Lasers (again!)

[Dmills]

I'm Interested here, I understand what you are saying, the laser generates light at 1064nm and by using a crystal actually emits the harmonic of this at 532nm.

 

However, I was under the impression that harmonics sat "On top" of the fundamental frequency (in this case 1/1064nm) Does this not mean that the actual light oputput is 1064nm AND the 532nm that we are actually interested in? - In fact the IR component will be somewhat higher, as it is the fundemental?

 

What actually happens is this:

 

A (in small stuff) laser diode at 808nm is used to pump the yag to lase at 1064nm (often the pump beam from the diode is admitted via a 1064nm reflective coating on rear of the laser crystal), the SHG crystal is installed intracavity (inside the 1064nm laser cavity), and the output mirror is (theoretically) totally reflective at 1064 and transmissive at 532. The 1064 leakage is due to imperfect coating on the output mirror.

 

The doubler needs to be intracavity because it needs massive optical power density in order to produce a reasonable amount of green.

 

These small systems often epoxy the doubler right onto a thin highly doped slice of laser crystal and coat the back of the laser crystal transmissive 808, reflective 1064, the front of the doubler is AR coated for both 1064 and 808. Finally a concave mirror reflective 1064, transmissive 532 is mounted in front of the assembly to form the laser cavity. Pump light is applied to the rear or the laser crystal via a simple lens (or something more complex in larger systems) from the diode laser.

 

In higher power systems an additional mirror (or two) is used to allow the green from both passes thru the multiplier to be extracted and to put a beam waist inside the KTP for better output. Temperature control of the multiplier is highly desirable for phase matching and a polarizing brewster plate intracavity will polarise the 1064 generation to optimise the SHG generation.

 

You can get a large increase in average power by running a Q switch, but that results in very high peak powers which can be hard on the optics (grey tracking in the multiplier), cannot be audience scanned and does not result in good graphics.

 

The canonical reference for this stuff is "Solid state lasers" by Koechner.

 

Just thought about it, as I guess that the IR component isn't a huge problem, as long as there is visible light along with it to initiate the blink response?

 

Except that for various reasons the IR beam can end up at a different angle to the main beam.

[JimWebber]

Thanks for that Dan! That is a pretty comprehensive description! I understood a few words in there

 

So If I understand it, there is a mirrored surface, that passes the visible light, and reflects the IR? - In theory...

 

It all sounds very complicated, and I'm going to go back and play with my train set...

 

Choo Choo...

 

Jim

[Dmills]

So If I understand it, there is a mirrored surface, that passes the visible light, and reflects the IR? - In theory...

 

Yea, the 1064 theoretically circulates between the rear (plane) mirror, thru the amplifier crystal then the SHG crystal and is then reflected by the concave mirror back again. Even in a small laser the optical power in this internal beam can be very large, and it doesn't take much transmission thru that output coupler at 1064 to give a fair amount of IR leakage.

 

The green light is produced by the shg crystal and escapes thru the concave mirror, with the green going the other way being wasted by being absorbed by the laser crystal.

 

All this applies to small units, something else is done in the big stuff.

 

Regards, Dan.

[jfitzpat]

Except that for various reasons the IR beam can end up at a different angle to the main beam.

 

It seems that because both wavelengths pass through a two plane substrate they would have to be at different angles, regardless of rather or not the second harmonic generation is type I or type II.

 

JimWebber: I found this article online: http://www.rp-photonics.com/frequency_doubling.html

 

But I don't think it is very clear. Reference 1 and 2 (especially 2) listed at the end are excellent if you are interested in the phenomena. I haven't read Dmill's reference, but judging from his knowledge I am sure it is excellent.

 

To be clear, Dmills is the expert. I have a background in lasers, but ION lasers ruled at that time. My understanding of nonlinear optics in these sorts of applications is purely theoretical. Some of the high resolution imaging work I did at Canon is related, but only in fundemental theory.

 

-jjf

[Dmills]

I have a background in lasers, but ION lasers ruled at that time. My understanding of nonlinear optics in these sorts of applications is purely theoretical. Some of the high resolution imaging work I did at Canon is related, but only in fundemental theory.

 

Ion (specifically Ar/Kr mixed gas) still rules if you care about pretty!

This solid state rubbish only wins on size, weight, power and cooling!

 

For a real display done right, give me a handful of I90s running mixed gas with broadband optics and PCAOMs, possibly backed up with a big copper vapour (for aerial beam effects).

 

Ok, so that is well over a hundred amps of three phase input (effectively all of which ends up as heat in the cooling water), but ooh man, pretty!

 

Regards, Dan.

[jfitzpat]

For a real display done right, give me a handful of I90s running mixed gas with broadband optics and PCAOMs, possibly backed up with a big copper vapour (for aerial beam effects).

 

I do hate that 532 green, though a friend calls it 'the color of money' (easy, no fuss gigs). Since he once was on a tour with me and in charge of 8 (yes 8) Spectraphysics 171s, I figure he has probably earned a few easy gigs. I did see four 171 purelights on an arena show last year and it looked fantastic. Personally, the affordability of a mixed gas 'white light' and acusto optic modulation to put that color to use, strikes me as the biggest real changes since LaserMedia and G120PD scanners ruled the earth.

 

-jjf