Do Atoms Literally Behave as Waves?

[cj2222]

Do physicists say that atoms behave as waves just because they can't look at them directly so they have to use math to predict probable areas where they could be?

or do atoms literally behave as waves?

 

[tom.stoer]

I prefer to talk about 'quantum objects' as it does neither use 'particle' nor 'wave'. Depending on the experimental setup atoms can behave as particles or waves, but they "are" quantum objects.

 

[Dmitry67]

Not only atoms, but even big molecules (C60) have wave properties.

 

[alxm]

According to QM, everything has wave-like properties to some extent.
The better question is perhaps: To what extent do atoms and molecules exhibit wave-like properties?

What Dimitry67 is alluding to, is that the C60 molecule is the biggest (to date) molecule for which we've been able to do a double-slit experiment and detect an interference pattern (a typical wave-like property). So the biggest thing we've actually seen act wave-ish.

But in general, atoms behave mostly 'particle-like' (seen as a whole, their electronic properties, OTOH, are more wave-like, since that's how electrons act). The reason I say this is because atoms tend to have quite definite positions in space (which is a 'particle-like' property).
After all, it's a pretty fundamental assumption of chemistry that molecules are more-or-less geometrically stable; that the atoms stay where they, and to the extent they move, they move more "classically" than quantum-mechanically.
This is almost always true, so chemistry works. Although we knew that already.

So the rare cases where atoms don't act very classically, or particle-like, are of great interest to physics. An example of this is Bose-Einstein condensates.

 

[GreenLantern]

A very simple answer to your question of "Do electrons/atoms/light quanta (photons) behave like particles or like waves?" YES

They behave both. All of them are dualistic in nature, meaning they exhibit the whole particle-wave duality thing :)

Sometimes they behave like waves and other times like particles. I think a lot of people get hung up on this apparent wave-particle conflict when it is first introduced and it seems all well and good for a little while but then as you dive further into QM you find yourself going...holdup...wait...WHAT?! how can it be both?

Neils Bohr first stated the answer to this apparent conflict between wave-particle ideas with the principle of complementarity. These two descriptions, wave and particle, are complementary. We are never going to need BOTH descriptions at the same time to describe the same part of a phenomenon but we do need both in order to have the complete our picture of nature.

So in terms of the de Broglie electron waves verification method using a single slit experiment with an electron beam:

We do a normal single slit experiment but instead of using monochromatic light we use a narrow beam of electrons (which can be produced through similar method to the electron gun in a cathode-ray tube) and do it in a vacuum (so the electrons don't hit air particles, bouncing off them and screwing things up ;) )

We will find we get the same interference pattern as we would if we had used monochromatic light! With a strong central maximum (≈85% of the electrons) and subsequently smaller maxima off to each side of the central with minima in between.

Now, classical mechanics would run into some major problems trying interpreting this in particle terms. Like answering, why don't the electrons all follow the same path (the same path of motion they initially had, i.e. straight...!) All the electrons had the same initial motion in the beam before they went through the slit... why the change in direction/momentum once they pass through the slit?? Newtonian mechanics doesn't have (at least not that i know of! pretty sure that's why QM is here to save the day :) ) a way to explain this change in state of motion.

So:
You cannot predict where any individual electron will strike the screen from knowing its initial state of motion before it reached the slit! Therefore you cannot be 100% certain of the position AND the momentum of each particle once it passes through the slit.

The best you can do is give probabilities that most of the electrons will go here, less will go here, and even less over here etc... which then gets into the explanation of Heisenburg Uncertainty principle..

anyway, all of this craziness (from the "eyes" of Newtonian mechanics) can be easily resolved to making sense if we think of the electrons as behaving as waves in this situation. This gives plenty of reason for the interference/diffraction pattern to be present! (if you don't know why you got to go look at normal single slit experiment with light)

The particle nature of electrons is not completely lost though in this experiment. (heres where it can get confusing for some but i'll try my best to keep things strait)

If we could slow down the whole experiment and had our screen, where the interference pattern is produced, be made of a black piece of photographic paper that turns white where ever an electron hits it.

We turn on the electron beam and say (gedanken experiment here ;) ) let out only a couple electrons. They fly through the slit and hit our black paper making their little white spots where they hit. What will it look like?!? You can't say for sure. But what you do know is with a HUGE number of electrons you get your nice strong central max and weaker side maximums.

What will happen is, with the first few, say five, it will potentially seem like random scattered dots. but when you leave the beam on, in slow motion again remember, over time your familiar interference pattern will slowly be "built" by tons and tons of those white dots on your black photo paper.

This is the electron particle behavior. They strike at points on the screen.

so WOAHHHH! Complementarity! We NEED both particle behavior AND wave behavior to explain completely what is going down in the slit experiments with a beam of electrons. With only one (or the other) of the behaviors, we cannot fully explain why both of these phenomena happen:

1. The inconsistent state of motion pre- and post slit
2. Our interference pattern gets "built" in small pieces or "dots" and seemingly randomly so (its not really exactly random per-se, in theory its built probabilistically. I don't know what name to put to natures method of painting with quanta lol :) )

We know this happens because we have seen it in experiments A LOT. lots of similar diffraction experiments with electron beams have been done and they all essentially build the same principle summed in the principle of complementarity.

If i made a mistake or wasn't exactly clear please someone let me know so i can correct!

-GreenLantern

 

[GreenLantern]

I really hope you read all that because, though it is long, it should help sort out the confusion. AND i spent wayyy longer than i should have when I should really be studying for my QM final :)

-GL

 

[tom.stoer]

Good summary. I always liked this explanation.

Let me add one comment: the whole trouble is due to the fact that we are used to think in terms of either particles or waves. Your everydays experience is full of particles, but we are not used to observe quantum objects, our senses, our language etc. is not capable of dealing with them. So the problem is not the quantum object, it's the human mind.

 

[LostConjugate]

Error 17 too many characters

So what Green is saying is that yes, they literally behave as waves. The mathematics is an attempt to understand.

 

[GreenLantern]

they literally behave as particles as well!

 

[tom.stoer]

As GreenLantern explained: it depends what you want to measure / observe. If you measure position (like the electron hitting the screen) you see one single particle; if you measure a wave (like in the overall interference pattern) you see a wave. The interpretation is that the particle interefered with itself while traveling so you can't say that it always behaved as a particle, because classical particles do not interfere; so somehow it behaved as a wave as well.

They are not both particles and waves - because this would be contradictory. They appear either as particle or wave - depending on the context. It is rather hard to say what they are, but if you look at the formalism of quantum mechanics, there are neither particles in the classical sense nor waves in the classical sense. There are abstract quantum objects living in infinite dimensional Hilbert spaces. Everything else is our interpretation.

The main prolem is that we try to discribe a quantum object in terms of a language that is appropriate for the world of classical objects but not for quantum objects. If I give you

 

[zonde]

Do physicists say that atoms behave as waves just because they can't look at them directly so they have to use math to predict probable areas where they could be?

or do atoms literally behave as waves?

I would say there is a bit of problem with semantics.
Classically word wave is reserved for collective behavior of many entities i.e. you should be able to observe concurrently different phases of some periodicity.
For single entities there simply exist other words to describe periodicity like vibration, oscillation and such. So you can't meaningfully talk about waving of single entity.

So if you ask if ensemble of atoms behave as wave then - yes. If you ask if each atom separately behave as wave it's quite meaningless question (at least from perspective of "classical" semantics).

 

[tom.stoer]

They show behaviour that can only be explained using "waves". This is valid even for one single atoms.

 

[zonde]

They show behaviour that can only be explained using "waves". This is valid even for one single atoms.

Do you suggest that it is possible to make more than one concurrent observation of one single atom showing it to be in different phases of some periodicity?
Or something else?

 

[tom.stoer]

I am saying that even the observation of a single atom on the screen cannot always be explained based on classical particles. One can set up the experiment in such a way that an atom is registered at a position which is classically forbidden. It is allowed as soon as one takes interference (based on waves) into account.

That means that the wave-like behaviour is not only a statistical or collective phenomenon but has some "reality" (I know that "reality" is a dangerous wording) even for an individual atom.

 

[LostConjugate]

they literally behave as particles as well!

Agreed

 

[SpectraCat]

Do you suggest that it is possible to make more than one concurrent observation of one single atom showing it to be in different phases of some periodicity?
Or something else?

What about the double slit experiment? It has been shown repeatedly that interference patterns are observed for single entities (photons, electrons, atoms, C60 molecules, whatever) with space-like separations. Yes, the interference patterns are technically only observable after an ensemble of particles have been measured, but that is irrelevant in this case, since the particles never interact with each other.

However, if your main point is that quantum objects do not in general display the same characteristics as classical waves, then I think I would agree with that.

 

[GreenLantern]

Okay, when people say "an interference pattern is observed for a double or single slit experiment when you just fire 1 electron" I think leaves wayy to much room for possibility of misunderstanding the true idea here.

WHEN YOU FIRE ONE ELECTRON at a double slit the interference pattern is not present. Interference pattern being bright and dark lines on your screen resembling constructive and destructive interference respectively. YOU WILL GET A SINGLE SPOT (saying your screen is a piece of say photo paper that is all black and turns white where electrons contact). Now the question is WHICH SLIT DID THE ELECTRON GO THROUGH??! classical mechanics cannot say why it would go through one verses the other. And furthermore classical cannot say why the electron doesn't just travel straight through whichever slit it goes through. because any given electron won't necessarily go through the slit and continue on straight to the screen. Any given electron is more likely to hit the spots where you (eventually) see constructive spots in your interference pattern (and even more likely still to strike along the central max than say the first order max)

Your interference pattern is slowly (okay not really slowly but hopefully you know what i mean) built up by little spots over time. This time interval that i am calling "over time," at the normal speed (the speed with which you experience is all in front of your eyes) seems instant. BAM! you have pretty bright and dark lines. but if you slowed time wayyyyyyy down and zoomed wayyyy in and started at the instant you flick on the light or electron beam emitter, you would see the pattern slowly take form by random speckles on your screen. Speckles that "appear" most frequently at your central max position, less frequently larger order maxima, and very infrequently in the minima regions.

 

[zonde]

What about the double slit experiment? It has been shown repeatedly that interference patterns are observed for single entities (photons, electrons, atoms, C60 molecules, whatever) with space-like separations. Yes, the interference patterns are technically only observable after an ensemble of particles have been measured, but that is irrelevant in this case, since the particles never interact with each other.

About double-slit.
We make double-slit experiment with photons. We can hardly observe anything about single photon. So it's logical that we make some assumption and compose interpretation about this double-slit with photons.
Next we make double-slit experiment with electrons. They are much more solid then photons but still we have only vague observations of them and it's still elementary particle. Fine we apply the same interpretation as for photons using the same assumptions.

Now we go further and further until we have double-slit experiment with buckyballs. And here I say full stop for this "photon" interpretation and assumptions that are used for this interpretation. Buckyball is nowhere near to vagueness of electrons and photons. It consists of complex structure of 60 carbon atoms that in turn have complex structure of 6 electrons and ~12 nucleons that again in turn consist of 3 quarks. So it has 4 levels of structure (quarks-nucleons-nucleus-electron shells-molecule) consisting of different elementary particles with different charges and mass. And buckyballs are observable quite fine.
We have entire (and incredibly successful) discipline - chemistry that relays on existence of exact molecular structure and in turn our own DNA is encoded in exact molecular structure of amino acids. DNA does not exists in superposition of being DNA for cat and being DNA for dog.

So I say sorry but it is other way around. It's not that experiments with buckyballs demonstrate that the same picture of photons is applicable to buckyballs but it's that experiments with buckyballs cast very serious doubt on correctness of interpretation for electrons and photons.

 

[alxm]

We have entire (and incredibly successful) discipline - chemistry that relays on existence of exact molecular structure and in turn our own DNA is encoded in exact molecular structure of amino acids. DNA does not exists in superposition of being DNA for cat and being DNA for dog.

So I say sorry but it is other way around. It's not that experiments with buckyballs demonstrate that the same picture of photons is applicable to buckyballs but it's that experiments with buckyballs cast very serious doubt on correctness of interpretation for electrons and photons.

That's just wrong. There is no conflict here.
Atoms and molecules as a whole behave mostly classically (as I already said in my first post in this thread) and you're correct that to the extent that they do behave non-classically, it's chemically insignificant (except for some rare circumstances for the lightest atoms). That does not mean the non-classical behavior can't be measured. In fact, there's a whole horde of atomic/molecular properties which are chemically insignificant (e.g. nuclear spin states).

The answer to why atoms and molecules do not exist in a structural superposition was answered by Hund right at the start of quantum mechanics (late 20's). There's nothing stopping such a superposition per se, it's just that their mass means it takes far too long for an atom/molecule to evolve into such a superposition. And more recently they've been able to show that the decoherence times are likely far too short.

There's nothing weird going on here, it's exactly as theory predicts.

 

[zonde]

That's just wrong. There is no conflict here.
Atoms and molecules as a whole behave mostly classically (as I already said in my first post in this thread) and you're correct that to the extent that they do behave non-classically, it's chemically insignificant (except for some rare circumstances for the lightest atoms). That does not mean the non-classical behavior can't be measured. In fact, there's a whole horde of atomic/molecular properties which are chemically insignificant (e.g. nuclear spin states).

The answer to why atoms and molecules do not exist in a structural superposition was answered by Hund right at the start of quantum mechanics (late 20's). There's nothing stopping such a superposition per se, it's just that their mass means it takes far too long for an atom/molecule to evolve into such a superposition. And more recently they've been able to show that the decoherence times are likely far too short.

There's nothing weird going on here, it's exactly as theory predicts.

Hmm, then please tell me if I understood you correctly.
If some chemist takes bunch of his favorite molecules fires them at double-slit and observes something like interference pattern behind double-slit and if he says: "Hey, it looks like these molecules are not taking straight paths behind double-slit." then he is dead wrong because he is sticking his nose in non-classical domain and observing non-classical behavior.
Did I understood you right?

 

[tom.stoer]

I would like to repeat what I said acouple of days ago:

Even the observation of a single atom on the screen cannot always be explained based on classical particles. One can set up the experiment in such a way that an atom is registered at a position which is classically forbidden. It is allowed only if one takes interference into account.

That means that the wave-like behaviour is not only a statistical or collective phenomenon but has some "reality" (I know that "reality" is a dangerous wording) even for an individual atom.

 

[alxm]

Did I understood you right?

No. You understood me dramatically wrong.

What I'm saying is that chemists, doing chemistry (almost) never need to take into account non-classical behavior of molecules/nuclei etc. The effects are not significant to chemistry. The De Broglie wavelength of a proton is on the order of 1/10,000 Å, chemical bonds are on the order of 1 Å. For all chemical purposes, nuclei can be modeled as classical particles, and are - even when doing quantum chemistry.

I don't see how you could possibly interpret that as chemists somehow denying the fact that nuclei and atoms as a whole, can behave quantum-mechanically. I mentioned nuclear spin as an example of a quantum property of nuclei which is chemically insignificant. - Which is not the same thing as being insignificant to chemistry! It's the basis for NMR spectroscopy (and MRI) which they use all the time, and whose usefulness is derived precisely from the fact that these interactions don't change the chemistry one bit.

 

[zonde]

No. You understood me dramatically wrong.

What I'm saying is that chemists, doing chemistry (almost) never need to take into account non-classical behavior of molecules/nuclei etc. The effects are not significant to chemistry. The De Broglie wavelength of a proton is on the order of 1/10,000 Å, chemical bonds are on the order of 1 Å. For all chemical purposes, nuclei can be modeled as classical particles, and are - even when doing quantum chemistry.

I don't see how you could possibly interpret that as chemists somehow denying the fact that nuclei and atoms as a whole, can behave quantum-mechanically. I mentioned nuclear spin as an example of a quantum property of nuclei which is chemically insignificant. - Which is not the same thing as being insignificant to chemistry! It's the basis for NMR spectroscopy (and MRI) which they use all the time, and whose usefulness is derived precisely from the fact that these interactions don't change the chemistry one bit.

There is difference when we talk about local quantum-mechanical behavior and not-so-local quantum-mechanical behavior.
I do not what to say that chemistry is somehow in conflict with existence of spin property or De Broglie wavelength property but that these properties are local to quite classically localized particles and are better described by word "oscillation" than by word "wave".

 

[zonde]

Even the observation of a single atom on the screen cannot always be explained based on classical particles. One can set up the experiment in such a way that an atom is registered at a position which is classically forbidden. It is allowed only if one takes interference into account.

If you want to say that there is real experiment that lends support to idea about waving of single atom then please be more specific about that experiment.

 

[alxm]

I do not what to say that chemistry is somehow in conflict with existence of spin property or De Broglie wavelength property but that these properties are local to quite classically localized particles and are better described by word "oscillation" than by word "wave".

These are not classical properties nor "local" properties, whatever you're pretending to mean by that. The De Broglie wavelength is directly related to the double-slit diffraction pattern which you brought up in the first place as some kind of 'contradiction'.

You're not making any sense here whatsoever.

 

[tom.stoer]

If you want to say that there is real experiment that lends support to idea about waving of single atom then please be more specific about that experiment.

The simplest example is quantum tunneling. w/o something like a wave-like behaviour we are unable to explain how a particle can penetrate a barrier which is classically forbidden.

Another example is the double-slit experiment - but you have to start with an analysis of a single slit. With a single slit you get a nearly continuous distribution (of dots) on the screen. Now opening the second slit you reduce the probability of observing a particle on the screen at certain areas (the minima of the interference pattern) to zero. This cannot be explained w/o an intereference mechanism.

 

[zonde]

These are not classical properties nor "local" properties, whatever you're pretending to mean by that. The De Broglie wavelength is directly related to the double-slit diffraction pattern which you brought up in the first place as some kind of 'contradiction'.

You're not making any sense here whatsoever.

This is not quite true that I brought up double-slit as some kind of 'contradiction'.
It was SpectraCat who brought this up as king of contradiction between concept of localized single entities (buckyballs as one of the examples) and interference pattern on the screen.

What about the double slit experiment? It has been shown repeatedly that interference patterns are observed for single entities (photons, electrons, atoms, C60 molecules, whatever) with space-like separations.

And I intended to argue that there is no contradiction but only poor assumptions.

Yes, the interference patterns are technically only observable after an ensemble of particles have been measured, but that is irrelevant in this case, since the particles never interact with each other.

Well, particles never interact directly with each other but it is fine (from classical perspective) to assume that they can interact indirectly through environment. So I say that contradiction comes from "non-contextuality" assumption rather than from classical locality of buckyballs.

 

[zonde]

The simplest example is quantum tunneling. w/o something like a wave-like behaviour we are unable to explain how a particle can penetrate a barrier which is classically forbidden.

As you are still giving only very general example I took this description of "Quantum tunneling of Hydrogen in Niobium" as more exact example:
http://tabish.freeshell.org/physics/nboh/"
I would agree that billiard ball picture can not provide obvious way how hydrogen can penetrate a barrier. However my objection was about "waving" versus "oscillating" as pertaining to single entity.

And to me it seems that this example does not give clear reason to prefer "waving" over "oscillating". Even the opposite.
In this link certain aspect of results is explained that way: "it means that as the temperature is increased, the atom tends to sit more on one site rather than jump around!"

Tunneling happens at lower temperature when according to this explanation atoms tend to jump around more. So from perspective of "oscillating" particle it has better chance to hit other stable site and settle for oscillation between the two stable sites.
Waves on the other hand do not jump around but settle for preferred direction smoothly.

 

[tom.stoer]

Tunneling can be observed even at zero temperature and for barriers with infinite height. Therefore tunneling means that a quantum object tunnels through a barrier; the picture of a classical particle that somehow jumps over the barrier is wrong.

If you study quantum tunneling in the double-well potential which contains one single quantum particle you see that the single "particle" spreads of both minima even if it is classically confined in one well and even if the height of the barrier exceeds the particles energy. This affects e.g. the energy levels of the particle. In addition there are so-called instanton solutions describing the tunneling process through the barrier.

So again: any attempt that tries to explain these effects based on a collection of interfering particles alone falls short in explaining effects which you can observe for one single quantum particle.

Another warning: it is misleading to try to explain the behaviour of a quantum object either based on waves or based on particles; there are quantum apsects that cannot be explained by any classical picture - neither wave nor particle (nor both which immediately leads to contradictions).

 

[zonde]

Tunneling can be observed even at zero temperature and for barriers with infinite height. Therefore tunneling means that a quantum object tunnels through a barrier; the picture of a classical particle that somehow jumps over the barrier is wrong.

If you study quantum tunneling in the double-well potential which contains one single quantum particle you see that the single "particle" spreads of both minima even if it is classically confined in one well and even if the height of the barrier exceeds the particles energy. This affects e.g. the energy levels of the particle. In addition there are so-called instanton solutions describing the tunneling process through the barrier.

So again: any attempt that tries to explain these effects based on a collection of interfering particles alone falls short in explaining effects which you can observe for one single quantum particle.

Another warning: it is misleading to try to explain the behaviour of a quantum object either based on waves or based on particles; there are quantum apsects that cannot be explained by any classical picture - neither wave nor particle (nor both which immediately leads to contradictions).

... and??
How all this is related to my arguments?
If you intend to fight straw man do it without my assistance.

 

[tom.stoer]

It seems to me that you don't want to understand my arguments.

I do not care about your "wave", "oscillation", "particle", "jumping around" etc. These are just words sufficient to describe classical entities but insufficient for qm objects.

If you look at the formalism plus the experiments you will learn that you have to take into account "interference-like terms" of different states in a Hilbert spaces (or path integrals, if you like) in order to calculate (predict) the experimental result - even for single particle phenomena! And you will see that the formalism does not allow for an interpretation in purely classical terms, regardless if you call it "particle", "jumping", "oscillatory", "wave-like" or any thing else.

 

[zonde]

It seems to me that you don't want to understand my arguments.

If you set formalism above experiment and quantum weirdness is basic guideline for your thinking then it might be quite so.

I do not care about your "wave", "oscillation", "particle", "jumping around" etc. These are just words sufficient to describe classical entities but insufficient for qm objects.

If you look at the formalism plus the experiments you will learn that you have to take into account "interference-like terms" of different states in a Hilbert spaces (or path integrals, if you like) in order to calculate (predict) the experimental result - even for single particle phenomena! And you will see that the formalism does not allow for an interpretation in purely classical terms, regardless if you call it "particle", "jumping", "oscillatory", "wave-like" or any thing else.

 

[tom.stoer]

If you set formalism above experiment and quantum weirdness is basic guideline for your thinking then it might be quite so.

I do not set formalism above experiment. I look at experimental results and at the (many many) attempts to explain them based on classical wording. As all these attempts failed so far, I accept that the formalism is sufficient to predict experimental results - being aware of the fact that it is notsufficient to provide an ontological explanation.

I do not know if such an ontological explanation will ever be available. But I am absolutely sure that we will not be able to find it using classical reasoning and wording and denying "quantum weirdness".