[CP2K-user] [CP2K:12145] Spin polarised calculation for bulk iron (16 atoms)

[Ari Paavo Seitsonen]

Dear Daniele,

  My few thoughts,

  What you could use instead of "MULTIPLICITY" is the keyword
"FIXED_MAGNETIC_MOMENT",
https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/DFT/SCF/SMEAR.html#list_FIXED_MAGNETIC_MOMENT
, if you indeed try to force a magnetic moment. Usually this is, however,
not done but the system is let to relax its electronic structure and find
the final magnetisation self-consistently. "MULTPLICITY" is meant for
systems with well-defined electronic shells (open- or closed-shell), with a
gap between the occupied and un-occupied states, whereas with the smearing
one goes for different tools and keywords. The suggestion was to used the
keyword "MAGNETIZATION" to get a reasonable initial guess and then let the
system relax and find its magnetic moment; if you have convergence
problems, you might try with "FIXED_MAGNETIC_MOMENT" in the first
calculation, once that converges you read the previous density (densities,
including the spin polarisation) and try without any constraint on the
magnetisation.

    Good Luck from Half-Sunny Paris,

       apsi

Le mer. 28 août 2019 à 11:42, Daniele Ongari <daniel... at gmail.com> a
écrit :

> Hi Tiziano,
> I'm interested in this topic because I have problem to make materials
> (MOFs) with Fe converge and I want to start to the bulk Fe as a test case.
>
> I wonder how is your suggestion supposed to work.
> 1) Without specifying the system MULTIPLICITY this is by default 0 (which
> is treated as 1 in CP2K) that means that no broken symmetry is occurring:
> spin1 and spin2 will be rescaled and basically the effect of specifying the
> MAGNETIZATION is null
> 2) If considering that the system has 16 Fe atoms, one used 1+16*2.3= 37.8
> as MULTIPLICITY (assuming ferromagnetic configuration), CP2K aborts with
> the error "An integer type object was expected".
>
> So, how can someone specify a MAGNETIZATION that doesn't sum up into an
> integer total magnetization?
>
> The problem I see is another, that I'm not able to make the calculation
> converge with MAGNETIZATION 2 and MULTIPICITY 1+16*2=33 (so no problems
> with non-integer numbers), but this is a problem to solve in the &MIXING
> section, right?
> Do you have any working settings for make the bulk Fe converge? I think
> you managed to do so for the Delta-test of Stefaan Cottenier.
>
> Thank you very much for your help
>
> Daniele
>
>
> Il giorno giovedì 26 aprile 2018 17:35:33 UTC+2, Tiziano Müller ha scritto:
>>
>> Hi Unmesh
>>
>> Am 25.04.2018 um 12:18 schrieb Unmesh Mondal:
>> > I am new to cp2k and wish to do a spin polarised calculation of bulk
>> > Iron  (Fe : 16 atoms, gamma point). Switching on UKS (spin unrestricted
>> > Kohn Sham) ensures the spin polarised part  and the MULTIPLICITY " "
>> > ensures the initial spin state of the whole system.
>>
>> Are you sure? Check your output right below "Guess for atomic kind: Fe".
>> This should tell you the used initial spin state.
>>
>> With CP2K 5.1 and later you can use the MAGNETIZATION keyword for the
>> KIND section (with a value of 2.3 for iron) to get a proper initial
>> state, see
>>
>> https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/SUBSYS/KIND.html
>>
>> If you are using an older version (<5.1) use the Broken Symmetry section
>> instead to achieve the same thing:
>>
>>
>> https://manual.cp2k.org/cp2k-4_1-branch/CP2K_INPUT/FORCE_EVAL/SUBSYS/KIND/BS.html
>>
>> Best regards,
>> Tiziano
>>
>> --
>> Tiziano Müller
>> University of Zurich
>> Department of Chemistry
>> Winterthurerstrasse 190
>> CH-8057 Zürich
>>
>> Tel: +41 44 63 54234
>> www.chem.uzh.ch
>> tiz... at chem.uzh.ch
>>
> --
> 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/706f2d35-d8b0-4368-8b24-8c764adcb809%40googlegroups.com
> <https://groups.google.com/d/msgid/cp2k/706f2d35-d8b0-4368-8b24-8c764adcb809%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>


-- 
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
  Ari Paavo Seitsonen / Ari.P.... at iki.fi / http://www.iki.fi/~apsi/
    Ecole Normale Supérieure (ENS), Département de Chimie, Paris
    Mobile (F) : +33 789 37 24 25    (CH) : +41 79 71 90 935
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190901/7f179cf8/attachment.htm>