[CPMD-list] PBC in MetaDynamics

[Nisanth Nair]

One correction to my previous email.

> If you want to introduce the PBC for DIST, you have to implement it for the
> collective variables and collective coordinates. So, it is bit involved,
On second thought, I think that you don't need the PBC for the collective 
variables (CV), only needed for collective coordinates (CC); the CV should 
any way follow CC by the restraints. Thus implementation should be very easy; 
it is only PBC adding to SUBROUTINE COLVAROFR in the distance calculation!

Nisanth


On Thursday 22 November 2007 10:23, Nisanth Nair wrote:
> Dear Ali,
>
> For distance (DIST) CV, the distance is NOT calculated with PBC,
> irrespective of the SYMMETRY.
>
> For all coordination number types of CVs, the individual distances in
> rational functions/fermi functions are calculated WITH  PBC.
>
> For dihedral angle, the PBC, is bit different. The collective coordinates
> are within -Pi and +Pi, while collective variables are within 0 and 2Pi.
>
> If you want to introduce the PBC for DIST, you have to implement it for the
> collective variables and collective coordinates. So, it is bit involved,
> and difficult to explain in single email. The best will be you start tyring
> out yourself, and ask doubts if you have any. You have to start looking at
> the routines meta_colvar_inp.F, especially the part  SUBROUTINE COLVAROFR.
> Look at some of the routines in meta_cv.F how PBC is incorporated while
> calculating distances. In meta_exlagr.F, and meta_hpot.F for implementing
> PBC for distances for collective coordinates and their displacements.
>
> Now, if you use distance between atoms which are not bonded, in
> metadynamics, you will be sampling almost all possible distances before you
> get them bonded; meaning you may have to wait for ages, unless you are
> lucky or they are some radicals.
> To sample only the relevant part of the free energy surface, i.e. to
> restrict the system to sample only specific range of distance, use keywords
> WALL+ and/or WALL- . Wrap you coordinates before you start, and keep the
> distances within say 4 or 5 Angstrom.
>
> Best of luck.
> Nisanth
>
> On Wednesday 21 November 2007 19:40, Ali Hassanali wrote:
> > Dear CPMD users,
> >
> > I haven't had any response on this issue so I'll try again. Does anyone
> > know how CPMD handles PBC conditions in metadynamics? So lets say you had
> > a collective variable being the distance between two atoms and the two
> > atoms sampled the distance moving away from each other. Within normal
> > PBC, the bond (distance between the atoms) will close in again.
> >
> > Or does metadynamics not deal with PBC in its implementation?
> >
> > Ali
> >
> > _______________________________________________
> > CPMD-list mailing list
> > CPMD-list at cpmd.org
> > http://cpmd.org/mailman/listinfo/cpmd-list

-- 
=========================================================================== 
 Nisanth N. Nair
 Lehrstuhl fuer Theoretische Chemie   Tel: ++49 (0)234 32 22121
 Ruhr-Universitaet Bochum                     Fax: ++49 (0)234 32 14045
 Universitaetstrasse 150
 D-44801 Bochum, Germany
 Email: nisanth.nair at theochem.ruhr-uni-bochum.de
 Web  : http://www.theochem.ruhr-uni-bochum.de
===========================================================================