as mentioned, I'm building a molecular dynamics simulation of decanoic acid and decanoate (with neutralising sodium cations) in water. I've minimized and equilibrated the simulation up to 295 K and I've generated a little movie of the run (0.002 ns steps for about 1 ms). I have noticed a couple of issues I was hoping to get advice on:

water seems to be getting into the lipid bilayer at points. Should I be concerned about this? I know passive transport through bilayers is a thing, but I'm unsure if water penetration happening frequently is something to be concerned about.

Lipids occasionally dissociate from the bilayer into the solution. Is this reasonable? I'm not sure if it's something I should be concerned about. I know I could potentially restrain the lipids, but given the purpose of this simulation is to determine the stability of the bilayer and determining area per lipid, restraining those lipids seems, to me, like defeating the purpose of the experiment.

Here is a linke to the first few ns of simulation so you can see what I mean. stackoverflow won't let me post a gif, I think, so I've linked my Linkedin page where I posted it:


I'm happy to provide more info but as this is a conceptual issue, less a coding issue, I didn't want to be dumping mdp files in here.

Note: sodium cations aggregate at the bilayer surface. This is something that the forcefields usually suffer and I will be transitioning to K+ ions next.

I'm aiming to simulate the stability of decanoic acid lipid bilayers in aqueous environments over temperature.

  • $\begingroup$ Are you trying to duplicate a published result, or are you using force fields / simulation settings that are not found in literature? $\endgroup$ Jan 1 at 1:19
  • $\begingroup$ @ShernRenTee is not in literature. There are a few unknowns in this process, including combining Lipid21 with GAFF2 in order to incorporate Lipid21's dihedral calculations, as using GAFF2 by itself makes the bilayer too stable. My supervisor said this is a reasonable thing to do, but given the number of unknowns, I think you've brought up a good point: I should use the forcefield on a published bilayer model first, then I can compare. I assume the effect is due to something in the electrostatics of decanoic acid, but with multiple unknowns I should start ruling things out. Thanks! $\endgroup$
    – Thomas
    Jan 1 at 11:05

1 Answer 1


I have been given a way forward by a user in the comments, so I am marking this question as completed. Please feel free to add extra answers if you think of something, but I have a path forward now. Thank you!


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .