There’s nothing quite like a game of tag – tossing a baseball back and forth promises old-fashioned fun with little effort. But it’s a tough proposition when it comes to frosty atoms and lasers.
In a new study, scientists set up a tiny baseball game in which laser beams flung and trapped atoms. This was the first example of laser-powered “optical traps” — which work much like a Star Wars lightsaber to manipulate tiny objects — successfully throwing and receiving atoms, according to a recent study published in the journal Optica. The speedball atoms were moving 4.2 micrometers at a speed of 65 centimeters per second, researchers found. In comparison, the fastest spider can crawl up to about 50 centimeters per second.
This new mechanism could eventually help power quantum computers, which have the potential to run models thousands of times faster than today’s machines — potentially paving the way for achievements such as new life-saving drugs, smarter AI, and improved cybersecurity.
More specifically, the optical traps could quickly rearrange qubits, the quantum-mechanical version of the bits found in regular computers, which hold chunks of information.
“There are ways to move qubits to enable more efficient and faster quantum computing,” Jaewook Ahn, a physicist at the Korea Advanced Institute of Science and Technology and a co-author of the new study, tells Inverse.
What is a quantum computer?
This is a superconducting quantum processor developed by D-Wave Systems Inc., one of the various approaches to running quantum computers.
D-Wave Systems, Inc./Wikimedia Commons
Scientists are still working to perfect this type of futuristic device that will take advantage of quantum mechanics to store data. As things develop, these hyped machines could solve problems that overwhelm even today’s supercomputers – and at exponentially faster speeds to boot.
Today, ordinary computers work by encoding information in content such as news articles, emails, and tweets into bits that represent the states in which electrical signals are turned “on” or “off” using combinations of ones and zeros. Meanwhile, a quantum machine places that information into qubits, which can be made from tiny objects that fit inside atoms — including electrons or photons.
Qubits can represent both 0 and 1 at the same time, a stunning quantum property called “superposition”. This means that one of these devices can essentially do the work of four regular computers. And as scientists add more qubits to a computer, processing capabilities grow exponentially.
Possible designs for quantum computers include trapping ions with electromagnetic fields and using powerful superconducting electrical circuits (currently being worked on by IBM).
Another category, called neutral atom computing, relies on super-focused laser beams that suspend atoms and manipulate them to act as qubits.
In practice, qubits are tricky — they usually need to be arranged in evenly spaced arrays to process data properly, and gaps often arise when scientists try to carefully load atoms into a device. But it’s difficult to precisely move individual atoms to fill these gaps.
“Arrays of neutral atoms trapped in focused laser beams are one of the leading quantum computing platforms,” Brian Leeds DeMarco, a physicist at the University of Illinois Urbana-Champaign, tells Inverse. “A challenge for this architecture is to efficiently build a bug-free array that doesn’t have any missing atoms.”
An ideal technique might address each defect individually, rather than having to move all the atoms in an array at once. The latter “is currently a time-consuming operation,” Robert Niffenegger, a physicist at the University of Massachusetts Amherst, tells Inverse.
Laser baseball to the rescue
Research team members Hansub Hwang (left) and Andrew Byun (right) are pictured with the optical setup used to create free-flying atoms.
Jaewook Ahn, Korea Advanced Institute of Science and Technology
To address this problem, Ahn’s team realized they could use optical traps to throw and catch specific atoms.
They began by cooling atoms of a metal called rubidium to near absolute zero and creating optical traps with an 800-nanometer laser. To “throw” an atom, they held the atom in place while accelerating the trap, turning them off and hurling the atom like a catapult. Then another laser trap is turned on to capture the atom and is slowed down to stop the atom in its orbits.
The scientists reported that their method worked 94 percent of the time in the most recent study. But “with a lower atomic temperature and more stable laser operation, this technique will be able to achieve near 100 percent accuracy,” says Ahn.
Ahn and his colleagues are not the first researchers to approach this problem with light traps. Previous studies have attempted to guide atoms between positions using lasers – but this new method differs in that it causes the atoms to rise themselves, which could prove faster and more effective.
Subatomic skepticism
While experts agree that this breakthrough could potentially boost quantum computing, it may also come with downsides. Niffenegger points out that this new method may not work as well as similar techniques applied to ions: In trapped ion designs for quantum computers, electromagnetic fields can orbit ions at higher velocities and longer distances to fix defects, he says, how shown in a study published last month.
DeMarco is also skeptical. “I’m not sure if this technique will be adopted by others in this field or by companies working on quantum computers with neutral atoms,” he says. “The probability of success in throwing and catching atoms is relatively low compared to standard techniques involving dynamic rearrangement.”
Also, this new technique introduces a lot of additional technical complexity compared to other approaches, adds DeMarco.
But Ahn concedes that this work is in its early stages and won’t necessarily find its way into computers any time soon.
“There are ways to move qubits to enable more efficient and faster quantum computing because they can be dynamically rearranged during quantum computing,” says Ahn. “For now, however, that’s too much of a claim, so we maintain that our method… could be more efficient and faster for qubit preparation (not directly for quantum computing).”
LEARN SOMETHING NEW EVERY DAY
