Multiscale Modeling Approach toward the Prediction of Viscoelastic Properties of Polymers

Abstract

We report a multiscale modeling approach to study static and dynamical properties of polymer melts at large time and length scales. We use a bottom-up approach consisting of deriving coarse-grained models from an atomistic description of the polymer melt. We use the iterative Boltzmann inversion (IBI) procedure and a pressure-correction function to map the thermodynamic conditions of the atomistic configurations. The coarse-grained models are incorporated in the dissipative particle dynamics (DPD) method. The thermodynamic, structural, and dynamical properties of the cis-1,4-polybutadiene melt are very well reproduced by the coarse-grained DPD models with a significative computational gain. We complete this study by addressing the challenging question of the investigation of the shear modulus evolution. As expected from experiments, the stress correlation functions show behaviors that are dependent on the molecular weights defining unentangled and weakly entangled polymer melts.

Developing coarse-grained potentials for the prediction of multi-properties of trans-1,4-polybutadiene melt

“…[31][32][33][34][35][36] The quantitative reproduction of material properties such as stress-strain response, rheological behavior and several key macroscopic observables (elastic modulus and shear viscosity) as well as the dynamical properties without any time-scale calibration has not been addressed. Furthermore, the representability of the resulting CG FFs, which remains a significant challenge within the multiscale modeling community but deserves stronger methodological efforts, has not been thoroughly evaluated too.…”

Prediction of structural and thermomechanical properties of polymers from multiscale simulations

“…48 The bulk atomistic congurations consisted of 40 chains of 200 monomers. These atomistic simulations 18 have established a quantitative prediction of the thermal expansion coefficient of 6.5 Â 10 À4 K À1 against an experimental value of 6.7 Â 10 À4 K À1 . The equations of motion were integrated using the Verlet leapfrog algorithm scheme with a timestep of 2 fs.…”

“…32,64 The procedure applied here allows to reproduce some thermodynamic properties at a target pressure but excludes any transferability to other pressures due to the deciency of reproducing the incompressibility of the polymer material at equilibrium. 18. First, concerning the degree of coarse-graining, we develop CG potentials for a level of coarsegraining of 4 leading to harder potentials.…”

“…However, many problems at the leading edge of materials science involve collective phenomena that occur over a range of time and length scales that are difficult to capture in atomistic simulations. These simplied/coarse-grained (CG) models allow for longer length and time scales than atomistic models that are unable to reach the complete relaxation of the polymer melts 7,8,18 with a reasonable computational effort. Different approaches have been developed to model the matter at two or more different length and time scales and participate to the overall strategy of multiscale modelling.…”

“…[4][5][6][7][8][9][10][11][12][13][14][15][16] Indeed, a powerful investigation of the relationship between molecular structure and mechanical properties of polymer melts requires statistical averages on well-equilibrated melts of very long chains. 5,9,16,18,[32][33][34] It represents an attractive alternative for designing new polymeric materials from the structure-properties relationship. S1 (ESI †) for a comparison between atomistic and mesoscopic simulations for the calculation of the end-to-end vector autocorrelation function).…”