High sensitivity spin measurements in atomic precision devices

In order to achieve fault-tolerant quantum computation we must meet a very demanding set of requirements, including a very low error rate for qubit readout, much less than 1%. The error rate for ancilla qubit readout includes decoherence of data qubits that occurs during the readout time, which means that readout must be not just accurate, but also much faster than the qubit T2 time. I will review two major approaches we have taken to spin measurements in atomic precision electron spin qubit devices: capacitive charge detection with the single-electron transistor (SET), and single-electron susceptibility measurements (“dispersive” readout). Measuring the SET charge detector at radio frequencies we have recently demonstrated charge detection with a signal-to-noise ratio of 12.7 at 9.5 MHz bandwidth, which allows us to perform high-fidelity electron spin readout at the 10 microsecond timescale. I will discuss these results, how they compare with the theoretical sensitivity limit, and what we need to do to achieve that limit. Susceptibility readout enables us to make devices without a charge sensing structure in them, but it is a higher impedance measurement, which has so far limited its sensitivity. I will discuss our experimental achievements to date and the prospects for high sensitivity susceptibility readout in the future.