Scientists have overcome a major obstacle in the development of quantum computers by demonstrating the ability to control multiple silicon qubits simultaneously with high fidelity. A team from the RIKEN Center for Emergent Matter Science and its collaborators achieved approximately 99.9% accuracy when operating five qubits at once, a significant step toward building larger and more capable quantum processors.
Silicon spin qubits are attractive for quantum computing due to their compatibility with existing semiconductor manufacturing techniques. However, a persistent challenge has been maintaining accuracy when controlling several qubits concurrently. As more qubits operate in parallel, their operations tend to become less precise. This degradation in accuracy has been a key limitation in scaling up quantum computing systems.
The research team identified a primary cause of errors during simultaneous operation: AC Stark shifts, which are frequency shifts in a qubit caused by the microwave pulses used for control. These shifts can alter the timing of a qubit's quantum phase, thereby reducing the accuracy of subsequent operations. To address this, the scientists developed a pairwise calibration protocol. This method corrects for unwanted phase shifts by measuring the accumulation of phase between individual qubit pairs.
In their experiments, the researchers used a five-qubit silicon quantum dot array. When operating the qubits individually, each achieved single-qubit gate fidelities exceeding 99.99%, with some approaching 99.999%. After applying their calibration technique, fidelities remained above 99.99% when up to three qubits operated simultaneously. Crucially, even when all five qubits were controlled at once, the accuracy remained at approximately 99.9%. This level of precision approaches the threshold required for fault-tolerant quantum computing, a critical goal for the field.
This advancement is particularly important for scaling quantum processors. Relying on individual control lines for each qubit would become unmanageable in larger systems. The use of shared microwave control lines, as demonstrated in this work, offers a more practical approach to expanding quantum computer architectures while preserving necessary precision. The findings suggest that simultaneous control of silicon spin qubits does not necessitate a substantial sacrifice in accuracy, offering a clearer path toward the development of large-scale quantum processors.