[QE-users] Incosistent energy contributions for 2D system
Lenz Fiedler
l.fiedler at hzdr.de
Fri Jul 26 09:48:02 CEST 2024
Dear Professor Giannozzi,
Thanks for the clarification - that makes a lot of sense! In that case I
can carry on with my calculations, using the total energy as a metric
for cutting my cells.
Kind regards,
Lenz
--
Lenz Fiedler, M. Sc.
PhD Candidate | Machine Learning for Materials Design
Tel.: +49 3581 37523 55
E-Mail: l.fiedler at hzdr.de
https://www.casus.science
CASUS - Center for Advanced Systems Understanding
Helmholtz-Zentrum Dresden-Rossendorf e.V. (HZDR)
Conrad-Schiedt-Straße 20
02826 Görlitz
Vorstand: Prof. Dr. Sebastian M. Schmidt, Dr. Diana Stiller
Vereinsregister: VR 1693 beim Amtsgericht Dresden
On 7/25/24 19:11, Paolo Giannozzi wrote:
> The "Hartree", "one-electron", "Ewald" contributions to the energy
> have something in common: the G=0 contribution to the energy
> separately diverges. Their sum however does not diverge for neutral
> cells. So one subtracts out the diverging terms and is left with a
> finite G=0 contribution. This may however vary a lot between different
> structures, even if the sum of the three contributions is quite the
> same as in this case.
>
> Paolo
>
> On 25/07/2024 18:43, Lenz Fiedler wrote:
>> [Non ricevi spesso messaggi di posta elettronica da
>> l.fiedler at hzdr.de. Per informazioni sull'importanza di questo fatto,
>> visita https://aka.ms/LearnAboutSenderIdentification.]
>>
>> Dear QE users,
>>
>> I have a question regarding a 2D QE calculation. Consider a very simple
>> bilayer graphene system, which I calculate either with a "big" (23
>> Angstrom in z-direction) or "small" (17 Angstrom in z-direction) cell
>> (Input files are at the end of the message). The C atoms are in two
>> planes in x-y-direction.
>>
>> Now both calculations are the same except for their z-dimension. In the
>> "small" case I make the cell a bit smaller in z-direction. I did careful
>> calculations to determine by how much I can make it smaller without the
>> total energy deviating too much from the original cell. The input I
>> share here is still well within the range where the total energy error
>> is very small. Of course, eventually one hits a point where there is a
>> noticeable energy difference.
>>
>> Now to my question: I have noticed that for both calculations, total
>> energies are essentially the same, but what differs quite drastically,
>> are the individual energy contributions. Namely I get:
>>
>> "big":
>> ! total energy = -48.18986746 Ry
>> estimated scf accuracy < 0.00003985 Ry
>> smearing contrib. (-TS) = -0.00000000 Ry
>> internal energy E=F+TS = -48.18986746 Ry
>>
>> The total energy is F=E-TS. E is the sum of the following terms:
>> one-electron contribution = -1245.11933758 Ry
>> hartree contribution = 624.03689026 Ry
>> xc contribution = -17.17420972 Ry
>> ewald contribution = 590.06678957 Ry
>>
>> "small":
>> ! total energy = -48.18985199 Ry
>> estimated scf accuracy < 0.00007424 Ry
>> smearing contrib. (-TS) = -0.00000000 Ry
>> internal energy E=F+TS = -48.18985199 Ry
>>
>> The total energy is F=E-TS. E is the sum of the following terms:
>> one-electron contribution = -759.11502918 Ry
>> hartree contribution = 381.02359494 Ry
>> xc contribution = -17.17243795 Ry
>> ewald contribution = 347.07402021 Ry
>>
>>
>> As you can see, the total energies match up nicely, but all other terms
>> except for the XC energy are different. Is this by design? Is there a
>> numerical reason for it? I am trying to come up with a way to determine
>> the minimal vacuum in z-direction I have to leave below and above my
>> graphene planes, and had originally only checked the total energy. Now I
>> am wondering if this is enough?
>>
>> Kind regards
>> Lenz Fiedler
>> PhD student (Helmholtz-Zentrum Dresden-Rossendorf)
>>
>> PS: I also disabled "assume_isolated = '2D'" - I still see the same
>> behavior, albeit with different numbers for the energy contributions,
>> the total energies are still the same.
>>
>> Input files:
>>
>> (PSP taken from http://www.pseudo-dojo.org/index.html):
>>
>>
>>
>> "big":
>> &CONTROL
>> calculation = 'scf'
>> verbosity = 'high'
>> restart_mode = 'from_scratch'
>> outdir = 'temp'
>> prefix = 'C'
>> pseudo_dir = './'
>> /
>> &SYSTEM
>> ibrav = 0
>> nbnd = 10
>> ecutwfc = 80
>> ecutrho = 320
>> nosym = .true.
>> noinv = .true.
>> occupations = 'smearing'
>> degauss = 0.0018874
>> smearing = 'fermi-dirac'
>> ntyp = 1
>> nat = 4
>> assume_isolated = '2D'
>> /
>> &ELECTRONS
>> conv_thr = 7.599999999999999e-05
>> mixing_mode = 'plain'
>> mixing_beta = 0.1
>> /
>> &IONS
>> /
>> &CELL
>> /
>>
>> ATOMIC_SPECIES
>> C 12.011 C_ncsr_0.5_pbe.upf
>>
>> K_POINTS automatic
>> 8 8 1 0 0 0
>>
>> CELL_PARAMETERS angstrom
>> 2.4622899999999999 0.0000000000000000 0.0000000000000000
>> 1.2310682316596653 2.1324961988004443 0.0000000000000000
>> 0.0000000000000000 0.0000000000000000 23.3000000000000007
>>
>> ATOMIC_POSITIONS angstrom
>> C 2.4622407526646328 0.0000426499239760 9.9535270000000011
>> C 1.2310701657767995 0.7108036329841640 9.9532939999999996
>> C 1.2310805438773487 2.1324748738384565 13.3464730000000014
>> C 2.4620787720005488 1.4216712408542924 13.3467060000000011
>>
>>
>> "small":
>> same as big but with:
>>
>> CELL_PARAMETERS angstrom
>> 2.4622899999999999 0.0000000000000000 0.0000000000000000
>> 1.2310682316596653 2.1324961988004443 0.0000000000000000
>> 0.0000000000000000 0.0000000000000000 17.3040246000000018
>>
>> ATOMIC_POSITIONS angstrom
>> C 2.4622407526646328 0.0000426499239760 6.9555393000000016
>> C 1.2310701657767995 0.7108036329841640 6.9553063000000002
>> C 1.2310805438773487 2.1324748738384565 10.3484853000000019
>> C 2.4620787720005488 1.4216712408542924 10.3487183000000016
>>
>> --
>> Lenz Fiedler, M. Sc.
>> PhD Candidate | Machine Learning for Materials Design
>>
>> Tel.: +49 3581 37523 55
>> E-Mail: l.fiedler at hzdr.de
>> https://www.casus.science
>>
>> CASUS - Center for Advanced Systems Understanding
>> Helmholtz-Zentrum Dresden-Rossendorf e.V. (HZDR)
>> Conrad-Schiedt-Straße 20
>> 02826 Görlitz
>>
>> Vorstand: Prof. Dr. Sebastian M. Schmidt, Dr. Diana Stiller
>> Vereinsregister: VR 1693 beim Amtsgericht Dresden
>>
>>
>> _______________________________________________
>> 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 --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5401 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20240726/133278f4/attachment.p7s>
More information about the users
mailing list