RIM & BSE

[WEI Wei]

Dear all,

For a reduced dimensionality, e.g. sheet in 2D or nanowire in 1D, box-shaped Coulomb interaction is suggested to included. So,
1) it can be introduced in GW and BSE calculation by variables as follows:
-----------------------------------------------------------------------------------------------------------
RandQpts= 3000000 # [RIM] Number of random q-points in the BZ
RandGvec= 1 RL # [RIM] Coulomb interaction RS components
CUTGeo= "box z" # [CUT] Coulomb Cutoff geometry: box/cylinder/sphere
% CUTBox
0.00000 | 0.00000 | 20.00000 | # [CUT] [au] Box sides
%
CUTRadius= 0.000000 # [CUT] [au] Sphere/Cylinder radius
CUTCylLen= 0.000000 # [CUT] [au] Cylinder length
-----------------------------------------------------------------------------------------------------------
My question is that should I omit the varibales of "CUTRadius" and "CUTCylLen" or just set them as zero?

2) Since RIM is an approximation about self-energy (am I right?), should I introduce a box-shaped Coulomb interaction in a RPA calculation (yambo -o c)?

3) How can I check and change the variables about the exchange and correlation parts for a G0W0 calculation?

Many thanks in advance!

[Daniele Varsano]

Dear Wei Wei,

My question is that should I omit the varibales of "CUTRadius" and "CUTCylLen" or just set them as zero

just leave them with zero value. Anyway in the report you find information if the information you gave in input are correct.

) Since RIM is an approximation about self-energy (am I right?), should I introduce a box-shaped Coulomb interaction in a RPA calculation (yambo -o c)?

No, it is not an apporzimation about self-energy, but a just a way to perform integrations of the coulomb potential in the Brillouin zone in the
correct way. Anyway, In RPA calculation you do not need to integrate the Coulomb potential.

How can I check and change the variables about the exchange and correlation parts for a G0W0 calculation?

I'm not sure I understood what you mean. In the report you have the matrix elements of the exchange and correlation part
separately, so you can check them and look if they are at convergences.

Cheers,

Daniele

[WEI Wei]

I'm not sure I understood what you mean. In the report you have the matrix elements of the exchange and correlation part
separately, so you can check them and look if they are at convergences.

I mean I am not sure what are the variables determining the exchange and correlation parts in input file.
It reports in "r_HF_and_locXC_em1d_ppa_gw0" that
-----------------------------------------------------------------------------
EXXRLvcs= 11733 RL # [XX] Exchange RL components
-----------------------------------------------------------------------------
Is this "EXXRLvcs" using to control the exchange part? What is for correlation part?

Thank you very much!

[Daniele Varsano]

Dear Wei Wei,
if you go through the documentation: here you can find the variable controlling each runlevels. As you can see there, EXXRLvcs controls the G-vectors entering in the exchange (Hartree-Fock) part. For the correlation you will have to
converge the Gvectors "NGsBlkXp" size in the response for the plasmon pole approximation as well as the number of bands: BndsRnXp (if you are using the PP approximation).
Next you will need to converge the bands GbndRnge in the GW summation. In the link I sent you a brief description of all the variables
Cheers,

Daniele

[WEI Wei]

Dear Daniele,

Thank you for your reply!