Multiscale modelling strategy using the lattice Boltzmann method for polymer dynamics in a turbulent flow

TL;DR

This paper introduces a multiscale simulation strategy combining direct numerical simulation and lattice Boltzmann methods to model polymer dynamics in turbulent flows across multiple scales, enhancing resolution while acknowledging some limitations.

Contribution

The paper presents a novel multiscale simulation approach that couples DNS and lattice Boltzmann methods for detailed polymer dynamics in turbulence.

Findings

Successfully reproduces local turbulent flow conditions in lattice Boltzmann simulations.

Enables passive polymer dynamics simulation across multiple scales.

Provides a framework for realistic macromolecule modeling in turbulent environments.

Abstract

Polymer dynamics in a turbulent flow is a problem spanning several orders of magnitude of length and time scales. A microscopic simulation covering all those scales from the polymer segment to the inertial scale of turbulence seems impossible within the foreseeable future. We propose a multiscale simulation strategy to enhance the spatio-temporal resolution of local Lagrangian turbulent flow by matching two different simulation techniques, i. e. direct numerical simulation for the flow as a whole, and the lattice Boltzmann method coupled to polymer dynamics at the Kolmogorov dissipation scale. Local turbulent flows sampled by Lagrangian tracer particles in the direct numerical simulation are reproduced in the lattice Boltzmann model with a finer resolution, by supplying the latter with both the correct initial condition as well as the correct time-dependent boundary condition, sampled from the former. When combined with a Molecular Dynamics simulation of a polymer chain in the lattice Boltzmann model, it provides a strategy to simulate the \emph{passive} dynamics of a polymer chain in a turbulent flow covering all those scales. While this approach allows for a fairly realistic model of the macromolecule, the back-coupling to the flow on the large scales is missing.