Pump and probe simulation

[sdwang]

Dear developers,
I have some questions about the yambo_nl for the P&P simulations:
(1). The wiki at https://www.yambo-code.eu/wiki/index.p ... _and_Probe give an example of P&P, where is:
'In our example we choose direction [0,1,0] for the pump and [1,0,0] for the probe' and for the initial step,
fixsyms # [R] Remove symmetries not consistent with an external perturbation
% Efield1
0.000000 | 1.000000 | 0.000000 | # First external Electric Field
%
% Efield2
1.000000 | 0.000000 | 0.000000 | # Additional external Electric Field

However, in the following in pump section, the field2 is:
% Field2_Dir
0.000000 | 1.000000 | 0.000000 | # [RT Field2]

And, in pump_and_probe section, the fields are:
% Field1_Dir
1.000000 | 0.000000 | 0.000000 | # [RT Field1]
% Field2_Dir
0.000000 | 1.000000 | 0.000000 | # [RT Field2]
which are both not consistent to the initial and pump step.
My question is how to define the field direction in pump, probe and pump&probe, considering the initial step?

(2) I have converged my GW calculations with the k-points, say about 102 points in the irreducible BZ, and when I perform the collision calculations, I have to remove the symmetry using ypp, and the new SAVE file contains more than 1000 points corresponding the converged 102k, since there is absent of the symmetry in the FixSymm directory, which is very huge for the collision calculations even for the single precision yambo_rt.
My question is: is there an efficient way to deal with the huge k-points for yambo_nl?

Thanks!

Best Regards,

Shudong

[claudio]

Dear Shudong

there were some errors in the wiki that now I fixed.

In the wiki example we choose the probe in x (1,0,0) and the pump in y (0,1,0).
This choice in general depends from the experiments you want to reproduce.


>(2) I have converged my GW calculations with the k-points, say about 102 points in the irreducible BZ, and when I perform the collision calculations, I have to remove the symmetry using ypp, and the new SAVE file contains
> more than 1000 points corresponding the converged 102k, since there is absent of the symmetry in the FixSymm directory, which is very huge for the collision calculations even for the single precision yambo_rt.
>My question is: is there an efficient way to deal with the huge k-points for yambo_nl?

this is really a problem, collisions are very expensive. You can try to parallelize collision calculations as much as possible, and to reduce
the number of G-vectors in collisions to few. Or you can try to use simpler model as for example

https://www.yambo-code.eu/wiki/index.ph ... _(TD-DPFT)

best
Claudio

[sdwang]

Dear Claudio,
Thank you for your reply. I have another question about the unit of filed like
Field1_Int= 1.00 kWLm2 # [RT Field1] Intensity

What does the unit of kWLM2 mean? Is it the pulse intensity of P/pi*omega^2 or the peak laser power as Q/T (Q is pulse energy)？

Thanks!

Shudong

[claudio]

Dear Shudong

kWLM2 is the field intensity and stays for kW/cm^2

best
Claudio

[sdwang]

Dear all,
I performed p&p calculations with collisions, and it is OK for hBN. However, when I move to a system with spin-orbit coupling, it stops after collision calculations (for probe process) with yambo_nl executable without any error in the log files. Attached are the input, report and log files.

Thanks in advance!

SD

[claudio]

Dear Shudong

thank you for reporting this, I will check make you know

best
Claudio