IRA FLATOW, host:
You're listening to TALK OF THE NATION: SCIENCE FRIDAY. I'm Ira Flatow.
This week researchers at the University of California Santa Barbara and Intel announced that they had made a new type of silicon chip. Now this is a chip that can produce laser light right on the surface of the chip. Wow. That could use to shuttle the 1s and 0s of computer code inside a computer circuit with light the same way that flowing electrons move through wires.
Joining me now to talk about is Mario Paniccia. Mario Paniccia is director of photonics technology labs at Intel in Santa Clara and one of the authors of a paper on the research being released in the journal Optics Express. Welcome to the program, Dr. Paniccia.
Dr. MARIO PANICCIA (Director of Photonics Technology Labs, Intel): Thanks for having me.
FLATOW: Do you literally just turn on the juice and the light comes out?
Dr. PANICCIA: At a high level you apply voltage and out comes a stream of light that's directly coupled into our silicon chip, yeah.
FLATOW: Is it in a form of a - if I looked at it, would I see light coming out of the chip?
Dr. PANICCIA: No. Actually the light that we generate, it's in the infrared. So most of the communication light, or laser beams, are at 1550 or 1.5 microns. So it's invisible to the human eye, but in the telecom wavelengths that's the lowest absorption in fiber.
FLATOW: So it flows through the silicon, though?
Dr. PANICCIA: Yeah. So what we actually have done is - you know, it's well known that, you know, most telecom lasers are made our of this material called indium phosphide. It's a material that emits light very efficiently. Most computer chips and communications technologies are made out of silicon, right. This is the mass majority of technology.
And what we've done with this collaboration, this development, is we've combined uniquely those two combination of materials: the light-emitting properties of indium phosphide with the silicon and high-volume capabilities of silicon and silicon photonics.
FLATOW: Now why - you know, this was written up as being a terrific breakthrough. Why is this so exciting to scientists?
Dr. PANICCIA: Well, there's a couple of things here. One is if you look at silicon photonics in general - silicon photonics is an area that's, you know, I think is going to revolutionize communications. It's an area that's evolving. But the concept is can we build optical devices using standard silicon and silicon manufacturing techniques.
And, you know, two or three years ago no one thought you could do it. The developments over the last, you know, one or two years - we've had phenomenal breakthroughs proving that you can build devices out of silicon.
And why is that important? Just like the vacuum tube transition to the plain or integrated circuit, which, you know, Intel and others have dramatically changed the way we live today. Taking the same analogy of taking these bulky, big components that are optical today and being able to put that down on silicon will transform the way people communicate, the way you download movies, the way you can transmit information in the future, because now you can do this using light instead of electrons or fiber optics.
FLATOW: And the advantage that light has is?
Dr. PANICCIA: Well, the advantages are, A, photons unlike electrons don't have any charge. So they don't interact. So I can send an enormous amount of data down a fiber. I can send it very far distances and I can send multiple beams simultaneously. So you get an enormous amount of bandwidth capability.
You know, we're talking potentially in the future with this laser development, you know, 1000x faster bandwidth speeds than what's done today at, you know, even at 10 gigabytes per second. So the remaining hurdle, or one of the keys issues was: If we can build these devices out of silicon, how do we bring the laser onto the chip?
And if we take the existing commercial lasers that are very expensive, it's not the optimal integration solution. And so this laser development almost gave us the last piece. You know, again, this is research. You know, I want to really thank our collaborators down - you know, Professor John Bowers and the University of Santa Barbara, who are experts in indium phosphide.
But this last remaining hurdle, we think we now have a way not only to get a laser on a chip but we're talking hundreds, if not thousands of lasers that can be assembled and manufactured using our standard silicon manufacturing techniques. So this will be enable us to drive high volume, optical technology to the home, in or around the PC.
You know, just imagine, you know, downloading movies that just take a couple of seconds and you can download movies for the rest of your week. You know, enormous, enormous improvements in bandwidth capability.
FLATOW: I understand the ramifications; they're incredible. Talk to me about the technology a bit more. How does the laser get combined with everything else that you need on a chip - the wires, the transistors, this stuff - how do you interface that with the light?
Dr. PANICCIA: Well, that's sort of the nitty-gritty details. We've developed a process that all the complexity is on the silicon side. So we manufacture the silicon photonics, the silicon wave guides. Silicon wave guides are devices where we can actually guide and direct the light. That's done first, and then we bond to that, using a unique process of, you know, it's a called a plasma oxidation. So we actually oxidize both materials. We then bond them together at a low temperature, and these oxide layers act as a glass glue.
So since all the patterning is on the silicon side, when you, you know, pattern and put contacts on the indium phosphide, it's just a light source. So it emits light. The light is coupled right into the silicon and the silicon chip acts as your laser cavity.
FLATOW: Wow, sounds great. We wish you good luck with it, and we'll watch for those movies. I'll try to keep my kids away from downloading a movie in two seconds, so.
Dr. PANICCIA: Terrific. Thank you, Ira.
FLATOW: Thank you very much, Mario. Mario Paniccia is director of the photonics technology lab at Intel, based in Santa Clara, California.
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.