Dear Accelrys Forum,

I am trying to figure out length distributions of hydrogen bonds. Maybe I am doing something totally wrong but its very simple and its giving me strange behavior, I hope someone has an idea.

First, I build my hydrogen bonded molecules as one fragment by drawing an extra bond. For example, I draw the -C=O---H-O- hydrogen bond as -C=O-H-O- . I put this into my Amorphous Cell environment to build a cell. Then I proceed to delete the covalent bond and re-optimise the geometry. I then run the Forcite Dynamics calculation for 600 ps in NPT ensemble.

What I get is really strange. The h-bond length stays at right around 2.2-2.4 A for oxygen to oxygen distance. The bond is very tight and in perfect alignment. This is quite strange since only the strngest hydrogen bonds can reach 2.4 A and anything below that is just impossible. Even if I run the dynamics at 800 C the bonds stay right together at this distance. So, what is going on? Am I getting stuck in some false local minimum of the forcefield? Is there a better way to build the system perhaps?

Also I might add that it would be nice if hydrogen bond objects could be analysed with Forcite-Analysis-Length_Evolution-Bond_Lengths since right now it cannot be done and it seems like an oversight.

Best regards,

Constantine

Details: I am running MS 6.12 Forcite with COMPASS forcefield 1 ps time step NPT ensemble at 1 bar usualyl 600 psec simulations.