John Rogers, professor of both chemistry and materials science and engineering at the University of Illinois at Urbana-Champaign, US, has been at the forefront of flexible electronics for almost 20 years. As a consequence, his research has mirrored the progression of the field as a whole.
‘Over the years, we’ve looked at pretty much all of the options for flexible electronics,’ he says. ‘In the early days, we were really focused on organic molecular systems such as thin-cast polymers as the semiconductors. We shifted our emphasis somewhat to carbon nanotubes in the early 2000s and now almost all our efforts take a hybrid approach.’
Pathway to success
The field of flexible electronics was really kick-started by the development of electrically-conducting polymers in the 1990s. Conducting polymers, such as polyaniline, seemed to promise everything from bendy electronic devices that can withstand being sat on to medical implants that mould themselves around the contours of the body. This is because polymers can be processed in solution, potentially allowing them to be printed onto almost any substrate. So rather than fabricating electrical circuits by depositing metal-based semiconductors and connectors onto a rigid silicon wafer, as happens at the moment, they could simply be printed onto a flexible material, such as rubber.
The problem with conducting polymers, however, is that despite much research effort they still don’t make as effective conductors and semiconductors as conventional metal-based materials, limiting their applications. This led to efforts to improve their electrical properties, as well as to induce electrical conductivity in normally non-conducting polymers, by incorporating tiny conducting particles, such as carbon nanotubes (CNTs) or metal nanoparticles, within them. Unfortunately, these modified polymers tend to lose their conductivity as they are repeatedly stretched.
Over the past few years the focus has shifted to hybrid systems that try to take advantage of the differing strengths of organic and inorganic materials. ‘We feel that a lot of advantages accrue from using high performance inorganic materials, like those that are already well-developed for conventional electronics, heterogeneously integrated with rubber or plastic substrates,’ says Rogers. ‘We’re trying to leverage the best of both worlds of materials for devices that not only afford mechanical flexibility but can also support very high performance functionality by using the materials for what they’re naturally good at.’
Several recent advances have allowed this shift to take place. One is the continued shrinking of electrical circuits and components, which are now so small that their rigidity is no longer an issue for many applications – a flexible substrate can still bend and stretch when they are attached to it. This is especially the case for components made from thin films, which can now be fabricated using low-temperature processes that don’t damage flexible substrates.
The other advance is in geometry rather than technology. Even though tiny circuits and components can now be fabricated on flexible substrates, they still need to be linked up with each other using metal connects, which quickly tear and break when stretched.
In 2008, a team led by Rogers came up with an elegantly simple solution to this problem by replacing the conventional straight paths taken by the connects with wavy, serpentine-like paths. These paths can withstand much more stretching before breaking than the previous ones, as a result of the extra length provided by their shape. ‘It’s like an accordion bellows or a spooled-up telephone cord, you can stretch it back and forth because you’re exploiting geometry changes,’ explains Rogers.
The wavy paths can be deposited directly onto a relaxed substrate using a mask or can be produced by fabricating a straight path on a stretched substrate, with the path then buckling to form a wavy configuration as the substrate relaxes. These meandering paths are now central to Rogers’ work in flexible electronics, although he has since gone on to experiment with more complex designs that can withstand even greater stretching. ‘That becomes the foundational science for how you put these systems together,’ he says. ‘We’ve found that ideas in fractal geometry actually lead to important insights into integrating hard filaments with soft supports.’
Together with colleagues, Rogers demonstrated the potential of this hybrid approach in 2015 by developing a flexible electronic system comprising tiny but commercially-available computer chips, sensors and power supplies embedded in a flexible silicone polymer1. The hard electronic components are attached to a bottom layer of silicone and then connected with wavy lines made of copper. They are covered in an upper layer of the same polymer and the space between the two layers is filled with a dielectric fluid that supports the rigid components, protecting them and allowing them to move relative to each other as the silicone bends.
As well as demonstrating the potential of the hybrid approach, this flexible electronic system also had a practical application. It was designed as a sensor to be attached to the skin for monitoring various physiological parameters, including heartbeat. Such biomedical applications are the main focus of Rogers’ work on flexible electronics. ‘We’ve really shifted our emphasis to devices that interface with the human body, either as implants or wearables or advanced surgical tools,’ says Rogers. ‘In that space, mechanical flexibility becomes not only a convenience, as it would in a flexible display, but it’s really indispensable in terms of the kind of integration you need to do in order to collect high quality data or perform therapy.’
Exploiting wavy lines
The hybrid approach is also leading to flexible electronics with unprecedented physical properties. By adapting the wavy lines concept, a team led by Ray Baughman, professor of chemistry at the University of Texas at Dallas, US, has produced conducting polymer fibres that are incredibly stretchy2.
This work takes advantage of Baughman’s expertise in producing CNTs – the electrical conductivity is provided by coating the polymer fibres with sheets of multi-walled CNTs. ‘[This work] was enabled by our previous invention of means to make highly oriented, flexible, electrically-conducting CNT sheets that are lighter than air and stronger pound-per-pound than steel,’ Baughman explains.
Baughman and his team wrap these sheets around the polymer fibre while it is stretched to its full extent. When the fibre relaxes, the CNT sheets buckle both horizontally and vertically while still being able to conduct electricity, and they retain the vast majority of this conductivity as the polymer is repeatedly stretched. Baughman and his team found they could stretch these CNT-coated polymers to over 10 times their original length for thousands of cycles, during which the conductivity fell by less than 5%.
According to Baughman, such extra-stretchy conducting polymers could find a wide range of uses. ‘Demonstrated patent-pending applications range from super-stretchable electronic interconnects and artificial muscles to giant-stroke strain gauges and mechanically robust cables for pacemakers,’ says Baughman.
The wavy lines concept can even work in two dimensions, as a team led by Zhifeng Ren, professor of physics at the University of Houston, US, has now shown. ‘We wanted to make a stretchable, transparent and conducting film for applications where they are needed,’ says Ren. ‘It has been known that one-dimensional serpentines are stretchable, so we tried to make a two-dimensional spring that is made of nanoscale metal serpentines and is transparent and conducting. We chose gold because gold is highly conducting and is stable under ambient conditions.’
Since there was no easy way to make such a two-dimensional spring, Ren and his team had to come up with their own technique. Termed ‘grain boundary lithography’, it involves exposing an indium-based film around one inch square in size to nitric acid, which etches a rough series of channels less than 100nm wide into the film, like the cracks in dried mud. Next, they fill these channels with gold and then remove the indium-based film to leave behind a two-dimensional gold nanomesh.
When loosely attached to a stretchy polymer such as polydimethylsiloxane (PDMS), this nanomesh is able to deform as the polymer is stretched. Ren and his team showed that the nanomesh could be stretched to over twice its original length for 50,000 cycles without any damage or loss of conductivity3.
Again, this conducting material could find many uses, although Ren is also focusing on biomedical applications. ‘We are trying to apply the electrode to stretchable solar cells or stretchable OLEDs [organic light emitting diodes],’ he says. ‘But the most desired application for our gold nanomesh is to be used as an implantable electrode because the gold nanomesh can match the native tissues in mechanical properties and exhibit good biocompatibility as well.’
The potential of the latest thin-film fabrication techniques for flexible electronics was recently demonstrated by Japanese scientists led by Kenji Hata, director of the CNT-Application Research Center at the National Institute of Advanced Industrial Science and Technology in Tokyo. By depositing a series of thin-film transistors organised into logic gates onto a silicone polymer, they were able to produce an electric circuit that had no solid components, making it highly flexible and robust4.
Single-walled CNTs formed the basis for most of the components making up each transistor. The CNTs were either printed directly onto the silicone or mixed with a liquid rubber and then sprayed onto a substrate, where the components were lithographically patterned before being transferred to the silicone polymer.
The lack of any solid components meant that the resultant electric circuit could be stretched, bent and twisted. It could also withstand a whole host of impacts without suffering any damage, including being run over by a car, stepped on by high-heeled shoes, hit with a hammer and washed in a washing machine. With this level of robustness, which had never been reported before, Hata envisages these flexible circuits being incorporated into clothing, where they would regularly be exposed to just this kind of stretching, bending, twisting and washing.
The ability to fabricate thin-film transistors at low temperatures means the hybrid approach can take advantage of other types of substrates besides polymers, producing flexible electronics with additional useful properties. For example, a team led by Zhenqiang Ma, professor in engineering at the University of Wisconsin–Madison, US, has made thin-film transistors from conventional gallium arsenide semiconductors on cellulose nanofibril paper5.
Although several groups have deposited organic semiconductors onto paper, this represents one of the first times anyone has tried doing it with conventional semiconductors, although Ma and his team did use a special kind of paper that is transparent and highly flexible. The process involved initially fabricating the transistors on a rigid gallium substrate and then transferring them to the paper with a polymer stamp.
Using this technique, Ma and his team were able to fabricate the kind of microwave circuits used for wireless communication in mobile phones, which require the superior electrical properties of gallium arsenide semiconductors. By using cellulose nanofibril paper derived from wood as the substrate, Ma ensured that not only were these circuits flexible but also biodegradable. Devices made from such flexible circuits would be much easier to dispose of at the end of their life than conventional electrical devices.
Indeed, thin-film transistors can now be fabricated on almost any flexible substrate, as a team led by Kilwon Cho, professor of chemical engineering at Pohang University of Science and Technology in Korea, has shown by putting them on sticky tape. This opens up the possibility of producing low-cost flexible devices, such as radiofrequency identification tags, that can easily be attached to almost anything.
Cho and his team made the transistors by depositing thin films of aluminium, gold and graphene through a mask onto sticky tape attached to a silicon wafer. They could then simply peel the tape off and stick it to whatever they wanted, including a banknote, a pen and someone’s hand6.
All in all, by moving on from conducting polymers, this new hybrid approach is allowing flexible electronics to become a lot more flexible.
1 Science, 2014, 344, 70
2 Science, 2015, 349, 400
3 Proceedings of the National Academy of Sciences, 2015, 112, 12332
4 Nano Letters, 2015, 15, 5716
5 Nature Communications, 2015, 6, 7170
6 Scientific Reports, 2015, 5, 12575
Jon Evans is a science writer based in Chichester, UK