Nowadays only a very small quantity of hydrogen can be found in the gas pipes at the moment (ppm order of magnitude). Yet, those quantities are going to increase drastically in the mid term with the emergence of “power to gas”. Hydrogen concentrations at ppm levels have no impact on the measurement of the calorific value of natural gas, but they have to be taken into consideration if they become more present in the natural gas grid. As of today, in the gas quality prescriptions of the French transmission system operator, a gas mixture containing up to 6% of H2 can be injected into the gas grid. At the moment, hydrogen is not being measured continuously by the online gas chromatographs installed on the field for billing purposes in France and ENGIE Lab CRIGEN was interested to see what exactly was the error in the calorific value measurement if hydrogen was not taken into account in the normalized composition of natural gas, or if it was not taken into account in the calculation of the physical and chemical properties of the gas. After a theoretical study on the subject, presented at GAS2015, ENGIE Lab CRIGEN decided in 2016 to corroborate this study with an experimental trial. Two process GCs, also used on the gas grid, were tested in the CRIGEN laboratories with binary mixtures (H2 in methane), tertiary mixtures (H2/N2 in methane) and synthetic natural gas containing 6% H2. For all the mixtures analysed, the measured GCVs and their uncertainties were compared to the theoretical ones obtained with the equation published in ISO 6976 : 2015. It turns out that the sensitivity of the process GCs for hydrogen is lower than that was expected in the theoretical study and as a result the difference between the measured GCV and the theoretical one is higher than previously calculated. Pushing the reasoning even further, ENGIE Lab CRIGEN questioned the viability of the ISO 6976 equation for gases containing high levels of hydrogen. After all, the state equation was validated only for all the current natural gases transiting through Europe and ENGIE Lab CRIGEN wanted to make sure that the equation would respond correctly to other types of gases. To demonstrate this, the CRIGEN used a calorimeter (supposed to give the “true GCV value”) and compared its measurement and its uncertainty to the calorific value calculation obtained via ISO 6976 : 2015 equation.
Beatrice is a research engineer at the Gas Analysis and Metering Section of ENGIE LAB CRIGEN (ENGIE) in Saint-Denis (France).
Beatrice has a master’s degree in analytical chemistry. She has worked in the field of analytical chemistry for 17 years, first in the pharmaceutical industry (SGS Life Science Services, quality control services) then in the gas industry (ENGIE). At the ENGIE Lab CRIGEN (ENGIE R&D centre for gas and new energies ) for the past 2 years, she is involved in the assessment of on-line and laboratory devices for Gas Quality Analysis (H2S, COS, mercaptans and tetrahydrothiophene monitoring, GCV measurement).