I am trying to perform MD simulation of Co2 molecule in gromacs (4.3.4). There is a nice example of the same here but suggests adding virtual sites (http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin/gmx-tutorials/vsites/03_simulation.html). I do not want to add virtual sites but rather add a constrain on angle ie O=C=O (180 degree) or a distance constrain between two oxygens to be (2*1.16)Å so that it remain liner. Here 1.16 Å is the bond length of O=C bond in Co2
I tried both ways but in the NPT or NVT simulation, the angle O=C=O fluctuates between 175 to 180 (attached image)
My questions are as follows
1- How to keep the angle fixed ? How to make the molecule rigid ?
2- What is the max deviation of angle that can happen / accepted when applying such constrain in gromacs
Force field used - TrappE
Update :-
I ran longer and the angle deviates much less (attached) and remain mostly from 179 to 180 degree. This points to my second question about how much deviate is acceptable ?
I have to use this topology to study nucleation behaviour.
Thanks
The "canonical" GROMACS answer (at least from the mailing list) is to indeed use vitual sites for a rigid CO2 model. See for instance: https://www.mail-archive.com/[email protected]/msg09979.html
In more recent versions of GROMACS there is an "linear angle" potential which you could possibly use? I have not used this myself, and unfortunately it is *not* documented so you might have to read the source code to get info on how to use it, see e.g.: gromacs/listed-forces/bonded.cpp
- Badges
- Science topic