IBM stores one bit of data using 12 atoms

IBM has discovered a way to store one bit of data using only 12 atoms, rather than the around one million atoms needed in typical electronics today.

IBM has developed a method to store one bit of data in an array of 12 atoms, rather than the million atoms currently needed. Image credit: IBM

The technique could pave the way for magnetic storage that is ultra-small or ultra-dense, and devices could hold dramatically greater amounts of data, IBM said in its announcement on Thursday.

"Scientists demonstrated magnetic storage that is at least 100 times denser than today's hard-disk drives and solid-state memory chips," the company said in a statement. "Future applications of nanostructures, built one atom at a time, and that apply an unconventional form of magnetism called antiferromagnetism, could allow people and businesses to store 100 times more information in the same space."

What the IBM Research team achieved was a way of storing a byte of information across eight groups of antiferromagnetic atoms, organised in a two-by-six pattern. The team, based in Almaden, California, outlined its method in a paper published in the journal Science.

"Two key elements are novel — one of them is to answer how many atoms you need to store a magnetic bit of information," Andreas Heinrich, the lead investigator into atomic storage at the IBM Research labs, told ZDNet UK. "On the slightly more technology side is using an antiferromagnetic structure."

Antiferromagnets

Antiferromagnets, as opposed to the ferromagnets used more typically in electronics, exist in a state where each atom's magnetic force, or magnetic moment, is opposed to that of its neighbour. Because of this, groups of antiferromagnets are less likely to disrupt their neighbours, as their net magnetic moment is relatively even. Ferromagnets have a unified field that can interfere with neighbours by changing their magnetic moment in turn.

With ferromagnets, "all the atoms are aligned, magnetically speaking, so they make a large magnetic moment. That's good, because it allows you to read the state from far away; but it's bad, because they are talking to each other when you put two next to each other", Heinrich said. "It's potentially a problem."

By contrast, antiferromagnets cancel each other out locally, so there is no long-range magnetic field, he noted. "It's good, because you can pack them very close, but bad, because they are more difficult to study," Heinrich said.

Practical considerations

The technique lends itself to the development of new magnetic storage and spintronic devices, the IBM researchers wrote in the paper. However, the use of the method for commercial production faces some hurdles: for example, small, energy-thrifty mobile devices cannot support the infrastructure needed for the scanning and tunnelling microscope required

We don't know how to do it on a mass-manufacturing scale. Making something that relies on atomic-scale precision.

– Andreas Heinrich, IBM

Another issue is that the test device works at 1 Kelvin, just a degree above absolute zero. Even so, the researchers found that a bit of information can be encoded at room temperature with groups of 150 atoms, which is still much smaller than the million atoms required in contemporary electronics.

In addition, IBM's breakthrough relies on the use of advanced equipment that is not likely to make it into consumer electronics any time soon. Working out how to build atomic-scale devices using mass-manufacturing processes is a challenge at the moment, Heinrich acknowledged.

"It's difficult. We don't know how to do it on a mass-manufacturing scale," he said. "Making something that relies on atomic-scale precision... that goes beyond the concepts we know today. We think of this as part of exploratory research."