Measurements of resistivity switching in Josephson junctions and nanowires are widely used in the experimental superconductivity community, to attain insight into decay processes of metastable states, to test qubits, or to detect single photons. Such measurements are commonplace, but not easy. The authors formulate and experimentally verify the criterion for clean, independent switching measurement. They study the correlated regime, where the result of the test for one pulse affects the outcome for the following pulses, showing how to introduce thermal correlation into the ``flipping coin'' experiment in a controlled manner, and analyzing its implications for the switching statistics.
Measurements of resistivity switching in Josephson junctions and nanowires are widely used in the experimental superconductivity community, to attain insight into decay processes of metastable states, to test qubits, or to detect single photons. Such measurements are commonplace, but not easy. The authors formulate and experimentally verify the criterion for clean, independent switching measurement. They study the correlated regime, where the result of the test for one pulse affects the outcome for the following pulses, showing how to introduce thermal correlation into the ``flipping coin'' experiment in a controlled manner, and analyzing its implications for the switching statistics.
Superconducting weak link (WL), acting as a Josephson junction (JJ), is one of the widely used elements in superconductor science and quantum circuits. A hysteretic JJ with robust switching between its superconducting and resistive state is an excellent candidate for single-photon detection. However, the ubiquitous fluctuations in the junction strongly influence the stability of the states and, thus, the transition from one to the other. Here, we present an experimental study of switching statistics of critical and retrapping currents of a JJ based on niobium WL in its hysteretic regime. The mean lifetimes of the two metastable states, namely, the zero-voltage superconducting state and finite-voltage resistive state, are estimated from the distributions. Further, close to the hysteresis crossover temperature, observed telegraphic noise in the time domain due to random switching between the states provides their lifetimes directly. We present a thermal model introducing a double-well (bistable) feature with an intriguing quantity with respect to the devices’ temperature states. The effects of temperature fluctuations on the stability of the states are shown. We discuss our results toward further improvement of the efficiency of superconducting WL or nanowire single-photon detectors.
Measuring the dynamics of a quantum bit (qubit) relies on the accurate detection of the quantum state of the system. A widely used method to measure the state of a solid state Josephson junction qubit is to measure the switching current of a Josephson device. This work investigates the measurement of the switching current of SQUID samples by means of fast current pulses. The response of a SQUID to a square current pulse has to be measured at the top of a dilution refrigerator through long cables, resulting in bandwidth limitations. A switch in the last instance of a pulse will not be detected, resulting in uncertainties in the detection. We explain how a square bias pulse that is directly followed by a hold level of lower amplitude can be used to circumvent the bandwidth limitations by latching the state of the system it was in after the bias pulse. This corresponds to a sample and hold measurement. Every single measurement in a quantum mechanical probability measurement has to be statistically independent. We show correlation measurements for di erent settings of the pulse parameters and at di erent magnitudes of the switching current. A gure of merit for a quantum detector is its resolution. The measurements show that with the sample and hold technique good current resolutions can be obtained, even at very small magnitudes and short pulse durations. In order to make a fast measurement of the switching current, the switching process must occur during the bias pulse. We show in both measurements and computer simulations that a fast switch pulse can induce switching by the hold level,even when the hold level was initially adjusted to a value where it never switched the sample. The computer simulations show that by choosing the hold amplitude low enough, switching occurs rapidly, determined by the bias pulse alone.
Quantum phase slips (QPS) or the macroscopic quantum tunneling (MQT) of a nanowire’s order parameter through an activation energy barrier have remained the subject of intense debate for many years. They are expected to occur at low enough temperature where thermally activated phase slips (TAPS) have been frozen out in analogy with Josephson junctions where the macroscopic tunneling of phase at low temperatures has been conclusively experimentally demonstrated. We address this question by following a similar experimental strategy to that employed for establishing MQT in JJs. By measuring switching current distributions, we can probe the phase slip rate and determine if it corresponds to thermal activation or quantum tunneling. Having established that the behavior we see is consistent with being in the quantum regime, we can alter properties of the nanowires and see if the response is consistent with the expectations of the quantum model. To do this we employ an in-situ modification technique using high bias voltage pulses. Using these pulses, we can change resistance, critical temperature, critical current and morphology of the nanowire. We can also change the shunting capacitance of the nanowire by altering the photolithography step used to create the nanowires electrodes. The resulting response of the nanowires agrees well with being in the quantum tunneling dominated regime. An interesting side benefit of the pulsing technique is that we can exactly set the nanowires switching current to a desired value. This may be instrumental in the development of superconducting nanowire qubits.
Superconducting quantum circuits (SQCs) are being explored as model systems for scalable quantum computing architectures. Josephson junctions are extensively used in superconducting quantum interference devices (SQUIDs) and in persistent-current qubit systems. Noise excitations, however, have a critical influence on their dynamics. Thus, the primary focus of this research was to investigate the effects of thermal ac-tivation on the superconducting properties of Josephson junctions. Specifically, ther-mal noise tends to result in a range of switching currents, values less than the critical current at which a junction switches from the superconducting to the normal state. First, a general review of superconductivity concepts is given, including a treatment of the Josephson phenomena. Next, I describe some of my work on characterizing the current-voltage traces of Josephson junctions tested at 4 K with a Multi-Chip Probe (MCP). Then, I describe thermal activation theory and examine the equations useful for modeling switching current distributions. The Josephson junctions of a SQUID with a ramped bias current were tested for numerous temperatures T < 4.5 K (and
Macroscopic quantum tunneling is a process by which a macroscopic object, rather than a single electron, tunnels between two macroscopically distinct quantum states. In this dissertation, we will present the study of macroscopic quantum tunneling of phase-slips in superconducting devices, including one-dimensional nanowires, quasi-two-dimensional superconducting strips and doubly connected superconducting devices composed of two nanowires connected in parallel. We observe macroscopic quantum tunneling in these devices by measuring switching current distributions. It is known that the standard deviation of switching current distributions can be measured on a single nanowire to reveal temperatures at which macroscopic quantum tunneling is responsible for phase-slips. Therefore we begin by studying higher moments of the switching current distributions, namely the skewness S (a measure of the asymmetry of a distribution) and kurtosis K (a measure of its peakedness). We find that the skewness and kurtosis of the switching current distributions obtained via the Kurkijarvi process on devices composed of single nanowires do not depend on whether the switching events are initiated by quantum or thermal phase-slips. Skewness and kurtosis deviate considerably from the values associated with a Gaussian distribution (S=0 and K=3). If, in an experiment these higher moments approach Gaussian values they indicate the existence of noise. Next we study macroscopic quantum tunneling of phase-slips in quasi-two-dimensional superconducting strips, which are commercially available devices used as single-photon detectors. These devices are composed of (250 um) long, (120 nm) wide, meandering superconducting strips. Each time a photon with sufficient energy strikes the strip a voltage pulse is produced and counted. However, the accuracy of these detectors is limited by a rate of dark counts (false events). We find that at sufficiently low temperatures, the macroscopic quantum tunneling of a vortex through an edge barrier on the wire (which results in a phase-slip as the vortex crosses the wide wire) contributes a base-level dark count rate in these detectors which must be considered during operation at low enough temperatures. After studying tunneling in single one-dimensional nanowires and wide, quasi-two-dimensional strips, we move our focus to doubly connected superconducting samples composed of two nanowires connected in parallel. We characterize and model these devices, and show their applicability as nanometer-scale superconducting memory cells. We develop precise algorithms allowing us to write and read the information onto such memory cells. We also observe signatures of macroscopic quantum tunneling in these doubly connected devices by observing a saturation in the standard deviation of switching current distributions at low temperatures. We then discuss how macroscopic quantum tunneling of the memory state can lead to a dissipationless operation of the superconducting memory cell. In each of these devices, whether single one-dimensional nanowires, wide quasi-two-dimensional superconducting strips, or two nanowires in parallel, we find, quite surprisingly, that the same equations can be used to model macroscopic quantum tunneling at the high currents we apply. Therefore we argue that the physics of macroscopic quantum tunneling in wide quasi-two-dimensional strips and doubly connected nanowire devices can be well understood in the context of tunneling in a single, one-dimensional nanowire.
Gate-tunable Josephson junctions (JJs) are the backbone of superconducting classical and quantum computation. Typically, these systems exploit low-charge-concentration materials and present technological difficulties limiting their scalability. Surprisingly, electric field modulation of a supercurrent in metallic wires and JJs has been recently demonstrated. Here, we report the realization of titanium-based monolithic interferometers which allow tuning both JJs independently via voltage bias applied to capacitively coupled electrodes. Our experiments demonstrate full control of the amplitude of the switching current ( I s ) and of the superconducting phase across the single JJ in a wide range of temperatures. Astoundingly, by gate-biasing a single junction, the maximum achievable total I s is suppressed down to values much lower than the critical current of a single JJ. A theoretical model including gate-induced phase fluctuations on a single junction accounts for our experimental findings. This class of quantum interferometers could represent a breakthrough for several applications such as digital electronics, quantum computing, sensitive magnetometry, and single-photon detection.