Atomic-level computing comes a step closer as IBM scientists learn to measure how long information can stay in an individual atom
In an article published today in Science magazine, IBM researchers have outlined a new technique that can measure how long a single atom can hold data. This could enable scientists to record, study and visualise the magnetism of these atoms extremely quickly.
Using IBM’s Scanning Tunnelling Microscope (STM) in a fashion similar to a high-speed camera, researchers at IBM Research’s Almaden Lab can study the behaviour of atoms at a speed a million times faster than before.
Nanoscale Holy Grail
As the computer industry continues chasing Moore’s Law and components continue to shrink, the inevitable end point is the atomic level. The IBM researchers believe they have now made a significant breakthrough that brings the industry a step closer to that goal.
The ability to record and study atom behaviour at nanosecond speeds now opens up several avenues of research for the scientists because they can add time as a dimension in their experiments. The result of all this could impact everything from solar cells, through nanoscale data storage, to quantum computing.
“If you take Moore’s Law to the end, where you end up at is the atom,” Andreas Heinrich, IBM Research staff member and group leader of nanoscale science at the Almaden Lab, said in an interview with eWEEK, adding that the question then becomes: can you do computing and other work at that scale?
“If you can do this, you reach… not only for IBM, but for the entire industry, the holy grail,” he said.
In the data storage arena, now that scientist can essentially see an atom’s electronic and magnetic properties, they can study whether information can reliability be stored on a single atom.
Previously, scientists had determined that an iron atom could hold data for a nanosecond. Now, with the new STM technique, they found that when putting a non-magnetic copper atom next to the iron atom, that iron atom could retain data for up to 200 nanoseconds. Being able to see that change means scientists can now experiment to see how they can impact the behaviour of atoms to improve the results.
According to IBM, the breakthrough could impact quantum computers, which are systems that are not bound by the binary nature of traditional computers. Gaining a greater understanding of the nature of atoms could lead researchers to perform advanced computations that are not possible today.
Turning Atoms Into Movie Stars
The new technique the IBM scientists have used enables the STM to record the behaviour of atoms stroboscopically, similar to how the first movies were created or to the workings of time-lapse photography. This was needed because the magnetic spin of an atom changes too fast to measure directly using the STM, according to the company.
Researchers use a “pump-probe” measurement technique, where a fast voltage pulse excites the atom. Then a weaker voltage pulse measures the nature of the atom’s magnetism at a certain time after the excitation. The time delay between the two pulses creates a time frame of each measurement. The delay is then varied and the average magnetic motion is recorded in small time increments.
Taken together, the recorded increments give the scientists a more complete picture of the magnetic motion of the atom, similar to how a series of incremental photos can create a motion picture. For each time increment, the alternating pulses are repeated about 100,000 times, which takes less than a second.
“This breakthrough allows us—for the first time—to understand how long information can be stored in an individual atom, Sebastian Loth, at IBM Research, said in a statement. “Beyond this, the technique has great potential because it is applicable to many types of physics happening on the nanoscale.”IBM scientists have been using the STM for two decades to study matter at the atomic level, which could lead to innovations around computing and data storage. With the new technique using the STM, researchers can study the behaviour of atoms at the nanosecond rate, where before it was at the millisecond level, Heinrich said.
That’s important, according to IBM. The difference between a nanosecond and a second is equivalent to the difference between a second and 30 years. Now, because of the new technique, scientists can observe the physics that happens within a time sector they could not have seen before.
“This technique, developed by the IBM Research team, is a very important new capability for characterising small structures and understanding what is happening at fast time scales,” Michael Crommie, professor of physics at the University of California Berkeley and a faculty researcher at the Lawrence Berkeley National Labs, wrote in a statement. “I am particularly excited by the possibility of generalising it to other systems, such as photovoltaics, where a combination of high spatial and time resolution will help us to better understand various nanoscale processes important for solar energy, including light absorption and separation of charge.”
Heinrich said it is far too early to tell if or how this will translate to products. It will probably take another two to five years to determine whether atoms can be manipulated to store data for hours or days, rather than nanoseconds. It will take even longer, maybe 15 years or more, to determine whether any of this research will result in products.
“Jumping to the scale of a single atom, that is clearly at the end of the roadmap,” Heinrich concluded.