April 1, 2007
A memory chip the size of a white blood cell has profound implications for the future of computing.
This month Metropolis takes detailed looks at objects ranging in scale from the microscopic to the architectural. Links to the related stories can be found at the bottom of this article.
Earlier this year, two scientists announced that they had built a working memory chip about the size of a white blood cell. This was not only a computing breakthrough but something of a design accomplishment. Conventional memory chips are made exponentially smaller every year, à la Moore’s Law, but those are “just a smaller version of what Intel made five years ago,” explains James R. Heath, of the California Institute of Technology, who, along with UCLA’s J. Fraser Stoddart, created the device. Heath and Stoddart, by contrast, were designing on such a minute scale that they had to rethink completely what a memory chip could be. Indeed, the two men are chemists, not engineers, and they built their chips out of clusters of molecules instead of silicon.
The new chip is so small that they needed to devise molecules that would self-assemble in the required ways. Do terms like structure and architecture still apply at this scale? Heath says they do, noting that the chip’s circuitry, which is highly tolerant of defects, was based on some of the same principles as the Los Angeles freeway system’s. “When we first were exploring this, we actually made a computer [chip] that was designed along these levels—with streets, boulevards, and interstates.” Although the technology is about 15 years away from being integrated into your laptop, Heath is optimistic that the chip will ultimately have important applications. In the captions, he explains the details of his revolutionary design.