GW+BSE

[lamia]

After seeing the hbn tutorial. I used the following command line for bse input :

../yambo -o b -b -k sex -y d -V qp and I got the input posted before. I don't know why it was full of inconsistencies!

How can I correct this input?

my best regards,
Lamia

[Daniele Varsano]

Dear Lamia,
if you already calculated the screening (plasmon pole for the GW) you do not need to recalculate so "-b" is not needed.
Next the problem is in the parameter you set:
1) You included also deep states in the BSE which are not needed and make your matrix very large
2) You included QP energies both as database and scissor, they are exclusive
3) BSENGBlk cannot be larger of the g vectors you considered for the screening.

Finally, please inspect your report to see what went wrong, if you cannot spot it please post the report/input files.
Best,
Daniele

[lamia]

Dear Daniele,

I tried to correct the input file following your suggestions .However, the job is done very quickly with a very small values of dielectric function .

Best,
Lamia

[Daniele Varsano]

Dear Lamia,
I suggest you to do a clean calculation, ie removing previous BS databases, if you inspect your report file you can see full of *ERR* due to previous databases inconsistent with your input file.
Anyway the BSE has been calculated.

Best,
Daniele

[lamia]

Dear Daniele,

Thank you for your suggestion. Actually, I get those files o.eps_q1_diago_bse , o.alpha_q1_diago_bse and o.jdos_q1_diago_bse. Which one should I plot to calculate dielectric function and the absorption ? what information could I get from those files ? and What is the difference between them?

Best,
Lamia

[Daniele Varsano]

Dear Lamia,
in order to look at the absorption you should look at imaginary part the polarisability of the system (as it is a 2D system).
The file is organized as written in the legend:

# E/ev[1] ALPHA-Im[2] ALPHA-Re[3] ALPHAo-Im[4] ALPHAo-Re[5]

Column 2 and 3 are the imaginary and real part of alpha. The column 3 and 4 are the imaginary and real part of the independent particle polarisability (no excitonic effects).

For the interpretation of your results I suggest you to look carefully at the tutorials.
Best,
Daniele

[lamia]

Dear Daniele,

You helped me a lot thank you very much for all.


Best,

Lamia

[lamia]

One more question please
what about dielectric function ? How could I extract it? And what is o.eps and o.jdos ?

Best,

Lamia

[Daniele Varsano]

The dielectric function itself it is report on o.eps, but it is not well defined in a D system, and you will find very small values that actually goes to Re(eps)=1 Im(eps)=0, i.e. the vacuum value. Spectroscopical quantities (absorption) needs to be extracted from the polarisability.

o.jdos is the joint density of states, you can find its definition in textbooks and papers, e.g. also here you can look at the definition:
https://www.tddft.org/bmg/files/seminarios/24830.pdf


Best,
Daniele

[lamia]

Dear YAMBO developpers,

I ran BSE . However, I get the ERROR below:


''
<01h-44m-49s> Xo@q[29] |########################################| [100%] 02m-30s(E) 02m-30s(X)
<01h-44m-49s> X@q[29] |########################################| [100%] --(E) --(X)
<01h-44m-49s> [X-CG] R(p) Tot o/o(of R) : 48206 167156 100
<01h-47m-19s> Xo@q[30] |########################################| [100%] 02m-29s(E) 02m-29s(X)
<01h-47m-19s> X@q[30] |########################################| [100%] --(E) --(X)
<01h-47m-19s> [08] Bethe Salpeter Equation @q1
<01h-47m-19s> [WARNING] Null BSE kernel block dimension(s) found.

[ERROR] STOP signal received while in :[08] Bethe Salpeter Equation @q1
[ERROR] The whole BSE has zero dimension !
''


Best,
Lamia