The thesis describes the development of porous ion-transport oxide interconnects that allow molecular communication between microchannels in complex microfluidic architectures. New methods to control the permeability of interconnects to certain species by externally tuneable parameters was investigated. DC electric fields were used to impose a driving force for the transport of selected cationic and anionic species. Electric fields are preferred over pressure gradients in nanochannels, because very large pressure drops are required to drive flow in small channels, while typical operating voltages are below the potential difference required for decomposition of water. Applying an external electric field across the interconnects, a potential difference is created across the membrane, which makes it is possible to selectively drive charged species from one liquid into the other through interconnects, by means of ion migration, Fick diffusion and/or electroosmotic flow.
An important potential application field of switchable gates is in microfluidic devices. However, there are potential difficulties associated with the implementation and incorporation of the state of the art mesoporous oxide membranes into these mostly Si-based devices. It was therefore essential to develop a new class of Si-compatible thin oxide films, which combines mesoporous and microporous oxide film technology with existing silicon technology. The general fabrication method of such Si-compatible thin oxide films on micromachined silicon microsieve support structures is described. This procedure to manufacture different types of microporous and mesoporous oxide films enables the incorporation of sol-gel type fabrication techniques into existing silicon technology; opening the way for constructing Si-supported oxide films.
Si-compatible oxide interconnects with external platinum electrodes were used as interconnects for selective gating of anionic and cationic species. The electric field-driven transport of species through three different types of oxide interconnects was measured. The influence of the applied potential difference on the transport rates of ionic probe species was also investigated along with other factors that play an important part in the transport such as ionic strength and pH, which regulate the tuneability and selectivity of the system. In these cases where external electrodes were used, the main transport mechanism of ionic species was ion migration and not Fick diffusion. This transport mechanism draws ionic species through the membrane to the oppositely charged electrode at the permeate side of the membrane.
Integrating the Si-supported oxide films into smaller device type set-ups is a potential way to establish communication between two microfluidic channels and may be an important component in future microfluidic devices since it allows extension of microfluidic architectures into the third dimension. Two designs are presented in this thesis.
In order to explore new methods to control the transport rate of ionic species through a mesoporous oxide layer, as well as the ion-selectivity of the layer, transport studies were performed on conventional mesoporous oxide membrane, as a model system. In one study a bias potential was imposed at the feed side surface of the membrane, and a potential difference was generated over the membrane. State of the art a-alumina supported g-alumina membranes were modified by depositing thin porous gold films on either side of the membrane. These gold layers were utilized as electrodes to apply a dc potential difference over the membrane. It was found that the membrane is completely cation-selective and that the potential difference can be used to promote or inhibit the transport of cations through the membrane. The transport rate is influenced by a number of external parameters, which include ionic strength, pH, concentration and ion valence. By controlling the sign of the applied potential over the membrane it is possible to either promote or stop the permselective transport of cations through the membrane.
From these investigations it became clear that there are a number of ways to open and close switchable mesoporous gates for certain ionic species. These include pH, ionic strength (double layer overlap) and applied electrical potential difference. An alternative way to control the permeability of a gate is by the co-addition of molecules to the system that alter the properties of the membrane. The use of different surfactants to influence the transport of ionic species through a membrane by steric blocking is investigated. Surfactants are known to adsorb strongly on oxide surfaces, and adsorption in membrane mesopores reduces the porosity and permeability of the membrane. Near-complete blocking of ion transport was achieved by formation of surfactant bilayers on or near the membrane surface at relatively high surfactant concentrations, while increased ion fluxes were obtained at lower concentrations. The latter phenomenon is thought to be due to the adsorption of a sub-monolayer of surfactant molecules that neutralizes the membrane surface charges, thus decreasing the electrostatic interaction between the ionic probe species and the membrane pore wall.