arXiv: Millivolt Bias Shifts Cut Gate Fidelity Below 99% in Simulations

Millivolt-level fluctuations in electrical bias can significantly degrade the performance of silicon spin qubits, according to new simulations detailed by Trung Nguyen, Sarah Dweik, and Hiu Yung Wong. The researchers used the Quantum Technology Computer-Aided Design (QTCAD) simulation suite, integrating three-dimensional Poisson and Schroedinger solvers with a many-body solver, to model double quantum dots built on nanosheet technology, an approach extending previous two-dimensional approximations. Their results demonstrate that even slight variations at the plunger and middle barrier gates can reduce two-qubit gate fidelity below 99%, a critical threshold for many fault-tolerant quantum-computing algorithms. The team reports that they evaluated the exchange energy sensitivity to process and bias variations, also analyzing gate-referred 1/f charge-noise effects to better understand qubit coherence.

Nanosheet Technology Simulates Silicon Spin-Qubit Interactions

The researchers focused on silicon spin-qubit double quantum dots constructed using nanosheet technology, a promising architecture for scaling quantum computers because of its compatibility with existing semiconductor manufacturing processes. This finding underscores the need for precise control over gate voltages during qubit operation. The study also investigated the impact of gate-referred 1/f charge-noise effects on qubit coherence, aiming to pinpoint specific noise sources that limit performance. This detailed modeling approach allows for a deeper understanding of how process and bias variations impact exchange interactions, crucial for entangling qubits, and provides valuable insights for optimizing qubit designs and control strategies for more robust and scalable quantum computers built on silicon nanosheet technology.

QuTiP Analysis Reveals Gate Fidelity Sensitivity to Bias Variations

Recent advances in simulating silicon spin qubits are moving beyond traditional two-dimensional approximations, offering a more realistic assessment of gate fidelity. This approach allows for a detailed examination of exchange interactions, crucial for entanglement operations, and extends previous work largely confined to silicon-on-insulator platforms. This detailed modeling provides insights into how process variations and bias settings impact performance, and the researchers believe these findings will contribute to more robust qubit designs and control strategies for scalable quantum computers.