Hybrid molecular-continuum modeling of diffusioosmotic flow in nanochannels

Diffsuioosmosis, the interfacially driven fluid motion in the presence of a concentration gradient, can be effectively modulated by leveraging various surface effects that gain prominence at the nanoscale. However, continuum modeling of nanoscale flows is often performed with certain a priori abstractions of pre-defined boundary conditions and interfacial fluid properties, making it highly inaccurate in capturing the sub-continuum effects. We optimize the magnitude and direction of diffusioosmotic mobility of solvent molecules in a negatively charged graphene nanochannel, resorting to full-scale molecular dynamics simulations by modulating the surface wettability and salt concentration of an aqueous NaCl solution. Subsequently, a pseudo-continuum model is proposed for extracting slip length, effective wall position, and interfacial viscosity from the molecular simulations, yielding an accurate prediction of diffusioosmotic flow at the nanoscale. Our results turn out to be of significant implications not only for the fundamental understanding of sub-continuum effects in diffusioosmotic flows in graphene nanochannels but also carry immense implications in applications ranging from blue energy harvesting, targeted drug delivery, and water desalination.