Periodicity

Chris christopher... at chem.ox.ac.uk
Tue Jul 15 12:51:09 CEST 2008


Hi,

I am trying to run QM/MM with CP2K but get the following message:

CP2K| A non-periodic calculation has been requested but the system
size exceeds the cell size in at least one of the non-periodic
directions!

However, I think that I am asking for a periodic calculation.  In my
input I have the following:

  &SUBSYS
    &CELL
      ABC 82.228 89.091 85.890
      UNITS ANGSTROM
      PERIODIC XYZ
    &END CELL

And, for example, if in the QMMM section I set the command

E_COUPL NONE

I then get the message at the start of the output

 *** PERIODIC CALCULATION WITH NONE COUPLING WAS REQUESTED! SWITCHING
OFF   ***
 *** THE PERIODIC KEYWORD SINCE PERIODIC AND NON-PERIODIC CALCULATION
WITH  ***
 *** NONE COUPLING REPRESENT THE SAME
METHOD!                               ***

Why does CP2K think that I am asking for a non-periodic calculation,
while simultaneously implying that I am asking for a periodic
calculation?

In addition, suppose that I was asking for a non-periodic
calculation.  Why is is that when I increase the cell size the problem
does not go away?

To define the MM cell size, I used vmd to find minimum and maximum xyz
coordinates for the system and worked out how big the system was from
those.  Adding 20 to each of these coordinates does not solve the
error.  The system was centred on the origin before generating pdb and
psf files.

To define the QM cell size I did the same thing for the atoms in the
QM region, and added 10 to each of the xyz coordinates to give the box
size.  Adding another 10 did not fix the problem.

Thanks,

Chris





More information about the CP2K-user mailing list