Randomized benchmarking of single-qubit gates in a 2D array of neutral-atom qubits

Area of Science:

• Quantum Computing
• Atomic Physics
• Quantum Information Science

Background:

• Neutral-atom arrays are a promising platform for scalable quantum computing.
• Accurate control of individual qubits is essential for implementing quantum algorithms.
• Randomized benchmarking is a standard technique for characterizing gate fidelities.

Purpose of the Study:

• To characterize single-qubit Clifford gate operations in a 2D neutral-atom array.
• To demonstrate high-fidelity global and site-selective gates.
• To quantify gate fidelity and crosstalk errors.

Main Methods:

• Utilized randomized benchmarking to assess gate performance.
• Implemented global microwave-driven gates across a 49-qubit array.
• Employed a focused laser beam for site-selective Stark shifting to activate single-site gates.

Main Results:

• Achieved an average fidelity of F2=0.9983(14) for global microwave-driven gates.
• Demonstrated site-selective gates with an average fidelity of F2=0.9923(7).
• Measured an average spin-flip crosstalk error of 0.002(9) at non-targeted sites.

Conclusions:

• High-fidelity single-qubit Clifford gates are achievable in neutral-atom arrays.
• Site-selective control enables precise manipulation of individual qubits within the array.
• The demonstrated fidelities and low crosstalk are promising for building larger, fault-tolerant quantum computers.