The net charge of MULLIKEN analysis for SE method depend on number of cores used

[姚懿]

Hi,

I am using cp2k/2.6.0

It seems the net charge calculated by SE method will depend on the number 
of cores used for the calculation.

Here is the outputs and input of the calculation, it seems only 1 core give 
the correct net charge value.

1 core
===
 POPULATION ANALYSIS

  # Atom   Element   Kind        Atomic population                    Net 
charge
       1      Ox        1                 6.618533                     
-0.618533
       2      Hy        2                 0.690733                     
 0.309267
       3      Hy        2                 0.690733                     
 0.309267
 # Total charge                           8.000000                     
-0.000000
===

2 cores
===
 POPULATION ANALYSIS

  # Atom   Element   Kind        Atomic population                    Net 
charge
       1      Ox        1                 6.618533                     
 5.381467
       2      Hy        2                 0.690733                     
 1.309267
       3      Hy        2                 0.690733                     
 1.309267
 # Total charge                           8.000000                     
 8.000000
===

3 cores
===
 POPULATION ANALYSIS

  # Atom   Element   Kind        Atomic population                    Net 
charge
       1      Ox        1                 6.618533                     
11.381467
       2      Hy        2                 0.690733                     
 2.309267
       3      Hy        2                 0.690733                     
 2.309267
 # Total charge                           8.000000                     
16.000000
===

input
===
&FORCE_EVAL
  METHOD Quickstep
  &DFT
    &QS
      METHOD PM6
      &SE
       ANALYTICAL_GRADIENTS F
      &END
    &END QS
    &SCF
      SCF_GUESS ATOMIC
    &END SCF
  &END DFT
  &SUBSYS
    &CELL
      ABC 20.0 20.0 20.0
      PERIODIC NONE
    &END CELL
    &COORD
          O                   .00000000     .00000000     .00000000
          H                  1.05000000     .00000000     .00000000
          H                  -.26289900    1.01655502     .00000000
    &END COORD
  &END SUBSYS
&END FORCE_EVAL
&GLOBAL
  PROJECT h2o
  RUN_TYPE GEO_OPT
&END GLOBAL

====

Yi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20150521/51e4f33c/attachment.htm>