Microfluidic Transistors for the Control of Liquids and Particles

Abstract

Microfluidics has played a significant role in the development of high-throughput and quantitative technologies for biology, medicine, genomics, and chemistry. Rapid growth in these fields have prompted a rising need in microfluidics to automatically control the movement of chemicals and biological samples with high levels of precision and complexity. In the field of electronics, complex automatic controllers for electrical signals were revolutionized by the invention of amplifying elements such as the electronic transistor. A microfluidic analogue to the transistor may similarly enhance the level of automatic control and processing of chemicals, droplets, and cells, in lab-on-a-chip systems. Prior works to address the problem of complex flow control in microfluidics resulted in the development of valves that could switch flows on or off, but did not exhibit the saturation phenomenon of the transistor and were not capable of proportional amplification- the defining function of the transistor. The absence of this amplification capability greatly limited the ability of these valve platforms to replicate electronic circuits and to create complex automatic fluid controllers for lab-on-a-chip processing applications. In this work, we exploit the fluid-mechanics phenomenon of flow-limitation to create a microfluidic element that is capable of proportional amplification of liquid signals and whose flow-pressure characteristics are strikingly analogous to the current-voltage characteristics of the field effect transistor. After fully characterizing this microfluidic transistor, we use the work of Shapiro to develop a preliminary model for its behavior and perform a scaling analysis. Next, we implement the microfluidic transistor in a set of building block circuits that demonstrate all three canonical transistor topologies from analog circuit design (common-source, common-gate, and common-drain), as well as digital logic and memory circuits. We then demonstrate how these building blocks can be combined and scaled to perform more complex signal processing operations. We also demonstrate the unique ability of microfluidic circuits to directly manipulate matter by creating a smart particle dispenser. This circuit senses individual particles suspended in fluid, autonomously performs fluidic signal processing, and dispenses these particles in a new stream in a programmable fashion without the use of electronics. Finally, we develop new fabrication techniques for manufacturing the next iteration of microfluidic transistor-based circuits that exhibit superior transistor performance and integrated circuit fabrication features. The microfluidic transistor developed in this work leverages the vast set of circuit designs, tools, and analysis techniques of electronics to enable uniquely complex fluidic signal processing and single-particle manipulation for the next generation of chemical, biological, and clinical platforms.