Q&A: Relying on Mother Nature to Make a Quantum Computer

Quantum computing research is still in its infancy. Isaac Chuang, a researcher at IBM's Almaden Research Center, has the most successful design so far. The working memory of his computer is a pencil-sized tube filled with liquid containing molecules of chloroform, alanine, or some other organic compound. The quantum "program" is a series of pulses from an NMR machine, which flip and jiggle the atoms inside the molecules. Business Week reporter Evelyn Wright spoke with Chuang about his research and his vision for the future of quantum computing.

Q: How did you get interested in working with NMR systems?
A: NMR is a very old technique. It's been around since the 1950s. Chemists have known all this time about the quantum mechanical effects that take place in NMR systems, but since they interfere with the results they're looking for, they've always tried to get rid of them.

We realized that these molecules that people had been working with for so many years had been doing quantum computations all the time, without anyone noticing it. We just tried to find ways to amplify the quantum effects, rather than trying to get rid of them.

The great thing about working with molecules is that they're always made perfectly. We are using nature to create the computer for us, rather than relying on some human manufacturing process. Even in silicon lithography, there's always some error, at the very least some atomic wobble in the lines. But nature makes molecules perfectly every time.

Q: Where does your research stand? What's next?
A: We've published results on quantum computers with two and three quantum bits, using molecules like chloroform and alanine. Now we're working with molecules that will get us to four, or maybe seven, bits. To get beyond that, we're trying to create polymer chains of molecules that will give us perhaps 10 or more bits.

Q: Do you think that your approach will eventually lead to a quantum computer that can be mass-marketed? Could I have one sitting on my desk in 20 years?
A: No, probably not. Probably something mass-produced would have to be made in solid-state, taking advantage of all the technology we have for working with silicon. An NMR design like mine might be a predecessor to a quantum supercomputer, that would be available for specialized tasks, like database search and security services.

Q: What needs to be done to get to that point?
A: When we interact with an NMR molecule, we're limited by our NMR technology, because it was designed for chemistry, not quantum computing. Most of my research focuses on overcoming these technical deficits.

We also need to find new molecules that will give us a larger number of bits. This is also a big challenge. It may be that we will need to use new techniques to design molecules of choice. There a technology called a quantum dot, which is basically an electron confined to a small box. It behaves like an atom, but we can make it out of gallium, arsenic, and other solid-state materials that we understand well. We may be able to make artificial molecules this way, and control them with NMR pulses like we do now. The trick will be trying to make our techniques insensitive to imperfections in the artificial molecules so that they work as well as natural molecules.

Q: What do you see quantum computers doing for us, if and when they become available?
A: Well, first of all, they'll give us a totally new way of seeing the world around us, as doing computations all the time. Nature is always doing computations. We just haven't noticed them.

Q: And on a more practical level?
A: I hope they will help us with things like speech and face recognition. If only my workstation could say, "Hello Ike," when I come in in the morning, and not when anyone else comes in, I'd be absolutely delighted.

We don't know how to solve problems like these yet. But by mapping them onto the physical world in new ways, we may find new ways to solve them.

Q: Physicists often say that quantum computers will be great for simulating quantum systems. Is this just something physicists would want to do, or will there be some practical application for it?
A: Absolutely there will be practical applications. For example, computer disk drives are rapidly approaching something called the superparamagnetic limit. This is a quantum limit, where the disk drive components will begin to behave quantum-mechanically. There are similar limits for many parts of computers.

When we get to these quantum limits, we will need to simulate these quantum systems, in order to be able to design systems and use them. But our current computers are unable to simulate them. Quantum systems are just too complicated for an ordinary computer simulation to handle. A quantum computer would allow us to design devices that are much smaller than those we currently can.

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

21: The Toughest Problems Will Be Solved With a Roll of the Dice

PHOTO: Research Scientist Isaac Chang Holding Quantum Computer Liquid

ONLINE ORIGINAL: Relying on Mother Nature to Make a Quantum Computer

Quantum Computing Information

NMR Quantum Computing Home Page

E-Mail to Business Week Online

Copyright 1999, Bloomberg L.P.
Terms of Use   Privacy Policy