DOI: 10.3303/CET2290061 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Paper Received: 22 November 2021; Revised: 9 March 2022; Accepted: 14 May 2022 
Please cite this article as: Eckart S., Pio G., Krause H., Salzano E., 2022, Chemical and thermal effects of trace components in hydrogen rich 
gases on combustion, Chemical Engineering Transactions, 90, 361-366  DOI:10.3303/CET2290061 
  

 CHEMICAL ENGINEERING TRANSACTIONS  
 

VOL. 90, 2022 

A publication of 

 
The Italian Association 

of Chemical Engineering 
Online at www.cetjournal.it 

Guest Editors: Aleš Bernatík, Bruno Fabiano 
Copyright © 2022, AIDIC Servizi S.r.l. 
ISBN 978-88-95608-88-4; ISSN 2283-9216 

Chemical and Thermal Effects of Trace Components in 
Hydrogen Rich Gases on Combustion  

Sven Eckarta, Gianmaria Piob, Hartmut Krausea, and Ernesto Salzanob,*  
a Institute of Thermal Engineering, TU Bergakademie Freiberg, Freiberg, Germany  
b Department of Civil, Chemical, Environmental and Materials Engineering, University of Bologna, Italy 
ernesto.salzano@unibo.it 

The production of carbon-neutral fuels through clean energy has been defined as a target by the European 
Union and by several international institutions. If the concepts are available, hydrogen, in particular, is 
considered to be one of the most target-oriented, ecologically and economically realizable approaches. In terms 
of safety, long-time storage and long-distance transport of hydrogen are still under development. However, 
pipeline systems similar to those for natural gas are being considered. Gas quality criteria will have to be 
developed for this case. The effects of trace components in the hydrogen on chemical and thermal aspects are 
not yet sufficiently understood and need to be characterized more precisely. For these reasons, this work 
presents a detailed analysis for a more complete understanding of the phenomena involved. More specifically, 
the flame structure, temperature profile and overall reactivity were first determined for gas mixtures analyzed 
under four varying dilutions of carbon dioxide, carbon oxide, nitrogen and methane in hydrogen. The 
characterization of the total reactivity and the laminar burning velocity offers an appealing solution to quantify 
the effects of dilution. The most distinctive effects of the operating conditions on the ignition phenomena have 
been worked out numerically for the lower and upper boundaries and have been discussed. The results collected 
in this work provide a robust feature for a detailed evaluation of normal operation as well as the accidental 
release of the hydrogen-rich fuels. 

1. Introduction 
In the future, carbon-free fuels such as hydrogen and ammonia will play a major role in energy production. As 
a compressible gas, hydrogen can have a significant role in supplying Europe because of the actual 
infrastructures, although its different properties with respect to the mixture currently employed. For this purpose, 
the existing gas transport grid will have to be adapted. On the one hand, existing pipelines will have to be 
prepared for the new challenges, and on the other hand, they will have to meet the high safety standards and 
be able to supply the end customers with a consistently high quality of fuel. For this reason, international 
regulations are being established to control the changes in the gas network for 100 % hydrogen as well. 
Currently, in Europe, both in France and in Germany, 2 vol.-% dilution in hydrogen is under discussion, whereas 
worldwide also 5 vol.-% are assumed. This change in trace gas concentration is to be investigated about its 
influence on fundamental combustion properties. Among the others, the laminar burning velocity offers the 
possibility to disregard turbulence by physic-chemical phenomena. Indeed, this parameter is commonly adopted 
for the evaluation of the overall reactivity, severity of explosions, and characterization of accidental scenarios in 
computational fluid dynamics (Skjold et al., 2018). Several approaches can be used for the evaluation of this 
fundamental parameter, including either experimental or numerical strategies. As a way of example, counter 
flow flame (Eckart et al., 2021c), heat flux burner (Eckart et al., 2017), or spherical vessel (Eckart et al., 2021b) 
can be used to measure it. Additional information on practical and theoretical aspects involving these systems 
can be found elsewhere (Konnov et al., 2018). However, it is worth reporting that the heat flux burner represents 
the only option where extrapolation is not needed to obtain stretch-free data, with a considerable reduction in 
uncertainties associated with the measurements (van den Schoor et al., 2008). Indeed, the heat flux burner has 
been largely adopted to measure the laminar burning velocity of different fuels, including hydrogen, under a 

361



wide range of conditions (Alekseev and Konnov, 2018). On the other hand, the numerical estimation of laminar 
burning velocity mostly relies on the accuracy of detailed kinetic mechanisms, usually consisting of hundreds of 
species and thousands of reactions (Egolfopoulos et al., 2014). The strategies implemented for the genesis of 
a detailed kinetic mechanism can variate by several aspects. Indeed, these models can be produced by 
automated or expert-driven procedures or can use different target species, sources for thermodynamic and 
kinetic parameters (Curran, 2019). Besides, an additional source of differences in kinetic mechanisms is 
represented by the approach adopted for the realization of transport models. Among the others, the Soret 
diffusion affects the mass burning rate also in the case of mono-dimensional flames (Zhou et al., 2017), 
especially if hydrogen is directly involved in the unburned mixture (Law, 2010). Each possible approach can 
potentially generate different mechanisms regarding accuracy and size, suggesting the comparison with 
experimental data for the sake of validation. 
In this work, both the influence on the burning velocity and the Wobbe number are investigated in more detail. 
Finally, the findings will be substantiated numerically to shed more light on the chemical changes. 

2. Methodology 
2.1 Theoretical background 

The trace gases carbon dioxide, nitrogen, carbon monoxide, and methane were considered in the studies. In 
each case, 0 to 5 vol.-% were added to hydrogen. The mixture compositions are summarized in Table 1.  
 

Table 1. Mixture composition of the trace gases in pure hydrogen at p = 1 bar and T = 298 K. 
CO [%v/v] CO2 [%v/v] CH4 [%v/v] N2 [%v/v] 

1-5 0 0 0 
0 1-5 0 0 
0 0 1-5 0 
0 0 0 1-5 

 
Further, the Wobbe number is a gradiometer of the interchangeability of gases and is frequently defined in the 
specifications of gas supply, transport utilities and heat load of fuel gas consumers with different gas 
compositions. The Wobbe number expresses both the energy content and the flow velocity-determining 
variables (density and pressure loss). It is the central key figure in several international technical rules. Gases 
with the same Wobbe index can be exchanged without design changes to the burner or conversion of the burner 
nozzle. The fluid mechanical principles are derived from the Bernoulli equation for the pressure drop across a 
burner nozzle which discharges the gas from a line into a room with ambient pressure. The pressure drop ∆p 
across the burner nozzle corresponds in a first approximation to the line overpressure pl. Via the continuity law, 
the ratio of the velocities (u) in line and nozzle can be derived from the diameter (d) ratio. 
𝜋𝜋
4
𝑑𝑑𝑙𝑙
2 ∙ 𝜌𝜌𝑔𝑔 ∙ 𝑢𝑢𝑙𝑙 =

𝜋𝜋
4
𝑑𝑑𝑑𝑑
2 ∙ 𝜌𝜌𝑔𝑔 ∙ 𝑢𝑢𝑑𝑑 (1) 

 
From this follows for the pressure drop  

∆𝑝𝑝 =
𝜌𝜌𝑔𝑔
2
∙ 𝑢𝑢𝑑𝑑

2 ∙ �1 − 𝑑𝑑𝑑𝑑
4

𝑑𝑑𝑙𝑙
4� (2) 

 
Thus, if the general conditions at the burner nozzle (pressure drop as well as line and nozzle diameter) remain 
unchanged, the exit velocity inevitably increases as the fuel gas density decreases. The following then applies 
to the heat load �̇�𝑄𝐵𝐵 of a burner or end device.  

�̇�𝑄𝐵𝐵 ≈ 𝛼𝛼
𝜋𝜋
4
𝑑𝑑𝑑𝑑
2 ∙ �

2∆𝑝𝑝
𝜌𝜌𝑔𝑔

∙ 𝐻𝐻𝑠𝑠 = 𝐶𝐶 ∙ �∆𝑝𝑝 ∙
𝐻𝐻𝑠𝑠
�𝜌𝜌𝑔𝑔

 (3) 

 
With the introduction of the density ratio k, the Wobbe index is obtained as a parameter: 

�̇�𝑄𝐵𝐵 ≈
𝐶𝐶

�𝜌𝜌𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙
∙ �∆𝑝𝑝 ∙

𝐻𝐻𝑠𝑠

�
𝜌𝜌𝑔𝑔

𝜌𝜌𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙

= 𝐶𝐶∗ ∙ �∆𝑝𝑝 ∙
𝐻𝐻𝑠𝑠
√𝑘𝑘

= 𝐶𝐶∗ ∙ �∆𝑝𝑝 ∙ 𝑊𝑊𝑠𝑠        or       �̇�𝑄𝐵𝐵~�∆𝑝𝑝 ∙ 𝑊𝑊𝑠𝑠 (4) 

362



As an option for the technical adaptation of gas appliances (burners) to a gas type, the nozzle inlet pressure 
(pressure drop ∆p) can be used, from which the Extended Wobbe Index �∆𝑝𝑝 ∙ 𝑊𝑊𝑠𝑠 is then derived.  

2.2 Numerical Approach 

In this work, laminar burning velocity was estimated through several detailed kinetic mechanisms existing in the 
current literature under different initial conditions and gaseous compositions of interest for the investigated 
system and representative of the experimental data collected. 
The freely propagating, steady and one-dimensional flames were investigated in a doubly infinite domain using 
the premixed laminar flame-speed code of the Chemkin-PRO 2020R2 software. Using the multi-component 
transport coefficients and thermal diffusion option the LBV the final results were calculated. A typical number of 
grid points of around 1000 with respect to the gradient and curvature adaptive mesh parameters of 
GRAD = 0.02-0.04 and CURV = 0.05-0.08 were used. Further details are explained in Eckart et al. 2021 (Eckart 
et al., 2021a). A shortlist of kinetic mechanisms considered in this work is reported in Table 2. 
 

Table 2. Kinetic mechanisms used in this study. 
Model reactions species source 
Glarborg 1407 154 Glarborg et al., 2018 
USC II 784 111 Wang et al., 2007 
Shrestha  1125 129 Shrestha et al., 2018 
KiBo 321 73 Pio et al., 2019 
GRI 3.0 325 53 Smith et al., 2000 

 
More specifically, the detailed kinetic mechanism developed by the Gas Research Institute, GRI 3.0 (Smith et 
al., 2000), was adopted since it has been largely considered as a benchmark for the description of the main 
combustion-related phenomena of light hydrocarbons. Besides, the detailed kinetic model developed at the 
University of Bologna (KIBO) was adopted since previous studies have indicated KIBO as a suitable mechanism 
for light hydrocarbons at the investigated conditions (Pio et al., 2019). Further, for comparison the Shrestha 
mechanism (Shrestha et al., 2018), Glarborg mechanism (Glarborg et al., 2018), and the USC 2016 (Wang et 
al., 2007) were included, as well. Eventually, the effect of different approaches for the generation of transport 
models were compared, as well.  

3. Results and discussion 
A comparison of the experimental and numerical laminar burning velocity of pure hydrogen, obtained at 298 K 
and 1 bar, is presented in Figure 1. Figure 1 shows that for pure hydrogen combustion, the measured data for 
LBV scatter significantly; this was also discussed extensively in Konnov et al. (Konnov et al., 2018). In the upper 
graph, a 95 % confidence interval was entered. The simulations with the three different transport mechanisms 
can predict within this interval. Nevertheless, the complex multi-component approach considering the Soret 
effect for thermal diffusion gives the best results and was used for the comparison in the lower graph for all 
mechanisms. Although the detailed reaction mechanism for hydrogen-air flames among each other shows a 
variation in the range up to 40 cm/s, this is comparable to the variations in the experimental data (Konnov et al., 
2018). The maximum deviation between the predictions of the mechanisms is 20 %. This should be considered 
for further evaluation and improvements. As well-known, the hydrogen is highly reactive with respect to the 
given conditions, as confirmed by higher values of LBV at an equivalence ratio ϕ, defined as the fuel to oxygen 
ratio in feeding stream to the ratio of the stoichiometric coefficients of fuel to oxygen, higher than 1.0 
(stoichiometry). Thereby, the Glarborg mechanism shows higher values over the whole range than the other 
four. The USC II from ϕ > 0.8 shows the lowest values. The KIBO mechanism shows over a wide range a good 
average value between these two and we will show for the following considerations. 

363



Figure 1 Comparison of experimental data (Konnov et al., 2018) for 100 % hydrogen with numerical results 
of different reaction mechanisms, Top: comparison mixed average, mixed average+ Soret effect and multi-
component + Soret effect, Bottom: compare the mechanisms USC II, Shrestha, GRI 3.0, Glarborg and KIBO 
for T=298 K and 1 bar. 

For future studies, this range will gain additional high importance, since traces of other gases from 2 vol.-% to 
5 vol.-% could be allowed in pure hydrogen networks. Both traces of fuels such as propane, butane or methane 
are conceivable, but nitrogen or carbon dioxide from biogenic fuel gas production. This would have a 
considerable influence on the burning velocity, which is expected to be much broader than the admixture up to 
50 %. No experimental observations are currently known on this subject. Numerically, the changes in burning 
velocity were plotted in Figure 2.  

Figure 2. Change in LBV resulting from changes in gas composition, based on a hydrogen flame 

364



It can be seen that CH4 has the largest effect on LBV followed by CO2, N2 and CO. Methane changes the trend 
of LBV above the equivalence ratio in this case, but the other trace gases do not. As expected, the addition of 
trace gases influences the laminar burning velocity of hydrogen flames. As shown in Figure 2, CO2 and N2 as 
inert gases as well as CO as fuel do not influence the global trend over the equivalence ratio, whereas methane 
leads to significant changes.  
For the admixture of 5 vol.-% methane, there is already a clear reduction in the burning velocity in the rich 
region, which leads to a shift of the point of maximum burning velocity to the region around ϕ = 1.3. It should be 
noted that the maximum point for methane is around ϕ = 1.1 (Konnov et al., 2018). The slope of the curve is 
also more consistent with the common slope for hydrocarbon flames. Likewise, the maximum velocity is also 
significantly reduced by methane to approx. 220 cm/s. Which still corresponds to factor 6 of the maximum 
velocity of methane. The operating map of a gas spans between the Wobbe number and the calorific value. 
This map defines the permissible limits for the distributed gases in an economic area. The characteristic diagram 
is shown in Figure 3.  

Figure 3 Change in Wobbe number due to changes in gas composition, based on natural gas high 

It includes both the fluctuation ranges of typical natural gas sources from the European environment and the 
characteristic curves for the admixture of hydrogen up to 100 % by volume. Depending on the type of base gas, 
the mixing characteristic curve covers a wide Wobbe number range. For pure hydrogen, however, the 
characteristic curve reaches the lower Wobbe number limit of the natural gas H range with a residual deviation 
from the base gas G 20 (methane) of -9.6 %. For mixtures with a typical natural gas L, there is a deviation from 
the base gas (natural gas L-Holland) of +5.4 %. For the sake of completeness, the hydrogen gas type A from 
ISO/DIS 14687:2018 (ISO/DIS 14687, 2018) with permissible inert impurities of up to 2 vol.-% shall also be 
included in the analysis. Since the effects of trace components have a significantly different density than 
hydrogen, there is also a non-negligible fluctuation range for this case. 

4. Conclusions
The limits of hydrogen and mixtures with minor rarefactions of nitrogen, methane and carbon dioxide were 
calculated using a numerical method based on a detailed kinetic model. The numerical model was validated 
against experimental data for pure hydrogen to provide reliable predictions for the dilution cases. 
It was shown that the KIBO mechanism, using the MC approach and the Soret effect, was able to reproduce 
the LBV values for hydrogen-air flames very well. Subsequently, it was shown that all 4 admixtures reduce the 
LBV. Significant changes in the trend above the equivalence ratio were only seen for CH4, whereas in the case 
of CO, CO2 and N2 there was only a reduction. It should be taken into account that the study carried out is 

365



strongly affected by the lack of experimental LBV data in the conditions studied. This could be a scientific 
objective of further interest. Furthermore, the shift in the Wobbe number was calculated for all 4 trace 
compounds. There are clear shifts in the upper Wobbe number towards lower values due to the addition of trace 
gases. Here, CO2 has the greatest influence and CO the least. However, the value for 2 vol.-% CO2 does not 
leave the range of the current natural gas low at below 11 kWh/m³. 

References 

Alekseev V.A., Konnov A.A., 2018, Data consistency of the burning velocity measurements using the heat flux 
method: Hydrogen flames, Combustion and Flame, 194, 28–36. 

Curran H.J., 2019, Developing detailed chemical kinetic mechanisms for fuel combustion, Proceedings of the 
Combustion Institute. International Symposium on Combustion, 37, 57–81. 

Eckart S., Cai L., Fritsche C., vom Lehn F., Pitsch H., Krause H., 2021a, Laminar burning velocities, CO and 
NOx emissions of premixed polyoxymethylene dimethyl ether flames, Fuel. 

Eckart S., Penke C., Voss S., Krause H., 2017, Laminar burning velocities of low calorific and hydrogen 
containing fuel blends, Energy Procedia, 149–156. 

Eckart S., Pizzuti L., Fritsche C., Krause H., 2021b, Experimental study and proposed power correlation for 
laminar burning velocity of hydrogen-diluted methane with respect to pressure and temperature variation 
investigated with multiple burner setups, submitted to International Journal of Hydrogen Energy. 

Eckart S., Yu C., Mass U., Krause H., 2021c, Extinction limits in non-premixed hydrocarbon counterflow 
flames in an oxygenreduced environment, Fuel, 120781. 

Egolfopoulos F.N., Hansen N., Ju Y., Kohse-Höinghaus K., Law C.K., Qi F., 2014, Advances and challenges 
in laminar flame experiments and implications for combustion chemistry, Progress in Energy and 
Combustion Science, 43, 36–67. 

Glarborg P., Miller J.A., Ruscic B., Klippenstein S.J., 2018, Modeling nitrogen chemistry in combustion, 
Progress in Energy and Combustion Science, 67, 31–68. 

ISO/DIS 14687, 2018, Hydrogen fuel quality - Product specification. 
Konnov A.A., Mohammad A., Kishore V.R., Kim N.I., Prathap C., Kumar S., 2018, A comprehensive review of 

measurements and data analysis of laminar burning velocities for various fuel+air mixtures, Progress in 
Energy and Combustion Science, 68, 197–267. 

Law C.K., 2010, Combustion physics, Cambridge University Press, Cambridge. 
Pio G., Barba D., Palma V., Salzano E., 2019, A Numerical Study on the Effect of Temperature and 

Composition on the Flammability of Methane–Hydrogen Sulfide Mixtures, Combustion Science and 
Technology, 191, 1541–1557. 

Shrestha K.P., Seidel L., Zeuch T., Mauss F., 2018, Detailed kinetic mechanism for the oxidation of ammonia 
including the formation and reduction of nitrogen oxides, Energy & Fuels, 32, 10202–10217. 

Skjold T., Souprayen C., Dorofeev S., 2018, Fires and explosions, Progress in Energy and Combustion 
Science, 64, 2–3. 

Smith G.P., Bowman T., Frenklach M., Golden G.M., Moriarty N., Eiteneer B., Goldenberg M., Hansan R., 
Song S., Gardiner W., Lissianski V., Qin Z., 2000, GRI-MECH 3.0 http://combustion.berkeley.edu/gri-
mech/. 

van den Schoor F., Hermanns R.T.E., van Oijen J.A., Verplaetsen F., Goey L.P.H. de, 2008, Comparison and 
evaluation of methods for the determination of flammability limits, applied to methane/hydrogen/air 
mixtures, Journal of Hazardous materials, 150, 573–581. 

Wang H., You X., Joshi V., Davis G., Laskin A., Egolfopoulos F., Law C.K., 2007, USC Mech Version II. High-
Temperature Combustion Reaction Model of H2/CO/C1-C4 Compounds., 
http://ignis.usc.edu/USC_Mech_II.htm. 

Zhou Z., Hernández-Pérez F.E., Shoshin Y., van Oijen J.A., Goey L.P. de, 2017, Effect of Soret diffusion on 
lean hydrogen/air flames at normal and elevated pressure and temperature, Combustion Theory and 
Modelling, 21, 879–896. 

366


	lp-2022-abstract-092.pdf
	Chemical and Thermal Effects of Trace Components in Hydrogen Rich Gases on Combustion
	References