This thread is locked.Only browsing is available.
Top Page > Browsing
Postprocessing - "Analysis of ..."
Date: 2011/10/13 22:10
Name: N.Kolchenko   <nkolchenko@mail.ru>

Dear prof. T.Ozaki,

It seems that simple (any dimer or even H2 molecule) tests of "Analysis of difference charge density induced by the interaction" force to change couple of lines in file add_gcube.c.
(All Manual's recommendations remain the same).

In the part:
/*******************************************
make a file
*******************************************/

1. To take into account full number of atoms:

fprintf(fp1,"%5d %12.6f %12.6f %12.6f\n",atomnum1+atomnum2,Grid_Origin1[1],Grid_Origin1[2],Grid_Origin1[3]);

instead of

fprintf(fp1,"%5d %12.6f %12.6f %12.6f\n",atomnum1,Grid_Origin1[1],Grid_Origin1[2],Grid_Origin1[3]);
/************************


2. After cycle
/***********************************
for (i=1; i<=atomnum1; i++){
fprintf(fp1,"%5.0f %12.6f %12.6f %12.6f %12.6f\n",
Gxyz1[i][0],0.0,Gxyz1[i][1],Gxyz1[i][2],Gxyz1[i][3]);
}
*************************************/
insert the same for second subsystem :
/***************************************
for (i=1; i<=atomnum2; i++){
fprintf(fp1,"%5.0f %12.6f %12.6f %12.6f %12.6f\n",
Gxyz2[i][0],0.0,Gxyz2[i][1],Gxyz2[i][2],Gxyz2[i][3]);
}
****************************************/

Possibly, I miss anything, although for series of test these corrections are sufficient.

On the other hand, there is Fig.5 in [Phys.Rev. B83, 104413 (2011)], but I can't obtain similar result by initial add_gcube.c (+diff_gcube, of course) version.

Regards,

NK
メンテ
Page: [1]

Re: Postprocessing - "Analysis of ..." ( No.1 )
Date: 2011/11/11 22:37
Name: T.Ozaki

Hi,

If you follow the way shown in the manual, that is,
./diff_gcube AB.cube A_B.cube dAB.cube
No problem happens.

On the other hand, if you do
./diff_gcube A_B.cube AB.cube dAB.cube
Then, the geometrical structure in dAB.cube contains only that of
either A or B. I guess that this is your case.

Regards,

TO
メンテ
Re: Postprocessing - "Analysis of ..." ( No.2 )
Date: 2011/11/13 02:26
Name: N.Kolchenko  <nkolchenko@mail.ru>

Dear prof. T.Ozaki,

Thank you for reply. The situation is clear and question is closed.

I lost the way when I've received message from diff_gcube - "Found a difference in the number of atoms", and "error" was confirmed by visualization of add_gcube result (A_B.cube). Unfortunetely I didn't make the visualization of dAB.cube immediately and began to search in add_gcube...

So,

1. I have the version of add_gcube which permit the proper visualization of intermediate A_B.cube and optional <input args> ordering for diff_gcube;

2. now one can read recommendation - don't worry about WARNING message "Found a difference in the number of atoms" in "Analysis of difference charge density induced by the interaction".

Regards,

NK
メンテ

Page: [1]