A qubit made of superfluid helium could cut quantum error rates 100-fold

Researchers at the University of Surrey have proposed the first qubit design based on a superfluid: a charge-neutral helium-3 device whose calculations suggest roughly 100 times lower error rates than today's superconducting qubits. A prototype has yet to be built.

Key Takeaways

Researchers at the University of Surrey have proposed the first qubit design based on a superfluid: a charge-neutral helium-3 device whose calculations suggest roughly 100 times lower error rates than today's superconducting qubits.
A prototype has yet to be built.

AI synthesis3 sources1

Quantum computers have a scaling problem that is really an error problem: superconducting qubits are so sensitive to electromagnetic noise and stray charge that, as more of them are packed together, keeping the information intact becomes the dominant engineering challenge. A team at the University of Surrey has now proposed a qubit that, by design, is deaf to some of that noise.

Their device, described in the journal npj Quantum Information, is called the Superfluid Helium Oscillator Quantum (SHOQ) device — the researchers say it is the first reported qubit design built on a superfluid. It uses superfluid helium-3, a frictionless quantum liquid, in a microfluidic structure. Because the helium is charge-neutral, the qubit is naturally shielded from the electrical disturbances that plague superconducting circuits.

The team's calculations suggest error rates about 100 times lower than conventional superconducting qubits. "The maths tells us that it should work. We have taken what we already know about superfluid helium and quantum technologies and turned that into an educated design, with the parameters and specifications needed to build one," said lead author Priya Sharma, a Daphne Jackson Fellow in hybrid quantum systems. "The next step is to make a prototype and put those predictions to the test."

The researchers are not pitching SHOQ as a replacement for existing hardware. They suggest it could be integrated alongside superconducting qubits — different qubit types doing different jobs — or, in the longer term, serve as quantum memory that stores information while other hardware computes. Co-author Eran Ginossar framed the work as widening the menu of quantum hardware rather than picking a winner. Run with Jens Koch of Northwestern University, a co-developer of the widely used transmon qubit, the project is now working towards a physical prototype, supported by an IAA Commercialisation Fellowship.