Direct DC 10 V comparison between two programmable Josephson voltage standards made of niobium nitride (NbN)-based and niobium (Nb)-based Josephson junctions

Abstract

BIPM’s new transportable programmable Josephson voltage standard (PJVS) has been used for an on-site comparison at the National Metrology Institute of Japan (NMIJ) and the National Institute of Advanced Industrial Science and Technology (AIST) (NMIJ/AIST, hereafter called just NMIJ unless otherwise noted).This is the first time that an array of niobium-based Josephson junctions with amorphous niobium silicon NbxSi1−x barriers, developed by the National Institute of Standards and Technology (NIST), has been directly compared to an array of niobium nitride (NbN)-based junctions (developed by the NMIJ in collaboration with the Nanoelectronics Research Institute (NeRI), AIST).Nominally identical voltages produced by both systems agreed within 5 parts in 1012 (0.05 nV at 10 V) with a combined relative uncertainty of 7.9 × 10−11 (0.79 nV).The low side of the NMIJ apparatus is, by design, referred to the ground potential. An analysis of the systematic errors due to the leakage current to ground was conducted for this ground configuration. The influence of a multi-stage low-pass filter installed at the output measurement leads of the NMIJ primary standard was also investigated. The number of capacitances in parallel in the filter and their insulation resistance have a direct impact on the amplitude of the systematic voltage error introduced by the leakage current, even if the current does not necessarily return to ground. The filtering of the output of the PJVS voltage leads has the positive consequence of protecting the array from external sources of noise. Current noise, when coupled to the array, reduces the width or current range of the quantized voltage steps.The voltage error induced by the leakage current in the filter is an order of magnitude larger than the voltage error in the absence of all filtering, even though the current range of steps is significantly decreased without filtering.