Materials that move on their own
Most materials stay put unless a motor pushes them. A small group of polymers, gels and fibers behaves differently: they change shape when given an electrical signal, a chemical change or heat. Researchers call many of these electroactive polymers, and they are the basis for much of the work on artificial muscles that does not rely on air pressure or metal wires. This article is about how those materials came to light, which was gradual and often accidental. The wider story of the muscle-like devices built from them has its own history.
A nineteenth-century observation
An often-cited early observation comes from Wilhelm Röntgen, better known for X-rays. In 1880 he reported that a strip of rubber changed length when it was electrically charged. The effect was small and, at the time, a curiosity. It is now understood in terms of electrostatic forces squeezing the soft material, the same principle that later underlies dielectric elastomers.
For many decades after that, little followed. The tools needed to make and measure new polymers did not yet exist, and few people had a practical reason to build a muscle from plastic.
Chemistry-driven contraction
In the late 1940s, chemists including Aharon Katchalsky in Israel showed that fibers of collagen, the main protein in tendon and skin, would lengthen or shorten when immersed in solutions of different acidity. The change came from chemistry rather than electricity, and it was slow, but it showed that a non-living material could shorten and relax in a controlled way, much as muscle does. Researchers later made synthetic polymer gels that swell and shrink in response to acidity, temperature or an electric field. Gels can undergo large volume changes, but they are generally slow and weak, since the movement depends on fluid moving in and out.
Conductive and ionic polymers
The 1970s brought a major advance in materials: polymers that conduct electricity. The work on polyacetylene recognized by the 2000 Nobel Prize in Chemistry opened a class of plastics whose properties could be changed by applying a voltage. Some conductive polymers, such as polypyrrole, were found to expand and contract as ions moved in and out of them, and by the 1990s researchers were building simple bending actuators from them.
Related ionic polymer-metal composites were developed in Japan and elsewhere. These are thin polymer membranes coated with metal electrodes that bend when a low voltage is applied while wet. They bend noticeably at low voltage, but usually produce modest force and need moisture.
Dielectric elastomers
A different mechanism developed in the 1990s. A dielectric elastomer is a soft rubbery film between two flexible electrodes. When a high voltage is applied, the electrodes attract each other and squeeze the film, which then spreads sideways. Researchers at SRI International, including Ron Pelrine and Roy Kornbluh, reported in 2000 that carefully designed films could stretch to very large degrees. This work drew attention because it appeared to rival muscle on some measures, notably strain.
The drawbacks are a requirement for high voltage, which raises safety and electronics issues, and concerns about durability and consistent behavior.
The 1999 arm-wrestling challenge
In 1999 the researcher Yoseph Bar-Cohen, then at NASA’s Jet Propulsion Laboratory, proposed a public challenge: build a robotic arm using electroactive polymer muscle that could win an arm-wrestling contest against a human. A first demonstration took place at a conference in 2005, where a human participant easily beat the entrants. The goal was partly to bring attention and funding to the field. Its outcome showed how far the materials still were from natural muscle in force and reliability, and it is a useful reminder that laboratory demonstrations of strain or bending do not necessarily translate into strong, durable actuators.
Carbon nanotubes and coiled fibers
In the late 2000s a group including Ray Baughman at the University of Texas at Dallas reported yarns spun from carbon nanotubes that contracted or twisted when stimulated electrically or chemically. In 2014 the same group published a result that surprised many: ordinary polymer fibers, such as nylon sewing thread or fishing line, could be twisted until they coiled, and then shortened strongly when heated. Because nylon is cheap and widely available, the discovery drew wide interest. Nylon had been among the first mass-produced synthetic fibers, and here it turned out to have an unexpected second use. As with other thermal actuators, the main difficulties are the speed of cooling and the efficiency of heating.
What the discoveries have in common
Nearly all of these findings began as basic science observations: a rubber strip, a collagen fiber, a plastic that conducts. Each opened a way to make a material that moves, but each also had a limitation, such as high voltage, slow response, a need for moisture or poor efficiency. For that reason researchers rarely speak of a single “artificial muscle”. They speak of a toolbox, and engineers choose from it according to the job.