Question of TB2J-OpenMX

[Tong Zhao]

Hello，

I am trying to calculate J using  tb2j-openmx   and to calculate curie temperature by Vampire. I have obtained 'exchange.out' and input files of Vampire. And I have following questions:

1. The J_ani term in 'exchange.out' does not correspond to the tensorial interactions in the 'vampire.UCF' input files. Were any processes undertaken in addition to unit conversion that could have led to this inconsistency?

2. How can I derive the single-ion anisotropy term using tb2j-openmx? I have obtained the magnetic anisotropy energy per unit cell using VASP. However, since the three magnetic atoms in the unit cell are not equivalent, I cannot simply divide by three.

3. Is tb2j-openmx  a linear response method based on magnetic force theory to solve the exchange constant？

 I have attached here the exchange.out and vampire.UCF files. Thank you.  

[Xu He]

Hello, 

-1 Indeed from the Vampire J tensor you don't recover the exchange.out Jani. That is because the trace of the Jani is not 0 in exchange.out. A trace(Jani)/3 is therefore moved from the anisotropic part to the isotropic part.  Indeed the Jani should have trace 0, and perhaps it is good to do the same thing in the exchange.out as in the the vampire files.

-2. Currently the SIA is not implemented in TB2J. 

-3. Yes. 

Best regards, 

HeXu

--
You received this message because you are subscribed to the Google Groups "TB2J" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tb2j+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/tb2j/f6c256ee-00bd-4424-8a00-75d2dd72456bn%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.