Page 1 of 1

Nan value of polaristaion in ral time BSE

Posted: Thu Jan 21, 2021 11:58 am
by Bramhachari Khamari
Dear Developers,
I tried to run a real time simulation using yambo_rt. I applied a field kind delta to get the spectrum. I would like to study the real time simulation by applying a different field kind (GAUSS or QSSIN) and set Field1_Freq to a value as peak of absorption. Following parameter related to field is used. However, after 46 fs I got NAN value of polarisation. Could you explain the different parameter such as

i) PhLifeTime. do we need to use for field other than delta.

ii) Field1_Width, is it the standard deviation of the pulse?

ii) is there any smart way of choosing RTstep, NEtime to avoid NAN error


PhLifeTime=100.000000 fs # [RT] Constant Dephasing Time
RTstep= 10.00000 as # [RT] Real Time step length
NETime=100.000000 fs # [RT] Simulation Time
% IOtime
2.000000 | 1.000000 | 0.500000 | fs # [RT] Time between two consecutive I/O (CARRIERs - GF - OUTPUT)
%
% Field1_Freq
1.900000 | 1.900000 | eV # [RT Field1] Frequency
%
Field1_Int=10000000.000000 kWLm2 # [RT Field1] Intensity
Field1_Width=100.000000 fs # [RT Field1] Width
Field1_kind= "GAUSS" # [RT Field1] Kind(SIN|RES|ANTIRES|GAUSS|DELTA|QSSIN)
Field1_pol= "linear" # [RT Field1] Pol(linear|circular)
% Field1_Dir
0.000000 | 1.000000 | 0.000000 | # [RT Field1] Versor
%
% Field1_Dir_circ
0.000000 | 1.000000 | 0.000000 | # [RT Field1] Versor_circ
%
Field1_Tstart= 0.0100000 fs # [RT Field1] Initial Time

Regards,

Re: Nan value of polaristaion in ral time BSE

Posted: Thu Jan 21, 2021 12:43 pm
by Davide Sangalli

Code: Select all

PhLifeTime
do we need to use for field other than delta.
Dephasing time for the off-diagonal components of the density matrix
It is independent from the pulse.

Code: Select all

Field1_Width
, is it the standard deviation of the pulse?
It is the duration for the pulse in time (only works for some kind of pulses.
You can also use, as an alternative

Code: Select all

Field1_FWHM
The first controls the duration of the field profile, the second of the intensity profile
is there any smart way of choosing RTstep, NEtime to avoid NAN error
Not really. It is related to the maximum frequency in the simulation. Try to lower it if you get NaN.
Changing the integrator may also help.

Best,
D.

Re: Nan value of polaristaion in ral time BSE

Posted: Thu Jan 21, 2021 1:04 pm
by Bramhachari Khamari
Dear Sir,
Thanks a lot for the reply. One more thing I would like to know related to output of the real time. I get a output file which has the information about the carriers.Its first line says
Time[fs] d(N_hole-N_elec) dN_hole dN_elec

Could you tell me what is dN_elec here.


Regards,

Re: Nan value of polaristaion in ral time BSE

Posted: Sat Jan 23, 2021 7:20 am
by Bramhachari Khamari
Dear Sir,
There are many possibility of choosing integrator given by the input. Which combination is better to avoid NaN error.



Integrator= " " # [RT] Integrator. Use keywords space separated ( "EULER/EXPn/INV" "SIMPLE/RK2/RK4/HEUN" "RWA")

Regards,