Researchers at Harvard University achieved 99.941(3)% fidelity in entangling CZ gates using a neutral atom quantum processor, a result enabled by a technique. This performance, detailed in a new paper, represents stable operation for ten hours and builds on a raw fidelity of 99.854(4)%. The team then used these low-error gates to both create and disentangle cluster states, demonstrating control that is important for complex quantum algorithms and ultimately enabling efficient fault-tolerant quantum computation.
Neutral Atom Processor Architecture for Entanglement
Employing a technique called loss postselection, scientists achieved 99.941(3)% fidelity in entangling CZ gates, a substantial improvement over the 99.854(4)% fidelity attained without the method and demonstrating a potential pathway to mitigating errors in quantum systems. This refinement involved selectively accepting only those instances where no atoms were lost during the gate operation, effectively filtering out error-inducing events and boosting the overall accuracy of the entanglement process.
The ability to enhance gate fidelity through postselection offers a valuable tool for improving the reliability of quantum computations without necessarily altering the underlying hardware. A reconfigurable architecture underpinned this sustained operation, using long-lived hyperfine clock states of rubidium-87 atoms for encoding quantum information. This design incorporates a large reservoir zone for rapid cycling of atoms, alongside dedicated sites for both entangling gates and state-selective readout.
Beyond simply creating entanglement, the team demonstrated precise control over its destruction by both creating and disentangling cluster states, a benchmark that is important for developing robust quantum algorithms and error correction protocols. The work also extended to implementing scrambling circuits with longer-range connectivity, allowing researchers to study non-locally entangled states generated through chaotic dynamics. “High-fidelity entanglement is at the heart of quantum science and technology, enabling wide-ranging applications in computing, simulation, metrology, and sensing,” the researchers state in their report.
Analysis of these complex circuits involved sampling from the output distribution and comparing the results to numerical simulations, confirming the consistency of the observed chaotic dynamics. These advances, the scientists suggest, “open the door for efficient realization of quantum algorithms with nonlocal connectivity as well as deep-circuit fault tolerant quantum computation with high-rate QEC codes.” The processor’s architecture, featuring a reservoir zone and optimized gate pulses, facilitated fast cycle rates and minimized pulse imperfections, contributing to the overall stability and performance of the system.
Rydberg Blockade and High-Frequency CZ Gate Pulses
The team’s approach uses the detection of atom loss, converting it into erasure information that is then used to refine the overall fidelity of the quantum circuit. The enhanced gate performance was achieved through several key optimizations, including a 1.7-times increase in Rabi frequency, demanding suppression of coupling to a specific Rydberg state and stronger blockade effects with closer atom spacing.
Improved Rydberg coherence times also contributed to the overall fidelity, alongside rapid calibration of the gate pulse itself. We use these gates to realize based on coherent atom motion, first benchmarking similarly high-fidelity CZ gate performance in the making and unmaking of cluster states.
941% Fidelity Achieved with Loss Postselection
Extracting fidelity from circuits with a full qubit array revealed a raw CZ gate performance of 99.854(4)%, a figure that climbed to 99.941(3)% when calculations accounted for instances where no atom was lost during the process. The results demonstrate the ability to maintain high-fidelity entanglement even with longer-range qubit connectivity, an important step for scaling up quantum computation. The observed performance suggests that the techniques employed could be instrumental in realizing deep-circuit, efficient fault-tolerant quantum computation.
Ten-Hour Stability of Entangled Qubit Performance
Neutral atom circuits maintained a consistent level of performance for over ten hours without recalibration, a duration suggesting increased viability for complex quantum computations. The researchers demonstrate how this fidelity can be maintained for over 10 hours without recalibration. Researchers verified the stability by repeatedly creating and disentangling 20-atom cluster states, a benchmark designed to assess the reliability of entangling gates. Cluster states were not merely created, but actively disassembled as a benchmark of control over entanglement, demonstrating the ability to both generate and extinguish quantum correlations within the neutral atom system.
Cluster State Creation Benchmarks CZ Gate Fidelity
To assess the robustness of gate operations, the team measured CZ fidelity using several benchmarking and readout methods, demonstrating consistent performance across different protocols. The results suggest that maintaining low-error regimes during is achievable, even with complex atom motion. Numerical simulations corroborated the experimental findings and identified key parameters influencing gate errors. Finite coherence time between the qubit state and the Rydberg state, limited by atom temperature and laser intensity fluctuations, also contributes to gate errors.
Residual errors arise from off-resonant coupling to another Rydberg level, but the simulations project that a raw CZ gate fidelity of 99.% is within reach with the current gate scheme. The AOD-SLM trap configuration was used during the creation of both cluster states and nonlocal circuits, ensuring consistent atom positioning throughout the experiments.
Nonlocal Circuits Demonstrate Super-Ballistic Scrambling
This sustained operation was demonstrated while running circuits designed to test the stability of CZ gates, critical components for building larger quantum systems. The ability to maintain gate fidelity over extended periods is essential for realizing deep-circuit quantum computation, where numerous operations are chained together to solve complex problems. These circuits were not merely static. They incorporated coherent atom motion and increasingly long-range connectivity, leading to a phenomenon known as super-ballistic scrambling of quantum information.
Scrambling, in this context, refers to the rapid and efficient distribution of quantum correlations throughout the system, a process vital for certain quantum algorithms and for mitigating the effects of errors. The circuits were designed to achieve entanglement entropy close to the Page entropy, approximately 9.28 for 20 qubits, a benchmark indicating near-maximal randomness expected from a truly scrambled quantum state.
Numerical simulations confirmed this behavior, showing that the entanglement entropy saturates at the expected value in a depth scaling with the square root of the system size. “Owing to the high degree of connectivity, such circuits are known to be effective in scrambling quantum information, and can be difficult to simulate classically,” according to the research.
Further validation came from examining the distribution of eigenvalues of reduced density matrices, which closely matched the Marchenko-Pastur distribution expected for Haar-random pure states, a hallmark of maximal entanglement. This agreement, observed with circuits featuring a nonlocal connectivity of 11, provides strong evidence for the successful implementation of super-ballistic scrambling.
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
