[Wannier] Fwd: Extracting tight binding Parameters

Christopher Pashartis cpashartis at gmail.com
Tue Jun 28 16:36:52 CEST 2016


Hi Jesse,

On Tue, Jun 28, 2016 at 3:15 AM, Jesse Vaitkus <jesse.vaitkus at rmit.edu.au>
wrote:

> Hello Christopher,
>
> a few things. Firstly, did you check that before truncating your
> interactions whether the band structure was reconstructed? This is a very
> important first step, if these are not recreated then you have bigger
> problems.
>

*The band structure before truncating the interactions was very close to
the DFT calculation, I'm simply using nearest neighbour interactions right
now so I can attempt to reproduce (as a first step) the band structure
generated by the wannier90 hamiltonian. I want to be sure I extract the
parameters (which judging by your response there seems to be no issue?)
right as well as my code works.*

Secondly, you have not reported the average spread of the WFs which is the
> metric used to work out how localised they are, if it exceeds the nearest
> neighbour distance then you likely need next nearest neighbour.
>

*Understanding what you are saying: to make my band structure as accurate
as possible what you are saying is that I should at least include the
spread of the WFs, that makes sense. I'll keep that in mind when
reproducing my full result. I would tell you what they are, but my
computing cluster is currently undergoing maintenance.*

Thirdly, you have not shown us which bands you wish to disentangle, if
> they're entangled you might need to finesse your frozen window.
>


> Fourthly, 2,8 and 3,8 don't necessarily need to be the same; 2,8 and 8,2
> and 3,8 and 8,3 need to be hermitian conjugates of each other and no, w90
> does not impose symmetry, something that I would so absolutely love.
>

*With regards to the 2,8 and 3,8 , how do I then know if the tight binding
factor is negative or positive?*


> Lastly, unfortunately the way in which w90 orders your wannier functions
> need not be how you want them ordered so what you call anion and cation
> might not be so (just check their output positions and then you'll know,
> assuming that you haven't already done this and I'm out of line).
>

*I did check this yes, but I will now remember it for sure :)*


> An 'easy way' to enforce your own personal ordering is to start a new run
> using the positions of the converged WFs you should be able to tell which
> atoms these correspond to and then use a strong guiding centre constraint.
>
> I hope my list has helped you. Best of luck.
>
> Cheers,
> Jesse Vaitkus
>

*Thank you for the kind response, and thank you additionally for covering
many of these topics for a new user.*

*Christopher Pashartis*

>
> On 28 June 2016 at 03:13, Christopher Pashartis <cpashartis at gmail.com>
> wrote:
>
>> Hello,
>>
>> I'm new to using wannier90 and I am trying to extract tight binding
>> parameters for use in my code. Unfortunately, I cannot seem to reproduce
>> the correct tight binding result that I get from my DFT code (not QE). If
>> someone can verify, point me to a proper example, or correct me it would be
>> greatly appreciated.
>>
>> I've reduced my Hamiltonian to nearest neighbour interactions only,
>> therefore, since I am using a zinc blende structure these are the
>> corresponding matrix elements I am interested in:
>>
>>     0    0    0    1    1   -4.324417   -0.000000
>>     0    0    0    2    1    0.000002   -0.000000
>>     0    0    0    3    1   -0.000000    0.000000
>>     0    0    0    4    1   -0.000001    0.000000
>>     0    0    0    5    1   -1.472555    0.000000
>>     0    0    0    6    1   -1.157538    0.000000
>>     0    0    0    7    1    1.157536   -0.000000
>>     0    0    0    8    1    1.157537   -0.000000
>>     0    0    0    1    2    0.000002    0.000000
>>     0    0    0    2    2    2.843276   -0.000000
>>     0    0    0    3    2    0.000000   -0.000000
>>     0    0    0    4    2    0.000000    0.000000
>>     0    0    0    5    2    1.067338    0.000000
>>     0    0    0    6    2    0.180666    0.000000
>>     0    0    0    7    2   -1.181808   -0.000000
>>     0    0    0    8    2   -1.181808   -0.000000
>>     0    0    0    1    3   -0.000000   -0.000000
>>     0    0    0    2    3    0.000000    0.000000
>>     0    0    0    3    3    2.843276   -0.000000
>>     0    0    0    4    3    0.000000   -0.000000
>>     0    0    0    5    3   -1.067337   -0.000000
>>     0    0    0    6    3   -1.181809   -0.000000
>>     0    0    0    7    3    0.180665    0.000000
>>     0    0    0    8    3    1.181808    0.000000
>>     0    0    0    1    4   -0.000001   -0.000000
>>     0    0    0    2    4    0.000000   -0.000000
>>     0    0    0    3    4    0.000000    0.000000
>>     0    0    0    4    4    2.843276   -0.000000
>>     0    0    0    5    4   -1.067337   -0.000000
>>     0    0    0    6    4   -1.181809   -0.000000
>>     0    0    0    7    4    1.181808    0.000000
>>     0    0    0    8    4    0.180665    0.000000
>>     0    0    0    1    5   -1.472555   -0.000000
>>     0    0    0    2    5    1.067338   -0.000000
>>     0    0    0    3    5   -1.067337    0.000000
>>     0    0    0    4    5   -1.067337    0.000000
>>     0    0    0    5    5   -8.772213    0.000000
>>     0    0    0    6    5   -0.000003    0.000000
>>     0    0    0    7    5   -0.000003   -0.000000
>>     0    0    0    8    5   -0.000003   -0.000000
>>     0    0    0    1    6   -1.157538   -0.000000
>>     0    0    0    2    6    0.180666   -0.000000
>>     0    0    0    3    6   -1.181809    0.000000
>>     0    0    0    4    6   -1.181809    0.000000
>>     0    0    0    5    6   -0.000003   -0.000000
>>     0    0    0    6    6    0.872588    0.000000
>>     0    0    0    7    6    0.000000    0.000000
>>     0    0    0    8    6    0.000000   -0.000000
>>     0    0    0    1    7    1.157536    0.000000
>>     0    0    0    2    7   -1.181808    0.000000
>>     0    0    0    3    7    0.180665   -0.000000
>>     0    0    0    4    7    1.181808   -0.000000
>>     0    0    0    5    7   -0.000003    0.000000
>>     0    0    0    6    7    0.000000   -0.000000
>>     0    0    0    7    7    0.872588    0.000000
>>     0    0    0    8    7   -0.000000    0.000000
>>     0    0    0    1    8    1.157537    0.000000
>>     0    0    0    2    8   -1.181808    0.000000
>>     0    0    0    3    8    1.181808   -0.000000
>>     0    0    0    4    8    0.180665   -0.000000
>>     0    0    0    5    8   -0.000003    0.000000
>>     0    0    0    6    8    0.000000    0.000000
>>     0    0    0    7    8   -0.000000   -0.000000
>>     0    0    0    8    8    0.872588    0.000000
>>
>> From my interpretation, the corresponding LCAO parameters would be:
>> E_sa E_sc E_pa E_pc ssSigma sapcSigma scpaSigma ppSigma ppPi
>> -8.77210 -4.32453 0.87263 2.84323 -1.47257 1.06736 1.15752 0.18067 1.1818
>>
>> Where c corresponds to the cation (in this case it should be the 1-4
>> functions) and a is the anion (5-8 functions).
>>
>> With regards to my parameters above, does Wannier90 already build in
>> symmetry operations? For example at (2,8) vs (3,8) these should be the same
>> value (+ or - not both) unless I'm misinterpreting what this file truly
>> presents, so which should it be?
>>
>> Any help would be greatly appreciated!
>>
>> Christopher Pashartis
>> Masters Candidate
>> McMaster University
>>
>>
>>
>>
>>
>> _______________________________________________
>> Wannier mailing list
>> Wannier at quantum-espresso.org
>> http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier
>>
>>
>
>
> --
>
> ————————————————————————————————
> Mr. Jesse Vaitkus MAIP
> BAppSc(Phys) Hons, BAppSc(Nano)
>
> PhD candidate
> Room 14.6.02
>
> Chemical and Quantum Physics
> School of Applied Sciences
> RMIT University
> Victoria 3001
> Australia
>
> jesse.vaitkus at rmit.edu.au
> ————————————————————————————————
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20160628/f528b520/attachment.html>


More information about the Wannier mailing list