ISO/TR 24094:2006

TECHNICAL ISO/TR
REPORT 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 2006
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 © ISO 2006 – All rights reserved

Contents Page
Foreword. iv
1 Scope . 1
2 Normative references . 1
3 Development of the validation methods. 2
4 Results of the VAMGAS project . 5
Annex A (informative) Report on the validation methods for gaseous reference materials . 6
Bibliography . 47

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 2.
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 © ISO 2006 – All rights reserved

TECHNICAL REPORT ISO/TR 24094:2006(E)

Analysis of natural gas — Validation methods for gaseous
reference materials
1 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.
2 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 6974-5, 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 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
3 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.
2 © ISO 2006 – All rights reserved

Figure 1 — Schematic concept of part 1 of the VAMGAS project
Figure 2 — Schematic concept of part 2 of the VAMGAS project
4 © ISO 2006 – All rights reserved

4 Results of the VAMGAS project
The project report provides results on two sets of comparisons.
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.
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, H , molar mass, M, and density at normal conditions of the mixtures were
s
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
[1] [2]
measurement of physical properties were reference calorimetry and precision densitometry . Statistically
significant agreement was found between the calculated data and the measurements.
Table A.1 — Comparison of experimental (M ) and calculated (M ) values
exp calc
a
of the molar mass for different PRMs
Gas mixture Type of gas M M Relative difference
exp calc
g/mol g/mol %
BAM 9605 4933 L 18,564 3 18,564 6 0,002
NMi 0602E L 18,542 7 18,543 0 0,002
BAM 9605 4902 H 18,793 1 18,796 6 0,018
NMi 9497C H 18,946 5 18,946 9 0,002
a
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 (ρ ) and calculated (ρ ) values
exp calc
a
of the gas density at standard conditions for different PRMs
Gas mixture Type of gas ρ ρ Relative difference
exp calc
3 3
kg/m kg/m %
BAM 9605 4933 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
a
Calculations are made in accordance with ISO 6976.
6 © ISO 2006 – All rights reserved

Table A.3 — Comparison of experimental (CV ) and calculated (CV ) values
exp calc
a
of the superior calorific value for different PRMs
Gas mixture Type of gas CV CV Relative difference
exp calc
MJ/kg MJ/kg %
BAM 9605 4933 L 44,061 44,068 0,015
NMi 0603E L 44,222 44,220 0,006
BAM 9605 4902 H 51,896 51,887 0,017
NMi 9498C H 51,910 51,895 0,03
a
Calculations are made in accordance with ISO 6976.

Table A.4 — Expanded uncertainties (95 % confidence interval) of the experimental reference values
and the calculated physical properties
Parameter Gas mixture
Type H Type L
relative % relative %
Calculated density 0,01 0,01
Measured density 0,015 0,015
Calculated molar mass 0,007 0,007
Measured molar mass 0,015 0,015
Calculated calorific value 0,1 0,1
Measured calorific value 0,035 0,035
For these gas samples, primary reference gas mixtures were once again produced. The composition of these
primary reference gas mixtures was selected so that they could be used for “bracketing calibration”. These
gas mixtures were used in a round-robin test series with a total of 18 participants from nine European
countries (see A.2.8.2). The test program was designed to ensure that the repeatability and comparability of
the results obtained by each individual participant could be determined by statistical methods with a view to
allowing an assessment of the uncertainty of all the individual results. Analytical results were transmitted as
raw data for uniform evaluation. Once again, the superior calorific value, molar mass and density at normal
conditions were calculated in accordance with ISO 6976.
The results of the round-robin test series are summarized in Table A.5:
Table A.5 — Comparison of the values of the physical properties calculated
from the mean of the 18 participating laboratories
with the values obtained from direct measurement by the reference methods
Parameter Type of gas Mean of the laboratories Reference method Relative difference
%
Calorific value, MJ/kg H 52,561 52,563 0,003
L 44,701 44,688 0,027
Molar mass, kg/kmol H 18,115 18,122 0,036
L 18,604 18,612 0,045
Density, kg/m H 0,7549 0,7551 0,034
L 0,7748 0,7752 0,048
A.1.2 Background
Chemical composition analysis represents a special case in the field of metrology as it is not possible to
ensure traceability to the SI unit “mole”. The objective is to avoid this problem by creating PRMs. PRMs
represent the best possible realization of the composition of a material. The primary reference gas mixtures
used in this project were produced by gravimetry, by successively weighing the individual pure components.
However, the significance of PRMs for chemical composition analysis is disputed because of the difficulty of
estimating cost, which is unsatisfactory for general use, and the often confusing terminology employed. In this
context, “traceability” means no more and no less than the statement of a result with documented uncertainty.
It is important not to confuse this quality target with the minimization of measurement error.
In view of the associated advantages, traceability is of very considerable importance for the European natural
gas industry, which operates a highly complex pipeline system with a comparatively large number of gas
compositions. Traceability becomes even more significant in the framework of the liberalized market. As the
value of gas supplied to a customer is calculated from the superior calorific value and volume flow measured,
measurement uncertainties have considerable financial impact.
This is why European gas companies have assumed the role of pioneers in this field, a role which is evident
from their participation in the ISO TC/193 and ISO TC/158 International Standards committees working on
traceability in natural gas analysis and gas analysis in general. Metrological institutes are also paying
increased attention to this requirement of their customers.
NOTE The participants in this joint research project under the leadership of Ruhrgas AG of Germany were Gasunie
of The Netherlands, Gaz de France of France, Nederlands Meetinstituut of The Netherlands, Bundesanstalt für
Materialforschung und -prüfung of Germany and Ofgem (previously: Ofgas) of the United Kingdom.
A.2 Material and methods
A.2.1 Primary reference gas mixtures
Primary reference gas mixtures (PRMs) are prepared by a gravimetric procedure as described in
ISO 6142 and are verified using the Dutch (NMi) or German (BAM) national primary standard gas mixtures
(PSMs). PRMs prepared by this method show the highest accuracy of gas standards and can be used as
calibration gases by the industry and calibration laboratories.
The production of a primary calibration gas mixture consists of a number of stages:
a) purity analysis of the starting components (pure gases) by FTIR, GC and MS methods;
b) gravimetric preparation of the gas mixture in passivated cylinders;
c) validation of the mixture using analytical methods to ensure that no errors have occurred during the
preparation process;
d) issue of the certificate.
A.2.1.1 Purity analysis of the starting components
The gases from which the mixture is prepared should be of known high purity and should preferably not be
contaminated by any of the other component gases that are to be part of the final mixture. The most accurate
method for determining purity is to quantify the impurities and to calculate the purity on a molar basis by
difference (purity is equal to 100 % minus the impurities). If high-purity starting gases are used, this means

−9 −9
that it is important that the concentrations of impurities be determined to at least the 10 × 10 to 1 000 × 10
mole fraction level in fairly pure gases. High-resolution Fourier transformation spectrophotometers equipped
with a gas cell of 100 m optical path length and several gas chromatographic methods (such as GC-MS, GC-
ECD, GC-DID) are available for carrying out these analyses.
8 © ISO 2006 – All rights reserved

Component purity and the associated uncertainty are estimated on the basis of estimates of impurity levels
and the uncertainty associated with these values. All the data obtained in this purity analysis are used in the
final calculation of the composition of the gas mixture prepared.
A.2.1.2 Gravimetric preparation of gas mixtures
For the preparation of a calibration gas mixture (see Figure A.1), a pre-treated aluminium cylinder with a mass
of approximately 8 kg is used. The cylinder is evacuated overnight using a turbo-molecular pump to achieve a
−6
vacuum of about 10 mbar. The gas remaining in the cylinder is usually the same as the matrix gas and,
therefore, makes a negligible contribution to the uncertainty in the composition of the final mixture.
Using a quadrupole mass spectrometer attached to the vacuum system, it is possible to analyse the
composition of the gas remaining in the evacuated cylinder. This is especially important when gas mixtures
with very low concentrations (nannolitres per litre levels) are prepared. In such cases, traces of moisture or
oxygen can cause instability of the final mixture.
The various high-purity gases are transferred to the sample cylinder in such a way that no (extra) impurities
are added from the materials used. For this purpose, a special assembly of electro-polished tubing, valves,
pressure and vacuum gauges and turbo molecular vacuum pumps with metal membranes is used.
To clean the system, the tubing connecting the sample cylinder to the starting cylinder is evacuated and
subsequently pressurized with the gas to be filled in. Experiments have shown that it is sufficient to repeat this
procedure eight times in order to remove all the contaminants present in the system. Since the system does
not include a compressor, the actual (vapour) pressure of the starting gases is used to pressurize the system.
If a refinery gas or natural gas mixture is prepared, the first component to be introduced to the cylinder is,
therefore, that with the lowest (vapour) pressure. Among other things, a compressor is not used, in order to
avoid possible contamination of the system with oil vapour or metal particles. For the same reason, the
vacuum system used consists of an oil-free membrane pre-vacuum pump in combination with a turbo-
molecular pump. After the tubing has been cleaned, an amount of the “pure” gas is added to the sample
cylinder in a controlled way using a fine metering valve. The amount of gas added to the sample cylinder is
monitored by placing the sample cylinder on a top weighing balance during the filling process.
This way of adding components to the cylinder allows considerable flexibility for the preparation of all kinds of
gas mixtures and results in very good target precision.
The precise mass of the gas introduced into the cylinder is determined by weighing the cylinder before and
after introduction of the component and comparing the weight of the sample cylinder several times with the
weight of a reference cylinder (in accordance with the Borda weighing scheme). Using a reference cylinder,
corrections for zero drift of the balance used, and influences of changing atmospheric conditions (temperature,
atmospheric pressure and humidity changes, which can cause a change in buoyancy) are minimized. The
mass comparison is performed on a 10 kg mass comparator with a resolution of 1 mg by calibrated mass
pieces. The typical uncertainty of mass determination is about 1,5 mg.
The traceability of gas composition to the SI system is ensured by using mass pieces directly calibrated
against the Dutch national standard of the kilogram.
After the mass determination of the first component, the sample cylinder is connected to the filling station
again for the introduction of the second component. This sequence of adding components and weighing of the
cylinder is repeated until all the components required have been introduced to the sample cylinder. The
introduction of large quantities of gas (e.g. matrix gas) to a cylinder results in a rise in the temperature of that
cylinder. As the difference in temperature between the sample cylinder and the reference cylinder has an
influence on weighing, it is necessary to observe a cool-down period. After the final component has been
added to the cylinder and the final weighing operation has been completed, the gas mixture, which now has a
pressure of about 10 MPa to 12 MPa (100 bar to 120 bar), is homogenized by rolling the cylinder for a few
hours.
The exact mixture composition and the associated uncertainties can be calculated from the data of the purity
analysis of the starting gases and the results of weighing. Typical uncertainties for minor components in the
mixture are of the order of 0,03 % (relative to the concentration). For components with high concentrations,
even lower uncertainties can be achieved.
Key
1 standing cylinder
2 needle valve and manometer
3 valve A
4 vacuum indicator
5 power supply
6 filter
7 vent
8 valve C
9 valve D
10 vacuum sensor
11 turbopump
12 vacuum pump
13 pressure sensor
14 pressure indicator
15 valve B
16 sample cylinder
Tubing
Electrical connections
Figure A.1 — Diagram of the gas filling station used for the preparation of PRMs
10 © ISO 2006 – All rights reserved

A.2.1.3 Validation of the gas mixture
Although the entire preparation procedure is defined and all uncertainty sources are identified and quantified,
the composition of the final mixture is verified to ensure that no errors have occurred during the preparation
process.
After the introduction of each component to the cylinder, the pressure of the cylinder is recorded and
compared with the calculated (predicted) pressure. However, this is a very rough method, which gives only a
preliminary indication of the reliability of mixture preparation.
A more accurate method for the validation of the gas mixture composition is the analysis of the mixture. For
analysis, a suitable analyser is selected and calibrated in the range of interest using primary standard gas
mixtures containing the same components as the mixture to be verified.
With appropriate PSMs for the calibration of the analyser, calibration curves can be calculated for each
component. The analysed concentration of a component in the freshly prepared mixture is determined using
the mathematical formula of a calibration curve and compared with the gravimetric concentration. If the
difference between these two values is larger than the uncertainties associated with these values, the gas
mixture is rejected and the entire preparation and verification cycle must be repeated
A.2.1.4 Issue of a certificate
After verification of the gravimetric data by analysis, the gas mixture is approved as a PRM and a certificate is
issued. This certificate includes information for the user of the calibration gas mixture such as the
concentrations and associated uncertainties, period of expected stability, information about side connections,
cylinder pressure, etc.
This certificate can also be used for demonstrating to accreditation organizations and trading partners that the
results of the measurements are traceable to accepted International Standards and are, therefore, accurate
and comparable with other measurements.
A.2.2 Preparation of compressed natural gas samples
A.2.2.1 Objective
The objective was to produce two sets of cylinders filled with compressed natural gas samples (min. 6 × type
L-gas and 6 × type H-gas) with identical compositions for use in the VAMGAS interlaboratory comparison.
11 type L and 11 type H natural gas samples were taken from the Ruhrgas pipeline system. The cylinders
were pressurized using an oil-free compressor. The sample gas was filtered through a molecular sieve filter
followed by a particulate filter to remove any residual moisture. To prevent contamination of the sample
cylinders with higher hydrocarbons and mineral oil, the cleanliness of the sampling system and the natural gas
stream were checked before sampling by performing extended hydrocarbon analyses up to C . The
condensation behaviour of the natural gas samples was calculated on the basis of the combined natural gas
and higher hydrocarbon analysis to ensure that no condensation occurred inside the cylinder.
A.2.2.2 Cylinder preparation
1)
The 10 l aluminium cylinders [Luxfer ] equipped with stainless steel valves were purchased from Messer-
2)
Griesheim . The cylinders were cleaned, heat-treated and filled with dry nitrogen upon delivery. The cylinder
contents were initially homogenized by rolling and heating for 6 h each and afterwards checked for residual
moisture by a routine Karl-Fischer method. Moisture was always found to be below the detection limit of
3 −3
0,01 mg/m in the gas phase. The cylinders were evacuated to < 0,1 Pa (10 mbar) (Leybold Thermovac)

1) Luxfer is an example of a suitable product available commercially. This information is given for the convenience of
users of this part of this International Standard and does not constitute an endorsement by ISO of this product.
2) Messer-Griesheim is an example of a suitable supplier. This information is given for the convenience of users of this
part of this International Standard and does not constitute an endorsement by ISO.
−2 -4
3)
using a vacuum pump [VacuuBrand RZ8 ; p < 4 × 10 Pa (4 × 10 mbar)]. Back diffusion of oil from the
vacuum pump was prevented by a trap cooled by liquid nitrogen. During evacuation, the cylinders were
4)
heated to 60 °C by a jacket heater [Isopad ]. For matrix conditioning, the cylinders were filled with high-purity
methane (Messer Griesheim) up to a pressure of approximately 200 kPa (2 bar), homogenized, heated and
evacuated again. Afterwards, the cylinders were shipped to the sampling site. Sampling was performed within
five days after evacuation.
A.2.2.3 Sampling system and method
The main component of the sampling system, which was assembled in-house, was an oil-free high-pressure
5)
compressor [Desgranges & Huot , p = 50 MPa (500 bar)], which was used to increase the cylinder
max
pressure by [10 MPa to 12 MPa (100 bar to 120 bar)] above pipeline pressure [4 MPa to 6 MPa (40 bar to
60 bar)].
The gas cylinders were connected to a closed loop made of pre-cleaned stainless steel tubing using
6) 7)
Swagelock tees. Purpose-built cylinder connectors with flush lines [Hage ], which protruded into the interior
of the cylinder valves, allowed this dead volume to be flushed with sample gas. The gas was sampled through
8)
two large-volume high-pressure filters. The first filter was filled with 1,5 kg of molecular sieve 3A [Fluka ], the
9)
second was used as a particulate filter [Headline filters efficiency > 99,9 % for particulates > 0,1 µm].
10)
Flexible tubing with Minimess connectors was used to connect the sampling system to the sampling station
and the cylinder arrangement.
The sampling system, all connecting lines and the cylinder valves were extensively flushed with sample gas.
After a leak check had been performed, the valve connecting the cylinder arrangement to the sampling system
was closed and the cylinder valves were opened. The valve was opened slowly and the cylinders were
pressurized by slowly increasing the back-pressure of the sampling system. When the pipeline pressure was
reached, the pressure booster was started up automatically. The final pressure was set to 10 MPa (100 bar).
The entire sampling procedure took approximately eight hours.
The cylinders were homogenized twice after sampling, by heating and rolling.
A.2.2.4 Sampling sites
Sampling sites were located at Dorsten (type L gas from the Groningen field) and Krummhörn (type H gas
from the Ekofisk field). Both sites are located in Germany. The sampling system was connected to sampling
units that are also used for custody transfer measurements. These units are, therefore, continuously flushed
with fresh gas and can be considered clean.

3) VacuuBrand RZ8 is an example of a suitable product available commercially. This information is given for the
convenience of users of this part of this International Standard and does not constitute an endorsement by ISO of this
product.
4) Isopad is an example of a suitable product available commercially. This information is given for the convenience of
users of this part of this International Standard and does not constitute an endorsement by ISO of this product.
5) The Desgranges & Huot compressor is an example of a suitable product available commercially. This information is
given for the convenience of users of this part of this International Standard and does not constitute an endorsement by
ISO of this product.
6) Swagelock is an example of a suitable product available commercially. This information is given for the convenience
of users of this part of this International Standard and does not constitute an endorsement by ISO of this product.
7) Hage is an example of a suitable product available commercially. This information is given for the convenience of
users of this part of this International Standard and does not constitute an endorsement by ISO of this product.
8) Fluka is an example of a suitable product available commercially. This information is given for the convenience of
users of this part of this International Standard and does not constitute an endorsement by ISO of this product.
9) Headline is an example of a suitable supplier. This information is given for the convenience of users of this part of this
International Standard and does not constitute an endorsement by ISO.
10) Minimess is an example of a suitable product available commercially. This information is given for the convenience of
users of this part of this International Standard and does not constitute an endorsement by ISO of this product.
12 © ISO 2006 – All rights reserved

A.2.2.5 Additional quality assurance measures
The natural gas streams were checked for higher hydrocarbons using analytical methods developed by the
Ruhrgas laboratory. The cleanliness of the sampling system was also ensured by this method.
The analysis method involves the use of a stainless steel trap (80 cm × 1,2 cm OD) filled with 40 g dehydrated
silica gel 50 Ǻ (200 mesh to 500 mesh). Hydrocarbons above C are quantitatively trapped on the solid
sorbent. Breakthrough volumes were shown to be > 1,2 m (N). Approximately 400 l (N) of natural gas were
sampled. The trapped components were eluted using 400 ml of a mixture of CH Cl and pentane (60:40 by
2 2
volume). The eluate was then concentrated to 10 ml and analysed by GC/FID. The detection limit of the
method is 0,001 mg/m (N) per component. Hydrocarbons up to C can be detected.
A routine natural gas analysis for components up to C was also performed on this occasion. The results of
the natural gas analysis and the higher hydrocarbon analysis up to C were combined for calculating the
11)
phase behaviour of the natural gas stream. The calculation was performed using the Hysim commercial
software package. A modified Soave-Redlich-Kwong equation of state was selected for calculation. The
calculated phase envelope of the type H gas is shown in Figure A.2.
The tests were performed before sampling in order to ensure that the sampling cylinders were not
contaminated with oil and no liquid phase was formed inside the cylinder.

Key
X temperature, expressed in degrees Celsius
Y pressure, expressed in kilopascals
1 cricondenbar
2 cricondentherm
NOTE Higher hydrocarbon content up to C are included in the calculation.
Figure A.2 — Phase envelope of the type H natural gas samples calculated
by a Soave-Redlich-Kwong equation of state

11) Hysim is an example of a suitable product available commercially. This information is given for the convenience of
users of this part of this International Standard and does not constitute an endorsement by ISO of this product.
A.2.2.6 Certification of the cylinders
One cylinder of each sample batch was investigated by precision densitometry. The procedure corresponded
exactly to the procedure followed when measuring the reference gases. 16 single measurements were
performed for each gas and fitted to the virial equation (Equation A.2). The uncertainty of the fit was not
increased compared with the reference gases, a good indication of mixture homogeneity.
The measurement results are listed in Table A.6.
Table A.6 — Measurement results from the density balance for the natural gas samples
Sample Molar mass Density
g/mol [at 20 °C, 101,325 kPa (1,013 25 bar)]
Type L gas, cylinder 3590 1,612 0,775 2
Type H gas, cylinder 3592 18,122 0,755 1
A.2.2.7 Homogeneity and stability tests
Homogeneity tests were first performed by routine GC analysis. All cylinders were analysed twice under
repeatability conditions. The analysis runs were performed uncalibrated (only a check sample was analysed
together with the samples), since the task was to detect differences between the sample cylinders and there
was insufficient time for calibration runs (11 cylinders had to be analysed on the same day). The statistical
analysis indicated no detectable differences between the cylinders.
For six cylinders of each batch, the gas density was then measured by precision densitometry at
approximately 1,5 MPa (15 bar) and 3 MPa (30 bar), with four individual measurement points per cylinder. The
measurements show satisfactory agreement to within ± 0,003 %.
Sample stability testing was performed by repeating the densitometric measurements after approximately six
months storage. The results of these repeated tests were the same as those obtained during the first analysis.
Finally one of the cylinders of each batch was completely consumed and gas density was determined at
different pressure levels. This test also gave no indication of any change in gas composition. It can, therefore,
be assumed that the natural gas samples are stable.
A.2.3 Reference calorimeter
A.2.3.1 Design
The reference calorimeter at the Technical Directorate of the Office of Gas Supply (Ofgas) was designed to be
the primary standard for determining the heat of combustion of natural gas samples. The instrument is based
on that used by Pittam and Pilcher at the University of Manchester in the late 1960s for studying the heat of
[1], [2]
combustion of methane and other hydrocarbons . This instrument was, in turn, based on that built by
Rossini at the National Bureau of Standards in the USA in the early 1930s for studying the heat of formation of
[3] [4]
water and the heats of combustion of methane and carbon monoxide . The Ofgas reference calorimeter
was constructed and refined by Mr C. Lythall. The experiment produces a superior heat of combustion in
kilojoules per gram, at a constant pressure, for combustion at 25 °C.
There have been three major changes from the designs of previous workers:
a) The sample of gas burnt is weighed directly.
b) The experiment is controlled and data are collected automatically by computer.
c) Measurements are made at a faster rate.
14 © ISO 2006 – All rights reserved

A.2.3.2 Calorimeter Theory
The objective of the Ofgas reference calorimeter is to measure the quantity of energy liberated in the complete
combustion of a hydrocarbon fuel gas. This is achieved by allowing the energy liberated in the reaction to be
transferred to a well-stirred liquid, in a calorimeter, and measuring its temperature rise. Multiplying this
temperature rise by the energy equivalent of the calorimeter gives the amount of energy liberated in the
reaction. The energy equivalent is the energy required to raise the temperature of the calorimeter by one
degree Celsius at the same mean temperature as the combustion experiment. It is determined by electrical
calibration experiments.
An ideal calorimeter would be thermally isolated from its environment so that the temperature change
observed is due solely to the reaction. As complete isolation from the environment is not possible in practice,
a calorimeter is usually surrounded by a thermostatically controlled jacket and allowance is made for the
various energy sources and sinks. The reference calorimeter is designed as an isoperibolic instrument.
There are three external influences and they are all sources of energy:
a) the water stirrer;
b) the self-heating of the temperature-measuring device;
c) energy flowing from the jacket to the calorimeter as a result of the temperature difference.
Figure A.3 shows a temperature versus time curve for a typical experiment (combustion or calibration). Data
collection starts at a predetermined temperature.
...