Frequency Doubling: Typical Conversion, Input Power & CW Laser?

[Thermodave]

I'm trying to get an idea of how efficient frequency doubling methods are. I quick google search gives back everything from .01% to above 50%. What is the typical conversion from 1064 to 532? Is this common for other wavelengths/crystals? More importantly, why does it sound like the efficiency is related to input power? Can low power cw laser be doubled? Thanks for any info.

 

[Cthugha]

In nonlinear optics you develop the response P to a light field of amplitude E in a power series of the field amplitude:
P(t)=\chi^{(1)} E(t) +\chi^{(2)}E^2 (t) +\chi^{(3)}E^3 (t)+...

Second harmonic generation/frequency doubling is a second order process and therefore proportional to the square of the amplitude of the incoming field. Therefore its efficiency increases also quadratically with the amplitude.

The efficiency also depends strongly on what nonlinear crystal you use and which wavelength you are aiming at. There are good crystals cut for 800nm->400nm achieving up to 20% efficiency. I am sure the record values are way higher, but not representative.

As a consequence SHG works best for short pulses. SHG for cw is of course also possible but not very effective, even worse for low input powers. What input power and wavelength are you considering exactly?

 

[Dr Lots-o'watts]

Higher peak powers generally allow higher efficiencies. But they can also give rise to other issues.

 

[Thermodave]

In nonlinear optics you develop the response P to a light field of amplitude E in a power series of the field amplitude:
P(t)=\chi^{(1)} E(t) +\chi^{(2)}E^2 (t) +\chi^{(3)}E^3 (t)+...

Second harmonic generation/frequency doubling is a second order process and therefore proportional to the square of the amplitude of the incoming field. Therefore its efficiency increases also quadratically with the amplitude.

The efficiency also depends strongly on what nonlinear crystal you use and which wavelength you are aiming at. There are good crystals cut for 800nm->400nm achieving up to 20% efficiency. I am sure the record values are way higher, but not representative.

As a consequence SHG works best for short pulses. SHG for cw is of course also possible but not very effective, even worse for low input powers. What input power and wavelength are you considering exactly?

Thank you for the explanation. I wasn't thinking a specific experiment though. I've just noticed that there are a lot of diode lasers (e.g. New Focus 25mW cw) models on the market now with narrow linewidths that can be used for absorption experiments but that they are all in the red. If there was an efficient doubling approach than I'm sure we'd be seeing a broader range of wavelengths. Doing absorption spec in the 300s would be convenient, if it was tunable of course.