Energy of atomic oxgyen and smearing

[Matthias Krack]

Dear Natalie

you may try "direct_p_mixing" instead of "broyden_mixing". In addition, you 
may increase ALPHA to 0.4 for a faster initial convergence and thus earlier 
onset of DIIS/Diag. This mixing approach works often well for molecular as 
well as non-metallic systems when using diagonalisation.

Best regards

Matthias



On Thursday, 7 July 2016 21:29:41 UTC+2, Natalie Austin wrote:
>
> Hello,
>
> I wanted to calculate the atomic energy of oxygen. I used the CP2K 
> tutorial for GEO_OPT of H2O as an example to generate the input file.  What 
> I noticed is that the energy calculation for atomic did not converge after 
> 500 SCF steps if I did not use smearing. Would anyone be able to explain 
> why this is? Because I thought for such a small atom smearing would not be 
> necessary. The energy for atomic oxygen also converged when I used the OT 
> method but I thought it best to use diagonalization to keep consistent with 
> the input I use for calculations on my metal systems. 
>   
> This is what I believe is the relevant part of the input form the 
> calculation that converged. I indicated which lines were omitted from the 
> calculation that did not coverge: 
> &FORCE_EVAL
>   METHOD QS
>   &DFT
>     BASIS_SET_FILE_NAME BASIS_MOLOPT
>     POTENTIAL_FILE_NAME GTH_POTENTIALS
>     &MGRID
>       CUTOFF 400
>       REL_CUTOFF 60
>       NGRIDS 4
>     &END MGRID
>     &QS
>        EPS_DEFAULT 1.0E-14
>        MAP_CONSISTENT
>     &END QS
>   &SCF
>      SCF_GUESS  ATOMIC
>      EPS_SCF 1.0E-7
>      MAX_SCF 500
>      ADDED_MOS  200                                       *Added MOS was 
> omitted in the calculation that did not converge.* 
>      CHOLESKY INVERSE
>      &SMEAR  ON
>          METHOD FERMI_DIRAC
>          ELECTRONIC_TEMPERATURE [K] 700     * This smear section was 
> omitted in the calculation that did not converge. *
>      &END SMEAR
>      &DIAGONALIZATION
>           ALGORITHM STANDARD
>      &END DIAGONALIZATION
>      &MIXING
>           METHOD BROYDEN_MIXING
>           ALPHA   0.1
>           BETA    1.5
>           NBROYDEN  8
>      &END MIXING
>   &END SCF
>     &XC
>       &XC_FUNCTIONAL PBE
>       &PBE
>       PARAMETRIZATION REVPBE
>       &END PBE
>       &END XC_FUNCTIONAL
>       &VDW_POTENTIAL
>          DISPERSION_FUNCTIONAL PAIR_POTENTIAL
>          &PAIR_POTENTIAL
>            TYPE DFTD3
>             REFERENCE_FUNCTIONAL PBE
>             CALCULATE_C9_TERM .TRUE.
>             PARAMETER_FILE_NAME dftd3.dat
>             R_CUTOFF 15.0
>             VERBOSE_OUTPUT .TRUE.
>          &END PAIR_POTENTIAL
>       &END vdw_POTENTIAL
>     &END XC
> UKS .TRUE.
> MULTIPLICITY 3
>
>
> Any help with this would be appreciated. 
> Thanks
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20160708/40aa01df/attachment.htm>