# When a photon is released, which way does it head?

## Recommended Posts

Strange,

Except I am reminded of a time I watched a solar eclipse by poking a pinhole in a piece of paper and let the sunlight pass through the hole and held a dark piece of paper a few feet from the hole, as a screen. I saw the moon encroach on the Sun (although upsidedown and backward) on the black piece of paper. The pinhole was a point source, but the photons coming from the leftside of the Sun and those coming from the rightside of the Sun wound up in exactly opposite spots on the black peice of paper.

Regards, TAR

• Replies 192
• Created

#### Popular Days

A pinhole camera is the model for an "ideal camera" that can be used to quantify the errors introduced by optical systems. However, real world pinhole cameras are not perfect. One reason is the diffraction at the edges of the hole.

Within limits, a smaller pinhole (with a thinner surface that the hole goes through) will result in sharper image resolution because the projected circle of confusion at the image plane is practically the same size as the pinhole. An extremely small hole, however, can produce significant diffraction effects and a less clear image due to the wave properties of light.

http://en.wikipedia.org/wiki/Pinhole_camera#Selection_of_pinhole_size

We seem to be getting ever further of topic. You would be better off asking these questions in the Physics sections of the forum as you would get more knowledgeable people responding.

##### Share on other sites

Strange,

Further off topic, perhaps, but the thread question is particularly addressing geometry. That is, if you put your screen one millimeter away from the source, or one meter, or one kilometer, the place where the photon will land, on each screen, is already determined, once you know which way it is heading, upon release.

There is only one spot on each screen that corresponds to that direction.

Regards, TAR

Any deviation from this direction, would have to have a cause, like a deflection, or an absorbtion and reemission.

##### Share on other sites

That is, if you put your screen one millimeter away from the source, or one meter, or one kilometer, the place where the photon will land, on each screen, is already determined, once you know which way it is heading, upon release.

But you don't know where it will land until you detect it. There is just a probability of it landing somewhere (anywhere) within the area illuminated by the source.

And if you limit the range of possible positions (e.g. by using a laser) such that you know which slit it will go through, then there will be no interference pattern.

##### Share on other sites

Strange,

Well what if you set up the experiment so only photons on the exact trajectory toward a chosen spot on the screen were allowed to proceed, and all others where absorbed by some dark nonreflective material?

Like if you put a black screen with a pin hole one meter from the source and then another black screen with a pin hole 2 meters from the source and then another black screen with a pin hole three meters from the source. Would the position that any photon that made it through all three pin holes not be exactly the position on the final screen that was geometrically in line with the three holes?

There would be only one place on the final screen that any photon could land, and no interference pattern, even though you did not detect a photon passing through hole number one, or number two or number three, you would still be able to predict the spot on the final screen that would be hit by a photon. Wouldn't you?

Regards, TAR

##### Share on other sites

There would be only one place on the final screen that any photon could land, and no interference pattern, even though you did not detect a photon passing through hole number one, or number two or number three, you would still be able to predict the spot on the final screen that would be hit by a photon. Wouldn't you?

Yes. (Although it wouldn't be a single spot, there would still be a slight distribution as it is not possible to specify the position that exactly). As you are not illuminating all slits / holes, there will be no interference pattern. (Apart from that caused by diffraction at the edges of the hole. Unless the hole is large enough that this doesn't happen.)

You no longer have a two-slit experiment. You have a one-hole experiment. I'm not sure I see the point of this.

Interestingly, if you wanted to use QED to calculate the results, you would have to include the paths through the other holes (as well as the photon bouncing off the moom, etc) but these contributions would be cancelled out in the final result.

##### Share on other sites

Strange,

Well what if you set up the experiment so only photons on the exact trajectory toward a chosen spot on the screen were allowed to proceed, and all others where absorbed by some dark nonreflective material?

Like if you put a black screen with a pin hole one meter from the source and then another black screen with a pin hole 2 meters from the source and then another black screen with a pin hole three meters from the source. Would the position that any photon that made it through all three pin holes not be exactly the position on the final screen that was geometrically in line with the three holes?

There would be only one place on the final screen that any photon could land, and no interference pattern, even though you did not detect a photon passing through hole number one, or number two or number three, you would still be able to predict the spot on the final screen that would be hit by a photon. Wouldn't you?

Regards, TAR

Tar, a single slit still produces a diffraction pattern http://www.math.ubc.ca/~cass/courses/m309-03a/m309-projects/krzak/

This is also true of a pin hole, so as the beam of photons pass through each pin hole, rather than filter only those photons heading for a spot on a screen, the pin holes will cause the beam of photons to spread out.

And this problem would also be present if trying to direct a beam of photons to only one slit of two slits.

##### Share on other sites

Robinpike,

Thanks for the link. There were a couple of considerations I was trying out as reading it. One was the thought of cylindrical waves he brought up in the description. This seemed reasonable from one perspective, in terms of the shape of the slits, but not reasonable from the perspective of a photon, which would not know the shape of the slit it was passing through. From the perspective of a single photon, if there were to be a new point source wavefront occurring at the slit, it would be more sensible to me, to expect a spherical wave front than a cylindrical one.

Another aspect hard to true up, was a line drawn normal to the top beam line (which was angled up) which intersected the bottom beam line two thirds of the way through the sine wave cycle, to show that the photon at the top was in a different phase than the photon at the bottom. This seemed to switch timing frames, as both the beam at the top, and the one at the bottom where in phase when they reached the slit, and timing wise would, in the next moment, still be in sync.

There is a requirement, I think, to not visualize a photon as traveling along a sine wave, as that would take the thing on a path longer than the distance between the source and the destination. Better to visualize a particle being at one place on a straight line between source and destination, and having an electric field and a magnetic field at right angles to each other, that is being carried along with the particle. Sort of like the particle has its four arms out and is sort of spinning like a pinwheel as it proceeds in a straight line to its destination. With this picture, a sine wave is an afterthought as you trace the tip of one of the arms through space. Not the path of the photon itself, but the path of the effects of its magnetic and electric fields.

Have to go to work, but my thought is that the particle wave duality of light, does not mean it has to be one sometime and the other at other times, but that it has to have the characteristics of both, all the time.

Regards, TAR

##### Share on other sites

This seemed reasonable from one perspective, in terms of the shape of the slits, but not reasonable from the perspective of a photon, which would not know the shape of the slit it was passing through. From the perspective of a single photon, if there were to be a new point source wavefront occurring at the slit, it would be more sensible to me, to expect a spherical wave front than a cylindrical one.

In a sense, the photon does "know" the shape of the slit. The probability of a photon ending up in a particular location is determined by non-local effects (which is why the whole two-slit thing works with individual photons). After all, this must be true because calculating anything in photon terms MUST produce the same results as calculating it using the classical (EM wave) view.

So, in QED when you calculate the possible places where the photon will be detected, you have to take into account the entire shape of the slit, and any adjacent slits, the edges of the barrier with the slits in, the shape of the room, the distance to Jupiter, ...

You are still thinking of photons as little projectiles that can be sent from A to B and we know that they went in a straight line between those two points. That isn't how they behave.

Better to visualize a particle being at one place on a straight line between source and destination

Absolutely not! You know where the photon started. You know where it is detected. You can calculate the probability of it being detected at any location, based on what you know about the environment. That is it. If you start assuming it travelled in a straight line from A to B, you will cause yourself all sorts of grief (starting with: how does the two slit experiment work).

You don't (can't) know what happened in between. If you do anything to find out, then you change the environment and change the probabilities (and, for example, stop the interference pattern forming).

Have to go to work, but my thought is that the particle wave duality of light, does not mean it has to be one sometime and the other at other times, but that it has to have the characteristics of both, all the time.

Absolutely correct!

##### Share on other sites

Strange,

Well OK, so I can't explain the double slit to your satisfaction, but are there not constraints as to where a particular photon that was released from source atom A at time T must be, if it has not run into anything that deflected it, or absorbed it? Since it must travel at C after 1 secound it must be in a spherical shell area very thin with atom A at its center, with a radius of 186 thousand miles. It cannot be right next to atom A, or at Jupiter, yet, because not enough seconds have gone by for it to make it to Jupiter, and it had all the time it needed to clear the area around A.

If we could limit the possible locations of the photon to the spherical shell one light unit away from the source, one unit of time after the emission, then we have half the location problem solved. Then the other half of the problem would be which way did it head when it was released. Which is the thread question. Once the thread question is answered, then the location of a photon can be predicted exactly, without detecting it. As in, if I release a photon at time T in the direction D, all I need to do is draw a line in the direction D the distance a photon would travel in the duration of time under consideration, and that point is where the photon should be. Whether we look at it, or let it run into a screen there, or measure it in any way. We already know where it should be.

Regards, TAR

Secondary question just occurred to me. If a photon will travel at C away from an atom, regardless of the motion of the atom, then the momentum that a body like TAR has, when he jumps up in the air (when I could actually do that) and lands in the spot he jumped from, that works regardless of the fact that the Sun is speeding around the center of the Milky Way and the Earth is speeding around the Sun, and the Earth is spinning on its axis is not something that a photon possesses. That is, a photon would not be expected to carry the momentum that a body like TAR would. It therefore, if it headed off in a direction would have to have its predicted position be figured by where the Sun and the Earth and the surface of the Earth was, when it was released, not where the Earth and Sun are a second after it was released. That is, the lab has moved quite a few miles in the second. I think I remember reading that the speed of the Sun around the center of the galaxy was about one thirteenth the speed of light. So in my direction plus distance formula, while predicting the postion the photon MUST be in according to the direction it left the atom and the time since it left the atom, the orientation of the lab to the direction of motion of the surface of the Earth, and the Earth around the Sun and the Sun around the center of the Galaxy and the Galaxy around or toward the great attractor would have to be factored in.

Maybe that is why we have such a hard time keeping track of the little fellas. The lab does not have a stationary reference point in it, with which to orient the position of the photon, after it has left the atom.

##### Share on other sites

Well OK, so I can't explain the double slit to your satisfaction

I'm not sure why you are trying to explain it. It has a perfectly good explanation already.

Since it must travel at C ...

Actually, when calculating the probabilities for where a photon will be found, the velocity is not restricted to c; you have to consider paths where the photon travelled faster and slower than this. That is because quantum interactions are non-local i both time and space. That is why the results of the double slit experiment can be affected by events that happen after the photons have hit the detector.

Once the thread question is answered, then the location of a photon can be predicted exactly, without detecting it. As in, if I release a photon at time T in the direction D, all I need to do is draw a line in the direction D the distance a photon would travel in the duration of time under consideration, and that point is where the photon should be.

You have gone from the light being being emitted evenly, in a sphere (in which case you have no idea which way the photon went) to sending the photons in a single direction. And, of course, you can't send them in a single direction, just in a narrower range of directions.

Whether we look at it, or let it run into a screen there, or measure it in any way. We already know where it should be.

Within some range, yes. But so what? I thought you wanted to explain the double slit experiment, but now you have moved away from that to projecting photons in a single direction. So I'm not really sure what your are trying to say or explain.

But despite all that, if you send a photon in direction D from A and detect it at B all you can say is that it left A and arrived at B. You cannot say anything about what happened in the meantime. You can assume it travelled in straight line if you wish. But then you can't explain things like the double slit experiment (if it travelled in a straight line from the source to a slit and from there to the screen, then how is an interference pattern formed). You have to take into account the fact that the photon is affected by everything around it, even if those things are nowhere near that straight-line path.

##### Share on other sites

Strange,

Well the explaination you give of probability and things effecting things in impossible ways is not very satisfying to me. It does not make sense. The interference pattern could be explained just as well by photons reflecting off the edge of the slit, and by considering the width of the source and such. And by figuring out what a packet of energy like a photon is effected by and how it is affected. What size slits it can go through and how it reforms on the other side and such. There are things about photons that can be looked at in several ways, and its all happening very quickly. Sometimes the cumulative effects are more an emergent property than an actual property of a single photon...and such.

I will keep looking for a reasonable explaination. You can believe we cannot understand what is going on, if you like.

Regards, TAR

##### Share on other sites

Well the explaination you give of probability and things effecting things in impossible ways is not very satisfying to me.

Well, that's a shame. <shrug>

You can believe we cannot understand what is going on, if you like.

I don't believe that, at all.

##### Share on other sites

Strange,

Well then why do you tell me photons behave in a manner that makes no sense?

Regards, TAR

And there is nothing wrong with looking for a satisying answer. Nothing wrong at all. The right answer will fit with everything we know, and it will be true in more than one way. You will be able to characterise it this way, or that way, and it will add back up to be true in both and all ways that it works.

Strange,

From what you know, does a photon have any particular reason to go in a straight line, that is straight in the lab and curved through space, from the center of the galaxy's perspective, or does it have a particular reason to go in a straight line from the center of the galaxy's perspective, which would make it a curved line in a moving lab?

Regards, TAR

##### Share on other sites

Well then why do you tell me photons behave in a manner that makes no sense?

When have I done that?

You will be able to characterise it this way, or that way, and it will add back up to be true in both and all ways that it works.

That sounds a bit like different interpretations of quantum theory. You can choose the explanation you want, but they are all describing the same underlying theory.

From what you know, does a photon have any particular reason to go in a straight line

From what I know (which is not much) we cannot say anything about the path of a photon.

##### Share on other sites

Strange,

Can we not say at least something about the path? If we would know for instance the source, and we would know the destination, and we would know the time it took to transverse the space between, and we would know the speed and vector direction of the lab, (figuring the Earth's spin and revolution around the Sun and the Sun's path around the center of the Galaxy and the Galaxy's movement amoungst the local cluster) we should be able to draw a path that would take the photon from the source to the destination in exactly the time we clocked the transistion. For simplicity we could entertain only the first photon to arrive at the destination, ignoring the late arrivals that have deflected or refracted or reflected off or around something, or by taking only the first, discount ones that are the result of absorbtion and reemission. And thereby be able to say at least something about the path of a photon. That being the total plot of the positions in space that it held on its way from source to destination.

Regards, TAR

##### Share on other sites

Again, if you assume that the photon travelled in a straight lines from the source to one of the slits and then to the destination then you cannot explain how an interference pattern is formed. Why is a particular photon more likely to arrive at some places than others?

Watch the Feynman QED lectures (or read the book). He gives a simpler example that is even more difficult to explain treating photons as particles that travel in straight lines: light being reflected off a glass surface.

In the classical view, we can consider half (say) the light travels through the glass and half is reflected. In the photon view, we simply say that the probability of any photon being reflected is 50:50 and it all works out the same.

But the amount of light reflected depends on the thickness of the glass. In the classical view, this is again easy to explain: part of the light is reflected from the front surface and part of the light is reflected from the rear surface. Depending on the thickness, these will interfere either constructively or destructively changing the amount of light reflected.

But how does a single photon, reflected from the front surface, know what its probability of being reflected should be?

You can't explain this by a classical view of photons as little balls travelling in straight lines.

(Watch the Feynman videos, he explains it better than me. Obviously. Or ask questions in the relevant part of the forum. You might get some expert answers.)

##### Share on other sites

I should note the phenomena is not just seen in the slit experiments. We have the technology now to send single photon. As well as detect them. Even without the slit the wave particle duality is present.

Even with single photons

##### Share on other sites

I should note the phenomena is not just seen in the slit experiments. We have the technology now to send single photon. As well as detect them. Even without the slit the wave particle duality is present.

Even with single photons

When you say "single photon" what do you have in mind?

A point particle?

Or a wave, like a line? or the circumference of a circle?

Or a wobbling surface?

##### Share on other sites

Mordred said:

The answer is clear like crystal water.

##### Share on other sites

No this company has a single photon emitter and detector. They also manufacture particle entanglement diodes.

http://www.toshiba-europe.com/research/crl/qig/entangledled.html

The quantum dot detector is the single photon detector.

Particles have both pointlike and wave properties. This detector is no exception.

The wave like properties affects where on the quantum dot detector the particle strikes. Just as in the double slit experiment. Their is some technical papers on the site. Some free some must be paid for.

##### Share on other sites

Mordred,

The diagram shows an R photon going one way and an L photon going the other way out of the entangled pair creation device.

They are shown as going at a similar downward from horizon angle in both directions. If there were to be a screen a distance to the left and a screen a distance to the right, would the incidence on each screen be geometrically related, every time a pair was created? That is, would you know where the left one hit the screen, by where the right one did?

Regards, TAR

##### Share on other sites

A good question I will have think on that. The design of those devices is to preserve the entanglement. Primarily for quantum communications and possible medical applications.

##### Share on other sites

Mordred,

Oddly enough, the downward angle of BOTH entangled photons, shown in the diagram, was suggestive of my earlier speculation that a photon is not bound by the environment of the lab, but goes of in a direction independent of the lab. This based on the thought that two entangled photons might should go off in opposite directions. If they were bound to the lab, they would be shown diametrically opposed direction wise. Since they both angled down, that might be an indication that they headed off in exactly opposite directions, and the lab moved, making it look like they both angled down.

Regards TAR

Mordred,

When these experiments are done, do they log the geographical location and orientation of the lab? As well as the time of day and year?

Perhaps they should. That or design the experiment so the whole experiment stays pointing in the same vector direction that the Lab, Surface, Earth, Sun is going in, so that any apparent "off line" movement of photons could be easily accounted for.

There might even be a way to harness some energy from such a situation, should it be well understood, since you could easily construct a device that pointed in the direction of our movement through the void, and another that pointed away, in such a manner as to create some differencial that could be taken advantage of.

Regards, TAR

##### Share on other sites

Mordred,

Oddly enough, the downward angle of BOTH entangled photons, shown in the diagram, was suggestive of my earlier speculation that a photon is not bound by the environment of the lab, but goes of in a direction independent of the lab. This based on the thought that two entangled photons might should go off in opposite directions. If they were bound to the lab, they would be shown diametrically opposed direction wise. Since they both angled down, that might be an indication that they headed off in exactly opposite directions, and the lab moved, making it look like they both angled down.

Regards TAR

Mordred,

When these experiments are done, do they log the geographical location and orientation of the lab? As well as the time of day and year?

Perhaps they should. That or design the experiment so the whole experiment stays pointing in the same vector direction that the Lab, Surface, Earth, Sun is going in, so that any apparent "off line" movement of photons could be easily accounted for.

There might even be a way to harness some energy from such a situation, should it be well understood, since you could easily construct a device that pointed in the direction of our movement through the void, and another that pointed away, in such a manner as to create some differencial that could be taken advantage of.

Regards, TAR

That sounds like a search for absolute motion.

## Create an account

Register a new account