Current quantum processors are fragile, noisy and fairly limited in both quantity and quality with tens of qubits and physical error rates of around 10^-3. To realize practical quantum applications, however, error rates need to be below 10^-15 across millions of qubits. To bridge this gap and fully harness the potential of quantum computers, quantum error correction (QEC) is essential. QEC codes are designed to protect quantum information by redundantly encoding it onto multiple physical qubits. This encoding allows for the detection and correction of local errors affecting individual qubits, e.g., through stabilizer measurements. Importantly, if the physical error rates are below a specific threshold, QEC codes can exponentially suppress logical error rates by increasing the number of physical qubits involved. This is essential for achieving fault-tolerant computations, which are key to unlocking the full potential of quantum computers. The work presented in this thesis focuses on the implementation and optimization of small-scale QEC experiments using the surface code and flux-tunable superconducting qubits (Transmons). It addresses several key challenges: enhancing two-qubit gate fidelity in Surface-4 (Chapter 2), implementing an error-detection code with Surface-7 (Chapter 3), automating the calibration and benchmarking of the building blocks in Surface-17 (Chapter 4), reducing leakage into higher excited states with leakage reduction units (Chapter 5), assessing and enhancing the performance of logical qubits (Chapters 7 and 8).
Current quantum processors are fragile, noisy and fairly limited in both quantity and quality with tens of qubits and physical error rates of around 10^-3. To realize practical quantum applications, however, error rates need to be below 10^-15 across millions of qubits. To bridge this gap and fully harness the potential of quantum computers, quantum error correction (QEC) is essential. QEC codes are designed to protect quantum information by redundantly encoding it onto multiple physical qubits. This encoding allows for the detection and correction of local errors affecting individual qubits, e.g., through stabilizer measurements. Importantly, if the physical error rates are below a specific threshold, QEC codes can exponentially suppress logical error rates by increasing the number of physical qubits involved. This is essential for achieving fault-tolerant computations, which are key to unlocking the full potential of quantum computers. The work presented in this thesis focuses on the implementation and optimization of small-scale QEC experiments using the surface code and flux-tunable superconducting qubits (Transmons). It addresses several key challenges: enhancing two-qubit gate fidelity in Surface-4 (Chapter 2), implementing an error-detection code with Surface-7 (Chapter 3), automating the calibration and benchmarking of the building blocks in Surface-17 (Chapter 4), reducing leakage into higher excited states with leakage reduction units (Chapter 5), assessing and enhancing the performance of logical qubits (Chapters 7 and 8).
Quantum error correction (QEC) represents a critical enabler for fault-tolerant quantum computing, protecting fragile quantum states from decoherence and operational errors. This paper provides a comprehensive analysis of leading quantum error correction schemes, focusing on surface codes and topological codes within the context of Noisy Intermediate-Scale Quantum (NISQ) era processors. We examine the theoretical foundations of stabilizer codes, compare surface code implementations with code distances ranging from d=3 to d=17, and analyze topological codes including toric codes and color codes. Our investigation encompasses syndrome extraction circuits, logical qubit encoding strategies, and threshold analysis for contemporary quantum hardware platforms. Performance metrics reveal that surface codes achieve error thresholds of approximately 1% with realistic noise models, while topological codes offer architectural advantages for specific qubit connectivity constraints. We discuss implementation challenges on current NISQ processors including IBM Quantum, Google Sycamore, and IonQ systems, addressing qubit overhead, gate fidelity requirements, and decoding algorithms. This analysis provides insights for selecting appropriate error correction schemes based on application requirements and hardware capabilities.
Quantum error correction [1–4] is essential for bridging the gap between the error rates of physical devices and the extremely low error rates required for quantum algorithms. Recent error-correction demonstrations on superconducting processors [5–8] have focused primarily on the surface code [9], which offers a high error threshold but poses limitations for logical operations. The color code [10] enables more efficient logic, but it requires more complex stabilizer measurements and decoding. Measuring these stabilizers in planar architectures like superconducting qubits is challenging, and realizations of color codes [11–19] have not addressed performance scaling with code size on any platform. Here, we present a comprehensive demonstration of the color code on a superconducting processor [8]. Scaling the code distance from three to five suppresses logical errors by a factor of Λ 3/5 = 1.56(4). Simulations indicate this performance is below the threshold of the color code, and the color code may become more efficient than the surface code following modest device improvements. We test transversal Clifford gates with logical randomized benchmarking [20] and inject magic states [21], a key resource for universal computation, achieving fidelities exceeding 99 % with post-selection. Finally, we teleport logical states between color codes using lattice surgery [22]. This work establishes the color code as a compelling research direction to realize fault-tolerant quantum computation on superconducting processors in the near future.
Superconducting quantum computing has become one of the leading technological routes toward practical quantum computing due to its relatively long coherence times and compatibility with scalable integration. However, the high error rates of physical qubits create a scaling paradox that limits further system expansion, making quantum error correction (QEC) a core enabler of fault-tolerant quantum computing. This paper briefly introduces superconducting quantum noise models, stabilizer-code principles, and representative research achievements from 2022 to 2023 as necessary theoretical background and benchmarking references, while mainly focusing on the latest research advances in quantum error correction codes in the field of superconducting quantum computing from 2024 to 2026. It elaborates on the foundational principles of superconducting quantum noise models and stabilizer codes, with a focus on analyzing the architectures and mechanisms of surface codes, color codes, high-rate qLDPC codes, and bosonic codes. This work also summarizes recent milestone achievements, including the first exponential suppression of logical errors achieved by surface codes, the significant reduction of hardware redundancy enabled by high-rate codes, and the break-even point crossed by bosonic codes. Through multidimensional comparisons, it is shown that achieving fault tolerance at the million-qubit scale will require deep synergistic optimization of hardware, codes, and decoders. On this basis, pathways and guidance are provided for building low-overhead, high-efficiency universal quantum computers.
Quantum error correction (QEC) is one of the crucial building blocks for developing quantum computers that have significant potential for reaching a quantum advantage in applications. Prominent candidates for QEC are stabilizer codes for which periodic readout of stabilizer operators is typically implemented via successive two-qubit entangling gates, and is repeated many times during a computation. To improve QEC performance, it is thus beneficial to make the stabilizer readout faster and less prone to fault-tolerance-breaking errors. Here we design a 3-qubit CZZ gate for superconducting transmon qubits that maps the parity of two data qubits onto one measurement qubit in a single step. We find that the gate can be executed in a duration of 35ns with a fidelity of F=99.96%. To optimize the gate, we use an error model obtained from the microscopic gate simulation to systematically suppress Pauli errors that are particularly harmful to the QEC protocol. Using this error model, we investigate the implementation of this 3-qubit gate in a surface code syndrome readout schedule. We find that for the rotated surface code, the implementation of CZZ gates increases the error threshold by nearly 50\% to ≈1.2% and decreases the logical error rate, in the experimental relevant regime, by up to one order of magnitude, compared to the standard CZ readout protocol. We also show that for the unrotated surface code, strictly fault-tolerant readout schedules can be found. This opens a new perspective for below-threshold surface-code error correction, where it can be advantageous to use multi-qubit gates instead of two-qubit gates to obtain a better QEC performance.
Quantum error correction [1, 2, 3, 4] provides a path to reach practical quantum computing by combining multiple physical qubits into a logical qubit, where the logical error rate is suppressed exponentially as more qubits are added. However, this exponential suppression only occurs if the physical error rate is below a critical threshold. Here, we present two below-threshold surface code memories on our newest generation of superconducting processors, Willow: a distance-7 code, and a distance-5 code integrated with a real-time decoder. The logical error rate of our larger quantum memory is suppressed by a factor of Λ=2.14±0.02 when increasing the code distance by two, culminating in a 101-qubit distance-7 code with 0.143% ± 0.003% error per cycle of error correction. This logical memory is also beyond break-even, exceeding its best physical qubit’s lifetime by a factor of 2.4±0.3 . Our system maintains below-threshold performance when decoding in real time, achieving an average decoder latency of 63 μs at distance-5 up to a million cycles, with a cycle time of 1.1 μs. We also run repetition codes up to distance-29 and find that logical performance is limited by rare correlated error events occurring approximately once every hour, or 3×109 cycles. Our results present device performance that, if scaled, could realize the operational requirements of large scale fault-tolerant quantum algorithms.
For quantum computing to become viable, the inherently fallible nature of qubits must be overcome with quantum error correction (QEC). QEC requires coherent qubits, in a configuration compatible with a given QEC scheme, and quantum logic operations with sufficiently low error. In this thesis, we perform a series of targeted experiments to achieve these goals on a path toward realizing the surface code QEC scheme. We first develop the Xmon variant of the transmon qubit, a highly coherent, planar, and frequency tunable superconducting qubit. With coherence demonstrated, we build an array of five Xmon qubits in a configuration compatible with the surface code, and demonstrate quantum logic operations with sufficiently low error to employ the surface code. These logic gates are characterized with randomized benchmarking, a protocol for determining gate error. We find applications of randomized benchmarking beyond the intended use in gate optimization and decoherence characterization, in addition to exploring the fundamental assumptions of randomized benchmarking. Lastly, we build a nine qubit Xmon transmon array and demonstrate correction of environmental bit-flip errors in a precursor to the surface code.