Abstract
Energy strategies at the international level increasingly support the utilization of hydrogen for energy purposes. One way to do this is to mix hydrogen into natural gas, which is delivered by the network to the combustion equipment of consumers. The paper examines the expected changes that will occur if the maximum amount of hydrogen permitted by law is mixed into the natural gas network. According to our results, the inflowing heat quantity in the gas has decreased, which is compensated by the increased flow rate due to the reduced density. The flame image changes spectacularly, the flame becomes lower, the half-cone angle increases in case of hydrogen mixing. Another noteworthy result is that the temperature of the equipment's burner did not change significantly as a result of mixing.
1 Introduction
The paper examines the effects of mixing hydrogen into natural gas, especially with regard to household combustion equipment. The effects of mixing hydrogen were also dealt with by other research groups, several researches also focused on household combustion devices.
X. Huang et al. [1] according to his experiment, the flame height in the equipment he tested will decrease as a result of hydrogen mixing, and he also states that there is a risk of back-ignition at a hydrogen mixing ratio of 23%, so he closed his experiments around this target value (25%) for safety reasons. Their measurements also showed that as a result of mixing in hydrogen, although the nominal heat input decreases, the overall thermal efficiency increases. Similar results were obtained in their publication by Zhao et al. [2], according to whom the addition of hydrogen also results in an increase in thermal efficiency. Zhao et al. [2] performed their measurements with a lower volume mixing of 20% maximum. They also highlight the safety risks of back-ignition; however, they indicate a much higher limit value, 55% volume ratio, in the most critical case, measured with a free jet flame loaded with a cooking pot. Another important result is that there was no major change in the heating time, although expectations showed this due to the lower calorific value of hydrogen per volume unit. Measurements were also carried out in the same place, which investigate the effect of mixing hydrogen on the temperature of gas burners. According to their experience, as a result of 75% mixing, the burner temperature increased by 25 °C, which is a remarkable result, compared to the result that the thermal efficiency increases in the meantime (so a relatively high hydrogen mixing ratio, the temperature increase of 25 °C under a reactive combustion medium is considered a relatively low value). In order to be able to interpret the results more easily, it is worth comparing the properties of methane gas, which is the basis of natural gas, and hydrogen gas, which are more important in terms of combustion. These are summarized in Table 1.
Comparison of the most important properties of hydrogen and methane gas, Higher Heating Value (HHV), Lower Heating Value (LHV)
| Methane | Hydrogen | |
| HHV (MJ kg−1) [3] | 55.8 | 141.90 |
| LHV (MJ kg−1) [3] | 50.0 | 119.90 |
| HHV (MJ m−3) [3] | 39.8 | 12.70 |
| LHV (MJ m−3) [3] | 35.8 | 10.80 |
| Density (kg m−3) (at 0 °C, 1 bar) [4] | 0.7 | 0.10 |
| Lower explosion limit (%) [4] | 5.0 | 4.00 |
| Upper explosion limit (%) [4] | 15.0 | 75.00 |
| Adiabatical Flame temperature (K) (Stochiometric mixtures) [5, 6] | 2.2 | 2.40 |
| Lower Wobbe number (MJ m−3) [6, 7] | 48.2 | 40.08 |
| Upper Wobbe number (MJ m−3) [6, 7] | 53.5 | 48.40 |
From these data, it is visible that according to the expectations, because of the higher adiabatical flame temperature, the burner will be warmer in case of hydrogen-natural gas mixture, than in case of pure natural gas. An other expectation that - because of the lower heating values - when hydrogen is adding to the natural gas pipeline, the thermal loading of the system will be lower. An other important data is the Wobbe-number range, as it is mentioned above. In the topic of the of the legal limits for hydrogen mixing made at the University of Colorado [8], the authors establish a general limit of 20 V/V%. According to the paper, it is a theoretical maximum, which was further reduced by the pipes and auxiliary equipment used material and their tolerances (mainly brittleness). Birkitt, Loo-Morrey, Sanchez and O'Sulliven founded the same limit in their paper [9]. At the University of California [10] authors get a significantly different result regarding the warming of gas equipment. They found that adding 10% hydrogen increases the temperature of the burner by 63% (and reduces the ignition time at the same time). From these measurements, it can be found that the experimental results are significantly different depending on where is from the gas, how is the gas burner designed, and how is the heating equipment was constructed. At this point it also visible that it could be important work to show the theoretic background of the dependency of the flame shape from the component (and component ratio) of the inlet gas, and after that show the measurement result. It has to be mentioned that after checking the references of the previous papers, it can say that if “thermal efficiency” is mentioned, it usually means a value, that is calculated according to a given standard, which can be different according to for example the geological area where it is used. It means that the theoretical calculations can be different according to the used standard.
Depending on whether the counter has a lower or upper calorific value, a lower or upper Wobbe number is distinguished. These two values are important because one of the criteria for gases flowing in gas pipes connected to households in the countries belonging to the European Union is that the lower and upper Wobbe numbers remain in a specific critical range. The EU standard only allows values between 45.66 MJ m−3 and 54.76 MJ m−3. It is the reason that initially only 23% hydrogen gas was mixed with methane, this mixture belongs to the lower Wobbe number, and even if a higher percentage measurement would yield positive results (e.g., a small decrease in efficiency), the practical implementation would encounter legal obstacles.
In this case, it is not the assumed back-ignition limit, but the legal regulation that defines the limit.
The modified Wobbe number considers the lower calorific value, but the temperature of the mixture is also included in the denominator, since the amount of energy flowing through the cross section depends on the gas temperature [11]. This solution opens up new possibilities for alternative energy storage in the form of hydrogen, and performing the calculation makes residential supply gas exchange safe even in general cases [12].
Although this study does not cover the aspects related to gas pipelines, models have recently been created in this area, which can be used to calculate the expected results even in the case of new gas mixtures [13].
2 Materials and methods
2.1 Theoretical background
Gas-fired combustion devices operate with a free-jet flame. The velocity distribution in the free jet is not uniform. Assuming that the pressure drop across the nozzle is really small and that no expansion occurs, a laminar inner core approximately 5d long is formed after the nozzle with diameter d [14]. The velocity distribution along the axis of symmetry led through this can be considered constant, and on the mantle surface of the forming cone it is close to 0. The radial velocity distribution within the flame is shown in Fig. 1. In the axial direction, the distribution decreases hyperbolically starting from the flame cone tip. The end of the free jet in the geometrical and firing technology sense is always where the flow velocity is 2 m s−1, this also called penetration depth. Looking at the volume flow along the axis, it can be said that just as the accelerated air flow in front of the nozzle did with the combustion air, the free jet forming the flame also draws in gases and gas mixtures from its surroundings. That is why the approximate line formed by a set of points with a speed is zero and is not parallel to the horizontal axis, but deviates from it by approximately 4–10° (Fig. 1). Since the pressure in the jet - apart from the dynamic component, which is negligible - is freely constant, the entrainment of the medium from the environment here is rather a consequence of friction (Fig. 2).
General velocity distribution after the nozzle with cross-section d
Citation: Pollack Periodica 2025; 10.1556/606.2024.01027
Typically, in the case of closed jet flows, it happens that the flames are not swirl-free (the flames of gas-powered cooking and heating equipment). Then, after a local velocity reduction at the outlet, an internal flow (recirculation) is created inside the flame cone, which stabilizes the flame.
Understanding the structure of the free jet flame is a great help in understanding the changes in the heating of the equipment. There are three separable parts in the free jet flame. The innermost part of the flame has a relatively low temperature and is called the inner flame cone. In the case of a stoichiometric or air-rich mixture, we find a layer on the surface of the inner cone that can be easily separated optically, since its light emission is significantly higher than that of the inner flame cone. Combustion takes place in this thin layer, so its temperature is also significant exceeds the inner flame cone. In the case when the mixture is lean (when not all gas molecules can burn during the reactions), the particles of the combustible gas pass through the inner flame cone and, as in the previous case, the so-called they arrive in a transition zone. Here they meet the mixed (or diffused) ambient air [14]. They then proceed to the third zone (burnout, also known as similarity zone) where they burn.
In Fig. 2 it is visible that the different types of free flames, more precisely expressing the effect of the speed conditions of the free jet on the geometry of the flame images. When the speed of the outgoing gas mixture (here: natural gas + air) belongs to the range below the critical Reynolds number, and the gas arrives in the combustion chamber with an excess of air, a line-like, yellowish flame is obtained upon ignition. This is a very “safe” case from the point of view of emissions, because with a good approximation (since technically we can only talk about probabilities in the case of gas molecules), here every molecule belonging to the gas mixture has found its oxygen gas pair necessary for combustion. In the event that this is not the case and the mixture is rich in fuel, the situation arises that even with perfect mixing there will be fuel molecules that do not have a combustion oxygen partner. These unburned molecules freely escape from the surface of the inner flame cone to the outer one into a free jet cone.
Flame types as a function of flame propagation speed and flow velocities (x is the location coordinate along the flame length)
Citation: Pollack Periodica 2025; 10.1556/606.2024.01027
2.2 Warming of the burner
Regardless of the type of combustion equipment, contact with flames will uniformly cause heating on the contact surfaces, however, in order to be able to judge, which surfaces are exposed to a smaller or larger heat load, the above relationships must be known. For example, it can be seen that the visible flame does not have a homogeneous temperature, and also that the relationship and balance of the flow velocities is not only responsible for the flame stability, but also for the temperature developed. A good example of this is measuring the burner temperature of a gas stove.
On the brass perforated gas rose cover plate, a thermocouple was placed through a hole, the end of which extends down into the incoming gas flow along the symmetry axis of the cylindrical inlet pipe (of the two typical constructions, this was a vertically injected device, unlike the drawing presented at the beginning of the paper, which is a horizontal shows an injection example). Another temperature measuring point is placed in the center of the inner disc of the burner itself. Therefore, in the first case, the gas inlet temperature can be measured, while in the second case, the temperature prevailing on the burner itself. The nominal output of the burner is 2.6 kW, which is also burdened by the usual loss of the equipment, flow losses and heat loss. The device was started from a cold state (room temperature), upon ignition; first the maximum burner output was measured, and then lowered it to an economy flame. During the measurement, the device pressure was kept constant, regardless of the gas or gas mixture used.
According to the results of the measurement (Fig. 3), the temperature measured at the inlet gas at nominal power was 27% lower than the burner temperature, which was in line with expectations, since during combustion the entire flame touches much more directly and is much closer to the measurement point located on the burner, and the metal burner it is also realistic that it heats up more than a gas (mixture) in flow.
Measurement systems schematic drawing
Citation: Pollack Periodica 2025; 10.1556/606.2024.01027
A more interesting observation was that the results did not meet expectations when the system was adjusted to a low flame (Fig. 4). According to the preconceptions, the heat input is also lower at lower power, so by definition the heating of the system will also be lower. However, in reality, this has two important effects working against it. On the one hand, the flame image in the case of the burner flame is smaller, and its shape differs from the axially symmetrical one due to the interposed obstacle, and it almost “encircles” the burner, so it touches it on a larger surface and is able to transfer more heat to it. Also, due to the change in geometry, the tip of the hotter outer flame cone is much closer to the burner head in the economy case; here the system also tends towards greater heat transfer. An even more important process takes place on the other side of the burner, which also helps the economy of heating: when the flame becomes smaller, the fuel fed in per unit of time decreases, along with the flow rate. As the flow rate decreases, the flow slows down, which cools the burner due to its speed, so the temperature actually increases due to the slowing down of the inflow.
Flame height in case of natural gas and 23 V/V% hydrogen
Citation: Pollack Periodica 2025; 10.1556/606.2024.01027
3 Results and discussion
Compared to the nominal temperature, the temperature of the economy power was about 30–33% higher both on the inlet pipe and on the burner.
It was mentioned that two types of gas mixtures has been measured. The two types of measurements were performed with 100% natural gas and 23% hydrogen and 77% methane mixture. The different properties of these gases cause changes in the combustion; one of the important elements of the research is the extent of this difference determined by practical measurement. Among the essentially different properties, calorific value and density stand out. The importance of the difference in calorific value can be explained in the section on the Wobbe number, and the difference in density - according to Bernoulli's equation - is the explanation of the different flow rate at the same pressure in the backbone, thus the increased volume flow due to the mixing of hydrogen. In addition, the density difference and the Bernoulli equation are also related to the phenomena described in the chapter detailing the geometric characteristics of the flame cone: the hydrogen flame makes a lower, sharper angle with the burner than the natural gas flame.
4 Conclusions
In the case of hydrogen mixing, the volume flow of the gas generally increases if the system pressure does not change; the value of the increase in this specific equipment is 27%. In general, the temperature of the equipment increases with conventional burners on an economy flame, with this particular equipment the increase is between 30% and 33% (depending on the measurement conditions). The calorific value of natural gas per volume is higher than the permissible 23% hydrogen - 77% methane mixture, the difference between the measurements made with the two gases in terms of the combustion temperature is less than 1%, in general it can be said that with this type of combustion the temperatures are close to they will stand up to each other in value.
Acknowledgements
Project no. RRF-2.3.1-21-2022-00009, titled National Laboratory for Renewable Energy has been implemented with the support provided by the Recovery and Resilience Facility of the Europian Union within the framework of Programme Széchenyi Plan Plus.
Acknowledgment goes to Gábor Bassa and her colleagues, whose book “Burning in Flow” [14] (in Hungarian) helped to prepare the figures.
References
- [1]↑
M. Sun, X. Huang, Y. Hu, and S. Lyu, “Effects on the performance of domestic gas appliances operated on natural gas mixed with hydrogen,” Energy, vol. 244, pp. 1–15, 2022.
- [2]↑
Y. Zhao, V. McDonell, and S. Samuelsen, “Influence of hydrogen addition to pipeline natural gas on the combustion performance of a cooktop burner,” Hydrogen Energy, vol. 44, pp. 12239–12253, 2019.
- [3]↑
I. Staffel, “The energy and fuel data sheet,” University of Birmingham, W1P1-Revision 1, 2011, pp. 1-22. [Online]. Available: claverton-energy.com/wordpress/wp-content/uploads/2012/08/the_energy_and_fuel_data_sheet1.pdf. Accessed: Nov. 22, 2023.
- [4]↑
I. Szunyog and A. B. Szombati-Galyas, “The effect of biomethane and hydrogen on natural gas in regulated gas appliances” (in Hungarian), in Proceedings of the Miskolc IPW – IV. Sustanaible Raw Materials International Project Week, Session 12, Miskolc, Hungary, November 25–27, 2020, pp. 1‒10.
- [5]↑
G. Federer-Kovacsne, H. Al Khalaf, E. Yalman, and N. A. Mohammed, “Api cement behavior under thermal conditions, Part I, 120 °C test,” Műszaki Földtudományi Közlemények, vol. 89, no. 2, pp. 189–202, 2020.
- [6]↑
G. Federer-Kovácsné, H. Al Khalaf, E. Yalman, and N. A. Mohammed, “Causes and mitigation of trapped annular pressure,” Műszaki Földtudományi Közlemények, vol. 89, no. 2, pp. 164–176, 2020.
- [7]↑
Y. Zhao, V. McDonell, and S. Samuelsen, “Experimental assessment of the combustion performance of an oven burner operated on pipeline natural gas mixed with hydrogen,” Int. J. Hydrogen Energy, vol. 44, no. 47, pp. 26049–26062, 2019.
- [8]↑
B. C. Erdener, B. Gergi, O. J. Guerra, A. L. Chueca, K. Pambour, C. Brancucci, and B. M. Hodge, “A review of technical and regulatory limits for hydrogen blending in natural gas pipelines,” Int. J. Hydrogen Energy, vol. 48, no. 14, pp. 5595–5617, 2022.
- [9]↑
K. Birkitt, M. Loo-Morrey, C. Sanchez, and L. O’Sullivan, “Materials aspects associated with the addition of up to 20 mol% hydrogen into an existing natural gas distribution network,” Int. J. Hydrogen Energy, vol. 46, no. 23, pp. 12290–12299, 2021.
- [10]↑
Y. Zhao, V. McDonell, and S. Samuelsen, “Experimental assessment of the combustion performance of an oven burner operated on pipeline natural gas mixed with hydrogen,” Int. J. Hydrogen Energy, vol. 44, no. 47, pp. 26049–26062, 2019.
- [11]↑
M. A. Nemitallah, A. A. Abdelhafez, A. Ali, I. Mansir, and M. A. Habib, “Frontiers in combustion techniques and burner designs for emissions control and CO2 capture: A review,” Int. J. Energy Res., vol. 43, no. 14, pp. 7790–7822, 2019.
- [12]↑
O. Savchenko, V. Zhelykh, Y. Yurkevych, S. Saphoval, and K. Kozak, “Using vortex tube for decreasing losses of natural gas in engineering systems of gas supply,” Pollack Period., vol. 13, no. 3, pp. 241–250, 2018.
- [13]↑
G. Bencsik, I. Haber, B. Naili, and I. Szabo, “Building thermal capacity for peak shifting, based on PV surplus production,” Pollack Period., vol. 16, no. 2, pp. 117–123, 2021.
