ARPES Plot problem

[abduljalils]

Dear Users: The output of bandstructure is not consistent to ARPES .wt.in file is given below,both files are generated from wannier Tool. &TB_FILE Hrfile = 'wannier90_hr.dat' Package = 'VASP' ! obtained from VASP, it could be 'VASP', 'QE', 'Wien2k', 'OpenMx' /

LATTICE Angstrom 15.184 0.000000 0.000000 ! crystal lattice information -7.5921 13.14997 0.000000 0.0000 0.000000 14.835313

ATOM_POSITIONS 36 ! number of atoms for projectors Direct ! Direct or Cartisen coordinate x 0.9073873983827454 0.4783230057052990 0.4582741926581662 x 0.0926126016172617 0.5216769942947010 0.5417258363418327 x 0.4783230057052990 0.5709356063225499 0.5417258363418327 x0.5709356063225499 0.0926126016172617 0.4582741926581662 x 0.4290643936774501 0.9073873983827454 0.5417258363418327 x 0.5216769942947010 0.0926126016172617 0.5417258363418327 x 0.4783230057052990 0.9073873983827454 0.4582741926581662 x 0.5709356063225499 0.4783230057052990 0.5417258363418327 x 0.4290643936774501 0.5216769942947010 0.4582741926581662 x 0.9073873983827454 0.4290643936774501 0.5417258363418327 x 0.0000000000000000 0.4512739261667633 0.5811722416809815 x 0.4512739261667633 0.4512739261667633 0.4188277583190114 x 0.5487260738332367 0.5487260738332367 0.5811722416809815 x 0.5487260738332367 0.0000000000000000 0.4188277583190114 x 0.4512739261667633 0.0000000000000000 0.5811722416809815 x 0.7947660712127202 0.3439924943485835 0.5925519645880470 x0.2052339287872726 0.6560075056514165 0.4074480354119530 x 0.6560075056514165 0.4507735758641473 0.5925519645880470 x 0.3439924943485835 0.5492264241358527 0.4074480354119530 x 0.5492264241358527 0.2052339287872798 0.5925519645880470 x0.4507735758641473 0.7947660712127202 0.4074480354119530 x 0.6560075056514165 0.2052339287872798 0.4074480354119530 x 0.3439924943485835 0.7947660712127202 0.5925519645880470 x 0.5492264241358527 0.3439924943485835 0.4074480354119530 x 0.4507735758641473 0.6560075056514165 0.5925519645880470 x 0.7947660712127202 0.4507735758641473 0.4074480354119530 x 0.2052339287872798 0.5492264241358527 0.5925519645880470 x 0.0000000000000000 0.5859110105721967 0.3559146910004571 x 0.0000000000000000 0.4140889894277962 0.6440853089995429 x 0.4140889894277962 0.4140889894277962 0.3559146910004571 x0.4140889894277962 0.4140889894277962 0.3559146910004571 x 0.5859110105721967 0.5859110105721967 0.6440853089995429 x 0.5859110105721967 0.0000000000000000 0.3559146910004571 x 0.4140889894277962 0.0000000000000000 0.6440853089995429 PROJECTORS x px py pz ! projectors x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz x px py pz

MILLER_INDICES ! now you can specify surface by Miller indices 1 0 0

!SURFACE ! Specify surface with three vectors, see doc 1 0 0 0 1 0 0 0 1

!> bulk band structure calculation flag &CONTROL BulkBand_calc = F DOS_calc = F BulkFS_calc = F BulkGap_cube_calc = F BulkGap_plane_calc = F SlabBand_calc = F WireBand_calc = F SlabSS_calc = T SlabArc_calc = F SlabQPI_calc = F Z2_3D_calc = F SlabSpintexture_calc = F Wanniercenter_calc = F FindNodes_calc = F /

&SYSTEM NumOccupied = 36 ! NumOccupied SOC = 1 ! soc E_FERMI =-4.1328 ! e-fermi /

&PARAMETERS OmegaNum = 251 ! omega number OmegaMin = -1 ! energy interval OmegaMax = 1 ! energy interval Nk1 = 101 ! number k points odd number would be better NP = 2 ! number of principle layers /

KPATH_SLAB 2 ! numker of k line for 2D case M 0.0 0.5 K -0.33 0.67 ! k path for 2D case K -0.33 0.67 G 0.0 0.0

wannier90.win file: num_wann = 216 ! set to NBANDS by VASP

use_bloch_phases = T hr_plot=.TRUE.. spinors = .true.

dis_num_iter=1000 num_iter=0 iprint=2

!min of outer window dis_win_min = -2.0 dis_win_max = 18.0

!inner - dis_froz_min = -2.0000 dis_froz_max = 5.5000 !begin projections x : px; py; pz x : px; py; pz x : px; py; pz Mn: px; py; pz end projections spinors = .true.

[abduljalils]

Secondly ,its necessary to do Hybrid functional calculations before winnerization?can we perform GGA-PBE calculations before winnerization?will it effect the output?

[quanshengwu]

Dear @abduljalils ,

One important point is that the surface states calculated from WannierTools in the framework of a tight-binding model could probably totally not be comparable with the ARPES measurement. The surface states from WannierTools are only used for analyzing the topological properties if the TB model is not well set. The reasons are:

1. If there are dangling bonds on the surface, there will be all kinds of possibilities of the surface states due to the different TB models constructed from MLWFs. The MLWFTB models constructed from Wannier90 could be different even if the band structures are the same. If the TB parameters on the surfaces are different, then the surface states could be different too. However, if there is no dangling bond on the surface, such as the (0001) surface of the layered compounds Bi2Se3, then the TB parameters on the surface is well comparable with the real situation (I mean the clean sample in the experiment). The surface states calculated from the WannierTools is trustable in such cases. Another example is the WTe2, MoTe2, which are also layered structures.
2. Even if there is no dangling bond on the surface, the SSs calculated from WannierTools could also be different from ARPES measurement. The key is the surface potential. We don't have charge relaxation in the tight-binding model. The TB parameters could be very different after and before the charge relaxation. TiB2 is such an example. The surface states of B terminated surface can be well reproduced from WannierTools, while the one of Ti-terminated surface can't.

I am sorry that I will skip the question about Wannier functions itself if it is not related to the problem of WannierTools. I suggest you put the questions related to Wannier90 to the wannier90's forum.

Best, QuanSheng

[quanshengwu]

By the way, I would not suggest you using the random projectors in wannier90.win for the topological properties analysis. Please set the projectors explicitly in the wannier90.win. The num_wann should be less then num_bands in wannier90.win if the band sets are entangled with other bands.

[abduljalils]

Many Thanks for the comprehensive and valuable reply.The num_wann are mentioned in wannier.win output file whats about num_bands and Numoccupied values?

[quanshengwu]

@abduljalils Please read the documentation of Wannier90 to get the definition of num_wann and num_bands. There are many examples in the Wannier90's package.