Even though bottom-up Coarse-Grained (CG) models can reproduce the structural correlations from the atomistic reference, their general applicability is highly limited due to the nontransferable nature of bottom-up parametrization. This transferability issue is generally understood by the effective bottom-up CG interactions are essentially the many-body CG potential of mean forces, which is a free energy quantity. Thus, one needs to adjust the changes in the CG interaction when applying the existing bottom-up CG models to non-parametrized systems, but this is a non-trivial and difficult process. In this talk, I will present a novel approach by developing the Ultra-Coarse-Graining (UCG) methodology in the mean-field limit that enables transferring CG interactions to different temperatures and pressures. Unlike conventional CG models, the UCG methodology allows for mixing different CG force fields coupled by the order parameter. In the mean-field description, it is shown that the UCG force field can be decomposed into the energy and enthalpy functional by choosing the effective semi-global parameter to reflect the global changes in the system conditions. I further demonstrate that the mean-field UCG ansatz can bridge three distinct systems, which are not possible from using a single CG force field. Moreover, the UCG force fields based on the single phase can be transferred to the different temperature and pressure conditions, including the gas-liquid phases, and can even be applied for the liquid-vapor interfaces. These findings suggest a predicable bottom-up CG approach for high fidelity multi-scale modeling.

Toward temperature and phase transferable coarse-grained models

ACS Fall 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.