Your most cited paper is your 1990 Nature paper on light emitting diodes—J.H. Burroughes, et al., "Light-emitting diodes based on conjugated polymers," (347[6293]:539-41, 11 October 1990). Could you describe the thinking that led up to that paper?

Well, in the 1980s, we set out deliberately to see whether we might be able to make structures, similar to what was done with inorganic semiconductors such as silicon. We started looking originally for field-effect transistors. We thought they would be relatively easy to make and it turns out they are. They work quite well, but there’s not a huge amount of interest in them. But then we were able to interest colleagues in the chemistry department at Cambridge, particularly Andrew Holmes, and the polymer we got through that collaboration was the famous PPV, p-polyphenylene vinylene. Originally we were trying to use that polymer to make transistors but we had difficulties getting it to behave. We then tested it as an insulator, seeing how much electric field it could withstand. We wanted to sandwich it between two electrodes and use it as an insulator in these field-effect transistors. We were seeing how much voltage we could put across it and we saw a light emerging through this structure, actually through one of the electrodes, which was thin enough to be partially transparent. That was in February 1989, and that was the beginning of the polymer light-emitting diode.

  Why did you wait more then two years to publish?

We did race off and file a patent and then we worked out why it was working. It wasn’t really obvious. Filing the patent first was probably a good idea, since the LED created a huge amount of interest when we finally published it in Nature in the fall of 1990.

  How would you describe the evolution of the field in the years since then?

It’s been a bit of a roller coaster. Obviously there’s been a lot of competition from other groups and a lot of new chemistry. The science has turned out to be far more interesting than people had thought it was going to be. The performance of these diodes is now spectacularly good, much better than we had any reason to hope when we started. That’s why there’s this huge amount of interest in the first paper. Rather miraculous that the very first structure turned out, with different materials and some tweaks, to be exactly how devices are made today and the right way to do it.

  Why is there such enormous interest in organic polymer LEDs?

The reason why is also the example par excellence of why organics are potentially so important. You can solution-process or paint or spin-coat them over a large area. You can use whatever technique you want. The diodes work well. The semiconductors are not particularly bothered by the disorder inherently present in such structures, disorder that would normally destroy semiconductor devices. The general perception is that there is a very high probability this will be a huge display industry technology.

  What is the killer application at the moment for the LEDs?

What you want to do is make full-color, high-resolution displays, which will be lightweight. A liquid crystal display, for example, is actually quite fragile. You have a liquid crystal layer sandwiched between two glass sheets, and that’s a remarkable piece of technology but not particularly robust. It requires that you have quite a rigid structure to protect the glass. Something that is all solid-state is different and more appealing. But what’s turned out to be very important is that the diode by itself doesn’t make the display. What’s important is being able to wire it all up and have millions of pixels of red green and blue and get them all to work. That has turned out to be possible and we have been fairly influential in setting some of that agenda in Cambridge. This is also through a company we set up from that first patent, Cambridge Display Technology, and working in concert with Seico Epson. What we’ve demonstrated is that we can actually formulate polymer semiconductors as though they were the inks in an ink jet printer. We can then print them in the three colors—red, blue, and green—into the correct position on a screen. That notion of printing rather than using photolithography is very powerful. It is hugely attractive if it turns to be scalable in a manufacturing way. There are huge cost reductions in manufacturing these devices. I think that’s one of those key concepts which the outside world finds both rather easy to grasp and rather appealing.

The interesting thing with materials technology is that the time span to get something to the marketplace is quite extended. There’s a lot of infrastructure needed to do it. What has emerged in the last couple of years as the application everyone is talking about is the color display for cell phones. It’s the convergence of the cell phone with the Palm Pilot. And there is a perception among manufacturers that the display is the thing that makes or breaks the commercial success of the product. So just at the moment it is actually that cell phone/PDA market where everyone sees the action. It wasn’t always thus. Maybe in a couple of years all the action will be in laptop displays. One has to be cautious about predicting these things.

  Was there ever any controversy about whether yours was the first organic polymer LED?

That turns out to be something I’m very certain about. Because there was a lot of commercial interest, our patents were very vigorously contended around the world. There have been extremely ingenious efforts to try to demonstrate that other people had done enough prior to us that what we did was obvious. But our patents turned out to be quite robust. So, yes, we started it.

  Are you surprised by how quickly the technology has evolved?

In some sense, things have actually not been phenomenally quick. There are other promising technologies that moved more quickly in comparison. In this field, the commercial interest has been an enormous driver of better technology, but that was quite limited at first. And the research environment alone was limited in how much it could develop new and better materials: their purity, reproducibility, and so on. We have been waiting for a snowball effect. Now it’s been around long enough that the results are starting to look good, and the level of interest by companies prepared to put money down has gone up. Now if you look at the cumulative amount of money spent in the field, the majority has been spent very recently. It seems to be taking off.

  Has your own company proven to be a useful platform for developing the science as well as the technology?

In fact, the company has turned out to be an incredibly valuable research tool. Technology developed in the company has fed back and made better research possible. The groups that have not had that access to the industrial process have been more limited in what they do. Maybe you could call that an unfair advantage, but in this case we grab what we can. And it didn’t come without a lot of hard work.

What has happened technologically in this field is rather different, for example, than what happened with inorganic semiconductors. There, in many ways, the technology came first and the science came second. I think the big highlights of the last decade or two—the quantum Hall effect, for example—were possible because the technology for making devices had been developed already for commercial purposes. It provided this great lever to make wonderful experiments on beautifully controlled structures possible. With organic polymers, we’ve been much closer to the coal face. That’s an English expression. We have had to beat the drum and get the technology done ourselves in order to do the science. So we’re doing things together. That’s a relationship between science and commercialization that is much closer to the traditional model than has been the case with semiconductor research and development.

  Have there been any major obstacles you’ve had to overcome in getting this work done?

"Obstacle" is possibly the wrong way of expressing it. One of the big opportunities in this science is that it crosses traditional divides between subject areas. There’s communication between physics and chemistry and materials and device physics. Managing that communication has been hard work, but it’s been really rewarding, as well. It hasn’t felt at all like the ordinary mode of activity for a research program in the physics department. In fact, I spend more of my time going to chemistry conferences than I do going to physics conferences. I find what I can pick up at chemistry conferences to be extremely valuable. I’m constantly trying to better understand what chemists are trying to do. That has been a real bonus.

  What are your long-term goals for this research?

Well, I’ve never had goals that stretch longer than the duration of my ongoing research contracts. If you know where research is going in five years’ time then it isn’t research, it’s development. I can, however, describe some of the tools which I think are going to be valuable. A lot has to do with macromolecular chemistry and controlling structures. And certainly over here we refer to polymer science and colloids and assembly from liquid phases and so on as soft condensed matter, which will be important.

Soft condensed matter is basically materials that are not held together by strong chemical bonds. So the study of soft condensed matter would include liquid crystals. It would include polymers and structural aspects of polymers. It would include how molecular materials assemble onto substrates at interfaces. It’s slightly concerned with thermodynamics and statistical mechanics of how materials assemble themselves, where what you’re not doing is forming strong chemical bonds. Rather, everything is held together by the assembly of weaker bonds as, in fact, we are. If you like, it’s the materials science of life. These are things actually rather well known to the life sciences community and they would probably regard what I do as pretty naïve. But there is a general sense now that soft condensed matter is important in terms of the interface between biology and physical science. A lot of understanding about how to make structures is going to come from that side. We’re going to turn much more to soft condensed matter to find inspiration for working out what we can do and what we can make. I see that as being a great powerhouse of concepts that will enable us to make new things.

Other than that, I know what I’m doing now, and I have some idea what I’ll probably be doing in the next few years. Beyond that I think research is too opportunistic to predict. You set about arming yourself with the best set of techniques, in the broad sense, both in experiments and as an approach to thinking about the problem, and then you look to see where the problems are at a later stage. One of the very nice features of this field is this sense of putting different bits together, trying to explore different sciences, to see whether there is something at the interfaces. That’s where the discoveries come from.

Prof. Richard Friend, FRS
University of Cambridge
Department of Physics
Optoelectronics Group
Cambridge, England