[Wannier] ERROR: unitarity of initial U

Gianluca Giovannetti gianluca.giovannetti at gmail.com
Tue Nov 6 20:44:07 CET 2012


Dear Arash,

the error i mentioned before it is solved using different version of LAPACK
and BLAS on the machines i`m using currenlty.
the post on this mailing list is then correct.
thank you for the advice.

However the convergence in getting the Wannier functions in the run is bad.
I`m wondering what is the problem related to this...probably the initial
guess....

Best Regards,

Gianluca Giovannetti

On Sun, Nov 4, 2012 at 3:07 PM, Arash Mostofi <a.mostofi at imperial.ac.uk>wrote:

> Dear Gianluca,
>
> There was a related post on this mailing list last year:
>
> http://www.democritos.it/pipermail/wannier/2011-October/000461.html
>
> Please could you check whether the error is reproducible with different
> libraries/compilers, and report back?
>
> Best wishes,
>
> Arash
>
> --
> Dr Arash A Mostofi
> Departments of Materials and Physics
> Deputy Director, CDT on Theory and Simulation of Materials
> Imperial College London, London SW7 2AZ, UK
> +44 (0)207 594 8154 | www.cmth.ph.ic.ac.uk/people/a.mostofi
>
> On 02/11/2012 14:32, Gianluca Giovannetti wrote:
> > Dear All,
> >
> > i`m working on extracting the Wannier functions of LiV2O4.
> > The trigonal symmetry splits the low-lying t_{2g} levels into one a_{1g}
> > and two degenerate e_{g}^pi levels, with the former slightly lower in
> > energy and narrower in bandwidth. The density of states looks very
> simple.
> > At Fermi level there are Vanadium d bands having a bandwidth of ~ 2eV.
> > Below in energy separated from the Vanadium d bands there are bands with
> > predominant Oxygen p bands.
> > I can extract with no problem the Vanadium d bands using an energy
> > window sitting at the Vanadium d bands.
> > Wannier90 bands are on top of DFT bands (i also compared mine with some
> > known in literature).
> > As a starting point for the projection i used:
> >
> > begin projections
> > V:dxz;dyz;dxy
> > end projections
> > Once i try to calculate Wannierfunctions for "d" and "p" orbitals i get
> > the following error:
> >
> >     A_mn = <psi_m|g_n> --> S = A.A^+ --> U = S^-1/2.A
> >     In dis_project...  ERROR: unitarity of initial U
> >   nkp=  1
> >   i=  2  j=  4
> >   [u_matrix_opt.transpose(u_matrix_opt)]_ij=     0.114936    -0.050292
> >   Exiting.......
> >   dis_project: Error in unitarity of initial U in dis_project
> >
> > Do you have any idea for solving it?
> > Below it is my input file.
> >
> > thank you in advance.
> >
> > Gianluca Giovannetti
> >
> > =========================
> >
> > iprint = 3
> > num_bands       = 70
> > num_wann        = 36
> >
> > num_iter        =  2000
> > num_print_cycles = 100
> >
> > dis_win_max     = 9.1
> > dis_win_min     = -0.5
> >
> > dis_num_iter    = 200
> > dis_mix_ratio   = 0.75
> >
> > write_r2mn=.true.
> > hr_plot=.true.
> >
> > kmesh_tol=0.00001
> >
> > ! SYSTEM
> >
> > begin unit_cell_cart
> > bohr
> > 7.789167066 0.0000 7.789167066
> > 0.0000 7.789167066 7.789167066
> > 7.789167066 7.789167066 0.0000
> > end unit_cell_cart
> >
> > begin atoms_cart
> > Ang
> > V    0.000000    0.000000    0.000000
> > V    2.060925    2.060925    0.000000
> > V    2.060925    0.000000    2.060925
> > V    0.000000    2.060925    2.060925
> > O     0.094555000   0.094555000  1.966370000
> > O     1.966370000   1.966370000  1.966370000
> > O     0.094555000   1.966370000  0.094555000
> > O     1.966370000   0.094555000  0.094555000
> > O    -0.094556000   4.027295000  2.155480000
> > O     2.155480000  -1.966370000  2.155480000
> > O    -0.094556000  -1.966370000 -0.094556000
> > O     2.155480000   4.027295000 -0.094556000
> > Li     1.030463000   5.152312000  1.030463000
> > Li     3.091388000   3.091388000  3.091388000
> > end atoms_cart
> >
> > begin projections
> > V:dxz;dyz;dxy
> > O:px;py;pz
> > end projections
> >
> > begin kpoint_path
> > G 0.00  0.00  0.00    X 0.00  0.50  0.50
> > X 0.00  0.50  0.50    W 0.25  0.75  0.50
> > W 0.25  0.75  0.50    L 0.00  0.50  0.00
> > L 0.00  0.50  0.00    G 0.00  0.00  0.00
> > end kpoint_path
> > bands_plot = .true.
> >
> > !To plot the WF
> > !wannier_plot = T
> > !wannier_plot_supercell = 2
> > !wannier_plot_list = 1-12
> > !restart = plot
> >
> > ! KPOINTS
> >
> > mp_grid : 8 8 8
> >
> > begin kpoints
> > .............
> >
> >
> > _______________________________________________
> > Wannier mailing list
> > Wannier at quantum-espresso.org
> > http://www.democritos.it/mailman/listinfo/wannier
> >
> _______________________________________________
> Wannier mailing list
> Wannier at quantum-espresso.org
> http://www.democritos.it/mailman/listinfo/wannier
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/wannier/attachments/20121106/ec8808f6/attachment.htm 


More information about the Wannier mailing list