problem of band structure( with SO coupling) interpolation

External optional components of Abinit (BigDFT, Wannier90...) and codes using/providing data from/to Abinit (AtomPAW, ONCVPSP, EXC, DP, Yambo...)

Moderators: ebousquet, bxu

Locked
tanyaohua
Posts: 1
Joined: Mon Feb 27, 2012 11:01 am

problem of band structure( with SO coupling) interpolation

Post by tanyaohua » Mon Feb 27, 2012 2:04 pm

Hello all.
I had a problem when doing band interpolation for a band with spin orbit coupling using Abinit-Wannier90 interpolation.
In the dft calculation I use option "nspinor 2" and using the "hgh" type pseudo potentials.

In the input files for wannier90, I set "num_wan" to be twice as many as the number of projectors according to the users guide of wannier90.
And I also set "spinors = .true.". Actually this input is neglected because I use W90 library mode.
The W90 with stopped because of error like this:
" param_get_projections: Fewer projections defined than the number of Wannier functions requested"

If I set "num_wan" to a number equal to the number of projectors.
The error will be like

"Exiting.......
dis_windows: More states in the frozen window than target WFs"

Can any one tell me what the problem is? Or did anyone successfully do the similar thing that I am doing?

Locked