Researchers at the University of Pennsylvania have created artificial muscle strands capable of tracking their own shortening and maintaining requested intermediate positions.
The fibers, named CoilLCE, integrate a liquid crystal elastomer with multilayer graphene powder and an ultrathin copper coil running along the fiber length. This composite translates electrical signals directly into muscle-like movement.
Containing just 0.25 wt.% graphene, the material reaches a toughness of 632 MJ/m³, more than triple the 192 MJ/m³ of a pure liquid crystal elastomer fiber. Rather than increasing, its stiffness decreases to a Young’s modulus of 0.1 to 0.3 MPa, aligning closer to skeletal muscle.
The research team attributes this behavior to graphene sheets functioning as physical barriers between oligomer chains, which lowers effective crosslink density without harming the liquid crystal alignment required for actuation. When loading increases to 0.75 wt.%, alignment is substantially impaired, causing actuation strain and breaking stress to decline.
In antagonistic pairs, the graphene-loaded fibers achieve an actuation strain of 16.1%, compared to 11.2% for pure liquid crystal elastomer. This represents an improvement of about 44%, alongside more uniform heating along the fiber.
An embedded copper coil measuring 50 µm in diameter acts as both a heater and a sensor. It Joule-heats the fiber to trigger contraction, while the linear variation of copper resistivity with temperature reports the fiber state without requiring a separate sensor.
“We used the resistance of the embedded copper coil as a feedback signal and adjusted the heating power in real time to keep the resistance near a target value,” said Xuelin Wu, a co-author on the study.
Using a PID controller to close the loop, the fibers hit strain targets within about 2%. Under a 13 g load, they maintained a 30% contraction for 30 minutes, and bounced back from blasts of hot air and liquid nitrogen in under five seconds.
“This is the first time that feedback control has been demonstrated on an LCE actuator, and it is a necessary step for these sorts of artificial muscles to be able to be used in future robotic applications,” said Sung.
Measuring 0.55 to 0.73 mm across, the fibers match small human muscle bundles. They deliver actuation strains above 40% at rates up to around 32% per second, featuring a work capacity of 343 J/kg with no measurable degradation across 116 heating and cooling cycles.
Demonstrations of the technology included 3D-printed robotic fingers playing piano keys with tunable force and duration. The fingers produced about 48 mN for a short note and 120 mN for a sustained note.
In a second demonstration, a pair of artificial eyes driven by eight fibers in four antagonistic pairs rotated up to 30°. The measured angles closely matched geometric predictions made by the team.
Researchers identify humanoid faces, which require dozens of small coordinated muscles to fit inside a confined space, as a primary target application for the technology. Funding for the project was provided by the Army Research Office’s MURI program, the National Science Foundation alongside its Graduate Research Fellowship Program, and the National GEM Consortium.
