12
$\begingroup$

I have a superconducting system with tens of qubits, each of which can be tuned using DC flux.

One of the main tasks for coherent manipulation of the qubits is to find good idling frequencies and operating points for entangling gates. This effort is confounded by two-level systems (TLS), which cause rapid energy relaxation, and wreak general havok on coherent manipulation.

I spent a long time finding a good set of idling frequencies and operating points, all the while considering the locations of the TLS's, and then one day I came in the the lab and they had moved around! I had to start all over again.

I want to learn more about how and why TLS's move, and whether it's possible to maybe control the movement. As part of my research, I want to poll the community and see what other people's experience with this problem is like.

Improve this question
$\endgroup$
1
  • 4
    $\begingroup$ See this blog post from the Google AI research website (and the associated journal paper ) on exactly this issue. I'm going to email my group with a link to this Stack Exchange post and ask Dr. Klimov to respond. $\endgroup$
    – DanielSank
    Sep 8, 2018 at 5:43

1 Answer 1

Reset to default
5
$\begingroup$

The resonance frequencies of TLS fluctuate due to their interaction with neighboring TLS, which occurs through electric dipole interaction or the local mechanical strain in the material. If a TLS at low energy (below kB*T) is involved, this one may change its state randomly due to thermal activation. The resulting change in local electric field or strain can detune also TLS at higher energy which are within the qubit tuning range. Here's a theory paper on that: https://arxiv.org/abs/1503.01637

This process is called 'spectral diffusion' and can probably only be avoided by improving circuit materials to reduce the TLS density (and thus interactions). But even without thermal processes, a TLS may be trapped in a long-living metastable potential well from which escape by tunneling could take hours, days, or even years.

TLS can be detuned in frequency by controlling the mechanical strain in the sample, see this paper which also discusses TLS interactions: https://www.nature.com/articles/ncomms7182

Here's a review article which summarizes decoherence effects from TLS on superconducting qubits and resonators: https://arxiv.org/abs/1705.01108

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.