A Practical Path to High-Fidelity Magic States in Silicon
ORAL
Abstract
We present a compact resource analysis for producing high-fidelity logical magic states on silicon spin-qubit platforms. Our analysis evaluates a spectrum of connectivity architectures. These include a shuttling-based SpinBus design, where interactions are mediated by conveyor-mode electron trans- port, a dense nearest-neighbour layout, and several intermediate hybrid-connectivity models. For error correction, we consider both surface and color codes. We evaluate two magic-state-production models: the 5-to-1 and 15-to-1 distillation protocols. Our distillation-factory framework accounts for layout inefficiency, shuttling latency, and realistic space–time-volume costs. We extend our study beyond simple depolarizing-noise analyses to provide theoretical insight into realistic hardware con- straints. We also invert the resource-estimation process: given a target logical magic-state fidelity, we determine the corresponding experimental error rates and physical parameters required to achieve it. Thus, we offer practical guidance for experimentalists. Together, these results bridge the gap between hardware-aware quantum-resource theory and near-term silicon spin-qubit architectures, outlining optimal strategies for scalable, high-fidelity magic-state generation.
–
Presenters
-
SONGQINGHAO YANG
- University of Cambridge