To create a bit, the researchers place a small amount of PCM in a nanoscale gap formed in the middle of a carbon nanotube. They can switch the bit “on” and “off” by passing small currents through the nanotube.
“Carbon nanotubes are the smallest known electronic conductors,” Pop said. “They are better than any metal at delivering a little jolt of electricity to zap the PCM bit.”
Nanotubes also boast an extraordinary stability, as they are not susceptible to the degradation that can plague metal wires. In addition, the PCM that functions as the actual bit is immune to accidental erasure from a passing scanner or magnet.
The low-power PCM bits could be used in existing devices with a significant increase in battery life. Right now, a smart phone uses about a watt of energy and a laptop runs on more than 25 watts. Some of that energy goes to the display, but an increasing percentage is dedicated to memory.
“Anytime you’re running an app, or storing MP3s, or streaming videos, it’s draining the battery,” said Albert Liao, a graduate student. “The memory and the processor are working hard retrieving data. As people use their phones to place calls less and use them for computing more, improving the data storage and retrieval operations is important.”
Pop believes that, along with improvements in display technology, the nanotube PCM memory could increase an iPhone’s energy efficiency so it could run for a longer time on a smaller battery, or even to the point where it could run simply by harvesting its own thermal, mechanical, or solar energy – no battery required.
“We’re not just talking about lightening our pockets or purses,” Pop said. “This is also important for anything that has to operate on a battery, such as satellites, telecommunications equipment in remote locations, or any number of scientific and military applications.”
In addition, ultra-low-power memory could cut the energy consumption – and thus the expense – of data storage or supercomputing centers by a large percentage. The low-power memory could also enable three-dimensional integration, a stacking of chips that has eluded researchers because of fabrication and heat problems.
The team has made and tested a few hundred bits so far, and they want to scale up production to create arrays of memory bits that operate together. They also hope to achieve greater data density through clever programming such that each physical PCM bit can program two data bits, called multibit memory.
The team is continuing to work to reduce power consumption and increase energy efficiency even beyond the savings they’ve already demonstrated.