A small strip of rubbery material sits motionless on a lab bench at Lawrence Livermore National Laboratory. A researcher aims a laser at one end, and the strip curls, lifts, and begins to crawl forward with no wires, no battery, and no motor. It moves because the light told it to.
That deceptively simple demonstration, published in the peer-reviewed journal Matter, represents one of the most advanced examples yet of light-driven artificial muscles. The LLNL team 3D-printed soft actuators from liquid crystal elastomers, a class of polymer whose molecular chains snap into new arrangements when heated, and laced them with gold nanorods that absorb light and convert it into heat at precise locations within the material. The result: printed structures that bend, crawl, roll, and oscillate on command, all controlled by where and how a beam of light lands on their surface.
How light becomes motion
Liquid crystal elastomers, or LCEs, occupy a middle ground between rigid plastics and floppy rubber. Their polymer chains contain rod-shaped molecular units that can be locked into specific orientations during manufacturing. When heated past a transition temperature, those aligned molecules lose their order and the material contracts along the alignment direction, much like a muscle fiber shortening when activated.
The LLNL researchers exploited this property by using a 3D-printing technique called direct ink writing. As the printer lays down each line of LCE ink, the flow through the nozzle aligns the liquid crystal molecules along the print path. By varying the direction of those paths across a structure, the team programs exactly where and how the material will deform when triggered.
The gold nanorods serve as microscopic antennas for light. When illuminated, they absorb photons and release heat into the surrounding elastomer, raising the local temperature past the transition point. Because the heating is localized to wherever the light falls, the researchers can activate different regions of a single printed structure independently, producing coordinated movements like the alternating contractions of a crawling gait or the sequential curling needed for a rolling motion.
“We can essentially program the motion into the material during printing and then trigger it remotely with nothing more than a beam of light,” said Nikola Dudukovic, a materials scientist at LLNL and a lead author on the Matter study. The U.S. Department of Energy’s Office of Scientific and Technical Information cataloged the study, confirming LLNL authorship and federal contract oversight. The project sits within a broader DOE-supported portfolio of responsive-materials research at the laboratory.
Why it matters beyond the lab bench
Most artificial muscles today rely on pneumatic pressure, electrical voltage, or tethered heating elements, all of which require physical connections to a power source or control system. Light-driven actuation sidesteps that constraint entirely. In principle, a soft robot built from these materials could be steered by a lamp, a laser pointer, or even focused sunlight, with no onboard electronics.
LLNL’s institutional summary of the work frames the advance as a step toward untethered soft machines that respond to light the way living tissues respond to nerve signals. According to that summary, the lab’s researchers describe the actuators as a platform for future soft robotics, prosthetics, and disaster-response devices, scenarios where rigid motors and wiring bundles are liabilities rather than assets.
The idea is not as far-fetched as it sounds. Soft robotic grippers are already used in food handling and warehouse automation, and researchers worldwide are racing to build compliant devices that can navigate rubble, squeeze through tight spaces, or conform gently to human anatomy. A material that moves on its own when illuminated could simplify the design of such systems dramatically.
Independent work strengthens the foundation
The LLNL results do not exist in isolation. A separate study published in Science (doi:10.1126/science.adf1525) demonstrated that liquid crystal elastomers can be engineered for programmable, multistep, and bidirectional deformation by designing specific internal mesophase structures. In that work, a single piece of material followed a choreographed sequence of shape changes as it was heated and cooled, behavior that goes well beyond simple bending.
Researchers at Brookhaven National Laboratory used the National Synchrotron Light Source II beamline to corroborate those findings with high-resolution synchrotron X-ray scattering measurements, directly tracking how shifts in molecular orientation translate into macroscopic motion. Their data, reported in the same Science paper, provided independent, instrument-level validation of the physics underlying programmable LCE deformation.
Additional published work in ACS Applied Materials and Interfaces has shown that embedded 4D printing, a technique that builds alignment cues directly into complex free-form geometries, can produce LCE structures with time-dependent transformations triggered by external stimuli. And earlier foundational research published in Nature Communications (doi:10.1038/s41467-022-32374-z) established that molecular switches embedded in polymer networks allow light to tune not only shape but also mechanical stiffness, meaning future devices could potentially stiffen on demand to bear loads and then soften again to move.
A second Nature Communications paper (doi:10.1038/s41467-023-37936-z) mapped how three-dimensional programming of the internal “director field,” the orientation pattern of liquid crystal domains, encodes shape changes directly into the bulk material. Rather than relying on external hinges or joints, the molecular orientation itself dictates where and how the elastomer deforms. This concept underpins the LLNL team’s ability to design specific crawling gaits or rolling behaviors into a single printed structure simply by varying print paths.
What still needs to happen
For all the promise, significant gaps remain between a tabletop demonstration and a working device.
Durability is unproven. None of the published studies report long-term fatigue data. It is unclear whether gold nanorods stay evenly dispersed after thousands of heating cycles, whether the elastomer’s programmed molecular alignment drifts with repeated use, or how the material holds up against humidity, dust, or mechanical wear. For any real-world application, from a prosthetic hand to a search-and-rescue crawler, those numbers will be essential.
Engineering integration is barely sketched. A light-driven actuator still needs a light source, and in most practical scenarios that means a control system to aim and modulate the beam. Pairing these actuators with sensors, feedback loops, and structural components will require substantial engineering work that the current research does not address.
Cost and scale are unknown. Gold nanorods are not cheap, and direct ink writing is a slow, precision process. The published literature does not discuss manufacturing throughput, material costs, or strategies for scaling production beyond single laboratory prototypes.
Safety and regulation are uncharted. Liquid crystal elastomers and gold nanoparticles are well characterized in materials science, but their behavior under prolonged exposure in medical or environmental settings would require dedicated toxicology and reliability studies. No regulatory pathway has been outlined for devices built from these composites.
Funding details are thin. The DOE bibliographic record confirms federal oversight and LLNL affiliation, but project budgets, the structure of collaborations with external nanorod suppliers, and potential industrial partnerships remain undisclosed. Without that transparency, it is hard to gauge how quickly the research might advance toward technology transfer.
Where the evidence stands as of May 2026
The strongest evidence here is primary and experimental. The Matter paper provides direct laboratory demonstrations of light-triggered locomotion in printed elastomer structures, with detailed descriptions of materials, printing parameters, and actuation conditions. The Science study and Brookhaven synchrotron measurements offer independent confirmation that LCEs can be programmed for complex, multistep deformations. The Nature Communications reports supply the theoretical and mechanical framework connecting molecular orientation to macroscopic motion and tunable stiffness.
Taken together, these converging lines of research justify confidence in a core claim: 3D-printed liquid crystal elastomers doped with photothermal additives can move in controlled, predesigned ways when illuminated. That is no longer speculative. What remains speculative is everything that comes after: robust devices, affordable manufacturing, certified products, and soft machines operating outside the controlled calm of a research lab. The gap between those two realities is where the hardest work still lies.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
