Micro-perforated membranes: versatile embodiments of non-1D-transport concept

A common definition of a membrane is something that in one dimension is much smaller than in the other two. Accordingly, membrane processes are often considered (locally) one- dimensional. This presentation will explore a relatively new class of strongly non-1D membrane processes occurring in membranes with relatively scarce micro-perforations. These can be, for example, tiny imperfections in barrier layers of composite RO membranes. Given that hydraulic permeances of barrier layers are several orders of magnitude lower than those of porous supports, flow patterns close to such imperfections are extremely inhomogeneous. We will consider what consequences this can have for concentration polarization close to the membrane surface and demonstrate an unusual lower rejection of larger solutes in this case. We will also see that in Forward Osmosis, close to such “openings” in the barrier layer, the direction of volume flow is reversed, which can have interesting consequences for the solute transfer from the feed as well as for the transfer of draw solute to the feed solution.

Another interesting embodiment of perforated membranes are composite structures where microperforated ion-exchange layers are put in series with nanoporous layers. We will see that such materials feature considerable and strongly asymmetric (depending on the current direction) electroosmosis, which means that they can be used as AC EO pumps. This behavior occurs due to the essentially different conditions for current-induced CP inside and outside such structures. Here, there already are some experimental data that qualitatively confirm model predictions.

In similar structures, there is also another interesting phenomenon of a strong counter- gradient accumulation of coionic species. Its pre-requisite is a simultaneous occurrence of noticeable advection and (space-dependent) electromigration. The former is made possible by the micro-perforations in the ion-exchange layer, while the latter occurs due to the current- induced CP of the surface of ion-exchange layer giving rise to strong conductivity gradients within the (nano)porous layer. Owing to them, the electromigration flow component becomes coordinate-dependent while the advective component is not (an some distances from the polarized interface). This can give rise to situations where an ionic species is brought into the system by advection but “stopped” somewhere inside due to a local increase in the electrical counterflow and strongly accumulates. Moreover, this turns out very sensitive to the species diffusion coefficient and charge, which can be a basis of a new separation process. Similar phenomena are already exploited in micro-analysis for species pre-concentration. However, the throughputs are extremely low. The corresponding membrane structures could make possible applications in downstream processing.