Why do the photons produced by a laser have the same phase?

[mrdopebunny]

Hi. This has been destroying my mind all day.

Take, for example, a semiconductor laser with no internal photons initially. By spontaneous emission, one electron falls from the conduction band to the valance band and emits a photon with the bandgap frequency and a random phase. As this photon stimulates the emission of others, they all have the same phase as the original and due to the gain from the optical cavity and the medium, a standing wave develops.

My confusion is that, since spontaneous emission never stops, another photon of that same frequency but a different random phase can be emitted. This photon can stimulate emission in the same way as the original, developing another standing wave with a different phase. Nothing in the gain medium or the optical cavity seems to be phase selective, so why wouldn't the output of a laser be a mixture of photons of random phase?

 

[alxm]

Well, your initial reasoning is correct here, and there is such a thing as http://en.wikipedia.org/wiki/Random_laser" , which doesn't have a coherent beam.

What you're missing is how the optical cavity works. You have reflection at both ends, so you have boundary conditions that 'select' the phase which has nodes at the ends, and thus resonate in the cavity. Which means rapid amplification for that phase, whereas all other phases quickly get washed out due to destructive interference.

 

[DrDu]

There is another point: The probability stimulated emission is proportional to the number of photons already in the mode. So there is a tendency for spontaneous emission to produce a photon which coincides with the mode already most populated.

 

[orenp]

I believe the main issue is that both spontaneous emission and stimulated emission processes need inverted population that is ready to drop from the higher state to the lower. Now these two processes compete on the same resources (they use the same gain) but with the difference that stimulated emission gives the new photon the same phase of the stimulator so the number of photons with the same phase is exponentially growing, while spontaneous emission gives it a random phase. The number of photon traveling in the cavity that were stimulated grows far beyond the number of the spontaneously emitted ones, given that we let this process enough time (that is we don't let the stimulated photons out of the cavity too fast = we control the losses).