An accurate understanding of the Josephson effect is the keystone of quantum information processing with superconducting hardware. Here we show that the celebrated sinφ current-phaserelation (CφR) of Josephson junctions (JJs) fails to fully describe the energy spectra of transmon artificial atoms across various samples and laboratories. While the microscopic theory of JJs contains higher harmonics in the CφR, these have generally been assumed to give insignificant corrections for tunnel JJs, due to the low transparency of the conduction channels. However, this assumption might not be justified given the disordered nature of the commonly used AlOx tunnel barriers. Indeed, a mesoscopic model of tunneling through an inhomogeneous AlOx barrier predicts contributions from higher Josephson harmonics of several %. By including these in the transmon Hamiltonian, we obtain orders of magnitude better agreement between the computed and measured energy spectra. The measurement of Josephson harmonics in the CφR of standard tunnel junctions prompts a reevaluation of current models for superconducting hardware and it offers a highly sensitive probe towards optimizing tunnel barrier uniformity.
Constructing a quantum computer requires immensely precise control over a quantum system. A lack of precision is often quantified by gate-error metrics, such as the average infidelityor the diamond distance. However, usually such gate-error metrics are only considered for individual gates, and not the errors that accumulate over consecutive gates. Furthermore, it is not well known how susceptible the metrics are to the assumptions which make up the model. Here, we investigate these issues using realistic simulation models of quantum computers with flux-tunable transmons and coupling resonators. We show that the gate-error metrics are susceptible to many of the assumptions which make up the model. Additionally, we find that consecutive gate errors do not accumulate linearly. Previous work showed that the gate-error metrics are poor predictors for the performance of consecutive gates. Here, we provide further evidence and a concise theoretical explanation for this finding. Furthermore, we discuss a problem that potentially limits the overall scaling capabilities of the device architecture we study in this work.
We develop a simulator for quantum computers composed of superconducting transmon qubits. The simulation model supports an arbitrary number of transmons and resonators. Quantum gatesare implemented by time-dependent pulses. Nontrivial effects such as crosstalk, leakage to non-computational states, entanglement between transmons and resonators, and control errors due to the pulses are inherently included. The time evolution of the quantum computer is obtained by solving the time-dependent Schrödinger equation. The simulation algorithm shows excellent scalability on high-performance supercomputers. We present results for the simulation of up to 16 transmons and resonators. Additionally, the model can be used to simulate environments, and we demonstrate the transition from an isolated system to an open quantum system governed by a Lindblad master equation. We also describe a procedure to extract model parameters from electromagnetic simulations or experiments. We compare simulation results to experiments on several NISQ processors of the IBM Q Experience. We find nearly perfect agreement between simulation and experiment for quantum circuits designed to probe crosstalk in transmon systems. By studying common gate metrics such as the fidelity or the diamond distance, we find that they cannot reliably predict the performance of repeated gate applications or practical quantum algorithms. As an alternative, we find that the results from two-transmon gate set tomography have an exceptional predictive power. Finally, we test a protocol from the theory of quantum error correction and fault tolerance. We find that the protocol systematically improves the performance of transmon quantum computers in the presence of characteristic control and measurement errors.
The real-time dynamics of systems with up to three SQUIDs is studied by numerically solving the time-dependent Schrödinger equation. The numerical results are used to scrutinize themapping of the flux degrees of freedom onto two-level systems (the qubits) as well as the performance of the intermediate SQUID as a tunable coupling element. It is shown that the two-level representation yields a good description of the flux dynamics during quantum annealing, and the presence of the tunable coupling element does not have negative effects on the overall performance. Additionally, data obtained from a two-level spin dynamics simulation of quantum annealing is compared to experimental data produced by the D-Wave 2000Q quantum annealer. The effects of finite temperature are incorporated in the simulation by coupling the qubit-system to a bath of spin-1/2 particles. It is shown that including an environment modeled as non-interacting two-level systems that couple only to the qubits can produce data which matches the experimental data much better than the simulation data of the isolated qubits, and better than data obtained from a simulation including an environment modeled as interacting two-level systems coupling to the qubits.