CCQM-P39.1 final report - BIPM · CCQM-P39.1 Final Report 1 Abstract CCQM-P39.1 was organised as a...

December 2005

CCQM-P39.1

As, Hg, Pb, Se and Methylmercury in Salmon

FINAL REPORT

Y. Aregbe and P.D.P. Taylor

With contributions from: D. Centioli, S. Barbizzi, M. Belli, C. Galas, S. Gaudino, S. Rosamilia from APAT J. Vogl, T. Win from BAM E. A. De Nadai Fernandes from CENA Zs. Jókai and P. Fodor from Corvinus University U. Sansone, M. J. Campell, A. Toervenyi from IAEA M. Horvat, V. Stibilj, V. Fajon, D. Gibičar, S. Žižek from JSI E. Krupp and O. Donard from LCABIE University Pau R. Hearn, J. Entwisle, D. Curtis and R. Wahlen from LGC Ltd J. Wang from NRCCRM B. Magnusson from SP H. Hintelmann from Trent University J. I. Garcia Alonso from University of Oviedo

The Mission of IRMM is to promote a common European measurement system in support of EU policies, especially health and consumer protection, environment, agriculture, internal market and industrial standards. European Commission Directorate-General Joint Research Centre Institute for Reference Materials and Measurements Contact information Yetunde Aregbe European Commission Directorate-General Joint Research Centre Institute for Reference Materials and Measurements Retieseweg 111 B-2440 Geel • Belgium Email: [email protected] Tel.: +32 (0)14 571 673 Fax: +32 (0)14 571 865 http://www.irmm.jrc.be http://www.jrc.cec.eu.int Legal Notice Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use, which might be made of the following information. A great deal of additional information on the European Union is available on the Internet. It can be accessed through the Europa server http://europa.eu.int Luxembourg: Office for Official Publications of the European Communities © European Communities, 2005 Reproduction is authorised provided the source is acknowledged Printed in Belgium

NOTICE: All the participants in CCQM-P39.1 gave their consent to the publication of this Final Report in the Metrologia Technical Supplement.

Contents ABSTRACT 1

INTRODUCTION 2

RATIONALE OF THIS COMPARISON 2

PARTICIPATION IN CCQM-P39.1 3

THE SALMON SAMPLE 4

HOMOGENEITY CHARACTERISATION 4

PARTICIPANT COORDINATION 4

CCQM-P39.1 INFORMATION DOCUMENTS 5 PROTOCOL FOR DRY-MASS CORRECTION 5 UNCERTAINTY EVALUATION 5 ANALYTICAL METHODS AND TECHNIQUES 7

CCQM-P39.1 PARTICIPANT’S RESULT 9

CCQM-P39.1 SUMMARY STATISTICS 10

THE CCQM-P39.1 GRAPHS 11

DISCUSSION 11

CONCLUSION 12

ACKNOWLEDGEMENTS 12

LIST OF ABBREVIATIONS 13

REFERENCES 14

Annex 1 15 Annex 2 27 Annex 3 37 Annex 4 47

CCQM-P39.1 Final Report

1

Abstract CCQM-P39.1 was organised as a follow-up pilot study in parallel to the key comparison CCQM-K43 after the previous pilot study on tuna fish. CCQM-P39.1 was an activity of the Inorganic Analysis Working Group (IAWG) of CCQM and was coordinated by the Joint Research Centre-Institute for Reference Materials and Measurements (IRMM, Geel, Belgium) of the European Commission (EC). In CCQM-P39.1 the amount contents of As, Hg, Pb, Se and Methylmercury (CH3Hg) in salmon (muscle and skin) were the measurands under investigation. Besides the National Metrology Institutes (NMIs) also non-IAWG members, expert laboratories for mercury and methylmercury measurements, were invited to participate in this pilot study. Results were reported by 6 IAWG members and 6 expert laboratories. During the CCQM-IAWG autumn meeting in Berlin, October 2005, it was agreed that in CCQM-K43 the KCRV is calculated as the mixture model median (MM-median) of all reported results. Therefore in CCQM-P39.1 the reported results are presented graphically with the KCRV from CCQM-K43. The reported results of the IAWG members fall within a range of ± 4% for arsenic and lead relative to the CCQM-K43 KCRV. For mercury, the spread was ±2%, but one IAWG member reported a very large uncertainty on the measurement result. For selenium the spread of IAGW members is ±2% deviation from the CCQM-K43 KCRV. Including the reported results from the invited expert laboratories, the spread of results increased for arsenic, lead and mercury to ±8%. The reported results including the experts fall within a range of ±20% for selenium and ±30% for methylmercury. The methods applied were isotope dilution mass spectrometry (IDMS) using sector field or quadrupole inductively coupled plasma-mass spectrometry (ICP-MS), external calibration or standard addition using ICP-MS, atomic

absorption spectrometry (AAS), atomic fluorescence spectrometry (AFS), atomic emission detection (AED) and electron capture detection (ECD). Instrumental neutron activation analysis (INAA) and k0-neutron activation analysis (k0-NAA) were also used as analytical techniques. Each of the 5 analytes under investigation was measured with at least 2 of these analytical techniques. IAEA reported 2 results for arsenic and lead. They were measured with ICP-MS and AAS, respectively. JSI reported 2 results for mercury, measured with NAA and AAS, and furthermore 2 results for methylmercury measured with GC-ECD and GC-CV-AFS. This report presents the participants’ results in CCQM-P39.1 for all analytes under investigation. In Annex 1, the results are displayed with the CCQM-K43 KCRV. In Annex 2, the different approaches for methlymercury measurements are presented in more detail. In Annex 3, the questionnaire data are presented. Annex 4 compiles all the CCQM-P39.1 information documents. The key comparison CCQM-K43 was carried out in parallel to this pilot study for the same measurands in the same salmon material Participation was meant for IAWG members, to support their CMCs in Appendix C.

CCQM-P39.1 As, Hg, Pb, Se and Methylmercury in Salmon

2

Introduction The previous pilot study CCQM-P39 on tuna fish [1] was proceeded in autumn 2004 to the key comparison, CCQM-K43 [2], and this pilot study, CCQM-P39.1. In alignment with the previous CCQM-P39 on tuna fish the choice of analytes under investigation in CCQM-P39.1 were exactly the same. The purpose of CCQM-K43&P39.1 was to demonstrate the measurement capability of determining the amount content of metals in a more challenging ‘real-life’ fish sample, with significantly lower mercury content as the preceding tuna fish sample, but with similar arsenic, lead and selenium content, in order to demonstrate ‘how far the light shines’. Participants in CCQM-P39.1 could demonstrate their measurement capabilities over an order of magnitude concentration range, particularly on total mercury and or-ganomercury. Furthermore the matrix ‘salmon muscle and skin’ was from an analytical sample pretreatment point of view more challenging than the freeze-dried tuna muscle powder used in CCQM P39. The same well known expert laboratories in mercury and methlymercury measurements as in CCQM-P39 were invited to join this study. Some of these expert laboratories also took the opportunity to report results for arsenic, lead and selenium, even if they might be less experienced in this kind of analysis. Rationale for this comparison In order to protect public health it is essential to keep contaminants at levels which are toxicologically acceptable. Thus surveillance measures are taken regarding the presence of contaminants in foodstuff, including fish. Mercury is a potential environmental toxin. The main source of human intake of mercury contaminants originates from methylmercury in fish and fish products. Methylmercury is particularly interesting due to its high toxicity compared to inorganic mercury and its high

proportion among organomercury species in the environment. Mercury species may induce alterations in the normal development of the brain of infants and may induce neurological changes in adults. Lead may induce reduced cognitive development in children and increased blood pressure and cardio-vascular diseases in adults. To protect public health, maximum levels of mercury and lead in fish products are laid down in relevant regulations. The EC Directive 2001/22 describes the community methods for the sampling, the sample preparation and the analysis of mercury and lead in fish [3]. The EC Regulation (466/2001) endorses officially the threshold value of 0.5 mg·kg-1 mercury and

0.2 mg·kg-1 lead in salmon [4]. Selenium is an essential trace element for human beings. Seafood is an important source of selenium intake for people in some regions. Certain forms of cancer and cardiovascular diseases have also been associated with selenium deficiency. Selenium is also counted among the most important elements in terms of food-chain contamination. Selenium has the narrowest plateau between concentrations that show deficiency and toxic effects, respectively. Recently, the EC requested the Scientific Committee on Food (SCF) to re-view the upper level of daily intake of individual vitamins and minerals, amongst them selenium, and to pro-vide the basis for the establishment of safety factors [5]. Rules for measurements of arsenic are set in the commission decision on implementing Council Directive 96/23/EC concerning the performance of analytical methods and the interpretation of results [6].


3

Participation in CCQM-P39.1

CC CC QQ MM -- PP 33 99 .. 11 PP AA RR TT II CC II PP AA NN TT CC OO UU NN TT RR YY

IAWG members

BAM Bundesanstalt für Materialforschung und –prüfung Germany

CENA Centro de Energia Nuclear na Agricultura

Brazil

IAEA International Atomic Energy Agency United Nations

LGC Ltd. Laboratory of the Government Chemist

United Kingdom

NRCCRM National Research Centre for Certified Reference Material

China

SP Sveriges Provnings- och Forskningsinstitut

Sweden

Invited expert laboratories

APAT Agenzia Protezione Ambiente e servizi Tecnici

Italy

Corvinus University of Budapest Hungary

JSI Jozef Stefan Institute

Slovenia

LCABIE – University of Pau Laboratory of analytical and bio-inorganic environmental chemistry

France

Trent University Canada

University of Oviedo Spain


4

The salmon sample The CCQM-P39.1 fish sample is a freeze dried and ground salmon muscle and skin powder bottled in amber glass vials each one containing about 2.2g. Each participant received 4 salmon samples (8-10g) to account for the small unit size. Participants were asked to analyse at least 2 sample aliquots from 2 bottles for each measured analyte. This salmon originates from Norway and is a real-life sample. The range of lead amount content in the salmon was just slightly exceeding the upper limit as stated in the EC Regulation (466/2001), whereas the mercury content was far below the upper legislation limit [4]. Therefore, this salmon material was perfectly appropriate for the purpose of a CCQM key comparison complementary to the previous pilot study on tuna fish, which exceeded the upper limit for mercury [1]. Homogeneity characterisation Within and between bottle homogeneity tests for Pb were carried out on 10 sub-samples of 10 bottles using solid sample Zeemann AAS (SS-ZAAS) and for Hg and MeHg on 3 sub-samples from 6 bottles applying IDMS. For As and Se, the homogeneity was assessed by analysing 2 sub-samples from 10 bottles applying k0-NAA. Results from these measurements were evaluated accordingly and compared to the procedures established in ISO Guide 35:1989 for the certification of reference materials based on analysis of variance ANOVA [7, 8]. The between bottle homogeneity of all 5 analytes was ranging from 0.5% - 2.5%, which was found to be adequate for this key comparison. These results were also confirmed by the observed spread of the CCQM-K43 participants’ results, where each reported result of the total amount content represents at least measurements of 2 sub-samples from 2 bottles [2].

Participant Coordination According to the rules for participation in key comparisons only IAWG members who meet the requirements given in paragraph 6 of the MRA could participate in CCQM-K43 [9]. The call for participation in CCQM-P39.1 was circulated to the IAWG members and expert laboratories in October 2004. The information package containing the salmon samples was sent to all registered participants in December 2004. The participants were asked to register for participation and to report their results and questionnaire answers online via the IRMM IT system for Interlaboratory Comparisons, called MILC. The first deadline for result reporting was 31st March 2005, which was finally extended to 4th April 2005. The results were presented at the Spring IAWG meeting in April at BIPM. In summer 2005, a summary report was circulated amongst all participants for confirmation of the reported results and applied analytical methods. Final results and proposal for the CCQM-K43 KCRV were presented at the last CCQM-IAWG meeting in Berlin in October 2005.


5

CCQM-P39.1 information documents

o Letter to the participants pointing out the deadlines

o Scope of the CCQM-P39.1 pilot study o General instructions o Instructions for online result and

questionnaire reporting in MILC o Instructions for uncertainty evaluation o Instructions for the dry-mass

correction and digestion of the salmon o Information for CCQM-P39.1

participants measuring methylmercury, including the ERM AE670 certificate and the publication on the preparation and certification of ERM AE670

Protocol for dry-mass correction The determination of the moisture content of the samples is to some extent “operationally defined” [10, 11]. In view of the comparability of the results, the protocol in CCQM-P39.1 for correction of the moisture was as follows: “For correction of the measured values to dry-mass, water content measurements should be made on a separate portion of the same material with a mass of 0.6 ± 0.1g. The material should be dried before weighing for a minimum of 24 hours in a ventilated oven at 102 ± 2°C. Cycles of drying and weighing should be repeated until a constant mass is attained. Each weighing has to be carried out after the sample has reached thermal equilibrium at room temperature in a dessicator, time about 30 min. Successive weights should not differ more than 0.001 g. The loss of mass corresponds to the “dry-mass correction factor” that should be applied”. The water content was also determined by Karl-Fischer titration. No significant difference was observed. The reported results in CCQM-P39.1 for the measured water content in the salmon samples are listed in Table 1.

Uncertainty evaluation The participants were asked to report a complete uncertainty budget with their results. The uncertainty statement should be evaluated and presented according to the principles outlined in, e.g. “ISO/GUM” [12] or the EURACHEM/CITAC Guide [13]. The organising laboratory asked all the participants in CCQM-P39.1 the following:

o state your measurement equation o identify all significant sources of

uncertainty o state your input quantities o quantify uncertainty components and

convert them to standard uncertainties o calculate the combined standard

uncertainty uc o present an expanded uncertainty U

with the coverage factor k=2 o include factors related to sample

treatment in your measurement equations

o describe the applied evaluation process and type of assumed distribution for your uncertainty estimation

The complete uncertainty statement should be forwarded to the organising laboratory as attachment to the result reporting sheets. 10 participants in CCQM-P39.1 provided an uncertainty budget together with their measurement results. University of Oviedo did not forward a complete uncertainty budget for MeHg. Corvinuns University did not provide an uncertainty budget but reported that their uncertainty was derived from results from method validation.


6

Table 1 Reported values for water content determination and dry-mass correction

CCCCQQMM--PP3399 ..11 CC oo rr rr ee cc tt ii oo nn ff oo rr dd rr yy -- mm aa ss ss

PP AA RR TT II CC II PP AA NN TT WW AA TT EE RR CC OO NN TT EE NN TT II NN %% CC OO RR RR .. FF AA CC TT OO RR CC OO RR RR EE CC TT II OO NN

APAT 2.355 ± 0.003 0.9765 ± 0.001

BAM 2.1 ± 0.1 0.979 ± 0.001

CENA 2.99 ± 0.06 0.9701 ± 0.0010

Corvinus University of Budapest 1.10 ± 0.09 0.9890 ± 0.0009

IAEA 2.2 ± 0.2 0.978 ± 0.002

JSI 2.425 ± 0.02 1.02485 ± 0.01

LCABIE – University of Pau 3.47 ; 2.91 ; 2.64 0.9653 ± 0.00122; 0.9709

± 0.00133; 0.9736 ± 0.000113

LGC Ltd. 1.84 ± 0.13 0.9816 ± 0.0013

NRCCRM 0.633 ± 0.10 0.9937 ± 0.0010

SP 2.6 ± 0.3 0.974 ± 0.003

Trent University 2.33 ± 0.26 0.9767 ± 0.0026

University of Oviedo 2.11 ± 0.04 0.9789 ± 0.0004


7

Analytical methods and techniques Different methods and instrumental techniques besides IDMS were applied in CCQM-P39.1. The methods applied were:

o isotope dilution mass spectrometry (IDMS) using sector field or quadrupole inductively coupled plasma-mass spectrometry (ICP-MS)

o external calibration using ICP-MS o atomic absorption spectrometry (AAS) o atomic fluorescence spectrometry

(AFS) o atomic emission detection (AED) o electron capture detection (ECD) o Instrumental neutron activation

analysis (INAA) and k0-neutron activation analysis (k0-NAA)

Each of the 5 analytes under investigation was measured with at least 2 of these analytical techniques. The analytical methods and instrumental techniques are listed in Table 2 and Table 3. In Annex 1 of this report, the CCQM-P39.1 results are graphically presented, as well separately as together with the CCQM-K43 results.

Table 2 Analytical methods and instrumental techniques in CCQM-P39.1 for methylmercury

CCCCQQMM--PP3399..11 aa nn aa ll yy tt ee :: MM ee HH gg

PP AA RR TT II CC II PP AA NN TT AA NN AA LL YY TT II CC AA LL MM EE TT HH OO DD II NN SS TT RR UU MM EE NN TT AA LL TT EE CC HH NN II QQ UU EE

BAM other method than species specific IDMS GC-AED

Corvinus University of Budapest other method than species specific IDMS SPME-GC-AFS

JSI other method than species specific IDMS GC-ECD, GC-CV-AFS

LCABIE_ University of Pau direct IDMS using ERM-AE670 as spike GC-ICP-MS

NRCCRM direct IDMS using ERM-AE670 as spike HPLC-ICP-MS

Trent University direct IDMS using ERM-AE670 as spike GC-ICP-MS

Universtity of Oviedo direct IDMS using ERM-AE670 as spike GC-ICP-MS


8

Table 3 Analytical methods and instrumental techniques in CCQM-P39.1 for As, Hg, Pb, Se

CCCCQQMM--PP3399..11 aa nn aa ll yy tt ee :: AA ss ,, HH gg ,, PP bb ,, SS ee

PP AA RR TT II CC II PP AA NN TT SS AA NN AA LL YY TT II CC AA LL MM EE TT HH OO DD II NN SS TT RR UU MM EE NN TT AA LL TT EE CC HH NN II QQ UU EE

APAT Pb, Se, As: External Calibration with internal standardisation Hg: AAS

Pb: ICP-MS Se, As: ICP-MS with octopole collision cell Hg: DMA

CENA NAA INAA

IAEA External calibration As, Pb: ICP-MS As, Pb:GF-AAS

JSI As, Hg, Se: RNAA Hg:AAS

As, Hg, Se: RNAA; Counting: Well type HP Ge detector, with 40 % efficiency connected to Genie 2000 multichannel analyser. Hg:CV-AAS

LGC Ltd. External Calibration with internal standardisation ICP-QMS

NRCCRM As: External calibration Se: double IDMS ICP-MS

SP IDMS ICP-MS

Trent University IDMS Continuous flow CV-ICP-MS


9

CCQM-P39.1 participant’s result The CCQM-P39.1 participants’ results, as reported to the organising institute, are listed in Table 4.

Table 4 CCQM-P39.1 participants reported results

REPORTED RESULTS AND EXPANDED UNCERTAINTY (k=2) in 10-6 mol·kg-1

Participant As Hg MeHg Pb Se

APAT 53.29 ± 9.20 0.418 ± 0.054 - 4.08 ± 0.58 5.90 ± 1.45

BAM - - 0.252 ± 0.031 - -

CENA 57.83 ± 1.04 0.390 ± 0.098 - - 7.44 ± 0.19

Corvinus University of Budapest

- - 0.391 ± 0.057 - -

IAEA_ICP-MS 59.38 ± 2.97 - - 3.64 ± 0.21 -

IAEA_GF-AAS 60.34 ± 3.98 - - 3.87 ± 0.10 -

JSI_NAA 56.50 ± 4.02 0.365 ± 0.063 - - 8.21 ± 0.42

JSI_AAS - 0.389 ± 0.042 - - -

JSI_AFS - - 0.300 ± 0.038 - -

JSI_ECD - - 0.320 ± 0.022 -

LCABIE_ University of Pau

- - 0.356 ± 0.070 - -

LGC 58.6 ± 2.9 - - - -

NRCCRM 60.18 ± 3.37 - 0.400 ± 0.024 - 7.38 ± 0.22

SP - 0.382 ± 0.015 - - -

Trent University - 0.394 ± 0.025 0.389 ± 0.023 - -

Univ. Oviedo - - 0.503 ± 0.112 - -


10

CCQM-P39.1 summary statistics At the CCQM meeting in Paris in April 2004, David L. Duewer from NIST presented a robust approach for the determination of CCQM Key Comparison Values and Uncertainties to the IAWG members [14]. He introduced the mixture model probability density function (MM-PDF) for each measurement population as a means of data analysis for key comparisons and pilot studies. MM-PDF based summary statistics enable estimation of the expected performance of the majority of participants of a key comparison or pilot study [15]. The “true value” of a measurand in a given material can be estimated in a robust way even when some of the results are not in accordance with the majority. During the last autumn IAWG meeting in Berlin 2005, the mixture model median (MM-median) and the MM-median based Standard Deviation S(MM-

Median) was chosen as robust estimation of the KCRV in CCQM-K43. In the table below the CCQM-K43 KCRV_MM-median is listed in bold for all the analytes as µ ± σ*ts/ √n which is the 95% confidence interval on µ for n reported results. Below the CCQM-P39.1_MM-median, derived from the CCQM-P39.1participants’ result, is given in the same way. (Due to the limited number of participants for Pb the dispersion multiplied by a factor of 2 was given instead of the 95% confidence interval on µ for n reported results). From Table 5 it can be seen that there is no significant difference of the CCQM-K43 KCRV compared to the CCQM-P39.1 MM-median.

Table 5 CCQM-K43_KCRV and CCQM-P39.1_MM-median

µ ± σ*ts / √n in 10-6 mol·kg-1

As Hg MeHg Pb Se

CCQM-K43-KCRV_MM-median 58.1 ± 1.5 0.3884 ± 0.0056 0.373 ± 0.023 3.82 ± 0.10 7.32 ± 0.28

CCQM-P39.1_MM-median 58.4 ± 2.0 0.385 ± 0.025 0.371 ± 0.057 3.79 ± 0.34* 7.41 ± 0.51

*Dispersion S(MM-Median) multiplied by 2 The CCQM-P39.1 graphs The CCQM-P39.1 results for As, Hg, Pb and Se are presented graphically in Annex 1 of this report. In the first 4 graphs, all CCQM P39.1 results are plotted in ascending order, including the KCRV. In the following 4 graphs the CCQM-P39.1 results are plotted together with the CCQM-K43 results. In Annex 2 the

results and questionnaire data for MeHg are presented and discussed in more detail. Table 6 summarises all the CCQM-P39.1 graphs of Annex 1 and Annex 2.


11

Table 6 CCQM-P39.1 graphs

General Graphs - Annex 1 As, Hg, Pb and Se

CCQM-P39.1 participants with KCRV CCQM-P39.1 and CCQM-K43 participants with KCRV

All participants – Amended results

Methylmercury Graphs - Annex 2 MeHg (CH3Hg)

CCQM-P39.1 with KCRV CCQM-P39.1 and CCQM-K43 participants with KCRV

Discussion All participants in CCQM-P39.1 had previously participated in the pilot study on tuna fish, except NRCCRM and APAT. As can be seen from Table 4 and the graphs in Annex 1 and Annex 2, the agreement of measurement results between IAWG members is very good, except for MeHg. It has to be mentioned that the 2 NMIs that participated in CCQM-P39.1 did not measure MeHg in the previous CCQM-P39 pilot study on tuna fish. The reported results of the IAWG members fall within a range of ± 4% for arsenic and lead relative to the CCQM-K43 KCRV. For mercury, the spread was ±2%, but CENA reported a very large uncertainty on the measurement result. After the results have been presented at the IAWG meeting in Spring 2005, CENA revised their uncertainty budget for Hg and found a relative expanded uncertainty of 18% instead of the reported 25% (see Figure 10, Annex 1). For selenium the spread of IAGW members is ±2% deviation from the CCQM-K43 KCRV. Amongst the 12 institutes that participated in CCQM P39.1 were 6 IAWG members and 6 invited expert because of their expertise in Hg and MeHg measurements. Including the reported results from the invited expert laboratories, the spread of results increased for arsenic, lead and mercury to ±8%, which is an improvement compared to the spread of results observed in CCQM-P39 [1]. The

reported CCQM-P39.1 results including the experts fall within a range of ±20% for selenium. For MeHg the overall spread of IAWG members and invited experts together is about ± 30%, which is larger than the spread that was observed in the previous pilot study on tuna fish [1]. After the CCQM-P39.1 summary report was distributed to the participants for comments, JSI investigated further their reported results. JSI encountered some problems with the software they used to calculate the results with uncertainty. They observed a slight difference when comparing the amount content of Hg calculated directly from the chromatograms (peak heights/area and calibration) with the Hg content as calculated by the software. JSI’s reported data in CCQM-P39.1 for Hg are based on the software calculation. Furthermore, there was an additional problem with the expression of uncertainty for the RNAA measurements. First JSI estimated that the recovery was 90 ± 5%, but after further investigation JSI found out that this recovery rate was overestimated. JSI asked to include in the final CCQM-P39.1 report results with amended expanded uncertainties, which are half of the originally reported uncertainties (see Figure 9 and Figure 10, Annex 1).


12

The University of Oviedo also reported that they encountered problems in changing the analytical procedure completely compared to the one they applied in CCQM-P39 on the tuna fish sample (see Annex 2). APAT assumes that the bias of their As, Se results could be due to the octopole collision cell, which on the one hand eliminates matrix interferences, but on the other hand probably reduces also the sensitivity of the ICP-MS analytical technique. As a consequence APAT has developed an up-grade of the ICP-MS with a new collision cell and improved sensitivity. Conclusion The ‘light shines very far’ for the determination of metals in fish tissue when drawing the conclusion from CCQM-K43, CCQM-P39.1 and the previous pilot study on the tuna fish, CCQM-P39. Due to the fact that this salmon material was a more challenging ‘fatty’ matrix compared to the dry tuna powder used in CCQM-P39 and due to the fact that the mercury amount content was 2 orders of magnitude lower than the one in the tuna fish sample, the salmon was an excellent choice for the proceeding of CCQM-P39 to a key comparison and this pilot study. This is a good example of a pilot study in parallel to a key comparison involving not only the NMIs, but also expert laboratories of their countries as it is one of the NMIs’ major tasks to disseminate metrological principles and good measurement practice to the laboratories in their countries. The analytes under investigation in CCQM-P39.1 have been measured with different techniques. Reliable measurements of highest metrological quality can be performed with various instrumental techniques (ICP-MS, AAS, AFS, NAA and k0 NAA) and analytical methods (IDMS, external calibration, standard addition and non destructive analysis). They are not method dependent. Nevertheless, it was observed that a few laboratories had problems with the MeHg analysis, most probably because of changes in sample preparation due to the high fat content of this salmon material. In addition

some of the CCQM-P39.1 participants still had difficulties in establishing a complete uncertainty budget and thus give a reliable estimate on the uncertainties of their reported measurement results. The main purpose of any Interlaboratory Comparison, thus also of CCQM comparisons, is to assess capabilities and to discover problems and correct analytical procedures accordingly. In this sense, IRMM would like to thank all the CCQM-P39.1 participants with a result significantly deviating from the CCQM-K43 KCRV or an overestimated uncertainty, that they provided information to be included in the final report on their investigation to find the origin of this bias. The majority of the participants have proven with their participation in CCQM-P39.1 (salmon) and the previous CCQM-P39 (tuna fish) not only their measurement capabilities on measurements close to legal limits, but also on low trace level measurements. In addition they have demonstrated their capability on appropriate sample treatment for a wide range of fish samples. Together with this report a consent form for publication will be distributed to the participants. After the agreement of all participants, the final report of the pilot study CCQM-P39.1 will also be published in the metrologia technical supplement. Acknowledgements Special thanks to F. Ulberth and A. Held from the RM unit for their support in providing the suitable salmon sample and in the design of the homogeneity study. The authors acknowledge furthermore the input of all the IRMM scientists who contributed to the reprocessing and characterisation of the tuna fish sample : - K. H. Grobecker, M. Bickel, A. Michiels, A. Bernreuther, S. Yazgan, P. Conneely. Furthermore the assistance of our MILC administrator S. Bynens is warmly acknowledged and the support of M. Sargent as chairman of the CCQM IAWG.


13

List of abbreviations

AAS Atomic Absorption Spectrometry AFS Atomic Fluorescence Spectrometry BIPM Bureau International des Poids et Mesures (Paris, France) CCQM Comité Consultatif pour la Quantité de Matière CITAC Co-operation for International Traceability in Analytical Chemistry CRMs Certified Reference Materials EC European Commission EU European Union EURACHEM A focus for Analytical Chemistry in Europe GC-AED Gas Chromatography Atomic Emission Detection GC-ID-ICP-MS Gas Chromatography Isotope Dilution Inductively Coupled Plasma Mass

Spectrometry GC-ECD Gas Chromatography Electron Capture Detection GUM Guide for expression for Uncertainty in Measurement IAWG Working Group on Inorganic Analysis ICP-MS Inductively Coupled Plasma-Mass Spectrometry IDMS Isotope Dilution Mass Spectrometry INAA Instrumental Neutron Activation Analysis IRMM Institute for Reference Materials and Measurements (EC, Geel, Belgium) ISO International Organisation for Standardisation JRC Joint Research Centre MM-median Mixture Model Median NAA Neutron Activation Analysis NMIs National Metrology Institutes RNAA Radiochemical neutron activation analysis SCF Scientific Committee on Food SS-ZAAS Solid Sample Zeemann Atomic Absorption Spectrometry


14

References

1. Aregbe Y., Quétel C., Taylor P.D.P., CCQM-P39: As, Hg, Pb, Se and methylmercury in tuna fish, Metrologia, 2004, 41, Tech. Suppl., 08004

2. Aregbe Y,.Taylor P.D.P., CCQM-K43: As, Hg, Pb, Se and methylmercury in Salmon, Metrologia, 2006, Tech. Suppl., in press

3. Commission Directive 2001/22/EC laying down the sampling methods of anlysis for the official control of lead, cadmium, nmercury and 3-MCPD in foodstuffs Official Journal of the European Communities L77/14, 08/03/2001

4. Commission Regulation (EC) No 466/2001 of 8 March 2001 setting maximum levels for certain contaminants in foodstuff, Official Journal of the European Communities L077, 16/03/2001, 0001-0013

5. SCF/CS/NUT/UPPLEV/11 FINAL, Nov/2000

6. Commission Decision No 2002/657/EC of 17 August 2002 implementing Coucil Directive 96/23/EC concerning performance of analytical methods and the interpretation of results, Official Journal of the European Communities L221, 17/08/2002

7. T. P. J. Linsinger et al. Accred.Qual.Assur (2001) 6:20-25

8. A.M.H. van der Veen et al. Accred.Qual.Assur (2001) 6:26-30

9. http://www.bipm.org/utils/en/pdf/mra.pdf 10. S. Rückhold et al. Fresenius J. Anal. Chem

(2000) 368: 522-527

11. S. Rückhold et al . Fresenius J. Anal. Chem (2000) 370: 189-193

12. International Organisation for Standardisation, “Guide to the Expression of Uncertainty in Measurement”, ©ISO, ISBN 92-67-10188-9, Geneva, Switzerland, 1993.

13. Eurachem/CITAC Guide Quantifying uncertainty in analytical measurement (2nd ed. 2000), www.eurachem.bam.de

14. David L. Duewer, A robust approach for the determination of CCQM key comparison reference values and uncertainties, Working document CCQM 04-15, BIPM, 2004

15. P. Ciarlini, M. Cox, F. Pavese, G. Regoliosi. The use of a mixture of probability distributions in temperature interlaboratory comparisons. Metrologia 2004;41:116-121

15

Annex 1 – Graphical presentation Contents CCQM-P39.1: As, Hg, Pb, Se and Methylmercury in Salmon

Figures 1 - 4 _________________________________________________________ 18-19 All participants with CCQM-K43 KCRV ________________________________________

Figures 5 - 8 _________________________________________________________ 22-23 CCQM-P39.1 and CCQM-K43 results with CCQM-K43 KCRV ______________________

Figures 9 - 10 ___________________________________________________________ 25 All participants - with JSI and CENA amended value ___________________________________

17

CCQM-P39.1: As, Hg, Pb, Se and Methylmercury in Salmon

Annex 1 – All participants CCQM-P39.1 results

Figure General Graphs Page number

Figure 1 All participants - As 18

Figure 2 All participants - Hg 18

Figure 3 All participants - Pb 19

Figure 4 All participants - Se 19

CCQM-P39.1 As, Hg, Pb, Se and Methylmercury in Salmon - Annex 1

18

Figure 1

IAEA_GF-AAS

NRCCRMIAEA_ICP-MS

LGCCENA

JSI

APAT

46.48

51.48

56.48

61.48

66.48

c

10

-6 m

ol·k

g-1

-20

-15

-10

-5

0

5

10

15

20

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-P39.1: As in salmonCCQM-K43 KCRV_Mixture Model-median: 58.1 ± 1.5 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]


Figure 2

APAT

Trent UnivCENA

JSI_AAS

SPJSI_NAA

0.31

0.33

0.35

0.37

0.39

0.41

0.43

0.45

c

10

-6 m

ol·k

g-1

-20

-15

-10

-5

0

5

10

15

20

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-P39.1: Hg in salmonCCQM-K43 KCRV_Mixture Model-median: 0.3884 ± 0.0056 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]


CCQM-P39.1 Final Report - Annex 1

19

Figure 3

APATIAEA_GF-AAS

IAEA_ICP-MS

3.1

3.3

3.5

3.7

3.9

4.1

4.3

4.5

c

10

-6 m

ol·k

g-1

-20

-15

-10

-5

0

5

10

15

20

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-P39.1: Pb in salmonCCQM-K43 KCRV_Mixture Model-median : 3.82 ± 0.10 ·10-6 mol·kg-1; [ µ ± σ*ts/√n]


Figure 4

JSI

CENANRCCRM

APAT

3.66

4.66

5.66

6.66

7.66

8.66

9.66

10.66

c

10

-6 m

ol·k

g-1

-50

-40

-30

-20

-10

0

10

20

30

40

50

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-P39.1: Se in salmonCCQM-K43 KCRV_Mixture Model-median: 7.32 ± 0 .28 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]



20

21


Annex 1 – All participants CCQM-P39.1 and K43 results

Figure Key comparison and pilot study graphs Page number

Figure 5 CCQM-P39.1 and CCQM-K43 - As 22

Figure 6 CCQM-P39.1 and CCQM-K43 - Hg 22

Figure 7 CCQM-P39.1 and CCQM-K43 - Pb 23

Figure 8 CCQM-P39.1 and CCQM-K43 - Se 23


22

Figure 5 CCQM-P39.1 and CCQM-K43 results with CCQK K43 KCRV – As

Figure 6 CCQM-P39.1 and CCQM-K43 results with CCQK K43 KCRV - Hg

NMIJLGC

INMETROCENA

IRMMNIST

JSILNE

IMGC

APAT

BAM

IAEA_ICP-MS

NMIA

IAEA

NRCCRM

IAEA_GF-AAS

46.5

51.5

56.5

61.5

66.5

c

10

-6 m

ol·k

g-1

-20

-15

-10

-5

0

5

10

15

20

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-K43 & P39.1: As in salmonMixture Model-median from CCQM-K43: 58.1 ± 1.5 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]

CCQM - P39.1 CCQM - K43

APATLGCLNE

Trent Univ

NMIJ

CENA

JSI_AAS

NMIANRCNIST

KRISSIRMM

SP

JSI_NAA

0.31

0.33

0.35

0.37

0.39

0.41

0.43

0.45

c

10

-6 m

ol·k

g-1

-20

-15

-10

-5

0

5

10

15

20

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-K43 & P-39.1 : Hg in salmonMixture Model-median from CCQM-K43: 0.3884 ± 0.0056 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]



23

Figure 7 CCQM-P39.1 and CCQM-K43 results with CCQK K43 KCRV – Pb

Figure 8 CCQM-P39.1 and CCQM-K43 results with CCQK K43 KCRV - Se

IAEA_ICP-MS

IAEABAM KRISS NMIJ

IAEA_GF-AAS NMIA

APAT

3.1

3.3

3.5

3.7

3.9

4.1

4.3

4.5

c

10

-6 m

ol·k

g-1

-20

-15

-10

-5

0

5

10

15

20

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-K43 & P39.1: Pb in salmonMixture Model-median from CCQM-K43: 3.82 ± 0.10 ·10-6 mol·kg-1; [ µ ± σ*ts/√n]


IAEA

JSILNE

NISTCENABAM

NRCCRMIMGCNMIANMIJIRMMLGC

APAT

INMETRO

3.66

4.66

5.66

6.66

7.66

8.66

9.66

10.66

c

10

-6 m

ol·k

g-1

-50

-40

-30

-20

-10

0

10

20

30

40

50

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-K43 & P39.1: Se in salmonMixture Model-median from CCQM-K43: 7.32 ± 0.28 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]



24


Annex 1 – All participants – amended results

Figure Amended graphs Page number

Figure 9 All participants - with JSI amended value - As 25

Figure 10 All participants - with CENA and JSI amended value s - Hg 25


25

Figure 9 All participants - with JSI amended value - As

Figure 10 All participants - with CENA and JSI amended value - Hg

IAEA_GF-AAS

NRCCRMIAEA_ICP-MS

LGCCENAJSIAPAT

46.48

51.48

56.48

61.48

66.48

c

10

-6 m

ol·k

g-1

-20

-15

-10

-5

0

5

10

15

20

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-P39.1: As in salmonCCQM-K43 KCRV_Mixture Model-median: 58.1 ± 1.5 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]


mean ± 2 STDEV

amended value

APAT

JSI_AASTrent UnivCENA

SP

JSI_NAA

0.31

0.33

0.35

0.37

0.39

0.41

0.43

0.45

c

10

-6 m

ol·k

g-1

-20

-15

-10

-5

0

5

10

15

20

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-P39.1: Hg in salmonCCQM-K43 KCRV_Mixture Model-median: 0.3884 ± 0.0056 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]


amended values

mean ± 2 STDEV

27

Annex 2 – Methylmercury (CH3Hg) Contents CCQM-P39.1: As, Hg, Pb, Se and Methylmercury in Salmon METHYLMERCURY MEASUREMENTS 29

DISCUSSION 30

QUESTIONNAIRE DATA 32-35

REFERENCES 36

29


Annex 2 – All participants results Methylmercury (CH3Hg)

Figure/ Table Page

number

Figure 1 All participants - MeHg 31

Figure 2 CCQM-P39.1 and CCQM-K43 - MeHg 31

Table 1 Measurement strategy 32

Table 2 Timing of the spike addition for ID or SA & ref. isotope used for IDMS 33

Table 3 Measurement process 34

Table 4 Experimental reproducibility from the different sample aliquots 35


30

Methylmercury measurementsAll participants who measured methylmercury in CCQM-P39.1 had previously participated in CCQM-P39 (tuna fish), except NRCCRM and BAM. A detailed evaluation of the CCQM-P39 results for the MeHg measurements was recently published in the open literature [1]. All participants measuring MeHg received the ERM AE670 202Hg enriched spike solution. Participants were not asked to use it, but were entirely free to implement the measurement method of their choice. The ERM AE670 material was offered for use by those applying species specific direct IDMS. In addition the final ERM AE670 certificate and the publication on the preparation and certification of the ERM AE670 [2], were also forwarded to all CCQM-P39.1 participants who measured methylmercury. ERM AE670 is distributed under the European Reference Materials programme [3]. 3 CCQM-P39.1 participants, including 1 NMI, relied on external calibration or the method of standard additions, whereas the other 4 implemented an IDMS approach and chose to use the ERM AE670. These participants used cold vapour atomic fluorescence spectrometry, atomic emission or electron capture for the detection step, with either gas chromatography or solid phase microextraction gas chromatography for species separation. JSI reported 2 results for MeHg measured with GC-ECD and GC-CV-AFS. In addition, MeHg was also done by distillation, to check on the extraction efficiency and similar results were obtained. All other CCQM-P39.1 participants implemented an IDMS approach with GC- or HPLC-ICP-MS. Participants measuring MeHg were asked to fill in a separate questionnaire. This Annex

summarises the questionnaire data and presents the results for methylmercury.

Discussion

As shown in Figure 1 , results for MeHg are within 30% relative to the CCQM-K43 KCRV taking into account all participants’ results. After the CCQM-P39.1 summary report was distributed to the participants for comments, University of Oviedo reported that they encountered problems with the procedure and their instrumentation, particularly the GC-ICP-MS. Furthermore due to the fatty matrix University of Oviedo changed from a HCl-NaCl extraction, as applied in the previous pilot study on tuna fish to an extraction with TMAH. The complete change of the analytical procedure compared to the one applied in CCQM-P39 on the tuna fish sample could explain the deviation of their result and the large uncertainty. JSI encountered some problems with the software they used to establish the result with uncertainty. They observed a slight difference when comparing the amount content of MeHg calculated directly from the chromatograms (peak heights/area and calibration) with the MeHg content as calculated by the software. This software is used to establish the final result with uncertainty when all input parameters are entered. The results for MeHg should not have been different from the one calculated directly from the chromatograms. JSI commented that they have had a small problem there, although the difference was not large but still significantly affected the result. JSI’s reported data in CCQM-P39.1 for MeHg are based on the software calculation.

CCQM-P39.1 final report – Annex 2

31

Figure 1 Results for MeHg from all CCQM-P39.1 participants

Figure 2 CCQM-P39.1 and CCQM-K43 – MeHg

BAM

JSI_AFS

Univ_Oviedo

Corvinus Univ

NRCCRM

JSI_ECD LCABIE_Univ_Pau

Trent Univ

0.19

0.24

0.29

0.34

0.39

0.44

0.49

0.54

c

10

-6 m

ol·k

g-1

-50

-40

-30

-20

-10

0

10

20

30

40

50

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-P39.1: MeHg in salmonCCQM-K43 KCRV_Mixture Model-median: 0.373 ± 0.023 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]


LCABIE_Univ_Pau

NISTJSI_ECD

LGCIRMM

JSI_AFS

BAM

NRC

NMIJTrent Univ

Corvinus Univ

NRCCRM

Univ_Oviedo

0.19

0.24

0.29

0.34

0.39

0.44

0.49

0.54

c

10

-6 m

ol·k

g-1

-50

-40

-30

-20

-10

0

10

20

30

40

50

Dev

iatio

n fr

om th

e K

43 K

CR

V in

%

CCQM-K43 & P39.1: MeHg in salmonMixture Model-median from CCQM-K43: 0.373 ± 0.023 ·10-6 mol·kg-1 ; [ µ ± σ*ts/√n]



32

Questionnaire Data Table 1 Measurement strategy

Participant Direct IDMS

with ERM-AE670

Double IDMS with I ERM-AE670

Other method

Calibration with

external matrix

matched standards

Without

an internal

standard

Method of standard addition

(SA) calibration

BAM


JSI

LCABIE – University of Pau

NRCCRM

Trent University

University of Oviedo

CCQM-P39.1 final report – Annex 2

33

Table 2 Timing of the spike addition for ID or SA & ref. isotope used for IDMS

REPORTED QUESTIONNAIRE DATA FOR METHYLMERCURY

Participant

Some hours

prior to MeHg

extraction

Immediately before MeHg

extraction

After MeHg extraction,

prior to measurement

Reference isotope used for

IDMS

Number of replicate

blend samples

BAM


JSI


202Hg 6

NRCCRM 200/202 5

Trent University 200

one blend per replicate

measurement; total of 4

blends (2 each for HgT and

MeHg)

University of Oviedo 200 4


34

Table 3 Measurement process


Participant Sample mass (g)

Extraction reagents

Extraction apparatus

Derivatisation reagents

Species

separation apparatus

Detector

BAM 0.6 acetic acid ultrasound bath 1h,

shaking, 1h NaB(C2H5)4 GC, DB 5 MS,

Helium AED


0.250 18 %

methanolic NaOH

ultrasound bath at 75 C NaBPh4

J&W, DB-1, 15m 0.53mm (megabore) 1.5um film thickness

(carrier : Ar)

AFS

JSI 0.05-0.4 KBr/H2SO4/CuSO4x5H2O/CH2Cl2

- NaB(C2H5)4 GC CV AFS


~ 0.4 TMAH 25%, 10 mL

microwave, open focused

NaBPr4, 1% solution,

GC, MXT-1, 0.53mm id,

30m length, 25 mL He carrier, inj T 220°C, Tinitial 60,

ramp 50, final T 220

ICP-MS

NRCCRM ~0.4g KOH/Methanol

Shaking table

CH2Cl2, Na2S2O3

HPLC, Zorbax-C8 ICP-MS

Trent University 0.4 HNO3 (4 M)

water bath, 50 degree

Celcius NaB(C2H5)

packed GC (15% OV-3 on acid-washed Chromosorb W, DMCS ),

50x0.4 cm, Ar carrier 40

ml/min, 100 degree Celcius

ICP/MS

University of Oviedo 0.25 TMAH 20%

in water Water bath

60 ºC NaB(C2H5)4 GC ICP-MS


35

Table 4 Experimental reproducibility from the different sample aliquots


Experimental reproducibility

BAM 3.5%

Corvinus University of Budapest 14.6 %

JSI -


-744-1: 3.05%, -744-2: 4.07%, -856-1: 4.82%, -856-2: 3.90%, -246-1: 4.17%, -246-2: 3.86%

NRCCRM 2.50%

Trent University Two replicates of the same blend did not deviate more than 1% from each other

University of Oviedo 12%


36

References 1 Methylmercury in tuna : demonstrating

measurement capabilities and evaluating comparability of results worldwide from the CCQM-P39 comparison,. C. R. Quétel et al., JAAS, 2005, 20, 1058-1066

2 Preparation and certification of ERM®-AE670, a 202Hg enriched methylmercury isotopic reference material. J. P. Snell et al., J. Anal. At. Spectrom., 2004, 19, 1315 – 1324

3 http://www.erm-crm.org

37

Annex 3 – Questionnaire data As, Hg, Pb, Se Contents


Tables 1 - 6 __________________________________________________________ 40-44 Questionnaire data________________________________________________________

39


Annex 3 – Questionnaire data

Table Page number

Table 1 Digestion method and acid mixture 40

Table 2 Reference isotope for IDMS 41

Table 3 Number of blends 41

Table 4 Experimental reproducibility 42

Table 5 (Isotopic) reference materials used for calibration 43

Table 6 Number of salmon vials used for the analysis and Use of square root of n for type A uncertainty contributions 44


40

Table 1 Digestion method and acid mixture

REPORTED QUESTIONNAIRE DATA

Participant As Hg Pb Se

APAT Microwave digestion H2O2, HNO3

- Microwave digestion H2O2, HNO3

Microwave digestion H2O2, HNO3

BAM - - - -

CENA - - - -


- - - -

IAEA Microwave digestion HNO3

- Microwave digestion HNO3

-

JSI Wet ashing of the irradiated sample with carrier HNO3, H2SO4, H2O2

wet digestion HNO3

-

Destruction of the irradiated sample with

added carrier using 3g Mg(NO3)2,

reduction with 6 M HCl, derivatization

with 4-nitro-1,2-diaminobenzene and

extraction of Se chelate with toluene


- - - -

LGC Ltd. wet digestion HNO3, H2SO4

- - -

NRCCRM microwave digestion HNO3

- - microwave digestion HNO3

SP - Microwave digestion H2O2, HNO3

- -

Trent University - Wet digestion

HNO3, H2SO4 - -

University of Oviedo - - - -


41

Table 2 Reference isotope for IDMS



APAT Not applicable - -

BAM Not applicable - Pb-208, Spike isotope was Pb-207 -

CENA/USP. Not applicable - - -


Not applicable - - -

IAEA Not applicable - - -

JSI Not applicable - - -

LCABIE – University of Pau Not applicable 202 - -

LGC Ltd. Not applicable - - -

NRCCRM Not applicable - - 78/80

SP Not applicable 196/202 mass

discrimination with thallium

- -

Trent University Not applicable 200 - -

University of Oviedo Not applicable - - -


42

Table 3 Number of blends



APAT Not applicable - - -

BAM Not applicable - - -

CENA/USP. Not applicable - - -


Not applicable - - -

IAEA Not applicable - - -

JSI Not applicable - - -

LCABIE – University of Pau Not applicable - - -

LGC Ltd. Not applicable - - -

NRCCRM Not applicable - - 5

SP Not applicable 8 - -

Trent University Not applicable one blend per

replicate measurement

- -

University of Oviedo Not applicable - - -


43

Table 4 Experimental reproducibility



APAT - - - -

BAM 1.8% - 1% 3.75%

CENA/USP. 0.62 % 5.1 % - 1.2 %


- - - -

IAEA - - - -

JSI - - - -

LCABIE – University of Pau - - - -

LGC Ltd. 2.1 % - - -

NRCCRM - - - 0.99%

SP - 0.64 % - -

Trent University -

Two replicates of the same blend did not

deviate more than 1% from each other

- -



44

Table 5 (Isotopic) reference materials used for calibration



APAT - - - -

BAM - - NBS 981 -

CENA/USP. - - - -


- - - -

IAEA - - - -

JSI Merck, Lot No. K25046202 - - Merck, Lot.No.

K25046555

LCABIE – University of Pau - - - -

LGC Ltd. - - - -

NRCCRM - - - Se Isotope Standard

(LIS78SE-50)

SP - ERM-AE640 and pure mercury - -

Trent University - DORM-3 - -



45

Table 6 Number of salmon vials used for the analysis and Use of square root of n for type A uncertainty

contributions


Participant Number of vials Sqrt (n)

APAT 4 Yes

BAM 2 Yes

CENA/USP. 2 Yes


2 No

IAEA 4 Yes

JSI 5 No

LCABIE – University of Pau 3 Yes

LGC Ltd. 1 Yes

NRCCRM 4 No

SP 2 No

Trent University 4 (2 vials for MeHg and 2 vials for HgT) Yes

University of Oviedo 2 Yes


46

47

Annex 4 – Documentation Contents


Documentation

Cover page including participants list__________________________________________

Accompanying letter_______________________________________________________

Scope of the study ________________________________________________________

General instructions _______________________________________________________

Instructions for determination of the dry-mas correction and the digestion of the Salmon _

Certification of MeHg in Salmon______________________________________________

Guidelines for result reporting _______________________________________________

Guidelines for questionnaire reporting _________________________________________

Questionnaire for measurements of Hg, Pb, Se and As ___________________________

Questionnaire for Methylmercury _____________________________________________

EUROPEAN COMMISSION DIRECTORATE GENERAL JRC JOINT RESEARCH CENTRE IRMM Insti tute for Reference Mater ials and Measurements

1st December 2004


Information Package

For further information please contact:

Dr. Y. Aregbe CCQM-K43 & CCQM-P39.1 Co-ordinator Isotope Measurement Unit tel.: +32 (0)14 571 673 fax: +32 (0)14 571 865 e-mail: [email protected]

Institute for Reference Materials and Measurements (IRMM) European Commission – JRC

Retieseweg 111, B-2440 GEEL (Belgium)

Distribution list

Registered CCQM_IAWG participants in CCQM-K43 and/or CCQM-P39.1: J. Vogl BAM E. Nadai-Fernandes CENA M. Gallorini IMGC C. Quétel IRMM P. Robouch IRMM E. Hwang KRISS R. Hearn LGC G. Labarraque LNE R. Greenberg NIST K. Inagaki NMIJ R. Sturgeon NRC J. Wang NRCCRM B. Magnusson SP L. Mackay NMIA T. Araujo INMETRO U. Sansone IAEA E. Wyse IAEA-MEL Registered invited participants to CCQM-P39.1: M. Horvat IJS E. Krupp LABEC Univ. Pau J.I. Garcia-Alonso Oviedo Univ. H. Hintelmann Trent Univ. Z. Jokai Corvinus Univ. D. Centioli APAT E. Bjorn Umea Univ. W. Hagan Frontier Geosciences Inc. CCQM and RMO delegates for info:

R. Kaarls President CCQM R. Wielgosz CCQM exec. Secretary M. Sargent CCQM IAWG chairperson J. McLaren CCQM KCWG chairperson P. Charlet EUROMET METCHEM

Retieseweg 111, B-2440 Geel, Belgium Tel.: +32-(0)14-571 673 • Fax: +32-(0)14-571 865 • [email protected] • http://www.irmm.jrc.be

EUROPEAN COMMISSION DIRECTORATE GENERAL JRC JOINT RESEARCH CENTRE Institute for Reference Materials and Measurements IRMM

Geel, 1st December 2004 L/IM/95/04 To: CCQM-P39.1 registered participants Subject: CCQM-P39.1 “As, Hg, Pb, Se and Methylmercury in Salmon” Dear CCQM-P39.1 participant, As proposed, discussed and decided in previous CCQM-IAWG meetings (Paris 2004), IRMM will act as the coordinating laboratory of the CCQM-P39.1 pilot study, “As, Hg, Pb, Se and Methylmercury in Salmon”. The salmon material originates from Norway. The sample is a freeze dried and ground salmon muscle and skin powder bottled in amber glass vials each one containing ~ 2.2 g of material. The material is tested for stability and homogeneity. It is appropriate for the needs of this comparison. The deadline for reporting the CCQM-P39.1 results is 31st March 2005. Subsequently the preliminary results can be presented and discussed at the CCQM-IAWG meeting in Spring, Paris. The information package for CCQM-P39.1 includes: • This letter • Scope of the CCQM-P39.1 pilot study • General instructions • Instructions for reporting the detailed measurement procedure including uncertainty evaluation • Instructions for the dry mass correction and digestion of the salmon • Information for CCQM-P39.1 participants measuring methylmercury • Instructions for online result and questionnaire reporting using the IT system of IRMM

Interlaboratory comparisons If any further information is required, please do not hesitate to contact us. Yours sincerely,

Dr. Y. Aregbe CCQM-P39.1 co-ordinator IRMM, IM Unit


EUROPEAN COMMISSION DIRECTORATE GENERAL JRC JOINT RESEARCH CENTRE IRMM Insti tute for Reference Mater ials and Measurements

Geel, 1st December 2004

CCQM-P39.1 “As, Hg, Pb, Se and Methylmercury in Salmon”

Scope of the pilot study

At the April meeting in Paris, 2004 it was decided to proceed the previous pilot study on tuna fish, CCQM-P39, to a key comparison and this pilot study CCQM-P39.1 using a salmon material. In order to protect public health it is essential to keep contaminants at levels, which are toxicologically acceptable, thus surveillance measures were taken regarding the presence of contaminants in foodstuff, including fish.

Arsenic is a mononuclidic toxic element and cannot be determined by IDMS. Rules for measurements of As are set in the commission decision on implementing council directive 96/23/EC concerning the performance of analytical methods and the interpretation of results. Including also total As measurements in CCQM-P39.1 can be seen as a contribution to broaden the scope and degree of difficulty of measurements addressed by the CCQM Inorganic Working Group. Mercury is a potential environmental toxicant. The main source of human intake of mercury contaminants originates from methylmercury in fish and fishery products. Methylmercury is particularly interesting due to its high toxicity compared to inorganic mercury and its high proportion among organomercury species in the environment. Mercury species, may induce alterations in the normal development of the brain of infants and may induce neurological changes in adults. To protect public health, maximum levels of mercury in fishery products are laid down in relevant regulations. Lead may induce reduced cognitive development in children and increased blood pressure and cardiovascular diseases in adults. The EC Directive 2001/22 describes the Community methods for the sampling, the sample preparation and the analysis of Hg and Pb in fish. Very recently another EC Regulation (466/2001) was introduced that endorses officially the threshold value of 0.5 mg Hg·Kg-1 and 0.2 mg Hg·Kg-1 in salmon [1]. Selenium is an essential trace element for human beings. Seafood is an important source of Se intake for people in some regions. Certain forms of cancer and cardiovascular diseases have also been associated with Se deficiency. On the other hand Se is counted among the most important elements in terms of food-chain contamination. Se has the narrowest plateau between concentrations that show deficiency and toxic effects, respectively. Recently the European Commission has requested the Scientific Committee on Food (SCF) to review the upper level of daily intake of individual vitamins and minerals, amongst them Se, and provide the basis for the establishment of safety factors [2]. The Hg concentration in the salmon material is lower than in the tuna fish material used in CCQM-P39. The first is a real life sample with a regular Hg content for non-contaminated fish, whereas the latter was a real life sample that has been taken off the market due to its elevated Hg content. Measurements of particularly the total amount content of Hg and MeHg in this salmon material would together with the results of the CCQM-P39 pilot study on tuna fish demonstrate the measurement capabilities of the participating institutes in this important field, not only for high Hg concentrations but also for lower levels of Hg in fish. This is important in view of the fact that the threshold value in the EC Regulation (466/2001) for Hg in salmon is lower than for Hg in tuna fish.


From a metrological point of view, the measurement of the amount content of mercury in salmon is representative for many similar measurements in fish and other food matrices. Laboratories who demonstrate their capability of measuring the As, Hg, Pb and Se amount content in the CCQM-P39.1 salmon samples, are likely to have the capability, knowledge and skills to measure the amount content of other elements at similar levels in other food matrices which require similar sample preparation.

Dr. Y. Aregbe CCQM-P39.1 co-ordinator IRMM, IM Unit 1 Commission Regulation (EC) No 466/2001 of 8 March 2001 setting maximum levels for certain

contaminants in foodstuff, Official Journal of the European Communities L077, 16/03/2001, 0001-0013

2 Guidelines of the Scientific Committee on Food for the development of tolerable upper intake levels for vitamins, minerals and trace elements SCF/CS/NUT/UPPLEV/11 Final, Nov/2000


EUROPEAN COMMISSION DIRECTORATE GENERAL JRC JOINT RESEARCH CENTRE IRMM Inst i tute for Reference Mater ials and Measurements


CCQM-P39.1

As, Hg, Pb, Se and Methylmercury in Salmon

General instructions and guidelines

♦ Each participant will receive 4 salmon samples. On arrival, it is recommended to store the glass vials containing the salmon material refrigerated at 4 oC.

♦ Please can you forward me confirmation the day of delivery, confirming the condition of the vials received to the following email address [email protected]

♦ The amount contents of Hg, Pb and MeHg in the salmon are approximately in the range of 0.1 - 5 µmol · Kg-1 and for As and Se approximately 5 - 50 µmol · Kg-1

♦ To account for the small unit size, each participant will receive 4 bottles (8-10g) of the salmon material where at least 2 sample aliquots from 2 bottles must be analysed for each measured analyte. In view of the limited number of salmon samples and the small unit size available, please plan carefully your measurements as far as the amount of sample needed is concerned.

♦ The participant is free to choose the analytical procedure for measurements in the salmon sample provided it is fit for purpose.

♦ Since the majority of the laboratories participating in CCQM-P39.1 will use IDMS, a few specific points for this analytical method are highlighted although most of the participants have considerable experience in isotope dilution measurements.

• minimise contamination (work in closed systems or class 100 clean bench, check

reagents and lab ware used) • prepare the blends and dilutions gravimetrically. Avoid weighing of too small

aliquots of solids or liquids in order to minimise the weighing uncertainty. • correct sample weighing for dry mass (see specific instructions) • spike the salmon prior to the digestion • make sure that the salmon digestion is complete (see specific instructions) • possible isobaric interferences for the As, Hg, Pb, and Se isotopes should be

investigated and treated accordingly • the correction factors for mass discrimination in the ratio measurements should be

measured repeatedly using materials of known isotopic composition and ratios similar to those in the blends or samples

• participants measuring MeHg, please consult the attached information letter “certification of MeHg in salmon”


Instructions for result reporting and uncertainty evaluation ♦ For the first time registration and result/questionnaire reporting is done online via the IT

system of IRMM interlaboratory comparisons. Please report your results and questionnaire replies according to the guidelines.

♦ The uncertainty statement should be evaluated and presented according to the principles outlined in, e.g. “ISO/GUM” [1] or the Eurachem/CITAC Guide [2]. This implies that you 1) state your measurement equation 2) identify all significant sources of uncertainty 3) state your input quantities 4) include factors related to sample treatment in your measurement equations 5) describe the applied evaluation process and type of assumed distribution for your uncertainty estimation 6) quantify uncertainty components and convert them to standard uncertainties 7) calculate the combined standard uncertainty uc 8) present an expanded uncertainty U with the coverage factor k=2

♦ As our IT system is still in developing production phase II it is not possible yet to attach documents to your individual result report sites. Therefore the description of the measurement procedure and the sample pre-treatment including the complete uncertainty statement needs to be sent as attachment by email to [email protected]

♦ The whole information package including guidelines for result reporting will also be sent to you via email.

If you require further information or assistance, do not hesitate to contact us. Please address directly Dr. Christophe Quétel ([email protected]) in case you need more specific information related to the MeHg measurements. Dr. Y. Aregbe CCQM-P39.1 co-ordinator IRMM, IM Unit 1. International Organisation for Standardisation, “Guide to the Expression of Uncertainty in Measurement”,

©ISO, ISBN 92-67-10188-9, Geneva, Switzerland, 1993.

2. Eurachem/CITAC Guide Quantifying uncertainty in analytical measurement (2nd ed. 2000), www.eurachem.bam.de.


EUROPEAN COMMISSION DIRECTORATE GENERAL JRC JOINT RESEARCH CENTRE IRMM Inst i tute for Reference Mater ials and Measurements


Instructions for determination of the dry-mass correction and the digestion

of the salmon sample in CCQM-P39.1 There are two potential problems in measurements performed on matrices like the salmon material used in this pilot study. Firstly, and most important, the water content of the powdered salmon will affect the salmon mass, and a correction for that is needed. From recent work in that area [1, 2] it seems that this correction is very important because depending on the method used a different result can be obtained. This means that the “measurand” is to some extent “operationally defined”. In order to achieve comparability of results the protocol as given below, must be followed by the CCQM-P39.1 participants. Secondly, the digestion of the salmon material (if needed for the procedure that you are following) is never a trivial exercise. This document intends to give guidance on the above matters. 1. Dry-mass correction The salmon will absorb ambient moisture at typical laboratory temperature and humidity conditions. Therefore the sample bottle should only be opened immediately before weighing aliquots for the IDMS blend preparation. For correction of the measured values to dry mass, water content measurements should be made on a separate portion of the same material with a mass of 0.6 ± 0.1g. The material should be dried before weighing for a minimum of 24 hours in a ventilated oven at 102 ± 2 oC. Cycles of drying and weighing should be repeated until a constant mass is attained. Each weighing has to be carried out after the sample has reached thermal equilibrium at room temperature in a dessicator, recommended for about 30 min. Successive weights should not differ more than 0.001 g. The loss of mass corresponds to the “dry-mass correction factor” that should be applied 2. Digestion of the salmon (only in case the method used requires digestion) There are a variety of digestion methods, employing various combinations of acids and for which different instruments are used. • use minimum sample mass of 0.4 g. • aim is to digest the salmon material completely. • should this fail, As, Hg, Pb and Se content in the residue should be measured in order to

estimate the correction needed and its uncertainty contribution. If you require any further assistance or information, please do not hesitate to contact us. Dr. Y. Aregbe CCQM-P39.1 co-ordinator IRMM, IM Unit 1. S. Rückold et. al. Water as a source of errors in reference materials,

Fresenius J Anal Chem (2001) 370: 189 - 193


2. S. Rückold et. al. Determination of the contents of water and moisture in milk powder, Fresenius J Anal

Chem (2000) 368: 522 - 527

Retieseweg 111, B-2440 Geel, Belgium Tel.: +32-(0)14-571 658 • Fax: +32-(0)14-571 673 • e-mail: [email protected] • http://www.irmm.jrc.be

EUROPEAN COMMISSION DIRECTORATE GENERAL JRC JOINT RESEARCH CENTRE Institute for Reference Materials and Measurements IRMM


CCQM-P39.1 Certification of MeHg in Salmon

Dear colleague,

You have registered for the CCQM-P39.1 interlaboratory comparison on the determination of the MeHg amount content in salmon. In total, there will be 16 participants including 9 from CCQM and 7 invited experts measuring MeHg. Since you have participated in the previous pilot study on tuna fish you have already received the IRMM-670 202Hg enriched MeHg isotopic CRM. The use of the IRMM-670 solution is intended to help you applying the isotope dilution mass spectrometry (IDMS) method for the determination of the MeHg content in the salmon sample. However, the use of IRMM-670 is only recommended and is by no means mandatory. It is expected that the participants will provide experimental results according to one of the 5 approaches described below. 1- Apply a direct IDMS (equation 1) approach using the 202Hg enriched MeHg IRMM-670 as a spike solution. 2- Apply a double IDMS (equation 2) approach using the 202Hg enriched MeHg IRMM-670 as a spike solution

(and thus chose to re-characterise it). 3- Apply another method than the two above. Equation 1.

Equation 2.

Considering that the characterisation of the MeHg content in the IDMS spike material is a crucial step, we invite the ‘IDMS participants’ to apply scenario 1, to make maximum use of the of IRMM-670 solution. In this way the number of potential sources of measurement biases among participants’ results can be minimised. At the same time, applying scenario 1 facilitates the final interpretation of possibly occurring significant differences between the participants’ results. Again, it is emphasised that the participants should remain entirely free to apply the scenario they like best and/or judge most appropriate.

Yours sincerely,

Dr. Christophe Quétel Isotope Measurements Unit EC-JRC-IRMM

∑∑⋅

−⋅

⋅−⋅⋅=

Yi

Xi

XBR

BRY

X

YYX R

RRRKRKR

mmcc

B

B

)()(

YB

BZ

XB

BY

Y

Y

X

ZZX RR

RRRRRR

mm

mmCC

−−

⋅−−

⋅⋅⋅= '

'

'

Retieseweg 111, B-2440 Geel, Belgium Page 1 Tel.: +32-(0)14-571 673 • Fax: +32-(0)14-571 865 • [email protected] • http://www.irmm.jrc.be• http://www.imep.ws

GUIDELINES to Participants on Reporting Results for CCQM-P39.1.

Our result reporting pages will only be open from the end of January 2005, we will notify you by e-mail when these pages are available. These guidelines will explain how you can report your measurement result and uncertainty. Further guidelines on how to complete the questionnaire information will be attached in the above mentioned e-mail. Participant’s name «SURNAME» Your password key: «PARTKEY» Sample 1 code no.: «SAMPLECODE»«Next Record» Sample 2 code no: «SAMPLECODE»«Next Record» Sample 3 code no: «SAMPLECODE»«Next Record» Sample 4 code no: «SAMPLECODE»«Next Record» Amount content code no: «SAMPLECODE» The result reporting is done on the Internet, the login page is located using the following URL http://www.irmm.jrc.be/imepapp/jsp/loginResult.jsp The following information page will appear. To obtain the login page, close down this screen.


This is the login page.

Please use your allocated password key

EXAMPLE:- Password Key – AYII1892393 Your Password Key – «PARTKEY» Once you have entered your password key, press the SUBMIT button (Please note that your password key is unique to the comparison you have registered to.) The RESULT REPORT FORM

Completing the RESULT REPORTING Page. 1. Please make sure that the correct result is entered for the correct sample code and for the correct analyte. In case you don’t measure all the samples or all the analytes, please leave the corresponding fields blank. Please be aware that for each analyte at least 2 replicates from 2 bottles have to be measured. 2. Indicated on the top right hand side of the screen, shows how many pages there are. Page 1 to 5. Per page displays the same sample code and the list of analytes.


3. For each analyte by sample code there are 2 replicates measurements: MEASUREMENT #1 and MEASUREMENT #2 Please ONLY enter information in the result field. (Leave the uncertainty and the coverage factor fields blank) Detailed information on the replicates measured needs to be included in the description of the measurement procedure that has to be sent as an attachment by email. (See also “instructions for result reporting and uncertainty evaluation”) 4. The measurement unit field has already been set. 5. In the field marked “Result value” enter your measurement result using the 2nd box to your left. 6. Select the field marked “Technique used” this will activate the drop down menu. Select the technique used. If the technique used is not listed, select the “OTHER” field and then specify When the “Technique used” field has been selected the “OTHER” field is disabled and no input can be entered. Likewise should you select the “OTHER” field then the “Technique used” field is disabled. 7. When completing the AMOUNT CONTENT on page five, in the field with your code no. «SAMPLECODE», you are required to complete ALL fields. RESULT, UNCERTAINTY and COVERAGE FACTOR. This is the overall result per analyte. Below is an example of a completed result screen for one sample and one analyte.

Once you have completed the first screen to proceed to the following screen, FIRST press the SAVE button followed by the NEXT.


SAVE – The SAVE button stores the entered data with the possibly to edit them as often as you need. To reconnect to our system use the same URL link http://www.irmm.jrc.be/imepapp/jsp/loginResult.jsp and re-supply your password key. The result form will appear with the data that has already been entered. Make the required changes and select either the SAVE button. This pop up page will appear advising you that your results have been saved.

The SUBMIT button is located at the button of each page, ONLY press this button when you have entered all you data. Remember to submit your results before the deadline date, as ONLY submitted results will be accepted. How to SUMBIT your results Once the SUBMIT button has been pressed, the questionnaire will appear ready for your input. Guidelines on how to complete the questionnaire information will follow shortly by e-mail


GUIDELINES to Participants on completing the Questionnaire for CCQM-K43 and CCQM-P39.1.

Once you have entered your results and pressed the SUBMIT button, the questionnaire information will appear ready for your input. The QUESTIONNAIRE FORM Completing the QUESTIONNAIRE Page. 1. You must enter or select data to every question, otherwise your questionnaire information will not be submitted. Should you not complete a question, or complete a question incorrectly, a message will appear directing you to that relevant question. IF YOU HAVE LEFT A QUESTION BLANK, WHERE AN ANSWER IS REQUESTED, YOU WILL RECEIVE AN ERROR MESSAGE, IF YOU HAVE NO COMMENTS RELATING TO THAT QUESTION, PLEASE CAN WE REQUEST THAT YOU INPUT ‘NOT APPLICABLE’. (PLEASE USE THE FOLLOWING FORMAT N/A) 2. Text fields are a maximum of 150 characters. 3. QUESTION 1 to 8 should be answered by all participants. QUESTION 9 to 17 should ONLY be answered if a measurement has been done for METHYLMERCURY, if no measurement has been done, please select for EACH question the option – ‘CH3Hg non Applicable measurement’ 4. Instructions for question no.10 PART A If you have selected ‘CH3Hg Applicable measurement for this question, you are now requested to select one of the three approaches listed. If selecting approach 3 (Approach 3 – another method than species specific IDMS) then you will need to select one of the three sub-questions in PART B.

ONLY WHEN SELECTING APPROACH 3 and OTHER, in PART B will question no. 11 need to be answered. Otherwise select ‘CH3Hg non Applicable measurement’

Part A

Part B


5. Instructions for question no.11 When PART A = approach 3 and PART B = OTHER, have been chosen. Please select ‘CH3Hg Applicable measurement’ and specify the used calibration strategy.

6. Instructions for question no.12 ONLY WHEN SELECTING APPROACH 3 and CALIBRATION WITH EXTERNAL MATRIX MATCHED STANDARDS, in PART B of question no.10, requires you to answer question no.12. Otherwise select ‘CH3Hg non Applicable measurement’

7. Instructions for question no.13 When PART A = approach 2 has been chosen in question no.10. Please select ‘CH3Hg Applicable measurement’ and specify your results for the determination of the MeHg content. Otherwise select ‘CH3Hg non Applicable measurement’ 8.Instructions for question no.15 When in PART A = approach 1 OR 2 has been chosen in question no.10. Please select ‘CH3Hg Applicable measurement’ and answer the next two questions with your comments. Otherwise select ‘CH3Hg non Applicable measurement’ Once you have completed the questionnaire, press the SUBMIT QUESTIONNAIRE button.


If you answered a question in the wrong format, or left a question blank, you will receive an error message, the system will direct you to that relevant question by adding a message in red text. Below is an example of an error message screen.

Before re-submitting your data, please make sure that the following has been applied:-

a) Ensure all questions have been completed. b) Ensure that the comment field has only been completed when asked.

When the requested changes have been made, press the SUMBIT QUESTIONNAIRE button again. If entries are acceptable by the system, the following confirmation screen will appear showing all the data entered.


Should any amendments need to be made, press the CHANGE RESULTS AND QUESTIONNAIRE button, this will return you to the previous screen. Make the required changes and submit your data again. Once again, the confirmation screen will appear, check your data again. When all data is correct, press the CONFIRM RESULTS AND QUESTIONNAIRE button. (Located at the bottom of the screen)

It is IMPORTANT that you print off your result report form, from the available print option. Please sign and fax this document to IRMM on Fax No. +32 (0)14 571 865


The final screen will conclude that your data has been accepted by IRMM, this message will appear at the top of the screen.

IMPORTANT : Disclaimer, Confidentiality Notice and rules on Privacy Protection IRMM Interlaboratory Comparison

> Login > Results > Questionnaire

Functions Results

Questionnaire for CCQM-P39.1

Ms. C Parker IRMM BELGIUM

1. Salmon moisture content

What was the moisture content by % of the weighed sample?

What was the uncertainty of the moisture content by % of the weighed sample?

2. Applied correction factor for dry-mass

What was the applied correction factor for dry-mass?

What was the uncertainty of the applied correction factor for dry-mass?

3. Which reference isotope was used (for IDMS)?

4. What was the number of blends (for IDMS)?

5. What was the number of vials from which the salmon was analysed?

6. Was the square root of n used for type A uncertainty contributions? Yes No 7. What was the experimental reproducibility of Cx from different sample aliquots in % (blends

standard deviation and NOT the standard deviation of the mean)?

8. Which (isotopic) reference materials were used for calibration purposes?

9. Moisture content for Methylmercury measurements CH3Hg Applicable measurement CH3Hg non Applicable measurement

What was the moisture content by % of the weighed sample?

What was the uncertainty of the moisture content by % of the weighed sample?

10.Measurement strategy? CH3Hg Applicable measurement CH3Hg non Applicable measurement

Page 1 of 3IRMM Interlaboratory Comparison

16/12/2005http://www.irmm.jrc.be/imepapp/result.action

Approach 1:direct IDMS with 202Hg enriched

MeHg ERM-AE670 as a spike solution

Approach 2:double IDMS with 202Hg enriched MeHg ERM-AE670 as a spike

solution (re-characterization of ERM-AE670)

If CH3Hg applicable measurement which measurement strategy was used?

Yes

If Approach 3, Calibration with external matrix matched standards

If Approach 3, Method of standard addition (SA) calibration

If Approach 3, Other

11.Measurement strategy continued CH3Hg Applicable measurement CH3Hg non Applicable measurement

If Approach 3 and Other is applicable, please specify the used calibration strategy

12.Internal standard? CH3Hg Applicable measurement CH3Hg non Applicable measurement

If Approach 3 and Calibration with external matrix standard was applied with an internal standard,internal standard used

13.MeHg content? CH3Hg Applicable measurement CH3Hg non Applicable measurement

In case Approach 2, was applied, please report your results for the determination of the MeHg conusing your own values for the ERM-AE 670 solution and using the values provided by IRMM for thsolution MeHg amount content (mol(CH3Hg).kg-1) obtained using own values for the ERM-AE670 solution

Uncertainty (k=2) (mol(CH3Hg).kg-1) obtained using own values for the ERM-AE670 solution

MeHg amount content (mol(CH3Hg).kg-1) obtained using the IRMM values for the ERM-AE670 so

Uncertainty (k=2) (mol(CH3Hg).kg-1) obtained using the IRMM values for the ERM-AE670 solutio

14.Measurement strategy based on SA or ID CH3Hg Applicable measurement CH3Hg non Applicable measurement

Some hours prior

to MeHg extraction

Immediately prior to MeHg

extraction

Aftprio

If a measurement strategy based on SA or ID was used which of the following options most closely fits the timing of the spike addition in your method?

15.Measurement strategy involving species specific IDMS Approaches 1 and 2 CH3Hg Applicable measurement CH3Hg non Applicable measurement

If the measurement strategy involved species specific IDMS Approaches 1 and 2, (as described inplease answer the following questions Reference isotope used during the measurements

Number of replicate blend samples prepared



16.Measurement process CH3Hg Applicable measurement CH3Hg non Applicable measurement

Please describe briefly your measurement process by listing the types of reagents and apparatus extraction and derivatisation of MeHg in the sample Sample mass (g)

Extraction reagents (e.g. HCl, TMAH etc.)

Extraction apparatus (e.g. microwave, ultrasound bath etc.)

Derivatisation reagents (e.g. NaB(C2H5)4, Grignard etc.)

Species separation apparatus (e.g. GC, LC, column type, carrier)

Detector (e.g. CVAAS, ICP-MS etc.)

17.Experimental reproducibility CH3Hg Applicable measurement CH3Hg non Applicable measurement

What was the reproducibility of Cx from the different sample aliquots measured in % (blends in castandard deviation and NOT the standard deviation of the mean)

Clear

Submit questionnaire

This site is managed by the Informatics Unit, IRMM. Last update : 16/12/2005



CCQM-P39.1 final report - BIPM · CCQM-P39.1 Final Report 1 Abstract CCQM-P39.1 was organised as a...

Documents

Transcript of CCQM-P39.1 final report - BIPM · CCQM-P39.1 Final Report 1 Abstract CCQM-P39.1 was organised as a...