(585e) Comparison of Friction Parameterization from Dynamics and Material Properties for a Coarse-Grained Polymer Melt | AIChE

(585e) Comparison of Friction Parameterization from Dynamics and Material Properties for a Coarse-Grained Polymer Melt


Johnson, L. - Presenter, National Institute of Standards and Technology
Phelan, F. Jr., National Institute of Standands & Technolog (NIST)
The hierarchy of length scales and thus time scales in polymeric materials often necessitate the use of coarse-grained (CG) simulation models to preserve the most pertinent features of the desired system while reducing computational effort. CG models of polymers typically focus on either preserving chemically specific features or reproducing viscoelastic behavior, or some combination of the two. In our recent work [1], we studied a CG method which combined a chemically specific potential with non-conservative forces to recover the dynamics for model high-temperature oligomer melts. There, we used iterative Boltzmann inversion (IBI) to parameterize the conservative potential to reproduce the structure of the melt from all-atom (AA) reference simulations; however, this smoothed potential landscape introduces unphysically fast dynamics. We utilized Langevin dynamics, which introduces friction and random forces set by a scalar friction parameter, as a means to recover the dynamics of the AA representation of the oligomeric systems. We found that the required friction depends on the dynamic quantity being parametrized, with chain center of mass diffusion requiring the lowest friction, monomer diffusion requiring the highest friction, and rotational diffusion falling in between. Here, we compare the parameterization of the Langevin friction parameter from material properties with the more fundamental dynamic measures utilized in our previous study. The calculation of material properties in AA representations of polymers is computationally expensive, while the measurements utilized to parameterize the dynamics in our prior work [1] were relatively more computationally simple and inexpensive. Here, we test the ability of short time, and thus less expensive, dynamics measures to parameterize CG models to recover material properties as well. We compare to experimental materials properties of polymer melts from literature.

[1] L.C. Johnson and F.R. Phelan Jr., “Dynamically consistent coarse-grain simulation model of chemically specific polymer melts via friction parameterization”, J. Chem. Phys. 154, 084114 (2021), https://doi.org/10.1063/5.0034910