Box (Z) truncation question

Concerns issues with computing quasiparticle corrections to the DFT eigenvalues - i.e., the self-energy within the GW approximation (-g n), or considering the Hartree-Fock exchange only (-x)

Moderators: Daniele Varsano, andrea.ferretti, andrea marini, myrta gruning

Box (Z) truncation question

Postby Fabiof » Wed May 23, 2018 4:35 pm

Dear all,

I have a 2D system (InSe) with a vacuum distance of 55 Bohr and I choose the CUTBox to be 52.

Code: Select all
% CUTBox
 0.00     | 0.00     | 52.00     |        # [CUT] [au] Box sides
%


Then I choose it to be

Code: Select all
% CUTBox
 0.00     | 0.00     | 54.00     |        # [CUT] [au] Box sides
%



and there were some big differences at some k-points. After doing the interpolation, the band with CUTBox of 54 is the one that seems correct. What is the criterium to choose the CUTBox sizes? I thought it was just a little less than the vacuum size.

Thank you,

Fabio
Fábio Ferreira, Graduate Student
University of Minho, Portugal
Fabiof
 
Posts: 68
Joined: Wed Feb 11, 2015 10:52 am

Re: Box (Z) truncation question

Postby Daniele Varsano » Wed May 23, 2018 5:16 pm

Dear Fabio,
It depends on the vacuum and also on the system size. If you have enough vacuum, the criterium is to set it slightly less than the box size and not the vacuum size. The more vacuum you have with respect your system size the more you have the freedom to reduce the z size. In your case, it seems that 52 is not enough and I would say that 54 is safer. In the next release of Yambo there will be a new implementation of the truncated coulomb potential that will not require geometrical setting in the input, so much easier to use.

Best,

Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/
User avatar
Daniele Varsano
 
Posts: 1950
Joined: Tue Mar 17, 2009 2:23 pm

Re: Box (Z) truncation question

Postby Fabiof » Fri May 25, 2018 4:00 pm

Dear Daniele,
If you have enough vacuum, the criterium is to set it slightly less than the box size and not the vacuum size.


So the box size (z) has to be larger than the vacuum distance? A distance z of 56 would be safer?

Fabio
Fábio Ferreira, Graduate Student
University of Minho, Portugal
Fabiof
 
Posts: 68
Joined: Wed Feb 11, 2015 10:52 am

Re: Box (Z) truncation question

Postby Daniele Varsano » Fri May 25, 2018 4:08 pm

Dear Fabio,
I do not know how large is your box.
The criteria is to set the size(z) just slightly smaller than your box size. This means that the interaction is cut above |r-r'| < z/2 (there is a factor 2 inside the code) and this ensures that you have interaction among all the part of your system and two replica do not interact if your vacuum size is larger than your system size (usually it is recommended to consider a vacuum size larger than the system size due to the spill out of the charge).

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/
User avatar
Daniele Varsano
 
Posts: 1950
Joined: Tue Mar 17, 2009 2:23 pm

Re: Box (Z) truncation question

Postby Fabiof » Fri May 25, 2018 7:28 pm

Dear Daniele,

I am not understanding the box size concept.

I have a 2D system which has a cell size (z) of 55 bohr. I call it vacuum distance because it is the interlayer distance between the replica.
I thought the box size is the size of the box that we choose to set when doing the yambo calculations.

In the 2D hBN tutorial they set the box dimensions to be (0,0,32) bohr. The cell size in the z direction is 33 bohr. I did the same to my system.

Thank you,

Fabio
Fábio Ferreira, Graduate Student
University of Minho, Portugal
Fabiof
 
Posts: 68
Joined: Wed Feb 11, 2015 10:52 am

Re: Box (Z) truncation question

Postby Daniele Varsano » Sat May 26, 2018 8:18 am

Dear Fabio,
ok there is a misunderstanding. Vacuum it is meant as the size of "vacuum" i.e. size of the supercell minus the size of the system. Coulomb cutoff technique you can apply for 0D, 1D or 2D systems. Think about a molecule or a nanotube, so it is important to evaluate the size of the system and the corresponding vacuum to add. Exact and rule of thumb conditions between supercell size, range of the cut can be found here:
Exact Coulomb cutoff technique for supercell calculations CA Rozzi, D Varsano, A Marini, EKU Gross, A Rubio Physical Review B 73 (20), 205119 (2006)
M. R. Jarvis, I. D. White, R. W. Godby and M. C. Payne. Phys. Rev. B , 56, 14072, 1997.


Briefly, just set your zcut slightly smaller than your cell as in the tutorial and as you already did.
I will definitely remove the choice of zcut from the input setting it automatically.

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/
User avatar
Daniele Varsano
 
Posts: 1950
Joined: Tue Mar 17, 2009 2:23 pm

Re: Box (Z) truncation question

Postby Fabiof » Sat May 26, 2018 5:14 pm

Dear Daniele,

Ok, it is clear now.

Thank you,

Fabio
Fábio Ferreira, Graduate Student
University of Minho, Portugal
Fabiof
 
Posts: 68
Joined: Wed Feb 11, 2015 10:52 am


Return to Self-Energy

Who is online

Users browsing this forum: No registered users and 2 guests

cron