Enabling Electroosmotic Flow for Chemical Retinal Stimulation

The objective of this dissertation is to develop strategies to electroosmotically actuate glutamate in chemical retinal prosthesis for restoring vision in photoreceptor degenerative diseases. Electroosmosis is a widely used propulsion mechanism in microfluidic devices and offers various advantages such as no moving components such as valves, diaphragms and plungers, low power consumption and ease of microfabrication. The first requirement for an electroosmotic flow (EOF) microfluidic device is to demonstrate glutamate actuation. This challenge is addressed in chapter 2, where electroosmotic actuation of glutamate is demonstrated in glass and SU8 microchannels. First, the experimental system is validated and corroborated with previous studies by using the current monitoring method to find the zeta potential between glass/ potassium chloride (KCl) interface. Zeta potential is the interfacial charge between the microchannel substrate and the solution. After corroboration, the solution and material are replaced with glutamate and SU8. It was found that the zeta potential for glass/glutamate and SU8/glutamate is -70 mV and -35 mV when 1 mM glutamate is used. The sign indicates that both glass and SU8 acquire a net negative charge and attract cations when filled with glutamate. These cations migrate when faced with an electric field (from positive to negative electrode) resulting in EOF of glutamate in the direction of the electric field. After establishing the feasibility of glutamate EOF, a prototype microfluidic device was fabricated to investigate the effect of glutamate EOF injections in a photoreceptor degenerated rat retinae. In chapter 3, a comparison between EOF glutamate injections and previously studied pneumatic glutamate injections is provided. It is observed that the EOF glutamate injections into the retina in-vitro generated a similar neural response characteristic comparable to pneumatic injections. These response characteristics consisted of glutamate evoke spike rate response width, amplitude and latency recorded from the retinal ganglion cells (RGCs) of the rat retina. Since neuromodulation of a retina using a prototype EOF device is feasible, the next objective was to miniature the EOF device further to close the gap between in-vitro and in-vivo experiments. This was achieved by using thin (tens of micrometers) SU8 microstructures coated with polymer electrodes. This resulted in a freestanding EOF micropump with a microchannel of 50 µm diameter and an integrated electrode. However, it was observed that quantifying such a micropump for EOF flow rate is non-trivial. Therefore, a novel flow quantification technique using microdroplets was developed. This technique is called as connected droplet shape analysis (CDSA). CDSA consists of two droplets connected by an EOF microchannel. By changing the EOF in the microchannel, the droplets’ shape can be changed which enables flow rate detection. The theory, numerical simulation and experimental findings are presented in chapter 4. It was observed that using this technique, SU8/glutamate zeta potential and nL/s flow rate can be determined simultaneously. This technique is further extended to the field of optics, wherein, the connected droplets act as a convex lens. By changing the shape of the droplets, one can change the focal length of the lens. The final chapter 5 presents a preliminary investigation of the effects of miniaturization and biocompatible electrodes on an EOF device for chemical retinal stimulation. The miniaturization requirements necessitate the inclusion of passive conduits in series with the EOF micropump. These conduits appear in the form of integrated electrodes, isolation and a microneedle layer. The microneedle layer is required to deliver glutamate deeper inside the retina tissue. These geometric modifications essentially reduce the fluid flow rate and conventional theory cannot predict the correct EOF rate. Furthermore, using biocompatible polarizable electrodes for voltage application in the EOF device gives rise to counterion screening, effectively reducing the electric field (and resulting EOF rate). Therefore, experiments are conducted to demonstrate the effect of geometry and electrode-electrolyte interaction, and numerical simulations are conducted to couple the two phenomena. In conclusion, this dissertation seeks to close the gap between in-vitro and in-vivo chemical retinal stimulation using electroosmosis. By systematically 1) establishing the feasibility of glutamate EOF, 2) neuromodulating rat retina using electroosmosis, and 3) developing flow quantification techniques for microfabricated EOF devices, a foundation is laid for future implantable EOF devices.