Australasian Science: Australia's authority on science since 1938

Microdevices Muscle Up

Muscle fibre

Muscle fibre

By Geoff Spinks

Artificial muscles are evolving from laboratory curiosities to serious applications in surprisingly diverse areas, from cochlear implants to robotic fish.

Deafness, cancer and pollution would seem to have very little in common, but in fact can all be helped by shape-shifting materials: artificial muscles that change their shape or size when stimulated. These fascinating materials are being developed for a wide range of micro-devices, including the next generation cochlear implant, a wearable garment to massage away excess limb fluid after cancer treatments, and a robotic fish that can swim about looking for water pollution.

Conventional motors and engines cannot be used because they are too complicated to miniaturise. Instead, researchers have turned to much simpler muscle-like materials.

We are developing microscopic guidance systems that can help surgeons accurately position the cochlear implant close to the nerves in the inner ear to improve sound quality. Our “lymph sleeve” device will use soft and flexible fabrics that can contract and squeeze the arm to prevent the build-up of lymphatic fluid, which causes considerable discomfort for up to one-third of Australian women diagnosed with invasive breast cancer. And the chemical sensors in our untethered robotic fish called WANDA (Wireless Aquatic Navigator for Detection and Analysis) offer a more efficient solution to detecting and tracking the source of water pollution than networks of thousands of fixed sensor stations.

The essential common element in all these applications is size, or lack of it. When space is limited, a good motor is hard to find.

The situation is very different at the macro-scale with commonplace engines for cars, planes and boats, and electric motors for pumps, vacuums, conveyors, lifts and a myriad of other domestic and industrial machines.

Engines and motors exist to perform mechanical work by producing forces or movement. The power-to-weight ratio of a motor is a key characteristic, and it is well-known that bigger is better: giant diesel motors for ships or jet-turbine engines for aeroplanes deliver the most power per kilogram.

The reverse is also true, with power output per kilogram dropping dramatically as we reduce the motor’s size.

The exception is muscle: nature’s ubiquitous motor. The power-to-weight ratio of muscle can be 100–1000 times greater than any man-made motor of equivalent size.

Of course, muscle is not useful for machines, so instead we need “artificial muscles” – synthetic materials that mimic the performance of real muscle.

Shape Shifters
While there are lots of different types of artificial muscle materials – like shape memory metals and piezoelectric ceramics – we prefer to work with soft plastics that resemble natural muscle more closely. In particular, we are developing polymer artificial muscles that directly convert electrical energy to mechanical work. These materials are electronically conducting and include conducting
polymers, nanomaterials like carbon nanotubes and graphene.

We formulate the conducting material into a sheet or fibre and use it as one electrode in an electrochemical cell with another electrode and electrolyte, which is an ion-filled solution like salt water.

The system is much like a rechargeable battery except that we are interested in the volume change of the electrode(s) that occurs during charging and discharging. Charging the cell causes ions from the surrounding electrolyte to enter the polymer electrode, causing it to swell. Shrinkage occurs during discharge as the ions are ejected back to the electrolyte solution.

We can quite routinely produce a change in length of 5% and a force of 1 N/mm2. Skeletal muscle produces larger contractions (20% or more) but lower force (0.3 N/mm2).

Natural muscle has the advantage of being very fast, with full contraction in one-tenth of a second. Our materials typically take 10–100 times longer to generate full contraction, which is a problem that we are trying to solve.

In the early days of developing polymer artificial muscles we became excited when we were able to produce simple bending movements. Here we would laminate the conducting polymer with another flexible plastic material so that the expansion or contraction of the conducting polymer would generate an overall bending motion. These systems generate large bending movements even though the actual volume change and force generated is small. The “benders” were great for demonstration purposes, but applications were limited because they had to be immersed in a beaker of electrolyte.

Our first step toward real applications happened when we succeeded in producing benders that operated outside the beaker. Our “trilayer” actuators consist of a sandwich structure with an internal porous film that is soaked in the liquid electrolyte and has a coating of conducting polymer on both sides. We have effectively packaged the entire electrochemical cell into a flexible, multi-layer sheet. By applying a small DC voltage between the outer layers, we can make them expand or contract and bend. Reversing the voltage causes them to bend back in the other direction.

The trilayer system is exactly the configuration that we use for WANDA the fish and for the cochlear positioners. WANDA is propelled by a single flapping tail driven by bender actuators. Its swimming speed is currently quite slow compared with similar-sized real fish, but WANDA’s muscle mass is also much less than 1% of its total body weight.

We have plans to “beef up” WANDA with more muscle and to build a flexible body that can undulate like real fish. The undulatory motion is believed to be crucial to the high swimming efficiency of fish.

Our first robotic fish, NEMO (novel electromechanical oscillator), carried no payload, but WANDA is fitted with a video camera and transmitter so that images can be beamed in real time to a remote computer. We can use image detection software to interrogate “sensor stations” that change colour in response to environmental pollutants. WANDA can swim from station to station and send an alert if a particular colour change indicates localised pollution.

Recently we discovered a new type of actuator configuration that may also be useful for aquabots. For the first time we were able to produce a very large and fast rotating motion from an artificial muscle. With our collaborators we produced a yarn fibre consisting of twisted carbon nanotubes. The yarn diameter is only a few microns – less than one-tenth the diameter of human hair. When charged up electrochemically, these yarns start to untwist in a process that can be reversed when the yarn is discharged.

At 5V the rotation rate for a 12cm long yarn was nearly 600 revolutions per minute. The amount of rotation was 1000 times higher than previously described rotating actuator materials, and the power-to-weight ratio was similar to large-scale commercial electric motors.

The twisting muscle is reminiscent of the rotating flagella that propel micro-organisms like bacteria. This similarity made us think of the possibility of one day creating torsional muscles from carbon nanotubes to power a microscopic aquabot.

The “lymph sleeve” application is a new project that aims to provide a useful solution to the problem of lymphoedema – a build-up of a lymphatic fluids in the arms and legs that can lead to swelling, heaviness, pain and discomfort. One effective treatment is manual massage, but this requires the patient to visit a lymphatic massage therapist. Compression garments are commonly prescribed, but these tend to be difficult to put on, hot in summer, and can cause rashes and other skin problems.

Our dream is to produce a light-weight sleeve that incorporates bands of our artificial muscles along with fabric pressure sensors. As the bands contract, the pressure will assist in the drainage of lymphatic fluids. We also hope that relaxing the bands will make putting the sleeve on a much easier process.

This project is in its infancy, but already we have developed a small prototype consisting of a single contracting band that is able to pump fluid.

The Future of Artificial Muscles
So what is the future for artificial muscles? Our main focus is the development of improved materials, as we cannot yet match the performance of natural muscle in terms of movement, force, speed and efficiency.

The use of a chemical “fuel” to power the muscle is likely to improve efficiency, and we are now investigating gel-like materials that respond to changing chemical environment (like pH) with very large volume changes. The trick is to make them go fast, and we have developed tough gels that can be made into thin films that will respond more quickly. Another challenge is to develop methods to deliver the fuel and remove the waste products.

As we improve performance, more and more applications will emerge. The miniaturisation of electronics is a tremendous technological success story that has transformed our lives in ways that were simply unimaginable 50 years or so ago.

However, we have not been so successful at miniaturising mechanical devices. Nature is so much better at making small “machines” than we are – ants carry heavy loads and mosquitoes can fly. We have not been able to make tiny swimming, walking or flying robots the size of these insects. To do so we must continue the quest for artificial muscles that can match the performance of natural muscle and that are suitable for both scaling down and assembling into machine systems.

In our laboratories and others around the world, we are continuing to improve artificial muscles with nature’s own systems both our guide and inspiration. We know that there is no shortage of serious applications for these fascinating materials, and their potential use in medical devices and environmental protection is great motivation.

Geoff Spinks is conducting his research at the University of Wollongong’s Intelligent Polymer Research Institute.