[QE-users] Convergence not achieved

Mon Mar 18 11:39:33 CET 2024

Dear Saiyed,
it is better if next time you also include output, just to understand what
is not converging.

In this case:
i) for convergence issues of the electronic self-consistent cycle, you
should always try to reduce mixing_beta to 0.3, 0.1, 0.05 ...

ii) before running ANY calculations, mostly if you're dealing with a
material that is novel for you, it is quite a good practice to visualize
your structure.
I'm not expert about this material, however what I can notice is that, if
you're running the same calculation as that you described in your first
email,
that is, 3 layers of calcite, I cannot see three layers but just one and
the one I see I suspect does not resemble that much to the models in the
paper you mention.
You could figure out that you're likely not simulating three layers, by
looking at your coordinates: the z coordinates range from 0.46234
to 0.53766, so they span an
interval 0.07532 wide. Because these are crystal units, this in Angstrom is
0.07532 *  43.2578.... Ang ~ 3.258 Ang. It is quite unlikely that three
layers of any material
(even graphene. the flattest you can think about) are just 3 Ang thick.

Giovanni
-- 

Giovanni Cantele, PhD
CNR-SPIN
c/o Dipartimento di Fisica
Universita' di Napoli "Federico II"
Complesso Universitario M. S. Angelo - Ed. 6
Via Cintia, I-80126, Napoli, Italy
e-mail: giovanni.cantele at spin.cnr.it <giovanni.cantele at spin.cnr.it>
Phone: +39 081 676910
Skype contact: giocan74

ResearcherID: http://www.researcherid.com/rid/A-1951-2009
Web page: https://sites.google.com/view/giovanni-cantele/home

Il giorno sab 16 mar 2024 alle ore 22:45 Saiyed Tasnim Md Fahim <
sfahim at usc.edu> ha scritto:

> Dear Dr. Cantele,
>
> Thank you for your response.
> I tried the steps that you mentioned. However, it still didn't converge.
> I am following this paper as the benchmark :
> https://doi.org/10.1016/j.cej.2024.148940
> I have used the norm conserving pseudopotentials from Pseudodojo website
> as USPP/PAW wasn't implemented for TS dispersion. I used the ASE (atomic
> simulation environment) to generate the atomic positions of the 2x2
> supercell.
>
> I have attached the new input file with this mail. Could you please tell
> me if any parameters I have missed or need to be changed or removed? Thank
> you.
>
>
> Sincerely,
> Saiyed Tasnim Md Fahim,
> Ph.D. Student,
> Environmental Engineering,
> University of Southern California.
>
>
> On Wed, Feb 28, 2024 at 12:27 AM Giovanni Cantele <
> giovanni.cantele at spin.cnr.it> wrote:
>
>> Dear Saiyed,
>>
>> when convergence issues show up there are several tricks you might
>> consider, one of these is certainly reducing mixing_beta,
>> another could be increasing degauss.
>>
>> However, before trying any trick to cure a calculation particularly
>> difficult to converge, as a rule of thumb the first step is to
>> check your input file parameters including cut-off, k-point grid,
>> degauss, GEOMETRY, pseudopotentials, etc.
>>
>> After a quick look to your input file I can notice:
>> - an incredibly huge value of degauss (1d-1 Ry = 0.1 Ry = 1.36 eV, this
>> would smear even the band gap of silicon I think)
>> - a mixing beta of 0.6, actually the advise of reducing it in cases of
>> difficult convergence is meant to go down to even 0.1 or 0.05
>> - but MOST IMPORTANTLY you specify ATOMIC_POSITIONS in crystal units,
>> that are usually expected to be numbers in the range
>> [-1,1] because in units of the lattice vectors there should be no reason
>> to specify the position of an atom in a unit cell far apart from the
>> origin. On the other hand, the numbers in your input file seem to
>> indicate that you're actually giving the positions in Angstrom, because
>> they are
>> huge. If you try to visualize your input file using, for example
>> XCrysDen, you can easily see atoms at incredibly huge distances. The
>> geometry
>> becomes much more reasonables if you replace "crystal" with angstrom.
>> After doing that, recheck the geometry (with XCrysDen) to understand if it
>> is
>> what you expect or if there are still issues, before running any
>> calculation.
>>
>> Giovanni
>>
>>
>> --
>>
>> Giovanni Cantele, PhD
>> CNR-SPIN
>> c/o Dipartimento di Fisica
>> Universita' di Napoli "Federico II"
>> Complesso Universitario M. S. Angelo - Ed. 6
>> Via Cintia, I-80126, Napoli, Italy
>> e-mail: giovanni.cantele at spin.cnr.it <giovanni.cantele at spin.cnr.it>
>> Phone: +39 081 676910
>> Skype contact: giocan74
>>
>> ResearcherID: http://www.researcherid.com/rid/A-1951-2009
>> <https://urldefense.com/v3/__http://www.researcherid.com/rid/A-1951-2009__;!!LIr3w8kk_Xxm!uHeu46wuGTYOEVDT0_V9s92ksXMy-0mbaezxExpDSYHQ7RiP0fPUrWo0ZpZ_1BDUcfbWaoy76a1xhBwsJ_ARu6Qgfu-r$>
>> Web page: https://sites.google.com/view/giovanni-cantele/home
>> <https://urldefense.com/v3/__https://sites.google.com/view/giovanni-cantele/home__;!!LIr3w8kk_Xxm!uHeu46wuGTYOEVDT0_V9s92ksXMy-0mbaezxExpDSYHQ7RiP0fPUrWo0ZpZ_1BDUcfbWaoy76a1xhBwsJ_ARuypOJcP_$>
>>
>>
>> Il giorno mer 28 feb 2024 alle ore 08:41 Saiyed Tasnim Md Fahim <
>> sfahim at usc.edu> ha scritto:
>>
>>> Dear users,
>>> I am working with calcite in Quantum Espresso 7.1. I want to perform a
>>> vc-relax calculation for 3 layers of calcite but after 100 iterations
>>> convergence is not achieved. I changed the mixing beta but it did not work.
>>>
>>> I have attached the input file with this mail. Kindly let me know how I
>>> should proceed for both the convergence.
>>>
>>> Sincerely,
>>> Saiyed Tasnim Md Fahim,
>>> Ph.D. Student,
>>> Environmental Engineering,
>>> University of Southern California.
>>> _______________________________________________
>>> 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
>>> <https://urldefense.com/v3/__http://www.max-centre.eu__;!!LIr3w8kk_Xxm!uHeu46wuGTYOEVDT0_V9s92ksXMy-0mbaezxExpDSYHQ7RiP0fPUrWo0ZpZ_1BDUcfbWaoy76a1xhBwsJ_ARu28Aj1dA$>
>>> )
>>> users mailing list users at lists.quantum-espresso.org
>>> https://lists.quantum-espresso.org/mailman/listinfo/users
>>> <https://urldefense.com/v3/__https://lists.quantum-espresso.org/mailman/listinfo/users__;!!LIr3w8kk_Xxm!uHeu46wuGTYOEVDT0_V9s92ksXMy-0mbaezxExpDSYHQ7RiP0fPUrWo0ZpZ_1BDUcfbWaoy76a1xhBwsJ_ARuyERRL04$>
>>
>> _______________________________________________
>> 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 (
>> https://urldefense.com/v3/__http://www.max-centre.eu__;!!LIr3w8kk_Xxm!uHeu46wuGTYOEVDT0_V9s92ksXMy-0mbaezxExpDSYHQ7RiP0fPUrWo0ZpZ_1BDUcfbWaoy76a1xhBwsJ_ARu28Aj1dA$
>> )
>> users mailing list users at lists.quantum-espresso.org
>>
>> https://urldefense.com/v3/__https://lists.quantum-espresso.org/mailman/listinfo/users__;!!LIr3w8kk_Xxm!uHeu46wuGTYOEVDT0_V9s92ksXMy-0mbaezxExpDSYHQ7RiP0fPUrWo0ZpZ_1BDUcfbWaoy76a1xhBwsJ_ARuyERRL04$
>
> _______________________________________________
> 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 list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20240318/ae1c0d5c/attachment.html>