Page 1 of 1
phase in PROCAR for LORBIT=12
Posted: Mon Feb 10, 2020 6:12 am
by xiaoming_wang
Hello,
Can you please confirm that if the second block of the PROCAR file for LORBIT = 12 are the QUANTITATIVE projections of <p_i | psi_nk>, where p_i is the projector function and psi_nk is the pseudo ks orbital?
Best,
Xiaoming
Re: phase in PROCAR for LORBIT=12
Posted: Fri Mar 06, 2020 2:37 pm
by henrique_miranda
Dear Xiaoming,
We updated the documentation of PROCAR AND LORBIT in our Wiki:
wiki/wiki/index.php/LORBIT
wiki/wiki/index.php/PROCAR
Let us know if with the updated documentation your question is answered.
Re: phase in PROCAR for LORBIT=12
Posted: Fri Mar 06, 2020 5:58 pm
by xiaoming_wang
Hi,
Thanks for your reply. It's really helpful. It seems to me that LORBIT=12 combine the projections from different PAW projectors while LORBIT=14 give a better projection on a single projector. Besides, I found that the projection for each projector can be printed separately with the keywork LOCPROJ. Are those wave functions to be projected (I mean both LORBIT and LOCPROJ) the AE wave funcitons psi_nk or the pseudo wave functions tiled_psi_nk in the following equation?
psi_nk = tiled_psi_nk + sum_i (phi_i - tiled_phi_i) < p_i | tiled_psi_nk >
What I really would like to know is the last term in the bracket which is frequently used in the vasp implementation. Any comments on how to print out that?
Best,
Xiaoming
Re: phase in PROCAR for LORBIT=12
Posted: Tue Mar 10, 2020 8:00 pm
by xiaoming_wang
Hi,
After looking at the source codes, the wave functions to be projected are the AE ones. The pseudo wave function characters <p_i | tilde_psi_nk > are stored in W%CPROJ(nproj,nb,nk,nspin).
Best,
Xiaoming