TECHNICAL REPORT ISO/TR 24094 First edition 2006-05-15 `,,```,,,,````-`-`,,`,,`,`,,` - Analysis of natural gas — Validation methods for gaseous reference materials Analyse du gaz naturel — Méthodes de validation pour matériaux de référence gazeux Reference number ISO/TR 24094:2006(E) Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 Not for Resale ISO/TR 24094:2006(E) PDF disclaimer This PDF file may contain embedded typefaces In accordance with Adobe's licensing policy, this file may be printed or viewed but shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing In downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy The ISO Central Secretariat accepts no liability in this area Adobe is a trademark of Adobe Systems Incorporated Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation parameters were optimized for printing Every care has been taken to ensure that the file is suitable for use by ISO member bodies In the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below `,,```,,,,````-`-`,,`,,`,`,,` - © ISO 2006 All rights reserved Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or ISO's member body in the country of the requester ISO copyright office Case postale 56 • CH-1211 Geneva 20 Tel + 41 22 749 01 11 Fax + 41 22 749 09 47 E-mail copyright@iso.org Web www.iso.org Published in Switzerland ii Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) Contents Page Foreword iv Scope Normative references Development of the validation methods Results of the VAMGAS project Annex A (informative) Report on the validation methods for gaseous reference materials `,,```,,,,````-`-`,,`,,`,`,,` - Bibliography 47 iii © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale ISO/TR 24094:2006(E) Foreword ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies (ISO member bodies) The work of preparing International Standards is normally carried out through ISO technical committees Each member body interested in a subject for which a technical committee has been established has the right to be represented on that committee International organizations, governmental and non-governmental, in liaison with ISO, also take part in the work ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization `,,```,,,,````-`-`,,`,,`,`,,` - International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part The main task of technical committees is to prepare International Standards Draft International Standards adopted by the technical committees are circulated to the member bodies for voting Publication as an International Standard requires approval by at least 75 % of the member bodies casting a vote In exceptional circumstances, when a technical committee has collected data of a different kind from that which is normally published as an International Standard (“state of the art”, for example), it may decide by a simple majority vote of its participating members to publish a Technical Report A Technical Report is entirely informative in nature and does not have to be reviewed until the data it provides are considered to be no longer valid or useful Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights ISO shall not be held responsible for identifying any or all such patent rights ISO/TR 24094 was prepared by Technical Committee ISO/TC 193, Natural gas, Subcommittee SC 1, Analysis of natural gas iv Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale TECHNICAL REPORT ISO/TR 24094:2006(E) Analysis of natural gas — Validation methods for gaseous reference materials Scope This Technical Report describes the validation of the calorific value and density calculated from current practice natural gas analysis by statistical comparison with values obtained by measurement using a reference calorimeter and a density balance Normative references The following referenced documents are indispensable for the application of this document For dated references, only the edition cited applies For undated references, the latest edition of the referenced document (including any amendments) applies ISO 6142, Gas analysis — Preparation of calibration gas mixtures — Gravimetric method ISO 6974-1, Natural gas — Determination of composition with defined uncertainty by gas chromatography — Part 1: Guidelines for tailored analysis ISO 6974-2, Natural gas — Determination of composition with defined uncertainty by gas chromatography — Part 2: Measuring-system characteristics and statistics for processing of data ISO 6974-3, Natural gas — Determination of composition with defined uncertainty by gas chromatography — Part 3: Determination of hydrogen, helium, oxygen, nitrogen, carbon dioxide and hydrocarbons up to C8 using two packed columns ISO 6974-4, Natural gas — Determination of composition with defined uncertainty by gas chromatography — Part 4: Determination of nitrogen, carbon dioxide and C1 to C5 and C6+ hydrocarbons for a laboratory and on-line measuring system using two columns ISO 6974-6, Natural gas — Determination of composition with defined uncertainty by gas chromatography — Part 6: Determination of hydrogen, helium, oxygen, nitrogen, carbon dioxide and C1 to C8 hydrocarbons using three capillary columns ISO 6976, Natural gas — Calculation of calorific values, density, relative density and Wobbe index from composition Guide to the expression of uncertainty in measurement (GUM), BIPM/IEC/IFCC/ISO/IUPAC/IUPAP/OIML, 1995 © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale `,,```,,,,````-`-`,,`,,`,`,,` - ISO 6974-5, Natural gas — Determination of composition with defined uncertainty by gas chromatography — Part 5: Determination of nitrogen, carbon dioxide and C1 to C5 and C6+ hydrocarbons for a laboratory and on-line process application using three columns ISO/TR 24094:2006(E) Development of the validation methods The validation methods for gaseous reference materials (VAMGAS) project was established by a group of European gas companies as an approach to confirming the practices used in natural gas analysis and physical property calculations The VAMGAS project proposed comparing the calorific value and density calculated from the current practices for natural gas analyses with values obtained by measurement using a reference calorimeter (located at the Ofgas, UK laboratory) and density balance (located at the Ruhrgas, Germany laboratory) Robust statistical comparisons allowed an assessment of the validity of the practices The natural gas analysis practice covered by the VAMGAS project can be divided into the following steps: ⎯ The gravimetric preparation of gas mixtures used as calibrants in the analysis of natural gas in accordance with ISO 6142 At the highest level, these mixtures are categorized as primary reference gas mixtures (PRMs) and are available from national institutes such as Bundesanstalt fur Materialforschung und -prüfung (BAM) of Germany and Nederlands Meetinstituut (NMi) of the Netherlands ⎯ The analysis of natural gas by gas chromatographic methods, such as those given is ISO 6974 (all parts) This is a multiple part International Standard that provides a number of different approaches to the gas chromatographic analysis of natural gas ISO 6974-2 describes the processing of calibration and analytical data to determine the uncertainties on sample component concentrations that are required for the calculation of uncertainties on calculated physical property values of the sample gas ⎯ The calculation of the values of physical properties from the results of the gas chromatographic analyses as described in ISO 6976 The VAMGAS project was divided in two parts: a) Part 1: comparison of the calorific values and densities of two PRMs calculated from the gravimetric preparation data against the values obtained from the reference calorimeter and density balance (see Figure 1); b) Part 2: gas chromatography intercomparison exercise, in which calorific values and densities calculated from the analyses of two natural gases (with bracketing calibration using PRMs) were compared to the values obtained from the reference calorimeter and density balance (see Figure 2) `,,```,,,,````-`-`,,`,,`,`,,` - The two separate exercises would enable problems arising from either the gravimetric preparation or the gas chromatographic analyses to be identified The participants in the VAMGAS project were Ruhrgas AG (Germany and project co-ordinator), Gasunie (the Netherlands), Gaz de France (France), BAM (Germany), NMi (the Netherlands) and Ofgem (previously Ofgas, the UK) In addition, a total of 18 laboratories participated in the gas chromatography intercomparison The technical report from the VAMGAS is given in Annex A Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) Figure — Schematic concept of part of the VAMGAS project © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS `,,```,,,,````-`-`,,`,,`,`,,` - Not for Resale ISO/TR 24094:2006(E) `,,```,,,,````-`-`,,`,,`,`,,` - Figure — Schematic concept of part of the VAMGAS project Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) Results of the VAMGAS project a) The results of the exercise using the PRMs showed statistical agreement between the calorific values and densities calculated from the gravimetric preparation data and the values of these physical properties obtained from direct measurement using reference instruments b) The results of the gas chromatographic intercomparison showed statistical agreement between the calorific values and densities calculated from gas chromatographic analyses, carried out using PRMs as calibrants, and the values of these physical properties obtained from direct measurements using reference instruments It can be concluded that the VAMGAS project has validated the current systems of natural gas analyses and calculation of physical property data involving the previously mentioned ISO International Standards As a result, all parties in the supply and use of natural gas, whether supplier or consumer, can have confidence in these The current ISO International Standards for calibration gas preparation and natural gas analysis, if carefully applied, give values of calorific value and density that are in agreement with values that were independently determined by reference measurements This also includes the tabulated values in ISO 6976, which are used in calculations of thermal energy for billing/fiscal transfer purposes The VAMGAS project was carried out as an integrated project to study the complete system of natural gas analysis involving the gravimetric preparation of calibration gas mixtures, the gas chromatographic analysis and calculation of physical properties Reference measurements of the physical properties were applied during the VAMGAS project as a means of assessing the system It is stressed that readers take account of the whole project; and it is totally wrong to take isolated parts and results of the project and use these for other purposes in the belief that the project results justify such an approach For example, in the first part of the project comparison was made between the physical property values calculated from the gravimetric preparation data of the PRMs and the values obtained from the reference measurements It is important not to use the results from this part of the project to justify using reference measurements of a physical property to validate the composition of a prepared natural gas mixture There are three reasons ⎯ The VAMGAS project was not designed to investigate the applicability, or otherwise, of such a procedure The VAMGAS project was designed to investigate whether or not a cylinder of gas designated as a PRM can provide gas of the composition given on the certificate attached to that cylinder ⎯ In the preparation of the PRMs, the national institutes have rigorous procedures including a system of validating the mixture composition by gas chromatographic analysis to give confidence in the composition of the gas mixture ⎯ Whereas it is true that a gas mixture of known composition has an unique calorific value or density, the same is not true of the reverse relationship: a specific calorific value or density does not have a corresponding unique gas composition; in fact a calorific value or density can result from an almost infinite number of different gas compositions Hence, it is not technically feasible to validate gas mixture compositions using measurements of a physical property As a simple illustration, consider the manufacture of a multi-component mixture containing both isomers of butane If, by mistake, the same isomer was added twice then the resulting mixture would have the same calorific value and density as the required mixture but the composition would be incorrect Measurements of the calorific value or density would appear to validate the mixture composition when it was, in fact, in error © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale `,,```,,,,````-`-`,,`,,`,`,,` - The project report provides results on two sets of comparisons ISO/TR 24094:2006(E) Annex A (informative) Report on the validation methods for gaseous reference materials A.1 General A.1.1 Summary In the first part of the project, 12 primary reference gas mixtures were produced by BAM and NMi As regards composition, the gas mixtures produced were similar to type L Groningen gas and type H North Sea gas The superior calorific value, Hs, molar mass, M, and density at normal conditions of the mixtures were calculated from the component concentrations specified by the producers The calculated data were then compared with the results of direct measurements of physical properties The methods used for direct measurement of physical properties were reference calorimetry[1] and precision densitometry[2] Statistically significant agreement was found between the calculated data and the measurements Table A.1 — Comparison of experimental (Mexp) and calculated (Mcalc) values of the molar mass for different PRMs a `,,```,,,,````-`-`,,`,,`,`,,` - a Gas mixture Type of gas Mexp g/mol Mcalc g/mol Relative difference % BAM 9605 4933 L 18,564 18,564 0,002 NMi 0602E L 18,542 18,543 0,002 BAM 9605 4902 H 18,793 18,796 0,018 NMi 9497C H 18,946 18,946 0,002 Calculations are made in accordance with ISO 6976 In the second stage, 20 natural gas samples was taken from the natural gas transmission system of Ruhrgas AG These samples included both type L Groningen gas and type H North Sea gas Gas samples were taken in batches, so that the compressed gas cylinders filled with each of the two types were of identical composition The homogeneity of the batches, i.e the agreement between the compositions of the samples in the various gas cylinders, was verified using the precision densitometer The stability of the gas samples during sampling was also tested Table A.2 — Comparison of experimental (ρexp) and calculated (ρcalc) values of the gas density at standard conditions for different PRMs a a kg/m3 kg/m3 ρcalc Relative difference % L 0,773 19 0,773 19 — NMi 0602E L 0,772 29 0,772 38 0,012 BAM 9605 4902 H 0,783 24 0,783 41 0,022 NMi 9497C H 0,789 67 0,789 72 0,006 Gas mixture Type of gas BAM 9605 4933 ρexp Calculations are made in accordance with ISO 6976 Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) A.3.3.3 Relative response factors `,,```,,,,````-`-`,,`,,`,`,,` - The indirect components, i e those that were directly calibrated, were evaluated using relative response factors in accordance with Equation (A.14): ⎛x ⎞ x i,indirect = Ai,indirect × ⎜ b ⎟ × Ri,f ⎝ Ab ⎠ (A.14) where xi,indirect concentration of the indirect component; Ai,indirect peak area of the indirect component; xb concentration of the bridge component; Ab peak area of the bridge component; Ri,f relative response factor The relative response factors were set to their theoretical values using a formula developed by Kaiser, as given in Equation (A.15): Ri,f = nb × M i,indirect (A.15) ni,indirect × M b where Mi,indirect, Mb are the molar masses of the indirect and bridge components, respectively; ni,indirect, nb A.3.3.4 are the number of carbon atoms in the respective components Evaluation of component groups Depending on the design of the gas chromatograph, several of the components of the natural gas samples are not separated Critical components in this category include 2,3-dimethylbutane/2-methylpentane/ 3-methylpentane and cyclopentane/2,2-dimethyl butane Gas chromatographs with column backflushing not separate components above a certain chain length at all Furthermore, the number of the isomers of hydrocarbons above C5 is too high to allow reporting of each individual component In these cases, “pseudo-components” were reported; these were evaluated in a special way Assumptions concerning the average number of carbon atoms of the pseudo-components, their enthalpy, density and molar mass were made on the basis of the results of the four laboratories that had reported all the individual components A.3.3.5 Detection of outliers A Grubbs test for the detection of outliers was performed for the calculated physical properties Two outliers were detected and removed from the data set before determining the mean values and the uncertainty of the mean for calorific value, density and molar mass A.3.3.6 Results The individual results of the participating laboratories are given in Figures A.7 to A.12 With respect to the desired method validation, however, the mean of the laboratories is of prime interest The values are summarized in Table A.13 34 Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) Table A.13 — Comparison of calorific value, molar mass and gas density of the mean of the laboratories and the results from the reference measurements Parameter Type of gas Mean of the laboratories Calorific value, MJ/kg H 52,561 52,563 0,003 L 44,701 44,688 0,027 H 18,115 18,122 0,036 L 18,604 18,612 0,045 H 0,7549 0,7551 0,034 L 0,7748 0,7752 0,048 Molar mass, kg/kmol Density, A.3.3.7 kg/m3 Reference method Relative difference % Calorific value The calorific value, calculated as the mean of the participating laboratories, was compared to the reference value by a z-test Agreement was within 0,003 % for the type H natural gas and within approximately 0,03 % for type L gas A z-test demonstrated that the mean of the laboratories was statistically equal (within 95 % confidence interval) to the reference It is noteworthy that both the uncertainty of the results of the individual laboratories and the lack of agreement with the reference value is higher for type L gas than for type H gas due to the larger amount of nitrogen A.3.3.8 Gas density The gas density derived from gas chromatography is also in statistically proven agreement with the reference measurements The agreement of the average of the laboratories is within 0,04 % for L gas and 0,05 % for H gas A.3.3.9 Molar mass Agreement with the reference values is essentially on the same level as for the gas density For type L gas, the values agree within 0,03 %; for type H gas, the agreement is 0,05 % The uncertainty for the individual laboratories, however, is very small compared to the inter-laboratory variance For this reason, the molar mass derived from composition analysis is statistically not equal to the reference value, if the uncertainty estimate is based upon the uncertainties of the individual laboratories If the uncertainty estimate is based upon the inter-laboratory variance, the z-test gives statistical agreement for type H gas, but not for type L gas It can, therefore, be concluded that an uncertainty estimate that is based upon the repeatability of the measurements most probably underestimates the real uncertainty of composition analysis A small but nevertheless notable bias of 0,044 % can be detected between gas chromatography and reference densitometry, with the value of the reference method being higher than the gas chromatographic value A possible reason for this bias can be that the concentrations of the minor components are underestimated in the gas analysis One contribution is certainly due to components that are below the detection limit of the respective methods Components that are not detected as a result of normalization are evaluated like the matrix gas, i.e methane The molar mass calculated from gas chromatography is likely, therefore, always be lower than the true value The natural gas samples contain hydrocarbons above C10, which are normally near or below the detection limit of the GC system As these components have a large molar mass, they contribute to the molar mass of the gas sample even at low concentrations These higher hydrocarbons were quantified in the natural gas samples by the Ruhrgas laboratory using the method described previously The analysis results are given in `,,```,,,,````-`-`,,`,,`,`,,` - 35 © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale ISO/TR 24094:2006(E) Table A.14 These values give rise to the assumption that approximately 50 % of the detected bias of 0,04 %, i.e 40 ppm, are due to the C10+ fraction On the other hand, the contribution of the higher hydrocarbons to the calorific value is certainly negligible on the uncertainty scale considered, as the variation of calorific value on a mass basis with hydrocarbon chain length is very small With respect to the gas density, these biased component uncertainties are obscured by the higher uncertainty level of the calculated density values Table A.14 — Component concentrations of the C10+ fraction of the natural gas samples as determined by a special analysis technique Component Type H natural gas mol/mol × 106 Type L natural gas mol/mol × 106 C10 1,55 4,71 C11 0,19 1,49 C12 0,03 0,49 C13 < 0,01 0,09 C14 < 0,01 0,01 C15 < 0,01 0,01 Key X laboratory number Y calorific value, CV, expressed in megajoules per kilogram (scale ± 0,2 %) result of the reference method: CVref = (52,563 ± 0,0134) MJ/kg confidence interval ■ average of single-laboratory results confidence interval of single-laboratory results Figure A.7 — Comparison of the results of the individual laboratories with the result of the reference method for the calorific value of the type H natural gas `,,```,,,,````-`-`,,`,,`,`,,` - 36 Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) Key X laboratory number Y calorific value, CV, expressed in megajoules per kilogram (scale ± 0,2 %) result of the reference method: CVref = (44,688 ± 0,012 1) MJ/kg confidence interval ■ average of single-laboratory results confidence interval of single-laboratory results Figure A.8 — Comparison of the results of the individual laboratories with the result of the reference method for the calorific value of the type L natural gas `,,```,,,,````-`-`,,`,,`,`,,` - 37 © ISO for 2006 – All rights reserved Copyright International Organization Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale ISO/TR 24094:2006(E) Key X Y laboratory number molar mass, M, expressed in kilograms per kilomole (scale ± 0,2 %) result of the reference method: Mref = (18,121 ± 0,001 8) kg/kmol confidence interval ■ average of single-laboratory results confidence interval of single-laboratory results Figure A.9 — Comparison of the results of the individual laboratories with the result of the reference method for the molar mass of the type H natural gas `,,```,,,,````-`-`,,`,,`,`,,` - 38 Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) Key X Y laboratory number molar mass, M, expressed in kilograms per kilomole (scale ± 0,2 %) `,,```,,,,````-`-`,,`,,`,`,,` - result of the reference method: Mref = (18,611 ± 0,001 9) kg/kmol confidence interval ■ average of single-laboratory results confidence interval of single-laboratory results Figure A.10 — Comparison of the results of the individual laboratories with the result of the reference method for the molar mass of the type L natural gas 39 © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale ISO/TR 24094:2006(E) Key X Y laboratory number density, D, expressed in kilogram per cubic metre (scale ± 0,8 %) result of the reference method: Dref = (0,755 11 ± 0,000 2) kg/m3 at T = 293,15 K and p = 101,325 kPa) confidence interval ■ average of single-laboratory results confidence interval of single-laboratory results `,,```,,,,````-`-`,,`,,`,`,,` - Figure A.11 — Comparison of the results of the individual laboratories with the result of the reference method for the density of the type H natural gas 40 Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) `,,```,,,,````-`-`,,`,,`,`,,` - Key X Y laboratory number density, D, expressed in kilogram per cubic metre (scale ± 0,8 %) result of the reference method: Dref = (0,755 21 ± 0,000 2) kg/m3 at T = 293,15 K and p = 101,325 kPa confidence interval ■ average of single-laboratory results confidence interval of single-laboratory results Figure A.12 — Comparison of the results of the individual laboratories with the result of the reference method for the density of the type L natural gas A.3.4 Uncertainty sources In this report, “traceability” is used to mean the statement of a measurement result together with the associated uncertainty On this basis, the necessity to pay considerable attention to the contribution made by each individual method used to the overall uncertainty level is clear In Annex A, only the results of the uncertainty calculations are summarized A.3.4.1 Gas composition, gravimetric preparation and calculated physical properties The uncertainties of the calculated physical properties of the PRMs were calculated using strict errorpropagation rules, starting from the gravimetric preparation process and taking the uncertainties of the tabulated values into account It was necessary to re-evaluate the uncertainties of the tabulated values with respect to ISO 6976, as the confidence intervals were not stated in some cases The overall uncertainties of the different properties are shown in Table A.15 Figure A.13 shows the different contributions to the overall uncertainties With respect to the composition uncertainty, the contributions of the individual components of the gas mixture are given in Figure A.14 However, as a result of the very low uncertainty of the gravimetric preparation process, the contribution of the uncertainty in composition to the overall uncertainty is relatively small (see Figure A.13) In Figure A.13, the uncertainty of the calorific value is dominated by the uncertainty of the calorific value of methane 41 © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale ISO/TR 24094:2006(E) Table A.15 — Composition uncertainties of PRMs manufactured by NMi Relative standard uncertainty of typical H-gas and L-gas given by NMi Component H-gas % L-gas % Methane, CH4 0,001 0,001 Carbon dioxide, CO2 0,005 0,006 Ethane, C2H6 0,006 0,009 Propane, C3H8 0,011 0,010 n-butane, n-C4H10 0,012 0,010 i-butane, iso-C4H10 0,012 0,011 n-pentane, n-C5H12 0,025 — Nitrogen, N2 0,014 0,005 Key X1 molar mass of H-gas X2 molar mass of L-gas X3 density of H-gas X4 density of L-gas X5 calorific value of H-gas X6 calorific value of L-gas xi, composition Mi, molar mass R, gas constant H, enthalply on a mass basis Figure A.13 — Assessment of the uncertainty contributions to the calculated physical properties 42 Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS `,,```,,,,````-`-`,,`,,`,`,,` - © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) Key X1 molar mass of H-gas X2 molar mass of L-gas X3 density of H-gas X4 density of L-gas X5 enthalply, mass of H-gas X6 enthalply, mass of L-gas `,,```,,,,````-`-`,,`,,`,`,,` - Y relative contribution, % CH4 C2H4 C3H8 i-C4H10 n-C4H10 n-C5H12 N2 CO2 Figure A.14 — Assessment of the source of uncertainty of calculated physical properties A.3.4.2 Densitometry As the uncertainty assessment of the densitometer is rather complex, only the results for a 95 % confidence interval are given here The expanded uncertainty of a single density measurement is u 0,02 % (relative) for ρ W kg/m3 and u 0,000 kg/m3 for ρ < kg/m3 In order to determine the molar mass from density measurements, a least squares fit to 14 single measurement is performed The expanded uncertainty of the measured molar mass is quoted as u 0,02 % (relative) 43 © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale ISO/TR 24094:2006(E) A.3.4.3 Reference calorimetry A full-error propagation calculation in accordance with GUM was performed by Ofgas for their reference calorimeter The expanded uncertainty of the calorific value as determined by the procedure given by Ofgas was calculated at 0,034 % (relative) for a 95 % confidence interval The sources of uncertainty are the following: ⎯ heat evolved during analysis, contributing % of the total uncertainty; ⎯ mass determination by weighing, contributing 94 % of the total uncertainty A.3.4.4 Gas chromatography of natural gas samples Initially, an uncertainty estimate of the gas chromatographic results was made on the basis of the repeatability that had been measured in part of the inter-comparison This uncertainty estimate can be found in the error bars in Figures A.7 to A.12 It is clear that that repeatability in analytical chemistry does not include all possible uncertainty sources A better uncertainty estimate also includes reproducibility Reproducibility can be derived only from a number of completely independent measurements, which are difficult to obtain The inter-laboratory variance, however, is a good estimate of reproducibility These values are listed in Table A.16 The inter-laboratory variance gives a good estimate of the method uncertainty of natural gas analysis by gas chromatography Table A.16 — Comparison of calculated (mean of the inter-comparison participants) physical property values for the natural gas samples with the reference values Parameter Sample Reference Reference uncertainty Laboratory Laboratory uncertainty Difference % Statistical z-test Calorific value H-NG 52,563 0,018 52,561 0,015 0,003 0,09 L-NG 44,688 0,015 44,701 0,024 0,027 0,79 Density H-NG 0,755 0,000 0,754 0,001 0,034 1,25 L-NG 0.775 0.000 0.774 0.001 0.048 1.85 H-NG 18,121 0,003 18,115 0,005 0,036 1,78 L-NG 18,611 0,003 18,603 0,006 0,045 2,16 Molar mass A.3.4.5 Method uncertainty of gas chromatography On the assumption that the best possible calibration technique is used, the calorific value of type H gases can determined by gas chromatography with a method uncertainty of 0,02 % (95 % confidence interval), if the uncertainty of the calorific value of methane (0,1 %) is neglected The uncertainty of the calorific value determined by gas chromatography increases with increasing inert gas content Density and molar mass can be determined with an uncertainty of 0,04 % to 0,05 % Most of the uncertainty of the molar mass is due to a bias resulting from components below the detection limit This bias is negative, the molar mass determined is always smaller than the true value A.4 Exploitation and dissemination of the results The VAMGAS project was presented at the Gas Analysis Symposium held in Eindhoven, in November 1999 The final project presentation was held in Brussels, in April 2000 Furthermore, the results of the VAMGAS project were presented at the IGRC, in November 2001, in Amsterdam `,,```,,,,````-`-`,,`,,`,`,,` - 44 Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) The results of the VAMGAS project are published as this Technical Report In order to disseminate the results into the European metrological society, the working group decided to submit a proposal to EUROMET to officially acknowledge the VAMGAS project Since a EUROMET proposal can only be submitted by a national metrological institute, the proposal was presented to EUROMET in January 1999, by Anton Alink of NMi EUROMET did well recognize the quality of the results presented However, EUROMET decided not to accept the project as an official EUROMET project, because no additional metrological institutes could participate in the project `,,```,,,,````-`-`,,`,,`,`,,` - As a spin-off of the VAMGAS project, a working group was formed within GERG to perform a feasibility study for reference calorimeter The companies involved in this GERG reference calorimeter study are PTB, Advantica, Ruhrgas and SNAM The study was concluded in October 2000 A.5 Conclusions A.5.1 Project The project has demonstrated that the confidence currently shown in state-of-the-art methods for the production of reference gas mixtures is justified For the first time, it was possible, with considerable effort, to verify gas mixtures of the highest quality produced by gravimetric methods using independent methods with an uncertainty equivalent to that of the production process In the future, calibration gas producers will be able to verify their uncertainty claims for the production of precision gas mixtures using the methods demonstrated in this project As discussed in A.5.2, customers have called for such verification This Technical Report contains the results obtained in the project The results of the project underline the importance of reference methods of gas density measurement and calorific value determination for gas analysis However, it is necessary to note that there has been a drastic fall in the number of reference measurement facilities in operation During work on the project, the reference calorimeter operated by Ofgem was the only reference calorimeter in service in the world In this context, the practical use of the procedure described for customers is naturally severely restricted National metrological laboratories are called upon to become active with respect to this or other reference methods with a view to restoring the variety of processes that was once available The superior calorific values listed in ISO 6976 are only based on two investigations The uncertainties 0,1 % stated are no longer acceptable It is also necessary to review and correct these values in order ensure fair and equitable treatment for consumers This is a future task for metrological institutes cooperation with the gas industry A project of this type has already been initiated under the auspices GERG of to in of The possibility demonstrated in this project of highly precise determination of the molar mass of a gas is of special scientific importance in the field of metrology In this way, it is possible to establish a metrological link between gas measurements and the SI unit, the mole The round-robin tests carried out as part of the product have shown that uncertainty levels in routine natural gas analyses in Europe are very low These results of the round-robin tests have given organizations involved in gas analysis considerable confidence in the comparability of their results In general, the participants expressed satisfaction with the results In the future, it will be necessary to conduct further high-quality roundrobin tests covering gas mixtures occurring naturally in addition to synthetic mixture A.5.2 Requirements of calibration gas users The VAMGAS project has shown that highly precise calibration gases on the level of primary reference materials can be produced by gravimetric methods The very low uncertainty obtained is feasible if extreme care is taken and highly precise balances and highly pure gases are used in an optimized production process for gases sufficiently far from the two-phase range 45 © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale ISO/TR 24094:2006(E) It is necessary that users take due care in handling these valuable gases with a view to ensuring that this very low uncertainty is maintained over the duration of use of the gas mixtures The rules to be followed were documented The uncertainties stated in manufacturers' certificates are often higher than the mathematical uncertainty of the weighing process These uncertainty values often include a component representing the uncertainty of analytical verification of the gas mixtures produced In this way, the manufacturers attempt to take account of all the error sources that cannot be assessed in quantitative terms, e.g minor deviations in the manufacturing process In addition, these uncertainties are intended to cover possible adverse affects of shipment (e.g by air freight) After all, manufacturers' certificates guarantee compliance with the uncertainty values stated `,,```,,,,````-`-`,,`,,`,`,,` - At the final project presentation, the round-robin test participants, who are all users of calibration gases, stated that they would prefer manufacturers to be less generous in their handling of uncertainty values in manufacturers' certificates In addition, users are interested in a further extension in the shelf lives (stability periods) of costly calibration gases In this context, calibration gas producers are called upon to engage in more intensive dialogue with their customers 46 Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS © ISO 2006 – All rights reserved Not for Resale ISO/TR 24094:2006(E) Bibliography [1] PITTAM, D.A., AND PILCHER, G., J Chem Soc Faraday Trans., I 68, 1972, pp 2224 to 2229 [2] PITTAM, D.A The Measurement of Heats of Combustion by Flame Calorimetry, MSc Thesis, University of Manchester,1971 [3] ROSSINI, F.D The Heat of Formation of Water, Journal of Research National Bureau of Standards, 6, 1931, pp to 35 [4] ROSSIN, F.D The Heats of Combustion of Methane and Carbon Monoxide, Journal of Research National Bureau of Standards, 6, 1931, pp 37-49 [5] PIEPERBECK, N., KLEINRAHM, R., W AGNER, W and JAESCHKE, M Results of (pressure, density, temperature) measurements of methane and on nitrogen in the temperature range from 273.15 K to 323.15 K at pressures up to 12 MPa using a new apparatus for accurate gas-density measurements, J Chem Thermodynamics, 23, 1991, pp 175 to 190 [6] GUO, X.Y., KLEINRAHM, R AND W AGNER, W Determination of the molar mass of two gas mixtures by means of accurate gas-density measurements on the 20° isotherm, Thermodynamics Department, Faculty of Mechanical Engineering at Bochum University, June 1995 [7] ISO 9001, Quality management systems — Requirements `,,```,,,,````-`-`,,`,,`,`,,` - 47 © ISO for 2006 – All rights reserved Copyright International Organization Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale `,,```,,,,````-`-`,,`,,`,`,,` - ISO/TR 24094:2006(E) ICS 75.060 Price based on 47 pages © ISO 2006 – All rights reserved Copyright International Organization for Standardization Provided by IHS under license with ISO No reproduction or networking permitted without license from IHS Not for Resale