Fresh-brewed. A prototype tabletop NMR quantum computer, showing the circuitry and the radio-frequency coil.

Quantum Computer in a Cup of Joe?

Researchers have come up with a way to turn coffee and other mundane liquids into primitive quantum computers. The findings, reported in today's Science and in an upcoming issue of the Proceedings of the National Academy of Sciences, suggest a new approach toward conquering the Mount Everest of computing: a quantum computer that in a few seconds could perform calculations that would take billions of years on an ordinary supercomputer.

The key is nuclear magnetic resonance (NMR), a technique for manipulating and measuring the spins of atomic nuclei that is already a mainstay of medical imaging and chemical analysis. The up-or-down spin of nuclei offers the two-bit logic system that a computer needs. What makes a quantum computer special, however, is quantum parallelism. In the strange world of quantum mechanics, the spin of a nucleus can be in both states, representing both bits, at the same time. That property should permit a quantum computer to perform calculations on incredibly large numbers of bits at once. And Neil Gershenfeld of the Massachusetts Institute of Technology and Isaac Chuang of the University of California, Santa Barbara, and a second team, Tim Havel and Amr Fahmy of Harvard Medical School and David Cory of MIT, have found that it's easier to achieve this parallelism by using NMR in an ordinary liquid than in the exotic systems researchers have tried so far.

Cory, Fahmy, and Havel have actually built quantum circuits using a standard NMR spectrometer, while Gershenfeld and Chuang are gearing up experiments and hope to demonstrate quantum circuits and maybe even a simple computation by next summer. What can be done now can be done with easily available and affordable equipment--"off-the-shelf coffee cups, off-the-shelf liquids, off-the-shelf magnets, etc.," says MIT computer scientist Seth Lloyd.

"This NMR scheme is pretty slick stuff, but in the long run, they're going to have to find a particularly special molecule or state that allows them to extend it to large numbers," says Chris Monroe of the National Institute of Standards and Technology.

Indeed, the researchers note that fundamental obstacles may prevent the scheme from ever producing a practical computer. But "from the point of view of verifying the basic ideas and doing interesting physics," says Lloyd, a pioneer theorist in quantum computing, "it's fantastic." For more details, Science Online subscribers can link to the News story in today's Science.