As a part of its roadmap towards the redefinition of the kilogram, the Consultative Committee for Mass and Related Quantities (CCM) has carried out a pilot comparison of mass calibrations based on methods that will be used to realize the kilogram after its redefinition. The objectives were to assess the consistency of future realizations of the kilogram by different NMIs, to test continuity with the present kilogram based on the International Prototype of the Kilogram (IPK) and to gain experience for future key comparisons. The BIPM acted as the pilot laboratory. The LNE, NIST and NRC participated with their Kibble balances, the PTB and NMIJ used 28Si spheres from the Avogadro coordination project.
Two sets of travelling standards had been used by each participant in order to test both the realization experiments and the dissemination. The standards of the first set had to be calibrated as directly as possible using the realization method. Four participants carried out these calibrations under vacuum. The standards of the second set had to be calibrated in air, which required additional vacuum-to-air transfers and buoyancy corrections. The uncertainties ranged from 15 µg to 140 µg. All the traveling standards were sent to the BIPM where they were compared with each other and with BIPM working standards, which had been calibrated and were traceable to the IPK. The measurements for the Pilot Study were carried out from January to November 2016.
The results are in good agreement for both sets of standards. Four of the five participants agree within one standard deviation; the fifth agrees at the level of k=2. The uncertainty of the weighted mean is 10 µg for both sets. The weighted mean agrees with the IPK well within the uncertainty. The results mean that the conditions for the redefinition set by the CCM in 2013 have been fulfilled.
The results of each comparison constitute a snapshot taken at the time of the comparison. In early 2017 new results for the determination of the Planck constant were published, which have smaller uncertainties than previous determinations and which demonstrate an agreement that is not as good as that found in the Pilot Study. The CCM, at its meeting in May 2017, has formulated a recommendation (RECOMMENDATION G 1 (2017))on how to deal with this situation.