What Will It Take To Build Good Qubits?

By Charlie Wood

Each week Quanta Magazine explains one of the most important ideas driving modern research. This week, physics staff writer Charlie Wood looks at recent high-profile news in the world of quantum computing.

Capable quantum computers would rewrite our understanding of electrons, atoms and other quantum particles, which behave in complicated ways that are tough for standard “classical” computers to simulate. As Richard Feynman memorably put it, “Nature isn’t classical, dammit, and if you want to make a simulation of nature, you’d better make it quantum mechanical.” Quantum computers might also be able to solve wider classes of problems that are hard for classical counterparts. All that’s missing is the quantum hardware to run it on.

Researchers are working hard to develop those machines, with Google, IBM, Microsoft and other institutions devoting billions of dollars to the enterprise. But the devices built so far are still only proof-of-concept science experiments. Each group is chipping away at the same colossal challenge: to marshal thousands of reliable “qubits” — the objects that power quantum computations.

A qubit, which is short for “quantum bit,” is any quantum object with two possible states. It could be a particle that spins clockwise or counterclockwise, an atom with an electron that sits close to or far from its nucleus, or a superconducting circuit acting in one of two different ways. Crucially, a qubit can exist in a murky middle ground — known as a quantum superposition — where it takes on a combination of its two states. This strange flexibility allows for algorithms that work entirely differently from those that rely on classical bits, which must exist in one state or the other.

The problem is that qubits are exceedingly wobbly. A classical magnet with its north pole pointing up (encoding one classical bit) tends to stay put. But a quantum particle in a superposition of spinning both clockwise and counterclockwise really wants to pick a lane and will readily do so if it has the slightest contact with the wider world. A stray cosmic ray, or a collision with a jittery oxygen molecule, will force it into one state or the other, instantly reducing it to a feeble bit. Researchers work hard to protect their qubits from the various tremors that permeate the world, but such efforts can go only so far. In state-of-the-art quantum physics simulations, researchers can generally keep a few dozen qubits in line for perhaps milliseconds before the whole operation falls apart.

To realize the dream of capable quantum computing, researchers need to go further. They need to either build — or simulate — more reliable, long-lived qubits.

What’s New and Noteworthy

One route is to engineer better hardware, a resilient qubit. For decades, theorists have known that under the right conditions, certain materials can ripple in a peculiar way. These ripples act rather as if an electron has been divided between two locations (although it is a collective effect, and all electrons individually remain whole). Since a disturbance is unlikely to jostle both places at once, this faux split electron — known as a Majorana zero mode — can serve as a relatively unflappable qubit. Such a “topological” qubit might function for weeks, the physicist Chetan Nayak told Quanta in 2013.

Or at least that’s the theory. Setting those ripples up has proved tough, to say the least. Programs at Bell Labs and Microsoft have devoted decades to the project, with no clear success. A Microsoft-backed team in the Netherlands announced that they had created and detected Majorana fermions in 2018, only for their claim to fall apart under further scrutiny.

The company announced a big step forward in February when it unveiled the Majorana 1, a quantum device that “not only [has] been able to create Majorana particles,” Microsoft claimed in a press release, “but can also reliably measure that information from them.” The announcement frustrated many academic researchers, because it appeared alongside a Microsoft publication in Nature that demonstrated the ability to detect particles similar to Majoranas but did not guarantee their identity. Nayak, who leads Microsoft’s efforts in this area, presented more data in an overflowing auditorium at a conference in March. But skeptics found the noisy data underwhelming and unconvincing. Microsoft researchers believe that the reality of their Majoranas will become apparent as they build bigger and more powerful devices, but most external researchers still consider Majorana zero modes to be uncaptured and untamed.

A more popular path is to develop what amounts to better quantum software. You use a bunch of shoddy qubits — atoms, say — to essentially simulate one bulletproof qubit, redundantly storing its information in multiple parts of the array. When some of the atoms inevitably fail, you use the states of their neighbors to reconstruct whatever parts of the simulated qubit got lost, a process known as quantum error correction.

Error correction is no small feat: Physicists sometimes liken it to continuously replacing all the parts in a jet plane midflight. And research milestones continue to fall. Last year, in a watershed moment, Google demonstrated that increasing the number of their physical qubits really did improve the quality of a simulated qubit. This achievement was a bit like building a wing that generates lift — a proof that flight is physically possible, but still no guarantee that a Boeing 747, a Douglas DC-3 or even a Wright Flyer is around the corner.

The last few years have seen significant advances in quantum computing. But they are modest steps down a long road. The powerful, world-changing machines that many researchers dream of are likely still decades away from being realized.

Around the Web

• Google’s Quantum AI team likens the development of quantum computers to scaling a mountain and highlights a few key milestones to watch out for along the way.
• For Science, Zack Savitsky wrote about the long, strange story of Majorana fermions and their role in quantum computing.
• Sophia Chen dove into the recent debates around Microsoft’s topological qubits for The Verge.