Why is a laser's beam parallel?

[fxdung]

If we do not consider diffraction,why lasers rays are parallel?Do atoms stimulatedly emission photons in same direction?It seems to me stimulated emission photons have same frequancy but random in direction?

 

[Dale]

Have you read anything about how lasers work? If so, what did you learn? Does laser light produce randomly oriented light? What did your initial research say?

 

[fxdung]

I have learned about stimulated emission in lasers. I know that laser is nearly monochromatic.But I do not know whether there is a random in direction in stimulated emission or there is only one direction in stimulated emission.

 

[DaveE]

Did you look at the Wikipedia page for "stimulated emission"? Read the first 3 sentences.

 

[fxdung]

I do not understand why stimulatedly emitted photon has the same phase and direction as the incident photon?

 

[DaveE]

I do not understand why stimulatedly emitted photon has the same phase and direction as the incident photon?

Me either. It seems like magic, doesn't it?

Maybe a real physicist will try to explain it, but I am sure it will be over my head. Definitely not beginner material. A good foundation in quantum mechanics is required at a minimum.

 

[atyy]

https://www.edmundoptics.com.sg/knowledge-center/application-notes/lasers/laser-resonator-modes/
"The shape of a laser beam is determined by the resonator cavity in which the laser light is amplified in a gain medium. Laser resonators are typically formed by using highly reflective dielectric mirrors or a monolithic crystal that utilizes total internal reflection to keep light from escaping ..."

https://en.wikipedia.org/wiki/Optical_cavity
Optical cavity

https://patents.google.com/patent/US7492805B2/en
Scalable spherical laser
Ronald LaComb, Sallie S. Townsend

 

[Dale]

I have learned about stimulated emission in lasers. I know that laser is nearly monochromatic.But I do not know whether there is a random in direction in stimulated emission or there is only one direction in stimulated emission.

What source did you read that didn’t mention the direction? Perhaps you should use a better source like Wikipedia

 

[Vanadium 50]

Perhaps you should use a better source like Wikipedia

He did and it is us.

 

[atyy]

I do not understand why stimulatedly emitted photon has the same phase and direction as the incident photon?

A good foundation in quantum mechanics is required at a minimum.

https://www.mpl.mpg.de/fileadmin/user_upload/LectureNotes.pdf
Quantum Physics of Light-Matter Interactions
Claudiu Genes

See Eq 1.40.
"The first one is a stimulated emission process where the first photon stimulates the emission of a second photon in exactly the same direction and with the same polarization as the first one. The second term is a spontaneous emission event where a photon is emitted in a random direction with any polarization."

 

[PeterDonis]

It seems to me stimulated emission photons have same frequancy but random in direction?

Heuristically, no, we would not expect this. Stimulated emission means the emitted photon is in the same state as the photons already present (because it comes from a property of bosons that the amplitude for them to be in the same state is enhanced). But "the same state" includes the photon's momentum, not just its energy; in other words, it includes the direction.

 

[fxdung]

Why the stimulated emission polarization of photon is the same as incident photon?And why stimulated emission photon is in phase with incident photon(Genes only explains the two photon have same direction)?

 

[anuttarasammyak]

If we do not consider diffraction,why lasers rays are parallel?

My intuitive thought is that photons go and back between two parallel mirrors many times with increasing their numbers. From this two-way route some of them escape beyond one mirror to form a straight beam.

 

[DaveE]

My intuitive thought is that photons go and back between two parallel mirrors many times with increasing numbers. From their route axis some of them escape to form a beam.

The effect of the laser resonator is really a different thing than stimulated emission. By providing a structure that selects a particular direction of radiation to send back to through the gain medium, the resonator just makes almost all of the photons that cause stimulated emission to be coherent.

But, on an atom-by-atom basis, any photon could stimulate emission, whether "on axis" (laser wise) or not.

There are basically three four things going on in your average laser:
1) Stimulated emission, which all excited can atoms do.
2) Population inversion, created by putting energy into a select group of atoms/molecules that can do this (some can't). This means that a passing photon is more likely to stimulate emission, and be amplified, than be absorbed. This is the genesis of "laser gain".
3) A resonator structure to provide feed back of desirable photons to be amplified. This increases the probability that the photons that do stimulate emission of other photons are the ones you want the laser to produce.
4) About a million other things that explain why lasers today are are better than lasers were 20 years ago, which were better than lasers from 20 years before that...

BTW, IIRC, astronomers have identified "laser" amplification of light traveling through excited gas in one pass, with no resonator. Light Amplification by Stimulated Emission of Radiation, doesn't have to have a resonator, even though >99.999% of the ones you see on Earth do.

edit: Also, there are Laser like machines that are called lasers, but don't use stimulated emission and population inversion. Things like free electron "lasers" and x-ray "lasers" that do produce coherent radiation like a laser, but with different physics. I'll choose to be pedantic and say they can't be LASERs because the "SE" is missing. Perhaps they should be called "LARs" or perhaps "LAs" or "ARs", since light and radiation are kind of redundant.

 

[PeterDonis]

Why the stimulated emission polarization of photon is the same as incident photon?

Same answer as I gave in post #11: the polarization is part of the photon's state, and the stimulated emission photon is emitted in the same state as the photons that are already present.

 

[Drakkith]

Same answer as I gave in post #11: the polarization is part of the photon's state, and the stimulated emission photon is emitted in the same state as the photons that are already present.

Within the medium, is there a particular axis along which most of the photons generated by spontaneous emission move? Or is it mostly random?

 

[PeterDonis]

Within the medium, is there a particular axis along which most of the photons generated by spontaneous emission move?

AFAIK spontaneous emission photons would be emitted with random momentum and polarization. However, I don't think spontaneous emission is significant in a laser compared to stimulated emission.

 

[Drakkith]

AFAIK spontaneous emission photons would be emitted with random momentum and polarization.

So then is it the cavity that causes the photons to get set up in a 'back and forth' manner along the axis that leads out of the laser? I guess my thinking was that the photons that don't get generated along this axis are quickly absorbed by the medium/cavity walls, leading to an avalanche effect where the on-axis photons end up stimulating the emission of the vast majority of photons by virtue of surviving longer than the off-axis ones.

Does that make sense?

 

[DaveE]

that the photons that don't get generated along this axis are quickly absorbed by the medium/cavity walls, leading to an avalanche effect where the on-axis photons end up stimulating the emission of the vast majority of photons by virtue of surviving longer than the off-axis ones.

Yes. The photons that start in the correct direction get fed back through the gain medium and are amplified by SE. Then there are a bunch more photons going in the correct direction and phase that also get amplified, etc. The gain is big for the desired photons and below 1 for the others (way below 1). So the ones that grow in numbers take over the distribution. For the transitions of interest the excited state doesn't have much spontaneous emission (long upper state lifetime) compared to the huge number of circulating photons. A typical output coupler is maybe 1-10%, which means a 1W laser output has 10-100W bouncing back and forth in the resonator constantly causing depopulation of the excited state by stimulated emission.

Some lasers that could support multiple longitudinal modes (like a Fabry-Perot etalon) will select the highest gain mode(s) by this sort of process. The photons with the best gain may take over the distribution. In any case if the round trip gain is less than 1 it won't lase.

edit: I said the gain is big, but that's because of multiple passes through the gain medium. Another way to think of the output coupling is that 1% output means an average photon makes 200 passes through the gain in a linear resonator. In practice the 1 pass gain isn't too much bigger than 1.

 

[atyy]

And why stimulated emission photon is in phase with incident photon(Genes only explains the two photon have same direction)?

They have the same state. But to see it in terms of electric field amplitude, one needs to take the state from Eq 1.40 and do something like Eq. 1.19 and Eq 1.20 in the notes by Genes. For the phase, I think there was a reference to a phase operator in one of your other threads, maybe see if you can do a similar calculation for the phase as for the electric field amplitude.

 

[vanhees71]

AFAIK spontaneous emission photons would be emitted with random momentum and polarization. However, I don't think spontaneous emission is significant in a laser compared to stimulated emission.

The spontaneous emission is one important source of the linewidth of the laser light. This minimal linewidth is inversely proportional to the lawer power, because the more intense the lase is the more the stimulated emission dominates over the spontaneous.

 

[Saw]

Yes. The photons that start in the correct direction get fed back through the gain medium and are amplified by SE. Then there are a bunch more photons going in the correct direction and phase that also get amplified, etc. The gain is big for the desired photons and below 1 for the others (way below 1). So the ones that grow in numbers take over the distribution. For the transitions of interest the excited state doesn't have much spontaneous emission (long upper state lifetime) compared to the huge number of circulating photons. A typical output coupler is maybe 1-10%, which means a 1W laser output has 10-100W bouncing back and forth in the resonator constantly causing depopulation of the excited state by stimulated emission.

This correct explanation that you give is the one that once gave me comfort as to why, if you shine two lasers upwards from the centers of two wagons in relative motion wrt each other that are momentarily side by side, they both hit the mid-points of the respective ceilings, whereas if the lasers point in the same direction, the beams will be traveling together. In the second case, the motion of the source does not affect the outcome, which is the same behavior in both cases (even if, due to te interplay of time and length, one can still hold that relativity holds in each wagon), but in the first case, the outcomes are actually different, as if the motion of the source did affect light's behavior.

The usual expalantion is: otherwise relativity would not work. But I felt a need for an explanation based on the physics of the mechanism.

This one gave me satisfaction.

At the root of the system you need spontaneous emission in random directions.

But then there is a sort of darwinian mechanism that ensures that photons end taking the direction in which the instrument, which does take the motion of the source, is always pointing.

The photons can only escape out of the instrument through the front hole, so only the ones have the correct direction along the axis of the instrument find there way out. It is survival of the fittest: photons that are not adapted to the environment (those crashing against the walls) are absorbed and die, those that are adapted (right direction) survive (escape out).

The stimulated emission trick only favor this process, because the photons endowed with surviving capacity reproduce into replicas of themselves.

Maybe one should add that another trick is that the hole has a half-silverly glass that is only transparent when many photons push through (more amplitude) and those big numbers only happen for photons moving to and fro along the axis.

I also heard that there are mirrors favoring the right direction.

In any case, these tricks are not perfect. Because the instrument is not infinitely thin or long, the laser is not totally directional and still suffers some divergence.

 

[Drakkith]

This correct explanation that you give is the one that once gave me comfort as to why, if you shine two lasers upwards from the centers of two wagons in relative motion wrt each other that are momentarily side by side, they both hit the mid-points of the respective ceilings, whereas if the lasers point in the same direction, the beams will be traveling together. In the second case, the motion of the source does not affect the outcome, which is the same behavior in both cases (even if, due to te interplay of time and length, one can still hold that relativity holds in each wagon), but in the first case, the outcomes are actually different, as if the motion of the source did affect light's behavior.

Uhh, I don't think his explanation has any bearing on this scenario. It just explains why the photons end up almost entirely on-axis of the laser.