Dear Dr. Axel Kohlmeyer and LAMMPS Users,
Hello All. First, thank you so much for your many advices on LAMMPS.
Is there a way in LAMMPS to implement atoms (assigned with a certain atom ID) at the center of mass of molecules where the implemented atoms retain the center of mass position throughout the dynamic run?
That is, I would like to track the trajectory of the center of mass of thiophenes moving during the dynamic run (shown in yellow dot in the below figure), but I do not want this “dummy” atoms located at the center of mass to interfere with the potential interactions and force fields.
I was thinking of implementing atoms with infinitesimal mass with a certain atom type. But, is there a way for these dummy atoms to retain the center of the mass as the thiophene molecules move around and still not interfere the actual molecular dynamics of the thiophene molecules and other side chain molecules?
Using fix rigid seems an idea, but I prefer the dummy atoms to move accordingly with the thiophene molecules.
Once again, I deeply appreciate for your advices during your busy work.
Sincerely,
Masato Koizumi
I don’t understand the following: “Using fix rigid seems an idea, but I prefer the dummy atoms to move accordingly with the thiophene molecules.”
Are the rings right now not rigid? If they are, you can simply follow the approach you mentioned and that should work I think. If they are not rigid, what exactly do you mean with " I prefer the dummy atoms to move accordingly with the thiophene molecules."
Finally, you could always just do this in post-processing if you just need the center of the rings for analysis and not for computation. That might actually be the easiest thing to do.
The thiophene rings have dynamic motion that would lead to a small change in the center of mass position (unless the bonds are constrained as rigid). In short, I do agree with Stefan though. This is most likely a type of calculation that could easily be done in post processing.
I am very unsure why you even want this information; if you are looking to use it for the radius of gyration might I recommend this: https://lammps.sandia.gov/doc/compute_gyration.html. While I think that the calculation may be “possible” in LAMMPS, the script would likely be rather complex. It is much much easier to post process the trajectory.
Best,
Dylan