I have been trying to calculate interfacial tension of H2S-H2O system at 313K and various pressure values. A number of researchers have simulated surface tension by first simulating a condensed system usually in NVT ensemble and then increase the size of the box in the z direction ( the system being in the middle). In my case, I tried simulating NPT and then NVT but the box simulation box increases continuously during NPT simulation. I tried Parrinello Rahman and Berendsen barostats. What can I do to control the continuous increase of the box?
What time step are you using and pressure coupling time constant?
Sometimes the initial configuration has too many overlaps resulting in a large energy and a large force/pressure inside the box. The box will expand to try and compensate this, but if you have a large timestep then the box doesn't always stop itself from expanding (Normally the box does stop expanding and start to compress back down but there are usually crashes before this happens).
If you lower the timestep the problem shouldn't be so bad, of if you reduce the pressure coupling time constant then the box will react more quickly to changes in the energy /pressure of the box and shouldn't expand to infinity. Try the timestep first.
Alternatively, this is a bit fiddly, but try running in NVT first to allow the molecules to remove high energy overlaps themselves and then try and run NPT when the system has lowered in energy. Then you can run in NVT again after if you want.
Hope this helps!
Hi Sakiru,
To add to what James suggested you to fix your problems I can first advise you to make sure that your force-field parameters have been set properly. Secondly, you check the topology of your system. If any of these two has not been fixed the atoms in your simulation cell may feel very huge forces, which in turn may lead some atoms to move continuously away from other atoms, and your simulation cell will continuously expand.
baris.
Thank you very much James & Baris. The time step reduction worked.
Cheers!
Dear Sakiru Badmos
As Baris Demir mentioned, it seems that your force field parameters are not accurate; from the other side of view, it may be probable that the force field applied is not consistent with the system.
If you are sure on your force field and its parameters, run your job at smaller time step and after achieving equilibrium condition run it at your considered time step.
Best regards
Set up your system properly.
Then, minimise the system energy using backbone restraints on your protein. Carefully check that the system energy has reached a plateau.
Finally, you may start NPT dynamics, but be sure to monitor the volume of your box. This should initially decrease, but then plateau and remain stable.
- Badges
- Science topic