Physics

Triple entanglement in silicon marks major quantum computer breakthrough

Triple entanglement in silicon marks major quantum computer breakthrough
A scanning electron microscope image of the new device, with the purple and green false colors added to highlight aluminum gates
A scanning electron microscope image of the new device, with the purple and green false colors added to highlight aluminum gates
View 1 Image
A scanning electron microscope image of the new device, with the purple and green false colors added to highlight aluminum gates
1/1
A scanning electron microscope image of the new device, with the purple and green false colors added to highlight aluminum gates

Quantum computers could one day outperform traditional machines at many types of tasks, but hurdles remain. Now, physicists in Japan have successfully entangled groups of three silicon quantum dots for the first time, in a breakthrough that could help make quantum computers more practical.

Quantum computers tap into the weird world of quantum physics to drastically boost the processing power and speed of computers. Information is encoded in quantum bits (qubits) in a similar way to the bits in traditional computers, except that qubits can be manipulated in a few unexpected ways.

One of these is quantum entanglement, which describes the phenomenon where groups of particles can become so intertwined that if you check the properties of one, you can not only infer that property of its partner (or partners) but actually influence it, no matter how far apart they may be. Einstein himself was baffled by the idea, referring to it as “spooky action at a distance” and originally took it as evidence that models of quantum mechanics were incomplete.

In the context of quantum computers, entangling qubits allows data to be transferred through them and processed much faster, and improves error correction. Most of the time qubits are entangled in pairs, but now researchers at RIKEN in Japan have successfully entangled three silicon qubits together.

In this case, the qubits are made of small circles of silicon called quantum dots. They’re one of the leading candidates for qubits in quantum computers, not just because silicon is already in wide use in electronics but because these quantum dots are stable for long periods of time, can be controlled precisely, operate at higher temperatures and could be scaled relatively easily. Entangling three silicon qubits is an important step towards all of these benefits, but has so far remained out of reach, although past studies have managed to entangle three photons together.

“Two-qubit operation is good enough to perform fundamental logical calculations,” says Seigo Tarucha, lead author of the study. “But a three-qubit system is the minimum unit for scaling up and implementing error correction.”

The new device is made up of three quantum dots, controlled through aluminum gates. Each of the quantum dots contains a single electron, which represents a binary one or zero through its spin state, whether it’s up or down at any given time. A magnetic field gradient keeps the qubits’ resonance frequencies separate, so they can be addressed individually.

To get the three qubits entangled together, the team started by entangling two of them, using a common unit of quantum computers called a two-qubit gate, then they entangled the third qubit with this gate. The resulting three-qubit array had a high fidelity of 88 percent, which indicates the probability that a qubit would be in the “correct” state when measured.

This robust entanglement would be most useful for error correction, the team says. In quantum computers, qubits have a tendency to randomly flip states and lose their stored information, and correction methods that work fine on traditional computers don’t work on quantum systems. Other quantum chip designs use grids of nine qubits to keep an eye on each other, while IBM’s error correction uses non-entangled qubits that perform checks on their entangled neighbors.

“We plan to demonstrate primitive error correction using the three-qubit device and to fabricate devices with 10 or more qubits,” says Tarucha. “We then plan to develop 50 to 100 qubits and implement more sophisticated error-correction protocols, paving the way to a large-scale quantum computer within a decade.”

The research was published in the journal Nature Nanotechnology.

Source: RIKEN

2 comments
2 comments
paul314
It's easy to see the need for error correction. At 88%, it only takes six gates (or six readings of one gate) to have a better-than-even chance of a mistaken readout. (Compare to modern silicon, where one error in a billion is considered dicey.)
Karmudjun
Amazing breakthrough - but lets wait for the proof of concept of 100 or 1000 qubits functioning reliably before even guestimating the necessary error correction protocols. After all, even the wrong bit states are likely correct in a few of the alternate universes of our multiverse.