News from the BIPM laboratories: 2014

In December 2013 the Mass Department started a campaign of calibrations (Extraordinary Calibrations) against the International Prototype of the Kilogram (IPK) in anticipation of the planned redefinition of the kilogram. This campaign will meet one of the prerequisites for the redefinition of the kilogram requested by the Consultative Committee for Mass and Related Quantities (CCM) [1]. It will enable the mass standards used in the watt balance and x-ray crystal density (XRCD) experiments, which will contribute to fixing the numerical value of the Planck constant, as well as the Bureau International des Poids et Mesures (BIPM) reference and working standards, to be compared as directly as possible with the IPK. In the first phase of the campaign, the six official copies of the IPK and the BIPM working standards have been compared with the IPK for the first time since the 3rd Periodic Verification (3rd PV, 1988–1992). The results show that the differences in mass between the IPK and the official copies have changed on average by only 1 µg since the 3rd PV. These results do not confirm the trend observed during the 2nd and 3rd PVs by which the masses of the six official copies diverged from the mass of the IPK. All BIPM working standards have been calibrated with respect to the IPK and all were found to have lower masses than those established when they were calibrated during the 3rd PV. As a consequence, the BIPM's 'as-maintained' mass unit has been found to be offset by 35 µg with respect to the IPK.

To understand how this offset of 35 µg in 2014 has evolved over time, extensive mathematical modelling has been carried out. The results of all mass comparisons between the BIPM working standards carried out since the 3rd PV have been used as input data. Different deterministic models were tested against these data and the one that describes the data most satisfactorily has been identified. The phenomena included in the model are: the mass increase of a standard after cleaning and washing, and mass changes proportional to the number of weighings in the two comparators used most often during this time. As boundary conditions, the fit assumes that the mass of the IPK is exactly 1 kg after cleaning and washing, in 1992 and in 2014. This model allows retrospective calculation of revised mass values for previous calibrations. The uncertainty of the revised mass values has been estimated to be 3 µg. The National Metrology Institutes (NMIs) participating in accurate determinations of the Planck or the Avogadro constant have been informed of the revised mass values for previous mass calibrations. This will allow them to recalculate their results, if necessary. The second phase of the Extraordinary Calibrations, consisting of calibrations of transfer standards from these NMIs started in December 2014 and will be completed in January 2015.

Preparations for a new version of the BIPM watt balance are actively under way with the specific objective of reducing the type B uncertainty due to misalignment. The new apparatus will comprise several new and improved measurement facilities. Some of them have already been assembled and tested in situ, most importantly a dynamic alignment mechanism for the correction of the coil trajectory. Its performance has been successfully validated under real working conditions. Alignment techniques for the coil and the magnetic field continue to be refined. Development of a new interferometer has advanced well. It is based on space-separated heterodyning techniques to allow for small non-linear error and high phase resolution. In order to facilitate the alignment of the interferometer, a new open support structure allowing easy access to the watt balance apparatus has been designed. In addition to being rigid and stable, the design has been refined using finite element analysis to avoid low frequency resonance peaks, especially those present in the environment. The Josephson voltage standard for current measurement has been successfully tested in the watt balance.

In 2011, the CGPM encouraged the BIPM to develop 'a pool of reference standards to facilitate the dissemination of the unit of mass when redefined' (Resolution 1 (2011)). Since then, the Mass Department has been assembling a new Ensemble of 12 Reference Mass Standards and four stacks of disks. The storage network for the Ensemble of Reference Mass Standards (ERMS) has undergone several improvements in 2014. The gas flow has been reduced and is now uniform within the different gas lines. New mass holders for the standards stored in gas allow easier manipulation in the glove box and are safer for the standards. New holders have been designed for the stacks, which allow the transfer to the mass comparator while keeping them in their storage medium. The first standards will be placed in their containers in early 2015.

Four new mass prototypes and one stack of eight discs have been fabricated during 2014. The four mass prototypes have been calibrated and, together with an already existing stack, have been characterized for their vacuum-to-air transfer behaviour. Three more prototypes are under fabrication. The department has determined the masses of the two repolished ²⁸Si spheres AVO-S5 and AVO-S8, in air and in vacuum. A study was also undertaken to determine the mass of the chemisorbed water layer present on the surface of a natural silicon sphere. The mean chemical adsorption coefficient obtained by two different methods was 0.026 µg cm⁻² with a standard uncertainty of 0.012 µg cm⁻². The BIPM results confirm those obtained by the National Metrology Institute of Japan/National Institute of Advanced Industrial Science and Technology (NMIJ/AIST), Japan, which had measured the adsorption isotherms on SiO₂/Si(1 0 0) plane surfaces. The mass results as well as the chemical adsorption coefficient will contribute to a new determination of the Avogadro constant. The Mass Department has published the final report on the CCM comparison, CCM.M-K4, of stainless steel masses.

A major achievement in the Time Department has been the implementation of a new algorithm for the regular calculation of Coordinated Universal Time (UTC). This algorithm is based on a parabolic model for the prediction of the participating clocks' frequency, and a weighting procedure which uses the clock predictability as the criteria. The time scale computed with the new algorithm has an improved stability of about 20% in both the short and long term.

Time transfer methods used for clock comparison are the major component in the uncertainty of [UTC − UTC(k)] in the key comparison CCTF-K001.UTC. Improving time transfer with the study and implementation of new methods has been a substantial part of the work programme of the Department, with the target of reducing the statistical uncertainty of time transfer to a few hundred ps from the present best value of 0.3 ns for 46% of time links, and also to enhance the accuracy of time transfer with more refined and frequent calibrations.

Coordination with the Regional Metrology Organizations (RMOs) for the calibration of time transfer links and equipment in UTC contributing laboratories was a major theme for the department in 2014. The BIPM revised the guidelines on the procedures to be followed in the regional calibrations; a scheme fixes the responsibilities of the BIPM for the calibration of a set of selected laboratories and the RMOs for the calibration of the others. The measurements that are the responsibility of the BIPM are close to being concluded. At the end of the first round of the process, all time links in UTC will be calibrated, and the Type B uncertainty (uB) value characterizing the links and impacting on the uncertainty of [UTC − UTC(k)] is expected to decrease by a factor of 2.

Four new caesium fountains have been incorporated in 2014 to improve the frequency accuracy of International Atomic Time (TAI) following approval from the relevant Consultative Committee for Time and Frequency (CCTF) Working Group. Eleven primary frequency standards, including nine caesium fountains, and one secondary representation of the second contributed to TAI in 2014, with an average of three to four caesium fountains reporting measurements each month. The increasing number of measurements reported for these standards reinforces the accuracy estimation of TAI, which is in the low 10⁻¹⁶.

BIPM Circular T continues to be published monthly, giving traceability to the SI second via UTC to its local realizations in national laboratories. It is the most frequent key comparison, with one evaluation of the key comparison reference value UTC and the degrees of equivalence [UTC − UTC(k)] every five days for the 73 participants that together contributed data from more than 450 atomic clocks at the end of 2014 compared to 400 in 2013.

Regular publication of rapid UTC (UTCr) has continued in 2014, providing a weekly solution that is consistently better than ±2 ns peak to peak with the final UTC published in Circular T. This rapid solution supports the quality of the representations of UTC in national laboratories and the steering of the Global Navigation Satellite Systems' times to local representations of UTC.

A first uncertainty budget has been established for the calculable capacitor, which includes the results of a theoretical model of the frequency dependence. The present alignment accuracy of the capacitor's electrode bars limits the relative uncertainty of determinations of the von Klitzing constant to about 1 part in 10⁷. A number of measurements involving the calculable capacitor and the dedicated coaxial ac-bridge have been carried out to investigate possible systematic errors and noise sources. These investigations allowed the elimination of several noise sources from the measurement chain which has improved the measurement repeatability to 1 part in 10⁸ or better. Two causes of systematic errors have been detected in the calculable capacitor. One of them is related to a mechanical instability of the electrode bars and the second to an instability of the interferometer laser, which has recently been solved. These difficulties have delayed the planned determination of the von Klitzing constant at the 10⁻⁷ level, which is now scheduled for early 2015. This will be followed by a careful re-alignment of the electrode bars with a new alignment probe which is at present being fabricated in the BIPM workshop. The ultimate goal remains a determination of the von Klitzing constant at the 10⁻⁸ level.

A Josephson Voltage Standard (JVS) dedicated to the measurement of the current sent to the coil of the BIPM watt balance experiment has been assembled, tested and delivered. It was demonstrated that the array voltage remained quantized and that the performance of the array did not suffer from the noise level of the watt balance environment. It has been successfully implemented on the watt balance and demonstrated for the first time that the stability of the current source needed to be improved as this was the limiting factor.

The BIPM series of onsite quantum Hall comparisons (BIPM.EM-K12) restarted with a successful visit to the Physikalisch-Technische Bundesanstalt (PTB) in November 2013 [2]. This comparison highlighted a limiting factor for 1 Ω measurements, which has been the subject of further study in 2014 in collaboration with the PTB. The effect of Peltier heating in a range of different 1 Ω standards has now been comprehensively characterized, with the results submitted for publication. With this improved knowledge it should be possible to minimize these effects in future comparisons, and ensure comparability of 1 Ω measurements to the level of around 10 nΩ/Ω. (The effect is less significant for higher value resistors, and comparisons of 100 and 10 kΩ standards against the quantum Hall reference are possible at uncertainties approaching 1 nΩ/Ω, as seen in the PTB comparison). The comparison series will continue with two onsite visits planned for 2015.

The very low frequency (1 Hz or below) room temperature current comparator bridge technology that the BIPM uses for onsite resistance comparisons forms the basis of a EURAMET research project on next generation quantum Hall reference systems based on graphene. The BIPM has been collaborating with the PTB and the MIKES in tests of new resistance bridges that can give primary standards level accuracy without the need for liquid helium. Initial tests have shown successful ratio measurements using MIKES-built current comparators and the existing BIPM electronics. This technology, combined with the graphene devices is now in development and promises simpler, more robust primary standards for resistance.

The project to develop an absorbed-dose standard for medium-energy x-rays has made significant progress in 2014. The transfer standards constructed at the BIPM have been measured in air and in the new water-phantom arrangement over the range of x-ray energies. Corresponding Monte Carlo calculations of the relative response of the chambers under these conditions indicate that the absorbed dose is likely to be determined to well within the target uncertainty of 1%.

The Ionizing Radiation Department carried out the seventh and eighth comparisons in the series BIPM.RI(I)-K6 for absorbed dose to water in high-energy photon beams, with the NPL and the VSL, in the 6, 10 and 25 MV beams of the NPL. The BIPM used a remote-controlled motorized monitoring and shutter system for the first time, to reliably track and correct for intrinsic intensity variations in the beams. The new design is compact and showed a high reproducibility in the positioning of the shutter. Being controlled remotely, the system avoids the need for staff to enter the radiation area between irradiations, which saves time and improves radiation protection for the operator.

The accumulated data from the BIPM.RI(I)-K6 comparison series were used for a re-determination of the value for W_air, the mean energy required to create an ion pair in air, with a world-wide impact on primary ionometric dosimetry measurements [3].

The comparison series BIPM.RI(I)-K8 for reference air kerma rate for HDR ¹⁹²Ir brachytherapy sources was re-launched: a new protocol was adopted, the results of the three previous comparisons were re-analysed and the corresponding comparison reports were produced and published. Two new comparisons with the NRC and the LNE-LNHB were carried out.

In total, sixteen dosimetry comparisons and thirteen calibrations of national secondary standards were made, which are underpinned by a significant effort in equipment calibration and maintenance. In addition, eleven comparison reports were submitted for publication.

Within the radionuclides measurements programme, the Système International de Référence (SIR) received 16 ampoules of 11 different radionuclides from seven laboratories, all oriented to generate equivalence values in the ongoing BIPM key comparison BIPM.RI(II)-K1. Measurements of ²²²Rn, a radioactive gas with a short half-life (T_1/2 = 3.8235 d, u = 0.0003 d), contributed to a co-authored article on geometry and type of containers. Monitoring of ²²²Rn has been implemented to study possible correlations of the background on SIR measurements.

This year marked a significant milestone with the effective use of the SIRTI to extend the SIR to ¹⁸F (T_1/2 = 1.8 h), one of the most frequently used radionuclides in positron emission tomography (PET), through a new BIPM.RI(II)-K4.F-18 ongoing comparison, onsite at the NMIs' premises. The new specific protocol was established and the link SIRTI-SIR was measured for ¹⁸F using both a commercial solution and a solution from the LNE-LNHB. The validation was made at the NPL by comparing the SIRTI result and the NPL SIR result from 2003. This new ongoing comparison series started with comparisons at the VNIIM, the NPL and the ENEA-INMRI.

In parallel, the BIPM.RI(II)-K4.Tc-99 m (T_1/2 = 6.0 h) comparison using the SIRTI continued with the VNIIM and the ENEA-INMRI participating in 2014. It is remarkable that, in spite of being transported around the world, the SIRTI has shown a very high reproducibility since 2007, with a relative standard deviation of 2  ×  10⁻⁴ for the count rate of the ⁹⁴Nb reference source measured world-wide.

Another significant step concerns the effective start in November 2014 of the trial exercise for the Extension of the SIR to beta emitters, with liquid scintillation measurements of ³H, ¹⁴C, ⁵⁵Fe and ⁶³Ni in different scintillators, volumes and counters and the participation of 14 NMI/DIs. The trial exercise will last 5 to 7 months and will serve to establish the procedures for a new ongoing, 'on demand' key comparison that will allow NMIs to obtain equivalence rapidly for beta emitters without the need for large-scale time-consuming comparisons.

In total, eighteen radionuclide activity comparisons were undertaken, and update reports of six BIPM.RI(II)-K1 comparisons and one BIPM.RI(II)-K4 comparison were submitted for publication in the Metrologia Technical Supplement covering ¹³¹I, ¹³³Ba, ¹⁵²Eu, ¹⁷⁷Lu, ^166mHo and ^99mTc, and including the links from APMP.RI(II)-K2.I-131, COOMET.RI(II)-K2.Eu-152, CCRI(II)-K2.Lu-177 and EURAMET.RI(II)-K2.Ho-166 m comparisons.

In summary, thirty-four comparisons, thirteen calibrations and seventeen comparison reports were generated by the BIPM in 2014. Additionally, four CCRI and five RMO comparison reports were reviewed and published.

In the field of gas metrology, the BIPM continued to coordinate comparisons of standards for greenhouse gases and air quality monitoring in 2014. In the area of greenhouse gases and their precursors, the final report of CCQM-K82 on methane in air at ambient levels was published and the results were presented at the European Geophysical Union General Assembly. A paper demonstrating equivalence between standards made in whole and synthetic air and measured by cavity ring-down spectroscopy (CRDS) and gas chromatography-flame ionization detector (GC-FID) for atmospheric monitoring applications was submitted to Analytical Chemistry in November 2014. A new vacuum Fourier transform infrared spectroscope (FTIR) was acquired and validated in preparation of the comparison CCQM-K120 on carbon dioxide in air standards. Standards chosen to span both the target amount fraction and isotope ratio ranges have been obtained through collaborative work with partner NMIs (NIST and NPL). A new methodology to deduce the isotopic shift δ¹³C in the CO₂/air mixtures by FTIR has been developed during the secondment of Dr Marta Doval Minarro from the NPL. A new system to measure carbon dioxide amount fractions with traceability to the BIPM pressure standards has also been installed in a new laboratory. The system makes use of cryogenic separation of carbon dioxide from its air matrix and accurate measurements of pressure, temperature and volume. Corrections for the presence of N₂O will be made based on gas chromatography-electron capture detector (GC-ECD) measurements, with the funds for the purchase of a system donated from the KRISS. The system will also underpin future key comparisons on nitrous oxide standards.

In the area of air quality gas standard comparisons, the ongoing surface ozone reference standard comparison (BIPM.QM-K1) has continued with four laboratories participating in 2014, bringing the total number of participants to 21. Work on new ozone absorption cross-section measurements was completed, leading to the proposal of a new value of 11.27  ×  10⁻¹⁸ cm² molecule⁻¹ with an expanded relative uncertainty of 0.92%. This is lower than the conventional value currently in use and measured by Hearn in 1961 with a relative difference of 1.8%, with the consequence that historically reported ozone concentrations should be increased by 1.8%. A paper describing the measurements was published online in Atmospheric Measurement Techniques Discussions [4]. The comparison CCQM-K90 on formaldehyde (HCHO) in nitrogen standards has started after the delivery of the transfer standards mixtures selected by BIPM for their stability and the final validation of two different systems to dynamically generate HCHO, one by permeation from paraformaldehyde, the other by diffusion of trioxane. The development of a dynamic generation system for nitric acid in nitrogen standards was carried out during the secondment of Mrs Céline Pascal from METAS. It was used to calibrate the FTIR spectrometer, reducing the relative uncertainty of measurements of nitric acid by a factor of three. This will reduce the uncertainty in the reference value for nitrogen dioxide standards in future key comparisons (CCQM-K74.2018).

The BIPM's organic programme forms an essential part of the CCQM Working Group on Organic Analysis' (OAWG) strategy for Core Competency comparisons. A white paper outlining an approach for the linkage of organic purity key comparison results to calibration and measurement capability (CMC) claims for organic analysis, which included an extensive compilation of the results by all NMIs that have participated in BIPM coordinated purity comparisons, was prepared. It has been accepted for implementation by the OAWG and the CCQM Working Group on Key Comparisons and CMC Quality (KCWG).

The BIPM's laboratory facilities for supporting purity key comparisons were augmented through the donation of a high field nuclear magnetic resonance (NMR) spectrometer from an instrument manufacturer. Obtaining the donation was the first outcome of the BIPM-NMIJ collaboration on the development of quantitative NMR (qNMR) for organic purity assignment which started in 2014. This was followed by training of BIPM staff at both the NMIJ and the NRC, and the installation and commissioning of the NMR instrument at the BIPM. In a lead up to this collaboration the BIPM participated successfully in the CCQM-P150 pilot study, coordinated by the NMIJ, on use of qNMR for a purity assignment of dimethyl sulfone with bis-3,5-triflurormethylbenzoic acid as the internal standard.

The final report of the CCQM-K55.c [(L) Valine purity] key comparison was published in the key comparison database (KCDB) and the preparative work on samples for CCQM-K55.d [Folic acid purity] is continuing, with homogeneity and stability measurements under way. The normal phase-liquid chromatography-tandem mass spectrometry method with atmospheric pressure photoionization developed for the purity assessment of 17β-estradiol by the BIPM as part of CCQM-K55.a was published in Analytical and Bioanalytical Chemistry [5]. The BIPM has led an International Union of Pure and Applied Chemistry (IUPAC) working group, with members from 12 NMIs and two international organizations, with two meetings held in 2014 to draft technical guidelines on 'Methods for the SI Value Assignment of the Purity of Organic Compounds for use as Primary Reference Materials and Calibrators'. A draft report from this working group is expected in 2015, which will describe validated methods and achievable measurements uncertainties for organic primary calibrator material characterization. The implementation of these methods is expected to improve the comparability of analytical chemical measurements world-wide.

The BIPM's organic programme on large organic molecule purity has continued with the start of the first CCQM key comparison (CCQM-K115/P55.2) on peptide purity (C-peptide (hCP)) coordinated by the BIPM in collaboration with the NIM, China, supported by the secondments of Dr Ming Li and Dr Dewei Song. Preparatory work for the comparison included the successful completion of the cross-validation of different approaches for the purity mass fraction value assignment of a model decapeptide (Angiotensin I in collaboration with NIST) as well as insulin and three external publications, currently in preparation describe this work.

The 25th meeting of the General Conference on Weights and Measures (CGPM) was held in Versailles from 18 to 20 November 2014. Five resolutions were adopted by the CGPM during the meeting.

On the future revision of the International System of Units, the SI

On the election of the International Committee for Weights and Measures

On the Pension and Provident Fund of the BIPM

Dotation of the BIPM for the years 2016–2019

On the importance of the CIPM Mutual Recognition Arrangement