[CP2K-user] [CP2K:12499] SCF problem on single atom adsorption

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

Hi

I think the problem is the initial guess. You can go with the default
(as you did) and use a robust method to start optimization. 
Later on you can switch back to DIIS for geometry optimizations.

This worked for me (100 steps, then restart using DIIS needed another 75 steps)

     &OT
        MINIMIZER CG
        LINESEARCH 3PNT
        PRECONDITIONER FULL_SINGLE_INVERSE
     &END OT

Expert method would be to create an advanced initial guess. Single out
the adatom O as a new kind and define an initial guess using
MAGNETIZATION.

Using a convergence of 10^-6 or even 10^-5 for starting the geometry optimization
would also be an option. Tighten the criteria towards the end to get better forces.

regards

Juerg Hutter
--------------------------------------------------------------
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: "cp2k" <cp... at googlegroups.com>
From: "Stefano Ferrero" 
Sent by: cp... at googlegroups.com
Date: 11/16/2019 09:20PM
Subject: [CP2K:12499] SCF problem on single atom adsorption

Dear CP2K experts,

I have just started using CP2K and I would be really thankful if anybody can help me.
I have tried to run single point calculations of single atoms (H, O, N, C) and then geometry optimizations at PBE-D3 level of theory on a periodic ice slab.
 I managed to optimize the adsorption of the Hydrogen atom without problems, therefore I just used the same input (making the required adjustments) for the optimization of the Oxygen atom but then some problems arose: the SCF procedure couldn't converge even after many cycles.
I am not very familiar with all the options that CP2K has. I just wanted to know if there is a particular set up for the SCF procedure (and for geometry optimizations), to use when working with single atoms and their relative adsorption on a non metallic system like an ice slab.
I'll attach the relevant files for the geometry optimization of the oxygen atom below.     
Thank you for your help!


 
 
------------------------

Indirizzo istituzionale di posta elettronica degli studenti e dei laureati dell'Università degli Studi di Torino
Official University of Turin email address for students and graduates   
  -- 
 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/87f94ebf-36cb-4c2e-bd6e-ae514b7e935f%40googlegroups.com.
 

[attachment "o_atom_slab_ih.inp" removed by Jürg Hutter/at/UZH]
[attachment "o_atom_slab_ih.xyz" removed by Jürg Hutter/at/UZH]