Scaling DFT exchange values for MnFe2O4 input

[Maryam]

Hi everyone,

I am simulating MnFe2O4 using VAMPIRE and have calculated exchange interactions using DFT (values in meV).
Could anyone clarify whether I should input the raw DFT energy values directly into the 'exchange-matrix', or if they require scaling (e.g., by spin moments) to match the VAMPIRE Hamiltonian?
My calculated values are:
Mn-Mn (A-site): +1.28 meV
Fe-Fe (B-site): -1.84 meV
Mn-Fe (A-B): -1.84 meV

Thanks for your help.

[gabo...@gmail.com]

I believe the Hamiltonian and exchange energy equations should be known from what was used to compute the exchange values from your DFT calculation.  The DFT Hamiltonian might be defined differently depending on what program you have used (VASP, QE, WIEN2k, SIESTA, or other).  By comparing that to VAMPIRE's Hamiltonian definition, the factor and/or unit differences between the two should be determinable.  

The VAMPIRE Hamiltonian given under the FAQ question "There are numerous definitions of the exchange in the literature. How do I calculate the exchange constant for vampire?" that should be at [1], you might find helpful.

[1] https://vampire.york.ac.uk/support/

Kind Regards,

Gavin

VAMPIRE user