Mesoscopic Simulations of the Polymer and Microfluidics.

Abstract

This thesis focuses on the dynamics and transport of flowing polymers in microfluidic devices. Using different mesoscopic simulation methods, we explore the dynamics of polymers in dilute and non-dilute, but unentangled, solutions under flow in confined geometries, namely periodic pressure-driven sudden contraction-expansion channel with contraction dimension comparable to Rg, the polymer radius of gyration. We first choose the method Stochastic Rotation Dynamics (SRD) to study this problem. But before SRD can be confidently used for quantitative calculations,there is a need to benchmark SRD for both fluid dynamics and polymer dynamics. We first examine the accuracy of SRD for contraction flow against results from the finite element method. We show that SRD results are influenced by unphysical compressibility effects due to the ghost-like SRD fluid particles, and we can minimize this effect by lowering the Mach number via adjusting different SRD parameters. We next examine the accuracy of SRD for isolated polymers against standard theoretical and Brownian dynamics (BD) results. We show that the main error is due to an inertial effect that finite bead mass has on polymer hydrodynamics, and we find that this effect is negilible at the ratio, of the distance over which polymer bead inertia is lost due to collisions with solvent to Rg, is less than 0.1. We finally apply SRD to study polymer migration in microfluidic contraction flow. The similarity in results from SRD and BD without hydrodynamic interaction (HI) at low Weissenberg number Wi (<10) indicates that HI has only a weak effect on polymer migration in our geometry. We find that the polymer migration is primarily due to streamline curvature on a length scale comparable to Rg, which produces a migration velocity that is proportional to the square of Wi. And using the central limit theorem, we show that streamline-curvature-induced (SCI) migration can, in long periodic geometries, lead to clear separation of polymers by molecular length. We find that while there are other mechanisms that can also cause polymer migration, SCI migration is the dominant mechanism for the polymer migration in our contraction flow at Wi<10.