[QE-users] A question on the volume_factor flag in d3_tk.x

[Lorenzo Paulatto]

On 6/15/23 21:12, Elio Physics wrote:
> Dear professor Paulatto,
>
> Thanks for the information.
>
> 1)Would that divergence produce negative line widths? I have been 
> trying to obtain some decent results for graphene but when the 
> smearing is above 1.0 , the code crashes stating that the linewidth is 
> negative! i tried to increase the grid dimension but in vain.


To get a negative linewidth it takes a seriously broken calculation, as 
the last therm can mathematically be negative


But that would require that the population of the higher phonon (q'') is 
higher than the lower phonon (q') as w'' = w+w' > w'.


Also a broken compilation of the code, or a bug, but to say anything I 
would need a specific case to test and reproduce.


>
> 2) Is there a flag in the input that ensures regularization or one 
> that imposes a cutoff?


No, it has been done ad-hoc for the specific paper about 2D systems, 
although I did not do those calculations, I'm not sure of the details. 
It is very easily done in post-processing, if store_lw=.true. is used to 
save the linewidths to file, using the matlab script recompute_sma.m, or 
any home-made script. A bit more complicated in the exact minimization 
method, setting a reasonable Casimir scattering length may be the best 
bet. Exact algorithm is very expensive to converge anyway, because one 
cannot use shifted grid (and symmetry is not exploited at the moment)



>
>
> Thank you
> ------------------------------------------------------------------------
> *From:* users <users-bounces at lists.quantum-espresso.org> on behalf of 
> Lorenzo Paulatto <lorenzo.paulatto at cnrs.fr>
> *Sent:* Thursday, June 15, 2023 5:02 PM
> *To:* Quantum ESPRESSO users Forum <users at lists.quantum-espresso.org>
> *Subject:* Re: [QE-users] A question on the volume_factor flag in d3_tk.x
> Hello Elio,
> The problem is that tk is given for bulk per unit of volume. When you 
> do a 2D material in plane waves, you need to put a lot of vacuum. The 
> volume factor is used to make the value independent on the amount of 
> vacuum. It can  is the ratio of the bulk uni cell to the 2d unit cell, 
> or the ratio between the c axis in bulk and in the film, it is the 
> same thing.
>
> Please not that 2d materials are very tricky: the thermal conductivity 
> from the quadratic band is integrable but diverging. Using some kind 
> of regularization or a cutoff may be required.
>
> Cheets
>
>
>
> On June 15, 2023 8:48:16 p.m. GMT+02:00, Elio Physics 
> <Elio-Physics at live.com> wrote:
>
>     Dear all (and specifically Professor Paulatto),
>
>     Sorry to bother you again with another question regarding the d3q
>     code. To calculate the lattice thermal conductivity for 2D
>     materials, we need to define a volume_factor =H/V. H is defined as
>     the interlayer distance for the corresponding bulk material. If I
>     understood correctly, let us say we have a 1H MoS2, we know that
>     the bulk can be modeled as two layers (a total of six atoms per
>     unit cell). H would then be the distance between the upper S layer
>     of the bottom MoS2 and the lower S layer of the MoS2 on top. Correct?
>
>
>     Now what if we have a 1T 2D structure. Up to my knowledge, the
>     bulk is modeled as one layer ( 3 atoms per unit cell).  In this
>     case, what would H be ? The thickness of the structure?
>
>
>
>     Regards
>
>     Elie Moujaes
>     Federal University of Rondonia
>     Brazil
>     Porto Velho
>
>
> _______________________________________________
> The Quantum ESPRESSO community stands by the Ukrainian
> people and expresses its concerns about the devastating
> effects that the Russian military offensive has on their
> country and on the free and peaceful scientific, cultural,
> and economic cooperation amongst peoples
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing listusers at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
-- 
Dr. Lorenzo Paulatto
IdR @ IMPMC - CNRS UMR 7590 & Sorbonne Université
phone: +33 (0)1 442 79822 / skype: paulatz
http://www.impmc.upmc.fr/~paulatto/ - https://anharmonic.github.io/
23-24/423 B115, 4 place Jussieu 75252 Paris CX 05
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20230616/7d0553e5/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rcd1KcF2uVGiYMJh.png
Type: image/png
Size: 8558 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20230616/7d0553e5/attachment.png>