TORONTO — Every innovation in memory technology begins with basic research, and a team at IBM Research has developed new technique to control the magnetism of a single copper atom. The technology could one day allow individual atomic nuclei to store and process information, but there’s a long path ahead to any form of commercialization.

In a paper recently published in the journal Nature Nanotechnology, IBM Research scientists Dr. Christopher Lutz and Dr. Kai Yang demonstrated how they can control the magnetism of a single atom’s nucleus by performing Nuclear Magnetic Resonance (NMR) one atom at a time. NMR is an essential tool for determining the structures of molecules, but the work by Lutz and Yang is the first time NMR has been achieved using a Scanning Tunneling Microscope (STM), the Nobel Prize-winning IBM invention that allows atoms to be viewed and moved individually.

“We're doing basic research in nanotechnology taken to the ultimate limit of the scale of the individual atoms,” explained Lutz in a telephone interview with EE Times. “This is the first time this has been achieved in an environment where we can see the atom and reposition it because we're using a Scanning Tunneling Microscope.”

The STM lets researchers build structures from atoms to test them out so they can understand what they want to build in the future by using a technique called spin resonance.

The STM can image and position each atom to study how the NMR changes and responds to the local environment. By scanning the ultra-sharp tip of the STM's metal needle across the surface, the STM can sense the shape of single atoms and can pull or carry atoms into desired arrangements.

“We are exploring what happens when we probe atoms one at a time and look at their magnetic properties,” said Lutz. “We’ve learned first to sense the magnetic state of a nucleus and then to control it.”

It’s a two-step process. “First we need to align it, so it's not just pointed in random directions,” he said. From there, the researchers manipulated the magnetism of the nucleus by applying radio waves emanating from the tip of a sharp metal needle. The radio waves are tuned precisely to the natural frequency of the nucleus. “Here we're getting access to the nucleus with electric current from the tip,” Lutz said.

An artist's view of the nuclear magnetism of a single copper atom. Cones represent different orientations of the magnetic north pole of the nucleus (left) and the electron (right) within the copper atom. The nuclear and the electron are magnetically linked (red spring). Electric current from the STM tip (shown at right) controls the atom's magnetism.
An artist’s view of the nuclear magnetism of a single copper atom. Cones represent different orientations of the magnetic north pole of the nucleus (left) and the electron (right) within the copper atom. The nuclear and the electron are magnetically linked (red spring). Electric current from the STM tip (shown at right) controls the atom’s magnetism.

The researchers started by looking at nuclear magnetism in iron and titanium atoms, but moved to copper, which of course is found everywhere, largely because it’s so good at conducting electricity. However, its magnetic properties are less understood. And while we never see a penny attracted by a magnet, copper’s magnetism becomes apparent when individual copper atoms are not surrounded by other copper atoms, said Lutz. “Now we've gone on to copper because copper has very strong interactions between the nucleus and its outer electrons.”

Lutz said this nucleus has four different quantum states. They’re dealing with the same ingredients for quantum computing. However, they are accessing it an environment with shorter coherence times than required for quantum computing. To put this in perspective to what it might mean for memory, he said, a magnetic memory such as M-RAM takes about a hundred thousand atoms to hold one bit. “And that's in a two-state device where the magnetic orientation gives us a one or a zero. Here we are hundred thousand times smaller, so it's really reaching as far into the future as we can.”

Lutz couldn’t speculate as to when this basic research might yield even the beginnings of a commercial application as it’s very early days. “Our next steps are going to be to build arrays of magnetic atoms. We've already started practicing assembling atoms,” he said. “It's a new effort for each new kind of atom on each new surface. But we have a long history of building structures atom by atom and showing new phenomena.”