Laminar burning velocities of methane-hydrogen-air mixtures

Abstract

In a future sustainable society, hydrogen is likely to play an important role as an energy carrier. In an EET-project called Greening of Gas (VG2) the transition path from pure natural gas towards the use of mixtures containing more and more hydrogen is investigated. The research carried out at the TU/e is focused on the safety of burner devices. A crucial parameter for the safety of burner devices is the laminar burning velocity. In this thesis the laminar burning velocity of methane-hydrogen mixtures is experimentally determined and compared to numerical data using several combustion reactionmechanisms. An asymptotic theory for stoichiometric methane hydrogen flames is presented. This theory is validated with the experimental and numerical data. To measure the laminar burning velocity accurately the heat flux burner is used, which is developed previously at TU/e. Based on the earlier works of van Maaren and Bosschaart the heat flux method is further analysed in this thesis. This analysis results in a better understanding of several aspects of themethod. For example it is shown that the influence of the heating jacket is negligible when using a temperature difference of at least 30 K between the unburnt gas temperature and the temperature of the heating jacket should be maintained. Furthermore, it is not likely that the burner surface influences the heat flux experiments in the presented measurement range. However when higher unburnt gas temperatures will be used this influence should be regarded. In the present research, three sets of laminar adiabatic burning velocities have been measured and presented using 95% confidence error intervals. The first set consists of hydrogen-oxygen-nitrogen mixtures at various fuel equivalence ratios and several nitrogen dilutions. The second set of measurements deals with methane-hydrogen-air mixtures at various fuel equivalence ratios and hydrogen contents up to 40%. The last set of measurements show a glimpse towards gas turbine situations. Here the unburnt gas temperature is increased from298 K up to 420 K for methane-hydrogen-airmixtures. The laminar burning velocity measurement data of hydrogen-oxygen-nitrogen mixtures, show significant differences with experimental results of other authors. This discrepancy is probably related to the non-linear stretch correction performed by them. The differences between the combustion reaction mechanisms and the heat flux data show significant differences in the performance of the methane based combustion reaction mechanisms in the case of hydrogen-oxygen-nitrogenmixtures. Especially the commonly used GRI-mechanism deviates from the experimental data. Remarkably the performance of the methane based SKG03 mechanism is comparable or even better compared to hydrogen based mechanisms for fuel lean flames to slightly rich hydrogen-oxygennitrogen flames. Generally, the hydrogen based kinetic mechanisms perform quite well for the investigated parameter range; especially the Konnov mechanism. When comparing the measurements of the laminar burning velocities at ambient conditions as well as increased unburnt gas temperatures of methane-hydrogen-air mixtures with numerical combustion mechanisms it is shown that both the SKG03 mechanism and the GRI-mechanism perform very well. Experimental data of the laminar burning velocities of methane-air flames show that the measurements of Bosschaart give comparable results with the present measurements. Regrettably experimental data of methanehydrogen- air flames is scarce; the data of Halter et al. show comparable results. In order to get more insight in the basic properties describing methane-hydrogen-air flames, the asymptotic theory of Peters and Williams for stoichiometric methane-air flames is adapted to stoichiometricmethane-hydrogen-air flames. This theory is validated both with experiments performed using the heat flux burner and numerical simulations using CHEM1D. With this theory for stoichiometric flames the laminar burning velocity as a function of the hydrogen content can be predicted qualitatively even for higher pressures and temperatures. The resulting equations show that the driving force for the increase in burning velocity of a methane-hydrogen flame is the increase in temperature difference between the inner layer temperature and the adiabatic flame temperature.