Multiscale simulation of soft matter systems – from the atomistic to the coarse-grained level and back

TLDR

Soft matter phenomena span wide length and time scales, requiring simulation models that cover these scales; multiscale methods link multiple hierarchies to address phenomena at various resolutions. The paper reviews and develops coarse‑grained simulation models that bridge atomistic to mesoscopic levels to reach relevant time and length scales, and discusses methodological aspects and future challenges. The authors focus on structure‑based coarse‑graining methods linked to atomistic descriptions, covering polymer and biomolecular systems, and examine time‑scale, dynamics, resolution‑exchange, and adaptive‑resolution approaches. Coarse‑grained simulation combined with efficient backmapping produces well‑equilibrated atomistic structures of polymeric melts or biomolecular aggregates at long times and large scales, enabling comparison with experimental data.

Abstract

Many physical phenomena and properties of soft matter systems such as synthetic or biological materials are governed by interactions and processes on a wide range of length- and time-scales. Computer simulation approaches that are targeted at questions in these systems require models which cover these scales and the respective levels of resolution. Multiscale simulation methods combine and systematically link several simulation hierarchies so that they can address phenomena at multiple levels of resolution. In order to reach the mesoscopic time- and length-scales important for many material properties, methods that bridge from the atomistic (microscopic) to a coarser (mesocopic) level are developed. Here, we review coarse-grained simulation models that are linked to a higher resolution atomistic description. In particular, we focus on structure-based coarse-graining methods which are used for a variety of soft matter problems – ranging from structure-formation in amorphous polymers to biomolecular aggregation. It is shown that by coarse-grained simulation in combination with an efficient backmapping methodology one can obtain well-equilibrated long time- and large length-scale atomistic structures of polymeric melts or biomolecular aggregates which can be used for comparison to experimental data. Methodological aspects are addressed such as the question of the time-scales and dynamics in the different simulation hierarchies and an outlook to future challenges in the area of resolution exchange approaches and adaptive resolution models is presented.