Schoinas, N; Rath, Y; Norimoto, S; Xie, W; See, P; Griffiths, J P; Chen, C; Ritchie, D A; Kataoka, M; Rossi, A; Rungger, I (2024) Fast characterization of multiplexed single-electron pumps with machine learning. Applied Physics Letters, 125 (12). 124001

Preview

Text eid10100.pdf - Published Version Available under License Creative Commons Attribution. Download (3MB) | Preview

Abstract

We present an efficient machine learning based automated framework for the fast characterization of current quantization areas in single-electron pump devices. It uses an iterative active learning algorithm sparse measurement approach to acquire measurements in the parameter space that provide targeted information to locate and characterize the plateau where a single electron is transferred per cycle, including an extrapolation of the quantization error based on an exponential approximation of the approach to the plateau. We find a decrease by an order of magnitude in the time required to achieve scanning resolution high enough for us to be able to estimate the plateau flatness at a level of 1 μA/A by our exponential approximation method when compared to traditional parameter scans. We demonstrate the robustness of the framework by characterizing 28 individual devices arranged in a GaAs/AlGaAs multiplexer array, which we use to identify a subset of devices suitable for parallel operation at communal gate voltages. The method opens up the possibility to efficiently scale the characterization of such multiplexed devices to large number of pumps.

Item Type:	Article
Keywords:	Machine Learning Single-Electron Pumps Quantum Dots Experimental Automation
Subjects:	Quantum Phenomena > Nanophysics
Divisions:	Quantum Technologies
Identification number/DOI:	10.1063/5.0221387
Last Modified:	27 Nov 2024 11:23
URI:	https://eprintspublications.npl.co.uk/id/eprint/10100