[CP2K-user] [CP2K:12214] Re: Spin Polaization in Kohn-Sham calculations
Krack Matthias (PSI)
matthi... at psi.ch
Thu Sep 12 13:01:42 UTC 2019
Dear Ron
Please, could you provide the full input file for testing. A value of 3 Hartree for RELAX_MULTIPLICITY is most likely inappropriate. The argument of that keyword is not the (preferred) multiplicity but an energy tolerance allowing for a spin flip between alpha and beta electrons. The assumed multiplicity should still be given via the MULTIPLICITY keyword. In the case of O2 starting with an initial (guessed) multiplicity of 1, the spin flip should happen already in the first SCF iteration step and it should not change thereafter.
You should check the output line starting with “Ideal and single determinant S**2” for the final multiplicity. I get for a gas phase O2 molecule starting with
MULTPLICITY 1
RELAX_MULTIPLICTY 0.01
something like:
Ideal and single determinant S**2 : 2.000000 2.000784
which looks reasonable to me.
HTH
Matthias
From: cp... at googlegroups.com <cp... at googlegroups.com> On Behalf Of Ronald Cohen
Sent: Donnerstag, 12. September 2019 14:21
To: cp... at googlegroups.com
Subject: Re: [CP2K:12214] Re: Spin Polaization in Kohn-Sham calculations
Dear Juerg,
Thank you so much. Here is what I get using the latest version on github:
When I have in the input:
UKS T
# MULTIPLICITY 3
RELAX_MULTIPLICITY 3
&KIND O2
BASIS_SET DZVP-MOLOPT-GTH-q6
ELEMENT O
POTENTIAL GTH-PBE-q6
&POTENTIAL
2 4
0.2445543000000000E+00 2 -0.1666721480000000E+02 0.2487311320000000E+01
2
0.2209559200000000E+00 1 0.1833745811000000E+02
0.2113324700000000E+00 0
&END POTENTIAL
MAGNETIZATION 2
&END KIND
The output shows:
DFT| Multiplicity 1
DFT| Number of spin states 2
Spin 1
Number of electrons: 38
Number of occupied orbitals: 38
Number of molecular orbitals: 40
Spin 2
Number of electrons: 38
Number of occupied orbitals: 38
Number of molecular orbitals: 40
Number of orbital functions: 210
Number of independent orbital functions: 210
and this number of electrons is repeated, even though there is a moment:
grep -B 3 "Number of electrons:" Opt.out
Spin 1
Number of electrons: 38
--
Spin 2
Number of electrons: 38
--
Spin 1
Number of electrons: 38
--
Spin 2
Number of electrons: 38
--
Spin 1
Number of electrons: 38
--
Spin 2
Number of electrons: 38
--
…
When I have this in the input:
UKS T
MULTIPLICITY 3
RELAX_MULTIPLICITY 3
I see:
DFT| Multiplicity 3
DFT| Number of spin states 2
and
grep -B 1 -A 2 "Re-scal" Opt.out.1
Spin 1
Re-scaling the density matrix to get the right number of electrons for spin 1
# Electrons Trace(P) Scaling factor
39 40.000 0.975
--
Spin 2
Re-scaling the density matrix to get the right number of electrons for spin 2
# Electrons Trace(P) Scaling factor
37 36.000 1.028
--
Spin 1
Re-scaling the density matrix to get the right number of electrons for spin 1
# Electrons Trace(P) Scaling factor
39 40.000 0.975
--
Spin 2
Re-scaling the density matrix to get the right number of electrons for spin 2
# Electrons Trace(P) Scaling factor
37 36.000 1.028
--
Spin 1
Re-scaling the density matrix to get the right number of electrons for spin 1
# Electrons Trace(P) Scaling factor
39 40.000 0.975
--
Spin 2
Re-scaling the density matrix to get the right number of electrons for spin 2
# Electrons Trace(P) Scaling factor
37 36.000 1.028
--
Spin 1
Re-scaling the density matrix to get the right number of electrons for spin 1
# Electrons Trace(P) Scaling factor
39 40.000 0.975
--
Spin 2
Re-scaling the density matrix to get the right number of electrons for spin 2
# Electrons Trace(P) Scaling factor
37 36.000 1.028
--
Spin 1
Re-scaling the density matrix to get the right number of electrons for spin 1
# Electrons Trace(P) Scaling factor
39 40.000 0.975
--
Spin 2
Re-scaling the density matrix to get the right number of electrons for spin 2
# Electrons Trace(P) Scaling factor
37 36.000 1.028
--
Spin 1
Re-scaling the density matrix to get the right number of electrons for spin 1
# Electrons Trace(P) Scaling factor
39 40.000 0.975
…
So it seems it is rescaling over and over? Or is there just a problem with the printout?
Thank you again,
Ron
---
Ronald Cohen
Extreme Materials Initiative
Geophysical Laboratory
Carnegie Institution
5251 Broad Branch Rd., N.W.
Washington, D.C. 20015
rco... at carnegiescience.edu<mailto:rco... at carnegiescience.edu>
office: 202-478-8937
skype: ronaldcohen
twitter: @recohen3
On 12. Sep 2019, at 07:29, hut... at chem.uzh.ch<mailto:hut... at chem.uzh.ch> wrote:
Hi
the renormalization is just information from the initial guess.
The superposition of atomic densities can result in initial
densities that are not representing the requested spin state.
A rescaling is used to fix that. As you can see in the output, all
of this happens before KS interations start.
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<mailto:hut... at chem.uzh.ch>
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------
-----cp... at googlegroups.com<mailto:cp... at googlegroups.com> wrote: -----
To: cp... at googlegroups.com<mailto:cp... at googlegroups.com>
From: "Ronald Cohen"
Sent by: cp... at googlegroups.com<mailto:cp... at googlegroups.com>
Date: 09/12/2019 01:22PM
Subject: Re: [CP2K:12212] Re: Spin Polaization in Kohn-Sham calculations
Oh I understand. Of course that need not be integer. So the printed multiplicity when relax multiplicity is on is incorrect. What about the number of electrons? It prints a renormalization when Multiplicity =3 and also shows the same number of up anddien when multiplicity =1 even with relax multiplicity on. I can send that output if you like. Thank you! Ron
On Sep 12, 2019, at 07:15, hut... at chem.uzh.ch<mailto:hut... at chem.uzh.ch> wrote:
Hi
as Vladimir wrote, the value printed is the integrated ABSOLUTE
spin value. The integrated spin value is integer and given by
construction. The printed absolute value gives additional information
on localization of the different spins.
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<mailto:hut... at chem.uzh.ch>
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------
-----cp... at googlegroups.com<mailto:cp... at googlegroups.com> wrote: -----
To: cp... at googlegroups.com<mailto:cp... at googlegroups.com>
From: "Ronald Cohen"
Sent by: cp... at googlegroups.com<mailto:cp... at googlegroups.com>
Date: 09/12/2019 01:12PM
Subject: Re: [CP2K:12210] Re: Spin Polaization in Kohn-Sham calculations
I do not understand how you get a non-integer spin if you fill states one electron at a time for up and down states. Perhaps I do not understand ‘Integrated absolute spin density is non-integer.’ Thank you,
Ron
Sent from my iPhone
On Sep 12, 2019, at 06:45, Vladimir Rybkin <rybk... at gmail.com<mailto:rybk... at gmail.com>> wrote:
Integrated absolute spin density is non-integer.
--
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<mailto:cp... at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/836B5B54-F287-4500-BF4B-B1B93249CFC9%40carnegiescience.edu.
--
You received this message because you are subscribed to a topic in the Google Groups "cp2k" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cp2k/dFUIwOAfmWM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cp... at googlegroups.com<mailto:cp... at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/OF8C565986.64274F05-ONC1258473.003DD972-C1258473.003DD974%40lotus.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<mailto:cp... at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/C374FD68-AF16-4372-A542-E2AD68CE38C3%40carnegiescience.edu.
--
You received this message because you are subscribed to a topic in the Google Groups "cp2k" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cp2k/dFUIwOAfmWM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cp... at googlegroups.com<mailto:cp... at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/OF62AFAD73.D676EC20-ONC1258473.003F173F-C1258473.003F1743%40lotus.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<mailto:cp... at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/07C85059-A478-4A1A-B0E7-3ED7D0CD650D%40carnegiescience.edu<https://groups.google.com/d/msgid/cp2k/07C85059-A478-4A1A-B0E7-3ED7D0CD650D%40carnegiescience.edu?utm_medium=email&utm_source=footer>.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190912/69999a4b/attachment.htm>
More information about the CP2K-user
mailing list