This thread is locked.Only browsing is available.
Top Page > Browsing
Pseudo-potential for fictitious atoms
Date: 2015/02/12 21:29
Name: PR

Dear Prof. Ozaki,

I have used Adpack 2.2 to generate the pseudo-potential for fictitious atoms like pseudo-hydrogen with a fractional charge of 0.75 and 1.25 in order to saturate the dangling bonds on the surface of III-V materials. For example, I have used the following setting for the atoms definition in Adpack for pseudo-hydrogen with a fractional charge of 0.75:

#
# Atom
#

AtomSpecies 0.75
max.occupied.N 1
total.electron 0.75
valence.electron 0.75
<occupied.electrons
1 0.75
occupied.electrons>


Although it seems to work for band calculations, there is a problem when I use the generated pseudo-hydrogens in geometry optimization. It seems that OpenMX can not handle this pseudo-hydrogen correctly and in first geometry optimization step I get the following output:

*******************************************************
MD or geometry opt. at MD = 1
*******************************************************

<Steepest_Descent> SD_scaling= 0.150011287118
<Steepest_Descent> |Maximum force| (Hartree/Bohr) = 0.000000000000
<Steepest_Descent> Criterion (Hartree/Bohr) = 0.000700000000

atom= 1, XYZ(ang) Fxyz(a.u.)= 11.2217 11.2036 5.5408 nan nan nan
atom= 2, XYZ(ang) Fxyz(a.u.)= 13.9748 13.9439 5.5337 nan nan nan
atom= 3, XYZ(ang) Fxyz(a.u.)= 9.8615 9.8529 1.3193 nan nan nan
atom= 4, XYZ(ang) Fxyz(a.u.)= 15.3161 9.8387 1.3330 nan nan nan
atom= 5, XYZ(ang) Fxyz(a.u.)= 9.8695 15.2878 1.3368 nan nan nan
atom= 6, XYZ(ang) Fxyz(a.u.)= 12.5994 12.5743 1.3359 nan nan nan
atom= 7, XYZ(ang) Fxyz(a.u.)= 15.3304 15.2948 1.3265 nan nan nan
atom= 8, XYZ(ang) Fxyz(a.u.)= 13.9653 11.1875 2.7844 nan nan nan
......

and program stops soon after.

Would you kindly help me regarding this, and let me know what the problem is?

Many thanks,

Best regards,

PR
メンテ
Page: [1]

Re: Pseudo-potential for fictitious atoms ( No.1 )
Date: 2015/02/20 17:32
Name: T. Ozaki

Hi,

Thank you very much for your report.
This is a bug and happens if the fractional nuclear charge is less than 1.
A routine 'SetPara_DFT.c' was modified so that the case can be treated.
Please apply the patch Ver. 3.7.9 including the modification, downloadable from
http://www.openmx-square.org/download.html

Regards,

TO
メンテ
Re: Pseudo-potential for fictitious atoms ( No.2 )
Date: 2015/02/21 00:52
Name: PR

Dear Prof. Ozaki,

Thanks a lot for your fast reply and providing the patch,

There is small issue, I found that there is a problem with the patch released, file "Band_DFT_Col.c" is corrupted and it seems that different files have been copied in to this file mistakenly so it generates error while compiling, I replaced the file with the one in patch 3.7.8 while using the rest in patch 3.7.9 and then it worked,

Just wanted to report this that you can fix this issue,

If the file Band_DFT_Col.c should be different from the one in version 3.7.8 then I suppose just replacing it with the one in patch 3.7.8 would not suffice,


Thanks for your support,

P.R.
メンテ
Re: Pseudo-potential for fictitious atoms ( No.3 )
Date: 2015/02/23 03:21
Name: PR

Many thanks for the new patch.

It has solved the issue,

Regards,

P.R.
メンテ

Page: [1]