Coarse-graining strategies in polymer solutions

TL;DR

This paper reviews a multiblob coarse-graining approach for polymer solutions, compares models to full-monomer results, and introduces a new tetramer model that better predicts thermodynamics and structure in the semidilute regime.

Contribution

The paper critically evaluates transferability in coarse-grained models and introduces a new tetramer model that improves predictions for polymer solutions.

Findings

Transferability has limited predictive power for thermodynamics.

The tetramer model accurately predicts properties up to .

More complex models are needed for higher concentration regimes.

Abstract

We review a coarse-graining strategy (multiblob approach) for polymer solutions in which groups of monomers are mapped onto a single atom (a blob) and effective blob-blob interactions are obtained by requiring the coarse-grained model to reproduce some coarse-grained features of the zero-density isolated-chain structure. By tuning the level of coarse graining, i.e. the number of monomers to be mapped onto a single blob, the model should be adequate to explore the semidilute regime above the collapse transition, since in this case the monomer density is very small if chains are long enough. The implementation of these ideas has been previously based on a transferability hypothesis, which was not completely tested against full-monomer results (Pierleoni et al., J. Chem. Phys, 127, 171102 (2007)). We study different models proposed in the past and we compare their predictions to full-monomer results for the chain structure and the thermodynamics in the range of polymer volume fractions \Phi between 0 and 8. We find that the transferability assumption has a limited predictive power if a thermodynamically consistent model is required. We introduce a new tetramer model parametrized in such a way to reproduce not only zero-density intramolecular and intermolecular two-body probabilities, but also some intramolecular three-body and four-body distributions. We find that such a model correctly predicts three-chain effects, the structure and the thermodynamics up to \Phi ~ 2, a range considerably larger than that obtained with previous simpler models using zero-density potentials. Our results show the correctness of the ideas behind the multiblob approach but also that more work is needed to understand how to develop models with more effective monomers which would allow us to explore the semidilute regime at larger chain volume fractions.