The 1981 movie The Graduate ends with a remarkable scene in which Dustin Hoffman (the young college graduate Benjamin) is advised to continue his carrier with plastics. A quarter century later, Benjamin would not have regretted following this advice, especially after the exciting developments in plastic electronics.

Many people probably would agree that the transistor was the greatest technological invention of the twentieth century. The first transistor, invented by Jack Kilby and Robert Noyce in 1958, was made from silicon. Even though integrated circuit technology has advanced to the level of putting millions of transistors on a fingernail-sized chip, transistors are still made from silicon. Now, however, both the scientific community and the high-tech industry are very excited about something new: plastic electronics. For many of us, this might sound like an oxymoron, for we know plastic only as an insulator. So how can a protective substrate or a carriage box to the actual electronic device be converted into an electronic device?

The story of plastic electronics started late 1970s with Alan Heeger, Alan Macdiarmid, and Hideki Shirakawa. These scientists demonstrated that the molecular structure of certain polymers (plastics) can be manipulated and then used as conductors. The Swiss Academy of Sciences was somewhat slow in recognizing their work, for they were awarded the Nobel prize in chemistry only in 2000. Nevertheless, the scientific community did not wait for the Nobel committee's recognition of plastics. Since 1977, plastic has become probably the most common material in our daily lives.

Silicon versus plastic

All computer chips are made out of silicon (semiconductor) and aluminum (metal). Silicon is a great material for integrated circuits, because it is available, can be acquired in an extremely pure state (single crystal or amorphous), and has a very high mobility (the speed that electrons can travel through material). This mobility, in turn, determines the device's switching speed. However, silicon has one important drawback: It is not easy to process.

Integrated circuit technology can deposit millions of silicon transistors on a single chip. However, the procedure for making these devices usually requires facilities worth billions of dollars, for silicon has to go through complicated photolithography procedures under clean room conditions before it can be incorporated into a device. But is it really worthwhile to spend billions of dollars on such facilities? The answer probably looks obvious, since Intel remains one of the world's largest companies. However, we do not really need such a high quality in many of the applications for which silicon is used. So if there is something cheaper that can do the job perfectly, why not use it? Plastic is far cheaper, but not so sophisticated an alternative.

Plastic is cheap because billion-dollar facilities are not required to convert it into a device. In fact, the technology needed to process plastic into an electronic device is only slightly more advanced than an ink-jet printer. Electronic circuits are printed on an insulating polymer, and then certain parts of the polymer are exposed to UV light in order to convert the insulating polymer into a conducting polymer. The result is a device in which only the parts that we want to conduct are conducting. Moreover, these conducting parts sit on a protective insulating sheet of plastic. That is pretty much all we need for many applications.

The case for plastic

Plastic has two advantages over silicon: price and flexibility. Like other inorganic elements, silicon has strong covalent bonds between its atoms. As these bonds are rigid, they cannot bend or stretch. Plastic has very loose molecular bonds that can tolerate a significant amount of bending and stretching. Flexibility combined with electronics implies applications like flexible displays that can be rolled up and taken somewhere else, reloadable electronic newspapers that can be bent like paper, disposable mobile phones, and many others.

So if plastic is that good, why is it not the electronics industry's standard material? For one simple reason: Plastic's loose molecular bonds, which make the material so flexible, make it more difficult for the electrons to travel through it. Thus, plastic devices are slower than silicon devices. Until several years ago, the mobility of a typical conducting plastic used to be around 0.1 cm2/volts, whereas crystalline (best) silicon could reach 1000 cm2/volts at room temperature. Recently, a new class of polymers (pentacene) has been found in which molecules tend to self-organize. As a result, the mobility has been pushed up to 3 cm2/volts. Scientist working on pentacene estimate a number close to 50 cm2/volts as the limit of achievable mobility for this special polymer.

These expectations are not just wishful thinking of some optimistic scientists. In fact, even now there are some significant outcomes of plastic technology.

Current developments

John Rogers and coworkers from Bell Labs (Lucent Technologies) have patterned 256 polymer transistors on the back-plane of a flexible optical display. Richard Friend and coworkers from University of Cambridge have produced thin film transistor circuits using a high-resolution inkjet printing. Dago de Leeuw and colleagues at Philips Research Laboratories in Eindhoven, The Netherlands, have developed a new technique called photochemical patterning. In photochemical patterning, a light sensitive-polymer is exposed to ultraviolet light through a mask shaped in the form of the desired circuit. The ultraviolet light changes the polymer from a conducting state to a non-conducting state. In this process, the polymer's resistance can increase as much as 11 orders of magnitude (100000000000). The advantage of this technique over the patterning techniques used for silicon is that it does not need any vacuum and can be used on flexible substrates. The problems with photochemical patterning are that it is not significantly cheaper than photolithography and etching used for silicon, and it can be used only for light-sensitive polymers. Different research groups have developed various techniques that have pros and cons compared to photochemical patterning. However, many of the techniques cannot print features that are small enough for electronic circuits. The critical length is the distance between the transistor's source and drain, typically 0.01 mm. This is the distance that the field-induced charges have to travel. As the drive current and switching speed of the device depend on this distance, having too large of a distance reduces the capabilities of the device. Another technique that pursues quite a different approach is microcontact printing. Developed by the Bell Labs group, microcontact printing with rubber-like stamps can make small enough features. Scientists have used this technique to make a flexible display in which a transistor controls each pixel. The key point in developing this technique was using gold pads, instead of a polymeric material, to deposit the transistor's source and drain. Even though this device is not completely plastic, it is a step toward that goal. When a special kind of ink was applied to a thin film of gold, it formed some sort of self-assembled layer on the gold, which then produced well-defined patterns and sharp edges. This provided the required resolution to make small enough features necessary for an electronic device having a reasonable speed. These two techniques show two important aspects of the problem. In photochemical printing, we have a device that is completely plastic and so has the important advantage of flexibility. However, it is not as cheap as it could be and does not have the required resolution. In the second technique, the outcome is not a 100 percent plastic device, so it is not as flexible as a purely plastic circuit. However, it can be manufactured very cheaply and has a better resolution (and thus a higher switching speed).

Conclusion

Many other approaches are being employed to develop this new and exciting technology. If plastic electronics does become standard for at least some applications, it probably will be a hybrid of these different techniques. If the optimistic group of scientists working on plastic electronics prove to be right, one day we might see TV screens curling around the walls of our rooms and even reloadable electronic newspapers that can be folded and carried like regular newspapers. Who knows what new inventions will come with this new technology?

References

• Garnier, F., (et al). Science 265 (1994): 1684-86.
• Gelinck, G., T. Geuns, and D. de Leuw. Applied Physics Letters 77 (2000): 406-8.
• Levi, Barbara G. 'New Printing Technologies Raise Hopes for Cheap Plastic Electronics.' Physics Today (February 2001). Online at: www.physicstoday.org/pt/vol-54/iss-2/p20.html.
• Nobel Focus: Electricity through Plastic.' Physical Review Focus (24 October 2000). Online at: http://focus.aps.org/v6/st18.html.
• Scott, Campbell. 'Electronics Put It on Plastics.' Physics in Action. (October 1998). Online at: www.physicsweb.org/article/world/11/30/3/1.
• Voss, David. 'Cheap and Cheerful Circuits.' Nature 407 (28 September 2000).