[gmx-users] Defining strongly interdependent dihedral force constants

[mts503]

I'm trying to model a species for which the force field I'm using (OPLS-AA)
doesn't contain all the force constants needed for some of the groups
present. I've successfully defined several angle and dihedral force
constants using quantum mechanical simulations to generate potential energy
curves and I'm happy that the method works, however there are a few that I'm
struggling with. 

The specific dihedrals are adjacent ones between aromatic rings with a
bridging atom between, e.g CA-CA-X-CA' and CA-X-CA'-CA', where the CA atoms
are part of a phenyl ring and CA' atoms are in a napthyl group (Napth-X-Ph). 

It seems that there is a large amount of inter-dependence between the two
dihedrals to the degree that the energy difference I'm trying to fit for one
varies significantly with the value of the other angle. Fixing one dihedral
at its most favourable angle and fitting the other (and vice-versa) results
in a potential energy surface that is quite different to that obtained by
quantum mechanics. 

Has anyone else managed to fit a similar system with strongly coupled
dihedrals? Any suggestions or advice would be much appreciated!



--
View this message in context: http://gromacs.5086.n6.nabble.com/Defining-strongly-interdependent-dihedral-force-constants-tp5002539.html
Sent from the GROMACS Users Forum mailing list archive at Nabble.com.