Periodic Reporting for period 1 - HYDROGENATE (Hydrogen-Based Intrinsic-Flame-Instability-Controlled Clean and Efficient Combustion)
Berichtszeitraum: 2022-06-01 bis 2024-11-30
Chemical energy carriers will be play an essential role for future energy systems, where the generation and use of renewable energy do not always occur at the same time or location, necessitating long-term storage and long-distance transport of energy. Hydrogen-based energy carriers, such as hydrogen and ammonia, are particularly promising in this regard. Their utilization by combustion-based energy conversion offers many advantages, including versatile applications for heat and power, robust and flexible technologies, and suitability for a continuous energy transition. However, combustion of hydrogen and ammonia poses significant challenges. Under technically relevant conditions, both fuels exhibit intrinsic thermo-diffusive instabilities (very different from the often-discussed thermo-acoustic instabilities), which can increase burn rates by an impressive factor of three to five. Ignoring these instabilities makes computational design impossible. While linear theories exist, there is limited understanding of the more relevant non-linear regime, and aside from some data and observations, virtually nothing is known about the interactions between intrinsic flame instabilities (IFI) and turbulence.
This project aims to conduct a rigorous analysis of new data from simulations and experiments on pure H2 and NH3/H2 blends to achieve a quantitative understanding of these critical aspects. Based on this analysis, a novel modeling framework with uncertainty estimates will be developed. The key hypothesis is that combustion processes involving hydrogen-based fuels can be optimized by selectively weakening or promoting IFI. This instability-controlled combustion approach could enhance efficiency, emissions performance, stability, and fuel flexibility across various combustion devices such as spark-ignition engines, gas turbines, and industrial burners. Guided by the developed knowledge and tools, this concept will be demonstrated both computationally and experimentally.
This project aims to conduct a rigorous analysis of new data from simulations and experiments on pure H2 and NH3/H2 blends to achieve a quantitative understanding of these critical aspects. Based on this analysis, a novel modeling framework with uncertainty estimates will be developed. The key hypothesis is that combustion processes involving hydrogen-based fuels can be optimized by selectively weakening or promoting IFI. This instability-controlled combustion approach could enhance efficiency, emissions performance, stability, and fuel flexibility across various combustion devices such as spark-ignition engines, gas turbines, and industrial burners. Guided by the developed knowledge and tools, this concept will be demonstrated both computationally and experimentally.
Direct numerical simulations (DNS) resolve all relevant scales of flame structure and turbulence, enabling detailed assessment and modeling of intrinsic flame instabilities under controlled conditions. Through the project, one of the most comprehensive data bases relating to thermodiffusively unstable flames has been generated using DNS. It covers a wide range of equivalence ratios, temperatures, pressures, and (where applicable) fuel mixture compositions. Furthermore, four different configurations are embodied: two-dimensional laminar flames, three-dimensional laminar flames, three-dimensional turbulent flames, and three-dimensional flame kernels. The generation of this database is paramount for the following data analysis and, hence, for successful progress within the project.
Using part of this data base, the analysis revealed that IFIs have a leading order effect on the flame dynamics and that these effects are even amplified and prevail over a large range of Karlovitz and Reynolds numbers. Most noteworthy, it was demonstrated that the total impact of reactivity, referred to as the stretch factor, is found to increase monotonically with increasing Karlovitz number. This results from an increase in the mean equivalence ratio within the flame due to turbulent strain, which compensates for the reduction of equivalence ratio fluctuations at high Karlovitz numbers. On the other hand, an increase in Reynolds number increases the flame wrinkling but does not change the local reactivity further. This understanding will be crucial for the upcoming modeling of IFI.
Novel methodologies were developed to decompose the velocity field to uncover turbulence statistics unpolluted by dilatation and to identify the general effect of turbulence on the tangential strain rate of premixed flames. From the analysis of the DNS data, it was shown that the tangential strain rate of premixed flames is consistent with incompressible flows and independent of the Lewis number and the reaction progress variable. This remarkable finding indicates that models of the tangential strain rate developed based on incompressible flows also apply to premixed flames with thermodiffusive instabilities, and only the solenoidal turbulence needs to be considered for modeling.
Using part of this data base, the analysis revealed that IFIs have a leading order effect on the flame dynamics and that these effects are even amplified and prevail over a large range of Karlovitz and Reynolds numbers. Most noteworthy, it was demonstrated that the total impact of reactivity, referred to as the stretch factor, is found to increase monotonically with increasing Karlovitz number. This results from an increase in the mean equivalence ratio within the flame due to turbulent strain, which compensates for the reduction of equivalence ratio fluctuations at high Karlovitz numbers. On the other hand, an increase in Reynolds number increases the flame wrinkling but does not change the local reactivity further. This understanding will be crucial for the upcoming modeling of IFI.
Novel methodologies were developed to decompose the velocity field to uncover turbulence statistics unpolluted by dilatation and to identify the general effect of turbulence on the tangential strain rate of premixed flames. From the analysis of the DNS data, it was shown that the tangential strain rate of premixed flames is consistent with incompressible flows and independent of the Lewis number and the reaction progress variable. This remarkable finding indicates that models of the tangential strain rate developed based on incompressible flows also apply to premixed flames with thermodiffusive instabilities, and only the solenoidal turbulence needs to be considered for modeling.
Within the project, one of the largest DNS databases of H2-based flames on Europe’s fastest supercomputers was generated. The data can be summarized as follows:
• 2D laminar flames:
o H2/air, with/without Soret effects, f = 0.3-0.5 T = 300-400 K, p = 1-20 atm
o H2/air, EGR fractions 0-20%, f = 0.3-0.5 T = 400 K, p =1-6 atm
o CH4/H2/air, H2 fractions 55-100%, f = 0.35-0.65 T = 300 K, p = 1-32 atm
• 3D laminar flames:
o H2/air, f = 0.4 p = 1 atm, T = 298 K, incl. NOx chemistry.
• 3D turbulent flame kernels
o H2/air, f = 0.4 p = 40 atm, T = 800 K
• 3D turbulent slot jet flames
o H2/air, Re=5,500; 11,000; 22,000, f = 0.4 T = 298 K, Ka=25, 1 atm
o H2/air, Ka=25; 50; 230, f = 0.4 T = 298 K, Re=11,000, 1 atm
o H2/air, Re=11,000, f = 0.4 T = 298 K, p = 1; 10 atm.
Each simulation required careful planning and massive effort to produce and analyze. The simulations have been performed on the supercomputers SuperMUC (LRZ), JUWELS (JSC), and CLAIX (RWTH Aachen University) and require more than 400 million core hours of computing time. The database enables the detailed analysis of intrinsic flame instabilities and their effects within this project's scope and beyond. A visualization of the temperature field in turbulent premixed hydrogen jet flames at Reynolds numbers ranging from 5,500 (left) to 22,000 (right) is displayed in the Figure below.
The second breakthrough involves progress in understanding the Reynolds and Karlovitz number effects in lean premixed hydrogen/air flames. These flames exhibit synergistic interaction between turbulence and thermodiffusive instabilities, which have a leading order effect on flame dynamics and drastically increase flame speed. In the context of this project, massive-parallel high-fidelity simulations enabled precise insight into the flame structure and the turbulence/chemistry interactions. Experimental investigations suggested that this synergistic interaction diminishes at high Reynolds numbers due to enhanced penetration of turbulence into the flame. However, the conducted analysis revealed that synergistic interactions persist under these conditions, which has important implication for the modelling. Next, a novel analysis methodology was employed, featuring a rigorous decomposition of the turbulent flame speed into contributions from flame wrinkling and variations of local reactivity. It was shown that the instability-related terms of the flame surface area generation and reaction rates feature different trends with increasing Karlovitz number. This finding is highly relevant for the modeling and prediction of hydrogen flames.
• 2D laminar flames:
o H2/air, with/without Soret effects, f = 0.3-0.5 T = 300-400 K, p = 1-20 atm
o H2/air, EGR fractions 0-20%, f = 0.3-0.5 T = 400 K, p =1-6 atm
o CH4/H2/air, H2 fractions 55-100%, f = 0.35-0.65 T = 300 K, p = 1-32 atm
• 3D laminar flames:
o H2/air, f = 0.4 p = 1 atm, T = 298 K, incl. NOx chemistry.
• 3D turbulent flame kernels
o H2/air, f = 0.4 p = 40 atm, T = 800 K
• 3D turbulent slot jet flames
o H2/air, Re=5,500; 11,000; 22,000, f = 0.4 T = 298 K, Ka=25, 1 atm
o H2/air, Ka=25; 50; 230, f = 0.4 T = 298 K, Re=11,000, 1 atm
o H2/air, Re=11,000, f = 0.4 T = 298 K, p = 1; 10 atm.
Each simulation required careful planning and massive effort to produce and analyze. The simulations have been performed on the supercomputers SuperMUC (LRZ), JUWELS (JSC), and CLAIX (RWTH Aachen University) and require more than 400 million core hours of computing time. The database enables the detailed analysis of intrinsic flame instabilities and their effects within this project's scope and beyond. A visualization of the temperature field in turbulent premixed hydrogen jet flames at Reynolds numbers ranging from 5,500 (left) to 22,000 (right) is displayed in the Figure below.
The second breakthrough involves progress in understanding the Reynolds and Karlovitz number effects in lean premixed hydrogen/air flames. These flames exhibit synergistic interaction between turbulence and thermodiffusive instabilities, which have a leading order effect on flame dynamics and drastically increase flame speed. In the context of this project, massive-parallel high-fidelity simulations enabled precise insight into the flame structure and the turbulence/chemistry interactions. Experimental investigations suggested that this synergistic interaction diminishes at high Reynolds numbers due to enhanced penetration of turbulence into the flame. However, the conducted analysis revealed that synergistic interactions persist under these conditions, which has important implication for the modelling. Next, a novel analysis methodology was employed, featuring a rigorous decomposition of the turbulent flame speed into contributions from flame wrinkling and variations of local reactivity. It was shown that the instability-related terms of the flame surface area generation and reaction rates feature different trends with increasing Karlovitz number. This finding is highly relevant for the modeling and prediction of hydrogen flames.