[CP2K:616] The center of masses drift problem

[Teodoro Laino]

Hi,

The COM in cp2k should ALWAYS stay fixed whenever velocities are  
initialized by cp2k, since we zero the COM velocity, after
initialization.
If you see that the COM drifts this is very probably due to the poor  
quality with which you're computing the forces..
..
Specifically with your input you're using a CUTOFF of 100 Ryd.  
Remember that the CUTOFF is the density cutoff and not as
for example in CPMD the WF cutoff. So 100 is absolutely too small.
If you use a cutoff of at least 200-250 Ryd the COM does not drift  
anymore.

Moreover.. you're input file is quite old.. Especially for the  
localization part (I posted a message here few weeks ago) it is highly
suggested to use an updated CVS version (there were quite few bug  
fixes and reorganization of the input..)..

The translated input file with a new (updated today) version is:

&FORCE_EVAL
   METHOD Quickstep
  &DFT
       BASIS_SET_FILE_NAME GTH_BASIS_SETS
       POTENTIAL_FILE_NAME POTENTIAL
      &MGRID
       CUTOFF 250
      &END MGRID
      &QS
      &END QS
      &SCF
       SCF_GUESS ATOMIC
      &END SCF
       &XC
       &XC_FUNCTIONAL PADE
       &END XC_FUNCTIONAL
     &END XC
     &POISSON
         PERIODIC NONE
         POISSON_SOLVER MULTIPOLE
        &MULTIPOLE
           EWALD_PRECISION 0.00000001
           RCUT 4.0
           ANALYTICAL_GTERM
        &END MULTIPOLE
     &END POISSON
     &PRINT
       &LOCALIZATION
           EACH 1
           &LOCALIZE
             METHOD CRAZY
           &END
           MOLECULAR_DIPOLES T
           TOTAL_DIPOLE T
       &END LOCALIZATION
     &END PRINT
  &END DFT
  &SUBSYS
      &CELL
         ABC 8.0 8.0 8.0
         UNIT ANGSTROM
         PERIODIC NONE
      &END CELL
      &COORD
         O        -1.3459376534       -0.0354156443        0.1465219667
         H        -1.6005239833        0.7471984224       -0.3684408375
         H        -1.6119945737       -0.7853664624       -0.4092127268
      &END COORD
      &KIND O
         BASIS_SET DZVP-GTH
         POTENTIAL GTH-BLYP-q6
      &END KIND
      &KIND H
         BASIS_SET DZVP-GTH
         POTENTIAL GTH-BLYP-q1
      &END KIND
      &TOPOLOGY
        CENTER_COORDINATES
      &END
   &END SUBSYS
&END FORCE_EVAL
&GLOBAL
    PROJECT water
    RUN_TYPE MD
    PRINT_LEVEL LOW
&END GLOBAL
&MOTION
   &MD
     ENSEMBLE NVE
     STEPS 10000
     TIMESTEP 0.2
     TEMPERATURE 300
     &PRINT
       &CENTER_OF_MASS
       &END
     &END
  &END MD
   &PRINT
     &RESTART
       EACH 1
     &END
     &RESTART_HISTORY OFF
     &END
     &TRAJECTORY
       EACH 1
     &END
     &VELOCITIES
       EACH 1
     &END
    &END
&END

cheers,
Teo

On 21 Jan 2008, at 11:12, mat1... at gmail.com wrote:

>
> Hello,
>
> I'm performing ab initio MD run to calculate further the IR spectrum.
> I've done a non-periodic NVE test simulation of a molecule of water
> (the input below), but I found that there is a drift of the center of
> masses. Is there any way to avoid it?
>
>
> Regards,
>
>
>
>
> &FORCE_EVAL
>   METHOD Quickstep
>  &DFT
>       BASIS_SET_FILE_NAME GTH_BASIS_SETS
>       POTENTIAL_FILE_NAME POTENTIAL
>      &MGRID
>       CUTOFF 300
>      &END MGRID
>      &QS
>      &END QS
>      &SCF
>       SCF_GUESS ATOMIC
>      &END SCF
>       &XC
>       &XC_FUNCTIONAL BLYP
>       &END XC_FUNCTIONAL
>     &END XC
>     &POISSON
>         PERIODIC NONE
>         POISSON_SOLVER MULTIPOLE
>        &MULTIPOLE
>           EWALD_PRECISION 0.00000001
>           RCUT 4.0
>           ANALYTICAL_GTERM
>        &END MULTIPOLE
>     &END POISSON
>     &PRINT
>       &LOCALIZATION
>           EACH 1
>           LOC_TYPE CRAZY
>           MOLECULAR_DIPOLES T
>           TOTAL_DIPOLE T
>       &END LOCALIZATION
>     &END PRINT
>  &END DFT
>  &SUBSYS
>      &CELL
>         ABC 8.0 8.0 8.0
>         UNIT ANGSTROM
>         PERIODIC NONE
>      &END CELL
>      &COORD
>         O        -1.3459376534       -0.0354156443        0.1465219667
>         H        -1.6005239833        0.7471984224       -0.3684408375
>         H        -1.6119945737       -0.7853664624       -0.4092127268
>      &END COORD
>      &KIND O
>         BASIS_SET DZVP-GTH
>         POTENTIAL GTH-BLYP-q6
>      &END KIND
>      &KIND H
>         BASIS_SET DZVP-GTH
>         POTENTIAL GTH-BLYP-q1
>      &END KIND
>   &END SUBSYS
> &END FORCE_EVAL
> &GLOBAL
>    FFTLIB FFTSG
>    PROJECT water
>    RUN_TYPE MD
>    PRINT_LEVEL LOW
> &END GLOBAL
> &MOTION
>   &MD
>     ENSEMBLE NVE
>     STEPS 10000
>     TIMESTEP 0.2
>     TEMPERATURE 300
>  &END MD
>   &PRINT
>     &RESTART
>       EACH 1
>     &END
>     &RESTART_HISTORY OFF
>     &END
>     &TRAJECTORY
>       EACH 1
>     &END
>     &VELOCITIES
>       EACH 1
>     &END
>    &END
> &END
> >