Skip to main content

Taking Microfluidics to New Lengths

Posted on by Dr. Francis Collins

Fiber Microfluidics

Caption: Microfluidic fiber sorting a solution containing either live or dead cells. The type of cell being imaged and the real time voltage (30v) is displayed at bottom. It is easy to imagine how this could be used to sort a mixture of live and dead cells. Credit: Yuan et al., PNAS

Microfluidics—the manipulation of fluids on a microscopic scale— has made it possible to produce “lab-on-a-chip” devices that detect, for instance, the presence of Ebola virus in a single drop of blood. Now, researchers hope to apply the precision of microfluidics to a much broader range of biomedical problems. Their secret? Move the microlab from chips to fibers.

To do this, an NIH-funded team builds microscopic channels into individual synthetic polymer fibers reaching 525 feet, or nearly two football fields long! As shown in this video, the team has already used such fibers to sort live cells from dead ones about 100 times faster than current methods, relying only on natural differences in the cells’ electrical properties. With further design and development, the new, fiber-based systems hold great promise for, among other things, improving kidney dialysis and detecting metastatic cancer cells in a patient’s bloodstream.

The new work published recently in Proceedings of the National Academy of Sciences [1] is the result of an interesting collaboration between the labs of Joel Voldman and Yoel Fink, both at the Massachusetts Institute of Technology, Cambridge. Voldman’s electrical engineering lab has an interest in cell sorting. In fact, his team earlier developed a chip-based microfluidic device that sorts blood cells with sound waves [2]. Fink’s material sciences lab focuses on the design and fabrication of multi-functional fibers and their potential to become highly functional devices [3]. Fink envisions a new generation of high-tech fibers and fabrics that can sense their surroundings, communicate, and monitor health.

The idea for the two labs to pair up arose out of a “speedstorming” event at MIT. Speedstorming—a cross between modern speed dating and good, old-fashioned brainstorming—places students and postdocs in a room, where they sit down in pairs for a few minutes to discuss possible collaborations, and then rotate to the next colleague for more innovative, outside-the-box conversation.

For members of the Voldman and Fink labs, the speedstorming event led to an intriguing question: Why not develop a novel microfluidics approach for cell sorting that uses a new class of fibers? Rodger Yuan, a doctoral student in Fink’s lab and first author of the new study, took the idea and ran with it.

Yuan and colleagues created microfluidic fibers much like others devised in the Fink lab. They started with a preformed, millimeter-thick cylinder, made of the thermoplastic polymer polycarbonate, and fashioned it into a microscale fiber.

To do so, all that’s needed is to stretch out the cylinder lengthwise. The researchers start by heating the polymer, causing it to soften. They then slowly pull and constrict the polymer into a thin fiber. What’s quite remarkable about this process is that the internal features are uniformly preserved but microscaled during this dramatic lengthening. Yuan likens it to the way pulled taffy candy maintains a consistent internal design of colorful shapes and swirls. Ultimately, the sturdy preformed cylinder, just a few inches tall to start, is stretched to a fully flexible 525-foot fiber—1,600 times its original length!

These microfluidic fibers have several advantages over “lab-on-a-chip” devices. For one, the internal channels can be fashioned into any internal shape or configuration. By comparison, traditional microfluidic devices are generally limited to rectangular channels.

To demonstrate this feature, the researchers created fibers with internal crosses, stars, and bow ties. These cross-sectional designs aren’t just meant to be cute. The design flexibility promises to enable researchers to influence the internal flow within a microfluidic device in virtually limitless ways.

While chip-based microfluidic devices have tight space restrictions, microfluidic fibers can be stretched to incredible lengths. That will help to screen large samples of blood or other fluids for rare items, such as metastatic cancer cells, or potentially to filter waste out of the bloodstream of a person whose kidneys no longer work properly.

The researchers say they can integrate electrical components into the fibers in any location to manipulate cells directly. It’s possible to include elements for sensing, heating, inducing electrical fields, or sound waves.

As proof-of-principle, the researchers have already shown their fibers can efficiently sort living and dead cells based on their natural response to electrical fields. With the help of 3D-printed connector components, they could get those cells back out of the fibers once separated, as illustrated in the diagram. Such connectors also will make it possible to string together fibers with different characteristics and elements in any configuration one might dream up.

Fiber Microfluidics Device Schematic

Credit: Adapted from Yuan et al., PNAS

In its current form, the fibers sorted cells within a 100-microliter sample, or about two drops, in one minute. That’s already a huge improvement over current methods, which sort about one microliter per minute. But, for the clinical applications that the researchers envision, they’d like to speed things up to about 1 milliliter, or 20 drops, per minute.

While “lab-on-a-chip” devices will remain the go-to choice for many applications, the newfound design freedom of these microfluidic fibers will now allow scientists to fabricate truly novel devices. Once further developed, tested, and scaled, their uses will likely stretch well beyond our current capabilities in microfluidics.


[1] Microfluidics in structured multimaterial fibers. Yuan R, Lee J, Su HW, Levy E, Khudiyev T, Voldman J, Fink Y. Proc Natl Acad Sci U S A. 2018 Oct 29. [Epub ahead of print]

[2] Iso-acoustic focusing of cells for size-insensitive acousto-mechanical phenotyping. Augustsson P, Karlsen JT, Su HW, Bruus H, Voldman J. Nat Commun. 2016 May 16;7:11556.

[3] New institute will accelerate innovations in fibers and fabrics. MIT News. April 1, 2016.


Speedstorming (Massachusetts Institute of Technology, Cambridge)

Yoel Fink (MIT, Cambridge, MA)

Joel Voldman (MIT, Cambridge, MA)

NIH Support: National Institute of Biomedical Imaging and Bioengineering

Leave a Comment