Re: Convergence problem in magnetic slab calculation ( No.1 ) |
- Date: 2018/01/22 09:16
- Name: Samuel Dechamps <samuel.dechamps@uclouvain.be>
- Dear Eike,
this section might help you : http://www.openmx-square.org/openmx_man3.8/node175.html
Best
|
Re: Convergence problem in magnetic slab calculation ( No.2 ) |
- Date: 2018/01/23 15:09
- Name: Eike F. Schwier <schwier@hiroshima-u.ac.jp>
- Dear Samuel,
thanks for the link. I already tried the last two options previously but forgot the VNA / overcompleteness. Maybe it helps.
best, Eike
|
Re: Convergence problem in magnetic slab calculation ( No.3 ) |
- Date: 2018/03/03 21:29
- Name: T. Ozaki
- References: t-ozaki@issp.u-tokyo.ac.jp
- Hi,
I also tested your system and found that the system is very difficult to get convergence due to the anisotropic structure that one of unit vector is much longer than the others. I noticed that even careful tuning several parameters of mixing may not help acceleration of the convergence, and that your parameter settings seem to be almost best choice. This may suggest a need for further improvement of mixing method for such an anisotropic systems.
Anyway, I was able to manage to get the convergent structure as can be seen at http://t-ozaki.issp.u-tokyo.ac.jp/openmx-files/Fe001.dat http://t-ozaki.issp.u-tokyo.ac.jp/openmx-files/fe001.out
Thank for sharing your experience.
Regards,
TO
|
Re: Convergence problem in magnetic slab calculation ( No.4 ) |
- Date: 2018/04/05 07:32
- Name: Eike F Schwier <schwier@hiroshima-u.ac.jp>
- Dear Ozaki-sensei,
thank you very much for testing my structure. In the meantime I found some indication that the problem of convergence seems to be related to the magnetic structure / spin properties. When I was relaxing a Pb(110) slab I had no problems with convergence (even up to 65 ML), but once I included NC and Spin-Orbit Coupling even restarting a converged "nomd" calculation can take many SCF iterations, while the nonmagnetic slab seems to restart fine. I also found during some recent testing that the EF relaxation seems to converge to a relaxed geometry faster compared to the RF I initially used, but I only tried for thinner Fe slabs so far.
best regards, Eike
|
Re: Convergence problem in magnetic slab calculation ( No.5 ) |
- Date: 2018/04/22 22:34
- Name: reza <e_majidinia@yahoo.com>
- hello .
I have encountered with scf convergence problem in my work.please give me some suggestion to improve my calculation. # # File Name #
System.CurrrentDirectory ./ # default=./ System.Name GFO level.of.stdout 2 # default=1 (1-3) level.of.fileout 2 # default=1 (1-3)
# # restart using a restart file, *.rst #
scf.restart on # on|off,default=off
# # Definition of Atomic Species #
Species.Number 3 <Definition.of.Atomic.Species Ga Ga7.0-s1p1d1 Ga_CA13 Fe Fe6.0S-s1p1d1 Fe_CA13S O O5.0-s1p1 O_CA13 Definition.of.Atomic.Species>
# # Atoms #
Atoms.Number 40 Atoms.SpeciesAndCoordinates.Unit Ang # Ang|AU <Atoms.SpeciesAndCoordinates # Unit=Ang. 1 Ga 1.335592 9.381600 0.908711 6.5 6.5 2 Ga 7.398908 4.690800 4.167889 6.5 6.5 3 Ga 3.031658 9.381600 3.447011 6.5 6.5 4 Ga 5.702842 4.690800 1.629588 6.5 6.5 5 Ga 1.388960 2.853226 4.134687 6.5 6.5 6 Ga 7.345540 7.544026 0.941912 6.5 6.5 7 Ga 2.978290 2.853226 1.596388 6.5 6.5 8 Ga 5.756210 7.544026 3.480212 6.5 6.5 9 Fe 1.336291 5.448740 1.030093 8 6 10 Fe 7.398209 0.757940 4.046507 8 6 11 Fe 3.030959 5.448740 3.568393 8 6 12 Fe 5.703541 0.757940 1.508207 8 6 13 Fe 0.279242 7.496555 3.403860 6 8 14 Fe 8.455257 2.805755 1.672740 6 8 15 Fe 4.088008 7.496555 0.865561 6 8 16 Fe 4.646492 2.805755 4.211040 6 8 17 O 2.805521 4.000127 4.987760 3 3 18 O 5.928978 8.690928 0.088840 3 3 19 O 1.561728 4.000127 2.449460 3 3 20 O 7.172771 8.690928 2.627141 3 3 21 O 8.639730 4.054446 2.620693 3 3 22 O 0.094769 8.745247 2.455907 3 3 23 O 4.462019 4.054446 0.082393 3 3 24 O 4.272481 8.745247 4.994207 3 3 25 O 8.710917 1.856994 3.367360 3 3 26 O 0.023583 6.547794 1.709241 3 3 27 O 4.390833 1.856994 0.829060 3 3 28 O 4.343667 6.547794 4.247540 3 3 29 O 1.398830 1.869190 0.737275 3 3 30 O 7.335670 6.559990 4.339325 3 3 31 O 2.968420 1.869190 3.275575 3 3 32 O 5.766080 6.559990 1.801025 3 3 33 O 1.388873 6.238013 4.282163 3 3 34 O 7.345627 1.547213 0.794437 3 3 35 O 2.978377 6.238013 1.743862 3 3 36 O 5.756123 1.547213 3.332737 3 3 37 O 1.420230 8.874337 2.660849 3 3 38 O 7.314270 4.183537 2.415751 3 3 39 O 2.947021 8.874337 0.122549 3 3 40 O 5.787479 4.183537 4.954051 3 3 Atoms.SpeciesAndCoordinates> Atoms.UnitVectors.Unit Ang # Ang|AU <Atoms.UnitVectors # unit=Ang. 8.7345 0.0000 0.0000 0.0000 9.3816 0.0000 0.0000 0.0000 5.0766 Atoms.UnitVectors>
# # SCF or Electronic System # scf.XcType LSDA-CA # LDA|LSDA-CA|LSDA-PW|GGA-PBE scf.SpinPolarization On # On|Off|NC scf.ElectronicTemperature 300.0 # default=300 (K) scf.energycutoff 150.0 # default=150 (Ry) scf.maxIter 400 # default=40 scf.EigenvalueSolver band # DC|GDC|Cluster|Band scf.Kgrid 7 7 12 # means n1 x n2 x n3 scf.Mixing.Type rmm-diisk # Simple|Rmm-Diis|Gr-Pulay|Kerker|Rmm-Diisk scf.Init.Mixing.Weight 0.30 # default=0.30 scf.Min.Mixing.Weight 0.001 # default=0.001 scf.Max.Mixing.Weight 0.40 # default=0.40 scf.kerker.factor 1 # default=1 scf.Mixing.History 5 # default=5 scf.Mixing.StartPulay 6 # default=6 scf.Mixing.EveryPulay 1 # default=6 scf.criterion 1.0e-6 # default=1.0e-6 (Hartree)
# # # MD or Geometry Optimization #
MD.Type nomd # Nomd|Opt|DIIS|NVE|NVT_VS|NVT_NH MD.Opt.DIIS.History 7 # default=7 MD.maxIter 1 # default=1 MD.TimeStep 0.5 # default=0.5 (fs) MD.Opt.criterion 1.0e-4 # default=1.0e-4 (Hartree/bohr)
# # Band dispersion #
Band.dispersion on # on|off, default=off
Band.Nkpath 5 <Band.kpath 25 0.0000000000 0.0000000000 0.0000000000 0.6199859200 0.0000000000 0.0000000000 g x 25 0.6199859200 0.0000000000 0.0000000000 0.6199859200 0.3603033103 0.0000000000 X S 25 0.6199859200 0.3603033103 0.0000000000 0.0000000000 0.0000000000 0.0000000000 S g 25 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.0000000000 0.3353392952 g Z 25 0.0000000000 0.0000000000 0.3353392952 0.6199859200 0.3603033103 0.3353392952 Z R Band.kpath>
# # MO output #
MO.fileout off # on|off num.HOMOs 3 # default=1 num.LUMOs 3 # default=1 MO.Nkpoint 1 # default=1 <MO.kpoint 0.0 0.0 0.0 MO.kpoint>
# # DOS and PDOS #
Dos.fileout on # on|off, default=off Dos.Erange -10.0 10.0 # default = -20 20 Dos.Kgrid 7 7 12 # default = Kgrid1 Kgrid2 Kgrid3 Voronoi.charge off # on|off, default = off
# # output Hamiltonian and overlap #
HS.fileout on # on|off, default=off
|