IRA FLATOW, host:
This is TALK OF THE NATION: SCIENCE FRIDAY. I'm Ira Flatow.
Unidentified Man: We're a few minutes between Park Heights and Falls(ph) and then again between York and Delaney Valley according to another driver. And we're do have some heavy traffic out along the bottom into the JFX from an earlier crash…
FLATOW: This may sound like a traffic report, but it's actually - could be the sound of a radio revolution, a radio unlike any other. It does not use standard parts made out of silicon semiconductors and transistors. This is a radio made out of carbon nanotube electronics, and is described last week in an article in the Proceedings of the National Academy of Sciences.
This transistor radio based on carbon nanotubes is so small that all of the active parts would fit on the head of a pin. Now, the point of the work is not to build a tiny radio this small, though it certainly could have practical applications.
The real point is to demonstrate something. It's a demonstration of carbon-based electronics - carbon-based transistors, carbon-based semiconductors. A field some researchers say will be the next thing after the silicon revolution of our last generation, moving toward carbon-based electronics.
John Rogers is a professor in the department of material science and engineering at the University of Illinois in Urbana-Champaign. He is one of the authors of the paper in the public - National Academy of Sciences, proceedings there. And he joins us by phone from Urbana.
Welcome to SCIENCE FRIDAY.
Professor JOHN ROGERS (Department of Material Science and Engineering, University of Illinois): Thanks. It's a pleasure to be on the show.
FLATOW: How small is this radio? I mean, you have to have a way to plug into it, right? It's got to have some bigger part to it.
Prof. ROGERS: Oh, well, in a way that's true. I mean, the active materials are of course carbon nanotubes and they're truly nanoscale in their dimensions. They have diameters that are comparable to the size of medium-sized molecule. They're pretty long, though.
And the devices that you build with them are quite a bit larger also because they involve electrodes and dialectrics(ph) and other components. So there's a range of size scales involved in the devices. And then, you know, several of those devices are assembled together to make the radio. So your previous comment that the, you know, all the active devices could fit on the heads of a pin is absolutely true. The active components are quite a bit smaller.
But, you know, when you think about a radio, you got to have an antenna. And for AMS and bands, those antennas are really going to determine the overall size, scale of the system. And, you know, together with the fact that you have to plug headphones in or plug them into a speaker to listen to them, those kinds of, you know, more conventional components...
Prof. ROGERS: ...set the overall scale of the system.
FLATOW: So you got to go to RadioShack to get some of these parts?
Prof. ROGERS: Oh, well, yeah. Actually, this is actually what we did quite literally. I mean, we're focused on the nanotube transistors themselves in showing that we could make amplifiers with gain and mixers and filters and so on. But the radio that we demonstrated, because, you know, we weren't focused on building a new kind of radio per se or a nanoscale radio, we actually assembled the other components of the, you know, final device, the resistors, capacitors, inductors, using a breadboard that we obtained at RadioShack. So it's, you know, maybe three inches by three inches, the whole thing assembled on that breadboard.
FLATOW: Mm-hmm. Well, so this is the first demonstration of using nanotubes for the semiconductor parts?
Prof. ROGERS: So - a couple of things need to be said about that. So I think you're accurate. This is the first nanotube transistor radio. The area of nanotubes is a highly competitive field. There are many groups worldwide that are working on nanotubes and trying to integrate them into electronics. And there were two other nanotube radio demonstrations where the nanotubes are implemented as individual tubes, as sort of passive oscillator elements to make a radio.
But this system that we've done is the first transistor radio where, you know, all the active components, including the audio amplifier, are all built with nanotube-based active materials.
FLATOW: So we would describe this era as sort of being very early on in the research of this?
Prof. ROGERS: Well, yeah. Early on, sort of an interesting historical perspective. You know, carbon nanotubes were discovered about in 1991, 1993, sort of in that time frame. So it's been 15 years since nanotubes were actually, you know, discovered.
And, you know, subsequent studies established their electronic properties are being quite remarkable, much better than silicon. It's taken a long time to really figure out how to configure them in a rational way that allow you to build something more than individual test devices and build, you know, circuits that do interesting things.
And I think, you know, from our standpoint, this paper is most significant in that sense is that it really demonstrates a realistic way to exploit tubes in circuits in a manner that fully, you know, takes advantage of their, you know, remarkable, intrinsic electronic properties.
And again, you know, the radio is just a demonstration of, you know, where we are in the state of development of the technology. In terms of the actual devices - and they're size scale and their performance - it's about where silicon was 15 years ago, in terms of the size scale. We don't think there's any fundamental reason why you can't, you know, push it to much higher levels of performances. But that's kind of where things are right now.
FLATOW: 1-800-989-8255 is our number. We're talking about the nanotube semi-conductor radio. Would you call it like the cat's whisker, your cat's whisker version of a radio or a crystal radio? I don't know if you're old enough to even remember those.
Prof. ROGERS: I'm not sure I know exactly what you're referring to there. Yes - fill me in…
(Soundbite of laughter)
Prof. ROGERS: Nice comment, all right?
FLATOW: Oh, we have a real generation gap.
(Soundbite of laughter)
FLATOW: The earliest radio - just to fill you in - the earliest radios have basically two or three parts.
Prof. ROGERS: Yeah.
FLATOW: And they were a crystal - called crystal radios, and they called it cat's whisker because it had a little whisker that you pushed on top of the crystal to make the contact. And it didn't have any batteries or anything in it. It uses - it used the electricity from the radio signal itself.
Prof. ROGERS: That's right, yeah. You can build a radio without an amplifier. I mean, ours incorporates an amplifier, so…
Prof. ROGERS: …you know, it produces a pretty, pretty audible, you know, loud output if you want that. But, yeah, you're exactly right. I mean, quartz crystal radios were some of the original ones. And you're right, they don't have to be very sophisticated. You can listen to the radio waves directly that way.
FLATOW: Now, in the silicon world, when you build a transistor, you put down a layer of semiconductor silicon material and you sort of etch into it the circuit board on top.
Prof. ROGERS: That's true.
FLATOW: Is that what you do in the nanotube world?
Prof. ROGERS: Well, almost. And so, you know, the reason why it's taken 15 years to sort of get to the point that we are now is that it's very difficult to control where the tubes are growing, you know?
So the tubes are great. You know, a lot of their, you know, remarkable property is derived from their nanoscale dimensions. But ultimately, you have to build a device that interfaces with humans, you know, at human scale dimensions. And figuring out how to go from the nanotubes in their usually as grown configuration, which is quite frankly a tangled rat's nest of tubes, to something that's more coherent and rational that would allow sort of scalable integration into circuits is a big challenge.
So the way that we've tackled that problem - we've been working on it for a while. And we really just kind of, by chance, stumbled onto to a solution is to grow the nanotubes on a crystalline wafer that kind of guides their orientation. So when we grow on a quartz wafer, in fact, the tubes, instead of growing in this kind of rat's nest will grow in perfectly linear, perfectly aligned arrays, sitting on the surface of the quartz, you know, over the entire wafer in a single growth step.
And so once you've grown the nanotubes in that configuration, you don't care that much that they're nanoscale in dimension anymore. You view that aligned array of tubes as a uniform thin film of a semiconductor material that, you know, has properties that are much better than silicon, and therefore, the interest in it.
And so once you've grown these aligned arrays, the rest of the processing to build circuits and devices and systems looks pretty similar to the way that you would process silicon in a sense that, you know, you deposit your metal, you put your dialectric down, you can etch away the nanotubes where you don't want them, you can do all the kind of photolithography processing that has been built up around silicon technology over the years, and really do things in a batched process, in a really sort of rational engineering controlled manner.
FLATOW: Now, you made two interesting points - let me get to both of them. One, you said that have advantages over silicon. What do nanotubes have advantages?
Prof. ROGERS: Well, they're charge-carrier mobility is quite a bit higher. And so this was made clear, I think a few years ago, by a group at Intel.
FLATOW: I got - I have to put a jargon alert in here.
Prof. ROGERS: Okay. I'm going to explain on the mobility, so I'll get to that. Intel published a paper where they…
(Soundbite of bell ringing)
FLATOW: That's the jargon - it's our jargon alert. Go ahead.
Prof. ROGERS: ...where they kind of benchmarked all, you know, different kinds of nanomaterials against silicon. Of course, they're interested in, you know, what comes after silicon. And they found that, you know, of the nanomaterials that they investigated at that time, carbon nanotubes were the one clear example of a class of nanomaterial that significantly outperformed silicon. And when I mention charge carrier mobility, basically, it tells you for a given device geometry, how fast can you switch it on and off?
Prof. ROGERS: And that switching speed - and together with the overall current output is telling you, you know, what level of sophistication you can achieve in terms of a circuit system. And the intrinsic mobility, that parameter that determines speed is about a factor of 10 higher than silicon. And so there's a general belief that if you can figure out a way to build circuits out of nanotubes, there would be advantages than over silicon.
And I think, you know, what we've tried to do is take a step down that path, to really build nanotubes out as in engineering level, you know, electronic material for next-generation circuits. And so I think we've taken a, you know, important step in that direction. There's still a lot of challenges and, you know, a lot of work that still needs to be done.
But at least it begins to look like something that's not, you know, science project. You know, it begins to look like a piece of technology. And where we think we really begin to get some traction here is we have the tubes now in a configuration that we can hand it off to electrical engineers such as our collaborators at Northrop Grumman, who know everything about RF electronics, can actually build radios and other kind of communication devices.
And I think when you get to that level, you've moved the ball forward in kind of a qualitative way and moved it into the hands of the people that need to work with that material if you ever want to achieve a sort of commercial level success.
FLATOW: Mm-hmm. You also said something that you stumbled on this by accident.
Prof. ROGERS: Yeah. So, you know what, I think a sort of random discovery, at least in my lab, it's sometimes a good thing, you know? And so I always tell my students, you know, the way to be successful is to stay active in the lab and be observant of what's going on. And so in this particular case, I have a student who was growing nanotubes in the standard, you know, rat's nest configuration. We're growing it on quartz because we were doing a laser type of processing of the tubes, where you can shine high-powered laser pulses through the tubes to burn them out. And we try to selectively burn certain kinds of tubes.
And we decided we had to grow on quartz because we wanted a substrate that was highly transparent to the laser that we were using. So we're growing on quartz. And you'll see mostly the rat's nest. A student came by one day with, you know, a micrograph image of a certain region of that quartz substrate where he pointed out that the tubes were aligned. And we thought, well, that - yeah, that's kind of neat. And we begin to work on it. And, you know, two years later, we had fully optimized and developed a, you know, a good level of understanding of what was going on and it becomes a platform for, you know, doing circuits and the sorts of things that we describe in this paper.
FLATOW: So you basically have discovered and created the template that everybody now can use to use nanotubes to create devices.
Prof. ROGERS: Well, we think so, you know? And I think our work demonstrates that we now ship out wafers to other people around the world. We, you know, we've hosted people who've come here and learned the growth techniques. People from Northrop Grumman for example have come here and they've learned how to do it. Our collaborators, Motorola learned how to do it and take it back to their own labs. And so we're very much in that mode, you know?
We've stumbled across something. We want to distribute it to the people who want to do it.
FLATOW: This is sort of like what Bell Labs did in 1948, around there when they invented the transistor?
Prof. ROGERS: That's true, yeah.
FLATOW: And allowed everybody to come in - for a fee, of course.
Prof. ROGERS: Well, we don't make much money out of it.
FLATOW: Well, it wasn't a whole lot of money back there either, you know, but comparatively speaking.
Prof. ROGERS: Yeah.
FLATOW: Considering the industry…
Prof. ROGERS: And if your listeners are interested in the quartz wafer with nanotubes on it, contact me, we'll send them…
(Soundbite of laughter)
FLATOW: You get a discount from SCIENCE FRIDAY.
Prof. ROGERS: Yeah. All right.
FLATOW: Just mention SCIENCE FRIDAY and you'll get a discount.
All right. Let's go to the phones. Let's go to Edward(ph) in Berkeley. Hi, Edward.
EDWARD (Caller): Hi. It is a wonderful program. I love it. What a neat thing, discovering it in a lab and then it becomes this thing and you make a radio out of it. I want to use the carbon nanotubes as a biological - neurobiological probe. I want to be able to stick it in the neuron and sense the current…
FLATOW: The firing.
EDWARD: Ira, do you know anything about its connection in aqueous solutions?
Prof. ROGERS: Well, we do some kind of electrochemical gating of these devices. So where the gate electrode is effectively an electrolyte in a water solution. So these things can definitely operate under water. The tubes, as they're configured on our quartz growth substrates are horizontal, they're lying in the plane of the surface of the quartz. So if you wanted to use them as probes or needles or something is what I'm envisioning based on your comment, you might want to cleave the quartz substrates so that the tubes are sort of jotting out over the edge. And then, you know, maybe, you tilt the thing on its side.
I mean, those kind of applications are very appealing to us. You know, we don't have really the bandwidth to do everything that kind of focused on these kind of electronic systems.
FLATOW: But theoretically, it's so small. Couldn't you have a tiny probe in the body? It's floating around by itself, maybe even radioing back what it finds.
Prof. ROGERS: Maybe so. Maybe so. You know, if - again, if anybody wants to learn how to do this growth, we're happy to teach them.
FLATOW: What's your next step?
Prof. ROGERS: Well, our next step, you know, we have a whole road map that we've built out with our collaborators at Northrop Grumman to go from, you know, we've demonstrated all the building blocks for R and RF electronics. But in its current form, you know, it's not competitive with current state of the art because of the scaling of the dimensions is not aggressive enough. So we have a road map that we've built out and we have some federal funding to pursue this that would allow us to go from where we are now to something in two years that would be competitive with the very best inorganic, you know, alternative or competing technology.
So we're trying to increase the density of the tubes that are in these arrays. Right now, they're fairly spaced apart. They're about a nanometer in diameter and the average spacing is about 100 nanometers. We have a lot of empty space we like to fill with tubes. So we're attempting to do that, and also sort of enhance or decrease the level of electronic heterogeneity in the properties of the tubes in these arrays.
Our collaborators in Northrop Grumman are really focused on circuit level implementations. They're optimized to exploit some of the particular properties of this class of device.
FLATOW: Mm-hmm. So you've got your work ahead of you, and that you're going to try to refine this a little better.
Prof. ROGERS: Yeah, yeah, but I think, you know, at least we have a platform. You will make things that are realistic today and it's more, you know, improving on what we have. I don't think it requires, you know, fundamental new discoveries of sort of…
FLATOW: Do your nanotubes work fast enough to be made into computer-switching devices?
Prof. ROGERS: Well, so this is a very interesting question. So we talked about, you know, we want to, you know, our research, you know, the success of a lot of the research that we do is measured ultimately by, you know, whether it becomes commercial and is useful to people, you know, more broadly speaking. And if you look at nanotubes, you say, okay, the mobility is higher than silicon. What's the best form of electronics to try to tackle first, you know, in terms of, you know, a development sequence.
We believe that RF analog is the best because it relies critically on its high switching speed, which is enabled by this high intrinsic mobility in the tubes. And probably more importantly, you can develop commercially relevant RF analog electronics for communication devices that don't involve extremely high levels of integration, in terms of - as measured in terms of numbers of devices per circuit.
So you can make a useful RF analog circuit that has hundred transistors, whereas in digital logic, you'd have to build hundred million transistors to make something that's competitive. And we just think that that's an awfully daunting engineering challenge to try to bite off at this relatively early juncture.
Having said that, we receive funding from Intel, and we do digital logic. And so we have, you know, our latest stuff, which is submitted recently, you know, involved digital logic circuits at the integration level of 100 transistors. So in many orders of magnitude from a pentium(ph), for example, but, you know, kind of trying to move the ball in that direction as well.
But for these reasons I've mentioned before, I think RF analog is probably more realistic.
FLATOW: Well, good luck to you.
Prof. ROGERS: Okay. Thanks a lot.
FLATOW: Thanks for taking time to be with us. John Rogers, professor in the department of material science and engineering at the University of Illinois in Urbana-Champaign.
NPR transcripts are created on a rush deadline by a contractor for NPR, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of NPR’s programming is the audio.