Building an Optical Cavity

[Limecat]

I just realized not too long ago that I can kill off higher transverse modes by sending a laser into an optical cavity of sufficient length, leaving me with just TEM00 mode (assuming the mirrors are relatively free of imperfections).

Anybody ever built one?

Namely, I am interested in the reflectivity of the mirrors required. I'm interested because as above, I would like the means to produce semi-clean gaussian beam profiles

 

[HIMNL9]

What type of optical cavity do you have in mind ?

 

[Limecat]

What type of optical cavity do you have in mind ?

Confocal in all likelihood. I've only observed a hemispherical one before. Basically I just want to use it to filter out higher order transverse modes. Its a little cheaper than using spatial filtering, plus I can employ the mirrors for other things too. Like I said I was mostly curious about the limits of reflectivity that I need. Thorlabs sells mirrors that have silver ~99+% reflectivity or <95% reflectivity with aluminum. I was curious if the 95% (or lower depending on wavelength) was enough and if there are any issues sending the beam into the mirrors.

In the optical cavity I have used once in a lab, one or both of the mirrors were from a HeNe laser, so I'm not sure where I'll get a mirror that will accept a beam from both directions.

 

[HIMNL9]

The main problem can be that, usually, those mirrors (i mean single layer reflective metalization ones) are reflective in both the directions ..... as far as i know, there's no way for make a mirror that reflect only in one (except ofcourse dichros for different wavelenghts), so expect some loss at incoming side .....

But, honestly, i never buyed mirrors from Thorlabs, so don't know all their products ..... i've looked at their site, but apparently they don't have listed any back-reflection specifications for their products ..... also, is better that you ask them first, cause for that what i can see in some of their site images, some of these mirrors looks transparent on the backside, but some others looks "sanded glass" on the back side .....

There are also dielectric mirrors, can be found both "broadband" and "tuned" for specific wavelenghts (usually old He-Ne mirrors was almost all dielectric ones, and old ruby rods face mirrors was also dielectric tuned ones), but i think they have a higher cost, due to the layering process .....

 

[Limecat]

Ahh. So where do you buy your mirrors?
Also, I know there would be some degree of back reflection.

 

[HIMNL9]

Well, i usually recover almost all my mirrors and lenses from mangled apparates ..... til now, i needed only a pair of times to buy something specific, and i found them on ebay ..... actually i'm searching a pair of curved mirrors (cylindrical), and a grating mirror, cause i want to build a monochromator, but til now i've found nothing ..... perhaps also Thorlabs don't have them .....

 

[Limecat]

Edmund Optics sells cylinder mirrors. Just the two though. Looks like they sell some sort of grating mirror too.
But yeah I tried looking at the dielectric ones and I'm not sure if those will do the trick haha.

 

[HIMNL9]

I see ..... but, apart their outrageous prices for gratings p), they have only two small cylindrical mirrors with short FL ..... for a decent "classis style" monochromator, i need a pair of square or rectangular mirrors, with a FL from 100 to 150 mm, and possibly large at least 50 or more mm ..... i think i probably have to reconsiderate the project, or to invent something new myself .....

 

[LSRFAQ]

Hum, your looking for a "Fabry Perot"

Since I am phasing out posting over here, you need to find and download

PSST! laser software from the University of St Andrews. It has the cavity simulator you need.

To get a single mode, you'll need a focusing lens to launch into the cavity, mirror reflections will be at least 90%, probably 95. Dont worry, the mode you want and perhaps a few modes you dont, will transmit with very low loss.

There is a tradeoff between finesse, loss, Q, and # of modes transmitted. Modes will be spaced N megahertz apart.

http://www.repairfaq.org/sam/laserlia.htm#liacisf

It is actually a easy experiment. When you get close to aligning the mirrors, you will see a lissajous pattern of dots moving around, then it they will suddenly merge into one dot.

Before you buy many expensive optics, make sure you understand single frequency does not mean you will get a tem00 beam out.

If your trying to clean up a diode, google Littman=Metcalf cavity + tunable + redlum

Good luck.

Steve

 

[Limecat]

Hum, your looking for a "Fabry Perot"

Since I am phasing out posting over here, you need to find and download

PSST! laser software from the University of St Andrews. It has the cavity simulator you need.

To get a single mode, you'll need a focusing lens to launch into the cavity, mirror reflections will be at least 90%, probably 95. Dont worry, the mode you want and perhaps a few modes you dont, will transmit with very low loss.

There is a tradeoff between finesse, loss, Q, and # of modes transmitted. Modes will be spaced N megahertz apart.

Sam's Laser FAQ - Laser Instruments and Applications

It is actually a easy experiment. When you get close to aligning the mirrors, you will see a lissajous pattern of dots moving around, then it they will suddenly merge into one dot.

Before you buy many expensive optics, make sure you understand single frequency does not mean you will get a tem00 beam out.

If your trying to clean up a diode, google Littman=Metcalf cavity + tunable + redlum

Good luck.

Steve

Actually I've done the experiment in lab. I observed cylindrically symmetric transverse modes via a beam profiler. I'm going after the fundamental gaussian TEM00 mode, hence a longer cavity length.

 

[Bluefan]

As Steve said, Fabry Perot cavities filters longitudinal modes, probably not transversal modes.
Thorlabs sells decent stuff, for high quality stuff you go to CVI Melles Griot, although for hobbyist Thorlabs will be more than good enough.
Note that metallic mirrors don't transmit anything, so they won't be usefull for a FP cavity.

 

[Limecat]

As Steve said, Fabry Perot cavities filters longitudinal modes, probably not transversal modes.
Thorlabs sells decent stuff, for high quality stuff you go to CVI Melles Griot, although for hobbyist Thorlabs will be more than good enough.
Note that metallic mirrors don't transmit anything, so they won't be usefull for a FP cavity.

Thanks, I'll keep that in mind.

However, about the filtering, I did in deed see a roughly TEM00 mode on the beam profiler with the cavity at some 25cm in length, vs many transverse modes while the cavity was shorter