Calculation of hopping integrals?

Message

maartendft · #1 Post by **maartendft** » Thu Apr 12, 2012 2:13 am

Dear VASP-users,

I'm studying bonding characteristics in HCP transition metals alloys with open d-shells. I'm interested to see how bonding and structural parameters change as a function of electron/atom-ratio.

Does anybody know how it would be possible to compute the 3 fundamental d-d hopping integrals between a given pair of atoms (p1,p2) in my HCP-lattice from VASP-data? I'm a bit worried this is a difficult exercise or maybe not possible with plane-wave codes.

I know how to calculate dz2, d(x2-y2) dxy, dxz, dyz etc. from VASP, but I'm not sure how to translate this into hopping integrals for HCP-lattices.

Many thanks!

#2 Post by **admin** » Thu Apr 12, 2012 4:48 pm

hopping integrals are typical to tight-binding approaches, based
on localized orbitals.

maartendft · #3 Post by **maartendft** » Thu May 03, 2012 6:41 am

Would it be feasible to obtain approximate hopping integrals via the Hamiltonian, in the Wannier function-approach using Wannier90? So, first running VASP and then running Wannier90 to parametrize the Hamiltonian.

I'm wondering if this would work for metals. Does anybody have experience with this?

Thanks!!