Researchers at Argonne National Laboratory have identified the primary cause of variability in silicon spin-qubits. By analyzing an Intel-manufactured 12-qubit processor, the team pinpointed atomic-scale disorder within silicon quantum wells as the source of fluctuating valley splitting. This phenomenon has long hindered the performance of silicon quantum devices, leading to increased error rates and loss of fidelity.
Silicon spin-qubits are a major focus for scalable computing because they share characteristics with current semiconductor manufacturing. However, the energy difference between valley states often interferes with the spin states used for calculation. When this valley splitting is inconsistent, electrons leak into unwanted states, which causes computing failures. Understanding the physical origin of this variation changes the issue from a mystery into a specific engineering target.
Using electrical spectroscopy at the Chicago Quantum Computing Testbed, the team mapped how valley splitting behaves as the position of a quantum dot shifts along the quantum well. The results show that random atomic fluctuations are responsible for the inconsistent performance observed across devices. This finding provides manufacturers with a clear direction for improving the reliability of quantum hardware.
This research represents a successful collaboration between a U.S. National Laboratory and industry. By treating qubit failure as a materials problem, the team has established a pathway to produce higher-fidelity processors. These insights are expected to inform future fabrication techniques for consistent, scalable silicon quantum computers.