[CPMD-list] periodic boundary conditions in metadynamics

[Axel Kohlmeyer]

On Sun, 18 Nov 2007, Ali Hassanali wrote:

AH> Dear CPMD Users,
AH> 
AH> Is anyone aware of how CPMD implements/handles PBC within the metadynamics
AH> method. Lets say for example I have 2 atoms and I sample along the
AH> distance between the 2 atoms in a periodic box. Does CPMD automatically
AH> "wrap around" the collective variable beyond a certain point? So if a bond
AH> breaks between 2 atoms during metadynamics, will the bonds reform due to
AH> PBC conditions?

this is a bit of a tricky issue and i had hoped
that one of the metadynamics developers had answered.
since this has not happened, i'll try my best based
on my observations (as a non-mtd user). 

as usual, the ultimate answer can be gathered from the 
source code and here you'll have to consider _two_
areas. 1) the electron structure calculation and 2)
the metadynamics force calculation (or constraint
force calculation).

in 1) the answer is simple. CPMD computes the electron
structure _always_ in PBC (even for isolated systems),
so all atomic coordinates are wrapped back into the 
principal unit cell.

in 2) it is not that straightforward. since CPMD usually
does not wrap atoms back in the trajectory files (unlike
most classical MD codes) since the cost of wrapping everything
back is negligible compared to the DFT part and this way it
is easier to compute mean squared displacements etc.
as far as i can see from the code, atom coordinates are not
treated according to minimum image conventions for calculation
of distance or related constraints (only for coordination number
constraints). so in your hypothetical case of a bond being broken
and reformed with a periodic image, the distance would not
be reset according to PBC and you should get a potential
profile with two minima. in practice this should not be
relevant, because if your contstraint (or collective variable)
would "wrap back" you'll have other problems. it is of course
important to know about the conventions used and knowing that
metadynamics/constraints and the electrons structure calculation
are fairly independent.

hope this helps,
   axel.

AH> Or are PBC conditions not accounted for in metadynamics?


AH> 
AH> Ali
AH> _______________________________________________
AH> CPMD-list mailing list
AH> CPMD-list at cpmd.org
AH> http://cpmd.org/mailman/listinfo/cpmd-list
AH> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.