As the name implies, artificial muscle extrapolates nature's own highly successful designs. When charge accumulates on a parallel plate capacitor, the electrostatic force of attraction generated by opposite charges tries to draw the two plates together. Place an elastic dielectric between movable plates, and the capacitor becomes an electromechanical actuator. If the dielectric is an electroactive polymer (EAP) and the metal plates are flexible conductors, then the actuator becomes a highly flexible transducer, producing mechanical motion from electric current.

Researchers believe that the concept could revolutionize bioengineering and robotics. Current robotic systems, from welding arms in automotive factories to Honda's humanoid Asimo, are driven by conventional servomotor technology. Mechanical actuators get the job done, but at high cost and high complexity. And although motors can supply the required mechanical energy, they do not perform other functions, such as vibration damping, binding joints together or cushioning a limb from physical shocks.

The idea of building robotics using nature's superior methods is attractive from the point of view of design simplicity and might reduce the cost of such systems while increasing their flexibility and usefulness. And medical applications, such as better prosthetics or mechanical assistance for weakened hearts, would be a compelling draw for medical equipment companies. For now, the young discipline is rapidly acquiring a technological base, as advanced research, coupled with government and venture backing, propels the field into application development territory.

Artificial muscle has followed the curve typical of laboratory curiosities that percolate in the back rooms of industrial labs for years and then seem to burst on the scene in the form of startups, annual conferences and industry associations dedicated to their development. Several conferences are now devoted to the subject, artificial-muscle institutes are being established, and companies offering design services and products in the field are open for business.

One is Artificial Muscle Inc. (Menlo Park, Calif.), which was recently spun off from SRI International to develop products using electroactive polymers. It received initial venture capital funding of $2.5 million, with $5 million more scheduled for later this year. Ophthalmotronics Corp. (Albuquerque, N.M.) was founded in 2000 to develop artificial-muscle systems for correcting vision problems. It is now seeking FDA approval for its flagship product, the Smart Eye Band, which is surgically implanted around the surface of the eyeball.

Environmental Robotics Inc. (Albuquerque) grew out of early work at Sandia National Laboratories and now has an extensive patent portfolio. The company offers a design service along with custom artificial-muscle systems for bioengineering and robotics.

Get a grip
In 1999, to stimulate the development of artificial-muscle systems, EAP pioneer Yoseph Bar-Cohen at the Jet Propulsion Laboratory in Pasadena, Calif., proposed a grand challenge contest for robotic designers: Build a human-like arm that can win an arm-wrestling contest with a human. Three challengers are scheduled to take part in the first contest, which will take place in San Diego in March at the International Society for Optical Engineering's (SPIE's) Electroactive Polymer Actuator and Devices (EAPAD) conference. Stepping into the ring to defend humankind will be Panna Felsen, a San Diego high school student interested in pursuing a career in the field.

When Bar-Cohen issued his challenge five years ago, EAP systems had many limitations. Force development was low, the distance over which the force could be actuated was small and energy conversion was inefficient. But progress since then has been rapid.

"Some EAP types generate forces that can meet or exceed the human muscle's capability," Bar-Cohen said, though there are "various issues that still need to be addressed with regard to their engineering implementation."

Some toy applications are reaching the commercial stage, such as an aquarium containing robotic fish that was recently introduced by Eamex Corp. The artificial fish swim realistically around the tank, powered by polymer muscles. Longer term, building robotic limbs using EAPs could yield realistic prosthetics to help disabled people.

Artificial-muscle actuators have one advantage that makes them attractive to NASA: a very high force-to-weight ratio. Polymers are much lighter than metals, and as progress is made in developing and articulating forces, the actuators would be attractive for building lightweight planetary rovers. With rocket payloads, weight is an overriding design constraint. And the flexibility of polymer actuators could help realize scaled-down crawlers inspired by insects and arachnids.

Indeed, NASA engineers have developed a small rover inspired by spider locomotion that they hope won't be prone to getting stuck — a design problem that plagues wheel-based rovers like the Mars Spirit and Opportunity.

Doing nature one better
"Generally, imitating nature offers many advantages, since nature came up with numerous inventions that work and last," observed Bar-Cohen. "But sometimes, it is better to be inspired by nature rather than make an exact copy. Examples of copying include the use of honeycomb, Velcro, fins for diving and many others. However, copying the wings of birds as a means of flying did not work, and we as humans had to learn the principles of aerodynamics in order to be able to fly."

Once those principles were proved out in rudimentary aircraft, subsequent engineering improvements produced machines that far exceeded the capabilities of birds in terms of speed, distance and load capacity.

"Making artificial muscles at this point falls into the latter category of being inspired by nature rather than imitating nature," Bar-Cohen said.

Technological advances in materials engineering promise to extend the capabilities of artificial muscle. As more is learned on how to structure materials from the atomic level up, targeted characteristics, such as the ability to deform in response to an electric field, could be built into novel compounds. One example is a new class of electroactive materials built with carbon nanotubes.

Other possibilities are resulting from nanostructured composites. In a serendipitous discovery, Richard Claus, a physicist at Virginia Polytechnic Institute (Blacksburgh), has come up with a metallic conducting material that can snap back into its original shape after being stretched easily into any configuration. Called Metal Rubber, the material would be ideal as the electrode for EAP actuators, which have been limited by the inability of typical conductors to stretch.

"We set out as part of a research program to create a flexible conducting material, but in the beginning we did not know that it would be as good as it is. We were pleasantly surprised," said Claus. "Metal Rubber has high electrical conductivity that is maintained at large strain and over many cycles of strain. In our lab, we have demonstrated strains as high as 300 percent while maintaining electrical conductivity, something other materials with similar conductivities and mechanical properties cannot achieve."

Claus had been working on a general process for building materials from nanoclusters when he developed Metal Rubber simply as one possible material derived from the process. He has since founded a startup, NanoSonic Inc., to continue developing the pro-cess that yielded Metal Rubber, tapping the process as a basis for supplying materials with customized properties.

"By varying the types of molecules we use to form the freestanding material, we can change its constitutive properties," he said. "The idea of engineering multifunctional macroscopic materials that have properties determined in part by their nanostructure is exciting."

The unusual combination of properties might solve a big problem facing developers of artificial muscle. EAPs are one class of a general group of conducting polymers, so it would be natural to try to bond a conducting polymer to an EAP to create flexible electrodes. Though classed as conducting, however, these polymers are far below metal conductors in their ability to transport electrons. They also exhibit mechanical fragility, tending to break and separate.

The same problem afflicts metal-based approaches, such as a metal paste and sputtered metallic coatings, which tend to separate or peel off when the polymer flexes.

"In comparison, Metal Rubber has low modulus that can be engineered for specific designs," Claus said. "Low modulus means that as an electrode is attached to an EAP actuator element, large deformations of the element will not produce large constricting stresses due to the electrode and will not lead to large interfacial stresses at the element/electrode interface. The latter avoidance of interfacial stresses means that the electrodes will not fall off, which is good."

Even more encouraging is the possibility that the underlying process that produced metal rubber could produce a hybrid material that has the characteristics of an EAP in its interior while having a conductive skin.

"We have several proposals pending in this area and certainly think it is an excellent idea," Claus said.