[Wannier] Gamma-point setting

[Stefan Badescu]

Dear Young-Su,

I will send you the files. Meanwhile I am checking something that I
might have done wrongly. Dr Giovanni Cantele and Professor Marzari drew
my attention to using K_POINTS {gamma} in the nscf calculation in
Quantum Espresso. This should provide real wavefunctions without the
problem of multiple branches of complex logarithm. I was using K_POINTS
{crystal} \ 1 \ 0.00 0.00 0.00 2.0 which may give other branches. I was
thinking that no matter how one specifies Gamma-sampling the result
should be the same.

I'll confirm whether this is makes the difference soon.

Thanks,
--Stefan

On 9/13/2010 5:02 PM, Young-Su Lee wrote:
>
> Dear Stefan Badescu,
>
> Could you send me the pwscf and wannier input of a system for which
> gamma-only does not work?
>
> For the cases where the gamma-only wannierisation fail, does the
> gamm-only version give you the same disentanglement result as the
> non-gamma version? I'm just wondering whether the error comes from the
> disentanglement step or it's purely from the wannierisation part.
>
> Thank you.
>
> Young-Su
>
>
>
>
> * ----- Original Message -----*
> From: Stefan Badescu
> To: wannier at quantum-espresso.org
> Date: 2010-09-13 21:46:43
> Subject: [Wannier] Gamma-point setting
> Dear users, I thought it may be of some interest to mention this
> Gamma-point calculations. Maybe this was signaled before on this
> forum. I found that using 'gamma_only = .true.' does not always
> converge for large systems. It may go past the disentaglement but in
> the wannierisation step it gives pretty large spreads and may not
> converge. If I use the default 'gamma_only = .false.' I avoid this
> problem most of the time, even if it somewhat slower. It also gives
> smaller spreads in the first step - disentanglement. The examples I
> used for this are graphene substrates with up to 288 C atoms, on which
> I have somewhere between 6 to 36 chemisorbed atoms. --Stefan
> _______________________________________________ Wannier mailing list
> Wannier at quantum-espresso.org
> http://www.democritos.it/mailman/listinfo/wannier
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20100913/d6d193a5/attachment.html>