Charge noise in quantum dot spin qubits

Short answer: Charge noise, meaning fluctuating electric fields in the semiconductor, limits many quantum dot spin qubits. A 2022 Nature Communications paper measured the charge-noise spectrum of a silicon-germanium singlet-triplet qubit over nearly 12 decades in frequency, a 2013 Nature Physics paper separated charge noise from spin noise using resonance fluorescence of a single quantum dot, and a 2018 Physical Review Applied paper compared a silicon dot qubit with a donor qubit. These are noise measurements, and they show where gate errors come from.

2022: silicon-germanium singlet-triplet qubit

The Nature Communications abstract says electron spins in silicon quantum dots are promising because of long coherence times, scalable fabrication and potential all-electrical control, but charge noise in the host semiconductor is a major obstacle to high-fidelity single- and two-qubit gates.

The authors measure the charge-noise spectrum of a Si/SiGe singlet-triplet qubit over nearly 12 decades in frequency, using several methods including dynamically decoupled exchange oscillations with up to 512 pi pulses. They find the noise is colored across the full range, although the spectral exponent changes with frequency. They also report that the spectrum inferred from conductance measurements of a nearby sensor dot agrees with the one inferred from the qubit's coherent oscillations. They suggest this means simple transport measurements can characterize charge noise over a wide frequency range in these devices.

2013: charge noise and spin noise in an optical dot

The Nature Physics abstract describes using resonance fluorescence from a single semiconductor quantum dot in an ultrapure device as a minimally invasive local probe. Charge noise is a fluctuating electric field and spin noise a fluctuating magnetic field, and both can dephase optical and spin states. The authors distinguish the two through a crucial difference in their optical signatures, and derive noise spectra for electric and magnetic fields from 0.1 Hz to 100 kHz. Charge noise dominates at low frequencies and spin noise at high frequencies. The noise falls rapidly with increasing frequency, which allowed them to show transform-limited optical linewidths when operating above 50 kHz.

2018: dot versus donor

The Physical Review Applied paper (Chan and coauthors) assesses a silicon quantum dot spin qubit through noise spectroscopy. The part of the abstract that was readable compares its sensitivity to electrical noise with an implanted silicon-donor qubit in the same environment and setup. The authors report that, as expected, the dot qubit is more sensitive to electrical noise because of a larger Stark shift, and that the noise-spectroscopy data show pronounced charge-noise contributions at intermediate frequencies of 2 to 20 kHz. The first part of the abstract was not captured, so this post does not describe it.

Limits

The three papers use different devices, an electrically controlled singlet-triplet qubit, an optical dot and a gated dot, so their noise levels should not be compared. None reports a qubit fidelity improvement from noise mitigation.

FAQ

What is charge noise?

Fluctuating electric fields from the host semiconductor, which affect gate operations, per the 2022 abstract.

Is the noise exactly 1/f?

The 2022 authors report colored noise with a spectral exponent that changes with frequency.

How can noise be measured simply?

The 2022 authors say sensor-dot conductance measurements agree with qubit-based measurements.

Are donor qubits less sensitive?

The 2018 authors report the dot qubit is more sensitive, due to a larger Stark shift.

Does this fix the noise?

No. These papers characterize it.