Spontaneous Emission and Amplifier Noise

Posted on 2006-07-16. Permanent link: http://www.rp-photonics.com/spotlight_2006_07_16.html

It is known that laser amplifiers exhibit some level of excess noise, which is often explained as the effect of spontaneous emission. Also, it has been reported that under some circumstances it is possible to largely suppress spontaneous emission. This has e.g. lead to thresholdless lasers. So shouldn't it be possible to somehow suppress a laser amplifier's excess noise?

Spontaneous emission generates radiation in all modes with suitable frequencies. In an amplifier, only a very small fraction of the power of spontaneous emission usually goes into the amplified mode, but this is what causes the excess noise. The emission going into all the other modes constitutes nearly all of the power loss, but as it can be separated from the amplified light, it is not relevant in the context of amplifier noise.

Now one can in principle surround the amplifying medium by a structure (e.g. a microcavity) which modifies the mode distribution so that there are fewer modes with frequencies which are suitable for the spontaneous emission. In this way one can reduce the power loss, but one cannot avoid emission into the amplified mode, and thus not reduce the amplifier noise.

There are even fundamental quantum-mechanical reasons why the excess noise of any phase-insensitive optical amplifier – whatever physical principle it may be based on – can never be lower than that of an ideal four-level laser amplifier, or of a degenerate optical parametric amplifier. (Quasi-three-level amplifiers don't even reach that limit.) So we see that any type of amplifier not exhibiting spontaneous emission must have some other noise influences which generate at least the same level of noise.

The only way around this limitation is to use a phase-sensitive amplifier, such as a degenerate parametric amplifier, which can in principle operate without any excess noise.