Molecular simulations of phenol and ibuprofen removal from water using multilayered graphene oxide membranes

We present here non-equilibrium molecular dynamic simulations concerning the separation of phenol and ibuprofen as impurities compounds (ICs) in water by novel graphene oxide (GO) membranes. The coupling between water permeability and impurity rejection is studied as a function of membrane thickness and concentration, focusing on the underlying molecular phenomena. Results show that water permeability decreases as the number of layers increases. Moreover, molecular sieving can be achieved by tuning the number of GO layers and the surface chemistry of the sheet: water flow through layers is up to 20% faster than that in graphene layers, because of strong hydrogen bonded interactions with the oxygenated groups. Analysis of the simulation results suggests that upon adsorbing on the GO surface, the translational motion of ICs in water would be supressed. Nevertheless, hydrophilicity affects the permeability for membranes with high O/C ratio, owing to these strong hydrogen bonds. Furthermore, 100% rejection for the ICs can be obtained for most of the GO membranes with four layers. This study elucidates the important role of hydrophilic interactions in GO membranes to become ideal candidates for removal of organic pollutants from water, showing the applicability of molecular simulations to obtain molecular insights into this problem.