Reining In a Chaotic Fluid
Ordinary fluids can flow when driven by pressure or gravity, but biological fluids, such as those inside cells, generate complex flows through internal sources of chemical energy. Flows of such “active fluids” could be extremely useful in robotics and other areas of engineering, but controlling them remains difficult. Now researchers have demonstrated a method of control that maintains a constant fluid speed despite changing conditions [1]. They hope that the approach can be used to stabilize active-matter flows in future technologies.
Life depends on biochemical processes that respond to many situations while maintaining fixed chemical conditions despite external and internal disruptions. Inspired by this impressive stability, researchers have been developing analogous artificial systems by assembling active fluids from key biochemical components akin to those inside cells. For example, they have created fluids that can generate their own bulk contractions or undergo spontaneous flows. Although these rudimentary designs mimic some features of living matter, researchers have so far failed to demonstrate techniques that keep properties such as fluid flow speeds stable over time.
“Our research is trying to understand how to control active matter—materials that, like living systems, generate their own motion by consuming energy,” says Seth Fraden, a specialist in soft matter at Brandeis University in Massachusetts. Spontaneous flows of active matter tend to be chaotic and difficult to harness for useful purposes. “To move from curiosity-driven science to technological function, we need some kind of control,” he says.
To test new control methods, he and colleagues experimented with a two-dimensional active fluid assembled from microtubules—long, cylindrical, hollow polymers made from the protein tubulin. The team trapped a dense collection of microtubules at the interface between layers of oil and water and added kinesin motor proteins that had been genetically engineered to be light activated. Shining light on the fluid activates these motors, causing them to bind pairs of parallel microtubules and drive one to move past the other, creating motion within the fluid. With light applied, the flow field is chaotic and highly time dependent, with a broad distribution of speeds as well as jets, swirls, and defect-driven bursts. In addition, the light level matters—the brighter the light, the faster the flow. When the light is turned off, the microtubules relax and disperse.
Using fluorescence microscopy, Fraden and colleagues could determine the average flow speed at any moment. They set up a control system to adjust the intensity of light in order to maintain a constant flow speed. With this feedback loop updating every 1–3 seconds, the researchers demonstrated that the system could hold the active fluid close to the target speed even when they disturbed the system. They warmed and cooled the sample to make the motors speed up and slow down. Without control, the flow sped up and collapsed unpredictably. But with the feedback system engaged, the measured speed remained stable. They also changed the target flow speed abruptly, forcing the fluid to jump between slow and fast states, and showed that the feedback system could make the fluid settle into its new behavior within seconds.
The researchers also tested the system’s response to another challenge: the gradual degradation of some of the biological components. Over time, motor proteins naturally clump together, and microtubules age, which means that the fluid’s natural response to light drifts. Even so, they found that their control system compensated automatically for this slow decay. Even after hours of aging, the flow speed remained constant.
“This paper shows that standard feedback methods for control work surprisingly well even in complicated settings,” says John Bechhoefer, an expert in control theory at Simon Fraser University in Canada. The system is messy, he says, because the flow depends on poorly known material parameters that also change with time because of aging. “One might expect that it would be hopeless to control a property such as the flow speed, but [Fraden and colleagues] show that it can be done using classic ideas of feedback, in much the same way that engineers control the speed of a motor under varying load,” Bechhoefer says. “This work reminds us that sometimes a very simple approach is useful.”
Fraden says that this work is part of a larger effort to embed real-time, adaptive decision-making into materials. “We envision using internal biochemical circuits—engineered genetic or chemical networks—to allow materials to sense, decide, and act autonomously,” he says.
–Mark Buchanan
Mark Buchanan is a freelance science writer who splits his time between Abergavenny, UK, and Notre Dame de Courson, France.
References
- K. Nishiyama et al., “Closed-loop control of active nematic flows,” Phys. Rev. X 15, 041053 (2025).