Two-chamber lattice model for thermodiffusion in polymer solutions

TL;DR

This paper introduces a two-chamber lattice model to understand thermodiffusion in polymer solutions, capturing the sign change of the Soret coefficient with solvent composition and temperature, aligning with experimental observations.

Contribution

The authors develop a novel two-chamber lattice model that accounts for compressibility and hydrogen bonding, providing insights into the sign change of the Soret coefficient in polymer solutions.

Findings

Soret coefficient changes sign with water content in the solution.

Temperature variations can induce sign changes in the Soret coefficient.

Model results agree with experimental data on polymer migration behavior.

Abstract

When a temperature gradient is applied to a polymer solution, the polymer typically migrates to the colder regions of the fluid as a result of thermal diffusion (Soret effect). However, in recent thermodiffusion experiments on poly(ethylene-oxide) (PEO) in a mixed ethanol/water solvent it is observed that for some solvent compositions the polymer migrates to the cold side, while for other compositions it migrates to the warm side. In order to understand this behavior, we have developed a two-chamber lattice model approach to investigate thermodiffusion in dilute polymer solutions. For a short polymer chain in an incompressible, one-component solvent we obtain exact results for the partitioning of the polymer between a warm and a cold chamber. In order to describe mixtures of PEO, ethanol, and water, we have extended this simple model to account for compressibility and hydrogen bonding between PEO and water molecules. For this complex system, we obtain approximate results for the composition in the warmer and cooler chambers that allow us to calculate Soret coefficients for given temperature, pressure, and solvent composition. The sign of the Soret coefficient is found to change from negative (polymer enriched in warmer region) to positive (polymer enriched in cooler region) as the water content of the solution is increased, in agreement with experimental data. We also investigate the temperature dependence of the Soret effect and find that a change in temperature can induce a change in the sign of the Soret coefficient. We note a close relationship between the solvent quality and the partitioning of the polymer between the two chambers, which may explain why negative Soret coefficients for polymers are so rarely observed.