Insights on the mechanism of water-alcohol separation in multilayer graphene oxide membranes: entropic versus enthalpic factors

Abstract

Experimental evidences have shown that graphene oxide (GO) can be impermeable to liquids, vapors and gases, while it allows a fast permeation of water molecules. The understanding of filtration mechanisms came mostly from studies dedicated to water desalination, while very few works have been dedicated to distilling alcohols. In this work, we have investigated the molecular level mechanism underlying the alcohol/water separation inside GO membranes. A series of molecular dynamics and Grand-Canonical Monte Carlo simulations were carried out to probe the ethanol/water and methanol/water separation through GO membranes composed of multiple layered graphene-based sheets with different interlayer distance values and number of oxygen-containing functional groups. Our results show that the size exclusion and membrane affinities are not sufficient to explain the selectivity. Besides that, the favorable water molecular arrangement inside GO 2D-channels forming a robust H-bond network and the fast water diffusion are crucial for an effective separation mechanism. In other words, the separation phenomenon is not only governed by affinities with the membrane (enthalpic mechanisms) but mainly by the geometry and size factors (entropic mechanisms). We verified that the 2D geometry channel with optimal interlayer distance are key factors for designing more efficient alcohol-water separation membranes. Our findings are consistent with the available experimental data and contribute to clarify important aspects of the separation behavior of confined alcohol/water in GO membranes.