The devices, known as memristors, or memory resistors, were conceived in 1971 by Leon O Chua, an electrical engineer at the University of California, Berkeley, but they were not put into effect until 2008 at the HP lab here.
They are simpler than today’s semiconducting transistors and can store information even in the absence of an electrical current and, according to a report in “Nature”, can be used for both data processing and storage applications.
The researchers previously reported in The Proceedings of the National Academy of Sciences that they had devised a new method for storing and retrieving information from a vast three-dimensional array of memristors. The scheme could potentially free designers to stack thousands of switches in a high-rise fashion, permitting a new class of ultradense computing devices even after two-dimensional scaling reaches fundamental limits.
Memristor-based systems also hold out the prospect of fashioning analog computing systems that function more like biological brains, Chua said.
In an interview at the HP research lab, Stan Williams, a company physicist, said that in the two years since announcing working devices, his team had increased their switching speed to match today’s conventional silicon transistors. The researchers had tested them in the laboratory, he added, proving they could reliably make hundreds of thousands of reads and writes.
That is a significant hurdle to overcome, indicating that it is now possible to consider memristor-based chips as an alternative to today’s transistor-based flash computer memories, which are widely used in consumer devices like MP3 players, portable computers and digital cameras.
A big deal
“Not only do we think that in three years we can be better than the competitors,” Dr Williams said. “The memristor technology really has the capacity to continue scaling for a very long time, and that’s really a big deal.”
As the semiconductor industry has approached fundamental physical limits in shrinking the size of the devices that represent digital 1’s and 0’s as on and off states, it has touched off an international race to find alternatives. New generations of semiconductor technology typically advance at three-year intervals, and today the industry can see no further than three and possibly four generations into the future.
The most advanced transistor technology today is based on minimum feature sizes of 30 to 40 nanometres — by contrast a biological virus is typically about 100 nanometres — and Williams said that HP now has working 3-nanometre memristors that can switch on and off in about a nanosecond, or a billionth of a second.
He said the company could have a competitor to flash memory in three years that would have a capacity of 20 gigabytes a square centimetre.
“We believe that that is at least a factor of two better storage than flash memory will be able to have in that time frame,” he said.
The H.P. technology is based on the ability to use an electrical current to move atoms within an ultrathin film of titanium dioxide. After the location of an atom has been shifted, even by as little as a nanometre, the result can be read as a change in the resistance of the material. That change persists even after the current is switched off, making it possible to build an extremely low-power device.
The new material offers an approach that is radically different from a promising type of storage called “phase-change memory” being pursued by IBM, Intel and other companies.
In a phase-change memory, heat is used to shift a glassy material from an amorphous to a crystalline state and back. The switching speed of these systems is slower and requires more power, the HP scientists say.
