Archimedes is said to have had his eureka moment while taking a bath. Henning Sirringhaus鈥檚 inspiration came in the shower.
As water droplets rained down on him, the University of Cambridge physicist came up with the idea that by dripping beads of conducting liquid polymer onto a surface, they could be made to slide off each other and align themselves a fraction of a micrometre apart. This simple trick solved a problem that Sirringhaus, in his role as chief scientist at Cambridge company Plastic Logic, had been wrestling with: how to build plastic transistors small enough and fast enough to use in electronic devices.
Plastic Logic is just one of a number of companies developing polymer-based or 鈥渙rganic鈥� electronic devices. Building transistors from polymer instead of silicon will allow today鈥檚 rigid chips to be replaced by flexible plastic ones. This will open the way to cheaper, lighter and more flexible electronic tags and sensors 鈥� and, first off, robust flexible displays for electronic books and newspapers.
Advertisement
Plastic chips promise a future in which everything from supermarket packaging to credit cards could carry electronic labels displaying messages and advice such as recipes, health warnings and credit limits. Plastic processors and switches could be embedded in furniture fabric to replace the remote controls that now litter our homes. Delicate clothes could contain chips that warn a smart washing machine not to use the spin cycle, or gloves could warn your cellphone that you are about to leave them on the bus, says Hermann Hauser of venture capital company Amadeus Capital Partners, which has invested in Plastic Logic. 鈥淧eople often think of electronic circuits doing very complex things, but very simple plastic ones, in which each device is reporting wirelessly where it is and what it is, over simple mesh networks, will be of very great benefit,鈥� Hauser says.
Devices like these may still be years away, but a small monochrome display screen based on plastic electronics is due to be launched later this year by the Netherlands-based company Polymer Vision, a spin-off from Philips. This 12-centimetre screen will be flexible enough to be stored rolled-up, and then unfurled when required from the side of its host gadget. The company will not reveal what kind of device the screen will be used in, other than that it will be an 鈥渋nfotainment appliance鈥�. 鈥淚t鈥檚 going to be aimed at people like commuters, people who travel a lot, giving them a large screen from a small device,鈥� says Thomas van der Zijden of Polymer Vision. One guess is that it could include a GPS reciever.
The early launch is being made possible by Polymer Vision鈥檚 use of lithographic manufacturing techniques similar to the tried and tested methods used for fabricating silicon chips. The polymer layers that build up to form the transistors that drive the display are deposited on a 50-micrometre-thick polyethylene naphthalate base layer or 鈥渟ubstrate鈥� using a sequence of stencil-like masks. The drawback of this approach is that it is hard to keep the flexible material aligned with each successive mask, which is why the screen has to be so small.
One of the big potential advantages of plastic electronics is that it should eventually be possible to make polymer-based 鈥渃hips鈥� extending over several square metres 鈥� far larger than a dense silicon circuit鈥檚 limit of a few square centimetres. This should make it possible to build cheap, large displays and solar panels from plastic. To avoid the need for a mask, the polymer layers can be deposited by spraying droplets from an adapted ink-jet printer, and this is the approach being adopted by Plastic Logic for its circuits.
Printing a plastic transistor involves depositing two droplets of conducting polymer as close together as possible 鈥� ideally less than 1 micrometre apart 鈥� and then connecting them with a strip of semiconducting plastic called a channel. The droplets act as the source and the drain terminals of a field-effect transistor. A voltage applied to a 鈥済ate鈥� terminal, a strip of conducting material on top of the channel, switches current flowing between the source and the drain on and off. Faster switching translates into faster data processing.
That鈥檚 the principle, but in practice this arrangement is hard to achieve. The problem is that normal ink-jet printing is not precise enough because no two squirts from an ink-jet head follow precisely the same trajectory, Sirringhaus says. 鈥淒oing this directly, by printing one droplet and then another next to it, generates short circuits because the ink-jet print head cannot position the droplets accurately enough to ensure there is always a small gap between them.鈥�
To ensure that there were no short circuits, Plastic Logic initially had to space the droplets about 100 micrometres apart. This meant the channels had to be far longer than the ideal, and led to appallingly slow transistor switching times of 1 to 10 hertz. To produce transistors with more useful switching speeds of tens of kilohertz and higher, much shorter channels were needed.
This is where Sirringhaus鈥檚 shower-time brainwave comes in. First he adds a surfactant called benzalkonium chloride to an aqueous solution of the conducting polymer, and prints a single droplet of it onto the substrate. As the polymer dries, the surfactant molecule migrates to the surface, making it water-repellent. When he then squirts a droplet of conducting polymer without surfactant on top of the first one, it slides to one side and dries on the substrate, aligning itself just 60 nanometres away. The surfactant makes the surface behave like Teflon does with water, Sirringhaus says. 鈥淭he surface of the first structure becomes repulsive to the second.鈥�
The first transistors that Plastic Logic built this way switch at around 50 kilohertz, and the team is now aiming to push this up to 1 megahertz by refining the surfactant it uses. The technology is promising enough for the company to announce this month that it will be investing $100 million in a factory in Dresden, Germany, that will make flexible, A4-sized electronic paper for displaying the pages of electronic books, magazines and newspapers. 鈥淚t鈥檚 the first factory to make commercial quantities of flexible displays made from genuine plastic electronics,鈥� says Jim Tully of information-technology analyst Gartner, based in Egham, Surrey, UK. 鈥淭his is a very significant move.鈥�
Existing e-book readers such as the Sony Reader and iRex iLiad are paperback-sized devices with rigid screens. They are built from a layer of electrically activated pixels containing black and white powders, beneath which is a matrix of pixel-switching transistors deposited on a glass surface. The glass layer makes them fragile and heavy. 鈥淯sing our plastic back planes, we think we can make a screen three times larger for the same weight as these first-generation e-books,鈥� says Simon Jones of Plastic Logic.
The devices will have a flexible back plane made of polyethylene terephthalate 鈥� the plastic many soft-drinks bottles are made of 鈥� peppered with plastic transistors developed by Sirringhaus and his team. They will be A4-sized, shaped rather like a rounded folder with the electronics stored at one end, tapering off to a flexible display around 5 millimetres thick at the other. The devices will be on the market by the end of 2008, Jones says.
Liquavista of Eindhoven, the Netherlands, is aiming to bring colour to plastic displays. At the moment, the screens that e-book readers use are monochrome-only, and refresh too slowly to display video. Liquavista uses electric fields to move droplets of coloured oils held in pixel-sized cells, displaying images at video speeds (New 杏吧原创, 27 September 2003, p 16). The technology, which will this year make its first appearance in wristwatches, can be used with rigid and flexible back planes, and plastic transistors.
Jones expects plastic displays of all shapes and sizes to become available. How plastic electronics will fare against silicon remains to be seen. 鈥淭here is a lot of hype about plastic taking over the world but there is no way it can really compete with silicon,鈥� Tully says. Hauser is more bullish. He points out that since plastic semiconductors appeared in the 1990s, their mobility 鈥� the measure of how well they transport electrons 鈥� has improved by a factor of 10,000, and is still increasing rapidly. 鈥淚n 20 years there really doesn鈥檛 look like anything they won鈥檛 be able to do,鈥� he says.
鈥淚n 20 years it looks like there鈥檒l be nothing polymer electronics won鈥檛 be able to do鈥�
