Single photon question?

sbsquarepants

Apologies in advance if this is a silly question!

If a light source emits a single photon - what determines the direction that photon travels in? In other words if, somewhere in the blackness of space, a single photon is generated and it could in theory head of in more or less any direction what is the mechanism that actually aims it somewhere in particular?

Labarbapostiza

There isn't a mechanism that determines it will be detected from any particular direction. There's just a probability it will be detected at some point. It appears to travel in a straight line from the point of emission to the point it's detected, because a straight line is the path of least time. If you sum all the other paths it could take, you get a straight line, which is the shortest path from the point of emission.

To put it simply. The photon is in lots of different places, but it will only be detected at one point; then the photon vanishes from everywhere else it was in space.

sbsquarepants

Cheers Labarbapostiza. I get the wave of probability part, but my question is altogether more stupid eh, basic.
What I mean is when the photon is produced and this wave comes into existence does it propagate as a sphere (well within reason, whatever created it must block it's path to some degree) or is there some mechanism that aims it towards a particular spot somehow? Obviously (I think anyway) if you just stick 2 or 200 or 2 million detectors at random locations the probability of the photon hitting any one in particular is not uniform? If it does hit one, why that particular one, why not it's neighbour or one in the complete opposite direction?

dlouth15

Have a look at this answers in this thread:

http://www.fnal.gov/pub/science/inquiring/questions/radiationdir.html

Whilst the precise direction of the photon emitted will be random, there is a probability distribution and this depends on the nature of the electron transition i.e. what orbitals are involved.

So it's not completely random apparently.

Labarbapostiza

sbsquarepants wrote: »

What I mean is when the photon is produced and this wave comes into existence does it propagate as a sphere (well within reason, whatever created it must block it's path to some degree) or is there some mechanism that aims it towards a particular spot somehow?

A major point about all this. If you're not confused by it, you don't understand it. Difference between classical mechanics and quantum mechanics is such, that even using the term mechanism puts you on thin ice.

When the photon is produced. it propagates like a sphere, growing at the speed of light, from the point it was produced, which is an electron releasing it.

Before the wave nature of photons was observed (or before even the concept of photons was conceived) light was observed to travel in straight lines. Then once you do any of the experiments which reveal the wave nature of light; like the twin slits experiment. Then you see light travels in every direction.

The straight line rays that we see when we observe light in a particular way is caused by summing, adding up all the possible paths and we get a straight line. Imagine if you had a magical ball, and you threw it at a wall, but every time you threw it, it would take a completely curly random path to the wall and back. If you repeated the exercise long enough, and then added up all the paths (some will be negative and others positive), you get a straight line.

Obviously (I think anyway) if you just stick 2 or 200 or 2 million detectors at random locations the probability of the photon hitting any one in particular is not uniform? If it does hit one, why that particular one, why not it's neighbour or one in the complete opposite direction?

If you randomly space the detectors, depending where the detectors are they will have different probabilities of detecting the photon than others. Some detectors will have the same probability as others; such as two detectors who are exactly the same distance from the point of origin as the photon.

From observances of the light wave, the only thing we can tell about it, is that the wave is a probability distribution. If you have a detector at some point where the wave is high, there's a higher probability of the photon being detected, where the wave is low, a lower probability of the photon being detected. If two detectors have equal probability it will be detected at one, or the other, but not both. There are two probabilities that are not allowed; the probability of 1; which is the absolute certainty of a photon being in a particular point in space, and the probability of 0, which is the absolute certainty of it not being in a particular point in space. The answer to the question "What the hell is actually going on?", is that nature at it's most fundamental level does not have a deterministic mechanism, and instead it is composed of a near infinite number of coin flips and dice rolls, that add up in such a way as to give us the appearance of the classical mechanics world, where everything appears deterministic.

sbsquarepants

Labarbapostiza wrote: »

A major point about all this. If you're not confused by it, you don't understand it..

If only the reverse of that statement were true!

What I'm getting stuck on is this.

I can (shakily) grasp the probability wave concept. Technically the world is our photons oyster it could go anywhere and everywhere, and indeed it seemingly does, until we go all paparazzi on it, as per the 2 slits experiment.
But as you said 2 or more detectors could have the exact same probability of detecting it but only 1 can. My question is why that one? There has to be a reason, doesn't there?

As per the 2 slits experiment, when the detectors are present the photon will go through only 1 slit or the other. I know THAT it behaves that way but I have no idea WHY. What makes it "choose" slit one over slit 2.
I suppose in many ways it's the same question as why not both? Or neither?

dlouth15

sbsquarepants wrote: »

If only the reverse of that statement were true!

What I'm getting stuck on is this.

I can (shakily) grasp the probability wave concept. Technically the world is our photons oyster it could go anywhere and everywhere, and indeed it seemingly does, until we go all paparazzi on it, as per the 2 slits experiment.
But as you said 2 or more detectors could have the exact same probability of detecting it but only 1 can. My question is why that one? There has to be a reason, doesn't there?

As per the 2 slits experiment, when the detectors are present the photon will go through only 1 slit or the other. I know THAT it behaves that way but I have no idea WHY. What makes it "choose" slit one over slit 2.
I suppose in many ways it's the same question as why not both? Or neither?

The answer (at least according to one interpretation) is that it is the act of measurement itself that forces the system into one state. The photon only goes through one slit only because we have set up our detectors to measure the system in that way.

You get the same issue if you try to measure the position of an electron at a certain energy in, for example, an atom. Quantum mechanics can predict the probability of finding the electron in any given region, but it can't predict that we will find it in that particular region. Only act of measurement forces the electron to be either in the region or not.

This is known as the measurement problem in quantum mechanics. Although various interpretations have been proposed, the problem never fully goes away. Maybe the problem is with our insistence that reality must be a certain way when in fact it is not.

sbsquarepants

Now again, this may be just displaying my ignorance, but if, when we look, we in fact find our photon to be in only one location in space does that not rule out the possibility that if ever could have been in some of the others?
This would be so much easier to explain if I could draw a diagram!

For example, photon is created at a certain spot and propagates spherically for x amount of time, lets say 1 year. It is then detected at a certain spot on the surface of our sphere which has a 1 light year radius. To get to that spot by any route other than a straight line would take more than 1 year or involve faster than light travel surely? Does that not imply that it wasn't actually anywhere else other than along that straight line. It could have been - but it wasn't! I could have been out for dinner last night for all you know, but in fact I wasn't. Now that you know I wasn't, no other possibilities exist.

I think i'm starting to see the appeal of " god made that way, so that's how it is":D

dlouth15

sbsquarepants wrote: »

Now again, this may be just displaying my ignorance, but if, when we look, we in fact find our photon to be in only one location in space does that not rule out the possibility that if ever could have been in some of the others?
This would be so much easier to explain if I could draw a diagram!

For example, photon is created at a certain spot and propagates spherically for x amount of time, lets say 1 year. It is then detected at a certain spot on the surface of our sphere which has a 1 light year radius. To get to that spot by any route other than a straight line would take more than 1 year or involve faster than light travel surely? Does that not imply that it wasn't actually anywhere else other than along that straight line. It could have been - but it wasn't! I could have been out for dinner last night for all you know, but in fact I wasn't. Now that you know I wasn't, no other possibilities exist.

I don't know too much about the "sum over histories" formulation, but as far as I'm aware, the non-straight paths cancel each other out and only the straight line path remains. So long as the theory doesn't violate the idea that useful information can't travel faster than c, it is OK. It predicts then that the photon will only be detected in a spherical shell of one light year radius. Outside or inside this shell, the probability of detecting it is zero.

It is still only a probability distribution however. We can't say where on the shell the photon will be detected although we can calculate the probability that it will be detected in any defined region.

I think i'm starting to see the appeal of " god made that way, so that's how it is":D

Yes, I think at present you have to take it that this is the way things are, funny as it seems. Quantum mechanics is still very powerful and explains a whole lot such as the energy levels of atoms and the shape of electron orbitals, why gold is a yellow colour and so on.

It also predicts that on the ordinary human scale, we don't generally see the strangeness which is possibly why we haven't evolved to deal with it and so find it strange.

Labarbapostiza

sbsquarepants wrote: »

For example, photon is created at a certain spot and propagates spherically for x amount of time, lets say 1 year. It is then detected at a certain spot on the surface of our sphere which has a 1 light year radius. To get to that spot by any route other than a straight line would take more than 1 year or involve faster than light travel surely? Does that not imply that it wasn't actually anywhere else other than along that straight line. It could have been - but it wasn't! I could have been out for dinner last night for all you know, but in fact I wasn't. Now that you know I wasn't, no other possibilities exist.

Yes, that is a good question, and you can see by the size of your sphere, quantum mechanics is not simply the science of the very small.

The probability distribution is usually called the wave function. When an electron is moved in the detector, by a photon, we'd normally say the photons wave function has collapsed. But before it collapses, the photon has the probability of ringing a detector somewhere else.

Think about it more, and read up on the twin slits experiment. It's even better to look at the twin slits experiment done with electrons. The experiment demonstrates the electron passing through both slits, and then hitting the detector in only one place. (I have a pet theory, which I think is pretty good but won't state here. That accounts for wave function collapse; but don't hold your breath for nice sounding answer; I believe it has to do with Quantum Vacuum Dynamics)

sbsquarepants

Don't suppose there's a quantum mechanics for dummies book on amazon is there!

Maybe I'm trying to compare apples and oranges, or I just don't understand probability but this is the way I look at it.

Say you play the lotto - before the draw you can't know how many numbers you'll match if any so you say you have a probability of winning of lets say 1 in 10 million for arguments sake. After the draw you look and see that you have actually matched all 6. When you sober up (a few months later!) you then could say your probability of winning was actually 1, you did win so it had to be in retrospect - you just couldn't know that at the time.
Is that not the same thing, probability only exists as a concept when trying to make predictions, when you have an actual event to describe it's probability is 1 by default and therefore the probability of any other event was 0. You can tell that looking back but not looking forward.
Therefore the photon wasn't in "reality" (I'm not sure that word has much meaning by the way!) actually anywhere else at all, despite the possibility that it could have been had things been different somehow. No?

dlouth15

sbsquarepants wrote: »

Don't suppose there's a quantum mechanics for dummies book on amazon is there!

Maybe I'm trying to compare apples and oranges, or I just don't understand probability but this is the way I look at it.

Say you play the lotto - before the draw you can't know how many numbers you'll match if any so you say you have a probability of winning of lets say 1 in 10 million for arguments sake. After the draw you look and see that you have actually matched all 6. When you sober up (a few months later!) you then could say your probability of winning was actually 1, you did win so it had to be in retrospect - you just couldn't know that at the time.
Is that not the same thing, probability only exists as a concept when trying to make predictions, when you have an actual event to describe it's probability is 1 by default and therefore the probability of any other event was 0. You can tell that looking back but not looking forward.
Therefore the photon wasn't in "reality" (I'm not sure that word has much meaning by the way!) actually anywhere else at all, despite the possibility that it could have been had things been different somehow. No?

I think that is basically a correct understanding of the Copenhagen interpretation. The act of measuring is equivalent to the drawing of the lottery in this analogy. The only thing I would say is that the photon is in "reality" it is just that its position is uncertain prior to measurement.

Labarbapostiza

sbsquarepants wrote: »

Don't suppose there's a quantum mechanics for dummies book on amazon is there!

There is. But it's not very good. It's aimed at college students who are doing a course in quantum mechanics. If you're doing this in college, you'll have maths class which will cover the requisite maths. Otherwise it looks like meaningless gibberish.

Some text books can be quite snotty, in not telling you something that a tutor would tell you in passing, but without which you'd be perplexed. Like the reason the probability is given by the square of the wave function result (It's because it's a complex number - one part is a real number like 1, 2, 3, the other part is an imaginary number, that is a number with a factor of the square route of minus one. Squaring it, multiplying it by itself, makes the imaginary numbers vanish, and gives you the probability as a real number)

Sensay

A photon chooses one slit over the other because it has the invisible quark stamp on it