Skip to main content
eScholarship
Open Access Publications from the University of California

UC Irvine

UC Irvine Electronic Theses and Dissertations bannerUC Irvine

A Programmable Droplet Microfluidic Liquid Handling Platform

  • Author(s): Werner, Erik Morgan
  • Advisor(s): Hui, Elliot E
  • et al.
No data is associated with this publication.
Abstract

Droplet microfluidics offer a variety of advantages over traditional robotic liquid handling and in recent years have enabled many new research and commercial applications. While existing droplet devices excel at generating millions of droplets with well-defined volumes and compositions, generating large numbers of droplets with unique contents remains a challenge. In addition, droplet microfluidic devices are typically built for a single purpose, in contrast to the arbitrary and programmable liquid handling ability of multi-well plate-based robotic liquid handlers.

We demonstrate arrays of droplet-on-demand dispensers capable of generating droplets with a diverse variety of contents from an array of wells spaced to match standard multi-well plates. Integrated pneumatic microfluidic logic circuits are used to decode binary pneumatic inputs, allowing large arrays to be controlled from a small number of input signals. We show 64 independently addressable droplet dispensers, each drawing from a separate reagent reservoir, controlled by only 6 address signals. Dispensers produce droplets at up to 10 Hz and metered volumes from 50 nL to 600 nL. Droplets remain spatially indexed, eliminating the need for barcoding, and are tracked on-chip by machine vision, allowing the sequential injection of reagents from multiple dispensers into a single droplet or splitting of droplets on demand. Hardware and software systems were developed to enable automated control of droplet-on-demand arrays and allow programable liquid handling operations on droplets.

Programmable droplet liquid handling allows large and diverse collections of droplets to be generated that would otherwise require significant time and cost to produce by robotic liquid handling. Moreover, droplets can be generated just-in-time for online machine learning applications. In future work, droplet-on-demand arrays may be a valuable tool in applications that include drug combination screening, studies into complex biological signaling systems, and personalized medicine.

Main Content

This item is under embargo until June 4, 2025.