Access Type

Open Access Dissertation

Date of Award

January 2014

Degree Type


Degree Name



Electrical and Computer Engineering

First Advisor

Amar S. Basu


The emerging field of droplet microfluidics deals with the manipulation of nL-fL droplets encapsulated within an immiscible carrier phase. The droplets are used as reaction containers for biochemical assays, enabling drastic reduction in assay volumes needed for modern life sciences research. To achieve this, basic laboratory processes such as mixing, detection, and metering must be emulated in the droplet format. Three important unit operations relevant to high throughput screening include 1) the concentration of particles and species within droplets, which is necessary for heterogeneous assays; 2) sensing the biochemical contents of a droplet; and 3) the sorting of droplets based on physical or chemical properties, which is important for single cell and proteomic assays. Currently, particle concentration in droplets requires active components, such as on-chip electrodes or magnets, along with charged or magnetic particles. Similarly, sensing and sorting droplets by chemical composition is based on flow cytometry, which also requires on-chip electrodes, feedback control, and chemical labeling. It is desirable to avoid active field techniques due to complexity, size, and cost constraints, and replace them with more simple and passive techniques. In this thesis, we utilize microvortices, the rotational motion of fluid, to enhance the capabilities of droplet microfluidics in the above three areas. The microvortices are generated using two methods: (i) hydrodynamic recirculation drag and (ii) tensiophoresis.

In the first method, species concentration is accomplished by exploiting the shear-induced vortices that occur naturally inside a droplet/plug as it moves through a microchannel. Prior studies utilized these flows for enhancing mixing or interphase mass transfer. This work exploits microvortices together with two other independent phenomena--sedimentation of particles and interfacial adsorption of proteins--to concentrate both types of species at the rear of the droplet, where they can be extracted from the drop. In the latter case, the protein localization at the rear of drop reduces the interfacial tension locally resulting in an asymmetry in the drop shape. Under laminar flow, the shape deformation is deterministic and can serve as a sensitive, label-free indicator of protein concentration in proteomic screening.

In the second method, label-free sorting of droplets is accomplished by a novel droplet actuation technique termed Tensiophoresis. A microchemical gradient across the droplet is transduced into a microvortex flow which propels the droplets up the chemical gradient. Using laminar flow to precisely control the gradient, droplets can be sorted by size with 3.3% resolution over a wide turning range. Droplets can be also sorted based on chemical composition because tensiophoresis is inhibited by surface active agents adsorbed on the droplet surface. Studies conducted using Bovine Serum Albumin (BSA) show that the droplet migration velocity scales inversely with protein concentration in the droplet, and migration velocity can be correlated to protein concentration with a 1 femtomole limit of detection.

As modern life sciences research becomes increasingly reliant on high throughput workflows, microdroplet technology can meet the growing demand to perform screening at ultra-high throughputs with reduced sample volume. This thesis contributes three important unit operations which expand the capabilities of droplet-based workflows in proteomics, cell biology, and other areas of biomedical research.

Included in

Engineering Commons