Graduate Program Events |
Events on Friday, May 16th, 2025
- Noise characterization for Spin and Charge qubits
- Time: 10:00 am - 12:00 pm
- Place: 5280 CH
- Speaker: Deepak Mallubhotla, Physics PhD student
- Abstract: Quantum computing offers substantial advantages over classical computers for certain problems, but a fundamental challenge persists: qubits suffer from environmental noise causing decoherence. To achieve scales required for practical applications, error correction techniques must overcome these noise limitations. Studying various noise types is essential for developing effective mitigation strategies.
This dissertation first analyzes evanescent-wave Johnson noise (EWJN) near small metallic devices. Caused by thermal and quantum charge motion in conductors, this noise creates field fluctuations beyond the conductor's surface. Noise correlations B(x, t) B(x', t') are calculated when device size is smaller than material skin depth, yielding closed-form solutions via multipole expansion.
Next, EWJN is examined near BCS superconductors using a half-space geometry where superconductor dimensions exceed qubit distance. Superconductors generate less noise than normal conductors at temperatures well below critical temperature, for both magnetic and electric fields. A Hebel-Slichter peak with enhanced noise appears just below critical temperature, dependent on qubit orientation.
Finally, this dissertation discusses 1/f charge noise.
It has been hypothesized that this noise is caused by fluctuating two-level systems (TLSs). We show that measurements of noise power spectral density do not fully determine TLS parameters exactly, and present a Bayesian technique of assigning likelihoods to different parameter ranges instead. This allows for partial, statistical information to be extracted, giving predictions of TLS size and density. Two recent measurements are analyzed, giving predictions consistent with both each other and with measurements in the literature obtained through other techniques. - Host: Maxim Vavilov/Bob Joynt