Top Page > Browsing

problems on the geometry optimization of graphene Date: 2013/09/16 22:10 Name: yugd   <yugd12@mails.jlu.edu.cn> Hello, everyone, i am a novice at OpenMX, i have compiled the 3.7 version successfully. I want to relax the geometry of graphene(it is well known that all the C-C bond length is about 1.42 Ang). The tag of MD.Opt.criterion was set to be 1e-7 hartree/ang, and the calculation stopped normally. After repeating the unit cell, which gives a graphene sheet, the C-C bond lengths composing of the six-membered ring are different, and they are 1.466, 1.398, and 1.398 Ang respectively. The difference are notably large. Following is the input: # # File Name # System.CurrrentDirectory ./ # default=./ System.Name graphene level.of.stdout 1 # default=0 (1-3) level.of.fileout 1 # default=1 (1-3) DATA.PATH ../../DFT_DATA13 # # Definition of Atomic Species # Species.Number 1 <Definition.of.Atomic.Species C C6.0-s2p2 C_PBE13 Definition.of.Atomic.Species> # # Atoms # Atoms.Number 2 Atoms.SpeciesAndCoordinates.Unit FRAC # Ang|AU <Atoms.SpeciesAndCoordinates # Unit=Ang. 1 C 0.3333333 0.3333333 0.500000000000 2.0 2.0 2 C 0.6666667 0.6666667 0.500000000000 2.0 2.0 Atoms.SpeciesAndCoordinates> Atoms.UnitVectors.Unit Ang # Ang|AU <Atoms.UnitVectors # unit=Ang. 2.460000000000 0.000000000000 0.000000000000 1.230000000000 2.130420000000 0.000000000000 0.000000000000 0.000000000000 15.0000000000 Atoms.UnitVectors> # # SCF or Electronic System # scf.XcType GGA-PBE # LDA|LSDA-CA|LSDA-PW scf.SpinPolarization off # On|Off scf.ElectronicTemperature 300.0 # default=300 (K) scf.energycutoff 160.0 # default=150 (Ry) scf.maxIter 100 # default=40 scf.EigenvalueSolver band # Recursion|Cluster|Band #scf.lapack.dste dstevx # dstegr|dstedc|dstevx, default=dstegr scf.Kgrid 3 3 3 # means 4x4x4 scf.Mixing.Type rmm-diisk # Simple|Rmm-Diis|Gr-Pulay scf.Init.Mixing.Weight 0.010 # default=0.30 scf.Min.Mixing.Weight 0.001 # default=0.001 scf.Max.Mixing.Weight 0.200 # default=0.40 scf.Mixing.History 15 # default=5 scf.Mixing.StartPulay 5 # default=6 scf.criterion 1.0e-6 # default=1.0e-6 (Hartree) # # MD or Geometry Optimization # MD.Initial.MaxStep 0.002 MD.Type EF # Nomd|Opt|DIIS|NVE|NVT_VS|NVT_NH MD.Opt.DIIS.History 7 # default=7 MD.Opt.StartDIIS 10 # default=5 MD.maxIter 100 # default=1 MD.TimeStep 0.1 # default=0.5 (fs) MD.Opt.criterion 1.0e-7 # default=1.0e-4 (Hartree/bohr)

Page: [1]

Re: problems on the geometry optimization of graphene ( No.1 ) Date: 2013/09/17 15:21 Name: yugd  <yugd12@mails.jlu.edu.cn> When i change the scf.EigenvalueSolver tag from band to cluster. openmx gives the correct result. i am confused, grapheen is 2D, and i think the band method is more conformable. But reality beat me . what's going wrong? Re: problems on the geometry optimization of graphene ( No.2 ) Date: 2013/09/17 22:43 Name: T. Ozaki Hi, This is due to the small number of k-points. Increasing the number of k-points recovers the symmetry. Note that only the single k-point is enough to get the convergent result along the c-axis because of no dispersion along the axis. Also, the keyword 'scf.energycutoff' is relevant to preservation of the symmetry. In fact, my result with the k-points of 11x11x1 almost preserved the symmetry. Regards, TO Re: problems on the geometry optimization of graphene ( No.3 ) Date: 2013/09/18 08:51 Name: yugd  <yugd12@mails.jlu.edu.cn> Thank you very much, i will retry my task with more k points. many thanks to prof. T. Ozaki.

Page: [1]