Dear all Users,
I want to perform a NVT simulation of a aromatic molecule (HMF) in water. Now, The HMF molecule has two parts, one is planar ring and other are two functional group (carbonyl and -CH2OH group) attached with it. My plan is to treat the planar part as rigid entity for that purpose I am only mentioning L-J parameters and for the attached functional groups I want to make them flexible so I have to put all the bonded parameters for that.
From the manual I got that, “fix shake” and “fix rigid” are two option to run the simulation in this fashion.
Now my query is that,
- Is is sufficient to run the rigid pat with L-J parameters only. (no bond angle or torsional parameter)
- If not, do I need to provide all bonded parameters for the rigid part also to constrain the structure?
2)On the other hand if I use fix rigid along with fix NVT, will I be able to run the system correctly?
Any expert suggestion will be highly appreciated.
Thank you
Rameswar
Dear all Users,
I want to perform a NVT simulation of a aromatic molecule (HMF) in water. Now, The HMF molecule has two parts, one is planar ring and other are two functional group (carbonyl and -CH2OH group) attached with it. My plan is to treat the planar part as rigid entity for that purpose I am only mentioning L-J parameters and for the attached functional groups I want to make them flexible so I have to put all the bonded parameters for that.
From the manual I got that, “fix shake” and “fix rigid” are two option to run the simulation in this fashion.
fix shake won’t work for aromatic rings or any construct with more than one “central” atoms.
Now my query is that,
- Is is sufficient to run the rigid pat with L-J parameters only. (no bond angle or torsional parameter)
yes. it is still useful to set the force constants. best to something strong, like 10 or 100 times larger than otherwise common, so that the structure will stay (near) rigid during minimization, which fix rigid cannot be used.
- If not, do I need to provide all bonded parameters for the rigid part also to constrain the structure?
fix rigid doesn’t care. fix shake would require the bond length, but it is not applicable.
2)On the other hand if I use fix rigid along with fix NVT, will I be able to run the system correctly?
that should be fine for as long as the fix group for fix nvt and the fix group for fix rigid do not overlap. atoms may be time integrated only once.
axel.
Thank you very much Axel for such a nice clarification. I would perform the simulation with “fix rigid” or with a higher value of force constant. Thanks for your help.
Rameswar
Please read my reply more carefully.