Researchers affiliated with Forschungszentrum Jülich, Aarhus University, and Chalmers University of Technology have demonstrated a dispersive measurement method for superconducting qubits that simultaneously measures a qubit’s state and resets the readout resonator. This combined operation addresses a key bottleneck in quantum computing speed by eliminating the need for separate measurement and reset steps.

The work demonstrates driving the resonator to approximately 10^2 photons and back to approximately 10^(-3) photons in less than three times the resonator’s natural decay rate while maintaining an assignment error below 1% due to T_1 limitations. The approach utilizes analytically defined pulses and extends to measuring an arbitrary number of qubit states, including qutrits.

Superconducting Qubit Readout via Dispersive Interaction

This rapid cavity reset is achieved through a technique the researchers term “Derivative Removal for Annulment of Cavity-Hybridized Measurement Adiabats.” The method involves calculating an input signal that effectively nulls the cavity field at the conclusion of the measurement pulse. The underlying principle relies on linear response theory, where the relationship between the input signal and the cavity field is described by a time-domain response function.

By carefully tailoring the input signal, the team could manipulate the cavity field to return to its initial state, eliminating the waiting times previously required to allow the cavity to naturally decay. Details of the experimental setup, including the utilized fixed-frequency transmon qubit coupled to a half-wavelength resonator, are available in their published work.

Measurements revealed a qubit resonance frequency of 3947 MHz with an anharmonicity of -232 MHz, a T_1 relaxation time of 40-75 microseconds and a T_2* dephasing time of 20-40 microseconds, varying over time. The resonator exhibited a resonance frequency of 6041.200 MHz and a linewidth of 0.5647(3) MHz, corresponding to a cavity decay time of 281.2(8) nanoseconds. The dispersive shift, a measure of the qubit’s influence on the resonator, was determined to be 0.299 MHz.

These parameters, while subject to slight variations between experimental runs, provided a stable platform for demonstrating the readout. The team successfully demonstrated the technique for both single-state measurements and for distinguishing between the ground and first excited states of the transmon, and also presented results for qutrit readout, demonstrating the ability to resolve three distinct quantum states.

Analysis of the experimental data showed an assignment error of 0.6% while achieving more than 30 decibels of cavity reset for the three-level system. This level of performance, achieved within a total readout duration of three times the resonator’s natural decay rate, represents an advancement in the speed and efficiency of quantum measurement and initialization. Device fabrication took place at Myfab Chalmers, as detailed in the published findings.

Analytical Pulse Shapes Enable Resonator Reset

This advancement, detailed in recent findings, streamlines quantum error correction by eliminating delays associated with resonator ring-down, a period where residual energy interferes with subsequent operations. The technique relies on precisely shaped microwave pulses applied to a superconducting qubit coupled to a half-wavelength resonator, allowing for a rapid transition from measurement to the next computational step. Researchers carefully characterized the qubit’s coherence times, finding T_1 relaxation times between 40-75 microseconds and pure dephasing times, T_2*, ranging from 20-40 microseconds, varying over time.

The resonator itself resonated at 6041.200 MHz, a value crucial for achieving fast readout. The core of this innovation lies in the application of meticulously designed pulses to null the intra-cavity field after measurement. These pulses are not derived through iterative optimization, but rather calculated directly from the known qubit and resonator parameters, simplifying implementation and enhancing scalability.

Researchers can drive the resonator to approximately 10^2 photons and back to approximately 10^(-3) photons, a more than 10^3-fold reduction, within a time frame less than three times the resonator’s natural decay rate, where the decay rate is represented by κ. This rapid reset minimizes decoherence and allows for faster cycling through quantum error correction routines. The team’s success hinges on a precise understanding of the system’s Hamiltonian, described as H_d = ℏ(Δ + χσ_z)a^+ + a, where Δ is the qubit-cavity detuning and χ represents the dispersive shift.

By leveraging this model and applying Duhamel’s integral, they could calculate the input pulse shape needed to achieve the desired cavity reset. This analytical approach circumvents the need for computationally intensive optimization algorithms, making it particularly attractive for complex quantum circuits, and offers a pathway toward scalable quantum computing by addressing a key bottleneck in measurement and initialization speed.

Demonstration of Sub-Cycle Readout Timing: 3κ^(-1)

Pavel Bushev and colleagues at Forschungszentrum Jülich have demonstrated a method for simultaneously reading the state of a superconducting qubit and resetting the resonator used for readout to its initial condition, a combined operation previously requiring sequential steps. The team’s approach leverages precisely shaped microwave pulses to not only extract qubit information but also rapidly deplete the energy stored within the readout resonator.

Conventional dispersive qubit readout relies on measuring a shift in the resonator’s frequency dependent on the qubit’s state; however, this process leaves residual photons within the resonator, creating interference for subsequent measurements and necessitating waiting periods equivalent to several cavity decay times. These pulses are designed to drive the resonator to approximately 10^2 photons and back to approximately 10^(-3) photons.

Experimental results confirm the theoretical predictions, demonstrating a T_1-limited assignment error below 1% within a total readout duration of three times the resonator’s natural decay rate while achieving the sub-cycle readout timing. This generalization to an arbitrary number of modes and states underscores the versatility of the approach and its potential for implementation in more complex quantum systems. The authors provide further details in their published work.

Transmon Qubit and Resonator Parameter Characterization

The conventional approach to qubit readout involves exciting a resonator, observing the resulting signal, and then waiting for the resonator to naturally decay before the next operation can begin. This decay time, characterized by the inverse of the decay rate, often limits the speed of quantum processing. Experimental results show the resonator can be driven to approximately 10^2 photons and then reduced to approximately 10^(-3) photons in less than three times the resonator’s natural decay rate, a substantial improvement over passive decay methods.

This rapid reset is crucial for maintaining qubit coherence and minimizing errors in subsequent operations. The method’s success depends on a detailed understanding of the system’s Hamiltonian and the application of carefully crafted analytical pulses.

Unlike iterative optimization techniques, this approach derives pulse shapes directly from theoretical calculations, making it computationally efficient and adaptable to complex systems. The team demonstrated this versatility by extending the technique beyond qubit readout to successfully measure a qutrit, a quantum system with three levels, further showcasing the method’s potential for scaling up quantum circuits. A pulse duration of 1 microsecond was used, providing a precise baseline for the pulse shaping and measurement fidelity.

Crucially, the researchers achieved a T_1-limited assignment error of 0.6%, indicating that the measurement process itself does not introduce significant errors beyond those inherent in the qubit’s natural decay. This high fidelity is essential for reliable quantum computation. The team’s approach also accounts for the qubit-induced Kerr nonlinearity, a phenomenon that can distort the resonator’s response and introduce errors. “We demonstrate more than a 10^3-fold residual cavity field reduction immediately after the pulse,” the authors write, highlighting the efficiency of their reset mechanism.

The team successfully demonstrated this by achieving a greater than 30 decibel cavity reset for the three-level system, with a pulse duration of 1 microsecond. This ability to rapidly reset the resonator, coupled with the analytical nature of the pulse generation, positions this technique as a promising approach for building faster and more reliable quantum computers. The device used in the experiment was fabricated at Myfab Chalmers, and a detailed description of the experimental setup is available in the supporting information.

Qutrit Readout Generalization & Cavity Field Reduction

Researchers have achieved this combined operation using superconducting qubits and a dispersive readout method, moving beyond the need for sequential measurement and reset steps. This approach relies on precisely shaped microwave pulses, analytical pulses, designed to null the cavity field, and crucially, requires only knowledge of the qubit and resonator parameters, bypassing complex pulse optimization routines. The team’s method extends beyond simple qubit, or two-level, systems.

This ability to rapidly return the resonator to its initial state is vital for maintaining coherence and accelerating quantum computations. The process begins with defining a desired cavity response, a, that is zero before and after a specific pulse duration. By applying an inverse Fourier transform, the necessary input field is determined, effectively canceling out the residual cavity field. This substantial reduction in residual field noise directly translates to improved measurement accuracy and reduced decoherence. The resonator resonated at 6041.200 MHz with a linewidth of 0.5647(3) MHz, corresponding to a decay time of 281.2(8) nanoseconds.

👉 More information
🗞 Dispersive Qubit Readout with Intrinsic Resonator Reset
✍️ M. Jerger et al.
🧠 DOI: http://link.aps.org/doi/10.1103/z5x1-2mr9

More like this