Session A39: Superconducting Circuits: Measurement I

Fast, high-fidelity, QND measurements of superconducting qubits using a transverse interaction?

The standard procedure for measuring superconducting qubits is to couple them to a cavity mode via a tranverse interaction, and to detune the cavity from the qubit to obtain an effective "dispersive" interaction in which the qubit shifts the frequency of the cavity. This allows the state of the qubit to be read out by measuring the phase shift of a signal that is passed through the cavity. This technique is often referred to as providing a quantum non-demolition (QND) measurement, since the effective dispersive interaction has the QND form, but in fact it will only approximate a QND measurement to any reasonable degree when used with a detuning so large as to significantly reduce the effective interaction strength and thus the measurement speed. As a result, measurements that use this technique so as to achieve state-of-the-art speeds (e.g., below 100 ns) are not operated in the QND regime. Here we show that there is signifcantly more to understand about the transverse interaction than the received wisdom that we have just described. In particualr we show that by using a circuit configuration in which the qubit/cavity interaction strength can be controlled in a time-dependent fashion, and by carefully choosing the time-envelope of this strength, it is possible to exploit a recurrance phenomena in the transferverse interaction to acheive fast QND measurements. This technique appears to be quite feasible even though it requires fairly precise control. We will discuss a circuit design that could be used to implement it as well as some further properties of the transverse interaction.   

High Fidelity Measurement of Single and Multi-qubit Superconducting Quantum Circuits

Fast, high fidelity measurements of multi-qubit devices are crucial for quantum error correction and fault tolerant quantum computation. In superconducting qubits, dispersive circuit-QED readout using a Josephson parametric amplifier (JPA) has been the common approach to obtain high-fidelity QND measurement. A standard way to improve speed and/or measurement fidelity is to use higher measurement power. However, that often leads to unwanted transitions both within and outside of the computational subspace, resulting in reduced fidelity. Moreover, this problem can get exacerbated when performing joint readout of multiple qubits. Using the 3D circuit-QED architecture, we explore the optimization of various parameters like measurement power, qubit-cavity coupling, qubit-cavity detuning to enhance measurement fidelity in both single and multi-qubit circuits. We will also discuss the optimization of the JPA so that it can handle the larger measurement signals needed for high fidelity.   

High fidelity dispersive qubit readout with two-mode squeezed light

High fidelity qubit measurement is essential for scalable, fault-tolerant quantum computing. In superconducting circuits, qubit readout with fidelity above 99% has been achieved by using a quantum-limited parametric amplifier such as the Josephson Parametric Converter (JPC) as the first stage amplifier. However, the Signal-to-Noise Ratio (SNR) of such readout is fundamentally limited by quantum fluctuations in the coherent readout pulse. Alternatively, readout with squeezed light can be used to reduce fluctuation along certain quadratures and thus improve the SNR. In this talk, we investigate a readout scheme with two-mode squeezed light both produced and amplified by JPCs in a simple interferometer unbalanced by a transmon qubit/cavity. This configuration has been predicted to improve the SNR compared to readout with both coherent states and single-mode squeezed light [1]. This readout method, which sends two-mode squeezed light along two paths simultaneously is also of interest for remote entanglement of qubits. We will present preliminary data and discuss optimization in the presence of JPC imperfections and photon loss.

[1] Sh. Barzanjeh et al., PRB 90, 134515 (2014).   

To catch and reverse a quantum jump mid-flight

A quantum system driven by a weak deterministic force while under strong continuous observation exhibits quantum jumps between its energy levels. Employing a three-level superconducting artificial atom of the V-type involved in the original observation of quantum jumps, we show that quantum jumps can be caught and even reversed mid-flight. The three required levels are: G (for Ground), B (for Bright), and D (for Dark). The D level is engineered to be decoupled from both any dissipative environment and any measurement apparatus. Quantum jumps between G and D are monitored indirectly by the combination of a Rabi drive between the G and B levels, together with the monitoring of the occupation of B, itself tracked by a dispersively-coupled readout cavity. Using digital low-latency feedback electronics, we demonstrate the catch of a quantum jump mid-flight, i.e. a coherent superposition of D (corresponding to having jumped) and G (corresponding to having not jumped). The fidelity of the mid-flight state is above 70%, in agreement with quantum trajectory theory simulations. Our monitoring scheme can be useful for other quantum information tasks, such as the continuous monitoring of error syndromes.   

Qubit State Measurement using the Josephson Photomultiplier

We describe results of high-fidelity qubit state measurement based on microwave photon counting. The protocol involves mapping the qubit state onto high-contrast microwave cavity pointer states in a linear readout resonator followed by subsequent photodetection using the Josephson Photomultiplier (JPM). Our technique provides access to the classical binary outcome of qubit state measurement at the millikelvin stage of a dilution refrigerator and does not use any non-reciprocal components between the qubit and JPM. We exploit the intrinsic damping of the JPM to “clean up” following photodetection so that the measurement is highly QND. This approach is well-suited for integration with proximal cryogenic digital logic.   

Non-Dispersive Limit of the Qubit Readout with a Microwave Photon Counter.

The Josephson photomultiplier (JPM) provides a path to fast, high fidelity qubit readout [Govia et al, Phys. Rev. A 90, 062307 (2014)]. The measurement protocol requires the creation of cavity pointer states with high contrast in photon occupation. For such large pointer states the dispersive approximation to the Jaynes-Cummings Hamiltonian breaks down. In this regime, the non-dispersive terms reduce the pointer-states contrast and lead to backaction of the measurement on the qubit state, reducing the readout accuracy and affecting its QND nature. We discuss the optimal conditions for JPM-based qubit readout in terms of the microwave drive parameters, the cavity photon decay time, and the detuning between the cavity and qubit. We also show that strong cavity drive facilitates reset of the readout system.   

Single-Shot Quantum Non-Demolition Detection of Itinerant Microwave Photons

We realize a quantum non-demolition detector for propagating single photons in the microwave domain. By using a cavity-assisted conditional phase flip gate between a photon and a superconducting artificial atom, the information of the presence of an itinerant photon is mapped into a definite state of the qubit [1]. We test the detector with a single photon source and perform single-shot readout of the state of the atom, whose first to second excited state transition is tuned to be resonant with the photon source. We demonstrate the quantum non-demolition nature of the detector by characterizing the field of the reflected photon, which is never absorbed. The internal detection fidelity is limited by the coherence properties of the qubit, but is independent of the temporal shape of the photon as long as its frequency bandwidth is smaller than the coupling of the atom to the assisting cavity.
[1] Reiserer et al., Nondestructive Detection of an Optical Photon, Science 342, 1349 (2013).   

Quantum non-demolition detection of an itinerant microwave photon

Microwave quantum optics in superconducting circuits enables us to investigate unprecedented regimes of quantum optics. The strong nonlinearity brought by Josephson junctions together with the strong coupling of the qubits with microwave modes reveals rich physics not seen in the optical domain before. However, single-photon detection in the microwave domain is still a challenging task because of the photon energy four to five orders of magnitude smaller than in optics.
Here, we demonstrate a quantum non-demolition detection of an itinerant microwave photon using a circuit QED architecture with a transmon qubit in a largely detuned 3D cavity. When an itinerant photon is reflected by the cavity, it interacts dispersively with the qubit. With an appropriate adjustment of the cavity bandwidth, the qubit deterministically acquires a phase flip upon the reflection. With a single-shot measurement of the phase flip, we detect the existence of the single photon without destroying it. We evaluate the quantum efficiency of the detection and characterize the reflected photon by using conditional quantum state tomography based on the outcome of the qubit readout.   

Parity measurements using parametrically driven resonators: Part I

Multi-qubit parity measurements are indispensable for quantum error correction. A possible way to realize these is by homodyne measurement of the field of a cavity which is dispersively coupled to multiple qubits. However such an interaction can lead to dephasing within the parity subspace and results in only half-parity measurement. With the aim of overcoming this shortcoming, in this talk we will introduce the parametrically driven nonlinear resonator as an alternate approach to measure multi-qubit parity. With analytical results, we will show that this approach leads to significant reduction of parity-subspace dephasing making full-parity measurements possible.   

Parity measurement using parametrically driven resonators - part 2

Non-destructive measurement of multi-qubit parity stabilizers is essential in quantum error correction. For superconducting qubits, one way to perform such parity measurements is to use an ancilla qubit and multiple two-qubit gates. However, this increases the required number of qubits for a given code and the need for two-qubit gates makes this measurement slow.
We propose to directly measure the parity of multiple qubits through a single parametrically driven nonlinear resonator. This would reduce the experimental footprint and at the same time make the parity measurements faster. In this talk, we will present numerical simulations showing that this measurement is fast, high-fidelity and that it preserves the measurement eigenspaces.   

Implementation of Continuous Parity Measurements and Error Correction

Continuous monitoring of quantum systems has been successfully used to examine state collapse and quantum jumps in single qubits. Continuous parity measurements allow for the observation of collapse dynamics in multiqubit systems, and naturally motivate a strategy for performing quantum error correction. One can observe the parity of two dispersive superconducting transmons without the need for ancilla qubits by strongly coupling the former to a joint readout resonator. With three qubits pairwise coupled to two resonators, we can measure two generators of the conventional three-qubit bit-flip code simultaneously. Using high-speed field programmable gate array electronics to continuously monitor these parities, we observe multipartite collapse and can apply correction pulses when an error is detected.   

Parametrically induced readout of a superconducting qubit

One of the basic operations of quantum information is to non-destructively identify the state of a quantum system with high fidelity and within a time that is short compared to its natural decay. In circuit quantum electro-dynamics, this goal is achieved by the dispersive readout scheme, in which the frequency of an ancillary cavity is pulled by an amount depending on the qubit state. Despite its successes, this scheme suffers from the Purcell limit set on the decay time, and from the dephasing induced by spurious photons in the readout cavity. We address these shortcomings with a parametrically driven 3rd order interaction between the qubit excitation number and the cavity coherent state displacement. This interaction can also be described as a so-called “longitudinal” coupling between the qubit z operator and the readout cavity field amplitude, and it differs markedly from the usual "transversal" Jaynes-Cummings coupling between the qubit x operator and the field amplitude. This interaction can be tuned in-situ, and can be turned off altogether. The presentation will report experimental progress towards the realization of this scheme.
  

Fast and Efficient All-Microwave Reset of a Transmon-Qutrit Coupled to a Large Bandwidth Resonator

Fast and efficient reset of qubits is a key operation in many quantum algorithms, and particularly in error correction codes. We experimentally demonstrate a reset scheme of a qutrit coupled to a low Q resonator. The reset protocol uses the microwave induced interaction between the |f,0〉 and |g,1〉 states of the system, with |g〉 and |f〉 denoting the ground and second excited state of the qutrit, and |0〉 and |1〉 the photon Fock states of the resonator. We characterize the reset process and demonstrate reinitialization of the transmon-resonator system to its ground state with 99.8% fidelity in less than 500 ns. Our protocol is of practical interest as it has no requirements on the chip design, in addition to those for fast and efficient single-shot readout of the transmon [1], and does not require any feedback for initialization.
[1] T. Walter et al., Phys. Rev. Applied 7, 054020