Which material will kick start the market for printed electronics?
For Professor Henning Sirringhaus of the University of Cambridge and founder of Plastic Logic, the aim of printed electronics is to get much cheaper electronics. But it is a slow process: "The vision is that, one day, we will be able to use graphic-art processes to print these devices and make circuits that meet real-world circuit requirements. But using these really low-performance materials with limited printed techniques really restricts the performance of fully printed transistor-based circuits."
Frustrated by the performance of organic semiconductors, some have gone back to silicon technology. LG Display opted to use thin-film transistors based on amorphous silicon in its A4 printed colour display. But even then, performance is constrained. A move to higher-performance polycrystalline TFTs was not possible because it demanded too high a temperature during production for the other materials to bear.
Sirringhaus claims that the quest for better performance has yielded improvements that mean companies do not have to go back to traditional silicon-based technologies.
"People have developed better materials. The conventional approach is to use a solution-processable polymer. And there are materials that can reach mobilities of somewhere between 0.1 and 1cm2/Vs," Sirringhaus adds, putting them in the same range as amorphous silicon. "But people have realised that solution processing does not necessarily mean polymers. People have attached soluble side chains to a crystalline core. You can solution-process those small molecules and get good uniformity."
Peter Harrop, chairman of specialist consultancy IDTechEx, argues that the future will not see a competition between organic and inorganic technologies: "We are going to see the distinction between organic and inorganic as completely broken down. Both will be used. There is no point in arguing about the semantics: is it organic or inorganic? The real question is: will it sell?"
Teams have been working on novel inorganic oxides that could work as better semiconductors than their organic counterparts. Sirringhaus says one problem with inorganic compounds is their reliance on high-temperature deposition processes. "We can get quite decent performance on the order of 1.5cm2/Vs. But it is hard to lower the temperature below 150°C. The question is: can we get to room temperature?"
At Oregon State University, Professor Douglas Keszler's team has found new types of inorganic oxide that can be laid onto a surface much more easily. "For the first time in oxide chemistry, we are producing vapour-quality films using completely new chemistries. We have developed some completely new ways to do direct patterning."
To get low-temperature films, Keszler explains, many researchers have resorted to using sol-gels but they tend to be porous, allowing in moisture that ultimately wrecks the transistors. The approach taken by the Oregon team is to use chemical reactions to have the oxide solidify out of solution as a continuous film. "With this new route, we try to control the chemistry on a smaller scale. And we have built a reasonable toolbox of chemistries right now. We can prepare these oxides with pretty good quality just putting them in a beaker and spin coating them in air," Keszler claims. With some films, the materials have reached mobilities of more than 25cm2/Vs. However, the processes still need temperatures of around 150°C.
"We think this is an enabling technology. We are now trying to push the temperature as low as we can," Keszler says.
Although the materials used for the transistor channel look to be obvious candidates for improvement work, researchers such as Sirringhaus are focusing on other parts of the transistor structure. It turns out that the gate dielectrics and contact materials are far from ideal and offer better prospects for improving speed in low-temperature printing processes.
Professor Daniel Frisbie of the University of Minnesota has been working on novel dielectrics that polarise when charged. That gives them a much higher dielectric constant than regular materials. The polarising process takes time, which slows down the switching speed of the transistor. But the dielectric constant is so high that it makes it possible to have the gate pushed to the side, instead of sitting directly on top of the channel. This could lead to very cheap printed-electronics circuits as it means entire circuits could be defined using just a couple of stages. This is the kind of process that will be useful in very
cheap circuits such as printed RFID tags.
Frisbie says the team has been working on improving device speed using the polarisable polymers. One device, he claims, can track a 10kHz. "The device at the beginning could not have done that, so it is a big improvement. We have now printed inverters that track at 1kHz. There is more work to be done to optimise speed, but it is promising," he notes.
How you make contact to the wiring in the printed circuit makes a difference. At Novaled, Ansgar Werner has been tuning the electron levels in organic materials to reduce the contact resistance between circuit elements such as the transistor source and drain and the conductive wires. By selectively doping the organic molecules, in a process analogous to the doping used in silicon transistors, it is possible to reduce contact resistance and improve the mobility of electrons through the interfaces between different parts of the circuit.
Jiro Kasahara, R&D director of Sony's Fusion Domain Laboratory, found that a silver-based organic ink turned in a surprisingly good performance when used to form a contact between the conducting wires and the pentacene used in printed transistors. "Somehow, using the silver ink, we get better contact resistance than just using the metals on their own," he says. The team is now working to understand why silver ink turned out to be much better than expected.
The result of all this work is that attention is now shifting away from the materials used in the early days of printed electronics to a much more exotic mix, all in the quest to make fast transistors on the cheapest process possible.