Imagine a tiny donut shaped robot that swims through thick fluids simply by basking in a beam of light. It needs no batteries, no motors, and no wires. Just constant illumination to keep it moving, twisting, and navigating through space with surprising agility. Researchers have now built exactly that: a light driven torus made from soft liquid crystal elastomer that achieves self sustained motion and, critically, can be steered in any direction by changing where the light shines.

This breakthrough addresses one of the toughest challenges in soft robotics. Getting materials to move continuously when driven far from equilibrium is one thing. Controlling where they go once they start moving is another. Most autonomous soft robots follow preprogrammed paths dictated by their shape or material properties, features locked in during fabrication. Dynamic steerability, the ability to change direction on demand, has remained elusive. The new torus overcomes this limitation by exploiting friction and drag forces that vary depending on which side of the structure absorbs more light. The result is a millimeter scale swimmer that operates in the Stokes regime, where viscosity dominates and Reynolds numbers drop to around 0.0001, a realm where most propulsion strategies fail.

The Challenge of Steering Self Sustained Motion

Living systems thrive by operating out of equilibrium. Cells pump ions, muscles contract, hearts beat. All require constant energy input and all involve feedback loops that regulate their own activity. Replicating this in synthetic materials means creating systems that can sustain motion under steady energy feed, not just twitch once and stop.

Recent advances in active matter have produced soft actuators that oscillate, crawl, or rotate when exposed to light, heat, or chemical gradients. Many rely on a principle known as zero elastic energy modes, or ZEEMs. These are mechanical configurations where a structure can deform continuously with little or no energy cost, thanks to built in mechanical frustration. A classic example is a prestrained torus: when heated or illuminated, the strain mismatch between different parts of the ring creates a torque that flips the torus inside out, over and over, as long as the stimulus persists.

The problem is control. If the motion direction is baked into the geometry or material alignment, the robot goes where physics dictates, not where you want. For applications ranging from targeted drug delivery to environmental sensing, a soft robot needs to adapt its path in real time. Achieving this requires not just self sustained motion but a mechanism to steer it.

Building a Torus That Responds to Light

The researchers fabricated their torus from liquid crystal elastomer, a rubbery material whose molecular chains align in ordered patterns. By controlling the alignment during fabrication, they could program how the material expands or contracts when heated. Stretching the elastomer in one direction during polymerization produces a fiber that shrinks along its length when warmed. Twisting it during curing creates a fiber that expands instead. This tunability is key: depending on whether the thermal expansion coefficient is positive or negative, the torus will evert or invert when illuminated.

To make the material light responsive, the team soaked the fibers in a solution containing Disperse Red 1, a dye that absorbs green light and converts it to heat. Under a 532 nanometer laser, the illuminated side of the torus heats up, softens, and deforms. The temperature difference between the lit and shadowed portions drives continuous rotation. Infrared imaging shows the surface temperature oscillating in a sawtooth pattern, peaking around 80 degrees Celsius above ambient, then dropping as the hot segment rotates into shadow and a cooler segment rotates into the light. This cycle repeats indefinitely as long as the beam stays on.

The rotation itself arises from the interplay between static prestrain, locked into the ring when the fiber is looped, and dynamic strain induced by the thermal gradient. When the driving torque exceeds internal losses from viscosity and friction, the torus enters a state of spontaneous, self sustained rotation. Theoretical modeling matches the experimental data closely, confirming that the system operates as a true ZEEM, with rotation speed scaling predictably with light intensity and fiber radius.

Steering on Land and in Liquids

Rotation alone does not produce locomotion. To convert spin into translation, the researchers exploited asymmetric forces. On a solid surface, oblique illumination heats one side of the torus more than the other. The warmer side softens, increasing its friction coefficient by roughly threefold. During inversion, friction forces on the two sides differ, producing a net force that pushes the torus in a predictable direction. Shine the light from the left, and the torus crawls left. Shift the beam, and it changes course.

The same principle works when the torus hangs on a thin thread or climbs the inside of a glass tube filled with glycerol. In each case, the contact point experiencing the highest photothermal heating becomes the anchor, and the rest of the structure rotates around it. The motion is not fast, typically fractions of a millimeter per second, but it is steady and fully steerable.

The most striking demonstration comes in a viscous fluid. Submerged in non polymerized polydimethylsiloxane, the torus swims by generating flow around itself. When the ring everts, the outer surface moves in one direction and the inner surface in the opposite direction. Because the outer surface has a larger area, it experiences a stronger viscous drag. The imbalance creates a net force pointing toward the light source. By adjusting the angle and position of the beam, the researchers could make the torus swim up, down, left, or right.

This is significant because swimming at low Reynolds numbers, where inertia is negligible and viscosity dominates, is notoriously difficult. The scallop theorem, formulated by physicist Edward Purcell in 1977, states that reciprocal motions, like opening and closing a shell, produce no net displacement in such regimes. You need a motion that breaks time reversal symmetry. The toroidal swimmer does exactly that: its continuous, unidirectional rotation coupled with asymmetric drag forces allows it to generate thrust even when Reynolds numbers are four orders of magnitude below one.

Three Dimensional Navigation

To demonstrate full spatial control, the team performed a series of maneuvers. First, they made the torus swim horizontally for 15 millimeters along the x axis. Then they optically reoriented it by shifting the light beam, causing the torus to tilt until it faced a new direction. The torus then swam vertically along the z axis, paused for another reorientation, and finally traveled along the y axis. The entire trajectory traced a three dimensional path through the fluid, controlled entirely by adjusting the light.

The reorientation process exploits another consequence of photothermal heating: viscosity changes. When one side of the torus absorbs more light, the local viscosity of the surrounding fluid drops. The uneven drag distribution causes the torus to tilt until the light intensity becomes uniform across its surface. This self regulating behavior happens at a rate of about seven degrees per second, fast enough to redirect the swimmer without waiting for the fluid to settle.

The relationship between rotation and translation is predictable. Plotting displacement against rotational angle yields a straight line, matching theoretical predictions derived from hydrodynamic models of Stokes flow. The constant of proportionality depends on the slenderness of the torus, the ratio of fiber radius to ring radius, and a sliding coefficient that quantifies how much the fluid slips along the surface. These parameters can be tuned by adjusting the fabrication process, offering a design space for optimizing performance.

Why It Matters

This work represents more than a clever demonstration. It shows that prestrained topological structures, driven out of equilibrium by external energy, can achieve agile locomotion with on demand steerability. The key insight is that friction and drag, often seen as obstacles, become tools for control when spatially modulated by a moving light field.

The implications extend beyond swimming. Terrestrial versions of the torus can navigate sand, cloth, or curved surfaces. Confined environments, like the inside of a tube or along a suspended thread, are accessible. The same physical principles apply across scales and media, suggesting that similar designs could operate in biological fluids, chemical reactors, or microfluidic channels.

The approach also sidesteps many limitations of conventional microswimmers. Magnetic or acoustic swimmers require external fields that penetrate the entire workspace. Chemical swimmers depend on fuel gradients that are hard to control. Light offers spatial precision, easy modulation, and no material contamination. It can be focused, scanned, and switched on or off instantly.

From a fundamental perspective, the work validates decades old theoretical predictions. Purcell imagined a toroidal swimmer in his 1977 lecture but lacked the materials to build one. Subsequent models explored the hydrodynamics of rotating tori, predicting that continuous rotation combined with limited fluid slip could enable efficient propulsion. The experiments confirm these ideas and reveal new behaviors, like the self reversal of locomotion direction on a tapered pipette, where the transition from friction dominated to drag dominated motion occurs as the geometry changes.

Beyond the Torus

The torus is just one example of a broader class of structures that support ZEEMs. Möbius strips, crumpled sheets, knotted loops, and helical coils can all be driven into self sustained motion under the right conditions. Each geometry offers different mechanical modes and different possibilities for encoding functionality. The choice of material is equally flexible: hydrogels, shape memory polymers, and piezoelectric films have all been explored. Liquid crystal elastomers stand out for their large, reversible strains and programmable alignment, but the principles generalize.

Future versions could incorporate sensing, enabling the robot to detect chemical gradients or temperature changes and adjust its path autonomously. Multiple tori could interact, forming swarms that coordinate through shared light fields or hydrodynamic coupling. Scaling down to tens of micrometers might allow operation inside living tissue or microfluidic chips, though maintaining photothermal efficiency at smaller sizes will require careful optimization.

The work also opens questions about efficiency. The torus converts less than one percent of absorbed light into mechanical work, the rest dissipates as heat. Improving this ratio will require better thermal insulation, faster response times, or hybrid designs that combine photothermal actuation with other transduction mechanisms. Understanding the fundamental limits of ZEEM based locomotion, and how they compare to biological motors or other artificial swimmers, remains an active area of research.

Conclusion

A light powered torus that swims, climbs, and navigates in three dimensions may seem like a novelty, but it demonstrates principles that could reshape how we think about autonomous soft matter. By leveraging topological prestrains, photothermal heating, and asymmetric forces, the system achieves self sustained motion and dynamic steerability without electronics, batteries, or complex control algorithms. It operates in regimes where traditional propulsion fails, offering a glimpse of what becomes possible when materials are designed not just to respond to stimuli but to sustain activity far from equilibrium.

The path from lab demonstration to practical application is long, but the foundation is solid. Soft robots that move by light, adapt to their surroundings, and require no tether or fuel could find roles in medicine, environmental monitoring, and manufacturing. The toroidal swimmer shows that such devices are not only feasible but controllable, opening the door to a new generation of autonomous materials that move, sense, and act with life like fluidity.

Credit & Disclaimer: This article is a popular science summary written to make peer-reviewed research accessible to a broad audience. All scientific facts, findings, and conclusions presented here are drawn directly and accurately from the original research paper. Readers are strongly encouraged to consult the full research article for complete data, methodologies, and scientific detail. The article can be accessed through https://doi.org/10.1038/s41563-024-02026-4