Silicon qubit fidelities approaching incoherent noise limits via pulse engineering

TL;DR

This paper demonstrates that pulse engineering can significantly reduce error rates in silicon quantum dot spin qubits, approaching the incoherent noise limit and enhancing prospects for fault-tolerant quantum computing.

Contribution

The study introduces pulse engineering techniques to silicon qubits, achieving a threefold reduction in gate error rates and providing a theoretical prediction for even lower achievable errors.

Findings

Achieved 0.043% average Clifford gate error rate with pulse engineering.

Inferred noise unitarity indicating coherence properties of the noise.

Predicted potential for 0.026% error rates with further pulse improvements.

Abstract

The performance requirements for fault-tolerant quantum computing are very stringent. Qubits must be manipulated, coupled, and measured with error rates well below 1%. For semiconductor implementations, silicon quantum dot spin qubits have demonstrated average single-qubit Clifford gate error rates that approach this threshold, notably with error rates of 0.14% in isotopically enriched $^{28}$Si/SiGe devices. This gate performance, together with high-fidelity two-qubit gates and measurements, is only known to meet the threshold for fault-tolerant quantum computing in some architectures when assuming that the noise is incoherent, and still lower error rates are needed to reduce overhead. Here we experimentally show that pulse engineering techniques, widely used in magnetic resonance, improve average Clifford gate error rates for silicon quantum dot spin qubits to 0.043%,a factor of 3 improvement on previous best results for silicon quantum dot devices. By including tomographically complete measurements in randomised benchmarking, we infer a higher-order feature of the noise called the unitarity, which measures the coherence of noise. This in turn allows us to theoretically predict that average gate error rates as low as 0.026% may be achievable with further pulse improvements. These fidelities are ultimately limited by Markovian noise, which we attribute to charge noise emanating from the silicon device structure itself, or the environment.