Skip to Content

Measuring Atomic Memory with Nano Precision

Researchers at IBM now know how long a single atom can “remember” its state.
September 27, 2010

The events that take place inside of atoms occur at speeds that are normally much too fast to capture. Now researchers at IBM’s Almaden Research Center have developed a technique that lets them watch this atomic action with unprecedented resolution.

Memory machine: IBM researcher Sebastian Loth operates the scanning tunneling microscope that his team used to measure how long a single atom can store information.

The researchers used the technique to flip the orientation of an atom’s spin, a fundamental quantum property, and then to measure how long the atom “remembered” this state before returning to its natural spin state. This is a first step toward developing a kind of computer memory that works on the atomic scale, and the technique could also be used by materials scientists to perform the basic research necessary in making more efficient organic solar materials.

Influencing and measuring an atom’s spin state is one way to make a quantum bit, or qubit, which can simultaneously serve as both a 1 and a 0 in a quantum computer. It is possible to take a static measurement of an atom’s spin, but until now it hasn’t been possible to watch an atom’s spin change over time.

Researchers led by Don Eigler and Andreas Heinrich at IBM’s lab in San Jose, California, were able to watch atomic spins flip, or “relax,” over time using a modified scanning tunneling microscope, or STM–an instrument IBM researchers invented in 1981. They captured images of the atom’s state every five nanoseconds–a million times faster than before.

The IBM researchers found that a single iron atom can store magnetic information in the form of spin for about one nanosecond. However, when the iron atom is near a copper atom, its quantum memory is prolonged, so that it takes about 200 nanoseconds for the spin to relax. The results were published last week in the journal Science.

“The information decays in 200 nanoseconds, but that’s a lot of time,” says Sebastian Loth, a member of the research team. “Current processors do several hundred cycles of calculations in that time.”

When the tip of an STM is brought very close to a surface, electrical current can flow between atoms on the surface and its tip. By moving over a surface, the microscope can generate a picture of it. And by analyzing the flow of current, it’s possible to learn about the atom’s magnetic state, including its spin.

To improve the time resolution of the STM, the researchers modified the tip so that it not only measured electrical current but also supplied it. They fed current to an atom and then measured its state after a fixed period of time. For each such time period, they took 100,000 measurements. They varied the time between pulses and measurements, repeating the process again and again. The images from each measurement were combined as frames in a video. By putting these frames together, the researchers created a moving picture of the spin state of the atom, with a frame taken every five nanoseconds or so.

Loth says the IBM researchers hope to use the fast STM technique for two basic areas of research. First, they’ll continue using it to determine whether different combinations of atoms can store quantum information for longer. Second, by using a stream of photons instead of a stream of electrons as the pulse signal, says Loth, the researchers hope to gain a better understanding of how some organic molecules convert light into electrical energy. This could lead to better solar cells.

Systems like IBM’s for flipping and measuring atomic spins could potentially be part of a future quantum computer, says Alán Aspuru-Guzik, professor of chemistry and chemical biology at Harvard University. Altering and measuring the spin of atoms, and being able to predict how atoms will behave, is an important step towards this goal, he says. Most of the devices that have been made so far, he says, are more like “quantum toys” than computers. But the field is moving steadily forward, he says. “Every week someone demonstrates manipulating the qubit a little better.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.