[Wannier] about initial projections in seedname.win

[Duk-Hyun Choe]

Dear Hung Pham,

Thank you very much for your reply.

I haven't checked the MLWFs but I expect them to be different, because the
calculated Wannier centers are very different in those two calculations.

However, no matter how localized the WF is, I think the "sum of the Wannier
centers" should be the same, because the bulk polarization is gauge
invariant.

In addition, can you give me any suggestions for setting up the initial
projection in BaTiO3?

I thought the below would be reasonable.

Begin Projections
Ba: s;px;py;pz
O: sp3
End Projections

Thank you in advance.


*|** Duk-Hyun Choe*, Post-doctoral fellow,
  Department of Physics, Applied Physics, and Astronomy
  Rensselaer Polytechnic Institute, Troy, NY, USA
  E-mail: hoalad at gmail.com,  Tel: +1-518-687-4389
  Home: https://sites.google.com/site/dhchoeprofile/


2019년 3월 25일 (월) 오후 3:00, Hung Pham <phamx494 at umn.edu>님이 작성:

> Did you check if two sets of projection functions lead to the same set of
> MLWFs? (do they look the same? with the same spread?)
> Is it possible that two inputs converged to two different local minima?
>
>
>
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> Virus-free.
> www.avast.com
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link>
> <#m_-5397300850589285672_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
> On Mon, Mar 25, 2019 at 1:31 PM Duk-Hyun Choe <hoalad at gmail.com> wrote:
>
>> Dear Wannier developers and users,
>>
>>  I am new to the Wannier90 code and trying to calculate the polarization
>> of bulk BaTiO3.
>>
>>  I think the bulk polarization should not depend on the choice of initial
>> projections once the convergence has reached but, strangely enough, the
>> results seem to depend on the initial projections schemes.
>>
>>  For example, I tested two initial projections for BaTiO3.
>>
>> Case #1: atomic orbital projections
>> dis_num_iter = 80000
>> dis_mix_ratio = 0.9d0
>> dis_conv_tol = 1.0d-9
>>
>> conv_window  = 100
>> conv_tol = 1.0d-9
>>
>> search_shells = 50
>>
>> *Begin Projections*
>> *Ba: s;px;py;pz*
>> *O: sp3*
>> *End Projections*
>>
>> guiding_centres = true
>> #num_no_guide_iter = 20
>>
>> num_iter = 80000
>> num_print_cycles = 100
>>
>> use_bloch_phases = F
>>
>> write_r2mn = true
>> write_xyz = true
>>
>> num_bands = 32
>> num_wann = 16
>> ===========================================
>>
>> Case #2: random projections
>> dis_num_iter = 80000
>> dis_mix_ratio = 0.9d0
>> dis_conv_tol = 1.0d-9
>>
>> conv_window  = 100
>> conv_tol = 1.0d-9
>>
>> search_shells = 50
>>
>> *Begin Projections*
>> *Random*
>> *End Projections*
>>
>> guiding_centres = true
>> #num_no_guide_iter = 20
>>
>> num_iter = 80000
>> num_print_cycles = 100
>>
>> use_bloch_phases = F
>>
>> write_r2mn = true
>> write_xyz = true
>>
>> num_bands = 32
>> num_wann = 16
>>
>> ===================================================
>>
>> The only difference in the two input files is the Projections block.
>>
>> Although both cases are converged, the sum of wannier centers for the
>> occupied bands are different for case 1 and case 2.
>>
>> Can anyone help me with this?
>>
>> Any comment and/or suggenstions would be very much appreciated.
>>
>> Thank you.
>>
>> Best,
>>
>> *|** Duk-Hyun Choe*, Post-doctoral fellow,
>>   Department of Physics, Applied Physics, and Astronomy
>>   Rensselaer Polytechnic Institute, Troy, NY, USA
>>   E-mail: hoalad at gmail.com,  Tel: +1-518-687-4389
>>   Home: https://sites.google.com/site/dhchoeprofile/
>> _______________________________________________
>> Wannier mailing list
>> Wannier at lists.quantum-espresso.org
>> https://lists.quantum-espresso.org/mailman/listinfo/wannier
>>
>
>
> --
>
> Hung Q. Pham
> Gagliardi Group
> Office: Smith 101
> Email: phamx494 at umn.edu
>
> Department of Chemistry
> University of Minnesota - Twin Cities, Minneapolis, MN 55455
>
>
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> Virus-free.
> www.avast.com
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link>
> <#m_-5397300850589285672_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20190325/183f7429/attachment.html>