[Wannier] error in wannier functions results for the specified projection orbitals

[刘乐雨]

Dear all,


1) Recently, I did a MLWF calculation for a single CrF4 molecule using VASP+Wannier90,and I wanted to get the final wannier functions for all the 3d orbitals and 4s orbital of Cr atom in CrF4, so I specifed projection flags in wannier90.win input file like this:


"
Begin Projections
Cr:s;d
End Projections
"


and I also set the "num_wann = 6", which I thought it meant one 4s orbital and five 3d orbitals, the detailed input contents in wannier90.win can be found below. 
Finally, I did get these six wannier functions for s and d orbitals, however, the wannier function center location for the s orbital did not stay on the Cr atom but seemed to stay on the F atom when I got their plots(xsf files) using "wannier_plot=true" command,it seemed like the wannier function belonged to the s orbital of F atom, I wondered whether this was reasonable?
Then when I got their dos caculation results using "dos=true" command in wannier90.win file, I found the energy location of the peak for s orbital in the dos from the six wannier functions was far lower than the peaks of these 3d orbitals, which seemed to mean the 4s orbital energy was far lower than 3d orbital?(I thought this was unreasonble), so I wondered whether I really got the correct wannier function for 4s orbital and I felt very confused about this question, if I did not got the 4s orbital in fact, What should I do to get the right wannier funtion?


2) The second question I want to ask is how to specify the 4s and 3s orbital projections simultaneously in wannier90.win, if I just set "Cr:s;d" like above, does the "s" mean "4s" or "3s"? I have read the "User Guide" tutorial of Wannier90 but I still feel confused about this question, I will be very grateful if someone can help me to figure out these questions. 






My wannier90.win input file:


num_bands =    24  ! set to NBANDS by VASP
num_wann=6


num_iter=1000
dis_num_iter      =  100000
Begin Projections
Cr:s;d
End Projections


guiding_centres=T


wannier_plot=true




begin unit_cell_cart
    15.9735003     0.0000000     0.0000000
     0.0000000    15.9735003     0.0000000
     0.0000000     0.0000000    15.9735003
end unit_cell_cart


begin atoms_cart
F        6.5098907     7.2668224     6.9851800
F        6.4041086     9.8161871     8.2105678
F        7.4299150     9.5488493     5.5858959
F        4.6616692     9.2034493     6.0655672
Cr       6.2513620     8.9588588     6.7118351
end atoms_cart


mp_grid =     1     1     1


begin kpoints
      0.000000000000      0.000000000000      0.000000000000
end kpoints






The final output results:
 
Final State
  WF centre and spread    1  (  7.631570, -6.265553,  5.467947 )     0.30701220  
  WF centre and spread    2  (  6.250670, -7.012090,  6.708068 )     0.49090043
  WF centre and spread    3  (  6.271644, -7.024878,  6.755821 )     0.50972991
  WF centre and spread    4  (  6.242683, -7.052756,  6.670884 )     0.51011098
  WF centre and spread    5  (  6.240996, -6.959348,  6.713718 )     0.50907534
  WF centre and spread    6  (  6.250609, -7.024307,  6.710767 )     0.49228399
  Sum of centres and spreads ( 38.888172,-41.338931, 39.027205 )     2.81911285




#the first wannier function is for s orbital (we can find its WF center coordinates is more closer to the third F atom coordinate, not the Cr atom)


--

Leyu Liu
Department of Chemistry
Tsinghua University
E-mail:liu-ly19 at mails.tsinghua.edu.cn
TEL：18911653273     
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20210418/a9d02af3/attachment.html>