I would like to simulate a liquid solvent (such as TEGDME) embedded into a polymeric cage (such as PEGDM). What is the most meaningful approach to estimate transport properties in such a system?

Three options that I have in mind are:

  • Using a FF typically used for liquid electrolytes such as GAFF or OPLS/A,
  • Use a machine learning interatomic potential such as ANI, M3Gnet or MTP,
  • Being more cautious and looking into transition state theory (such as NEB).

But I am open to others.


  • $\begingroup$ I gave my +1 long ago, but I just wanted to check up on you since it has been more than 9 months since this question was asked. Have you found an answer? Are you still actively in need of an answer? $\endgroup$ Sep 30, 2023 at 19:52
  • $\begingroup$ Hello @NikeDattani, I am still working on this and using TB (GFN1-xTB) within ASE $\endgroup$ Oct 2, 2023 at 7:52
  • $\begingroup$ Okay, I have bumped it up to the top of the front page by editing it. Hopefully this helps the question to get an answer! $\endgroup$ Nov 5, 2023 at 15:24
  • $\begingroup$ I added point 2, also pertinent to the discussion $\endgroup$ Nov 6, 2023 at 15:01


You must log in to answer this question.