Visible to the public Ionic Liquid and Amorphous Metal-Oxide Semiconductor Interactions: Towards a New Programmable Neuromorphic Platform

This project will design and implement a domain-specific language and compiler for microfluidic laboratory-on-a-chip (LoC) devices based on electrowetting-on-dielectric (EWoD) technology. The Lead PI's team has designed and implemented BioScript, a domain-specific programming language for programmable microfluidics. The BioScript syntax is programmer friendly, with the intention of being accessible to biologists and other researchers and practitioners in the life sciences. BioScript is a CPS language, in the sense that it targets EWoD LoCs that include integrated sensors, which enables specification of biological assays that feature real-time feedback based on sensory data. BioScript includes a novel type system, based on union types, which interfaces with the EPA/NOAA CameoChemicals database. The type system ensures that biological programs do not mix substances that may be poisonous, harmful, or fatal, as classified by the CameoChemicals reaction groups. The PI's team has produced a static code generator which interfaces with a software-based EWoD simulator. Future work will port the language and compiler to target a real-world device.

The PI's team at the University of Tennessee has successfully fabricated an EWoD LoC based on active-matrix addressing, which provides independent control over a quadratic number of pixels using a linear number of control pins. The technology underlying the pixel array is indium gallium zinc oxide (IGZO) thin film transistors (TFTs). A critical bottleneck to widespread adoption of IGZO TFTs is the need to activate the materials at high temperatures. The PI's team was able to control activation of amorphous IGZO semiconductor channels using ionic liquid gating at room temperature, where activation is controlled by electric field-induced oxygen migration across the ionic liquid-semiconductor interface. Using a hydrated Ionic liquid (BMIM-TFSI), the PI's team demonstrated control over an amorphous metal oxide transistor threshold voltage and on-current via H+ injection, yielding "neuromorphic" behavior that was previously unanticipated.

Combining this behavior with a pixelated electrowetting array results in a programmable neuromorphic platform which can be scaled to high pixel counts. In the future, this platform could be used to model neuro-biological architectures using analog electronic signals. In the future, researchers could program this behavior using BioScript, compile it to run on the electrowetting platform, and observe experimental results.

License: 
Creative Commons 2.5
Switch to experimental viewer