Patent Filed in Nanotechnology
Researchers at Hewlett-Packard Co. have patented a potential breakthrough in their quest to develop computer circuits made merely of individual molecules. HP hopes to refine the process over the next decade to create microchips as powerful as the next generation of silicon-based chips - but 1,000 times smaller and less expensive.
That advance and others in the burgeoning field of nanotechnology could make computers small enough to be worn, embedded in materials or perhaps even injected into the bloodstream to serve as diagnostic sensors.
Computers are run by silicon chips crammed with millions of transistors that turn on or off extremely quickly. Because of the physical limitations of silicon, there is only so much room for more and more transistors - so there is believed to be a limit to how much silicon chips will be able to be improved after the middle of the next decade.
Hoping to get around that blockade, scientists at HP and other labs have already shown that parts of some molecules can be made to swing open or shut, making the molecule itself a switch that can be turned on or off.
HP researchers Stan Williams and Phil Kuekes, in conjunction with scientists at UCLA, also have shown that the molecular-scale switches can be connected with chemical ``wires'' that are just six to 10 atoms wide and two atoms tall. They also have found a way to make the molecular systems run even with the minute imperfections found in nature.
But researchers have still been struggling to find an efficient way to connect the microscopic systems to the much larger integrated circuits that operate computers, so that data can flow in and out when needed.
One possible solution developed by Williams and Kuekes was awarded a patent this month. They have developed a chemical process and computer program that create some order to the molecular system and allow it to be mapped like a city grid. That would allow the brains of a computer to know exactly where certain information is being stored.
Molecular computing researchers from Yale University and Rice University - who, like the HP-UCLA team are partially funded by the military's Defense Advanced Research Project Agency - have come up with a different solution to the same problem. Kuekes believes HP's method is simpler.
``We think it's sort of the only serious solution so far that's manufacturable,'' Kuekes said Tuesday. ``If this is to become a commercially viable technology, we have to be able to do this on a mass-production basis.''
A leader of the Yale-Rice project, Rice chemistry professor James M. Tour, called the newly disclosed HP process ``an interesting approach'' and ``another piece in the puzzle.''
``They've mapped it out, which is important,'' said Tour, who also is a co-founder of Molecular Electronics Corp., which plans to exploit developments in molecular computing. ``But there's still some missing pieces that need to be filled in.''
