String Calculation issue with pawovlp  [SOLVED]

Total energy, geometry optimization, DFT+U, spin....

Moderator: bguster

Locked
ketong
Posts: 11
Joined: Thu Apr 25, 2019 10:44 am

String Calculation issue with pawovlp

Post by ketong » Wed May 08, 2019 4:16 am

Dear ABINIT friends,

I want to calculate the energy barrier of a hopping oxygen atom on a ZrO2(111) surface. The initial and final configurations have been optimized. When I do the string calculation, a strange error occurs.

Code: Select all

--- !ERROR
src_file: m_paw_tools.F90
src_line: 211
mpi_rank: 0
message: |
    PAW COMPENSATION DENSITIES ARE OVERLAPPING !!!!
       There are     4 pairs of overlapping atoms.
       The maximum overlap percentage is obtained for the atoms  55 and  58.
        | Distance between atoms  55 and  58 is  :   3.29921
        | PAW radius of the sphere around atom  55 is:   2.50000
        | PAW radius of the sphere around atom  58 is:   1.41465
        | This leads to a (voluminal) overlap ratio of  0.93 %
    THIS IS DANGEROUS !, as PAW formalism assumes non-overlapping compensation densities.
...


--- !ERROR
src_file: m_paw_tools.F90
src_line: 223
mpi_rank: 0
message: |
    Action: 1- decrease cutoff radius of PAW dataset
        OR  2- ajust "pawovlp" input variable to allow overlap (risky)
...


 abi_abort: decision taken to exit ...
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
with errorcode 14.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------

The version of ABINIT I used is 8.10.1. The PAW potential is JTH-v1.1 PBE table from https://www.abinit.org/psp-tables. In addition, the input file ZrO2surf.in, log file log (renamed log.out for uploading), out file ZrO2surf.out and file file (renamed cZrO2_s.in for uploading) have been attached in order to reproduce this problem.

It is noted that I find some related topic:
1. viewtopic.php?f=8&t=3974&p=12054&hilit=string+calculation#p12054
I regenerate the PAW file. However, this problem isn't solved by this method.
2. viewtopic.php?f=8&t=3986&p=12328#p12328
The pawovlp value has been increased to 25 as suggested Eric. The calculation is running.

Has anyone else ever encountered such an issue? Any comments would be most appreciated.
Attachments
ZrO2surf.in
input file
(9.89 KiB) Downloaded 222 times
log.out
log file
(819.95 KiB) Downloaded 236 times
ZrO2surf.out
out file
(131.19 KiB) Downloaded 195 times
cZrO2_s.in
file file
(284 Bytes) Downloaded 199 times

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: String Calculation issue with pawovlp

Post by ebousquet » Fri May 10, 2019 8:09 am

Dear Ke Tong,
Quick response: there is also the option pawovlp=-1, which I've forgot to mention in the last posts.
Best wishes,
Eric

ketong
Posts: 11
Joined: Thu Apr 25, 2019 10:44 am

Re: String Calculation issue with pawovlp  [SOLVED]

Post by ketong » Sun May 12, 2019 10:13 am

ebousquet wrote:Dear Ke Tong,
Quick response: there is also the option pawovlp=-1, which I've forgot to mention in the last posts.
Best wishes,
Eric

Dear Eric

Thank you very much for your suggestion.

Here is a piece of good news. The problem in this topic has been solved when I set the "pawovlp" to 25. I also believe that the problem will be solved if we use the option pawovlp=-1.

In addition, I don't know where can we find the pawovlp value of the last step? log file? or out file? In other words, if there is no warning about pawovlp, can we think no overlap among the calculated atoms? Your reply is appreciated.

Best wishes,
Ke Tong

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: String Calculation issue with pawovlp

Post by ebousquet » Mon May 13, 2019 3:22 pm

Dear Ke Tong,
I don't see an option to print the overlap of the spheres when they don't go beyond the max...
You could add some writing lines in the source file "src/65_paw/m_paw_tools.F90" where you have the chkpawovlp module, which makes the test of overlaping and print the error message. This is something we could add as a printing option in future version.
Best wishes,
Eric

ketong
Posts: 11
Joined: Thu Apr 25, 2019 10:44 am

Re: String Calculation issue with pawovlp

Post by ketong » Tue May 14, 2019 8:33 am

ebousquet wrote:Dear Ke Tong,
I don't see an option to print the overlap of the spheres when they don't go beyond the max...
You could add some writing lines in the source file "src/65_paw/m_paw_tools.F90" where you have the chkpawovlp module, which makes the test of overlaping and print the error message. This is something we could add as a printing option in future version.
Best wishes,
Eric

Dear Eric

I have a piece of good news. I'm so happy that I can't help sharing with you.

I search "paw_tool" in log file in an occasional chance and find the "pawovlp" value, like this

Code: Select all

--- !WARNING
src_file: m_paw_tools.F90
src_line: 213
message: |
    PAW SPHERES ARE OVERLAPPING!
       There are     2 pairs of overlapping atoms.
       The maximum overlap percentage is obtained for the atoms  14 and  51.
        | Distance between atoms  14 and  51 is  :   3.57360
        | PAW radius of the sphere around atom  14 is:   1.41465
        | PAW radius of the sphere around atom  51 is:   2.21000
        | This leads to a (voluminal) overlap ratio of  0.06 %
even if they don't go beyond the setting value (Thanks for the developers of ABINIT). My Abinit version is 8.10.1.

Best wishes
Ke Tong

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: String Calculation issue with pawovlp

Post by ebousquet » Wed May 15, 2019 4:57 pm

OK, great!
Thanks,
Eric

Locked