[en] It is sketched how a monostable rf- or dc-SQUID can mediate an inductive coupling between two adjacent flux qubits. The non-trivial dependence of the SQUID's susceptibility on external flux makes it possible to continuously tune the induced coupling from antiferromagnetic (AF) to ferromagnetic (FM). In particular, for suitable parameters, the induced FM coupling can be sufficiently large to overcome any possible direct AF inductive coupling between the qubits. The main features follow from a classical analysis of the multi-qubit potential. A fully quantum treatment yields similar results, but with a modified expression for the SQUID susceptibility. Since the latter is exact, it can also be used to evaluate the susceptibility-or, equivalently, energy-level curvature-of an isolated rf-SQUID for larger shielding and at degenerate flux bias, i.e., a (bistable) qubit. The result is compared to the standard two-level (pseudospin) treatment of the anticrossing, and the ensuing conclusions are verified numerically

[en] We have designed, fabricated and tested an XY-addressable readout system that is specifically tailored for the reading of superconducting flux qubits in an integrated circuit that could enable adiabatic quantum optimization. In such a system, the flux qubits only need to be read at the end of an adiabatic evolution when quantum mechanical tunneling has been suppressed, thus simplifying many aspects of the readout process. The readout architecture for an N-qubit adiabatic quantum optimization system comprises N hysteretic dc SQUIDs and N rf SQUID latches controlled by 2√N+2 bias lines. The latching elements are coupled to the qubits and the dc SQUIDs are then coupled to the latching elements. This readout scheme provides two key advantages: first, the latching elements provide exceptional flux sensitivity that significantly exceeds what may be achieved by directly coupling the flux qubits to the dc SQUIDs using a practical mutual inductance. Second, the states of the latching elements are robust against the influence of ac currents generated by the switching of the hysteretic dc SQUIDs, thus allowing one to interrogate the latching elements repeatedly so as to mitigate the effects of stochastic switching of the dc SQUIDs. We demonstrate that it is possible to achieve single-qubit read error rates of < 10^-6 with this readout scheme. We have characterized the system level performance of a 128-qubit readout system and have measured a readout error probability of 8 x 10^-5 in the presence of optimal latching element bias conditions.

[en] We have designed, fabricated and operated a scalable system for applying independently programmable time-independent, and limited time-dependent flux biases to control superconducting devices in an integrated circuit. Here we report on the operation of a system designed to supply 64 flux biases to devices in a circuit designed to be a unit cell for a superconducting adiabatic quantum optimization system. The system requires six digital address lines, two power lines, and a handful of global analog lines.

[en] A practical strategy for synchronizing the properties of compound Josephson junction (CJJ) radio frequency monitored superconducting quantum interference device (rf-SQUID) qubits on a multi-qubit chip has been demonstrated. The impact of small (∼1%) fabrication variations in qubit inductance and critical current can be minimized by the application of a custom-tuned flux offset to the CJJ structure of each qubit. This strategy allows for a simultaneous synchronization of the qubit persistent current and tunnel splitting over a range of external bias parameters that is relevant for the implementation of an adiabatic quantum processor.

[en] Superconducting microresonators have been successfully utilized as detection elements for a wide variety of applications. With multiplexing factors exceeding 1000 detectors per transmission line, they are the most scalable low-temperature detector technology demonstrated to date. For high-throughput applications, fewer detectors can be coupled to a single wire but utilize a larger per-detector bandwidth. For all existing designs, fluctuations in fabrication tolerances result in a non-uniform shift in resonance frequency and sensitivity, which ultimately limits the efficiency of bandwidth utilization. Here, we present the design, implementation, and initial characterization of a superconducting microresonator readout integrating two tunable inductances per detector. We demonstrate that these tuning elements provide independent control of both the detector frequency and sensitivity, allowing us to maximize the transmission line bandwidth utilization. Finally, we discuss the integration of these detectors in a multilayer fabrication stack for high-speed readout of the D-Wave quantum processor, highlighting the use of control and routing circuitry composed of single-flux-quantum loops to minimize the number of control wires at the lowest temperature stage