Researchers have developed a 3-millimeter area on a chip that creates and directs microbubbles on demand using electrodes and ultrasound. Published on October 14, 2026, the technology eliminates the need for physical trapping, opening paths for targeted drug delivery and reconfigurable microfluidic systems.
Why are we trying to trap these bubbles?
asked Itai Cohen, the Josephson Family Professor of physics in the College of Arts and Sciences at Cornell University. His lab set out to answer that question by building a platform that makes bubbles exactly where and when they are needed. In a separate domain of microfluidics, researchers have also advanced fluid control using flexible microfibers. Other teams are exploring microrobotic cilia that integrate directly with complementary metal–oxide–semiconductor circuitry to sense temperature and actively pump fluid.
How the Nine-Electrode Chip Generates and Resonates Bubbles
Fabricated at the Cornell NanoScale Science and Technology Facility, the device uses a square array of nine electrodes fitted within a 3-millimeter area on a chip. Each electrode measures six microns in diameter. To start the process, an investigator applies a voltage that breaks apart a water molecule, leaving behind a single hydrogen gas bubble attached to the electrode tip.
Once the hydrogen bubbles are in place, the system introduces acoustic energy via ultrasound. The size of the bubble determines the resonant frequency of the bubble,
Cohen explained. When the ultrasound causes the bubble or multiple bubbles to resonate, the oscillating bubbles move the fluid around them. Zhiyuan Zhang, a postdoctoral fellow and first author of the study published in Science Advances on October 14, 2026, noted that the platform controls bubble size, numbers, and placement in real time to dictate fluid paths.
Programmable Fluid Flows and the Bubble Shooting Function
Different arrangements of active bubbles produce distinct hydrodynamic behaviors across the chip surface. For example, three bubbles in a horizontal row can create a flow from left to right. Another three bubbles in a diagonal line create a flow up to the left. Five bubbles arranged in an L-shape direct a flow in a curve. Researchers also placed flexible fibers within microfluidic channels—each channel measures roughly half the width of a human hair in depth—allowing physical components to store information and remember past flows.
Researchers can coordinate multiple elements to achieve complex routing. During testing, investigators also discovered a bubble shooting
capability. A higher voltage generates a cloud of bubbles in a rapid flow, which can then be propelled in a specific direction by turning on the ultrasound source. Abhineet Rajput, a graduate student in the lab of Amir Pahlavan, an assistant professor of mechanical engineering at Yale, stated: While conventional microfluidics relies on active elements, this system uses microfibers to passively store information.
Pahlavan added: We could design a fully fluidic controller that can be integrated into a soft robot and effectively act as its brain.
If, for some reason, one of the bubbles isn’t working, you just inflate it until it detaches. And it goes away and then you build another bubble.
Itai Cohen, Josephson Family Professor of Physics, Cornell University
Collaboration and Future Applications in Microfluidics
The study, titled Ultrasound-Driven Electronically Programmable Microbubbles
, was co-authored by Zhiyuan Zhang and Itai Cohen. Alyosha Molnar, the Ida and Charles Lee Professor of Engineering in the School of Electrical and Computer Engineering at Cornell, also contributed to the project alongside postdoctoral researcher Jinsong Zhang. Financial support for the work came from the National Science Foundation and the Army Research Office.
By turning the ephemerality of hydrogen bubbles into a regenerative feature, the platform avoids the mechanical failures that plague traditional microfluidic traps. Future steps for the team involve designing circuits that create the bubbles through this mechanism on robots, potentially enabling medical applications such as super-targeted drug delivery.