Issues with high energy pulses in YAG @ 3100nm

[mill0851]

Currently I am trying to simulate pulse compression @ 3100 nm in YAG, but I am having the issue that when my pulse energy is set above about 2 uJ the simulation results blow up. Instead of seeing the usual artifacts of nonlinear interaction like SHG and compression. The E field data becomes extremely noisy, and K-space is unreadable. Maybe I am missing something, but I am not sure why this is occurring, any help is much appreciated. For now I have simply scaled back my pulse energies.

[mill0851]

[mill0851]

Attempting to recreate this experiment: DOI:10.1364/OE.21.028095

[NickKarpowicz]

Hi there,
what I think is happening is that since the nonlinear absorption is zero, the Kerr effect is causing the beam to collapse with no plasma generation to stop it. It's self-focusing enough that it's over-filling the k-space of the grid (upper-right plot) - so it would need a smaller spatial step size.

It's a bit tricky to get a convergent simulation with plasma generation - if you post the YAG crystal definition you're using I can try to get something close...

Best wishes,
Nick

[mill0851]

Hello Nick,

Sorry for the delayed response, here is my crystal definition for YAG:

Name:
YAG
Type:
0
Sellmeier equation:
0
1st axis coefficients:
1 0 2.282 -0.01185 0 3.27644 -282.734 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
2nd axis coefficients:
1 0 2.282 -0.01185 0 3.27644 -282.734 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
3rd axis coefficients:
1 0 2.282 -0.01185 0 3.27644 -282.734 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Sellmeier reference:
D. E. Zelmon, D. L. Small and R. Page., Refractive-index measurements of undoped yttrium aluminum garnet from 0.4 to 5.0 um
chi2 type:
0
d:
0.0 0.0 0.0 0.0 -2.405 0.0
0.0 0.0 0.0 2.605 0.0 0.0
-2.405 2.605 0.15 0.0 0.0 0.0
d reference:
...
chi3 type:
2
chi3:
6.719720332155478e-22
0
0
chi3 reference:
M. Hemmer, Self-compression to sub-3-cycle duration of MIR optical pulses in dielectrics
Spectral file:
None
Nonlinear reference frequencies:
9.67e+13 9.67e+13 1.934e+14 9.67e+13 9.67e+13 9.67e+13 9.67e+13

Thank you for helping me!

[NickKarpowicz]

Hi there,
sorry for the long delay, there were a few things that I wanted to check, and I'm working on a new release of the code, which will have a different YAG crystal defined inside to make it a bit easier to work with.

Here's a simulation I did that qualitatively seems to agree with what they show in the paper: https://datashare.mpcdf.mpg.de/s/E8A8XiWuSSTWSha

I modified a few things to make it more stable:

• The photoionization/nonlinear absorption model I use is based on fairly simple multiphoton absorption. Since YAG has a pretty large bandgap and the photon energy is pretty small (0.4 eV), this means that it needs 17 (!!) photons. The resulting nonlinear polarization is extremely nonlinear, and hard to work with numerically. In this regime, one would be better off with a different photoionization model. I have a colleague who is working on integrating the semiconductor Bloch equations into the propagation, which would do a better job, but that's a while in the future. To have something to work with, I lowered the bandgap, with the justification that these fields are in the regime where a more complete Keldysh-type response would be dropping in slope.
• I add a small perturbation to the beam shape. Since they mention that their M^2 is 1.8, the perfect TEM00 that one gets in the simulation likely self-focuses too efficiently. I add a smaller, out-of-phase duplicate of the beam; this has the effect of giving a more flat-top focus.
• I broaden the bandwidth and add a little third-order dispersion since their pulse is asymmetric.
• I simulate a z-scan by first instantiating the beam in air (first line of the sequence). This allows it to refract at the front surface of the crystal.

I'd say this simulation isn't perfect, since physically it's at the edge of the photoionization model in the code, but maybe it can give a qualitative sense - it does replicate the space-time structure they show in the paper, as well as the behavior of the spectrum over the course of the z-scan.

I'll upload the next release soon, which will have the YAG definition I've been using (It's in the database in my 'dev' branch on the repo here).

Cheers,
Nick

[mill0851]

Hello Nick,

Thank you for all of your help. It is greatly appreciated! I will let you know if some good experiments come from our simulations.

Best,
Robert Miller

[mill0851]

Hello Nick,

May I ask why you decreased the value of the kerr coefficient in your YAG crystal definition?

Best,
Robert Miller

[NickKarpowicz]

Which value are you referring to? In the current CrystalDatabase.txt it's 6.719720332155478e-22, which is the same as what you posted