[Wien] Hund's Exchange parameter (J)

[Gavin Abo]

Sorry, I don't know, but maybe more details about your calculation would 
be needed for someone else to answer your question.

For example, are you doing something similar to the Constraint U 
exercise [ 
http://susi.theochem.tuwien.ac.at/reg_user/textbooks/Constraint_U.pdf ], 
which uses the F^0_eff formula of EPL vol. 69, 777 (2005) [ 
https://doi.org/10.1209/epl/i2004-10416-x , 
https://arxiv.org/abs/cond-mat/0412560v1 (equation 4)]?

For NiO [ 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg14460.html 
, 
http://wien2k-algerien1970.blogspot.com/2016/08/how-to-calculate-u-parameter-in-nickel.html 
], I think that gave about:

ε3d_up = 0.119

ε3d_dn = -0.282

If those are the same parameters as in equation 11 of Phys. Rev. B vol. 
44, 943 (1991) [ https://doi.org/10.1103/PhysRevB.44.943 ], then it 
looks like that would be:

J = ε3d_up - ε3d_dn = 0.12 - (-0.28) = 0.401

Though, one has to keep in mind that the Ueff values obtained from the 
Constrained method are known to be of questionable accuracy [ 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg15599.html 
].  So, the J values might be of questionable accuracy as well. In 
addition, well converge results can be needed for such calculations [ 
https://www.researchgate.net/post/How_to_calculate_U_and_or_J_values_of_DFT_U 
].

On 5/12/2018 6:28 AM, Wien2k User wrote:
> Dear Prof. Tran and WIEN2k users;
>
> I found in Anisimov's paper the following relation to calculate J
> J=E3d'up'(n+0.5,n-0.5)-E3d'dn'(n+0.5,n-0.5)
> but I think that the energy of the spin-up electrons is more stable 
> compared to that of Dn electrons and that's why I found a negative 
> energy (Do you think that this relationship is reversed?) or my 
> reasoning is wrong and I have to review my calculations?
>
> Thank you in advance
>
> 2018-05-11 18:47 GMT+02:00 Wien2k User <wien2k.user at gmail.com 
> <mailto:wien2k.user at gmail.com>>:
>
>     Thank you very much
>
>     2018-05-11 16:09 GMT+02:00 Wien2k User <wien2k.user at gmail.com
>     <mailto:wien2k.user at gmail.com>>:
>
>         Dear Prof. P. BLAHA and WIEN2k users;
>
>         I want to estimate the effective value of Hubbard term  "Ueff=U-J"
>
>         Can the Hund's Exchange parameter (J) be negative?
>
>         Thank you very much.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20180512/1b378908/attachment-0001.html>