Top Page > Browsing

scf convergence problem in soc calculation Date: 2020/10/12 04:28 Name: reza   <majidiniareza@gmail.com> hello to all. I am doing soc calculation . this is my input file. # # File Name # System.CurrrentDirectory ./ # default=./ System.Name GFO level.of.stdout 1 # default=1 (1-3) level.of.fileout 1 # 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-s2p2d2f1 Ga_PBE13 Fe Fe6.0S-s2p2d2f1 Fe_PBE13S O O5.0-s2p2d1 O_PBE13 Definition.of.Atomic.Species> <Hubbard.U.values # eV Ga 1s 0.0 2s 0.0 1p 0.0 2p 0.0 1d 0.0 2d 0.0 1f 0.0 Fe 1s 0.0 2s 0.0 1p 0.0 2p 0.0 1d 4.0 2d 0.0 1f 0.0 O 1s 0.0 2s 0.0 1p 0.0 2p 0.0 1d 0.0 Hubbard.U.values> # # Atoms # Atoms.Number 40 Atoms.SpeciesAndCoordinates.Unit Ang # Ang|AU <Atoms.SpeciesAndCoordinates # Unit=Ang. 1 Ga 1.28066 9.38418 0.88738 6.5 6.5 0.0 0.0 0.0 0.0 1 off 2 Ga 7.43807 4.67018 4.17712 6.5 6.5 0.0 0.0 0.0 0.0 1 off 3 Ga 3.04057 9.38419 3.44838 6.5 6.5 0.0 0.0 0.0 0.0 1 off 4 Ga 5.67816 4.67019 1.61608 6.5 6.5 0.0 0.0 0.0 0.0 1 off 5 Ga 1.38022 2.85354 4.11797 6.5 6.5 0.0 0.0 0.0 0.0 1 off 6 Ga 7.33849 7.56756 0.94650 6.5 6.5 0.0 0.0 0.0 0.0 1 off 7 Ga 2.94099 2.85354 1.55696 6.5 6.5 0.0 0.0 0.0 0.0 1 off 8 Ga 5.77773 7.56756 3.50751 6.5 6.5 0.0 0.0 0.0 0.0 1 off 9 Fe 1.29944 5.45554 0.94921 9 5 0.0 0.0 0.0 0.0 1 off 10 Fe 7.41927 0.74155 4.11527 9 5 0.0 0.0 0.0 0.0 1 off 11 Fe 3.02180 5.45553 3.51022 9 5 0.0 0.0 0.0 0.0 1 off 12 Fe 5.69694 0.74153 1.55427 9 5 0.0 0.0 0.0 0.0 1 off 13 Fe 0.25846 7.47108 3.44746 5 9 0.0 0.0 0.0 0.0 1 off 14 Fe 8.46030 2.75713 1.61699 5 9 0.0 0.0 0.0 0.0 1 off 15 Fe 4.06277 7.47104 0.88644 5 9 0.0 0.0 0.0 0.0 1 off 16 Fe 4.65593 2.75709 4.17801 5 9 0.0 0.0 0.0 0.0 1 off 17 O 2.81474 3.97881 4.98049 3 3 0.0 0.0 0.0 0.0 1 off 18 O 5.90399 8.69286 0.08385 3 3 0.0 0.0 0.0 0.0 1 off 19 O 1.50641 3.97879 2.41946 3 3 0.0 0.0 0.0 0.0 1 off 20 O 7.21217 8.69290 2.64483 3 3 0.0 0.0 0.0 0.0 1 off 21 O 4.23188 4.03270 2.58837 3 3 0.0 0.0 0.0 0.0 1 off 22 O 4.48676 8.74669 2.47600 3 3 0.0 0.0 0.0 0.0 1 off 23 O 0.08933 4.03275 0.02736 3 3 0.0 0.0 0.0 0.0 1 off 24 O 8.62943 8.74675 5.03700 3 3 0.0 0.0 0.0 0.0 1 off 25 O 8.75954 1.84244 3.30464 3 3 0.0 0.0 0.0 0.0 1 off 26 O -0.04086 6.55641 1.75979 3 3 0.0 0.0 0.0 0.0 1 off 27 O 4.35660 1.84236 0.74363 3 3 0.0 0.0 0.0 0.0 1 off 28 O 4.36214 6.55633 4.32081 3 3 0.0 0.0 0.0 0.0 1 off 29 O 1.37497 1.83009 0.74846 3 3 0.0 0.0 0.0 0.0 1 off 30 O 7.34377 6.54412 4.31594 3 3 0.0 0.0 0.0 0.0 1 off 31 O 2.94622 1.83010 3.30948 3 3 0.0 0.0 0.0 0.0 1 off 32 O 5.77246 6.54411 1.75496 3 3 0.0 0.0 0.0 0.0 1 off 33 O 1.46273 6.32215 4.31628 3 3 0.0 0.0 0.0 0.0 1 off 34 O 7.25601 1.60819 0.74814 3 3 0.0 0.0 0.0 0.0 1 off 35 O 2.85852 6.32211 1.75530 3 3 0.0 0.0 0.0 0.0 1 off 36 O 5.86016 1.60815 3.30913 3 3 0.0 0.0 0.0 0.0 1 off 37 O 1.45280 8.81816 2.63396 3 3 0.0 0.0 0.0 0.0 1 off 38 O 7.26594 4.10423 2.43045 3 3 0.0 0.0 0.0 0.0 1 off 39 O 2.86840 8.81810 0.07294 3 3 0.0 0.0 0.0 0.0 1 off 40 O 5.85026 4.10422 4.99147 3 3 0.0 0.0 0.0 0.0 1 off Atoms.SpeciesAndCoordinates> Atoms.UnitVectors.Unit Ang # Ang|AU <Atoms.UnitVectors # unit=Ang. 8.795 0.000 0.000 0.000 9.428 0.000 0.000 0.000 5.122 Atoms.UnitVectors> # # SCF or Electronic System # scf.XcType GGA-PBE # LDA|LSDA-CA|LSDA-PW|GGA-PBE scf.SpinPolarization NC # On|Off|NC scf.SpinOrbit.Coupling On # On|Off, default=off scf.Hubbard.U On # On|Off , default=off scf.Hubbard.Occupation dual # onsite|full|dual, default=dual scf.ElectronicTemperature 300.0 # default=300 (K) scf.energycutoff 300.0 # default=150 (Ry) scf.maxIter 400 # default=40 scf.EigenvalueSolver band # DC|GDC|Cluster|Band scf.Kgrid 4 4 4 # means n1 x n2 x n3 scf.Mixing.Type RMM-DIISH # Simple|Rmm-Diis|Gr-Pulay|Kerker|Rmm-Diisk scf.Init.Mixing.Weight 0.01 # default=0.30 scf.Min.Mixing.Weight 0.001 # default=0.001 scf.Max.Mixing.Weight 0.1 # default=0.40 scf.kerker.factor 1 # default=1 scf.Mixing.History 20 # default=5 scf.Mixing.StartPulay 20 # default=6 scf.Mixing.EveryPulay 1 # default=6 scf.criterion 1.0e-6 # default=1.0e-6 (Hartree) the problem is scf convergence. I have changed all mixing tags but I couldn't get convergence. the result of run is below. SCF= 1 NormRD= 1.000000000000 Uele= -269.664255521798 SCF= 2 NormRD= 10.309747184817 Uele= -269.668709936491 SCF= 3 NormRD= 10.278210065166 Uele= -270.232961978430 SCF= 4 NormRD= 7.166102641252 Uele= -270.883631062642 SCF= 5 NormRD= 4.808044981317 Uele= -271.464741641596 SCF= 6 NormRD= 3.084985583104 Uele= -271.896597555983 SCF= 7 NormRD= 1.961024283813 Uele= -272.193141822655 SCF= 8 NormRD= 1.303378418510 Uele= -272.405854215985 SCF= 9 NormRD= 0.933293566019 Uele= -272.577538228405 SCF= 10 NormRD= 0.715916414347 Uele= -272.731416520919 SCF= 11 NormRD= 0.577907935386 Uele= -272.877109709695 SCF= 12 NormRD= 0.483952085681 Uele= -273.017481510306 SCF= 13 NormRD= 0.416523951913 Uele= -273.152665734697 SCF= 14 NormRD= 0.366005212745 Uele= -273.281966572280 SCF= 15 NormRD= 0.326360325866 Uele= -273.404628279805 SCF= 16 NormRD= 0.294104043607 Uele= -273.520101130608 SCF= 17 NormRD= 0.266923717959 Uele= -273.628090284795 SCF= 18 NormRD= 0.243186898956 Uele= -273.728528970931 SCF= 19 NormRD= 0.222178350089 Uele= -273.821538049540 SCF= 20 NormRD= 0.197584333845 Uele= -274.766239768796 SCF= 21 NormRD= 0.058660988196 Uele= -274.804626993261 SCF= 22 NormRD= 0.027436352523 Uele= -274.815427669340 SCF= 23 NormRD= 0.011976895561 Uele= -274.822600371427 SCF= 24 NormRD= 0.012097426493 Uele= -274.838137351446 SCF= 25 NormRD= 0.019281399077 Uele= -274.824366091982 SCF= 26 NormRD= 0.010845472097 Uele= -274.814422287522 SCF= 27 NormRD= 0.004991953296 Uele= -274.812906241372 SCF= 28 NormRD= 0.003992271451 Uele= -274.816237576200 SCF= 29 NormRD= 0.003500381311 Uele= -274.814581199419 SCF= 30 NormRD= 0.002962810145 Uele= -274.814263694339 SCF= 31 NormRD= 0.002827147329 Uele= -274.814385892165 SCF= 32 NormRD= 0.002790353198 Uele= -274.814406857743 SCF= 33 NormRD= 0.002641756787 Uele= -274.813986949263 SCF= 34 NormRD= 0.002823847492 Uele= -274.814987576565 SCF= 35 NormRD= 0.002127998926 Uele= -274.814312180435 SCF= 36 NormRD= 0.002104249743 Uele= -274.813793206073 SCF= 37 NormRD= 0.001376574043 Uele= -274.814332247897 SCF= 38 NormRD= 0.001045077061 Uele= -274.814654547871 SCF= 39 NormRD= 0.000822583556 Uele= -274.814345506827 SCF= 40 NormRD= 0.000697846793 Uele= -274.814322377955 SCF= 41 NormRD= 0.000267471179 Uele= -274.814407116988 SCF= 42 NormRD= 0.000176042088 Uele= -274.814446666711 SCF= 43 NormRD= 0.000141055295 Uele= -274.814442756257 SCF= 44 NormRD= 0.000123795366 Uele= -274.814425192838 SCF= 45 NormRD= 0.000120570645 Uele= -274.814425862137 could some one give me any idea for better convergence? thanks a lot.

Page: [1]

Re: scf convergence problem in soc calculation ( No.1 ) Date: 2020/10/12 12:41 Name: Naoya Yamaguchi Hi, >scf.criterion 1.0e-6 # default=1.0e-6 (Hartree) >SCF= 44 NormRD= 0.000123795366 Uele= -274.814425192838 >SCF= 45 NormRD= 0.000120570645 Uele= -274.814425862137 The difference of "Uele" between "SCF=44" and "SCF=45" is 6.69299e-7 Hartree, and it is smaller than the value for "scf.criterion" of 1.0e-6 you set, that is, the calculation seems to be finished because it satisfies the condition of "scf.criterion". You can give a more strict value for "scf.criterion" to get better convergence. Regards, Naoya Yamaguchi Re: scf convergence problem in soc calculation ( No.2 ) Date: 2020/10/12 19:09 Name: reza  <majidiniareza@gmail.com> Hello mr.yamaguchi Thanks for your reply I only looked at the final Norm RD and compare it with the criterion at the front of it. but the other problem is that if we compare the energies of different lattice constants, we see an irregularity that we can not distinguish the optimized lattice constant. can this problem be solved by higher scf.criterion? the results are below. a=8.817 -1722.14268279 a=8.816 -1722.14268792 a=8.815 -1722.14268621 a=8.814 -1722.14269281 a=8.813 -1722.14269358 a=8.812 -1722.14268982 Thanks a lot. Re: scf convergence problem in soc calculation ( No.3 ) Date: 2020/10/12 21:31 Name: Naoya Yamaguchi Dear reza, Please take a look at the following. http://www.openmx-square.org/openmx_man3.9/node37.html Regards, Naoya Yamaguchi Re: scf convergence problem in soc calculation ( No.4 ) Date: 2020/10/21 20:26 Name: reza  <majidiniareza@gmail.com> thank you for your reply mr.yamaguchi.

Page: [1]