Multi-qubit and multi-dot reflectometry measurements in arrays of semiconductor quantum dots

My talk on high-frequency reflectometry measurements has two parts, explaining its use to tune gate voltages into desired charge configurations of quantum-dot arrays, and to perform simultaneous single-shot readout of multiple qubits.

For gate-voltage tuning, the traditional dense raster scans of charge stability diagrams cannot be scaled to larger arrays with many gate electrodes, due to the prohibitively large number of pixels contained in naive high-dimensional gate-voltage cubes. To mitigate this challenge, we developed sparse acquisition methods in which the boundaries of Coulomb-blockade polytopes are reconstructed from a discrete set of time stamps, each triggered by a simple thresholding circuit that detects abrupt changes in the reflectometry signal without ever digitizing it. We combine this method with an active-learning algorithm and a large-margin classifier to adaptively propose and execute new line searches in gate-voltage space. We implement this approach on a 2x2 array of quantum dots implemented in fully-depleted silicon-on-insulator devices.

For the simultaneous single-shot readout of four singlet-triplet qubits, implemented in gate-controlled gallium-arsenide heterostructures, we utilize frequency multiplexing and four radiofrequency resonators with distinct resonance frequencies, each monitoring via time-resolved reflectometry the state of a sensor dot associated with a qubit. I will explain the simultaneous and interlaced detection of exchange-rotation and Overhauser-dephasing operations, as well as various coherent spin-exchange processes within the quantum dot array that reveal mechanisms related superexchange, electron-electron correlations, and negative (multi-electron) exchange coupling.