[QE-users] NV0 Center - Absolute Magnetization

[Paolo Giannozzi]

Funny. Look at the electronic structure: with smearing, you have two
degenerate states with up spin at the Fermi energy, atoms n.71, 125, 179
(those close to the vacancy) with the same spin polarization, 0.15 or so
(whatever it means: atomic charges and polarizations reprinted by the code
should not be taken too seriously). By fixing occupancies, you break the
degeneracy and get a sort of "broken-symmetry" solution, with a gap, two of
the three atoms with a larger polarization (0.29) and one with opposite
polarization (-0.18). I guess that in order to reproduce the former
solution with fixed occupancies, you have to set occupancies (from input)
as 1 1 1 ... 0.5 0.5 0 0 ... for spin up, 1 1 1 ... 0 0 ... for spin down.

Paolo

On Sun, Oct 13, 2019 at 4:03 AM Brendan Smith <bsmith24 at buffalo.edu> wrote:

> Hi QE Experts,
>
> Please know that before sending this question, I have done my best to
> extensively read the existing body of past forum messages and other QE
> related learning materials.
>
> I am currently studying NV centers in bulk diamond, and I have a question
> regarding a single NV0 (charge neutral) center I have made in a 3x3x3 (216)
> atom supercell of bulk diamond. In this system, we have an odd number of
> electrons, leaving a single lone electron (spin-up in this case).
> Therefore, I expect that my total magnetization should be 1.0 Bohr
> mag/cell, and I see this consistently in my QE outputs. I am comfortable in
> thinking that my system is ferromagnetic, based on my deductions from
> reading the literature. However, I have some discrepancy when it comes to
> the absolute magnetization, and it is here where my question lies.
>
> When I fix the total magnetization to be = 1.0 Bohr mag/cell, and use a
> fixed occupations scheme, the absolute magnetization for some reason
> converges to a value of 2.11 Bohr mag/cell. I find this to be strange,
> since it should be around the value of the total magnetization, which is 1
> .0 Bohr mag/cell. What this tells me is that for some reason, perhaps I am
> converging to some anti-ferromagnetic solution. Then again, I don't know if
> this is even possible given that I am using fixed occupations.
>
> Next, when I use occupations = "smearing", ("cold") and set only a guess
> starting magnetization (do not fix total magnetization), I find that I
> actually get a value for BOTH the total and absolute magnetization to be
> around 1.0 Bohr mag/cell, which is more or less what I expect to see, given
> that I believe that my system is rightfully a simple ferromagnetic system.
>
> So my question is, what the heck is going on in the case of fixed
> occupations? why am I converging to  solution in which the total
> magnetization is 1.0 Bohr mag/cell, and the absolute magnetization is 2.11
> Bohr mag/cell? Is there any apparent misunderstanding I may be having in my
> analysis? I seek your expertise in this regard.
>
> I have attached my PWscf input and output files. The output file that was
> used with smearing is labeled "x0.scf_smear.out", and the output file in
> which fixed occupations were used along with setting tot_magnetization is
> labeled "x0.scf_fixed.out". The input file is labeled "x0.scf.in". To run
> the calculations with the fixed occupations, I just comment out the
> smearing part and uncomment the tot_magnetization line.
>
> Thank you for your time in reading my email. I look forward to any
> suggestions.
>
> Best,
> Brendan A. Smith
> Ph.D candidate at the State University of New York at Buffalo
>
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users



-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20191014/b200bcfb/attachment.html>