The best way to build muscle

22 January 2020



A wide variety of polymers are used as biomaterials. Biostable polymers are commonly used for medical device applications because of their resilience, especially within the human body. Although there are a number of these on the market, efforts to improve their physicochemical and biological properties are ongoing. Mehmet Kanik, a postdoctoral fellow at MIT’s Simon Centre for the Social Brain, speaks to Lynette Eyb about the use of thermoplastic polymers in artificial muscles.


Polymers have become a mainstay of medical innovation, their range and scope offering manufacturers a plethora of options in the development of new devices. It is an exciting and fast-paced sector that is producing cutting-edge devices for use across a range of healthcare fields. At the forefront of the development of polymers in medicine is the prosthetics and artificial muscle research field. Here, the lines are increasingly being blurred with robotics to create artificial devices that aim to mimic the performance and durability of the human body.

The versatility of thermoplastic polymers in particular is leading to exciting developments. Soft, easy to mould and highly biocompatible, thermoplastic polymers are excellent candidates for use in prosthetics and artificial muscles.

One research project at MIT is focused on the development of thermoplastic polymer-based artificial muscles for biological applications.

The team – brought together by Polina Anikeeva, assistant professor in materials science and Engineering at MIT’s Department of Materials Science and Engineering – is multidisciplinary, with input from different fields including: the materials science, physics, mechanical engineering and electrical engineering.

While the focus of the researchers has been on developing artificial muscles, their inspiration has come from the natural world. To be more precise, they have looked to the cucumber plant.

These plants produce tightly coiled tendrils that work to support the plant as it grows, positioning itself to maximise sunlight.

The MIT team observed that if they could create an artificial fibre that could mimic the plant’s ‘coiling and pulling’ mechanism, then this artificial fibre could also act in a similar way to the human muscle. The artificial fibre could then be applied to prosthetic limbs and other biomedical devices, as well as being used in robotics.

Assemble the team

Mehmet Kanik, a postdoctoral fellow at MIT’s Simon Centre for the Social Brain, and MIT graduate student Sirma Orguc, working in collaboration with professors Polina Anikeeva, Yoel Fink, Anantha Chandrakasan, C Cem Tasan and five others, used a polymer-processing method called thermal size reduction (TSR) to bond two different polymers together, creating a bilayer or sandwich of the two materials.

The first was a highly stretchable elastomer with low thermal-expansion coefficient (cyclic olefin copolymer elastomer, COCE). The second was a nonstretchable polymer with a higher thermal-expansion coefficient (polyethylene, PE).

When combined, the two polymers produced a new fibre that could be stretched several times its original length. “The elastomer stretches but the other polymer undergoes permanent plastic deformation,” says Kanik. “Once we release the tension, the elastomer wants to contract, while the other polymer cannot, which results in the fibre curling into a spring, which is our fibre-based muscle.”

The fibre therefore acts in the same way as the tendrils of cucumber plants do when they coil. The fibre coils tighter when a small amount of heat is applied but retracts again after the temperature decreases – acting as a thermoresponsive artificial muscle. This movement – combined with the durability and strength of the fibre – suggests that it has the properties required to perform similar functions to a human muscle.

Their research paper, ‘Strainprogrammable fibre-based artificial muscle’, has been published in Science journal.

“We have been working on this project since 2016,” says Kanik. “Our platform can create artificial muscles from any thermoplastic and elastomer combination with distinct thermal expansion coefficient. Important features of this artificial muscle are its simplicity, fast responsiveness and low actuation energy requirement for large actuation strain. It’s lightweight but has a high weightlifting capacity of around 650 times its own weight. Its fast response time is similar to biological muscles.”

With every fibre

Fibre-based muscles offer significant advantages over other artificial muscle systems, including servo motors, hydraulic systems and polymers based on stimuli. MIT’s artificial muscle is more lightweight and more responsive than any of these existing systems. But there are other benefits to using polymer-based technology in this way.

“The type of coil formation we have developed eliminates the motorised coiling process used in traditional fibre muscles,” says Kanik. “We can also add coils to our fibre, free from tethering or twisting processes, and we can programme the fibre with applied strain and deformation speed. Unlike conventional artificial muscles, fibre-based muscles do not require rotational constraints while operating. Last but not least, our devices can demonstrate large strains at low temperatures, leading to very low actuation energy consumption with respect to conventional artificial muscles.”

Kanik describes current artificial muscle research as “only the beginning of the new robotics-based age” for medical device development. He predicts there will be rapid development in textile-based robotics and artificial muscles in the next 10 years. Larger-scale applications with wireless and autonomous control are also set for development, though more extensive work will be required before artificial muscles can be used in daily robotic applications.

Even so, Kanik believes artificial muscles will one day form a cornerstone of human mobility. However, for this to happen, they need to be developed to a level that can compete commercially with servo motors and other common actuators. The signs are positive – fibre-based artificial muscles are produced via thermal size-reduction techniques that are cost-effective. This means fibre muscles can be developed for around $0.01/m – meaning the technology is ripe for industrial scale-up.

“One of the key applications could be weaving artificial muscle fibre in textiles for use as a wearable physical therapy instrument for patients with partial stroke.”

“The field of fibre-based artificial muscles is new, and we haven’t finished our research on many different polymers yet,” he says. “However, the possibilities for materials of this type are virtually limitless because almost any combination of two materials with different thermal expansion rates could work, leaving a vast realm of possible combinations to explore.

“We have shown proof of principle fibre design and basic application. We have demonstrated that by bundling fibres we can increase the weight-lift capacity. In short, fibre muscles are ready for the next stage of development to create larger-scale prosthetics and soft robotics devices.”

He describes TSR as a convenient fabrication technique in comparison with the other fibre production methods that limit multi-material in-fibre device manufacturing.

“We’ve developed a framework for a new type of device in this study, and it can be applied to a variety of materials, but extreme functionalities can also be added to our device by introducing new materials and systems for implantable sensors and actuator applications. These kinds of unique fibre devices with extended lengths and a tiny footprint can only be obtained using the TSR method and TSR-compatible materials.”

For man and machine

The MIT team plans to use its artificial muscles for robotic and prosthetic applications, with the implantation of the muscles into human beings the long-term ambition. There is also wider scope for innovation. “One of the key applications could be weaving artificial muscle fibre in textiles for use as a wearable physical therapy instrument for patients with partial stroke,” says Kanik.

Further research and more development of the artificial muscle technology is required to move innovations like these forward. Although fibrebased muscles are state of the art, they are still no match for the human muscle.

“There is no artificial muscle yet that surpasses the functionalities and the performance of the human muscle,” Kanik says.

However, this natural-artificial divide will close with the engineering of new materials stimulated using different energy sources.

“For example, we plan to develop several different artificial muscles that can be stimulated from different actuation sources, such as light, electric and chemical,” explains Kanik. “Fibre-based fabrication is applicable to many different materials with versatile properties.”

There is enormous scope for developing devices with functional features that mimic human muscle performance. “We are looking forward to carrying these concepts further and developing thermally, chemically and electrically responsive closed-loop systems for soft robotic applications.”


Nature as a biologically inspired model for artificial muscles

Muscles are drivers of human and animal movement; they are the biological elastic actuator. A glance at the complexity and variety of the generated movements shows that muscles are a versatile, powerful and flexible actuator, capable of working in different modes modulating the dynamics of contraction and structural implementation. The development of artificial muscles is a challenging topic for a widespread number of applications, and in particular, in biomedical science for the construction of orthotics and prosthetics for rehabilitation purposes, as well as for minimally invasive surgical and diagnostic tools, and in robotics science for developing human-assistance devices and walking machines. One of the key aspects of making biologically inspired robots is the development of actuators that allow the emulation of the behaviour and performance of real muscles. A muscle-like technology would be an enormous benefit for medical implants.

Muscles are driven by a complex mechanism and are capable of lifting large loads with short response time (milliseconds). The operation of muscles depends on the chemically driven reversible hydrogen-bonding between two polymers – actin and myosin. A peak stress of 150–300kPa is developed at a strain of about 25%, while the maximum power output is 150–225W/kg. The average power is about 50W/kg, with an energy density of 20–70J/kg that decreases as speed increases.

Although muscles produce linear forces, all motions at joints are rotary. Therefore, the strength of an animal is not just a muscle force, but a muscle force modified by the mechanical advantage of the joint, which usually varies with the joint rotation. Finally, muscles are able to operate for billions of cycles over a period of a hundred years or more, due to the ability to regenerate proteins in situ. The desirable properties of the biological muscles are, in summary, that they are energy efficient, have a high contraction ratio, are intrinsically compliant, their stiffness can be varied smoothly and dynamically, and they also incorporate a sensory part.

Source: The Journal of Applied Biomaterials & Functional Materials

A range of plastic polymers.
A miniature arm with a bundled muscle.
Mehmet Kanik holding a 1m long muscle fibre.


Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.