[CP2K-user] [CP2K:13842] Re: NaN for Iodine atoms in Hirshfeld charges

[hut... at chem.uzh.ch]

Hi

I don't know the history of this problem, but I have fixed it in
the development version available on github. If you want to
fix a recent version yourself you can do so by adding a single line 
to the file

atom_fit.F

      ostate%state = 8
      CALL powell_optimize(ostate%nvar, x, ostate)
      CALL density_fit(density, atom, num_gto, x(1), x(2), co, ostate%f)      <--- new line

      CALL release_opgrid(density)

best regards

Juerg Hutter

PS The proble is related to use SHAPE_FUNCTION DENSITY. It should work
   in all versions with the default SHAPE_FUNCTION GAUSSIAN


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

-----cp... at googlegroups.com wrote: -----
To: cp... at googlegroups.com
From: "Lucas Lodeiro" 
Sent by: cp... at googlegroups.com
Date: 09/06/2020 11:25PM
Subject: [CP2K:13842] Re: NaN for Iodine atoms in Hirshfeld charges

I forgot to attach the files.

El dom., 6 sept. 2020 a las 4:16, Lucas Lodeiro (<eluni... at gmail.com>) escribió:
Hi all,

I want to obtain Hirshfeld-I charges for my system, as I do many times for other systems using self_consistent and density for shape function, in CP2K7.0. But the output charges prints NaN for the trace and the iodine charge.
I prove different combinations of self_consistent and shape function, and the only one which gives non NaN values is when Gaussian is used. I attach the files and input.

I use the same input using density for the shape function, in my laptop  with CP2K 6.1, and it works fine. Also, for CP2K7.0, if I change the iodine atoms to bromine ones, there is no problem... 
Which could be the problem?

I try to get more information with different level printing for hirshfeld property, but MEDIUM is the only one which works for 7.0 and 6.1.

I want to run AIMD and get the charges for each frame, and I only have the 7.0 version on the HPC...

Regards - Lucas Lodeiro
   
  -- 
 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 cp... at googlegroups.com.
 To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CAOFT4PLhqYTV4MpxuC7MXM82papX%2B8ng5xzsmohRB8%3D5CznBOg%40mail.gmail.com.
 

[attachment "bulk.zip" removed by Jürg Hutter/at/UZH]