Strange behavior when using PM6

[Abdullah Bin Faheem]

Greetings everyone,

I was using PM6 to run semi-empirical MD of LiTFSI in DME and I noticed some strange behavior. The net charge obtained from population analysis after each step is dependent on the number of processors I use.

If I use a single processor, then the output file shows a net charge of 0, which is normal.

 

     682      Hy        7                 0.853347                      0.146653

     683      Hy        7                 0.892861                      0.107139

     686      Hy        7                 0.891219                      0.108781

     687      Hy        7                 0.853434                      0.146566

     688      Hy        7                 0.830966                      0.169034

 # Total charge                        1850.000000                     -0.000000

However, when using 48 processors, the net charge is 86950.00.

     682      Hy        7                 0.853347                     47.146653

     683      Hy        7                 0.892861                     47.107139

     686      Hy        7                 0.891219                     47.108781

     687      Hy        7                 0.853434                     47.146566

     688      Hy        7                 0.830966                     47.169034

 # Total charge                        1850.000000                  86950.000000

I observed similar behavior when using the example \cp2k\tests\SE\regtest-3-3\wfn_ex1.inp.

Is this normal? I have attached both the output and input files.

Thank you for your time.

Best regards,

Abdullah

[hut...@chem.uzh.ch]

Hi

thank you for reporting. This is a bug in the code. It has been fixed
in the last Trunk version with PR 1040.

regards

Juerg Hutter
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie C FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: "cp2k" <cp...@googlegroups.com>
From: "Abdullah Bin Faheem"
Sent by: cp...@googlegroups.com
Date: 08/14/2020 05:38PM
Subject: [CP2K:13781] Strange behavior when using PM6

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/f7297918-516c-49c4-97f0-0ca9398c5be4o%40googlegroups.com.


[attachment "litfsi.inp" removed by Jürg Hutter/at/UZH]
[attachment "litfsi_1_proc.out" removed by Jürg Hutter/at/UZH]
[attachment "litfsi_48_proc.out" removed by Jürg Hutter/at/UZH]

[abdullahb...@gmail.com]

Is it a bug in just printing out the partial charges in the output file or is it a bug in the inherent calculation. I ask as the values of partial charges are not required in my work.

Regards,

Abdullah

[hut...@chem.uzh.ch]

The bug is just in printing the Mulliken charges.

--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie C FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: "cp2k" <cp...@googlegroups.com>

From: "abdullahb...@gmail.com"
Sent by: cp...@googlegroups.com
Date: 08/17/2020 05:21PM
Subject: Re: [CP2K:13787] Strange behavior when using PM6

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/a8d07c16-db2c-45c1-8b87-83b8dab209fan%40googlegroups.com.

[abdullahb...@gmail.com]

Thank you for your help.