Double quantum dots (DQDs), sometimes referred to as ``artificial molecules'', are highly controllable quantum systems that serve as building blocks for spin-based quantum computation. A fast, high-fidelity charge sensor is needed for determining the number of electrons trapped in the DQD, and for spin-state readout. Here the authors integrate a Josephson parametric amplifier into the readout chain, improving the signal to noise ratio by a factor of 2000 and enabling live tuning of the DQD potential well, without the slow, painstaking adjustment of voltages on several electrodes.
Double quantum dots (DQDs), sometimes referred to as ``artificial molecules'', are highly controllable quantum systems that serve as building blocks for spin-based quantum computation. A fast, high-fidelity charge sensor is needed for determining the number of electrons trapped in the DQD, and for spin-state readout. Here the authors integrate a Josephson parametric amplifier into the readout chain, improving the signal to noise ratio by a factor of 2000 and enabling live tuning of the DQD potential well, without the slow, painstaking adjustment of voltages on several electrodes.
Spins in silicon quantum devices are promising candidates for large-scale quantum computing. Gate-based sensing of spin qubits offers a compact and scalable readout with high fidelity, however, further improvements in sensitivity are required to meet the fidelity thresholds and measurement timescales needed for the implementation of fast feedback in error correction protocols. Here, we combine radio-frequency gate-based sensing at 622 MHz with a Josephson parametric amplifier, that operates in the 500-800 MHz band, to reduce the integration time required to read the state of a silicon double quantum dot formed in a nanowire transistor. Based on our achieved signal-to-noise ratio, we estimate that singlet-triplet single-shot readout with an average fidelity of 99.7% could be performed in 1 μs, well below the requirements for fault-tolerant readout and 30 times faster than without the Josephson parametric amplifier. Additionally, the Josephson parametric amplifier allows operation at a lower radio-frequency power while maintaining identical signal-to-noise ratio. We determine a noise temperature of 200 mK with a contribution from the Josephson parametric amplifier (25%), cryogenic amplifier (25%) and the resonator (50%), showing routes to further increase the readout speed.
In microwave-based quantum circuits, including double quantum dots (DQDs), superconducting qubits and spin qubits, parametric amplifiers are indispensable in achieving high-fidelity qubit readouts. Despite its importance, the application of parametric amplifiers is hampered by several challenges, such as high insertion losses, constrained tunability, and a pronounced vulnerability to magnetic fields. Here, we demonstrate an on-site single-atom parametric amplifier (SAPA) within a reconfigurable quantum circuit, which consists of a superconducting microwave cavity and two GaAs gate-defined DQDs. Leveraging the inherent nonlinearity of the DQD, a parametric gain exceeding 11 dB is achieved. This gain contributes to enhance the qubit readout, as evidenced by exceeding two times improvement in the signal-to-noise ratio (SNR) when employing the DQD-based amplifier for reading out another DQD. Our work not only presents a versatile experimental platform with enhanced readout capabilities in quantum computing, but also introduces alternative choices of parametric amplifiers for a variety of microwave-based quantum circuits.
The need for fast readout to measure the state of a quantum dot device has important applications in the capabilities of using them as a means for quantum information processing. Traditional charge sensors used to readout the charge state of a quantum dot device are very robust and sensitive electrometers that have been in use since the late 1980s. They pose an issue to the future of quantum dots as a viable means of quantum computing, and that is in their slow acquisition times of the charge state of the quantum dot. This rate of measurement means that implementing any quantum error correction scheme becomes very difficult or even unfeasible as the coherence times of these semiconductor systems is short compared to charge sensor’s acquisition times. A possibility to speed up the measurement process of the charge state is possible thanks to the atomic molecular and optical (AMO) physics community.Cavity quantum electrodynamics (CQED) is the study of light-matter interactions when both the light and matter are taken to be quantized. A photon may interact with an atom via a dipole interaction to excite it from its ground state to its excited state, and the reverse can occur where the atom emits a photon and decays from its excited state ii to the ground state. This fundamental interaction which has been extensively studied in the AMO community since the early 1960s had been adopted by the superconducting qubit community to measure the state of their qubits. These superconducting qubits, representing atoms, which are coupled to a resonant structure on chip, which represents the photons that can interact with the artificial atom. Since these systems were not traditional trapped or neutral atoms interacting with a photon via a laser, but instead planar circuit structures, the superconducting qubit community re-coined the term cavity quantum electrodynamics to circuit quantum electrodynamics, which we will refer to as cQED.The quantum dot community adapted these methods of what is known as dispersive readout; a capacitive change of the atomic like system, results in a dispersive shift which then alters the characteristic frequency response of the resonant structure. This has allowed for a new kind of charge sensor by means of measuring the resonators response due to its coupling to a quantum dot system. This can be done with very high frequency microwave circuitry, allowing much faster readout than is capable via traditional charge sensors. Thus, allowing a pathway to implement quantum error correction schemes with reasonable times compared to any dephasing or decoherence times of the semiconductor system. Furthermore, such fast readout allows for feedback protocols to be implemented to the system to correct for certain undesirable effects due to instruments or the system itself. Our work focuses on a limit of this cQED interaction of the quantum dots to a resonant structure allowing us to do fast readout of the quantum dot charge states via a dispersive shift.The work I will demonstrate in this dissertation will be in the efforts of fast readout on the charge occupation of a double quantum dot system which is galvanically coupled to a superconducting microwave resonator. This was done in a Si/SiGe heterostructure system where I designed, simulated, and fabricated the device which was optimized to have a stronger dispersive shift signal due to an enhanced gate lever arm that couples to the resonator. I will discuss our results of attaining a minimum integration time of 34.54 ns for a unity signal to noise ratio. This demonstrates a fast readout system in the community iii which could be extended to allow for even better readout speeds with faster instruments than what were available to me for the experiment. I will further discuss our characterization of the hybrid system and how we extracted a charge noise due to charging events between the two dots in our system up to around 10 kHz. This work demonstrates a robust dispersive readout scheme which helps in the scalability of hybrid superconducting-semiconducting architectures for means of quantum information processing.
We demonstrate the parametric modulation of a double quantum dot charge dipole coupled to a cavity, at the cavity frequency, achieving an amplified readout signal compared to conventional dispersive protocols. Our findings show that the observed cavity field displacement originates from dipole radiation within the cavity, rather than from a longitudinal coupling mechanism, yet exhibits the same signatures while relying on a transverse coupling. By carefully tuning the phase and amplitude of the intracavity field, we achieve a π-phase shift between two dipole states, resulting in a substantial enhancement of the signal-to-noise ratio. In addition to its applications in quantum dot based qubits in circuit QED architectures, this protocol could serve as a new promising tool for probing exotic electronic states in mesoscopic circuits embedded in cavities.
Spins in semiconductor quantum dots hold great promise as building blocks of quantum processors. Trapping them in Si MOS transistorlike devices eases future industrial-scale fabrication. Among the potentially scalable readout solutions, gate-based dispersive radio-frequency reflectometry only requires the already existing transistor gates to read a quantum dot state, which alleviates the need for additional elements. In this effort toward scalability, traveling-wave superconducting parametric amplifiers significantly enhance the readout signal-to-noise ratio (SNR) by reducing the noise below that of typical cryogenic low-noise amplifiers, while offering a broad amplification band, essential to multiplex the readout of multiple resonators. In this work, we demonstrate a 3-GHz gate-based reflectometry readout of electron charge states trapped in quantum dots formed in Si MOS multigate devices, with SNR enhanced due to a Josephson traveling-wave parametric amplifier (JTWPA). The broad tunable 2-GHz amplification bandwidth combined with more than 10-dB ON-OFF SNR improvement of the JTWPA enables frequency and time-division-multiplexed readout of interdot transitions, and noise performance consistent with JTWPA characteristics at higher frequencies. In addition, owing to a design without superconducting loops and with a metallic ground plane, the JTWPA is flux insensitive and shows stable performance up to a magnetic field of 1.1 T at the quantum dot device, compatible with standard Si MOS spin-qubit experiments.
Josephson parametric amplifiers (JPAs) approaching quantum-limited noise performance have been instrumental in enabling high fidelity readout of superconducting qubits and, recently, semiconductor quantum dots (QDs). We propose that the quantum capacitance arising in electronic two-level systems (the dual of Josephson inductance) can provide an alternative dissipation-less non-linear element for parametric amplification. We experimentally demonstrate phase-sensitive parametric amplification using a QD-reservoir electron transition in a CMOS nanowire split-gate transistor embedded in a 1.8~GHz superconducting lumped-element microwave cavity, achieving parametric gains of -3 to +3 dB, limited by Sisyphus dissipation. Using a semi-classical model, we find an optimised design within current technological capabilities could achieve gains and bandwidths comparable to JPAs, while providing complementary specifications with respect to integration in semiconductor platforms or operation at higher magnetic fields.