about the hamiltonian convention
chao zhou
Dear Vampire Development Team,
Greetings! My name is zhouchao, and I am currently using Vampire for atomistic spin dynamics simulations in combination with density functional theory (DFT) calculations to study the properties of magnetic materials. Thank you for developing and maintaining such a powerful tool, which has been immensely helpful for our research!
I am writing to seek your guidance on a few technical questions regarding the Heisenberg Hamiltonian convention in Vampire and the setup of related parameters. Below are my specific inquiries:
Consistency of the Heisenberg Hamiltonian with DFT-extracted J values
In the Vampire documentation , the spin Hamiltonian is shown in Figure.
In DFT calculations, we typically extract the exchange coupling Jij via energy mapping with the Hamiltonian defined as (shown in Figure 2):
Both formulations appear to avoid double counting . Could you confirm whether these two definitions are theoretically equivalent? If there are subtle differences (e.g., in summation conventions or sign definitions), is any adjustment needed when using DFT-extracted Jij values as input for Vampire?Spin normalization and units of exchange interactions
In Vampire’s material definitions, if we normalize the spins to 1 and define the atomic magnetic moment as 2 muB (Bohr magneton) in the material file, can the exchange coupling in the unified calculation format (UCF) directly use the DFT-fitted Jij values (assuming they are converted to Joules)?
Thank you very much for your time and assistance! I look forward to your response. If further details are needed (e.g., specific DFT methods or Vampire input files), I’d be happy to provide them.
zhouchao
