[Wannier] Phases of Wannier functions in generated Hamiltonian

[Seonghoon Jang]

Dear Mr. Zwartsenberg

They are all purely real.
If you take a look at "Cubic Harmonic" page wikipedia, you will see
Wannier90's projections (as in manual) are the same with the wiki
definitions, except for py, dyz, and dxz. As to py, dyz, and dxz, you
should multiply them by -1.

That is, py(Wannier90) = -py(wiki)
Other orbitals including all f orbitals remain the same. (Wannier90
definition = wiki definition).

Sincerely,
Jang
==========================
Seong-Hoon Jang (D1)

Motome Lab.,
Department of Applied Physics,
University of Tokyo, Hongo 7-3-1,
Bunkyo-ku, Tokyo 113-8656, JAPAN
TEL +81-70-2193-3176



2018/01/23 5:20 "Yunpeng Wang" <yunpengwang85 at gmail.com>:

If I remember correctly, the Wannier functions using wannier90 are always
real, although in the original wannier90 paper, the realness of wannier
functions are not proved from a math point of view.



Yun-Peng Wang





*From: *Berend Zwartsenberg <bzwartsenberg at phas.ubc.ca>
*Sent: *Monday, January 22, 2018 2:16 PM
*To: *wannier at quantum-espresso.org
*Subject: *[Wannier] Phases of Wannier functions in generated Hamiltonian



Dear Wannier experts,



I am using a combination of Wien2k, Wien2Wannier and Wannier90 to obtain
MLWFS. I use the generated Hamiltonian to obtain an effective model of the
system. This all works very well, and gives me the correct band structure.
The purpose I am using this model for however, is sensitive to the phase of
the orbitals too (by ‘phase’ here I mean a complex factor of unit
magnitude, for example: dxy -> c dxy, with c some complex number satisfying
|c| =1.)



I can see from plotting the orbitals that I get the requested d:t2g
orbitals, and the manual describes the functional form for the d-orbitals
used in the calculations, but it is not clear to me wether these orbitals
are always purely real? If the generated orbitals had some arbitrary
complex phase it would not change the eigenvalues of the Hamiltonian, but
it would change the matrix elements and the eigenvectors. Is there a way to
check in the output what the actual phases of the orbitals are that are
used to calculated the Hamiltonian?





Kind regards,



Berend Zwartsenberg



PhD candidate

University of British Columbia

Vancouver



_______________________________________________

Wannier mailing list

Wannier at quantum-espresso.org

http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier



_______________________________________________
Wannier mailing list
Wannier at quantum-espresso.org
http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20180123/1673c961/attachment.html>