarized photons passng throuhg the apertures.

EDIT: Addition: "A quantum particle is not a wave." True. Photons are quanta which have wave and particle -like properties. They are neither waves, NOR particles and feel free to call them "quanta" not "quantum particles". That latter is... meaningless and semi-redundant.

 

[eaglelake]

Annnnd... you're wrong. In fact, even when passing C60 or Rubidium atoms (one at a time) there is interference consistant with passage through both apertres of the experiment. There are sites on the web that show how 8th dimension lizards run the country... that doesn't make them accurate. Interference patterns of the type you describe are not consistant with a purely particle-theory of light. Light's wave-like properties are as well established as its particle-like properties.

As for experiment with photons... you'rre wrong again? I don't know what else to say... you can in fact set up simple controls for these experiements which eliminate interence by more than specifically polarized photons passng throuhg the apertures.

EDIT: Addition: "A quantum particle is not a wave." True. Photons are quanta which have wave and particle -like properties. They are neither waves, NOR particles and feel free to call them "quanta" not "quantum particles". That latter is... meaningless and semi-redundant.

No need to be a wiseguy! If you do not understand something, just ask for a clarification.

I responded to the statement, “if we fire one photon at the slitted screen, we'll get an interference pattern on the rear screen.”, which is not true. A single photon is always detected as a single dot on the detection screen. A single photon is never smeared over the screen in an interference pattern, as a wave would be. If you want to see wave effects then you must repeat the same experiment many times.

The original statement by mintparasol referred to a single photon, while you are talking about, “passing C60 or Rubidium atoms (one at a time)”, which is about MANY atoms. “One at a time” means more than one! You seem to imply that the interference pattern is built up “one (atom) at a time”. I agree! That is exactly what I said. So, where is the disagreement?

The basic question is, “ When does a photon, or any quantum particle, look like a particle and when does it look like a wave”? We either observe one or the other, but never both at the same time. (Bohr’s complimentarity principle)

The answer is – if you detect a single particle it looks like a particle. It looks like a wave only after you have detected many of them. [1] (and, then, only in special circumstances.)

Best wishes.

[1] A. Tonomura, et al, Amer. J. Phys., 57, 117-120 (1989)

 

[Frame Dragger]

No need to be a wiseguy! If you do not understand something, just ask for a clarification.

I responded to the statement, “if we fire one photon at the slitted screen, we'll get an interference pattern on the rear screen.”, which is not true. A single photon is always detected as a single dot on the detection screen. A single photon is never smeared over the screen in an interference pattern, as a wave would be. If you want to see wave effects then you must repeat the same experiment many times.

The original statement by mintparasol referred to a single photon, while you are talking about, “passing C60 or Rubidium atoms (one at a time)”, which is about MANY atoms. “One at a time” means more than one! You seem to imply that the interference pattern is built up “one (atom) at a time”. I agree! That is exactly what I said. So, where is the disagreement?

The basic question is, “ When does a photon, or any quantum particle, look like a particle and when does it look like a wave”? We either observe one or the other, but never both at the same time. (Bohr’s complimentarity principle)

The answer is – if you detect a single particle it looks like a particle. It looks like a wave only after you have detected many of them. [1] (and, then, only in special circumstances.)

Best wishes.

[1] A. Tonomura, et al, Amer. J. Phys., 57, 117-120 (1989)

No... you missed my point entirely. The experiment involving Rubidium showed a single atom having a unique wavefront just as a single photon does. The fact that it takes multiple passes (as you say, a buildup) to make the pattern visible is a limitation of our detection methods. If one could image a photon more exactly there would be a wavefront causing an interference pattern, visible or not. The dual nature of the quanta seems pretty clear. That's a limitation of the experimental apparaturs, but it's clear from the distribution... built over time as you say... that each individual photon, atom, etc, while observed at any given time to be particle or wave -like... has both properties at all times.

 

[Mentz114]

FrameDragger:

The fact that it takes multiple passes (as you say, a buildup) to make the pattern visible is a limitation of our detection methods.

I don't see how. A particle detector is designed to detect particles and that's what it does. What apparatus would you use to detect a 'Rubidium wave' ?

 

[Frame Dragger]

FrameDragger:


I don't see how. A particle detector is designed to detect particles and that's what it does. What apparatus would you use to detect a 'Rubidium wave' ?

The same way you do with buckyballs (C60) a la http://www.users.csbsju.edu/~frioux/two-slit/c60-slit.htm

 

[Mentz114]

The same way you do with buckyballs (C60) a la http://www.users.csbsju.edu/~frioux/two-slit/c60-slit.htm

That doesn't answer the question. The note you linked to shows the wave function that models the experiment. The wave function gives us probabilities of detecting the particle at certain locations. If you send one particle through, you detect one at the screen. If you send a 100 particles through one by one, they will make a pattern like the square of the wave function at the screen. The particle goes through one slit or the other but we cannot find out (even in principle) which slit, and still get the interference pattern building up.

I enjoyed that little note.

For more on single-photon interference look up Hong-Ou-Mandel.

 

[Dmitry67]

I wanted to remind that the balance between particle and wave behavior is interpretation-dependent.

For example, in BM there are real particles, just guided by the wave.
In MWI, there are only waves (and dots we see on the screen are the result of quantum decoherence)

 

[Frame Dragger]

That doesn't answer the question. The note you linked to shows the wave function that models the experiment. The wave function gives us probabilities of detecting the particle at certain locations. If you send one particle through, you detect one at the screen. If you send a 100 particles through one by one, they will make a pattern like the square of the wave function at the screen. The particle goes through one slit or the other but we cannot find out (even in principle) which slit, and still get the interference pattern building up.

I enjoyed that little note.

For more on single-photon interference look up Hong-Ou-Mandel.

Understood, clearly this was a misinterpretation on my part. However that does not eliminate the fact that in the absence of observation even a single photon must have properties which that experiment allows to build into an a clear interference pattern.

The distribution of many (as you say) photons allows for the interference pattern to be observed, but presumably both properties ARE always present. Even when light appear to travel as a wave, it is subjected to the effects of gravity, and when it appear to register as a single particle on a screen, it clearly has wave-like properties. The pattern at the end of the experiment can be predicted based on the arrangement of the test, and the material being tested, if not the position of each individual "strike" on the screen.

The belief the the pattern is simply a function of many particles acting without wave-like behaviour flies in the face of experimental evidence, but is compatible with non SQM or dBB theories. Even dBB postulates a pilot wave to explain experimental evidence.

Dimitry67: Well, it's interpretation dependant in the case of MWI and some others, but that is a conjecture and not really a mechanical theory. Right or wrong, it's purely ad hoc. dBB... well.. that's a separate theory that emerged before SQM, and not just a separate interpreation of QM behaviour. I still think it's wrong and somewhart contrived, but to relegate it to an interpreation of the theory it rejects is probably unfair.

 

[mintparasol]

Thanks to all who've replied here. I think it's kind of cool that a question from someone like myself who has very little understanding of the technical aspects of QM can get such a lively debate going!
I've been reading and re-reading the thread and some of the links posted here and I think I have a slightly better understanding of what's going on here than I had at the beginning.
Thanks again
ad

 

[Frame Dragger]

Thanks to all who've replied here. I think it's kind of cool that a question from someone like myself who has very little understanding of the technical aspects of QM can get such a lively debate going!
I've been reading and re-reading the thread and some of the links posted here and I think I have a slightly better understanding of what's going on here than I had at the beginning.
Thanks again
ad

Well, your question is at the heart of one of the major unsolved questions in modern physics. It was an interesting question, and bound to lead to some complex ruminations on the subject. And fireworks... musn't forget the fireworks lol.

 

[Mentz114]

FrameDragger:

The belief the the pattern is simply a function of many particles acting without wave-like behaviour flies in the face of experimental evidence, but is compatible with non SQM or dBB theories. Even dBB postulates a pilot wave to explain experimental evidence.

But SQM is saying that the pattern is determined by the wave-function otherwise there would be no interference pattern. SQM does not say that the particles behave like classical particles.

I still think it's wrong and somewhart contrived, but to relegate it [dBB] to an interpreation of the theory it rejects is probably unfair.

Saying that dBB is 'wrong' is somewhat wild and unconsidered. The same wave-function is used to calculate the trajectories and the same predictions are made. I don't think you know enough about dBB to say these things.

If you reject all this - do you think a Rubidium atom can somehow pass through both slits at the same time ?

 

[Frame Dragger]

FrameDragger:

But SQM is saying that the pattern is determined by the wave-function otherwise there would be no interference pattern. SQM does not say that the particles behave like classical particles.


Saying that dBB is 'wrong' is somewhat wild and unconsidered. The same wave-function is used to calculate the trajectories and the same predictions are made. I don't think you know enough about dBB to say these things.

If you reject all this - do you think a Rubidium atom can somehow pass through both slits at the same time ?

I realize (and have stated a couple of times) that wave-particle duality in SQM isn't simply classical waves and particles, but something else combining properties of both. I realize that the distribution on the screen is a result of the wave function of each non-classical particle. That said, depending on which theory and interpretation you subscribe to, yes... the rubidium atom, C60, or photon has wave-like properties that interefere with itself unless you place a detector in the path of the aperture.

It may be that an individual particle or atom interferes with itself, but that's impossible to confirm or refute right now as far as I know. The emergence of the pattern is the result of properties that you find in a single photon, and in some interpretations that would seem to indicate that a single particle in QM behaves like a wave at that point.

From what I've been reading of dBB (and a lot of it in the last few days after a rightous and right chastisement from Zenith) doesn't change my view of the theory. It still strikes me as last gasp of Classicism, but I've said all of this before. That said, it's the only respectable theory to survive as far as SQM re: Bell's Theorem. Personally, I don't believe the inverse of the inequalities is the case, and I think that experimental evidence found and analyzed over the next decade will eliminate it. If not, as I've stated before, and SQM fails to deliver then people will be open to different theories of the microscopic.

AS for the question about the rubidium atom, no, I don't believe one atom passes through two slits simultaneously, but it's certainly possible. What I believe has little impact on reality, and yes, I realize how ironic that statement is. I DO believe that a counterintuitive reality shouldn't be terribly shocking, and entanglement is surely as counterintuitive as it gets. Alas, there it is, with one simple, but unpalatable explanation, and dBB with a far more palatable image, but now depends on non-local hidden variables. I'm sorry, but I see that as retreat in the face of new thinking, not just academic pressures or the lack of popularity of a given theory.

 

[Mentz114]

Frame Dragger,

Fair enough. Quantum phenomena are certainly weird and trying to find intuitive meaning behind the equations is probably futile.

 

[wofsy]

I am just learning physics. Correct me please but it seems like the whole thing makes sense when you look at the quantum amplitudes as evolving according to a particular stochastic process that is similar to a Markov process.The amplitude for finding a particle in a state is the sum of the conditional amplitudes that it will land it that state given that it is in the possible previous states. If these amplitudes were real numbers rather than complex this would be a Markov process.

From this point of view, interference is not really a wave interference but rather a linear combination of conditional amplitudes. this is similar to a Markov process where the probability of finding a particle in a particular state is a sum of conditional probabilities.

Any stochastic process depends upon its initial conditions. The single and double slit are two different initial conditions.

The double slit has two initial states that generate the possible future states
according to this Markov like process of amplitudes.The single slit has only one. A triple slit would have three initial states and give a even different amplitude combinations.

When one measures which slit the particle has come out of you actually reduce the problem to a one slit case so there is no amplitude contribution any longer from the other slit. The stochastic process of amplitudes has a new initial condition that is one slit rather than two. Thus there is no interference.

This point of view explains why you see interference even when you send one particle at a time through the slits. A single particle evolves according to this Markov process of amplitudes. Measuring where the particle hits a detecting screen is much like measuring where a dust particle in a fluid will hit a barrier. Measuring many particles will produce a distribution of positions just as with dust particles. Different initial conditions will produce different distributions.

From what I can gather from reading, it seems that the Shroedinger wave equation should be called the Shroedinger diffusion equation. In fact, formally the Shroedinger equation for a free particle is a complex heat equation and just as the heat equation can be derived from a continuous Brownian motion, the Shroedinger equation can be derived form a continuous stochastic process of complex amplitudes.

The incredible thing about all of this to me is that there really is nothing else to say. That is the way it is. The classical picture of the dynamics of a particle is just inapplicable here. This quantum particle is not a particle at all but something different and ordinary usual concepts are not appropriate for it.

I would add that in Brownian motion we actually imagine a dust particle bouncing around randomly. In QM it seems that there there is no physical object bouncing around but merely a mathematical formalism that gives the right answer - every time. I think this is why Einstein disliked Quantum Mechanics whereas the theory of Brownian motion was his idea.

 

[Frame Dragger]

Frame Dragger,

Fair enough. Quantum phenomena are certainly weird and trying to find intuitive meaning behind the equations is probably futile.

I agree with literally every word in that quote. Maybe whatever supplants SQM/dBB/etc and GR will be more elucidating.

 

[Mentz114]

The amplitude for finding a particle in a state is the sum of the conditional amplitudes that it will land it that state given that it is in the possible previous states. If these amplitudes were real numbers rather than complex this would be a Markov process.

From this point of view, interference is not really a wave interference but rather a linear combination of conditional amplitudes. this is similar to a Markov process where the probability of finding a particle in a particular state is a sum of conditional probabilities.

OK, but if the amplitudes are governed by a wave function, then there is path dependent phase, and so interference will take place when you sum the amplitudes from the different paths. This can only happen with waves, so a wave phenomenon is happening. You've just rephrased it in path integral terms.

 

[wofsy]

OK, but if the amplitudes are governed by a wave function, then there is path dependent phase, and so interference will take place when you sum the amplitudes from the different paths. This can only happen with waves, so a wave phenomenon is happening. You've just rephrased it in path integral terms.

I was trying to point out that this is really a diffusion process. Waves that satisfy the wave equation are not diffusion processes.

The wave function does not satisfy the wave equation but rather a complex heat equation. I guess you do get waves when you look at stationary solutions in the presence of potentials.

 

[Mentz114]

I was trying to point out that this is really a diffusion process. Waves that satisfy the wave equation are not diffusion processes.

The wave function does not satisfy the wave equation but rather a complex heat equation. I guess you do get waves when you look at stationary solutions in the presence of potentials.

I don't know if we're talking about the same thing here. Solutions to the Schroedinger equation for realistic situations are always of the form

<br /> \psi(x,t)=Ae^{\frac{i}{\hbar}(Et+px)<br />

where E and p are the energy and momentum. This is a wave even for a free particle.

 

[wofsy]

I don't know if we're talking about the same thing here. Solutions to the Schroedinger equation for realistic situations are always of the form

<br /> \psi(x,t)=Ae^{\frac{i}{\hbar}(Et+px)<br />

where E and p are the energy and momentum. This is a wave even for a free particle.

To me, the broglie wave is a trivial case. The general Fourier transform is a super position of formal de Broglie waves and can no longer be looked at as a wave - for instance a de broglie wave has a definite momentum whereas a general Fourier transform does not -unless you want to take the mathematical view that any function that can be written as a Fourier transform is really a superposition of waves. There are some simple cases where you get a finite number of de Broglie wave glued together along their boundaries. But this picture is not right for complicated wave functions.


My point still stands that the wave equation is not satisfied here but rather a different kind of equation altogether.

 

[Mentz114]

wofsy,
you've probably got an interesting case to argue but this thread is not the place, so maybe you could start a new one about diffusion.

 

[Frame Dragger]

I don't know if we're talking about the same thing here. Solutions to the Schroedinger equation for realistic situations are always of the form

<br /> \psi(x,t)=Ae^{\frac{i}{\hbar}(Et+px)<br />

where E and p are the energy and momentum. This is a wave even for a free particle.

@wofsy: ...And not to put to fine a point on it, but that is a central tenant of SQM. If you're (Wofsy, not Mentz114) advocating a different theory that's one thing, but if you believe your description is in line with QM... it simply isn't. In fact, if you want an example of just such a momentum distribution graph, the link I provided earlier for Mentz about C60 is a visual of just this equation in action.

 

[Mentz114]

To be more precise, the plane-wave solution I quoted above won't be enough for most cases, but a linear combination of plane waves makes a nice wave-packet.

Just nit-picking ...

 

[wofsy]

I agree that this is no place to argue about QM formalism. I was just trying to explain the double slit experiment in terms of the dynamics of QM laws. The stochastic process view helped me a lot.

All I was saying is that the Shroedinger equation is not the wave equation. The wave equation involves the second derivative in time as well as space. The Shroedinger equation only uses the first derivative of time. This is a huge difference. When one uses only the first derivative you get entirely different dynamics. Instead of wave propogation you get stochastic processes. For instance, the heat equation only uses the first derivative in time.

It was revealing for me to read in Feynmann's Lectures On Physics that the Shroedinger equation falls simply out of the Markov like process of amplitude evolution. The hidden reality here was the stochastic process of complex amplitudes just as the hidden reality in heat flow is Brownian motion. This hidden reality generalizes to all QM situations such as spin or chemical bonds. In these cases,wave-particle duality does not apply.

For me, thinking about what is really meant by a wave here has been key for trying to understand the theory. It seems that simple wave packets - a finite number of superposed de Broglie waves - can be thought of as waves formally since they look like waves with localized amplitudes - but for the life of me I do not know what it means to think of de Broglie waves as particles and then superpose these particles to get another particle. That just isn't right. However with standard linear waves such superposition makes total sense.

 

[Frame Dragger]

To be more precise, the plane-wave solution I quoted above won't be enough for most cases, but a linear combination of plane waves makes a nice wave-packet.

Just nit-picking ...

"...just this equation" as in, "This very equation" not, "This equation alone". ;)

 

[mintparasol]

Well, your question is at the heart of one of the major unsolved questions in modern physics. It was an interesting question, and bound to lead to some complex ruminations on the subject. And fireworks... musn't forget the fireworks lol.

The lack of consensus is illuminating (pardon the pun!)

I had been under the impression that QM models were as immutable as, say, special relativity or the laws of thermodynamics

 

[Frame Dragger]

The lack of consensus is illuminating (pardon the pun!)

I had been under the impression that QM models were as immutable as, say, special relativity or the laws of thermodynamics

That definitely depends on who you ask, but I doubt that anyone here is filled with a deep and abiding certainty about QM,dBB,etc... etc... and when you get into the realm of interpretations, well... the only concesus is probably forced. I think everyone here has done a fine job of advocating a viewpoint, but that is all that most people can claim. QM and GR both cry out for either unification, or a new theory to replace them (crazy unlikely in the case of GR at least ;) ). The question of what will emerge from an understanding of how the world of the very small is 1.) unlike the world we observe, but from which that world emerges 2.) Like the world we observe, but for various reasons appears to be unlike 3.) OTHER... is unsolved and open for debate.

That's a pretty gross simplification on my part, but that's the kicker... without the math none of these theories make much sense using Classical analogues past the introductory period. The fact that terms such as "Observer" and "Information" take on new meanings as they become terms of art in the field, doesn't help for some people either. That's a work in progress for you though!

 

[DrChinese]

I had been under the impression that QM models were as immutable as, say, special relativity or the laws of thermodynamics

The mathematical formalism itself is essentially immutable (of course that could change) and has remained in place for over 80 years. What changes is the mapping of the formalism to underlying mechanical processes, something which is not strictly required for any theory. This "mapping" is the source of the debate and confusion, just in case that point was not clear from the above.

And by the way, there is a similar debate raging about relativity and its interpretations. Although the one about QM is more well known and tends to have more *robust* debate.

 

[Frame Dragger]

The mathematical formalism itself is essentially immutable (of course that could change) and has remained in place for over 80 years. What changes is the mapping of the formalism to underlying mechanical processes, something which is not strictly required for any theory. This "mapping" is the source of the debate and confusion, just in case that point was not clear from the above.

And by the way, there is a similar debate raging about relativity and its interpretations. Although the one about QM is more well known and tends to have more *robust* debate.

To be fair to those scientists who stake their careers and reputations on various theories... it's easy to see politics stifling science in hindsight, but when your *** is on the line... not so easy. So, in some cases the debate is spirited for the sake of retaining one's viewpoint or standing, and sometimes it's spirited because the math says very strange things about the universe that we as humans do not see in our everyday lives (and recognize as such at least).

GR and SR have plenty of debate, including ideas such as treating time as separate from space. Einstein's theories however, have had the benefit of experiments which refute some counterclaims and support it. Time dilation, gravitational lensing, and more have been directly observed. By its very nature, QM defies that same degree of precision in the absence of new thinking, math, and technology.

Want to test GR? Make some really great telescopes and wait for the right time (or make some really good gyros and lasers in the extreme). Want to test SQM? Build the Large Hadron Collider and cross your fingers. You see the problem... ;)