Título/s: | CCQM key comparison – organic solutions : CCQM-K47 volatile organic compounds in methanol. Final Report |
Fuente: | Metrología, 50 |
Autor/es: | Pérez Urquiza, Melina; Maldonado Torres, Mauricio; Mitani, Yoshito; Schantz, Michele M.; Duewer, David L.; May, Wille E.; Parris, Reenie M.; Wise, Stephen A.; Kaminski, Katja; Philipp, Rosemarie; Win, Tin; Rosso, Adriana; Kim, Dal Ho; Ishikawa, Keiichiro; Krylov, A. I.; Kustikov, Y. A.; Baldan, Annarita |
Editor: | IOP Publishing |
Palabras clave: | Metrología; Alcohol metílico; Compuestos orgánicos; Calibración; Compuestos volátiles; Interferencia; Soluciones |
Idioma: | eng |
Fecha: | 2013 |
Ver+/- 1 / 24
CCQM Key Comparison – Organic Solutions CCQM-K47 Volatile Organic Compounds in Methanol Final Report April 8, 2009 Melina Pérez Urquiza, Mauricio Maldonado Torres and Yoshito Mitani Centro Nacional de Metrologia (CENAM) Querétaro, Mexico Michele M. Schantz, David L. Duewer, Wille E. May, Reenie M. Parris and Stephen A. Wise National Institute of Standards and Technology (NIST) Gaithersburg, MD USA Katja Kaminski, Rosemarie Philipp and Tin Win Federal Institute for Materials Research and Testing (BAM) Berlin, Germany Adriana Rosso Instituto Nacional de Tecnología Industrial (INTI) Buenos Aires, Argentina Dal Ho Kim Korea Research Institute of Standards and Science (KRISS) Taejon, Korea Keiichiro Ishikawa National Metrology Institute of Japan (NMIJ) Ibaraki, Japan A.I. Krylov and Y.A. Kustikov D.I. Mendeleyev Institute for Metrology (VNIIM) St. Petersburg, Russia Annarita Baldan Van Swinden Laboratorium (VSL) Delft, the Netherlands 2 / 24
Introduction At the October 2005 CCQM Organic Analysis Working Group Meeting (IRMM, Belgium), the decision was made to proceed with a Key Comparison study (CCQM-K47) and a concurrent subsequent pilot study (CCQM-P61.1) addressing the calibration function for the determination of volatile organic compounds (VOCs) used for water quality monitoring. Both studies were coordinated by CENAM and NIST. Benzene, o-xylene, m-xylene, and p-xylene were chosen as representative VOCs. The solvent of choice was methanol. This report summarizes the results for CCQM-K47. Summary of Pilot Study CCQM-P61 At the October 2004 CCQM Organic Analysis Working Group Meeting (NRCCRM, China), the decision was made to proceed with a pilot study addressing the calibration function for the determination of volatile organic compounds (VOCs) used for water quality monitoring. Benzene, toluene, ethylbenzene, and three xylene isomers were chosen as representative VOCs. The solvent of choice was methanol. The pilot study CCQM-P61 Volatile Organic Compounds in Methanol was coordinated by CENAM and NIST. Each CCQM-P61 participant received five ampoules of the a gravimetrically prepared solution. Each 2 mL ampoule contained approximately 1.2 mL of a methanol solution including benzene, toluene, ethylbenzene, o-xylene, m-xylene, and p-xylene at concentrations between 13.37 µg/g (o-xylene) to 49.55 µg/g (toluene). The instructions requested the analysis of duplicate subsamples from each of four ampoules using the laboratory’s analytical procedure for determination of the mass fraction concentrations of the target analytes. Nineteen laboratories received samples for CCQM-P61; eighteen laboratories reported measurement results to the coordinators. Two laboratories used IDMS, six laboratories used internal standards, six laboratories used external standards, and the other four laboratories did not report the method of quantification used. Most of the laboratories used gas chromatography with mass spectrometry (GC/MS) and/or GC with flame ionization detection (GC-FID). The majority of reported results agreed with the gravimetric preparation values to within ± 5 %, with a minority of values ranging up to ± 20 %. Design, Conduct, and Results from CCQM-K47 Study Material The material used in CCQM-K47 was similar to that used in the CCQM-P61 pilot study: a gravimetrically prepared methanol solution of the four target VOCs (benzene and the three xylene isomers) plus toluene and ethylbenzene. The solution was prepared from neat materials procured from commercial sources. The purities of the starting materials were determined by GC-FID. Table 1 lists the gravimetric preparation impurity-adjusted mass fraction of the target VOCs in the study solution. 3 / 24
Table 1. Gravimetric Preparation Mass Fraction and Homogeneity Evaluation of Target VOCs Mass Fraction, μg/g ANOVA Analyte Value U95(Value) Fc Ft benzene 39.40 0.39 1.83 2.39 o-xylene 15.77 0.28 1.39 2.39 m-xylene 20.87 0.20 1.16 2.39 p-xylene 28.25 0.28 1.04 2.39 Homogeneity Evaluation Table 1 also summarizes the results of an Analysis of Variance (ANOVA) homogeneity evaluation for the target VOCs. Ten ampoules (A043, A073, A173, A203, A373, A381, A486, A502, A579, A581) were measured in triplicate using GC-FID. The null hypothesis was that the measurement means in all of the ampoules are identical; the alternate hypothesis was that the mean in at least one of the ampoules differed from the others. For all four of the target VOCs, the null hypothesis of homogeneity is accepted. The relative standard deviation of the measurements ranged from 0.2 % to 0.5 %. Study Design Each participant in CCQM-K47 received five ampoules of the study solution. Each 2-mL ampoule contained approximately 1.2 mL of solution. Participants were informed that the mass fractions of all target VOCs ranged between 10 µg/g and 50 µg/g. Participants were requested to analyze duplicate aliquots from each of four ampoules using an analytical procedure validated in CCQM-P61. Participants The eight laboratories listed in Table 2 received samples for CCQM-K47. All eight laboratories reported measurement results to the coordinators. Table 2. CCQM-K47 Participants NMI Submitted by Country Email Contact BAM Tin Win, Rosemarie Philipp,Katja Kaminski Germany tin.win@bam.de CENAM Mauricio Maldonado Torres, Evangelina Camacho Frías, Melina Pérez Urquiza México meperez@cenam.mx INTI Adriana Rosso Argentina mariare@inti.gov.ar, cpuglisi@inti.gov.ar KRISS Dal Ho Kim Korea byungjoo@kriss.re.kr NIST Michele Schantz USA michele.schantz@nist.gov NMIJ K. Ishikawa Japan ishikawa-keiichiro@aist.go.jp VNIIM A.I. Krylov, Y.A. Kustikov Russia lkonop@b10.vniim.ru VSL Annarita Baldan Netherlands abaldan@nmi.nl 4 / 24
Analytical Methods Table 3 summarizes the analytical methods used by the eight participants. Table 4 summarizes the amount of sample used in each analysis. Tables 5 and 6 summarize the materials used to prepare calibration solutions. Table 7 summarizes how the calibration solutions were used. Table 3. Analytical Methods Column NMI Method Phase L, m ID, mm FT, µm BAM GC/MS: benzene, o-xylene, m&p-xylene (coelution) GC-FID: m/p-xylene ratio DB-624 DB-WAX 30 60 0.32 0.32 1.8 0.5 CENAM GC/MS DB-WAX 60 0.25 0.25 INTI GC-FID PEG (Supelcowax) 60 0.32 0.5 KRISS GC-FID PEG (5% phenyl) methyl polysiloxane 30 0.53 1 2.65 NIST GC-FID: xylenes GC/MS: benzene DB-WAX DB-VRX 15 60 0.45 0.25 0.85 1.4 NMIJ GC/MS DB-WAX 60 0.32 0.5 VNIIM GC/MS HP 5 MS 30 0.25 0.25 VSL GC-FID CPWAX 52 60 0.53 2 Table 4. Sample Mass of CCQM-K47 Solution used for Analysis NMI Mass solution used, g BAM 0.05 to 0.09 CENAM 0.3 INTI 0.5 KRISS 0.2 NIST 0.4 NMIJ 0.4 VNIIM 0.5 VSL 0.7 Table 5. Certified Reference Materials Used as Calibrants Certified Mass Fraction, μg/g NMI Solvent Source benzene o-xylene m-xylene p-xylene BAM Methanol KRISS 40.6 ±0.6 102.0±1.6 100.2±0.6 100.2±0.6 5 / 24
Table 6. Neat Materials Used to Prepare Calibrants Mass Fraction, mg/g NMI Source benzene o-xylene m-xylene p-xylene CENAM Aldrich 0.9988 0.9915 0.9968 0.9967 INTI Merck >0.99 INTI AnalityCals Carlo Erba 0.988 0.978 >0.99 KRISS Sigma-Aldrich 0.9995 0.9916 0.9977 0.9986 NIST Sigma Aldrich 1.00 0.9363 1.00 0.9924 NMIJ NIM CRM 4002-a * 0.99996 NMIJ NIM CRM 4011-a * 0.9994 NMIJ NIM CRM 4012-a * 0.9981 NMIJ NIM CRM 4013-a * 0.9988 VNIIM SOP 0003-03 STC 0.9997 VNIIM SOP 0020-03 STC 0.997 VNIIM SOP 0015-03 STC 0.9966 VNIIM SOP 0016-03 STC 0.999 VSL Aldrich 0.999 VSL Fluka 0.995 0.995 0.995 * Deuterated Certified Reference material Table 7. Calibration Method used, Number of Calibration Levels, and Calibration Range benzene o-xylene m,p-xylene NMI Use Internal standard was added # mg/g # mg/g # mg/g BAM Internal at 2nd of 2 dilutions 5 0.3 - 4.8 5 0.7 - 12.0 5 0.7 - 12.0 CENAM Internal before injection 5 1 - 60 5 1 - 60 5 1 - 60 INTI External 4 3.5 - 53 3 3.5 - 53 3 3.5 - 53 KRISS External 1 1 1 NIST Internal during sample preparation 1 1 1 NMIJ IDMS to sample (deuterated VOC) VNIIM Internal before subsample analysis 1 1 1 VSL Internal to autosampler vials 6 4 - 60 6 1 - 35 6 1 - 35 Results The values, the combined uncertainties on the values, u, and the 95 % expanded uncertainties, U95, as submitted by the participants are summarized in Table 8 and displayed in Figure 1. Each panel of Figure 1 displays the results, their consensus summary, and the gravimetric preparation values for one of the target VOCs. Each “dot” denotes a value and each vertical “bar” denotes the 95% confidence interval on the value. The solid black horizontal line represents the consensus median, the red horizontal lines represent the robust 95 % confidence interval about the median, and the dotted black line goes through the gravimetric value. The black curve to the right of each panel is the additive probability density function for the reported results; the light blue curve is the Gaussian defined by the consensus summary values. In all four panels, the mass-fraction axis spans a 30 % interval about the consensus value. 6 / 24
Table 8. Submitted Values, µg/g benzene o-xylene m-xylene p-xylene NMI Value u U95 Value u U95 Value u U95 Value u U95 BAM 40.38 0.31 0.62 16.07 0.14 0.28 19.56 0.43 0.86 27.95 0.61 1.22 CENAM 38.26 0.27 0.53 15.93 0.05 0.10 19.58 0.08 0.16 27.42 0.48 0.97 INTI 39.17 0.49 1.21 16.42 0.24 0.68 18.95 0.16 0.46 27.32 0.16 0.38 KRISS 38.58 0.31 0.62 16.33 0.12 0.25 18.83 0.13 0.27 26.71 0.20 0.40 NIST 39.38 0.43 0.85 15.60 0.17 0.34 20.19 0.29 0.61 28.26 0.35 0.70 NMIJ 39.26 0.14 0.27 16.42 0.06 0.12 19.92 0.07 0.14 28.12 0.10 0.20 VNIIM 32.64 0.76 1.52 19.04 0.42 0.84 Reported as sum of m- & p-xylene VSL 39.83 0.69 1.37 16.77 0.32 0.64 20.11 0.27 0.54 28.55 0.36 0.72 Consensus* 39.22 0.33 0.76 16.38 0.18 0.42 19.58 0.30 0.70 27.95 0.30 0.70 * Robust consensus values calculated as: Value = median, u = median absolute deviation from the median (MADe), and U95 = u * t0.025,n-1 where t0.025,n-1 is the two-sided Student’s t for 95 % confidence and n measurements. Figure 1. Dot-and-Bar Plots for the Four Target VOCs <G rav Pr ep > BA M VS L NI ST NM IJ IN TI KR ISS CE NA M VN IIM 32 34 36 38 40 42 Be nz en e, µg /g <G rav Pr ep > VN IIMVS L NM IJ IN TI KR ISSBA M CE NA M NI ST 16 17 18 19 o-X yle ne , µ g/g <G rav Pr ep > NI STVS L NM IJ CE NA M BA M IN TI KR ISS 17 18 19 20 21 22 m- Xy len e, µg /g <G rav Pr ep > VS L NI ST NM IJ BA M CE NA M IN TI KR ISS 24 26 28 30 32 p-X yle ne , µ g/g Additional information on the methods and uncertainty calculations used by the participants is included in Appendix A. Representative chromatograms are provided in Appendix B. 7 / 24
Key Comparison Reference Values As displayed in Figure 1, the consensus summary values agree well with the gravimetric preparation values for benzene and p-xylene but not for o-xylene nor m-xylene. Review of the solution preparation records failed to identify any experimental or data analysis oversight. Inspection of the chromatograms in Appendix B suggests that discrepancy for m-xylene is not caused by insufficient chromatographic separation of m- and p-xylene. At the April 2006 CCQM Meeting in Paris, VSL was asked to analyze the CCQM-47 solution with thermal desorption GC-FID. Dynamic dilution was used to prepare six calibration levels of gas mixtures following the procedure of ISO 6145-8. These reference gas mixtures were loaded onto tenax TA sorbent tubes. The CCQM-K47 solution was loaded onto tenax TA sorbent tubes using a calibrated 10 μL syringe, three tubes each for five ampoules. Table 9 summarizes the results of this evaluation; Figure 2 displays the results relative to the measurement and gravimetric preparation values Table 9. Results from VSL Thermal Desorption Analysis, μg/g benzene o-xylene m-xylene p-xylene toluene Value U95 Value U95 Value U95 Value U95 Value U95 39.75 1.03 17.06 0.42 20.47 0.53 28.71 0.72 20.80 0.53 Figure 2. Dot-and-Bar Plots for the Four target VOCs with VSL Thermal Desorption Results <T hrm Ds rp> <G rav Pre p> BA M VS L NI ST NM IJ INT I KR ISS CE NA M VN IIM 32 34 36 38 40 42 Be nz en e, µg /g <Th rm Ds rp> <G rav Pre p> VN IIMVS L NM IJ INT I KR ISSBA M CE NA M NI ST 16 17 18 19 o-X yle ne , µ g/g <T hrm Ds rp> <G rav Pr ep > NIS T VS L NM IJ CE NA M BA M INT I KR ISS 17 18 19 20 21 22 m- Xy len e, µg /g <Th rm Ds rp> <G rav Pr ep > VS L NIS T NM IJ BA M CE NA M INT I KR ISS 24 26 28 30 32 p-X yle ne , µ g/g 8 / 24
The thermal desorption results agree well with the gravimetric preparation values for benzene, m-xylene, and p-xylene. The result for o-xylene does not agree well with either the gravimetric or consensus value. At the April 2008 CCQM Meeting in Paris, consideration was given to defining the Key Comparison Reference Value (KCRV) from the consensus, gravimetric preparation, thermal desorption, or some combination of the three values. Consideration was also given to voiding the study completely and to dropping o-xylene as a target measurand. On the basis of the agreement between the gravimetric preparation and thermal desorption values for m-xylene and the absence of any evidence of an analytical blunder that would be specific to o-xylene, the decision was made to accept the complete study as valid and use the gravimetric preparation values as the KCRV for all for four target VOCs. Key Comparison Reference Value Uncertainties Recognizing that the uncertainty estimated from the gravimetric preparation process may not fully represent the variability of the VOC levels in the solution levels as delivered to the participants, the combined uncertainty for the KCRV is estimated as: ns suu MADeConsensus ConsensusGravPrepKCRV 22 . where MADe is a robust estimate for the standard deviation of the n reported values. The 95 % level of confidence expanded uncertainty is estimated with the standard metrological factor, k=2: KCRV2KCRV95 uU . Table 10 lists the KCRV, u(KCRV), and U95(KCRV) values for the four target VOCs. Table 10. Key Comparison Reference Values and Their Uncertainties, μg/g VOC KCRV u(KCRV) U95(KCRV) benzene 39.4 0.4 0.8 o-xylene 15.8 0.2 0.5 m-xylene 20.9 0.3 0.6 p-xylene 28.3 0.3 0.7 Degrees of Equivalence Since participant results are not used in the estimation of the KCRV, the degree of equivalence for a given participant value for a given VOC, d ±U95(d), is estimated as: )();(2;KCRVValue)( KCRV;Value 9522 dudRddudUuudud where Value and u(Value) are the participant’s reported value and combined uncertainty. Table 11 lists the degrees of equivalence for the four target VOCs. 9 / 24
Table 11. Degrees of Equivalence benzene o-xylene m-xylene p-xylene NMI d U95 Rd d U95 Rd d U95 Rd d U95 Rd BAM 1.0 1.0 2.0 0.3 0.5 1.1 -1.3 1.1 -2.5 -0.3 1.4 -0.4 CENAM -1.1 0.9 -2.5 0.2 0.5 0.7 -1.3 0.6 -4.0 -0.8 1.2 -1.4 INTI -0.2 1.4 -0.3 0.7 0.8 1.6 -1.9 0.8 -4.9 -0.9 0.8 -2.5 KRISS -0.8 1.0 -1.7 0.6 0.5 2.1 -2.0 0.7 -6.0 -1.5 0.8 -4.0 NIST 0.0 1.1 0.0 -0.2 0.6 -0.6 -0.7 0.9 -1.6 0.0 1.0 0.0 NMIJ -0.1 0.8 -0.3 0.7 0.5 2.7 -0.9 0.6 -3.0 -0.1 0.7 -0.4 VNIIM -6.8 1.7 -8.0 3.3 1.0 6.8 VSL 0.4 1.6 0.5 1.0 0.8 2.5 -0.8 0.8 -1.8 0.3 1.0 0.6 Values of Rd less than -2 or greater than +2 indicate results that deviate from the KCRV by more than is likely given the estimated uncertainties. Figure 3 provides an approximate graphical display of the degrees of equivalence. Note that in the panels of this Figure the participant results are listed in alphabetical order, the solid black line represents the KCRV, and the solid red lines bound the symmetric interval KCRV ±U95(KCRV). Figure 3. Approximate Degrees of Equivalence <T hrm Ds rp> <G rav Pre p>VS L VN IIMNM IJ NI ST KR ISSIN TI CE NA M BA M 32 34 36 38 40 42 Be nz en e, µg /g <Th rm Ds rp> <G rav Pre p>VS L VN IIMNM IJ NI ST KR ISSIN TI CE NA M BA M 16 17 18 19 o-X yle ne , µ g/g <T hrm Ds rp> <G rav Pre p>VS L <V NI IM > NM IJ NI ST KR ISSIN TI CE NA M BA M 17 18 19 20 21 22 m- Xy len e, µg /g <Th rm Ds rp> <G rav Pre p>VS L <V NI IM > NM IJ NI ST KR ISSIN TI CE NA M BA M 24 26 28 30 32 p-X yle ne , µ g/g 10 / 24
How Far Does the Light Shine? Key Comparison CCQM-K47 demonstrated the capabilities of participating NMIs to identify and measure the four target VOCs benzene, o-xylene, m-xylene, and p-xylene in a calibration solution using GC-based methods. These measurands were selected to be representative of VOCs monitored in water quality assessments. The measurement challenges in CCQM-K47, such as avoiding volatility loss, achieving adequate chromatographic resolution, and isolating potential interferences, are typical of those required for value-assigning volatile reference materials. Participants achieving comparable measurements for all four VOCs in this Key Comparison should be capable of providing reference materials and measurements for VOCs in solutions when present at concentration levels greater than 10 µg/g. 11 / 24
Appendix A: NMI Reports BAM CENAM Uncertainty was calculated by using ISO-GUM guide 1993 "Quantifying Uncertainty in Analytical Measurement" standard deviation between ampoules, same ampoule standard deviation, calibration curve error and uncertainty by solvent purity was combined and the combined uncertainty multiplied by k=2. INTI Uncertainty estimated from the calibration curve (sample variability agrees with the variability of the calibration curve 12 / 24
KRISS Source of Uncertainty Relative standard uncertainty (rel %) from each Source Benzene o-Xylene m-Xylene p-Xylene Standard Solution (rel%) 0.55% 0.50% 0.56% 0.56% Purity (rel%) 0.005% 0.005% 0.008% 0.005% Repeatability of gravimetric dilution of Stock solution (rel%) 0.52% 0.44% 0.18% 0.18% Repeatability of gravimetric preparation of Stock solution 0.04% 0.24% 0.53% 0.53% Measurement (rel%) Repeatability of anaylsys(Within sampl 0.1-0.8% 0.1-0.8% 0.1-0.8% 0.1-0.8% Between sample 0.21-0.22% 0.03-0.15% 0.06-0.28% 0.09-0.32% Combined Uncertainty 0.79% 0.75% 0.71% 0.74% Expended Uncertainty(95%) 1.61% 1.52% 1.43% 1.49% NIST benzene o-xylene m-xylene p-xylene degrees of freedom Measured Value (mean) 39.38 15.60 20.19 28.26 Uncertainty Components Measurement of Samples 0.177 0.070 0.214 0.208 7 Measurement of Calib Stds 0.002 0.001 0.001 0.006 5 purity of compounds used for prep of calibration solutions 0.002 0.003 0.001 0.001 2 Conc of Calib Soln 0.394 0.156 0.202 0.283 inf Combined Standard Unc. 0.432 0.171 0.294 0.351 Effective degrees of freedom 249 254 25 56 k (from t-distribution) 1.97 1.97 2.06 2.00 Expanded Uncertainty 0.85029 0.34 0.61 0.70 Expanded Unc as % 2.16% 2.16% 3.00% 2.49% NMIJ Rsample Rstd Mx Weight of sample solution (g) Mz Weight of standard solution (g) MYSmp Weight of spike solution (g) added to sample solution MYStd Weight of spike solution (g) added to standard solution Ccalib Concentration of calibration solution (ug/g) Analyte CX (ug/g): Analyte/D‑analyte ratio for sample blend Analyte/D‑analyte ratio for standard blend YStdXstd YSmpcalibZsample X MMR MCMRC = 13 / 24
NMIJ: Benzene Parameter Source of Uncertainty (∂f/∂xi) (∂f/∂xi)u(xi) Degrees of Freedom (i) Type Source of Data Pmethod Precision for the entire method 39.26 ug/g 0.02 ug/g 1 0.0160 39 A Replicate analysis of the 4 ampoules Rsample Measurement of Rsample 1.15 Covered withinPmethod - - - - - Rstandard Measurement of Rstd 1.05 Covered withinPmethod - - - - - MZ (L) Balance linearity 0.42 g 0.0003 g 93.477 0.028 Large B Balance calibration certificate Ccalib 39.3 ug/g 0.09825 ug/g 0.999 0.0982 Large B Supplier’s specification MYsmp (L) Balance linearity 0.2 g 0.0003 g 196.303 0.0589 Large B Balance calibration certificate MYstd (L) Balance linearity 0.2 g 0.0003 g -196.303 -0.0589 Large B Balance calibration certificate MX (L) Balance linearity 0.44 g 0.0003 g -89.228 -0.0268 Large B Balance calibration certificate Xi U(xi) NMIJ: o-xylene Parameter Source of Uncertainty (∂f/∂xi) (∂f/∂xi)u(xi) Degrees of Freedom (i) Type Source of Data Pmethod Precision for the entire method 16.42 ug/g 0.01 ug/g 1 0.0128 39 A Replicate analysis of the 4 ampoules Rsample Measurement of Rsample 1.07 Covered withinPmethod - - - - - Rstandard Measurement of Rstd 1.08 Covered withinPmethod - - - - - MZ (L) Balance linearity 0.42 g 0.0003 g 39.095 0.012 Large B Balance calibration certificate Ccalib 16.6 ug/g 0.0415 ug/g 0.989 0.0410 Large B Supplier’s specification MYsmp (L) Balance linearity 0.2 g 0.0003 g 82.099 0.0246 Large B Balance calibration certificate MYstd (L) Balance linearity 0.2 g 0.0003 g -82.099 -0.0246 Large B Balance calibration certificate MX (L) Balance linearity 0.44 g 0.0003 g -37.318 -0.0112 Large B Balance calibration certificate uc= 0.06 Xi U(xi) 14 / 24
NMIJ: m-xylene Parameter Source of Uncertainty (∂f/∂xi) (∂f/∂xi)u(xi) Degrees of Freedom (i) Type Source of Data Pmethod Precision for the entire method 19.92 ug/g 0.01 ug/g 1 0.0144 39 A Replicate analysis of the 4 ampoules Rsample Measurement of Rsample 0.92 Covered withinPmethod - - - - - Rstandard Measurement of Rstd 0.94 Covered withinPmethod - - - - - MZ (L) Balance linearity 0.42 g 0.0003 g 47.417 0.014 Large B Balance calibration certificate Ccalib 20.1 ug/g 0.05025 ug/g 0.991 0.0498 Large B Supplier’s specification MYsmp (L) Balance linearity 0.2 g 0.0003 g 99.575 0.0299 Large B Balance calibration certificate MYstd (L) Balance linearity 0.2 g 0.0003 g -99.575 -0.0299 Large B Balance calibration certificate MX (L) Balance linearity 0.44 g 0.0003 g -45.262 -0.0136 Large B Balance calibration certificate uc= 0.07 Xi U(xi) NMIJ: p-xylene Parameter Source of Uncertainty (∂f/∂xi) (∂f/∂xi)u(xi) Degrees of Freedom (i) Type Source of Data Pmethod Precision for the entire method 28.12 ug/g 0.02 ug/g 1 0.0192 39 A Replicate analysis of the 4 ampoules Rsample Measurement of Rsample 0.97 Covered withinPmethod - - - - - Rstandard Measurement of Rstd 1 Covered withinPmethod - - - - - MZ (L) Balance linearity 0.42 g 0.0003 g 66.959 0.020 Large B Balance calibration certificate Ccalib 28.4 ug/g 0.071 ug/g 0.990 0.0703 Large B Supplier’s specification MYsmp (L) Balance linearity 0.2 g 0.0003 g 140.614 0.0422 Large B Balance calibration certificate MYstd (L) Balance linearity 0.2 g 0.0003 g -140.614 -0.0422 Large B Balance calibration certificate MX (L) Balance linearity 0.44 g 0.0003 g -63.915 -0.0192 Large B Balance calibration certificate uc= 0.10 Xi U(xi) 15 / 24
VSL The content of each ampoule is divided in two aliquotes and placed in an autosampler vial and added of internal standard. 2 series of measurements are carried out (samples + gravimetrically prepared liquid calibrants (BTEX in MeOH)) A weighted line of regression is obtained for each component in each series. The component average result per series (4 samples analysed twice) and corresponding uncertainty are further calculated The contributions to the standard uncertainty of each series (U (k=2) between 1-5%) are mainly: a- calibrants preparation (weighing, purity and potential losses due to evaporation) b- lack of fit and regression coefficient uncertainties (main uncertainty source) c- repeatability of the sample measurement The final value is the mean value of the two series of measurements. The uncertainty in the final result is the pooled estimate of uncertainty of the series of measurements added of the mean standard deviation of the two series results. The reported expanded standard uncertainty has a coverage factor of two. 16 / 24
APPENDIX B: Chromatograms BAM GC-MS, K47 sample 17 / 24
GC-FID K47 sample min20 20.25 20.5 20.75 21 21.25 21.5 21.75 22 pA 20 25 30 35 40 45 FID2 B, (2006_S\KAMINSKI\008F0801.D) p-Xylene m-Xylene o-Xylene CENAM 18 / 24
19 / 24
20 / 24
INTI 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 standard 35 ug/g 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 sample CCQM-K47 21 / 24
KRISS 22 / 24
NIST GC-FID for Xylenes 23 / 24
NMIJ GC/MS for Benzene – Calibration Solution 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.000 50000 100000 150000 200000 Time--> Abundance Ion 78.00 (77.70 to 78.70): MK47007.D 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.000 10000 20000 30000 40000 50000 60000 Time--> Abundance Ion 116.00 (115.70 to 116.70): MK47007.D GC/MS for Benzene – CCQM-K47 Solution 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.000 1000 2000 3000 4000 5000 6000 7000 Time--> Abundance Ion 78.00 (77.70 to 78.70): MK47008.D 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00 0 5000 10000 15000 20000 25000 30000 35000 Time--> Abundance Ion 116.00 (115.70 to 116.70): MK47008.D 24 / 24
VSL Ver+/- | |
![]() | Descargar |
Atrás |