Recovery dynamics of a gap-engineered transmon after a quasiparticle burst

Ionizing radiation impacts create bursts of quasiparticle density in superconducting qubits. These bursts temporarily degrade qubit coherence which can be detrimental for quantum error correction. Here, we experimentally resolve quasiparticle bursts in 3D gap-engineered transmon qubits by continuously monitoring qubit transitions. Gap engineering allows us to reduce the burst detection rate by a factor of five. This reduction falls four orders of magnitude short of that expected if the quasiparticles were to quickly thermalize to the cryostat temperature. We associate the limited effect of gap engineering with the slow thermalization of the phonons in our chips after the burst.

💡 Research Summary

This paper investigates how engineering a superconducting gap difference across the Josephson junction of a 3‑dimensional aluminum transmon can mitigate the deleterious effects of quasiparticle (QP) bursts caused by ionizing radiation. The authors fabricate three transmons with different aluminum film thickness pairs, thereby creating gap differences (δΔ) of roughly 0.5 GHz (small), 5 GHz (medium), and 10 GHz (large). By measuring parity‑switching rates (Γ₀ for the ground state and Γ₁ for the excited state) as a function of fridge temperature, they confirm that the QP tunneling rates follow the expected Arrhenius behavior, allowing extraction of both δΔ and the non‑equilibrium QP density x_neQP for each device.

To detect QP bursts, the qubit is repeatedly prepared in the excited state and read out every 5.7 µs. In the absence of bursts the number of relaxation events per 1 ms window follows a Poisson distribution determined by the steady‑state T₁ (~100 µs). Bursts manifest as a heavy, non‑Poissonian tail in the histogram of event counts. By applying a common threshold (eight standard deviations above the Poisson mean) the authors identify burst windows and compute burst detection rates: 0.57 min⁻¹ for the large‑δΔ device and 2.76 min⁻¹ for the small‑δΔ device. Thus, gap engineering reduces the burst rate by only a factor of five, far short of the ≈10⁴ suppression predicted if QPs instantly thermalized to the base temperature (25 mK).

The authors attribute this discrepancy to slow phonon thermalization. High‑energy particles generate a cascade of energetic phonons that heat the entire chip. During a burst the effective qubit temperature, inferred from the excess relaxation rate ΔΓ₁₀, rises from the background ~50 mK to ~90 mK and remains elevated for several milliseconds. This temperature increase provides QPs with enough energy to overcome the engineered gap, allowing them to tunnel despite the large δΔ. The burst duration (~5 ms) is essentially independent of δΔ, confirming that the limiting factor is the phonon bath, not the gap itself.

Consequently, while gap engineering is highly effective at suppressing the steady‑state “resident” QP tunneling (as shown by the exponential suppression of Γ₁ with δΔ), it offers limited protection against radiation‑induced bursts. The paper suggests that improving phonon evacuation—through better thermal anchoring, additional phonon‑absorbing layers, or substrate engineering—will be necessary to achieve the full potential of gap engineering for large‑scale quantum error correction. The work also introduces a robust, continuous‑monitoring protocol for real‑time burst detection, providing a valuable tool for future studies of QP and phonon dynamics in superconducting quantum processors.