[CP2K-user] PM6 wrong atom in population analysis

[Stanislav Šimko]

Dear developers,

I'm not sure if it's a problem in my input or there indeed is a bug, but I 
noticed that atom type(s) are wrongly recognized in the population analysis 
section when I try to use PM6 method.

In my specific case, N (Nitrogen) is printed out as Ni (I assume Nickel).
The same behavior is observed for one of the older dev branches 
(v8.0, git:c1d396c) and the stable 6.1 version. 

I haven't observed such issues for an equivalent system with DFT/PBE.

I attach my input + a random a test dummy structure that reproduces the 
problem.

Thank you for help!
Best regards,
Stan.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20201027/14f8c134/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test2.xyz
Type: chemical/x-xyz
Size: 162 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20201027/14f8c134/attachment.xyz>
-------------- next part --------------
&GLOBAL
  PRINT_LEVEL MEDIUM
  PROJECT sio2
  RUN_TYPE CELL_OPT                # Geometry optimization calculation
#  RUN_TYPE GEO_OPT                # Geometry optimization calculation
&END GLOBAL

&MOTION
  &GEO_OPT                        # Parameters for GEO_OPT convergence
    MAX_FORCE 1.0E-4
    MAX_ITER 1000
    OPTIMIZER BFGS
    &BFGS
      TRUST_RADIUS [bohr] 0.1
    &END
  &END GEO_OPT
  &CELL_OPT
    OPTIMIZER BFGS
    TYPE DIRECT_CELL_OPT
    KEEP_ANGLES T
#    KEEP_SYMMETRY T
    &BFGS
    &END BFGS
  &END CELL_OPT
  &PRINT
    &FORCES
      FORMAT DCD 
    &END FORCES
    &RESTART
      BACKUP_COPIES 1
    &END RESTART
    &RESTART_HISTORY OFF 
    &END RESTART_HISTORY
    &TRAJECTORY
      FORMAT DCD_ALIGNED_CELL
      #FORMAT XYZ 
    &END TRAJECTORY
  &END PRINT
&END MOTION

&FORCE_EVAL
  METHOD Quickstep                # Quickstep - Electronic structure methods
  STRESS_TENSOR ANALYTICAL
  &DFT
    CHARGE 0                       # There is a negatively charged anion
    &QS
      METHOD PM6                  # Parametrization Method 6
      &SE
        PERIODIC EWALD
        INTEGRAL_SCREENING KDSO-D
        &COULOMB
           CUTOFF   8.0
        &END COULOMB
      &END SE
    &END QS
    &SCF                          # Convergence parameters for force evaluation
      SCF_GUESS ATOMIC
      EPS_SCF 1.0E-5
      MAX_SCF 100
#       &OUTER_SCF
#          EPS_SCF 1.0E-7
#          MAX_SCF 500
#       &END OUTER_SCF
      &DIAGONALIZATION T
          ALGORITHM STANDARD
          EPS_JACOBI   1.E-1
          JACOBI_THRESHOLD   1.E-7
      &END DIAGONALIZATION
      &PRINT
        &RESTART OFF
        &END
      &END PRINT
    &END SCF
    &POISSON                      # POISSON solver
      PERIODIC XYZ
      &EWALD
        EWALD_TYPE EWALD
        ALPHA  0.5
        GMAX 41 41 41
        &MULTIPOLES
           MAX_MULTIPOLE_EXPANSION QUADRUPOLE
        &END MULTIPOLES
      &END EWALD
    &END POISSON
    &PRINT
        &MULLIKEN
        &END MULLIKEN
    &END
  &END DFT

  &SUBSYS
    &CELL
      ABC 10.076 10.551 10.951
      ALPHA_BETA_GAMMA 89.50 85.72 88.00
      PERIODIC XYZ
      MULTIPLE_UNIT_CELL 1 1 1
      &CELL_REF
        ABC 10.076 10.551 10.951
        ALPHA_BETA_GAMMA 89.50 85.72 88.00
        MULTIPLE_UNIT_CELL 1 1 1
      &END CELL_REF
    &END CELL
    &TOPOLOGY
      COORD_FILE_FORMAT XYZ
      COORD_FILE_NAME test2.xyz
      MULTIPLE_UNIT_CELL 1 1 1
    &END TOPOLOGY
   &END SUBSYS
&END FORCE_EVAL