Scientists Build Memory Chip as Small as Blood Cell

 
Post new topic   Reply to topic    Institute of Nanotechnology Forum Index -> Nano Computing
View previous topic :: View next topic  
Author Message
nanoorg
Site Admin


Joined: 16 Mar 2004
Posts: 1392

PostPosted: Thu Jul 05, 2007 10:42 am    Post subject: Scientists Build Memory Chip as Small as Blood Cell Reply with quote

Scientists have built a working memory chip that is roughly the size of a white blood cell — about 1/2000th of an inch on a side.

Although the memory chip is modest in capacity with 160,000 bits of information, the bits were crammed together so tightly that it is the densest ever made. The achievement points to one possible path for continuing the exponential growth of computing power even after current silicon chip-making technology hits fundamental limits in 10 to 20 years.

The scientists, led by James R. Heath of the California Institute of Technology and J. Fraser Stoddart of the University of California, Los Angeles, are to report their findings in Thursday’s issue of the journal Nature.

As far back as 1999, Dr. Heath and Dr. Stoddart have reported on aspects of their report, like specially designed molecular switches and a novel technique for making ultrathin wires. The new work pulls the components into an integrated circuit.

“Our goal always was to develop a manufacturing technique that works at the molecular scale,” said Dr. Heath, a professor of chemistry at Caltech. “It’s a scientific demonstration, but it’s a sort of a stake in the ground. It’s really small. It’s about the size of a white blood cell. We can store information in it, and we can read it out.”

The density of bits — about 100 billion per square centimeter — is about 20 times greater than current memory chips, Dr. Heath said. Improvements to the technique could push the density up by another factor of 10, he said.

But Dr. Heath said he did not know if this technique would ever be useful commercially. “I don’t know if the world needs memory like this,” he said. “I do know if you can manufacture at these dimensions, it’s a fundamentally enabling capability.”

For example, the wires used in the chip are about the same width as proteins, and that could make possible tiny circuits that detect cancer or other diseases. The researchers are making transistors and simple logic circuits using similar techniques, Dr. Heath said.

A key component of the memory chip is the molecular switch designed by Dr. Stoddart. The molecule, which belongs to a class of molecules known as rotaxanes, looks like a dumbbell with a ring that can slide along the central bar. Voltage pulses push the ring between two positions on the bar, which represent the 0’s and 1’s used by computers to store data. The dumbbell shape keeps the ring from sliding off.

To build the memory chip, the researchers first etched 400 parallel wires, each less than a millionth of an inch wide and separated by about 1/750,000th of an inch from its neighbors. On top of the wires, they deposited a layer of the molecular switches, the dumbbells standing vertically, and then a second set of 400 wires turned 90 degrees to the first set. Each crossing point between two perpendicular wires, with about 100 of the molecular switches wedged in between, is the storage location of one bit of information.

Current methods for drawing circuits on silicon will bump against the laws of physics, because the circuit components will be much smaller than the wavelength of light used to draw them. That has led researchers to look for alternative approaches to build molecular-size electronics. Most are still building circuits containing only a handful of components compared to the 160,000 in the new memory chip. That suggests the new process Dr. Heath and Dr. Stoddart developed can be scaled up to a viable manufacturing process, said Vivek Subramanian, a professor of electrical engineering and computer sciences at the University of California, Berkeley.

“This is sort of the capstone in that they’ve pulled all this together,” said Dr. Subramanian, who was not involved in the research.

Not everything works yet. When the researchers tested a small portion of the chip, they found that only 30 percent of the bits actually worked. But it is possible to use only the working parts of the chip, and the researchers successfully wrote and read information to those parts, though even there the success was temporary. The switches routinely broke after being flipped about 10 times.

The researchers readily admit that their chip is a demonstration, not imminent technology.

“We’re just happy it works,” Dr. Heath said.




Source: http://www.nytimes.com/2007/01/25/science/25chip.html



Story first posted: 25th January 2007.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    Institute of Nanotechnology Forum Index -> Nano Computing All times are GMT
Page 1 of 1
 

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2002 phpBB Group