The presentation will be in English.
The push towards a scalable quantum computer is entering a crucial phase, with several different solid-state qubit designs demonstrated as strong candidates for the basis of a future quantum computer. One such candidate is silicon-based quantum dot spin qubits, a relatively recent entry into the field but with potential for long coherence times and high-fidelity measurement and manipulation. One of the major environmental interactions experienced by quantum dots is the characteristic 1/𝑓 charge noise present in all electronic devices. In this thesis, the charge noise experienced by quantum dots in CMOS silicon nanowire devices is investigated. Frequency-domain analysis is used to determine the variability in the charge noise experienced by a quantum dot due to its position within the nanowire and its interaction with different interfaces. A novel technique of measuring the charge noise experienced by a single electron is demonstrated, and extract the single-electron charge noise value of (130 ± 60) µeV2/Hz at a typical measurement temperature of 400 mK. Secondly, single-shot measurement of the spin of a single electron is enacted and the spin physics in a nanowire quantum dot characterized. The spin-lattice relaxation time 𝑇1 is measured with a fidelity greater than 70%. The spin-valley relaxation hotspot is detected via magnetic field relaxometry, finding a valley splitting energy of (297 ± 5) µeV. Finally, the magnetic field anisotropy of the spin-valley mixing is analyzed to assess the local disorder of the quantum dot, demonstrating suppression of the relaxation mechanism by an order of magnitude in a field oriented along the main symmetry axis of the device.
These experiments have developed our understanding of the noise environment and spin physics present in CMOS nanowire quantum dots, and form the foundations of in-depth characterization that can be applied at scale to direct future development of spin qubits in silicon.
Directeur de thèse : Tristan MEUNIER (CNRS, Institut Néel)
Co-directeur : Matias URDAMPILLETA (CNRS, Institut Néel)