Loading a quantum-dot based \u201cQubyte\u201d register

C Volk,P T Eendebak,C J Van Diepen,W Wegscheider,U Mukhopadhyay,L M K Vandersypen,A M J Zwerver,T Hensgens,C Reichl,J P Dehollain,T Fujita

doi:10.1038/s41534-019-0146-y

Abstract

Electrostatically defined quantum dot arrays offer a compelling platform for quantum computation and simulation. However, tuning up such arrays with existing techniques becomes impractical when going beyond a handful of quantum dots. Here, we present a method for systematically adding quantum dots to an array one dot at a time, in such a way that the number of electrons on previously formed dots is unaffected. The method allows individual control of the number of electrons on each of the dots, as well as of the interdot tunnel rates. We use this technique to tune up a linear array of eight GaAs quantum dots such that they are occupied by one electron each. This new method overcomes a critical bottleneck in scaling up quantum-dot based qubit registers.

Highlights

Quantum-dot based electron spin qubit systems[1,2,3] have made significant steps towards becoming a scalable platform for quantum computation
26) We show that we can locally control the number of electrons on each dot down to the last electron, and that we can set all interdot tunnel couplings to typical values used in spin qubit experiments
We probe the linear quantum dot (QD) array via the two sensing dots, which are sensitive to the number of electrons in the array, as well as to their position in the array

Summary

INTRODUCTION

Quantum-dot based electron spin qubit systems[1,2,3] have made significant steps towards becoming a scalable platform for quantum computation. For more than three or four dots, the co-tunnel rates become impractically low These challenges present themselves when measuring the charge occupation in quantum dot arrays through conventional charge stability diagrams. We explored several approaches to form long linear arrays in a controlled way, such as forming individual single dots first and stitching them together, stitching together double dots, or starting with a large QD and splitting it up into an array of separate dots We found it difficult to make these approaches work well. We discuss the limitations and potential pitfalls of the n + 1 method

RESULTS

P2 B2 P3 B3 P4 B4

Measure the cross-capacitance between all gates and the

METHODS