Hello Adam,
sorry for the delay in my reply.
Somehow it finally worked without errors. Anyway I wonder why would the removal of the gmxlib property be a reasonable approach. If I remove gmxlib I would not be using the modified force field including the Zn parameters that I need right?
On a different note, despite including positional restraints on my Zn ion I can see it moving outside its coordination site during the NVT equilibration, so this is quite puzzling…
Thank you
Regards
Juan