Electroosmotic Flow, Ionic Currents, and Pressure-Induced Flow in Microsystem Channel Networks: NMR Mapping and Computational Fluid Dynamics Simulations

Abstract

Using nuclear magnetic resonance (NMR) velocity mapping and NMR current density mapping as well as finite-element computational fluid dynamics methods, transport in microsystem electrolytic cells with increasing complexity has been examined ranging from single straight channels to random-site percolation clusters. This sort of system is considered as paradigm for more or less complex devices in microsystem technology. Corresponding model objects were designed on a computer and milled into ceramic (polar) or polystyrene (nonpolar) matrices. The pore space was filled with electrolyte solutions. Maps of the following quantities were recorded: velocity of hydrodynamic flow driven by external pressure gradients, velocity of electroosmotic flow (EOF), ionic current density in the presence of EOF, ionic current density in the absence of EOF. As far as possible, the experiments were supplemented by computational fluid dynamics simulations. It is shown that EOF, as well as the electric current, leads to recirculation patterns in closed complex structures such as percolation clusters. Remarkably, all transport patterns turned out to be dissimilar, and the occurrence and positions of eddies do not coincide in the different maps. Velocity histograms and the mean velocity as a function of the porosity have been evaluated. An EOF percolation transition is found at the geometrical percolation threshold. The combined application of NMR techniques for the quantitative, noninvasive visualization of the total variety of hydroand electrodynamics in the same channel system promises to become a powerful tool for design purposes in microsystem technology.