Simulation of Individual Polymer Chains and Polymer Solutions with Smoothed Dissipative Particle Dynamics

Sergey Litvinov, Qingguang Xie, Xiangyu Hu, Nikolaus A. Adams, Marco ElleroFluids 2016 · DOI

TL;DR

This paper extends the Smoothed Dissipative Particle Dynamics (SDPD) method to simulate polymer solutions and melts in 3D. The authors demonstrate that SDPD accurately captures both static and dynamic properties of single polymer chains, including crucial hydrodynamic interactions, and reproduces rheological behavior of concentrated solutions and melts. This work offers a robust mesoscopic simulation tool for complex polymer systems.

What Problem Does This Paper Address?

Simulating the behavior of polymers, especially in solutions and concentrated states, is a long-standing challenge in computational rheology. Accurately capturing the complex interplay of polymer chain dynamics and fluid hydrodynamics is computationally expensive with traditional methods like Molecular Dynamics (MD). While mesoscopic methods like Dissipative Particle Dynamics (DPD) offer a compromise, incorporating realistic hydrodynamic interactions and avoiding artifacts like polymer bond crossing remains difficult. This paper addresses the need for a mesoscopic simulation method that can accurately and efficiently model polymer solutions and melts, capturing both individual chain behavior and bulk rheological properties, particularly the effects of hydrodynamic interactions.

Methods

The core of this work is the application and extension of the Smoothed Dissipative Particle Dynamics (SDPD) method. SDPD is a mesoscopic particle-based method that builds upon Smoothed Particle Hydrodynamics (SPH). Key methodological aspects include:

  • SDPD for Solvent: The solvent is modeled using Navier-Stokes equations discretized in a Lagrangian framework, incorporating conservative, dissipative, and random forces between particles. Crucially, thermal fluctuations are included in a physically consistent manner, satisfying the fluctuation-dissipation theorem (FDT).
  • Coarse-Grained Polymer Model: Polymers are represented as chains of beads connected by Finite Extensible Non-linear Elastic (FENE) potentials. This potential prevents bond crossing and, combined with the stiff equation of state for the solvent, effectively accounts for excluded volume effects.
  • Hydrodynamic Interactions: Hydrodynamic interactions between solvent and polymer particles, and among polymer beads themselves, are implicitly handled through the SDPD forces. This is a key advantage over methods that require explicit modeling of these interactions.
  • Reverse Poiseuille Flow (RPF): For simulating concentrated polymer solutions and melts, the authors employ a "reverse Poiseuille flow" setup. This involves applying a constant force to drive flow through a channel, allowing for the measurement of steady-state rheological properties like viscosity and normal stress coefficients.

Datasets

Not discussed in the paper. The simulations are performed on generated systems with varying numbers of polymer beads and concentrations.

Key Insights and Ideas

  • Accurate Hydrodynamic Interactions: The SDPD model, as implemented, successfully captures the full development of hydrodynamic interactions for single polymer chains, aligning with theoretical predictions from the Zimm model. This is a significant achievement for a mesoscopic method.
  • Exclusion of Artifacts: The FENE potential and the stiff equation of state effectively prevent polymer bond crossing and bead overlap, respectively, ensuring more realistic simulations compared to some other mesoscopic methods.
  • Zimm vs. Rouse Behavior: The dynamic properties of single polymer chains (relaxation times) strongly follow the Zimm model, which explicitly accounts for hydrodynamic interactions, rather than the Rouse model, which neglects them. This provides a strong validation of the SDPD's ability to incorporate HI.
  • Rheological Prediction Power: The RPF setup coupled with SDPD effectively predicts steady-state rheological properties of polymer melts and solutions, showing good agreement with previous studies.
  • Cross-Stream Migration: The method qualitatively reproduces the polymer cross-stream migration phenomenon observed in experiments and other simulations, highlighting its utility for complex flow behaviors.

Key Results

  • Single Polymer Properties: Static properties (radius of gyration, static structure factor) and dynamic properties (diffusion coefficient, longest relaxation time, Rouse mode relaxation times) of single polymer chains were calculated and found to be in excellent agreement with theoretical predictions (Zimm model, self-avoiding walk exponents).
  • Polymer Melts: Velocity profiles for polymer melts under RPF showed increasingly non-Newtonian behavior with increasing chain length, consistent with expectations. Shear stress distributions were also accurately reproduced.
  • Polymer Solutions (50% concentration): The simulation of a concentrated polymer solution demonstrated a Newtonian-like behavior with slightly shear-rate dependent viscosity and qualitatively reproduced the cross-stream migration phenomenon.

Limitations and Future Work

  • Computational Cost: While SDPD is more efficient than MD, simulating very large systems or very long timescales can still be computationally demanding.
  • Parameter Sensitivity: The paper notes that differences in parameters (e.g., spring constant, maximum spring length) between SDPD and previous DPD simulations might explain minor discrepancies in rheological behavior. Further investigation into parameter optimization and transferability would be beneficial.
  • Future Studies: The authors suggest that the SDPD method is well-suited for studying polymer-wall interactions and transitions in polymer conformation, which are areas currently under investigation. The ability to accurately capture excluded volume effects is particularly highlighted for these future studies.

Summary updated

Abstract

In an earlier work (Litvinov et al., Phys.Rev.E 77, 066703 (2008)), a model for a polymer molecule in solution based on the smoothed dissipative particle dynamics method (SDPD) has been presented. In the present paper, we show that the model can be extended to three-dimensional situations and simulate effectively diluted and concentrated polymer solutions. For an isolated suspended polymer, calculated static and dynamic properties agree well with previous numerical studies and theoretical predictions based on the Zimm model. This implies that hydrodynamic interactions are fully developed and correctly reproduced under the current simulated conditions. Simulations of polymer solutions and melts are also performed using a reverse Poiseuille flow setup. The resulting steady rheological properties (viscosity, normal stress coefficients) are extracted from the simulations and the results are compared with the previous numerical studies, showing good results.
BibTeX
@inproceedings{litvinov2016simulation,
  author = {Sergey Litvinov and Qingguang Xie and Xiangyu Hu and Nikolaus A. Adams and Marco Ellero},
  title = {{Simulation of Individual Polymer Chains and Polymer Solutions with Smoothed Dissipative Particle Dynamics}},
  booktitle = {Fluids},
  year = {2016},
  doi = {10.3390/fluids1010007},
}
Figure