[CP2K-user] convergence problem of spin unrestricted calculation of geo_opt

[Li Mengxuan]

 Dear cp2k users,

I built a Mg slab with Fe attachment using vesta and tried to optimize it 
with cp2k. It works well for spin restricted calculation but when it comes 
to spin unrestricted calculation, it didn't converge.  I set UKS .True. , 
multipilicity =5( spin moment is roughly 2.3 for iron and 2 iron in my 
slab), maginezation=2.3. 

Here is my input.  I appreciated all your suggestions.

Best regards,
Mengxuan

&GLOBAL
  PROJECT_NAME Mg_Feslab
  PRINT_LEVEL MEDIUM
  RUN_TYPE GEO_OPT
&END GLOBAL
&MOTION
  &GEO_OPT
  &END GEO_OPT
&END MOTION
&FORCE_EVAL
  METHOD Quickstep
  STRESS_TENSOR ANALYTICAL
  &PRINT
    &FORCES On
    &END FORCES
  &END PRINT
  &SUBSYS
    &COORD
Mg    0.000001    1.860046   11.675848
Mg    0.000001    1.860046   16.865114
Mg    0.000001    1.860046   22.054379
Mg    0.000001    1.860046   27.243645
Fe    0.000001    1.860046   32.432911
Fe    1.610850    0.930024    9.081215
Mg    1.610850    0.930024   14.270481
Mg    1.610850    0.930024   19.459747
Mg    1.610850    0.930024   24.649012
Mg    1.610850    0.930024   29.838278
    &END COORD
    &CELL
      A             3.2216992378         0.0000000000         0.0000000000
      B  -1.6108503481         2.7900737882         0.0000000000
      C  0.0000000000         0.0000000000        41.5141258240
    &END CELL
    &KIND Mg
      POTENTIAL GTH-PBE-q2
      ELEMENT Mg
      BASIS_SET spd-1z-s-2z-rc7pt5-r06pt0-plus-diffuse-s-8pt0
    &END KIND
 &KIND Fe
      POTENTIAL GTH-PBE-q16
      ELEMENT Fe
      BASIS_SET TZV2P-MOLOPT-SR-GTH-q16
      MAGNETIZATION 2.5
    &END KIND
  &END SUBSYS
  &DFT
    POTENTIAL_FILE_NAME GTH_POTENTIALS
    BASIS_SET_FILE_NAME PLATO_BASIS
    BASIS_SET_FILE_NAME BASIS_MOLOPT_UCL
    UKS .TRUE.
    MULTIPLICITY 5
    RELAX_MULTIPLICITY 0.05

    &MGRID
      CUTOFF [eV] 10000
      REL_CUTOFF [eV] 700
      NGRIDS 4
    &END MGRID
    &KPOINTS
      SCHEME MONKHORST-PACK 10 10 1
    &END KPOINTS
    &XC
      &XC_FUNCTIONAL PBE
      &END XC_FUNCTIONAL
    &END XC
    &SCF
      SCF_GUESS ATOMIC
      ADDED_MOS 8
      MAX_SCF 300
      EPS_SCF 1.0E-7
      &MIXING T
        METHOD BROYDEN_MIXING
        ALPHA 0.1
        NBUFFER 8
      &END MIXING
      &DIAGONALIZATION ON
        ALGORITHM Standard
      &END DIAGONALIZATION
      &SMEAR ON
        METHOD FERMI_DIRAC
        ELECTRONIC_TEMPERATURE [K] 300
      &END SMEAR
    &END SCF
    &QS
      EPS_DEFAULT 1.0E-10
    &END QS
  &END DFT
&END FORCE_EVAL

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20201208/bdcc0c75/attachment.htm>