[Wien] About the magnetic moment of vanadium in vanadium sulphide

[Fecher, Gerhard]

Now I am a little confused
do you think that it was the convergence criteria that I found a magnetic moment of 2.5 mu_B instead of 0.05 mu_B at the V ?
or did I make another mistake ;-)

Indeed, you are right, the 0.05 mu_B might be a result of to bad convergence
however, to check if such a moment is something reliable you will need brute force with respect to the number of k-points,
that is, if you think you have enough k-points, then just double the number or use ten times more.

Ciao
Gerhard

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
and
Max Planck Institute for Chemical Physics of Solids
01187 Dresden
________________________________________
Von: Wien [wien-bounces at zeus.theochem.tuwien.ac.at] im Auftrag von Gavin Abo [gsabo at crimson.ua.edu]
Gesendet: Freitag, 8. September 2017 03:14
An: wien at zeus.theochem.tuwien.ac.at
Betreff: Re: [Wien] About the magnetic moment of vanadium in vanadium sulphide

Perhaps the -ec 0.001 and -cc 0.001 are too large of values.

As I recall, to be well-converged, it is usually best to use about the default values seen in the post [1] or WIEN2k 17.1 usersguide [2] as:

-cc 0.0001
-ec 0.0001

It sounded like about the default value for -ec was good unless something like -ec 0.00001 was desired to reduce numerical noise, but anything smaller seemed useless [3,4].

For -cc, 0.00001 also may be the lowest limit [5] and quite ambitious to try to use [6].

In section "4.5.4 Antiferromagnetic (AFM) calculations" on page 46 in the usersguide [2], there is the statement:

"If nothing changes (E-tot and other properties), then you are ok, otherwise make sure the scf calculation is well converged (-cc 0.0001 or better)."

If I read from the statement correctly, a well converged scf calculation typically uses -cc with a value of 0.0001 or smaller.  Though, there is a realistic limit as mentioned above on how small it could be set.

In [7], it sounded important that the -cc value was low for the :MMIn values.

So, maybe the calculation can still converge further such that possibly the "MMI for V1" and "MMI for V2" will both become zero when they reach better convergence.

[1] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg12620.html
[2] http://susi.theochem.tuwien.ac.at/reg_user/textbooks/usersguide.pdf
[3] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg10650.html
[4] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg16077.html
[5] http://zeus.theochem.tuwien.ac.at/pipermail/wien/2008-November/011797.html
[6] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg11558.html
[7] https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg09231.html

On 9/7/2017 9:11 AM, Abderrahmane Reggad wrote:
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 compounds.

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