Gas phase geometry optimization using wavelet - warning

[Philipp Marquetand]

Hi,
can anyone help me to understand what I am doing wrong here? Probably, I have to set some cut-offs differently but since I am new to CP2K, I don't really know how to proceed. Please find the warning and my input file below.
Thanks,
Philipp

 *** 13:59:12 WARNING in ps_wavelet_types:cp2k_distribution_to_z_slices    ***
 *** Density hits the upper boundary of the system in                      ***
 *** XY-plane/user/phchm/cp2k-cvs/cp2k/makefiles/../src/ps_wavelet_types.F ***
 *** line    331                                                           ***


&GLOBAL
  PROJECT pyridine
  RUN_TYPE GEO_OPT
  PRINT_LEVEL MEDIUM
&END GLOBAL
&FORCE_EVAL
  METHOD Quickstep
  &DFT
    BASIS_SET_FILE_NAME ../basissets/BASIS_SET
    POTENTIAL_FILE_NAME ../basissets/GTH_POTENTIALS   
    &MGRID
      CUTOFF 400
    &END MGRID   
    &XC
      &XC_FUNCTIONAL PBE
      &END XC_FUNCTIONAL
    &END XC  
    &POISSON
     POISSON_SOLVER WAVELET
     PERIODIC NONE
    &END POISSON
    &PRINT
      &MO
        EIGENVALUES
        OCCUPATION_NUMBERS
      &END MO
    &END PRINT
  &END DFT
  &SUBSYS
    &CELL
      ABC 12.0 12.0 12.0
      PERIODIC NONE
    &END CELL
    &COORD
 N     0.000000     0.000000     0.000000
 C     0.000000     0.000000     1.300000
 C     1.255737     0.000000     2.025000
 C     2.411881     0.000000     1.357500
 C     2.411881     0.000000    -0.092500
 C     1.255737     0.000000    -0.760000
 H    -0.943102     0.000000     1.844500
 H     1.255737     0.000000     3.114000
 H     3.354982     0.000000     1.902000
 H     3.354982     0.000000    -0.637000
 H     1.454729     0.000000    -1.830665
    &END COORD
    &KIND C
      BASIS_SET DZVP-GTH-PBE
      POTENTIAL GTH-PBE-q4
    &END KIND
    &KIND N
      BASIS_SET DZVP-GTH-PBE
      POTENTIAL GTH-PBE-q5
    &END KIND
    &KIND H
      BASIS_SET DZVP-GTH-PBE
      POTENTIAL GTH-PBE-q1
    &END KIND
  &END SUBSYS
  &PRINT 
    &GRID_INFORMATION ON
    &END
  &END
&END FORCE_EVAL
&MOTION
  &GEO_OPT
    MAX_FORCE 0.0004
    MAX_ITER 500
    OPTIMIZER BFGS
    TYPE MINIMIZATION
  &END GEO_OPT
&END MOTION

[Teodoro Laino]

Hi Philip,

increase the box size in the corresponding direction.
Teo

> --
> You received this message because you are subscribed to the Google Groups "cp2k" group.
> To post to this group, send email to cp...@googlegroups.com.
> To unsubscribe from this group, send email to cp2k+uns...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/cp2k?hl=en.

[Ross, Sun]

Hi,
1. POISSON_SOLVER WAVELET: the molecular is on the edge of the box and
this is the problem. Two methods cound handle this, First, use
POISSON_SOLVER MT and use large box; Second, still use WAVELET and add
follow in SUBSYS block:
&TOPOLOGY
&CENTER_COORDINATES
&END
&END

2. Add follow in the XC block.
&XC_GRID
XC_SMOOTH_RHO NN10
XC_DERIV SPLINE2_SMOOTH
&END XC_GRID

Regards.

[Philipp Marquetand]

Thank you, Sun and Teo!
Now, it works.