The reason we started it is we wanted to know about how electron transfer occurs in materials used in polymer solar cells. So we decided to attach the conjugated polymer and the fullerene covalently because that enables you to take these materials into different conditions, different solvents, and you can always be sure that the fullerene and the polymer remain close and connected properly. In a normal solar cell, the best way to use the materials now is in a mixture, where the fullerene and the polymer, for instance, are not really bound together, but mixed together. By bonding them together covalently, it allowed us to study them in detail and gave us these new insights.

“The nice thing about this work is the realization that one day it might make significant contributions to the goal of renewable energies.”

Of our three or four most-cited papers, I think this one has the most probably because it’s the oldest. I should add that this was a collaboration between our group, Serdar Sariciftci’s in Linz and Kees Hummelen’s group at Groningen University. What happened is that together with the Groningen group, we prepared the molecules; we did the photophysics, and the group at Linz made working photocells out of these molecules.

In terms of performance, the cells were not that great. That wasn’t the point. I think the reason they got a lot of citations was because we were able to pinpoint some very interesting photophysics happening with these materials.

In this field, we tend to approach the science in two different ways. On one side, we try, of course, to optimize these organic solar cells and so constantly improve them. At the same time, we are continuously engaged in trying to understand why and how they work. This particular paper was really a study of the latter, rather than an attempt to improve performance. It’s a scientific analysis.

We showed that in many of these systems, there is an electron-transfer reaction between the polymer and the fullerene, and in that particular study, we showed that the electron-transfer reaction actually competes with energy transfer. That has turned out to be common in many of these materials. It’s still an issue that hasn’t been fully resolved, and not everyone agrees on what’s happening, but from our point of view, there is a very strong competition between energy and electron transfer. That was the message of that paper and that’s why it is so highly cited.

That is another story. Again, it’s part of a collaboration, this time with the same group in Groningen and now with the Energy Research Centre of the Netherlands. That paper is very interesting. In a normal polymer solar cell, the polymer is used together with C60 or a derivative of C60. And C60 has a number of very good properties for solar cells, but one is not so good: the compound itself really does not absorb too much light. The basic reason for that is that it is a spherical compound, which means many of the optical absorptions are what we call forbidden. The result is a low coefficient of absorption.

We thought that if we replaced C60 with one of the higher fullerenes—say, C70, which is not round but more egg-shaped and less symmetric—then some of these optical absorptions that are forbidden with C60 become allowed and we could get a higher coefficient of absorption. That was the idea. That means now that the fullerene can contribute to the current of the photocell, which it doesn’t in a typical polymer solar cell.

Prof. Dr. René Janssen's most-cited paper with 746 cites to date: Sirringhaus H, et al., "Two-dimensional charge transport in self-organized, high-mobility conjugated polymers," Nature 401(6754): 685-8, 14 October 1999. Source: Essential Science Indicators

So that paper reports on the first combination of a polymer with the C70 molecule and, indeed, we demonstrated a 50% higher current than in the same cell using C60. That was a new insight for many people. I don’t remember if it was a world record at the time, but it was one of the best-performing polymer solar cells of that year. It was certainly one of the very first or very few rational improvements to a photocell.

A lot of intuition and art goes into this business. Improvements are usually a combination of many intuitive guesses. This was one of the few times that a simple change resulted in such a big change in absorption and paid off at the end in efficiency.

It plays the role of accepting an electron from the polymer. The polymer and the fullerenes are mixed together; the mixture is excited with light, and in most cases, the polymer absorbs most of the light, which brings it to an excited state. Then it gives an electron to the fullerene; the negative charge is transported by the fullerene, and the positive charge by the polymer. With two materials, you get this charge separation. It doesn’t occur in the polymer alone and it doesn’t occur in the fullerene alone.

First of all, fullerenes have quite good electron-accepting properties. Many molecules can do that, so in that sense C60 isn’t unique, but it has much better electron-transport properties than many of the other materials that have been tried. It’s quite good at what we call charge-carrier mobility; that is, the ability of materials to allow electrons to migrate or flow under the influence of an external electric field. It allows us to extract charges from the solar cell quickly. What you’re doing in a solar cell is making positive and negative charges. The trick is to get them out as quickly as possible. If you don’t, there’s a fair chance they will recombine and you end up with heat. They only contribute to the electrical power if you extract them from the cell. So charge generation and collection is key and fullerenes allow you do it.

Basically there are four steps that are important in solar cells: absorption of light, generation of charges, transport of charges to electrodes, and then collection of charges. If you can control each of these four processes you can make a reasonably efficient organic solar cell.

That’s the basic principle of solar cells. We know those principles, but it’s finding the details of how to organize these materials to make cells better and better over the years that has always been as much art and intuition as science.

In 2003, most people were still working with a polymer called PPV. At the time it gave efficiencies of maybe 2.5%. With C70, we could improve that to 3%. In 2003, the first reports came out of another polymer, P3HT. That material is now kind of the state of the art in this field. With C60 or C70, it reaches efficiencies of slightly above 4%. There are other reports in the literature claiming 5% or more, but I think most people agree that efficiencies between 4 and 5% are realistic measurements.

It’s rather difficult to measure these efficiencies very accurately. There’s no real fundamental reason why it shouldn’t be higher in the future with other materials. People are working on that, of course, and there are options that might get efficiencies well above 7 or 8%, maybe even higher. Although the better they get, the more difficult it becomes to make them even better.

I would hope that by then there might be some first products, small-scale products, that use organic solar cells for generating power. I hope that by then the state of the art will have improved to 8%, but, again, that’s just a guess.

I think we know what kind of specifications we need for the materials that will get us there—in terms of how much light they should absorb, where the energy levels should be, what the charge-carrier mobility should be. If we can do all that in one material, we can get an efficiency of 7 or 8 or even 10%.

The problem, the heart of the issue, is that these material parameters are very difficult to predict when you’re making a new polymer. It’s a multi-parameter question and if you change one thing, you simultaneously change three or four other things in ways you can’t predict. So we have to be clever and maybe get a bit lucky.

Yes and no, of course. In the early 1990s we were dreaming that maybe one day you could go to a home improvement market and buy a bottle of paint, and paint a photovoltaic wall on the side of your house. It’s a nice dream, but the further along we get the more we learn how difficult that is. We’ve made very good progress over the years; there’s a lot we now understand about these materials—how and why they work.

But, as a scientist, you’re never satisfied in the end. If I were content with what we’ve accomplished, there would be nothing for me to do. I always say that you need a kind of discontent in a field to make progress. The field has made nice progress over the last decade, but it’s never fast enough. The nice thing about this work is the realization that one day it might make significant contributions to the goal of renewable energies. There, too, progress cannot be fast enough.

It’s still a dream, but it’s not out of the question. We now know just little more that it’s not going to be easy. It’s not so difficult to make solar cells or conducting paint. The challenge is making something really efficient that lasts for a long time. But never say never. All scientists somehow like to prove those things that other people say are impossible.

Prof. Dr. Ir. R.A.J. Janssen
Eindhoven University of Technology
Chemical Engineering and Chemistry
Molecular Materials and Nanosystems
Eindhoven, The Netherlands