Shigeo Maruyama
Carbon Nanotubes Session, MRS Memorial


I would like to talk about solar cells and the possibility of using carbon nanotube films for some energy devices. But first, I would like to show you a few pictures of me and Millie.  Not many! Maybe two or four.

This was taken some time ago.  You can see me and Saito-sensei, here. Now, before I explain this photo further, I’d like to remind you all — Millie was everywhere! She was in Japan, in China, in Malaysia, and everywhere she went, she’d make a point to talk to young students along the way. Of course, Millie’s trips to Asia started a long time ago, so by now, most of those students she spoke to are professors, with research groups of their own.

Millie's trips to Asia started a long time ago.  Photo courtesy of Dresselhaus Family
Millie's trips to Asia started a long time ago. Photo courtesy of Dresselhaus Family

But Millie wasn’t just everywhere — she also seemed eternal, to us.  As we all grew up and our funding changed and our lives changed — Millie didn't change. She always looked just the same and came at the science with the same sense of energy and passion.  The research might go in a new direction, but Millie didn’t change.

Photo courtesy of the Dresselhaus Family
Photo courtesy of the Dresselhaus Family

Millie always looked forwards — inspiring students to research the next big thing in carbon — but she never forgot people, and the people she knew were in every generation.  She always asked us how Kamimura-sensei was doing. Now, for us, Kamimura-sensei retired a long time ago. But for her, he is a friend and collaborator and colleague. All these different generations of people — to Millie, they were all just the same! Every generation was Millie’s generation, and every person Millie worked with would become a part of her family.

I intentionally brought this in Japanese, by the way, even though Chinese people may think it is in Chinese.

Last year, Millie gave a very big talk in our Fullerene, Nanotube, and Graphene Research Society in Japan. We have a recording of this talk, and we hope to make it available on the web, sometime.  I bring it up because it shows Millie’s engagement with the other side of the world and her enthusiasm for scientific discoveries made so far away from her own home! She had so much enthusiasm, all the time.  We really enjoyed it.

Millie bows to the audience before her talk
Millie bows to the audience before her talk

For me, I had been working on carbon nanotubes, nano-heat transfer, and energy topics. So Millie and I would sometimes meet while giving talks at conferences. Actually, it’s a funny thing but — if you look at all the fields I just mentioned, the overlap of those fields is very small. Whenever we came to the Nano Heat Transfer Conference (with Millie, of course), we always looked around and noticed that there were no other nanocarbon physics people there! And the opposite was true.  When we went to carbon nanotube or physics conferences, we never saw any heat transfer people. What I’m saying is that it's not easy to combine all these different fields. Also, I’ll just point out that it’s actually a little difficult to look at the energy field, because it's a very different and very diverse field.

But I digress.  Millie! Millie was everywhere and she had responsibilities everywhere. We, in Japan, have this Fullerene, Nanotube, and Graphene Research Society. I'm now the president of this Society.



We were so happy that Millie would come to Japan so often and give so many talks. Of course, Saito-sensei and Endo-sensei were always very good friends of hers — but Millie was always eager to meet students and reach out to the next generation. Millie inspired so many people and taught us so much about how to teach our own students, that I have no doubt Millie's impact and legacy will continue for a long, long time.

She always cared about the future of carbon research. I will discuss this more, in a little while.

In this picture, you can see the older members of our lab. Now, the important thing to notice, here, is that these are the professors, but we also have visiting professors!  So Anano-sensei, Professor Yan Li, Marcos-sensei — they are all in our lab as visiting professors.

I would like to thank all these visiting professors, and also the IRENA project that we worked on with Esko — we finished this with quite good success! We created controlled growth nanotubes in some devices, in transistors to the solar cell.

I would like to discuss solar cells for a little bit. Now, I don't think I need to explain to anyone in this room what a solar cell is or why we are working on it. Solar cells are necessary items in the next generation. And it is true that the silicon solar cell is quite well understood — but it is not enough.  We need to have something more. So in the organic and the perovskite… wait a minute, I will zoom in…

So, in this century, people are still doing research on many new types of solar cells, and the number and variety of solar cells is growing.  Take a look at the line for the perovskite solar cell. This is an interesting one. Last year, we had about 22% per combustion efficiency. So you can see that something new is appearing. Now, historically, new techniques may appear in solar cells but will quickly disappear when they are not adopted by industry. But at some point, we will come up with a solar cell that industry will find irresistible, so there is a lot of merit in looking for other types of solar cells.  In order to create the technology of the future, we must first invest in the science. And at our lab, we want to show that carbon nanotube and graphene can create extremely useful and important technology in this field. That is something I really want to stress.

We’ve been spending a lot of time, in our lab, working on the science of carbon nanotubes and how to grow them consistently, but we need to ask ourselves, as we’re doing all this research — what are the unique applications for carbon nanotubes?  I believe these solar cells may be one of the answers.

In organic source thin film solar cells, most people have been using fullerenes along with some polymers (like P3HT) mixed together to absorb the light.  We need 2 semiconductor layers — carrier selective layers. If we have some mixture that absorbs the light, we need some material to filter only holes (first) and then electrons. And one of the electrons must be transferred, as well! Overall, as you can see, organic thin film solar cells are really quite complicated.

We can describe the band as shown in the slide above, though we need electrons and two semiconductors.  

But actually, carbon nanotubes can work as the transparent electrode and as the carrier transport selective layer. So we demonstrated, in these two different types of the organic solar cells, that carbon nanotubes can work as the transparent conductive film and also (a little bit) as the hole transport layer. We thought of that some time ago.  And if we used the other side, we could make some transparent solar cells. The power combustion efficiency is not so high — transparent doesn't mean it's working very well — but still, it's a very unique solar cell that we can make. And the organic thin film is quite good!

But we want create new types of perovskite solar cells. So we’re going to use the exact same technique, but replace Iridium T Oxide with carbon nanotubes — or, if we’d rather, we can use carbon nanotubes on the top. This is a different design.

I want to show that carbon nanotubes can be used for both.

It's very simple! Nanotubes, nanotubes, and perovskite. This is one example. We go from here, ITO, and we have already shown that carbon nanotubes can be transferred in conductive film. Why can we not replace ITO, since in previous experiments, we used carbon nanotubes instead of ITO and established that this works?

You can see that with ITO, we have a 9% efficiency, whereas with carbon nanotubes, we only have 6%. So while it’s not perfect yet, we can make it flexible. Right?

This efficiency of 9% is not good enough for the perovskite solar cells. Research always takes time to develop. We are currently working to improve the perovskite, itself. This is a new paper that we have published with the Mansu-Choi Korean group. Now, the power conversion efficiency with graphene or nanotubes might go all the way to 14% or 15%, which would be more useful for real-world applications.  

So either carbon nanotubes or graphene, both can work. Depending on how you engineer them, it’s possible to make both of these flexible. The ITO cannot really create anything as flexible as graphene or nanotubes. Therefore, using the nanotube/graphene field, you can see the potential to create something truly unique.  

This is something we believe is very useful. We can make it flexible, we make it as a transparent conductive film; in short, we can make it do all of the things that graphene/nanotubes are known for.  This is not surprising for us. We can engineer the nanotubes/graphene on the surface, specifically, but it's still not so surprising. However, if we create the transport layer out of perovskite, it's a little more surprising. I'm not going to go into too much detail. Below is a very typical perovskite solar cell.

We use FTO, TiO2, and a perovskite layer. Usually, if we go upwards, we get the transport layer, then a layer of polymer, then one of gold. Now, we want to replace two of these layers with carbon nanotubes. Why do we want to replace these layers?  Well, the whole transport layer is rather unstable to begin with — and, of course, there are problems that arise from using gold. Gold is expensive, and it can invade the perovskite to make it less stable. So we really need to replace these layers.

When we simply take those layers completely and replace them with only carbon nanotubes, it turns out that we can get 10% power combustion efficiency — which is surprising. Encouraged, we created an inert polymer to make the nanotubes more stable, and that got us up to 13%.  Why does this happen?

We’re not exactly sure.  Carbon nanotubes have some possibility to select the holes, but it's not actually that straight-forward for us. The carbon nanotube combination that we use is a mixture of metallic and semiconductor nanotubes. The composition of our mixture means that the mixture behaves in a metallic fashion — actually, in other density of states, it can behave like a semiconductor. This a very strange thing, but it works. And I'm really trying to figure out how we can understand this phenomena and if we can use some C60 or carbon nanotubes to create higher and higher power combustion efficiency. But we don’t fully understand the physics behind what are we looking at and why the carbon nanotubes behave the way they do.  Why does this metallic carbon nanotube mixture sometimes behave like a semiconductor? It is more and more strange for us, the more research we do.

The more interesting thing is that we can mix the carbon nanotubes with PCBM — the solution-based fluorine. It behaves like the electron extracting layer. Let’s look at the other side. Normally, we use carbon nanotubes on the bottom as a hole transport layer. Nanotubes naturally tend to have great affection for holes. But sometimes, we can make the nanotube composite have great affection for electrons, instead.

Then, we can make some film out of the carbon nanotubes with PCH3 or the other carbon nanotubes with PCBM. We may have to add some additional layers, but you can see how this process creates a film that can be used to create nanotube-nanotube flexible solar cells.

Now, this bond energy diagram is just crazy. I really cannot understand it. Even though we just published a paper about it, and we walk around pretending to understand it, I'm actually afraid that I really do not understand what is going on. It simply doesn't make any sense.

But the device does work!

Maybe this is an indication of something new and useful at work that we don’t know about. Normally, in this situation, I’d call up Millie and talk to her and ask her for advice.  But I guess we’ll need to work it out for ourselves, now.

So many things are happening that are very interesting, at the moment!  This is the latest NT, in Beijing. I hope that everybody will come to this, next year, and also there will be a poster award named after Millie Dresselhaus, which will occur at this NT conference.

This is the new NT home page. We have collections of all the old NT’s, along with NT18 — and actually, we have new NT19 which is already announced, right here!

It’s a tradition — we must just keep moving forward. All this nanocarbon science is going the right direction, and I predict the future will be bright.

Photo courtesy of the Dresselhaus Family
Photo courtesy of the Dresselhaus Family

What do I remember most about Millie?  Her dedication to her work. Millie always sat in the front row, at all the conferences, taking notes.  At meetings, she’d read all the memos and was the most serious contributor there. With the students, she always took care to read every single paper and listen to run-throughs of all the talks — which is surprising for a professor of her level of fame.  

Now, it is up to us to continue this legacy, and I am prepared to take up that mantle.

Thank you very much.


Esko Kauppinen: Thank you, Shigeo, for a very interesting talk. Let’s open it up to questions.

Question Asker: Maruyama-sensei, what kind of carbon nanotube film is best for your solar cells?

Shigeo Maruyama: So far, we use Esko’s film. The reason is that it's very reproducible. We can play with other films, such as those produced by filtering water dispersed nanotubes (buckypaper) or by using direct growth of nanotubes on a substrate.

Question Asker: I meant to ask what kinds of properties are best for your solar cell? After all, if we know, then we can try to make it! Right?

Shigeo Maruyama: I think that we are still not sure.  The transparency must be good. The conductivity is very important. But actually, right now, we are tuning the diameter of the nanotubes. So this is something... it's just... The thing is, because we don't understand the mechanism, we don't know what we can do. But at least we have a lot of channels with many different carbon nanotubes. Of course, we used some purely semiconductor sorted ones, but — sometimes they're good, sometimes not. This field is extremely interesting.

Esko Kauppinen: Okay. Another question, please.

Photo courtesy of Dresselhaus Family
Photo courtesy of Dresselhaus Family

Question Asker: Hi. That was a wonderful talk. I had some questions about the defect structures in the solar cells that you mentioned. Specifically, I was wondering — you saw some very interesting electronic behavior, and I was wondering if you had seen what the fatigue rates and defect structures that typically dominated were?

Shigeo Maruyama: I think that, at this stage, we don't think that the defect is so important. We believe that all this is a very simple system, so we don't think the defect is very important but — all the interface of carbon nanotubes with some molecules, there, it's like a defect and also some interactions. They will do a lot of jobs, but that is the reason that we do not really understand what is going on.

Question Asker: So you mentioned, specifically, that they're flexible. I know many times in inflexible electronics — for example, in molecular electronics — then you would typically see that, as you flex it, then the defect structure changes. Therefore, I was wondering about fatigue with that material.

Shigeo Maruyama: The nanotube system is very good. They can move and still have contact, and the other molecules can simply adjust. You can just keep working. You can see that the defects are generally not really moving.  They can adjust. This is one of the things that makes this such a unique system.