All of tech-dom’s aflutter with news that Stanford University engineers have built a functional processor out of carbon nanotubes. A very primitive processor, mind you, with number-juggling capabilities one of the Stanford engineers admits are roughly on par with the Intel 4004 — you know, Intel’s very first microprocessor, released nearly four decades ago.
But the breakthrough here, outlined in the science journal Nature, is the shift from silicon to carbon, which could be both faster and more efficient when you’re gauging the demands of tomorrow’s exponentially more powerful computers. That, you’re going to hear a lot of people claim, could mean carbon nanotubes are the magic bullet we’ve been waiting for to stave off the imminent collapse of Moore’s Law.
More on that in a moment, but first, what the heck are carbon nanotubes anyway?
Imagine atom-thin layers of carbon — a chemical element that among other things allows us and all other forms of organic matter to exist — rolled into incredibly tiny tubes. Imagine something that looks a bit like the wall of a beehive, only made of Tinker Toys, folded around until it forms a cylinder, then shrunk to the size of nanometers, and you’re in the visual vicinity. According to MIT Technology Review, the Stanford-built processor is composed of 142 transistors, each in turn made up of carbon nanotubes between 10 and 200 nanometers long (a nanometer is one-billionth of a meter).
The future of electronics has always been about smaller and faster whether you’re future-casting protein computers, DNA computers, optical computers, quantum computers, molecular computers, whatever. But with silicon’s viability ebbing, the question’s which horse to bet on. If you can’t go smaller and faster with silicon, what can you go smaller and faster with, and soon enough to keep Moore’s Law cooking?
We might have an answer in carbon nanotubes. We’re talking about technology that’s an order of magnitude more energy efficient than silicon, says MIT Technology Review. Processors composed of carbon nanotubes would have extraordinary thermal capabilities, dissipating heat so quickly you could really crank up the speed, and it’s that speed-to-thermal-dissipation ratio that’s at the crux of silicon’s looming demise.
The Stanford researchers’ breakthrough doesn’t pull Moore’s Law out of its tailspin, but it does illustrate that carbon nanotubes — notoriously tricky to work with — may be a plausible computing solution after years of skepticism. How plausible depends on researchers’ ability to overcome considerable manufacturing obstacles in a crazy-sounding timeframe: estimates have Moore’s Law “flattening” by 2022, less than a decade away.