Parametrization of H+ and OH-

GROMACS version:
GROMACS modification: Yes/No
Dear all
I want to simulate OH- and H+ diffusion in polyelectrolyte using OPLS-AA force field. But, I could not find parameters of these ions in ion.itp of gromacs files. What should I do for that?

Thank you very much

If the parameters are not defined in the available ff, you need to parameterize it using QM tools. Check in the ff file whether these atoms are available or not. or use gmx insert-molecule tool to insert

Thank you very much for your response.

This is not a straightforward issue, as both OH- and H+ are not as simple as they seem. They both will have well defined hydration structure, and in fact, H+ doesn’t exist in solution; it’s H3O+. The proton is also quite quickly exchanged between water molecules, so studying such a system assuming a discrete H3O+ entity is flawed. OH- is similar. These are probably systems that would be better handled by ab initio MD. The classical assumptions break down.

1 Like

Is your problem solved? i also want to add OH- but don’t find how.

Hello,
As I understand it is not possible to parametrize OH- and study its behavior in classical MD. However, I used H+ in the form of H3O+ and skipped OH-.

Best

Thank you for your kind response. I have guessed that maybe only MD is not enough to parametrize OH-, but I am not sure because I am new to MD. I am really grateful for your reply which makes me more sure about this guess.

There are several quantities of interest that could be obtained from MD for OH- and H3O+ and the corresponding parameters exist in the literature. See for example here as well as cited works. It’s just that in general these ions are a bit too subtle for non-reactive classical MD, e.g., see Justin’s post above.

1 Like

You are welcome!

It has been mentioned in several articles that parametrization of OH- is a bit tricky. However, Impossible is impossible…!
I suggets you to try forcefield which are parametrized for electrolytes like APPLE&P mentioned here . It may help.

You can also refer to the response of Mr. Lemkul above in the current post.

Best,

1 Like

Thanks a lot again, and I’ll take a try as your suggetions.