Quantum computing relies on the operation of qubits in an environment as free of noise as possible. Assessing the quality of this environment is a key aspect of ensuring high-fidelity implementations based on superconducting qubits. Relaxation, decoherence, dephasing, and quasiparticle tunneling rates have been measured for various shielding configurations used in the measurement environment for state-of-the-art transmon qubits. An ensemble of approximately 120 control devices was used for this study, with five different capacitor pad designs. The shielding elements varied in the configuration included an indium gasket at the qubit can's lid, Cryoperm magnetic shielding, the mixing chamber shield of the dilution refrigerator, the inclusion of a vacuum pump-out port, and capping unused subminiature version A connectors at the top of the measurement can's lid. It was found that the qubit lifetimes T1, T2, and Tϕ are robust to the all of configuration changes tried until the mixing chamber shield was removed, significantly increasing blackbody radiation levels in the qubit measurement space, where in that limit it was found that tapering the qubit pads reduced the amount of loss. In contrast, the quasiparticle tunneling rates were found to be extremely sensitive to all configuration changes tested. Consistent with earlier reports [McEwen et al., arXiv:2104.05219 (2021); Cardani et al., Nat. Commun. 12, 2733 (2021); Wilen et al., Nature 594, 369–373 (2021); Ristè et al. Nat. Commun. 4, 1913 (2013)], the findings from this study indicate that non-equilibrium quasiparticles do not currently limit the lifetimes of well-shielded transmon qubits.
Read full abstract