[CP2K-user] [CP2K:18175] GEO_OPT too long
Augustin Bussy
augustin.bussy at chem.uzh.ch
Mon Dec 12 10:02:44 UTC 2022
Dear Ednilsom,
I'am happy to hear that! Generally, the default parameters for GAPW are quite well tuned. In GAPW, different grids are used depending on the region of space. Where the electronic density is smooth, standard regular grids are employed. Where the density varies sharply (i.e. core region around atoms), atomic grids are used. See https://link.springer.com/article/10.1007/s002140050523 for more details on this.
With this in mind, here are some tips:
1. since core orbitals are dealt with on the atomic grids, it is typically not necessary to use high CUTOFF for the regular grid. In your original input, I see you had a cutoff of 1200 Ry. Reducing that number would speed up your calculation. A convergence test is probably still advised, but you could probably go with 400-600 Ry instead.
2. The fineness of the atomic grid is controlled by the LEBEDEV_GRID and RADIAL_GRID in the KIND section defining each atom. In my experience, default values are usually good enough.
3. There are a bunch of keywords controlling GAPW in the DFT%QS section. Understanding what they do exactly is tricky, and I also think that defaults are good.
I hope that helps. Best,
Augustin
________________________________
From: cp2k at googlegroups.com <cp2k at googlegroups.com> on behalf of Ednilsom Orestes <eorestes at gmail.com>
Sent: Monday, December 12, 2022 12:04 AM
To: cp2k at googlegroups.com <cp2k at googlegroups.com>
Subject: Re: [CP2K:18172] GEO_OPT too long
Dear all (and Augustin).
Yes, your indication not only improved but also corrected the optimization procedure.
But another question arose.
Is there any specific parameter(s) which should be set (or optimize) when GAPW is in use?
Thanks in advance
Best,
Ednilsom
Prof. Dr. Ednilsom Orestes
Departamento de Ciências Exatas - Sala C19
Escola de Engenharia Industrial Metalúrgica de Volta Redonda
Universidade Federal Fluminense
R. dos Trabalhadores, 420 - Vila Santa Cecília
Volta Redonda, Rio de Janeiro, Brasil
CEP 27255-125
Tel: +55-24-2107-3777
Em ter., 29 de nov. de 2022 às 11:59, Ednilsom Orestes <eorestes at gmail.com<mailto:eorestes at gmail.com>> escreveu:
Hi Augustin,
In this case, its important to use all-electron BS to CO2 (maybe to some Pt atoms underneath, in the future).
Hence, I'll use GAPW with ALL potential as you recommended and promise be back here to tell how it changes the optimization overall.
Thank you very much by the hints.
Best,
Ednilsom
Em terça-feira, 29 de novembro de 2022 às 11:40:31 UTC-3, Augustin Bussy escreveu:
Hi Ednilsom,
I notice that you are using the GPW method together with the all-electron 6-311++Gxx basis set for carbon and oxygen. GPW is designed to work with pseudo-potentials and corresponding basis sets. I suggest you switch to the DZVP-MOLOPT-SR-GTH basis set for these two atomic kinds. If it is important to you that carbon and oxygen atoms be described at the all-electron level, then you should switch to the GAPW method and use the ALL potential for these atoms.
I am not sure that this will change the convergence rate of your geometry optimization.
Best,
Augustin
________________________________
From: cp... at googlegroups.com <cp... at googlegroups.com> on behalf of Ednilsom Orestes <eore... at gmail.com>
Sent: Tuesday, November 29, 2022 3:27 PM
To: cp2k <cp... at googlegroups.com>
Subject: [CP2K:18112] GEO_OPT too long
Dear all,
I'm performing a simple GEO_OPT test of CO2 molecule on a small (27 atoms) Pt(111) slab, which are freezed, by the way. But, the optimization is taking too long and, in most of the steps, the energy did NOT decreased.
What am I doing wrong and how to improve (actually normalize) the optimization time in this case? Thanks in advance.
&GLOBAL
PROJECT Ptlibxc4
RUN_TYPE GEO_OPT
PRINT_LEVEL MEDIUM
&END GLOBAL
&FORCE_EVAL
METHOD QS
STRESS_TENSOR ANALYTICAL
&SUBSYS
&CELL
A 8.313195 0.000000 0.000000
B 4.156600 7.199440 0.000000
C 0.000000 0.000000 47.000000
ALPHA_BETA_GAMMA 90.0 90.0 60.0
&END CELL
&COORD
Pt 8.3131998680 1.5998760605 0.2625664229<tel:(262)%20566-4229>
Pt 9.6987331793 3.9996901513 0.2625664229<tel:(262)%20566-4229>
Pt 2.7710666227 1.5998760605 0.2625664229<tel:(262)%20566-4229>
Pt 5.5421332453 1.5998760605 0.2625664229<tel:(262)%20566-4229>
Pt 4.1565999340 3.9996901513 0.2625664229<tel:(262)%20566-4229>
Pt 5.5421332453 6.3995042421 0.2625664229<tel:(262)%20566-4229>
Pt 6.9276665566 3.9996901513 0.2625664229<tel:(262)%20566-4229>
Pt 8.3131998680 6.3995042421 0.2625664229<tel:(262)%20566-4229>
Pt 11.0842664906 6.3995042421 0.2625664229<tel:(262)%20566-4229>
Pt 0.0000000000 0.0000000000 2.5251328458
Pt 5.5421332453 0.0000000000 2.5251328458
Pt 6.9276665566 2.3998140908 2.5251328458
Pt 2.7710666227 4.7996281816 2.5251328458
Pt 2.7710666227 0.0000000000 2.5251328458
Pt 4.1565999340 2.3998140908 2.5251328458
Pt 1.3855333113<tel:(385)%20533-3113> 2.3998140908 2.5251328458
Pt 5.5421332453 4.7996281816 2.5251328458
Pt 8.3131998680 4.7996281816 2.5251328458
Pt 6.9276665566 5.5995662119 4.7876992688<tel:(787)%20699-2688>
Pt 6.9276665566 0.7999380303 4.7876992688<tel:(787)%20699-2688>
Pt 4.1565999340 5.5995662119 4.7876992688<tel:(787)%20699-2688>
Pt 1.3855333113<tel:(385)%20533-3113> 0.7999380303 4.7876992688
Pt 4.1565999340 0.7999380303 4.7876992688<tel:(787)%20699-2688>
Pt 5.5421332453 3.1997521211 4.7876992688<tel:(787)%20699-2688>
Pt 8.3131998680 3.1997521211 4.7876992688<tel:(787)%20699-2688>
Pt 9.6987331793 5.5995662119 4.7876992688<tel:(787)%20699-2688>
Pt 2.7710666227 3.1997521211 4.7876992688<tel:(787)%20699-2688>
C 5.5398760357<tel:(539)%20876-0357> 3.1918740289 6.9058747884
O 5.4880348378 1.8992270578 7.3570952227
O 4.4466764893 3.8794527937 7.3633313274
&END COORD
&KIND Pt
BASIS_SET DZVP-MOLOPT-SR-GTH-q18
POTENTIAL GTH-PBE-q18
&END KIND
&KIND C
BASIS_SET 6-311++Gxx
POTENTIAL GTH-PBE-q4
&END KIND
&KIND O
BASIS_SET 6-311++Gxx
POTENTIAL GTH-PBE-q6
&END KIND
&END SUBSYS
&DFT
CHARGE 0
MULTIPLICITY 1
BASIS_SET_FILE_NAME BASIS_MOLOPT
BASIS_SET_FILE_NAME BASIS_MOLOPT_UCL
BASIS_SET_FILE_NAME EMSL_BASIS_SETS
POTENTIAL_FILE_NAME POTENTIAL
&POISSON
PERIODIC XYZ
PSOLVER PERIODIC
&END POISSON
&QS
METHOD GPW
EPS_DEFAULT 1.0E-12
&END QS
&MGRID
CUTOFF 1200
NGRIDS 4
REL_CUTOFF 100
&END MGRID
&KPOINTS
SCHEME MONKHORST-PACK 8 8 8
&END KPOINTS
&SCF
SCF_GUESS ATOMIC
EPS_SCF 1.0E-06
MAX_SCF 300
ADDED_MOS 50
&OUTER_SCF
MAX_SCF 50
EPS_SCF 1.0E-6
&END OUTER_SCF
&DIAGONALIZATION
ALGORITHM STANDARD
&END DIAGONALIZATION
&MIXING
METHOD BROYDEN_MIXING
ALPHA 0.070
BETA 0.030
NBROYDEN 8
&END MIXING
&SMEAR ON
METHOD FERMI_DIRAC
ELECTRONIC_TEMPERATURE [K] 300
&END SMEAR
&END SCF
&XC
&XC_FUNCTIONAL
&LIBXC
FUNCTIONAL GGA_X_PBE
&END LIBXC
&LIBXC
FUNCTIONAL GGA_X_PBE
&END LIBXC
&END XC_FUNCTIONAL
&VDW_POTENTIAL
POTENTIAL_TYPE PAIR_POTENTIAL
&PAIR_POTENTIAL
LONG_RANGE_CORRECTION
PARAMETER_FILE_NAME dftd3.dat
TYPE DFTD3
REFERENCE_FUNCTIONAL PBE
&END PAIR_POTENTIAL
&END VDW_POTENTIAL
&XC_GRID
XC_DERIV NN50_SMOOTH
&END XC_GRID
&END XC
&END DFT
&END FORCE_EVAL
&MOTION
&GEO_OPT
OPTIMIZER BFGS
TYPE MINIMIZATION
&END GEO_OPT
&CONSTRAINT
&FIXED_ATOMS
COMPONENTS_TO_FIX XYZ
LIST 1..27
&END FIXED_ATOMS
&END CONSTRAINT
&END MOTION
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns... at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/d875b9ed-28f4-452d-999b-f50868685365n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/d875b9ed-28f4-452d-999b-f50868685365n%40googlegroups.com?utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com<mailto:cp2k+unsubscribe at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/ad3482fb-0b37-4d0e-a8a9-27f893ddfc48n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/ad3482fb-0b37-4d0e-a8a9-27f893ddfc48n%40googlegroups.com?utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com<mailto:cp2k+unsubscribe at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CABpROdBKgXZrQNmQ_nn2vuMxdpiLLm0bNW1LiZkXGdgAjdRRmA%40mail.gmail.com<https://groups.google.com/d/msgid/cp2k/CABpROdBKgXZrQNmQ_nn2vuMxdpiLLm0bNW1LiZkXGdgAjdRRmA%40mail.gmail.com?utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/GV0P278MB0370F5F4BBE0954D3C67D97FD3E29%40GV0P278MB0370.CHEP278.PROD.OUTLOOK.COM.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20221212/8d513593/attachment-0001.htm>
More information about the CP2K-user
mailing list