GROMACS version:2022.4
Dear Gromacs users,
I’m studying the process of micelles diffusing from the liquid phase to the surface.
In the simulation, I hope to keep water layer (resname SOL) staying in the region of 5<z<15 nm, but I’m not sure how to write the posre.itp to realize this setting. Thanks for your help.
Hi @minqi
I have some experience in this. I tried to help a few other people in writing down restraint, take a look at some other posts, starting from the manual, then on this forum here, and on researchgate.
I don’t understand why the water layer would move. I assume you have center of mass motion removal on, then there will be no net motion, apart from a water molecule evaporating now and then and crossing the periodic boundary.
Hi hess, thanks for your reply. In the simulation I set COM pulling to the micelle to drag it out of water. and the micelle is a bit large so when it reached the interface, the water layer also moved upwards with micelle, I’m not sure whether setting COMM grps is physical is this case.
Thanks obZehn, before asking this question I have read the manual and your reply in forum before, they are very nice guidance.
But I suddenly realized that perhaps using a flat-bottom potential cannot achieve my target, because I want to keep all water molecules in the range of 5 < z < 15 nm, while the layer form of the flat-bottom restraint can only restrict each water molecule to its adjacent layer. Is my understanding correct? If so is there any other way to solve this problem?
But that doesn’t sound like you are studying diffusion, as you wrote in your first post.
What geometry are use using for the pulling and what are the pull groups?
Sorry, I didn’t make it clear. Here I am investigating in the process of perfume molecule (like Limonene) diffusing from solvent to air, as in solvent they often aggregate into micelles, so I set the micelle and a dummy atom as pull groups, which located at the center and top of box respectively, the pull-geometry is direction.
Still the process you are simulating is not diffusion. I assume you want to compute the free-energy profile of the molecule moving into/out of the solvent.
I think the best solution is to use pull-geometry=cylinder and select the whole solvent as the first pull group. Doing something else, like fixing the whole solvent, can lead to strong local deformation of the interface around the molecule.
Now not relevant anymore, but I don’t understand why you are using a dummy atom. You could pull with respect to an absolute reference in space.
Thank you, Hess. I will try using pull-geometry=cylinder as you suggested. Indeed, I want to simulate the process of a molecule moving out of the solvent into the air.
Regarding your last point about pulling with respect to an absolute reference in space, I am not quite sure I understand. Do you mean that I can directly specify a coordinate as the reference group for the pulling? I have seen tutorials online that often set a dummy atom as the reference group and then pull the pulling group towards it.
Ah, then the people writing those tutorials are unaware that you can simply specify pull group 0 to have an absolute reference.
Got it, thank you so much :)