Why the photon is disturbed by a hole?

[zrek]

Please help me to understand why the path of the photon changes when travels close to a material (for example the edge of a hole)?
I'm aware that the path of the photon can be calculated by the QM principles as a probability wave and by this the change of the wavefront follows the Huygens–Fresnel principle. Also I know that because of the uncertainity the smaller size of the hole results bigger possible change in the path.

But what is exactly the physical effect that causes the photon to change its path? Maybe the electrostatic force of the protons and electrons that builds up the edge of the hole? Or simply the spacetime curvature that is stronger because of the mass of the close atoms? If there would be a hole inside a plate made of neutrons (like the neutron stars), the photon path would be disturbed also in it?

Thank you for your help.

 

[SteamKing]

Are you talking, perhaps, about a photon passing close by a 'black' hole? Or just any old hole?

 

[berkeman]

Please help me to understand why the path of the photon changes when travels close to a material (for example the edge of a hole)?
I'm aware that the path of the photon can be calculated by the QM principles as a probability wave and by this the change of the wavefront follows the Huygens–Fresnel principle. Also I know that because of the uncertainity the smaller size of the hole results bigger possible change in the path.

But what is exactly the physical effect that causes the photon to change its path? Maybe the electrostatic force of the protons and electrons that builds up the edge of the hole? Or simply the spacetime curvature that is stronger because of the mass of the close atoms? If there would be a hole inside a plate made of neutrons (like the neutron stars), the photon path would be disturbed also in it?

Thank you for your help.

Sounds like you are asking about the physics of diffraction?

http://en.wikipedia.org/wiki/Diffraction

https://encrypted-tbn3.gstatic.com/...GvbM_0A9YHnSmzi-no0CQfuAlweQL3xTEta_Hzv0yiSyg

 

[Drakkith]

But what is exactly the physical effect that causes the photon to change its path?

I don't think the photon ever changed paths. I'm not sure you can say where the photon is in the time between emission and absorption.

 

[sophiecentaur]

This is a problem that needs to be approached from the direction of Waves and not Photons. You never have any idea of where a Photon actually went between being emitted and detected (that's one of the basics of QM). Applying the principles of Diffraction will give you the probability that the photon arrived (or will arrive) at a particular spot but, if you were actually to intercept and detect a photon on the way, it would not arrive at the target so you would have destroyed the experiment.
There is nothing 'wrong' with using the appropriate analysis for a particular situation and it makes sense to treat this sort of problem without trying to introduce photons. They just confuse the issue and get you no further, here.

 

[zrek]

Thanks to all of you for the replies.

I'm aware that we have information only about (and at) the emission and the detection. I also know that (only) the probability of the place of the detection can be calculated.
But I'm talking about the case when the photon is already arrived. By applying some filters, we can collect some informations (for example in a double-slit experiment we can find out in which slit (hole) the photon was going through, and by doing this we lose the interference pattern).
And for example also we usually say that the path of the photon is changed by the curvature of spacetime. So you can find lots of places when we are talking about the path or trajectory of the photon (even if it is only a presumption)

So if we have already collected the data and know the staring-ending point of the path and we also know that the photon had the only opportunity to go through the hole, we assume its path. Also clear that normally (without any disturbance) the path of the light is a straight line.
But in case of diffraction caused by a hole the path was obviously not a straight line, I assume that we need to have an explanation what effect caused the break in the path.

Spacetime curvature? Electric force? Other?

(Let's consider a gravity lens caused interference. In this case I know for sure that the path of the photon was disturbed by a spacetime curvature. Is this the same when the diffraction is caused by a slit in a lab? Is there any other effect that can disturb the path of the photon?)

Thank you!

 

[sophiecentaur]

Why are you distinguishing between the effect of two slits and a hole? Each slit has a finite width and will introduce its own contribution to the overall diffraction pattern. (Wave approach says it all.)
I also don[t see how you can say the photon has already arrived, yet go back and find which way it went. This is contradictory. I see no point or meaning in any discussion about the 'path taken' by a photon, in any case. There is no model, afaik, for describing the extent or location of a photon except when it is interacting with something. If the interaction is very subtle (as in refraction or relativistic bending) then the extent of the photon is over the whole refracting region and, again, is much better described in terms of waves.
I think it is very risky to look for a "what's really happening" kind of answer at this level of Physics.

 

[dauto]

Your assumption that the photon ought to follow a straight line just isn't valid. Not in Quantum Mechanics. The Path integral formulation of quantum mechanics, for instance, is founded in the postulate that all possible paths contribute equally to the propagation of the photon. Let me say that again: ALL possible paths must be taken into account. The straight paths have no (a priory) privilege. The propagation of light following a straight line (sort of), that we observe in our daily lives is obtained as a consequence of the calculation (for free space), not as a postulate, and won't be valid if obstacles are present.

 

[zrek]

Your assumption that the photon ought to follow a straight line just isn't valid. Not in Quantum Mechanics. The Path integral formulation of quantum mechanics, for instance, is founded in the postulate that all possible paths contribute equally to the propagation of the photon. Let me say that again: ALL possible paths must be taken into account. The straight paths have no (a priory) privilege. The propagation of light following a straight line (sort of), that we observe in our daily lives is obtained as a consequence of the calculation (for free space), not as a postulate, and won't be valid if obstacles are present.

I think that your state is about how to calculate the (probable) behavior (path) of the photon when it is going to be released. However if it is already arrived, we have certain (more or less) sure points of the path. The straight line comes to our calculation if we count with the all possible paths and most of that are destructed by the effect of the (self) interference. So finally the staight line will be the privileged shape.
If there are (disturbing) obstacles, only then we can calculate with other kinds of path shapes than (more or less) straight line.

However the question is not this problem, I'd like to know exactly why the obstacles changing the situation? What kind of the property of the obstacles count in this situation?

For example when a photon hits an atom, we can say that the atom is a "disturbing" obstacle because of its proper electron which will receive the energy packet (and the atom gets excited). And what kind of force that we are talking about in this case? The electric force, because surely the photon is not disturbed by the spacetime curve in this case, but it is interacted by the other property of the electron: its electric field. This is a concrete explanation why an atom disturbes the path of the photon.

But what happens in case of a hole?

Let's imagine a two-slit experiment with electrons. There is a wall with two slits. In this case I'm sure that we have to count with the static electricity of the wall, because there will be some kind of electrostatic lens inside the holes (slits). But if the wall is neutral, still there are atoms at the edge of the slits that may have effect of the path of the electron by their non-consistent electric shape. For example if the wall is made of ice, then there are H2O molecules inside and surely we know that they are polar. I don't know if this counts or not. If the electric field of the atoms (molecules) are too weak to count with then what force disturbes the electron inside the slit? (Obvious that if the slit is too big, then there will be noting to disturb the straight line path of the electron). And to top it all, the experiment can be done with photons also, so we must say something what force or effect counts as "disturbing effect" in case of photon.

 

[zrek]

... I see no point or meaning in any discussion about the 'path taken' by a photon, in any case.

I agree that it is difficult to say something about the photon if it is not released yet. Also the "path" of the photon can be only an imaginary property.
But we can try to imagine a path if we already know that the photon was released and arrived. And we can easily exclude places where we can be sure that the photon was not there. For example if you (laser) light through a tube (pipe) and the arrived measured energy fits to the released energy of the laser machine, then we can know for sure that the path of the photons was through of the tube.
Also if we create a wall from the same material (as the tube) and we found that the photon (energy) can't across it, then we'd be stupid to not assume that the material of the wall is not blocking the path of the photon in any other case. This two simple experiment ensures that the path of the photon was inside the tube.
Also known the refreaction and the refractive index of materials. These effects are well known from their ability to change the path of the light.
Take a look at this picture: light path
Why we can see a path made of straight lines? Because there were lots of single photons, trying to behave similarly, following a path, and some of them were "disturbed" by the atoms of the material and by this they arrived to the camera.
I don't want to explain more why we can talk about "path" in case of photons and why it is important to know what effect is changing it. Please agree if you can accept this.

 

[jartsa]

How obstacles mangle a wave function: (what is a wave function, that I don't necessarily know)Photon and a small hole in a glass pane: Momentary slowing of propagation of wave function and partial reflection of wave function.

Photon and a small hole in a mirror: Partial reflection of a wave function.

Photon and a small hole in a black cardboard: Partial collapse of a wave function by many measurements, where the result of each measurement is: "no photon here".

 

[sophiecentaur]

But what happens in case of a hole?

Let's imagine a two-slit experiment with electrons. There is a wall with two slits. In this case I'm sure that we have to count with the static electricity of the wall, because there will be some kind of electrostatic lens inside the holes (slits). But if the wall is neutral, still there are atoms at the edge of the slits that may have effect of the path of the electron by their non-consistent electric shape. For example if the wall is made of ice, then there are H2O molecules inside and surely we know that they are polar. I don't know if this counts or not. If the electric field of the atoms (molecules) are too weak to count with then what force disturbes the electron inside the slit? (Obvious that if the slit is too big, then there will be noting to disturb the straight line path of the electron). And to top it all, the experiment can be done with photons also, so we must say something what force or effect counts as "disturbing effect" in case of photon.

I think you are bound for confusion if you want to maintain the Wave / particle ideas at the same time (that's a general principle). How can you think that there is any fundamental difference between the two slit and the single hole situation in principle (except in a 'nonsense' ideal case for infinitely thin slits)? If the slits are of finite width then you are just dealing with a diffraction pattern in both cases. The nuts and bolts of what happens to the photon (or electron) as you consider its path through the screen are not relevant (just adding details of complication but nothing in principle). Introducing electrons doesn't help, actually. It just involves a different set of parameter values. Of course the electric fields of the atoms around the edge are relevant. Without those fields, the electron would sail through the screen like a neutron and we would have a totally different situation. (Likewise for gamma photons, actually)
It is important to avoid red herrings and to keep to the essence of the situation. An ideal screen with an ideal hole or finite slots is quite acceptable to get the principle - which is much too philosophical to bother with the odd practical detail like screen material.

 

[zrek]

I think you are bound for confusion if you want to maintain the Wave / particle ideas at the same time (that's a general principle).

I mentioned the (double) slit experiment(s) because they are well known. Also known that it is an important question (in this case) that we know the "which slit" information or not. But I didn't wanted to discuss this topic, I only used this experiment to show that the path of the photon is important (and the term "path" is important). In this topic I don't want to make difference between "two slits" or "one hole" or any kind of close material that makes the photon to participate to produce diffraction (interference) pattern. The important thing is now that without any disturbance (big hole or no barrier (obstacle) at all), we can count with a straight line as path (and on the detector screen there will be a point-like-shape). But if there is a wall with a hole (or two slits, doesn't matter) we will definitely get a different shape on the detector screen (interference pattern or probability distribution), but in fact we can count every single photons released and arrived. By this if we'd like to imagine a path for the photons, we must consider that at least some of them was not straight.

I think I'm not confused by the duality. The wave function describes how to count with the path, and the obstacles between the source and the destination point cause that the path is not straight.

Problems:
1: We can not talk about "path" of the photons. I agree that maybe there is no "movement". Then what happens? Maybe the photon teleports from the source to the destination. But this "teleportation" have strict rules (depending on the obstacles, spacetime curve, etc)... but these rules define a path. I think that there is no other way than considering a path. Let's close this issue.

2: Wave behavior. Doesn't matter, the photons can be counted one by one even if the wave property is visible (the arrive points of the single photons form an interference pattern). Let's close this issue too.

3: The path is obviously can be changed by the effect of material. No more words needed.

4: The path is also changed even if we know that the path is close to a material, but not "crossed" it. Why? The near obstacles disturb the path. But what kind of force counts in this case? A little help: we know that the spacetime curvature can provide a phenomenon like this, also the electric field can bend the path of the electron, both of these cases are similar to ours.

(It can be that the answer is not a direct effect, but probabilistic, but if this is the case, I need an explanation how this have connection to the wave function. For example I've heard that someone tried to explain gravity with the entropy change, maybe my problem can be explained this way, but I think that there is a simple answer how the physicists are thinking about when they are try to explain why the path is changed by a close atom)

 

[sophiecentaur]

1. The question about "movement". There is a delay, due to c and the actual time taken for the resulting detection will have a statistical spread, just as the point of arrival has. (You still seem to demand that the photon has to be in existence whilst the energy is traveling from A to B; why?; it doesn't have to)
2. You may be able to tell when a photon arrives but you can't tell when it left without disturbing it or destroying it.
3. No more words - I agree.
4. The concept of the need for an actual 'path' is flawed', I feel. The energy flows through an aperture, which may vary in size on the way, from an infinitely wide space to number of restricted apertures. From the wave approach, it is the surface integral over the whole of space that determines the way the wave develops along its journey (even going back to Huygens) so why should a particular 'ray' be relevant? The actual nature of the 'disturbance' or obscuring of the light between source and detector doesn't really affect any of this. Why can't you allow a photon to be just a quantity of energy and leave it at that? We surely acknowledge that we can't tell where it is when it goes through two slits so why should we ever know?
Yes, it is of interest to know how the beam / wave / photon is affected by the presence of matter and this implies that mass and charge will have an effect. It is usual to split the problem into two and to look at the propagation separately to the effects introduced by obstacles or apertures or even when there is scattering (another separate issue).

 

[zrek]

1. (...) From the wave approach (...) why should a particular 'ray' be relevant?

The photon can be detected and the light can be emitted only in "packages", quantums, photons (not as waves). The wave function can not be detected. The only reason of the wave function is to calculate the probability the place where the photon will arrive. But we can detect a photon, (no more) in a specific and almost exact place (compared to the spread size of an interference pattern for example). If we build up an experiment that consequently tries to find the path of the photon, we can see the path of it (as I mentioned the picture above, it is an easy experiment for it). One arrived photon represents one particular ray. This is why I think that it is relevant. (And also relevant because of the fact that by knowing the part of the path may cause the change the result of the double-slit experiments, etc, so it is important because it have some kind of connection with the behavior of the light)

Despite this I'd like to try to accept your explanation. Waves. Then first of all we need to know what disturb the wave function, what kind of property have the material that counts? You can calculate the wave function if you know the followings: the places of the obstacles, the behavior of the obstacles (reflective or absorber, etc)... so how the photon would behave if the material of the obstacle has been hit by the photon. If we assume that this is based on the electric force of the electon (of an atom) then we can say: the path of the photon through the hole is disturbed by some kind of a "remainig" electrostatic effect of the material that surrounds the hole. (and the spacetime curvature is negligeable) Is this close to your opinion?

We surely acknowledge that we can't tell where it is when it goes through two slits so why should we ever know?

I think that we can tell that on which slit it went through, so we know some point(s) of the path, the only problem with it that in this case the interference pattern will disappear. But this doesn't matter in our topic because in this case the path is still not a straight line in every case.

 

[zrek]

How obstacles mangle a wave function: (what is a wave function, that I don't necessarily know)


Photon and a small hole in a glass pane: Momentary slowing of propagation of wave function and partial reflection of wave function.

Photon and a small hole in a mirror: Partial reflection of a wave function.

Photon and a small hole in a black cardboard: Partial collapse of a wave function by many measurements, where the result of each measurement is: "no photon here".

Thanks jarsta, this helped me a little.
What is your opinion, why the obstacle slows down/reflects/or collapses the wave function?

 

[sophiecentaur]

I think that we can tell that on which slit it went through, so we know some point(s) of the path, the only problem with it that in this case the interference pattern will disappear. But this doesn't matter in our topic because in this case the path is still not a straight line in every case.

I can't think why you are drawing a distinction between the diffraction phenomenon due to two slits and for a single hole or slit. In neither case is there a 'direct path' through the aperture(s).

I have a feeling that you are still hanging on to the 'little bullet' model of a photon, in your head. You need to ditch this completely. If you still feel inclined to hang onto it then I suggest you try to apply this view to the photons of a long wave radio transmission, with a wavelength of 1.5km. How big would your photons be in that case? Any model you have for light and gamma must also fit in with these and even longer wavelengths. The concept of the 'Extent of a photon' is basically a non-starter. Whether you call the interference pattern a wave phenomenon or a probability distribution, neither description involves a photon of defined size and they are, of course, mathematically equivalent.
People like to use the term 'wave packet' as if that implies some small entity but it's just playing with words and really has no implication about size.

 

[jartsa]

Thanks jarsta, this helped me a little.
What is your opinion, why the obstacle slows down/reflects/or collapses the wave function?

The black cardboard is sitting there and observing photons. The wave function of a photon collapses into that point on the cardboard were a photon is observed.

When no point of the cardboard observes a photon, then the photon wave function collapses into the hole on the cardboard. Before the collapse the wave might have been 10 cm wide, after the collapse the wave is as wide as the hole.


A wave function of a black cat and a wave function of a photon collide. The result is a more energetic black cat wave function. The possible energy levels of the cat were such that aforementiond result is possible. Actually pigment molecules on the surface of the cat have those energy levels. Pigment and photon interact somehow.

Or should I say pigment wave function and photon wave function interact somehow. Do you know how black paint and light interact?

 

[zrek]

I have a feeling that you are still hanging on to the 'little bullet' model of a photon, in your head. (...) How big would your photons be in that case? Any model you have for light and gamma must also fit in with these and even longer wavelengths. (...) interference pattern a wave phenomenon or a probability distribution

I don't hanging on the "bullet" model of the photon, this is why I mentioned the energy transfer as "teleportation", it doesn't matter. Even if it is a (delayed by c) teleportation, in correct experiments you can (and logical to) consider a path for it. I hope you can understand this, please let me know if this is not acceptable for you. The "bullet form" is an easy way to imagine that if the freq/energy is high, the effect of the photon can be limited to a small area (point-like spot on the detector screen) which is much more exact than what the wave function can describe -- exactly because the reason of the wave function is not to define exactly the annihilation place, but gives only its probability, so the "result" of the calculation of the wave fuction is obviously a wider range than when what we can determine as "path" or "ray".
If we are counting with long wavelengths (low freq/low energy) then this is the same: you can define a path for it, which will be a (more-or less) straight line, especially if we compare it to some kind of an interference pattern of it (which will be a very wide range in case of long waves, I think). We can say the same: if we "disturb" the log waves, we will get a different path. (however I have heard about several models where the phenomenon of the "bullet-like" photon is fit also for the long waves, but doesn't matter now, this is different topic)

"I can't think why you are drawing a distinction between the diffraction phenomenon due to two slits and for a single hole or slit. In neither case is there a 'direct path' through the aperture(s)."

I don't wanted to draw a distinction between these, I wanted to show that in both of the cases we can imagine the effect as the path was disturbed.
Now I'll try to explain once again why I think that it is OK, if we want to describe a 'direct path', for any cases, sorry for my bad english. I'll try to make a numbered list (steps), please let me know in which point you don't agree.

1. We can define 2 point-like areas/spots for the photon: the source and the destination. We can detect the photon in an (more or less) exact place when it is arrived to the detector screen. By the wave function we can't determine this place so exactly, because the reason of the wave function is not to give a place, but a wider range and probability levels in this range.
2. These 2 points can be connected. First of all physically: we can make it sure that the source energy arrived to the destination. Secondly: we can imagine a path, for example we can draw a straight line between this 2 points. (in the beginning this is up to our imagination)
3. We can make lots of similar experiments, and we can conclude that the energy transfer (lets say photon) behave similarly every time.
4. By putting obstacles between or near to this imagined straight line, we can aware the followings: there are obstacles/material that reflects/blocks the energy transfer (the photon). By several experiments with obstacles, we can draw a conclusion: sometimes the path can be a straight line, sometimes not. For example in case of a pipe, or a mirror...
5. If there is no disturbing obstacle, we can measure the size/width of the area, where we can detect the photons destination places. By this we can assume a width of a straight path (maybe we are wrong, but we can assume), by this we get a "more or less" straight path area.
6. Now let's take an obstacle made of a special material. We experienced earlier that if we block fully the more-or-less straight path with this obstacle, then it will fully blocks the transfer and absorbs the energy. But now we will just put this obstacle not exactly between the 2 points, but close to this more-or-less straight path. The obstacle will not even touch this imagined path.
7. Now if we measure the energy transfer we will find 2 things: 1: a very little energy portion will be absorbed by this obstace. (or: if we do the experiment photon by photon then sometimes it will block the photon before it arrives to the screen) 2: the destination place area is bigger compred to the earlier mesurement (step 5). We will call this phenomenon as "diffraction"
8. On the new destination place we will find such arriving points of individual photons, which can not be connected by a straight line from the emission point.
9. We can now conclude the followings for sure: A): If there is no disturbing obstacle, we can imagine a straight line as path for every individual photon every time. B): If there is a disturbing obstace, then there are cases when there is no straight line available (the material disturbs the path, this is ok) C): if the obstace is close enough to the imagined path that we described in A) (but not blocks it at all) it may result that some start-end points can't be connected by a straight line.
10. There are other scientifically accepted cases, when we can talk about the modified path of the photon (without "direct touch" of an obstacle/material), for example one of the proof of the relativity when the curved path of the photon was detected near to the Sun, and the modification was caused by the spacetime curvature. In this case it was exactly given why the path was not straight. Also it was important to know what was exactly the force/field/effect that caused the "disturbance".
11. It is not a stupid state if someone assumes that in case of 9.C the photon path was disturbed not by a "direct touch" of the material of the obstacle, but there is some effect near to it (in the hole) that caused the disturbance of the path.
12. The case 9.C and 10 is similar, for the first view it differs only by the size of the experiment (ok, the similarity can be only an illusion)
13. I found two cases for 9.C: 1: at the surround of the obstacle we can imagine some effect that causes the path disturbance 2: the obstacles(material) have an additional feature -- it can not only reflect/absorb the energy but also can "draw it closer" (ok, this sounds stupid :-) )
14. It is clear that 2 kinds of "force" can cause any "disturbance" for the photon itself (electric and spacetime curvature). By this if we want to imagine a path change of it and this seems to be not a "direct touch" then the cause of the change is also some kind of an (indirect) effect of these forces.

Thank you for your patience.

 

[sophiecentaur]

The above is fairly reassuring (that the little bullets are not part of your model )
I don't think we are disagreeing much about this, in fact.

If there is no obstruction between source and detector then, once the photon has been detected, you can draw a metaphorical line between them. Does that really mean that a photon has actually traveled along this line? I don't think so. It's just a way of looking at things, imo. It is true that, with no significant restriction to the wave front in an optical system, you can draw a straight geometrical line between emitter and detector, with a ruler and that 'ray' is a fair description of how the optics have worked. That is the most trivial case and it is because all other possible points of arrival have zero probability. Any 'selection' by blocking or reflection will spread the probability function out and the simple 'ray' will not give a satisfactory answer. With enough computation, you can calculate the diffraction pattern (or whatever else you care to call it). I don't agree that this method could be any less accurate than any other method as they all boil down to the same thing - involving calculations across the whole of the width of the optical system and the detail of interaction every 'obstacle'. Whether you discuss what goes on in terms of collapsing wave functions or diffraction, there is no basic difference afaics.

This thread seems to be about "what's really happening" more than anything else and you seem to be bringing in various practical situations in order to resolve the question. I guess it's always true to say that the specific problem may call for a particular 'best' approach. But I don't think it's any deeper than that. There are many cases when a very conventional Ray Tracing method will yield excellent predictions (that's going to the extreme classical approach, if you like). Just because it doesn't consider waves or QM doesn't invalidate it, when it gives good results. Likewise, one would choose between different modern methods to suit the problem. I say this because I believe that there is no absolute truth in this (or any Science ftm) and we just use the best available model for the purpose.

 

[zrek]

Does that really mean that a photon has actually traveled along this line? I don't think so.

Ok, but why not? What makes you prevent thinking on it? I agree that it is not necessary, but I think that it is not a problem if someone thinking like that.
You said that I maybe hanging on the "bullet" model, but now I think that you hanging on the wave model, but it is not better.
Finally I think that imaging a path have certain advantages, especially when the endpoint is known. We can build up a model from the experienced, collected data, and the known endpoint helps us.

The wave model and the path integral have equal strength and (if I know well) basically these two approach result the same mathematic solution.

But when the endpoint is given, and you want to determine what happened (why the obstacles resulted this endpoint) then the wave approach is less illustrative. (Why I should imagine "concentric" waves, when its final conclusion is an area with probabilities? It will not give a concrete endpoint.) I like the "lots of paths" and if there is an endpoint, I know that one of them was the winner.
And why it is a good idea to thinking on that the energy travels on the path? Because in general if we block somewhere (anywhere) the line, we will find there an endpoint.

The only problem that in this case we have to tell something that why the path is disturbed by a hole :-) LOL

Thank you for your answers, I think that basically we try to agree with each other. I'm happy that you devote time for my stupid toughts.

 

[sophiecentaur]

Reading your last post I conclude that you are wanting the equivalent of an 'explanation' for something that has already occurred, rather than being able to predict what will happen. Surely that's what Science tries to do. Once the photon has been detected, then only one thing could have happened. If you were to put a counter in one place on a screen, that would tell you very little about the diffraction pattern that the 'obstruction' caused. The "winning" path is hardly relevant - just the fortunes of a single foot soldier, rather than the progress of the whole battle.
Knowing about the distribution of photons on a screen can give information about the layout of the obstruction, of course, but all that stuff is arrived at by methods such as Fourier Analysis, which are basically, wave-methods.

BTW, I would not dream of applying the wave model where the photon model is appropriate. I am just arguing about where each particular model happens to be appropriate.

PS. They are NOT "stupid thoughts", by any stretch of the imagination.

 

[zrek]

Thank you sophiecentaur, I'll still thinking on this topic and will post here if I've found out something. Have a nice day!

 

[meBigGuy]

You seem to be struggling with the concept of a particle being affected by the material near its path to cause it to take a curved path and cause diffraction. But you can't think of it that way. If you want to understand diffraction in terms of particles you need to think in terms of feynmans path integral. The particle takes all paths and interferes with itself. If you compute all the possible paths through the hole you get the diffraction pattern.

GO to this lecture by feynman and jump forward to about 50 min. http://vega.org.uk/video/programme/46

 

[sophiecentaur]

Feynman was very keen on talking in terms of photons being particles but you will note that in his lecture (and elsewhere) he allows the 'particle' to be pretty much 'anywhere'. Despite giving it the name 'particle' he does not assign it a size or a location until it arrives at a target. In his head, he clearly had it all sussed out but I think that is is not sussed out in the heads of all of his audience (for generations, now!). He never, ever talks of a photon particle as being like a little bullet.
I wish he had used another term for the thing that he understood so well. I think he must have been too smart to realize how relatively unsmart many people are about this matter. Strange, because he spent hours ranting (with good reason) about the meaninglessness of expecting to know exactly how things really are.

Reading this through, I can see it looks a bit like some religious scholar explaining what was meant by the author of some portion of scripture. I don't want it to be taken that way.

 

[zrek]

The particle takes all paths and interferes with itself. If you compute all the possible paths through the hole you get the diffraction pattern.

GO to this lecture by feynman and jump forward to about 50 min. http://vega.org.uk/video/programme/46

Thank you for the video!
I know the concept of the path integral, and that the self-interfere destroys most of the possibilities (I mentioned this earlier). This is not a problem for me. But, if we are talking about "paths" and there is a start and an already known endpoint, then we can conlude that one (and no more) path finally won the fight of the interference battle.
If we calculate the intergral of all possible paths, then we have to select which paths are NOT possible. Everyone says: it is obvious, since there is a blocking obstace.
The question is what blocks exactly a path? What kind of force? The blocking obstacle is a real material with real properties. It is necessary to discover every properties of it.

The path integral method and the way how we calculate it (the mathematical formulae that results the final, winner path(s) ) is working with lots of imagined paths, which will finally have no effect on the real material (the obstacle). This is only a mathematical trick. Noone explains how an imagined path interacts with the real material (ok, I know it is not necessary beacuse this is only a model), and no one tells what kind of property counts when the material of the obstacle makes a path "not possible" --- and this is the question.
Now let say that those imagined paths are labelled "not possible" which if this path would be a path of a real photon (?) then an obstacle would block the energy transfer (the photon). The block caused by the electric property of an electron of an atom (which would absorb the energy if it would be a real photon) Then 1) by this we agree that the remaining, winner path is the path of the photon. 2) the winner path is curved, avoids the obstacles and goes through the hole. 3) the path is curved because of some kind of "far, remaining, collective, indirect" effect of the atoms of the obstacles around. This effect disturbes the path of the photon inside the hole.

This would be a concrete answer for the topic question. I'm not sure that this is acceptable.

The wave function is similar: the waves are no real constructions, have no real effect on the obstacles, but we still have to define why we can count on the obstacles as blocking object in case our waves. The obstacles are made of real material. If we are talking about these imagined waves, then we don't have to talk about real properties, but if we count with real materials, then every property of it have to be connected to real properties.

The path integral is mathematically the same as the wave function (as far as I know), it is a twist of it, a model to imagine and calculate easier the phenomenon.
Maybe if we name this "far, remaining, collective, indirect" effect, then we will arrive to another (third) model. Maybe this will lead nowhere. Maybe this is only a part of the two other modells, but no one yet named it. Maybe someone already named it, but I don't know about it.

 

[meBigGuy]

The path blocking is just that -- a total blockage of the path. A perfect absorber. Either the path is allowed, or it is totally blocked. To the extent that the particle is affected by the blocking barrier in some other way, it is noise (it actually becomes entangled with the barrier) and coherence is either destroyed or weakened.

Again, the path is not curved by anything. The particle takes all paths. The barrier is considered a perfect barrier. It either absorbs or allows passage of the particle. Anything else is beyond the scope of the problem.

So again, if the barrier in anyway affects the particle while allowing it to pass, the problem has just become different. The particle is now entangled with the barrier and coherence is weakened accordingly.

Moving on to particle wave duality, I think resorting to the "wave nature of a particle" to explain particle interference/diffraction is old school. Particles are particles. We could go on and on about what a particle is and where it is, and previous threads have done so very well.

 

[sophiecentaur]

A particle is a particle, of course. But that statement is not enough until you describe the nature of this thing. The name "particle" is very misleading in the case of photons because it carries wth it, a lot of confusing baggage. It implies things about a photon which are not appropriate.
If one uses that word in any other context, the implication is a small entity. There is nothing small or large about a photon so I say it is not a good term to use, without a lot of care.

 

[zrek]

Again, the path is not curved by anything. The particle takes all paths.

(I'm not talking about a particle, but it doesn't matter, you can imagine like that now. It doesn't matter if there is something "moving" on the path or not.)
I'm aware this, but I have a different point of view of this (but the same thing, only the view different). Allow me to use your words.
"The particle takes all paths" So there are paths. Some of them are straight (more or less), some of them not. These are (lets say) curved paths.
(The "straight" path not necessarily perfectly straight. But I'd like to use this word (straight) to show the difference between the undisturbed line of ray and the diffracted one, which is for sure not straight)
All of these paths are imaginary lines, they have no real effect to the matter (can't be measured).
But once the endpoint is known, then we can choose a "valid" (still imagined) path form the all (a final result, the "winner"), and sometimes it is not possible to choose a straight one -- only a curved one. This, the final, curved/kinked path is what I'm talking about.

(If this above is not true then maybe I misunderstand the path integral)

The particle is now entangled with the barrier and coherence is weakened accordingly.

Sorry, but I don't understand this. Is this have any connection with the path integral?

 

[sophiecentaur]

Any "path" that you bring into this discussion will just be a mathematical construct, it will just refer to some method of integration over all space (or a practical sub-set). Whether you view it as a sum of all 'paths' or something along the lines of a Huygens construction, there is no implication of one particular path being particularly special.

I realize that we are probably stuck with using the term 'particle' but the very fact that this thread has grown so long, goes to show the sort of problems the term introduces into an already difficult subject. After is a 'particle of sand' not, by definition and universal usage, the way that some small entity or spatial part contributes to the whole of a concrete wall? It doesn't occupy any more than a well defined, tiny volume. So why ever would one use the term 'particle' to describe something, in the context of EM, which has the direct inverse of that nature? The only times that a photon fits the conventional description (i.e. appears at just one particular point in space) is at each end of its life.
It's all the fault of that ancient corpuscular theory, I think.

 

[meBigGuy]

"

But once the endpoint is known, then we can choose a "valid" (still imagined) path form the all (a final result, the "winner"), and sometimes it is not possible to choose a straight one -- only a curved one. This, the final, curved/kinked path is what I'm talking about."

That is where you are missing the point. There is no single "final" path. There is only the sum of all the paths which sum to a probability amplitude for a given target location. If you only considered a "final" path you would see no amplitude variation. Think about it this way: In order to get cancellation, it MUST take at least two paths AT THE SAME TIME to the target point. Sometimes the two paths add, sometimes subtract to any given target. In reality there are infinite paths to every target, none of which is "the" path.

You can make up a curved path to the target if you like, but it will tell you nothing about the probability (brightness of the interference bands). Your made up path tells you nothing about the hole. The only way is to sum infinite paths through the hole to the target point (a path integral) to get a probability amplitude for the target point.

 

[Drakkith]

Personally I prefer to imagine that photons, and indeed all particles, do not exist as point particles until they interact with something. Easier than imagining a particle taking multiple paths at once and all that. But I'm not sure how "accurate" that is with regards to current quantum theory.

 

[zrek]

"
That is where you are missing the point. There is no single "final" path. There is only the sum of all the paths which sum to a probability amplitude for a given target location.

Yes you are right (thank you for spotting out), this is what was not clear and I forgot to explain. Now it is easier to realize and explain what is my problem.
I wanted to explain but somehow I missed to tell that (as I think) it doesn't matter if the path integral gives a final "single" path or not. Three things matter:

1. We are talking about paths (lots of).
2: We calculate them one by one, and during this process we count with obstacles. We must provide the concrete effect that why an obstacle blocks a path. Mathematically this is not necessary, but phisically this is important. I assumed that maybe this is the same effect as if we assume that if the single path was connected to a real photon, then it would lead the photon to be absorbed by the obstacle.
3: We know that the photon finally absorbed by the final obstacle, the detector screen. So if we are using lots of "single paths" during our calculations (point 2), and we are counting with them as assuming that what would be if that was the "final path" and the real photon, then it is logical to thinking on that the real final absorbtion point with the really real photon is also connected to a single path. (The final obstacle, the screen have no privilegued state, we are counting with it the same way.) (The conclusion is that you are right, we need at least two paths for the calculation, but as we are counting with the paths one by one the same way as "if it would be a real path", the final path is also a single one, and the interference which chooses it is only a side effect, an indirect effect.)

I know that my english is very bad, please forgive me and try to understand. Please let me know if it is confusing and I'll try to explain it on other way.

Also it is not necessary to agree with me in the point 3. But I think that you should consider agreeing that the point 2 must be clearly described, and we must choose a physical effect even if the paths we are calculating with are only imaginary constructions.

 

[zrek]

Personally I prefer to imagine that photons, and indeed all particles, do not exist as point particles until they interact with something. Easier than imagining a particle taking multiple paths at once and all that. But I'm not sure how "accurate" that is with regards to current quantum theory.

This is a good viewpoint, but you still need a modell that describes why the photon interacts finally with a specific point and counts with the obstacles. This modell can be wave based or path based or anything else, but it must involve real properties, real effects of the obstacles, even if the waves or paths are only imaginary.

 

[meBigGuy]

There is no final path. Period. End of story. There is no model of why it ended up at a point. Only a probability for each point that can easily be calculated. There is no way to determine where it will go or, afterwards, how or why it got there. The waves or infinite paths are not imaginary. The effects you are trying to create are what are imaginary. They do not exist. You want a cause for the photon to arrive at a specific point, and it does not exist. You want to rationalize how it might have gotten there, but there is no rationale.

I've pretty much said it in every way I know how. You obviously don't understand, and want to assign a cause where none exists.

 

[zrek]

I've pretty much said it in every way I know how. You obviously don't understand, and want to assign a cause where none exists.

OK I accept that I don't understand, and what do you think about the point 2? It is not about the "one final path", but the way to tell why the paths are blocked by an obstacle.

 

[sophiecentaur]

We are on the 37th post now and you still don't want to let go of the idea of 'paths' being relevant, in a 'physical' way. Imagine you had a cloud (in the sky) and it was moving across the sky, changing shape all the time (as they do). Also, accept that you can only see it in 2d. How would you describe its path in terms of a single raindrop that lands on your head? You cannot know which bit of the cloud produced the raindrop and you cannot specify the 'path' of the drop or the cloud other than, perhaps, from left to right. Because you cannot identify the 3D shape at any time, you can't know where the CM of the cloud is and, because it is in a turbulent atmosphere, you can't apply Newton's Laws to what you see. Then try to define the path to include the evaporation from the sea that formed the cloud in the first place. The precise path would be a meaningless thing to discuss - and so it is with a photon. You are hanging onto too many classical notions and insisting on applying them in a situation where they just aren't relevant.
You keep reacting to the views of all who have answered you by 'Yes, i get that but I want the story told my way'. It is you who need to change your ideas and you cannot expect Physics to change to fit in with your personal view. (that is rule number one in Science). Just consider the fact that your idea may not actually work.

 

[zrek]

We are on the 37th post now and you still don't want to let go of the idea of 'paths' being relevant, in a 'physical' way.

Please read back if you don't believe me, but I never said that I want to give any physical meaning to paths. I always said that it is imaginary and have no effect on the material. I also said that this is only a model from the several possible ones (which are maybe all good ones, and results the same mathematical result: only the probability of the place and not the place itself can be calculated)

Don't think of this. Please don't try to describe this again, since I accepted this long years ago (10?), this is not a problem at all.

Let's take this as a pure mathematical model. Imagine a sketch, a piece of paper, a drawn 1 slit experiment and let's try to calculate the probability distribution on the drawn "screen" which is a line.
The "light" source is a selected point, no more. The wall that blocks the imagined "light" is a drawn, long rectangle. There are two of them, there is a "hole" between them. I think that we need no more thing. We can define the sizes, distances about everything on this paper.
Now everything is given to calculate the probability distribution.
We can choose the wave model, or the path integral, no matter, the final result will be the same.
If we choose the path integral, then we will draw lots of lines, some of them will be blocked by the "wall" "rectangles".

As you can see that there is nothing real in this experiment. Now this have no real meaning at all.
This is a thought experiment, a mathematics task. Mathematical tought experiment. (I have no doubt that this modells well the real behavior of light.)

How can we turn this to physical tought experiment? We have to give some imagined real meaning into it.
To give meaning for this, we need to replace few things to imagined real objects. These objects have to behave in our imagination as we know about the really real ones. For example we put a laser apparatus to the startpoint. Now we have to choose some matherial for the screen and the wall. Now some restrictions arise. For example we can't choose such material that completely transparent. (But why? How we define the opaque material?) Also it is not enough to give the size of the material. A drawn square in a paper as wall - this is OK for a mathematics task. But now we know that our material consists of atoms. We need to define why it is an obstacle in our case.
Still we can leave our "paths" as it was in our paper, since we know that they have no effect on the material. But, the material have an effect in our imagined paths. The material have real properties, for example it is opaque. It consists of atoms. We have to define which property of the atoms counts to make it behave in our physical tought experiment the same as the drawn rectangle behaved in our mathematical tought experiment.

I hope that you can understand that this point of view doesn't mean to destruct the mathematical experiment and its results. Also I have no doubt that the physical tought experiment will have the same results: only probabilities. I need to know only what effect of an atom (or more complex material structure) counts in case of the imagined paths, and if I know I can imagine an obstacle, a non-transparent wall made of this material for a physical tought experiment.

The same question arise if we choose the wave model: we have to define what real property we count with in case of "blocking the imagined wave".

This is independent to the fact that maybe there are no paths or waves at all.
I hope now that you know what I mean on this. If not, then let's close this meaningless topic.

 

[sophiecentaur]

You can 'block' a 'beam' of EM energy by absorbing it, absorbing it or refracting it away. There is a load of information about the details of all these three effects at many different levels. Are you expecting something 'new' from this discussion? Just read around, choosing sources that appeal to you and are at the appropriate level. Just try to avoid imposing the 'path' thing except in the context of any Maths you might use in your analysis. It has no deeper significance and, in any case, nothing is 'really real'; all Science is a model.

 

[zrek]

nothing is 'really real'; all Science is a model.

A scientific model have to fit to the experimented, measured results. That is real.

You can 'block' a 'beam' of EM energy by absorbing it, absorbing it or refracting it away.

Since you stepped back to this, I assume that you still not understand my view or simply don't want to thinking on it. I accept this, no problem, still I thank you for your help. Have a nice day! :-)

 

[sophiecentaur]

A scientific model have to fit to the experimented, measured results. That is real.


Since you stepped back to this, I assume that you still not understand my view or simply don't want to thinking on it. I accept this, no problem, still I thank you for your help. Have a nice day! :-)

Absolutely. And there is no experiment that will reveal what you are looking for. That is my whole point. An experiment can just confirm of contradict the results that are predicted by a hypothesis. Any experiment you propose (and you have not yet proposed one) will not identify any particular photon as coming by any particular path. You keep telling me that you know that so what would your experiment consist of? i.e what actual model are you hoping to demonstrate? You seem to be a bit vague on that point.

There is plenty of information about the interaction of photons with matter, on the QM level and the interaction of waves with material obstacles. Which did you want? I really can't be sure, from what you write.

 

[meBigGuy]

I've already described this, but I'll say it again with more and slightly different words. (I recommend you reread what I have posted already)

Assume the blocking material is an infinitely thin perfect absorber with a perfect hole and a perfect boundary, and assume the particle has 0 dimensions. It can either pass or not pass based on the path. There are no boundary conditions. The barrier cannot interact with the particle except to let it pass, or not pass, based on the particles infinite possible trajectories. To the extent that the conditions of this ideal barrier and particle are met, the patterns will match simple predictions. To the extent that there are any boundary effects, or other particle interactions with the barrier, the pattern will be distorted. Anything varying from ideal will cause anomolies. The basic math assumes ideal conditions. You can adapt the math to account for a realistic barrier (and I'm sure there are ample papers doing so) and the result will be a non-ideal interference/diffraction pattern.

Edited slightly for grammar

 

[sophiecentaur]

I've already described this, but I'll say it again with more and slightly different words. (I recommend you reread what I have posted already)

Assume the blocking material is an infinitely thin perfect absorber with a perfect hole and a perfect boundary, and assume the particle has 0 dimensions. It can either pass or not pass based on the path. There are no boundary conditions. The barrier cannot interact with the particle except to let it pass, or not pass, based on its infinite possible trajectories. To the extent that the conditions of this ideal barrier and particle are met, the patterns will patch simple predictions. To the extent that there are any boundary effects or other particle interactions with the barrier, the pattern will be distorted. Anything varying from ideal will cause anomolies. The basic math assumes ideal conditions. You can adapt the math to account for a realistic barrier (and I'm sure there are ample papers doing so) and the result will be a non-ideal interference/diffraction pattern.

Yes, absolutely. There is little to be gained, in the way of understanding how EM 'waves' behave in the presence of obstacles (I'm trying to be as general as possible here) by going outside the ideal case. The basic diffraction calculations pretty much say it all.
There are some practical occasions where the situation is not ideal, such as in radio wave propagation, where the ground, atmosphere or ionosphere do not behave as simple components and a further layer of analysis is required but that is definitely 'third year work', as we say.

 

[Claude Bile]

Forgive me, I did not read through the thread in its entirety...this is a direct response to the OP.

Firstly, forget QM - you don't need it. This is a classical EM problem if there ever was one...

The reason diffraction occurs - Waves get partially blocked (truncated) by an aperture which generates NEW plane wave components (in the same way, truncating a pulse in time generate new frequencies), causing the wavefront to spread.

The material properties of the obstacle are important as well; e.g. a dielectric will diffract light differently to a conductor due to how the electrons in the contrasting materials respond to the incident field.

Claude.