Bureau International des Poids et Mesures (BIPM), Sèvres. National Institute of Standards and Technology (NIST), USA. Becquerel (LNE-LNHB), France

Transcription

1 BIPM comparison BIPM.RI(II)-K1.I-131 of activity measurements of the radionuclide 131 I for the NMIJ (Japan), the NIST (USA) and the LNE-LNHB (France), with linked results for the APMP.RI(II)-K2.I-131 comparison C. Michotte, G. Ratel, S. Courte, A. Yunoki 1, Y. Unno 1, R. Fitzgerald 2, L. Pibida 2, C. Fréchou 3, C. Bobin 3, C. Thiam 3 Bureau International des Poids et Mesures (BIPM), Sèvres 1 National Metrology Institute of Japan (NMIJ), Japan 2 National Institute of Standards and Technology (NIST), USA 3 Laboratoire National de Métrologie et d Essais Laboratoire National Henri Becquerel (LNE-LNHB), France Abstract Three new participations in the BIPM.RI(II)-K1.I-131 comparison have been added to the previous results and this has produced a revised value for the key comparison reference value (KCRV), calculated using the power-moderated weighted mean. A link has been made to the APMP.RI(II)-K2.I-131 comparison held in 2009 through the NMIJ who participated in both comparisons. Three national metrology institutes (NMIs) used the Asia Pacific Metrology Programme (APMP) comparison to update their degree of equivalence. The degrees of equivalence between each equivalent activity measured in the International Reference System (SIR) and the KCRV have been calculated and the results are given in the form of a table for twelve NMIs in the BIPM.RI(II)-K1.I-131 comparison and five other participants in the APMP.RI(II)-K2.I- 131 comparison. A graphical presentation is also given. 1. Introduction The SIR for activity measurements of -ray-emitting radionuclides was established in Each national metrology institute (NMI) may request a standard ampoule from the BIPM that is then filled with 3.6 g of the radioactive solution, or a different standard ampoule for radioactive gases. Each NMI completes a submission form that details the standardization method used to determine the absolute activity of the radionuclide and the full uncertainty budget for the evaluation. The ampoules are sent to the BIPM where they are compared with standard sources of 226 Ra using pressurized ionization chambers. Details of the SIR method, experimental set-up and the determination of the equivalent activity, A e, are all given in [1]. 1/25

2 From its inception until 31 December 2012, the SIR has measured 966 ampoules to give 721 independent results for 67 different radionuclides. The SIR makes it possible for national laboratories to check the reliability of their activity measurements at any time. This is achieved by the determination of the equivalent activity of the radionuclide and by comparison of the result with the key comparison reference value determined from the results of primary standardizations. These comparisons are described as BIPM ongoing comparisons and the results form the basis of the BIPM key comparison database (KCDB) of the Comité International des Poids et Mesures Mutual Recognition Arrangement (CIPM MRA) [2]. The comparison described in this report is known as the BIPM.RI(II)-K1.I-131 key comparison and includes results published previously [3-5]. In addition, an international comparison was held in 2009 for this radionuclide, APMP.RI(II)-K2.I-131 [14]. Seven laboratories took part in this comparison including the NMIJ who participated in the SIR at the same time, enabling to link the APMP.RI(II)-K2 comparison to the BIPM.RI(II)-K1 comparison. Three NMIs had previously submitted ampoules to the SIR and have updated their results through this APMP comparison. 2. Participants Eighteen NMIs and one other laboratory have submitted forty-five ampoules for the comparison of 131 I activity measurements since As the key comparison reference value has been re-evaluated for this comparison all the participants details are given in Table 1a. In cases where the laboratory has changed its name since the original submission, both the earlier and the current acronyms are given, as it is the latter that are used in the KCDB. Table 1a. Details of the participants in the BIPM.RI(II)-K1.I-131 Original acronym NMI Full name Country Regional metrology organization Date of measurement at the BIPM NBS NIST National Institute of Standards and Technology ASMW * PTB continued overleaf Physikalisch- Technische Bundesanstalt United States SIM Germany EUROMET /25

3 Table 1a continued. Details of the participants in the BIPM.RI(II)-K1.I-131 Original acronym NMI Full name Country Regional metrology organization Date of measurement at the BIPM BARC Bhabha Atomic Research Centre India APMP NAC * NMISA National Metrology Institute, South Africa NPL National Physical Laboratory South Africa United Kingdom SADCMET EUROMET OMH MKEH Magyar Kereskedelmi Engedéyezési Hivatal IER IRA Institut de Radiophysique Appliquée PDS PTKMR Pusat Teknologi Keselamatan dan Metrologi Radiasi ETL NMIJ National Metrology Institute of Japan Hungary EUROMET Switzerland EUROMET Indonesia APMP Japan APMP UVVVR CMI-IIR Český Metrologický Institut -Inspectorate for Ionizing Radiation Czech Republic EUROMET LMRI LNE- LNHB Bureau national de métrologie- Laboratoire national Henri Becquerel France EUROMET NIRH National Institute of Radiation Hygiene ANSTO Australian Nuclear Science and Technology Organisation continued overleaf Denmark EUROMET Australia APMP /25

4 Table 1a continued. Details of the participants in the BIPM.RI(II)-K1.I-131 Original acronym NMI Full name Country Regional metrology organization Date of measurement at the BIPM LNMRI /IRD Laboratorio Nacional de Metrologia das Radiaçoes Ionizantes/Instituto de Radioproteção e Dosimetria CIEMAT Centro de Investigaciones Energéticas, Medioambientales y Tecnològicas IFIN-HH Institutul de Fizica si Inginerie Nucleara- Horia Hulubei KRISS Korea Research Institute of Standards and Science BEV Bundesamt für Eichund Vermessungswesen * a previous standards laboratory in the country Brazil SIM Spain EUROMET Romania EUROMET Republic of Korea APMP Austria EUROMET The INER and OAP that took part in the APMP international comparison, APMP.RI(II)-K2.I-131 in 2009 [14] and are also eligible for the KCDB are shown in Table 1b together with the ANSTO, BARC and KRISS that used this comparison to update their results. 3. NMI standardization methods Each NMI that submits ampoules to the SIR has measured the activity either by a primary standardization method or by using a secondary method, for example a calibrated ionization chamber. In the latter case, the traceability of the calibration needs to be clearly identified to ensure that any correlations are taken into account. A brief description of the standardization methods used by the laboratories, the activities submitted, the relative standard uncertainties (k = 1) and the half-life used by the participants are given in Table 2. The uncertainty budgets for the three new submissions are given in Appendix 1, previous uncertainty budgets are given in the earlier K1 report [3-5]. The uncertainty budgets for all the participants in the 4/25

5 APMP.RI(II)-K2.I-131 comparison were published in the final report [14]. The acronyms used for the measurement methods are given in Appendix 2. Table 1b. Details of the participants in the 2009 APMP.RI(II)-K2.I-131 to be linked to BIPM.RI(II)-K1.I-131 NMI Full name Country Regional metrology organization ANSTO BARC INER KRISS Australian Nuclear Science and Technology Organisation Bhabha Atomic Research Centre Institute of Nuclear Energy Research Korea Research Institute of Standards and Science Australia India Chinese Taipei Republic of Korea APMP APMP APMP APMP OAP Office of Atoms for Peace Thailand APMP The half-life used by the BIPM and in the APMP.RI(II)-K2 comparison is (19) d from the BIPM Monographie 5 [6]. Details regarding the solutions submitted are shown in Table 3, including any impurities, when present, as identified by the laboratories. When given, the standard uncertainties on the evaluations are shown. The BIPM has developed a standard method for evaluating the activity of impurities using a calibrated Ge(Li) spectrometer [7]. The CCRI(II) agreed in 1999 [8] that this method should be followed according to the protocol described in [9] when an NMI makes such a request or when there appear to be discrepancies. No impurities were identified in the latest NMIJ, NIST and LNE- LNHB submissions. 4. Results All the submissions to the SIR since its inception in 1976 are maintained in a database known as the "master-file". The SIR equivalent activity, A ei, for each ampoule for the previous and new results is given in Table 4a. The date of measurement in the SIR is also given and is used in the KCDB and all references in this report. The relative standard uncertainty arising from the measurements in the SIR is also shown. This uncertainty is additional to that declared by the NMI for the activity measurement shown in Table 2. Although submitted activities are compared with a given source of 226 Ra, all the SIR results are normalized to the radium source number 5 [1]. 5/25

6 Table 2. Standardization methods of the participants for 131 I Metrologia 2014, 51, Tech. Suppl., NMI Method used and acronym (see Appendix 2) Half-life / d Activity A i / kbq Reference date Relative standard uncertainty 100 by method of evaluation YY-MM-DD A B NIST PTB BARC continued overleaf Pressurized IC calibrated by 4 (PC) 4P-PC-BP Pressurized IC calibrated in 1979 by 4 - coincidence a 17 h UT 8.02 (1) h UT 8.02 (1) h UT h UT 4P-PP-BP-NA-GR-CO b h UT h UT 4 - anti-coinc. 4P-LS-BP-NA-GR-AC Pressurized IC d 4 - coincidence Pressurized IC d Pressurized IC calibrated in 2004 by 4 - coinc. and CIEMAT/NIST 4P-LS-MX CN 4 - coincidence (19) [6] h UT h UT h UT a 0 h UT (9) a 0 h UTC h 30 UT h 30 UT /25

7 Table 2 continued. Standardization methods of the participants for 131 I NMI Method used and acronym (see Appendix 2) Half-life / d Metrologia 2014, 51, Tech. Suppl., Activity A i / kbq Reference date Relative standard uncertainty 100 by method of evaluation YY-MM-DD A B NMISA NPL MKEH IRA PTKMR NMIJ CMI-IIR LNE- LNHB 4 (LS) - coinc. 4P-LS-BP-NA-GR-CO Pressurized IC e 4 (PPC)- coinc. 4P-PP-BP-NA-GR-CO 4 - coincidence 4 (PC)- coinc. 4 - coincidence Pressurized IC d Pressurized IC 4 (PC) - coinc. 4 - coincidence Pressurized IC d 4 (PC) - coinc. 4 - coincidence 4 - anticoincidence 4P-PC-BP-NA-GR-AC 4 counting 4P-NA-GR HE continued overleaf a 10 h UT h UT a 12 h UT (2) [10] (1) [11] (1) [11] h h UT h UT h UT h UT a 2 h UT a h UT h UT h UT (1) h UT (19) [6] h UT (1) a 12 h UT (19) [6] h UT /25

8 Table 2 continued. Standardization methods of the participants for 131 I NMI NIRH ANSTO LNMRI /IRD CIEMAT IFIN-HH KRISS BEV Method used and acronym (see Appendix 2) Pressurized IC Pressurized IC c Pressurized IC f 4 (PC) - (NaI) coincidence 4 - coincidence 4 - coincidence Pressurized IC g Half-life / d Metrologia 2014, 51, Tech. Suppl., Activity A i / kbq h UT h UT (1) [11] Reference date Relative standard uncertainty 100 by method of evaluation YY-MM-DD A B h UT h UT h UT (19) [6] (19) [6] h UT h UT h UT a two ampoules submitted b corrected for the growth of 131m Xe which was assumed to be zero at the time of sealing the ampoule c calibrated by 4 - coincidence measurements of 131 I in1978 d traceable to 4 - coincidence measurements of 131 I (in 2001 for the CMI-IIR) e calibrated by primary measurements of 131 I f calibrated by 4 - coincidence measurements of 131 I in1993 g traceable to NPL 8/25

9 Table 3. Details of the solution of 131 I submitted NMI Chemical composition Solubilization medium [12] conc. Carrier conc. /( g g 1 ) Density /(g cm 3 ) Relative activity of impurity NIST 1977 KI and Na 2 SO 3 in diluted LiOH and NaOH LiOH: mol dm 3 NaOH: mol dm 3 Na 2 SO 3 : 30 g g LiOH: mol dm 3 NaOH: mol dm 3 KI: (2) 125 I : 0.3 % * KI: (2) Na 2 SO 3 : 3 g g (2) 1980 KI and Na 2 SO 3 in diluted LiOH.H 2 O and NaOH LiOH.H 2 O: mol dm 3 NaOH: mol dm 3 Na 2 SO 3 : 2 g g LiOH.H 2 O: mol dm 3 NaOH: mol dm 3 KI: (2) KI: (2) Na 2 SO 3 : 3 g g KI in LiOH LiOH : 0.01 mol dm 3 KI: KI, Na 2 SO 3 in diluted LiOH Na 2 SO 3 : 70 g g 1 Continued overleaf LiOH: 234 g g 1 KI: m Xe: 0.36 (15) % * 9/25

10 Table 3 continued. Details of the solution of 131 I submitted NMI Chemical composition Solubilization medium [12] conc. Carrier conc. /( g g 1 ) PTB 1978 NaI and Na 2 S 2 O 3 in water with 0.1 % Formalin water with 0.1 % Formalin Na 2 S 2 O 3 : 100 g g 1 Density /(g cm 3 ) NaI: < 0.01 % 1989 KI and Na 2 S 2 O 3 in water Na 2 S 2 O 3 : 50 g g 1 KI: < 0.03 % 1993 NaI and Na 2 S 2 O 3 and 0.1 % Formalin in NaOH 2004 NaI and Na 2 S 2 O 3 in HCHO (formaldehyde) Relative activity of impurity NaOH: 1 mol dm 3 with 0.1 % Formalin Na 2 S 2 O 3 : 45 g g 1 NaI: HCHO: 0.04 mol dm 3 NaI: Na 2 S 2 O 3 : 45 g g 1 BARC 1979 KI and Na 2 S 2 O 3 in water Na 2 S 2 O 3 : 60 g g 1 KI: NaI and Na 2 S 2 O 3 in water KI: NMISA 1980 KI and Na 2 SO 3 in diluted NaOH NaOH: mol dm 3 KI: Na 2 SO 3 : 1600 g g 1 NPL 1980 KI in NaOH NaOH: 0.01 mol dm 3 KI: NaI: NaI and Na 2 S 2 O 3 in diluted NaOH NaOH: mol dm 3 Na 2 S 2 O 3 : 25 g g 1 MKEH 1983 KI; KIO 3 and Na 2 S 2 O 3 in water Na 2 S 2 O 3 : 50 g g 1 KI: 50 < 0.01 % KIO 3 : I in NaOH NaOH: 0.01 mol dm 3 Continued overleaf 10/25

11 Table 3 continued. Details of the solution of 131 I submitted NMI Chemical composition Solubilization medium [12] conc. Carrier conc. /( g g 1 ) Density /(g cm 3 ) Relative activity of impurity IRA 1984 KI and Na 2 SO 3 LiOH: mol dm 3 KI: in diluted LiOH Na 2 SO 3 : 20 g g KI: 20 PTKMR1985 KI and Na 2 SO 3 in diluted LiOH LiOH: mol dm 3 NMIJ 1986 NaI and Na 2 SO 3 in diluted LiOH LiOH: mol dm # NaI and Na 2 S 2 O 3 in diluted LiOH LiOH: 20 g g 1 Na 2 SO 3 : 20 g g 1 KI: (1) Na 2 SO 3 : 30 g g 1 NaI: Na 2 S 2 O 3 : 20 g g 1 NaI: CMI-IIR1986 KI and Na 2 S 2 O 3 in NaHCO 3 NaHCO mol dm 3 Na 2 S 2 O 3 : 50 g g 1 KI: 50 < 0.1 % LNE-LNHB KI and Na 2 S 2 O 3 in water Na 2 S 2 O 3 : 50 g g 1 KI: 50 < 0.1 % 2006 KI and Na 2 S 2 O 3 in NaHCO 3 NaHCO 3 : mol dm 3 Na 2 S 2 O 3 : 50 g g 1 KI: 50 KI and Na 2 S 2 O 3 in diluted NaOH NaOH: mol dm 3 Na 2 S 2 O 3 : 50 g g 1 KI: < NaI and Na 2 S 2 O 3 5H 2 O in water Na 2 S 2 O 3 5H 2 O : mol dm 3 NaI: Continued overleaf -3 % 11/25

12 Table 3 continued. Details of the solution of 131 I submitted NMI Chemical composition Solubilization medium [12] conc. Carrier conc. /( g g 1 ) NIRH 1991 ANSTO 1994 LNMRI/IRD 1999 CIEMAT 2002 IFIN-HH 2005 Density /(g cm 3 ) Relative activity of impurity 1 NaI NaCl in water 1.01 Na 2 HPO 4 in water Na 2 HPO 4 : 6800 g g 1 NaI and Na 2 S 2 O 3 in NaOH NaOH: 0.01 mol dm 3 Na 2 S 2 O 3 : 80 g g 1 NaI: m Xe: 2.69 (45) % * NaI: 50 1 NaI: NaI and Na 2 S 2 O 3 in NaOH NaOH: 0.01 mol dm 3 Na 2 S 2 O 3 : 80 g g 1 KRISS NaI in NaOH NaOH: 0.1 mol dm 3 NaI: BEV NaI and Na 2 S 2 O 3 in HCHO HCHO: 0.04 mol dm Na 2 S 2 O 3 : 130 g g 1 the ratio of the activity of the impurity to the activity of 131 I at the reference date. * negligible effect on the SIR measurement # same solution as for the APMP.RI(II)-K2.I-131 comparison. 12/25

13 Table 4a. Results of SIR measurements of 131 I NMI NIST 1977 Mass of solution m i / g Activity submitted A i / kbq N of Ra source used SIR A e / kbq Relative uncertainty from SIR Combined uncertainty u c,i / kbq PTB (1) (9) (9) BARC 1979 NMISA * NPL MKEH IRA (1) PTKMR (1) Continued overleaf /25

14 Table 4a continued. NMI NMIJ 1986 Mass of solution m i / g Results of SIR measurements of 131 I Activity submitted A i / kbq N of Ra source used Metrologia 2014, 51, Tech. Suppl., SIR A e / kbq Relative uncertaint y from SIR Combined uncertainty u c,i / kbq a CMI-IIR 1986 LNE- LNHB (47) # 6 10 NIRH ANSTO 1994 LNMRI /IRD 1999 CIEMAT 2002 IFIN-HH 2005 KRISS b BEV the mean of the two A e values is used with an averaged uncertainty, as attributed to an individual entry [13] * mass of active solution before dilution: g and g respectively. # weighted mean value and standard uncertainty of the two results obtained with two different methods given in Table 2 a results used to link the APMP.RI(II)-K2 comparison b updated result now calculated with the half-life [6] as all the other SIR results As the half-life of the daughter radionuclide 131m Xe (T 1/2 of d [6]) is longer than that of the parent, no transient equilibrium can be reached. However, the contribution of 131m Xe to the SIR ionization current is negligible for a period of up to about 14/25

15 8 weeks after the source production, because of the low branching ratio and gammaemission probability. Measurements repeated at the BIPM after periods of up to two weeks later produced comparison results in agreement within the combined standard uncertainty for the NMIJ(2009) and for the LNE-LNHB(2012), and within two combined standard uncertainties for the NIST(2012). The NMIJ(2009) and LNE-LNHB(2012) results agree within standard uncertainty with their earlier SIR result. The NIST(2012) result is in close agreement with the NIST(1980 and 1981) results. No recent submission has been identified as a pilot study so the result of each NMI is normally eligible for the key comparison database (KCDB) of the CIPM MRA. However, the NIRH no longer undertake the metrology of activity and the PTKMR is not a designated laboratory of the Puslit KIM-LIPI, Indonesia, therefore none of these results is included in the KCDB. An international comparison for this radionuclide, APMP.RI(II)-K2.I-131 was held in 2009 [14] and the five laboratories from this comparison to be added to the matrix of degrees of equivalence are given in Table 1b 1. The results (A/m) i of the APMP comparison have been linked to the BIPM.RI(II)- K1.I-131 comparison through the measurement in the SIR of one ampoule of the APMP solution standardized by the NMIJ. The link is made using a normalization ratio deduced from the values in the row indicated in Table 4a: [ ] (a) The details of the links are given in Table 4b. The uncertainties for the APMP.RI(II) comparison results linked to the SIR are comprised of the original uncertainties together with the uncertainty in the link, , given by the relative standard uncertainty of the SIR measurement of the NMIJ ampoule. The ANSTO result in the APMP.RI(II)-K2.I-131 comparison is in close agreement with the earlier SIR result of the ANSTO. The BARC result in the APMP.RI(II)-K2.I- 131 comparison is more than two standard uncertainties lower than the earlier SIR results of the BARC. The KRISS result in the APMP.RI(II)-K2.I-131 comparison differs by about from the earlier SIR result of the KRISS. Consequently, the KRISS degree of equivalence for 131 I changed from the lowest value amongst the participants in 2008 to the highest value in Indonesia participated in the APMP.RI(II) comparison. However the PTKMR is not a designated laboratory of the Puslit KIM-LIPI, Indonesia, therefore their result is not included in the KCDB. 15/25

16 Table 4b. NMI Results of the 2009 APMP.RI(II) comparison of 131 I and links to the SIR Measurement method and acronym (see Appendix 2 and [14]) Activity* concentration measured (A/m) i / (kbq g 1 ) Standard uncertainty u i / (kbq g 1 ) Equivalent SIR activity A ei / kbq Combined standard uncertainty u ci / kbq ANSTO BARC INER KRISS 4P-PP-BP-NA-GR-CO OAP # NMIJ see Table 4a *referenced to 0:00 UTC 20 March 2009 # calibrated by 51 Cr, 57 Co, 60 Co, 85 Sr, 134 Cs and 137 Cs solutions traceable to the NMIJ 4.1 The key comparison reference value In May 2013 the CCRI(II) decided to no longer calculate the key comparison reference value (KCRV) by using an unweighted mean but rather by using the powermoderated weighted mean [15]. This type of weighted mean is similar to a Mandel- Paule mean in that the NMIs uncertainties may be increased until the reduced chisquared value is one. In addition, it allows for a power smaller than two in the weighting factor. Therefore, all SIR key comparison results can be selected for the KCRV with the following provisions: a) only results for solutions standardized by primary techniques are accepted, with the exception of radioactive gas standards (for which results from transfer instrument measurements that are directly traceable to a primary measurement in the laboratory may be included); b) each NMI or other laboratory has only one result (normally the most recent result or the mean if more than one ampoule is submitted); c) possible outliers can be identified on a mathematical basis and excluded from the KCRV using the normalized error test with a test value of 2.5 and using the modified uncertainties; d) results can also be excluded for technical reasons. e) The CCRI(II) is always the final arbiter regarding excluding any data from the calculation of the KCRV. The data set used for the evaluation of the KCRVs is known as the KCRV file and is a reduced data set from the SIR master-file. Although the KCRV may be modified when other NMIs participate, on the advice of the Key Comparison Working Group of the CCRI(II), such modifications are made only by the CCRI(II) during one of its biennial meetings as for the case of 131 I in June 2011, or by consensus through electronic means (e.g., ) as discussed at the CCRI(II) meeting in /25

17 As mentioned above, the rules to calculate a KCRV have been recently modified. Consequently, the KCRV agreed in 2011 for 131 I has been re-calculated as (35) kbq on the basis of the SIR results from the NMISA, IRA(1984), ASMW(1989), ANSTO, BARC(1994), MKEH(1998), NPL(1999), LNMRI/IRD, CIEMAT, PTB(2004), CMI-IIR(2006), IFIN-HH(2008), NMIJ(2009) and the recent results of the NIST(2012) and the LNE-LNHB(2012). The KRISS SIR result is identified as an outlier. The updated KCRV can be compared with the previous KCRV value of (40) kbq published in 2008 [5]. 4.2 Degrees of equivalence Every participant in a comparison is entitled to have one result included in the KCDB as long as the NMI is a signatory or designated institute listed in the CIPM MRA, and the result is valid (i.e., not older than 20 years). Normally, the most recent result is the one included. An NMI may withdraw its result only if all other participants agree. The degree of equivalence of a given measurement standard is the degree to which this standard is consistent with the KCRV [2]. The degree of equivalence is expressed quantitatively in terms of the deviation from the key comparison reference value and the expanded uncertainty of this deviation (k = 2). The degree of equivalence between any pair of national measurement standards is expressed in terms of their difference and the expanded uncertainty of this difference and is independent of the choice of key comparison reference value Comparison of a given NMI result with the KCRV The degree of equivalence of the result of a particular NMI, i, with the key comparison reference value is expressed as the difference D i between the values D i Ae KCRV (1) i and the expanded uncertainty (k = 2) of this difference, U i, known as the equivalence uncertainty; hence U u( D ). (2) i 2 i When the result of the NMI i is included in the KCRV with a weight w i, then u 2 (D i ) = (1-2w i ) u i 2 + u 2 (KCRV). (3) However, when the result of the NMI i is not included in the KCRV, then u 2 (D i ) = u i 2 + u 2 (KCRV). (4) Comparison between pairs of NMI results The degree of equivalence between the results of any pair of NMIs, i and j, is expressed as the difference D ij in the values D ij D D A A (5) i j and the expanded uncertainty (k = 2) of this difference, U ij = 2u(D ij ), where ei e j 17/25

18 2 2 2 u( Dij ) ui u j - 2u( Ae, i, Ae, j ) (6) where any obvious correlations between the NMIs (such as a traceable calibration, or correlations normally coming from the SIR or from the linking factor in the case of linked comparison) are subtracted using the covariance u(a ei, A ej ) (see [16] for more detail). However, the CCRI decided in 2011 that these pair-wise degrees of equivalence no longer need to be published as long as the methodology is explained. Table 5 shows the matrix of all the degrees of equivalence as they will appear in the KCDB. It should be noted that for consistency within the KCDB, a simplified level of nomenclature is used with A ei replaced by x i. The introductory text is that agreed for the comparison. The graph of the results in Table 5, corresponding to the degrees of equivalence with respect to the KCRV (identified as x R in the KCDB), is shown in Figure 1. This graphical representation indicates in part the degree of equivalence between the NMIs but obviously does not take into account the correlations between the different NMIs. It should be noted that the final data in this paper, while correct at the time of publication, will become out-of-date as NMIs make new comparisons. The formal results under the CIPM MRA [2] are those available in the KCDB. Conclusion The BIPM ongoing key comparison for 131 I, BIPM.RI(II)-K1.I-131 currently comprises twelve results. These have been analysed with respect to the updated KCRV determined for this radionuclide. The results of the APMP.RI(II)-K2.I-131 comparison held in 2009 have been linked to the BIPM comparison through the mutual participations of the NMIJ. This has enabled the table of degrees of equivalence to include seventeen results in total. Further results may be added as and when other NMIs contribute 131 I activity measurements to the ongoing comparison or take part in other linked comparisons. Acknowledgements The authors would like to thank Dr J.M. Los Arcos of the BIPM for editorial assistance. References [1] Ratel G., The Système International de Référence and its application in key comparisons, Metrologia, 2007, 44(4), S7-S16. [2] CIPM MRA: Mutual recognition of national measurement standards and of calibration and measurement certificates issued by national metrology 18/25

19 institutes, International Committee for Weights and Measures, 1999, 45 pp. [3] Ratel G., Michotte C., BIPM comparison BIPM.RI(II)-K1.I-131 of activity measurements of the radionuclide 131 I, Metrologia, 2003, 40, Tech. Suppl., [4] Ratel G., Michotte C., Kossert K., Janßen H., Activity measurements of the radionuclide 131 I for the PTB, Germany in the ongoing comparison BIPM.RI(II)-K1.I-131, Metrologia, 2005, 42, Tech. Suppl., [5] Ratel G., Michotte C., Sahagia M., Park T.S., Dryak P., Sochorová J., Maringer F.J., Kreuziger M., Results of the IFIN-HH (Romania), KRISS (Republic of Korea), CMI (Czech Republic) and the BEV (Austria) in the BIPM comparison BIPM.RI(II)-K1.I-131 of activity measurements of the radionuclide 131 I, Metrologia, 2008, 45, Tech. Suppl., [6] Bé M.-M., Chisté V., Dulieu C., Browne E., Chechev V., Kuzmenko N., Helmer R., Nichols A., Schönfeld E., Dersch R., 2004, Table of Radionuclides, BIPM Monographie 5, Vol 1, 249. [7] Michotte C., Efficiency calibration of the Ge(Li) detector of the BIPM for SIRtype ampoules, Rapport BIPM-1999/03, 15 pp. [8] Comité Consultatif pour les Étalons de Mesures des Rayonnements Ionisants 16th meeting (1999), 2001, CCRI(II) [9] Michotte C., Protocol on the use of the calibrated spectrometer of the BIPM for the measurement of impurities in ampoules submitted to the SIR, CCRI(II)/01-01, 2001, 2 pp. [10] BNM-CEA, Table de radionucléides, version 1975, BNM-LNHB, Gif-sur- Yvette. [11] BNM-CEA/DTA/DAMRI/LPRI, Nucléide, Nuclear and Atomic Decay Data Version : /12/98 CD ROM, BNM-LNHB, Gif-sur-Yvette. [12] Iroulart M.-G., Thermodynamic stability of radioactivity standard solutions, 2007, Monographie BIPM-6, 52 pp. [13] Woods M.J., Reher D.F.G. and Ratel G., Equivalence in radionuclide metrology, Applied Radiation and Isotopes, 52, (2000) [14] Yunoki A. et al., APMP comparison of activity measurements of I-131 (APMP.RI(II)-K2.I-131), Metrologia, 2013, 50, Techn. Suppl., [15] Pommé S., Determination of a reference value, associated standard uncertainty and degrees of equivalence for CCRI(II) key comparison data, European Commission, Joint Research Centre, Institute for Reference Materials and Measurements, 2012, Report EUR EN. Errata notice published in the CCRI(II) working document, 2013, CCRI(II)/ [16] Michotte C. and Ratel G., Correlations taken into account in the KCDB, CCRI(II) working document, 2003, CCRI(II)/ /25

20 Table 5. Table of degrees of equivalence and introductory text for 131 I Key comparison BIPM.RI(II)-K1.I-131 MEASURAND : Equivalent activity of 131 I Key comparison reference value: the SIR reference value x R for this radionuclide is kbq, with a standard uncertainty u R of 35 kbq. The value x i is taken as the equivalent activity for laboratory i. The degree of equivalence of each laboratory with respect to the reference value is given by a pair of terms: D i = (x i - x R ) and U i, its expanded uncertainty (k = 2), both expressed in kbq, and U i = 2((1-2w i )u i 2 + u R 2 ) 1/2 when each laboratory has contributed to the calculation of x R. When required, the degree of equivalence between two laboratories is given by a pair of numbers: D ij = D i - D j = (x i - x j ) and U ij, its expanded uncertainty (k = 2), both expressed in kbq. The approximation U ij 2 ~ 2 2 (u i 2 + u j 2 ) may be used. Linking APMP.RI(II)-K2.I-131 to BIPM.RI(II)-K1.I-131 The value x i is the equivalent activity for laboratory i participant in APMP.RI(II)-K2.I-131 having been normalized to the value of the NMIJ as the linking laboratory. The degree of equivalence of laboratory i participant in APMP.RI(II)-K2.I-131 with respect to the key comparison reference value is given by a pair of terms: D i = (x i - x R ) and U i, its expanded uncertainty (k = 2), both expressed in MBq. The approximation U i = 2(u i 2 + u R 2 ) 1/2 is used in the following table as none of these D i contributed to the KCRV. When required, the degree of equivalence between two laboratories i and j, one participant in BIPM.RI(II)-K1.I-131 and one in APMP.RI(II)-K2.I-131, 20/25

21 or both participants in APMP.RI(II)-K2.I-131, is given by a pair of terms expressed in MBq: D ij = D i - D j and U ij, its expanded uncertainty (k = 2), approximated by U ij ~ 2(u i 2 + u j 2-2fu l 2 ) 1/2 with l being the linking laboratory when each laboratory is from the APMP-K2 and f is the correlation coefficient. These statements make it possible to extend the BIPM.RI(II)-K1.I-131 matrices of equivalence to the participants in APMP.RI(II)-K2.I-131. Lab i D i U i / MBq IRA MKEH NPL LNMRI/IRD CIEMAT PTB CMI-IIR BEV IFIN-HH NMIJ NIST LNE-LNHB ANSTO BARC INER KRISS OAP /25

22 IRA MKEH NPL LNMRI/IRD CIEMAT PTB CMI-IIR BEV IFIN-HH NMIJ NIST LNE-LNHB ANSTO BARC INER KRISS OAP D i = (x i - x R ) / (MBq) [D i / x R ] / (kbq/mbq) Metrologia 2014, 51, Tech. Suppl., Figure 1. Graph of degrees of equivalence with the KCRV for 131 I (as it appears in Appendix B of the MRA) BIPM.RI(II)-K1.I-131 and APMP.RI(II)-K2.I-131 Degrees of equivalence for equivalent activity of 131 I N.B. The Right hand axis shows approximate values only 22/25

23 Appendix 1 Uncertainty budgets for the activity of 131 I submitted to the SIR NMIJ 2009, Relative standard uncertainties u i 10 4 evaluated by method Contributions due to A B dead time 1 resolving time 1 Gandy effect 2 counting time 1 background 5 half-life 2 extrapolation of efficiency curve # 23 Quadratic summation 24 3 Relative combined standard uncertainty, u c 24 # including from counting statistics, from weighing and from pileup NIST 2012, 4P-LS-BP-NA-GR-AC Relative standard uncertainties u i 10 4 Comments Evaluation type Contributions due to measurement variability 9 Std. dev. of 3 sources A dead time 10 Previous measurements B gravimetric links 5 Dilutions and sources B background 0 Embodied in meas. var. A half-life 1 From DDEP B extrapolation of efficiency curve 25 From f(x) and -ray window impurities 15 x-ray and -ray spec. B Relative combined standard uncertainty, u c 33 B 23/25

24 LNE-LNHB 2012, 4P-NA-GR HE Relative standard uncertainties u i 10 4 Comments Evaluation type Contributions due to counting statistics 30 Uniform distrib. applied on 6 sources measured dead time 1 Live-time technique B weighing 10 Pycnometer technique B background 10 A half-life 4 B detection efficiency 20 M-C simul. with Geant4 B extrapolation of efficiency curve 10 B Relative combined standard uncertainty, u c 40 B LNE-LNHB 2012, 4P PC-BP-NA-GR-AC Relative standard uncertainties u i 10 4 Comments Evaluation type Contributions due to counting statistics 10 A dead time 5 Live-time technique B weighing 5 B background 5 A half-life 5 B extrapolation technique 20 A Relative combined standard uncertainty, u c 25 24/25

25 Appendix 2. Acronyms used to identify different measurement methods Each acronym has six components, geometry-detector (1)-radiation (1)-detector (2)-radiation (2)-mode. When a component is unknown,?? is used and when it is not applicable 00 is used. Geometry acronym Detector acronym 4 4P proportional counter PC defined solid angle SA press. prop. counter PP 2 2P liquid scintillation counting LS undefined solid angle UA NaI(Tl) NA Ge(HP) Ge(Li) Si(Li) CsI(Tl) ionization chamber grid ionization chamber bolometer calorimeter PIPS detector Radiation acronym Mode acronym positron PO efficiency tracing ET beta particle BP internal gas counting IG Auger electron AE CIEMAT/NIST CN conversion electron CE sum counting SC mixed electrons ME coincidence CO bremsstrahlung BS anti-coincidence AC gamma rays GR coincidence counting with efficiency tracing X - rays XR anti-coincidence counting with efficiency tracing photons (x + ) PH triple-to-double coincidence ratio counting alpha - particle AP selective sampling SS mixture of various radiations GH GL SL CS IC GC BO CA PS CT AT TD MX high efficiency HE Examples method acronym 4 (PC) -coincidence counting 4 (PPC) -coincidence counting eff. trac. 4P-PP-MX-NA-GR-CT defined solid angle -particle counting with a PIPS detector SA-PS-AP (PPC)AX- (Ge(HP))-anticoincidence counting 4P-PP-MX-GH-GR-AC 4 CsI-,AX, counting 4P-CS-MX HE calibrated IC internal gas counting 4P-PC-BP IG 25/25