A Numerical Procedure to Evaluate Memory Effects in Non-Equilibrium Coarse-Grained Models

When developing coarse-grained models of complex processes out of equilibrium, one encounters the non-stationary generalized Langevin equation. The most important feature of this equation is the presence of a non-stationary memory kernel. Here, a method is presented to infer this memory kernel from MD simulation data in non-equilibrium processes. The method provides an improvement of a previously published numerical scheme, the applicability of which is limited by a truncation problem. As an illustration, the method is applied to ion dissociation of NaCl in water, for which non-trivial dampened oscillations are observed in the memory kernel.

Automated summary

A method is presented to infer the memory kernel in non-equilibrium processes, using ion dissociation in water as an example. This method improves a previously published numerical scheme and addresses the issue of truncation.