ERROR: should be no new maxima in edge refinement

Bader charge density analysis

Moderator: moderators

jisoo93
Posts: 3
Joined: Thu Jan 19, 2017 6:45 am

ERROR: should be no new maxima in edge refinement

Post by jisoo93 » Thu Jan 19, 2017 7:00 am

Hi, I'm a beginner at bader charge analysis.
I tried to use this for the first time and when I type './bader CHGCAR -ref CHGCAR_sum'
I got a lot of 'ERROR: should be no new maxima in edge refinement' this error message.
how can I solve it please give me some help...
I also attach the files that I used
Thank you.

[njs11@clustermaster 20d]$ ./bader CHGCAR -ref CHGCAR_sum

GRID BASED BADER ANALYSIS (Version 0.95a 02/26/16)

OPEN ... CHGCAR
VASP5-STYLE INPUT FILE
DENSITY-GRID: 80 x 160 x 324
CLOSE ... CHGCAR
RUN TIME: 1.14 SECONDS

OPEN ... CHGCAR_sum
VASP5-STYLE INPUT FILE
DENSITY-GRID: 80 x 160 x 324
CLOSE ... CHGCAR_sum
RUN TIME: 1.20 SECONDS

CALCULATING BADER CHARGE DISTRIBUTION
0 10 25 50 75 100
PERCENT DONE: **********************

REFINING AUTOMATICALLY
ITERATION: 1
EDGE POINTS: 1827279
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement
ERROR: should be no new maxima in edge refinement

graeme
Site Admin
Posts: 1866
Joined: Tue Apr 26, 2005 4:25 am
Location: University of Texas at Austin
Contact:

Re: ERROR: should be no new maxima in edge refinement

Post by graeme » Thu Jan 19, 2017 7:56 pm

This often indicates a problem with the charge density file. If you attach it, I can try to see what's going wrong.

jisoo93
Posts: 3
Joined: Thu Jan 19, 2017 6:45 am

Re: ERROR: should be no new maxima in edge refinement

Post by jisoo93 » Fri Jan 20, 2017 4:28 am

I thought I attached the file but there wasn't
anyway here's the files
thank you so much
Attachments
bader.zip
(57.89 MiB) Downloaded 485 times

graeme
Site Admin
Posts: 1866
Joined: Tue Apr 26, 2005 4:25 am
Location: University of Texas at Austin
Contact:

Re: ERROR: should be no new maxima in edge refinement

Post by graeme » Fri Jan 20, 2017 5:06 pm

The CHGCAR looks ok but the CHGCAR_sum is not physical -- something is wrong with how it was generated. See the isosurface:
Attachments
chgcar_sum_iso.png
chgcar_sum_iso.png (622.98 KiB) Viewed 13827 times

jisoo93
Posts: 3
Joined: Thu Jan 19, 2017 6:45 am

Re: ERROR: should be no new maxima in edge refinement

Post by jisoo93 » Mon Jan 23, 2017 6:34 am

thank you so much
I will have a look the cHGCAR_sum
so do I have to fix the code that make CHGCAR_sum?

graeme
Site Admin
Posts: 1866
Joined: Tue Apr 26, 2005 4:25 am
Location: University of Texas at Austin
Contact:

Re: ERROR: should be no new maxima in edge refinement

Post by graeme » Mon Jan 23, 2017 1:09 pm

I would take a look at the files generated by your DFT code. They must be reasonable for the analysis to work properly.

jbornl
Posts: 1
Joined: Fri Jan 27, 2017 9:53 pm

Re: ERROR: should be no new maxima in edge refinement

Post by jbornl » Fri Jan 27, 2017 10:06 pm

I have seen this topic pop up a few times now.
(Since I have been looking at this site as I have the same error.)
Indeed, I can see very large neg values in my AECCAR0 file.
Still, I used a standard VASP INCAR performing a SCF w LAECHG =.TRUE. (and doubling my initial
values of NGX, NGY, NGZ, from 126-->252)
I am doing spin polarized calcs, but that shouldn't have any effect, with a 2 dim material with a large vacuum.
Perhaps rapid oscillations in the core wave functions for my PAW_PBE psps is the cause of the problem?
(Mo_sv being recommended for Mo).

I can't see anyway to overcome this.
And I think my calcs are good as I generate elec and phys structure very similar to that seen in the literature.

Any ideas?
tnx,
jb

graeme
Site Admin
Posts: 1866
Joined: Tue Apr 26, 2005 4:25 am
Location: University of Texas at Austin
Contact:

Re: ERROR: should be no new maxima in edge refinement

Post by graeme » Sat Jan 28, 2017 2:28 am

this might be an appropriate question for the vasp guys.

juganta
Posts: 9
Joined: Wed Nov 30, 2016 8:02 pm

Re: ERROR: should be no new maxima in edge refinement

Post by juganta » Wed Feb 15, 2017 6:32 am

I am also getting the same problem. But when I tried with "-vac off" flag, it gone well and I am getting vacuum charge zero (0). Is there any effect of vacuum charge on the final charge transfer quantity, in case of adsorbed system?
Thanks.

graeme
Site Admin
Posts: 1866
Joined: Tue Apr 26, 2005 4:25 am
Location: University of Texas at Austin
Contact:

Re: ERROR: should be no new maxima in edge refinement

Post by graeme » Wed Feb 15, 2017 1:08 pm

Any changes due to having a vacuum region should be minor. You can easily verify this.

juganta
Posts: 9
Joined: Wed Nov 30, 2016 8:02 pm

Re: ERROR: should be no new maxima in edge refinement

Post by juganta » Wed Feb 15, 2017 10:50 pm

Thanks a lot Professor. What I have mentioned regarding "-vac off", is it a standard procedure? Because you have mentioned that something wrong with CHGCAR_sum file, if this error pop up "ERROR: should be no new maxima in edge refinement" .
For my case I got:
VACUUM CHARGE: 0.5448
VACUUM VOLUME: 684.9021
NUMBER OF ELECTRONS: 1408.0000
Is it too much vacuum charge?
Would you comment on this situation? Thanks a lot!

juganta
Posts: 9
Joined: Wed Nov 30, 2016 8:02 pm

Re: ERROR: should be no new maxima in edge refinement

Post by juganta » Wed Feb 15, 2017 10:57 pm

Here is the ACF.dat.
Attachments
acf.tar.gz
(7.15 KiB) Downloaded 441 times

graeme
Site Admin
Posts: 1866
Joined: Tue Apr 26, 2005 4:25 am
Location: University of Texas at Austin
Contact:

Re: ERROR: should be no new maxima in edge refinement

Post by graeme » Thu Feb 16, 2017 12:38 am

I would need to see the charge density files to understand the error. But first, take a look at them in a visualizing program -- many of the problems that people report are related to unphysical charge density files.

juganta
Posts: 9
Joined: Wed Nov 30, 2016 8:02 pm

Re: ERROR: should be no new maxima in edge refinement

Post by juganta » Thu Feb 16, 2017 6:29 pm

I have tried to upload files but it failed. Would you please tell me what is the max size of the uploaded files? Thanks

graeme
Site Admin
Posts: 1866
Joined: Tue Apr 26, 2005 4:25 am
Location: University of Texas at Austin
Contact:

Re: ERROR: should be no new maxima in edge refinement

Post by graeme » Thu Feb 16, 2017 8:11 pm

Make sure they are compressed, and attach them separately. I'll check the limit ... Ok, I think it should be 2GB/file now.

You can also use a dropbox link, or any cloud based storage.

Post Reply