I know that the linear response approach is implemented in VASP but not yet documented. But, the instructions are as follows
i) A well-converged DFT calculation without any additional Hubbard potentials.
ii) Then a Non-SCF calculation with Hubbard potential turned on.
iii) Finally, a SCF calculation with the shifting potential turned on.
Now, how to compute \chi_IJ = dn^I_d / d \alpha_J to get the U value?
Does anyone have any idea to compute this from VASP results?
What for?
Imagine that you are dealing with CoFeMnSi. Here the 3 U values
are needed: Uco, Ufe, and Umn. I think it is better to use any hybrid functional for exchange-correlation energy. The Hubbard U means only just approach of the mean field theory.
Regards,
Dr., Prof. Associate S. Syrotyuk.
Thank you Prof. Syrotyuk for your answer. I need to know how to calculate the Hubbard U parameter precisely for each magnetic element present in a compound using linear response approach. Then we have to use the calculated U value for DFT+U calculations.
Anfortunately, the U parameter depends of system. It means that it is not transferable to another material. Researchers consider U as ajustable parameter. I could understand your desire to find the U unless in the application of it in the DMFT theory.
- Badges
- Science topic
- Similar topics
- Engineering
- Control Theory