Using TrajCast to accelerate modeling reactive systems #10
Description
Firstly, thanks for the effort made to provide TrajCast framework to the public. I have a question about the possibilty of using TrajCast to accelerate calculaitons that involves bond breaking/formation?
On page 3 of the preprint, the authors mentioned TrajCast is free of assumptions about chemical bonds, making it capable, in principle, of modeling reactive systems.
I am not fully able to understand how is that being possible and how is it related to some classical force fields like ReaxFF, If the framework is only updating the the atomic positions and velocities? Is there any information at the end about the partial charges on the atoms?
Thank you ~