First, as Gavin told, your convergence criteria are too bad to decide about 
your magnetic moments
just try to improve them in steps say up to ec 10^-6 cc 10^-3 and see what 
happens with the moments
(that was why I asked fror it)
however be sure you have enough k-points

I assume you used the same structure (with split V position) and you were 
setting the initial moments of V in case.inst to 
nm   0 MU_B both V
fm    3 mu_B both V
afm +3 mu_B -3 mu_B

now with GGA you find that all magnetic moments vanish (I assume 0.05 mu_B is 
just numerics)
that is you found a simple metal without magnetic moment that is most probably 
a paramagnet
(try to do a calculation for the fm start values using a very high magnetic 
field to check that)

If you plot the density of states or band structures, you will find that they 
are rather the same in all three cases.

What you missed  may be is a Mott insulator as for example NiO. In that case 
LDA+U is a good choice and you should find
fm = 2.5 mu_B   afm  =  +- 2.5 mu_B 
or at least something between 2 and 3 muB at the V atoms depending on your U 
you should NOT start from the GGA or any LDA solution as they are non magnetic 
and the minimum of the magnetic solution
might be to far from the nonmagnetic one to be found, just start from scratch.
(that's why I asked for it)
You will also find that at least the afm solution has a gap.

in the same way you should find the magnetic solutions when using EECE or HYBR
with magnetic moments of at least about 2 mu_B.

So far, the magnetic moments might be from the spheres, a better way would be 
to use a Baader AIM analysis
to find the moments in the atomic basins here you can use the aim programs 
coming with Wien2k or the program CRITIC 2 from the software goodies.

Actually the existence of the "magnetic" solutions do NOT mean that this are 
the ground states and that those exist !

a) you should do an optimization of the lattice to find the structure for each 
magnetic state (for nm you may force a non-magnetic solution)
and then check which one has the lowest energy.

b) even this does not mean that the so determined state exists. Here you have 
to check whether the crystal is stable at all.
This can be done by calculating the elastic constants and veryfing the 
Born-Huang criteria and by calculating the phonons
to check that no soft modes appear that make the crystal unstable. 
Further you need to check that no soft magnon modes appear, otherwise the 
magnetic state is not stable.

About Hund's rules, please read and understand F Hund; Z. Phys. 33 (1925) 345
to find that those empirical rules are for free ATOMS.
There is a big difference between atoms and solids.
In atoms you have definit states that are occupied by integer numbers of 
In solids you have energy bands that my be occupied only partially
Immidiately you see that the results are not necessarily the same.
Magnetism in solids is a COLLECTIVE phenomenon (that is not possible in atoms) 
and the results may coincide or not depending whether or not you have purely 
localized, purely itinerant or mixed cases.
If one assumes that there are only covalent or ionic bonds then one just misses 
metals (and also rare gas solids).

There are many models to describe atoms, molecules and solids, however, one 
also needs to check which of those approximations is valid and applicable.


DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy:
"I think the problem, to be quite honest with you,
is that you have never actually known what the question is."

Dr. Gerhard H. Fecher
Institut of Inorganic and Analytical Chemistry
Johannes Gutenberg - University
55099 Mainz
Max Planck Institute for Chemical Physics of Solids
01187 Dresden
Von: Wien [] im Auftrag von Abderrahmane 
Reggad []
Gesendet: Donnerstag, 7. September 2017 17:11
Betreff: Re: [Wien] About the magnetic moment of vanadium in vanadium sulphide

Hi All

I have used the PBE+EECC calculation for 3 configurations: nm, fm and afm I and 
I found that the afm I is the most stable.

The energy criterion and charge are 0.001 Ry and 0.001 e respectively.

I don't worry about if the material is really antiferromagnetic or paramagnetic 
because of:

1- I found only one experimental study that they found the compound to be pauli 
magnetic and one theoritical study which they found the compound to be non 
magnetic and these two studies are not sufficient to judge the compound to be 
in a such state. The theoritical study used the GGA method which is not good 
for correlated systems.

2- In the anfiferromagnetic state afm I in the NiAs structure for vanadium 
sulphide I found the following results:

MMI for V1: 0.05 MB
MMI for V2 :- 0.05 MB
MMI for S:    0 MB

My questions are now:

what's the definition of non magnetic compound ?

I think we can talk about non magnetic calculation and not about non magnetic 

As Blaha said we can't silulate the paramagnetic state or at at least it's 
difficult to do it because we can't orientate the spins randomly ang maintain 
the total magnetic moment equals to zero.

Because of the Hind's prediction and because the impaired number of the V2+ ion 
to equal 3 I believe the atomic magnetic moment to be different from zero.

Best regards
Wien mailing list

Reply via email to