magnetization of bcc-Fe

[wang... at gmail.com]

Hi everyone,

I'm using the new cp2k 3.0 version with its k points sampling method to 
model the bcc-Fe.
I hope the magnetization per atom can be properly modeled, so that I may 
use it for large cells with point defect, where distribution of 
magnetization per atom will be explored.
I used the following input, which produces a magnetization of 3.6 muB/atom, 
which is larger than experiments and other simulations of around 2.2 
muB/atom. (Similar in the case of fcc-Ni, which is 0.63 muB/atom > 0.52 
muB/atom in exp)
Is it rather a problem of cutoff and k points or the problem of potential 
and basis sets?
I'm quite new to cp2k, wish this question is not too naive.

Many thanks for any suggestion.



&GLOBAL
  PROJECT bcc-Fe
  RUN_TYPE ENERGY_FORCE
  PRINT_LEVEL LOW
  &FM
    TYPE_OF_MATRIX_MULTIPLICATION DBCSR_MM
    FORCE_BLOCK_SIZE T
  &END FM
  &DBCSR
    MM_DRIVER AUTO
  &END DBCSR
  FFTW_PLAN_TYPE MEASURE
&END GLOBAL
&FORCE_EVAL
  METHOD Quickstep
  &SUBSYS
    &KIND Fe
      ELEMENT   Fe
      BASIS_SET DZVP-MOLOPT-SR-GTH
      POTENTIAL GTH-PBE-q16
    &END KIND
    &CELL
      A     -1.43 1.43 1.43
      B     1.43 -1.43 1.43
      C     1.43 1.43 -1.43
      PERIODIC XYZ
    &END CELL
    &COORD
      Fe    0.000000000    0.000000000    0.000000000
    &END COORD
  &END SUBSYS
  &DFT
    BASIS_SET_FILE_NAME  BASIS_MOLOPT
    POTENTIAL_FILE_NAME  GTH_POTENTIALS
    SPIN_POLARIZED       T
    MULTIPLICITY 3
    RELAX_MULTIPLICITY 0.
    &PRINT
      &HIRSHFELD ON
         REFERENCE_CHARGE MULLIKEN
      &END HIRSHFELD
    &END PRINT
    &QS
      METHOD  GPW
      EPS_DEFAULT 1.0E-14
      EXTRAPOLATION USE_GUESS
    &END QS
    &MGRID
      NGRIDS 5
      CUTOFF 500
      REL_CUTOFF 100
    &END MGRID
    &XC
      &XC_FUNCTIONAL PBE
      &END XC_FUNCTIONAL
    &END XC
    &SCF
      SCF_GUESS ATOMIC
      EPS_SCF 2.0E-7
      MAX_SCF 300
      &DIAGONALIZATION  ON
        ALGORITHM STANDARD
      &END DIAGONALIZATION
      &MIXING  T
        METHOD BROYDEN_MIXING
        NSKIP        4
        N_SIMPLE_MIX 4
        ALPHA 0.08
        BETA  0.25
        NBUFFER 12
      &END MIXING
      ADDED_MOS 6 6
      CHOLESKY  INVERSE_DBCSR
      MAX_DIIS  6
      &SMEAR T
        METHOD FERMI_DIRAC
        ELECTRONIC_TEMPERATURE [K] 300
      &END SMEAR
    &END SCF
    &KPOINTS
      SCHEME MONKHORST-PACK 12 12 12
      FULL_GRID T
    &END KPOINTS
  &END DFT
  &PRINT
    &FORCES ON
    &END FORCES
  &END PRINT
&END FORCE_EVAL



Yi Wang
-------------------------------
Yi Wang
Ph.D student, Nanjing University of Sci. & Tech.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20160516/b69edb2f/attachment.htm>