Five silicon spin qubits kept about 99.9% gate fidelity when operated simultaneously in a RIKEN experiment, a key step toward scaling silicon-based quantum processors.
The team traced much of the accuracy loss to AC Stark shifts from shared microwave control signals, then offset those phase errors with a pairwise calibration method instead of calibrating every qubit combination.
Individual qubits exceeded 99.99% fidelity, with the best nearing 99.999%, and simultaneous operation of up to three qubits also stayed above 99.99%.
Remaining errors came mainly from drive-induced decoherence caused by the higher microwave power needed for parallel control, suggesting device-design improvements are still required.
Shared control lines could make large silicon quantum chips more practical than giving each qubit its own control hardware, supporting the longer-term push toward fault-tolerant quantum computing.
If silicon qubits can keep 99.9% fidelity under full parallel drive, what still prevents chip-industry-style quantum processors from scaling up?
Why did controlling five silicon spin qubits at once require correcting AC Stark shifts, and how close does this get quantum computing to fault tolerance?
Can a shared microwave line really scale silicon quantum chips, or will decoherence and calibration complexity block the path beyond five qubits?