In a stacked laser diode array, multiple beams are emitted parallel to each other. These beams can be very close to each other, which should results in interference patterns. How does this setup differ from the scenario when light travels through an aperture with multiple slits, producing a large diffraction pattern behind it (as in the picture below)?
$\begingroup$
$\endgroup$
1
-
$\begingroup$ It's the same thing, the stacked lasers, each of them will produce an outgoing wave which has a curved wavefront resulting in interference in the same way as with the diffraction grating where from each slit a spherical wavefront emerges due to diffraction, resulting in the interference of the different spherical wavefronts of the different slits. $\endgroup$– SimonJan 8, 2020 at 18:44
Add a comment
|
3 Answers
$\begingroup$
$\endgroup$
8
The difference is that unless the laser designer took special care to make it so, the output of the lasers in the array will not be mutually coherent. (Also, doing so would generally require making the laser array on a single chip, rather than stacking multiple chips as described in your linked document)
Therefore the interference pattern they produce will be constantly changing, on a time scale too fast for you to observe the interference by eye.
In the case described in the document you linked (multiple laser devices used together to produce high power), it wouldn't be desirable to have mutually coherent output (precisely because it would produce an interference pattern) so they certainly haven't gone to any special trouble to make them coherent.
-
$\begingroup$ This doesn't make sense. Making the sources incoherent doesn't make the diffraction less severe. And in any case, I doubt that it's true that they're not coherent with each other. A typical solid-state laser has a coherence length of centimeters. $\endgroup$– user4552Jan 8, 2020 at 22:35
-
2$\begingroup$ @BenCrowell, the question is, if you have 5 lasers in an array, why doesn't it make the same diffraction pattern as a plane wave illuminating a 5-slit aperture. The 5 lasers are totally separate sources on 5 separate chips. There's no reason for them to be coherent with each other. More likely, their center frequencies are at least 10's of MHz apart, and quite likely they are GHz apart. $\endgroup$ Jan 8, 2020 at 22:47
-
$\begingroup$ @BenCrowell I would agree with you, diffraction always occurs. Even if we could somehow make the sources coherent with each other, the sources are still separate geometrically/spatially and will not behave like one light source. I think we would still have 5 distinct wave functions each doing their own thing. Although in thinking about it deeply if we do somehow make them coherent with each other are we effectively making one light source? $\endgroup$ Jan 18, 2020 at 0:49
-
$\begingroup$ @PhysicsDave, but they will diffract to form (roughly) the pattern shown as "single slit envelope" in OP's image, not the 5-slit diffraction pattern. $\endgroup$ Jan 18, 2020 at 0:51
-
$\begingroup$ Yes 5 overlapping single slit patterns. Any thoughts on if the coherency is somehow shared then we are effectively making a singe light source .... $\endgroup$ Jan 18, 2020 at 0:55
$\begingroup$
$\endgroup$
I see three possible reasons to explain why you do not see the interference pattern shown in your picture.
1) The laser diodes are not coupled and hence there is no reason why the laser diodes should be in phase. The pattern in your picture is created if the slits are illuminated by a plane wave which means that the "light sources" represented by the slits are in phase. Any phase offset between the light sources will create a different interference pattern but will still differ from the interference pattern of a single slit. The single wavefront equivalent would be an illumination of the multiple slits with a warped wavefront.
2) Laser diodes are not stabilized and are therefore prone to mode hops (which are accompanied with phase changes) and wavelength shifts in combination with phase shifts. This temporal phase instability between the different light sources creates varying interference patterns (see reason 1) which might change faster than the eye can see. An example of fast mode hopping is e.g. given here on page three. The single slit interference is not influenced by the phase shifts (the wavefont is still be assumed as constant within the width of the slit). Therefore, in the extreme of fast random phase shifts, the observer will see the interference pattern of a single slit only.
3) Only light with identical wavelengths can create temporally stable interference. A set of laser diodes have usually slightly shifted spectra. Therefore, only the overlapping spectrum of the different light sources produces a multi-slit equivalent interference (and single split interference) and the non-overlapping spectrum single-slit interference only.
$\begingroup$
$\endgroup$
1
Diffraction does occur. Diffraction is one of the effects that limits how well collimated a laser beam can be.
-
$\begingroup$ The question isn't about the beam divergence due to the diffraction at a single aperture, it's about whether 5 lasers next to each other would produce the same multi-peaked pattern as a plane wave passing through 5 slits. $\endgroup$ Jan 8, 2020 at 22:48