What is the size and shape of single optical photon?

[exchemist]

On 4/30/2021 at 7:19 PM, SergUpstart said:

Imagine that a photon hits a semi-transparent mirror. What, it will split into two "ellipsoids"? But in the end, only one half will be absorbed, and the other half will disappear. And which specific half will be absorbed will be known only at the time of absorption.

Eh? No. A photon is not split in two by a semi-transparent mirror. It is either reflected or transmitted, with 50% probability of each outcome.  

[SergUpstart]

8 minutes ago, exchemist said:

Eh? No. A photon is not split in two by a semi-transparent mirror. It is either reflected or transmitted, with 50% probability of each outcome.  

Yes, so. But whether the photon was reflected or passed on will only become clear when it is absorbed in one of the paths. What is the shape of the photon between the mirror and the absorption point????

[exchemist]

27 minutes ago, SergUpstart said:

Yes, so. But whether the photon was reflected or passed on will only become clear when it is absorbed in one of the paths. What is the shape of the photon between the mirror and the absorption point????

Meaningless question.

Unless this "shape" has observable consequences, in the way the photon interacts, it is simply not real as far as QM is concerned. One could just as well hypothesise that there are pixies dancing round it on its way. 

[swansont]

2 hours ago, SergUpstart said:

Yes, so. But whether the photon was reflected or passed on will only become clear when it is absorbed in one of the paths. What is the shape of the photon between the mirror and the absorption point????

You could only determine this by interacting with the photon, which you don’t, as the scenario is described.

[SergUpstart]

1 hour ago, exchemist said:

Meaningless question.

Unless this "shape" has observable consequences, in the way the photon interacts, it is simply not real as far as QM is concerned. One could just as well hypothesise that there are pixies dancing round it on its way. 

The title of this topic is "What is the size and shape of a single optical photon?" This meaningless question was asked in order to show that the concept of "shape" does not apply to the photon.

[MigL]

I liked the ibsightful ( and humble ) answere given by exchemist early in the previous page ...

"My limited, chemist's understanding of QM is that you can't really speak of an "objective" EM field configuration for a single photon. If you could, it seems to me it would be a classical object rather than a QM one."

QM does not allow us to say anything about certain properties of a quantum particle, until we interact with that quantum particle, thereby changing that property.
If that were not the case we would be dealing with a classical particle.
There is a difference.

Edited May 2, 2021 by MigL

[Duda Jarek]

13 hours ago, SergUpstart said:

The title of this topic is "What is the size and shape of a single optical photon?" This meaningless question was asked in order to show that the concept of "shape" does not apply to the photon.

In quantum mechanics we have wavepackets - why shape of wavepacket from production of single photon is meaningless?

Size is e.g. mean deviation from the center of particle (E[(x-E[x])^2] in position space) for its energy density.

[John Cuthber]

If I want to measure the length of a table, the fist thing I need to do is define the edge of the table.

Does a photon have an edge?

[Duda Jarek]

As written e.g. E[(x-E[x])^2] - expected value of squared distance from the center of such wavepacket ... just any estimation of size photon, preferably experimental - e.g. minimal duration of laser pulse, weakening transmission through very narrow hole etc.

[Duda Jarek]

"The size and shape of single photon" http://dx.doi.org/10.4236/oalib.1107179 has nice looking models of photon:

The big question is how true they are???
Sadly, while many claim that physics is nearly solved, we know nearly nothing about such basic questions like EM field configuration and dynamics of photon (wavepacket) ...

[exchemist]

8 minutes ago, Duda Jarek said:

"The size and shape of single photon" http://dx.doi.org/10.4236/oalib.1107179 has nice looking models of photon:

The big question is how true they are???
Sadly, while many claim that physics is nearly solved, we know nearly nothing about such basic questions like EM field configuration and dynamics of photon (wavepacket) ...

Well by now you know my view: I suspect these are just semi-classical fictions, with no observable consequences. 

[Duda Jarek]

But quantum mechanics is built on the classical one, e.g. through the quantization procedure.

Classically we have single trajectory (or history of field) optimizing action, to take it to QM we e.g. take Feynman ensemble of such trajectories (field histories) instead (allowing to derive classical as approximation).

In QM we have e.g. wavepackets - isn't single optical photon such a wavepacket?

 

Photon is EM field - we should be able to say anything concrete about this EM field configuration/evolution - as wavepacket, or as quantum ensemble, or through distortion of position operator etc.

Otherwise it is very similar to creationism: just blindly repeating "God created"/"It is quantum" as universal answer supposed to resolve every question.

Edited May 6, 2021 by Duda Jarek

[swansont]

2 hours ago, Duda Jarek said:

Otherwise it is very similar to creationism: just blindly repeating "God created"/"It is quantum" as universal answer supposed to resolve every question.

How is insisting that it's really classical underneath not exactly the same thing?

[Duda Jarek]

The question is size in whatever view you want, like distortion of quantum position operator e.g. E[(x-E[x])^2].

I have posted at least 3 papers trying to do it from classical or semiclassical approximation of quantum mechanics, also started with and mostly ask for experimental suggestions.

Instead, as for creationists, there is no discussion, only some nonsense excesses that we cannot even ask such holy/quantum questions.

Being Dirac delta in standard Feynman diagrams: in momentum space, doesn't mean being perfect point but a plane wave: filling entire Universe.

[exchemist]

2 hours ago, Duda Jarek said:

But quantum mechanics is built on the classical one, e.g. through the quantization procedure.

Classically we have single trajectory (or history of field) optimizing action, to take it to QM we e.g. take Feynman ensemble of such trajectories (field histories) instead (allowing to derive classical as approximation).

In QM we have e.g. wavepackets - isn't single optical photon such a wavepacket?

 

Photon is EM field - we should be able to say anything concrete about this EM field configuration/evolution - as wavepacket, or as quantum ensemble, or through distortion of position operator etc.

Otherwise it is very similar to creationism: just blindly repeating "God created"/"It is quantum" as universal answer supposed to resolve every question.

Your attitude seems to betray a conviction that if only we could somehow "see behind" QM, we could restore the Newtonian world of exact, deterministic knowledge of physical systems. Einstein thought the same, so you are in good company. However, every attempt at restoring a deterministic universe, via "hidden variables" and so forth, has failed, to date. Einstein was wrong, apparently. 

Most physicists seem to think the QM picture, in which there are things that are intrinsically indeterminate (to do with Fourier transforms, non-commuting operators for the observables in question, and all that jazz) looks correct.   

As I understand it, Heisenberg's QM was deliberately built on modelling only observables - and avoiding what may or may not "go on" besides, because whatever it may be, it is not observable! So I do not think it is fair to say QM is built on classical mechanics, really. And it's not like creationism, because it is a model that works, experimentally. All theories come up against a limit at some stage, beyond which we can only shrug and say "that is just how nature seems to be. Sorry."

Edited May 6, 2021 by exchemist

[Duda Jarek]

From one side, can we directly measure processes inside a star? Rather not, but does it prevent us from searching for models of what actually happens there? Also rather not: we can build self-consistent models based on general knowledge, and tune details of what we don't know to get agreement with measurements of indirect consequences.

Similarly with microscopic physics - for which human observer is just a system of atoms governed by the same objective physics - on which we should focus on, like some properties of EM wave created by single atom deexication: optical photon.

 

From the other side, while QM might be the final description, it is built on classical mechanics: as Feynman ensemble, through quantization etc.

And we directly experience classical mechanics - emerging from the quantum one, also being its approximation - e.g. the initial one in WKB semiclassical approximation ( https://en.wikipedia.org/wiki/WKB_approximation ).

If there are technical difficulties with complete quantum description of shape/size of optical photon, a natural approach is starting with approximations: like classical, semiclassical - you are welcomed to extend to complete quantum description (or complement the set of articles).

Feynman diagrams are in another approximation: perturbative of QFT in momentum space. Their Dirac deltas in momentum space mean infinite size plane wave.

 

To model finite size in position and momentum space there are e.g. used wavepackets ( https://en.wikipedia.org/wiki/Wave_packet ) : of bounded product of widths in position and momentum space in Heisenberg principle/Fourier transform.

The question of size of single photon is e.g. about such width of wavepacket in position space.

Even better would be experimental arguments I am especially asking for - like this minimal duration of ultrashort laser pulses.

[swansont]

1 hour ago, Duda Jarek said:

mostly ask for experimental suggestions.

Different experiments give different results, owing to wave-particle duality

[Duda Jarek]

4 minutes ago, swansont said:

Different experiments give different results, owing to wave-particle duality

Great, so please elaborate which experiments you are referring to, and let us try to understand/discuss differences between their evaluations.

[swansont]

2 hours ago, Duda Jarek said:

Great, so please elaborate which experiments you are referring to, and let us try to understand/discuss differences between their evaluations.

You misunderstand. I was explaining why this is a waste of effort. I wasn’t offering to waste more of my time.

You should be able to evaluate some basic QM experiments. 

 

[Duda Jarek]

You write "Different experiments give different results", so I ask for elaboration, but instead of giving any example you write further "You should be able to evaluate some basic QM experiments." - which ones?

Like creationists claiming "many examples" which cannot be explained by evolution, asking for example they give e.g. eye ... https://en.wikipedia.org/wiki/Evolution_of_the_eye

Size of photon makes sense also in QM: e.g. as width of wavepacket, as uncertainty of quantum position operator, through adding terms in WKB approximation etc.

Please be more specific.

[exchemist]

41 minutes ago, Duda Jarek said:

You write "Different experiments give different results", so I ask for elaboration, but instead of giving any example you write further "You should be able to evaluate some basic QM experiments." - which ones?

Like creationists claiming "many examples" which cannot be explained by evolution, asking for example they give e.g. eye ... https://en.wikipedia.org/wiki/Evolution_of_the_eye

Size of photon makes sense also in QM: e.g. as width of wavepacket, as uncertainty of quantum position operator, through adding terms in WKB approximation etc.

Please be more specific.

I'm not sure the "width" of a wavepacket tells you anything about the size of a photon. Surely it just tells you the shape of the probability distribution of where you may expect to detect it, doesn't it? And QM tells you that anyway, so there is no issue there.  Whatever it is you are after must be something different, or we would not be having this discussion. 

[Duda Jarek]

If you prefer, can be "width" in momentum space - as they are related with Fourier transform/Heisenberg uncertainty.

Can we tell anything concrete about any of them - e.g. for optical photon produced by deexcitation of single atom?

 

Also, how to relate such probability distribution with EM field, e.g. expected value of its energy density?

[John Cuthber]

5 hours ago, Duda Jarek said:

 

 

 

Even better would be experimental arguments I am especially asking for - like this minimal duration of ultrashort laser pulses.

Nice graph.

But the y axis is electric field strength, not "height", so you can't say that the photon has a width.

You would be saying something like the width  is "x volts per meter".

Does that help in any way?

 

[Duda Jarek]

This gif is from https://en.wikipedia.org/wiki/Wave_packet

For optical photon it indeed should have some connection with (e.g. expected value of) EM field - what exactly?

The big size question is "how many periods are inside" - the mentioned 3 papers suggest it is approximately one period in propagation direction (and some fraction in perpendicular directions).

Also, is the envelope really Gaussian as mathematically convenient, or maybe has some different e.g. tail behavior, or even compact support?

These are absolutely fundamental questions of physics ... we know nearly nothing concrete about - due to only blindly repeating "it is quantum" as an excuse for searching for answers.

[exchemist]

22 minutes ago, Duda Jarek said:

This gif is from https://en.wikipedia.org/wiki/Wave_packet

For optical photon it indeed should have some connection with (e.g. expected value of) EM field - what exactly?

The big size question is "how many periods are inside" - the mentioned 3 papers suggest it is approximately one period in propagation direction (and some fraction in perpendicular directions).

Also, is the envelope really Gaussian as mathematically convenient, or maybe has some different e.g. tail behavior, or even compact support?

These are absolutely fundamental questions of physics ... we know nearly nothing concrete about - due to only blindly repeating "it is quantum" as an excuse for searching for answers.

These are only "fundamental questions of physics" if the questions have a meaning. And they only have a meaning in physics if they predict some observable result. This is where I struggle. I cannot see what observational outcomes can be dependent on this shape issue of yours. A photon is either detected, by absorption in an atom generally, or they are not.  Isn't it?