Paul McEuen
Graphene and 2D Materials Session, MRS Memorial

First, I want to say that it’s a real honor to be here, today.  And it was also a real honor to attend the events yesterday. This memorial has been really touching, and the entire last couple of days has been fantastic, so thank you to the organizers.

Millie and Gene, 1983.  Photo credit: Geof Cooper
Millie and Gene, 1983. Photo credit: Geof Cooper

I was going to start off by telling you about my first interaction with Millie and Gene, which actually came through one of their very, very early collaborations, together — by which I mean their son, Paul Dresselhaus.

Paul in the 1980's. Photo credit: Geof Cooper
Paul in the 1980's. Photo credit: Geof Cooper

It turns out that I went to grad school with Paul Dresselhaus. We were in the same group. We even lived together for a summer! And Paul had a big influence on me, for a lot of reasons. First, it was kind of strange, because Paul would go home on the weekends, and his mother would help him with his laundry and his group theory homework.

Photo courtesy of Paul McEuen
Photo courtesy of Paul McEuen

But the more important thing is what we learned from Paul.  Above, you can see a picture of me, back then, along with Bruce Alphenaar, another one of my friends. You may notice that we look a little confused by the equipment in front of us. Well, we’re not confused by the equipment itself — we’re just confused about how it got fixed after we’d just broken it.

This was, in fact, most of what we did back then — we’d go around breaking things. And what we found out was that, after we’d finished breaking things, Paul would come along, behind us, and fix it all, again.

It was kind of like magic gnomes: you would break something, then go home for the night, and the next morning you’d come into the lab and — as if by magic — it would be fixed, again! And Paul would do this all the time, and we were all stunned because we knew he had research of his own to do. We didn’t get it!

But after the events of... particularly yesterday... I kind of started to understand why. Millie helped everybody and loved fixing things — and, apparently, that got passed on to Paul.

So Paul’s fixing skills were a tremendous benefit to me, and even if I’d never met Millie, I think that Paul’s actions, alone, were enough to completely change the direction of my career. So thanks, Millie, for Paul.

Now onto the topic at hand, which is also, as it happens, the question that I know is on all of your minds at the moment — what is Paul holding in the above picture?

Well, Paul is holding a nanosubmarine.

Well, he didn’t know that, but I’m sure that’s what’s in his hands.

So the title of my talk, today, is going to be: “building the nanosubmarine” — or, more to the point, the parts for the nanosubmarine.

This project is done in the spirit of Millie inspiration’s message: "go where your interest is!"  For some reason, this is where my interests have gone — trying to build little nano-things.

And it’s a big group of people that is involved in this project, consisting of grad students, postdocs, and faculty — we want to include everyone who’s interested!  That’s another Millie trait, so you can see that this whole project really is being done with her blessing.

So what am I talking about when I say “nanosubmarine”?  If you really take it seriously, the idea that you want to build cell-sized (∼10 microns) active, intelligent microbots.

That sounds great!  So what do you need, in order to build this thing?  Well, you need a whole bunch of things: actuation sensing, communication, computation, memory, etc.

Where are you going to get all that? This is not a trivial thing to do. This took some head-scratching.  But Millie always told me that we should pay attention to art. Okay, well, Pablo Picasso once said:

"Good artists copy, great artists steal."

(And he actually stole quite a bit, over his career!)

So… great!  We’ll steal.  Problem solved.

The first thing we’re going to steal from, to achieve our objectives is... I hate to say it in a room full of carbon enthusiasts, but… we’re going to steal from silicon.

I know what you’re all thinking.  Not carbon? Silicon?! Yes, I know, it’s not the most popular place to enthuse over silicon, but… stick with me.  Because, yes, for a little while, silicon is going to take the lead role — but don’t worry! Carbon will have its time!

We’ve got this unbelievable technology — developed over the past 50 or 60 years — for making all sorts of devices. And so, in my own group, we’ve actually started to rediscover the things that I did back when I was a grad student at Yale University, if you can believe it.  Things such as making photovoltaics, LEDs, P-injunctions, etc. That thing that you see glowing, to the left, is a gallium arsenide LED that has been integrated into some silicon electronics.

So you can build all these things and then, with a release step — some sort of etch — you can set them free. Therefore, we figured that if we needed standard electronics, maybe we should just build the electronic parts of the nanosub out of silicon and then cut it loose at the end of the day.

Just to give you one example of where this is going in my own group, we’ve started to collaborate with Al Molnár’s group in electrical engineering, where they design a high level circuit that they then send off to the Silicon Foundry. The Foundry makes the thing. They ship it back to you, and then you do some post-processing.

In our case, the post-processing involves gluing on one of those LEDs that I was telling you about earlier.   Remember those?

Once you do that, voila!  We have a complicated integrated-circuit that can do whatever function we wish. In this particular case, the goal of this little thing on the right — which is a little tab that’s about the width of a human hair and a few hundred microns long — is to record the voltage between those two probes on either side as a function of time. It then digitizes the reading and pulses out light in a specific pattern, and the pulses of light will tell you what that voltage is. I’m not going to get into that, in detail.

What do we want to do with these nanosubs we’re building? Well, we want to shove them in our heads!

I mean that literally.  We would like to take these tiny little submarines and put them into our brains and then do wireless neural recording. Or, in other words, we want to create a cell phone inside our brains!  

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

So looking back at our master-list of things we need for the nanosubmarine, our current silicon technology will do three of those things — namely sensing, communication, computation/memory. I mean, if you’re just looking for something that’ll fulfill those criteria, you don’t have to develop new science.  It’s all there, right now — and all very stealable! Silicon will do it for you, or gallium arsenide, or any other standard technology.

If, however, you want to start making things that can go from a 2D world into a 3D one, you can’t just steal, anymore.  You actually do need a little bit of help. We don’t actually know how to do that, yet.

We also do not yet know how to make things that actuate.  What does that mean? Well, in short, if you want your little neuron-cell-phone not to have to be surgically implanted into your brain, and prefer to just swallow it in pill form and have the sub crawl up into your brain (which, for some reason, I do prefer), we need to develop some new technology.

Millie prefers to swallow the nanosub with a cup of tea, rather than surgically implant it into her brain!
Millie prefers to swallow the nanosub with a cup of tea, rather than surgically implant it into her brain!

The approach that we’ve adopted is to combine Smart Panels (which would have all the standard electronics on it that you want) with some sort of 2D actuator that can cause those panels to fold up into some kind of shape. And I can hear all you carbon-enthusiasts getting excited, because you’ve all glanced ahead at the next figure and noticed that there — in between those green panels — is something that looks like it might… just might… be an artist’s rendition of graphene.

What could be in between these green panels?  Is it graphene?
What could be in between these green panels? Is it graphene?

Well, shame on you for reading ahead.  But as to the question of what that stuff is in between the panels… well, for the moment, I’m not telling.

Now, you’re all thinking this thing must work, because there’s a little picture on the right of one of these things folding up!  See it? That little thing you’re seeing is — in actual scale — only about 10 microns across. And, yes, it really is folding itself up.

How is it doing that? And what is that stuff between the green panels?

That little folding thing is a bimorph! It’s something we’re going to steal from mechanical engineering. If you want to make something that bends, the simplest thing is to glue together two materials on top of each other, and then have one of the materials contract with respect to the other — and the system will bend!

Got that?  That’s what a bimorph is. It’s in everything — all kinds of devices — so it’s easy to steal.  All we need to do is make that technology as small as possible. If you want to make a biomorph that bends with a small radius of curvature — say, a micron — you need to scale down the thicknesses of the layers to the nanometer scale.

And now, we’re in the realm of 2D materials.

All we’ve done, here, is try to push the concept of a biomorph as far as we possibly could, and, as a result, we’ve created what we think might be the world’s thinnest bimorph.

Actually, we made a couple of different versions.  The version I’ll tell you about first is a sheet of graphene and a two nanometer thick layer of glass that was deposited by atomic layer deposition. The whole stack is just barely over 2 nanometers thick.

It turns out, if you take this thing and heat it up (as we do in the image on the left), it will curl up — because when it bends, if it bends tight enough, it just curls up. It’ll curl up with a radius of curvature on the order of a micron or so.

It doesn’t just curl up in response to heat, either! Another fun thing is that it curls up in response to pH. If you put it in water (and, in fact, everything I’m going to show you related to these experiments is all in water), and if you then change the pH of the water, ions will go into the glass part of our biomorph, causing the glass to swell and causing the biomorph to bend. So it’s a pH actuated bimorph.

So now that we have this material that knows how to bend, we can combine it with panels that can be made out of anything we want, and — finally — we can make these little things that fold up in response to pH. You change the pH, and the thing folds up. Boom!  Just like that.

And, with a little work and a little cherry picking, you can get nice pictures of things that fold up in all kinds of shapes. Boxes. Mountain Valley folds. A little book that has little latches that clamp it closed. It’s really a lot of fun!

If you’re interested in this work, it’s coming out soon in PNAS. The leader of the show, by the way, is Mark Miskin — an incredibly good postdoc. Word of advice for you faculty out there — you should hire this guy.  Seriously.

Anyway, we’re having a lot of fun. And we’ve made it most of the way through our list! We can make 3D structures out of 2D ones by folding these things up using pH or temperature. But we’ve still got one problem.

If you want something that’s going to swim or burrow or do something, you would like to have actuation. All the rest of the items on the list — items 3-5 — they’re all electronic. We’re physicists; we like electrons.

For item number 2, we want voltages. We want to apply voltages, in order to have things actuate. So the last thing I’m going to show you is the way we do that.

Turns out, it’s pretty easy. Look at the cartoon in the slide above.  This cartoon is completely different from the one I showed you earlier!  It’s completely different. That layer is now platinum instead of glass. And platinum also absorbs hydrogen out of water (at least, in this case), and, better still, it’s controlled by the voltage on the platinum. Therefore, if you apply a voltage, it will take ions in. If you change the voltage, it will push them out.

So this thing should bend in response to changes in the voltage applied to the membrane. And… does that really work?


So here’s one of these little things, on the left in the slide above. And you can see that when you apply a voltage, it curls up into a really tight curl. And it’s very fast! It’s only limited by the viscosity of water as it bends through. These layers are so thin that the ions can diffuse in and out, very quickly.

In short, it’s really fun. It’s micron scale, of course.  But it’s really fun.

And here’s the really cool thing: the actuation voltages that you need to make this thing actuate are hundreds of millivolts. Really KT limited. So it’ll go from the flat state to the curled state in a change of voltage of about 100-200 millivolts. And that’s very important! If you’re working in water, you are not allowed to apply voltages larger than a few hundred millivolts, or electrochemistry starts happening, and you can do all kinds of things.

So that’s really exciting! We have our actuator, there, and if you’re wanting to wave at Millie, you can do it with lots of them. You can have many of these arms running in parallel, going like that.

The actuator waves at Millie
The actuator waves at Millie
Millie waves back
Millie waves back

It’s pretty fun and very exciting, and I don’t know of another kind of actuator that will do this — that will give you these kinds of tight radii of curvature with such small voltages.

To wrap things up, let’s look back at our list.  And, yes, we now have basically all of the items we need. All the parts! We know how to make all the little bits and pieces to construct this tiny nanosubmarine. And what’s going to be really fun, over the next 50 years, is to figure out cool new ways of combining these elements to create ever more complex robots.

And I'll just tell you about the ones that we’re working on right now. We want to take those silicon photovoltaic devices that are absolutely routine to make and combine them with these graphene biomorphs to make an actual robot.

So that’s what we’re working on. These haven’t worked yet, but I’m sure they’ll work soon! The little fancy stuff in the middle, by the way, are a couple of photovoltaic cells that will apply voltages when you shine light at them.  The legs, you see, are these graphene-platinum biomorphs. So, what you’re supposed to do is, take your laser and wiggle it back and forth across the robot, and that should make it crawl across the screen.

Well, it hasn’t happened, yet, but we hope it will. And I have to say that, I think, if we can make that happen, I think Millie would have found that pretty cool.

Millie thinks that's pretty cool
Millie thinks that's pretty cool

So with that, I will stop and thank you.