Surrey Physicists Propose Superfluid Helium-3 Qubit That Could Cut Errors 100-Fold
The SHOQ design uses charge-neutral superfluid helium to shrug off the electromagnetic noise that plagues today's superconducting qubits.
Physicists at the University of Surrey have proposed a new kind of qubit built from superfluid helium-3 that they calculate could have error rates about 100 times lower than today's superconducting qubits.
The design, called the Superfluid Helium Oscillator Quantum device, or SHOQ, is described in a paper in npj Quantum Information. The work was done by Surrey's Quantum Sciences Group in collaboration with Northwestern University.
The problem it targets is noise. Superconducting qubits, the type used in most leading quantum computers, rely on electrical currents and are highly sensitive to electromagnetic interference and stray electric charges. "Tiny disturbances can cause errors and scramble the quantum information being held," said Dr. Priya Sharma, one of the researchers. Those errors are a main obstacle to scaling quantum machines up to useful size.
The SHOQ design sidesteps part of the issue by using superfluid helium-3, a rare form of liquid helium that flows without friction. Because the superfluid is charge-neutral, it should be naturally immune to some forms of electromagnetic noise. Theoretical calculations in the paper predict error rates around 100 times lower than conventional superconducting qubits.
The researchers do not present the device as a replacement for the current technology. They say it could work alongside superconducting qubits or serve as a new kind of quantum memory. "Combining different quantum technologies could allow us to take advantage of the strengths of each," said Dr. Eran Ginossar.
The catch is that the result is a prediction, not a measurement. No SHOQ device has been built yet. The team plans to make a prototype to test whether the calculated gains hold up in the lab. Operation would require extremely low temperatures, but the researchers note that such conditions have already been reached in earlier helium-3 research, which they say makes experimental testing feasible.
If prototype results match the theory, the approach would add a new option to a field that has been searching for ways to protect fragile quantum states without bulky error correction. Reducing the raw error rate of each qubit by two orders of magnitude would shrink the number of extra qubits needed to correct mistakes, which is one of the largest costs in designing a practical quantum computer.
For now the claim rests on theory, and the next step is the prototype.
Quantum computing groups worldwide are pursuing many qubit types, and the Surrey proposal adds a neutral-fluid approach to a field dominated by charged circuits.
The central test will be whether a real device can keep its quantum state long enough, and whether it can be connected to existing superconducting hardware without reintroducing the noise it was designed to avoid.
