North Carolina State University engineers have created a new material that could allow a fingernail-sized chip to store the equivalent of 20 high-definition DVDs or 250 million pages of text — fifty times the capacity of current memory chips.
“Instead of making a chip that stores 20 gigabytes, we have a created a prototype that can [potentially] handle one terabyte,” says Jagdish Narayan, a professor of materials science and engineering at NC State. That’s at least fifty times the capacity of the best current DRAM (Dynamic Random Access Memory) systems.
The key to the breakthrough is selective doping, the process by which an impurity is added to a material to change its properties. The researchers added nickel, a metal, to magnesium oxide, a ceramic. The result has clusters of nickel atoms no bigger than 10 square nanometers that can store data. Assuming a 7-nanometer magnetic nanodot could store one bit of information, this technique would enable storage density of more than 10 trillion bits per square inch, says Narayan.
Expanding current memory systems is a hot topic of research. At the University of California Berkeley, Ting Xu, an assistant professor of materials science, has also developed a way to guide the self-assembly of nano-sized elements in precise patterns. Xu is trying to extend the technique to create paper-thin, printable solar cells and ultra-small electronic devices.
Other researchers have shown a way to develop a carbon nanotube-based technique for storing data that could potentially last more than a billion years, thereby improving on the lifespan on storage.
A big challenge for Narayan and his team, who have been working on the topic for more than five years, was the creation of nanodots that can be aligned precisely.
“We need to be able to control the orientation of each nano dot,” says Narayan, “because any information that you store in it has to be read quickly and exactly the same way.” Earlier, the researchers could make only one-layer structures and 3-D self assembly of nano-dots wasn’t possible. But using pulsed lasers they have been able to achieve greater control over the process.
Unlike many research breakthroughs, Narayan says, his teams’ work is ready to go into manufacturing in just about a year or two. And memory systems based on doped nano-dots won’t be significantly more expensive than current systems.
“We haven’t scaled up our prototype but we don’t think it should cost a lot more to do this commercially,” he says. “The key is to find someone to start on the large-scale manufacturing process.”
See Also:
Photo: RAM (redjar/Flickr)
