Single-purpose quantum computers are helping physicists build simulations of nature's greatest hits and observe them up close.
INSIDE HIS LAB in Israel, Jeff Steinhauer crafts microscopic black holes. These objects are but humble specks, lacking the spaghettifying suction strength of an actual dead star. But Steinhauer, a physicist at the research university Technion, assures me that he’s constructed them mathematically to scale. Zoom in far enough, and you’ll see a miniature event horizon restaging the drama of a true black hole.
Each of these tiny blobs consists of 8,000 rubidium atoms that Steinhauer has cooled to near absolute zero and then swished around with a laser. Collectively, the atoms weigh about a thousandth of a single bacterium.
At a real black hole, gravity is so strong that once you cross its event horizon, not even light can escape. Steinhauer’s replica, technically called a Bose-Einstein condensate, has the same property but for sound waves. Past a boundary in the blob, no sonic vibrations can escape.
This work is an example of a new type of scientific experiment called a quantum simulator. Quantum simulators are small-scale replicas of complicated natural phenomena whose behavior obeys the rules of quantum mechanics. It’s the quantum equivalent of building a model airplane to predict how a real jet would fly, says physicist Ignacio Cirac of the Max Planck Institute for Quantum Optics. Steinhauer, for example, learned from his quantum replica that it emitted sonic waves analogous to the light waves that real black holes are supposed to produce, known as Hawking radiation. Because real black holes are so difficult to study, and Hawking radiation is so dim, researchers had never observed the radiation in outer space. But the sound waves in Steinhauer’s simulation offered some support to that idea. In another experiment involving cold atom blobs, physicists at the University of Chicago simulated a different extreme environment—what it would be like for a person to accelerate to billions of g’s. Theory predicts that a person accelerating this fast should be able to see objects emitting light, called Unruh radiation. It’s impossible to accelerate a person that much in the lab; for one, they’d crash into the walls almost instantly. So the researchers made the treadmill version of the scenario—everything stays in place, but they manufacture the illusion of the lab accelerating past their atom blob. “It’s like we put ourselves in a flight simulator,” says physicist Cheng Chin of the University of Chicago. “You think you’re driving a jet, but you’re really just in the laboratory.”
To create this feel, they use lasers and magnetic fields to mold the blob into the evolving shape that theory predicts it would take as an observer whips by. During this process, they use a special camera to watch the atom blob eject particles resembling the predicted behavior of Unruh radiation.
Other quantum simulators target more practical applications. For example, researchers who want to invent new materials and pharmaceutical drugs often turn to computer simulations of potential molecules as a first step. But these simulations take a lot of computing power and are not very accurate. Researchers like Cirac have proposed quantum simulator experiments to observe more closely how various geometries might lead to particular chemical properties.
These quantum simulators rely on the same techniques and hardware as quantum computers, but are tailor-made for narrow applications. They all owe a theoretical debt to the physicist Richard Feynman, who in the 1980s described a machine composed of quantum mechanical parts that could more precisely simulate reality than awkward ones and zeroes. “Nature isn’t classical, dammit, and if you want to make a simulation of nature, you’d better make it quantum mechanical,” he told a conference in 1981. Researchers at Google, IBM, and elsewhere are also trying to simulate complex molecules and other quantum objects with their so-called “universal” quantum computers, but they intend their machines be more general-purpose, capable of delivering better data encryption and speeding up artificial intelligence algorithms. Machines with this broader capability have proven to be much more challenging to build compared to the more limited-use quantum simulators.
It’s important to remember that quantum simulators are replicas of phenomena, not the phenomena themselves. Steinhauer’s atom blob relates to an astronomical black hole “as a water wave to a light wave,” says physicist Robert Wald of the University of Chicago. Both water waves and light waves consist of rippling crests and troughs, and they obey many of the same mathematical equations. Like the two types of waves, the black hole and its simulation are different objects made of different materials—which means it’s not always clear what a simulation says about the phenomenon it’s mimicking.
But these strange models are interesting in their own right, says Wald. It is a bizarre feat to replicate the properties of a black hole with ingredients as seemingly irrelevant as rubidium atoms and lasers. Regardless of what these simulations say about their natural counterparts, physicists have found a way to fit quantum components together in structures unlike any that came before.