PDA

View Full Version : wave/particle duality-- a clarification



Ken G
2006-Aug-13, 09:43 PM
I'm posting this in Q&A because there seem to be a lot of questions about it, and there are some oft-repeated answers that don't really hit the mark, which may contribute to why the questions are heard so often! The usual statement is that macro particles, like cannon balls, obey particle mechanics, which is basically Newton's laws applied to trajectories, while micro particles, like electrons, sometimes do this, and sometimes obey wave mechanics. Hence the latter are said to exhibit "wave/particle duality".

This is a very confusing pedagogy, because it seems to arbitrarily insert different behaviors for different objects, based on their masses. But what about a particle the size of an atom with the mass of a cannon ball? Is that going to exhibit "duality"?

I feel the best answer to this can be seen using a mathematical analogy. Have you ever heard of "calculus/arithmetic duality"? Probably not, it wouldn't have the same mystique as "wave/particle duality". But it's the same thing. Here's what I mean. Let's say I told you that I travel at 50 miles per hour in a straight line, for an hour. How far did I go? As a calculus problem, this goes: the distance is the integral of v*dt over an hour of time. Since v=50 is constant, it comes out of the integral, dt integrates to t, and we get distance=50*t. Now I'll bet you were way ahead of me-- you used arithmetic to accomplish the same result. You see, that's "calculus/arithmetic duality". It was "really" a calculus problem, but you were able to solve it using arithmetic instead. Get it? So it is with wave/particle duality-- all things obey wave mechanics, you just don't always need to know this. It's not so mysterious when you look at it this way, which makes it a better pedagogy, I claim.

max8166
2006-Aug-14, 08:05 AM
That's fine where the result is the same (50 miles), but with wave/particle duality the results are not the same. With one reality (particles) you get two straight lines on a measuring surface (in the duel slit experiment) and with the other reality (wave) you get a series of banded lines, interferance. In fact (with light) you get a set of banded lines (therefore it's a wave), unless you attempt to measure which photon goes thru which slit at which point reality changes... and you get two lines (therefore it's Particles)

Ken G
2006-Aug-14, 11:22 AM
That's fine where the result is the same (50 miles), but with wave/particle duality the results are not the same.
Actually, they are the same, that's the point. The issue is, the de Broglie wavelength gets very small for macro particles. Wave mechanics with very small wavelengths is the same as particle trajectories, which is why trajectory physics was invented even though it isn't "exactly right" (what really is, in physics).


In fact (with light) you get a set of banded lines (therefore it's a wave), unless you attempt to measure which photon goes thru which slit at which point reality changes... and you get two lines (therefore it's Particles)

Not if you use gamma rays! It is just a question of the wavelength compared to the slit sizes. This is what I'm saying-- all this "duality" business is quite a bit overblown, what we are talking about is simply two limits of the same thing. We don't say particles are "dual" because sometimes they behave Newtonian, and sometimes relativistic, do we?

Spaceman Spiff
2006-Aug-14, 03:31 PM
Interesting thought, although in addition - large numbers of interacting quanta also tend to "decohere", but perhaps we're really talking about two sides of the same coin.

Cougar
2006-Aug-14, 04:27 PM
Actually, they are the same, that's the point. The issue is, the de Broglie wavelength gets very small for macro particles....

Well, I don't know about the de Broglie wavelength, but as Dr. Spiff notes, "large numbers of interacting quanta also tend to "decohere"...", which works out such that the probability(A and B) happening = P(A)*P(B), as we normally think. But when we don't have a large number of interacting quanta, then this is not the case. Gell-Mann's The Quark and the Jaguar has a pretty good analogy on this, dealing with horses at the racetrack....

Ken G
2006-Aug-15, 03:46 AM
But then we are back to the de Broglie wavelength, I don't know the racetrack analogy... sounds interesting. Especially if it helps dispell the mystique of "duality".

Peter Wilson
2006-Aug-16, 11:25 PM
Its like dogfish, catfish, and parrotfish. What are those? Animals that are part dog/cat/bird and part fish?

No, those are just names we label certain animals with for lack of better descriptors. Likewise, is an atom part particle and part wave? No, these are just clumsy labels we use for lack of better terms.

Quanta are what they are; the duality is what it is; and the mystique will not go away, no matter what we call it.

Ken G
2006-Aug-17, 05:11 AM
I agree, but what I'm saying is that the "duality" is often misunderstood-- what most people think of as particlelike behavior is simply quantized wave behavior in the limit of short wavelengths-- it is not something 'different'. It is just a simple limit, like arithmetic is to calculus. I suppose many people view those two as quite different, but we don't scratch our heads about the "duality" of mathematics as a result.

Peter Wilson
2006-Aug-17, 04:08 PM
In the Nature-vs-Nurture human behavior debate, are nature and nurture "really" the same thing? In one sense, Yes, because our nature has been nurtured, but what's wrong with saying there are two sides to our character?

Basically, if we want to account for human behavior, we have to take both nature and nurture into consideration. And if we want to understand the quanturm, we must acknowledge the duality of its nature. The great "leap" in understanding came when physicists accepted the reality of the duality.

Ken G
2006-Aug-18, 01:14 AM
No, the "leap" came when they understood the ubiquitous importance of wave mechanics, and the way particle mechanics was merely a convenient limiting case of this more general rule. Prior to quantum mechanics, we thought wave mechanics and particle mechanics were completely different things. That was duality. Quantum mechanics is unification, the "mysteriousness" of duality is missing the message, in my view. Of course, the role of decoherence due to internal interactions in macro objects has been mentioned, but that's a different issue because it does not relate to the duality question (nobody thinks a bowling ball has a "dual" description).

Peter Wilson
2006-Aug-18, 05:34 PM
I see your point.

I guess I've had a slightly different take on the history of it. I recall the particle approach and the wave approach yielding the same answer. Philosophically, I look at mathematics as a tool, not an answer book. If you can model one phenomenon two completely ways, and both give you the correct answer, which can be considered "more correct"? I do not think mathematics can answer that question; math is a tool for modeling reality, but it yields few clues to the deeper nature of reality.

And perhaps a bowling ball is a bowling ball, but "in reality," it is very different than what you see. To x-rays, the ball is soft and porous; to neutrinos, it is almost non-existant. You can only say "what" something is in the context of "how" you are looking at it. Is the door on the microwave oven clear or opaque? To my eyes, it is a metal plate so full of holes I can see right through it, and notice my re-heated soup is boiling over and making a big mess in there. But to microwave eyes, the door is completely opaque. So the answer to the question--is the door clear or opaque?--depends on how you look at it.

And what is the true nature of the quantum? (Everyone all together here) It depends on how you look at it.

To me, it is a deep mystery that I am comfortable with. It does not bother me that we cannot pigeon-hole it with absolute certainty into one category or another, "It is this..." or, "It is that." Somethings, you just can't explain. Se le vie, or something like that.

Ken G
2006-Aug-18, 07:43 PM
You can only say "what" something is in the context of "how" you are looking at it.
I agree, and with the role of mathematics. What I'm really saying is that we should not think there is anything special or mysterious about quantum mechanics simply because it admits to a "dual" picture of the behavior of particles. This is completely normal in physics, and what deeper theories like quantum mechanics really do is demonstrate the underlying unity in reality. It is this underlying unity, not its duality, that makes quantum mechanics such a breakthrough.

Bad jcsd
2006-Aug-18, 08:37 PM
I'll just say this wave particle duality is a pre-quantum hypothesis which can account for some observations. In quantum mechanics neither the classical pictur eof the particle or the wave applies.

Spaceman Spiff
2006-Aug-18, 09:42 PM
Of course, the role of decoherence due to internal interactions in macro objects has been mentioned, but that's a different issue because it does not relate to the duality question (nobody thinks a bowling ball has a "dual" description).

Right - but then there are the fuzzy things called nano-particles that live in the nether world of almost classical and not quite quantum world descriptions. And as we investigate their properties and interactions with other force and matter fields, we'll probably need to scratch our heads a little harder.

But maybe I've side tracked you a bit. I think you're wanting to better define the "duality" question for those entities that are clearly dominated by their quantum/wave properties. If I understand you correctly, I think what you are suggesting is indeed a better way to think of it. An electron is a quantum of the "electron field". A photon is a quantum of the electromagnetic field. These things are fields (wave-like) as they travel (as disturbances in the field) through space(**). When such fields interact, the interaction is localized and an exchange of quantized energy we call photons or electrons (or both or other quanta, as the case may be) occurs.

(** Of course, what the heck that actually means and the relation of these quantized fields to the space-time of General Relativity is an unsolved problem).

Spaceman Spiff
2006-Aug-18, 09:49 PM
I see your point.

I guess I've had a slightly different take on the history of it. I recall the particle approach and the wave approach yielding the same answer. Philosophically, I look at mathematics as a tool, not an answer book. If you can model one phenomenon two completely ways, and both give you the correct answer, which can be considered "more correct"? I do not think mathematics can answer that question; math is a tool for modeling reality, but it yields few clues to the deeper nature of reality.

And perhaps a bowling ball is a bowling ball, but "in reality," it is very different than what you see. To x-rays, the ball is soft and porous; to neutrinos, it is almost non-existant. You can only say "what" something is in the context of "how" you are looking at it. Is the door on the microwave oven clear or opaque? To my eyes, it is a metal plate so full of holes I can see right through it, and notice my re-heated soup is boiling over and making a big mess in there. But to microwave eyes, the door is completely opaque. So the answer to the question--is the door clear or opaque?--depends on how you look at it.

And what is the true nature of the quantum? (Everyone all together here) It depends on how you look at it.

To me, it is a deep mystery that I am comfortable with. It does not bother me that we cannot pigeon-hole it with absolute certainty into one category or another, "It is this..." or, "It is that." Somethings, you just can't explain. Se le vie, or something like that.

Nice points.
As for your last one....it kinda reminds me of the problem of pigeon-holing going on over in Prague, regarding when to call something a planet...but I digress.

Ken G
2006-Aug-19, 01:37 AM
Right - but then there are the fuzzy things called nano-particles that live in the nether world of almost classical and not quite quantum world descriptions.
Yes, if you have a little decoherence, but not complete decoherence, that's the real place to look for dual behavior, in the sense of a mixture of classical and quantum ideas.



I think you're wanting to better define the "duality" question for those entities that are clearly dominated by their quantum/wave properties.
Yes, for single particles, but with large enough energies to express themselves classically in some applications. This is where people start hemming and hawing about "duality", instead of just saying we are talking about simple limits of the same unified theory. Newtonian gravity is an example of a simple limit of a unified theory of gravity (unified with light, I mean, not quantum mechanics-- I have no idea how that might require further unifications), yet no one is mystified by the force vs. curvature "duality" of gravity.


An electron is a quantum of the "electron field". A photon is a quantum of the electromagnetic field. These things are fields (wave-like) as they travel (as disturbances in the field) through space(**). When such fields interact, the interaction is localized and an exchange of quantized energy we call photons or electrons (or both or other quanta, as the case may be) occurs.
Precisely, it's a self-contained approach that applies all the time, everywhere. Duality is merely a side effect of simplified approaches that we use purely for convenience.

Digix
2006-Aug-19, 01:40 AM
But maybe I've side tracked you a bit. I think you're wanting to better define the "duality" question for those entities that are clearly dominated by their quantum/wave properties. If I understand you correctly, I think what you are suggesting is indeed a better way to think of it. An electron is a quantum of the "electron field". A photon is a quantum of the electromagnetic field. These things are fields (wave-like) as they travel (as disturbances in the field) through space(**). When such fields interact, the interaction is localized and an exchange of quantized energy we call photons or electrons (or both or other quanta, as the case may be) occurs.


I dont understand why do you say "fields". quantization is all about energies, fields are not quantizied. Planck constant is J*s energy and time.

Anyway I think, that QM is newtonisn physic modified with probabilistic wavefunction. but here we got nasty word "ptobability"

Was there any atempts to do oposite: modify wavefunction to express newtonian and relativistic physic?
I found this atricle which derives all newtonian physic from wavefunctions (at least I think so)
http://www.calphysics.org/articles/gravity_arxiv.pdf

Ken G
2006-Aug-19, 02:45 AM
That's a cute article, thanks for that Digix. It provides an even deeper way to unify Newtonian ideas and quantum mechanics.

publius
2006-Aug-19, 03:47 AM
Digix,

Fields are indeed quantized. That's what "Quantum Field Theory" (QFT) is all about. :) Quantum Mechanics (the "first quantization") was about quantizing mechanics. QFT, the "second quantization" is about quantizing classical fields. QED, Quantum Electrodynamics, is the quantum version of EM theory.

This was a point I was trying to get you to appreciate in the "radio frequency photon" thread.

The field does not have a definite value, there is only a probability distribution of what value of E and B one might measure at a point in space. A (real) photon is a contribution to that probability distribution. When that photon interacts, its contribution is removed, taking a "chunk" out of the fields.

-Richard

Digix
2006-Aug-19, 04:02 AM
publius, from what you say I am strting to think that I dont undertand what is quantization.

what I know base of QM is planck constant h(J*s), which means that within 1 second you can transfer n*h J amount of energy, in 0.5 seconds it will be n*h/2 J
of course from energy we can derive nearly any physical value. and we can derive any other interpretations expressing energy and time values by anything we like.

But how can you have quantization which does not depend on energy and time? To use planck constant you must have energy and time values, or else your formula will be wrong.

Anyway, i see that all QM is trying to use mechanic with probabilities, do you know, why phisicists chose particles, istead of waves afterall in case of duality, we are free to choose best.
I found some articles like that one before
http://www.calphysics.org/zpe.html
This seems better than various string and membrane theories, and they even does not introduce anything new .

publius
2006-Aug-19, 04:29 AM
Digix,

Planck's constant is indeed the "fundamental quantum of nature", but it is much more than simply the product of energy and time in the way you're thinking about. What is energy times time? That is one way to express the dimension of a very deep quantity known as "action". Action can also be expressed as momentum times distance(position, actually).

You'll note those are two pairs that come up in the uncertainty principle. Two dynamical variables whose product is action are called "canonical conjugates" and the uncertainty principle applies to them. Electric and magnetic flux are such variables as well -- their product is action.

Action is not something one learns about in Physics 101 courses, because it gets pretty deep and one needs to learn a lot of physics to appreciate it, as well as a lot of math -- the "Calculus of Variations" is the math associated with it. It is involved in a very fundamental principle known as "The principle of Least Action", which, roughly, states the evolution of a dynamical system will be such that the action is minimized. (actually, the action is techinically only "stationary", it could be maximum as well as minimum)

Enter quantum theory. Besides being stationary, the action is also quantized. Planck's constant is the fundamental chunk of action.

One can derive the laws of nature by requiring the action be stationary. That is, the equations we're familiar with can be derived by using the Calculus of Variations to find the differential equations that minimize the action integral. Look up the variational calculus to see the formalism.

Basically it goes like this. The action of dymanical system can be expressed as a general "path integral", some function integrated over a path. Conside a function y(x). We know from simple calculus that function can have local extrema.

Now, consider not finding what value of x produces extreme values of y, but consider finding which function y(x) will minimize the value of a path integral involving it and its derivatives. That is, the whole darn function is a variable.

The calculus of variations is all about solving problems like that. The result is a differential equation for the unknown function y(x).

Now, once you know a way to express the action of some physical system, you can apply that technique, and the resulting diffential equation gives you the "law of nature" for that system.

Trouble is, you have to first know that some quantity like energy, momentum, electric field, etc, etc, actually exists before you can know it "makes action". :lol:

Anyway, one can write an expression for the action of the electromagnetic field. Apply the Principle of Least Action and get some differential equations for those fields. Guess what those equations are? You guessed it, Maxwell.

That's how Einstein came up with the field equation for General Relativity. He minimized the action of the curvature of space-time.

Don't ask me to elaborate on this. I've told you as much as I know. :) This gets very deep and very fundamental and I have only a glimpse of how it all "clicks" on the grand scale.


-Richard

Digix
2006-Aug-19, 05:27 AM
All that is correct, but I assume that fields are stationary, or else they become waves.
stationary fields cant be quantized, because time comonent is infinite, so quantization energy is zero. If we talk about changing fields, then we actualy talk about waves with energy anyway, so it is obvious that they are quantized.

RussT
2006-Aug-19, 11:41 AM
So, how does the Fine Structure Constant (FSC) fit into all of this?

Where does a=~1/137 fit into the Planck size/length/time regime?

In other words the electron is ~10 ^-15 and Planck size is ~ 10 ^-33, so is there a way to compare this to the FSC a= ~1/137 scale?

Ken G
2006-Aug-19, 02:38 PM
stationary fields cant be quantized, because time comonent is infinite, so quantization energy is zero. If we talk about changing fields, then we actualy talk about waves with energy anyway, so it is obvious that they are quantized.

Your comments apply to real particles, but you overlook the role of virtual particles. The former represent the quantization of propagating energy, and the latter are the way to quantize the stationary fields that you claim cannot be quantized. The disconnect is that, according to the uncertainty principle, there is no such thing as a truly stationary field, because it would require an infinite energy uncertainty. This is also related to those zero-point energy articles you posted (which were quite interesting, by the way, but I did not see any of them mention "interference effects"-- a large oversight, as it is also fundamental to the aspects of quantum mechanics that you are missing.)

Digix
2006-Aug-20, 04:50 AM
Your comments apply to real particles, but you overlook the role of virtual particles. The former represent the quantization of propagating energy, and the latter are the way to quantize the stationary fields that you claim cannot be quantized. The disconnect is that, according to the uncertainty principle, there is no such thing as a truly stationary field, because it would require an infinite energy uncertainty. This is also related to those zero-point energy articles you posted (which were quite interesting, by the way, but I did not see any of them mention "interference effects"-- a large oversight, as it is also fundamental to the aspects of quantum mechanics that you are missing.)

You go in some absurd area if you say that pure static field is quantized. of course all our interactons are quantized, but if some elctron sits in the void from the begining of universe then how do you apply planck constant?
without motion we even dont have time definition.

That article only shows flexibility if wave theory, I am quite sure that I could derive any exsisting and nonexsisting physical law from wavefunction.

Ken G
2006-Aug-20, 02:04 PM
You of course all our interactons are quantized, but if some elctron sits in the void from the begining of universe then how do you apply planck constant?
without motion we even dont have time definition.

You have shown the flaw in your own argument. A universe with just one electron has no physics at all, so why use it show the "absurdity" of one particular physical insight?

Digix
2006-Aug-20, 06:48 PM
You have shown the flaw in your own argument. A universe with just one electron has no physics at all, so why use it show the "absurdity" of one particular physical insight?

Why no physic? all laws are same. but if you talks about or detecting that electron, then you talk about interaction anyway.

Ken G
2006-Aug-20, 07:01 PM
How would you know the laws are the same in a universe with nothing but a single electron? How would you know there are any laws at all?

Digix
2006-Aug-20, 08:56 PM
Because this is thought experiment.
and same will be if we just charge capacitor and leave it for 100 years.
in that case charge quantization should depend on how fast you charge and discharge capacitor. isnt it so?

Ken G
2006-Aug-20, 09:25 PM
No. Charge on a capacitor is quantized by the electric charge. You know that. You probably meant energy, and the uncertainty would depend on how fast you discharge it, but that would be a small fraction of the actual energy of the capacitor.

Digix
2006-Aug-21, 12:31 AM
You probably meant energy, and the uncertainty would depend on how fast you discharge it, but that would be a small fraction of the actual energy of the capacitor.

Yes, I was talking about this all time, but I think that uncertanity and quantization are same thing. energy or field strength(assuming that we change it be moving capacitor plates) is not quantized if time is nearly infinite, we can adjust any field strength over infinite time.

Ken G
2006-Aug-21, 01:43 AM
This is untrue. Photons at a frequency f still come in packets of energy h*f, even if those photons have existed forever (or practically forever). Indeed, photons are idealized entities that have no energy uncertainty whatsoever, the uncertainty appears in the fact that you don't know which photon mode is going to be excited in a given situation.

Peter Wilson
2006-Aug-22, 09:07 PM
I do not think the arithmatic/calculous analogy is good, because they are really two different realms. The example you give, distance vs. constant velocity, contains no dependent variables. The simple equation, dr/dt = kr, for example, contains a dependent variable, r, and cannot be solved using arithmetic (k is a constant; t is time; d is differentiator)

Nonetheless, taking into consideration the feedback generated here, can you take another stab at a succinct, one-size-fits-all explanation of the wave-particle duality?

Ken G
2006-Aug-23, 02:01 AM
Yes, here is the duality. Particles are what they are, waves are how they move. This is always true, the concept of a trajectory is not necessarily a particle concept, this is my point. Trajectories are simplified wave mechanics in the short wavelength limit, just as algebra is simplified calculus in the constant-v limit.

jack butler
2007-Mar-06, 06:34 PM
I'm posting this in Q&A because there seem to be a lot of questions about it, and there are some oft-repeated answers that don't really hit the mark, which may contribute to why the questions are heard so often! The usual statement is that macro particles, like cannon balls, obey particle mechanics, which is basically Newton's laws applied to trajectories, while micro particles, like electrons, sometimes do this, and sometimes obey wave mechanics. Hence the latter are said to exhibit "wave/particle duality".

This is a very confusing pedagogy, because it seems to arbitrarily insert different behaviors for different objects, based on their masses. But what about a particle the size of an atom with the mass of a cannon ball? Is that going to exhibit "duality"?

I feel the best answer to this can be seen using a mathematical analogy. Have you ever heard of "calculus/arithmetic duality"? Probably not, it wouldn't have the same mystique as "wave/particle duality". But it's the same thing. Here's what I mean. Let's say I told you that I travel at 50 miles per hour in a straight line, for an hour. How far did I go? As a calculus problem, this goes: the distance is the integral of v*dt over an hour of time. Since v=50 is constant, it comes out of the integral, dt integrates to t, and we get distance=50*t. Now I'll bet you were way ahead of me-- you used arithmetic to accomplish the same result. You see, that's "calculus/arithmetic duality". It was "really" a calculus problem, but you were able to solve it using arithmetic instead. Get it? So it is with wave/particle duality-- all things obey wave mechanics, you just don't always need to know this. It's not so mysterious when you look at it this way, which makes it a better pedagogy, I claim.

I like very much the analogy of the supposed calculus/arithmetic duality. Find the general explanation quite useful, and agree it would make for better teaching. Philosophically, one might observe that "waves" and "particles" are each human categorizations of phenomena. Why should it be surprising to find that at the limits of our perception the categories are not absolute? Why should it be surprising that fundamental reality gives rise to each of the characteristics we perceive as distinct?

The same may be said--and has been, mathematically, in Einstein's famous one-line trimeter poem--of the distinction between energy and matter. It's a distinction we have found very useful, but not one the universe at large appears to trouble itself over.

Ken G
2007-Mar-07, 10:53 PM
Well put-- I think we agonize a bit too much over our interpretations of reality. Another perfect example is the two-slit experiment of quantum mechanics. For a while there, the "Copenhagen interpretation" of this phenomenon, which basically asserts that if you prepare a system according to one measurement than there become a whole set of new measurements which can do no better than ascribing a probability to, was viewed as sacrilege. How could science be allowed to set such limitations on itself? It seemed like throwing in the towel. But when you realize that we have simply chosen a particular way of knowing reality, it should not surprise us that in applications that are sublimely separate from the experiences that trained our brains, we should encounter inabilities to answer every question that we have language to pose. The universe is not beholden to our language, like "particle" and "wave" and "noncommuting observables". So then the big question is, should we be bothered that quantum mechanics and general relativity are inconsistent at the Planck scale? I would tend to say no, but I do agree that unification is a fundamental goal of physics.

jack butler
2007-Mar-08, 01:34 AM
So then the big question is, should we be bothered that quantum mechanics and general relativity are inconsistent at the Planck scale? I would tend to say no, but I do agree that unification is a fundamental goal of physics.

Agree again, hope not too boringly. Why should quantum mechanics and gr agree at the Planck scale? On the other hand, as you point out, unification is a fine goal. Keeps one busy thinking.

sirius0
2007-Mar-08, 06:07 AM
I challenge anyone to demonstrate quatisation in free space of any field that doesn't require a quantised device to measure it. Sure there is a glaring flaw in my science here but not in my logic. By the way the wave packet that I am made of has an evanescent wave or particle uncertainty of 10^(-28) metres (I calculated this once).
I guess this means that if I am motionless and positioned 10^(-28) metres from another object then I cannot be certain if I am touching it (or slightly inside it) or not (for any given instant).

SockMonkey
2007-Mar-08, 11:43 AM
I always interpreted the wave/particle paradox as descriptions of how photons *behave* as opposed to a definition of what they *are*.
I'm no physicist but that always worked for me.
That's my two cents for whatever it's worth.

Ken G
2007-Mar-08, 11:47 AM
I challenge anyone to demonstrate quatisation in free space of any field that doesn't require a quantised device to measure it. I'm not sure what you mean here. Are you saying that light "requires a quantized device to measure it?" The universe existed long before there was any way to measure light, yet it was quantized (that's how you avoid the "ultraviolet catastrophe" of thermal radiation, see the resolution by Planck). Or are you saying that the "quantized measuring devices" you have in mind can be hypothetical as well? If so, then you are not really saying much of anything-- science just says that light energy is quantized, it doesn't say why. You are welcome to assert that the "why" is because there exists hypothetical "quantized measurers", but the fact is, all of science rests on measurers (real ones in the case of experimental science, hypothetical ones in the case of theoretical science), so this is not exactly a news flash.

By the way the wave packet that I am made of has an evanescent wave or particle uncertainty of 10^(-28) metres (I calculated this once).It sounds like you are associating your characteristic energy with some kind of deBroglie wavelength, but this is just bad quantum mechanics. You are a macroscopic system comprised of zillions of microscopic quantum systems that all have random coherence relations. There is no way you can treat yourself as a single quantum system, nor associate with yourself a meaningful deBroglie wavelength, because that only describes your bulk motion but your internal degrees of freedom are far more important over such tiny scales. Put differently, the state of any of your atoms will have more impact on "what you are touching" than your overall bulk deBroglie wavelength treating yourself as all one particle. It's basically the "Shroedinger Cat Paradox" baloney.

Ken G
2007-Mar-08, 11:50 AM
I always interpreted the wave/particle paradox as descriptions of how photons *behave* as opposed to a definition of what they *are*.Yes, I think that's a perfectly good way to think about it. The point is, science is not really about discerning the fundamental essence of things, it's about figuring out a working model for understanding or predicting how things behave. The word "duality" sounds like a statement of essence, not just a model of behavior, which is my objection to it. If I sometimes drive to work, and other times cycle, am I exhibiting "car/bicycle duality"?

kzb
2007-Mar-08, 12:58 PM
I think Ken G's idea is a very good way of helping people understand this area a bit better, at least if those people actually have basic understanding of calculus, (which is not a given these days).

Wave/particle duality is a bit of a misleading phrase, and I think a lot of people tend to misinterpret it.

SockMonkey
2007-Mar-08, 02:50 PM
Good analogies can be pretty helpful in keeping the general public from being spooked or alienated by science. I've noticed that a lot of the ones commonly used on the news and in basic textbooks aren't as apt as they could be.

korjik
2007-Mar-08, 04:22 PM
There is a really simple clarification.

All particles of any size always act as waves.

Which is true. It is just once you get out of the nanoscale, the wavelenghts are so short that there is pretty much no effect to the wave by anything.

Ken G
2007-Mar-08, 04:56 PM
Indeed, see posts #1 and #3. The way I like to think of it is, particles "travel" like waves, and "arrive" like particles (i.e., little localized quanta of energy).

sirius0
2007-Mar-08, 07:49 PM
I'm not sure what you mean here. Are you saying that light "requires a quantized device to measure it?" The universe existed long before there was any way to measure light, yet it was quantized (that's how you avoid the "ultraviolet catastrophe" of thermal radiation, see the resolution by Planck). Or are you saying that the "quantized measuring devices" you have in mind can be hypothetical as well? If so, then you are not really saying much of anything-- science just says that light energy is quantized, it doesn't say why. You are welcome to assert that the "why" is because there exists hypothetical "quantized measurers", but the fact is, all of science rests on measurers (real ones in the case of experimental science, hypothetical ones in the case of theoretical science), so this is not exactly a news flash.
It sounds like you are associating your characteristic energy with some kind of deBroglie wavelength, but this is just bad quantum mechanics. You are a macroscopic system comprised of zillions of microscopic quantum systems that all have random coherence relations. There is no way you can treat yourself as a single quantum system, nor associate with yourself a meaningful deBroglie wavelength, because that only describes your bulk motion but your internal degrees of freedom are far more important over such tiny scales. Put differently, the state of any of your atoms will have more impact on "what you are touching" than your overall bulk deBroglie wavelength treating yourself as all one particle. It's basically the "Shroedinger Cat Paradox" baloney.


:) Ken I wasn't all that sirius with this post. But now you have asked questions (sigh). I can assure you that despite some door ways being very close together that I have never gone through both at once. the interference pattern would take a long time to clean up!:)

A lot of your points I agree with which is why I said that there is a flaw in my science, but not my logic.

I am interested in the ultraviolet catastrophe and will look this up (many of your corrections have had imense pedagogical value to myself).

What I meant about that challenge regarding quantisation of a field (not actually light) has a similar feel to what I think when you say.

"The way I like to think of it is, particles "travel" like waves, and "arrive" like particles (i.e., little localized quanta of energy)."

I thought that the whole body as a particle/wave might have had some very limited pedagogical value. Remember I said I was motionless (0 Kelvin) (still doesn't work though!). My point is that the problem with this is the same as the duality issue. It is only an issue with things the size of humans trying to understand things that behave as they are; but don't match the veneer of mental analogies we all try to impose.

SockMonkey
2007-Mar-09, 03:11 AM
I think it's important to remember the time dialation effect of lightspeed when pondering the concept.
From the photon's perspective it's entire path is shrunk down to a single point so it exists everywhere along it in the same instant.

I was going over how this affects what a photon is before going to bed last night and I think there is a conceptual answer in there somewhere but my brain said "Sorry buddy, I'm going on a break."

Darn lazy brain, and after I fed it all that Asimov too...

Ken G
2007-Mar-09, 04:28 AM
I am interested in the ultraviolet catastrophe and will look this up (many of your corrections have had imense pedagogical value to myself).Thanks, I didn't mean to be critical, just clearing up some points. The ultraviolet catastrophe is an infinite energy flux from a thermal radiation source in classical thermodynamics, and the quantum effects fix it. It was really the birth of quantum mechanics, and Planck's constant.


My point is that the problem with this is the same as the duality issue. It is only an issue with things the size of humans trying to understand things that behave as they are; but don't match the veneer of mental analogies we all try to impose.

I see, you're saying that science is a patchwork of mental veneers, and there are a few "creases" that we shouldn't be terribly surprised about. I agree there.

Ken G
2007-Mar-09, 04:30 AM
I think it's important to remember the time dialation effect of lightspeed when pondering the concept.
From the photon's perspective it's entire path is shrunk down to a single point so it exists everywhere along it in the same instant.


Yeah, that's another issue too-- if we're bothered by wave/particle duality of light, we should be downright tortured by the relativity implications.

SockMonkey
2007-Mar-09, 06:58 AM
I see, you're saying that science is a patchwork of mental veneers, and there are a few "creases" that we shouldn't be terribly surprised about. I agree there.

True, I like to phrase it like this:
Do not fear paradox because it does not exist, there is only the *appearance* of paradox caused by incomplete knowledge.

Ken G
2007-Mar-09, 08:23 AM
How do you know? Perhaps paradox is a fundamental element of reality, but only at the deepest levels.

SockMonkey
2007-Mar-09, 09:56 AM
How do you know? Perhaps paradox is a fundamental element of reality, but only at the deepest levels.

Hmm, I think I'm gonna need a while on this one.

Ken G
2007-Mar-09, 01:11 PM
I didn't really expect an answer, it's just something to think about. Kind of "zen", I imagine.

jack butler
2007-Mar-10, 03:10 PM
How do you know? Perhaps paradox is a fundamental element of reality, but only at the deepest levels.

I have found that in my thinking paradox always signals mystery. Either my conceptions are not adequate or there is a totality which includes both of my apparent answers (could be both of course). Isn't the root meaning of the word itself something like "beyond teaching"?

Ken G
2007-Mar-11, 01:01 AM
Good point, I don't know the etymology but judging from organic chemistry it seems that "orthodox" means "thinking on the same side" and "paradox" would be "thinking on opposite sides". So an orthodox view fits in with prevailing opinion, and a paradoxical view is opposed to prevailing thinking. I think it also generally carries the connotation of being in opposition with itself, so that it signals a conflict no matter what you do. Generally that's a bad thing, and science always tries to avoid it, but maybe it sits just outside science's gates, waiting for unwary passers to fail to notice the "exit" sign. Mystery, yes-- the goal of science being to eke out a perimeter against mystery, but not to eliminate it altogether, as that boundary helps to define what science is.

SockMonkey
2007-Mar-11, 12:06 PM
That's what I was thinking. I just didn't know how to word it coherently.

Michael Noonan
2007-Mar-11, 03:11 PM
Has there ever been a duality test?

I was thinking create one or two slits but in each have something in the middle like a convex lens focused at the exit or thereabout.

If it worked you would see if the particle could enter in whatever form be required to act as a wave then appear to exit as a particle.

Is this possible?

Ken G
2007-Mar-11, 03:28 PM
Particles are affected by lenses according to wave theory, so it would be easy to predict the result of such an experiment, and it would still exhibit similar wave/particle "duality". It might be an interesting problem to work out, but there aren't going to be any surprises there. Diffraction and refraction get combined in calculations all the time.

sirius0
2007-Mar-12, 11:06 PM
We are blind to the intrinsic nature of these particle/waves why don't we try the logic model path? My concern is that we are approximating their nature from our macroscopic point of view. Why not make up a few axioms and try to build a consistent model. Allow the model to readjust the axioms a little and possibly even allow a little well thought out paradox. Always trying to match actual observations too.

I know this is what many of us try to do but in this area we seem to have trouble because we all have a point of view and can start to look like ATMers.

My cards on the table; I prefer the wavepacket idea, just that some of these packets exhibit gravity and pauli repulsion etc

StupendousMan
2007-Mar-13, 12:01 AM
Has there ever been a duality test?

I was thinking create one or two slits but in each have something in the middle like a convex lens focused at the exit or thereabout.

If it worked you would see if the particle could enter in whatever form be required to act as a wave then appear to exit as a particle.

Is this possible?

It sounds like you are asking about the "delayed choice" experiment.

Lecture notes from my course on Modern Physics (http://spiff.rit.edu/classes/phys314/lectures/dual4/dual4.html)

Ken G
2007-Mar-13, 02:44 AM
My cards on the table; I prefer the wavepacket idea, just that some of these packets exhibit gravity and pauli repulsion etc

I'm not sure I understand, the "wavepacket idea" is the only viable one, it's the only way to get agreement with observation. It isn't an interpretation of quantum mechanics, it is quantum mechanics, for realizable systems.

Ken G
2007-Mar-13, 03:19 AM
It sounds like you are asking about the "delayed choice" experiment.

Lecture notes from my course on Modern Physics (http://spiff.rit.edu/classes/phys314/lectures/dual4/dual4.html)

I'm afraid I don't understand-- the data seems to show that you get an interference pattern whether you make the delayed choice or not. It is as though the Pockell's cell is simply not working, what are we to conclude about the wave/particle duality from this experiment?

sirius0
2007-Mar-13, 06:01 AM
I'm not sure I understand, the "wavepacket idea" is the only viable one, it's the only way to get agreement with observation. It isn't an interpretation of quantum mechanics, it is quantum mechanics, for realizable systems.

So we agree. Why is duality an issue?

Or is that actually the question behind this thread "Why is duality an issue?"

Ken G
2007-Mar-13, 06:45 AM
Duality is often treated as a deep mystery, that energy quanta can have wave and particle properties at once. The resolution is that they move like waves, but arrive like particles. Real waves are always wavepackets, and the uncertainty principle is a result of that. But the Wheeler delayed choice experiment is the closest to something that actually seems weird of paradoxical, I'm trying to understand that.

Peter Wilson
2007-Mar-13, 05:27 PM
I'm afraid I don't understand-- the data seems to show that you get an interference pattern whether you make the delayed choice or not...

No, it is shadow or interference.

If it is possible to know which way the photon went, it leaves a shadow (particle-like); if it is not possible to know which way the photon went, it leaves an interference pattern (wave-like).

Either way, the detection is always a quantum event, but the pattern of detection produces either shadow or interference pattern.

Ken G
2007-Mar-13, 08:50 PM
No, it is shadow or interference.
But the data is identical either way, that's what I don't understand here.

Grey
2007-Mar-13, 09:33 PM
But the data is identical either way, that's what I don't understand here.This doesn't look like the usual delayed choice experiment. As far as I can tell, the upper path was always opened, the only question is whether it was opened ahead of time, or only after the pulse had a chance to travel down a certain path. But it's not surprising that this wouldn't change the results. It's never valid to make an assumption about where a photon is in between interactions with it. The photon can just as easily take both paths whether you open the upper path early or late.

Ken G
2007-Mar-13, 09:44 PM
I agree this doesn't seem like the normal situation in "delayed choice", and I considered the possibility that you couldn't tell where the photon was, but I decided that a wavepacket could be used for the photon with plenty of peaks and valleys with which to get interference, yet a well-limited uncertainty in the photon's location and time of emission. With all that in place, it does seem that you could close the upper path after the photon leaves the laser but before it gets to the Pockell's cell, not that it should be any different then closing it yesterday. I just don't see why they get interference, it's as though they are closing the Pockell's cell later than they think, or they are mistaken about the uncertainty in when the photon is really being emitted. I guess this last possibility is what you are suggesting, and I can't rule it out-- it just seems like a botched experiment, rather than a display of some weird photon properties!

Grey
2007-Mar-13, 10:17 PM
I...it just seems like a botched experiment, rather than a display of some weird photon properties!I agree. Maybe there's more data here that isn't presented, but this doesn't have all the (usually really weird!) hallmarks of a delay-choice quantum experiment.

Ken G
2007-Mar-13, 10:19 PM
We need StupendousMan to tell us how he uses this in his lectures on delayed choice. Are there better examples of this type of experiment, perhaps that Wheeler was referring to in his rather scientifically thin essay on the topic?

StupendousMan
2007-Mar-13, 10:51 PM
We need StupendousMan to tell us how he uses this in his lectures on delayed choice. Are there better examples of this type of experiment, perhaps that Wheeler was referring to in his rather scientifically thin essay on the topic?

(I've edited my original posting to remove an error I made -- sorry!)

You can find a longer and better description of the experiment in an article in Scientific American, "The Reality of the Quantum World", by Abner Shimony; see volume 258, pp 46-53 (1988).

I guess my little web page works better when I project it on the screen in front of the class and discuss it with my students (no surprise there). Perhaps the written words fail to get the point across.

The point of the experiment is to let a single photon enter the apparatus, wait for it to move a certain distance -- far enough to go past the first beamsplitter, and thus have to "choose" to be a wave (and enter both arms) or a particle (and enter only one arm) -- and THEN to modify the Pockels cell. One arranges things carefully so that the photon has already

a) encountered either a system with two open legs, or just a single open leg

So, if you believe that the photon has to make a decision at some point, "Am I a wave or a particle? Should I produce an interference pattern, or not?" then you have forced the photon to make this decision ... and THEN you modify the apparatus. You modify the apparatus so that the photon was asking the wrong question when it made the decision, so that (if the hidden variables idea were correct, and the photon really did have to make a decision) it would be too late for the photon to change its mind.

But what happens in practice is that, even though one leg of the apparatus was closed when the wave/photon encountered it, the wave/photon acts just as if the leg had been opened the whole time.

If the results of the experiment still don't puzzle you, then I would guess that you are either really, really, REALLY understanding the properties of light (much better than I every will), or I haven't explained it well enough.

Ken G
2007-Mar-13, 11:43 PM
My problem with that explanation is that I don't believe the experiment, but it really depends quite a bit on what the photon's wave function is, i.e., the coherence length of the wave packet (which hasn't been specified and needs to be). If you are saying you wait until the front of the wavepacket encounters the Pockell's cell, and then you open it, but the photon wave function is a very long train that takes much longer to pass through, then the result isn't surprising. If the photon wave function is a tight packet that has completely passed the Pockell's cell, then opening it can't do anything, it's too late. It's just Shroedinger's equation and quantum mechanics, it sounds like something is not being explained fully here.

StupendousMan
2007-Mar-13, 11:47 PM
If the photon wave function is a tight packet that has completely passed the Pockell's cell,


(edited)



then opening it can't do anything, it's too late.

You are correct. The "choice" referred to in the title was the choice the photon made when it encountered the first beamsplitter.

Ken G
2007-Mar-14, 02:39 AM
I'm afraid I'm not going to believe that until I hear the details of the timing and how the photon wave packet is prepared. Quantum mechanics has odd elements, like quantum entanglement, but they are all a part of standard quantum mechanics. What you are describing sounds like it would require an entirely new theory to handle. For example, how long can you wait and still get the interference?

I note that in both Wheeler's description and on your lecture notes, all that is said is that the photon "has entered the apparatus", or "has entered the optical fiber". But now you are saying something very different, that it has "passed the Pockell cell". That makes all the difference, because the "photon" doesn't do anything until you measure it-- what "does something" in the conceptualization of quantum mechanics is the photon wave amplitude. That's a lot different from the photon, because it receives contribution from all kinds of impossible sounding things, but it is necessary to sum over everything. So until the wave amplitude of the wave packet "reaches" the Pockell cell, not the fiber optic cable, the "photon" still hasn't "decided" if there are two paths or one, so there's still the potential for interference, despite the implication that it has to decide when it reaches the first beam splitter. The wave function is time dependent, that's all, it's still normal quantum mechanics. If you can really wait until after the majority of the wavepacket has passed the actual location of the Pockell cell, and then open it and get interference, then I say there is something screwy with the experiment itself, like the coherence length of the photon or the timing of the Pockell cell.

sirius0
2007-Mar-15, 12:39 AM
But the Wheeler delayed choice experiment is the closest to something that actually seems weird of paradoxical, I'm trying to understand that.

This is why I like this forum; I learn so much. I hadn't heard of the Wheeler experiment and have looked it up. Fascinating!

Some thoughts I am having now in a desperate attempt to fix the Universe as it has obviously broken :D.

Warning these thoughts are forming as I speak and are likely fairly incorrect!

Perhaps this is a SR issue? If I say do a thought experiment where I am the photon I have a null time for my travel so what does the delay part mean to me? We in our frame see SR as shifting or dilating time between a cause and it's effect. Perhaps for light it is space that gets dilated. You Ken have posted before relating distance to time. Oh it is so weird what my mind is doing now! I need to think some more. But really the final measuremnet is simultaneous to the double slit for the light so the difficulty is working out how this makes sense for us!

Ken G
2007-Mar-15, 02:37 AM
I think that's an interesting insight, but note that the key issue here is not how much time something takes, it is a concept of an order of events, a sequence. Even though all time exists "at once" for a photon, it still encounters your eyelid before it encounters your optic nerve. You should be able to look at it in any frame and get the same result. I do think it's interesting to note that from the point of view of a massless particles, all its causes and effects happen in the same place and at the same time (because, in one sense, a massless particle "doesn't really exist", it emerges as a result of this weird concept of spatial distance that we have invented, which you alluded to), but I don't think that can resolve the delayed choice issue. I think the mistake in this "paradox" is thinking the key order of events involves the state of the Pockell cell when the photon encounters the first beam splitter, when actually it is the state when it encounters the Pockell cell itself!

StupendousMan
2007-Mar-15, 11:18 AM
.... but I don't think that can resolve the delayed choice issue. I think the mistake in this "paradox" is thinking the key order of events involves the state of the Pockell cell when the photon encounters the first beam splitter, when actually it is the state when it encounters the Pockell cell itself!

(edited to remove my error in description of the experiment)

It sounds to me as if you think the "delayed choice" experiment is easy to explain, hence your use of quotation marks around "paradox" above. Perhaps you are right.

On the other hand, the mainstream physics community thinks that this experiment does involve phenomena which defy common sense and even not-so-common sense.

As before, the problem may be that my notes don't explain the experiment properly. The device is designed so that the single photon which enters the device will pass through the first beamsplitter and THEN the cell changes its state -- that's the whole point of the experiment. The photon must "decide" whether to split into two waves which can interfere when it encounters the beamsplitter, which would yield one result, or remain as a single photon when it encounters the beamsplitter, which would yield a different result. After the photon has "made the decision", the Pockell cell changes state.

Ken G
2007-Mar-15, 12:26 PM
It sounds to me as if you think the "delayed choice" experiment is easy to explain, hence your use of quotation marks around "paradox" above. Perhaps you are right.
It all depends on whether I'm right about the way the timing has to work. If I am, it is easy to explain-- you just follow the time evolution of the wave function. That's just pure Shroedinger equation, it's nothing odd at all. What's odd is that the location of the photon is not resolved until you measure it-- before that all you have is a wave function. But that's just the oddness of quantum mechanics, it's not a new oddness. I think the experiment is a nice way to underscore the way you have to think about quantum mechanics, and it really brings out that oddness, but personally I find quantum entanglement to be more bizarre because that isn't just Shroedinger's equation, you have to know what to add manually.


On the other hand, the mainstream physics community thinks that this experiment does involve phenomena which defy common sense and even not-so-common sense. But the point of this thread is that it is the "mainstream physics community" that promotes some almost quasi-mystical views about quantum mechanics (the "sometimes a particle, sometimes a wave" business), when it's all just the way you use a wave function. That's why this experiment is a great example of just this. It's interesting, I'm glad you brought it up, but again it's an example of how one must, and must not, think about quantum mechanical wave functions.


As before, the problem may be that my notes don't explain the experiment properly. The device is designed so that the single photon which enters the device will pass through the Pockell cell and THEN the cell changes its state -- that's the whole point of the experiment.

This is the key issue. But I note that both your notes and Wheeler's essay only talk about the photon passing the beam splitter that "decides" if the photon will go to the Pockell cell-- not the Pockell cell itself. That seems right to me, because in fact the beam splitter does not decide this, it doesn't break the photon superposition because it doesn't make a measurement, it only splits the wave function.

Grey
2007-Mar-15, 01:54 PM
What's odd is that the location of the photon is not resolved until you measure it-- before that all you have is a wave function. But that's just the oddness of quantum mechanics, it's not a new oddness.I'd say that this is the crux of all delayed choice type experiments. It's natural to assume that a photon (or other particle) can be treated as though it has a trajectory and follows a path, so that you can work out when it passes some element in your apparatus, and then you try to do something to change it. But Ken's got the righ tidea. Unless you actually measure where that photon is, it's not valid to say "well, we knew it was over here a millisecond ago, so it must be a thousandth of a light second that way right now". Quantum particles don't have well-defined definite positions unless you're measuring the positions. It is weird, but once you've accepted that, it doesn't seem so bad.


I think the experiment is a nice way to underscore the way you have to think about quantum mechanics, and it really brings out that oddness, but personally I find quantum entanglement to be more bizarre because that isn't just Shroedinger's equation, you have to know what to add manually.This isn't true, though. All the weird entanglement results come right out of using standard quantum wave mechanics to handle the situation. The fact that you have to describe interacting particles using a single wave function to get the right results isn't anything unique to EPR-type situations, that's really true of any particle interactions. Sometimes you can avoid it, but that's just that in many physical situations you can use a simplification of the full treatment and still get the right result. And really, it's largely the same issue, just in a different way. It's natural to imagine that, because the two particles have travelled a great distance apart, they should not be able to affect each other anymore. But that's making the same mistake that, prior to measuring them, the particles have definite positions. The wave function describing them still shows them to be connected.


But the point of this thread is that it is the "mainstream physics community" that promotes some almost quasi-mystical views about quantum mechanics (the "sometimes a particle, sometimes a wave" business), when it's all just the way you use a wave function.I think it's mostly that a typical layperson doesn't really understand just how strange things are, and physicists are trying to make it clear. For example, it's really common for people who know just a little quantum theory to try to come up with ways around the uncertainty principle, thinking that it's just a problem of measurement. Or imagining that particles really have a definite position, we just don't know what it is.


This is the key issue. But I note that both your notes and Wheeler's essay only talk about the photon passing the beam splitter that "decides" if the photon will go to the Pockell cell-- not the Pockell cell itself. That seems right to me, because in fact the beam splitter does not decide this, it doesn't break the photon superposition because it doesn't make a measurement, it only splits the wave function.Actually, from what I've seen of other experiments of this type, I don't think it would help if you change the state of the Pockell cell after the photon has "passed" it, either. It's easy to see why: that would still be assuming that the photon has a classical trajectory and a definite position.

Ken G
2007-Mar-15, 05:22 PM
This isn't true, though. All the weird entanglement results come right out of using standard quantum wave mechanics to handle the situation. Note I didn't say it wasn't "standard quantum mechanics", I said it wasn't the Shroedinger equation. And it's not-- quantum mechanics is not just Shroedinger, there is a "manual" component, which has to do with the role of measurement. That's where the weirdness in both quantum entanglement and delayed choice comes in, in a very analogous way, but the quantum entanglement seems even more bizarre to me, in terms of what that "manual" component requires you to do.


The fact that you have to describe interacting particles using a single wave function to get the right results isn't anything unique to EPR-type situations, that's really true of any particle interactions.You're right, but I wasn't just talking about the need for a single wave function, I was talking about the manual component of how you have to alter that wave function in response to measurements. There's no Shroedinger equation for that.


Actually, from what I've seen of other experiments of this type, I don't think it would help if you change the state of the Pockell cell after the photon has "passed" it, either. It's easy to see why: that would still be assuming that the photon has a classical trajectory and a definite position.No it wouldn't, it would only be assuming that the photon has a wave function that is a function of space and time, which it is. Once that wave function has passed the Pockell cell, such that you could no longer observe the quantum as being in the Pockell cell, then nothing you do to the Pockell cell should matter a whit. That would indeed be odd-- it would mean that quantum mechanics, and even the most basic concepts of time and cause and effect, would be wrong, and I'm still very skeptical that this is true.

Grey
2007-Mar-15, 05:40 PM
You're right, but I wasn't just talking about the need for a single wave function, I was talking about the manual component of how you have to alter that wave function in response to measurements. There's no Shroedinger equation for that.But you don't. That's my point. Calculate the wave function for an entangled pair of particles. Use that to determine the results of any potential measurements you might do. Look at the probabilities that those measurements will be correlated. Then do the experiment and find out whether it works. You don't have to modify the wave function based on which measurement you do first or anything like that.


Once that wave function has passed the Pockell cell, such that you could no longer observe the quantum as being in the Pockell cell, then nothing you do to the Pockell cell should matter a whit. That would indeed be odd-- it would mean that quantum mechanics, and even the most basic concepts of time and cause and effect, would be wrong, and I'm still very skeptical that this is true.Then you don't get it after all, because that's exactly the sort of thing that happens. If you constrain or measure the photon so that you know it's passed the Pockell cell (because you actually checked), then you can calculate how it will behave without taking it into account. But if you don't actually measure it, and you try to work out the results by assuming that it has already passed it in the way a classical particle would, then you get the wrong result. I'll look and see if I can find a better example.

Ken G
2007-Mar-15, 07:22 PM
You don't have to modify the wave function based on which measurement you do first or anything like that.I'm not sure what you mean by "have to". There are many situations where you have a two-particle wave function, and you make a measurement on one particle, and it changes the wave function of the other particle, even though there's no equation to describe that process-- it's all "manual". That's just EPR, is it not?


Then you don't get it after all, because that's exactly the sort of thing that happens. If you constrain or measure the photon so that you know it's passed the Pockell cell (because you actually checked), then you can calculate how it will behave without taking it into account. But you are overlooking the meaning of a wave function. If the photons are prepared with a certain coherence length (and they are), then you do know something about where the photon can and cannot be, without doing any measurements (other than the one that generated the photon state). No state that the Pockell cell is in at a time when the photon has a vanishingly small wavefunction in that cell can possibly affect anything, or else the correspondence principle and macroscopic cause and effect would break down-- literally anything would be possible. The photon wavefunction effectively exists over a finite region of space and time, that's certainly a central part of standard quantum mechanics.

I'll look and see if I can find a better example.
That would help a lot. Remember, the timing, and the coherence length of the photon wavefunction, are crucial issues here.

Peter Wilson
2007-Mar-15, 08:02 PM
...No state that the Pockell cell is in at a time when the photon has a vanishingly small wavefunction in that cell can possibly affect anything, or else the correspondence principle and macroscopic cause and effect would break down-- literally anything would be possible...

True if there are hidden variables...

The whole point of the delayed choice experiment is to show that there are no hidden variables.

Your refusing to accept/believe the experiment's results simply means you refuse to believe in indeterminism; the delayed choice experiment shows that at the quantum level, there are no "hidden variables," i.e. matter behaves indeterminately.

sirius0
2007-Mar-15, 08:39 PM
"Pockell cell" Loads of refferences on the net but I realise i am just guessing. What is it? I don't have a mention of one in the Wheeler experiment that I have.

Thinking a bit further on this wheeler thing. In SR if we have a cause and effect the two events can become simultaeneous in one frame while the other framed observer might see a substantial period of time between the two.

For light it has null time for it's trip. but what if there is a SR effect? Apparently the timelike dimension swaps with a spatial one for matter that has passed the event horizon of a black hole. Perhaps in the case of light this has already happened in our space-time. This might mean that SR style simultanaeity might occur spatially. What i am suggesting is that the photon, from its perpective, sees all real paths as one; they have become spatially simultaeneous. Therefore it only sees one slit. Just a thought and a pretty weird one.

Grey
2007-Mar-15, 10:09 PM
I'm not sure what you mean by "have to". There are many situations where you have a two-particle wave function, and you make a measurement on one particle, and it changes the wave function of the other particle, even though there's no equation to describe that process-- it's all "manual". That's just EPR, is it not?No. You can determine the expected results of the experiment at either end of an EPR type experiment without knowing whether or not the other measurement has been made. The wavefunction is unchanged.

I'll respond to the other half of your message when I get a chance to look up some other delayed choice experiments.

Ken G
2007-Mar-16, 01:44 AM
The whole point of the delayed choice experiment is to show that there are no hidden variables.I'm not invoking any hidden variables, just garden variety wavefunctions.

Ken G
2007-Mar-16, 01:47 AM
No. You can determine the expected results of the experiment at either end of an EPR type experiment without knowing whether or not the other measurement has been made. The wavefunction is unchanged.

Let me put it this way. Let's say you collide positrons and electrons at a given point, one pair at a time. You set up a photomultiplier tube in one direction, and place a mirror on the opposite side of the collision point. Eventually, you detect a photon from the collision-- is there or is there not another one coming soon? If there is, how is that not one measurement changing the wave function of the other? And if it is, how can you get this from the Shroedinger equation? You have to apply it manually.

sirius0
2007-Mar-16, 05:37 AM
It sounds like you are asking about the "delayed choice" experiment.

Lecture notes from my course on Modern Physics (http://spiff.rit.edu/classes/phys314/lectures/dual4/dual4.html)

Ahh Pockell cell Why didn't I follow your link earlier; thanks!

Ken G
2007-Mar-16, 04:10 PM
It is now clear to me that the problem with this experiment is that there is no quantum entanglement in it. That's when things get interesting, but it's only because entangled wave functions segment their information in subtle ways. This experiment only involves a single particle wave function, in the manner I've described. It gets a lot trickier with entangled particles, but even then "delayed choice" issues are not cause-and-effect issues, they are merely matters of data segmentation-- where data is not randomly sampled when it naively seems to be (in other words, any complete dataset that does not show interference is comprised of subsets that do, if a way to properly sample them can be found-- that's all you can do "after the fact".)

(edited for clarity.)

StupendousMan
2007-Mar-16, 06:45 PM
It is now clear to me that the problem with this experiment is that there is no quantum entanglement in it.

I guess it's a "problem" if you are hoping to find an experiment which involves quantum entanglement. On the other hand, if you are hoping to find an experiment that illustrates and tests one's ability to understand the behavior of light, I think this is a good one.


This experiment only involves a single particle wave function, in the manner I've described. It gets a lot trickier with entangled particles, but even then "delayed choice" issues are not cause-and-effect issues, they are merely matters of data segmentation-- where data is not randomly sampled when it naively seems to be.

I guess we'll have to differ on this point.

Ken G
2007-Mar-16, 07:21 PM
I guess it's a "problem" if you are hoping to find an experiment which involves quantum entanglement. On the other hand, if you are hoping to find an experiment that illustrates and tests one's ability to understand the behavior of light, I think this is a good one.
I agree it's an excellent illustration, but it's just quantum mechanics. There are no suprises in it, other than that you need to use a time dependent wave function. In other words, this is a great experiment for understanding what a single particle wave function is, but there's no subtleties if you do. The only "delayed choice" elements emerge if one does not understand a wave function. In other words, it's not an example of delayed choice, it's an example of why there's not delayed choice, there's a wave amplitude. If you close the Pockell cell after the particle wave function is past the Pockell cell, you will not get an interference pattern.


I guess we'll have to differ on this point.
But do we differ on my claim above? That's the physics here, it's not a matter of opinion. Remember, there may be lurkers around who are counting on us to get our physics right, and we want to know what's right anyway.

StupendousMan
2007-Mar-16, 11:41 PM
If you close the Pockell cell after the particle wave function is past the Pockell cell, you will not get an interference pattern.


(edited to prevent future readers of this thread from following my misconceptions)

You are correct. The Pockell cell is switched when the photon has passed the first beamsplitter, but has not reached the cell yet.

Ken G
2007-Mar-17, 12:57 AM
The experiment is designed to close to Pockell cell after the particle wave function is past the Pockell cell. And it still produces an interference pattern.
No, it won't. None of the "delayed choice experiments" work that way. Even your lecture notes, and Wheeler's essay that you link to, talk about the "choice" happening after the photon encounters the beam splitter, not the Pockell cell. That makes all the difference. The point here is that we haven't established the photon coherence length, and until we know that, we can't even tell when the photon has "passed the Pockell cell". Still, I'm pretty sure it has not, given the data. (Indeed, what would be the point of the long fibre optic cable, is not to delay the arrival at the Pockell cell relative to the beam splitter?)


You see, that's why it is called the "delayed choice experiment". Because common sense would claim that the photon must make a choice when it reaches and passes the Pockell cell: am I wave, or am I a photon? The only after-the-fact delays come when you have entangled particles, and you can "choose" after the fact how to segment the data that has already been collected. But it doesn't change the data, it just segments it in interesting ways, that will either pick out a subset that shows interference or not.
Common sense would claim -- as you do -- that if you close the Pockell cell after the wave function passes the cell, then you will not get an interference pattern. But you do.It's not common sense that dictates this, it's quantum mechanics. This is kind of what this thread is about-- quantum mechanics has bizarre elements, but a lot of the most "mystical" things said about it actually stem from an incomplete understanding.


John Wheeler wrote a short essay on the puzzling nature of this experiment for the New York Times -- alas, the link on my web page to that essay is dead. Wheeler won a Nobel Prize in physics. If he finds this experiment puzzling, then it's probably a sign that the matter isn't so simple.I'm sure that Wheeler understands a time-dependent wave function. His puzzlement is either something more complex than this experiment, or else he is just saying that quantum mechanics is strange until you understand the need for a wave function rather than macro concepts.


I apologize if this message appears snippy. I agree with you that it _is_ important that we eventually get the physics right in these discussions -- and many are the times that I've been wrong! -- and it's been a little frustrating for me to keep failing to find a convincing way to explain it.
No worries, snippiness is kind of a mode of debate really, and debate is the engine that fuels this whole place. I come off snippy many times, but at the end of the day, I just want everyone, including myself, to discover the truth!

StupendousMan
2007-Mar-17, 03:45 AM
I went back and read the article by the group who did the experiment, back in 1987. "Delayed-choice experiments in quantum interference", by Hellmuth, Walther, Zajonc and Schleich, Phys Rev A, 35, 2532 (1987). I should have done that at the start of this discussion. It's been over a year since I taught the Modern Physics course, and -- obviously -- that's too long for me to retain important concepts.

The experiment does work as Ken had been stating: there's an interferometer with two legs. A photon/wave approaches the first beamsplitter, which will direct it down one or both legs of the interferometer, then reaches a second beamsplitter which might combine two waves (but not one photon) and finally a pair of detectors.

Now, if both legs of the interferometer are open, the photon/wave hits the first beamsplitter, goes down _both_ legs, strikes the second beamsplitter, and interferes with itself. Under these circumstances, one can predict the signals which should be produced by the detectors.

On the other hand, if one closes the upper leg of the interferometer, by placing a brick or other opaque object in it, then the wave/photon will not split at the first beamsplitter; instead, it will act like a photon, and enter one leg only. If it chooses the upper leg, it hits the brick -- no signal from the detectors. If it chooses the lower leg, it goes to the second beamsplitter, and then to one of the detectors.

The statistical properties of the signals from the two detectors are quite different in the two cases. If you gather a bunch of data, you can figure out what happened: did the photon choose one leg only, or both legs at once?

Now, here's the important point which I screwed up earlier in this thread. Physicists placed a "Pockell cell" in the upper leg. This is simply a device which can change its polarization in a VERY short time -- about 4 ns. It allows one to open or close the upper leg to light quickly.

The experiment consists of (in simple terms) allowing a single photon to enter the apparatus, and go through the first beamsplitter, with the Pockell cell CLOSED. One might expect the photon/wave to choose to act as a photon: it will go down one leg only, and either slam into the closed cell, or reach the second interferometer from only one leg, as a photon.

However, after the photon has entered the device and passed the first beamsplitter, the Pockell cell is OPENED. This allows the photon/wave to pass through both legs, after all, and interfere with itself after it reaches the second beamsplitter. The measurements indicate that, in fact, it does.

The puzzling aspect occurs if one believes that the photon had to make a choice when it reached the first beamsplitter: act as a wave -- and go down both paths -- or act as a photon -- and go down only a single path. In this view, the photon/wave makes its choice, and then, afterwards, the upper leg is opened.

Ken G. was right -- this experiment wouldn't make sense at all if one waited until after the photon had passed the Pockell cell. I don't know why I was sticking to that claim so strongly. Sigh. In the interests of preventing future readers of this thread from gaining a similar misunderstanding, I've gone back to my own earlier posts and fixed my erroneous statements. I've indicated in each spot that I'm editing my earlier post to fix an error. I'm not trying to revise history or avoid blame for being wrong -- I just don't want someone a month from now to come across this old thread and pick up a bad idea.

So, Ken, your intuition is obviously a lot better than mine :-) This is really interesting stuff, and I'm glad you made me go back and re-read the original research.

Allow me to close by quoting from the 1987 paper.


"We therefore have no right to say what 'the photon is doing'
during its journey in the interferometer. During this time the
photon is 'a great smoky dragon'* which is only sharp at its
tail (at the beam splitter 1) and at is mouth where it bites
the detector. "

*The great smoky dragon is taken from an article by Wheeler, also published in 1978.

Ken G
2007-Mar-17, 02:00 PM
Thank you for clarifying all this-- and I hope I will respond with the same grace when you correct a misconception that has crept into my thinking on some subject; it serves as an example of what can make this place work more effectively for all.

In terms of Wheeler's "smoky dragon", I think it is a useful picture that a photon is created and destroyed like a particle, but travels like a wave, and this is the sense in which there is a duality, but it's not really a split personality (indeed, one might wish to imagine how this entire experiment could be conducted with water waves on a pond). The issue of combined wave functions of multiple particles (entanglement) came up here as well, and things get even weirder there, but again one must be on the lookout for overly mysterious claims about what happens to cause an effect.

sirius0
2007-Mar-18, 08:33 PM
In terms of Wheeler's "smoky dragon", I think it is a useful picture that a photon is created and destroyed like a particle, but travels like a wave, and this is the sense in which there is a duality, but it's not really a split personality (indeed, one might wish to imagine how this entire experiment could be conducted with water waves on a pond). The issue of combined wave functions of multiple particles (entanglement) came up here as well, and things get even weirder there, but again one must be on the lookout for overly mysterious claims about what happens to cause an effect.

I imagine a bucket (my pond) a drop of water (photon) lands in the bucket. A wave travels towards the edge of the bucket (the bucket is just on full) a drop just makes it over the edge........

I will have to think some more but that is a start.

Ken G
2007-Mar-19, 04:16 AM
Yes that's quite an interesting start at a macroscopic analogy-- when one recognizes that the water that falls off the edge of your bucket must also form a drop, and it is highly unpredictable in the idealized case just where on the edge that drop will form!

sirius0
2007-Mar-19, 05:00 AM
Yes that's quite an interesting start at a macroscopic analogy-- when one recognizes that the water that falls off the edge of your bucket must also form a drop, and it is highly unpredictable in the idealized case just where on the edge that drop will form!

Exactly; there will be either a wave pattern left around the flat of the rim or a drop will make it over. Not both if things are set up well.

Grey
2007-Mar-19, 01:11 PM
Let me put it this way. Let's say you collide positrons and electrons at a given point, one pair at a time. You set up a photomultiplier tube in one direction, and place a mirror on the opposite side of the collision point. Eventually, you detect a photon from the collision-- is there or is there not another one coming soon? If there is, how is that not one measurement changing the wave function of the other? And if it is, how can you get this from the Shroedinger equation? You have to apply it manually.This example misses the point of the "paradoxes" of entanglement. That is, I can make this work by simply saying that the "trajectory" of the photons (or, if you prefer, a wavefunction describing them) was fixed when the electron and positron collided; I just didn't know which way they were headed. There are plenty of circumstances under which my making a measurement here instantly tells me something about what the results of a hypothetical measurement there would have to be (where here and there are arbitrarily far apart), but most of those cases do not require postulating any superluminal influence to handle them. I can easily explain the correlation between measurements here while sticking to purely subluminal influence.

The cases where I run into problems are where I'm performing different measurements on, say, the spin of two correlated photons, perhaps trying to sneakily find out more about a photon than quantum mechanics says I can know. When I do that, I find out that I cannot simply suggest that the two photons had some value for spin when the collision happened; I can only explain the apparent correlation by positing some influence between the two.

Grey
2007-Mar-19, 01:15 PM
True if there are hidden variables...

The whole point of the delayed choice experiment is to show that there are no hidden variables.It doesn't work, though. The proof that there are no hidden variables relies on the assumption that influence is always subluminal. But Bell's Theorem shows that is not the case. No model of quantum behavior can explain the results without postulating that events are influenced superluminally by arbitrarily distant other events. Once you've given in to that, a hidden variables model can exist. Most physicists still don't think there are any, but it's perfectly possible to construct a model where there are. It's just that the values for those variables have to be able to change in response to things that "shouldn't" be able to affect them.

Ken G
2007-Mar-19, 02:03 PM
This example misses the point of the "paradoxes" of entanglement. That is, I can make this work by simply saying that the "trajectory" of the photons (or, if you prefer, a wavefunction describing them) was fixed when the electron and positron collided; I just didn't know which way they were headed. True, but you and I both know that would be the wrong wavefunction, prior to the detection of one.
The cases where I run into problems are where I'm performing different measurements on, say, the spin of two correlated photons, perhaps trying to sneakily find out more about a photon than quantum mechanics says I can know. When I do that, I find out that I cannot simply suggest that the two photons had some value for spin when the collision happened; I can only explain the apparent correlation by positing some influence between the two.

Not an influence exactly, just a combined wave function. But the issue here is, the experiment described in this thead (which we have agreed is flawed) does not have a combined wave function, there's only one particle. But this means, it does matter when the Pockell cell is closed-- you can't get interference by closing it after the wave function passes. I agree that entanglement is a lot more subtle, but it still has no causality violations-- it's just an interesting way that existing data is segmented that you don't notice until you try to extract more information than you are allowed, as you say.

I'm basically making two points about wave functions here. First, the standard rules of cause and effect don't apply to wave functions, they apply only to the measurables that wave functions allow us to predict. Thus there are separate rules to the way wave functions work, but once one understands them, there is never any cause-and-effect paradoxes. The second point is that not everthing that happens to wave functions is describable by the Shroedinger equation-- there also has to be included a "manual" component. One needs to know the rules of applying that manual part when dealing with measurement and information, like quantum entanglement (though there are far more everyday but still fascinating examples, such as the experiment in this thread). Have we agreed now that one does need to know when the time-dependent wave function passes the Pockell cell, and this is crucial to the timing of the closing of the cell, in order to get the interference observed in the experiment?