[Wien] Spin-polarized state not really spin-polarized

fabien.tran at vasp.at fabien.tran at vasp.at
Sun Oct 30 17:47:15 CET 2022 

I don't think that it is worth using the FSM method. The calculation 
started with non-zero moments (FM state) which at the end disappeared, 
which is already an indication (at least with PBE). In addition, 
magnetism in solids is usually expected when there are transition-metal 
atoms, which is not the case here. As Xavier mentioned, SOC should be 
considered for such heavy atoms.


On 30.10.2022 16:30, pboulet wrote:
> All right, so here are the MMTOT data:
> 
> Starting point of SCF: 123.85779
> Converged: 0.05631
> 
> And MMI ones:
> Starting point:
> 
> :MMINT:  MAGNETIC MOMENT IN INTERSTITIAL  =   71.11022
> :MMI001: MAGNETIC MOMENT IN SPHERE   1    =    1.03742
> :MMI002: MAGNETIC MOMENT IN SPHERE   2    =    1.03736
> :MMI003: MAGNETIC MOMENT IN SPHERE   3    =    0.62202
> :MMI004: MAGNETIC MOMENT IN SPHERE   4    =    0.62205
> :MMI005: MAGNETIC MOMENT IN SPHERE   5    =    1.03746
> :MMI006: MAGNETIC MOMENT IN SPHERE   6    =    0.62203
> :MMI007: MAGNETIC MOMENT IN SPHERE   7    =    0.62196
> :MMI008: MAGNETIC MOMENT IN SPHERE   8    =    1.03238
> :MMI009: MAGNETIC MOMENT IN SPHERE   9    =    0.62236
> :MMI010: MAGNETIC MOMENT IN SPHERE  10    =    0.29692
> 
> Converged:
> 
> :MMINT:  MAGNETIC MOMENT IN INTERSTITIAL  =    0.04102
> :MMI001: MAGNETIC MOMENT IN SPHERE   1    =    0.00000
> :MMI002: MAGNETIC MOMENT IN SPHERE   2    =   -0.00015
> :MMI003: MAGNETIC MOMENT IN SPHERE   3    =    0.00028
> :MMI004: MAGNETIC MOMENT IN SPHERE   4    =    0.00029
> :MMI005: MAGNETIC MOMENT IN SPHERE   5    =   -0.00003
> :MMI006: MAGNETIC MOMENT IN SPHERE   6    =    0.00030
> :MMI007: MAGNETIC MOMENT IN SPHERE   7    =    0.00027
> :MMI008: MAGNETIC MOMENT IN SPHERE   8    =    0.00104
> :MMI009: MAGNETIC MOMENT IN SPHERE   9    =    0.00038
> :MMI010: MAGNETIC MOMENT IN SPHERE  10    =    0.00128
> 
> Obviously the system converges towards a non-spin polarized state.
> 
> From the literature, there has been some experimental investigation
> on, e.g., Pb(1-x)Tl(x)Te (x=0.001-0.02). One can read: [..] Various
> mechanisms** which can lead to observable anomalies, including
> Kondo-like behavior of a non-magnetic degenerate two-level system are
> discussed.
> 
> So maybe the structure is non-magnetic.
> 
> ** related to thermoelectric power
> 
> Now let’s say I want to make sure this is a non-magnetic compound by
> enforcing a magnetic state (in which case the total energy should be
> higher than for the non-magnetic state), I should run runfsm_lapw and
> change case.inst to enforce a spin polarization right at the
> beginning, shouldn’t I?
> 
> Pascal
> 
>> Le 30 oct. 2022 à 14:04, fabien.tran at vasp.at a écrit :
>> 
>> Dear Pascal,
>> 
>> Depending on the system it may be possible to stabilize more than
>> one magnetic state. In such cases, the magnetic state obtained at
>> the end of the calculation typically depends on the initial magnetic
>> state when starting the calculation. What was the initial magnetic
>> state in your calculation? Grep for :MMTOT (total moment in cell) or
>> :MMI (moment on atoms) in case.scf to see how these quantities
>> evolved during the SCF procedure. Is Pb31TlTe32 supposed to be
>> magnetic according to experiment?
>> 
>> On 30.10.2022 13:07, pboulet wrote:
>> 
>>> Dear all,
>>> I am investigating Pb31TlTe32 in which Tl is the only element that
>>> bring an odd number of electrons.
>>> I have set up a spin-polarized calculation with init_lapw, but not
>>> with an anti-ferromagnetic state.
>>> As a starting point, I do not include spin-orbit and I use PBE.
>>> NOE=959 in the structure.
>>> After converging the SCF, I end up with the following (to me
>>> strange)
>>> occupation states:
>>> For spin up:
>>> :BAN00479: 479    0.272337    0.309267  1.00000000
>>> :BAN00480: 480    0.283605    0.328642  0.50431432
>>> :BAN00481: 481    0.371927    0.455285  0.00000000
>>> For spin down:
>>> :BAN00479: 479    0.272405    0.309306  1.00000000
>>> :BAN00480: 480    0.283720    0.328787  0.49568569
>>> :BAN00481: 481    0.372018    0.455369  0.00000000
>>> I rather expected to have 480 spin up occupied states with 1
>>> electron
>>> and 479 spin down occupied states with 1 electron, but I have
>>> something like a closed-shell spin polarized state.
>>> Is it what we should expect?
>>> If not, could you please explain me what happens and eventually
>>> how to
>>> remedy this to have a ‘real’ spin polarized state?
>>> Thank you
>>> Pascal
>>> Pascal Boulet
>>> —
>>> _Professor in computational materials chemistry - DEPARTMENT OF
>>> CHEMISTRY_
>>> University of Aix-Marseille - Avenue Escadrille Normandie Niemen -
>>> F-13013 Marseille - FRANCE
>>> Tél: +33(0)4 13 55 18 10 - Fax : +33(0)4 13 55 18 50
>>> Email : pascal.boulet at univ-amu.fr
>>> _______________________________________________
>>> Wien mailing list
>>> Wien at zeus.theochem.tuwien.ac.at
>>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>> SEARCH the MAILING-LIST at:
>>> 
>> 
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>> SEARCH the MAILING-LIST at:
>> 
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> 
> Pascal Boulet
> —
> _Professor in computational materials chemistry - DEPARTMENT OF
> CHEMISTRY_
> 
> University of Aix-Marseille - Avenue Escadrille Normandie Niemen -
> F-13013 Marseille - FRANCE
> Tél: +33(0)4 13 55 18 10 - Fax : +33(0)4 13 55 18 50
> Email : pascal.boulet at univ-amu.fr
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

More information about the Wien mailing list