Coarse-grained Modeling for Polymer Solutions via the Mori-Zwanzig formalism

In this talk, we present a new method to establish implicit-solvent coarse-grained (CG) modeling for polymer solutions to conserve the dynamical properties of polymers. In the CG modeling, tens to hundreds of atoms were grouped as one CG particle; and the CG dynamic equations were rigorously derived from the atomistic data. The solvent-mediated dynamics of polymers was accurately captured via the generalized Langevin equation (GLE) with a non-Markovian memory kernel based on the Mori-Zwanzig formalism. The computational cost for direct evaluation of the non-Markovian memory kernel and generation of colored noise was significantly reduced by exploiting the equivalence between the non-Markovian dynamics and Markovian dynamics in an extended space. A higher-order time-integration scheme was developed to further accelerate the CG simulations. To assess, validate, and demonstrate the established CG modeling, we have applied it to simulate four different types of polymer solution systems. We find that the proposed CG modeling effectively conserves the velocity autocorrelation function and diffusivity of polymers and enables two orders of speedup in computer time, compared with the reference molecular dynamics simulations.