Questions about how to set NLDamping in SHG calculations

[Dean]

Dear all,
I want to calculate 2D materials using Yambo. But, I find that the SHG coefficients are greatly affected by the parameter "NLDamping", as shown in the figure.
So, what's the principle of setting "NLDamping"? Why "NLDamping" is set as 0.2 eV in many Refs?
Thanks in advance!
Best,

[myrta gruning]

Hello,

the NLDamping parameter corresponds to the spectral broadening which is observed in experiment and accounts empirically for the interaction with other degrees of freedom (neglected from the ab-initio dynamics). The choice of 0.1-0.2 eV is a reasonable estimate of broadening that is usually observed, but as I said this is an empirical parameter, so it should be evaluated per case. As you correctly observed, by broadening the spectrum one reduces the intensities of the peak, so one should be aware of this and report the spectral broadening.

On the real-time dynamics perspective, the NLDamping translates into a decay time of the oscillation modes of the system. The larger the damping (and broadening), the shorter the decay time and vice versa. So, the choice of the NLDamping also influences the length of the simulation. The smaller the NLDamping, the longer the simulation.

[Dean]

Dr Myrta Grüning，
Thanks for your reply.